Breathalyzer accuracy

I saw an article about a pilot in Little Rock who got pulled aside because the TSA smelled alcohol on his breath. After a few shenanigans he took a breathalyzer and blew a .004 (note the extra 0 - not a .04, but a .004).

This is 1/10th of the FAA limit (.04).

I have two questions, I guess:

1.When BAC levels get very low, how accurate (or how sensitive) are the machines? Are all breathalyzers accurate to a .001 level? Is a .004 result accurately measured versus a .005 result?

  1. At these extremely low levels, are the machines actually detecting intoxicating alcohol? Could a Certs, Tic-Tac or Altoids cause a reading this low? How about using Scope or Listerine in the morning? If not these, then what could cause a (very low) positive reading? Is there some number that most police agencies use as a cutoff below which it is considered “nothing found”?

Actually, that now looks like seven questions! :rolleyes:

  1. Yes, they’re accurate; assuming that it’s calibrated properly, measuring hundredths of a percent is no problem (and quite common).

  2. The machines detect alcohol from the bloodstream, which is transferred to the air deep in the lungs. The lungs have to be practically emptied to get a reading, and there is nothing you can put in your mouth that will affect the result.

I’ll start off with the disclaimers and qualifications. IANAL, but my (soon to be ex-) wife is. She’s practiced this routine criminal law stuff for about 12 years now and I’ve heard it all from her. But that’s not all. I’m an analytical chemist with experience in instrumental detection, identification and quantification of chemical compounds, and I’ve attended the Borkenstein course at IU.

Probably not, although they could be quite easily. In analytical chemistry, you calibrate instruments for the range of interest. In the case of blood/breath alcohol detection and quanitization, such low levels are not worth worrying about. As for the accuracy at such low levels; again, probably not. These instruments are typically calibrated at one or two points in the analytical range of interest which is pretty far away from the low levels you’re asking about. And it is usually the case that beyond the range of calibration, accuracy suffers as sensitivity becomes squirrely and non-linear. That said, detecting compounds at such a level isn’t that tough. I’m an analytical chemist and I routinely use a similar instrument to measure compounds at those levels with great accuracy. But then, I’m also calibrated for those low levels.

Maybe. The instrument really isn’t even detecting alcohol at all, but instead, is detecting the presence of Methyl groups (-CH[sub]3[/sub]), one of which the ethanol molecule has. The instrument detects and quantitizes methyl groups. The gigantic assumption is that all of the methyl groups detected in a breath sample originate from ethanol molecules. However, methyl groups are found on lots of other compounds besides ethanol which might be inside a person. Acetone, common in diabetics (and Atkins dieters) during ketosis has two methyl groups. (Most breathalyzers detect and compensate for acetone) Vinegar is dilute acetic acid, which has a single methyl group. Some aerosol propellants have derivitaves of methane and ethane, which contain methyl groups. And the compounds responsible for some flavors are loaded with methyl groups. Onions, garlic and mint come to mind.

Yeah. Possibly. Foods and flavors would tend to impart a lesser signal on the instrument since they don’t much end up in the lungs via circulation, but tend to stay gastric. But it’s been pretty well established that “mouth alcohol”, if it were present, is “completely” dissipated after 15-20 minutes, thus the SOP of keeping the subject under observation for at least 15 minutes to make sure they don’t eat, drink or put anything in their mouth.

Most likely. But it’s not likely a firm and consistent number, even in the same department, or even with the same officer. If the breathalyzer result is above the legal limit for the jurisdiction, they can easily persue conviction. If the result is detectable but somewhat below, they can still try for a conviction based on “impairment”, but that is a lot more subjective and has a much lower chance of a successful conviction. With a blood/breath alcohol result of 0.004, unless, the subject was running over elderly women crossing the street, they’d probably apoligize for taking up the citizen’s valuable time.

Thank you for the replies - especially honkytonkwillie for that thorough answer.

Mods, don’t get mad at me for bumping a post with a thank you - that answer was worth one!:smiley: