Designing wearable microgrids: towards autonomous sustainable on-body energy management

Abstract

The rapid development of wearable sensing and interfacing electronics is facing challenges in sustainability and energy independence. The reliable and sustainable operation of such autonomous wearable electronics hinges on the rational integration of energy harvesting and storage modules, as well as their corresponding control and regulation circuitries. Such emerging self-powered wearable systems still lack systematic system-level energy management considerations and are thus limited in efficiency, reliability, and practicality for their targeted use cases. Viewing the scattered wearable energy technologies through the concept of independent microgrids allows us to reassess the goal of establishing a reliable, practical, and energy-economical wearable system. Powered by distributed on-body energy harvesting modules, the continuous operation of wearable devices can thus be realized with their strategic integration with energy storage managed by the decentralized, hierarchical control system. This Perspective discusses the vision of a wearable microgrid, based on a judicious scenario-specific selection of harvesting and storage modules, with commensurate performance, towards the rational design of practical wearable electronic systems with high energy autonomy and reliability. The energy supply via different sources and power demands of various wearable applications are examined for determining appropriate energy reserves, attainable functionalities, and suitable operation modes, along with key considerations in device form-factors for maximizing system efficiency. By applying the concept of a microgrid on miniaturized self-powered systems for wearables, we propose three system-level design guidelines – commensurate energy rating, complimentary device characteristics, and compatible form factors – towards the future development of reliable, self-sustainable on-body systems and their extension to autonomous implantable, ingestible, or small mobile devices. We conclude by discussing the prospects for developing more efficient and sustainable wearable microgrids for higher power applications, through accurate and smart energy budgeting and regulation involving artificial intelligence and advanced algorithms towards dynamic data-driven prediction of rapidly changing power supply and demands.

---

Lu Yin

Lu Yin received his BS degree in chemical engineering (2017), MS degree in nanoengineering (2018), and is continuing his PhD study at the University of California San Diego, CA, USA, in Dr Joseph Wang's group of nanobioelectronics. His current research interests are on the development of wearable sensors, batteries, biofuel cells, displays, integrated electronic systems for energy-autonomous wearable applications towards continuous and accessable health management.

Kyeong Nam Kim

Kyeong Nam Kim received his MS and PhD degree in Materials Science and engineering from Ulsan National Institute of Science and Technology (UNIST), Korea. From 2018 to 2020, he worked as postdoctoral fellow at University of California, San Diego (UCSD), United States in Prof. Joseph Wang's group of nanobioelectronics lab. And he is currently a postdoctoral researcher in Korea Research Institute of Chemical Technology (KRICT), Korea. His research interests are on the development of nanomaterials and flexible, stretchable energy-electronic devices for their integration into self-sustainable wearable system and multiarray patch.

Alexander Trifonov

Alexander Trifonov received his BS And MS in Chemistry from the Hebrew University of Jerusalem where he was working on the development of enzymatic and DNA-based biosensors. In 2020 he earned his PhD in Engineering at ETH Zurich and continued to UCSD as a postdoctoral fellow in the laboratory of Nanoengineering, working on enzymatic biosensors and biofuel cells.

Tatiana Podhajny

Tatian Podhajny received her BS degree in nanoengineering at the University of California San Diego in June 2021. She is continuing there to pursue a masters in nanoengineering with focus in biomedical nanotechnology and research in Dr Wang's group. Her current research interests include developing flexible wearable sensors and drug delivery devices for closed-loop health management.

Joseph Wang

Joseph Wang is a Distinguished Professor and the SAIC Endowed Chair at the University of California, San Diego (UCSD). He is also the Director of the UCSD Center of Wearable Sensors. He served as the chair of the Nanoengineering department of UCSD and as the director of the Center for Bioelectronics and Biosensors of Arizona State University (ASU) before joining UCSD. His research interests are nanomachines, wearable sensors, flexible electronics, electrochemical devices, and bioelectronics.

---

Broader context

While many energy harvesters and storage technologies have been developed for powering of wearable electronics, their rational assembly into self-sustainable energy systems has yet to be discussed. Thus far, energy devices in special wearable form factors have not been widely seen in commercial electronics due to limited practicality and reliability. We believe that strategies of integrating energy harvesting and storage modules in wearable applications must be applied to endow energy autonomy, which offers easy maintenance, extended operation, and enhanced reliability. Various harvesters and storage modules have distinct advantages and limitations, and their respective energy/power ratings, characteristics, applicable scenarios, and form factors must be taken into account. Similarly, the development of various generators for renewable energy has promoted the development of microgrids – small energy grids integrating various generators, energy storage facilities, and loads using accurate systematic energy controls. This perspective points out the similarity between self-sustainable wearable systems and independent microgrids, summarizes key system-level considerations in designing smart and reliable wearable microgrids with dynamic energy prediction and budgeting, and envision the future roadmap for the development of wearable electronics.

Introduction

Wearable electronics has recently witnessed spectacular growth due to its tremendous potential for a broad range of applications, including human health monitoring, robotics, human–machine interfacing, and informatics.^1–5 The sensing of motions, biomarker fluctuations, hemodynamic activities of humans, and our interaction with electronics through visual, audio, or tactile elements, along with various means of wireless data transmission, are changing our lifestyles dramatically.^6–10 Such tremendous progress has led to the development of corresponding wearable energy devices for addressing the rapidly growing power demands of on-body electronics systems.^11,12 Correspondingly, emerging wearable energy harvesting devices demonstrated the possibility of harvesting energy from the human body (e.g., motion, sweat, body heat) or from its surrounding environment (e.g., sunlight, heat, moisture), freeing wearable devices from drained bulky and rigid batteries and tethered or close-range power transmissions.^11,13–23 We have since seen scattered examples of using wearable energy harvesters for powering physiological and biochemical sensors, displays, wireless data transmissions, microcontroller units (MCU), and system-on-a-chip (SoC).^10,24–31 However, while most related studies have focused on the characterization of individual modules, the discussion of their reliability, practicality, and sustainability as part of an independent powering system is still lacking. Such individual powering modules rely on monolithic and volatile energy sources, operate intermittently in limited, specific scenarios, and demand high-intensity activities for energy input. Such operation leads to low harvesting reliability and poor energy return of investment (EROI), all of which limit the reliability and practicality of self-powered wearable systems.^32,33

The strategic energy utilization thus represents a critical challenge for advancing self-sustainable wearable technology. Such a challenge requires a systematic assessment of integration strategies of diverse energy harvesting, storage, and management modules. Seeking inspiration from large-scale deployed renewable energy generators and their successful integration in modern power grids, we envision the possibility of establishing on the human body a self-sustainable energy network through the concept of microgrids. Similar to the challenges of deploying renewable energy generators, such as wind turbines and solar cells, to power small isolated communities, wearable electronics are facing reliability and consistency challenges.³⁴ The concept of microgrid has thus been introduced to combine power generation, storage, distribution, and consumption in one network, which links the supply of ample renewable resources to the demand of local communities through advanced controlling algorithms.^35–37 In the microgrid, hierarchical control systems are employed to ensure the compatibility of modules with diverse characteristics, along with dynamic prediction and budgeting of energy income, storage, and output, thus ensuring reliable and sustainable operation of the system.^34,37–39 Viewing the systematic integration of various wearable devices through the concept of microgrid, we see the growing need for applying similar strategies towards designing reliable self-powered on-body systems: the leverage of multiple generation mechanisms to diverse energy inputs, appropriate storage modules to rectify energy demands, and the rational selection of modules customized for the specific environment and use case.

In this Perspective, we discuss the concept of “wearable microgrid” and summarize the system-level design considerations in the modular integration of energy harvesters and storage devices using components with complementary characteristics and commensurate performance. To guide the judicious selection of components essential for maximizing the practicality and reliability of the system, we propose the complementary pairing of multiple wearable harvesters and storage devices that are use-case specific. We also summarize and compare the power and energy ratings of various wearable energy harvesting and storage technologies, along with the power consumption required for achieving different functionalities in the wearable sensing system, for selecting modules with commensurate performance and optimal energy budgeting. Next, we discuss key engineering considerations in integrating various textile-based and epidermal energy systems, toward the secured connection of modules with compatible form factors. We also summarize the limitations of current wearable energy systems and discuss expanding the wearable “microgrid” design concept into a broader context of miniaturized energy systems for implantable and ingestible electronic systems. Lastly, we share our perspective on future developments of smart wearable microgrids based on accurate dynamic energy budgeting, leveraging data and algorithm-assisted user-behaviour analysis, for responding autonomously to changing energy demands. In this perspective, we wish to avoid the lengthy discussions and comparisons of individual components in a wearable system. Instead, we would like to share our vision of wearable microgrid with its related system-level design considerations. Compared to existing device-oriented review articles and perspectives summarizing the novel materials and structures of wearable platforms (e.g., energy harvesters,^{14,18–20,40} energy storage devices,^15,20,41 sensors^17,42), this perspective aims at stimulating in-depth discussion of applying the microgrid concept towards the rational design of self-powered, self-sustainable wearable systems with maximized efficiency, reliability, autonomy, and practicality.

Concept of wearable microgrid

The concept of the wearable microgrid is originated from the traditional isolated, “island-mode” microgrid – a small network of various power generation units, energy storage units, hierarchical control systems, and loads that can operate independently from the main power grid.^36,43,44 There is a significant social and economic impact of developing such independent microgrids for powering remote villages, underdeveloped countries, islands, or even in large ships, which have limited access to established main power grids.^45–48 Even for entities such as colleges, hospitals, and large facilities with direct access to the main power grids, the establishment of microgrids can be highly beneficial for endowing a higher level of energy sovereignty against unexpected system power outages or shutdowns during high peak power demands.^48–50 Not only relying on small-scale fuel-based generators, independent microgrids often involve renewable distributed energy generators, such as solar cells, wind turbines, small hydroelectric systems, whose energy input are mostly stochastic and uncontrollable, and unable to meet the energy demand of the loads in the network at all times.^44,51 For example, solar cells generate most energy during the mid of a day with sufficient sunlight, whereas the peak power demand period for residential use often occurs in the early morning and after sunset. To meet the energy demands and avoid power outages, microgrids powered by solar cells require energy storage solutions that bridge the gap between energy production and consumption.⁵² Larger independent microgrid systems, with diverse users and various distributed energy sources, rely on several essential elements for ensuring the system reliability, including advanced algorithms for predicting the energy supplies and demands, hierarchical supervisory control systems that accurately monitor and regulate the energy flow at each level, as well as commensurate energy reserves for energy security (Fig. 1A).^37,44,50,51

Fig. 1 Concept of wearable microgrids and similarity to traditional microgrids. (A) The structure of a traditional microgrid includes distributed energy sources and energy storage facilities that are managed by supervisory controls to sustainably power various loads within the grid. (B) The structure of a wearable microgrid includes wearable energy harvesters and energy storage devices managed by the control system to power wearable applications for sensing, display, data transmissions, interfacing, and control, and ensure system reliability.

System-level considerations of such distributed networks can easily be translated to wearable devices. Usually, wearable devices do not have all-time access to our household power, as the loads stay on the human body and have change locations constantly. Current wearable devices, such as headphones, smartwatches, or various on-body activity sensing devices mostly rely on Li-ion battery-based energy storage solutions. Yet, the need for frequent battery recharges requires constant (often daily) maintenance, hence negatively affecting the users’ daily workflow.^53,54 With the rapid rise of the new class of wearable sensors, displays, and human-machine interfacing devices, a larger number of discrete wearable modules will be added to enable comprehensive health monitoring and human-machine interactions. Correspondingly, the energy demands of these modules need to be addressed individually, resulting in more complex daily maintenance routines. Similar to the end-users in a microgrid, such diverse demands of individual devices can benefit from an integrated energy management system that handles the power distribution for the wearable device, thus enhancing the energy reliability while reducing the system complexity.²⁴

Energy harvesters represent another important constituent of the wearable analogue of microgrids. The main incentive of including wearable energy harvesters in the system is to introduce additional energy inputs to extend the system runtime and eventually replace the need for recharging, leading to partial to full autonomy. To this end, numerous wearable power generators, leveraging various mechanisms, such as photovoltaics, thermoelectric, pyroelectric, osmotic, piezoelectric, and triboelectric effects, have been proposed^{11–23,40,55–57} With extensive re-engineering of materials, structures, and form factors, such as wearable energy harvesting devices have been transformed from rigid and bulky forms to their durable, flexible, bendable, or even stretchable and washable counterparts.^16,17,58 However, the availability of diverse energy inputs raises challenges in their systematic integration since the power, voltage, and pattern of the harvested energy differ largely from each other. Each harvester also features various drawbacks, such as dependence on direct sunlight, perspiration, or movements, that often limit their practicality. Such inconsistency is analogous with that of many renewable energy sources, such as solar cells and wind turbines, whose reliability in a microgrid requires systematic control, regulation, and judicious integration with energy storage devices. To fully utilize the discrete and diverse wearable energy harvesters, the systematic integration of such harvesters would not only be helpful towards generating larger amounts of power but will also allow to mitigate the limitation of each harvesting mechanism and hence to enhance the practicality of the entire system in different usage scenarios.

These considerations indicate that it will be extremely beneficial to consider all wearable devices and energy harvesters in terms of an integrated system. This integrated approach will not only reduce the energy demand complexity but will also improve the reliability of the energy supply. Using the analogy of a self-powered microgrid, we conceptualized the idea of “wearable microgrids” – an integrated system connecting multiple on-body energy harvesting and storage devices, along with control systems for power regulation and redistribution to various applications. As illustrated in Fig. 1B, the wearable harvesters, as the energy input, resemble the power generation stations in a microgrid, providing the system with diverse yet irregular energy. Controlled by the energy regulation modules, the energy input with various voltages and currents in different durations can be stored in suitable energy storage devices, which offset differences in the energy supply and demand throughout the operation. The control system further distributes the stored energy to power the wearable applications, considering their individually ranked priority along with a budgeted energy allowance. We envision that the rational design of a wearable microgrid with modularized components distributed around our body can achieve high energy generation and utilization efficiency, along with good reliability and practicality, towards complete system energy autonomy.

Although numerous wearable sensors and energy devices have been developed, wearable electronics (other than smartwatches and wireless earphones) have rarely been commercialized. One of the major challenges for widespread commercialization is the lack of energy-wise systematic considerations, which greatly limits the practicality and reliability of individual devices. Transitioning from designing individual wearable devices to a practical integrated wearable platform is a challenging step that requires interdisciplinary collaborations. To this end, in the following sections, we discuss three important concepts, including:

(1) the accurate budgeting of the energy flow;

(2) the scenario-based prediction of the system operation and the corresponding selection of system modules; and

(3) the specific considerations in module development, depending on various wearable form factors.

We believe that these concepts can guide the conceptualization of a practical wearable system with high energy autonomy and reliability.

Energy-centred system-level budgeting

The rise of wearable energy harvesters offers an attractive solution to the increasingly demanding energy requirements of wearable applications. The integration of various wearable harvesters, depending on the level of power output, can extend the runtime of devices or eliminate the need for recharges. Although a plethora of wearable harvesters has been developed, the specific requirements for the wearable form factors (such as softness, flexibility, stretchability, washability) restrict the potential structures and materials and hence may compromise the generated power. Currently, only a limited amount of energy (sub-mW level) can be scavenged from such wearable harvesters; this amount cannot support the continuous power for the majority of wearable applications, which run on electronics demanding tens to hundreds of mW. Realizing absolute energy autonomy requires that the energy input from the harvesters, energy reserved in the batteries and capacitors/supercapacitors, and the energy demand from the electronics, are budgeted and balanced accurately to ensure the system operates with maximum continuity and stability. Thus, when designing a wearable microgrid for a specific application, their operation requirement in terms of time and power must be calculated. Then, energy harvesters must be selected based on their corresponding power ratings and the appropriate usage scenario (will be elaborated in the following section), and their sizes will be determined by the power requirement of the applications. At this point, the possibility of achieving energy autonomy can be analyzed, leveraging basic Fermi estimation – the estimation of order-of-magnitude of the energy and power input vs. output. Lastly, suitable energy storage units with sufficient (but not excessive) capacity must be selected to ensure that the device can operate for the intended duration. On the other hand, if the energy harvesting modules in the microgrid are initially identified with the target of achieving energy autonomy, then the order-of-magnitude estimation is due to determining the functionalities that such systems can support. This is followed by similar accurate selections of energy storage units with capacity ratings that are appropriate for both the harvesters and the applications.

Energy input from wearable harvesters

Fig. 2 (centre) summarizes the range of power inputs from different harvesters and the corresponding power demands of various common wearable applications to guide the order-of-magnitude estimation. As wearable devices are usually designed in planar configurations to conform to skin or textiles, their performance is usually reported in terms of areal power density. Due to the limited area available on the human body, this criterion is important not only to compare the performance of different technologies but is also helpful for estimating the harvester footprint towards balanced energy budgeting.

Fig. 2 Power and energy ratings of wearable devices. Left: Power requirement of various wearable electronic loads. Middle: Power and daily energy rating of various wearable energy harvesting technologies. Right: Energy rating of battery and supercapacitors. Images of light-emitting displays, copyright, Elesvier, 2018¹¹⁸ and Springer Nature, 2021.¹⁰ Refresh display images are adapted from ref. 24 and 32. Copyright, Elsevier, 2021.³² Stretchable battery images, copyright, Wiley VCH, 2017.¹⁴² Supercapacitors, images, copyright, Elsevier, 2018¹⁴⁷ and American Chemical Society, 2016.¹⁴⁶

Among different energy harvesters, photovoltaic solar cells (PVs) can provide the highest power density.⁴⁰ Considering the power conversion efficiency of ∼5–30% and the fact that the solar radiation on the earth's surface can reach a magnitude of 10² mW cm⁻², wearable solar panels can provide energy in the order of 10⁰–10¹ mW cm⁻² under direct outdoor sunlight.^5,59–62 It is worth noting that such performance is strongly dependent on many factors, such as time or weather, that determine the transmittance and the angle of incidence.⁶³ Hence, the harvested energy by wearable PVs is usually an order of magnitude lower in their daily usage. In most cases, the fraction of time people stay in the indoor environment is considerably longer than that of the outdoor. The indoor illumination level is significantly lower due to indirect sunlight with the filtering of windowpanes and glasses. In indoor settings, natural lighting can only provide ∼10¹ mW cm⁻² of irradiance, whereas artificial lighting provides a significantly lower irradiance level (10⁻² mW cm⁻²) with a narrower spectrum.⁶⁴ Thus, in reality, in indoor settings, wearable PVs present only a power rating of 10⁻¹–10⁻³ mW cm⁻² level, depending on the light source, setting their energy levels to a similar order as of other types of wearable harvesters. As PVs are used to harvest energy from the light, which usually comes from above, it is recommended to assume only a maximum device area of 10¹–10² cm² suitable for wearable form factors, which translates to the power of 10¹–10² mW outdoor and 10⁻¹–10¹ mW indoor per device, and to a maximum total of 10² mW h of energy throughout a day.

Wearable fuel cell energy harvesters represent a class of energy conversion devices that consumes fuel via catalytic reactions to generate energy. Some efforts have been made to convert hydrogen, methanol, or ethanol fuel cell into wearable form factors. However, such devices cannot be classified as wearable harvesters due to their need for external fuel and hence are not discussed here.^65,66 Wearable biochemical energy harvesting relies on enzymatic biofuel cells (BFCs) and microbial fuel cells (MFCs), which leverage bioelectrocatalytic reactions that consume metabolites present in body fluids to generate power.^67–71 Among these, sweat lactate-based BFCs have the highest reported power density (level of a few mW cm⁻²) due to the high sweat lactate concentration (2–50 mM). Theoretically, the maximum sweat rates of adults can reach the order 10⁴ mL per day, which translates to a total daily harvestable energy of 10³ mW h from lactate alone.^72,73 In addition, a wide range of other metabolites (e.g., glucose, urea, alcohol) present in various biofluids (e.g., saliva, tears, urine) can also be used as the source of energy, which further enhances the total amount of harvestable energy.⁶⁸ However, several other factors limit the power of wearable BFCs, including the collection of sweat, the fuel dilution due to excessive sweating, the limited reaction rates, as well as the stability and biocatalytic activity of enzymes in given operating conditions. Currently, the characterization of the BFC power primarily uses scanning voltametric methods, which report unrealistic high power, which can be 1–2 orders of magnitude higher than their actual power when operating over an extended time.^24,32 In general, as the lactate concentrations and sweat rates vary per different users, the actual obtainable power by BFCs is only assumed to be 1–10⁻² mW cm⁻². The large area of skin on the body can theoretically permit a device footprint ranging in 10⁰–10² cm², translating to the total power output of the device as high as 10² mW, with a total harvestable energy of 10³ mW h per day. Such level is along with the same order of magnitude as that of its theoretical maximum harvestable energy. Recent progress also demonstrated the continuous energy harvesting using passive, thermoregulatory sweat, which is extremely attractive due to its high energy return-on-investment and the possibility for day-long operation without the need for exercise, even while sleeping.³²

Piezoelectric and triboelectric nanogenerators (PENGs and TENGs) are among the most popular wearable energy harvester research topics.⁷⁴ Such energy harvesting relies on a charge exchange induced by various sliding, bending, stretching, and tapping motions in connection to a wide range of materials. As a result of such a high degree of design freedom, various form factors of wearable PENG and TENG harvesters have been developed. These nanogenerators can be located in different body locations to scavenge the kinetic energy from activities such as breathing, finger moving, running, or stepping. The generation mechanism of PENG relies on the strain-induced charge redistribution within the lattice of intrinsically polarized material, and has been discussed extensively in several reviews.^75–77 In their wearable implementations, PENGs typically deliver unrectified peak open-circuit voltages of 10⁰–10¹ V, peak short-circuit current density up to 10⁻¹ mA cm⁻², and a peak power of up to 10⁰ mW cm⁻².^18,77–79 It is worth noting that such peak power only occurs transiently upon applying and removing the strain on the PENG; thus, the time-averaged power generation is highly dependent on the load pairing, force applied, and frequency of movements, giving only up to 10⁻² mW cm⁻² under regular biomechanical inputs from users. Generating power by electrostatic charge exchange between two surfaces with different charge affinities, TENGs delivers superior performance, generating peak open-circuit voltage up to 10³ V, peak power up to 10⁶ mW, and charge density reaching 10² nC cm⁻².^79–83 Considering the average human movement frequency within the order of 10¹ Hz, with proper rectification and impedance matching, the averaged power (total energy over time) of such harvesters can reach the level of 10⁻²–10⁻¹ mW cm⁻² during active usage at certain locations of the body.^82,83 Considering the numerous locations on the body that can implement the TENG and PENG harvesters, we can safely assume the total device footprint to reach up to 10¹–10³ cm². A high-frequency and high-intensity movement can only account for a maximum of a few hours throughout a day, followed by mostly low-frequency and low-intensity dynamics that result in a lowering of 1–2 orders of magnitude in power generation. Therefore, the total amount of energy harvestable throughout a day via PENG and TENG can be estimated on the 10¹–10² mW h levels.

Thermoelectric generators (TEG) have been investigated for energy harvesting since the discovery of the Seebeck effect about two centuries ago.⁸⁴ Relying primarily on the junctions of n-type and p-type semiconductors, the TEGs are usually built with complicated serial-connection structures and rigid, bulky materials.⁸⁵ Recent progress on polymeric thermoelectric materials, as well as advances in printing fabrication techniques have resulted in TEGs with high flexibility, stretchability, and even self-healing functionalities, making them more suitable for wearable devices.^22,86–88 Wearable TEGs can generate a low voltage from the temperature gradient between the human body and the ambient environment and can thus continuously harvest energy as long as the temperature gradient persists. However, the power output of TEGs is generally low, delivering 10⁻³ mW cm⁻² power in a typical ambient environment (ΔT ∼ 10⁰–10¹ °C).²² Considering the large possible device footprint of 10²–10³ cm² and the day-long harvesting duration, the total amount of energy obtainable with TECs is only 10⁰ mW h.

In addition to these four types of prominent wearable energy harvesters, other types of harvesters have also been developed, including electromagnetic generators (EMGs) and magnetoelastic generators that generate power using the movements of permanent magnets,^89,90 pyroelectric nanogenerators that generate power from temperature fluctuations,^28,91,92 antennas that harvest from either directed long-range power delivery or radio-frequency radiations.^93,94 Due to the scope of this perspective, these harvesters are not discussed in detail, although similar methods of order-of-magnitude calculation can be applied to estimate their energy input to a wearable microgrid.

Energy demands of typical wearable applications

In general, most wearable applications can be classified as either control, sensing, displaying, and wireless communications. The operation of such electronics typically runs on integrated circuits (ICs) that: (1) rectify and regulate the energy input; (2) manage the power and voltage of the energy input and output; (3) use analogue-to-digital converters (ADC), digital-to-analogue converters (DAC), an amplifier for signal generation and processing; (4) execute, store, and compute codes, programs, and data; (5) transmit data or information wirelessly to other devices; (6) controls the connected displays for visual interactions. Depending on the functionalities of a wearable system, the power consumption can vary drastically over an extremely wide range of pW to W.^11,12,95,96 Knowledge of system-level power consumption is critical for balancing the power input and power output towards establishing a self-sufficient energy network. Conversely, knowing the system-level energy input is crucial also for budgeting the known amount of energy to support various functionalities with the desired amount of runtime. Fig. 2 (left) maps the general power consumption of various electronics for realizing different wearable applications.

As the heart of most wearable electronics, system-on-chips (SoCs), and microcontroller units (MCUs) enable a wide range of functionalities to execute from simple instructions to complex programs. In this context, MCUs refer to controllers with limited functionalities, memories, and input/output (I/O). It is commonly used in electronics with simple functionalities with small, embedded control systems, typically featuring low power consumption in the order of 10⁻¹–10³ mW when active for commercial MCUs.^97–99 Customized MCUs with very specific functionality can further reduce the need for unnecessary memories, processing speed, and I/Os, which can enable the power in the level of 10⁻¹ mW when active and 10⁻⁷ mW when in low-power/sleep modes. SoCs is a more encompassing term describing ICs with central computing units (CPUs), graphics processing unit (GPU), memory, and/or signal processing functions embedded, which features more computing power, I/O, and programmability. It is commonly seen on electronics with programmable interfaces or even operating systems and is widely used on smartphones, smartwatches, and even some personal computers. Although its power consumption is similar to that of MCUs, SoC is usually coupled with more peripherals, which leads to higher power consumption as a system. As a rule of thumb, the clock speed of the processors is positively correlated with their input voltage and current; thus, a faster processing speed would require larger power consumption. For sensors requiring a high data sampling rate, resolution, and signal-to-noise ratio (SNR), the integrated ADC and DAC can consume a significant amount of power.

Sensors stand at the forefront of wearable applications as they enable direct interaction of the human body with electronics, monitoring and translating a myriad of signals within and around our body. Sensors for the monitoring of touch, motion, temperature, chemicals, light, physiological signals, and biomarker levels have been developed in the wearable form factors, featuring a wide spectrum of signal transduction methods. In general, the sensors vary in their electrical properties (e.g., resistance, impedance, capacitance, voltage, current), or their optical properties (e.g., colour, transmittance) in response to the chemical or physical changes, and these analogue signals are eventually converted into digital electrical signals for further processing.^3,5,100,101 Depending on the type of signal transduction methods, the operation of the sensor results in drastically different power consumptions. Some sensors require high SNR and resolution to obtain high sensitivity but not high temporal resolution. For example, many electrochemical sensors measure nA–pA current signals at low time intervals (e.g., every few minutes) and thus require converters with high resolutions.^102,103 Other wearable sensors, based on acoustic or optical transducers, or electrocardiogram (ECG) sensors for physiological signal monitoring, require high temporal resolution and thus, sample thousands to millions of data points per second, which poses high demand on processor clock speed and system memory.¹⁰⁴ As another example, wearable devices with integrated optical sensors or cameras for image processing (e.g., bar-code scanning, feature recognition) can have high power demands associated with the graphical computing function. Many self-powered electrochemical sensors have been proposed to generate an analytical response with no energy input, which somewhat reduces the system complexity and power consumption.^92,105,106 In some rare cases, the self-powered sensor can harvest sufficient energy to power the electronics, which endows system energy autonomy.^25,107–110

As a channel for controlling the electronics and interfacing with the obtained data, wireless communication and displays are critical components of wearable electronic systems. The power consumption of the wireless module largely depends on the range and data size of the transmitted signal.^111,112 Bluetooth, which is widely used for data transmission over distances of tens of meters, consumes power in the range of 1–10¹ mW.^113,114 Less common in low-power wearable devices, WiFi and cellular technologies enable longer transmission range along with higher data speed and 10¹–10³ mW energy consumption range. Alternatively, near-field wireless technologies, such as near-field communication (NFC) and radio frequency identification (RFID), in addition to untethered power delivery, can also be used for data transmission that extracting power from the external reader and require no internal power supply; however, the readers require 1–10³ mW of power and such systems in general lack operational independence.^115,116 Displays are another critical element for many wearable devices as they offer direct visual interaction with users. Among them, active displays requiring light-emitting elements (e.g., liquid crystal displays (LCDs) with backlighting, quantum dot displays, electroluminescent displays, or light-emitting diodes (OLED)) consume a significant amount of power, in the range of 10²–10³ mW in the case of wearable electronics.^10,117–119 As low-power alternatives, low-resolution LCDs, electronic-ink (e-ink) displays, or electrochromic displays (ECD) consume significantly less power, down to the 10⁻³–10⁻¹ mW level, and thus provide a very attractive and elegant designing solution as self-sustainable wearable electronics.^32,120–122

Power regulation and energy storage

In a wearable microgrid system, energy management systems (EMS) are crucial for efficient energy regulation. Similar to an electric microgrid, a hierarchical control system is usually required, in which regulators regulate and maximize the power input from various wearable harvesters, battery manage circuits that control and monitor the currents and voltages in and out of energy storage modules, and controllers adjusting the availability, duration, and interval of different services of the wearable applications.

The regulator modules are usually dedicated for each type of energy conversion mechanism to regulate their current and voltage inputs to be compatible for storage in the energy storage modules or direct powering of wearable applications. For common low-voltage direct current (DC) harvesters, such as BFCs and TEGs that delivers sub-1 V input, a boost converter is commonly employed to elevate the voltages to levels that are compatible with the wearable applications in exchange of lower current. For high voltage DC harvesters, such as PV groups that typically deliver 12 V or 24 V input, buck converters that reduce the voltages in exchange for higher current can be used. Typically the use of these DC–DC converters involves some energy loss due to the non-ideality of the diodes, inductors, capacitors, and switches in the circuits.^123,124 For wearable electronics that typically delivers low current, step-down buck converters typically deliver lower efficiency with higher input voltages, whereas step-up boost converters deliver lower efficiency with lower input voltages; notably, for harvesters that deliver sub-mA level current, commercially available DC–DC converters can suffer from 50–90% energy loss.^125,126 The DC energy harvesters typically has optimal loads or operating voltages which allow maximum power output, correspondingly, the regulating circuits can employee maximum power point tracking (MPPT) algorithms that automatically and dynamically adjust the output by perturbing the output or measuring the harvesters’ conductance or parasitic capacitance.^127,128 For alternating current (AC) input harvesters, such as PENGs, TENGs, and EMGs, bridge rectifiers are typically used to convert AC inputs into DC inputs for storage. To further stabilize the rectified but rippling single-directional input, simple resistors and capacitors (RC) circuit can be used to reduce the ripples and stabilize the output voltage. More advanced power management modules were proposed that can further enhance the energy harvesting efficiency from the high-voltage, low-current PENGs and TENGs, including the use of serial and parallel switch mechanisms, inductor and capacitor (LC) circuits, transformers with primary and secondary coils for reducing voltage, and spark switches that reduce current leakages.^83,129–131 Similar to the DC harvesters, such AC harvesters also deliver higher power with proper impedance matching, which can be simulated using their corresponding equivalent circuits.^132,133 The regulated energy can thereafter be stored in capacitors, supercapacitors, and/or batteries, which are then discharged to power electronics.

Among the different energy storage modules in wearable devices, capacitors are the most essential element for circuits, allowing to regulate power and filter signals. Capacitors are typically integrated with the circuit, have the capacitance of 10⁻⁹–10⁻² F, and are rated at tens to hundreds of volts. Capacitors used for storing energy are typically rated at 10⁻⁴–10⁻² F and can store energy in the range of 10⁻⁴–10⁻¹ mW h. Such energy level can be used to power electronics only for a brief period (<1 s) and is typically recharged repeatedly throughout one usage session.^26,27,32 Supercapacitors can hold a significantly larger amount of charge, rated with the capacitance of 1–10⁴ F, albeit at a lower voltage (typically <5 V).^134–136 Wearable supercapacitors are typically limited by their form factors and can only store up to a few F, thus storing only 1–10¹ mW h of energy. Such energy levels are sufficient to power low-power electronics during a moderate amount of time (minutes to hours). In comparison, batteries feature significantly higher energy density, with their wearable version typically hold 1–10¹ mW h per cm².^137–140 Batteries can thus store up to 10³ mW h of energy with a device footprint of 10³ cm², which is sufficient for many wearable electronics in the market. Controlling circuits are usually necessary to control the voltage and current in and out of the wearable energy storage modules, and hence to avoid overcharge or over-discharge, which are detrimental to the device's performance and may pose safety concerns.^134,141 Depending on the battery chemistry, the maximum, minimum and nominal voltages are different, and should be closely managed. Typical Li-ion batteries operates between 2 V and 4 V, with the overcharging resulting in risk of thermal runaway and over-discharging resulting in capacity degradation, current collector corrosion, and internal short circuiting; rechargeable aqueous batteries typically operate at lower potential between 2 V to 0.5 V, although having less risk of explosion, still face the risk of electrolyte breakdown and gas evolution if overcharged or over-discharged due to the limited electrochemical window of water (1.21 V). The energy ratings of both energy storage technologies are represented in Fig. 2 (right); currently, many micro-SC, SC, and batteries in flexible and stretchable form factors have been developed for powering various wearable applications.^{137,142–148}

Accurate energy budgeting and selection of commensurate energy storage modules are crucial for enabling self-sustainable operation. A system should select storage modules with sufficient capacity to hold the harvested energy over the target use time. However, the excess capacity will result in a large device footprint, high self-discharge, long charging time, and is generally inefficient.²⁴ On the other hand, insufficient energy storage will bottleneck the power demand and operational runtime of the electronics, hence limit the possible functionalities of the system. As an example, for a wearable system integrated with 20 mW solar cell to harvest energy for 2 h in a day-long operation, a battery rated at 50 mW h would be sufficient to power an SoC with 8-bit ADC sampling and Bluetooth connectivity with a low-power e-ink display. Alternatively, for a wearable system integrated with a 5 μW thermoelectric harvester that harvests energy continuously throughout a day, a capacitor of 2–5 mF is sufficient for energy storage to briefly power an MCU-controlled thermometer several times every hour. A budget list calculating the power consumption and energy demand of the applications and the energy income from all harvesters within a usage period can not only guide the selection of compatible components but is also necessary for assessing the reliability of the entire system. Envisioning a typical usage scenario that reflects the peak demand and peak supply, along with the required energy storage to offset the gap between the two, towards the system's reliable continuous operation, as will be discussed in the following section.

Scenario-specific design based on complementary characteristic

Towards maximizing energy generation, the deployment of renewable energy was always coupled with thorough investigation on the corresponding meteorological, hydrographical and geological resource availability and its potential environmental impacts. While energy diversity is generally beneficial to the stability of a microgrid, the installation of various types of energy harvesters should always match realistic scenarios, e.g., the weather, usage pattern, land availability, module affordability, and ease of maintenance.^35,37,52 Such considerations are critical to guide the selection of modules and the design of the microgrid. Similarly, the scenario-specific selection of energy harvesting and storage modules is of tremendous importance towards improving the efficiency, reliability, and practicality of any autonomous system. Aside from energy budgeting considerations discussed in the previous section, factors such as the location of the device, available area, user's activity pattern, the surrounding environment, and impact on user's daily workflow are to be considered when proposing and designing a new integrated wearable system. Furthermore, depending on the characteristics of various components in a system, the coupling of complementary harvesters and/or storage modules can result in additive or synergistic behaviour, which can be highly desirable.

Currently, several studies have showcased the activity-specific operation of self-powered systems.^{13,24,26,27,30} These systems were coupled with low energy demand applications and only operate for a short amount of time (from tens of minutes to a few hours), thus having relatively low system energy rating requirements. In this context, self-powered sweat-sensing offers great examples for the scenario-specific design of a wearable system (Fig. 3A). The generation of sweat for subsequent sensing usually requires high-intensity, sweat-inducing exercises. Thus, BFC, which harvests energy from the sweat, and various motion-based harvesters (e.g., PENG, TENG, EMG), can be integrated into the same system compatible with the activity. Similarly, motion-induced energy can also be used for various physical sensors towards gait monitoring, pedometers, or motion tracking, which the energy harvesting is compatible with the usage scenario. Applying a similar strategy, one can also envision a system using PV and TEG to harvest energy from the radiation and thermal energy of the sunlight.²⁸ With sufficient power harvested, we can envision the use of such power for wearable Peltier cooler patches for personal temperature regulation on a hot, sunny day. Such scenario-specific selection of modules in a wearable system does not only enhance the practicality of the system but also reduces the system complexity and disruptions to the user's workflow.

Fig. 3 Complimentary, scenario-specific selection of modules. (A) Examples of the complementary pairing of wearable harvesters and applications in their proposed scenario, and synergistic pairing of energy harvesters for a proposed use case. (B) A proposed power supply and demand and energy reserve for an urban office worker. (C) A proposed power supply and demand and energy reserve for a soldier deployed in a mission.

Most current energy harvesting studies are focusing on energy harvesting from movements, based on various motion-based harvesters (e.g., PENG, TENG, EMG) that generate charge from various sliding, tapping, stretching, and shaking movements.^92,109,149 The deployment of such harvesters at different body locations can fully utilize the limited area and generate a sizable amount of energy. For example, an E-textile shirt for a running session can integrate the TENG modules on the side of the torso and the EMG modules on the wrists to harvest energy both from the sliding between the arm and the waist as well as from the swinging of the arm itself. Both TENG and EMG additively contribute to biomechanical energy harvesting using natural body movements, maximizing the overall efficiency. Moreover, in some scenarios, the coupling of two harvesters scavenging different types of energy, can act synergistically, offering additional benefits to the system. As an example, wearable PVs can be coupled with TEGs that harvest thermal gradients induced by the sunlight, thus delivering additional power using the same device footprint.³⁰ In a previous report, Yin et al. have demonstrated the pairing of TENG and BFCs, which harvest from the motion-induced biomechanical energy and the biochemical energy from the generated sweat, respectively.²⁴ Besides the additive effect of combining two harvesters, such complementary coupling of BFC and TENG offers advantages of faster system booting (compared to BFC along) and more extended harvesting (compared to TENG along).^24,150 The integration of multiple harvesters ensures the diversity in energy source, which enhance the reliability of a system when part of the energy input becomes unavailable. Furthermore, the coupling of complementary harvesters can address the limitations while amplifying the advantages of individual modules. Table 1 summarizes the advantages and limitations of wearable energy harvesting and storage technologies in different scenarios, which are crucial characteristics in establishing a synergistic, complementary wearable microgrid; accordingly, their maximum power and energy ratings and their applicable scenarios are also summarized.

Table 1 Characteristics of various wearable energy harvesters

	PV	BFC	PENG	TENG	TEG
Merits	• Harvest solar energy	• Harvest biochemical energy	• Harvest biomechanical energy	• Harvest biomechanical energy	• Continuous power
	• Continuous power	• Continuous power	• High voltage	• Less dependent on external environment	• DC output
	• High output power	• Less dependent on external environment	• Less dependent on external environment	• Wide material selections, low-cost	• Unsensitive to surface contamination
	• DC output	• DC output	• Unsensitive to surface contamination	• High voltage	• Large-area compatible
	• Large-area compatible		• Large-area compatible	• Large-area compatible
Limitations	• Low-performance indoor or without direct sunlight	• Low voltage	• Pulsed AC input	• Pulsed AC input	• Low voltage
	• Highly dependent on external environment	• Expensive material	• Require constant movement	• Require constant movement	• Low performance
	• Biocompatibility	• Limited fuel availability	• Limited material selections	• Sensitive to surface contamination/damage	• Limited material selections & biocompatibility
		• Enzyme stability	• Biocompatibility		• Complex structure
		• Sensitive to surface contamination			• Highly dependent on external environment
Max. power	• 10^0 mW cm^−2	• 10^0 mW cm^−2	• Up to 10^−2 mW cm^−2	• 10^−1 mW cm^−2	• 10^−3 mW cm^−2
Max. energy per day	• 10^1–10^2 mW h	• 10^3 mW h	• 10^1 mW h	• 10^2 mW h	• 10^0 mW h
Use-case	• Outdoor preferred	• Indoor & outdoor OK	• Indoor & outdoor OK	• Indoor & outdoor OK	• Indoor/outdoor (with temperature difference)
	• Stationary (sitting, standing)	• Rigorous full-body exercise	• Rigorous & moderate exercise	• Rigorous & moderate exercise	• Stationary (sitting/standing/sleeping)
	• Active	• Stationary (sitting/standing/sleeping)	• Localized movement	• Localized movement	• Active

---

Future development of an integrated system will focus on a more diverse system, integrating additional energy harvesting and storage modules, thus ensuring its day-long, week-long, and eventually year-long autonomous operation. Such a system will thus incorporate more scenario-specific considerations, as the pattern of a user's daily life will determine its energy supply and demand. Fig. 3C illustrates the energy flow in a proposed wearable system for an urban office worker throughout the day. We can see that in such use case, the user switches between different scenarios, from active to sedentary and from indoor to outdoor settings; correspondingly, the energy supply from several energy harvesters and the energy demand from a few wearable applications vary significantly throughout the day. Considering such a use case, the system will benefit significantly from diverse energy sources that ensure constant energy supply to the system, as well as the inclusion of supercapacitors or batteries that regulate the energy flow and offset the differences between the supply and demand. In another more extreme example, illustrated in Fig. 3D, involving a soldier deployed in a desert for a week-long mission, the energy supply can be drastically different due to the extreme environment. The corresponding microgrid design is optimized for such a scenario, focusing primarily on harvesting solar energy and sweat-based BFC bioenergy during the day and on high-capacity energy storage that ensures continuous operation of the device overnight.

Currently, self-sustainable wearable systems with an extended operation time do not exist due to the low energy/power rating, practicality, durability, and lack of compatibility among the modules. The limited power from the current wearable harvesting technologies thus demands the judicious, scenario-specific selection of modules to maximize the energy supply. Energy harvesters that can operate in all scenarios, independent of the external environment, and generate power regardless of user activity are highly desirable, as they will greatly improve the system's practicality and reliability. Recently, BFCs that harvest energy from passive thermoregulatory sweat from fingertips and operate in most of the scenarios, even during sleep, were developed and were thus considered a major improvement to the practicality of wearable energy harvesters.³² Future technological advances in developing more complementary harvesters and scenario-specific deployment of such harvesters will aid the realization of more practical wearable microgrids.

Compatible form factors towards practical wearable systems

The form factors of wearable devices are possibly the most well-considered aspect, as the majority of research reports predominantly focus on showcasing the flexibility, stretchability, durability, and washability of the developed wearable devices to effectively attract the public's attention. For designing a wearable microgrid, the form factor considerations are certainly of great importance, as they are deterministic to the wearable system's applicable scenarios, durability, and practicality. As many previous reviews have already discussed in length the advances in novel form factors in various wearable harvesters, storage devices, and applications,^{5,11,25,151–155} this perspective aims to discuss only briefly a few key points in logistically determining the necessary form factors in an integrated system.

Wearable electronic systems can be roughly discriminated into three groups: standalone rigid wearables, skin-based wearables, and textile-based wearables. Currently, the standalone rigid wearables are dominating the consumer wearable electronics market, with the greatest number of shipments in smartwatches, fitness trackers, wireless headphones/earbuds, hearing aids, virtual/augmented reality gears, and wearable cameras.^156,157 The rising market of industrial internet-of-things and medical wearables has also shown rapid development, delivering point-of-care products such as continuous glucose monitoring devices, insulin pumps, and drug-delivery patches for patients, or heated clothing, wearable scanners, and exoskeletons for users in special industries.^158–160 In addition, wearable electronics in less-common form factors, such as glasses, contact lenses, mouthguards, neckbands, or shoes, have also been investigated.^93,161–165 Such wearable devices have seen tremendous advances in functionalities and their market growth over the past decade and have fuelled the excitement of the public. However, such standalone technologies mostly lack the desirable flexible, stretchable, and conformal features expected for next-generation wearable electronics. In addition, this current generation of wearable electronics is still relying on rigid electronics, powered by Li-ion or disposable batteries, acting as foreign devices that we mindfully add to our daily workflows. Instead, many academic researchers would envision the next-generation wearable electronics as integrated electronic systems that are soft, unintrusive, intimately and seemingly integrated with the human body, physically unnoticeable, and logistically non-disruptive to the user's daily activities. To this end, many studies have chosen platforms that are already inseparable integral to people's daily lives – namely, the skin and the textiles.

Epidermal wearable electronics, sometimes also referred to as electronics skins, have grown with exciting development in biochemistry, material science, and structural engineering.^12,58,166 Conformal devices with mechanical properties similar to that of skin featuring novel structures or intrinsically stretchable materials to accommodate for the curved skins undergoing constant deformations have been developed.^167–171 Prominent examples include epidermal sensors worn as tattoos or patches for monitoring sweat and interstitial fluid biomarker fluctuations, physiological activities, physical movements, or surrounding environments (Fig. 4A).^{7,170,172,173} Similarly, skin-worn energy harvesters and energy storage devices activated by sweat have been proposed, providing potential energy solutions for the aforementioned sensing activities.^68,153,174 LED display on skin has also been reported, featuring futuristic visual elements on skin.¹⁷⁵ In general, such epidermal electronics often have a rather strict requirement on the mechanical modulus (10⁴–10⁶ Pa, similar to that of skin) to ensure device conformity.^176,177 In addition, ca. 10–30% stretchability would be required to ensure the device's uninterrupted operation during movements.^167,178 Safety is another important consideration as epidermal electronics require direct contact to the skin for an extended amount of time; sweat and other biofluids are typical of high salinity and corrosive; hence, such epidermal devices would require materials with high biocompatibility and chemical stability to avoid leakages or damage to the skin and the electronics.¹⁷⁹ Practically, epidermal electronics are designed to be used for a limited amount of time, which requires special consideration during product development. Due to the lack of battery technologies that meet these requirements in conformity and safety, such devices can only operate with wired connections or near-range power delivery, thus limiting the user's mobility and the device's level of independence. Furthermore, typical epidermal electronic devices are constructed onto a patch-sized geometric area (10⁰–10¹ cm²) which limits the area and types of modules that can be integrated. Lacking secure connections between individual patches, energy harvesting, storage and sensing distribution around the body also become challenging and require special engineering solutions. Owing to the above limitations, most epidermal systems are dependent on external devices and connections and used for stationary users, with only a few examples demonstrating a well-constructed miniaturized autonomous system, capable of energy harvesting, storage, and sensing.

Fig. 4 Wearable microgrid form factor considerations. (A) Form factor considerations for E-skin epidermal microgrid systems. (B) Form factor considerations for E-textile-based microgrid systems.

Alternative to epidermal electronics, smart electronic textiles (e-textile) expand the platform of wearable electronics from our skin to clothing (Fig. 4B). Textile-based wearable systems feature many advantages compared to their epidermal counterparts, including larger usable area, diverse types of fabrics, well-developed textile engineering technologies, and wider applicable use cases. The fabrication of textile-based wearable devices can be classified as “bottom-up” and “top-down” methods, where the former refers to the fabrication of functional devices into yarns and threads that thereafter weave, sew, or embroider into fabrics, while the latter refers to the direct printing, adhering, or stitching functional devices onto or into the textiles. Using these methods, a myriad of energy harvesters (e.g., PVs, TENGs, PENGs, TEGs, BFCs), batteries, SCs, physical and biochemical sensors, wireless antennas, and large-area displays have been developed as functional textiles and integrated e-textile systems.^{10,18–20,24,138,142,180–185} Endowed by various microstructures and patterns within textiles, textile-based electronics can feature flexibility and stretchability even without developing intrinsically stretchable materials.⁵⁸ Well-distributed wearable energy harvesting, storage, and applications around the body can be easily achieved through connections within the textile, thus allowing a wired but untethered system. Numerous textile-based wearable systems have been demonstrated, featuring the sensing, display, wireless communications powered by the integrated textile-based energy harvesting and storage devices. As textile-based electronic systems are intended to be used in normal clothing, the textile form factor poses strict requirements in terms of the durability of the devices against friction, repeated flexing, stretching, and washing while maintaining the softness and breathability of the textiles.

Overall, to ensure form-factor compatibility, one should firstly keep in mind the distinctions between epidermal and textile platforms and select the most appropriate platform and device architecture based on the use-case-specific requirements (e.g., reusability, energy consumption, device footprint) of the wearable applications. The second step requires the preliminary energy budgeting of the modules under the proposed applicable scenario to establish a balanced energy supply and demand within the system. The third step should involve selection of the material, structure, and fabrication process of each module, ensuring that the key requirements for the system (e.g., power, flexibility, durability) are satisfied by each individual module as well as the interconnections between them to remove any “Achilles’ heel” in the system. With preliminary performance of modules characterized, these steps should be reiterated to make adjustment on the use of materials, structures, sizes, etc., which leads to updated performance data for further adjustments. Only through such iterative process, a wearable microgrid system with compatible form factors, commensurate energy rating, and complimentary device characteristics can be obtained.

Outlook and summary

Through the above discussion of a wearable microgrid, this perspective aims to stimulate system-level discussions in designing an energy-autonomous wearable system. The integration of three pillars of wearable microgrids – the energy harvesting devices, the energy storage devices, and the applications – is to be implemented using key design considerations – accurate energy budgeting, scenario-specific complementary characteristics, and compatible form factors – towards practical execution of a seamlessly integrated system. In addition, efficient and reliable cooperation among modules requires systematic management that predicts, balances, and regulates the energy supply and demand of individual modules in the system. Such awareness of system-level energy considerations is crucial from the very beginning of building a wearable microgrid system, since the rational selection of modules depends strongly on its use case, specific energy supply and demands, and form factor requirements. The critical understanding of the energy ratings of the various components in specific scenarios will allow both retrospective planning of energy harvesting and storage modules for meeting the energy demands of wearable electronics, as well as the prospective feasibility assessment of adaptable applications based on the maximum energy supply from the harvesters. Considerations of the use-case and scenario of the system will further aid the accurate energy budgeting accounting for different usage patterns while simultaneously helping to determine the necessary form factors for its reliable on-body operation. The three key considerations are thus intertwined in establishing an efficient wearable microgrid towards the rational design of self-sustainable wearable energy systems (Fig. 5A).

Fig. 5 Summary and prospects of wearable microgrid systems. (A) Key component and corresponding design considerations in a wearable microgrid system. (B) Established smart wearable microgrid with advanced algorithm to improve system energy reliability. (C) Proposed developmental roadmap for energy-independent wearable electronics.

Recent advances in integrated wearable systems have demonstrated the possibilities in establishing such wearable microgrid systems. However, their future advances are hindered by the low performance of wearable energy harvesters that bottlenecked the overall energy demand of a system, resulting in a large mismatch in the power rating to various wearable applications. To bridge this major gap between energy demand and supply, major improvements are needed to further enhance the performance of existing energy harvesting technologies, expand the type of harvesters while reducing the power consumption of various wearable sensors, ICs, wireless technologies, and displays. Specifically, the improvements on wearable harvesters will rely on several aspects of technological advances, including the development of new materials featuring higher performance and stability, discovery of new synthesis and fabrication methods leading to lower cost, better scalability and biocompatibility, as well as research into new device structures and architectures for improved efficiency and robustness. Beyond incremental performance improvements, new exciting breakthroughs in proposing new use-cases and applicable scenarios of existing energy harvesters and the discovery of new energy conversion mechanisms can open new doors towards bioenergy harvesting with drastically higher practicality, energy return-on-investment, and performance. Similarly, wearable energy storage modules suffer from trade-offs between electrochemical performance, mechanical performance, and safety, which can be addressed with material and structural innovations. Efforts towards developing advanced and sophisticated fabrication processes for wearable batteries and SCs, compatible with such new materials and structures, will improve their scalability, reliability, and cost-effectiveness. In addition, wearable energy storage beyond electrochemical cells has rarely been explored, which may grant new possibilities in advanced energy regulation strategies for wearable microgrids. The addition of new and more powerful wearable harvesters and storage devices will greatly diversify the energy sources within a wearable microgrid system, towards reliable 24/7 uninterrupted operation of various on-body applications compatible with diverse scenarios and lifestyles.

Furthermore, the current energy budgeting concept in wearable microgrids is still presumptive, crude, and can be inaccurate, depending on the specific user and scenario. We envision that the next phase in developing wearable microgrids should resemble the traditional microgrids, which more often feature data-driven prediction of energy supply and demand in the grid and adjustable control of various loads toward effective energy budgeting and flow that ensure highly reliable and sustainable microgrid operation. Guided by modern artificial intelligence and machine learning techniques and advanced algorithms, smart energy budgeting for wearable systems will thus rely on the dynamic prediction of energy supply from weather, geological location, daily schedule, diet, or health conditions, is possible. This would allow adjusting the power consumption of wearable applications, e.g., by increasing or decreasing the frequency of sensing or the data transmission, in order to maintain continued sustainable system operation (Fig. 5B). Such design concepts can be also applied to the rising class of implantable and ingestible electronics for biomedical applications, in which autonomous and extended operations are highly desirable. Using the wide range of mechanical and biochemical energies within the body ready to be scavenged, electronics can become independent, battery-less, self-powered designs for various in vivo neurostimulation, drug delivery, and comprehensive monitoring applications.

The rise of the 5th generation mobile network and the internet-of-things has led to the development of many precursor technologies for the eventual evolution of wearable electronics. Endowed by advanced internet infrastructure, distributed edge computing has equipped our homes, cars, and pockets with physically standalone wirelessly connected devices as our daily workflows become increasingly informative and interactive. Traditional wearable electronic industries have slowly evolved around diverse form factors of wireless earbuds, smartwatches, epidermal devices that are functionally advanced yet rigid, bulky, battery-dependent, and lack independence. Despite the growing advocacy towards the transition from rigid and bulky platforms to soft and conformal ones, the development of flexible and stretchable wearable devices (in both e-skin and e-textile form factors) is still limited to the scope of initial proof-of-concept. We believe that this lagged development is due to lack of necessary “infrastructures”: standardized protocols for the inter-modular connections that ensure compatibility between various modules from different developers; standardized characterization method and performance reporting format for benchmarking each energy harvesting and storage technologies; commercially available mechanically robust soft energy storage devices that deliver performance comparable to that of their rigid counterparts for fast prototyping; low-cost, scalable assembly processes that connect current rigid, high-performance Si-based electronics and the new soft peripheral energy devices and applications; and eventually, new semiconductor technologies and standardized protocol for fabricating and packaging of intrinsically soft, high-performance integrated electronics.

Currently, the wearable market has shown appreciable growth in medical, personal wellness, and industrial applications and features relatively price-insensitive state-of-art soft electronics. We envision the roadmap of future wearable electronics development, where rigid wearable devices along with novel e-textile and e-skin platforms for biomedical applications lead the continued development of this field. This is followed by breakthroughs in low-cost, high-performance flexible or stretchable energy storage devices, that will enable a wider range of flexible electronics, which previously lack high-performance batteries with compatible form-factors, to quickly expand the wearable market and advanced functionalities in sensing, communication, and interaction. Lastly, with the development of low-power electronics and high-power energy harvesters, the gap between the power supply and demand in wearable systems will be narrowed, allowing applications to operate autonomously, independently, and maintenance-free (Fig. 5C). Towards this goal, the concept of the wearable microgrid will greatly promote the development of integrated wearable technologies featuring smart and efficient energy budgeting and management. We believe that such cross-disciplinary efforts for realizing the wearable microgrid vision will enhance the practicality of wearable electronic systems in different usage scenarios and will lead to next-generation electronics, characterized by high efficiency, reliability, autonomy and practicality.

Author contributions

L. Y., K. N. K., and J. W. drafted the outline for this perspective article. L. Y. and J. W. prepared the first draft of the manuscript. L. Y., K. N. K., T. P. and J. W. prepared the figures. L. Y., K. N. K., A. T. and J. W. wrote and revised the manuscript. All authors contributed to the discussions and revisions of the figures.

Conflicts of interest

The authors declare no competing interests.

Acknowledgements

This perspective is the result of multiple wearable electronics projects supported by the UC San Diego Center for Wearable Sensors (CWS).

References

1. J. Kim, A. S. Campbell, B. E.-F. de Ávila and J. Wang, Nat. Biotechnol., 2019, 37, 389–406.
2. J. Heikenfeld, A. Jajack, B. Feldman, S. W. Granger, S. Gaitonde, G. Begtrup and B. A. Katchman, Nat. Biotechnol., 2019, 37, 407–419.
3. A. J. Bandodkar, W. J. Jeang, R. Ghaffari and J. A. Rogers, Annu. Rev. Anal. Chem., 2019, 12, 1–22.
4. M. Bariya, H. Y. Y. Nyein and A. Javey, Nat. Electron., 2018, 1, 160–171.
5. T. R. Ray, J. Choi, A. J. Bandodkar, S. Krishnan, P. Gutruf, L. Tian, R. Ghaffari and J. A. Rogers, Chem. Rev., 2019, 119, 5461–5533.
6. W. Gao, S. Emaminejad, H. Y. Y. Nyein, S. Challa, K. Chen, A. Peck, H. M. Fahad, H. Ota, H. Shiraki, D. Kiriya, D.-H. Lien, G. A. Brooks, R. W. Davis and A. Javey, Nature, 2016, 529, 509–514.
7. J. R. Sempionatto, M. Lin, L. Yin, E. De la Paz, K. Pei, T. Sonsa-ard, A. N. de Loyola Silva, A. A. Khorshed, F. Zhang, N. Tostado, S. Xu and J. Wang, Nat. Biomed. Eng., 2021, 5, 737–748.
8. S. Sundaram, P. Kellnhofer, Y. Li, J.-Y. Zhu, A. Torralba and W. Matusik, Nature, 2019, 569, 698–702.
9. L. N. Awad, J. Bae, K. O’Donnell, S. M. M. D. Rossi, K. Hendron, L. H. Sloot, P. Kudzia, S. Allen, K. G. Holt, T. D. Ellis and C. J. Walsh, Sci. Transl. Med., 2017, 9, eaai9084.
10. X. Shi, Y. Zuo, P. Zhai, J. Shen, Y. Yang, Z. Gao, M. Liao, J. Wu, J. Wang, X. Xu, Q. Tong, B. Zhang, B. Wang, X. Sun, L. Zhang, Q. Pei, D. Jin, P. Chen and H. Peng, Nature, 2021, 591, 240–245.
11. M. Gao, P. Wang, L. Jiang, B. Wang, Y. Yao, S. Liu, D. Chu, W. Cheng and Y. Lu, Energy Environ. Sci., 2021, 14, 2114.
12. S. Gong and W. Cheng, Adv. Energy Mater., 2017, 7, 1700648.
13. X. Pu, W. Song, M. Liu, C. Sun, C. Du, C. Jiang, X. Huang, D. Zou, W. Hu and Z. L. Wang, Adv. Energy Mater., 2016, 6, 1601048.
14. M. Tebyetekerwa, I. Marriam, Z. Xu, S. Yang, H. Zhang, F. Zabihi, R. Jose, S. Peng, M. Zhu and S. Ramakrishna, Energy Environ. Sci., 2019, 12, 2148–2160.
15. L. Li, Z. Wu, S. Yuan and X.-B. Zhang, Energy Environ. Sci., 2014, 7, 2101–2122.
16. H. Wu, Y. Huang, F. Xu, Y. Duan and Z. Yin, Adv. Mater., 2016, 28, 9881–9919.
17. H.-R. Lim, H. S. Kim, R. Qazi, Y.-T. Kwon, J.-W. Jeong and W.-H. Yeo, Adv. Mater., 2020, 32, 1901924.
18. K. Dong, X. Peng and Z. L. Wang, Adv. Mater., 2019, 32, 1902549.
19. X.-L. Shi, W.-Y. Chen, T. Zhang, J. Zou and Z.-G. Chen, Energy Environ. Sci., 2021, 14, 729–764.
20. L. Huang, S. Lin, Z. Xu, H. Zhou, J. Duan, B. Hu and J. Zhou, Adv. Mater., 2020, 32, 1902034.
21. V. Vallem, Y. Sargolzaeiaval, M. Ozturk, Y.-C. Lai and M. D. Dickey, Adv. Mater., 2021, 33, 2004832.
22. A. Nozariasbmarz, H. Collins, K. Dsouza, M. H. Polash, M. Hosseini, M. Hyland, J. Liu, A. Malhotra, F. M. Ortiz, F. Mohaddes, V. P. Ramesh, Y. Sargolzaeiaval, N. Snouwaert, M. C. Özturk and D. Vashaee, Appl. Energy, 2020, 258, 114069.
23. Y. Liu, H. Khanbareh, M. A. Halim, A. Feeney, X. Zhang, H. Heidari and R. Ghannam, Nano Sel., 2021, 2, 1459–1479.
24. L. Yin, K. N. Kim, J. Lv, F. Tehrani, M. Lin, Z. Lin, J.-M. Moon, J. Ma, J. Yu, S. Xu and J. Wang, Nat. Commun., 2021, 12, 1542.
25. Y. Song, D. Mukasa, H. Zhang and W. Gao, Acc. Mater. Res., 2021, 2, 184–197.
26. Y. Yu, J. Nassar, C. Xu, J. Min, Y. Yang, A. Dai, R. Doshi, A. Huang, Y. Song, R. Gehlhar, A. D. Ames and W. Gao, Sci. Robot., 2020, 5, eaaz7946.
27. Y. Song, J. Min, Y. Yu, H. Wang, Y. Yang, H. Zhang and W. Gao, Sci. Adv., 2020, 6, eaay9842.
28. H. Li, C. S. L. Koh, Y. H. Lee, Y. Zhang, G. C. Phan-Quang, C. Zhu, Z. Liu, Z. Chen, H. Y. F. Sim, C. L. Lay, Q. An and X. Y. Ling, Nano Energy, 2020, 73, 104723.
29. H. Xue, Q. Yang, D. Wang, W. Luo, W. Wang, M. Lin, D. Liang and Q. Luo, Nano Energy, 2017, 38, 147–154.
30. J. Chen, Y. Huang, N. Zhang, H. Zou, R. Liu, C. Tao, X. Fan and Z. L. Wang, Nat. Energy, 2016, 1, 16138.
31. B. Seo, Y. Cha, S. Kim and W. Choi, ACS Energy Lett., 2019, 4, 2069–2074.
32. L. Yin, J.-M. Moon, J. R. Sempionatto, M. Lin, M. Cao, A. Trifonov, F. Zhang, Z. Lou, J.-M. Jeong, S.-J. Lee, S. Xu and J. Wang, Joule, 2021, 5, 1888–1904.
33. M. Raugei, Joule, 2019, 3, 1810–1811.
34. W. R. Issa, A. H. E. Khateb, M. A. Abusara and T. K. Mallick, IEEE Trans. Ind. Electron., 2018, 65, 4831–4839.
35. M. F. Zia, E. Elbouchikhi and M. Benbouzid, Appl. Energy, 2018, 222, 1033–1055.
36. B. Lasseter, 2001 IEEE Power Engineering Society Winter Meeting. Conference Proceedings, 2001, vol. 1, pp. 146–149.
37. D. E. Olivares, A. Mehrizi-Sani, A. H. Etemadi, C. A. Cañizares, R. Iravani, M. Kazerani, A. H. Hajimiragha, O. Gomis-Bellmunt, M. Saeedifard, R. Palma-Behnke, G. A. Jiménez-Estévez and N. D. Hatziargyriou, IEEE Trans. Smart Grid, 2014, 5, 1905–1919.
38. T. C. Ou and C. M. Hong, Energy, 2014, 66, 314–323.
39. Y. Li, Z. Yang, G. Li, D. Zhao and W. Tian, IEEE Trans. Ind. Electron., 2019, 66, 1565–1575.
40. S. Alireza Hashemi, S. Ramakrishna and A. Gerhard Aberle, Energy Environ. Sci., 2020, 13, 685–743.
41. J. Lv, J. Chen and P. S. Lee, SusMat, 2021, 1, 285–302.
42. Y. Khan, A. E. Ostfeld, C. M. Lochner, A. Pierre and A. C. Arias, Adv. Mater., 2016, 28, 4373–4395.
43. M. Ross, R. Hidalgo, C. Abbey and G. Joós, IET Renew. Power Gen., 2011, 5, 117–123.
44. Y. Kuang, Y. Zhang, B. Zhou, C. Li, Y. Cao, L. Li and L. Zeng, Renewable Sustainable Energy Rev., 2016, 59, 504–513.
45. O. M. Longe, N. Rao, F. Omowole, A. S. Oluwalami and O. T. Oni, Int. J. Energy Eng., 2017, 7, 55–63.
46. H. Xie, S. Zheng and M. Ni, IEEE Electrif. Mag., 2017, 5, 28–35.
47. S. G. Jayasinghe, L. Meegahapola, N. Fernando, Z. Jin and J. M. Guerrero, Inventions, 2017, 2, 4.
48. A. Bosisio, M. Moncecchi, G. Cassetti and M. Merlo, Sustain. Energy Technol. Assess., 2019, 36, 100535.
49. B. Washom, J. Dilliot, D. Weil, J. Kleissl, N. Balac, W. Torre and C. Richter, IEEE Power Energy Mag., 2013, 11, 28–32.
50. X. Jin, J. Wu, Y. Mu, M. Wang, X. Xu and H. Jia, Appl. Energy, 2017, 208, 480–494.
51. O. Hafez and K. Bhattacharya, Renewable Energy, 2012, 45, 7–15.
52. D. Murakami, Y. Yamagata, T. Yoshida and T. Matsui, Energy Procedia, 2019, 158, 4109–4114.
53. C. Min, S. Kang, C. Yoo, J. Cha, S. Choi, Y. Oh and J. Song, Proceedings of the 2015 ACM International Symposium on Wearable Computers - ISWC ’15, ACM Press, Osaka, Japan, 2015, pp. 11–18.
54. D. Ferreira, A. K. Dey and V. Kostakos, in Pervasive Computing, ed. K. Lyons, J. Hightower and E. M. Huang, Springer, Berlin, Heidelberg, 2011, vol. 6696, pp. 19–33.
55. B. J. Kim, D. H. Kim, Y. Y. Lee, H. W. Shin, G. S. Han, J. S. Hong, K. Mahmood, T. K. Ahn, Y. C. Joo, K. S. Hong, N. G. Park, S. Lee and H. S. Jung, Energy Environ. Sci., 2015, 8, 916–921.
56. A. Thakre, A. Kumar, H.-C. Song, D.-Y. Jeong and J. Ryu, Sensors, 2019, 19, 2170.
57. B. E. Logan and M. Elimelech, Nature, 2012, 488, 313–319.
58. L. Yin, J. Lv and J. Wang, Adv. Mater. Technol., 2020, 5, 2000694.
59. O. Coddington, J. L. Lean, D. Lindholm, P. Pilewskie and M. Snow, NOAA National Centers for Environmental Information DOI:10.7289/V51J97P6.
60. C. Yan, S. Barlow, Z. Wang, H. Yan, A. K.-Y. Jen, S. R. Marder and X. Zhan, Nat. Rev. Mater., 2018, 3, 1–19.
61. H. Jinno, K. Fukuda, X. Xu, S. Park, Y. Suzuki, M. Koizumi, T. Yokota, I. Osaka, K. Takimiya and T. Someya, Nat. Energy, 2017, 2, 780–785.
62. W. Huang, Z. Jiang, K. Fukuda, X. Jiao, C. R. McNeill, T. Yokota and T. Someya, Joule, 2020, 4, 128–141.
63. J. Page, in Practical Handbook of Photovoltaics, ed. A. McEvoy, T. Markvart and L. Castañer, Academic Press, Boston, 2nd edn, 2012, pp. 573–643.
64. C. A. Reynaud, R. Clerc, P. B. Lechêne, M. Hébert, A. Cazier and A. C. Arias, Sol. Energy Mater. Sol. Cells, 2019, 200, 110010.
65. Q. Zhai, Y. Liu, R. Wang, Y. Wang, Q. Lyu, S. Gong, J. Wang, G. P. Simon and W. Cheng, Adv. Energy Mater., 2020, 10, 1903512.
66. T. Thampan, D. Shah, C. Cook, J. Novoa and S. Shah, J. Power Sources, 2014, 259, 276–281.
67. I. Jeerapan, J. R. Sempionatto and J. Wang, Adv. Funct. Mater., 2020, 30, 1906243.
68. L. Manjakkal, L. Yin, A. Nathan, J. Wang and R. Dahiya, Adv. Mater., 2021, 33, 2100899.
69. A. J. Bandodkar, J. Electrochem. Soc., 2017, 164, H3007–H3014.
70. S. Choi, Biosens. Bioelectron., 2015, 69, 8–25.
71. S. Pang, Y. Gao and S. Choi, Adv. Energy Mater., 2018, 8, 1702261.
72. N. A. Taylor and C. A. Machado-Moreira, Extreme Physiol. Med., 2013, 2, 4.
73. M. N. Sawka, C. B. Wenger and K. B. Pandolf, Comprehensive Physiology, American Cancer Society, 2011, pp. 157–185.
74. Y. Zou, V. Raveendran and J. Chen, Nano Energy, 2020, 77, 105303.
75. S. R. Anton and H. A. Sodano, Smart Mater. Struct., 2007, 16, R1–R21.
76. M. Safaei, H. A. Sodano and S. R. Anton, Smart Mater. Struct., 2019, 28, 113001.
77. N. Sezer and M. Koç, Nano Energy, 2021, 80, 105567.
78. N. R. Alluri, A. Chandrasekhar, V. Vivekananthan, Y. Purusothaman, S. Selvarajan, J. H. Jeong and S.-J. Kim, ACS Sustainable Chem. Eng., 2017, 5, 4730–4738.
79. L. Gu, J. Liu, N. Cui, Q. Xu, T. Du, L. Zhang, Z. Wang, C. Long and Y. Qin, Nat. Commun., 2020, 11, 1030.
80. Y. Zi, S. Niu, J. Wang, Z. Wen, W. Tang and Z. L. Wang, Nat. Commun., 2015, 6, 8376.
81. H. Zou, L. Guo, H. Xue, Y. Zhang, X. Shen, X. Liu, P. Wang, X. He, G. Dai, P. Jiang, H. Zheng, B. Zhang, C. Xu and Z. L. Wang, Nat. Commun., 2020, 11, 2093.
82. H. Wu, S. Wang, Z. Wang and Y. Zi, Nat. Commun., 2021, 12, 5470.
83. Z. Wang, W. Liu, W. He, H. Guo, L. Long, Y. Xi, X. Wang, A. Liu and C. Hu, Joule, 2021, 5, 441–455.
84. T. J. Seebeck, Ann. Phys., 1826, 82, 253–286.
85. A. R. M. Siddique, S. Mahmud and B. V. Heyst, Renewable Sustainable Energy Rev., 2017, 73, 730–744.
86. W. Ren, Y. Sun, D. Zhao, A. Aili, S. Zhang, C. Shi, J. Zhang, H. Geng, J. Zhang, L. Zhang, J. Xiao and R. Yang, Sci. Adv., 2021, 7, eabe0586.
87. S. Shin, R. Kumar, J. W. Roh, D.-S. S. Ko, H.-S. S. Kim, S. I. Kim, L. Yin, S. M. Schlossberg, S. Cui, J.-M. M. You, S. Kwon, J. Zheng, J. Wang and R. Chen, Sci. Rep., 2017, 7, 7317.
88. Y. Shi, Y. Wang, D. Mei, B. Feng and Z. Chen, IEEE Robot. Autom. Lett., 2018, 3, 373–378.
89. Z. Wu, J. Tang, X. Zhang and Z. Yu, Appl. Phys. Lett., 2017, 111, 013903.
90. Y. Zhou, X. Zhao, J. Xu, Y. Fang, G. Chen, Y. Song, S. Li and J. Chen, Nat. Mater., 2021, 1–7.
91. T. Zhang, T. Yang, M. Zhang, C. R. Bowen and Y. Yang, iScience, 2020, 23, 101689.
92. K. Zhao, B. Ouyang, C. R. Bowen, Z. L. Wang and Y. Yang, Nano Energy, 2020, 71, 104632.
93. T. Takamatsu, Y. Chen, T. Yoshimasu, M. Nishizawa and T. Miyake, Adv. Mater. Technol., 2019, 4, 1800671.
94. D. Masotti, A. Costanzo and S. Adami, Proceedings of the 5th European Conference on Antennas and Propagation (EUCAP), 2011, pp. 517–520.
95. B. Zhao, J. Mao, J. Zhao, H. Yang and Y. Lian, IEEE Trans. Biomed. Circuits Syst., 2020, 14, 283–296.
96. Y.-W. Chong, W. Ismail, K. Ko and C.-Y. Lee, IEEE Sens. J., 2019, 19, 9047–9062.
97. D. Flynn, R. Aitken, A. Gibbons and K. Shi, Low Power Methodology Manual: For System-on-Chip Design, Springer Science & Business Media, 2007.
98. STM32 Ultra Low Power Microcontrollers, https://www.st.com/en/microcontrollers-microprocessors/stm32-ultra-low-power-mcus.html.
99. Low-Power Microcontrollers and Microprocessors, https://www.microchip.com/en-us/solutions/low-power.
100. S. Z. Homayounfar and T. L. Andrew, SLAS Technol., 2020, 25, 9–24.
101. A. S. Dahiya, J. Thireau, J. Boudaden, S. Lal, U. Gulzar, Y. Zhang, T. Gil, N. Azemard, P. Ramm, T. Kiessling, C. O’Murchu, F. Sebelius, J. Tilly, C. Glynn, S. Geary, C. O’Dwyer, K. M. Razeeb, A. Lacampagne, B. Charlot and A. Todri-Sanial, J. Electrochem. Soc., 2019, 167, 037516.
102. D. C. Klonoff, D. Ahn and A. Drincic, Diabetes Res. Clin. Pract., 2017, 133, 178–192.
103. J. Wang, Analytical Electrochemistry, John Wiley & Sons, Ltd, 2006, pp. 201–243.
104. S. C. Mukhopadhyay, IEEE Sens. J., 2015, 15, 1321–1330.
105. M. Grattieri and S. D. Minteer, ACS Sens., 2018, 3, 44–53.
106. G. Valdés-Ramírez, Y.-C. Li, J. Kim, W. Jia, A. J. Bandodkar, R. Nuñez-Flores, P. R. Miller, S.-Y. Wu, R. Narayan, J. R. Windmiller, R. Polsky and J. Wang, Electrochem. Commun., 2014, 47, 58–62.
107. Y. Shi, Y. Wang, Y. Deng, H. Gao, Z. Lin, W. Zhu and H. Ye, Energy Convers. Manage., 2014, 80, 110–116.
108. J. Zhao, Y. Lin, J. Wu, H. Y. Y. Nyein, M. Bariya, L.-C. Tai, M. Chao, W. Ji, G. Zhang, Z. Fan and A. Javey, ACS Sens., 2019, 4, 1925–1933.
109. M. Zhu, Q. Shi, T. He, Z. Yi, Y. Ma, B. Yang, T. Chen and C. Lee, ACS Nano, 2019, 13, 1940–1952.
110. I. Jeerapan, J. R. Sempionatto, A. Pavinatto, J.-M. You and J. Wang, J. Mater. Chem. A, 2016, 4, 18342–18353.
111. S. Seneviratne, Y. Hu, T. Nguyen, G. Lan, S. Khalifa, K. Thilakarathna, M. Hassan and A. Seneviratne, IEEE Commun. Surv. Tutor., 2017, 19, 2573–2620.
112. Q. Wang, M. Hempstead and W. Yang, 2006 3rd Annual IEEE Communications Society on Sensor and Ad Hoc Communications and Networks, 2006, vol. 1, pp. 286–295.
113. J.-S. Lee, Y.-W. Su and C.-C. Shen, IECON 2007 – 33rd Annual Conference of the IEEE Industrial Electronics Society, 2007, pp. 46–51.
114. A. Dementyev, S. Hodges, S. Taylor and J. Smith, 2013 IEEE International Wireless Symposium (IWS), 2013, pp. 1–4.
115. H.-J. Kim, H. Hirayama, S. Kim, K. J. Han, R. Zhang and J.-W. Choi, IEEE Access, 2017, 5, 21264–21285.
116. P. Lathiya and J. Wang, Near-Field Communications (NFC) for Wireless Power Transfer (WPT): An Overview, IntechOpen, 2021.
117. M. R. Fernández, E. Z. Casanova and I. G. Alonso, Sustainability, 2015, 7, 10854–10875.
118. J. Kim, H. J. Shim, J. Yang, M. K. Choi, D. C. Kim, J. Kim, T. Hyeon and D.-H. Kim, Adv. Mater., 2017, 29, 1700217.
119. Y. Zhou, C. Zhao, J. Wang, Y. Li, C. Li, H. Zhu, S. Feng, S. Cao and D. Kong, ACS Mater. Lett., 2019, 1, 511–518.
120. P. Andersson, R. Forchheimer, P. Tehrani and M. Berggren, Adv. Funct. Mater., 2007, 17, 3074–3082.
121. P. Andersson Ersman, J. Kawahara and M. Berggren, Org. Electron., 2013, 14, 3371–3378.
122. J. A. Rogers, Z. Bao, K. Baldwin, A. Dodabalapur, B. Crone, V. R. Raju, V. Kuck, H. Katz, K. Amundson, J. Ewing and P. Drzaic, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 4835–4840.
123. A. R. Nikhar, S. M. Apte and R. Somalwar, 2016 International Conference on Global Trends in Signal Processing, Information Computing and Communication (ICGTSPICC), 2016, pp. 432–437.
124. S. Ghosh, S. Satpathy, S. Das, S. Debbarma and B. K. Bhattacharyya, 2018 International Conference on Smart Systems and Inventive Technology (ICSSIT), 2018, pp. 29–33.
125. TPS628503-Q1 datasheet, https://www.ti.com/document-viewer/TPS628503-Q1/datasheet/GUID-DFA56A7D-14AB-4D63-88FA-4ADBBFA5F128#TITLE-SLUSDM0X188.
126. BQ25505 datasheet, https://www.ti.com/document-viewer/BQ25505/datasheet/electrical-characteristics-slusbj34357#SLUSBJ34357.
127. D. P. Hohm and M. E. Ropp, Prog. Photovoltaics Res. Appl., 2003, 11, 47–62.
128. Y. Chaibi, A. Allouhi, M. Salhi and A. El-jouni, Prot. Control Mod. Power Syst., 2019, 4, 15.
129. Y. Zi, J. Wang, S. Wang, S. Li, Z. Wen, H. Guo and Z. L. Wang, Nat. Commun., 2016, 7, 10987.
130. G. Cheng, Z.-H. Lin, L. Lin, Z. Du and Z. L. Wang, ACS Nano, 2013, 7, 7383–7391.
131. X. Cheng, L. Miao, Y. Song, Z. Su, H. Chen, X. Chen, J. Zhang and H. Zhang, Nano Energy, 2017, 38, 438–446.
132. S. Niu, Y. S. Zhou, S. Wang, Y. Liu, L. Lin, Y. Bando and Z. L. Wang, Nano Energy, 2014, 8, 150–156.
133. S. Niu, Y. Liu, Y. S. Zhou, S. Wang, L. Lin and Z. L. Wang, IEEE Trans. Electron Devices, 2015, 62, 641–647.
134. A. González, E. Goikolea, J. A. Barrena and R. Mysyk, Renewable Sustainable Energy Rev., 2016, 58, 1189–1206.
135. A. Borenstein, O. Hanna, R. Attias, S. Luski, T. Brousse and D. Aurbach, J. Mater. Chem. A, 2017, 5, 12653–12672.
136. Y. Zhang, H. Feng, X. Wu, L. Wang, A. Zhang, T. Xia, H. Dong, X. Li and L. Zhang, Int. J. Hydrogen Energy, 2009, 34, 4889–4899.
137. L. Yin, J. Scharf, J. Ma, J.-M. Doux, C. Redquest, V. L. Le, Y. Yin, J. Ortega, X. Wei, J. Wang and Y. S. Meng, Joule, 2021, 5, 228–248.
138. J. He, C. Lu, H. Jiang, F. Han, X. Shi, J. Wu, L. Wang, T. Chen, J. Wang, Y. Zhang, H. Yang, G. Zhang, X. Sun, B. Wang, P. Chen, Y. Wang, Y. Xia and H. Peng, Nature, 2021, 597, 57–63.
139. D. Wang, C. Han, F. Mo, Q. Yang, Y. Zhao, Q. Li, G. Liang, B. Dong and C. Zhi, Energy Storage Mater., 2020, 28, 264–292.
140. H. Li, L. Ma, C. Han, Z. Wang, Z. Liu, Z. Tang and C. Zhi, Nano Energy, 2019, 62, 550–587.
141. J. Wen, Y. Yu and C. Chen, Mater. Express, 2012, 2, 197–212.
142. R. Kumar, J. Shin, L. Yin, J. M. You, Y. S. Meng and J. Wang, Adv. Energy Mater., 2017, 7, 1602096.
143. X. Chen, H. Huang, L. Pan, T. Liu and M. Niederberger, Adv. Mater., 2019, 31, 1904648.
144. L. Yin, J. K. Seo, J. Kurniawan, R. Kumar, J. Lv, L. Xie, X. Liu, S. Xu, Y. S. Meng and J. Wang, Small, 2018, 14, 1800938.
145. F. Tehrani, M. Beltrán-Gastélum, K. Sheth, A. Karajic, L. Yin, R. Kumar, F. Soto, J. Kim, J. Wang, S. Barton, M. Mueller and J. Wang, Adv. Mater. Technol., 2019, 4, 1900162.
146. J. Yu, W. Lu, S. Pei, K. Gong, L. Wang, L. Meng, Y. Huang, J. P. Smith, K. S. Booksh, Q. Li, J.-H. Byun, Y. Oh, Y. Yan and T.-W. Chou, ACS Nano, 2016, 10, 5204–5211.
147. X. Liang, G. Long, C. Fu, M. Pang, Y. Xi, J. Li, W. Han, G. Wei and Y. Ji, Chem. Eng. J., 2018, 345, 186–195.
148. S. Xu, Y. Zhang, J. Cho, J. Lee, X. Huang, L. Jia, J. A. Fan, Y. Su, J. Su, H. Zhang, H. Cheng, B. Lu, C. Yu, C. Chuang, T. Kim, T. Song, K. Shigeta, S. Kang, C. Dagdeviren, I. Petrov, P. V. Braun, Y. Huang, U. Paik and J. A. Rogers, Nat. Commun., 2013, 4, 1543.
149. Y. Zi, L. Lin, J. Wang, S. Wang, J. Chen, X. Fan, P.-K. Yang, F. Yi and Z. L. Wang, Adv. Mater., 2015, 27, 2340–2347.
150. H. Li, X. Zhang, L. Zhao, D. Jiang, L. Xu, Z. Liu, Y. Wu, K. Hu, M.-R. Zhang, J. Wang, Y. Fan and Z. Li, Nano-Micro Lett., 2020, 12, 50.
151. J. S. Heo, J. Eom, Y.-H. Kim and S. K. Park, Small, 2018, 14, 1703034.
152. G. Rong, Y. Zheng and M. Sawan, Sensors, 2021, 21, 3806.
153. L. Manjakkal, A. Pullanchiyodan, N. Yogeswaran, E. S. Hosseini and R. Dahiya, Adv. Mater., 2020, 32, 1907254.
154. R. Mukherjee, P. Ganguly and R. Dahiya, Adv. Intell. Syst., 2021, 2100036.
155. M. Parrilla and K. De Wael, Adv. Funct. Mater., 2021, 2107042.
156. IDC – Wearable Devices Market Share, https://www.idc.com/promo/wearablevendor.
157. Wearable Technology Forecasts 2021–2031, https://www.idtechex.com/en/research-report/wearable-technology-forecasts-2021-2031/839.
158. M. Amjadi, S. Sheykhansari, B. J. Nelson and M. Sitti, Adv. Mater., 2018, 30, 1704530.
159. H. Lee, C. Song, S. Baik, D. Kim, T. Hyeon and D.-H. Kim, Adv. Drug Delivery Rev., 2018, 127, 35–45.
160. X. T. R. Kong, H. Luo, G. Q. Huang and X. Yang, J. Intell. Manuf., 2019, 30, 2853–2869.
161. J. Kim, S. Imani, W. R. de Araujo, J. Warchall, G. Valdés-Ramírez, T. R. L. C. Paixão, P. P. Mercier and J. Wang, Biosens. Bioelectron., 2015, 74, 1061–1068.
162. J. R. Sempionatto, T. Nakagawa, A. Pavinatto, S. T. Mensah, S. Imani, P. Mercier and J. Wang, Lab Chip, 2017, 17, 1834–1842.
163. J. R. Sempionatto, L. C. Brazaca, L. García-Carmona, G. Bolat, A. S. Campbell, A. Martin, G. Tang, R. Shah, R. K. Mishra, J. Kim, V. Zucolotto, A. Escarpa and J. Wang, Biosens. Bioelectron., 2019, 137, 161–170.
164. T.-C. Hou, Y. Yang, H. Zhang, J. Chen, L.-J. Chen and Z. Lin Wang, Nano Energy, 2013, 2, 856–862.
165. G. Zhu, P. Bai, J. Chen and Z. Lin Wang, Nano Energy, 2013, 2, 688–692.
166. S. J. Benight, C. Wang, J. B. H. Tok and Z. Bao, Prog. Polym. Sci., 2013, 38, 1961–1977.
167. D.-H. Kim, N. Lu, R. Ma, Y.-S. Kim, R.-H. Kim, S. Wang, J. Wu, S. M. Won, H. Tao, A. Islam, K. J. Yu, T.-I. Kim, R. Chowdhury, M. Ying, L. Xu, M. Li, H.-J. Chung, H. Keum, M. McCormick, P. Liu, Y.-W. Zhang, F. G. Omenetto, Y. Huang, T. Coleman and J. A. Rogers, Science, 2011, 333, 838–843.
168. D. J. Lipomi, M. Vosgueritchian, B. C.-K. Tee, S. L. Hellstrom, J. A. Lee, C. H. Fox and Z. Bao, Nat. Nanotechnol., 2011, 6, 788–792.
169. M. Kaltenbrunner, T. Sekitani, J. Reeder, T. Yokota, K. Kuribara, T. Tokuhara, M. Drack, R. Schwödiauer, I. Graz, S. Bauer-Gogonea, S. Bauer and T. Someya, Nature, 2013, 499, 458–463.
170. A. Miyamoto, S. Lee, N. F. Cooray, S. Lee, M. Mori, N. Matsuhisa, H. Jin, L. Yoda, T. Yokota, A. Itoh, M. Sekino, H. Kawasaki, T. Ebihara, M. Amagai and T. Someya, Nat. Nanotechnol., 2017, 12, 907–913.
171. W.-H. Yeo, Y.-S. Kim, J. Lee, A. Ameen, L. Shi, M. Li, S. Wang, R. Ma, S. H. Jin, Z. Kang, Y. Huang and J. A. Rogers, Adv. Mater., 2013, 25, 2773–2778.
172. H. U. Chung, B. H. Kim, J. Y. Lee, J. Lee, Z. Xie, E. M. Ibler, K. Lee, A. Banks, J. Y. Jeong, J. Kim, C. Ogle, D. Grande, Y. Yu, H. Jang, P. Assem, D. Ryu, J. W. Kwak, M. Namkoong, J. B. Park, Y. Lee, D. H. Kim, A. Ryu, J. Jeong, K. You, B. Ji, Z. Liu, Q. Huo, X. Feng, Y. Deng, Y. Xu, K.-I. Jang, J. Kim, Y. Zhang, R. Ghaffari, C. M. Rand, M. Schau, A. Hamvas, D. E. Weese-Mayer, Y. Huang, S. M. Lee, C. H. Lee, N. R. Shanbhag, A. S. Paller, S. Xu and J. A. Rogers, Science, 2019, 363, eaau0780.
173. C. Wang, X. Li, H. Hu, L. Zhang, Z. Huang, M. Lin, Z. Zhang, Z. Yin, B. Huang, H. Gong, S. Bhaskaran, Y. Gu, M. Makihata, Y. Guo, Y. Lei, Y. Chen, C. Wang, Y. Li, T. Zhang, Z. Chen, A. P. Pisano, L. Zhang, Q. Zhou and S. Xu, Nat. Biomed. Eng., 2018, 2, 687–695.
174. A. J. Bandodkar, S. P. Lee, I. Huang, W. Li, S. Wang, C.-J. Su, W. J. Jeang, T. Hang, S. Mehta, N. Nyberg, P. Gutruf, J. Choi, J. Koo, J. T. Reeder, R. Tseng, R. Ghaffari and J. A. Rogers, Nat. Electron., 2020, 3, 554–562.
175. T. Yokota, P. Zalar, M. Kaltenbrunner, H. Jinno, N. Matsuhisa, H. Kitanosako, Y. Tachibana, W. Yukita, M. Koizumi and T. Someya, Sci. Adv., 2016, 2, e1501856.
176. B. Wang, W. Huang, L. Chi, M. Al-Hashimi, T. J. Marks and A. Facchetti, Chem. Rev., 2018, 118, 5690–5754.
177. Y. Liu, M. Pharr and G. A. Salvatore, ACS Nano, 2017, 11, 9614–9635.
178. C. Wang, C. Wang, Z. Huang and S. Xu, Adv. Mater., 2018, 30, 1801368.
179. J. Y. Oh and Z. Bao, Adv. Sci., 2019, 6, 1900186.
180. G. Chen, Y. Li, M. Bick and J. Chen, Chem. Rev., 2020, 120, 3668–3720.
181. J. Lv, I. Jeerapan, F. Tehrani, L. Yin, C. A. Silva-Lopez, J.-H. Jang, D. Joshuia, R. Shah, Y. Liang, L. Xie, F. Soto, C. Chen, E. Karshalev, C. Kong, Z. Yang and J. Wang, Energy Environ. Sci., 2018, 11, 3431–3442.
182. W. Jia, X. Wang, S. Imani, A. J. Bandodkar, J. Ramírez, P. P. Mercier and J. Wang, J. Mater. Chem. A, 2014, 2, 18184–18189.
183. R. Lin, H.-J. Kim, S. Achavananthadith, S. A. Kurt, S. C. C. Tan, H. Yao, B. C. K. Tee, J. K. W. Lee and J. S. Ho, Nat. Commun., 2020, 11, 444.
184. X. Tian, P. M. Lee, Y. J. Tan, T. L. Y. Wu, H. Yao, M. Zhang, Z. Li, K. A. Ng, B. C. K. Tee and J. S. Ho, Nat. Electron., 2019, 2, 243–251.
185. L. Wang, L. Wang, Y. Zhang, J. Pan, S. Li, X. Sun, B. Zhang and H. Peng, Adv. Funct. Mater., 2018, 28, 1804456.

---