Page 249 - Radiochemistry and nuclear chemistry

Detection and Measurement Techniques 233

8.10. Sample preparation

From the discussion of the factors that enter into the counting efficiency it is obvious that
the preparation of the counting sample must be done with care and must be reproducible
if several samples are to be compared. Counting of a- and/~-emitters in solution is best
achieved by means of liquid scintillation counting. Because in this technique the emitters
are included in the detection system itself the efficiency is very high and reproducible.
The advantage of using solid samples for counting is that the samples can be made very
robust and small, allowing the use of either very simple counting systems (e.g. GM-
counters), or the use of commercial very efficient high-capacity (> 1000 samples/hour)
automatic counting systems. Alpha-emitters can only be counted efficiently if the sample
is very thin so that the self-absorbtion is eliminated. For a-spectrometry the surface density
should be < 0.1 mg cm- 2. Preferably, c~' s should be counted by surface barrier detectors,
windowless proportional counters or internal ion chambers. Counting of solid samples of
B-emitters may or may not be a problem depending on the energy of the /3-emission.
Again, care must be taken with uniform thickness of sample, backscattering, etc. The use
of energy sensitive detectors makes possible a reliable measurement of one particular
radioactive nuclide m the presence of other radiation of secondary importance.
Solid samples can be prepared by a variety of techniques such as precipitation,
evaporation, and electrolysis (Fig. 8.22). When the precipitation technique is used the
radioactive material must always be precipitated for comparative counting with the same
amount of carrier and all samples must have the same surface density. The precipitate is
filtered on a filter plate or filter paper of known reproducible geometry A. If filtration is
not feasible the precipitate may be centrifuged in special vials C, or the precipitate, after
centrifugation and decanting, may be slurried with ether or alcohol, and the slurry
transferred by pipette to a counting disc of fixed geometry; when the organic liquid
evaporates, it hopefully leaves a uniform deposit of the precipitate slurry D, F. Problems
are plentiful in obtaining a deposit of uniform thickness by evaporation of a solution.
However, an arrangement such as that shown in Figure 8.22 E has been found to be
suitable; slow and even evaporation of 0.1 - 1.0 ml samples result in an even deposit if the
amount of solid material is small. More even deposits can be obtained by electro-deposition
of samples from solution B. This method can be used also for nonaqueous solutions
provided that the organic solvents contain traces of water and a potential of several hundred
volts per centimeter is used. To speed-up sample preparation, different types of semi-
automatic sample preparation systems are commercially available. They are particularly
useful in bio-medical tracer research, where often large numbers of samples are produced.

8. II. Statistics of counting and associated error

Even if the experimental design and execution are perfect so that the determinant error
is eliminated in experiments involving radioactivity there is always a random error due to
the statistical nature of radioactive decay. Each radioactive atom has a certain probability
of decay within any one time interval. Consequently, since this probability allows unlikely
processes to occur occasionally and likely processes not to occur in any particular time
interval, the number of decays may be more or less than the number in another similar