where a large number of blanks (preferably more than 20) have been measured or
where a smaller number of paired measurements of sample and blank have been made.
Example 2.7
The determination of iron at trace levels can be accomplished by the spectrophotometric measurement
of its complex with the reagent o-phenanthroline. The sensitivity of a particular method is 53 ppm per
unit absorbance and the standard deviation of the blank estimated from 25 measurements is 0.002 of an
absorbance unit. The practical detection limit is therefore 1.64σB or 0.0033 of an absorbance unit which
corresponds to 0.17 ppm, and the theoretical detection limit is 3.28σB or 0.0066 of an absorbance unit
which corresponds to 0.35 ppm. (The value of μB in this case is assumed to be zero.) Hence if a result is
less than 0.17 ppm, the conclusion is that iron is 'not detected'. If the value lies between 0.17 ppm and
0.35 ppm, the iron content should be reported as 'less than 0.35 ppm'.
Quality Control Charts
Chemical analysis finds important applications in the quality control of industrial processes. In an ideal
situation a continuous analysis of the process stream is made and some aspects of this are discussed in
Chapter 12. However, such continuous analysis is by no means always possible, and it is common to
find a process being monitored by the analysis of separate samples taken at regular intervals. The
analytical data thus obtained need to be capable of quick and simple interpretation, so that rapid
warning is available if a process is going out of control and effective corrective action can be taken.
One method of data presentation which is in widespread use is the control chart. A number of types of
chart are used but where chemical data are concerned the most common types used are Shewhart charts
and cusum charts. Only these types are discussed here. The charts can also be used to monitor the
performance of analytical methods in analytical laboratories.
Shewhart Charts
In an explanation of the construction and operation of control charts it is helpful to consider a simple
example such as the mixing of two materials in a process stream. It is important to recognize that there
are two aspects of the composition of the process stream which need to be controlled. Firstly, the
overall composition of the mixture may be assessed by averaging the results on a run of about five
samples and plotting the results on an averages chart. Secondly, short term variations in composition
are reflected by the range of results in a run plotted on a ranges chart. Shewhart charts are thus used in
pairs (Figure 2.5(a) and 2.5(b)). It is usual to mark the charts with