3.2. Method development for slurry sampling
Preliminary experiments have shown that injection of 20L of
the wheat flour slurry (plus 10L of modifier solution) did not
provide sufficient sensitivity for the determination of cadmium in
the wheat flour samples. Significantly greater volumes of slurry
could not be injected without the risk of an overflow because of
the limited capacity of the platform. One possible solution of this
problem would have been to prepare more concentrated slurry;
however, difficulties have been encountered with the stability of
the slurry in this case so that this approach was not further considered.
A pre-concentration step using multiple injections was
therefore investigated in order to reach an appropriate level of
sensitivity. Twenty microliters of the slurry were injected into the
graphite tube, followed by a thorough drying stage (steps 1–3 in
Table 1) before the next aliquot was injected. After the last slurry
injection, 10L of the chemical modifier solution (30g Pd+12g
Mg+ 0.05% TritonX-100)was injected and the full temperature program
shown in Table 1 was executed. Fig. 5 shows the absorption
signals obtained for awheat flour sample after the injection of one,
three and five aliquots of 20L of slurry into the graphite tube.
The increase in integrated absorbance was directly proportional to
the number of aliquots injected into the graphite tube, i.e., to the
sample mass.
The pyrolysis curves established with an aqueous standard solution
and a wheat flour slurry, which are shown in Fig. 6, are
quite similar to those in Fig. 3 obtained for direct SS. The minor differences in the shape of the curves and in the maximum loss-free
pyrolysis temperatures (850 ◦C for the aqueous standard solution
compared to 900 ◦C in Fig. 3) are most likely due to the different
tubes and platforms used for SS and SlS, respectively. The SS platform
has a different mass and is inserted loosely into the SS tube,
whereas the PIN platform is permanently fixed in the tube. This
might result in a different heat transfer from the tube to the platform
and slightly different apparent pyrolysis temperatures for the
two platforms. The optimum pyrolysis temperature, however, was
the same as for direct SS analysis, i.e., 800 ◦C, and the optimum
atomization temperature was 1600 ◦C.
The atomization signals obtained under these conditions are
shown in Fig. 7. Like in the case of direct SS, the signals for the
aqueous standard and five injections of 20L aliquots of the wheat
flour slurry are very similar in appearance time and peak shape, and
there is essentially no background absorption visible. The significantly
earlier appearance time and sharper peak shape compared
to the signals obtained with SS (Fig. 4) are again due to the smaller
mass of the PIN compared to the SS platform and the different heat
transfer mechanisms. An additional factor that influences the peak
shape is the absence of a dosing hole in case of the SS tube, which
results in a longer residence time of the atoms in the absorption
volume.