Laserfiche WebLink
Section No: 8 <br />Page: 2 of 5 <br />Revision No: 2.0 <br />Date: January 15, 1994 <br />Calibration Procedures and Frequency <br />8.3.3 Selection of Quantitation Technique (Organics) <br />For organic analysis, a decision must be made during the validation process (and detailed <br />in the SOP) as to whether an internal or external quantitation technique will be routinely <br />employed. <br />The internal standard method of quantitation cannot be employed unless all of the <br />following conditions are met: <br />(1) The internal standard must be added post -extraction. <br />(2) The internal standard must be added quantitatively. <br />(3) Any analyte that is a target analyte in the method of interest may not be selected for <br />use as an internal standard. <br />(4) The concentration of the internal standard(s) must not exceed the calibration range <br />of the method target analyte. In cases where the target analytes are associated with <br />more than one calibration range (i.e. analytes 1114" are calibrated from 1 to 10 <br />ppb, while analyte °'S°' is calibrated from 10 to 100 ppb, all target analytes should be <br />prepared at a level between the highest and lowest calibration standard (e.g. <br />approximately 50 ppb in the example given). ' <br />The use of internal standard quantitation is of greatest benefit in those methods subject <br />to injection variability, and thus, variability in the absolute mass injected onto the <br />column(s) employed. The drawback to this technique, for GC methods, is that any <br />compound which exhibits a similar retention time as the compound used for the internal <br />standard will be identified as the internal standard, leading to erronous quantitation. <br />For this reason, the internal standard technique is most useful for GC/MS where <br />deuterated analytes not naturally occuring can be detected and quantified. <br />8.3.4 Selection of Calibration Method <br />As part of the validation process, the specific calibration range and calibration method <br />must be determined and documented in the SOP. Once determined in this manner, the <br />same protocols must be followed each time the method is employed. This will ensure that <br />data reduction is not performed differently on separate data sets by different analysts. <br />The calibration acceptance criteria are listed in section 8.3.8. A least squares (linear) <br />regression is initially tried as a calibration method. For organics, if a linear regression is <br />used, a single average response factor may be used. For inorganics, if the acceptance <br />criteria cannot be met using a linear regression, then a second order polynomial can be <br />used to fit the data, with the same acceptance criteria being applied. In the event that <br />neither a simple linear regression nor a second order polynomial fit result in an equation <br />which meets the calibration acceptance criteria, then the calibration range must be <br />reduced. <br />8.3.5 Minimum Number of Calibration Levels <br />Most calibrations include a minimum of three or five initial calibration standards plus a <br />blank. Specific SOPs may have other requirements. is <br />