Advancements in Energy Harvesting: Piezoelectric, Triboelectric, Pyroelectric, and Magnetoelectric Technologies for Self-powered Sensor Systems
Article information
Abstract
In the era of Internet of Things (IoT) advancements, millions of sensors and electronic devices are interconnected through the internet. A reliable power source is required for their continuous operation. Self-powered sensor systems have emerged as promising solutions for various applications including wearable devices and environmental monitoring, benefiting both society and the environment. These systems can maintain themselves by eliminating the need for external power sources or frequent battery replacements and converting discarded energy sources around them into electrical energy. Among various energy harvesting technologies, piezoelectric, triboelectric, pyroelectric, and magnetoelectric mechanisms have garnered significant attention due to their ability to convert mechanical, frictional, thermal, and magnetic energy into electrical power, respectively. This review aims to provide a comprehensive overview of recent advances in these technologies for self-powered sensor systems, catering to a wide audience. It explores fundamental principles of their-energy harvesting mechanisms, highlighting their strengths, limitations, and potential applications. In addition, this review discusses challenges related to the development of nanogenerator-based self-powered sensor systems and presents new opportunities for their advancement.
1 Introduction
Over the past few years, the Internet of Things (IoT) has permeated various sectors, including advanced agriculture through smart farming, innovative residential spaces via smart homes, automated industries with smart sensor technology, personal healthcare systems with wearable sensors, and even commonplace household items [1]. IoT network technology ensures that a wide range of sensors and electronics are continuously connected and operational 24/7. The uninterrupted and reliable operation of the smart sensors, which are key component of these systems, requires a continuous supply of electrical power. Traditionally, the use of various types of small batteries to power these devices has been effective to some extent. However, the widespread deployment of wireless sensors presents challenges such as the need for regular charging and the increased cost associated with battery replacement. Moreover, these batteries often contain chemicals that are not only toxic but also pose serious environmental and health risks [2]. Therefore, there is growing interest in recycling various discarded energy sources and developing self-powered electronics that use sustainable energy. Additionally, relevant innovations can provide environmentally friendly alternatives, reducing reliance on potentially harmful battery technologies and paving the way for more sustainable and secure implementation of IoT systems across different domains. Achieving sustainability in power sources for electronic devices involves harnessing ambient environmental energy such as vibrational or mechanical energy, thermal energy, and electromagnetic waves, and converting it into useful electrical energy. This promising technology is gaining traction through several innovative harvesting devices. Notably, piezoelectric nanogenerators (PENGs) and triboelectric nanogenerators (TENGs) are at the forefront of converting environmental mechanical or vibrational energy into electrical energy. On the other hand, pyroelectric nanogenerators (PyNGs) utilize the heat energy from human body warmth, vehicle exhausts, convection, and solar radiation, converting it into electrical energy through the pyroelectric effect [3–8]. Additionally, the magnetoelectric (ME) energy harvesters also play a crucial role by converting magnetic energy into electrical energy and vice versa [9,10]. In general, people are generally exposed to 50/60 Hz electromagnetic fields, which can be used as an energy source in their daily lives [10,11]. At the heart of the technology required to exploit discarded electromagnetic fields in this field is the development of superior ferroelectric materials based on ceramics, polymers, and magnetostrictive alloys. However, the development of new materials with superior properties is a challenging and time-consuming process. Among many examples, piezoelectric effects, one of the fastest techniques to enter the market, occur when an external force is applied to piezoelectric materials, causing the charges to move within the material and creating an electric dipole. This movement results in a voltage potential from the change in distance between the charges at the dipole’s ends. For PyNGs, the pyroelectric materials capitalize on time-dependent temperature changes (dT/dt) to generate electrical signals [9]. In magnetoelectric effects, the strain from the magnetic field is transferred to the piezoelectric layer, inducing electric polarization through the piezoelectric effect [12]. TENGs work on the principle of static electricity generated through contact electrification, where touching materials exchange charges, leading to one material becoming positively charged and the other negatively charged [1,7,8]. These devices typically utilize materials such as Fluorinated Ethylene Propylene (FEP), Silicone, and Polytetrafluoroethylene (PTFE) to achieve negative charging, and Polyamides (PA), Indium Tin Oxide (ITO), and Zinc Oxide (ZnO) for positive charges [1,13]. Each of these energy harvesters offers distinct advantages and challenges, and they are increasingly employed in energy harvesting and sensing technologies, as detailed in the provided tables. However, most nanogenerator technologies developed to date to build smart systems or environments still have problems in terms of device size, low voltage or current, and long-term reliability.
This comprehensive review offers insights into recent advancements in energy harvesting, emphasizing four types of energy harvesters: Piezoelectric Nanogenerators (PENGs), Triboelectric Nanogenerators (TENGs), Pyroelectric Nanogenerators (PyNGs), and Magnetoelectric (ME) energy harvesters. It thoroughly explores the materials involved, their practical applications in powering self-sufficient sensors, the inherent challenges they face, and the future prospects of these technologies. This paper aims to deepen the understanding of nanogenerators for those intrigued by the dynamic field of energy harvesting, providing both a foundational overview and a detailed examination of the current state and potential advancements in these critical technologies.
2 Classification of Energy Harvesters
2.1 Piezoelectric Energy Harvesters
2.1.1 Mechanism of Piezoelectric Effects
Mechanical energy harvesting is facilitated through the use of Piezoelectric Nanogenerators (PENGs). For a better understanding of piezoelectric effects, the working mechanism for the PENG is described in Fig. 1(a). The piezoelectric layer, when polarized, contains dipoles arranged in a specific direction, which enhances its piezoelectric performance. Initially, when the material is in a stress-free state (Fig. 1(a-I)), charges accumulate on the surface to maintain electrical balance. When compressive stress is applied (Fig. 1(a-II)), the polarization decreases, causing the surface charges to move and generate an electric current. Conversely, removing stress or applying tensile stress (Fig. 1(a-III)) increases the polarization, leading to a current flow in the opposite direction to restore balance [14].
2.1.2 Piezoelectric Materials and Its Progress
Ferroelectric materials are promising for energy harvesting through piezoelectricity. Various materials, including ceramics, polymers, and piezoelectret polymers, are used in PENG fabrication. The piezoelectric ceramics exhibit micro/nanostructures of wurtzite and perovskite. Examples of piezoelectric ceramics with wurtzite micro/nanostructures include ZnO NW, CdS, ZnS, and group III nitrides (InN, GaN, AlN). Examples of piezoelectric ceramics with perovskite micro/nanostructures include PZT, PMN-PT, BaTiO3, ZnSnO3, and K, NaNbO3. The performance of piezoelectric ceramics is superior due to their high piezoelectric coupling coefficient (d33), but they are brittle in nature and some contain lead, which is highly toxic to the environment. Therefore, their use in flexible sensors or certain biomedical applications is limited. In contrast, the ferroelectric polymer such as synthetic (e.g. PVDF and its copolymers) and natural (e.g. silk, cellulose, and collagen) are have gained intense interest in energy harvesting and sensor applications due to their exceptional mechanical flexibility, ease of fabrication, environment friendly, and biocompatibility [6,15]. In addition to that, ferroelectret polymers, like cellular PP foams, possess piezoelectret properties as a result of charge accumulation within their voids. To achieve high piezoelectric performance while preserving excellent device flexibility, a specific number of piezoelectric ceramics is incorporated into the polymer matrix [9,15]. The mechanism of charge generation in ferroelectret polymers differs from that of ferroelectric polymers [16,17]. Extensive research has been conducted to enhance the performance of piezoelectric nanogenerators (PENGs), focusing on parameters such as open circuit output voltage, short circuit current, power density, and energy conversion efficiency [2,6,15,16]. For example, Ghosh et al. proposed platinum (Pt) nanoparticles incorporated P(VDF-HFP) composite film for flexible ferroelectretic nanogenerator (FTNG) [18]. The as prepared FTNG generated an 18 V open circuit voltage with an energy conversion efficiency of 0.2%. The schematic of the capacitor charging process with an equivalent circuit diagram and several capacitor charging performances is illustrated in Fig. 1(I-b). Mahanty et al. proposed an all-fiber pyro-and piezoelectric nanogenerator (PPNG) by incorporating MWCNT into PVDF nanofiber [3]. The schematic of the fabricated device is shown in Fig. 1(I-c) with excellent flexibility in Fig. 1(I-d). The fabricated PPNG demonstrated improved performance with an open-circuit voltage of approximately 35 V, an instantaneous power density of around 34 μW·cm−2, and an energy conversion efficiency of 19.3%. The pressure dependent output voltage of PPNG is shown in Fig. 1(I-e). The fabricated device was able to detect human physiological signals wirelessly through a mobile device, enabling remote healthcare monitoring. Furthermore, Roy et al. adopted a scalable approach by incorporating a [Cd(II)-μ-I4] two-dimensional (2D) metal-organic framework (MOF) bridged by naphthylamine into PVDF composite nanofibers mat [19]. The schematic of electrospinning setup and fabricated nanofiber mat is shown in Figs. 1 II-f and 1II-g respectively. The fabricate device generated high electrical output (open circuit output voltage ~22 V (Fig. 1(I-h)), and power density ~24 μW·cm−2) under periodic imparting. The composite nanofibers showed a superior electroactive β-phase content of 98% with a high piezoelectric coefficient of 41 pC N−1. However, all these devices are composed of a single layer as the piezoelectric active layer. In order to improve the performance of the PENG layer-by-layer structure was adopted further [20–24]. Several studies have demonstrated that homo-layered structure PENGs can significantly improve output performance. To overcome the limitations of homo-layered PENGs, a hetero-structured PENG (HT-PENG) was created, showing enhanced voltage and current output. The schematic of the layer-by-layer hetero-structure PENG with interspace metal sheet and cross-sectional SEM images of the as-prepared HT-PENG are shown in Figs. 1(II-a) and 1(II-b), respectively. The open-circuit voltage and short-circuit current of the HT-PENG are shown in Figs. 1(II-c) and 1(II-d), respectively. Furthermore, the open-circuit voltage and short-circuit current of the optimized 6th layer HT-PENG are shown in Figs. 1(II-e) and 1(II-f), respectively. The optimized six-layered HT-PENG achieved an open circuit output voltage of 350 V, a short circuit current of 6 μA, and a power output of 3.62 W·m−2 [6].
2.2 Triboelectric Energy Harvesters
2.2.1 Mechanism of Triboelectric Effects
In addition to PENGs, mechanical energy can be efficiently converted into electrical energy using Triboelectric Nanogenerators (TENGs), which operate in four distinct modes: vertical contact separation (VCS), linear sliding mode (LSM), single-electrode mode (SEM), and free-standing mode (FSM), as depicted in Fig. 2(a) [14,25]. TENGs function based on the principles of the parallel-plate capacitor model and the dependence of the electric field on the distance to optimally match electrical loads [14]. Table 2 presents a detailed overview of the structural characteristics, advantages, disadvantages, and potential applications of each operational mode [14]. Fig. 2(b) demonstrates a TENG operating in the VCS mode. Initially (Fig. 2(b-I)), when the layers of the device are compressed together, triboelectric charges are generated. As these layers are released and separate (Fig. 2(b-II)), a potential difference is created between the charged layers. This difference drives the movement of free charges in the electrodes to balance the potential, thereby generating a current flow when the device is connected to a load resistance (R). Once the surface charge balances, the current flow ceases (Fig. 2(b-III)). Re-compressing the films allows the previously accumulated charges to flow back into the circuit, producing a current flow in the opposite direction.
2.2.2 Triboelectric Materials and Its Progress
TENGs generate static electricity through contact electrification. When two materials come into contact, charges are exchanged between them, resulting in one material becoming positively charged and the other becoming negatively charged [26–28]. Tremendous research work has been carried out in order to improve the electrical performance of TENGs for energy harvesting and sensor applications [25,29,30]. However, there is still room for further improvement in terms of energy conversion efficiency, open circuit output voltage, short circuit current, power density, and sensitivity for self-powered wearable sensors in human-machine interaction. The charge transfer mechanism occurring on the surface of the two tribo materials during physical contact stands as a pivotal factor in optimizing the output performance of a TENG. To increase the surface charge density, many methods have been adopted, including surface modifications (e.g., physical and chemical), material modifications (e.g., dielectric constant and mechanical properties) or surface and bulk modifications, and the layer-by-layer self-assembly approach has also been adopted [29]. For example, Prasad et al. investigated the self-powered triboelectric sensors (TES) through a straightforward poly-L-lysine (PLL) post-surface-modification method. This approach aimed to enhance the positive polarity of Nylon 11 electrospun sheets [31]. The Nylon 11 fabrication and post-surface-modification using PLL is shown in Fig. 2(c). The as fabricated TENG exhibited superior electrical output performance (such as, output voltage~270 V, short circuit current~7.2 μA, and power density~2 W·m−2) under 8.5 N imparting force (Figs. 2d and 2e). Ghosh et al. suggested modifying materials by creating a polyurethane polymer composite film infused with ferroelectric barium-titanate-coupled (BTO-coupled) 2D MXene (Ti3C2Tx) nanosheets [32]. MXene enhances the dielectric constant, while coupling with the ferroelectricity of BTO reduces dielectric loss, improving the nanogenerator’s output performance. The fabricated device uses quantum-mechanical calculations to convert biomechanical energy into electricity, producing 260 V open-circuit voltage, 160 mA·m−2 short-circuit current, and 6.65 W·m−2 power output, ideal for charging various consumer electronics. Fig. 2(f) illustrates the microporous hierarchical composite TENG, the dielectric permittivity of TPU and its various composite films, and the output voltage of TPU and its composite-based TENGs respectively. Interlayer engineering is a promising method for enhancing triboelectric performance, in addition to surface and bulk modifications. Feng et al. found that using a polyimide (PI) layer as a charge storage layer between a PVDF tribo-material and a Cu electrode enhanced the triboelectric performance significantly [33]. The PI layer stored negative charges, leading to a 9-fold improvement in triboelectric output compared to a PVDF monolayer-based TENG. The TENG with a PI transition layer reaches a peak output voltage of 1,010 V and a short-circuit current of 65 μA. Salauddin et al. developed a composite material called LC-MXene/ZiF-67 with a silicone friction layer [34]. The LC-MXene/ZiF-67 intermediate layer has abundant charge trapping sites and a porous structure, leading to improved charge trapping properties. The voltage and current densities of CM-TENG are 13.4 and 14.5 times higher than a single layer. The schematic illustration of the fabrication process of LC-MXene/ZiF-67 is shown in Fig. 2(g). The as prepared CM-TENG shown superior power density of 65W·m−2 under a matching load impedance of 1.3 MΩ as shown in Fig. 2(h).
2.3 Pyroelectric Energy Harvesters
2.3.1 Mechanism of Pyroelectric Effects
In addition to environmental mechanical energy harvesters, pyroelectric nanogenerators (PyNGs) convert environmental waste heat into electrical energy through pyroelectric effects [35]. PyNG technology integrated into wearable electronics can efficiently harness wasted heat from the environment. It can also be integrated into masks, which have become essential due to the Covid-19 pandemic, a global health crisis. Numerous research studies have been conducted to refine the design of PyNGs using pyroelectric materials [3,4,35–38]. Pyroelectric materials display inherent spontaneous polarization (Ps) even in the absence of an electric field. The pyroelectric effect is observed when Ps undergoes temporary changes in response to variations in temperature within the material [39]. When the temperature increases, the surface charges bound to the surface decrease due to thermal vibration, leading to a decrease in Ps. This generates an electrical potential across the pyroelectric material if it is under an open circuit condition. In the case of a short circuit condition, electrical current flows through the external circuit. The equation governing the pyroelectric current output is given as Eq. (1).
Here, “i” represents current, “Q” signifies pyroelectric charge, “p” is the pyroelectric coefficient, “A” is the surface area of pyroelectric materials, and “t” signifies time. Various methods are used to accurately measure the pyroelectric coefficient (p), including static and dynamic techniques [40]. Static methods include charge compensation, hysteresis measurement, and X-ray techniques. Dynamic methods involve temperature ramping and optical techniques. The practical assessment of the pyroelectric coefficient of PyNG can be attained by measuring the output current and relevant physical parameters. This coefficient is estimated using the following Eq. (2):
Fig. 3(a) depicts the energy harvesting system utilizing the pyroelectric effect. The working mechanism of PyNG is detailed in Figs. 3(b)–3(d), based on temperature-induced changes in spontaneous polarization (Ps) [37,41]. Polymers with crystalline structures, where molecular chains align, can exhibit (Ps) due to the alignment of polarized covalent bonds. Similarly, ceramics with ionic bonding can also show Ps due to polarization within their crystal lattice [42]. Ideally, internal dipoles should align in one direction, but wiggling atoms disrupt this. When polarization and wiggling angle (θ) remain constant at a constant temperature, no current flows through the external circuit (Fig. 3(b)). As the temperature rises (dT/dt > 0), thermal energy causes dipole alignment to change, with the dipoles wiggling around their respective pole axes (Fig. 3(c)). Higher temperatures result in larger wiggling angles for diverse polarizations, leading to a decrease in the intensity of Ps. The decrease in induced surface charges on pyroelectric material is a result of the rising temperature, prompting released surface charges to traverse through the external circuit. Upon returning to the initial state, the dipole alignment reverts, inducing reverse current flows in the external circuit. Conversely, a temperature decrease (dT/dt < 0) leads to internal dipoles with reduced angles (θ2 < θ), owing to diminished thermal energy. This, in turn, amplifies the intensity of Ps significantly, as illustrated in Fig. 3(d). The surface charges on pyroelectric material increase when temperature changes, causing them to flow through the external circuit. When the temperature returns to normal, the dipole alignment of the material is restored, leading to a decrease in surface charges and a reverse current flow through the external circuit. This process generates electrical energy through the pyroelectric effect. Fig. 3(e) shows the temperature-dependent Ps change in pyroelectric materials [43].
2.3.2 Pyroelectric Materials and Its Progress
Pyroelectric materials have shown great potential in sensors, energy harvesting devices, and even in medical applications. The ability of pyroelectric materials to convert temperature changes into electrical signals makes them valuable in a wide range of technologies. Researchers continue to explore new ways to enhance the performance and versatility of pyroelectric materials for future applications [35,43,44]. Selection of pyroelectric materials from a wide variety of materials, including both ferroelectric and non-ferroelectric materials, is a significant area of research. All ferroelectric materials exhibit pyroelectric properties, and all pyroelectric materials demonstrate piezoelectric behavior, but the reverse is not necessarily the case [43,45,46]. It can be observed that triglycine sulphide (TGS) is extremely potential materials for pyroelectric energy harvesters. TGS, with the chemical formula (NH2CH2COOH)3H2SO4, consists of crystals derived from the glycine group (NH2CH2COOH) that are polar and demonstrate exceptionally high pyroelectric figures of merit (FOM) [42,47]. TGS (Triglycine Sulfate) faces limited interest for energy harvesting applications primarily because of its low Curie temperature of 49°C, water solubility, hygroscopic nature, and relatively low mechanical strength [48]. Lead magnesium niobate-lead titanate (PMN-PT) and lead zirconate titanate (PZT) are popular ceramics known for their simple fabrication process and excellent piezoelectric properties. These materials have garnered attention for their potential in pyroelectric energy harvesting applications [49–51]. Yang et al. showcased the initial use of a PyNG as a self-powered sensor to detect temperature variations with a single PZT micro/nanowire. The sensor’s response time and reset time were approximately 0.9 and 3 seconds, respectively [51]. The schematic of a single PZT microwire PyNG is shown in Fig. 3(f). Figs. 3(g)–3(j) show the self-power sensor demonstrations. The sensor is attached to a metallic body as shown in Fig. 3(g) with an LCD screen in Fig. 3(h). When there is no change in temperature in the sensor, the LCD is off, whereas the LCD is on and showing a number on the screen as shown in Fig. 3(j) while the sensor is under heat treatment at a temperature of 473 K as shown in Fig. 3(j). Lead-free ferroelectric materials such as bismuth sodium titanate-barium titanate (BNT-BT) and Mn:BNT-BT offer high pyroelectric coefficients and figure of merit, making them appealing alternatives to lead-based materials [52,53]. Additionally, Ba0.65Sr0.35TiO3 (BST) thin films are commonly used to enhance pyroelectric energy harvesting [54]. Besides these, lithium tantalate (LiTaO3, LTO), lithium niobate (LiNbO3, LNO), and potassium sodium niobate-based materials have gained popularity over lead-based pyroelectric materials due to their low dielectric loss [43], high Curie temperature [55], and excellent pyroelectric coefficient [56], respectively. Besides ferroelectric materials, certain non-ferroelectric materials are also crucial in the design of a PyNG. These materials include wurtzite-based compounds like AlN, GaN, CdS, and ZnO [43]. However, their pyroelectric coefficients are lower than those of ferroelectric materials, but they have higher thermal conductivities [43].
The materials mentioned above are usually ceramic-like, resulting in high density, stiffness, and brittleness. If mechanical flexibility and toughness are desired, a polymeric pyroelectric material like polyvinylidene fluoride (PVDF) and its copolymers (e.g. polyvinylidene-difluoride trifluoro-ethane; P(VDFTrFE)) can be used for pyroelectric energy harvesting [3,9,57,58]. Mahanty et al. proposed an all-fiber pyro- and piezo-electric nanogenerator (PPNG) that was demonstrated using MWCNT-doped PVDF composite nanofiber [3]. The PVDF-MWCNT composite nanofiber possesses fifteenth times higher pyroelectric coefficient (~60 nC·m−2·K−1) compare to that of pure PVDF nanofibers (~4 nC·m−2·K−1). In another study, Sultana et al. proposed a composite electrospun nanofiber made of methylammonium lead iodide (CH3NH3PbI3) and PVDF to harvest thermal energy [57]. The prepared PyNG showed a fast response time of 1.14 s, a reset time of 1.25 s, a voltage of 41.78 mV, and a pyroelectric coefficient of approximately 44 pC ·m−2·K−1. Recently, there has been a growing demand for layered double hydroxide (LDH)-based pyroelectric nanogenerators (PyNGs) due to their unique properties, including high gradients of composition and energy levels, size effects, and the ease of fabrication and tuning [59]. Prestopino et al. demonstrated experiments where they fabricated several LDH-based PyNGs. Remarkably, they observed that without the need for poling treatment, the pyroelectric coefficient can range from positive, up to 150 μC·m−2·K−1 for PyNGs made of ZnAl-LDH, to negative, down to −160 μC·m−2·K−1 for those composed of MgAl-LDH [59].
2.4 Magnetoelectric Energy Harvesters
2.4.1 Mechanism of Magnetoelectric Effects
In addition to mechanical and heat energy converters, magnetoelectric (ME) energy harvesters stand out as a notable energy harvesting technology. These devices convert magnetic energy into electrical energy by leveraging mechanical strain/stress-mediated magnetoelectric (ME) coupling effect. Multiferroic composites, which exhibit both ferromagnetism and ferroelectricity, have attracted considerable interest for their magnetoelectric (ME) couplings, making them promising candidates for energy harvesting applications. The magnetoelectric (ME) effect emerges from the coupling between magnetic and polar sublattices within single-phase materials. It is uncommon to find long-range ordered magnetic moments and electric dipoles coexisting in the same phase, leading to the discovery of only a few single-phase compounds exhibiting the ME effect. Obtaining significant magnetoelectric (ME) coupling in single-phase compounds at room temperature has proven to be challenging due to the concurrent transitions from ferroelectric to paraelectric states and from ferro/ferri/antiferromagnetic to paramagnetic states [60–62]. However, these constraints can be surpassed through the integration of ferroelectric and ferromagnetic materials within ME composites, capitalizing on the remarkable properties inherent in these phases. Research has shown that multiphase ME composites can achieve a significantly improved ME response [63] compared to single-phase ME materials [64] at ambient temperature. Layered ME composites, which combine piezoelectric and magnetostrictive materials in an elastic structure, have been extensively researched due to their simple manufacturing process and adaptable design. Various factors such as strain, spin, or charge carrier exchange among the constituent phases are responsible for the interplay of magneto-electricity within the composite materials. While the mechanism of strain-mediated coupling is comprehensively understood, the exploration of the other two mechanisms remains ongoing [10,65,66]. In composites, strain-mediated coupling occurs due to the elastic interaction between piezoelectric and magnetostrictive components, as shown in Fig. 4(a) [63,67]. The detailed working mechanism is illustrated below. The ME effect encompasses the process wherein the imposition of a magnetic field induces mechanical strain within the magnetic layer, a phenomenon referred to as magnetostriction. Subsequently, this strain propagates to the piezoelectric layer, obtaining an electric displacement via the piezoelectric effect, as illustrated in Fig. 4(a). The ME response is quantified using the ME voltage coefficient
2.4.2 Magnetoelectric Materials and Its Progress
In the past decade, there has been significant research on magnetoelectric (ME) composites [9,68]. The ME effect involves converting magnetic energy into electric energy and vice versa. This transducer responds to changes in electric polarization due to a magnetic field (direct ME effect) or changes in magnetization due to an electric field (converse ME effect). These transducers consist of magnetostrictive and piezoelectric layers, which can function as sensors, actuators, or energy harvesters. These devices efficiently transform AC magnetic fields into vibration and electric charge, rendering them ideal for harvesting energy from diverse sources. This includes but is not limited to mobile base stations, Wi-Fi routers, satellite communications, radio and TV transmitters, and power distribution lines [69]. According to Ampere’s circuital law, current-carrying conductors connected to house hold appliances can generate a low-amplitude magnetic field with a low frequency. The ME effect has indeed been observed in single-phase materials like Cr2O3 and BiFeO3, though typically it is rather weak at room temperature [70]. Van Suchtelen [71] demonstrated that piezoelectric-piezomagnetic composites could manifest a more dominant ME effect. The constitutive relations for these composites are typically expressed using the following equations [72],
where B, and H are the magnetic induction and magnetic field intensity, σ, μ, and β are the elastic compliance, magnetoelastic constant, magnetic permeability, and ME susceptibility, respectively.
Tremendous research has been conducted in the field of magnetoelectric energy scavenging/harvesting technology using a variety of materials and composites [10,73–84]. Wu et al. developed a rotational parameter sensor using a FeNi/PZT magnetostrictive/piezoelectric laminated composite (MPLC) and a multi-pole magnetic ring [85]. Tang et al. developed a self-biased magnetoelectric (ME) charge coupling within a two-phase laminate stack termed Ni/PZT. This stack incorporates Nickel foils and a Pb (Zr,Ti)O3 (PZT) plate boasting high capacitance [74]. Experimental findings reveal that the Ni/PZT-stack with n = 0.4 showcases a remarkable zero-biased resonant ME charge coefficient (αQ,r) of 47.52 nC· Oe−1, notably surpassing the values reported for other ME laminates in prior research. Additionally, Ghosh et al. developed an increased magneto-electric (ME) effect at room temperature (RT) and above in a nanocomposite of LaYFe2O6/P(VDF-HFP) [81]. At room temperature, this nanocomposite exhibits a first-order ME coupling coefficient of approximately 2.92 mV·cm−1·Oe−1 and a second-order ME coupling coefficient of around 0.051 μV·cm−1·Oe−2. The experimental setup is illustrated in Fig. 4(c). Fig. 4(d) shows the schematic and digital photograph of the fabricated device, as well as the harvesting process. The harvested output voltage from the household device and the Cole-Cole plot of the impedance are depicted in Figs. 4(e) and 4(f), respectively. Recently, Ghosh et al. reported an enhanced ME effect in spin-photon coupled single-phase La1-x SmxYFe2O6 (0 ≤ x ≤ 1) [84]. The composition with x = 0.75 exhibited the most effective polarization and magnetoelectric response. It demonstrated a first-order ME coupling coefficient approximately 31% higher and a second-order ME coupling coefficient approximately 1 order higher than the original x = 0 sample. The spin-reorientation transition significantly impacts the magnetic and magnetoelectric properties, making it a valuable tool for adjusting these properties. With advancements in inorganic materials, polymer-based ME materials have achieved similar levels of ME coefficients, and their usage as energy harvesters is rising owing to their excellent flexibility, lightweight, and biocompatibility [10,86,87]. To date, many research works have been carried out using polymer-based materials to improve magnetoelectric energy harvesting [10,73,77,88–90]. Kwon et al. showed a bilayer composite sample consisting of Metglas and PVDF with epoxy bonding in their study [89]. DC and AC magnetic fields were produced using a pair of Helmholtz coils. The fields were adjusted from 0 to 3 kA·m−1 (~37.7 Oe), and the AC frequency was varied from 0 to 200 Hz. The magnetic field intensity was monitored using a Hall sensor to control the field level. Silva et al. proposed a three-layer (piezoelectric+epoxy+magnetostrictive) (PVDF/epoxy/Vitrovac) ME structure [73]. Experimental results demonstrate that the magnetostrictive effect response improves with the increase of the PVDF thickness. The highest response of 53 V·cm−1·Oe−1 was achieved with a 110 μm thick PVDF/M-Bond epoxy/Vitrovac laminate. This excellent ME response makes it a prominent material for use in sensors, actuators, memories, and energy harvesting devices. Additional research by Martins et al. explored the dispersion of cobalt ferrite (CoFe2O4) nanoparticles in a poly (vinylidene fluoride)-trifluoroethylene (P(VDF-TrFE)) matrix [79]. The study investigated the impact of nanoparticle dispersion on the piezoelectric, magnetic, and magnetoelectric characteristics of the nanocomposite. Two distinct dispersion techniques were employed in sample preparation: ultrasound and citric acid nanoparticle surfactation. No significant distinctions were found in the ferroelectric, piezoelectric, magnetic, and magnetoelectric properties between samples fabricated with or without surfactants, thereby streamlining the process for large-scale production. Ghosh et al. introduced a versatile and flexible magneto-mechano-electric nanogenerator (MMENG) that can be rolled up [10]. This study showcased the fabrication of a fully rollable MMENG employing P(VDF-TrFE)/nickel ferrite (NiFe2O4/NFO) 0–3-type magnetoelectric nanocomposites. This device is designed for wireless Internet of Things (IoT) sensors to detect and harness environmental magnetic noise even without a direct current magnetic field. The working principles of MMENG with TrFE/NFO 0–3 nanocomposites is illustrated in Fig. 4(g). When subjected to an AC magnetic field (Fig. 4(g-i)), the NFO nanoparticles in the MMENG undergo magnetostriction, causing them to elongate or contract in sync with the magnetic field frequency (Fig. 4(g-ii)). This strain is then transferred to the P (VDF-TrFE) polymer chain through strong interfacial interactions, resulting in stress application (Fig. 4(g-iii)) that generates electric charges due to polarization alignment. Subsequently, the generated voltage is delivered across an external electrical load via the direct piezoelectric effect (mechano-electric coupling). The harvested electrical power can be utilized to power consumer electronics (Fig. 4(g-iv)). Fig. 4(g-i) shows the fabricated MMENG. Fig. 4(h-ii) illustrates the demonstration of magnetic field energy harvesting from the stray magnetic field surrounding the power cable of an electric kettle (1 kW, 50/60 Hz). Additionally, Fig. 4(h-iii) presents a digital photograph of the MMENG. The MMENG is known for its flexibility, as depicted in Fig. 4(h-iv), and rollability, as showcased in Fig. 4(h-v). Fig. 4(i) illustrates the MMENG producing an AC voltage output with a sinusoidal waveform. The MMENG using TrFE/NFO1 yielded a peak-to-peak output voltage (Vpp) of approximately 640 mV, while TrFE/NFO2 produced a Vpp of around 1.4 V. Increasing the distance between the MMENG and power cable from 0.5 to 2 mm resulted in a decrease in voltage output from approximately 1.4 to 0.4 V as the AC magnetic field strength decreased from 1.7 to 0.4 mT (Fig. 4(j)). The power output of the MMENG was assessed by capturing peak output voltages across various external load resistances (RL), ranging from 0.25 MΩ to 1 TΩ, during the boiling of water in an electric kettle (Fig. 4(k)). The peak power density reached 0.05 μW·cm−3, across RL~100MΩ.
3 Applications in Self-powered Sensor Systems
Recent advancement in energy harvesting have gained momentum owing to the growing demand for self-powered portable and wireless electronics, as well as systems with extended lifespans, which have opened up a wide range of applications. In this section, we showcase some applications of self-/auto-powered wearable sensors, implantable medical devices, human machine interactions, and wireless healthcare systems utilizing energy harvesters. Significant advancements have been made in self-powered electronics through the use of energy harvesters [10,29,91–94]. Meanwhile, the energy harvesting outlooks for some of the potential applications in self-powered electronics will be illustrated in this section. For instance, Kim et al. proposed a wireless communication system for a healthcare system that utilized in vivo energy harvesting with excellent output performance PMN-PZT PENG technology [91]. Fig. 5(a) shows the experimental setup for self-powered wireless data transmission through in vivo energy harvesting. Energy generated by cardiac motions was stored in a 22 μF capacitor via a full-wave bridge rectifier. The transmitting part could operate wirelessly when connected to the charged capacitor. Data was wirelessly transmitted to the receiver utilizing a communication protocol known as Wireless Universal Serial Bus (WUSB). To visually confirm the data transmission, instructions were sent to switch a light bulb on and off at a distance of approximately 5 m. Gupta et al. demonstrated MXene-incorporated PVDF composite nanofiber based PyNG for advanced breathing sensors and IR data receivers with machine learning prediction [92]. To validate pyroelectric sensor data for advanced applications, a machine learning algorithm was adopted on the obtained data as shown in Fig. 5(b). The breathing response was captured using a pyroelectric sensor in two scenarios: before and after exercise. The breathing rate differed significantly between the two cases, with approximately 14 breaths per minute before exercise and around 30 breaths per minute after exercise. Additionally, the peaks in the breathing response varied between the two conditions. These sensor data were utilized for machine learning applications to differentiate between pre- and post-exercise breathing signals. Multiple data sets were generated, preprocessed, and split into training and test data for the development of machine learning algorithms. Four algorithms (Logistic Regression, K-Nearest Neighbors, Support Vector Machine, and Random Forest) were employed to train the models. The results indicated that K-Nearest Neighbors and Random Forest were the most effective in identifying breathing patterns based on the input features. Further, Roy et al. proposed a self-powered wearable pressure and pyroelectric breathing sensor based on graphene oxide (GO) interfaced with PVDF nanofibers [58]. The performance of this breathing sensor was enhanced by incorporating conductive GO nanofillers into the PVDF nanofibers. During the demonstration, the PyNG was mounted on an N95 mask worn on the human face. The temperature of the airflow during respiration was detected by the PyNG as voltage or current fluctuations. Thus, the breathing rate and pattern, vital signs of human health, especially regarding the cardiorespiratory system, were monitored. In addition, Xue et al. proposed a wearable PyNG as a self-powered breathing sensor using PVDF thin film [93]. Due to temperature fluctuations from human breathing at 5°C, the PyNG generates output signals with an open-circuit voltage of 42 V and a short-circuit current of 2.5 μA. The fabricated PyNG was attached inside a mask, mounted on the face, and driven by human breathing. The temperature sensor fixed on the PyNG recorded the real-time temperature during the respiratory process. The output electrical signals of the PyNG directly record the respiratory rate of the human being. Therefore, the prepared PyNG can be used as a breathing sensor for human health monitoring. Ghosh et al. engineered a flexible and rollable MMENG using nickel ferrite (NiFe2O4) nanoparticles alongside a piezoelectric PVDF polymer-based wireless IoT sensor. This device captures and utilizes environmental magnetic noise without the need for a direct current magnetic field [10]. The MMENG was showcased as a wireless IoT sensor within a position monitoring system, illustrated in Fig. 5(c). The signal generated by MMENG (using TrFE/NFO2) from an electric kettle’s power cable during water boiling was wirelessly transmitted to a smartphone via an Arduino MCU-based platform. The wireless data transmission circuit schematic is depicted in Fig. 5(d), alongside real-time images showcasing the Arduino MCU board connected to a Bluetooth module (HC-05) and an Android smartphone displaying the signals. The measured wireless signal output is illustrated in Figs. 5(e) and 5(f). This demonstrates the potential of the MMENG as a position monitoring IoT sensor. When a user approaches the Bluetooth transmitter with a smart device, the signal amplitude increases, and decreases when the user moves away, confirming the MMENG’s ability to harvest magnetic fields for IoT applications. Maharjan et al. developed a highly sensitive self-powered triboelectric flex sensor (STFS), capable of effectively detecting finger bending motion and monitoring hand gestures with high precision [94]. The flex sensor can detect pressure in a wide range from 0.2 to 500 kPa. It has been successfully used in a real-time sign language interpretation system that detects finger gestures and converts them into voice and text using a smartphone application. Sensors are attached to the human finger joints, and their raw output voltage signals are filtered and processed in the signal processor unit. Each alphabet letter in American Sign Language (ASL) corresponds to a specific hand gesture, generating unique sensor output voltages. The processed signals are converted from analog to digital using an ADC and transmitted to a microcontroller, which wirelessly sends the data to a smartphone via Bluetooth. A custom Android app displays the data from the smart glove in text and voice formats. The smart glove incorporates five STFS, a microcontroller unit equipped with Bluetooth Low Energy (BLE) capability, and a smartphone application. Due to its easy fabrication process, exceptional sensing capabilities, and robust mechanical durability, the STFS emerges as a cost-effective solution viable for mass production. It is ideal for assisting individuals with speech disabilities in interpreting sign language. Furthermore, Park et al. demonstrated frequency - selective acoustic and haptic smart skin for dual-mode dynamic/static human-machine interface system [95]. They demonstrated that their dual-mode sensor, featuring linear response and frequency selectivity across a broad range of dynamic pressures, effectively differentiates surface textures and controls an avatar robot using both acoustic and mechanical inputs without interference from ambient noise. They fabricated a smart glove by integrating triboelectric sensors (TES) onto the tips and joints of the fingers to detect surface texture and finger movements, respectively (Fig. 5(g)). To evaluate the texture perception capability, the proposed TES was placed on 3D-printed target surfaces with regular line patterns and scanned using a homebuilt system that controlled the scanning speed and micro-unit displacement. They collected large quantities of time-dependent current data corresponding to different surface textures (e.g., polyester, cotton, nylon, silk, glass, paper, and sandpaper) for ANN training and transformed the data into FFT spectra to extract frequency features. The resulting classification matrix indicated a high positive predictive value of 92.7% for the proposed training model, providing a simple platform for robotics requiring sophisticated tasks (Fig. 5(h)) [94]. In addition to the single-layer PENG, multilayer PENGs have also been demonstrated to improve device performance further [6]. Mahanty et al. proposed a multilayered PENG for real-time applications as self-powered wearable sensors on different parts of the human body to measure biomedical activity [6]. They also showcased a wireless remote healthcare monitoring system (Figs. 5(i) and 5(j)). The circuit diagram and schematic of the IoT-based remote health monitoring system are presented in Fig. 5(i), emphasizing key components such as a single-chip ESP8266A Wi-Fi module and a PC/smartphone running the Blynk app. Furthermore, the real-time practical circuit is illustrated in Fig. 5(j-i), demonstrating the output response of the multilayered PENG under gentle finger touch. This response is displayed on a smartphone screen (Fig. 5(j-ii) via the local server for the IoT-based remote healthcare monitoring system. Recently, Wang et al. demonstrated non-contact sensing technology aimed at seamless data acquisition and intelligent perception, which brings innovative interactive experiences to wearable human-machine interaction [96]. This study introduces triboelectric nanopaper prepared through a phase-directed assembly strategy, showcasing both low charge transfer mobility (1,618 cm2·V−1·s−1) and remarkable stability at high temperatures. The noncontact sensing based on triboelectric nanopaper is shown in Fig. 5(k-p). The schematic diagram of the noncontact sensors for wearable motion monitoring and spatial position perception is depicted in Fig. 5(k). The sensor was attached on the volunteer’s arms and inner sides of the shoes (Fig. 5(k)). By analyzing the peak pattern and numerical characteristics of the open-circuit voltage signal, the sensor can distinguish step frequency, foot distance, and movement speed (Fig. 5(l)), enabling monitoring of motion states such as stationary, walking, running, and jumping (Fig. 5(m)). The system uses wearable non-contact sensors to capture motion signals (e.g., walking, running) and transmits them wirelessly via Bluetooth to a mobile app, enabling real-time movement visualization (Figs. 5(n) and 5(o)). When the distance to the instrument fell below 30 cm, the non-contact sensor’s signal exceeded 4 V, warning workers of harmful temperatures and preventing thermal injuries (Fig. 5(p)). Thus, the triboelectric nanopaper-based sensor also enables spatial positioning, movement detection, and functions effectively in high-temperature environments. The examples mentioned here are only part of the relevant technologies, and the development of self-powered advanced sensor technologies for better lives is still growing remarkably from wearable applications to industrial applications.
4 Conclusions and Future Perspectives
This paper provides a concise review of recent advancements in piezo-, tribo-, pyro-, and magneto-electric energy harvesters and their applications in self-powered electronics. Self-powered sensor systems are poised for significant growth in the electronics sector, with potential applications ranging from monitoring human health to environmental conditions in the era of the Internet of Things (IoT). We highlight cutting-edge developments in energy harvesters concerning their materials, fabrication techniques, applications, and development processes. These advancements indicate substantial progress across various domains, suggesting imminent technological breakthroughs. Nanogenerator-based self-powered sensor systems have shown promising developments; however, several challenges hinder their widespread commercial adoption:
1) Efficiency: Enhancing the energy conversion efficiency of these devices.
2) Reliability and Durability: Ensuring consistent performance and long-term usability.
3) Compatibility and Interfacing: Integrating seamlessly with existing electronic systems.
4) Miniaturization and Integration: Reducing size while maintaining functionality.
5) Cost-effectiveness: Making the devices economically viable for mass production.
6) Wearability: Designing comfortable and practical wearables
Addressing these challenges requires a collaborative approach in interdisciplinary research, merging insights from materials science, mechanical engineering, electronics, and energy harvesting technologies. As we advance in these areas, nanogenerator-based sensor systems are anticipated to increasingly influence the development of sensing and IoT technologies. In conclusion, surmounting the obstacles in developing self-powered wearable devices with diverse energy harvesting and sensing capabilities could unlock new opportunities and insights for the future of these technologies.
Acknowledgements
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (RS-2022-00165505 and 2020R1A5A8018367).
List of Symbols
IOT
Internet of Things
PENG
Piezoelectric Nanogenerator
TENG
Triboelectric Nanogenerator
PyNG
Pyroelectric Nanogenerator
ME
Magnetoelectric
Notes
Declaration of Competing Interest
The authors declare no competing financial interest.
References
Biography
Dr. Biswajit Mahanty received his Bachelor’s degree in Electronics from Burdwan University, West Bengal, India, and a Master’s degree in Electronic Science from Jadavpur University, West Bengal, India. He completed his M.Tech in Software Engineering from MAKAUT, Kolkata, India, and received his Ph.D. in Physics from Jadavpur University, West Bengal, India. Currently, he is working as a post-doctoral researcher at the MEMS and Nanotechnology Laboratory, School of Mechanical Engineering, Chonnam National University, Republic of Korea. His research interests include the design and development of Piezo-, Tribo-, Pyro-electric, Tribo-voltaic Nanogenerators, Self-powered Flexible Sensors, Supercapacitors, and Wireless energy transfers.
Prof. Dong-Weon Lee received his Ph.D. degrees in Mechatronics Engineering from Tohoku University, Sendai, Japan in 2001. He has been a Professor of Mechanical Engineering at Chonnam National University (CNU), Republic of Korea since 2004. Previously, he was with the IBM Zurich Research Laboratory in Switzerland, working mainly on microcantilever devices for chemical AFM applications. At CNU, his research interests include smart cantilever devices, miniaturized energy harvesters & flexible supercapacitors, smart structures & materials, and nanoscale transducers. He is a member of the technical program committee of IEEE MEMS conference, IEEE Sensors Conference, Transducers, and Microprocesses and Nanotechnology Conference etc.