Page:NIOSH Manual of Analytical Methods - 9109.pdf/32

METHAMPHETAMINE . . . on Wipes by SPE: METHOD 9109, Issue 1, dated 17 October 2011 - Page 32 of 33

H. MAKING DILUTIONS: If the samples exceed the upper calibration range for the analysis, one of the following procedures may be used to estimate the high level concentrations. 1. Dilution procedure A (dilution of the derivatization mixture within a GC vial): Transfer an aliquot of the derivatization sample mixture from the GC vial to a clean low-volume GC vial and add acetonitrile, MSTFA, and MBHFBA. For example, for a 10:1 dilution transfer 20 μL of sample to a clean vial and add 120 μL of acetonitrile and 30 μL each of MSTFA and MBHFBA, for a total volume of 200 μL. For a 4:1 dilution, transfer 50 μL of sample to a clean vial and add 100 μL of acetonitrile and 25 μL each of MSTFA and MBHFBA, for a total volume of 200 μL. Cap the GC vial, mix by inversion a few times, and analyze diluted sample. Do not include the dilution factor in step 19 since the internal standard will be diluted along with the target analyte. NOTE: For dilutions greater than 10, the internal standard may become too diluted to quantify. In such a case, use the following procedure B. 2. Dilution procedure B (dilution of the original sample desorbate): In this procedure, an aliquot of the original sample desorbate is diluted with a simulated blank solution and then transferred to a SPE column in step 8d. For example, for a 10:1 dilution, dilute 0.5 mL of sample desorbate solution from step 7f in a clean test tube containing 4.5 mL of a simulated blank solution, mix, and then transfer the entire contents to a pre-conditioned SPE column. For a 50:1 dilution, dilute 0.1 mL of sample desorbate solution from step 7f in a clean test tube containing 4.9 mL of a simulated blank solution, mix, and then transfer the entire contents to a pre-conditioned SPE column. Proceed thereafter to step 8d as normal. The simulated sample blank should be prepared identically to the sample needing dilution, using the same volumes of internal standard spiking solution and desorption solution that were used with the sample in the original desorption. For example, if the original sample was desorbed with 40 mL desorption solution with 80 μL of added internal standard spiking solution, then prepare the simulated blank in the same way. The volume of wetting alcohol is estimated (e.g., about 3 mL per 3”x3” 12-ply cotton gauze wipe). Include a dilution factor (V3/V4) in the calculations in step 19 (e.g., V3/V4 = 5 mL divided by the volume in mL of original desorbate diluted to 5 mL with solution from the simulated blank). The dilution factor in the above examples are 5 mL/0.5 mL or 10 for a 10:1 dilution and 5 mL/0.1 mL or 50 for a 50:1 dilution. Correct for differences in internal standard spiking solution volumes in step 19 (if applicable) using for V1 the volume of internal standard spiking solution which was added to the original undiluted sample. Caution: This dilution procedure gives quantitative results only if the residual volume of methanol (or isopropanol) used for wetting the sample wipes was exactly the same as the volume used in preparing the calibration standards (normally about 3 mL, see Table 7). Deviations of a few milliliters in residual wetting alcohol will not affect the results for undiluted samples but will amount to an error of a few percent in the final results of samples that are diluted. The potential error due to differences in residual wetting solvent can be estimated for specific volumes of desorption solution and wetting alcohol. Assume the sample wipes and calibration standards are both desorbed in 30 mL of desorption solution and 3 mL of alcohol is added to the calibration standards. The potential error in volume (and final results) in the samples is approximately ±3.03% (inversely proportional) per mL difference in the residual alcohol in the samples (i.e. ±1 mL difference in 33 mL). For 40 mL of desorption solution and 4 mL of alcohol added to the calibration standards, the error is ±2.27% for every mL difference (i.e. ±1 mL difference in 44 mL). However, since the volume of residual wetting alcohol is not known and cannot be determined once the sample wipe has been desorbed, the actual error cannot be determined.

However, the maximum possible error can be calculated. Since the maximum amount of alcohol that a 3”x3” 12-ply (or 4”x4” 8-ply) cotton gauze can hold is about 6 mL when saturated (dripping wet), there can only be a deviation of plus or minus 3 mL from the 3 mL alcohol added to the calibration standards. Therefore, the maximum error in a result

Method rev. 1.1.1

NIOSH Manual of Analytical Methods (NMAM), Fifth Edition