Laserfiche WebLink
Page 4 of 24 <br />SOP NUMBER 28.0 Analytical Data Validation <br />4.1.5 Matrix-Dependent Quality Control <br />Matrix dependent quality control (QC) samples are used to evaluate how the sample matrix <br />affects the accuracy and precision of the analytical results. <br />In order to evaluate how the site-specific sample matrix affects the accuracy of the analysis; the <br />laboratory will spike one or two additional aliquots of a field sample with known amounts of <br />target analytes and prepare the spiked samples in a fashion identical to that of the field samples. <br />The amount of each spiked analyte recovered can be used to infer the accuracy of the analysis on <br />the site-specific sample matrix. <br />To assess the precision of the analysis on the site-specific sample matrix, a laboratory duplicate <br />or spike duplicate sample is prepared. A laboratory duplicate sample is a laboratory split of a <br />homogenized environmental sample that is prepared and analyzed in a manner identical to that of <br />the original sample. A matrix spike duplicate is similar with the exception that both aliquots are <br />spiked with known amounts of target analytes. The closeness of the agreement between the two <br />results can be used to infer the precision of the analysis on the site-specific sample matrix. <br />For inorganic methods, one aliquot is typically spiked and for organic methods, two aliquots are <br />typically spiked. For inorganic methods, a duplicate sample is typically used to assess precision <br />whereas for organic methods, a spiked duplicate is typically used. These conventions were <br />developed based on the probability of finding the target analytes in the sample matrix. However, <br />some laboratories choose to do matrix spike and matrix spike duplicates for some of their <br />inorganic analyses. <br />The subsections below describe how the results for matrix QC samples will be evaluated. <br />4.1.5.1 Matrix Spike (MS) Analysis <br />The matrix spike results, expressed as percent recovery of the spiked analytes, are used to assess <br />effects of the general sample matrix on the accuracy of the analysis. <br />The matrix spike recoveries are compared to the appropriate acceptance ranges per Section 4.1 <br />for instances in which the native sample concentration was less than four times the spike level. <br />When sample concentrations of an analyte are >_ four times the spiking concentration, the results <br />are considered to be inappropriate for assessing accuracy. The reviewer should also be aware <br />that a matrix spike recovery may be outside acceptance limits when the parent sample was <br />quantified by method of standard additions but the matrix spike was not. In such a case, the <br />matrix spike recovery is not an appropriate measure of accuracy. Data associated with matrix <br />spike recoveries that are outside the acceptance range will be qualified as follows using guidance <br />from Functional Guidelines. <br />• If the recovery of an inorganic matrix spike analyte exceeds the upper limit of the acceptance <br />range, suggesting a potential high bias in sample results, positive results for that target <br />analyte in all associated samples are qualified as estimated ("J"); whereas, nondetect results <br />for that analyte are considered to be acceptable for use without qualification. <br />• If the recovery of an inorganic matrix spike analyte is below the lower limit of the <br />acceptance range, but >_30% (>_l0% for organics), suggesting a potential low bias in sample <br />results, both positive and nondetect results for that analyte in all associated samples for <br />Powertech R Squared Inc <br />Rev 28-1 JAB Attachment A <br />4/23/2007