Schaffer Online Library of Drug Policy Sign the Resolution
Contents | Feedback | Search
DRCNet Home
| Join DRCNet
DRCNet Library | Schaffer Library | Drugs and Driving 


Breath Alcohol Instrumentation: A Proposal in Commercial Taxonomy

Professor R.J. Breakspere* and Dr P.M. Williams**

* Central Queensland University, Rockhampton, Queensland 4702, Australia

** Lion Laboratories plc, UK


Commercial descriptions of breath alcohol measuring instruments may sometimes mislead prospective purchasers as to the true capabilities of such products.

In this paper we propose an international taxonomical standard against which all such devices should be judged, and their relative merits then properly assessed. To class such instruments by application offers no easy answer. For example, some fuel cell based handheld instruments are used for both roadside screening and evidential testing applications. Conversely, certain infrared devices used evidentially in some countries find only screening test applications in others. Breath sampling systems have often been the subject of much debate and dispute, especially regarding what is meant by the term 'passive'. Confusion also exists with alcohol sensors: semiconductors and fuel cells have very different analytical capabilities, but these are not always well realised.

We propose there are four criteria of taxonomy against each of which manufacturers should classify their products. These criteria are: Application; Size, Weight and Mobility; Breath Sampling Method; and Principle of Alcohol Detection. This simple approach should assist purchasers of breath alcohol instruments in traffic law enforcement and the newer but expanding industrial and public markets make correct decisions concerning the appropriate instrument for their application.


Over recent years technological advances in breath alcohol analysis and an increase in innovation in occupational health and safety, police and public use have led to the development and marketing of well over forty different instruments world-wide.

This plethora of instruments and, in particular, the way they are sometimes described to the consumer have led to some degree of confusion in the market place. This confusion is not just with persons new in the field but also where a central tendering agency is used.

In this paper we are hoping to initiate discussion which may lead to accepted descriptors for breath alcohol testing equipment world-wide. We suggest that breath testing instruments can be comprehensively described using four descriptors, viz.

  1. Application(s)
  2. Portability and mobility
  3. Breath sampling method
  4. Alcohol measurement technology


This is, of course, the end use (or uses) for which the breath testing device is intended and can be covered by four principal applications.


Screening involves the testing of subjects to determine whether their body alcohol concentration (i.e. breath or blood) is above or below some particular defined level (the, so-called, per se level) or even to determine whether they have any alcohol in their body at all.

Breath alcohol screening may be carried out in several market sectors such as road traffic law enforcement, industrial safety programs and in various branches of medicine.


As the name suggests, evidential testing involves the collection of evidence of a subject's alcohol concentration in an acceptable form to be used in a Court or Industrial Tribunal.

The analytical situation is that, apart from some semiconductor based devices, most modern breath alcohol instruments are technically quite capable of giving breath alcohol readings to an acceptable degree of accuracy. It is, however, the level of documentation and proof of accuracy required by the relevant legislation, rather than its analytical capability, that generally determines whether a particular instrument can be used evidentially.

For instance, in some countries it is sufficient to record the reading given by a portable handheld instrument by simply handwriting the BAC level onto the appropriate document, which may then be countersigned by the subject. However, other authorities will only accept a timed and dated printout from a fully automated tamperproof instrument complete with proof of the accuracy of calibration using a traceable standard.


These are instruments specially designed and produced for permanent connection to the electrical system of a vehicle or machine in order to prevent or deter its operation by persons with more than a permitted (pre-set) level of alcohol in their system.

Instruments, to date, have been based on both fuel cell and semiconductor sensing technologies.

Personal Self-Testers

The right of people to test themselves for alcohol as a means of ensuring compliance with laws and regulations has meant a demand by the public to purchase or use such devices. Their need may be to check whether their alcohol level is above or below the driving per se limit or, as is now increasingly required by employers, that their alcohol level is zero before commencing work.

These instruments have, to date, been produced in two forms, namely, portable hand-held devices (usually based on semiconductor detectors) and wall mounted - usually coin operated - devices predominately using fuel cell sensors.

In Australia it has been recognised that it is necessary to ensure that any breath alcohol instrument used by the public must possess a minimum (but high) level of accuracy and reproducibility. To this end a national standard (Breath testing devices for public use, AS 3547-1993) has been produced to cover bag and tube, portable electronic, interlock and wall mounted devices.


Both the physical size and the power supply requirements of an instrument will determine the above. There are three possibilities:


These instruments are small enough to be carried in the field, on the user, and require no external source of mains power. Depending on their accuracy, and local legislation and regulations, they may be used for screening, public use and evidential applications - or both. Most of these instruments are small hand-held devices but others have accessories such as printers.


These instruments are generally produced for evidential testing in the field. They require a continual source of external power which may be derived from a vehicle's electrical system and are normally rugged enough to be carried and operated in mobile field applications. They are generally designed so that they can also be operated from the mains power supply and thus their use in police stations is also possible.

Stationery or Fixed

These are non-portable, non-mobile instruments, generally requiring a continual supply of mains electrical power. They are intended for use only in fixed locations such as Police Stations although they can sometimes be operated in specially designed Police vehicles or caravans.


There is much confusion and misunderstanding in the marketplace about what is meant by the terms active breath testing and passive breath testing.

Before discussing the meaning of these two terms it is important to consider the two possible results of a breath test:

  1. a quantitative measurement of the subjects body alcohol concentration by analysis of deep lung or alveolar breath. This form of testing is generally carried out to determine if the level of alcohol in the subject is above or below some defined value other than zero - usually known as the per se level; or
  2. a qualitative determination of alcohol to see if the subject has any alcohol at all in his or her body - but with no particular concern for how much. This form of test would be carried out in any country where it was forbidden to drive with any alcohol in body. It is also common in industry and the military.

Let us now look at each of the three possible types of breath sampling:


In the Active mode the subject blows into the instrument, through a mouthpiece, so as to provide a sample of deep lung or alveolar breath. It is, of course, only in the alveoli that full equilibration occurs between the lung air and blood, so only by analysing deep lung breath can a quantitative measurement of BAC or BrAC be obtained.

Because the subject has to cooperate and blow properly into the mouthpiece the method is called active.

Active, no mouthpiece

At police roadblocks or in high intensity testing at in factories, most subjects will have a zero alcohol level. The important factor is therefore to test as many subjects as quickly as possible and with minimum costs per test, in order to pick out those few subjects who do have alcohol in their bodies and who may therefore be required to undergo further testing.

This breath sampling method is achieved by having the subject blow actively (i.e. with co-operation) into the device but with no physical contact. This means that a new mouthpiece is not required for each test.


A true passive instrument, as the name suggests, requires the subject to do nothing. The device sucks into itself a sample of air taken from in front of the subjects face and measures the alcohol content. The length of the sampling time should ensure that the subject is breathing out or talking at the time the sample is taken. Co-operation of the subject is therefore not required.

Since there is no sample control, and hence no certainty that what is analysed is expired breath at all, it is possible for zero alcohol readings to be produced on subjects who may have significant levels of alcohol in their blood.

A number of instruments now combine a passive screening with an active (i.e. with mouthpiece) measurement in what we suggest should be called a dual sampling device. Furthermore, it is not inconceivable that instruments with triple sampling capabilities may be developed although we cannot, at this time, see the need for such a device.


In this paper we identify six modes of alcohol measurement:

Fuel Cells

A fuel cell sensor is an electrochemical device in which the substance of interest, e.g. alcohol, undergoes a chemical oxidation reaction at a catalytic electrode surface to generate a quantitative electrical response.

Fuel cells are characterised by the following positive analytical features towards breath alcohol testing:

Disadvantages with fuel cell sensors for breath alcohol analysis are:

It must also be remembered that a fuel cell requires a separate sampling system, to inject or draw into it a small but fixed volume of the sample vapour to be analysed. Fuel cells, unlike infrared systems, are not continuous flow analysers. This means they cannot monitor the expired alcohol concentration curve, either to determine the passage of deep lung air, or to detect the presence of mouth alcohol.


This sensor consists of a small bead of a transition metal oxide, heated to a temperature of around 300 C, across which a voltage is applied to produce a small standing current. The magnitude of this current is determined by the conductivity of the surface of the bead, which may be affected by the presence [and concentration] of any substances adsorbed on to it.

So when alcohol [or one of many other substances] comes into contact with this bead, it is adsorbed on to the surface, changes the surface resistivity, and hence the standing current. This change in current is taken as a measure of the concentration of alcohol in the sample. The effect is a purely physical one: it is not specific to the alcohol molecule.

Semiconductors are non-specific to alcohol, non-linear in response to alcohol vapour concentration and unstable in sensitivity with time: and their effective working life is rarely longer than one year, but this actually depends to a large extent on how often they are used.

Further, since the surface effect by which they operate is dependent on the atmospheric partial pressure of oxygen, semiconductors have been found to vary in sensitivity to alcohol with changes in climate, and even more so at changing altitudes of operation.

Furthermore, caution should be taken when reviewing the claims made by some suppliers and distributors about the type of alcohol sensor employed in their instruments. The term 'fuel cell' has been used to describe sensing devices which are clearly semiconductors, as opposed to true fuel cells operating on elctrocatalytic principles and being dependent on the chemical [as opposed to purely physical] nature of the process.

Infrared Absorption

These analysers operate on the principle that organic substances absorb infrared light at various wavelengths depending on their atomic make up and molecular structure. The quantity of radiation absorbed depends on the concentration of absorbing substance present in the sample, and is thus a measure of it.

In such circumstances, therefore, the difference between the amount of infrared light entering one end of the sample chamber from that received at the other end is measured and taken as being proportional to the concentration of absorbing chemical vapour [breath alcohol] present in that chamber.

One advantage that infrared systems have over and above fuel cell and semiconductor sensors is that they are continuous flow analysers. This means that they are able to track the shape of the alcohol concentration curve during the course of an expiration, which allows the presence of both deep lung breath and mouth alcohol to be detected.

There is an argument in the scientific world as to whether infrared light in the 3 or 9 microns region of the spectrum should be used for breath alcohol testing. Each has its benefits.

Gas Chromatography

This technique involves injecting a small sample of the substance of interest [in this case, breath] into a heated separating column, through which it is then forced by a carrier gas. . The volatile components are separated from each other as they pass through the column, and enter a detector in discrete bands. The time taken for the appearance of each substance may be used to help identify it, whereas the size of the detector response is used as a measurement of its concentration.


The oxidation of alcohol by an acidified solution of potassium dichromate, resulting in a quantitative yellow to green colour change, has been used in various early instruments.

The analytical principle is also still employed in disposable alcohol detector tubes (the bag and tube) used essentially for screening purposes.

Dual Sensing

Some companies are now combining two technologies - e.g. infrared and fuel cell - into one instrument.

In this paper we have discussed four major descriptors. It can be seen from descriptor (1) (Applications) that all instruments can be used for screening but that not all instruments can be used evidentially: some devices have multi applications. Thus in the instrument classifications proposed only (2), (3) and (4) are deemed appropriate to be descriptors.

It is informative to place these criteria in a simple matrix and classify instruments (Table 1).

Table 1
Classification of Instruments

    1 2 3 4 5 6
A Portability & Mobility Portable Mobile Stationery      
B Breath sampling methods Active Active no mouthpiece Passive Dual Sampling    
C Alcohol measurement Fuel cell Semi-conductor i.r g.c colourmetry dual sensing

Typical examples are:-

Drager 7410 A1 B1 C1
Drager 7110 A2 B1 C3 and A2 B1 C6 (1.3)
Lifelock PBA 3000 A2 B4 (1.3) C1
Intoximeter, Alcosensor IV A1 B1 C1
National Patent, Datamaster A3 B1 C3
Alcotech AR1005 A1 B4 (1.2) C1
Lion Alcometer SD-400 A1 B1 C1
Lion Alcometer SD-400P A2 B1 C1
Lion Intoxylizer 6000/5000 A3 B1 C3

Although we are not necessarily suggesting that the above be used as a manufacturers code, we feel that a simple and well understood set of criteria would be of assistance to manufacturers, purchasers and the ultimate end users of breath alcohol instruments.

The authors suggest that this matter is sufficiently important that it should be the subject of review by a special working group of ICADTS. We suggest that the Executive Committee may consider reviewing the subject at their meeting in Adelaide - August 1995.