Two-dimensional photodetectors enabling ultraminiaturized computational spectrometers

Miniaturized spectrometers that can be seamlessly integrated into on-chip, portable, and wearable systems are essential for next-generation sensing and information-processing technologies. Computational spectrometers, which reconstruct incident spectra by combining device-dependent spectral responses with reconstruction algorithms, offer a unique miniaturization route beyond conventional dispersive or interferometric architectures. Their ultimate performance is jointly governed by spectral encoding capability and reconstruction robustness. In the detector-encoding route, the choice of photodetector materials and tuning strategies directly determines the dimensionality, diversity, and stability of spectral encoding, thereby setting the upper limit of reconstruction accuracy and the attainable degree of system miniaturization. Two-dimensional (2D) van der Waals (vdW) materials have recently emerged as exceptional candidates for realizing high-performance ultraminiaturized computational spectrometers, owing to their atomic thickness, strong light-matter interactions, continuously tunable band structures, and vdW heterostructures compatibility. These properties make 2D photodetectors particularly attractive for approaching the ultimate limit of device-level miniaturization while maintaining rich spectral encoding capabilities. This review provides a comprehensive perspective on 2D photodetector-based computational spectrometers. We first analyze the theoretical foundations and operational schemes of linear and nonlinear computational spectrometers, highlighting the distinct roles of device physics in spectral encoding. We then discuss spectral response modulation strategies and the evaluation of spectral encoding effectiveness, followed by reconstruction algorithms with particular emphasis on compressive sensing and deep learning. Key performance metrics are summarized and compared across representative demonstrations. Emerging proof-of-concept applications are subsequently reviewed. Finally, we outline future challenges and opportunities that may advance this field from current lab-scale demonstrations toward practical edge-device deployment.