A BaTiO 3 @polyacrylonitrile/poly(vinylidene fluoride) nanofibrous composite membrane with high piezoelectricity based on the central combination design method and cross-electrospinning technology

Abstract

The rapid development of piezoelectric sensors has been studied extensively, owing to their good flexibility, wearability, high sensitivity and low cost. However, some inorganic materials with good piezoelectricity cannot make sensors flexible, and the organic materials with good flexibility have a weak output electrical signal and low strength. In order to explore and optimize the preparation technology of piezoelectric sensors, a BaTiO₃@polyacrylonitrile (PAN)/poly(vinylidene fluoride) (PVDF) nanofibrous composite membrane (NCM) was prepared by cross-electrospinning technology and the central combination design (CCD) method. The morphology, structure, hydrophobicity, mechanical properties and piezoelectricity of the BaTiO₃@PAN/PVDF NCMs were investigated. The BaTiO₃@PAN/PVDF NCMs had the better hydrophobicity and mechanical properties compared with the pure PAN/PVDF NCM. The 5BaTiO₃@PAN/PVDF NCM designed by CCD had a more uniform fiber diameter, and a more stable output voltage with a 46% improvement. With the help of cross-electrospinning technology and the CCD method, the NCM will be outstanding for the development of fabricating flexible wearable piezoelectric sensors.

Keywords

Nanofibrous composite membrane piezoelectric properties central combination design cross-electrospinning technology

In recent years, flexible and wearable piezoelectric sensors have become popular in the fields of medical monitoring, human–machine interaction and artificial intelligence,¹ because they can provide electricity for micro and nano electronic devices and reduce the use of chemical batteries.² In 1880, the piezoelectric effect was first discovered, which is that the piezoelectric potential would be generated owing to the break of central symmetry in the crystal structure under the action of the external force.³ In 1947, Roberts discovered the strong piezoelectric effect of BaTiO₃ material, which was a leap in the history of piezoelectric materials. The piezoelectric performance of BaTiO₃ is good, but it cannot make sensors flexible.⁴ In 1969, Kawai⁵ discovered the piezoelectric properties of poly(vinylidene fluoride) (PVDF). In particular, the increase of the β phase in PVDF can effectively improve its piezoelectric properties through inducing dipole alignment by stretching, polarization under a high electrical field, thermal annealing and filler incorporation.⁶ Recently, Wang et al.⁷ reported a new polyacrylonitrile (PAN) polymer piezoelectric material, the piezoelectric output signal of which was higher than that of PVDF. However, PVDF and PAN piezoelectric sensors^8,9 still have the problems¹⁰ of a weak output piezoelectric signal¹¹ and low strength.¹² Therefore, it is a research hotspot to explore the combination of organic and inorganic piezoelectric materials¹³ with a high dielectric constant¹⁴ and good flexibility.¹⁵

The flexible wearable applications of piezoelectric materials have been widely studied, especially in the design of nanofiber-based textile structures. Electrospinning technology^16,17 is the process of forming micro-nanofibers from polymer melts or solutions^18,19 under the action of a high-voltage electrostatic field. Under a high electrical field,²⁰ the phase structure of PVDF/PAN nanofibrous membranes will be affected, resulting in different piezoelectric properties. The conventional electrospinning technologies²¹ include single-needle²² and multi-needle technology.²³ Multi-needle electrospinning technology is often used to prepare multi-component or multifunctional nanofibrous composite membranes (NCMs), mainly due to the low efficiency of single-needle electrospinning technology. However, when multiple needles are used with composite solutions to electrospin multi-component fibers, it is difficult to find a common solvent or suitable spinning parameters. Cross-electrospinning technology can solve this problem to fabricate PVDF/PAN NCMs, the mechanism of which is that in the process of electrospinning, needles containing different solutions are cross-arranged along the spinneret, and the roller collects the nanofibers together.²⁴

Excellent technology and research methods often achieve twice the result with half the effort. Response surface analysis is a common method to optimize experimental conditions,²⁵ where the obtained value of each level in the prediction model is continuous.²⁶ At the same time, central combination design (CCD)²⁷ is the most widely used experimental design method in response surface analysis.²⁸ There are few researches on the optimal spinning parameters and response values of NCMs by CCD. To obtain a higher and more stable piezoelectricity, BaTiO₃@PAN/PVDF NCMs was fabricated based on the CCD method and cross-electrospinning technology in this paper. The effects of the addition of BaTiO₃ at different levels on the fiber morphology and distribution, crystal shape, hydrophobicity, mechanical properties and piezoelectric properties of the NCMs were investigated. The experimental results showed that the mechanical and piezoelectric properties of BaTiO₃@PAN/PVDF NCMs could be fabricated and significantly improved by cross-electrospinning and the CCD method.

Experimental details

Materials

PVDF with molecular weight of 1,200,000 was purchased from Solvay in the USA. PAN (molecular weight 80,000) was purchased from Qilu Petrochemical Company. N,N-dimethylformamide (DMF, ≥99.9%), anhydrous ethanol (≥99.9%) and deionized water (DI) were purchased from Sinopharm Chemical Reagent Co., Ltd. BaTiO₃ (B118840-100 g) and dopamine hydrochloride (DA·HCl, D103111-5 g) were purchased from the Aladdin Company. Trimethylol aminomethane (Tris) was obtained from the Biofroxx Company. All chemicals were used without further purification.

Modification of BaTiO₃

To prevent aggregation in the spinning solution, it was necessary to perform surface modification treatment on BaTiO₃. Each treatment of BaTiO₃ required DI, 1.2 g. Tris (0.24 g) was firstly dissolved in DI (200 mL) to form a buffer solution (PH: 8.5). Then BaTiO₃ (1.00 g) was added into the buffer solution, and the obtained solution was placed into an ultrasonic cell crusher for ultrasonic treatment for 1 h, so that the added BaTiO₃ nanoparticles were dispersed evenly. Then DA·HCl (1.20 g) was added to the solution, which would be stirred magnetically at room temperature for 24 h. The polydopamine modified BaTiO3 (PDA@BaTiO₃) nanoparticles formed by centrifugal separation were washed several times with DI and anhydrous ethanol. After being dried in the oven for 12 h, the modified BaTiO₃ were finally obtained.

Preparation of cross-electrospinning solutions

Modified BaTiO₃ was firstly dispersed in DMF and treated with ultrasound for 1 h, so that the modified BaTiO₃ was fully dispersed in the DMF. Then a quantitative PAN or PVDF powder was added and the solution was magnetically stirred at 40°C for 12 h. The uniform and stable modified BaTiO₃@PAN or BaTiO₃@PVDF spinning solutions were obtained, in which the levels of modified BaTiO₃ were 0%, 1%, 5%, 9% and 13%. According to the previous results from our group, the levels of PAN and PVDF were 12%, and 14%, respectively.

Preparation of cross-electrospun BaTiO3@PAN/PVDF NCMs

Under the environment temperature of 25–30°C and humidity of 30–35%, the spinning solutions were cross-arranged in four needles along the spinneret, where two needles contained BaTiO₃@PAN solution and the other two needles contained another solution. At an appropriate spinning voltage, liquid supply speed and receiving distance, NCMs with different levels of modified BaTiO₃ could be fabricated. After being dried in the oven at 60°C for 4 h, the modified BaTiO₃@PAN/PVDF NCMs were obtained, which were named 1BaTiO₃@PAN/PVDF, 5BaTiO₃@PAN/PVDF, 9BaTiO₃@PAN/PVDF and 13BaTiO₃@PAN/PVDF NCM. The blank sample was PAN/PVDF NCM. Then they were sealed with sealing tape and stored for subsequent testing.

CCD experiment

The three main process parameters of cross-electrospinning, the spinning voltage, receiving distance and liquid supply rate, were optimized by response surface analysis of the CCD. In the practical application, the smaller the diameter of the fiber, the more beneficial to obtain NCMs with a high sensitivity of piezoelectricity. Therefore, with average fiber diameter and output voltage as the response values of the CCD, Design Expert 8.0.5 software was used to design each factor, and the experimental results were analyzed to finally obtain the spinning parameters corresponding to the best response values. As the voltage increases, the electric field strength and the charge on the jet surface increase, which is conducive to the formation of nanofibers with smaller diameters. As the voltage continues to increase, the stretching time during jet flow spinning is further reduced, and the solvent cannot completely evaporate, resulting in the increase of the fiber average diameter. Therefore, based on previous experiments, the voltage for electrospinning was selected to be 28–32 kV. The receiving distance directly affects the electric field strength and the flight and stretching time of the jet in the electric field. The receiving distance is small and the average fiber diameter is small, but the solvent is not completely evaporated, and the fibers are uneven. The receiving distance is large, the fiber diameter is large and the uniformity decreases. Thus, we choose a receiving distance of 18–22 cm. There is a certain relationship between the liquid supply rate and the voltage during the spinning process. The higher the voltage, the greater the solution consumption, and a higher liquid supply rate is required to ensure the continuity of spinning. Thus, we choose a liquid supply rate of 0.5–1.0 mL/h.

Characterization

The surface morphology of the NCMs was observed by a scanning electron microscope (SEM, Nippon Electronics, JSM-IT300A). Image Pro Plus software was used to measure the fiber diameter of the recorded SEM photos and calculate the coefficient of variation (CV) value. According to the SEM results, the diameters of 100 fibers were randomly measured and then the average fiber diameter was calculated. Finally, the CV of the average fiber diameter was calculated using the following formula:

CV = standard deviation / mean value \times 100 %

The chemical structures of BaTiO₃ nanoparticles and BaTiO₃@PAN/PVDF NCMs were tested using a Fourier transform infrared spectrometer (FTIR, Bruker, Tensor27) with the wavenumber range of 650–4000 cm⁻¹ and an X-ray diffractometer (XRD, Panaco, Empyrean, the Netherlands) with the scanning range of 10–80°. The tensile tests of BaTiO3@PAN/PVDF NCMs were measured by an electronic universal testing machine (Shanghai Xeqiang Instrument Manufacturing Co., Ltd, CTM2100) at the speed of 50 mm/min. The piezoelectric properties of BaTiO₃@PAN/PVDF NCMs were tested under cyclic compression or the drop-hammer pattern and electrometer test. The piezoelectric sample was a sandwich structure, in which the middle layer was BaTiO₃@PAN/PVDF NCM and the upper and lower layers were packaged with polyethylene terephthalate (PET) film, protecting the NCM from damage.

Results and discussion

Morphology and structure

The scheme and physical photos of the cross-electrospun BaTiO₃@PAN/PVDF NCMs are shown in Figures 1(a) and (b). From Figure 1(b), at the level of BaTiO₃, the macroscopic color of the NCM gradually deepened from white to dark brown. Figures 1(c)–(g) show the SEM results of neat PAN/PVDF and BaTiO₃@PAN/PVDF NCMs. In Figure 1(c), it could be seen that the average fiber diameter of neat the PAN/PVDF nanofibrous membrane was 205 nm, and its CV value was 23.30%. From Figures 1(d)–(g), the average fiber diameter of NCMs became larger with the introduction of different levels of BaTiO₃. The average fiber diameter of the 5BaTiO3@PAN/PVDF NCM was 233.39 nm, and its CV value was 24.57%. In this case, the fiber of the NCMs was the thinnest and uniform with a smooth surface. The 13BaTiO3@PAN/PVDF NCM had an obvious beading phenomenon. This was because the excessive BaTiO₃ nanoparticles were agglomerated in the spinning solution, and fibers were not getting uniform drafting in the electric field during the cross-electrospinning process, resulting in a large difference in fiber diameter and the beading phenomenon.

Figure 1.

(a) Scheme of the cross-electrospinning process. (b) Physical photos of nanofibrous composite membranes (NCMs). From left to right, the levels of BaTiO₃ were 0%, 1%, 5%, 9% and 13%, respectively. The scanning electron microscopy results and diameter distribution of NCMs: (c) 0%; (d) 1%; (e) 5%; (f) 9%; (g) 13%. X-ray diffractometry of BaTiO₃ (h) and the NCM (i) and (j) Fourier transform infrared spectroscopy results. PVDF: poly(vinylidene fluoride); PAN: polyacrylonitrile.

In order to observe the structural changes of BaTiO₃ before and after modification and the effect of the BaTiO₃ level on the structure of NCMs, XRD and FTIR tests were conducted, and the results are shown in Figures 1(h)–(j). As shown in Figure 1(h), the BaTiO₃ had the same characteristic crystal structure, which indicated that the surface modification did not change the original crystal form of BaTiO₃, resulting in retaining excellent piezoelectric properties. As can be seen from Figure 1(i), the crystallinity of neat the PAN/PVDF NCM was very low. After BaTiO₃ was added as a nucleating agent, the diffraction peak of the (110) crystal plane of the BaTiO₃@PAN/PVDF NCM gradually increased. When the level of BaTiO₃ was 5%, the diffraction peak of the (110) crystal plane was the most obvious, which indicated that the 5BaTiO₃@PAN/PVDF NCM had the best crystallization and the fastest crystallization rate. When the level of BaTiO₃ exceeded 5%, the movement of macromolecular chain segments was inhibited by heterogeneous nucleation, resulting in the weakened diffraction peak of the (100) crystal plane and the deteriorated crystallization.

As shown in Figure 1(j), the absorption peak at 837 cm⁻¹ belonged to the bending vibration of -CH₂ in PAN, which represented the 31-helical conformation of PAN. Meanwhile, the absorption peak of 876 cm⁻¹ was attached to the joint action of the plane rocking vibration of -CH and the non-plane rocking vibration of -CH₂, which represented the planar zigzag conformation of PAN. The absorption peaks at 1076, 1173 and 1396 cm⁻¹ showed the β-crystal characteristic of PVDF in the cross-electrospun BaTiO₃@PAN/PVDF NCM. Under the action of electrostatic force generated in the electrostatic field during the cross-electrospinning process, PVDF was converted from the original α crystal form to a β crystal form, which would be helpful to improve the subsequent piezoelectric performance of NCMs.

Mechanical property and hydrophobicity

In order to investigate the effect of the BaTiO₃ level on the mechanical properties of NCMs, a tensile test was conducted and the results are shown in Figures 2(a)–(c). From Figures 2(a)–(c), the tensile strengths of the BaTiO₃@PAN/PVDF NCMs were higher than those of the pure PAN/PVDF NCMs, which indicated that the introduction of BaTiO₃ can significantly improve the mechanical properties of the NCMs. Furthermore, the 5BaTiO₃@PAN/PVDF NCM had the highest tensile strength of 3.12 MPa, which was 43% higher than that of the pure PAN/PVDF NCMs. The reason might be that the average fiber diameter and CV value of the 5BaTiO₃@PAN/PVDF NCM were both small at this time, resulting in no weak joint effect when most fibers withstand tensile stretching. In Figure 2(c), the addition of BaTiO₃ could slightly reduce the Young's modulus of the NCMs.

Figure 2.

Stress–strain curves (a), tensile strength (b), Young’s modulus (c) and water contact angle (WCA) (d) of the nanofibrous composite membrane (NCM). PVDF: poly(vinylidene fluoride); PAN: polyacrylonitrile.

Due to the fact that the prepared NCMs were used for piezoelectric sensors, the material should have a certain degree of hydrophobicity. The water contact angles (WCAs) of the NCMs are shown in Figure 2(d). From Figure 2(d), the WCA of the BaTiO₃@PAN/PVDF NCMs were higher than those of pure PAN/PVDF NCMs, which indicated that the introduction of BaTiO₃ can significantly improve the hydrophobicity of NCMs. Moreover, the WCAs of BaTiO₃@PAN/PVDF NCMs were above 90°, resulting in the better waterproofing.

Piezoelectricity

To characterize the piezoelectric properties, the NCMs were surrounded by a pair of copper electrodes on the top and bottom, and the outermost layer was a non-conductive flexible polyester film to prevent fiber damage²⁹ and improve stability during the piezoelectric test, according to the previous method,³⁰ as shown as in Figure 3(a). Figures 3(b)–(d) display the piezoelectricity of pure PAN/PVDF NCMs under different compression pressures, speeds and thicknesses of NCMs. From Figure 3(b), the highest output voltage of pure PAN/PVDF NCMs was 73.34 mV when the compression pressure was 5 N, and thus the perfect compression pressure of the piezoelectric test was selected to be 5 N. Figure 3(b) shows the piezoelectricity of pure PAN/PVDF NCMs at different compression speeds under the pressure of 5 N. In Figure 3(c), the output voltages of NCMs were 61.87, 57.73, 68.78, 73.34, and 63.35 mV. The maximum output voltage was 73.34 mV when the compression speed was 30 cm/min. As shown in Figure 3(d), the highest output voltage of NCMs with the thickness of 3 mm was 73.34 mV. Therefore, the best parameters in compression piezoelectric testing were a pressure of 5 N, a speed of 30 cm/min and a thickness of 3 mm.

Figure 3.

(a) The test equipment and sample. Piezoelectric properties of pure polyacrylonitrile (PAN)/poly(vinylidene fluoride) (PVDF) nanofibrous composite membranes (NCMs) under different pressures (b), speeds (c) and thicknesses (d). Piezoelectric properties of pure PAN/PVDF and the BaTiO₃@PAN/PVDF NCM under cyclic compression (e) and the drop-hammer (f) pattern. (g) Electrometer test and (h) Piezoelectric mechanism.

Due to the different force situations of people in motion in real life, cyclic compression and the drop-hammer pattern were conducted to characterize the piezoelectric properties of NCMs, as shown in Figures 3(e) and (f). From Figure 3(e), the output voltage of NCMs first increased and then decreased with the increase of the BaTiO₃ level. The main reason might be that excessive BaTiO₃ accumulated on the fiber surface and could not form a continuous and stable channel during the transmission of electrical signals. When added to 5%, BaTiO3@PAN/PVDF NCMs had the highest output voltage of 80.04 mV, which was 9.13% higher than the that of pure PAN/PVDF NCMs. In Figure 3(f), the piezoelectric properties of NCMs exhibit the same trend of change as the level of BaTiO₃ increased. Due to the limitation of the test equipment, the output voltages of NCMs with the best test parameters by directly connecting the electrometer were obtained and are shown in Figure 3(g). As can be seen from Figure 3(g), the 5BaTiO3@PAN/PVDF NCM had the highest output voltage of 2.613 V, which was 46.39% higher than the that of pure PAN/PVDF NCMs. Under the electrometer test conditions, the output voltages of NCMs increased significantly by nearly 10 times, but the change trend of the output voltage was the same.

Generally, if pressure is applied to a piezoelectric material, it will generate a potential difference, known as the positive piezoelectric effect. Conversely, applying voltage generates mechanical stress, known as the inverse piezoelectric effect. Therefore, piezoelectric materials have the function of converting and inverting mechanical and electrical energy. Figure 3(h) shows the piezoelectric mechanism of NCMs. When a NCM is subjected to a fixed direction of external force, internal polarization occurs and opposite charges are generated on two surfaces. After the external force is removed, the NCM returns to its uncharged state. Moreover, the polarity of the charge also changes with the direction of external force action, which represents the positive piezoelectric effect.

Theoretical and experimental validation by CCD

In order to optimize the preparation process of the 5BaTiO₃@PAN/PVDF NCM, the CCD was conducted using Design Expert software, aiming at three influential factors: spinning voltage (A), acceptance distance (B) and liquid supply rate (C). The average fiber diameter and output voltage were the response values, which are summarized in Table 1. The regression analysis of the data in Table 1 was conducted by using Design Expert and the multivariate quadratic regression equations of each factor on the average fiber diameter and output voltage of the response value were finally obtained:

\begin{array}{l} D = + 195.17 + 17.26 * A + 12.50 * B + 7.88 * C \\ + 7.12 * A * B + 1.24 * A * C - 0.95 * B * C \\ + 8.97 * A^{2} - 1.21 * B^{2} + 6.33 * C^{2} \end{array}

(1)

\begin{array}{l} E = + 89.20 + 1.19 * A + 0.89 * B - 0.69 * C \\ - 0.18 * A * B - 1.37 * A * C - 1.74 * B * C \end{array}

(2)

where A is the spinning voltage, B is the acceptance distance, C is the liquid supply rate, D is the average fiber diameter and E is the output voltage.

Table 1.

Experiments and results of the central combination design

No.	Voltage (kV)	Distance (cm)	Rate (mL/h)	Average fiber diameter (nm)	Output voltage (mV)
1	30.00	20.00	0.75	195.34	88.95
2	28.00	18.00	0.75	183.35	86.62
3	28.00	20.00	1.00	195.72	88.51
4	30.00	22.00	0.50	204.79	92.93
5	30.00	20.00	0.75	198.55	89.70
6	32.00	22.00	0.75	236.75	89.06
7	30.00	20.00	0.75	193.16	87.75
8	30.00	20.00	0.75	191.88	89.14
9	30.00	18.00	0.50	179.71	88.59
10	28.00	22.00	0.75	195.92	88.70
11	28.00	20.00	0.50	182.77	87.70
12	30.00	18.00	1.00	197.69	89.22
13	30.00	22.00	1.00	218.96	88.59
14	32.00	20.00	1.00	240.66	89.83
15	32.00	20.00	0.50	222.75	94.50
16	30.00	20.00	0.75	196.92	88.96
17	32.00	18.00	0.75	195.69	88.71

The variance analysis of multiple quadratic regression (Equations (1) and (2)) is given in Tables 2 and 3. In Table 2, the P-value of the model and the misfit term were 0.0002 (<0.05) and 0.0577 (>0.05), respectively. This indicated that the regression model was significant and effective. The signal-to-noise ratio was 16.480 (>4), which is an important index to measure the accuracy of the response value, resulting in significant discriminant ability. Similarly, Table 3 indicates that the model selection of this experiment was very appropriate, with significant discrimination ability, and could be used for theoretical prediction.

Table 2.

Regression model and analysis of the variance of average fiber diameter

Source	Sum of squares	Degree of freedom	Mean square	F-value	P-value
Model	4877.13	9	541.9	23.38	0.0002
A-Voltage (kV)	2383.61	1	2383.61	102.86	<0.0001
B-Distance (cm)	1249.5	1	1249.5	53.92	0.0002
C-Rate (mL/h)	496.28	1	496.28	21.42	0.0024
AB	202.92	1	202.92	8.76	0.0211
AC	6.15	1	6.15	0.27	0.6223
BC	3.63	1	3.63	0.16	0.7041
A ²	338.97	1	338.97	14.63	0.0065
B ²	6.22	1	6.22	0.27	0.6205
C ²	168.84	1	168.84	7.29	0.0307
Residual	162.22	7	23.17
Lack of fit	132.84	3	44.28	6.03	0.0577
Pure error	29.38	4	7.35
Cor total	5039.35	16
Adeq precision	16.480

Table 3.

Regression model and analysis of variance of output voltage

Source	Sum of squares	Degree of freedom	Mean square	F-value	P-value
Model	41.42	6	6.9	4.24	0.0221
A-Voltage (kV)	11.42	1	11.42	7.01	0.0244
B-Distance (cm)	6.38	1	6.38	3.91	0.0761
C-Rate (mL/h)	3.86	1	3.86	2.37	0.1547
AB	0.13	1	0.13	0.082	0.781
AC	7.5	1	7.5	4.6	0.0576
BC	12.13	1	12.13	7.44	0.0213
Residual	16.3	10	1.63
Lack of fit	14.27	6	2.38	4.69	0.0783
Pure error	2.03	4	0.51
Cor total	57.72	16
Adeq precision	6.835

In order to select the best comprehensive spinning parameters, response surface analysis of the average fiber diameter and output voltage were conducted, as shown as Figure 4. From Figures 4(a) and (d), under a constant liquid supply rate, the average fiber diameter and output voltage were enhanced with the increase of spinning voltage, and they reduced with the decrease of spinning distance. In Figures 4(b) and (e), at a certain spinning distance, the average fiber diameter and output voltage increased with the increase of spinning voltage. With the decrease of liquid supply rate, the average fiber diameter became smaller and the output voltage became out higher. This was because the reduction of spinning distance could easily mean NCMs are not formed. While the liquid supply rate was reduced, and it was difficult for drips to form. Under the condition that the forming was guaranteed and the liquid was not dripping, the increase of voltage helped the nanofibers to form uniformly, so that the piezoelectric properties of the NCMs were improved. Figures 4(c) and (f) show the relationship between the predicted and actual values of the average fiber diameter and output voltage, respectively. It could be found that the predicted value of the yield was very similar to the actual value, which proved a high match between the theoretical model and the actual situation.

Figure 4.

(a), (b) Response surface value of average fiber diameter with the spinning voltage, acceptance distance and liquid supply rate. (c) Comparison of predicted and actual average fiber diameter. (d), (e) Response surface value of output voltage with the spinning voltage, acceptance distance and liquid supply rate and (f) Comparison of predicted and actual output voltage.

According to the results of the CCD for obtaining a good piezoelectric and structurally stable NCM, the spinning parameters should be the spinning voltage of 30 kV, the acceptance distance of 22 cm and the liquid supply rate of 0.5 mL/h. In order to verify the accuracy of the theoretical value, the average fiber diameter and output voltage of the 5BaTiO₃@PAN/PVDF NCM under the optimal experimental conditions were compared with the experimental values without optimization and theoretical results, as shown in Figure 5. According to Equations (1) and (2), the predicted average fiber diameter and output voltage should be 200.2 nm and 93 mV, respectively. From Figure 5, the optimized experimental value of NCMs was 192 nm with the relative error of 10.1%, and 91 mV with the relative error of 3.3%, respectively. This indicated that the model could fit well with theoretical model and meet the experimental requirements. Furthermore, after optimizing the experimental conditions, the average fiber diameter of NCMs decreased by 20% from 240 to 192 nm, and the output voltage increased by 21% from 75 to 91 mV. With the help of the CCD, the average fiber diameter of NCMs was thinner and more uniform, resulting in improving the output voltage and piezoelectric sensitivity of the NCMs. Therefore, the piezoelectric properties of NCMs were significantly enhanced and relatively stable after the CCD.

Figure 5.

Average fiber diameter (a) and output voltage (b) of the 5BaTiO₃@polyacrylonitrile/poly(vinylidene fluoride) nanofibrous composite membrane.

Conclusions

In summary, modified BaTiO₃@PAN/PVDF NCMs were prepared by cross-electrospinning technology. The effects of the addition of BaTiO₃ at different levels on the fiber morphology and distribution, crystal shape, hydrophobicity, mechanical properties and piezoelectric properties of NCMs were investigated. The results showed that the mechanical properties and piezoelectricity of the 5BaTiO₃@PAN/PVDF NCM could be improved by over 40%. The effects of three spinning parameters on the piezoelectric properties of NCMs were recorded and analyzed by the CCD, and the optimum process parameters for the preparation of the 5BaTiO₃@PAN/PVDF NCM were determined: voltage of 30 kV, distance of 22 cm, rate of 0.5 mL/h. The piezoelectric performance of the optimized 5BaTiO₃@PAN/PVDF NCM was more stable and improved by 46%. Therefore, we believe that the modified BaTiO₃@PAN/PVDF NCM combined with cross-electrospinning technology and CCD can facilitate the development of flexible wearable piezoelectric sensors. anionic-based conductive organic nanofibrous sensor with high flexibility, high adhesion, recycling ability and good self-power supply performance under mechanical action was fabricated, which would expand the application scope of intelligent wearable devices, human motion and health monitoring devices and medical testing and flexible electronic devices.

Footnotes

Data availability

Data will be made available on request.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China (NSFC, No.51973168) and the Wuhan Yellow Crane Talents Program (Grant No. [2022]734).

ORCID iD

Bo Yang

References

Dong

Peng

Wang

ZL.

Fiber/fabric-based piezoelectric and triboelectric nanogenerators for flexible/stretchable and wearable electronics and artificial intelligence. Adv Mater 2020; 32: 1902549.

Mercante

Andre

Facure

MHM

, et al. Recent progress in conductive electrospun materials for flexible electronics: energy, sensing, and electromagnetic shielding applications. Chem Eng J 2023; 465: 142847.

Mason

WP.

Piezoelectricity, its history and applications. J Acoust Soc Am 1981; 70: 1561–1566.

Roberts

Dielectric and Piezoelectric Properties of Barium Titanate

Physical Review 1947; 71: 890–895. DOI: 10.1103/PhysRev.71.890.

Kawai

The piezoelectricity of poly (vinylidene fluoride). Jpn J Appl Phys 1969; 8: 975.

Ding

Liu

, et al. Flexible PVDF based piezoelectric nanogenerators. Nano Energ 2020; 78: 105251.

Wang

Zheng

Jin

, et al. Unexpectedly high piezoelectricity of electrospun polyacrylonitrile nanofiber membranes. Nano Energ 2019; 56: 588–594.

Tao

Wang

Zheng

, et al. A review: polyacrylonitrile as high-performance piezoelectric materials. Nano Energ 2023; 118: 108987.

Bhatt

Singh

Bamola

, et al. Enhanced piezoelectric response using TiO2/MoS2 heterostructure nanofillers in PVDF based nanogenerators. J Alloy Comp 2023; 960: 170664.

10.

Purushothaman

Tronco

Kottathodi

, et al. A review on electrospun PVDF-based nanocomposites: recent trends and developments in energy harvesting and sensing applications. Polymer 2023; 283: 126179.

11.

Uslu

Gavgali

Erdal

, et al. Determination of mechanical properties of polymer matrix composites reinforced with electrospinning N66, PAN, PVA and PVC nanofibers: a comparative study. Mater Tod Commun 2021; 26: 101939.

12.

Cheng

Jung

Heo

, et al. Enhanced piezoelectric performance of ZnSnO3@PVDF composite films by control of embedded contents of ZnSnO3 nanoparticles. Surf Interf 2023; 41: 103177.

13.

Bakhtar

Abdoos

Rashidi

A review on fabrication and in vivo applications of piezoelectric nanocomposites for energy harvesting. J Taiwan Inst Chem Eng 2023; 148: 104651.

14.

Savakus

Klicker

Newnham

RE.

PZT-epoxy piezoelectric transducers: a simplified fabrication procedure. Mater Res Bull 1981; 16: 677–680.

15.

Mezheritsky

Ultrasonic transceiver with a regular/periodic 1–3 piezocomposite based on the SAW resonance mode on damping backing. Acoustics 2020; 2: 110–127.

16.

Liao

Loh

C-H

Tian

, et al. Progress in electrospun polymeric nanofibrous membranes for water treatment: fabrication, modification and applications. Prog Polym Sci 2018; 77: 69–94.

17.

Zou

Wang

, et al. Dynamic hydrophobicity of superhydrophobic PTFE-SiO2 electrospun fibrous membranes. J Membr Sci 2021; 619: 118810.

18.

Xue

Dai

, et al. Electrospinning and electrospun nanofibers: methods, materials, and applications. Chem Rev 2019; 119: 5298–5415.

19.

Zou

Sun

Wang

Elastic, hydrophilic and biodegradable poly (1, 8-octanediol-co-citric acid)/polylactic acid nanofibrous membranes for potential wound dressing applications. Polym Degrad Stabil 2019; 166: 163–173.

20.

Xue

Xie

Liu

, et al. Electrospun nanofibers: new concepts, materials, and applications. Acc Chem Res 2017; 50: 1976–1987.

21.

Démuth

Farkas

Pataki

, et al. Detailed stability investigation of amorphous solid dispersions prepared by single-needle and high speed electrospinning. Int J Pharmaceut 2016; 498: 234–244.

22.

Shen

Peng

, et al. Preparation of single-ion conductor solid polymer electrolyte by multi-nozzle electrospinning process for lithium-ion batteries. J Phys Chem Solids 2021; 158: 110229.

23.

Varesano

Carletto

Mazzuchetti

Experimental investigations on the multi-jet electrospinning process. J Mater Proc Technol 2009; 209: 5178–5185.

24.

Yanbo

Xinyu

Yunxia

, et al. Cross-electrospun PVDF/PVDF-HFP nanofibrous membrane with central combination design and its waterproof and moisture permeable composite fabric. Text Res J 2022; 93: 2273–2289.

25.

Huang

Cao

, et al. Orthogonal experimental preparation of Sanguis Draconis- polyvinylpyrrolidone microfibers by electrospinning. J Biomater Sci Polym Ed 2019; 30: 308–321.

26.

Cui

Zhou

, et al. Investigation on process parameters of electrospinning system through orthogonal experimental design. J Appl Polym Sci 2007; 103: 3105–3112.

27.

Ameen

Mathivanan

Zhang

, et al. One factor at a time and two-factor optimization of transesterification parameters through central composite design (CCD) for the conversion of used peanut oil (UPNO) to biodiesel. Fuel 2023; 352: 129065.

28.

Sharma

Pattnaik

Behera

Development and in-vitro, in-vivo evaluation of Pioglitazone-loaded polymeric nanoparticles using central composite design surface response methodology. OpenNano 2023; 11: 100141.

29.

Kang

Liu

, et al. Polyvinylidene fluoride piezoelectric yarn for real-time damage monitoring of advanced 3D textile composites. Compos B Eng 2022; 245: 110229.

30.

Boettcher

Mills

Swerdloff

, et al. Auditory evoked potentials in aged gerbils: responses elicited by noises separated by a silent gap. Hear Res 1996; 102: 167–178.

A BaTiO 3 @polyacrylonitrile/poly(vinylidene fluoride) nanofibrous composite membrane with high piezoelectricity based on the central combination design method and cross-electrospinning technology

Abstract

Keywords

Experimental details

Materials

Modification of BaTiO3

Preparation of cross-electrospinning solutions

Preparation of cross-electrospun BaTiO3@PAN/PVDF NCMs

CCD experiment

Characterization

Results and discussion

Morphology and structure

Mechanical property and hydrophobicity

Piezoelectricity

Theoretical and experimental validation by CCD

Conclusions

Footnotes

Data availability

Declaration of conflicting interests

Funding

ORCID iD

References

Modification of BaTiO₃