Monitoring Contamination in the Laboratory

The only direct technique for monitoring contamination in the laboratory is to use various types of blanks (Section 9.5.6b) at strategic locations or steps in the method. Depending on the nature of the analyte and matrix, laboratory contamination may only be detected in extracted matrix blanks, and therefore at least one extracted matrix blank should be included with every method development, validation and sample analysis batch note that the preparation of extracted blanks in duplicate is often considered to be the minimum requirement for many methods. Solvent blanks and method blanks can also be invaluable with respect to detecting contamination in the laboratory, but these should always be used in conjunction with extracted blanks unless specific experiments or historical data show that they are equivalent to extracted blanks in this respect for the method in question. 

In some instances laboratory contamination may occur but not be detected in the blanks that are run concurrently with analytical samples. This situation is often discovered via unusually high concentrations of analyte detected in the standards or quality control samples at the lower end of the curve. This relatively common indicator is often overlooked until the problem is so bad that the incidence of run failures increases significantly. 

When contamination is detected in the laboratory, an investigation and corrective action strategy should be developed to determine the source of the contamination and what procedures will be necessary to control it. No samples should be analyzed until it is shown that the problem has been corrected and that blanks and curves can be generated successfully on a routine basis. Additional procedures or modifications to the analytical method may be required to prevent the reoccurrence of the contamination. If it was determined that the blanks used concurrently did not adequately indicate that contamination existed, an additional investigation should be conducted to determine why, and what additional steps or types of blanks will need to be generated for the method. 

Even if laboratory contamination is not suspected it may be deemed necessary to run one or more control sets with blanks only, prior to running samples. This is most common with extremely valuable samples where there is no archive and the analyst has essentially only one chance to generate reliable results. 

There is no one procedure to determine which types of blanks would be best to use. Nonetheless, a method depending on the analyte, matrix, method and LLOQ or LOD required will have to be developed to ensure that laboratory contamination is detected as early as possible. 

---