Page 81 - An Introduction to Analytical Atomic Spectrometry - L. Ebdon
P. 81
Page 63
3.6 Methods of Overcoming Interferences
3.6.1 Control of Furnace Temperature
The faster the rate of heating, the higher is the density of the atoms which will be formed in the
transient atomic cloud. This leads to improved analytical sensitivity, provided that the electronics can
follow the rapid signals. To obtain good reproducibility, it is necessary to control the temperature
actually reached by the tube. While it is relatively simple to stabilize the applied voltage, variations in
the resistance of individual tubes and the degradation of the tube as it is used mean that the temperature
achieved must be monitored. Thus the voltage applied should be controlled via a feed-back circuit,
linked to some method of sensing the tube temperature. This temperature may be measured using a
thermocouple, which unfortunately may suffer from temperature lag, or more commonly by using a
light sensor, e.g. an infrared sensor which views the tube-wall radiation via a fibre optic. In the latter
case, it is clearly important that the end of the fibre optic remains clean. In a well designed modern
furnace, therefore, the control settings 'dry', 'ash', 'atomize' and 'clean' should refer to reproducible
temperatures (probably inaccurately known but precise) rather than to different applied powers.
3.6.2 The Effect of the Orientation of Tube Heating.
Initially it was thought that maximum sensitivity would be obtained from larger tubes, as these would
retain the atomic vapour for a longer period. Long tubes, however, require more power to achieve a
given temperature, and smaller tubes can be heated more rapidly. Therefore, in the interests of rapid
heating and simplicity of power supply, small tubes, typically of 20-30 mm length, are preferred. The
tube should not be so short that the atoms escape from the tube and cool too rapidly; modern tubes give
residence times of about 0.5 s.
Most atomizers are heated longitudinally i.e. they are clamped at either end by electrodes and the
electricity is passed through the length of the tube. This leads to a temperature gradient along the tube,
with the central portion being several hundred degrees hotter than the ends. This can lead to
condensation of the analyte or recombination with other species at the cooler ends of the tube. Some
modern instruments heat the tube transversely, i.e. from the sides. With this method of heating, a
temperature gradient along the tube does not exist. Therefore, the efficiency