High-sensitivity computational miniaturized terahertz spectrometer using a plasmonic filter array and a modified multilayer residual CNN

Abstract

Spectrometer miniaturization is desired for handheld and portable applications, yet nearly no miniaturized spectrometer is reported operating within terahertz (THz) waveband. Computational strategy, which can acquire incident spectral information through encoding and decoding it using optical devices and reconstruction algorithms, respectively, is widely employed in spectrometer miniaturization as artificial intelligence emerges. We demonstrate a computational miniaturized THz spectrometer, where a plasmonic filter array tailors the spectral response of a blocked-impurity-band detector. Besides, an adaptive deep-learning algorithm is proposed for spectral reconstructions with curbing the negative impact from the optical property of the filter array. Our spectrometer achieves modest spectral resolution (2.3 cm^-1) compared with visible and infrared miniaturized spectrometers, outstanding sensitivity (e.g., signal-to-noise ratio, 6.4E6: 1) superior to common benchtop THz spectrometers. The combination of THz optical devices and reconstruction algorithms provides a route toward THz spectrometer miniaturization, and further extends the applicable sphere of the THz spectroscopy technique.

Keywords: adaptive deep-learning algorithm; computational miniaturized terahertz spectrometer; plasmonic filters.

Figure 1:

Theoretical simulations for the plasmonic THz filters. (a) Schematic diagram of a filter structure including a periodic square lattice array of Al holes and an ultrapure Si substrate. (b) Simulated transmission spectra of the filters with different periods in the array. Inset: the resonant wavelength in the transmission spectra as a function of the array period. (c) The electric field intensity mapping under the irradiation of monochromatic THz wave with the primary-peak frequency at the interface of the Al layer and Si substrate. (d) The electric field intensity mapping at the vertical cross-section. (e & f) the electric field intensity mapping corresponds to that in (c) and (d), respectively, at the non-resonant frequency. The color-scale bar represents the ratio of the enhancement electric-field and the incident electric-field magnitudes in (c–f).

Figure 2:

Optical properties of the fabricated hardware. (a) A skew-view SEM image of the periodic structure in a fabricated Al hole filter. (b) Comparison of experimental and simulated transmission spectra of the filters with different periods. (c) Measured transmission spectra of Al hole filters in the first row of the 16 × 15 filter array, with periods increasing from 14 μm to 36.5 μm with a step of 1.5 μm. (d) The measured and the simulated primary-peak wavelength as a function of the period. (e) Response spectrum of the Ge:P BIB detector. Inset: an SEM image of the detector with false-color AL and BL. (f) Source radiation spectrum within the working range of the spectrometer. Insert: the whole radiation spectrum of the Hg-arc light source, where the red shadow corresponds to the working range.

Figure 3:

Operational procedure of our miniaturized spectrometer system. (a) The operational scheme of the computational spectrometer, involving the hardware (a THz illuminant, a filter array, and a detector) and the software (reconstruction algorithms). (b) The flowchart of the proposed AD-ResNet algorithm, where “Conv”, “ReLU”, and “Deconv” represent convolution, rectified linear unit, and deconvolution, respectively.

Figure 4:

Spectral reconstruction performance of the spectrometer system. (a) The transmission spectra of the same cobalt violet deep analyte reconstructed by the spectrometer with the different algorithms and their ground truths. (b) The transmission spectra of five different analytes reconstructed by the spectrometer with the proposed AD-ResNet and their ground truths. In (a) and (b), the MSEs for evaluating the reconstruction quality are given. (c) MSE distributions for the 40 additional analytes obtained with the deep-learning algorithms. (d) The MSE loss versus epoch during the trainings with the AD-ResNet in the five-fold cross validation, where K1 represents the regular training, K2–K5 represent other folds; inset: the average MSE for the test dataset in different folds.

Figure 5:

Spectral resolution presentation. (a–c) The reconstructed spectra using the deep-learning algorithms when the simulated spectra are with the same FWHM of 0.6 cm⁻¹. (d) The reconstructed spectra using the AD-ResNet when the simulated FWHM is 3.0 cm⁻¹. (e) The FWHMs of the output spectra versus those of the simulated spectra for the deep-learning algorithms. (f) The comparison of the spectral RRB of our spectrometer with other miniaturized spectrometers [, , , , , –44].