An Overview of Basic Principles of Analytical Methods - Field Sampling: Principles and Practices in Environmental Analysis - page 272

encompasses the linear response range of the method. Often the range and number of
these calibration standards as well as the statistical form used to generate the response
curve will be dictated by published method conditions.
Ideally, the raw sample response will fall within the range of responses used in
determination of the instrument or method calibration range. If the raw response is
outside this range, there are two factors to consider. The first is what the concentration
limit is for the particular component or compound in the analytical method used. If the
analyte in question exceeds the calibration range of the analytical method or instrument,
the answer may be to dilute the sample by an accepted factor and reanalyze.
Diluting and rerunning a sample can introduce contamination from the diluting agent,
and it will introduce a dilution factor that must be added to the calculation when deriving
the true concentration of the sample. Dilution may mask or reduce other important
analytes in the sample to such an extent that they are not accurately determined using the
analytical method. Thus, great care and high-accuracy glassware and measurement must
be used in making dilutions. A general rule related to all analysis is that any additional
manipulation of the sample will introduce error.
Usually there are also other points to consider as part of this. One of these points is the
question of how far above the instrument limit the analysis went. This is important
because the instrument may now be contaminated. Contamination can be determined by
running a method blank to see if it is below detection levels for the compound in
question. If it is, there is no problem. However, if it is above detection limits, the
analytical system requires cleanup or maintenance to bring the level back below the
detection threshold for the compound in question. Finally, the system must be thoroughly
checked using a series of system blanks to ensure conformity to method requirements. In
extreme cases the system might require further cleaning or component replacement
before the instrument system is clean.
Another question to be considered is if the level of the analyte interferes with other
compounds that might be present. If so, the other compounds may not be reliably seen or
reportable. The sample in this case may need either dilution or further separation and
reanalysis with the reporting levels modified to reflect the dilution factor.
In extreme cases, the level of a compound may exceed its solubility in a sample or
solution. When saturation occurs excess material will separate out, as in the case of oil or
gasoline contamination of groundwater. Compounds that make up the gasoline dissolve
in water until saturation occurs, and the remainder of the gasoline then pools or forms a
layer on the top of the water sample. This can cause errors in the characterization of a
sample if part of the nonuniform sample is taken to represent the entire sample. This is an
example often seen in groundwater samples from a field contaminated by a leaking
gasoline storage tank. While the layer of free product on the surface may lead to very
high concentrations for compounds such as benzene or toluene in the analysis of the
water sample, they are often not the true concentrations of the compounds actually
dissolved in the water.
This is an example of a sample with multiple layers in which the concentration will
vary between the layers of the liquid sample. In the case of a soil, the analytes may, in
extreme cases, be seen as deposits within the soil matrix as a whole. Because this is not a
characteristic sample, it can lead to analytical problems unless care is taken in the