The Internet of Things has a fundamental problem: billions of connected devices need power, and traditional batteries create maintenance nightmares and environmental waste. Researchers have now developed a novel photocapacitor that harvests energy directly from ambient indoor light, enabling truly autonomous IoT devices that can run indefinitely without replacement batteries.

The advance combines a dye-sensitized solar cell with a specially engineered supercapacitor in a three-terminal architecture, allowing the system to both capture light and store energy with unprecedented efficiency in low-light conditions. Under typical office lighting of 1000 lux, the device achieved a photocharging voltage of 920 millivolts and overall efficiencies up to 18 percent. In practical testing, it powered a wireless IoT network running machine learning inference for 72 hours continuously, outperforming commercial modules by 3.5 times.

The Energy Harvesting Challenge

Today's Internet of Things extends far beyond smart speakers and thermostats. Imagine hospitals where thousands of wireless sensors monitor patient vitals, or smart buildings where distributed networks track occupancy, air quality, and energy use. Every sensor needs power. In outdoor settings, photovoltaic cells work reasonably well. But most IoT devices operate indoors, where light intensity is 100 to 1000 times lower than sunlight.

Batteries have long been the default solution, but they bring serious limitations. They degrade over time, require regular replacement, and create logistical headaches for large-scale deployments. For a building with thousands of sensors, battery replacement becomes a continuous maintenance burden. Worse, discarded batteries accumulate environmental waste.

Energy harvesting from ambient light offers an appealing alternative, but the physics is challenging. Indoor light is dim and intermittent. Standard photovoltaic technologies struggle to convert such weak signals into usable power. Even when they succeed, the energy must be stored efficiently to bridge gaps when the lights dim or turn off.

A Three-Part Strategy

The research team addressed these challenges by engineering three key components. First, they used dye-sensitized solar cells, a technology known for superior performance under indoor lighting. Unlike conventional silicon panels optimized for bright sunlight, these cells use organic dyes that absorb the specific wavelengths present in office fluorescent and LED light. They've achieved power conversion efficiencies exceeding 38 percent under standard indoor illumination—better than any competing photovoltaic technology in low-light conditions.

Second, they developed novel energy storage materials called polyviologens. These organic compounds undergo reversible electrochemical reactions, storing energy electrostatically through what scientists call the electrical double layer effect. The innovation lay in creating two variants: a linear form called PV1 and a newly synthesized V-shaped form called PVN.

The V-shaped structure of PVN proved crucial. Computational modeling revealed that its biphenyl-dioxy bridges enhanced ion accessibility and charge transfer. More practically, PVN demonstrated nearly perfect stability over 3,000 charge-discharge cycles, retaining 100 percent of its original storage capacity where the linear variant lost about 60 percent. The team attributes this durability to lower mechanical stress during cycling, a consequence of the V-shaped geometry.

Third, they replaced standard commercial membranes with bioderived fungal-based films. These materials proved superior at ion transport and electrolyte retention, significantly boosting overall performance compared to conventional alternatives.

Integration and Architecture

The critical insight was integrating these components into a three-terminal architecture rather than the traditional two-terminal design. In conventional systems, the solar cell and supercapacitor share two terminals, creating a bottleneck. The extracted charges want to recombine, wasting energy and reducing efficiency.

By adding a third terminal, the researchers enabled independent control of the solar cell and storage functions. Energy harvested by the photovoltaic unit transfers directly to the supercapacitor through a shared electrode without interference. The design minimizes charge recombination and maximizes storage voltage. In essence, the photocapacitor charges continuously under light while the storage compartment releases power on demand without disrupting the harvesting process.

Measured Performance

Under simulated 1 sun illumination (the standard laboratory benchmark), 24 small test devices achieved maximum voltages exceeding 1.1 volts and overall device efficiencies of 3.5 percent. These figures appear modest compared to pure solar cells, but they represent a genuine advance for integrated storage systems. The loss in efficiency reflects the energy cost of dual functionality—harvesting and storing—which researchers note is far outweighed by having both capabilities in one compact device.

Under indoor lighting conditions (1000 lux from fluorescent and LED sources), larger devices achieved the headline results: 920 millivolts photocharging voltage and 18 percent photocharging efficiency. A single device retained 95 percent capacity after 1000 hours of continuous operation at room temperature.

Real-World Edge Computing Test

The ultimate test of any energy device is whether it can actually power real applications. The researchers built a three-layer wireless IoT network simulating the control and data systems found in smart buildings.

Layer 3 acted as a data-acquisition node, preloaded with 100 handwritten digit images (14 by 14 pixels). It sent the image data one row at a time to the compute layer.

Layer 2 housed a small microcontroller running a pretrained neural network for digit classification. When it received a complete image, it performed inference and sent the result forward.

Layer 1 acted as a data relay, forwarding results to a main power-connected logging system.

All three layers were powered entirely by photocapacitors or, in comparison trials, commercial amorphous silicon modules. They implemented adaptive sleep algorithms, allowing nodes to conserve power based on voltage levels of their storage capacitors.

At 1000 lux, photocapacitor-powered nodes achieved 1.9 predictions per hour—137 predictions total over 72 hours. The commercial silicon modules managed just 0.54 predictions per hour—39 total. That's a 3.5-fold advantage for the photocapacitor system.

The edge-computing inference itself ran with 93 percent accuracy on the standard CIFAR-10 image classification benchmark, using merely 0.81 millijoules per prediction. This demonstrates not just that the devices can power IoT networks, but that they can sustain real artificial intelligence workloads at the edge of networks, where computational decisions happen locally rather than in the cloud.

Why Fungal Membranes Matter

One detail worth highlighting: the role of fungal-derived membranes. Conventional supercapacitors use Nafion, a synthetic polymer developed for hydrogen fuel cells. The research compared devices side-by-side. Those with fungal-based separators consistently outperformed Nafion counterparts, showing higher capacitance, longer discharge times, and better cycling stability.

This wasn't accidental. The fungal membranes possess inherent properties—porosity, ion conductivity, electrolyte retention—that happened to align well with polyviologen electrochemistry. The result demonstrates a broader principle: matching materials at the molecular level, rather than simply adapting commercial components, often yields surprising performance gains.

A Path Toward Battery-Free Deployment

For IoT infrastructure, this work offers a concrete path forward. Instead of maintaining battery inventories and managing replacement schedules, facility managers could deploy wireless sensors once and leave them to harvest energy indefinitely.

The implications extend beyond buildings. Outdoor sensor networks in fields, forests, or disaster areas could operate autonomously. Wearable devices could potentially collect power from ambient light throughout the day. Smart city infrastructure could scale without the logistics burden of millions of batteries.

The research team noted that future improvements should focus on reducing voltage drops in larger devices, scaling manufacturing, and testing long-term stability under varied environmental conditions. Current efficiencies, while record-breaking for integrated photocapacitors, still lag behind pure solar cells or commercial supercapacitors in isolation. But for applications that demand both functions in one package, the trade-off makes sense.

The Larger Picture

This work aligns with broader efforts to decouple IoT infrastructure from grid electricity and battery supply chains. As device counts approach 30 billion, environmental and logistical pressures demand solutions that don't simply shift the burden elsewhere. Battery mining carries ecological costs. Manufacturing and transportation add emissions. Disposal creates persistent waste streams.

Photocapacitors that harvest ambient indoor light offer a different model entirely. They require no moving parts, no consumables, and no periodic replacement. They convert waste light—the illumination we generate anyway—into stored electricity.

The research demonstrates that such devices are no longer theoretical. With careful material engineering and system architecture, they can power real applications in real indoor environments. The next step is scaling from laboratory prototypes to manufacturing deployments, and from specialized research applications to commodity IoT networks.

For anyone tracking sustainable technology, this represents a meaningful shift: energy harvesting devices are finally crossing the threshold from novelty to practical utility.

Credit & Disclaimer: This article is a popular science summary written to make peer-reviewed research accessible to a broad audience. All scientific facts, findings, and conclusions presented here are drawn directly and accurately from the original research paper. Readers are strongly encouraged to consult the full research article for complete data, methodologies, and scientific detail. The article can be accessed through https://doi.org/10.1039/D5EE01052G