Electrospun Fabrication of 1–3-Type PVP/SbSI and PVP/SbSeI Nanocomposites with Excellent Piezoelectric Properties for Nanogenerators and Sensors

Abstract

Electrospun one-dimensional nanocomposites composed of polyvinylpyrrolidone (PVP) matrices reinforced with antimony sulphoiodide (SbSI) or antimony selenoiodide (SbSeI) nanowires were fabricated for the first time. Their properties were investigated for applications in piezoelectric sensors and nanogenerators. Precise control of the electrospinning parameters produced nanofibres with diameters comparable to the lateral dimensions of the nanowires, ensuring parallel alignment and a 1–3 composite structure. Structural analysis confirmed uniform nanowire distribution and stoichiometry retention. In both nanocomposites, the alignment of the nanowires enables clear observation of the anisotropy of their piezoelectric properties. PVP/SbSI nanocomposites exhibited a ferroelectric–paraelectric transition near 290 K. Under air-pressure excitation of 17.03 bar, they generated a maximum piezoelectric voltage of 2.09 V, with a sensitivity of 229 mV/bar and a surface power density of 12.0 µW/cm² for sandwich-type samples with nanowires aligned perpendicularly to the electrodes. PVP/SbSeI composites demonstrated stable semiconducting behaviour with a maximum piezoelectric voltage of 1.56 V, sensitivity of 130 mV/bar, and surface power density of 2.3 µW/cm² for the same type of sample and excitation. The high piezoelectric coefficients d₃₃ of 98 pC/N and 64 pC/N for PVP/SbSI and PVP/SbSeI, respectively, combined with mechanical flexibility, confirm the effectiveness of these nanocomposites as a practical solution for mechanical energy harvesting and pressure sensing in nanogenerators and sensors.

1. Introduction

In recent years, there has been a growing demand for the development of novel materials with attractive practical applications. Among these materials, composites, particularly nanocomposites, have garnered significant attention, thanks to advancements in nanotechnology. Nanodimensional materials exhibit distinct properties compared to their bulk counterparts and, due to their minute dimensions, can serve as a functional component within various nanocomposite matrix materials. The reinforcement of composites can be tailored to meet specific mechanical requirements. There is a particular need for lead-free piezoelectric materials that combine high energy-harvesting efficiency with chemical stability and easy integration into flexible devices. SbSI and SbSeI nanostructures, with their high piezoelectric coefficients, quasi-one-dimensional chain-like crystal structures, and robust chemical stability, provide a promising solution. Embedding these nanostructures into polymer matrices enables the fabrication of efficient, flexible, and environmentally stable nanocomposite-based piezoelectric sensors and energy harvesters.

In this study, we embedded piezoelectrically active components, namely antimony sulphoiodide (SbSI) and antimony selenoiodide (SbSeI), into a Polyvinylpyrrolidone (PVP) matrix using the electrospinning method. PVP is a versatile polymer with a wide range of applications, and its flexibility and toughness make it an excellent choice for enhancing the mechanical properties of nanocomposites. It exhibits compatibility with nanofillers, such as nanoparticles [1], nanowires [2], nanotubes [3], and nanosheets [4], ensuring good dispersion and property enhancement. Its high solubility in both water and organic solvents makes it suitable for dispersing and processing nanofillers or nanoparticles. To date, PVP has been used as a reinforcing nanocomposite matrix, with active constituents including ZnO [5], ZnSe [6], SiO₂ [7], Fe₂O₃ [2], and other compounds.

Additionally, PVP boasts excellent stability, crucial for maintaining the structural integrity of nanocomposites under elevated temperatures during processing or use. Its resistance to significant chemical degradation or reactivity with nanofillers is vital to ensure the long-term stability of nanocomposite properties. Depending on the application, PVP’s dielectric properties can be advantageous, particularly in electronics and sensor applications. Additionally, depending on the nanofillers used, PVP can be modified to enhance or control electrical conduction [8], which is crucial for specific applications, such as sensors, electronics, or conductive coatings. Furthermore, PVP’s transparency in its pure form is advantageous for applications requiring optical clarity, such as in nanocomposite films or coatings [5,7].

The PVP matrix was filled with two functional compounds, antimony sulphoiodide (SbSI) [9] and antimony selenoiodide (SbSeI) [10], representing A¹⁵B¹⁶C¹⁷ ternary chalcohalides composed of elements from groups 15, 16, and 17 of the periodic table. These semiconducting materials [11] have gained attention in recent years due to their various properties, including photoconductivity [12], ferroelectricity [13], piezoelectricity [14,15,16], electromechanical behaviour, pyroelectricity [17], pyrooptic, and electrooptic. A¹⁵B¹⁶C¹⁷ ternary chalcohalides are distinguished by their chain-like crystal structures, in which the element (A, group 15) forms covalent bonds with chalcogen (B, group 16) and halogen (C, group 17) atoms, creating quasi-one-dimensional chains along specific crystallographic axes. In our case of SbSI and SbSeI, A corresponds to Sb, B corresponds to S or Se, respectively, and C corresponds to I, with the chains oriented along the c-axis. This arrangement yields highly anisotropic electrical, optical, and mechanical properties, with strong directional bonding within the chains contributing to pronounced ferroelectric and piezoelectric responses. Meanwhile, van der Waals-like interactions between chains confer structural stability and low leachability. The ternary composition allows fine-tuning of electronic bandgaps, dielectric constants, and polarization characteristics [18], enabling these materials to serve as efficient energy harvesters, sensors, and optoelectronic components. Their semiconducting nature, combined with robust environmental stability, makes A¹⁵B¹⁶C¹⁷ chalcohalides a compelling choice for advanced functional nanocomposites, especially when embedded in polymer matrices for device applications. Interest in A¹⁵B¹⁶C¹⁷ ternary chalcohalides dates back to the early 1960s when they were recognized as promising materials for optoelectronic applications. Research on bulk SbSI and SbSeI single crystals continued over subsequent decades [19]. With the advent of nanotechnology, nanostructures, particularly SbSI nanowires, have found applications in gas and humidity sensors [20], as well as in solar cells [21]. In contrast, SbSeI nanowires have potential applications in photovoltaic devices [22], thermoelectric elements [17], and ionizing radiation detectors [23]. However, first of all, due to their exceptionally high piezoelectric and electromechanical coefficients (SbSI single crystals: d₃₃ = 1 nC/N [19] and k₃₃ = 0.9 [24]), these materials are ideal candidates for use in piezoelectric sensors and nanogenerator structures. The piezoelectric effect has practical applications in measuring mechanical properties using its ability to respond to deformation. The measurement of physical quantities holds paramount importance across various scientific disciplines and industrial applications. Recently, SbSI nanowires have been integrated into a tension sensor embedded within a Fibre Reinforced Polymer (FRP) laminate, catering to construction monitoring needs [25]. Moreover, up to now, SbSI nanowires have been applied to create the nanocomposites for energy harvesting by the piezoelectric effect with Epoxy resin [26], Cellulose [27], Polyvinylidene fluoride (PVDF) [28], Poly(methyl methacrylate) (PMMA) [29], Polyacrylonitrile (PAN) [30]. Due to the preparation technique, these composites present the 0–3 type form [31] in which the nanowires are randomly dispersed in the matrix material.

The nanocomposite was fabricated using the electrospinning method. Electrospinning is a remarkably straightforward and adaptable technique that has been effectively employed to create fibres or fibre mats from a diverse range of organic polymers. Its origins trace back to 1934 when the first US patent was developed, although there was little interest in this process until the nineties [32]. In recent years, this versatile and scalable manufacturing technique has been used to produce nanofibres from a wide range of materials, including composites [7], polymers [33], and ceramics [34]. This technique has been widely employed in various fields, such as materials science, nanotechnology, biotechnology, and tissue engineering, as it allows for the creation of nanofibres with high surface area-to-volume ratios and unique properties. With the ability to precisely control the parameters of the electrospinning process, this method enables the production of nanofibres with diameters comparable to the lateral dimensions of nanowires, forming one-dimensional composite nanostructures [35]. To obtain composite nanofibres with precisely defined physical properties, various types of nanostructures can be added to the spinning solution, ensuring their even dispersion within individual fibres [36,37,38,39]. Thanks to the possibility of using any polymer as a matrix and various types of nanofillers, composite nanofibres produced by the electrospinning method offer a wide range of applications. Composite nanofibres are already used in the production of, among others, supercapacitors [40], lithium-ion batteries [41], and photovoltaic cells [42].

This article presents, for the first time, the synthesis, properties, and application of one-dimensional PVP/SbSI and PVP/SbSeI nanocomposites in piezoelectric devices, designed to convert mechanical energy into electricity. An advantageous feature of these nanocomposites lies in the electric field generated during the electrospinning process, facilitating the creation of a 1–3 type [43] nanocomposite efficiently and straightforwardly. This composite configuration aligns nanowires parallel to each other throughout the volume, endowing it with superior properties compared to previously studied, less ordered 0–3-type composites.

2. Materials Preparation and Their Properties

2.1. Nanowires Preparation

The SbSI and SbSeI nanowires were synthesized using sonochemical methods developed by Nowak et al., as detailed in [9] and [10], respectively. In the standard protocol, a stoichiometric mixture of elemental constituents (from groups 15, 16, and 17) was dispersed in ethyl alcohol and enclosed in a sealed container. The container was immersed in water (Figure 1a) within the cup-horn of a 750 W ultrasonic processor VCX-750, equipped with a sealed converter VC-334 (provided by Sonics & Materials, Inc., Newtown, CT, USA). The ultrasonic frequency employed was 20 kHz, and the manufacturer ensured a power density of 565 W/cm². To maintain a stable temperature of 313 K, a refrigerated circulating bath AD07R (PolyScience, Niles, IL, USA) was used. The sonochemical reaction continued until nanowire gelation took place (Figure 1b). Consequently, ultrasonic irradiation was applied for two hours. The resulting ethanogel underwent rinsing and drying under a pressure of 60 Pa and a temperature of 353 K for 2 days, yielding a xerogel of A¹⁵B¹⁶C¹⁷ nanowires.

2.2. Characterization of Synthesized Nanowires

The morphology and chemical composition of the prepared xerogels were examined using a scanning electron microscope (SEM), specifically the Phenom PRO X (model by Thermo Fisher Scientific, Waltham, MA, USA), equipped with an energy-dispersive X-ray spectroscopy (EDS) detector. SEM micrographs and EDS spectra for SbSI and SbSeI nanowires are depicted in Figure 1c and Figure 1d, respectively. In Figure 1c, the EDS spectrum of the SbSI nanowires reveals discernible peaks corresponding to antimony, sulphur, and iodide. Furthermore, it confirms an elemental atomic ratio of 0.35:0.31:0.34 for Sb, S, and I, respectively, indicative of a stoichiometric SbSI composition, considering experimental error. In Figure 1d, the EDS spectrum of the SbSeI nanowires displays evident peaks associated with antimony, selenium, and iodide. Additionally, it validates an elemental atomic ratio of 0.34:0.33:0.33 for Sb, Se, and I, respectively, suggesting a stoichiometric SbSeI composition, within uncertainties. For a comprehensive understanding of the procedure and properties of SbSI and SbSeI nanowires, refer to [9] and [10], respectively. Following synthesis, SbSI and SbSeI compounds were further processed to prepare nanocomposites.

2.3. Nanocomposites Preparation

The new PVP/SbSI and PVP/SbSeI nanocomposites were prepared using the electrospinning method. For the spinning solutions, the following reagents were employed: poly(vinylpyrrolidone) (PVP, purity 99%, Mw = 1,300,000 g/mol, Sigma-Aldrich, Burlington, VT, USA), ethanol (EtOH, purity 99.8%, Avantor Performance Materials, Gliwice, Poland), and the previously synthesized SbSI and SbSeI nanowires. A 15% (w/w) PVP solution in EtOH was used for the synthesis of nanofibres. The reinforcing phase comprised sequentially generated SbSI and SbSeI nanowires, resulting in EtOH/PVP/nanowires (SbSI or SbSeI) with a 15% weight concentration. This included a 70% mass concentration of nanowires relative to the polymer mass. To break agglomerates in the reinforcing phase, we added the measured mass of nanowires (3.22 g) to 10.77 mL of EtOH, subjecting these solutions to one hour of sonification with continuous solvent replenishment. Immediately after sonification, a measured PVP amount (1.5 g) was added, and the solutions were stirred magnetically for 24 h at room temperature. Post-stirring, the solution was introduced into the device’s pump using a sterile syringe.

Composite nanofibres were formed using an electrospinning process from a solution using the FLOW—Nanotechnology Solutions Electrospinner 2.2.0-500 device (YFlow, Malaga, Spain), equipped with a drum collector (Figure 2a). The process employed constant electrospinning parameters, with a solution flow rate of 3.5 mL/h, a nozzle-to-collector distance of 10 cm, and a potential difference of 20 kV. The drum collector’s rotational speed was 800 RPMs. These endeavours were intended to cause at least partial ordering of the electroactive nanofibres.

Figure 2 presents the fabricated PVP/SbSI fibrous mat and close-up images of PVP/SbSI and PVP/SbSeI fibrous mats. The photograph depicts the three-dimensional composite nanomaterial represented by the PVP/SbSI fibrous mat (Figure 2b). This mat was obtained through electrospinning, utilizing a drum collector. Composite nanofibres are embedded in a poly(vinylpyrrolidone) (PVP) matrix and dispersed uniformly. As expected, they are oriented lengthwise, containing piezoelectric antimony sulphoiodide at a concentration of 70 wt% relative to the polymer mass. Figure 2d presents an SEM image illustrating the cross-section of the PVP/SbSI fibrous mat. Analysis of the obtained nanomaterials reveals that, depending on the electrospinning process’s duration and chosen parameters, effective three-dimensional, elastic piezoelectric nanomaterials can be produced in the form of PVP/SbSI or PVP/SbSeI composite fibrous mats. These mats exhibit thicknesses ranging from approximately 10 to 1000 µm, widths between 8 and 26 cm, and lengths extending up to 80 cm.

2.4. Analysis of Morphology and Structure of the Fabricated Nanofibres

The resulting nanofibres were subjected to quantitative and qualitative analysis through X-ray EDS microanalysis and surface topography imaging. This analysis was conducted using the Zeiss Supra 35 scanning electron microscope (Carl Zeiss AG, Oberkochen, Germany) equipped with the Trident XM4 X-ray spectrometer (Trident, Castenedolo, Italy), featuring a backscattered electrons detector (QBSD) supplied by Energy Dispersive X-ray Analysis (EDAX). To examine the morphology and structure of the obtained PVP composite nanofibres, which included SbSI and SbSeI nanowires, surface topography imaging at a magnification of 10,000 times was employed. This was achieved using the scanning electron microscope with a backscattered electrons detector (Figure 3a,b). The utilization of the QBSD enabled the observation of individual nanowires embedded within the polymer nanofibres (Figure 3b,f).

The analysis of the morphology and structure of PVP nanofibres, obtained with a 15% solution of poly(vinylpyrrolidone) in ethanol, containing 70 wt% of SbSeI or SbSI nanowires, indicated that these nanofibres are free of structural defects and are characterized by a uniform diameter value along their entire length. Moreover, the nanowires are uniformly dispersed throughout the whole volume of the nanofibres; they do not agglomerate and are arranged in parallel to each other (Figure 3b,f). The use of the drum collector during the electrospinning process allowed for orienting the majority of nanofibres in a chosen direction, a key factor when using this type of material for the manufacturing of nanogenerators.

One hundred measurements of the diameters of the obtained PVP/SbSI and PVP/SbSeI composite nanofibres indicated that the measured diameters varied in the range of 20–450 nm and 100–550 nm, respectively. In the case of PVP/SbSI composite nanofibres, the average diameter values were 240(66) nm, with the most prevalent diameters falling in the range of 150–200 nm, representing 28% of all measured values (Figure 3c). For PVP/SbSeI composite nanofibres, the most common diameter values were in the range of 300–350 nm, accounting for 24% of all measured diameter values for this sample (Figure 3g). The average diameter values of the analysed PVP/SbSeI nanofibres were 333(67) nm. It is noticeable that the average diameter of nanofibres is about 4 times higher than the average nanowire diameter and lower than their average length (see, e.g., SbSeI nanowires dimensions distribution in [16]). This guarantees that the c-axis of nanowires is parallel to the nanofibres axis. The illustrated structure of PVP nanofibres filled with nanowires, maintaining scale, is presented in Figure 3i. It clearly demonstrates that the nanowires are surrounded by a polymer nanofibre.

EDS spectra of fabricated PVP/SbSI and PVP/SbSeI nanofibres are presented in Figure 3d and Figure 3h, respectively. The derived sample compositions are presented in

Table 1. Weight and atomic composition of PVP/SbSI and PVP/SbSeI nanofibre mats.

PVP/SbSI Nanofibres			PVP/SbSeI Nanofibres
Element-Line	Wt%	At%	Element-Line	Wt%	At%
C-K	72.6	86.85	C-K	61.86	82.81
O-K	3.28	2.95	O-K	2.98	2.99
Al-K	17.8	9.48	Al-K	17.77	11.84
Au-M	3.19	0.23	Au-M	2.75	0.22
S-K	0.39	0.17	Se-L	3.66	0.74
Sb-L	1.23	0.15	Sb-L	5.36	0.71
I-L	1.51	0.17	I-L	5.62	0.69

.

Carbon and oxygen originate from the PVP matrix. The Al peak results from the foil onto which the fibres are deposited, while the Au peak arises from the conductive layer sputter-coated onto the sample for imaging purposes. It is evident that the stoichiometric composition of antimony, iodine, and sulphur in the case of PVP/SbSI, or selenium in the case of PVP/SbSeI samples, persists, indicating that the composition of nanowires remains stable throughout the electrospinning process.

2.5. Sample Preparation

The prepared nanocomposites were utilized to create both planar-type and sandwich-type samples. Fibrous mats with a thickness of 0.1 mm were used to prepare each sample. To assess the piezoelectric properties, the samples were constructed with nanofibres and nanowires aligned perpendicular to the electrodes, exploiting the anisotropy of the piezoelectric coefficient in SbSI and SbSeI. The A¹⁵B¹⁶C¹⁷ compounds exhibited a substantial longitudinal d₃₃ coefficient, representing the longitudinal component of the piezoelectric tensor, with smaller values for its transverse components, d₃₁ and d₃₂ [44,45,46]. A sheet of nanocomposite was cut and placed in a pumped coating system (Quorum Q150T ES) to fabricate planar-type samples. Gold electrodes, 150 nm thick, were sputtered on one side of PVP/SbSI and PVP/SbSeI nanocomposites, maintaining a 2 mm distance between them. The electrodes were sputtered perpendicular to the nanofibres and, consequently, to the c-axis of the nanowires. Copper wires were then affixed to the electrodes using 05002-AB silver paste (SPI Supplies, West Chester, PA, USA) to obtain reliable electrical contact. Photographs of the planar-type samples with PVP/SbSeI nanofibres and PVP/SbSI nanofibres parallel to electrodes are presented in Figure 4a and Figure 4c, respectively. This type of sample was also utilized in optical and AC measurements. Furthermore, planar-type samples were fabricated with nanofibres aligned parallel to the electrodes in a similar manner to enable a direct comparison of piezoelectric responses. These samples shared identical dimensions with those featuring nanofibres oriented perpendicularly to the electrodes.

To construct sandwich-type samples, sheets of PVP/SbSI and PVP/SbSeI nanocomposites were rolled cylindrically and cut to achieve a thin sandwich-type sample with nanofibres oriented perpendicular to the electrodes. The same spotter coater was used to deposit a 150 nm thickness of gold electrodes. Electrical contacts were attached similarly to those of planar-type samples. Photographs of the sandwich-type samples are shown in Figure 4b and Figure 4d, respectively. This type of sample was employed in DC measurements due to the current flowing throughout the entire sample volume, as well as in piezoelectric measurements.

Stemi 2000-C microscope (Carl Zeiss AG, Oberkochen, Germany), equipped with an Olympus DP25 camera (Olympus, Tokyo, Japan), was used to acquire photographs of nanocomposites. The sample dimensions were estimated based on acquired images by using the ImageJ software (National Institutes of Health, Rockville Pike, MD, USA). The crucial sample dimensions are presented in

Table 2. Dimensions of the examined samples.

Sample Type	Thickness; d [mm]	The Area Between Electrodes; A [mm2]
PVP/SbSI sandwich	0.5	5.1
PVP/SbSI planar	0.1	1.9 × 18.3
PVP/SbSeI sandwich	0.5	5.1
PVP/SbSeI planar	0.1	1.9 × 19.2

.

3. Experimental Methods and Measurement Setups

3.1. Diffuse Reflectance Spectroscopy

Spectral characteristics of PVP/SbSI and PVP/SbSeI nanocomposites were captured through diffuse reflectance spectroscopy (DRS). It was composed of the Ulbricht sphere ISP-REF (Ocean Optics Inc., Orlando, FL, USA). The Ulbricht sphere was connected via an optical fibre to the HR-4000 spectrophotometric card (Ocean Optics Inc., Orlando, FL, USA), facilitating spectrum recording across the 200 nm to 1100 nm wavelength range. Before the actual measurements, calibration of the measurement system was applied by measuring the background level and the intensity of light reflected from the WS1 standard (Ocean Optics Inc., Orlando, FL, USA).

3.2. DC Conduction Measurements

The DC measurements of PVP/SbSI and PVP/SbSeI nanocomposites were carried out using a Keithley 6517B electrometer (Tektronix, Beaveron, OR, USA) which functions as both a voltage source and an ammeter. The internal connection of these components minimizes the need for extra wiring. The temperature dependences of conductivity were investigated on samples placed in the environmental test chamber SH-242 (Espec, Osaka, Japan). Measurements were taken in the range from 278 K to 325 K with a slow heating rate of 0.5 K/min. The temperature was monitored using a thermocouple positioned adjacent to the sample and connected to a Metex-M 4660A (Metex, Shenzhen, China). The measurements were controlled and recorded through a LabVIEW program installed on a PC.

3.3. Impedance Spectroscopy

The prepared samples were securely positioned in a rigid sample holder, which was then installed in a high-efficiency Janis cryostat STVP-200-XG (Janis Research Company, Woburn, MA, USA). This system utilizes static helium exchange gas for cooling or warming the sample within the operational temperature range of 80–500 K. Consequently, the sample attained mechanical freedom and established a robust electrical contact. The impedance magnitude |Z| and phase shift φ were measured using a Solartron 1260 impedance analyser (Ametek Scientific Instruments, Leicester, UK) equipped with a 1296 dielectric interface. Measurements were carried out across frequencies 10⁻³ ≤ f ≤ 10⁶ Hz at a temperature of 270 K. The cryostat’s temperature was closely controlled within ±0.01 K by a Lake Shore 335 temperature controller (Lake Shore Cryotronics, Inc., Westerville, OH, USA). The experimental procedure was overseen using a SMaRT Solartron software (Ametek Scientific Instruments, Leicester, UK).

3.4. Piezoelectric Measurements

The piezoelectric measurements were performed to estimate the piezoelectric coefficient and to assess the possibility of practical use of the material. The measurements of the piezoelectric coefficient were performed based on the recently published direct method of piezoelectric coefficient measurement [47]. The samples were positioned on a hard surface, and the pneumatic air gun Zoraki HP-01-2 (Atak Arms Industry Co. Ltd., Istanbul, Turkey) was mounted on a vertical lever above. This air gun is equipped with a compressed air container, allowing for the compression of varying air amounts based on the number of pumps, thereby altering the airflow pressure. The sample was connected to the input channel of a Keysight DSOX 3104T oscilloscope (Keysight Technologies, Inc., Santa Rosa, CA, USA), which has an input impedance of 1 MΩ and a capacitance of 14 pF. The sampling frequency was 200 kS/s, ensuring a reliable measurement of the sample’s output voltage. A WIKA S-10 pressure sensor (WIKA Alexander Wiegand SE & Co. KG, Klingenberg am Main, Germany), replacing the sample, was employed to measure the differential pressure of a shock wave in a range of 0–100 bars, with a sensitivity of 0.1 V/bar.

4. Results

4.1. Light Absorption Properties and Kubelka-Munk Analysis

Diffuse reflectance spectroscopy was carried out at a temperature of 293 K on the prepared nanocomposite samples without electrodes. The registered diffuse reflectance spectra are presented in Figure 5.

The absorption edge for the PVP composite with SbSeI is observed to shift towards longer wavelengths compared to the PVP composite with SbSI. This shift is ascribed to the broader band gap of SbSI relative to the SbSeI compound. Generally, the band gap exhibits a monotonic increase with higher sulphur molar composition in the four-component SbS_1−xSe_xI compounds [18].

The Kubelka-Munk theory was employed to determine the energy gap of PVP filled with SbSI and SbSeI nanowires. Kubelka-Munk functions (Figure 5b), directly proportional to the light absorption coefficient, were calculated based on the measured spectra of the diffuse reflectance coefficient R_d(λ) using the relationship [48]:

$$F_{K-M}(R_d)={\displaystyle \frac{{\left(1-R_d\right)}^2}{2R_d}}={\displaystyle \frac{\alpha }{S}},$$

(1)

where α and S represent the absorption and scattering coefficients, respectively. It is noteworthy that the light scattering coefficient S remains constant for samples with thickness exceeding particle dimensions and is approximately equal to unity for nanoparticles [48].

For SbSI and SbSeI nanowires, the absorption mechanisms were identified as a sum of an additive constant α₀, absorption in the energy gap for forbidden quantum transitions without the involvement of excitons α₁, and Urbach absorption α₂ [9,10,48]. The same absorption mechanisms were fitted in the case of PVP nanocomposites:

$$\begin{array}{l}{\alpha }_0=A_0,\\ {\alpha }_1=\left\{\begin{array}{l}0, \mathrm{for} h\nu <E_{gIf}\\ A_1{\left(h\nu -E_{gIf}\right)}^3, \mathrm{for} h\nu \ge E_{gIf}\end{array}\right.,\\ {\alpha }_2=A_2\exp \left(h\nu /E_U\right),\end{array}$$

(2)

where A₀—an additive constant independent of the wavelength, h—Planck’s constant, ν—frequency of the light wave, E_gIf—width of the indirect forbidden energy bandgap, A₁, A₂—proportionality factors for bandgap absorption and Urbach absorption, respectively, and E_U—Urbach energy.

The absorption mechanisms were fitted for wavelengths above the inflection points on absorption edges of the Kubelka-Munk function, i.e., in the ranges of 570–950 nm and 650–950 nm for PVP/SbSI and PVP/SbSeI nanocomposites, respectively. It was observed that the most optimal fit, characterized by the minimal value of the determination coefficient, was attained when considering the combination of energy gap absorption and constant absorption for the PVP/SbSI nanocomposite:

$${\alpha }_{\mathrm{SbSI}/\mathrm{PVP}}={\alpha }_0+{\alpha }_1.$$

(3)

Similarly, for the PVP/SbSeI nanocomposite, the best fit was achieved by combining energy gap absorption, Urbach absorption, and constant absorption:

$${\alpha }_{\mathrm{SbSeI}/\mathrm{PVP}}={\alpha }_0+{\alpha }_1+{\alpha }_2$$

(4)

The fitted dependences are shown in Figure 5b. The fitting parameters are presented in

Table 3. Best fitted absorption parameters for PVP/SbSI and PVP/SbSeI nanocomposites. A₀—an additive constant, E_gIf—width of the indirect forbidden energy bandgap, A₁, A₂—proportionality factors for bandgap absorption and Urbach absorption, respectively, E_U—Urbach energy.

	PVP/SbSI	PVP/SbSeI
A0 [m−1]	0.5382(42)·10−2	0.5398(80)·10−2
A1 [m−1eV−3]	77.4(43)	34.09(57)
EgIf [eV]	1.9091(20)	1.6142(13)
A2 [m−1]	-	0.34(12)·10−8
EU [eV]	-	0.1067(25)

.

The fitting parameters can be compared to the absorption parameters of SbSI [9,48] and SbSeI [10] nanowires. The constant absorption, independent of the wavelength, increases for both nanocomposites compared to pure materials. The observed phenomenon can be explained by the presence of the PVP matrix, which enhances the scattering factor and demonstrates consistent, low-level absorption across the examined wavelength range [49]. Moreover, as an insulator, PVP exhibits a considerably higher bandgap exceeding 3.5 eV [50], making it invisible within the examined wavelength range. The estimated energy gaps for the fillers in both nanocomposites are in good agreement with previously reported data for both chalcohalides. The slight shift in the energy gap for SbSI, approximately 70 meV towards higher energy, may be attributed to the slightly lower measurement temperature compared to [39], where room temperature is reported. The overall Urbach absorption level in the PVP/SbSeI nanocomposite is comparable to the data reported in [10]. Surprisingly, the Urbach absorption in the PVP/SbSI nanocomposite is non-existent. This absence results from a lower number of localized states in the energy gap due to greater order and fewer band distortions in the material structure, confirming that the prepared SbSI is of outstanding quality.

4.2. Temperature Dependence of DC Conductivity

The DC measurements of temperature-dependent conductivity for PVP/SbSI and PVP/SbSeI nanocomposites were carried out from 278 K to 325 K with a slow heating rate of 0.5 K/min. The temperature was measured by a thermocouple placed next to the sample. The slow heating rate and use of a thermocouple allowed for precise temperature measurement. The DC current was measured under an applied voltage of 1 V, which is equivalent to an electric field strength E = 100 V/cm. The registered current characteristics are presented in Figure 6a.

One observes a monotonic increase in the current flowing through the PVP/SbSeI nanocomposite as the temperature is raised. This is a characteristic behaviour of semiconducting materials. In the case of the SbSI-based nanocomposite, two distinct regions are revealed. For temperatures lower than approximately 290 K, the current exhibits a steeper increase than for temperatures above 290 K (see Figure 6a). Moreover, the visible peak indicates the Curie point (T_C), signifying the transition of the material from a ferroelectric state below that temperature to a paraelectric state above this temperature. It is worth noting that the current values are slightly higher in the ferroelectric phase of the PVP/SbSI composite than in its paraelectric phase or in the case of the PVP/SbSeI nanocomposite, as evident from the distortion of the current data points in this temperature range. It is not caused by a characteristic of the measurement method but by additional carriers released from boundaries between ferroelectric domains [51]. The ferroelectric domains disappear at the Curie temperature, resulting in the observed current peak at this temperature (Figure 6a).

In order to extract the activation energy of conductivity (σ), the following formula has been applied:

$$\sigma ={\displaystyle \frac{Id}{UA}},$$

(5)

where I is the measured current, U is the applied voltage of 1 V, d is the sample thickness of 0.5 mm, and A is the electrode surface of 5.1 mm². Considering the exponential relationship of semiconductor-like conductivity vs. temperature, the activation energy (E_a) can be determined using the well-known Arrhenius equation:

$$\sigma ={\sigma }_0\exp \left(-{\displaystyle \frac{E_a}{kT}}\right),$$

(6)

where σ₀ is a pre-exponential factor, k is the Boltzmann constant, and T is the temperature in Kelvin, since σ has already been defined. Taking the logarithm of the above equation, a straight line can be fitted with a slope equal to the activation energy divided by the Boltzmann constant (Figure 6b):

$$\ln \sigma =-{\displaystyle \frac{E_a}{k}}{\displaystyle \frac{1}{T}}+\ln {\sigma }_0.$$

(7)

This linear relationship provides a convenient method for extracting the activation energy from experimental conductivity data, offering valuable insights into the semiconductor behaviour under varying temperature conditions. The best-fitting parameters and the extracted activation energies are presented in

Table 4. Best-fitting parameters for Equation (7) in PVP/SbSI and PVP/SbSeI nanocomposites. For the SbSI-based nanocomposite, the fitting is performed separately for temperatures below and above the Curie temperature (T_C). a—slope, b—intercept, and E_a—activation energy for the conductivity.

	PVP/SbSI		PVP/SbSeI
	T < TC	T > TC
a [K]	−6845(34)	−528.4(16)	−366.5(15)
b	7.51(12)	−14.3112(51)	−16.4360(50)
Ea [meV]	589.9(29)	45.53(14)	31.58(13)

.

From Figure 6b, one may easily read that the conductivity of PVP/SbSI and PVP/SbSeI at room temperature equals to 1·10⁻⁷ S/m and 2·10⁻⁸ S/m, respectively. The conductivity of SbSI and SbSeI semiconducting nanowires is equal to 1.1·10⁻⁵ S/m (determined from the data provided in [52]) and 2.5·10⁻⁷ S/m [16], respectively. Mere PVP exhibits a conductivity ranging from 2.32·10⁻⁷ [53] up to 1·10⁻⁶ S/m [54]. When comparing the conductivity of separated constituents with the entire nanocomposite, one may conclude that the nanocomposite conductivity is primarily dominated by the high resistivity of the polymeric composite matrix. This influence is also evident in the temperature dependences of conductivity. The slope of σ(T) differs for the nanocomposite compared to detached semiconductors (confront Figure 6b with dependences in [16,52]).

For the PVP/SbSeI and paraelectric phase of PVP/SbSI nanocomposite, the σ(T) slope is smaller for the nanocomposite in comparison to separated semiconducting nanowires. Only in the case of the ferroelectric phase of PVP/SbSI nanocomposite, the σ(T) slope is like the typical characteristic of pure composite SbSI reinforcement nanowires. This dependence may result from the influence of ferroelectric domains and charges accumulated on their boundaries. In the absence of ferroelectric domains, the semiconducting-ferroelectric behaviour gives way to the insulating properties of the PVP matrix. This is reflected in the activation energy values of the nanocomposite (Table 4).

For the ferroelectric phase, the conductivity activation energy of the PVP/SbSI nanocomposite is comparable to the value of detached SbSI nanowires [52]. For the paraelectric phase of PVP/SbSI and PVP/SbSeI nanocomposites, the activation energy of conductivity is much smaller than for the individual materials. In this case, it ranges within the order of dozens of meV (Table 4), whereas for separate materials, it remains within the scope of 0.1–0.25 eV [52] for SbSI and around 0.6 eV [16] for SbSeI nanowires, respectively. In the case of the PVP composite matrix, the activation energy for conductivity is 0.85 eV [53]. In a previous study [53], the authors observed a decrease in activation energy with the addition of filler in their PVP-based nanocomposite. They proposed that the incorporated component may lead to the formation of charge complexes within the host lattice, resulting in a reduction in activation energy. The same explanation may be applied here. In the case of ferroelectric SbSI, the influence of these charge complexes may be obscured by numerous charges trapped on ferroelectric domain boundaries; hence, the activation energy is comparable to pure ferroelectric SbSI nanowires. In the paraelectric phase and in the non-ferroelectric SbSeI composite, the formed charge complexes play a crucial role in electrical conduction.

4.3. Impedance Spectroscopy Results and Equivalent Circuits

Impedance spectroscopy was used to comprehensively characterize electrical properties over a wide frequency range of the driving electric field. The output impedance magnitude |Z| and phase shift φ data for PVP/SbSeI and PVP/SbSI nanocomposites were then rescaled to obtain the real (Z′) and imaginary (Z″) parts of the complex impedance and real (C′) and imaginary (C″) parts of the complex capacitance. The frequency dependences of these components are presented in Figure 7.

One can easily notice in Figure 7a that, at low frequencies, the real part of the impedance of PVP/SbSeI is significantly higher than that of the PVP/SbSI nanocomposite by approximately two orders of magnitude, and aligns with constant conductivity. As expected, the real part of the impedance decreases as the frequency increases, converging to nearly the same values at 1 MHz for both composites. One can anticipate that the electric properties are primarily dominated by the filler (composite reinforcement) constituent at lower frequencies. In contrast, the PVP composite matrix at higher frequencies mainly determines the dielectric response.

In Figure 7b, the imaginary part of impedance exhibits one peak for the PVP/SbSeI composite and two for PVP/SbSI. The maximum for PVP/SbSeI nanocomposite takes place at a frequency of 1 Hz, while for PVP/SbSI, the peaks are observed at 1.5 kHz and 20 mHz.

Figure 7c,d present the frequency dependences of real and imaginary parts of complex capacitance C* = C′ − iC″. Here, due to the planar configuration of electrodes on the investigated sample, we decided not to rescale the capacitance data into permittivity. The value of C’ decreases as the frequency increases and becomes almost constant from 0.1 Hz and 100 Hz for PVP/SbSeI and PVP/SbSI, respectively. An enhancement in C’ values with frequency decrease has already been observed in the composite dielectric materials [55]. Here, we dispute the opinion sometimes found in the literature (see, e.g., [56]) that the low-frequency behaviour of such systems may originate from the ac electric conduction caused by doping PVP with semiconducting SbSI and SbSeI. The point is that other authors (see, e.g., [57]) and our own verifying experiments do not support the above statement. Namely, the electric conduction alone does not provoke any increase in the real part of the complex capacitance/permittivity. To cause an increase in the real part of the capacitance, in addition to the electric current, an additional process taking place in the explored system is necessary. The results of a detailed analysis of the behaviour of the investigated systems in the low-frequency range of the probing electric fields will be published elsewhere.

Figure 7e presents the frequency dependences of the phase shift φ data obtained in the experiment. Their values vary from 0° at lower frequencies to −90° at higher frequencies. This means that the electrical properties of the investigated nanocomposites related to electrical conductivity dominate at low frequencies, while dielectric properties dominate at high frequencies.

The display of impedance data in the form of a Nyquist plot Z′ (Z″) allows the sample to be simulated as an equivalent circuit in which various elements represent the appropriate phenomena occurring in the sample. Such plots generated for PVP/SbSeI and PVP/SbSI nanocomposites are presented in Figure 8. In the high-frequency range, we observe a semicircle for both samples. In the case of SbSeI, a long tail relates to decreasing impedance, visible also in Figure 7a. In the case of the PVP/SbSI nanocomposite (Figure 8b), a distinctive loop-like shape is observed. To the best of our knowledge and supported by an extensive literature survey, such behaviour has not been documented in available literature, prompting more comprehensive investigations and analysis. Furthermore, it gives rise to negative resistances in the equivalent circuit. Consequently, to evaluate the equivalent circuit, the theoretical dependence has been fitted up to 100 mHz and 50 Hz for PVP/SbSeI and PVP/SbSI nanocomposites, respectively.

The experimental data presented in Nyquist plots were fitted using an equivalent circuit proposed for different models. The fitting was accomplished with the LEVM program developed by Macdonald [58]. The most efficient result was achieved with a simple parallel RC model (inset in Figure 8a) in the case of PVP/SbSeI. The optimally fitted parameters for the equivalent circuits (as shown in the inset of Figure 8a) are R = 1.69(13)·10¹¹ Ω, C = 6.85(32)·10⁻¹³ F. In the case of the PVP/SbSI nanocomposite, the capacitance was replaced by a constant phase element (CPE). The impedance of the CPE is represented by the equation:

$$Z_{CPE}={\displaystyle \frac{1}{A_{CPE}{(i\omega )}^{n_{CPE}}}},$$

(8)

where i is an imaginary number, ω is an angular frequency, A_CPE is a frequency-independent factor, and n_CPE is a power with values ranging from zero to unity. A unity value corresponds to a pure capacitor. The CPE can generally be associated with inhomogeneous behaviour [59]. The optimally fitted parameters are R= 1.4299(79)·10⁸ Ω, A_CPE = 9.672(65)·10⁻¹³ Fsⁿ⁻¹, n_CPE = 0.98725(20). A clear single semicircle indicates that only one primary mechanism exists for the electrical conduction within the samples. Exploring the temperature dependencies of the obtained parameters is crucial for confirming their origin. These studies should also ideally extend to a broader frequency range. These investigations will be performed and reported shortly.

4.4. Piezoelectric Properties

In the aligned PVP/SbSI and PVP/SbSeI nanofibrous composites, the mechanism of piezoelectric output generation can be explained by considering the oriented one-dimensional SbSI or SbSeI nanowires as the primary active phase. Electrospinning in a strong electric field aligns the nanowires along the PVP fibre axis so that their crystallographic c-axis runs parallel to the fibre direction, while simultaneously inducing nanowire pooling under the influence of the electric field. This orientation is crucial because the polar c-axis corresponds to the direction of maximum spontaneous polarization and strongest piezoelectric response, with the d₃₃ coefficient significantly exceeding d₃₁ and d₃₂ in these compounds [60].

When the composite is mechanically deformed, the aligned nanowires experience strain that generates a piezopotential along the c-axis. Each nanowire behaves as an electric dipole, developing opposite bound charges at its ends and creating an internal electric field that drives charge carriers toward the electrodes. Because the nanowires are preferentially oriented, the individual dipole contributions add constructively, producing a higher output than in randomly dispersed systems.

The direction of applied stress relative to the fibre axis determines which piezoelectric coefficient dominates. In the sandwich-electrode configuration, stress is applied parallel to the c-axis, and voltage is measured along the same direction (Figure 9a). Axial compression or elongation produces a strong piezopotential corresponding to the longitudinal d₃₃ mode, the largest coefficient in these chalcohalides.

In the planar-electrode configuration, stress is applied perpendicular to the c-axis while the voltage is still collected along it, so the output arises mainly from the transverse d₃₁ and d₃₂ coefficients (Figure 9b). Although smaller than d₃₃, these coefficients still provide a measurable signal through polarization changes under lateral strain.

In addition to the previously discussed configurations, the piezoelectric response can also be measured along the fibre axis in a planar geometry, with electrodes positioned parallel to the aligned nanofibres. In this setup, the external stress is applied perpendicular to the fibre axis, and the induced voltage is collected along the nanofibres direction (Figure 9c). The resulting piezopotential corresponds to the in-plane longitudinal coefficients d₁₁ and d₁₂. Although the magnitude of d₁₁/d₁₂ is typically lower than d₃₃, this measurement provides additional insight into the anisotropic piezoelectric behaviour of the aligned nanowires. It complements the information obtained from the sandwich (d₃₃) and transverse planar (d₃₁/d₃₂) configurations.

The PVP matrix provides mechanical flexibility, efficiently transfers external stress to the nanowires, and serves as an insulating medium that limits charge leakage. Gold electrodes collect the induced charges, and the combined effects of nanowire alignment, strong c-axis polarization, and effective stress transfer from the PVP matrix underpin the enhanced energy-harvesting performance of these aligned chalcohalide–polymer composites.

4.4.1. Calculation of d₃₃ and d₃₁/d₃₂ Piezoelectric Coefficients

The SbSI and SbSeI, serving as the fillers in the nanocomposites, exhibit piezoelectric properties. Consequently, a recently published method [47] for estimating the piezoelectric coefficient has been employed. The piezoelectric response is generated by the material placed between electrodes. The piezoelectric coefficient, using Voigt notation, can be designated as [47]:

$$d_{ij}={\displaystyle \frac{U_{\max i}}{R_V\cdot A}}{\left({\displaystyle \frac{d{p}_j}{dt}}\right)}^{-1},$$

(9)

where U_max is the maximum voltage measured at a peak, R_V is the load resistance equal to the oscilloscope’s internal resistance, A is the active surface of the sample, and dp/dt is the speed of pressure changes. The indices i and j represent the directions of voltage measurement and the applied pressure, respectively, with respect to the crystallographic axes of the nanowires. The speed of pressure changes is equal to 23.61 bar/ms, 39.57 bar/ms, and 56.18 bar/ms for air pressures of 4.86 bar, 11.54 bar, and 17.03 bar, respectively [47]. The active sample area refers to the shared region where the sample interacts with the air stream. Figure 10 illustrates the time dependence of piezoelectric voltage responses measured on the nanocomposite samples with electrodes perpendicular to aligned nanofibres at a temperature of 298 K, thus just above the Curie point of the PVP/SbSI composite.

The piezoelectric coefficient indicated by Equation (9) represents the intrinsic properties of the sample and remains independent of air pressure. Therefore, based on three measurements, average values and their uncertainties can be used to estimate the piezoelectric coefficient for each sample. The complete set of values for the thus obtained piezoelectric coefficients d_ij is presented in

Table 5. Piezoelectric properties of the investigated nanocomposites in planar and sandwich configurations. From left to right, the columns indicate: nanocomposite type and contact configuration, maximum piezoelectric voltage (U_max), piezoelectric response duration (t₁), sample effective area (A), piezoelectric coefficient (d_ij), sensitivity (η), generated energy (E), and power surface density (P_S) for examined samples. In the absence of provided uncertainties, its values are negligibly small. For a detailed explanation, go into the body text.

Sample Type	Umax [V]	t1 [ms]	A [mm2]	dij [pC/N]	η [mV/bar]	E [nJ]	PS [µW/cm2]
PVP/SbSeI planar	0.526	2.5	5.34	d31 = 35.0(45)	84(22)	0.065	0.49
	0.806					0.115	0.86
	1.000					0.170	1.28
PVP/SbSeI sandwich	0.910	4.0	5.10	d33 = 64.4(73)	130(35)	0.289	1.42
	1.272					0.378	1.85
	1.564					0.468	2.29
PVP/SbSI planar	0.495	4.3	5.34	d31 = 37.2(34)	73(17)	0.155	0.67
	0.687					0.276	1.20
	1.116					1.080	4.71
PVP/SbSI sandwich	1.564	5.7	5.10	d33 = 98(20)	202(73)	1.465	5.04
	1.868					3.468	11.93
	2.092					3.499	12.04

. It is crucial to emphasize that in the case of sandwich samples, the pressure acts along the c-axis of nanocrystalline structures. Simultaneously, the voltage is measured in the same direction, allowing the estimation of the longitudinal d₃₃ component of the piezoelectric tensor. On the other hand, in the case of planar samples, the pressure is applied perpendicularly to the c-axis of the crystals, while the voltage is measured along this axis. In such instances, the resultant values of the transverse d₃₁ and d₃₂ components are obtained. For SbSI and, more generally, in A¹⁵B¹⁶C¹⁷ compounds, the d₃₃ component is approximately 10 times higher than d₃₁ and d₃₂, where the latter two exhibit relatively comparable values [45].

The potential application of these nanocomposites as pressure sensors can be evaluated by estimating sensitivity, which is calculated as the generated voltage (U_max) under impact pressure (p):

$$\eta ={\displaystyle \frac{U_{\max }}{p}}.$$

(10)

Like the piezoelectric coefficient, the values of η represent sample properties and are independent of shock pressure. Therefore, the mean values and their uncertainties, based on three measurements, have also been presented in Table 5.

To showcase a potential implementation of the presented nanocomposites as nanogenerators, the electrical energy generated under a single shock wave has been calculated. In this context, the registered output voltages (Figure 10) have been squared and numerically integrated, and in this way, the generated energy has been determined:

$$E={\displaystyle \underset{0}{\overset{t_1}{\int }}\frac{U^2}{R_V}}dt,$$

(11)

where t₁ stands for a time interval of the piezoelectric signal response. In addition, surface power densities have also been gained to determine the energy obtained from the surface unit:

$$P_S={\displaystyle \frac{1}{A}}\cdot {\displaystyle \frac{E}{t_1}},$$

(12)

where E is the energy determined by Equation (11), A is a surface active area, and t₁ is a piezoelectric response time interval. The calculated values are also presented in Table 5. Coefficient η, as well as d_ij, define sensing properties of the presented samples, while calculated energy (E) and power densities (P_S) describe the potential usefulness of samples as nanogenerators.

Each of the investigated samples generates voltage through the piezoelectric effect within a 5 ms timeframe (Figure 10). A positive voltage is induced when the sample, and consequently its piezoelectric component, is compressed, while a negative voltage is generated during sample extension. In general, the output voltage generated by sandwich-type samples, and consequently the energy and power, is higher than that produced by planar-type samples of the same composite. This difference is attributed to the distinct orientation of A¹⁵B¹⁶C¹⁷ nanowires in the nanocomposite. The planar samples exhibit a higher piezoelectric response due to the elevated longitudinal piezoelectric coefficient of d₃₃ compared to the resultant transverse d₃₁ and d₃₂. This is due to the high anisotropy of the piezoelectric properties of A¹⁵B¹⁶C¹⁷ compounds, especially SbSI [45].

The received value of the longitudinal d₃₃ coefficient for the PVP/SbSI nanocomposite reaches 100 pC/N, and transverse d₃₁ is equal to 35 pC/N. These values are understandably smaller than those for the SbSI single crystal, which reaches 1000 pC/N for d₃₃ [45] and 340 pC/N for d₃₁ [45,47]. The PVP/SbSI composite demonstrated a higher output voltage compared to the PVP/SbSeI composite. This is a consequence of the superior piezoelectric properties of SbSI. It is worth emphasizing that the generated power in the current study is the highest among all previously examined SbSI composites [26,27,28], and it compares favourably with contemporary achievements [61,62]. However, it is crucial to consider that, for a direct comparison of the generated values, one must also account for the excitation mechanism type and its broadly defined force.

4.4.2. Determination of the d₁₁/d₁₂ Piezoelectric Coefficients

The planar-type samples were meticulously crafted, featuring nanofibres aligned both perpendicular and parallel to the electrodes. This was done to investigate the previously mentioned anisotropic properties of PVP/SbSI and PVP/SbSeI nanocomposites. The two types of prepared samples shared identical dimensions, detailed in Table 2 and illustrated in Figure 4a and Figure 4c for PVP/SbSeI and PVP/SbSI nanocomposites, respectively. The sole distinction between them lay in the nanofibres’ alignment relative to the gold electrodes.

The recorded output voltages, excited by air pressure excitation, are graphically presented in Figure 11. These signals were acquired under the pressure of 17.03 bars at the temperature of 298 K.

The green curves in Figure 11a,b mirror the corresponding green curves in Figure 10b,d for PVP/SbSeI and PVP/SbSI nanocomposites. These curves are reiterated here to emphasize and highlight the observed changes. For planar-type samples with fibres aligned parallel to gold electrodes, both the excitation and voltage measurements occur perpendicular to the c-axis of the crystals. Consequently, the resultant values of the d₁₁ and d₁₂ components are under examination.

The registered signal is significantly smaller when nanofibres are aligned parallel to the electrodes in the PVP/SbSI sample. In the case of PVP/SbSeI, the signal experiences almost complete attenuation for this configuration. It provides clear evidence of the anisotropic properties inherent in these composites. Notably, SbSI demonstrates superior piezoelectric features compared to SbSeI, resulting in a higher signal for both sample configurations.

The response duration for both nanocomposites appears independent of the relative orientation of nanofibres and electrodes. However, it is noteworthy that the response duration is shorter for PVP/SbSeI. To facilitate a comparative analysis of the piezoelectric response for both sample configurations, the parameters from Equations (9)–(12) have been designated and are presented in

Table 6. Piezoelectric properties of the investigated nanocomposites in planar configurations with various nanowire alignments. From left to right, the columns indicate: nanocomposite type, nanowires alignment, maximum voltage (U_max), piezoelectric response duration (t₁), sample effective area (A), piezoelectric coefficient (d_ij), sensitivity (η), generated energy (E), and power surface density (P_S) for PVP/SbSeI and PVP/SbSI planar-type samples with electrodes perpendicular and parallel to nanofibres. In the absence of provided uncertainties, its values are negligibly small. Description in the body text.

Sample	Nanowires and Electrodes Alignment	Umax [V]	t1 [ms]	A [mm2]	dij [pC/N]	η [mV/bar]	E [nJ]	PS [µW/cm2]
PVP/SbSeI planar	parallel	0.212	2.5	5.34	d11 = 7.1	12.4	0.019	0.14
	perpendicular	1.000	2.5	5.34	d31 = 33.3	58.8	0.170	1.28
PVP/SbSI planar	parallel	0.720	4.3	5.34	d11 = 24.0	42.3	0.284	1.24
	perpendicular	1.116	4.3	5.34	d31 = 37.1	65.5	1.080	4.71

.

It is evident that, in the case of the PVP/SbSeI composite, both the energy and generated power are nearly 10 times smaller when the nanofibres are aligned in a parallel configuration compared to the perpendicular alignment. In contrast, for the PVP/SbSI composite, the difference is less pronounced. Unfortunately, the specific component values of the piezoelectric tensor remain unknown for both PVP/SbSI and PVP/SbSeI composites, as well as for SbSI and SbSeI nanowires. However, prior investigations into polycrystalline SbSI within a polymer matrix and SbSI ceramics have suggested that the transverse components of the piezoelectric tensor are approximately half the magnitude of the longitudinal values [46,63]. Furthermore, the spontaneous polarization aligns along the c-axis of SbSI crystals, thereby enlarging the piezoelectric properties in this specific direction [64]. It is reasonable to anticipate a similar dependence in the current case.

4.4.3. Influence of Temperature on Piezoelectric Properties

Antimony sulphoiodide nanowires exhibit ferroelectric properties with a phase transition occurring at T_C = 292(1) K [52]. In the case of the PVP/SbSI nanocomposite, the Curie temperature was previously estimated to be 290 K. To facilitate comparison with earlier measurements carried out at 298 K, just above the Curie point, additional measurements were carried out at 278 K within the ferroelectric phase. It is noteworthy that the piezoelectric coefficient in ferroelectrics is highly temperature-dependent, especially in proximity to the Curie point. In detail, SbSI exhibits a rapid decrease in the paraelectric phase, while a slower variation occurs in the ferroelectric phase [14]. In the case of SbSeI, if there is a Curie point, it is below 50 K [65]. Additionally, no phase transition is observed in SbSeI nanowires in the temperature range of 250–330 K [16].

The voltage responses recorded from planar-type samples of PVP/SbSI and PVP/SbSeI, with nanowires aligned perpendicularly to electrodes, are presented in Figure 12 for two different temperatures.

Once again, the green curves in Figure 12a,b precisely mirror the corresponding green curves in Figure 10b,d for PVP/SbSeI and PVP/SbSI nanocomposites, emphasizing and highlighting the observed changes. The measured signal values show an increase with a decrease in temperature for both nanocomposites. While the improvement is subtle for PVP/SbSeI, the signal measured in 278 K is twice as high as in 298 K for PVP/SbSI. At 278 K, the SbSI nanowires are in the ferroelectric phase, whereas at 298 K, they are in the paraelectric phase. In contrast, the SbSeI nanowires remain in the paraelectric phase at both temperatures.

To enable a comparative analysis of the piezoelectric response for both sample configurations, the parameters derived from Equations (9)–(12) have been determined and are presented in

Table 7. Piezoelectric properties of the investigated nanocomposites in planar configurations at different temperatures. From left to right, the columns indicate: nanocomposite type and contact configuration, temperature (T), maximum voltage (U_max), piezoelectric response duration (t₁), sample effective area (A), piezoelectric coefficient (d₃₁), sensitivity (η), generated energy (E), and power surface density (P_S) for PVP/SbSeI and PVP/SbSI planar-type samples with electrodes perpendicular to nanofibres at two temperatures. In the absence of provided uncertainties, its values are negligibly small. Description in the text.

Sample	T [K]	Umax [V]	t1 [ms]	A [mm2]	d31 [pC/N]	η [mV/bar]	E [nJ]	PS [µW/cm2]
PVP/SbSeI planar	278	1.140	0.8	5.34	38.0	66.9	0.242	5.67
	298	1.000	2.5	5.34	33.3	58.7	0.170	1.28
PVP/SbSI planar	278	2.105	4.3	5.34	70.2	123.6	4.780	20.82
	298	1.116	4.3	5.34	37.2	65.5	1.080	4.71

.

Significant improvements in the d₃₁ and η coefficients, as well as the generated energy and the gained power surface density, are noticeable with a decrease in temperature for the PVP/SbSI nanocomposite. At the same time, only slight enhancements are observed for the PVP/SbSeI nanocomposite. Furthermore, the PVP/SbSI composite demonstrates superior performance at 298 K compared to the PVP/SbSeI composite at 278 K. Considering that measurements for the PVP/SbSI composite at 298 K were carried out slightly above and below the Curie point, it is anticipated that the piezoelectric response of the PVP/SbSI composite would be even higher at temperatures closer to T_C.

4.4.4. Comparison with Other PVP-Based Materials

To better contextualize the performance of the material developed in this study, we conducted a comparison with selected PVP-based materials previously reported in the literature. This comparison highlights the relative improvements achieved in terms of key performance metrics and demonstrates the advantages of the current approach. The best piezoelectric performance was achieved for the sandwich-type samples, and therefore, these results were selected for comparison.

Table 8. Comparison of piezoelectric properties of PVP-based materials reported in the literature with the PVP/SbSI and PVP/SbSeI nanocomposite sandwich structures presented in this work.

Sample Type	Umax [V]	d33 [pC/N]	Comment	Reference
PVP/SbSeI	1.564	64.4(73)	Voltage for sandwich structure at 17.03 bar	This work
PVP/SbSI	2.092	98(20)	Voltage for sandwich structure at 17.03 bar	This work
PVP/(Ti + Zr)		23.2	For the composite with 1.25 PVP content.	[66]
Glycine/PVP		0.11–7.45	Depending on the glycine-to-PVP ratio	[67]
KNN–LN-PVP		140	0.94KNN–0.06LN with PVP 55K	[68]
PZT/MWCNT/PVP	16		Mechanical impact	[69]
P(VDF-TrFE)/PVP/AIL	6.5		Finger tapping	[70]
PVP/BaTiO3/MXene/PVDF-TrFE	3.3		Pressing with a force of 18 N	[71]
PVP/ZnO	45		Finger tapping	[72]
[PAN/BTO]@[PANI/PVP]	~1		Nanofibre membrane with 15% BTO contents	[73]

Table 8. Comparison of piezoelectric properties of PVP-based materials reported in the literature with the PVP/SbSI and PVP/SbSeI nanocomposite sandwich structures presented in this work.

Sample Type	Umax [V]	d33 [pC/N]	Comment	Reference
PVP/SbSeI	1.564	64.4(73)	Voltage for sandwich structure at 17.03 bar	This work
PVP/SbSI	2.092	98(20)	Voltage for sandwich structure at 17.03 bar	This work
PVP/(Ti + Zr)		23.2	For the composite with 1.25 PVP content.	[66]
Glycine/PVP		0.11–7.45	Depending on the glycine-to-PVP ratio	[67]
KNN–LN-PVP		140	0.94KNN–0.06LN with PVP 55K	[68]
PZT/MWCNT/PVP	16		Mechanical impact	[69]
P(VDF-TrFE)/PVP/AIL	6.5		Finger tapping	[70]
PVP/BaTiO3/MXene/PVDF-TrFE	3.3		Pressing with a force of 18 N	[71]
PVP/ZnO	45		Finger tapping	[72]
[PAN/BTO]@[PANI/PVP]	~1		Nanofibre membrane with 15% BTO contents	[73]

Table 8 summarizes the piezoelectric performance of various polymer-based composites reported in the literature, as well as those obtained in this work. The maximum piezoelectric coefficients (d₃₃) and output voltages of our PVP/SbSI and PVP/SbSeI nanocomposites are comparable to or higher than those reported for similar systems. For instance, the d₃₃ of the PVP/SbSI composite is significantly higher than the respective values for glycine/PVP composites [67] and PVP/(Ti + Zr) composites [66], and is on the order of the piezoelectric coefficient reported for KNN–LN-PVP composites [68]. Similarly, the piezoelectric performance of the PVP/SbSeI composite is within the range of or exceeds that of other lead-free polymer-based piezoelectric materials, such as PVP/(Ti + Zr) composites [66] or P(VDF-TrFE)/PVP-based systems [70,71], especially considering that the literature voltage values were measured under different forms of mechanical impact, whereas the results reported here were obtained under air pressure impact. These comparisons indicate that the aligned SbSI and SbSeI nanowire composites developed in this work achieve competitive piezoelectric performance while retaining the advantages of polymer flexibility and facile processing.

5. Conclusions and Discussion

In this study, we employed the electrospinning method to incorporate two piezoelectric compounds, antimony sulphoiodide (SbSI) and antimony selenoiodide (SbSeI), into a Polyvinylpyrrolidone (PVP) matrix, forming one-dimensional composite nanostructures. The active phases, SbSI and SbSeI, were introduced into PVP to serve as the reinforcing constituents. The resulting nanofibres were collected on a grounded collector, forming a nonwoven sheet whose morphology and alignment could be influenced by the collector’s design. Precise control over the electrospinning process parameters enabled the fabrication of nanofibres with diameters comparable to the lateral dimensions of chalcohalides nanowires.

The nanocomposite exhibits absorption characteristics comparable to individual nanowires. Additionally, the nearly transparent properties of PVP for visible light make it highly suitable for optoelectronic applications. Notably, the material exhibits enhanced mechanical properties, including flexibility, which enhances its utility in various applications.

The activation energy of the PVP/SbSI nanocomposite in paraelectric and ferroelectric phases was estimated, with the Curie point identified at 290 K. No phase transition was confirmed in the PVP/SbSeI nanocomposite, showing only one value of activation energy in the examined temperature range.

The impedance spectroscopy results were presented. The Nyquist plots enabled the simulation of the sample as an equivalent circuit, aiding in the determination of potential applications in various systems. However, the unexpectedly low-frequency dependencies observed in dielectric spectroscopy indicate the necessity for further investigation to fully describe the composite properties.

The application of the proposed nanocomposites as nanogenerators or nanosensors was tested. Both nanocomposites exhibited anisotropic properties, corresponding to those of A¹⁵B¹⁶C¹⁷ compounds. It was also evident that these properties are influenced by temperature, especially in the case of PVP/SbSI. While the composites’ piezoelectric coefficients are smaller than those of single crystals due to the presence of non-piezoelectric PVP, their mechanical properties and strength surpass those of single crystals, making them more likely and easier to apply.

The polymer matrix and nanowire filler critically determine the final properties of these newly developed one-dimensional composite nanomaterials. Moreover, preliminary studies have shown that the application potential of the developed PVP/piezoelectric nanowire composite materials can be significantly expanded by embedding the constructed piezoelectric systems in resin, which eliminates the issue of PVP solubility in water. Furthermore, the phase transition in the PVP/SbSI nanocomposite enables the adjustment of its properties through temperature changes. This feature can be especially valuable in certain applications.