Electric wiring in homes and factories generates an alternating magnetic field around it whenever current flows. Researchers are working on ways to harvest power from this faint field and run electronic devices without batteries. Sensors with extremely low power consumption, in particular, may be able to operate on small amounts of energy gathered from their surroundings.
On October 20, 2026, a joint research team from the University of Arkansas and the University of Michigan published experimental results in the peer-reviewed journal Journal of Vacuum Science & Technology B. They showed that power drawn from the magnetic field around electric wiring can run a custom-developed, low-power temperature sensor. The lead author is Md Rezaul Kabir, and the paper was accepted on September 25, 2026. The team combined a graphene inductor with passive rectifying and voltage-boosting circuits to build a sensor system that uses neither a conventional rechargeable battery nor a power management IC.
What the team harvested was not ionizing radiation or some unknown form of energy. It was the energy of the magnetic field created by alternating current in electric wiring. The experiment used 60Hz, the common mains frequency in the United States. The setup is distinctive for combining not only a graphene-based receiving element but also an iron core, a multistage voltage-multiplier circuit, and storage capacitors. However, what was demonstrated is sensor operation under specific experimental conditions. It has not been shown that the system can start autonomously from a completely depleted state, or that it can run maintenance-free for decades.
How the system harvests power from a 60Hz field around wiring
The research is motivated by the growing burden of battery replacement as Internet of Things (IoT) devices spread. When large numbers of sensors are placed in buildings, factories, and outdoor facilities, replacing batteries requires ongoing labor and cost even if each device consumes little power. In places that people can hardly reach, battery life becomes a factor that limits how long the whole device can operate. As a countermeasure, researchers have studied "energy harvesting," which collects small amounts of energy from ambient light, heat, vibration, electromagnetic fields, and other sources.
Paul M. Thibado, a professor of physics at the University of Arkansas, and his colleagues aim to build autonomous sensors that need neither external charging nor periodic battery replacement. In this study, they removed the rechargeable battery and the dedicated power management IC (PMIC), and instead ran the sensor on charge stored in three capacitors. "Battery-free," however, does not mean there are no energy-storing components at all. By using capacitors instead of chemical batteries, the team aims to reduce the capacity degradation and the replacement burden that come with batteries.
The energy source is the alternating magnetic field that forms around a conductor carrying current. When alternating current flows through a wire, it creates a magnetic field around it that varies with the current. Capturing this field with a coil produces a tiny alternating voltage through electromagnetic induction. The power that can be recovered depends on the current in the wire, the layout of the conductors, and the distance to the coil, so the same output cannot be expected near every kind of wiring.
In an announcement published on October 23, the University of Arkansas describes this energy source as "ambient radiation." What is actually used, however, is not ionizing radiation but the alternating magnetic field at mains frequency. The mechanism is also more accurately understood as drawing power by electromagnetic induction from the fluctuating field near the wiring, rather than receiving electromagnetic waves propagating from afar.
Power generation with an iron-core graphene inductor and a voltage-multiplier circuit
The system has three stages: receiving power with a graphene inductor, rectifying and boosting the AC voltage, and storing energy in capacitors. The team fabricated two kinds of graphene inductors: a two-dimensional spiral formed on a flat surface and a three-dimensional helix with a coiled structure. They measured current while varying the applied voltage and frequency to examine the inductance characteristics. They also placed the inductors near a power cable carrying alternating current and measured the induced voltage and current in response to changes in cable current.
An important point is that the receiving element is not made of graphene alone. According to the paper, to obtain enough voltage to operate the downstream circuitry, a steel core was built into a single graphene inductor. The iron core concentrates magnetic flux and raises the electromotive force generated by electromagnetic induction. The result should therefore be understood not as a special power-generation phenomenon of graphene by itself, but as energy harvesting by an inductor that combines a graphene conductor with an iron core.
The AC voltage generated by the inductor is small and cannot power a temperature sensor or digital circuitry as it is. The team therefore adopted a Cockcroft-Walton circuit, in which diodes and capacitors are connected in multiple stages. This circuit rectifies alternating current into direct current and, at the same time, raises the voltage according to the number of stages. It requires no active boost IC driven by an external power supply, but the output voltage and power actually obtained depend on the input conditions and the load.
In this design, voltages are taken from different stages of the voltage-multiplier circuit, and three storage capacitors are each charged to the required voltage. The custom-developed temperature sensor system needs several supply voltages, and by feeding it directly from each stage, the team made it work without a dedicated power management IC. The researchers report that this design lowered the overall power consumption of the system compared with a conventional setup using a rechargeable battery and a PMIC, and extended its expected operating life.
However, raising the voltage is not the same as increasing the power that can be extracted. The voltage-multiplier circuit has losses, and for the sensor to operate for a long time, the balance between harvested power and consumed power must hold. The key point of this technology is not that it generates large amounts of power from a faint magnetic field, but that it keeps the sensor's own consumption extremely low to match the energy that can be harvested.
Demonstrated performance and the university's vision for the future
What the paper directly reports is a proof of concept (PoC): a graphene inductor, passive rectifying and boosting circuits, and three storage capacitors were combined to run a purpose-designed, ultra-low-power temperature sensor. The abstract describes the characterization of the inductors, measurements of induced voltage and current near a power cable, and the basic operation of the sensor system. The published summary alone, however, does not confirm detailed performance comparisons with commercial sensors, statistical variation across many prototype elements, or results of independent verification by a third party.
In the university's announcement, three requirements set by the team are listed: using energy harvested from the surroundings, operating on nanowatt-level power, and running for long periods without human intervention. The university explains that the sensor met these requirements, but this does not mean that decades of continuous operation were confirmed by measurement. The paper summary also does not give enough specific figures for the current passed through the cable, the receiving distance, capacitor values, conversion efficiency, or temperature measurement error. Evaluating these conditions in detail requires checking the experimental data in the body of the paper.
The current demonstrated results and future vision can be summarized as follows.
| Item | Reported in the paper | University's explanation / future vision | Current assessment |
|---|---|---|---|
| Energy source | Harvests power from the magnetic field around a power cable | Uses electromagnetic fields around wiring in homes and facilities | Power reception from a 60Hz AC magnetic field demonstrated |
| Receiving element | Graphene inductor with a built-in iron core | Expansion to autonomous sensors using diverse environmental energy | Structure combining an iron core and a graphene conductor |
| Power conversion and storage | Cockcroft-Walton circuit and three capacitors | Aims for operation that eliminates battery replacement | Battery-free configuration using passive circuits demonstrated |
| Sensor function | Custom-designed low-power temperature sensor operated | Applications in logistics, environmental monitoring, wearables, and healthcare | Basic operation confirmed in the laboratory |
| Operation during power outages | The paper summary gives no details of long-duration outage tests | Expects 24 hours of operation, possibly about a week, depending on stored charge | Measured results showing long unpowered operation cannot be confirmed from the summary |
| Initial startup | Initial capacitor charging and programming of the CPU are required | Aims for long-term autonomous operation | Autonomous startup from a fully discharged state not verified |
| Service life | Expected life extended by eliminating the battery and PMIC | Goal of decades of maintenance-free operation | Long-term reliability needs further evaluation |
A design that removes the battery and the PMIC has the advantage of reducing the burden of battery replacement and the self-consumption of power management circuitry. That does not automatically extend the sensor's life to decades, however. Actual service life is also affected by changes in capacitor characteristics, degradation of circuit components, and the installation environment. This result demonstrates a promising circuit configuration for long-term operation, but it has not established the long-term reliability required of a commercial device.
Autonomous startup from a fully discharged state and outage endurance remain unverified
One key challenge for practical use is autonomous startup from a state where all power has been lost, known as a "cold start." The paper's abstract states clearly that, before the sensor could operate, the capacitors had to be initially charged and the CPU had to be programmed. What was confirmed, in other words, is that a system that has completed its initial setup can operate on harvested energy. The research team itself says it has not evaluated the ability to start up from a completely discharged state using only energy from the magnetic field.
This difference bears directly on the operational burden of installing sensors in the field. If charging and programming are done in advance, a certain amount of work is still required at installation, even if battery replacement is unnecessary afterward. And if the capacitors' charge were to run out completely during operation, the current results cannot tell us whether the sensor could restart on power harvested from the magnetic field alone. Programming as initial setup and restarting after power loss are separate issues, but both matter in evaluating a practical autonomous system.
The time the sensor can keep running during a power outage should also be considered separately for the university's explanation and the measured results. In the university's announcement, Professor Thibado explained that the inductor replenishes the capacitors' charge slowly but faster than the sensor consumes it. He added that, after an outage, the stored charge could easily keep the sensor running for about 24 hours and might last about a week. This statement, however, includes the researcher's own outlook and does not mean that seven days of continuous unpowered operation were confirmed in an experiment.
Actual outage endurance is determined by the relationship between the energy stored in the capacitors and the energy the sensor consumes per unit time. Lengthening the measurement interval and increasing standby time can reduce consumption, while raising measurement frequency or processing load drains the stored charge faster. Capacitor leakage current and temperature characteristics also affect how long the sensor can keep running. With these conditions clearly specified, it will be necessary to actually cut off the supply from the magnetic field and measure how long the sensor keeps operating.
Another challenge is whether power can be harvested stably in real wiring environments. The experiment targets 60Hz, the standard frequency in the United States, but the induced voltage changes with the magnitude of the current, the layout of the wiring, and the distance from the receiving element. The same performance is not guaranteed in environments where current rises and falls with the operation of electrical equipment, or in 50Hz environments such as eastern Japan. To deploy the system in real environments, it is essential to examine power generation, charging speed, and the stability of sensor operation as wiring conditions change.
Remaining challenges for battery-free IoT sensors
The significance of this study is that it actually ran a battery-free temperature sensor using the faint alternating magnetic field around ordinary electric wiring. In particular, a configuration that secures multiple supply voltages with a passive voltage-multiplier circuit and capacitors, without a dedicated power management IC, offers a new option for designing ultra-low-power devices. The important point is that it shows that even when the recoverable power is small, the energy needed for operation can be covered if the sensor's own consumption is kept low enough.
As future applications, the research team cites package tracking in logistics, environmental monitoring of air and soil, vehicle management, wearable devices, and healthcare-related monitoring. They are also pursuing the idea of combining multiple environmental energy sources, including not only electromagnetic fields but also sunlight, heat, sound, and mechanical vibration. For example, if one energy source becomes unavailable, power from another could raise the device's uptime. These are, however, future plans set out by the team, and the paper has not demonstrated a system that integrates multiple energy sources.
To reach practical use, a circuit capable of autonomous startup from a fully discharged state must first be implemented and verified. It must also be examined whether the required voltage and power can be secured reliably when the cable current, frequency, and receiving distance change. Whether the energy balance can be maintained while preserving temperature measurement accuracy and measurement frequency is another important evaluation item. In addition, degradation and reliability over long-term operation must be verified for the whole circuit, including the capacitors and the graphene inductor.
This result is significant as a circuit design for realizing battery-free IoT sensors, but it is not a technology that can draw unlimited power anywhere. The energy that can be harvested from the surroundings is limited, and a device's power consumption and operating frequency must be designed around those conditions. If initial startup, outage endurance, and power generation performance in real environments are verified in the future, the approach could become a strong option for low-power sensors installed where battery replacement is difficult. This demonstration is best seen as a step toward that practical use.