Preliminary alcohol screening (PAS) devices and the results they purport to provide has been the subject of much debate. In some jurisdictions they are called preliminary breath tests (PBT). Either way, they refer to the instruments used by police to conduct tests in the field in order to try to determine whether there is alcohol present in the blood. While some tests only detect the existence of alcohol (these "flashlight" sensors are essentially detecting if there is any alcohol present, but not how much) most commonly they are used to determine a specific level of alcohol in the blood. There are two types of PAS devices that accomplish this: fuel cell devices and Taguchi gas sensor devices.
PAS Devices: Fuel Cell and Taguchi Gas Sensor
Fuel cell devices work by electrochemically oxidizing ethanol in the subject's breath. Ethanol (or ethyl alcohol) is the type of alcohol found in wine, beer, and distilled spirits. The electrical current created by the process is then measured and the resulting reading gives the officer an indication of the amount of alcohol present. There are many inherent problems with these devices, the most prominent being the lack of specificity. Specificity merely refers to the fact that these devices, while working to detect the presence of ethanol, have also been known to show similar results with the presence of a variety of other chemicals.
Researchers have found acetaldehyde, methanol, isopropanol, and n-propanol have all reacted positively to tests while using the Alcometer Pocket Model. Aside from specificity, there is an array of other issues that can cause false positives. For instance "mouth alcohol" can result in artificially raising the levels of alcohol in the breath. If the test has been administered recently after eating, drinking, smoking, a burp, belch, or vomiting, the PAS can detect more ethanol in the breath than is actually impairing the individual.
As such, a 15-20 minute observation period is necessary to ensure that no burping or similar act has occurred to result in extra mouth alcohol. Other issues, such as oxide buildup, improper care for units, how recently the unit was used (an issue in sobriety check points) and battery life has also been known to skew test results. The Alco-Sensor III, along with Alco-Sensor IV and Alcotest 7410 Plus, are the most commonly used devices by the California Highway Patrol. The Alco-Sensor III has been demonstrated insufficiently accurate for use as evidentiary testing.
Along with fuel cell devices, the Taguchi gas sensors are also commonly used. Chemicals in the sensor are attracted to alcohol and result in increased electrical conductivity. The two types of units are active and passive gas sensors. The passive units (the aforementioned flashlight tests) merely take readings off of the air around the subject. This is typically accomplished by sticking the officer's "flashlight" into the car to read a driver's license, and the device detects alcohol in the air. This creates obvious issues, since it often detects a variety of things other than alcohol in the driver, such as alcohol from passengers and other products like aftershave and perfume. Both active and passive tests have the same issue with specificity as fuel cell devices, as both have been known to test positively for compounds other than ethanol. Lastly, temperature also affects readings for both active and passive units.
Though these tests are typically used as a preliminary indication of the presence of alcohol, this often will bias an officer's judgment on subsequent FST. California is currently moving towards adopting a single handheld device for all evidential breath tests, eliminating the current "bench top" machines found in police stations. This will possibly provide new angles for attorneys to explore, given the unreliable nature of the devices and the skepticism from most jurors towards these handheld devices. For example, the Alcotest 7410 Plus, the device currently being examined for this purpose, has been found to have specific faults, namely a lack of mouth alcohol detectors (slope-detectors) and no radio frequency interference (RFI) detector. This lack of RFI detection means problems can arise from the officer's radio transmitters, walkie-talkies, and as previously mentioned these fuel cell analysis devices do not detect only ethanol.