9.8 Control Samples
163
– Pure substance (material with a certified content of the substance)
– Certified reference material (CRM)
Both types of control samples are very useful for the evaluation of random variation and to obtain estimation of any systematic effect, as well as for monitoring
trends over time. The control samples prepared from pure chemical substances are
expected to be within the concentration range close to that of routine test samples,
but are not always able to mimic the composition of the matrix. Although CRMs,
purchased from the qualified producer, seem to be the best chose as a control samples, they are not always available for all possible analytical situations. It is important
to stress that CRMs are always of better homogeneity; thus the standard deviation
of measurements are always smaller compared with that obtained for routine test
samples. Thus, the natural samples of sufficient stability could be a good choice for
the evaluation of the typical spread of results for repetitive measurements.
9.8.2 Routine Laboratory Samples
Routinely analyzed laboratory test samples, whenever they exhibit sufficient stability
over time, could be very useful for the estimation of the random variation of the
results. They are of exactly the same matrix and exactly the same homogeneity;
thus, to the best extent, they mimic the routine analytical situation. When they are
used, the best estimation of repeatability and reproducibility can be obtained; thus,
they could be used for the day-to-day monitoring of the correctness of the analytical
procedure. However, the use of routine laboratory test samples, as obtained, do not
provide sufficient information about the accuracy of the measurement results. For
this purpose, the spike samples can be used instead.
When the known amount of standard is added to the sample, the increase in
the analytical signal can be used for the evaluation of the matrix effects. This is
done mostly by monitoring the recovery for the given type of matrix. The most
valid information of the influence of matrix is possible when the spike sample is
subjected to the same analytical processing as the sample. The analyst should be
aware, however, that the spiked standard cannot always undergo the same binding as
the originally occurring substance in the natural sample.
9.8.3 Blanks
Blanks are those samples that do not contain the substance of interest. Blank samples
could mimic the standard solution of pure substance (e.g., for standard solutions of
cadmium in 0.5 mol/L HNO 3 , a blank will be a solution of 0.5 mol/L HNO 3 ), or
163
– Pure substance (material with a certified content of the substance)
– Certified reference material (CRM)
Both types of control samples are very useful for the evaluation of random variation and to obtain estimation of any systematic effect, as well as for monitoring
trends over time. The control samples prepared from pure chemical substances are
expected to be within the concentration range close to that of routine test samples,
but are not always able to mimic the composition of the matrix. Although CRMs,
purchased from the qualified producer, seem to be the best chose as a control samples, they are not always available for all possible analytical situations. It is important
to stress that CRMs are always of better homogeneity; thus the standard deviation
of measurements are always smaller compared with that obtained for routine test
samples. Thus, the natural samples of sufficient stability could be a good choice for
the evaluation of the typical spread of results for repetitive measurements.
9.8.2 Routine Laboratory Samples
Routinely analyzed laboratory test samples, whenever they exhibit sufficient stability
over time, could be very useful for the estimation of the random variation of the
results. They are of exactly the same matrix and exactly the same homogeneity;
thus, to the best extent, they mimic the routine analytical situation. When they are
used, the best estimation of repeatability and reproducibility can be obtained; thus,
they could be used for the day-to-day monitoring of the correctness of the analytical
procedure. However, the use of routine laboratory test samples, as obtained, do not
provide sufficient information about the accuracy of the measurement results. For
this purpose, the spike samples can be used instead.
When the known amount of standard is added to the sample, the increase in
the analytical signal can be used for the evaluation of the matrix effects. This is
done mostly by monitoring the recovery for the given type of matrix. The most
valid information of the influence of matrix is possible when the spike sample is
subjected to the same analytical processing as the sample. The analyst should be
aware, however, that the spiked standard cannot always undergo the same binding as
the originally occurring substance in the natural sample.
9.8.3 Blanks
Blanks are those samples that do not contain the substance of interest. Blank samples
could mimic the standard solution of pure substance (e.g., for standard solutions of
cadmium in 0.5 mol/L HNO 3 , a blank will be a solution of 0.5 mol/L HNO 3 ), or