Comparability across infrared spectroscopy techniques and substrates for bioanalytical applications.

The development of quantum cascade lasers (QCLs) brought new possibilities in infrared (IR) spectroscopy for bioanalytical applications, offering highly enhanced resolution, sensitivity, speed, and multi-modal capability. This has been supported by QCL-based IR techniques, including the laser direct infrared (LDIR) and optical photothermal IR (O-PTIR) methods. Despite their vast strengths, the comparability of these IR techniques with the established Fourier transform infrared (FTIR) spectroscopy technique remains an open question. In this study, we systematically evaluate and compare O-PTIR, LDIR, and FTIR spectroscopy using six representative biologically relevant samples (ATP, brain lipid extract, cholesterol, glycogen, lysozyme, and urea) and three commonly used substrates (CaF2, silicon, and MirrIR). The measurements were generally performed in the single-point mode, with brain lipid and urea measured additionally in the mapping mode to investigate the influence of sample heterogeneity. On this basis, we examined three sources of variability in this study: sample heterogeneity, IR technique, and substrate. The variability was characterized from the perspective of the band intensities, band positions, band widths, and spectral similarities based on the Pearson's correlation coefficient. Overall, the measurements demonstrate clear differences across both techniques and substrates. The LDIR measurements are largely different from FTIR and O-PTIR measurements, while CaF2 exhibited better stability than MirrIR (Kevley) and silicon. More importantly, we found that the substrate plays a role in band positions, while the variability across IR techniques does not show clear patterns. With this study, we expect to draw attention to the unavoidable spectral variability in IR measurements across both IR techniques and substrates. This should be carefully considered, especially in multi-center studies.