Page:NIOSH Manual of Analytical Methods - 7504.pdf/5

 VANADIUM OXIDES: METHOD 7504, Issue 2, dated 15 August 1994 - Page 5 of 10 14.

Scan each field blank over the same 2 θ range used for the analyte and silver peaks. The analyte peak should be absent. The normalized intensity of the silver peak of the field blanks should match that of the media blanks. NOTE: These analyses serve only to verify that contamination of the filters has not occurred.

CALCULATIONS: 15.

Calculate absorption correction factors (Tables 2, 3, and 4) [5]:

where: R = (sin θAg)/(sin θx) and T = / = transmittance of sample. 16.

Calculate concentration, C, of analyte in air volume sampled, V (L):

where:

Îx = normalized intensity of sample peak Îb = normalized intensity of blank m = slope of calibration graph, counts/µg

EVALUATION OF METHOD: VANADIUM PENTOXIDE: The V 2O 5 method was evaluated [1] with both spiked and generated samples of V 2O 5. With the spiked samples, using the 31.05° line and loadings of 433 to 699 µg/filter, the overall SˆrT was 8.3% with an average bias of -7.3%. Five sets of generated samples, ranging from 184 to 2400 µg V 2O 5 per filter, gave an average bias of 10.12% with a pooled Sr of 6.9%. The generated samples used an aerosol which had been sized with a cyclone with characteristics similar to the 10-mm nylon cylone; however, individual 10-mm nylon cyclones on the cassettes were not used. VANADIUM TRIOXIDE: The method was evaluated [1] with both spiked and generated samples of V2O 3. With the spiked samples, and the 24.39° line and loadings of 107 to 321 µg per filter, the overall Sr was 8.2% with an average bias of 12.2%. Five sets of generated samples, from atmospheres of 0.24 to 2.57 mg V 2O 3/m 3, gave an average bias of -10.1% with a pooled Sr of 13.4% (24.39° line). As above, the samples were generated by using an aerosol, which had been sized by means of a cyclone with characteristics similar to the 10-mm nylon cylone; however, individual 10-mm nylon cyclones on the cassettes were not used. AMMONIUM METAVANADATE: The method was evaluated [1] with spiked NH 4VO 3 samples over the range of 109 to 277 µg NH 4VO 3 per filter, and with samples generated as described above, with atmospheres of 0.037 to 0.435 mg/m 3 of NH 4VO 3. The 18.10 °2 θ line provided an overall SˆrT of 26.3%, with a bias of -10.2% over the generation range. The stability of the compound is critical and analysis should occur within two weeks of sampling, unless refrigeration storage is employed. Sampling is quantitative using the two-filter technique described.

NIOSH Manual of Analytical Methods (NMAM), Fourth Edition, 8/15/94