Scholarlink

<p>Compact mid-infrared (MIR) spectrometers based on tunable interband cascade lasers (tICLs), which are integrated with thin-film waveguides, represent a promising platform for discrete-emission spectroscopic measurements in aqueous systems, operating across the amide I region (1692–1640&#xa0;cm<sup>− 1</sup>). This study presents a wavenumber-resolved performance evaluation of a prototype tICL-based spectrometer using bovine serum albumin (BSA) standards at different concentrations to assess the limit of detection (LOD), limit of quantification (LOQ), and signal-to-noise ratio (SNR), while benchmarking its performance against a compact Fourier-transform infrared attenuated total reflectance (FTIR-ATR) system to contextualize analytical behavior. Although the prototype showed acceptable initial performance, system-induced variability affected the analytical sensitivity and reproducibility of the measurements. To address these limitations, a wavenumber-resolved analytical framework was applied to quantify sample-induced spectral variations. Absorbance variation was analyzed at individual emission points, focusing on concentration dependence and replicate consistency. Using the strictly standardized mean difference (SSMD) metric and <i>z</i>-score standardization, wavenumbers were ranked by concentration-level discriminability, with those having positive <i>z</i>-scores selected for further analysis. When applied to an independent dataset, the SSMD-based selection yielded lower LOD (1.24% to 0.71%) and LOQ (3.75% to 2.17%), reflecting improved analytical sensitivity through channel selection. Partial least-squares regression (PLSR) score-space clustering further demonstrated enhanced separation of BSA concentrations with SSMD-filtered subsets. Overall, the proposed wavenumber-resolved performance framework enabled the selection of high-quality emission points, supporting the optimization of compact tICL-based MIR spectrometers for protein-relevant analytical measurements and industrial process-monitoring contexts.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Tunable Interband Cascade Laser-Based Mid-Infrared Spectroscopy for Protein Sensing: Wavenumber-Resolved Performance Evaluation within the Amide I Region

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月31日

查看全文
  • Mehmet Can Erdem,
  • Bijay Kafle,
  • Pranish Karki,
  • Uladzislau Blazhko,
  • Boris Zimmermann,
  • Maren Anna Brandsrud,
  • Julian Scheuermann,
  • Robert Weih,
  • Johannes Koeth,
  • Boris Mizaikoff,
  • Achim Kohler

摘要

Compact mid-infrared (MIR) spectrometers based on tunable interband cascade lasers (tICLs), which are integrated with thin-film waveguides, represent a promising platform for discrete-emission spectroscopic measurements in aqueous systems, operating across the amide I region (1692–1640 cm− 1). This study presents a wavenumber-resolved performance evaluation of a prototype tICL-based spectrometer using bovine serum albumin (BSA) standards at different concentrations to assess the limit of detection (LOD), limit of quantification (LOQ), and signal-to-noise ratio (SNR), while benchmarking its performance against a compact Fourier-transform infrared attenuated total reflectance (FTIR-ATR) system to contextualize analytical behavior. Although the prototype showed acceptable initial performance, system-induced variability affected the analytical sensitivity and reproducibility of the measurements. To address these limitations, a wavenumber-resolved analytical framework was applied to quantify sample-induced spectral variations. Absorbance variation was analyzed at individual emission points, focusing on concentration dependence and replicate consistency. Using the strictly standardized mean difference (SSMD) metric and z-score standardization, wavenumbers were ranked by concentration-level discriminability, with those having positive z-scores selected for further analysis. When applied to an independent dataset, the SSMD-based selection yielded lower LOD (1.24% to 0.71%) and LOQ (3.75% to 2.17%), reflecting improved analytical sensitivity through channel selection. Partial least-squares regression (PLSR) score-space clustering further demonstrated enhanced separation of BSA concentrations with SSMD-filtered subsets. Overall, the proposed wavenumber-resolved performance framework enabled the selection of high-quality emission points, supporting the optimization of compact tICL-based MIR spectrometers for protein-relevant analytical measurements and industrial process-monitoring contexts.