In a study published in Nature, an international team led by Prof. LI Wei from the Changchun Institute of Optics, Fine Mechanics and Physics (CIOMP) of the Chinese Academy of Sciences, introduced a novel miniaturized photodetector capable of characterizing arbitrary polarization states across a broadband spectrum with a single device and a single measurement.

The intricate nature of light, characterized by its intensity, polarization, and spectrum composition, holds great importance across a range of scientific and technological disciplines. Traditional photodetectors can only measure light intensity, while current polarization and spectrum photodetectors often enhance detection capabilities by integrating multiple polarization- or wavelength-sensitive elements in time or space.

In many natural scenarios, light may carry high-dimensional information, i.e., arbitrary changes in polarization and intensity across a broad spectrum. Current photodetectors typically sacrifice one dimension of information for another as they can measure either intensity and polarization at a fixed wavelength or intensity and wavelength under uniform polarization.

Researchers in this study proposed an innovative idea of using spatial dispersion and frequency dispersion of optical interfaces to modulate polarization and spectrum response in wavevector space. With the deep learning method to decode the polarization and spectrum information, they realized the detection of high-dimensional light field information by a single measurement using a single device.

Researchers theoretically and experimentally verified that the proposed photodetector has the ability to detect high-dimensional information, and that its polarization detection accuracy and spectral resolution are comparable to those of current sole-functional polarimeters and spectrometers.

In addition, the ultra-integrated high-dimensional imager can be realized by simply combining a thin film with a microlens array and an imaging sensor array in a “sandwich” style, which requires no alignment and a single measurement. This provides a way for ultra-compact, high-dimensional information detection and imaging.

Prof. LI envisioned that ultra-broadband detection could be achieved by integrating broadband commercial photodetectors, the detection resolution could be further improved by using photonic crystals, metasurfaces, and two-dimensional materials instead of existing thin film schemes, and the detection capability could be stepped up in higher dimensions by integrating functionalities like image processing and distance measurement.