Development of a flexible piezoelectric and triboelectric energy harvester with piezo capacitive sensing ability from barium tungstate nanorod-dispersed PVDF nanofabrics

Lead-free flexible piezoelectric nanogenerator (PNG) and triboelectric nanogenerator (TENG) are sought after due to their ability to produce electricity by harnessing wasteful mechanical energy. A comprehensive understanding of additives and processing techniques is crucial for fine-tuning the performance of such energy systems. We have investigated in detail the effect of the addition of reverse microemulsion synthesized barium tungstate nanorods (BWN) on morphology, crystallinity, polymorphism of electrospun nanofabrics of poly(vinylidene fluoride) (PVDF). The electroactive phase content of the nanofabrics was enhanced upon the addition of BWN and the highest electroactive phase content of 86.5% was observed in the nanofabric containing 3 wt% of BWN. The dielectric constant of the nanofabric containing 5 wt% BWN was ∼1.96 times higher than that of pristine electrospun PVDF nanofabric (EPVDF). The ratio of relative change in the capacitance to initial capacitance of the sensor fabricated from the same system was ∼4 times greater than that of EPVDF. Consequently, its piezoelectric and triboelectric performances were improved. The PNG fabricated using the nanofabric containing 3 wt% BWN produced the highest open-circuit voltage of 8 V under an applied load of 8 N. A TENG made using the same system was able to produce a voltage output of 200 V, which was 1.77 times as high as that of EPVDF under one-finger tapping in contact-separation mode. The same composite nanofabric produced piezoelectric and triboelectric power densities of 4.3 µW cm−2 and 646 µW cm−2, respectively. The TENG was able to light 40 LEDs under one finger tapping. Fluttering-driven TENG fabricated using the aforementioned nanofabric was able to produce a triboelectric voltage of 84 V at a wind speed of 7 m s−1. Overall, these nanofabrics could be a potential material for energy harvesting devices for powering wearable devices, environmental sensors, and internet of things.


Introduction
Nowadays, with the rise in miniature devices and the internet of things, energy harvesters have gained a great deal of attention as sustainable, environmentfriendly alternative power sources and can effectively reduce the dependency on batteries [1][2][3]. Piezoelectric nanogenerator (PNG) and triboelectric nanogenerator (TENG) produce power in the range of mW-µW, which could be sufficient to power the low-power miniature devices. Conventional PNGs based on ceramics such as barium titanate [4] and lead zirconate titanate (PZT) [5] have been used as energy harvesters, but their high processing temperature, brittleness, and design restriction limit the use of the same in flexible electronics and miniature devices. On the other hand, piezoelectric polymers are considered as an alternative to piezoceramic materials due to their excellent flexibility. However, piezoelectric polymers possess low piezoelectric coefficient, which hinders their use in practical applications. Therefore, there has been numerous efforts to improve the piezoelectric properties of polymers by various strategies that include the incorporation of nanoparticles in the polymer matrix and use of various processing techniques [6,7].
The triboelectric effect is another mechanism that can scavenge waste ambient energy more efficiently. During the triboelectric process, two dissimilar materials, namely tribo-positive and tribonegative, are repeatedly brought in contact with each other and separated, which leads to electrical discharges [3]. TENGs have been used to generate electricity through the combination of the triboelectric effect and electrostatic induction. Several strategies have been explored to enhance the triboelectric performance of nanogenerators, such as meticulous selection of materials [8][9][10][11][12], surface and interface modification [13][14][15][16], and different design configurations [17][18][19]. Material properties such as dielectric constant, surface charge density, chemical nature of the surface, and effective frictional contact area are the governing parameters that determine the triboelectric performance of a TENG [20][21][22].
Material selection is crucial for effectively utilizing the piezoelectric and triboelectric effects. The reasonably good piezoelectric response, high dielectric constant, flexibility, and ease of processing of poly(vinylidene fluoride) (PVDF) have attracted the attention of researchers. PVDF is also a wellknown potential tribo-negative material for highperformance TENGs [23][24][25]. PVDF is a semicrystalline polymer with a simple repeating unit of CH 2 and CF 2 . Among its five phases, β and γ are polar, so efforts have been made to enhance the electroactive phases using different processing techniques (mechanical, thermal, and electrical processing techniques) and by the addition of reinforcements that can act as nucleating agents selectively for the electroactive phases of PVDF. The incorporation of nanoscale fillers such as bismuth ferrite (BiFeO 3 ) [26], cobalt ferrite (CoFe 2 O 4 ) [27], zinc oxide (ZnO) [28], nanoclay [29], halloysite nanotube [30], graphitic carbon nitride (g-C 3 N 4 ) [31], organo-modified nickel cobalt layered double hydroxide (Ni-Co LDH) [32] has been proven to enhance the electroactive phase in PVDF. Also, the formation of the electroactive phase results in increased surface charge density due to high rapid polarization contributing to the enhancement of its triboelectric performance [33][34][35][36]. The enhanced dielectric property could also add to the advancement of triboelectric performance [20,37]. The incorporation of reinforcements such as MXene (Ti 3 C 2 T x ) [38], cobalt ferrite (CoFe 2 O 4 ) [39], dopamine-treated tin oxide (SnO 2 ) [34], iron oxide (Fe 3 O 4 ) [33], graphene quantum dots (GQDs) [40], cellulose nanocrystal [41], and multiwall carbon nanotubes [42] has been demonstrated in enhancing triboelectric performance through stabilizing and improving the β-phase and dielectric constant.
Electrospinning is a simple, versatile, and costeffective method to fabricate polymeric nanofabrics. During the electrospinning process, uniaxial mechanical stretching combined with electrical poling induces an electroactive phase in PVDF [43][44][45]. Also, it is an effective method to fabricate nanofabrics with high aspect ratio, adequate roughness, and enhanced effective frictional contact area, which together contribute to enhanced triboelectric performance [13,40,46].
Barium tungstate (BaWO 4 ) is one of the members of the alkaline earth tungstate family, with a tetragonal scheelite-type structure [47]. BaWO 4 has excellent dielectric property [48], and could act as a nucleating agent by interacting with PVDF chains, which makes it a suitable filler material for uplifting the piezoelectric and triboelectric behavior of PVDF. Therefore, combining electrospinning with addition of nanoscale BaWO 4 to PVDF could possibly enhance the piezoelectric and triboelectric properties of PVDF.
In the present work, the interfacial/electrostatic interaction between PVDF chains and barium tungstate nanorods (BWN) combined with electrospinning was used to induce the electroactive phase of PVDF; the superior dielectric property of BaWO 4 was taken into advantage to enhance the dielectric characteristics of PVDF. A comprehensive study was carried out to explore the effect of BWN on the morphology, polymorphism, dielectric, piezoelectric, and triboelectric properties of electrospun PVDF nanofabrics.

Synthesis of BWN
The BWN were synthesized using the reverse microemulsion method mentioned elsewhere [49].

Part 1: synthesis of Na 2 WO 4 microemulsion
A beaker was filled with 25 ml of cyclohexane, 3.39 ml of Triton X-100 (surfactant), and 2 ml of 0.2 M aqueous Na 2 WO 4 solution in sequence under constant magnetic stirring (500 rpm). Then, 1.38 ml of 1-hexanol (co-surfactant) was added, and the resultant mixture was vigorously stirred at 1000 rpm for 15 min to create a transparent reverse microemulsion.

Part 2: synthesis of Ba(NO 3 ) 2 microemulsion
Ba(NO 3 ) 2 microemulsion was prepared using a similar method as mentioned in part 1, replacing Na 2 WO 4 solution with Ba(NO 3 ) 2 solution.
The Ba(NO 3 ) 2 microemulsion was gently added to the Na 2 WO 4 microemulsion under gentle stirring, and the mixture was allowed to turn into white and turbid. The resultant mixture was kept at 50 • C for 24 h and allowed to form a white precipitate. The precipitate was separated using a centrifuge, washed and cleaned using ethanol five times to remove the residual surfactant and organic solvent, and then dried at 60 • C in a hot air oven for 12 h to obtain the BWN.

Electrospinning of PVDF/BWN composite nanofabrics
About 13.5 w/v% solution of PVDF in a 9:1 mixture of DMF and acetone was prepared to synthesize pure electrospun PVDF nanofabric (EPVDF). For synthesizing the composite nanofabrics, accurately weighed BWN (0.5, 1, 1.5, 2, 3 and 5 wt% on the basis of the wt of PVDF) was added to 10 ml of the mixed solvents. The resultant dispersion was ultrasonicated for 90 min, then 1.35 g of PVDF was slowly added to it under constant magnetic stirring. The mixture was magnetically stirred for 10 h to ensure homogeneity. The aforementioned PVDF solution/homogenized mixture of PVDF and BWN were transferred to 10 ml syringes fitted with 22-gauge stainless steel needles, and subjected to electrospinning at 28 ± 2 • C under the optimized conditions as follows: applied voltage 18 kV; flow rate 0.6 ml h −1 ; spinneret tip to collector distance 17 cm; rotating drum collector speed 1500 rpm; relative humidity 50 ± 2%.

Characterization
Particle size and Zeta potential analyzer (Litesizer ™ 500, Anton-Paar, Austria) was used to measure the zeta potential of BWN. A particular amount of BWN was dispersed in water so that at least 70% transmittance was maintained to carry out the measurements.
The morphologies of BWN and nanofabrics were investigated by field emission scanning electron microscopy (Zeiss GeminiSEM 300 FESEM, Germany) and transmission electron microscopy (JEOL-2100 TEM, Japan). The nanofabrics were gold sputtered prior to FESEM imaging. Image-J software [50] was used for the measurement of fiber diameter. To calculate the average fiber diameter (AFD) and standard deviation (SD) of the diameter of the nanofabrics, 50 fibers were taken for the measurement, and three readings were taken for each fiber at different locations.
Fourier transform infrared (FTIR) spectroscopy (FTIR-4200, JASCO, Japan) was used to investigate the polymorphism of PVDF and the interaction between the polymer and filler. The measurement was carried out in the range of 4000-650 cm −1 at a resolution of 4 cm −1 with 32 scans in the ATR mode.
X-ray diffraction (XRD) (Empyrean-Malvern PANalytical, UK) (Cu K α = 0.154 nm, 45 kV, 40 mA) was done to study the polymorphic phases of PVDF in the nanofabrics. The samples were scanned in a 2θ range of 10 • -45 • at a step size of 0.013 • , and the time per step was 24 s.
The dielectric properties of the samples were examined using an inductance-capacitanceresistance (LCR) meter (RS PRO: LCR-6300, UK). The samples were kept between copper electrodes for the measurement, and the test was carried out in the frequency range of 1 kHz-200 kHz at a bias voltage of 1 V. The following equation was used for calculating dielectric constant: where ε r and ε 0 are the dielectric constant composite nanofabrics and dielectric permittivity of air (8.854 × 10 −12 F m −1 ), C and d are the capacitance, and thickness of nanofiber mat, respectively, and A is the surface area of the electrode. The changes in the capacitance of the composite nanofabrics were also examined under the application of an external load (1,3,5,8,9, and 10 N) using a dielectric analyzer (Alpha A, Novocontrol Technologies, Germany). The sensitivity was obtained using the following equation: where ∆C is the relative change in capacitance under external load (C − C 0 ), C 0 is the initial capacitance, and ∆L is the applied load. The slope of the response curve was used to determine the sensitivity of the device. The finger tapping force was measured using a force sensor (Model-9712B5, Kistler, Austria).

Evaluation of piezoelectric performance
The piezoelectric performances of the EPVDF and PVDF/BWN composite nanofabrics were evaluated using a digital oscilloscope (RS PRO-1052, UK) and a standard linear motor kit (Cyltronic AG, Switzerland). A copper electrode was attached to a nonconducting plastic film, and the nanofabric was sandwiched between these copper electrodes to fabricate the PNG device. The electrodes were connected to the oscilloscope to measure the voltage output. Then, the nanogenerator was tested under the application of an external load (figure 1). A slight modification was made for the piezo-capacitance sensor, as shown in figure S1 (ESI).

Evaluation of triboelectric performance
The triboelectric performances of EPVDF and PVD-F/BWN composite nanofabrics were studied using a digital oscilloscope (Tektronix, TDS 2012C, USA). The TENG was fabricated using the nanofabric as the tribo-negative material (2 cm × 2 cm) and a copper strip as the tribo-positive material (2 cm × 2 cm). The nanofabric was pasted onto a copper electrode, and another copper electrode was attached to the transparent acrylic sheet as shown in figure 2. The top copper electrode, i.e. the tribo-positive material, was glued to the acrylic sheet such that it creates an arc shape above the tribo-negative material. The distance between the tribo-positive and tribo-negative materials was 10 mm. The average diameter and length of these rods were 252 ± 47 nm and 1421 ± 370 nm, respectively (the measurements were done on 50 nanorods using image-J tool).

SEM and TEM analysis of the nanofabrics
The AFD of composite nanofabrics increased upon increasing the loading of BWN (figure 4), which could be attributed to the increased viscosity of the polymer   solution upon the addition of BWN [52,53]. The diameter of fiber signifies the amount of stretching that occurred during the electrospinning, and this stretching could induce the formation of the electroactive phase in PVDF as discussed in our previous reports [54,55]. However, apart from the stretching effect, the interaction between the nanofiller particles and polymer chains also has a significant role in inducing the electroactive phase. Herein, the interaction between the PVDF polymer chains and the surface of BWN could be the dominating factor over the stretching effect, leading to enhancement in the electroactive phase. Figure 5 shows the representative TEM images of PVDF/BWN-3 composite nanofibers imaged at two locations. The BWN were finely dispersed and encapsulated by the PVDF (figure 5(a)). Also, the particles were aligned along the fiber axis due to the stretching of the fiber jets during electrospinning. Some nanorods are observed also on the surface of the nanofiber, as shown in figure 5(b).

FTIR spectroscopy of the nanofabrics
FTIR spectroscopy was carried out to study the polymorphism in the EPVDF and PVDF/BWN composite nanofabrics (figure 6). The peaks at 762, 796, and 975 cm −1 are the signatures of the α-phase. The peaks 840 and 1275 cm −1 represent the β-phase. The characteristic peak of the γ-phase did not appear at 1232 cm −1 in EPVDF [6,56]. The FTIR spectra of PVDF/BWN composite nanofabrics reveal that the intensities of the peaks representing the α-phase, i.e. those at 763, 795, and 975 cm −1 were diminished upon the addition of BWN. Also, the intensity of the polar phase peak at 840 cm −1 was improved, and a small shoulder peak of the γ-phase appeared at 1232 cm −1 , which can be seen in the enlarged view of FTIR spectra in figure 6(b). However, the significant peak at 1275 cm −1 of the β-phase and the absence of peaks at 776, 811, and 833 cm −1 of the γ-phase hint on the predominance of the β-phase in the polar phase content [6,56,57]. The peaks at 3022 cm −1 and 2981 cm −1 correspond to the CH 2 stretching vibrations of the PVDF polymer chain. The peaks of CH 2 were shifted to lower wavenumber (figure 6(c)), which may be attributed to the interfacial/electrostatic interaction between the negatively charged surface of the BWN particle (with a zeta potential of −38.8 mV) and CH 2 group of the PVDF chains [58][59][60].
The following equation was used for the calculation of the electroactive phase content [F(EA)] in EPVDF and PVDF/BWN composite nanofabrics [56]: where A 840 and A 762 are the absorbances at 840 cm −1 and 762 cm −1, respectively. The K 840 and K 762 are the absorbance coefficients at 840 cm −1 and 762 cm −1 , respectively, with values of 7.7 × 10 4 cm 2 mol −1 and 6.6 × 10 4 cm 2 mol −1 . The intensities at 1275 cm −1 and 1234 cm −1 were used for the quantitative measurement, as the absorbance intensities of these peaks are related to the amount of the β and γ-phases present, respectively [56] × 100% (5) where ∆H β is the height difference (absorbance intensity) between the peak at 1275 cm −1 and its nearest valley at 1260 cm −1 and ∆Hγ is the height  difference between the peak at 1234 cm −1 and the valley at 1225 cm −1 [56]. The F(EA) was improved with the BWN loading up to 3 wt%; after which the polar phase content was decreased (table 1). The decrease in the F(EA) at higher BWN loading may be attributed to the agglomeration of particles. The highest electroactive phase content of 86.5% was observed for PVDF/BWN-3 composite nanofabrics. The enhancement in the polar phase could be attributed to the following reasons: first, the C=O group of the polar solvent DMF might have weakly interacted with CH 2 of PVDF chains, lowering the energy barrier for the formation of the electroactive phase [54,61]. Second, the interaction between the negatively charged surface of BWN and the positively charged CH 2 group of PVDF might have compelled to arrange the PVDF chains in the polar conformation [58-60, 62, 63].
However, the variation in surface charge distribution over the nanoparticles causes repulsion between CH 2 and positive charge, and attraction between CF 2 and positive charge of BWN led to the formation of a small fraction of the γ-phase along with its βcounterpart ( figure 7). Similar results were also reported by other researchers [64]. Finally, the electrospinning process is proven to enhance the electroactive phase by mechanical stretching and local electrical poling of the fiber. The synergism of all these factors could have contributed to enhancement in the electroactive phase.    [6,65]. The peaks at 26.6 • , 28.1 • , and 32 • correspond to BWN. The intensity of the α-phase peak was reduced while that of the β-phase was enhanced upon the addition of BWN, which may be due to the improved electroactive phase in PVDF. The results suggest that the electroactive phase of PVDF was enhanced upon the addition of BWN, which agrees with the FTIR results. In XRD curves, the peaks of the γ-phase were not observed. The γ-phase peaks overlap with that of the α-phase at 18.5 • , 20 • , and 26.7 • , and these overlapped peaks hinder the identification of a small amount of the γ-phase present in the PVDF/BWN composite nanofabrics using only the XRD data [6,66]. However, the FTIR data confirms the absence of the γ-phase in EPVDF, which concludes that the peak at 18.5 • belongs to the αphase in EPVDF. So, the FTIR spectral results were taken into account for the phase study along with that of XRD.

XRD of the composite nanofabrics
The total crystallinity was calculated by deconvoluting the XRD curves, as depicted in figure 9. As the peaks of the γ-phase were absent in XRD, it was not considered for calculating the total crystallinity.
The crystallinity of the composites increased upon the addition of BWN. The higher total crystallinity of composite nanofabrics is due to the nucleating effect of the BWN for the PVDF macromolecules [67][68][69][70].

Dielectric properties
The dielectric properties of the EPVDF and PVD-F/BWN composite nanofabrics were measured with respect to frequency at room temperature (figure 10). In the composites, the dielectric constant increased slowly with an increase in the filler concentration at all frequencies. The resultant improvement in the dielectric constant of composite nanofabrics is caused by the electroactive phase content [55,71], the addition of the filler with a high dielectric constant [48,72], and the charge accumulation at the interface of PVDF and BWN particles in the low-frequency region [73]. At 1 kHz, the highest dielectric constants of 17.64 and 17.68 were observed for the PVDF/BWN-3 and PVDF/BWN-5 composite nanofabrics, respectively. The dielectric constant decreased as the frequency was increased, which could be due to frequencydependent dipolar and space charge polarization; when dipoles fail to align themselves to the applied field as frequency increases, the dielectric constant  decreases [74]. The conduction loss and polarization (dipolar, distortion, and interfacial) determine the dielectric loss in polymer composites [75]. The dielectric loss of EPVDF was low compared to that of the composites. The composites have a high dielectric loss, which could be due to the interfacial polarization and conductive loss caused by the formation of smaller domains of the network [76]. The increase in the concentration of the filler leads to an increase in heterogeneity, which leads to an increase in interfacial polarization resulting in the high dielectric constant and dielectric loss at low frequency. In addition to the dielectric properties of the filler and polymer, the morphology, dispersion, and polymer-filler interaction also affect the dielectric properties of the composites [72]. Also, the dipole density of nanofabrics varies due to their porous structure, affecting the dielectric property [32]. The dielectric constant represents the charge storage ability, whereas the dielectric loss represents the leaky nature or inefficacy of the material to maintain the charges. A material with a low dielectric loss can maintain the charges efficiently, and the high dielectric constant can improve triboelectric charge density due to enhanced capacitance of the material. Therefore, the high dielectric constant and low dielectric loss of the composite nanofabrics makes them ideal for good triboelectric performance.

Piezo capacitance
The capacitive-based sensor was fabricated using two copper electrodes and PVDF/BWN-5 nanofabric as the dielectric material (parallel plate configuration) and linear motor setup was used for the application external force (figure S1). The PVDF/BWN-5 nanofabric was used to fabricate the capacitive-based pressure sensor as it has the highest dielectric constant among all the composite nanofabrics. The performance of the sensor was examined using the linear motor kit, and the force was applied normal to the sensor. Figure 11 depicts the performance of the capacitive-based sensor of EPVDF and PVDF/BWN-5-based nanofabric.
The capacitance can be calculated using equation: where C is the capacitance of the nanofiber, ε r and ε 0 are the dielectric constant composite nanofabrics and dielectric permittivity of air (8.854 × 10 −12 F m −1 ), A and d are the area of the electrode and thickness of nanofiber mat, respectively. The capacitance of the PVDF/BWN nanofabricbased sensor was higher than that of the EPVDF nanofabric-based one. The enhanced capacitive performance of former may be due to its higher dielectric constant value. The ratio of change in capacitance to the initial capacitance of the composite nanofabric based sensor increased linearly upon the application of external force and its value increased by ∼4 times that of EPVDF based sensor under a load of 8 N. The increase in capacitance of the sensor could be due to the decrease in the distance between the electrode and the improved effective dielectric constant of the PVDF/BWN-5 nanofabric. The increase in dielectric constant could be due to improved packing density of this nanofabric upon application of external load, which gradually eliminated the pores [77][78][79]. The sensitivity is the relative change in capacitance value with respect to applied load and was measured by taking the slope of the curve. The sensitivity of the fabricated device was found to be 0.66 N −1 . The PVDF/BWN-5 composite nanofabric-based capacitive sensor with good sensitivity could find application in touch sensing, wearable devices, and medical devices [80,81].

Piezoelectric properties
The piezoelectric response of the PNG of EPVDF and PVDF/BWN composite nanofabrics was assessed using a standard linear motor kit. Figure 12 shows the piezoelectric performance of the PNG of EPVDF and PVDF/BWN composite nanofabrics. The PNG was tested under an applied force of 8 N (at a frequency of 1.25 Hz), and force was applied normal to the plane of the nanogenerator. The piezoelectric open-circuit voltage (V OC ) was increased upon the addition of filler, and the V OC continued to increase up to 3 wt% loading of the BWN, beyond which it decreased. The highest V OC recorded was 8 V for the PVDF/BWN-3 composite nanofabricbased PNG, which was ∼6.4 times that of the EPVDF based device. The enhanced piezoelectric response may be attributed to the improved electroactive phase in the PVDF [55,82]. The morphology and surface nature of additives and relative humidity may also contribute to the enhanced piezoelectric performance [83][84][85]. The V OC was decreased at 5 wt% loading of the filler, which may be due to the reduction of the electroactive phase content. The piezoelectric voltage generation in EPVDF and PVDF/BWN composite nanofabrics is due to the piezoelectric potential developed across the surface of the nanogenerator upon the application of external force. The developed piezoelectric potential causes electrons to flow, which are collected by the electrode and supplied to the external load. The reverse output voltage is produced due to the release of external load that brings the nanogenerator back to its initial position. Figure 12(b) shows the change in V OC and power density across the different load resistances for the PVDF/BWN-3 composite nanofabric-based PNG. The voltage and power density were increased as the load resistance increased.
The power density was calculated using the following equation: where V is V OC across the load resistance, A is the area of PNG, and R l is load resistance across the circuit. The maximum power density of 4.3 µW cm −2 was recorded across the load resistance of 8 MΩ.

Triboelectric properties
TENG based on EPVDF and PVDF/BWN composite nanofabrics were fabricated as mentioned in the experimental section ( figure 2). In order to evaluate the influence of the BWN loading on the triboelectric performance of the PVDF/BWN composite nanofabrics, TENG was fabricated using composite nanofabrics with different loading of BWN. The triboelectric performance of the fabricated TENG was evaluated under one finger tapping (∼avg 3.8 N@4.5 ± 0.5 Hz) in contact-separation mode ( figure 13). Figure 14 (Video) shows the PVDF/BWN-3 based TENG tested under one finger tapping and its corresponding response in oscilloscope. The EPVDF-based TENG showed a triboelectric V OC of 112.8 V, where the pressing signal was in upward direction and the releasing signal was in the downward direction. The PVDF/BWN composite nanofabric TENGs showed better triboelectric performance than EPVDF-based TENG. The triboelectric V OC was improved upon an increase in the loading of the BWN up to 3 wt%, and at higher filler loading, it was decreased. The TENG based on PVDF/BWN-3 composite nanofabrics displayed the highest V OC with a maximum peak output voltage of 200 V. The V OC of the TENG based on PVDF/BWN-3 composite nanofabrics has improved performance by 1.77 times as high as that of EPVDF.
The following equation can be used for expressing the triboelectric output voltage of TENG [20]: where ε 0 and ε r are the permittivity of vacuum and the nanofiber mat, respectively. The σ 0 and ∆σ are the triboelectric charge density on the nanofabric and the transferred charge density on the electrode, respectively. x(t) is the interlayer distance, d c is the thickness of the nanofabric, and t is time.
For the open-circuit voltage (V OC ), the above equation will be given as follows: The above equation implies that the V OC depends on the distance of separation and triboelectric charge density. The capacitance of the triboelectric material determines the surface charge density in contact mode TENG because the TENG acts both as the energy storage and the energy producing device. Therefore, the surface charge density is proportional to the capacitance of the triboelectric material [20,37].
The capacitance is directly proportional to the dielectric constant (ε r ) and the contact surface area for the same thickness of tribo-materials, so it can  be concluded that the improvement in the dielectric constant and contact surface area could enhance the triboelectric performance of the TENG. The PVD-F/BWN composite nanofabric-based TENG showed an increase in the triboelectric performance with an increase in the loading of BWN up to 3 wt%, beyond which the output decreased. The increase in the triboelectric output of PVDF/BWN composite nanofabrics-based TENG could be due to the rise in the dielectric constant. In addition, the enhanced electroactive phase will cause prompt alignment of dipoles, leading to the quick introduction of negative charges onto the surface of composite nanofabrics, which could have improved triboelectric performance [33][34][35].
However, at a higher concentration of the filler, even though the dielectric constant was high, the triboelectric performance was reduced. The decrease in triboelectric performance at higher loading could be due to the high dielectric loss and reduction in the electroactive phase [86]. Also, at higher loading, fillers present on the surface of nanofiber will be more, which could potentially reduce the surface contact area of PVDF with copper, causing the reduction in triboelectric performance at higher loading [36].
The instant power density of PVDF/BWN-3 was calculated by using the measured voltage across various resistances, i.e. 0.39-13 MΩ. As depicted in figure 13(b), the voltage across resistance increased as the resistance was increased. The maximum power density of 646 µW cm −2 was obtained at 5.2 MΩ. The triboelectric performance of TENG based on PVDF/BWN-3 was comparable with that of some similar systems reported in literature (table 2). Figure 13(c) shows the working mechanism of TENG, where the combined effect of the triboelectric effect and piezoelectric effect of EPVDF and PVDF/BWN TENG under contact-separation mode is demonstrated. The fabricated TENG was archshaped at one side, where the copper tape was used as the top electrode as well as tribo-positive material for the device. The other side was flat, where the PVDF/BWN nanofabric was placed over the copper tape (bottom electrode), which acts as tribo-negative material. The copper tape is positively charged as it gives the electrons, and the PVDF/BWN nanofabric is negatively charged as it accepts the electrons. At the initial position, there was no contact between the triboelectric surfaces, which led to no charge transfer/generation resulting in no electric potential across the electrodes. Upon the application of external force, the top copper electrode approaches the PVD-F/BWN composite nanofabric and when it comes in contact with the nanofabric causes friction/rubbing between both the materials, which leads to the generation of charges at the interface due to the triboelectric effect. The piezoelectric material generates charges under the application of an external load due to the alignment of dipoles in the direction of the applied force. The enhanced negative charge generation due to the piezoelectric effect could promote triboelectric performance. When the force is released, both surfaces are separated, and the charges generated on the surface introduce the opposite charges on the copper electrodes due to electrostatic induction that causes the development of great electrical potential difference between the electrodes. In order to balance the developed electrical potential, the electrons move from the bottom electrode to the top electrode through the external circuit resulting in an electrical signal. Once the top electrode reaches its top position, electrical potential comes to equilibrium, resulting in no further flow of electrons. As the external force is applied on the top surface again, the top copper electrode approaches the bottom layer recreating the opposite potential difference between the electrodes, which forces charges to flow back to the bottom electrode resulting in an electrical signal [34,36]. Furthermore, the structural integrity and triboelectric performance of TENG remained almost unchanged even after ∼600 tapping cycles, suggesting the excellent durability of the device ( figure 13(d)). The PVDF/BWN-3 nanofabric based TENG was used to light commercial LEDs and a maximum of 40 LEDs were illuminated (figures 13(e) and 15).
The fluttering-driven triboelectric nanogenerator (FDTENG) was fabricated to harvest energy from wind, as illustrated in figure 16(a). An electric fan (Tornado II, Orient electric, India) was used to blow wind at different speeds (measured using Professional Instruments Anemometer, India). FDTENG is the small rectangular wind channel of acrylic sheets attached. The length of the wind channel was 9 cm, and its cross sectional area was 2.8 × 2 cm 2 . At the top and bottom of the wind channel, PVDF/BWN-3 composite nanofabric with copper electrodes was attached. The copper foil of a thickness of 30 µm was used as a flag for the FDTENG device, which was attached horizontally at the inlet of the wind channel. The copper flag acts as an electrode as well as a fluttering body in the FDTENG device. Figure 16(b) shows the voltage output of FDTENG at different wind speeds. The voltage output increased as the wind speed increased. The increase in wind speed causes an increase in the fluttering frequency. With the restricted amplitude of the flag, increased frequency causes the contact force to rise with the nanofiber mat. Due to the high contact force, the surface charge density increases, which will lead to the higher output voltage of FDTENG at the highest wind speed [87]. The highest V OC of 84 V at a wind speed of 7 m s −1 was observed. Meanwhile, V OC values of 48.4 and 72.8 V were recorded at the wind speeds of 6.3 and 6.6 m s −1 , respectively. Figure 16(c) shows the working principle of the FDTENG. As wind flows through the channel at a particular speed, the copper flag starts fluttering, leading to alternate contact and separation between the tribo-positive and tribo-negative materials. As the copper flag comes in contact with PVD-F/BWN composite nanofabric, charge transfer occurs between the tribo-positive copper flag and tribonegative nanofabric due to the triboelectric effect. An electric potential is developed across the copper flag and nanofiber mat due to electrostatic induction as soon as the flag is separated. This leads to the flow of electrons through the external circuit. When the flag approaches the nanofiber mat again, the reverse potential is created, causing charges to flow in the reverse direction, thus developing an alternating voltage [87,88]. Figure 17 shows the FDTENG in operation and its corresponding response.

Summary and conclusions
Barium tungstate nanorod-dispersed PVDF nanofabrics with enhanced electroactivity for energy harvesting applications have been developed using electrospinning. The highest electroactive phase content of 86.5% was attained in the composite nanofabric containing 3 wt% of BWN. The dielectric constant of the composite nanofabric containing 5 wt% of BWN was ∼1.96 times as high as that of EPVDF. The piezo capacitive sensing of the PVDF nanofabric containing 5 wt% BWN was improved by ∼4 times compared with EPVDF. Most importantly, the piezoelectric and triboelectric performances of the PVDF nanofabrics were significantly boosted upon the incorporation of BWN due to a well-enhanced electroactive phase content and improved dielectric properties. The PNG based on the nanofabric containing 3 wt% BWN was capable of producing V OC up to 8 V (∼6.4 times that of EPVDF) and power density up to 4.3 µW cm −2 under 8 N force. The TENG fabricated using the nanofabric incorporating 3 wt% BWN was able to produce a maximum V OC of 200 V (∼1.77 times that of EPVDF) and a maximum power density of 646 µW cm −2 under one finger tapping. Also, 40 LEDs were lit up by the TENG fabricated using the same nanofabric. Fluttering-driven energy harvester based on the nanofabric with 3 wt% BWN was capable of producing ∼84 V at a wind speed of 7 m s −1 . This study has paved way for the use of the newly developed composite nanofabrics in piezo capacitive sensing, piezoelectric and triboelectric energy harvesters that could be useful in self-powered sensors and miniature devices.

Data availability statement
All data that support the findings of this study are included within the article (and any supplementary files).