JBCS

INTRODUCTION

Electrospinning is a well-established and versatile technique for producing nanofibers from polymer solutions by applying electrostatic forces. This method has gained prominence in nanomaterials manufacturing due to its ability to generate fibers with high mechanical strength, flexibility, simple processing, and relatively low cost.^1-4 The resulting nanofibrous structures exhibit high porosity and an extensive surface area, properties that enhance their potential for applications in drug delivery, biosensing, and filtration.^5-10 Among the various polymers utilized in electrospinning, polyamide 6 (PA-6) has been extensively explored due to its outstanding mechanical resistance, thermal stability, flexibility, biocompatibility, and biological resistance.^11,12 These attributes make PA-6 nanofibers an ideal platform for advanced biomedical applications, including controlled drug delivery and photodynamic therapy (PDT).

One of the main advantages of electrospinning in PDT applications is its ability to immobilize photosensitizers (PSs) within nanofibrous matrices, providing controlled and localized drug delivery.^13-16 Electrospun fibers offer a high surface area-to-volume ratio, enhancing PS loading capacity while promoting efficient oxygen diffusion, an essential factor for PDT efficacy.¹⁷ Furthermore, the tunability of nanofiber architecture enables the precise modulation of drug release kinetics, ensuring sustained availability of the PS at the target site while mitigating premature degradation or systemic diffusion.¹⁵ This approach enhances therapeutic outcomes by improving PS stability, prolonging its photodynamic activity, and reducing off-target effects.^18,19 However, despite these advantages, challenges remain, including optimizing fiber composition to balance biodegradability and mechanical stability, ensuring homogeneous PS distribution within the fibers, and addressing potential issues of aggregation that could impact therapeutic performance.^20-22 Future research should focus on refining electrospinning parameters and developing hybrid nanofibrous systems incorporating functional biomaterials to enhance biocompatibility, improve PS retention, and maximize PDT efficacy.^23,24

PDT has emerged as a promising therapeutic strategy for treating cancer and microbial infections by utilizing light-activated PSs to induce reactive oxygen species (ROS) generation.¹⁷ Upon light absorption at specific wavelengths, photosensitizers transition to an excited state, initiating photochemical reactions classified as type I or II. Type I reactions involve direct electron transfer, generating radical species such as hydroxyl radicals (OH•) and superoxide anions (O₂^*-), while type II reactions primarily lead to the formation of singlet oxygen (¹O₂), a highly reactive species responsible for cytotoxic effects.²³ Among these pathways, singlet oxygen production via type II reactions is widely recognized as the primary mechanism for PDT-induced tumor cell apoptosis and necrosis.²⁴

PDT depends on administering a photosensitizer, which selectively accumulates in diseased tissue, and then activating it with light to produce cytotoxic reactive oxygen species. Over recent decades, various classes of PSs have been studied, each with unique physicochemical and therapeutic properties. Porphyrins and related derivatives constitute the first generation of PDT agents. They exhibit strong absorption bands in the visible spectrum and have been approved for the treatment of certain cancers and skin conditions. However, they have limitations such as prolonged skin photosensitivity, uneven distribution, and low singlet oxygen quantum yields. Phthalocyanines, the second generation of PSs, exhibit higher molar absorptivity and absorb light in the red to near-infrared range (600-700 nm), allowing for deeper tissue penetration. They are more photostable and produce ROS more efficiently than porphyrins, although their hydrophobic nature often necessitates encapsulation within delivery systems to address poor solubility and aggregation. Chlorins and bacteriochlorins are also extensively researched due to their strong absorption within the therapeutic window (650-750 nm) and advantageous photophysical properties. Chlorin e6, for instance, has been used in both cancer and antimicrobial PDT. Despite these advantages, challenges such as large-scale synthesis, stability issues, and cost barriers remain obstacles to widespread clinical use.^25-30

Methylene blue (MB), a phenothiazine-derived dye, has garnered significant interest as a photosensitizer in PDT due to its strong absorption in the red and near-infrared spectral regions (630-680 nm), low toxicity, and well-established pharmacological profile.^31,32 MB has demonstrated effectiveness in PDT against melanoma, adenocarcinoma, and other tumor types, as well as in antimicrobial PDT against a range of bacterial and fungal pathogens.³³ However, the hydrophilic nature of MB and tendency to aggregate in aqueous environments limit its bioavailability and overall efficacy, underscoring the need for innovative delivery strategies to enhance its therapeutic potential. The incorporation of MB into nanofiber matrices, such as electrospun PA-6, represents a promising solution for overcoming these challenges, enabling localized and sustained PS release with improved tissue penetration and photodynamic response.³⁴

The present study aims to investigate the structural and morphological characteristics of electrospun PA-6 membranes loaded with MB and assess their potential as PDT agents in tumor and non-tumor cell models. By utilizing the unique properties of PA-6 nanofibers, this approach aims to enhance the stability, bioavailability, and controlled release of MB, thereby improving the efficacy of PDT for superficial tumor treatment in tissues such as the skin, nails, and mucosal surfaces. The findings of this study contribute to the continuous development of nanostructured biomaterials for targeted and efficient photodynamic applications, opening new possibilities for novel therapeutic strategies in oncology and infectious disease management.^35,36

 

EXPERIMENTAL

Reagents and chemicals

All reagents used in the experiments were of analytical grade. Formic acid (85% P.A., A.C.S.) and absolute ethyl alcohol (99.5%) were obtained from Synth (Brazil). Polyamide 6 (PA-6), with a density of 1.084 g cm^-3 and a molecular weight (Mw) of 12,000 g mol^-1, was supplied by Rhodia (Brazil). 1,3-Diphenylbenzofuran (DPBF) was procured from Sigma-Aldrich (USA), and methylene blue (MB) hydrate (purity ≥ 82%) was purchased from Neon (Brazil).

Design of an experiment for electrospinning

A fractional factorial (2^5-1) design was employed to establish a quantitative relationship between reactive oxygen species (singlet oxygen) and spinning parameters. Additionally, it aimed to establish a quantitative relationship between rheology and spinning parameters. Accordingly, five parameters were selected: polymeric solution concentration, flow rate, PS concentration, needle-to-collector distance, and applied voltage. PA-6 solutions of 18 and 30 wt.% were used to study how polymer concentration affects electrospinning. The 18% solution represents a lower viscosity system, which tends to produce thinner fibers. In contrast, the 30% solution, with its higher viscosity, promotes the formation of more uniform and continuous fibers with larger diameters. Beyond viscosity, polymer concentration also affects electrical conductivity, surface tension, and solvent evaporation rate, which collectively influence jet stability and fiber solidification during the electrospinning process. Therefore, comparing these two concentrations provides insights into how these parameters interact and influence fiber morphology, contributing to the optimization of electrospinning conditions. The selected MB concentrations were used to evaluate how different levels of this PS influence the in vitro release profiles, allowing for the identification of optimal conditions for controlled release and ensuring bioavailability for therapeutic applications. The objective was to investigate the retention of MB in the fibers and the impact of different PS concentrations on the stability of PA-6 fibers, ensuring adequate levels for therapeutic efficacy and clinical applicability.

In the adopted experimental design, 19 experiments were initially conducted (identified as 1 to 19), as detailed in Table 1S (Supplementary Material). Each condition was performed in triplicate to verify the reproducibility of the process. To minimize systematic errors, all experiments were conducted in a random order. The ideal conditions were defined and optimized using a proposed algorithm for regression and analysis of variance (ANOVA). The Octave program, version 6.4.0 (Free Software Foundation, USA) was used to perform the calculations. For viscosity, it was observed that effects 1 (variable 1) and 3 (variable 3) were the most significant (Figure 1Sa, Supplementary Material). The probability plot in Figure 1Sb confirms that these two factors are the most relevant, as they are positioned far from zero on the x-axis. The effects of variables located close to zero were neglected and removed from the factorial design.

After identifying PA-6 and PS concentrations as the most relevant variables, the next step was to determine the optimal working conditions. The ANOVA results for rheology are summarized in Table 2S (Supplementary Material). The ratio of the mean square of the regression to the mean square of the residual produced an F-calculated value (64.30) that exceeded the F-tabulated value (3.64) at the 95% confidence level, indicating a good regression model. However, for the ratio of the mean square of the lack of fit to the mean square of the pure error, the F calculated value (81.58) was much higher than the F tabulated value (3.74), indicating a significant lack of fit. Despite this, the model was retained due to its exploratory nature and simple structure, which facilitates interpretation and aligns with the goals of the study. As shown in Figure 2S (Supplementary Material), viscosity increased with polymer concentration. The PA-6 and MB concentrations that contributed most to this increase were 30% (m/v) with 28 mM MB and 30% (m/v) with 10 mM MB, respectively.

Figure 3Sa (Supplementary Material) shows the relative percentage of calculated effects, displaying the individual values of the significant factors affecting singlet oxygen release, represented by bars. MB concentration (variable 3) and PA-6 concentration (variable 1) were the most influential. The probability plot in Figure 3Sb confirmed their importance, as they are located at a considerable distance from the x-axis zero line. The vertical blue line indicates zero, and effects 1 and 3 lie outside the confidence interval, marked by vertical red lines. Hence, to maximize singlet oxygen formation by nanofibers, it is necessary to decrease the conditions of variable 1 and increase those of variable 3.

Optimal working conditions were determined after identifying PA-6 and MB concentrations as key variables. The ROS production results assessed by ANOVA are shown in Table 3S (Supplementary Material). The regression model provided an adequate fit to the data, as indicated by the F test: F calculated (17) > F tabulated (3.65) at a 95% confidence level. The lack-of-fit test also confirmed the suitability of the model. Figure 4S (Supplementary Material) demonstrates the effect of PA-6 (%) and PS (mM) concentrations on ROS production. It was found that increasing polymer concentration decreases singlet oxygen production, while increasing PS concentration enhances ¹O₂ generation. The optimal values were determined to be 18 wt.% for PA-6 and 28 mM for PS. The region with 30 wt.% and 28 mM also yielded satisfactory results.

Preparation of electrospinning solution

PA-6 pellets were dissolved in 8 mL of formic acid at room temperature with magnetic stirring for 24 h to ensure complete dissolution. The polymer concentration was adjusted between 18 and 30 wt.%, representing lower- and higher-viscosity regimes, allowing for a comparison of fiber formation under different electrospinning conditions. Methylene blue (MB) was dissolved in formic acid at two concentrations, 10 mM (low concentration, LC) and 28 mM (high concentration, HC). These levels were chosen based on preliminary screening experiments that established the operational limits for stable jet formation and consistent fiber deposition. The 28 mM concentration, although not part of the formal factorial design, was included to explore the upper limit of MB incorporation into the PA-6 matrix, since higher concentrations are known to reduce solution spinnability and promote dye aggregation.^1,2 The MB solution was mixed with PA-6 and loaded into a 10 mL glass syringe with a 20-gauge needle. Electrospinning was performed onto 15 × 15 cm² aluminum foil-covered collector plates in a controlled environment. The working distance (10-20 cm), applied voltage (20-28 kV), and flow rate (5-15 mg min^-1) were varied within ranges previously reported to ensure stable fluidic regimes for continuous fiber formation.

Measurement of solution viscosity

Viscosity measurements were conducted at room temperature using a rotational rheometer (Viscotester IQ, Thermo Haake, Germany). The measurements were performed using a double-cone sensor in controlled rate (CR) mode. The shear rate was increased from 0 to 1000 s^-1 over 500 s, maintained for 5 min, and then decreased to 0 s^-1 over another 5 min. Rheological data were extracted from the flow curves using the Haake RheoWin Data software, version 4.63.0003 (Thermo Fisher Scientific, Germany, 2016).

Scanning electron microscopy (SEM)

The morphology of the electrospun fibers was examined using scanning electron microscopy (SEM) (FEI, model Quanta 250, FEI Company, USA). Fiber samples were mounted on metal stubs using carbon double-sided adhesive tape and sputter-coated with a 10 nm gold layer. SEM images were acquired using a backscattered electron detector at a working distance (WD) of 10 mm, an accelerating voltage of 10 kV, and a magnification of 5000×. Fiber diameter analysis was performed using ImageJ software, version 1.54f (National Institutes of Health, USA, 2025).

Contact angle measurements

Water static contact angle measurements were conducted using an SEO (model Phoenix MT) contact angle goniometer at room temperature (20 °C). Each sample was measured in three independent trials. Distilled water (10 µL) was carefully placed onto the scaffold surface, and the contact angle was determined using the Surfaceware software package (SEO, South Korea). The mats were attached to glass slides to hold them flat during analysis.

Determination of thermal properties

Thermogravimetric analysis (TGA) was performed to evaluate the temperature-dependent mass variation resulting from thermal decomposition. The thermograms were obtained using a Q500 analyzer (TA Instruments, USA). The analyses were performed under a nitrogen flow rate of 60 mL min^-1, with an initial temperature of 25 °C, a final temperature of 600 °C, and a heating rate of 10 °C min^-1. Differential scanning calorimetry (DSC) analyses were performed using a Netzsch 204 F1 Phoenix DSC apparatus under a nitrogen atmosphere with a flow rate of 50 mL min^-1. Approximately 10 mg of fiber samples were first heated from 30 to 300 °C at a rate of 10 °C min^-1 and held at this level for 5 min. The temperature was then reduced to -20 °C at a rate of 10 °C min^-1, held for 5 min, and subsequently increased to 300 °C at a rate of 10 °C min^-1 to obtain the second melting cycle.

Detection and quantification of methylene blue release

To determine the MB release, fibers containing PS at concentrations of 10 mM (LC) and 28 mM (HC) were cut into smaller pieces, each weighing 10 mg. These samples were placed on a mesh mounted in the cuvette to prevent fibers from floating in the beam path, which could interfere with and distort the ultraviolet (UV) signal. Stock solutions at 10 and 28 mmol L^-1 MB were prepared, and the MB solution was dropped onto 10 × 10 mm white membranes and air-dried. The amount of MB released was measured using an ultraviolet-visible (UV-Vis) spectrophotometer at a wavelength of 665 nm. All solutions were prepared with distilled water, and absorption scans were recorded at room temperature (25 °C) with stirring at 500 rpm at 1-min intervals over a total of 1 h. To analyze the MB release, mathematical models were used to understand the release mechanism.^37,38 Equation 1 is a zero-order model describing a concentration-independent release.

where Q is the amount of MB released at time t, Q₀ is the initial MB amount, and K₀ is the zero-order release constant.

The first-order model (Equation 2) describes the diffusion or dissolution processes of concentration-dependent release:

where K₁ is the first-order release constant.

The Higuchi model is based on a Fickian diffusion (Equation 3):

where K_H is Higuchi release constant.

The Korsmeyer-Peppas model (Equation 4) is related to release in a polymeric matrix, and it is defined as:

where is the fraction of MB released at time t, K^KP is Korsmeyer-Peppas release constant, and n is the release exponent, when n ≤ 0.5 indicates Fickian diffusion and 0.5 < n < 1.0 corresponds to anoumalous transport.

Detection and quantification of reactive oxygen species (ROS)

The generation of reactive oxygen species, specifically singlet oxygen (¹O₂), was determined using the DPBF method, as described in the literature.^39,40 First, a DPBF stock solution with a concentration of 50 µM was prepared by dissolving DPBF powder in absolute ethyl alcohol at room temperature (25 °C). Then, 15 µL of DPBF stock solution was transferred to a Falcon tube, and absolute ethyl alcohol was added to complete a total volume of 3 mL. This solution was transferred to a quartz cuvette, followed by the addition of nanofibrous mats (10 mg). The photosensitizer was excited using a 640 nm light-emitting diode (LED) lamp (10 mW cm^-2) directly in the cuvette for 30 min. Subsequently, the decrease in DPBF absorbance (417 nm) was recorded using a UV-Vis spectrometer. The group without nanofibrous mats served as the control.

To determine the rate of DPBF degradation, a one-phase kinetic decay model was applied to the variation of molar concentration (y) over time, as described by Equation 5:

where K is the rate of DPBF degradation (min^-1), t represents time (in min), y₀ is the initial concentration, and y is the concentration at infinity.

Cell culture of HEK293 and SW480 cell lines and photodynamic therapy tests

HEK293 and SW480 cell cultures were maintained in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and penicillin-streptomycin. A total of 10⁴ cells per well were plated and incubated at 37 °C under 5% CO₂ for 24 h. Electrospun fibers, cut into circular shapes, were added to the cell cultures, followed by an additional 24-h incubation at 37 °C under 5% CO₂. The next day, electrospun fibers were removed, and one plate was subjected to irradiation using a homemade LED irradiation system developed in our laboratory, equipped with an array of 24 LEDs emitting at 660 nm. The system ensured uniform light distribution over the entire plate, with a fluence rate of 36.5 mW cm^-2, and irradiation was maintained for 10 min. Another plate was kept in the dark as a control.

After treatment, plates were incubated at 37 °C under 5% CO₂ for 24 h. Subsequently, cells were washed with 200 µL PBS and treated with MTT reagent (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide) at a concentration of 0.3 mg mL^-1. After 3 h of incubation, the MTT medium was removed, and the wells were treated with dimethyl sulfoxide (DMSO). Absorbance readings were recorded at 570 nm using a microplate reader (Biochrom, Model EZ Read 400, USA).

 

RESULTS AND DISCUSSIONS

Determination of the viscosity of pure and PS-loaded electrospinning PA-6 solutions

The viscosity of a polymer solution is directly proportional to its concentration. It plays a crucial role in determining the diameter of electrospun fibers, which in turn significantly affects the release kinetics of photosensitizers (PSs). The viscosity of PA-6 electrospinning solutions, both with and without PS, was analyzed to assess this relationship.

Figure 1 presents the electrospinning flow curves of PS-free and PS-loaded polymer solutions, prepared with either 18 or 30 wt.% PA-6 and MB concentrations of 10 and 28 mM. All solutions exhibited Newtonian behavior, indicating that stress increases linearly with increasing shear rate. The mechanical resistance of the formulations depends on PA concentration as well as the presence of MB. Solutions containing 30 wt.% PA exhibited higher resistance compared to the 18 wt.% formulation; however, the addition of MB diminished this resistance. This suggests that MB molecules, at both low and high concentrations, interfere with polymer-polymer interactions. Interestingly, the hysteresis of the formulations with MB-LC was low, suggesting that low MB concentrations integrate more uniformly within the PA-6 matrix.⁴¹ Specifically, the viscosity values for the 18 wt.% PA-6 solutions loaded with 10 mM MB and 30 wt.% PA-6 solutions loaded with 10 and 28 mM MB were 0.1428, 0.7163, and 0.7185 Pa s, respectively. In contrast, PS-free polymer solutions had viscosities of 0.231 and 1.177 Pa s for the 18 and 30 wt.% PA-6 solutions, respectively (Figure 1).

 

 

This reduction in viscosity can be attributed to the hydrophobic interactions between MB molecules and PA chains, which disrupt chain entanglements and thereby reduce the overall resistance to flow. Such rheological changes directly influence fiber formation: lower-viscosity systems (18 wt.% PA-6 with MB) tend to produce thinner fibers with higher surface area-to-volume ratios, favoring rapid MB release. Conversely, higher-viscosity systems (30 wt.% PA-6) yield thicker, more entangled fibers that retard MB diffusion, enabling a more controlled and sustained release profile.^42,43

From a therapeutic perspective, this diffusion-dominated release mechanism has critical implications for topical PDT. A faster release may lead to an initial burst of MB, which can enhance immediate ROS production but risks photobleaching and reduced long-term efficacy. In contrast, slower and sustained release ensures prolonged PS availability, potentially maintaining ROS generation over extended irradiation cycles and improving treatment outcomes for superficial tumors or infected tissues. Thus, tuning solution viscosity not only optimizes fiber morphology but also provides a means to balance rapid initial drug availability with sustained therapeutic action.⁴⁴

Morphological analysis of MB-electrospun PA-6 membranes

The surface morphology and ultrastructural properties of electrospun scaffolds play a crucial role in determining their functionality. To assess these characteristics, fiber diameter, diameter distribution, and bead density were evaluated using SEM analysis.

Figure 2 presents SEM images of PS-free and PS-encapsulated fibers, and their respective size distributions are shown in Figure 5S (Supplementary Material). Fibers obtained from an 18 wt.% pure PA-6 solution or those incorporating 10 mM PS, electrospun at a flow rate of 5 mg min^-1 with a fixed voltage of 28 kV, exhibited a high density of granules (Figures 2a and 2b). The inclusion of PS in 18 wt.% PA-6 fibers reduced mean fiber diameter from 353 ± 3.21 nm to 316 ± 12.16 nm (Figure 5S). This decrease in fiber diameter may be attributed to the electrostatic repulsion of the positively charged MB molecules, which affects the fiber formation process.^45,46

 

 

In contrast, fibers obtained from a 30 wt.% pure PA-6 solution displayed well-defined nanofiber morphology (Figure 2c). The fibers prepared with 30 wt.% PA-6 and PS concentrations of 10 and 28 mM, electrospun at voltages between 20 and 28 kV with a fixed flow rate of 15 mg min^-1, exhibited high homogeneity and smoothness, regardless of the applied voltage and flow rate (Figures 2d and 2e).

The increased PA-6 concentration contributed to the formation of nanofibers with minimal imperfections, as higher polymer content enhances solution viscosity and facilitates polymer chain entanglement. This improved fiber formation created three-dimensional, homogeneous networks with significantly reduced structural defects.

Comparing PS-free and MB-encapsulated fibers, no apparent morphological imperfections were observed due to PS incorporation. Moreover, the inclusion of PS increased the mean fiber diameter, reaching 458 ± 4.5 and 473 ± 12.26 nm for 30 wt.% PA-6 fibers loaded with 10 and 28 mM PS, respectively. In contrast, the mean diameter of PS-free fibers was recorded at 332 ± 14.74 nm (Figure 5S).

The reduction in fiber diameter observed with the incorporation of MB (PA-6) solutions can be attributed to the combined effects of increased solution conductivity and electrostatic interactions during electrospinning. MB, being a cationic dye, raises the ionic strength of the solution, thereby increasing its electrical conductivity. This enhancement in conductivity promotes greater charge density on the electrospinning jet, intensifying the whipping instability and resulting in stronger elongational forces that lead to thinner fibers. Additionally, positively charged MB molecules may contribute to intra-jet charge repulsion, further facilitating jet thinning. Importantly, fibers produced from 30 wt.% PA-6 solutions, both with and without MB, exhibited uniform morphology and high homogeneity across a range of applied voltages (20-28 kV) at a constant flow rate of 15 mg min^-1. This indicates that the observed changes in fiber diameter are not due to variations in jet stability or solution spinnability, but are more likely linked to the electrostatic effects induced by MB. These findings are consistent with previous reports^3,47 that correlate increased solution conductivity with reduced fiber diameters in electrospun materials.

Determination of wettability of MB-electrospun PA-6 membrane

Wettability assessment is important in biomaterial selection, as it directly affects PS diffusion and cell adhesion to fiber surfaces.⁴⁸ The contact angle (θ) varies depending on the surface hydrophilicity: a surface is hydrophobic when θ > 90° and hydrophilic when θ < 90°. A completely hydrophilic surface has θ = 0°, whereas a completely hydrophobic surface has θ = 180°.⁴⁹

The apparent contact angles measured on PA-6/MB nanofiber mats were interpreted using the Cassie-Baxter framework, which assumes a composite interface between liquid, solid, and trapped air.⁵⁰ This model describes an approximation to rationalize the effect of the fiber morphology on wettability.^50-52 PA-6 fiber samples exhibited water contact angles below 90° under all experimental conditions (Figure 3), confirming their general hydrophilic nature. A modest but statistically significant correlation was observed between fiber diameter and contact angle, with the contact angle decreasing as the fiber diameter increased (see Table 1). For fibers fabricated from 3 wt.% PA-6 solutions loaded with 10 and 28 mM MB, the contact angle decreased from 34.01 ± 0.44° to 33.19 ± 0.67°, while fiber diameter increased from 458 ± 9.5 to 473 ± 12.3 nm (p < 0.05, unpaired t-test).

 

 

 

 

Fibers from the less concentrated PA-6 solution (18 wt.%) exhibited higher surface defect density and smaller diameters than those from the 30 wt.% solution. This morphology may enhance surface roughness, affecting apparent wettability. According to the Wenzel and Cassie-Baxter models, micro- and nano-scale roughness can either amplify or reduce hydrophilicity depending on surface chemistry and topography. In this case, we suggest that increased surface smoothness at higher fiber diameters contributes to the observed decrease in contact angle.⁵³

Although the absolute differences in contact angle were small (ca. 1°), they were consistent and statistically significant. Therefore, we interpret these findings as evidence of subtle but measurable changes in surface wettability, likely linked to fiber morphology rather than intrinsic chemical hydrophilicity alone.

Thermal analysis of MB-electrospun PA-6 membranes

DSC analyses were employed to investigate the melting, crystallization, and thermodynamic behavior of the nanofibers, with a focus on the influence of PS on the semi-crystalline structure of PA-6 electrospun fibers. The structure of semi-crystalline polymers is complex and influenced by processing methods and thermo-mechanical treatments. The degree of crystallinity (X_c) is particularly sensitive to processing conditions due to molecular alignment and texture variations. These structural differences significantly affect overall material behavior, as crystalline and amorphous regions exhibit distinct physical properties. DSC consists of a heating/cooling/heating cycle, with the initial heating cycle aimed at eliminating any existing thermal history of fibers.³⁴ An observed endothermic peak at 220 °C corresponds to the melting point of pristine PA-6.

The degree of crystallinity was deduced from the melting enthalpies (ΔH_m) using the Equation 6:

where is the heat of fusion for 100% crystalline PA-6 (213 J g^-1 in this study), and w is the weight fraction of the polymer in the sample.⁵⁴ Here, we assume w = 1 since the MB content in the electrospun fibers is very low relative to the total sample mass (below 5 wt.%) and was therefore considered negligible for crystallinity estimation.

Figure 4 presents the DSC curves, and Table 2 lists the thermal characteristics of MB-electrospun PA-6 fibers. The results indicate that MB concentration influences the thermal behavior of PA-6. The glass transition temperature (T_g), which marks the transition of the amorphous phase from a glassy to a rubbery state, varies between 53 and 59 °C (data not shown), depending on MB concentration.⁵⁵ The highest T_g (59 °C) is observed in 30 wt.% PA-6 samples with 10 mM MB, suggesting that MB at this concentration restricts chain mobility due to molecular interactions. One possible explanation for this restricted mobility is the interaction of MB with polymer chains, solvent residue, and moisture absorption. During electrospinning, residual solvents or environmental moisture can act as secondary plasticizers. Due to the hydrophilic nature of PA-6, water molecules readily penetrate the polymer matrix, disrupting intermolecular hydrogen bonds and enhancing the mobility of polymer chain segments. In contrast, at 28 mM MB, T_g decreases to 53 °C, indicating a plasticizing effect that enhances chain flexibility.⁵⁶

 

 

 

 

The melting temperature (T_m), reflecting the stability and perfection of crystalline structures, varies slightly between 218 and 223 °C. The highest T_m (223 °C) is found in MB-free samples, suggesting the formation of more stable crystalline regions. MB presence slightly decreases T_m, which may be associated with disruptions in the crystalline arrangement. Similarly, the crystallization temperature (T_c) declines slightly, ranging from 177 to 181 °C, with increasing MB concentrations.⁵⁵ This suggests that MB interferes with molecular organization during cooling, delaying crystallization.

The enthalpy of fusion (ΔH_m), which provides insights into the energy required for melting, also reflects variations in crystallinity. The highest ΔH_m (82.24 J g^-1) is observed in 30 wt.% PA-6 with 10 mM MB, indicating a slight increase in crystallinity at this concentration. However, at 28 mM MB, ΔH_m decreases to 76.91 J g^-1, supporting the idea that higher MB concentrations hinder the formation of well-ordered crystalline structures.⁵⁵ This trend is further confirmed by the degree of crystallinity (X_c), which ranges from 36 to 39%. The highest crystallinity is observed at 10 mM MB, while the lowest occurs at 28 mM MB, reinforcing the hypothesis that excessive MB disrupts molecular packing and acts as a plasticizer.

TGA was conducted to evaluate the thermal stability and decomposition behavior of MB-electrospun PA-6 fibers. This technique provides insights into mass loss as a function of temperature, identifying key thermal events such as solvent evaporation, the onset of degradation, and the principal decomposition stages. Multiple factors, including polymer composition, molecular interactions, and the concentration of MB, influence the thermal behavior of PA-6.

Figure 5 illustrates the variation in mass (%) as a function of temperature, highlighting the decomposition rate of each sample. The thermal degradation of all samples occurs predominantly in a single stage between 450 and 500 °C, characteristic of the decomposition of PA-6. A minor weight loss at lower temperatures (< 200 °C) is observed in some samples, likely due to residual solvent evaporation and moisture absorption, expected given the hydrophilic nature of PA-6.⁵⁷

 

 

The presence of MB influences thermal stability. The PA-6 30% MB-LC (black curve) and PA-6 18% MB-LC (red curve) exhibit similar degradation profiles, with slightly different weight loss behaviors before primary decomposition. In contrast, the PA-6 30% MB-HC (blue curve) demonstrates a minor reduction in thermal stability, suggesting that higher MB concentrations alter the degradation pathway of the polymer. This effect may be attributed to the plasticizing role of MB, which disrupts intermolecular interactions and leads to earlier thermal decomposition.

Compared to pure PA-6 (green dashed line), the unmodified polymer presents a slightly higher onset decomposition temperature and a sharper degradation transition, indicating greater thermal stability. MB incorporation modifies thermal behavior, possibly due to interactions between MB molecules and the polymer matrix. However, the residual mass fraction at the end of the analysis remains relatively consistent across samples, suggesting that MB does not significantly contribute to the formation of char.

MB release profile

The release profiles of methylene blue (MB) from electrospun PA-6 fibers were evaluated for three formulations containing either low (LC) or high (HC) concentrations of MB. All conditions exhibited rapid release during the first 10 min, followed by a plateau, consistent with a burst release typical of hydrophilic drug-loaded nanofibers. The cumulative release reached approximately 95% of the total MB content within 60 min for the 18% PA-6 MB-LC group, while a slower release was observed for the 30% PA-6 MB-HC group (Figure 6).

 

 

To better understand the release mechanism, the release data were fitted to four kinetic models: zero-order, first-order, Higuchi, and Korsmeyer-Peppas. The cumulative average data from the replicates were used for fitting, and the release plateau was considered to be the maximum release (Q), lower than the infinite line in the Korsmeyer-Peppas and first-order models. The model parameters and goodness of fit (R²) are summarized in Table 3.

 

 

The Higuchi and Korsmeyer-Peppas models provided the best fit to the experimental data (R² > 0.98), indicating that MB release was mainly controlled by Fickian diffusion. This is further confirmed by the release exponent values (n) from the Korsmeyer-Peppas model, which were all below 0.45, aligning with a diffusion-controlled process typical of polymeric matrices. In contrast, the zero-order and first-order models showed lower R² values and were less suitable for describing MB release from the nanofibers.

These findings align with the existing literature^38,58 on electrospun hydrophilic polymer matrices, where the high surface area and porous structure facilitate the rapid diffusion of small hydrophilic molecules. Therefore, we conclude that the release of MB from PA-6 electrospun fibers follows a diffusion-dominated mechanism, with minimal contribution from polymer erosion or degradation within the tested time frame. These results indicate that MB release was more pronounced in fibers prepared from 18 wt.% PA-6 solutions. Additionally, this effect is attributed to the lower degree of entanglement between PA-6 chains.⁵⁸ Several factors may influence the rate of photosensitizer release. Studies⁵⁹ have demonstrated the effect of pH on MB release from nanographene oxide sheets: at physiological pH, only 25% of MB was released after 72 h, while at acidic pH, approximately 40% was released within 6 h. Although the direct impact of release speed on treatment efficacy remains unclear, a controlled release can contribute to improved treatment outcomes. Therefore, MB-encapsulated PA-6 fibers present an attractive material for tumor treatment applications.

Quantification of reactive oxygen species (ROS) generation

The ability of MB-loaded electrospun membranes to produce singlet oxygen (¹O₂) under red light exposure was tested using the DPBF degradation assay. Control experiments with polymer-only PA-6 membranes under the same irradiation conditions showed no DPBF degradation, confirming that ROS generation came from MB and not the polymer matrix.

For membranes containing 30 wt.% PA-6 and the highest of MB concentration (MB-HC), 50% of the DPBF signal was quenched within 5 min, whereas in all other formulations this threshold was reached only after approximately 10 min (Figure 7). These results demonstrate that increasing MB concentration enhances the rate of ¹O₂ generation.

 

 

The degradation kinetics were analyzed using a one-phase exponential decay model, which provided statistically acceptable fits across all samples (R² = 0.873-0.979; Table 4). The apparent rate constants increased with MB concentration: in 18 wt.% PA-6 membranes, from 0.1298 ± 0.0018 min^-1 (10 mM MB) to 0.2236 ± 0.0037 min^-1 (28 mM MB), and in 30 wt.% PA-6 membranes, from 0.0713 ± 0.0035 min^-1 (10 mM MB) to 0.2425 ± 0.0033 min^-1 (28 mM MB). These trends indicate a positive correlation between MB loading and photosensitizer activity.

 

 

Closer inspection of the degradation curves suggests that, at higher MB concentrations, the process may deviate from simple single-exponential kinetics. This behavior could result from MB aggregation within the fiber matrix, localized oxygen depletion, or a heterogeneous distribution of reactive species. While the single-phase model provided a consistent basis for comparing formulations, future studies will explore more complex kinetic models (e.g., biphasic or diffusion-influenced), supported by model selection criteria such as the Akaike information criterion (AIC), to refine mechanistic understanding.

These findings confirm that incorporating MB into PA-6 nanofibers allows for effective generation of ¹O₂ upon red light irradiation, with ROS production heavily dependent on dye loading. Including polymer-only controls shows that the photodynamic activity is solely due to MB, emphasizing the potential of these electrospun systems for topical PDT use.

Evaluation of cell viability after PDT treatment

To assess the photodynamic performance of MB-loaded electrospun membranes, cytotoxicity assays were carried out in both non-tumor (HEK293) and tumor (SW480) cell lines. Cells were incubated with PA-6 fibers containing two different MB concentrations for 24 h, with PA-6 membranes without MB serving as controls. Cell viability was measured using the MTT reduction assay after red-light irradiation.

As shown in Figure 8, PA-6 membranes alone did not affect cell viability in either cell line, confirming the biocompatibility of the polymer matrix. In contrast, MB-loaded membranes caused a significant reduction in viability in both HEK293 and SW480 cells. This effect was further intensified by red-light irradiation, consistent with MB-mediated photodynamic action. Notably, even without light exposure, MB-containing fibers exhibited measurable cytotoxicity, which may result from dark interactions between MB and cellular biomolecules such as purine nucleotides, as previously reported.⁶⁰

 

 

The observation that HEK293 cells are also sensitive to MB-loaded fibers highlights a potential limitation of this approach: the absence of strict tumor selectivity. Although the cytotoxic effect was more significant in the SW480 tumor model (Figure 8b), suggesting preferential activity against cancer cells, the data imply that damage to non-tumor tissue cannot be ruled out. This issue is particularly relevant for topical PDT treatments, where the accumulation of photosensitizers in healthy surrounding tissue may lead to unwanted side effects.

Future strategies to enhance tumor selectivity could involve functionalizing nanofiber surfaces with tumor-targeting ligands (such as folic acid, arginine-glycine-aspartic acid (RGD) peptides, or antibodies) to encourage preferential binding and uptake by malignant cells. Alternatively, co-loading with tumor-microenvironment-responsive elements (like pH- or enzyme-sensitive linkers) can restrict MB release to diseased tissues. These approaches may reduce off-target cytotoxicity and improve the therapeutic index of MB-based electrospun systems.

 

CONCLUSIONS

In this work, MB-loaded PA-6 electrospun nanofibers were successfully fabricated and evaluated as potential platforms for topical photodynamic therapy. The incorporation of MB altered solution viscosity, influenced fiber morphology, and enabled diffusion-controlled release of the photosensitizer. Release was faster from fibers produced with lower polymer concentration (18 wt.% PA-6) and more sustained from denser mats (30 wt.% PA-6). Upon irradiation with red light, the membranes efficiently generated singlet oxygen, with ROS yields increasing in proportion to MB content. In vitro, MB-loaded membranes reduced viability in both tumor (SW480) and non-tumor (HEK293) cells, with more pronounced effects in the tumor model. This underscores their potential for PDT while also highlighting the need to improve tumor selectivity, for example, by introducing targeting ligands or microenvironment-responsive release mechanisms.

When compared with other photosensitizer-loaded electrospun systems, the MB-PA-6 fibers demonstrated competitive performance. For example, cationic porphyrin-loaded polyacrylonitrile (PAN) nanofibers incorporated photosensitizers at ca. 3-4 wt.% and showed strong photodynamic activity.²⁹ Chlorin e6-loaded polycaprolactone (PCL)/gelatin fibers achieved sustained release and efficient ROS generation suitable for antibacterial PDT.³⁰ Similarly, zinc phthalocyanine embedded in biodegradable poly(lactic acid) (PLA)/poly(butylene adipate-co-terephthalate) (PBAT) electrospun mats retained strong singlet oxygen generation.²⁸ Taken together, these comparisons place MB-PA-6 fibers within the same performance range as more complex porphyrin- and phthalocyanine-based nanofiber systems, while offering the advantages of lower cost and simpler chemistry.

Overall, this study highlights the advantages of electrospun MB-PA-6 membranes as scalable, low-cost, and photodynamically active systems, while identifying selectivity as the primary limitation to be addressed in future work. With targeted modifications, these membranes could represent a competitive alternative to porphyrin-, chlorin-, and phthalocyanine-based nanofiber platforms for topical PDT applications.

 

SUPPLEMENTARY MATERIAL

Complementary material for this work is available at http://quimicanova.sbq.org.br/, as a PDF file, with free access.

 

DATA AVAILABILITY STATEMENT

All data generated or analyzed during this study are included in this published article and its supplementary material.

 

ACKNOWLEDGMENTS

This study was partially funded by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), Brazil (process No. 2024/00989-7 and 2022/14753-0), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, grant 305574/2023-0), and Instituto Nacional de Ciência e Tecnologia de Bioanalítica - Lauro Kubota (INCTBio-LK) (CNPq grant 408338/2024-5). Additionally, this study was partially funded by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES, finance code 001). The authors are grateful to the Multiuser Central Facilities at Universidade Federal do ABC (UFABC).

 

AUTHOR CONTRIBUTIONS

Wendel A. Alves was responsible for conceptualization, data curation, funding acquisition, project administration, validation, writing (original draft, review and editing); Juliana Pirajá for data curation, formal analysis, investigation, project administration, writing (original draft, review and editing); Aline R. F. Teixeira for data curation, investigation, Irina M. Factori for data curation, investigation, Anderson F. Sepulveda for data curation, writing (review and editing); Luiz F. G. Setz for data curation, validation; Anderson O. Ribeiro for data curation, validation.

 

REFERENCES

1. Huang, Z. M.; Zhang, Y. Z.; Kotaki, M.; Ramakrishna, S.; Compos. Sci. Technol. 2003, 63, 2223. [Crossref]

2. Li, D.; Xia, Y.; Adv. Mater. 2004, 16, 1151. [Crossref]

3. Bhardwaj, N.; Kundu, S. C.; Biotechnol. Adv. 2010, 28, 325. [Crossref]

4. Xue, J.; Wu, T.; Dai, Y.; Xia, Y.; Chem. Rev. 2019, 119, 5298. [Crossref]

5. Rogalski, J. J.; Bastiaansen, C. W. M.; Peijs, T.; Fibers 2018, 6, 37. [Crossref]

6. Lim, D. J.; Polymers 2021, 13, 3955. [Crossref]

7. Factori, I. M.; da Silva, E. R.; Gerbelli, B. B.; Liberato, M. S.; de Araujo, D. R.; Alves, W. A.; J. Appl. Polym. Sci. 2024, 141, e56226. [Crossref]

8. Liberato, M. S.; Mancini, R. S. N.; Factori, I. M.; Ferreira, F. F.; de Oliveira, V. L.; Carnielli, J. B. T.; Guha, S.; Peroni, L. A.; Oliveira, M. A. L.; Alves, W. A.; ACS Appl. Electron. Mater. 2019, 1, 2086. [Crossref]

9. Liberato, M. S.; Kogikoski, S.; da Silva, E. R.; de Araujo, D. R.; Guha, S.; Alves, W. A.; J. Mater. Chem. B 2016, 4, 1405. [Crossref]

10. Junior Kogikoski, S.; Liberato, M. S.; Factori, I. M.; da Silva, E. R.; Oliveira, C. L. P.; Ando, R. A.; Alves, W. A.; J. Phys. Chem. C 2017, 121, 863. [Crossref]

11. Jalalah, M.; Ahmad, A.; Saleem, A.; Qadir, M. B.; Khaliq, Z.; Khan, M. Q.; Nazir, A.; Faisal, M.; Alsaiari, M.; Irfan, M.; Alsareii, S. A.; Harraz, F. A.; Membranes 2022, 12, 1158. [Crossref]

12. Garkal, A.; Kulkarni, D.; Musale, S.; Mehta, T.; Giram, P.; New J. Chem. 2021, 45, 21508. [Crossref]

13. Henke, P.; Lang, K.; Kubát, P.; Sýkora, J.; Slouf, M.; Mosinger, J.; ACS Appl. Mater. Interfaces 2013, 5, 3776. [Crossref]

14. Henke, P.; Dolanský, J.; Kubát, P.; Mosinger, J.; ACS Appl. Mater. Interfaces 2020, 12, 18792. [Crossref]

15. Kabay, G.; Meydan, A. E.; Eom, T.; Shim, B. S.; Mutlu, M.; Kaleli-Can, G.; Int. J. Pharm. 2023, 630, 122442. [Crossref]

16. Yang, L.; Zhang, Y.; Hu, X.; Shiu, B. C.; Lou, C. W.; Lin, J. H.; Li, T. T.; J. Ind. Eng. Chem. 2023, 120, 410. [Crossref]

17. Monroe, J. D.; Belekov, E.; Er, A. O.; Smith, M. E.; Photochem. Photobiol. 2019, 95, 1473. [Crossref]

18. Cui, M.; Wu, J.; Wei, J.; Wei, Z.; Surf. Coat. Technol. 2023, 463, 129497. [Crossref]

19. Chang, W. Y.; Chen, C. Y.; ACS Omega 2023, 8, 36906. [Crossref]

20. Jain, S.; Meka, S. R. K.; Chatterjee, K.; ACS Biomater. Sci. Eng. 2016, 2, 1376. [Crossref]

21. Qian, S.; Song, L.; Sun, L.; Zhang, X.; Xin, Z.; Yin, J.; Luan, S.; J. Photochem. Photobiol., A 2020, 400, 112626. [Crossref]

22. Costa, S. M.; Fangueiro, R.; Ferreira, D. P.; Macromol. Biosci. 2022, 22, 2100512. [Crossref]

23. Fan, Z.; Dai, X.; Lu, Y.; Yu, E.; Brahmbatt, N.; Carter, N.; Tchouwou, C.; Singh, A. K.; Jones, Y.; Yu, H.; Mol. Pharmaceutics 2014, 11, 1109. [Crossref]

24. Meesaragandla, B.; Sarkar, D.; Mahalingam, V.; ACS Omega 2019, 4, 3169. [Crossref]

25. Nigoghossian, K.; Saska, S.; Christovam, L. M.; Coelho, F.; Beatrice, C. A. G.; Lucas, A. A.; Neto, P. I.; da Silva, J. V. L.; Tercjak, A.; Baptista, M. S.; J. Braz. Chem. Soc. 2020, 31, 638. [Crossref]

26. Oliveira, R.; dos Santos, L. H. C.; Ghasemishahrestani, Z.; Pereira, M. D.; Forero, J. S. B.; Simões, G.; Ribeiro, E. S.; Corrêa, R. J.; J. Braz. Chem. Soc. 2024, 35, e-20240036. [Crossref]

27. Rossin, A. R. S.; de Oliveira, E. L.; de Moraes, F. A. P.; Silva Júnior, R. C.; Scheidt, D. T.; Caetano, W.; Hioka, N.; Dragunski, D. C.; Quim. Nova 2020, 43, 613. [Crossref]

28. Rossin, A. R. S.; Caetano, J.; Zanella, H. G.; Bariccatti, R. A.; Gaffo, L.; Muniz, E. C.; Caetano, W.; Stolf, S. F.; Dragunski, D. C.; J. Therm. Anal. Calorim. 2022, 147, 4579. [Crossref]

29. Mendes, A. C.; Moura, N. M. M.; Neves, M. G. P. M. S.; Chronakis, I. S.; BBA Adv. 2025, 7, 100160. [Crossref]

30. Michalak, M.; Szymczyk, J.; Pawska, A.; Wysocki, M.; Janiak, D.; Ziental, D.; Ptaszek, M.; Güzel, E.; Sobotta, L.; Molecules 2025, 30, 2810. [Crossref]

31. da Collina, G. A.; Freire, F.; Santos, T. P. C.; Sobrinho, N. G.; Aquino, S.; Prates, R. A.; da Silva, D. F. T.; Horliana, A. C. R. T.; Pavani, C.; Photochem. Photobiol. Sci. 2018, 17, 1355. [Crossref]

32. Sohrabi, M.; Razbin, M.; Pourtavvaf, M.; Bagherzadeh, R.; Mirmahale, M. M.; The Journal of the Textile Institute 2023, 114, 1667. [Crossref]

33. Glass, S.; Rüdiger, T.; Griebel, J.; Abel, B.; Schulze, A.; RSC Adv. 2018, 8, 41624. [Crossref]

34. Simon, T.; Boca‐Farcau, S.; Gabudean, A.; Baldeck, P.; Astilean, S.; J. Biophotonics 2013, 6, 950. [Crossref]

35. Alberdi, E.; Gómez, C.; Dermatol. Ther. 2020, 33, e14280. [Crossref]

36. Kirakci, K.; Kubát, P.; Fejfarová, K.; Martincik, J.; Nikl, M.; Lang, K.; Inorg. Chem. 2016, 55, 803. [Crossref]

37. Costa, P.; Sousa Lobo, J. M.; Eur. J. Pharm. Sci. 2001, 13, 123. [Crossref]

38. Sepulveda, A. F.; Kumpgdee-Vollrath, M.; Franco, M. K. K. D.; Yokaichiya, F.; de Araujo, D. R.; J. Colloid Interface Sci. 2023, 630, 328. [Crossref]

39. Souza, M. I.; Jaques, Y. M.; de Andrade, G. P.; Ribeiro, A. O.; da Silva, E. R.; Fileti, E. E.; Ávilla, E. S.; Pinheiro, M. V. B.; Krambrock, K.; Alves, W. A.; J. Phys. Chem. B 2013, 117, 2605. [Crossref]

40. Souza, M. I.; Silva, E. R.; Jaques, Y. M.; Ferreira, F. F.; Fileti, E. E.; Alves, W. A.; J. Pept. Sci. 2014, 20, 554. [Crossref]

41. Zhang, P.; Zhou, W.; He, Y.; Xu, Z.; Li, M.; Hong, W.; Yang, C.; ACS Appl. Mater. Interfaces 2022, 14, 49264. [Crossref]

42. Wang, M.; Ge, R. L.; Zhang, F.; Yu, D. G.; Liu, Z. P.; Li, X.; Shen, H.; Williams, G. R.; Biomater. Adv. 2023, 150, 213404. [Crossref]

43. Wang, Y.; Yu, D. G.; Liu, Y.; Liu, Y. N.; J. Funct. Biomater. 2022, 13, 289. [Crossref]

44. Meng, X.; Schiffman, J. D.; Perry, S. L.; Macromolecules 2018, 51, 8821. [Crossref]

45. Nirmala, R.; Woo-il, B.; Navamathavan, R.; Kalpana, D.; Lee, Y. S.; Kim, H. Y.; Colloids Surf., B 2013, 104, 262. [Crossref]

46. Angammana, C. J.; Jayaram, S. H.; IEEE Trans. Ind. Appl. 2011, 47, 1109. [Crossref]

47. Deitzel, J. M.; Kleinmeyer, J.; Harris, D.; Beck Tan, N. C.; Polymer 2001, 42, 261. [Crossref]

48. Dowling, D. P.; Miller, I. S.; Ardhaoui, M.; Gallagher, W. M.; J. Biomater. Appl. 2011, 26, 327. [Crossref]

49. Law, K. Y.; J. Phys. Chem. Lett. 2014, 5, 686. [Crossref]

50. Rauter, M. T.; Schnell, S. K.; Kjelstrup, S.; J. Phys. Chem. B 2021, 125, 12730. [Crossref]

51. Stachewicz, U.; Bailey, R. J.; Zhang, H.; Stone, C. A.; Willis, C. R.; Barber, A. H.; ACS Appl. Mater. Interfaces 2015, 7, 16645. [Crossref]

52. Rawal, A.; Shukla, S.; Sharma, S.; Singh, D.; Lin, Y. M.; Hao, J.; Rutledge, G. C.; Vásárhelyi, L.; Kozma, G.; Kukovecz, A.; Janovák, L.; Appl. Surf. Sci. 2021, 551, 149147. [Crossref]

53. Park, I. W.; Ribe, J. M.; Fernandino, M.; Dorao, C. A.; Adv. Mater. Interfaces 2023, 10, 2202439. [Crossref]

54. Millot, C.; Fillot, L. A.; Lame, O.; Sotta, P.; Seguela, R.; J. Therm. Anal. Calorim. 2015, 122, 307. [Crossref]

55. Dujmić, A. L.; Radičić, R.; Razić, S. E.; Cingesar, I. K.; Glogar, M.; Jurov, A.; Krstulović, N.; Polymers 2024, 16, 1883. [Crossref]

56. Morales, J.; Rodrigue, D.; Macromol. Mater. Eng. 2025, 310, 2400304. [Crossref]

57. Anjeh, A. M.; Nabavi, S. R.; J. Polym. Environ. 2022, 30, 3934. [Crossref]

58. Nascimento, M. H. M.; Franco, M. K. K. D.; Yokaichyia, F.; de Paula, E.; Lombello, C. B.; de Araujo, D. R.; Int. J. Biol. Macromol. 2018, 111, 1245. [Crossref]

59. Jiang, W.; Chen, X.; Yan, X.; Li, J.; Xie, L.; Yu, Y.; Colloids Surf., A 2024, 702, 135098. [Crossref]

60. Tuite, E. M.; Kelly, J. M.; J. Photochem. Photobiol., B 1993, 21, 103. [Crossref]