Laserfiche WebLink
s <br /> E <br /> F <br /> CALIFORNIA WATER LAOS <br /> QUALITY ASSURANCFJRUALITY CONTROL <br /> i MANAGEMENT SUMMARY <br /> (Continued) <br /> Matrix Spikes <br /> To assess aCCumcy. a sample is chosen at random to spike with an analyte of <br /> interest. Matrix spikes must be analyzed for every 20 samples or similar matrix or per <br /> batch,whichever is more frequent. Spiked samples are reviewed in terms of.percent <br /> recovery. <br /> Spike Duplicates <br /> To assess precision,the matrix spike above is prepared and analyzed in duplicate. <br /> y This is done because many samples will not contain the analyte of interest and may <br /> not provide precision data. Spike duplicates are analyzed at the same frequency as <br /> matrix spikes, and results are compared in terms of the relative percent difference <br /> (RPD)between the percent recovered for each of the duplicates. <br /> Acceptance criteria are as follows: <br /> Inorganic. Warning limits for RPD and%Recovery are dented as±2X the standard <br /> deviation of the mean. Out-of-control limits for RPD and%Recovery are defined as <br /> ±3X the standard deviation of the mean. <br /> f 3 Organic. Out-of-control limits for RPD and %Recovery are defined as±2X the <br /> standard deviation of the mean with some exceptions that are ± 3X the standard <br /> deviation. <br /> Note: These limits do not apply to Ieachate extractions. <br /> f Instrument Calibrations <br /> All analytical tnstruments are calibrated using standards of at least three different <br /> concentrations, including a standard at the concentration of the hrr ft of detection, <br /> Samples must be diluted so that the raw data fail within the concentration of the <br /> f r, calibration. Acceptable methods for linearity include cori�iadon coefficient, %RSD, <br /> and least squares curve fitting. <br /> CWL uses the protocol outlined in USEPA Contract Laboratory Program, State- <br /> ment of Work for Organic Analysis Multi-Medla, Multi-Concentration, 10/86 <br /> revision, for tuning and calibration of all GC/MS systems. <br /> Method Spikes <br /> When samples are extracted or digested,a method blank is spiked with an analyte <br /> f of interest. The analysis of this sample may provide information regarding the <br /> preparatior efficiency when matrix interferences are present. <br /> i <br /> t:s 5 <br /> Xx x! <br />