Fabrication of piezoelectric fiber composites by the double cutting–filling method and properties characterisation

Abstract

Fine-scaled piezoelectric fiber composites (PFCs) were successfully fabricated by the so-called double cutting–filling method developed in this research. The fabricated PFCs consisted of interdigitated electrodes, epoxy resin and piezoelectric fibers. The microstructure, piezoelectric properties, mechanical properties, free strain and sensing properties of the fabricated PFCs were characterized. It has been found that the fabricated PFCs with an active area of 50 × 30 mm² exhibited good coupling between the electrodes and fibers, which is highly required. The relative dielectric constant (ε_r) and dielectric loss (tan δ) of the fabricated PFCs were 1961 and 0.054, respectively. The tensile strength reached 49.28 MPa and 36.18 MPa in the longitudinal and transverse directions, respectively, indicating that high and comparable tensile strength was achieved in both directions. The longitudinal free strain of the PFCs reached about 1400 με and the transverse free strain 680 με at a 2 kV peak-to-peak sinusoidal alternating voltage, suggesting that the PFCs exhibited good orthotropic properties, as desired. The longitudinal piezoelectric coefficient (d₃₃) of the PFCs reached 518 pC N⁻¹. The fabricated PFCs also exhibited desirable anisotropic sensing properties which can be utilized to quantify the origin of excitation in damage detection of composites. The PFCs developed in this study featured excellent piezoelectric and mechanical properties, high free strain, desirable orthotropic mechanical properties and excellent anisotropic sensing properties, which helps to extend their applications in self-diagnosis and adaptive smart structures.

---

1. Introduction

Piezoelectric fiber composites (PFCs) have unique thin layered structures, consisting of unidirectional piezoceramic fibers embedded in an epoxy resin matrix and sandwiched between interdigitated electrodes.¹ The PFCs can overcome the inherent brittleness of piezoceramics and the temperature limitation of polymers. They present desirable features such as being lightweight and flexible with high strength and toughness, and exhibit high actuation capability and high sensitivity.^2,3 Owing to these advantages, PFCs have received wider attention in smart applications, such as structure control,⁴ structural health monitoring,⁵ energy harvesting⁶ and flight control through active wind design.⁷

The advantages of PFCs motivate researchers worldwide to fabricate and optimize smart composites for manufacturing PFCs, among which active fiber composites (AFCs) and macro fiber composites (MFCs) have been verified as very promising composites for such purposes so far. AFCs were developed by the Active Materials and Structures Laboratory at MIT,⁸ in which the interdigitated electrodes were used to improve the efficiency of electromechanical conversion of the composites which exploited the d₃₃ piezoelectric effect along the longitudinal direction of PZT fibers.⁹ Based on this, NASA Langley Research Center¹⁰ developed MFCs with rectangular cross-sectional PZT fibers, which increased the contact area between the rectangular piezoceramic fibers and electrode fingers to ensure the most efficient transfer of electric field into the fibers.^11,12 However, researchers prepared piezoceramic wafer using a tape casting method and then fabricated rectangular fibers by using a dicing method during the preparation of composites. However, it has been found that it was difficult to sinter and ensure the smoothness of thin piezoceramic wafers with larger area using that method.¹³

In this study, a new fabrication method, called the double cutting–filling method, was developed for PFCs. In this method, first piezoceramic wafers were cut from a piezoceramic block with good smoothness and large area using a multi-wire cutting machine. Then the piezoceramic wafer was cut into fibers with a rectangular cross-section. Epoxy resin was utilized to coat electrodes and fill in the space between piezoceramic fibers arranged in parallel. After that they were packaged and cured to form the PFCs. Finally the fabricated PFCs were characterized and their piezoelectric, mechanical and sensing properties as well as free strain were figured out and/or compared with those of comparable PFCs reported in literature.

2. Experimental

2.1 Sample preparation

In this study, piezoceramic PZT-51 (Zibo Yuhai Electronic Ceramic Co. Ltd), interdigitated electrodes (Shanghai Cengtuo Electronics Technology Co., Ltd) and epoxy resin matrix (Beijing Jinhong Century Technology Co., Ltd) were selected as the raw materials for fabrication of PFCs.¹⁴

First, non-polarized PZT-51 piezoceramic blocks were fabricated by the solid phase reaction method. Then the piezoceramic block was cut using a multi-wire cutting machine to make piezoceramic wafers. The fabricated piezoceramic wafers (as shown in Fig. 1) were further cut into piezoceramic fibers with rectangular cross-sections using a precision dicing saw. This is why the “double cutting” is called in the double cutting–filling method developed in this study. Next, a right amount of epoxy resin which features good fluidity and short curing time was applied on the surfaces of electrodes and filled the space between the piezoceramic fibers to fabricate the PFCs. Therefore, a part of epoxy resin was used to fill the gap of 100 μm between fibers, another part of epoxy acted as the intermediate bonding layer of 2 μm between interdigitated electrodes and fibers. It is worth noting that the epoxy used in this study with good fluidity was able to completely fill the space between the piezoceramic fibers and there were no bubbles found existing between the fibers or the thin layer between the interdigitated electrodes and the piezoceramic fibers. Then the vacuum hot pressing process was utilized to cure the epoxy resin. Next, the piezoceramic fibers, epoxy resin and interdigitated electrodes were packaged together to form the PFCs. Noting that the longitudinal direction of the piezoceramic fibers was perpendicular to the electrode fingers. The packaging process was carried out under vacuum for 3 hours at 60 °C under the environmental pressure about 1.5 MPa. Finally, polarization of the PFCs were performed with the high DC voltage of 3 kV mm⁻¹ in methyl silicone oil for 10 minutes at room temperature along the longitudinal direction of the piezoceramic fibers. Fig. 2 shows the fabricated PFC specimens with the 1–2 coordinate system, i.e. the 1-direction was the longitudinal direction of piezoceramic fibers and the 2-direction was perpendicular to the 1-direction (i.e., parallel to the direction of electrode fingers). The processing flow for fabrication of the PFCs is presented in Fig. 3.

Fig. 1 Piezoceramic wafers cut from piezoceramic block.

Fig. 2 PFC specimen with the 1–2 coordinate system.

Fig. 3 Processing flow for fabrication of PFCs.

2.2 Characterization

The dimensions of the fabricated PFCs were measured by a metallurgical microscope (BA310, Motic China Group Co., Ltd, China). The microstructures of the cross-sectional PFCs were examined by a scanning electron microscope (FEG 250, FEI Quanta Co., Ltd, United States). The piezoelectric properties were measured and calculated by a precision impedance analyzer (Agilent 4294A, Agilent Technologies Co. Ltd, United States). The properties analyzed include the static capacitance and dielectric loss (tan δ) at 1 kHz and the relative dielectric constant (ε_r). Tensile test of the PFC specimens in the longitudinal and transverse directions of the embedded piezoceramic fibers was performed using a universal testing machine (CMT 5504, Shenzhen Sans Material Test Instrument Co., Ltd, China) under a displacement control mode with a rate of 1 mm min⁻¹. The ASTM standard, D 3039/D 3039M-00 specified testing and data analysis methodology was utilized to conduct the tensile test.¹⁵ The first series of tensile tests loaded the PFC specimens in the longitudinal direction of the piezoceramic fibers (i.e. along 1-direction). The second series of tensile tests loaded the PFC specimens in the direction of the electrode fingers (i.e. along the 2-direction). The tabs near the ends of the PFC specimens were used for mounting the specimens into the grips of the universal testing machine. A silicone pad and abrasive paper were inserted between the PFC specimens and the grips to prevent slipping during testing. The free strain of the composites was measured by a dynamic strain acquisition system (Jinan Sigma Technology Co., Ltd, China). The longitudinal piezoelectric coefficient (d₃₃) of PFCs was calculated based on the longitudinal free strain and the applied voltages using the E–S method. The sensing properties of the fabricated PFCs were measured by an acoustic emission system (Physical Acoustic Corp., America).

3. Results and discussions

The dimensions of the fabricated PFCs are listed in Table 1. It can be seen that the active area of the fabricated PFCs reached 50 × 30 mm² and the piezoceramic fiber thickness was smaller than 300 μm. Compared with the macro piezoceramic fiber composites fabricated using the cutting–filling method by Wen et al.,¹³ the PFCs fabricated in this study had finer scales and larger effective area and were lighter as desirable for various applications.

Table 1 Dimensions of the fabricated piezoelectric fiber composites

Property	Dimension
Piezoceramic fiber width, wf/μm	300
Piezoceramic fiber thickness, tf/μm	280
Piezoceramic fiber spacing, kf/μm	100
Interdigitated electrode finger width, we/μm	75
Interdigitated electrode gap, center to center, pe/mm	0.75
Interdigitated electrode thickness, te/μm	20
Sample active area width, wpfc/mm	30
Sample active area length, lpfc/mm	50
Sample maximum thickness, tpfc/μm	360

---

3.1 Microstructure

Fig. 4(a) shows the metallographic micrograph of the precisely cut piezoceramic fibers with rectangular cross-sections which were arranged in parallel in the resin matrix. It can be seen from this SEM image that the width of piezoceramic fiber and spacing between fibers were uniform and they were accurate as designed. Fig. 4(b) and (c) present the cross-sectional micrograph of the assembled PFCs, which was examined from the fracture surface of PFCs along the transverse direction of the piezoceramic fibers. The bright area was the piezoceramic fibers while the dark area was the epoxy resin. The SEM micrographs showed that the interface among fiber and epoxy resin were well compacted where scarcely existed visible gaps and holes, which proved the vacuum treatment was necessary and had a good effect during PFCs' assembling procedure. And, there were no cracks in fibers which indicated the applied pressure was eligible and further ensured PFCs exhibited the optimal performance. It can be concluded that good coupling among the piezoceramic fibers, epoxy resin and interdigitated electrodes was achieved, which was very important to ensure the integrity and good performance of PFCs fabricated.

Fig. 4 (a) Metallographic micrograph of the PZT fibers (b) and (c) SEM micrographs of the cross-section of PFCs.

3.2 Free strain

Fig. 5 shows the free strain of the PFCs, which was obtained under a sinusoidal alternating voltage with the range of −500 V to +1500 V tested in the electric field with the frequency of 0.1 Hz. From the strain curve, it can be seen that a nonlinear hysteretic relationship existed between free strain of PFCs and the applied voltage. The longitudinal free strain of PFCs reached about 1400 με while the transversal free strain reached about 680 με. The results indicated that the PFCs exhibited well orthotropic property, i.e. significantly different free strain thus piezoelectric properties in the longitudinal and transverse directions. As the longitudinal piezoelectric coefficient d₃₃ was exploited via the interdigitated electrodes in the PFCs along the longitudinal direction of the piezoceramic fibers, it was bigger than the transverse piezoelectric coefficient d₃₁ as expected.

Fig. 5 Free strain of PFCs under a sinusoidal alternating voltage with the range of −500 V to +1500 V at 0.1 Hz.

3.3 Dielectric and piezoelectric properties

The relative dielectric constant (ε_r) of the PFCs was calculated from the measured static capacitance using eqn (1) as following,where C is the measured static capacitance of the PFCs at 1 kHz, t and A are the thickness and the active area of the composite, and ε₀ is the permittivity of vacuum (in this case ε₀ = 0.08854 pF cm⁻¹).

The measured dimensions and calculated dielectric properties of the PFCs are listed in Table 2.

Table 2 Dielectric properties of the PFCs

Parameter	t/(mm)	A/(mm)	εr	tan δ/%
Value	0.36	1500	1961	5.4

---

The longitudinal piezoelectric coefficient (d₃₃) of the PFCs was calculated from the longitudinal free strain and the applied voltages by the E–S method, using eqn (2) as following according to the inverse piezoelectric effect.^16,17

where S₃ is the peak-to-peak free strain in the longitudinal direction and E₃ is the amplitude of the electric field.

As the gap between the interdigitated electrode fingers was 0.75 mm and the longitudinal free strain of the composite was 1400 με at the peak-to-peak voltage 2000 V (i.e. 1500 − (−500) = 2000 V), the longitudinal piezoelectric coefficient (d₃₃) of the PFCs was 518 pC N⁻¹ as calculated from eqn (2) with the voltage amplitude of 2000 V at 0.1 Hz. By contrast, the d₃₃ coefficient of PFCs obtained by Williams was 451 pC N⁻¹ at 2000 V.¹⁸

3.4 Tensile strength

Fig. 6 shows the tensile stress–displacement curves for the PFCs in both the longitudinal and transverse directions. From the tensile stress–displacement curves, it can be seen that the PFCs exhibited linear elastic behavior during initial deformation stage in which tensile stress increased linearly with respect to displacement. After the elastic limit was reached, the PFCs exhibited nonlinear behavior featured slower increasing in stress with respect to strain until the fracture limit was reached upon which the PFCs were broken. The tensile behaviors of the PFCs were similar to those of the PFCs fabricated by Williams et al.¹⁵ Excellent flexibility and high tensile strength exhibited by the fabricated PFCs is essential to extend their application in new areas such as structures with curved or irregularly shaped surfaces.

Fig. 6 Tensile stress–displacement curves for PFCs.

The tensile behaviors of the PFCs could be associated with fiber pull-out phenomenon. Furthermore, it is postulated that weak interfacial adhesion between fiber and matrix could result in fiber pull-out process and plastic behavior of material. Fig. 7 shows the SEM micrograph of the cross-section of PFCs after tensile test. Hereinbefore, the Fig. 4(b) of SEM micrograph showed that the interface did not exist obvious defects, which only guarantee the interface could contact with each other as much as possible and generate good mechanical adhesion. The interdigitated electrode and epoxy resin have the characteristic behavior of polymer, e.g., higher tensile and plastic, compared to piezoceramic fiber. At the same time, the internal stress exists in the interface due to the difference of thermal expansion coefficient of fiber and epoxy resin during the resin curing process. Therefore, all those factors induced the stress concentrated and fracture generated in the piezoceramic fibers and interfaces. Fig. 7 also shows that both the fiber pull-out phenomenon and interfacial debonding, which further proved the flexibility of the deduction.

Fig. 7 SEM micrograph of the cross-section of PFCs after tensile test.

The tensile strength was calculated according to eqn (3).

where σ_s is the tensile strength, F_max is the maximum tensile load obtained from experiment, A is the area of the specimen bonding surface, a and b are the width and length of the bonding surface, respectively.

The measured parameters and tensile strength in the longitudinal and transverse directions of the PFCs are listed in Table 3. By contrast, the maximum tensile strength of PFCs obtained by Williams was 45 MPa.¹⁸ It can be seen that the tensile strength along the longitudinal direction (i.e. along the 1-direction) was comparable to that along the transverse direction (i.e. along the 2-direction) which was highly desirable.

Table 3 Measured parameters and tensile strength of PFCs

Test type	Length (mm)	Width (mm)	Thickness (mm)	Cross section area (mm^2)	Fmax (N)	σs (MPa)
1-Direction test	50	30	0.36	10.8	532.2	49.28
2-Direction test	30	30	0.36	10.8	390.8	36.18

---

3.5 Sensing properties

Fig. 8 shows the amplitudes and voltages of acoustic emission (AE) signals received by various piezoelectric materials in this case PZT, 1–3 piezoelectric composites and the PFCs fabricated in this study. It can be seen that the amplitudes and voltages of the AE signals received by the fabricated PFCs varied with the excitation angle in roughly a cosine mode. In comparison, the amplitude and voltage of the AE signal received by the PZT and the 1–3 piezoelectric composites remained almost constant when excited at different angles. These results indicated that the PFCs exhibited anisotropic sensing properties, which can potentially be utilized in quantifying the amplitude and angle of stress waves generated by, for instance, fracture of materials in damage detection of composites and composite structures. In other words, the fabricated PFCs in this study are able to quantify the position of the origin of excitation. In comparison, the PZT and the 1–3 piezoelectric composites can only detect occurrence of excitation but cannot quantify the origin of excitation.

Fig. 8 (a) Amplitudes and (b) voltages of acoustic emission signals received by various piezoelectric materials.

4. Conclusions

In this study, fine-scaled PFCs with active area of 50 × 30 mm² were successfully fabricated by the double cutting–filling method proposed. The combination of the PZT fibers and epoxy resin was found to be very good, which ensured high integrity and good mechanical performance of the fabricated PFCs. Various properties of the fabricated PFCs were characterized. It has been found that:

(1) The relative dielectric constant (ε_r) and dielectric loss (tan δ) of the fabricated PFCs were 1961 and 0.054;

(2) The tensile strength of the fabricated PFCs reached 49.28 MPa and 36.18 MPa in the longitudinal and transverse direction, respectively, indicating high and comparable tensile strength achieved in both directions;

(3) The longitudinal free strain of the PFCs reached about 1400 με and the transverse free strain 680 με at a 2 kV peak-to-peak alternating voltage, suggesting that the PFCs exhibited good orthotropic property;

(4) The longitudinal piezoelectric coefficient (d₃₃) of the PFCs reached 518 pC N⁻¹;

(5) The fabricated PFCs exhibited desirable anisotropic sensing properties, suggesting that they might be able to quantify the origin of damage in composites.

It can thus be able to conclude that the PFCs developed in this study featured excellent piezoelectric properties, high free strain, high tensile strength, desirable orthotropic mechanical properties and excellent anisotropic sensing properties, which helps to extend their applications in self-diagnosis, damage detection of composites and adaptive smart structures.

Acknowledgements

This work presented in this paper is supported by the National Nature Science Foundation of China under the grant of 51172097, Shandong Provincial Natural Science Foundation under the grant of ZR2014EMP002, National High Technology Research and Development Program (“863 Program”) under the grant of 2015AA034701 and Program for Scientific Research Innovation Team in Colleges and Universities of Shandong Province.

References

1. C. R. Bowen, L. J. Nelson, R. Stevens, M. G. Cain and M. Stewart, Optimisation of interdigitated electrodes for piezoelectric actuators and active fibre composites, J. Electroceram., 2006, 16(4), 263–269.
2. X. J. Lin, D. Zhang and X. Y. Zhang, Modeling and optimization of piezoelectric fiber composites based on finite element method, Chin. J. of Nonferrous Met., 2013, 22(6), 1748–1753.
3. X. J. Lin, K. C. Zhou, X. Y. Zhang and D. Zhang, Development, modeling and application of piezoelectric fiber composites, Trans. Nonferrous Met. Soc. China, 2013, 23(1), 98–107.
4. H. S. Kim, J. W. Sohn and W. K. Wilkie, Vibration control of a cylindrical shell structure using macro fiber composite actuators, Mech. Base. Des. Struct. Mach., 2011, 39(4), 491–506.
5. A. J. Brunner, The potential of active fiber composites made from piezoelectric fibers for actuating and sensing applications in structural health monitoring, Mater. Struct., 2005, 38(279), 561–567.
6. X. B. Shan, R. J. Song, B. Liu and T. Xie, Novel energy harvesting: a macro fiber composite piezoelectric energy harvester in the water vortex, Ceram. Int., 2015, 41(1), S763–S767.
7. R. Paradies and P. Ciresa, Active wing design with integrated flight control using piezoelectric macro fiber composites, Smart Mater. Struct., 2009, 18(3), 1–9.
8. A. A. Bent and N. W. Hagood, Piezoelectric fiber composites with interdigitated electrodes, J. Intell. Mater. Syst. Struct., 1997, 8(11), 903–919.
9. G. A. Rossetti, A. Pizzochero and A. A. Bent, Recent advances in active fiber composites technology, Applications of Ferroelectrics, 2000, vol. 2, pp. 73–756.
10. W. K. Wilkie, R. G. Bryant, J. W. High and R. L. Fox, Low-cost piezocomposite actuator for structural control applications, Industrial and Commercial Applications of Smart Structures Technologies, 2000, pp. 323–324.
11. R. B. Willams, B. W. Grimsley and D. J. Inman, Manufacturing and mechanics-based characterization of macro fiber composite actuators, Proceedings of IMECE, ASME International Mechanical Engineering Congress & Exposition, 2002, pp. 17–22.
12. L. J. Nelson, Smart piezoelectric fiber composites, Mater. Sci. Technol., 2002, 18(11), 1245–1256.
13. K. Wen, J. H. Qiu, H. L. Ji and K. J. Zhu, Fabrication of macro piezoelectric fiber composite actuators by cutting–filling method, J. Mater. Eng., 2015, 43(1), 72–76.
14. S. C. Li, K. J. Zhu, J. H. Qiu and X. M. Pang, Fabrication of flexible piezoelectric fiber composites actuator by arrangement-casting method, J. Inorg. Mater., 2013, 28(3), 331–335.
15. R. B. Williams, D. J. Inman and M. R. Schultz, Nonlinear tensile and shear behavior of macro fiber composite actuators, J. Compos. Mater., 2004, 38(10), 855–869.
16. R. B. Williams, D. J. Inman and W. K. Wilkie, Nonlinear response of the macro fiber composites actuator to monotonically increasing excitation voltage, J. Intell. Mater. Syst. Struct., 2006, 17(17), 601–608.
17. A. Deraemaeker, N. Nasser and A. Benjeddou, Mixing rules for the piezoelectric properties of macro fiber composites, J. Intell. Mater. Syst. Struct., 2009, 20(12), 1475–1482.
18. R. B. Williams, Nonlinear mechanical and actuation characterization of piezoceramic fiber composites, Virginia Polytechnic Institute and State University, Virginia, 2004, pp. 160–164.

---