2.1 Materials

l-aspartic acid (≥98%, Sigma) and o-phosphoric acid (≥99%, Sigma-Aldrich) were used in PSI synthesis. N,N-dimethylformamide (≥99.5%, Fisher Scientific), and CaCl₂ (≥ 99.0%, Reanal) were used in the electrospinning.

All chemicals were used during the experiments without further purification.

2.2 Preparation of the Electrospun Meshes

In the first step, PSI was synthesized according to the previously published process^{[24, 26]} and used to generate fiber meshes. Table 1 shows the composition of the different PSI solutions used in electrospinning.

For electrospinning, two devices were used: a homemade one connected to a fume hood and a Spincube (Spinsplit, Hungary) type compact device with a built-in power supply and sensors for measuring RH and temperature to achieve high RH. The homemade device included a Genvolt 73030P (Genvolt Ltd., Shropshire, United Kingdom) type high-voltage DC power supply together with a KDS 100 (KD Scientific Inc., Holliston, USA) type syringe pump. The distance between the needle and the collector was always set to 15 cm and the volumetric flow rate was 1 mL h⁻¹. The RH and temperature were measured with a Testo 610 manual hygro and thermometer (Testo Ltd., Alton Hampshire, United Kingdom) within the electrospinning chamber of the homemade device. The humidity was controlled between 35% and 65% relative humidity level by using a Trotec B7E ultrasonic humidifier (Trotec, Heinsberg, Germany). Silica gel was put into the electrospinning device to achieve a low relative humidity level. The applied voltages during electrospinning are summarized in Table 1.

2.3 Scanning Electron Microscopy

For SEM, pieces of fiber samples were placed on metal studs with carbon-loaded double-sided tape and conductive silver paint. In the case of insulating samples (such as polymers), the coating was necessary to perform high-resolution imaging. Here, the samples were carbon coated (thickness of about 200 Å) using an AUTO 306 carbon evaporator (BOC Edwards, West Sussex, United Kingdom). Observations were performed with a JEOL JSM 7800F LV field emission SEM (JEOL, Akishima, Tokyo, Japan). Secondary electron images were acquired with an accelerating voltage of 2 keV, a low probe current value, the smallest objective lens aperture, and a working distance of 10 mm to limit polymer degradation. The determination of the average fiber diameter was carried out by the ImageJ program (Figure S1, Supporting Information). The diameter of 100 fibers was measured from at least three different images with a magnification of 5000× or 2500×, from which the average of the diameters, as well as the standard deviation, were calculated.

2.4 Conductivity Measurement

Conductivity measurements were carried out for two different purposes. First, the CaCl₂ content of the fibrous meshes was determined by measuring the CaCl₂ release from the meshes with a Mettler–Toledo SevenCompact Duo S213 Benchtop pH/Conductivity meter (Mettler-Toledo, Switzerland). After weighing the samples, it was immersed in 20 mL of distilled water and stirred at 100 rpm with an IKA RCT basic magnetic stirrer (IKA-Werke GmbH & Co., Staufen, Germany). The conductivity was measured every 10 min in the first hour, after that, once every hour until it reached saturation. Second, the precursor polymer solutions used for electrospinning were measured with the same equipment to determine the effect of the salt concentration on the conductivity.

2.5 Viscosity Measurement

To measure the viscosity of the precursor solutions, an SV-10 Sine-wave Vibro-type vibrational viscometer was used (A&D Company Ltd, Japan). The precursor solutions were loaded into a cuvette which was thermostated at 25 °C (similar to the temperature applied during electrospinning). The equipment measured viscosity by detecting the electric current necessary to resonate the sensor plates at a constant frequency of 30 Hz and amplitude of less than 1 mm. The measurement was carried out until the viscosity remained constant.

2.6 Determination of the Pore Size Distribution Based on SEM Image Analysis

The pore size determination was carried out by analyzing the SEM images with 2500× magnifications by using ImageJ based on the protocol published by Hojat et al.^[27] First, the Set Scale tool was used to establish an accurate measurement scale for the image. Then, a cropping was performed to eliminate the bottom information band. The Window/Level tool was then applied to optimize the brightness and contrast ratios. Following this, the Threshold tool was employed to identify the specific intensity range (55–65%) that retained focused fibers while eliminating background elements. After thresholding, the Make Binary tool facilitated the conversion of the image to binary, which was then inverted to make fibers black against a white background. Finally, the Analyze Particle tool was employed to create fiber outlines and define the inter-fiber spaces. The presence of small fibers prevented the use of Erode or Dilate tools, resulting in artifacts resembling small pores. To mitigate these artifacts, the Z-Scoring method was applied, with a defined threshold of 0.05 points to provide accurate results. Statistical evaluation of the pore size data was carried out using the Mann–Whitney U test. A difference was considered statistically significant if p < 0.05.

2.7 Confocal Microscopy Measurement

To demonstrate the 3D structure of the fibrous matrices on a microscopic level, the autofluorescence property of PSI^[20] was measured by confocal microscopy. The Z-stack measurements were carried out by Nikon Eclipse Ti2 Inverted Research Microscope (Nikon, Japan) with a 20× objective. The 3D images were taken by the Abberior Imspector software and processed by the ImageJ program. To create 3D images, first, the Window/Level tool was applied to optimize the brightness and contrast ratios of each stack. Afterward, the Smoothing plugin was used and Gaussian smoothing with sigma = 1 was applied. The 3D images were constructed by 3Dviewer plugin.

2.8 Attenuated Total Reflectance Fourier-Transformed Infrared Spectroscopy

The measurements were conducted with a Jasco FTIR-4700 spectrometer with an ATR PRO ONE Single-Reflection Diamond head (Able&Jasco, Japan). The wavenumber range was 4000–400 cm⁻¹, the resolution was 2 cm⁻¹, and every spectrum was calculated from the average of 128 scans. Before every sample, the background was measured to get a more accurate spectrum. The results were evaluated by using the Jasco Spectra Manager program, by applying baseline, ATR head correction, and means-movement smoothing.

2.9 Thermogravimetry Measurement

TGA spectra of the electrospun meshes and powder mixtures were carried out by using a TGA Q500 thermogravimeter (TA instruments, USA). The tests were carried out on samples with a mass of ≈5 mg between 30 and 700 °C at a heating rate of 10 °C min⁻¹ under nitrogen purge flow (60 mL min⁻¹).

2.10 X-ray Diffraction Measurement

XRD studies were performed using a D8 Advance type diffractometer (Bruker AXS, Karlsruhe, Germany) with a Cu-Kα cathode (λ = 1.54 Å) operating at 40 mA current and 40 kV voltage. The collection of data has been performed by a LYNXEYE XE-T type linear detector. The scan rate was 4.8° min⁻¹ with a scanning step of 0.02° in the 2Θ range of 5° to 75°.

2.11 Statistical Analysis

Statistical evaluation of the fiber diameter data was carried out using the Kruskal–Wallis nonparametric ANOVA followed by a median test. A difference was considered statistically significant if p < 0.05.

For the presentation of the effect of different factors “The Design & Analysis of Experiments: Central composite, non-factorial, surface designs” was used by Statistica 14.0.1.25 (Copyright 1984–2020 TIBCO Software Inc.). The dependent factor was the diameter (µm), while the independent factors were humidity (%), and the CaCl₂/PSI (m/m) ratio.

3.1 Effect of the Polymer Concentration on the Microstructure and Diameter of the Fibers

The SEM images in Figure 1 show the dependence of the fiber morphology on the PSI concentration.

As the results show, the PSI concentration highly influences both the microstructure and the average diameter of the electrospun fibers. Significant bead formation can be observed on the SEM image at 20 w/w% PSI concentration, but the increase in the PSI concentration eliminated these structural artifacts (Figure 1). On the other hand, the increase in the PSI concentration resulted in a higher average fiber diameter and a wider diameter distribution which differences proved to be significant according to the statistical analysis (Table 2). These effects can be explained by the higher viscosity of the precursor polymer solution since the polymer concentration has no noticeable effect on the conductivity. These tendencies correlate well with the literature data about the precursor solution viscosity's effect and the previous findings related to PSI fibers.^[28]

3.2 Effect of the CaCl₂ Concentration on the Micro and Macrostructure of the Electrospun Meshes

To investigate the effect of the concentration of CaCl₂ at different polymer concentrations, consistent ambient RH (48–55%) and electrospinning parameters were used. Adding CaCl₂ to the 25 w/w% PSI solution drastically increased the viscosity and induced gelation, thus fiber formation could not be carried out. Therefore, the effect of CaCl₂ on the fiber formation, as well as on the micro and macrostructure of the fibers was investigated only at 20 and 22.5 w/w% PSI concentration. SEM micrographs were used to present the morphological changes, while the macroscopical appearance of the samples was demonstrated by photos taken directly after electrospinning (Figure 2).

As the SEM images show, introducing CaCl₂ to the PSI solutions resulted in mostly homogenous fibers and significantly reduced the bead formation in the case of 20 w/w% PSI concentration, especially at 5% CaCl₂ (Figures 1 and 2 and Figure S2, Supporting Information, and Table 3). The stabilization effect of the salt can be explained by the enhanced conductivity of the precursor solution. Adding salt to the polymer solution increases the charge density on the polymer and the surface of the polymer jet; therefore, the electrostatic force pulling the jet toward the collector is higher. This causes a higher elongation of the polymer jet which can eliminate the bead formation.^{[29, 30]} Besides this effect, the presence of CaCl₂ increased the viscosity of the precursor solution, which resulted in a higher average fiber diameter in almost all the cases (Table 3). Furthermore, at the conditions where a 3D mesh formation was induced by CaCl₂, the concentration of the salt did not have a significant effect. However, the effect of the concentration of CaCl₂ on the microscopic and macroscopic structures of the samples proved to be slightly different in the case of the two PSI concentrations. At 20 w/w% PSI concentration, adding CaCl₂ at 5 w/w% concentration resulted in homogenous fibers with smooth surfaces and higher average fiber diameter, while adding 7 w/w% CaCl_2, the average fiber diameter decreased. However, we can observe on the SEM images (Figure 2 and Figure S2, Supporting Information) that the higher CaCl₂ concentration induced significant changes in the microstructure of the PSI mesh. Several fused fibers can be observed. Furthermore, big spherical blobs appeared that also contained several fused fibers, which change made the determination of the fiber diameter challenging (Figure 2 and Figures S2 and S3, Supporting Information). This change can be explained by the slow gelation of the precursor solution, which mainly occurred during fiber formation. This gelation can be explained by the complex formation between the C═O group of the PSI molecules and the Ca²⁺ ions, which becomes more significant at higher CaCl₂ concentrations.^{[31, 32]} At 22.5 w/w% PSI concentration, 5 w/w% of CaCl₂ increased the average fiber diameter and resulted in a wider distribution (Figure S3, Supporting Information) compared to 3 w/w%, which can be explained by the higher viscosity.^[33] He et al. experienced a similar effect in the case of the electrospinning of polyvinylidene fluoride nanofibers at different FeCl₃ concentrations.^[34] At this composition (22.5%PSI/5%CaCl₂) the gelation did not occur.

Besides the change in the average fiber diameter, the addition of CaCl₂ to the precursor solution affected the surface morphology of the fibers and induced wrinkled fiber formation, which is more significant at higher CaCl₂ concentrations (Figure 2). The occurrence of the wrinkled fiber surface is explained by the formation of thin skin on the surface of the liquid jet, which collapses,^[35] by the formation and collapse of droplets formed on the surface of the jet,^[36] or by the phase separation on the surface of the polymer jet.^[37] However, inorganic salt additives are able to precipitate on the surface of the fibers as was demonstrated in the case of nylon-6/GaCl₃,^[38] polyamide-6/CaCl₂,^[39] and chitosan/different salt fibers^[40] resulting in amorphous surface structure. Pai and co-workers also demonstrated that the surface morphology of polystyrene fibers significantly depends on the relative humidity of the environment, and low humidity favors the formation of wrinkled fibers.^[41]

The macroscopic appearance was also influenced by the concentration of CaCl₂. As can be seen in Figure 2, at higher CaCl₂ concentrations, the fibers were collected at a lower position than the spinning needle, which indicates heavier fibers than at lower CaCl₂ concentrations. This can be explained by the higher density of the CaCl₂ compared to the PSI. Higher CaCl₂ concentrations induced gelation of the PSI solutions after a certain time which can also explain the microscopic and macroscopic inhomogeneities.

According to the previous observations and the literature, CaCl₂ and PSI concentration, their ratio are all crucial parameters, influencing the micro and macrostructure of the electrospun meshes. Furthermore, RH can influence these fiber properties which effect can be also more relevant when CaCl₂ is present due to its strong interaction with vapor water. For a better understanding of the effect of humidity and to exert the gelation effect, only 20%PSI/5%CaCl₂ and 22.5%PSI/3%CaCl₂ samples were used in the following experiments.

3.3 Effect of the Relative Humidity on the 3D Structure of Electrospun PSI/CaCl₂ Fibers

In the previous experiment, we demonstrated that adding CaCl₂ to the precursor solution induced 3D fiber formation regardless of the different PSI/CaCl₂ ratios, and also affected the microscopic structure of the fibers. In this section, our intention is to demonstrate that the RH has an important role in 3D fiber formation, therefore fibers were produced at different RH levels and the obtained structures were analyzed (Figures 3 and S4, Supporting Information).

As the SEM images show in Figure 3, fibers have no defect at any RH level at the microscopic scale, but the effect of humidity has a high impact on the surface morphology and macroscopic structure of the meshes. Under 48% RH, there was no 3D formation of the meshes and the fibers obtained have smooth surfaces with constant diameters lengthwise. Above 48% RH, and if CaCl₂ was added to the precursor solution, meshes started to grow out from the collector plane and formed a random 3D structure. According to the pictures, the higher RH caused a more robust 3D formation, and at some point, the meshes reached the tip of the needle (Video S1, Supporting Information). On the other hand, no 3D formation was experienced without adding CaCl₂ to the precursor solution at any RH level (Figure S5, Supporting Information), which results indicate that interaction between the water in the vapor phase and CaCl₂ can be crucial for 3D formation of electrospun PSI meshes.

The effect of both the humidity and inorganic additive on the 3D formation of electrospun meshes was already demonstrated in the literature, however, the connection between the two parameters has not been presented.^{[17, 42]} Cheng et al. demonstrated that RH level significantly influences the macroscopic appearance of the cellulose acetate electrospun fibers. They experienced that under 50% RH, there was no 3D formation; furthermore, the increase of RH significantly influenced the macroscopic morphology of the meshes, which findings are in good correlation with our results.^[15] Huang and co-workers presented similar results during the production of silk-fibroin fibers at low temperatures.^[43] To produce 3D fiber macrostructures, inorganic or polymer additives are mainly used in the sacrificial agent strategies,^[44] although few examples can be found where the mixture of polymer and inorganic precursor solution resulted in 3D inorganic nanofibrous scaffolds.^{[13, 45, 46]}

The microscopic 3D structure of the 20%PSI/5%CaCl₂ and 22.5%PSI/3%CaCl₂ fibrous matrices is demonstrated by using Z-stack images captured by confocal microscopy (Figure 4A). The images show the top 40 µm depth of the samples since at more depth no fluorescence signal was detected due to the scattering and absorption of the emitted photons. The images represent the loose structure of the fibers and show several voids between the fibers with a size of several tens of micrometers (Figure S6, Supporting Information), which is ideal for example biomedical applications in wound healing or tissue engineering.^[11] As we represent in Figure 4B, the random structure of the fibrous matrix can be shaped according to the final application.

To determine the pore-size distribution of the samples and compare the 3D structure of the samples on a microscopic level, the SEM images were analyzed according to the protocol published by Hotaj and co-workers^[27] (Table 4). The distribution of the pore sizes shows that each sample contains a high number of small pores (1–3 µm) and the number of the pores decreases by their size quasi exponentially (Figure S6, Supporting Information) similar to the literature data.^{[47, 48]} The results show that the RH did not affect the pore size when CaCl₂ was not applied during the electrospinning while RH showed a slight effect on the pore size when CaCl₂ was mixed in the precursor solution (Table 4). In these cases at 48% RH level the mean, the median, and the maximum value of the pores were increased significantly according to the statistical analysis which is the consequence of the 3D mesh formation.^[42]

3.4 Effect of the Relative Humidity on the Microstructure of Electrospun PSI/CaCl₂ Fibers

The effect of the humidity on the microstructure and diameter distribution of electrospun polymer fibers is thoroughly investigated and presented in the literature,^[17] but not in connection with salt additives. To demonstrate the effect of the humidity on the fiber diameter, its distribution, and the average fiber diameter were determined (Figure 5).

As the results show, the higher the humidity, the higher the average fiber diameter is (Figure 5A) in the case of both compositions of the precursor solution with CaCl₂. However, distribution graphs show that at higher RH, the distribution of the fiber diameter is getting wider, and for example, in the case of the 22.5%PSI/3%CaCl₂ two populations can be seen at 65% RH. On the other hand, if no CaCl₂ was added to the precursor solution, the humidity did not show any influence on the average fiber diameter (Figure 5A) but higher RH significantly reduced the bead formation (Figure S5, Supporting Information). A similar effect on the bead formation was observed in the case of the electrospinning of poly(carbonate urethane),^[49] polystyrene,^[50] and poly(vinyl chloride) fibers.^[50] Besides the fiber diameter, the humidity also had a high influence on the microstructure of the fibers containing CaCl₂. As we described previously, the addition of CaCl₂ induced the formation of wrinkled fibers (Figure 2), which effect proved to be RH dependent. Similar dependency on the RH was already described in the case of polyetherimide,^[51] polyacrylonitrile,^[52] and polycaprolactone fibers.^[53] Besides, wrinkled fibers at 55% RH ribbon-like structure with a wrinkled surface were obtained, which was more relevant in the case of 20%PSI/5%CaCl₂ (Figure 5B). This finding correlates well with the fact that humidity-induced change in the surface morphology of hydrophobic fibers was described between 40% and 50% RH.^[17] These results indicate that the change in the surface morphology is caused by the non-solvent effect of the vapor water, which at a low RH level induces a wrinkled surface, and later on ribbon-like structure.^[17] At a high RH, the solidification of the PSI fiber is significantly faster, which results in the complete solidification of the polymer jet therefore the fiber diameter also increases^[54] (Figure 5A). According to these results, the joint effect of the humidity and CaCl₂ is affecting both the macroscopic appearance of the fibers (3D structure) and the microscopic structure as well, since the crystallization and water adsorption of CaCl₂ is highly humidity dependent. Abid et al. also described that the presence of a certain hydrophobic molecule, namely phenylalanine affects the size and structure of CaCl₂ nanoparticles during crystallization from aerosol.^[55] Our results and the previously mentioned finding highly indicate the interaction between the CaCl₂ and vapor state water which can occur not just in crystal formation but also at a molecular level.

According to the surface design fitting and the statistical analysis (Figure 5C and Table S1, Supporting Information) the two independent factors, namely the CaCl₂/PSI (m/m) and RH (%), show a significant linear effect (L) on the dependent factor, namely the fiber diameter (µm). Moreover, the interaction of the linear part of the two independent factors also significantly affects the fiber diameter. These are indicated by the low p-values which are lower than 0.05. By comparing the two independent factors with each other we can recognize that the quadratic factor (Q) of the CaCl₂/PSI (m/m) has a significant effect on the fiber diameter while the Q factor of RH is not significant. This result indicates that the CaCl₂/PSI (m/m) has a higher influence on the fiber diameter than the RH (Table S1, Supporting Information).

We also made an effort in the statistical analysis to investigate the effect of the PSI and CaCl₂ content on the fiber diameter (Table S2, Supporting Information). This analysis proved that the linear effect of all the independent factors, as well as the interaction of the linear parts of the PSI content (%) and humidity (%) (1L by 3L), and CaCl2 (%) and humidity (%) (2L by 3L) on the fiber diameter, is significant. These results prove that both CaCl₂ and RH are key elements in the fine-tuning of the microscopic structure of the PSI fibers and show a strong interaction between the parameters.

3.5 FTIR, TGA, and XRD Analysis of PSI/CaCl₂ Fibers and Powder Mixtures with the Same Composition

Since CaCl₂ proved to be crucial for the 3D fiber formation, we aimed to investigate the relationship between the polymer, CaCl_2, and vapor water. At first, the presence of CaCl₂ in the fibrous meshes was proved by measuring the release of the CaCl₂ by conductometry (Figure S7, Supporting Information). As the results show, almost all of the CaCl₂ was released from both samples after 2 h (20%PSI/5%CaCl₂ 96%, 22.5%PSI/3%CaCl₂ 82%) which results prove the successful incorporation of the CaCl₂ into the PSI fibers.

To investigate the composition of the different fibers and the interaction between the water and CaCl₂, FTIR spectroscopy (Figures 6A and S8, Supporting Information), XRD (Figure 6B), and TGA analysis were carried out (Figure 5C). FTIR measurement of dehydrated and hydrated CaCl₂ shows that pure CaCl₂ signals are very weak, and a small amount of water can cover the characteristic peaks of CaCl₂. The peaks of the water at 1630 and 3250 cm⁻¹ increased very quickly after the dehydration of CaCl₂, which can be explained by the water absorbed from the air (Figure S8A, Supporting Information).

The appearance of the peak at 1630 cm⁻¹ in every electrospun sample proves that all of the samples contain some water, and its quantity shows an increasing tendency with the CaCl₂ concentration_. If we compare the spectrum of the electrospun fibers with a mixture of the same components in powder form (Figure 6A and Figure S8B, Supporting Information), we can observe that the relative intensity of the water peak compared to the intensity of the peak at 1710 cm⁻¹ is higher and it increases more with the concentration of the CaCl₂ (Figure 6A). The peak at 1710 cm⁻¹ peak represents the asymmetric vibration of the C═O group in the PSI. This means that the water content of the electrospun fibers is smaller than that of the same materials mixed together in a powder form, which indicates that the water content of the electrospun fibers may be found mainly in the crystalline state in CaCl₂·XH₂O. The other relevant difference between the spectra of the electrospun fibers and the powder mixtures is the exact position of the vibration peak of the C═O group at around 1713 cm⁻¹. While it shifts to lower wavenumbers in the case of the electrospun fibers (Figure 6A), it does not shift for the powder samples (Figure S8B, Supporting Information). This change indicates the interaction between the PSI and the CaCl₂ at a molecular level. Wei et al. described that the addition of CaCl₂ to polyamide-6 fibers induces the shift of the C═O characteristic peak on the FTIR spectra, which shift is proportional to the CaCl₂ content.^[56] They hypothesized that the Ca atom coordinated with the O atom in the C═O group, therefore, the Ca atom became negatively, while the O atom became positively charged. Since PSI contains two C═O groups in every monomer, we can apply the same assumption to our system.

To support the FTIR results, XRD measurements were carried out on the fibrous samples with and without CaCl₂. Inorganic additives can alter the crystal structure of polymers resulting in hybrid crystal structures.^[57] As we can see in Figure 6B, the CaCl₂ applied in our experiment is completely crystalline and contains a mixture of CaCl₂·4H₂O and CaCl₂·6H₂O, which were formed due to interaction with humidity. As we can see, pure PSI fibers can be considered almost completely amorphous, as was also described in our previous paper.^[24]

The addition of calcium chloride to PSI fibers resulted in an increase of amorphous halo between 40 and 60 2Θ degrees, with some small peaks corresponding to the crystalline state of CaCl₂-based hydrates. The lack of a high number of peaks shows that the dispersion of CaCl₂ and its hydrates is uniform, similarly as was described by Jaworska et al.^[58] These findings support our previous hypothesis and indicate that the solidification of PSI and CaCl₂ mainly occurs simultaneously. These results lead us to the assumption that CaCl₂ induces the early solidification of the polymer jet, which can be related to the 3D mesh formation.

The more relevant difference can be seen on the TGA curve of the electrospun fibers with different compositions and the powder mixture of the same compositions (Figure 6C). The results show that the CaCl₂/PSI powder mixtures decompose in two steps around 345 and 410 °C which are related to the decomposition of PSI.^[59] In contrast, the main decomposition step of the electrospun fibers can be found at the lower temperature (345 °C), while the main decomposition peak at 410 °C completely disappears from the DTGA curve (Figure 6C). CaCl₂ can be used to catalyze the pyrolysis of cellulose as was described by Shimada and co-workers.^[60] The mechanism can be explained by the coordination bonds between the polymer C═O groups and the Ca atom,^{[61, 62]} which catalyzes the decomposition of the amide groups or, in our cases, the imide groups as well. A slight decrease can be seen on the TGA curve of the electrospun fibers between 100 and 300 °C while no change can be explored in this temperature region on the TGA curves of the powder mixture. This result indicates that crystalline water bounded stronger in the fiber's structure than in the powder mixtures in which there is no strong chemical interaction between CaCl₂ and PSI.^[60] The XRD results also support the hypothesis of molecular interaction between the crystalline water and PSI since the characteristic Bragg peaks of CaCl₂ almost completely disappeared (Figure 6B). The TGA curves of the fibers prepared on the different humidity levels show that the environmental humidity has no tendentious effect on the water content of the polymer fibers (Figure S9, Supporting Information). We can also see that the residual mass is higher in the case of electrospun fibers than in the case of powder mixtures which can also indicate a strong interaction on the molecular level between the CaCl₂ and PSI.