Environmental Engineering Reference
In-Depth Information
are used as recommended by the appropriate authority. In addition to the above,
QA also includes control of the following: calibration and standardization of
instruments, preventive and remedial maintenance, proper instrument selection
and use, quality laboratory water, clean laboratory environment, replicate analysis,
spiking of samples, holding facilities for samples, responsible evaluation of data,
and recording and maintaining a quality control (QC) database.
Quality control is the system of practices and procedures that provides the
measure of precision, accuracy, detection limits and completeness of the testing
facility. Precision measures the degree of agreement among replicate analyses of a
sample. It quantifies the repeatability of a given measurement. The precision is
calculated as relative percent difference of duplicates. The precision for three or
more replicates is estimated by calculating the relative standard deviation (RSD) as
100
s
X
RSD
¼
where: s
¼
standard deviation of replicate analysis, and X
¼
mean of replicate
analysis.
The best mechanism to evaluate precision is the examination of relative percent
difference of duplicate samples in the analytical run. This is expressed in the
formula:
¼
½
ð
Þ=
ð
Þ=
RPD
100 X1
X2
f
X1
X2
2
g
where: RPD
¼
relative percent difference, X1
¼
first observation of unknown X,
and X2
¼
second observation of unknown X.
In analytical chemistry, the detection limit (also called the lower limit of
detection or limit of detection) is the minimum concentration of a substance that
can be determined with a given level (usually 99 %) of confidence. That is, the true
concentration of the substance in question is greater than zero. There are several
types of detection limits. For example, some detection limits are set by the manu-
facturer of a specific piece of analytical equipment. The method detection limit
(MDL) is unique in that it is designed for each individual laboratory. A sample
containing a known amount of the compound being measured is analyzed by the
laboratory seven or more times and the standard deviation of those measurements is
determined. The MDL is calculated according to the formula: MDL
¼
Student's
t value
the standard deviation.
Sample unknowns are duplicated based on the assay, commonly at the rate of
1 per every 10-20 unknowns, depending on the standard operating procedure.
Relative percent differences of 10 % are expected at levels of ten times the method
detection limit (MDL) and above. Another mechanism to evaluate precision
involves a comparison of a check sample run daily with each batch of samples.
If the check sample is run several times during the analytical run, then an estimate
of replicability of the run can be obtained. The standard deviation of these results is
an estimate of daily precision. The repeatability of the procedure over time can be
