ACETIC ANHYDRIDE: METHOD 3506, Issue 2, dated 15 August 1994 - Page 3 of 4 SAMPLE PREPARATION: 7. Transfer sample solution to a 50-mL volumetric flask. 8. Rinse bubbler twice with 1 mL distilled water. Add rinses to the flask. NOTE: For samples containing > 4 mg acetic anhydride, dilute an aliquot of the bubbler solution to 50 mL at this step. Apply the appropriate dilution factor in calculations. 9. Pipet 5 mL ferric chloride solution into the flask. 10. In an ice bath, mix equal volumes of sodium hydroxide solution, hydroxylamine hydrochloride solution and ferric chloride solution. Add this solution to bring the solution in the 50-mL volumetric flask to the mark. The purple complex is formed rapidly.
CALIBRATION AND QUALITY CONTROL: 11. Calibrate daily with at least six working standards over the range 0.05 to 4 mg acetic anhydride per sample. a. Pipet 10 mL absorbing solution into each of six 50-mL volumetric flasks. b. Transfer 0 (reagent blank), 10, 20, 40, 80 and 100 µL acetic anhydride calibration stock solution into the flasks. c. Add reagents (steps 9 and 10). d. Analyze with samples and blanks (steps 12 through 14). e. Construct calibration graph (absorbance vs. mg acetic anhydride).
MEASUREMENT: 12. Set spectrophotometer to read at 540 nm. 13. Adjust the baseline to zero with distilled water in both cells. 14. Read absorbance of the sample vs. reagent blank in the reference cell.
CALCULATIONS: 15. Determine the mass, mg, of acetic anhydride in the bubbler, W, and in the average media blank, B, from the calibration graph. 16. Calculate the concentration, C (mg/m 3), of acetic anhydride in the air volume sampled, V (L):
EVALUATION OF METHOD: Method S170 was issued on August 1, 1975 [2], and validated using 100-L air samples over the range 9.35 to 37.4 mg/m 3 at 22 °C and 761 mm Hg [1]. Overall precision, SˆrT, was 0.060, with a collection efficiency of 1.00 ± 0.01. The average of concentrations obtained at 20 mg/m 3 for the overall sampling and measurement technique was 2.5% higher than the true concentrations representing a non-significant bias for a limited number of experiments [1]. Sample atmospheres were generated using the apparatus and procedure for epichlorohydrin [5] and a total hydrocarbon analyzer was used to monitor the generated vapor concentrations [1].
NIOSH Manual of Analytical Methods (NMAM), Fourth Edition, 8/15/94
