Halochromic composite nanofibrous mat for wound healing monitoring

This study was focused on the development and characterization of a wound dressing for pH monitoring based on a sodium alginate (SA) and polyvinyl alcohol (PVA) nanofibrous mat including anthocyanins extracted from black carrot (BC). The pH-sensitive mat was prepared through the electrospinning method. The crosslinking of the nanofibrous mat was carried out using glutaraldehyde (GA). The functionalized PVA/SA/BC nanofibrous mat was characterized by scanning electron microscopy (SEM), Fourier-transform infrared spectroscopy (FTIR), and thermogravimetric analysis (TGA). The colors of the nanofibrous mat resulting from pH changes were evaluated using a digital camera. The images were processed using Adobe Photoshop software, CS5 version, and analyzed through RGB and CIELAB. When the dried nanofibrous mat with a light pink color was immersed in buffer solutions at different pHs, the color of the nanofibrous mat ranged from red (in acidic medium) to blue (in neutral medium) and then to black-green (in basic medium). The response time and temperature stability of the nanofibrous mat were also investigated. It was concluded that the developed pH-sensing composite nanofibrous mat could be successfully used as wound dressing for monitoring healing.


Introduction
Chromic textiles, a subclass of smart textiles, change color depending on environmental changes. The number of studies focusing on the fabrication of chromic textiles has rapidly increased in recent years [1,2]. Numerous external stimuli such as temperature, light, pressure, and pH can lead to changes in color. Halochromism is defined as a reversible color change arising from a pH change [3].Although halochromic textiles are lesser known than other chromic textiles [4], they present an increasing usage potential, especially in the future of medical textiles.
Electrospinning is a straightforward and affordable method used to generate ultrathin polymer fibers via an electric field. Electrospun nanofibrous mats have distinctive characteristics, such as high specific surface area, high porosity, small pore sizes, permeability to moisture and air, and resistance to microorganisms [16,17]. They present a wide range of application possibilities, especially in biomedical applications such as medical 1.18 g ml −1 ) and GA (50%) were supplied by Sigma-Aldrich, and acetone purchased from Merck was used in the crosslinking process. All other chemicals were of analytical grade. Elga Flex3 (Veolia Water Solutions & Technologies, France) was used for water purification.

Extraction of BC pigment anthocyanins
Fresh BCs were milled into a uniform sample using a blender. After that, ultrasonic-assisted extraction was performed in a temperature-controlled ultrasonic bath (Wiseclean Ultrasonic Cleaner). In the procedure, approximately 350 g of the homogenized sample was placed into a glass Erlenmeyer flask containing 500 ml of the ethanol/water (80:20) mixture. Then, the sample was exposed to ultrasonic treatment at 35 kHz for 60 min The obtained mixture was filtered through a filter paper. The ethanol/water mixture containing BC anthocyanins was kept in a refrigerator at 4°C. A second portion of ethanol/water mixture (500 ml) was replenished to the filtered solid, and ultrasonic-assisted extraction was performed using the same procedure. After the completion of the extraction procedure, the two portions were put together. The total extract was lyophilized at 0.188 mBar pressure at −45°C for 72 h using a freeze drier (Labconco Freezone Freeze Dry Systems). Finally, the residual solid was dissolved in pure water (100 ml) and stored at −24°C until use.
The amount of anthocyanins extracted from BCs was determined via the pH differential method described in the AOAC Official Method 2005.02 [59] and expressed as milligrams of cyanidin-3-glucoside equivalent per 1 g dried weight.

Preparation of the pH-sensing PVA/SA/BC nanofibrous mat
The electrospinning solution of the PVA/SA/BC nanofibrous mat was prepared by mixing SA, PVA, and BC extract. A magnetic stirrer (Cleaver Scientific LTD) was used to prepare solutions. PVA and SA solutions were separately prepared under a reflux condenser to prevent water from boiling away. Firstly, the SA solution (1%, w/w) was prepared by stirring the aqueous solution for 6 h at a 400 rpm stirring rate at 60°C. PVA (12%, w/w) was dissolved in distilled water by stirring for 12 h at a 400 rpm stirring rate at 80°C. After that, a PVA/SA mixture was prepared by mixing the PVA and SA solutions at a volume ratio of 2/1 (v/v). BC extract was added to the PVA/SA mixture at a concentration of 1% (w/v). The obtained PVA/SA/BC mixture was gently stirred for 3 h at room temperature to obtain a homogeneous mixture. The PVA/SA mixture was also prepared using the same procedure without addition of BC extract.
The pH, surface tension, electrical conductivity, and viscosity of the electrospun solutions were rigorously determined before the electrospinning process. The pH and electrical conductivity measurements were performed using HANNA pH and Conductivity Meter (HI-98129). A Brookfield viscometer (RV-DV II) was used to perform viscosity measurements. An Attension Theta KSV instrument was used to determine the surface tension. All measurements were repeated three times.
The electrospinning of the PVA/SA and PVA/SA/BC solutions (20 ml for each) was conducted using an electrospinning system (Inovenso Technology Inc.) including a high-voltage DC power supply. Electrospinning parameters for PVA/SA and PVA/SA/BC solutions were the same, with the exception of applied voltage. In the electrospinning process, the feed rate and tip-to-collector distance were 0.7 ml h −1 and 12 cm, respectively. The electrospinning procedure was carried out by applying a voltage of 18 kV and 14 kV for PVA/SA and PVA/SA/BC solutions, respectively. The PVA/SA and PVA/SA/BC electrospun nanofibrous mats were collected on aluminum foil wrapped around a cylindrical drum rotating at 200 rpm.
The PVA/SA/BC and PVA/SA electrospun nanofibrous mats were crosslinked by GA to obtain waterinsoluble crosslinked nanofibers. The crosslinking was achieved by immersing the electrospun nanofibrous mats in acetone solution containing 0.5 ml of GA and 0.2 ml of concentrated HCl at room temperature for 10 min. After that, the crosslinked nanofibrous mat was washed with ethyl alcohol and phosphate buffer (pH 7.0) and dried at room temperature for 24 h.  The morphology of the PVA/SA, PVA/SA/BC, crosslinked PVA/SA, and PVA/SA/BC electrospun nanofibrous mats were examined using a scanning electron microscope (CARL ZEISS AG-EVO 40 XVP). The mean nanofiber diameter and its distribution were determined from 100 random measurements using ImageJ software (National Institute of Health, USA) on SEM images of each sample. The thickness of the electrospun mats was measured using a digital micrometer (INSIZE Thickness Gauge) with 0.005 mm accuracy at five points. The tests were conducted at 25±2°C. The average values and standard deviations were calculated from the obtained data.
An FTIR spectrometer (Perkin Elmer, Spectrum100) equipped with an attenuated total reflectance (ATR) apparatus was used to study the chemistry of the PVA/SA, PVA/SA/BC, and crosslinked PVA/SA/BC nanofibrous mats within the range of 650-4000 cm −1 .
The TGA curves of the PVA/SA, PVA/SA/BC and crosslinked PVA/SA/BC nanofibrous mats were obtained using a thermogravimeter (Perkin Elmer, STA 6000). TGA was performed by heating the sample (10 mg) in the temperature range of 30°C to 850°C under a nitrogen flow of 20 ml min −1 and at a heating rate of 20°C min −1 .

Color measurements
The color schemes of the PVA/SA/BC electrospinning solutions and the crosslinked PVA/SA/BC nanofibrous mats at different pHs were tested by color measurements. The PVA/SA/BC electrospinning solutions were diluted with prepared buffer solutions in the ratio of 1:6. The PVA/SA/BC nanofibrous mats (cut into 2×2 cm squares) were soaked in the buffer solutions for 1 min. The color schemes of the PVA/SA/BC electrospinning solutions and the PVA/SA/BC nanofibrous mats were compared through the CIELAB (International Commission on Illumination) color space coordinates. The colors were recorded as digital photographs using a camera (CANON EOS 5D Mark III camera). RGB and CIE L * a * b * values of the PVA/SA/BC solutions and the PVA/SA/BC nanofibrous mats were determined using the Adobe Photoshop CS5 program. Color analyses were carried out by using the photographs obtained under the same illumination conditions. An average of ten readouts each comprising an average of 100 pixels were considered for analysis. The color of each electrospinning solution and nanofibrous mat at different pHs was compared through the total color difference (DE ab * ) using equation (1): where D is th difference between the numeric value of the sample and the numeric value of the standard; DE ab * is the distance between the two colors in the CIELAB color space [57].

Stability of the pH-sensing PVA/SA/BC nanofibrous mat
The stability of the pH-sensing PVA/SA/BC nanofibrous mat was evaluated under different conditions with respect to temperature. The nanofibrous mat samples were incubated at two temperatures of −20°C (storage temperature) and 37°C (body temperature) for 24 h. Subsequently, the responses to pH changes were investigated by soaking the samples in the buffer solutions (pH 4.0, 7.0 and 9.0). RGB values were calculated from the photographs of nanofibrous mats. The measurements were repeated three times.

Solution properties
The electrospinning process and nanofibrous surface morphology are significantly affected by the properties of electrospinning solutions such as viscosity, conductivity, pH, and surface tension [60,61]. The properties of the PVA/SA and PVA/SA/BC electrospinning solutions are summarized in table 1.
The solution parameters clearly showed that addition of BC extract to the PVA/SA solution had a remarkable effect on viscosity and conductivity values. The 40% decrease observed in solution viscosity was associated with the decrease in polymer concentration owing to the addition of BC extract to the PVA/SA solution. On the contrary, the 20% increase in the conductivity resulted from the existence of ions such as calcium, iron, and sodium in the BC extract solution, besides anthocyanins [62]. The pH and surface tension values of the PVA/SA solution did not significantly change, but an inconsiderable decrease was detected with the addition of BC extract.
During the production of the PVA/SA and PVA/SA/BC nanofibrous mats, the process parameters other than applied voltage were kept completely equal. Therefore, the solution properties played a determinant role in the morphology of the electrospun mats. The addition of aqueous BC extract into the PVA/SA solution caused a decrease in viscosity because the polymer concentration decreased. The effect of the concentration/viscosity on the morphology of the nanofibers was reported in many previous studies [64][65][66]. When the viscosity is low, weak chain entanglement among the polymer chains is obtained and the effect of surface tension force on the jet becomes dominating during electrospinning. Therefore, continuous jet formation may not be obtained, and the formation of beads or beaded nanofibers might be observed.
The conductivity of electrospinning solution also plays a significant role in the electrospinning process. Usually, an increase in conductivity of the solution causes the formation of fine fibers because the polymer solution is exposed to more stretching under the high electrical field [67][68][69]. The addition of BC extract to the PVA/SA polymer solution caused an increase in conductivity due to the existence of ions in the BC extract, which results in an increase in the surface charge density of the polymer jet. Therefore, the PVA/SA/BC nanofibrous mat had approximately a 20% lower fiber diameter (211±39 nm) and lower fiber diameter variation (18%) than the PVA/SA nanofibrous mat. In addition, sticking was observed at the junction points of the nanofibers in the nanofibrous mat containing BC extract. When the concentration of free solvent molecules in the solution is high, solvent molecules have a tendency to accumulate because of surface tension, and they cannot evaporate [70]. This partial dissolving on the collector could lead to the fibers sticking to each other.
The PVA/SA or PVA/SA/BC nanofibrous mats were crosslinked using water-soluble GA. When the SEM photographs (figures 2(c),(d)) obtained after crosslinking were examined, it was observed that the integrity of the fibrous structure was kept but the fibers were flattened. In previous studies, it has been reported that the newly formed crosslinks lead to squeezing of the fibrous structure, and some amount of solvent remaining in the fibrous structure evaporates, causing the formation of flat fibers [70,71]. The morphologies obtained from SEM analyses of the crosslinked PVA/SA and PVA/SA/BC electrospun mats supported the results reported in previous studies. As a result of the flattening after the crosslinking process, the average diameter values of the crosslinked PVA/SA nanofibers and the crosslinked PVA/SA/BC nanofibers increased to 263±80 nm and 213±37 nm, respectively. Moreover, the porosity on the crosslinked mats decreased visibly. In addition to flattening, fusions at nanofiber junction points increased after crosslinking by GA. It has been reported that the nanofibrous structures crosslinked by GA in aqueous solution are immediately changed into a dense membrane [72]. In fact, vapor crosslinking by GA is generally preferred to avoid the collapse of the nanofibrous matrix in the aqueous system [73]. However, in this study, PVA/SA nanofibers and PVA/SA/BC nanofibers were successfully crosslinked by GA in aqueous medium without a significant morphological change, although fusions at nanofiber junction points slightly increased after crosslinking. It can be concluded that the crosslinking medium consisting of a small amount of water (0.5%) and a short crosslinking time (10 min) enabled the formation of a morphologically functional crosslinked PVA/SA/BC electrospun mat.

Thickness measurements
Thickness is an important characteristic of electrospun mats because the thickness of the mats directly affects performance properties such as permeability. Because the crosslinked PVA/SA/BC electrospun mat is expected to respond to different pHs with a color change, the thickness value becomes even more important. The thicknesses of the nanofibrous PVA/SA, PVA/SA/BC, crosslinked PVA/SA, and crosslinked PVA/SA/BC mats were measured as 82.4±3.4 μm, 84.0±4.8 μm, 78.4±2.4 μm, and 81.6±4.2 μm, respectively. The thickness was controlled through the volume of the electrospinning solutions. Although both PVA/SA and PVA/SA/BC nanofibrous mats were produced under the same conditions, the slightly higher thickness value of the PVA/SA/BC electrospun mat (84.0±4.8 μm) could be attributed to the existence of beaded nanofibers with thicker diameters in structure. In addition, the crosslinking process did not significantly affect the thicknesses of the PVA/SA and PVA/SA/BC nanofibrous mats, but a insignificant decrease was observed (3%-5%), probably associated with the flattening of the nanofibers after crosslinking. As a result, the crosslinked PVA/SA/BC electrospun mat with a convenient thickness was prepared for monitoring pH changes.

FTIR analysis
FTIR analysis was conducted to examine the effect of BC extract addition and the crosslinking procedure. FTIR spectra of PVA/SA, PVA/SA/BC, and crosslinked PVA/SA/BC nanofibrous mats are shown in figure 3.
In the FTIR spectrum of the PVA/SA nanofibrous mat, the broad adsorption band at 3399 cm −1 mainly resulted from the stretching of -OH groups of PVA [74]. The absorption bands at 2841 cm −1 and 2916 cm −1 are due to the symmetric and antisymmetric stretching vibrations of C-H from alkyl groups, respectively [75]. The absorption band at 1734 cm −1 is related to the stretching vibration band of the ester carbonyl (C=O) group of PVA because it is a semi-crystalline polymer prepared by hydrolysis of the poly(vinyl acetate) [76]. The characteristic band belonging to stretching vibrations of the carboxylic acid carbonyl group of SA appeared at 1713 cm −1 [45]. The absorption bands resulting from the C-O stretching of PVA and SA were also observed at 1091cm −1 [77].
The adsorption band at 3306 cm −1 resulting from the stretching of -OH groups more intensely appeared in the FTIR spectrum of the PVA/SA/BC nanofibrous mat because the anthocyanins in BC extract also have -OH groups and are present as glycosides [78]. Additionally, the characteristic adsorption band at 1046 cm −1 corresponding to aromatic ring C-H deformation of anthocyanins supported [79] that BC anthocyanins were successfully incorporated into the nanofibrous mat.
The GA crosslinking agent used in the crosslinking process activates the hydroxyl groups in the polymer molecules generating the nanofibrous structure and ensures the formation of acetal bonds between them [80]. In the FTIR spectrum of the crosslinked PVA/SA/BC mat, the absorption bands at 1130 cm −1 belonging to C-O-C stretching of acetal groups showed that the crosslinking with GA was successfully performed [81]. Moreover, the intensity of the absorption band at 3398 cm −1 considering -OH groups considerably reduced after crosslinking because the crosslinking reaction occurs via free hydroxyl groups. The aldehyde peak observed at 2867 cm −1 in the FTIR spectrum of the crosslinked PVA/SA/BC mat probably resulted from the participation of only one aldehyde group of some GA molecules in the crosslinking reaction, whereas the other remained unreacted. As a result, the crosslinking was successfully performed in the presence of BC extract.
The anthocyanin content of the nanofibers was also calculated as 1.37±0.06 mg anthocyanin/g for the PVA/SA/BC nanofibrous mat.

Thermal properties
Thermal degradation profiles of PVA/SA, PVA/SA/BC, and crosslinked PVA/SA/BC nanofibrous mats were examined by TGA to determine the thermal stability of their components ( figure 4).
TGA curves of the PVA/SA, PVA/SA/BC, and crosslinked PVA/SA/BC mats showed similar patterns. The degradation of the mats occurred in three stages of mass loss [74]. The first was between approximately 50 and 100°C due to loss of adsorbed water. The mass losses of PVA/SA and PVA/SA/BC nanofibrous mats were 10% and 4%, respectively. The mass loss of the PVA/SA/BC nanofibrous mat slightly decreased as compared to the PVA/SA. This was probably because the amount of water adsorbed onto the PVA/SA/BC mat comparatively decreased due to the existence of BC extract in the nanofibrous structure. The second mass losses, about 90%, occurred between 220.69°C and 251.07°C ( figure A.3). The second stage of thermal degradation includes both melting point and the degradation temperature of PVA. The result was considered a chemical degradation resulting from the breaking of carbon-carbon bonds in the polymeric backbone [82]. In the second stage, the mass losses of PVA/SA and PVA/SA/BC nanofibrous mats started at 251.07°C and 220.69°C, respectively. The addition of BC extract caused a slight decrease (∼30°C) in thermal stability, indicating that the extract has a low thermal stability. On the other hand, the mass loss of the crosslinked PVA/SA/BC started at 243.85°C. When compared to PVA/SA/BC, the thermal stability of the crosslinked PVA/SA/BC nanofibrous mat showed a slight increase (∼20°C), which suggests that the reaction between PVA and GA caused a decrease in the amount of hydroxyl groups involved in the polyene formation, leading to a decrease in weight loss [83]. In the third stage, the mass losses around 510°C-540°C were attributed to the fragmentation of the macromolecular structures of the mats. As a result, TGA clearly showed that the crosslinked PVA/SA/BC nanofibrous mat was thermally stable until the temperature of 50°C, thus having thermal stability to be applied in wound healing monitoring at body temperature.

RGB and CIELAB values
The visual color change of PVA/SA/BC electrospinning solutions and crosslinked PVA/SA/BC nanofibrous mats are presented in figure 5. The crosslinked PVA/SA/BC nanofibrous mats had a short response time (5-10 s) for each solution at different pHs (4.0-10.0). This result was consistent with the one reported by Devarayan and Kim [51], who investigated immobilization of natural pH indicator dyes extracted from red cabbage into a cellulose acetate nanofibrous mat. The visual color changes of electrospinning solutions and nanofibrous mats demonstrated that the color-changing property of BC extract in both cases resulted from the pH changes in the medium. The RGB values calculated for PVA/SA/BC electrospinning solutions and crosslinked PVA/SA/BC nanofibrous mats at each pH are presented in figure 6. RGB values and their corresponding changes in the pH range of 4.0-10.0 clearly showed the red color of the acidic samples at pH 4.0-6.0, the blue color of the samples at pH 7.0-9.0, and the black-green color at pH 10.0. The differences in RGB values were more visible in electrospinning solutions than nanofibrous mats, probably because of the light transmittance property of the solutions. In both cases, the R value showed the most remarkable changes between pH 4.0 and 10.0, especially for the electrospinning solutions. It could be concluded that the change in R value primarily determined the color change for the two cases. This could be because of the natural color of BC extract, which is pure red.
CIELAB color coordinates were calculated to evaluate the colors in the CIELAB color space. The CIELAB model is similar to human vision and is not dependent on the device on which the colors are displayed or the method of creation. The observable colors and the colors that are out of range for human vision are included in the CIELAB color space [84].
CIELAB L * a * b * coordinates were determined on the digital photographs of the electrospinning solutions and nanofibrous mats using Adobe Photoshop CS5. Total color differences (DE*) of the standard PVA/SA/ BC electrospinning solutions and crosslinked PVA/SA/BC nanofibrous mats were calculated and presented in table 2. It can be clearly seen that the DE* values for the PVA/SA/BC solutions were significantly higher than those of the crosslinked PVA/SA/BC mats between each pH. It is generally reported that the color differences are visibly distinguished when the DE* value is greater than 5. Furthermore, the colors generally belong to different color hues if DE* values are higher than 12 [51]. DE* values calculated for the crosslinked PVA/SA/BC nanofibrous mats were higher than 5 at all studied pHs. These results demonstrated that the colors obtained at different pHs belonged to different color hues. Moreover, they were easily detected by the unaided eye.

Stability of the pH-sensing nanofibrous mat
Long life, reusability, and photostability are the main requirements for halochromic materials. The crosslinked PVA/SA/BC electrospun mats were examined at two temperature conditions of −20°C (storage conditions) and 37°C (body temperature) for 24 h. The electrospun mats developed different colors for pH 4.0, 7.0, and 9.0 ( figure 7). Interestingly, there were no significant changes in color at −20°C and 37°C, demonstrating that the halochromic properties of the electrospun mats were not adversely influenced by temperature. The RGB values of the stored mats were also similar to those of the freshly prepared PVA/SA/BC electrospun mat ( figure 6(b)).

Conclusion
In this study, a PVA/SA-based halochromic nanofibrous mat for wound healing monitoring was prepared using the electrospinning process. The halochromic characteristic was added to the nanofibrous mat with the addition of natural BC anthocyanins into the electrospinning solution. The SEM images showed that electrospinning of the PVA/SA/BC solution produced rarely beaded continuous nanofibers. The PVA/SA/BC electrospun mat was successfully crosslinked by GA. The morphological structure of the electrospun mat did not change significantly after crosslinking processing. However, it was seen from SEM images that flattening and fusions of the nanofibers occurred, and the average diameter of the nanofibers increased to 213±37 nm. The thickness of the halochromic PVA/SA/BC nanofibrous mat was 81.6±4.2 μm and applicable for color changes at different pHs. The PVA/SA/BC nanofibrous mats exhibited halochromic behaviors exactly same to PVA/SA/BC electrospinning solutions, demonstrating that natural pH-indicator anthocyanins (extracted from BC) were successfully integrated into the nanofibrous mats. TGA demonstrated that the crosslinking of the PVA/SA/BC mat increased the thermal stability despite the fact that thermal stability was weakened by the addition of BC extract. The halochromic PVA/SA/BC nanofibrous mat with a short response time presented visibly  distinguishable colors, especially at pHs of 4.0-6.0 and 8.0-10.0, which are consistent with the pH changes in the wound healing process. As a result, the crosslinked PVA/SA/BC nanofibrous mat can be used as a pH sensitive wound dressing for monitoring the healing progress.