Ultrasensitive Flexible Magnetoelectric Sensor
Ever-evolving advances in flexible magnetic sensors are promising to fuel technological developments in the fields of touchless human–machine interaction, implantable medical diagnosis, and magnetoreception for artificial intelligence. However, the realization of highly flexible and extremely sensitive magnetic sensors remains a challenge. Here, we report a cost-effective, flexible, and ultra-sensitive heterostructural magnetoelectric (ME) sensor consisting of piezoelectric Pb(Zr0.52Ti0.48)O3 (PZT) thick films and Metglas foils. The flexible sensor exhibits a strong ME coefficient of 19.3 V cm−1 Oe−1 at low frequencies and 280.5 V cm−1 Oe−1 at resonance due to the exceptionally high piezoelectric coefficient d33 ∼ 72 pC N−1 of the constituent PZT thick films. The flexible ME sensor possesses not only ultrahigh sensitivities of 200 nT at low frequencies and 200 pT at resonance but also shows an excellent mechanical endurance. Through 5000 bending cycles (radii of ∼1 cm), the sensors showed no fatigue-induced performance degradation. This ultrasensitive flexible sensor provides a platform capable of sensing and responding to external magnetic fields and will find applications in soft robotics, wearable healthcare monitoring, and consumer electronics.
Flexible electronics have underpinned many technological innovations in the fields of paper-like rollable displays, wearable sensors, energy conversion devices, and ergonomic health monitoring systems.1–7 Emergent flexible magnetic sensors are poised to scaffold developments in promising applications in touchless human–machine interaction elements, flexible navigation modules for next generation consumer electronics, structural health condition monitoring for naturally curved electrical motors, and bio-inspired magnetoreception for human or soft artificial intelligence.8 To realize these promising applications, flexible magnetic sensors with performance characteristics commensurate with their rigid counterparts must be developed. In principle, there are several different approaches to realize such performance requirements, including deposition of ferromagnetic materials on giant magnetoresistive (GMR) flexible organic substrates,8–10 giant magnetoimpedance (GMI),11 anisotropic magnetoresistance (AMR),12 Hall effect coupled devices,13,14 magnetic resonance response organic diode,15 and magnetic-functionalized suspended gate based organic field effect transistors.16 Notably, an AMR-based flexible sensor with an ultrahigh sensitivity of 150 nT at low frequency was recently demonstrated. All flexible magnetic sensors, regardless of underlying technology, still suffer from high cost, low sensitivity, or high power consumption.17
Multiferroic magnetoelectric (ME) systems represent an appealing class of multifunctional materials that exhibit both ferromagnetic and ferroelectric orders simultaneously18,19 The ability of ME materials to control ferroelectric polarization by a magnetic field, and/or conversely manipulation of the magnetization by an electric field enables promising applications in passive magnetic field sensors and electric-write magnetic-read memory devices.20–24 Since no single-phase material has been reported with a practical intrinsic ME coefficient, the strain-mediated extrinsic ME effect in composites composed of magnetostrictive and piezoelectric components has been exploited23,24 through technologies ranging from laminated bulks to nano-structured thin films.20,25,26 Such devices are attractive for magnetic field sensors that could replace low-temperature superconducting quantum interference devices (SQUIDs).22,27 Furthermore, the ME sensor has notable advantages of low cost, light weight, high sensitivity, low power consumption, and room temperature operation.17
The extrinsic magnetoelectric properties of composite heterostructures can be realized by judiciously selecting both magnetostrictive and piezoelectric components and optimizing their interface.28,29 These heterostructure are limited by the piezomagnetic and piezoelectric properties of the constituent components.20 A technological milestone in the development of ME composites was the reporting of a giant ME coefficient αE of 52 V cm−1 Oe−1 and an extremely low equivalent magnetic noise of 5.1 pT Hz−1/2 at 1 Hz for an ME heterostructure based on flexible magnetostrictive Metglas alloys and Pb(Mg1/3Nb2/3)-PbTiO3 (PMN-PT) single crystals. These devices owe their performance largely to the giant piezomagnetic coefficient of Metglas (arising principally from a huge magnetic permeability) and the large piezoelectricity of the PMN-PT layers.30
The magnetostrictive Metglas foil exhibits not only giant piezomagnetic coefficient but also excellent mechanical flexibility [see Fig. 1(a)], which has been employed in electronic-skin sensors.31 However, most state-of-the-art high-performance piezoelectric materials are perovskite-structured oxides, which have brittle and rigid features due to their small maximum elastic strain.32 Consequently, the most severe limitation to the development of ultrasensitive flexible ME sensors arises from the intrinsic nature of brittleness and rigidity of high-performance piezoelectric components. To address this challenge, several different strategies have been attempted in the literature, such as organic piezoelectric materials epoxied with flexible magnetostrictive foils,33–36 piezoelectric thin film growth on flexible magnetic metal substrates,37 and core–shell structured nano-particle composites.38 Systems based on flexible poly(vinylidene fluoride) (PVDF) and its copolymers are considered to be the most popular approach due to not only their considerable ME response but also their low cost and ease of manufacture. Most recently, Zong et al. proposed an organic biopolymer cellulose based magnetoelectric composite with excellent flexibility, albeit with a low ME coefficient on the order of 1.5 V cm−1 Oe−1 at low frequencies.39 Palneedi et al. obtained a near-theoretical ME coefficient of 7 V cm−1 Oe−1 in longitudinal-transverse mode PZT/Metglas laminates by granule spray deposition and laser radiation annealing techniques,29 but the ME coefficient dramatically decreased as the PZT film thickness was decreased to increase mechanical flexibility.
To date, though many of the techniques reported in the literature show significant promise toward flexible ME laminate sensors, ME heterostructure laminates exhibit either low ME coefficient, low mechanical compliance, or both. The single most limiting factor in the realization of a flexible ME sensor lies in the nature of ceramic, perovskite-based piezoelectric components.
Here, we introduce a strategy that allows us to achieve highly flexible heterostructural ME magnetic field sensors by employing all-inorganic flexible piezoelectric PZT films, which are enabled by two-dimensional mica substrates. A large ME coefficient of 19.3 V cm−1 Oe−1 and an ultrahigh sensitivity of 200 nT have been obtained at low frequencies via the synergism of exceptionally high piezoelectric coefficient d33 ∼ 72 pC N−1 of flexible PZT films and strong piezomagnetic coefficient of flexible Metglas foils. Moreover, the proposed flexible ME sensors are remarkably resilient—withstanding over 5000 bending cycles under radii of ∼1 cm without evidence of degradation due to fatigue.
Mica-based PZT films preparation
PZT films were produced by the spin coating method combined with a sol–gel process. First, Pb(CH3COO)2·3H2O was injected to glacial acetic acid at 100 °C. Afterward, Zr(OnPr)4 and Ti(OnBu)4 were mixed to a 0.52:0.48 M ratio of Zr and Ti to maintain Pb(Zr0.52Ti0.48)O3 stoichiometry. Then, the above solutions were blended under continuous magnetic stirring with a 5% excess of Pb for compensating Pb loss caused during annealing. After that, ethylene glycol and distilled water were dripped into the solution to increase stability. The solution was stirred for 30 min at room temperature. After 48 h room-temperature storage, the PZT sol was filtered by using a filter paper with a pore size of 1 µm to obtain the final PZT sol with the concentration of ∼0.5M.
The PZT sol was dropped onto the two-dimensional mica substrate, which was spin-coated at 3000 rpm for 30 s. The obtained films were dried at 300 °C for 10 min on a hotplate to evaporate the solvent. Then, the films were inserted into a conventional box-type furnace and annealed at 700 °C for 10 min. The procedures from spin-coating to annealing were repeated for 15 times to obtain the final PZT thick films with a thickness of 2.2 µm. Afterward, 100 nm thick Pt interdigital electrodes (IDEs) were deposited onto PZT thick films by means of magnetron sputtering combined with a custom-designed mask. Then, the IDEs/PZT/mica heterostructures were poled at 120 °C for 10 min in silicon oil under an electric field of 100 kV cm−1. At last, the IDEs/PZT/mica heterostructures were thinned (∼200 µm–∼20 µm) from the bottom by scotch tape to improve the flexibility. To characterize the local ferroelectric switching behavior via PFM, 100 nm Pt films were deposited on mica by magnetron sputtering to serve as bottom electrodes. To directly measure the macroscopic piezoelectric and ferroelectric properties, another 100 nm thick Pt electrode was deposited on opposite surface of the mica, and the two Pt bottom electrodes were connected using silver paint.
Metglas/PZT magnetoelectric laminate composite fabrication
Metglas (FeSiB alloy) was commercially supplied (Vacuumscheltze GmbH & Co. KG, Germany) as a roll with a thickness of 25 µm. A six-layer Metglas stack (each layer made of a Metglas piece was cut to 40 × 8 mm2) was bonded to the bottom side of the mica-based PZT film with epoxy resin (West system 206, USA) using a vacuum bag pressure method at room temperature.
Characterization of PZT films
2θ scans of PZT films were obtained using an x-ray diffraction (XRD, Bruker D8), and a SEM (ZEISS Merlin) was used to analyze the cross-sectional quality. The surface morphology of PZT films was characterized by AFM (Bruker Multimode 8). Piezoresponse force microscopy (PFM, Bruker Multimode 8) was used to analyze the local piezoelectric response of the PZT films. The ferroelectricity of films was obtained using a ferroelectric tester (Radiant Technologies, USA).
ME effect measurements
All measurements were performed at room temperature in a standard lab environment. Magnetic fields including both the AC magnetic field (Hac) and DC bias field (Hdc) were applied along the length of the heterostructure. The AC magnetic field Hac was generated by a lab-made Helmholtz coil driven by a lock-in amplifier (Stanford Research, SR-850). The DC bias field Hdc was supplied by a water-cooled, U-shaped electromagnet (Myltem PEM-8005K) controlled by a DC current supply/amplifier.
The ME coefficient characterization was performed using an in-house automatic system. The quasi-static ME coefficient αE as a function of Hdc in response to a constant AC magnetic drive of Hac = 0.3 Oe at frequency fac = 1018 Hz was measured by a charge meter (Kistler type 5015), coupled with a lock-in amplifier (Stanford Research, SR-850). The quasi-static ME voltage coefficient αV was calculated by a derivative measurement determined from the ME charge coefficient αQ and capacitance C of the laminate. The dielectric properties of the heterostructure were measured using an impedance analyzer (Agilent 4294 A). The dynamic ME coefficients were measured by sweeping the Hac frequency over the range of 0.1 Hz < f < 100 kHz via a lock-in amplifier under constant amplitudes Hac and Hdc. The magnetic field sensitivity was obtained by measuring the ME output voltage as a function of AC magnetic field under fixed Hac frequency.
Fatigue properties of ME sensor
A lab-made linear stage, driven by a functional generator (Rigoal DG1022a) with a power amplifier (GST YE5872A), was employed to produce periodic bending/unbending motions for the ME sensor. The fatigue response of the ME sensor was performed by measuring the ME coefficient after certain bending cycles.
RESULTS AND DISCUSSION
Configuration of flexible ME sensor
Mica is a unique substrate to fabricate inorganic piezoelectric films. Layered mica exhibits excellent flexibility [Fig. 1(b)] and high chemical and thermal stability.32 The fabrication of a flexible ME laminate is illustrated in Fig. 1(c). The PZT thick films were spin-coated on mica substrates via the sol–gel process, and Pt IDEs were deposited on PZT films by magnetron sputtering via a custom-designed mask. The mica substrate was then thinned to ∼10 µm from the bottom by scotch tape to increase flexibility. Metglas foils were bonded to the bottom side of the mica-based PZT films with epoxy resin. The ME laminate can be attached on a glass bottle with a radius of 1 cm, showing an excellent flexibility [Fig. 1(d)]. The inset is the optical microscopy photograph of Pt IDEs, showing that the finger length, width, and spacing were 4.6 mm, 220 µm, and 80 µm, respectively. The flexible ME laminate heterostructure [Fig. 1(e)] is characterized using a lab-made automatic system [Fig. 1(f)].
Characterization of flexible PZT films
Figure 2(a) shows the x-ray diffraction (XRD) patterns of the pure mica substrate and PZT/mica and PZT/Pt/mica heterostructure, respectively. The results reveal that the PZT films exhibit a clear poly-crystalline pure perovskite structure in both PZT/mica and PZT/Pt/mica heterostructures. From the cross-sectional scanning electron microscopy (SEM) image [Fig. 2(b)], we can see that the PZT film has a uniform thickness of ∼2.2 µm with homogeneous density and an obvious interface between the PZT film and mica substrate. To further scrutinize the homogeneity of the flexible PZT films in large areas, atomic force microscopy (AFM) measurements were carried out. A typical AFM morphology image in an area of 30 × 30 µm2 is given in Fig. 2(c), showing an ultra-flat surface with a height waviness of ∼16 nm. A small root-mean-square roughness of ∼2.47 nm was calculated by the height distribution analysis [Fig. 2(d)].