The least invasive and most commonly used technique for evidential indicators of blood alcohol concentration (BAC) is breath analysis. An indication of 0.8 mg of ethanol (the statutorily regulated alcohol) in 2.1 L of breath, or one half part per million ethanol by volume, corresponds to the legislative recommendation for the legal maximum 0.08% (m/v) BAC in operators of motor vehicles. For this reason, infrared radiation (IR) absorbance, which is extremely sensitive to most chemical environments, relatively inexpensive and portable, is frequently is used in evidential breath analyzers.

The three-filter Intoxilyzer 5000 is an example of an evidential breath analysis devise that employs infrared radiation. It was designed to compare relative absorbance in three different spectral regions to distinguish between ethanol and acetone and ultimately determine breath ethanol concentration. The centers of the three spectral regions are 2950 cm-1 (3.39 µm), 2874 cm-1 (3.48 µm) and 2632 cm-1 (3.80 µm). The absorbance at the nominal 2632 cm-1 region is used to establish a baseline. In the ideal absence metabolites such as acetone or other IR absorbing molecules, ethanol concentration could be determined simply by measuring the IR absorbance at the nominal 2874 cm-1 region using Beer's law A= abc, with " a"(2874 cm-1) and "b" predetermined. However, acetone and the other molecules also absorb IR at 2874 cm-1. Their presence in the Intoxilyzer 5000 could elevate the reported BAC, unless identified as "Interferents".

Acetone is the prototypical Interferent in the Intoxilyzer 5000. To identify its presence, the three-filter Intoxilyzer 5000 adjusts the relative absorptivities of ethanol during calibration so that nominally a(2874 cm-1) = a(2950cm-1). Because the relative absorptivities of acetone significantly differ from that of ethanol in these two regions, it is identified as an Interferent.

Our research group compares the Intoxilyzer 5000 with FTIR in identifying other respired molecules that may contribute to false positive readings by the Intoxilyzer 5000, primarily due to a person's occupational exposure to such chemicals. This research has developed a formula where the Intoxilyzer's response to such chemicals can be predicted from the chemical's infrared absorbance as determined by an FTIR. Significantly, potential Intoxilyzer 5000 interferents can be identified in the absence of that instrument.