Seventh Session

(27 June - 1 July 1994)

FRANCE AND GERMANY

Working Paper

SENSITIVITY AND SPECIFICITY OF ANALYTICAL INSPECTION EQUIPMENT (GC-MS)

1. This paper lists some arguments intended to help clarify aspects which could be of importance in the discussion on analytical inspection equipment. This equipment must enable inspection teams to cope with the fine-tuned verification requirements in a cost-effective manner. A combined gas chromotograph/mass spectrometer (GC-MS) system seems to be the adequate response to a complex technological challenge. In this context, the preliminary comparison tests on GC-MS systems carried out by Switzerland (PC-VI/B/WP.21, dated 11 April 1994) are welcomed.

2. Making use of state-of-the-art analytical equipment does not, however, necessarily mean that everything that is possible is also practical or meaningful. As pointed out in the non-paper presented by Germany dated 26 January 1994, there is no analytical method or apparatus capable of acting as a "non-compliance" detector at any sensitivity level. For instance, evidence on the prohibited intent which could make any chemical a chemical weapon cannot be established by elucidation of the chemical structure of a compound.

3. On the basis of practical experience gained on computerised mobile GC-MS systems in recent years in chemical weapons (CW) defence, in a number of trial inspections and various environmental applications, but also in the light of discussions in expert groups so far, one could draw the following conclusions:

(a) GC-MS, because of its versatility and highly variable specificity, is the appropriate method and tool for both the non-intrusive screening for and the identification of a given set of chemicals, e.g. a set limited to scheduled chemicals. GC-MS systems can be tailored to the specific needs of verification under the Chemical Weapons Convention (CWC). A system to be approved as inspection equipment must demonstrate that it can be reliably adjusted to the specific needs of a particular type of inspection or inspection scenario or even inspection site.

(b) Overly ambitious specifications in the wrong technical context (e.g. sensitivity/detection limits or mass range) might be perceived as enhancing the potential of intrusion in a way which would be inconsistent with the CWC.

(c) Increase in sensitivity can exponentially increase the false-positives rate, thus requiring increased analysis times, i.e. drastically delayed results in many cases. As the experience from Round Robin tests has shown, misinterpretation of analytical data obtained on traces occurs even under optimal laboratory conditions. Independent spectroscopic methods (e.g. infra red (IR)) would not be available for confirmation at the same low sensitivity level.

(d) The CWC provides for analysis in a way which can be understood to include trace analysis, wherever appropriate. However, routine on-site verification (including on-site analysis) will constitute by far the largest portion in the overall verification effort, and the chemicals to be verified at inspection sites will typically be accounted for in tonnes or kilograms rather than nanograms or picograms.

(e) The detection at ultra-trace level of certain widely-spread scheduled chemicals - some of which occur naturally , such as ricin - could create, rather than solve, ambiguities to an extent which would be counter-productive. For example, alkylphosphonic acids on Schedule 2, which are degradation products of corresponding Schedule 1 chemicals, are also formed by degradation of flame retardants or an unscheduled pesticide; harmless ultra-traces of a Schedule 1 chemical could be found in Schedule 2 chemicals; impurities of Schedule 2 chemicals are contained in Schedule 3 chemicals; and impurities of scheduled chemicals could be found in discrete organic chemicals, a fact which often is unknown to the producer (or consumer or processor).

(f) Trace level identification could be required in certain challenge inspections. It is typically required in the investigation of alleged use. However, as recommended earlier by the Expert Group of Analytical Chemists, in most alleged-use investigations samples should be analysed off-site in designated laboratories.

(g) Elucidation of the structure of any compound will not be the task of on-site analysis.

(h) Reliability of the results and the speed (ideally real-time) at which these results are available to the inspection team would be of highest priority, in order to really facilitate the progress of the inspection.

4. France and Germany continue to support an effective and credible verification regime which puts "teeth" into the CWC. Meaningful use of state-of-the-art portable GC-MS will add to that credibility and effectiveness.