Development, Testing, and Validation of a Prototype for Qualification of Substances Based on Near-Infrared Spectroscopy

Abstract

Some commercial equipment used to qualify substances in complex mixtures employs the principle of observation of the interaction of these substances with electromagnetic radiation. The chemical bonds present in the molecules of the substance absorb specific wavelengths, allowing for extracting important information about the physical properties of the analyzed element. However, such equipment is expensive, hampering access to it. With new optoelectronic components available in the market, it is possible to develop devices with the same operating principle of commercial equipment. With this in mind, several authors have suggested prototypes with a lower cost that are applicable to the ultraviolet and visible range, making their development less complex. This paper presents the development of an optoelectronic prototype for substance qualification using LEDs as the light source, with emission in the range of 1400-2250 nm (nearinfrared range). A GaInAsSb-based photodiode was used to detect the signal. The sampling mode used in the development of this work was diffuse reflection, which is applicable to non-homogeneous substances. Bandpass filters were employed to validate the methodology, where the results show that the prototype can perceive the variation of the intensity of the spectrum received by the sensor. INDEX TERMS Optoelectronic devices, instrumentation and measurement, infrared sensor, optical filters, light emitting diodes. The associate editor coordinating the review of this manuscript and approving it for publication was Jiang Wu.

A Review of Data Analysis Techniques Used in Near-Infrared Spectroscopy

“…As mentioned above, NIRS data are acquired through photodiode-based sensors. These sensors generally output as analog signals and are transferred to the computer with the help of analog-digital converter circuits [17]. For this reason, deviations in the signal, called noise, which negatively affects the data analysis, occur during both the measurement phase and the conversion phase.…”