CHLORINATED AND ORGANONITROGEN HERBICIDES (HAND WASH): METHOD 9200 - Page 3 of 11
10. Filter an aliquot through a 0.45-µm PTFE filter into a 2-mL GC vial; label. NOTE: It is important to filter the samples because the silicic acid fines can build-up in the chromatographic system and cause deterioration of analytical column performance. CALIBRATION AND QUALITY CONTROL: 11. Calibrate daily with at least six working standards covering the analytical range of the method for individual analytes. Three standards (in duplicate) should cover the range from LOD to LOQ. a. Add known amounts of calibration stock solution to isopropanolin a 10-mL volumetric flask. Add 0.5 mL of the diazomethane derivatizing reagent. Let stand for 1 hour. b. Add 10 mg silicic acid to each standard vial and let stand for another hour. c. Filter through a 0.45-filter into a GC vial. d. Include a calibration blank of unspiked diazomethane derivatizing reagent solution. e. Analyze together with field samples and blanks (steps 13 and 14). f. Prepare calibration graph (peak height or area vs. µg analyte). 12. Analyze three quality control blind spikes and three analyst spikes to ensure that the calibration graphs are in control. MEASUREMENT: 13. Set gas chromatograph according to manufacturer’s recommendations and to conditions listed in Table 3. Inject 2-µL aliquot manually using solvent flush technique or with autosampler. See Table 4 for retention times of selected analytes. NOTE: If peak height is greater than the range of the working standards, dilute with isopropanol and reanalyze. Apply the appropriate dilution factor in calculations. 14. Measure peak height of analyte. CALCULATIONS: 15. Determine the concentration, C (µg/mL), of the herbicide found in the sample from the calibration graph. 16. Calculate the mass of herbicide, M (µg), found in the volume, V (mL), of hand wash solution.
M
C
V , µg
CONFIRMATION: Whenever the identity of an analyte is uncertain, confirmation may be achieved by analysis on a column of different polarity. If primary analysis was performed using a nonpolar or weakly polar column (e.g., DB-1 or DB-5) confirmation should be accomplished by reanalysis on a polar columne.g., ( DB-17 or DB-1701). See Table 4 for approximate retention times for each column type. For positive identification of high-level analytes (1 to 10 µg/mL or greater), GC/MS may be used. Table 5 provides notes on the analytical characteristics of the chlorinated and organonitrogen herbicides. EVALUATION OF METHOD: This method was evaluated over the ranges specified in Table 3. These ranges represent 3 x LOQ to 30 x LOQ for each of the compounds. The analytical conditions used in the evaluation are listed in Table 2 for the DB-5ms column. Humidity did not seem to have an effect on the analyte recoveries from the samplers. Measurement precision ( r), LOD, LOQ, and storage stability data are also presented in Table 3. The LOD and LOQ were determined bytaking a series of liquid standards, prepared in triplicate, analyzed, and the data fitted to a quadratic curve. The Limit of Detection (LOD) and Limit of Quantitation (LOQ) were estimated with Burkart’s Method [3]. A long-term storage study was done at the 10 x LOQ level. Handwash solutions were stored at 4 ± 2 oC for 30, 60, or 120 days in glass containers sealed with caps having PTFE-faced liners. The stored samples were analyzed on the DB-17 column using the conditions listed in Table 2. The results are summarized in Table 3. The 30-Day storage samples had acceptable recoveries of nearly 100%, with the extremes being simazine and alachlor having recoveries of 127.7% and 87.1%, respectively. Day NIOSH Manual of Analytical Methods (NMAM), Fourth Edition
