The Stability of Chlorite Ion in Electrospun Poly(Vinyl Alcohol) Fibers: pH-Responsive Chlorine Dioxide Release

Abstract

Poly­(vinyl alcohol) (PVA)-based electrospun fiber mats containing sodium chlorite were prepared and characterized to explore their structural properties and chemical behavior relevant to potential self-disinfecting wound dressing applications. Scanning electron microscopy (SEM) revealed that the diameter and uniformity of the fibers are influenced by both the chlorite ion concentration and the pH of the spinning solution. Elemental mapping and ion chromatography confirmed the homogeneous distribution of chlorite ions within the mats. The decomposition of the chlorite ion was found to be pH-dependent, with enhanced stability at alkaline pH and lower chlorite concentrations. At acidic pH, chlorite rapidly decomposes, primarily forming chlorine dioxide, which is a desirable disinfectant. Higher chlorite ion concentrations favor the formation of an unwanted chlorate ion. Gas chromatography confirmed the evolution of ClO₂ over extended time, and ¹H NMR analysis verified that side reactions with the PVA matrix contribute to chlorite ion depletion. The system’s response to skin-like pH conditions demonstrated relatively fast ClO₂ release, underlining the importance of local environmental factors. Overall, low chlorite ion concentrations and a slightly basic pH are required to produce stable mats. The functional electrospun PVA mats presented here are promising candidates for controlled antimicrobial release of ClO₂ in biomedical applications.

Introduction

Combating pathogens, such as bacteria or viruses, is a central issue in medical practice. − In this context, one of the most significant challenges of our time is the emergence and growth of antimicrobial resistance (AMR). − A commonly accepted principle is that the excessive use of antibiotics should be avoided to mitigate adverse effects, including AMR, in agriculture, veterinary care, and human medical practice. − A broad range of antiseptics and disinfectants can help to reduce the overuse of antibiotics and antiviral agents. However, the rise of pathogens being resistant to such agents is a notorious phenomenon either, especially in hospital environments. Numerous studies have addressed the problem of microbial resistance to disinfectants. − Earlier studies led to the conclusion that exploring new ways for delivering well-established biocides with high disinfection efficiency is of utmost importance.

Chlorine dioxide (ClO₂) is a strong, water-soluble oxidizing agent, which is widely used as a disinfectant in water treatment technologies, in the food industry, and in sterilizing medical devices. The effectiveness of ClO₂ against various microbes, including bacteria, fungi, protozoa, and viruses, has been extensively studied in recent years. In many cases, it delivers an excellent microbial kill rate compared to other common disinfectants. − Notably, it is a prominent agent for decontamination of buildings infected with Bacillus anthracis. Several research groups have demonstrated that ClO₂ has remarkable penetrative abilities; thus, it is capable of disinfecting hard-to-reach areas, such as biofilms. − Bacteria and viruses may be killed at low concentration levels of ClO₂ that are not harmful to human or animal cells. − This offers a possibility for in vivo medical applications of this disinfecting agent.

Chlorine dioxide is prone to spontaneous explosions; thus, it is not allowed to be stored or transported in the gaseous state or in high-concentration solutions. In practical applications, ClO₂ is generated on-site, typically by electrochemical oxidation or catalytic decomposition of chlorite ion or by the reduction of chlorate ion. − Custom-made functional materials can be loaded with NaClO₂ and used for delivering this agent to designated surface areas where its decomposition into chlorine dioxide is utilized for eliminating pathogens.

Managing microbial contamination on surfaces is a widespread challenge in biological, health, and consumer fields. − The use of antimicrobial polymer films presents a promising approach to address this issue. − Polymer films containing immobilized antimicrobial functional groups can destroy microbes upon contact. ,, Another strategy is to design polymer films that release biocides to kill microbes, although this method is less efficient, because antimicrobial agents are released even in the absence of microorganisms. − Several polymer systems are known to release antimicrobial ClO₂ gas in response to changes in pH, moisture, − or exposure to visible light. It has been demonstrated that poly­(acrylic acid) gels incorporate ClO₂ for a prolonged time and gradually release it during the disinfection process. Sodium-chlorite-filled electrospun nanofibers have also been produced for the generation and slow on-site release of ClO₂. The disinfection efficiency of this material was tested in the inactivation of Salmonella on tomato. In each of these systems, the spontaneous release of ClO₂ poses a significant challenge for controlling the disinfection process. −

In designing medical-grade wound dressings, it is a central issue how the wound environment, particularly the pH, changes compared to the normal conditions. The pH of healthy skin ranges from 4.2 to 5.8, , but elevated pH levels increase bacterial presence, alter microbial species, and impair the skin’s barrier function. Chronic wounds typically exhibit pH levels between 7.2 and 8.9, which change during the healing process depending on the wound type. −

Wellinghoff et al. produced a TiO₂–NaClO₂ powder by spray-drying aqueous NaClO₂ with a titanium dioxide photocatalyst and extruding this powder into polymer films. Under visible light irradiation, the films released ClO₂ gas. Other approaches include moisture-activated reactions between a polymer matrix-embedded chlorite salt and an acid, generating ClO₂. − , Studies have characterized PVA–NaClO₂ films that exhibit significantly higher ClO₂ production compared to larger NaClO₂ crystals under similar conditions.

Using electrospinning, Palcsó et al. created sodium-chlorite-containing poly­(ethylene oxide) (PEO)-based nanofibrous structures. These PEO–NaClO₂ fibers generated ClO₂ in humid environments and demonstrated the ability to produce ClO₂ gas through in situ carbonic acid formation. Ray et al. developed PLA-based films containing sodium chlorite and citric acid, which released ClO₂ gas upon exposure to moisture. These films were used as vegetable packaging, and the released ClO₂ effectively inactivated Escherichia coli O157:H7 and Salmonella spp. Du et al. created a multifunctional, freshness-preserving fabric using a trilayer nanofibrous membrane. This cellulose acetate/PVA-g-poly­(acrylic acid) /polyurethane (CA/PVA-g-PAA/PU) fabric exhibited prolonged and controlled ClO₂ release, showing excellent antimicrobial efficacy against E. coli, Staphylococcus aureus, and Aspergillus niger. Bai et al. designed a two-component polymer film system with NaClO₂-impregnated acrylate-based polymers and tartaric acid-containing PVA matrices. These films released ClO₂ gas upon moisture activation in various antimicrobial environments.

Now we report a systematic and detailed study on the stability of novel electrospun poly­(vinyl alcohol) (PVA) fibers containing chlorite ions to develop a functional material suitable for wound treatment based on its disinfectant properties. The basic concept is that the chlorite ion exhibits similar features in aqueous solution and embedded in electrospun mats. This makes it possible to control the stability of this compound in the mat and the release of chlorine dioxide. We explore how the concentration of the chlorite ion and the pH of the spinning solution affect the stability of the mats. In addition, it is also demonstrated that the new materials are suitable for delivering ClO₂ efficiently when skin-like pH is set in aqueous environment.

Experimental Section

Chemicals, Spinning Solutions

The poly­(vinyl alcohol) (PVA, 87–90% hydrolyzed, P8136, Sigma-Aldrich) used in this study has a molecular weight of 30,000–70,000 g/mol. Sodium chlorite (80% NaClO₂) was obtained from VWR (Hungary). PVA solution (30 w/v %) and sodium chlorite solutions (10.0, 5.0, 2.0, 1.0 w/v %) were prepared in a 9:1 water–ethanol (VWR) mixture. In the case of PVA, the dissolution was made at 60 °C. The solutions of sodium chlorite contained sodium hydroxide to ensure a slightly alkaline pH. In these experiments, the solutions were prepared using double-deionized and ultrafiltered water obtained from Sartorius Arium Mini Plus and Adrona Integrity+ water purification systems. The mixtures of PVA and sodium chlorite solutions (spinning solutions) were used as precursors for fiber drawing.

Methods

The exact NaClO₂ content of sodium chlorite was determined by iodometric titration using a Metrohm Titrino 721 NET automatic titrator connected to a Metrohm 6.0451.100 combined platinum electrode. The viscosity of the PVA solutions was measured by using an Anton Paar Modular Compact Rheometer 302 (MCR 302) equipped with a vapor hood (proUmid MHG 100) (Table S1). The measurement conditions were as follows: parallel plate (PP50) measuring head, 0.1–1000 s^–1, 25.0 ± 0.5 °C, and 40 ± 2% relative humidity. Because of the high viscosity of these solutions, a Metrohm Double Junction 6.0255.100 combined glass electrode was used to measure its pH.

A Contipro 4SPIN device was used for electrospinning. It is a compact tabletop laboratory instrument consisting of a transparent spinning chamber and a control unit. As an emitter, we used the E6 double jet emitter included with the device, along with a mixing attachment, and a Hamilton G17 needle that was connected to the emitter. A continuously rotating solid cylindrical C3 collector covered with baking paper was used for collecting the fiber mats that formed. During the optimization of the spinning process, we established that the ideal distance between the emitter and the collector was 18 cm, the applied voltage was 38 kV, and the feed rate was 8 μL/min. The collector was operated at 400 rpm during the fiber collection. The air temperature was maintained at 22–24 °C, while the relative humidity was kept between 40 and 45%.

The analysis of the finished fabrics was conducted over an extended period of time, 2–3 months. For this purpose, appropriate amounts of the mats (4–6 mg) were dissolved in 25 mL of water. The amount of sodium chlorite and the concentrations of chloride and chlorate ions formed during its decomposition were determined by ion chromatography using a Thermo Scientific Dionex ICS-5000+ DC-suppressed instrument. The separation was performed with a Dionex IonPac AS27 and IonPac AG27 column, using 0.02 M sodium hydroxide solution as an eluent. A conductivity detector was used in these studies. Retention times were determined using standard solutions as follows: ClO₂ ^–: 5.9 min, Cl^–: 7.5 min, and ClO₃ ^–: 10.9 min.

The formation of chlorine dioxide was monitored by a gas chromatograph coupled to a mass spectrometer (Thermo Scientific Trace GC Ultra Polaris Q). In these experiments, a known amount of fiber was sealed in a vial and stored at room temperature in the dark. The samples were transferred to a TriPlus headspace sampler after different incubation times. During the GC measurements, the samples were kept at 80 °C for 20 min prior to injection. After injection, the samples were kept at 60 °C for 2 min, then were heated to 100 °C at a rate of 16.0 °C/min, where they were held for 1 min. The peak of chlorine dioxide appeared at 1.8 min.

The morphology of the fibers was examined by high-resolution, low accelerating voltage scanning electron microscopy (LVSEM, Thermo Fisher Scientific Scios 2 dual beam). Previous studies on aerogel samples have demonstrated that gold coating affects the morphology of the samples. Therefore, our experiments were also performed without depositing a conducting coating on the surface. The samples were mounted on vacuum-resistant carbon tape; the accelerating voltage was 2 kV, and the working distance was 3 mm. On the basis of energy-dispersive X-ray analysis (EDX), the elemental map of the surface was constructed to provide insights into the homogeneity of the fabrics. The X-ray maps were taken from a relatively large area to obtain a general picture of the element distribution. The scale bar in the images is 40 μm. The accelerating voltage was 10 kV for the excitation of Cl and Na atoms.

NMR measurements were conducted on a Bruker DRX 400 (9.4 T) spectrometer with a Bruker VT-1000 temperature controller and a BB inverse z-gradient probe (5 mm). The water signal was suppressed in aqueous samples using a watergate pulse sequence (12.6 dB) to suppress the 4.8 ppm of proton signal. Each NMR sample included a sealed capillary containing DSS (4,4-dimethyl-4-silapentane-1-sulfonic acid) dissolved in D₂O as an external standard for ¹H NMR chemical shift calibration. The DSS peaks in the spectra were observed at 0.00 (singlet), 0.63 (triplet), 1.76 (multiplet), and 2.91 ppm (triplet). For ¹H NMR spectra, each experiment consisted of 32 scans with an acquisition time of 1.366 s and a 90° pulse. Spectra were analyzed using the Bruker WinNMR and MestReNova software packages.

The data obtained from viscosity measurements were analyzed by using RheoCompass (Anton Paar) software. Chromatograms from ion chromatography were evaluated with Chromeleon (Thermo Scientific), while gas chromatography results were analyzed with xCalibur (Thermo Scientific) software. The average fiber diameter of the fabrics was determined from SEM images using Fiji ImageJ software. Data visualization was performed with OriginPro 2018 (OriginLab Corp.) software.

Results and Discussion

Scanning electron microscopy (SEM) images of the produced fiber mats feature randomly arranged, smooth-surfaced nanoscale fibers (Figure a). The fibers appear less uniform at higher salt concentrations and tend to merge at crossing points. However, the fibers become less fused, more uniform, and distinct when the sodium chlorite content decreases in the samples (Figures S1–S5). The fiber diameters, determined from the SEM image, show normal distribution for all samples (Figure b), and the average diameter depends on the sodium chlorite content of the mats (Table and Figure ). At the highest salt concentration (14.89 w/w %), the average fiber diameter is 68 ± 13 nm, and it increases to 150 ± 29 nm as the salt content decreased to 2.44 w/w %.

1.

SEM image of the fiber mat containing 2.44 w/w % chlorite ion (a) and the distribution of fiber diameters (b). pH = 7.80 (spinning solution).

1. Fiber Diameter as a Function of Chlorite Ion Concentration and pH.

c NaClO2 (w/w %)	pH	fiber diameter (nm)
14.89	7.80	68 ± 13
8.89	7.80	77 ± 22
4.15	7.80	100 ± 21
2.44	7.80	150 ± 29
4.15	7.56	182 ± 40
4.15	7.12	110 ± 36

^apH of the spinning solution.

2.

Fiber diameter as a function of sodium chlorite concentration in the fiber. pH = 7.80 (spinning solution).

At a constant sodium chlorite concentration, the effect of the spinning solution’s pH on fiber diameter was also studied. Based on SEM images, the largest fiber diameters were obtained at pH = 7.56 (Table ).

The Na and Cl elemental maps were created using EDX-SEM (Figure ). The uniform distribution of Na and Cl atoms across each sample confirms that the impregnating agent is homogeneously distributed in the electrospun mats. This result was corroborated as follows. Samples were taken from different areas of a fiber mat and dissolved in water. The chlorite ion concentration of these solutions was measured by ion chromatography. The results confirmed that the chlorite ion concentration varied by less than 3% within a 15 × 30 cm fiber mat (Figure S6).

3.

Distribution of Cl (a) and Na (b) atoms in the fiber mat containing 2.44 w/w % chlorite ion constructed on the basis of EDX analysis. pH = 7.80 (spinning solution).

Under acidic conditions, the disproportionation of the chlorite ion occurs and produces chlorine dioxide (ClO₂), chlorate ion (ClO₃ ^–), and chloride ion (Cl^–). The actual stoichiometry of this reaction is strongly dependent on the concentration of the reactant and the pH. This feature can be interpreted by considering the linear combination of eqs and . In eq , only chlorine dioxide is formed alongside the chloride ion, while the chlorate ion is also a product in eq . At high chlorite ion concentrations, a further reaction path is also operative, in which only chlorate ion is formed in eq . −

At pH ∼7, the decomposition of ClO₂ ^– occurs in a similar manner, producing the same chlorine-containing products. These reactions cause a slight increase in pH. Measuring such small pH changes within the matrix is not feasible. Because of this effect, the rate of ClO₂ ^– decomposition slows and eventually stops.

$$5\mathrm{HCl}{\mathrm{O}}_2=4\mathrm{Cl}{\mathrm{O}}_2+{\mathrm{Cl}}^{-}+{\mathrm{H}}^{+}+2{\mathrm{H}}_2\mathrm{O}$$ 1

$$4\mathrm{HCl}{\mathrm{O}}_2=2\mathrm{Cl}{\mathrm{O}}_2+\mathrm{Cl}{\mathrm{O}}_3^{-}+{\mathrm{Cl}}^{-}+{2\mathrm{H}}^{+}+{\mathrm{H}}_2\mathrm{O}$$ 2

$$3\mathrm{HCl}{\mathrm{O}}_2=2\mathrm{Cl}{\mathrm{O}}_3^{-}+{\mathrm{Cl}}^{-}+{3\mathrm{H}}^{+}$$ 3

In the case of potentially self-disinfecting wound treatment materials, the production of chlorine dioxide is advantageous, while the formation of the antagonistic chlorate ion should be avoided.

The concentrations of chlorite, chloride, and chlorate ions in the electrospun mats were measured on the day of preparation, and their variation was monitored by ion chromatography over an extended period of time. In these studies, technical grade NaClO₂ was used that contained NaCl in substantial concentration, which was determined in separate experiments and used to correct the chloride ion concentrations determined in the samples. In additional experiments, the concentration of chlorine dioxide was also determined by gas chromatography. It needs to be noted that only a relatively small amount of ClO₂ is formed during the decomposition, and its concentration is far below its explosion limit (9.5% [ClO₂]/[air]). In these experiments, approximately equal amounts of electrospun mats were placed in a series of gastight vials and stored in the dark. The headspace of the individual containers was analyzed at increasing incubation times. All results were normalized to a unit mass of the samples, as shown in Figure .

4.

Change in the concentration of chlorite ion (ClO₂ ^– black square), chlorate ion (ClO₃ ^– blue triangle up), chloride ion (Cl^– red circle solid), and chlorine dioxide (ClO₂ pink triangle down) as a function of time. The amount of the given species (expressed in moles) is normalized to the unit mass of the sample. c _NaClO2 = 14.89 w/w %, pH = 7.80 (spinning solution).

The concentration of chlorite ion in the freshly prepared fiber mats was always about 30% less than expected on the basis of the composition of the spinning solutions. This deficit is partly due to the removal of an aqueous solution containing chlorite ion from the spinning chamber by continuous air exhaustion. The amount of PVA transferred to the fiber is less likely affected by this physical process. This concept was proven by putting filter paper impregnated by potassium iodide into the stream of exhaust gas. During the electrospinning process, the color of this paper turned yellowish because of the oxidation of the iodide ion to iodine by the chlorite ion.

PVA-based fiber sheets were prepared from solutions of different initial chlorite ion concentrations at pH 7.80 (Table S1). At higher concentrations of chlorite ion, its concentration decreases to less than half in 10–15 days. In contrast, the 2.44 and 4.15 w/w % samples remained stable, and the decomposition of chlorite ions did not exceed a 15 and 30% loss, after 2 months, respectively (Figure ).

5.

Concentration decay of chlorite ions as a function of time in electrospun mats at different initial concentrations of NaClO₂. c _NaClO2 = 14.89 w/w % (black square), 8.89 w/w % (red circle solid), 4.15 w/w % (blue triangle up), 2.44 w/w % (pink triangle down), pH = 7.80 (spinning solution).

It has long been known that the decomposition of chlorite ion is a strongly pH-dependent process. Therefore, the stability of the fiber sheets was studied by varying the pH at a constant chlorite ion concentration. Results of ion chromatographic measurements confirmed that the decomposition process occurs significantly faster at lower pH (Figure S7). The most stable fiber sheets were produced with relatively low chlorite ion concentrations (<4.15 w/w %) and at pH > 7.50.

We also studied the stoichiometric ratio of the products formed during the decomposition process. According to the results, chlorine dioxide formation is dominant at low chlorite ion concentrations (eq ), while at higher chlorite ion levels, significantly more chlorate ions are produced (eqs and ). Thus, the n _ClO2 /n _ClO3– ratio gradually increases when the chlorite ion concentration of the electrospun mats decreases (Table ).

2. Ratio of the Products at Different Chlorite Ion Concentrations in the Electrospun Mats at Two Different Incubation Times .

t (days)	c NaClO2 (w/w %)	n ClO2 / n ClO3–
14	14.89	0.30
14	8.89	0.87
14	4.15	1.90
14	2.44	2.60
36	14.89	0.48
37	8.89	0.96
36	4.15	1.40
35	2.44	2.20

^apH = 7.80 (spinning solution).

^bNormalized by the mass of the fiber mats.

As the reaction progresses, the amount of transformed chlorite ion should be equal to the total amount of products formed. Based on the ion and gas chromatographic results, the total amount of products was 10–30% less than expected (Table ). This can be explained by the fact that chlorite ion is not only consumed as shown in eqs and but also involved in oxidation side reactions with the PVA matrix. Analyzing the samples by ¹H NMR spectroscopy, a new peak was observed at 8.44 ppm that corresponds to an oxidation byproduct in the fiber in the presence of chlorite ion (Figure ). This peak was not observed in the ¹H NMR spectrum of the fiber sheet made from pure PVA. In addition to the noted exhaustion, this is another reason the measured chlorite ion concentration is significantly smaller in the electrospun mats than expected on the basis of the composition of the spinning solutions.

3. Amount of Transformed Chlorite Ion and the Products.

c NaClO2  (w/w %)	Δ c ClO2– (mol/g)	c products (mol/g)
14.89	1.44 × 10–3	1.01 × 10–3
8.89	4.61 × 10–4	3.38 × 10–4
4.15	1.30 × 10–4	8.53 × 10–5
2.44	4.24 × 10–5	3.77 × 10–5

6.

¹H NMR spectrum of a chlorite ion-containing PVA fiber mat (red) and a pure PVA fiber mat (black). c _NaClO2 = 14.89 w/w %, pH = 7.80 (spinning solution).

To test how chlorite-ion-containing PVA mats behave at around the pH of healthy skin, exactly known amounts (6–8 mg) of the mat containing 14.89 w/w % sodium chlorite were put in 20 mL vials. After adding 0.2 mL of acetate buffer (pH = 4.87, 0.015 M Na-acetate +0.015 M acetic acid), the vials were immediately sealed, and their headspace was analyzed for chlorine dioxide after different incubation times. As shown in Figure , a significant amount of ClO₂ forms already after ∼80 min (the first measured point), i.e., the decomposition of chlorite ion is orders of magnitude faster than in the solid mats at higher pH. This observation can straightforwardly be interpreted by considering the well-known acid-catalyzed decomposition of chlorite ion. In this case, the PVA mat gradually dissolves in the solution, thus assisting the decomposition process. It needs to be emphasized that the experiment clearly demonstrates the pH dependency of the decomposition of chlorite ion, but the results are not suitable for drawing direct conclusions regarding the interactions between the electrospun mats and skin. Under real conditions, the volume of the liquid phase is small on the surface of the skin, and due to its limited buffer capacity, the pH probably does not remain constant during the whole process. In any case, it is safe to conclude that the rate of chlorine dioxide formation is strongly influenced by the condition of the skin at the actual location of the disinfection.

7.

Formation of chlorine dioxide from an electrospun PVA mat containing 14.89 w/w % sodium chlorite upon addition of acetate buffer at pH 4.87. m _fiber = 6–8 mg (exactly known in the individual samples); acetate buffer: 0.015 M Na-acetate +0.015 M acetic acid; V _vial = 20 mL.

Conclusions

This study demonstrates that electrospun PVA-based fiber mats containing sodium chlorite exhibit tunable structural and chemical properties that are critically influenced by the salt concentration and pH of the spinning solution. The use of lower chlorite ion concentrations and slightly alkaline conditions (pH > 7.5) results in more uniform and stable fibers; i.e., the reaction with the PVA matrix and the decomposition of chlorite ion become considerably slower. The distribution of sodium and chlorite ions within the mats is homogeneous, and the mats efficiently release chlorine dioxide under acidic conditions, mimicking the pH of the skin. The pH-dependent decomposition can be utilized to control ClO₂ release, which is highly desirable in applications such as self-disinfecting wound dressings. At higher chlorite ion concentrations and lower pH, chlorate ion, an unwanted byproduct, also forms. This can diminish the effectiveness and safety of the material. The interaction between chlorite ion and the PVA matrix was evidenced by ¹H NMR, indicating oxidative side reactions that further impact the stability of the mats. Overall, careful optimization of the chlorite ion concentration and pH is essential to maximize the antimicrobial potential while minimizing undesired chemical transformations, paving the way for safe and effective biomedical use.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.5c13188.

• Viscosity of the spinning solutions (Table S1); SEM images of the fiber mats containing different chlorite ion concentrations and the distribution of fiber diameters; homogeneity of sodium chlorite content in the electrospun fiber; and concentration decay of chlorite ions as a function of time at different pHs (Figures S1–S9) (PDF)

The authors declare no competing financial interest.