Perovskite Image Sensor Delivers Better Images for Humans and Computers

Researchers from ETH Zurich and Empa have developed a novel image sensor based on perovskite, a semiconductor material that improves colour accuracy and reduces image artefacts.

The new technology offers significant advantages over traditional silicon-based sensors, including enhanced light sensitivity, better colour reproduction, and higher resolution. The perovskite image sensors also show great potential for machine vision applications.

The research is published in Nature.

Overcoming limitations of silicon sensors

Image sensors are crucial components in digital cameras and smartphones, enabling the detection of colour by capturing light through individual pixels. In traditional silicon-based sensors, light must pass through filters to detect specific colours like red, green, and blue (RGB). This process results in the loss of some light, reducing efficiency and limiting the sensor's performance. In contrast, the new perovskite-based sensors are designed to capture all available light without the need for filters.

The breakthrough involves using lead halide perovskite, a crystalline semiconductor material with unique properties that can be tuned to absorb specific wavelengths of light. By adjusting the chemical composition of the perovskite, the researchers can make the material absorb red, green, or blue light while remaining transparent to other wavelengths.

This approach eliminates the need for traditional RGB filters and allows the pixels to be stacked vertically within the sensor, resulting in increased light capture and improved spatial resolution.

Demonstrating the advantages

In initial experiments, the team demonstrated that the perovskite sensors can capture up to three times more light than conventional sensors of the same size. This results in greater sensitivity to light and the ability to deliver images with much higher resolution. The new image sensors are also capable of reducing common digital photography artefacts such as demosaicing and the moiré effect, which arise from the limitations of traditional image processing.

The researchers have successfully built two fully functional perovskite-based image sensors, marking a significant step forward in the development of this technology. While the technology is still in its early stages, the progress made in miniaturizing the sensors suggests that they could soon be ready for commercial use.

“We are developing the technology further from a rough proof of principle to a dimension where it could actually be used,” said study author Maksym Kovalenko, a professor of inorganic chemistry and the head of the Functional Inorganic Materials group at ETH Zurich. “The first transistor consisted of a large piece of germanium with a couple of connections. Today, 60 years later, transistors measure just a few nanometers.”

Potential applications in machine vision

Beyond consumer cameras, perovskite image sensors hold promise for use in machine vision systems, which are critical in fields such as medicine, agriculture, and environmental monitoring.

Unlike traditional RGB sensors, perovskite sensors can be customized to capture light in specific wavelength ranges, enabling hyperspectral imaging. This technique, which captures more than the standard three colour channels, can be used for tasks such as detecting specific biological markers in medical diagnostics or monitoring crop health in precision agriculture.

How To Detect Trace Metals During Semiconductor Manufacturing

This application note demonstrates how advanced multi-quadrupole ICP-MS technology can revolutionize ultra-pure NMP analysis for effective semiconductor manufacturing quality control.
View App Note / Case Study

Advertisement

The ability to precisely control the wavelength range absorbed by each layer of perovskite is a key advantage over silicon-based sensors, which rely on complex filters and algorithms to capture a broader spectrum of light. Perovskite sensors could therefore simplify and improve the performance of hyperspectral imaging systems.

Next steps and future prospects

The next goal for the researchers is to further reduce the size of the perovskite pixels and increase their density, with the aim of achieving even higher resolution. Current prototypes have pixel sizes ranging from 0.5 to 1 millimetre, but commercial sensors typically feature much smaller pixels in the micrometre range. The researchers are confident that perovskite-based sensors can be miniaturized further and are working on adapting the electronic connections and processing techniques to suit the new semiconductor material.

While challenges remain, such as optimizing the readout electronics for perovskite sensors, the team believes these hurdles can be overcome. The researchers are optimistic about the future of perovskite image sensors, which could offer significant improvements in both consumer imaging and industrial machine vision applications.

Reference: Tsarev S, Proniakova D, Liu X, et al. Vertically stacked monolithic perovskite colour photodetectors. Nature. 2025;642(8068):592-598. doi: 10.1038/s41586-025-09062-3

This article has been republished from the following materials. Note: material may have been edited for length and content. For further information, please contact the cited source. Our press release publishing policy can be accessed here.

This content includes text that has been generated with the assistance of AI. Technology Networks' AI policy can be found here.