ARCHIVED REPORTS_1988_2 - Laserfiche WebLink

Home Browse Search

SPIKED SAMPLES Spiked samples are prepared from commercially available standard mixtures in methanol. These are prepared by injecting 500 ng of each component into an electropolished canister containing zero air at ambient pressure. The container is then pressurized with additional zero air. The pressure is recorded and used, together with the volume of the canister, to calculate the concentration of the spike. Spiked samples prepared in this way are stable for a month or more and can therefore be used to develop acceptance criteria. We beleive that field spiking, particularly if done at ambient pressure, is itself an unacceptable source of error. Representative spiking data are appended. Accuracy is defined as the difference between the measured concentration and the actual concentration divided by the actual concentration and expressed as a percentage. In this case, the measurement is described as the percent relative error of the measurement. Smaller values, of course, represent better accuracy. DUPLICATE SAMPLES Duplicate integrated samples will be taken during the course of the proposed program. In order to ensure a greater abundance of non -zero values, it is recommended that all of the duplicates be downwind samples. Colocated samplers will be employed. Precision is calculated from duplicate sample data as the difference between the two results divided by the mean of the two results, expressed as a percentage. As in the case of the accuracy measurements, smaller values indicate greater precision. If more than two replicates are available, the standard deviation provides a statistically more valid means of expressing precision. BLANKS A zero air blank accompanies every set of samples. This blank is run first in order to verify that the valves and cryogenic systems are free from contamination. Unsatisfactory blanks are a cause for shutdown and system cleaning. Coinisters pressurized with zero air and sent into the field as field blanks ore returned to the laboratory unopened and analyzed to provide a measure of residual contamination which might escape removal during cleaning and subsequently desorb from the interior of the container during shipment and storage.