Nonlinear memristive computational spectrometer

Xin Li; Jie Wang; Feilong Yu; Jin Chen; Xiaoshuang Chen; Wei Lu; Guanhai Li

doi:10.1038/s41377-024-01703-y

Nonlinear memristive computational spectrometer

Light Sci Appl. 2025 Jan 14;14(1):47. doi: 10.1038/s41377-024-01703-y.

Authors

Xin Li^{1

2

3}, Jie Wang¹, Feilong Yu¹, Jin Chen¹, Xiaoshuang Chen^{1

3

4

5}, Wei Lu^{1

2

3

4

5}, Guanhai Li^{6

7

8

9}

Affiliations

¹ State Key Laboratory of Infrared Physics, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, 500 Yu-Tian Road, Shanghai, 200083, China.
² School of Physical Science and Technology, ShanghaiTech University, Shanghai, 201210, China.
³ University of Chinese Academy of Science, No. 19A Yuquan Road, Beijing, 100049, China.
⁴ Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, No.1 Sub-Lane Xiangshan, Hangzhou, 310024, China.
⁵ Shanghai Research Center for Quantum Sciences, 99 Xiupu Road, Shanghai, 201315, China.
⁶ State Key Laboratory of Infrared Physics, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, 500 Yu-Tian Road, Shanghai, 200083, China. ghli0120@mail.sitp.ac.cn.
⁷ University of Chinese Academy of Science, No. 19A Yuquan Road, Beijing, 100049, China. ghli0120@mail.sitp.ac.cn.
⁸ Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, No.1 Sub-Lane Xiangshan, Hangzhou, 310024, China. ghli0120@mail.sitp.ac.cn.
⁹ Shanghai Research Center for Quantum Sciences, 99 Xiupu Road, Shanghai, 201315, China. ghli0120@mail.sitp.ac.cn.

PMID: 39809745
DOI: 10.1038/s41377-024-01703-y

Abstract

In the domain of spectroscopy, miniaturization efforts often face significant challenges, particularly in achieving high spectral resolution and precise construction. Here, we introduce a computational spectrometer powered by a nonlinear photonic memristor with a WSe₂ homojunction. This approach overcomes traditional limitations, such as constrained Fermi level tunability, persistent dark current, and limited photoresponse dimensionality through dynamic energy band modulation driven by palladium (Pd) ion migration. The critical role of Pd ion migration is thoroughly supported by first-principles calculations, numerical simulations, and experimental verification, demonstrating its effectiveness in enhancing device performance. Additionally, we integrate this dynamic modulation with a specialized nonlinear neural network tailored to address the memristor's inherent nonlinear photoresponse. This combination enables our spectrometer to achieve an exceptional peak wavelength accuracy of 0.18 nm and a spectral resolution of 2 nm within the 630-640 nm range. This development marks a significant advancement in the creation of compact, high-efficiency spectroscopic instruments and offers a versatile platform for applications across diverse material systems.