Under these conditions, which are generally identical to those created in the
ordinary analytical procedure, we obtain a molar absorptivity coefficient for HPA
identical with that for heteroatom or analyte when the latter is determined with HPA.
We tried to correlate the molar coefficients for the dye, the dye in the aggregated form
and IA (which is identical with a molar absorptivity coefficient obtained for analyte)
obtained in analytical conditions.
It was established that the spectral changes of the same nature occurred by the
interaction of cationic dyes with tungsten HPAs of the Keggin structure having
different charge (PW
12
O
40
3-
, SiW
12
O
40
4-
, GaW
12
O
40
5-
). In all the studied cases by
mixing of the dye and HPA solutions, two new bands appear in the spectrum of IA,
which are displaced bato- and hypsochromically at 30-60 nm relative to the original
dye band (Fig. 3). In accordance with this, two maxima and one minimum are present
in the difference spectrum. Intensity of the red-shifted band of IA is, as a rule, less than
that for the blue-shifted one. However, absorption in this area can be preferentially
used to develop analytical methods for determining heteroatoms of HPA or other
elements or species, which can be converted in sufficiently large and hydrophobic
anions. Reason for this is that the hypsochromically shifted band is strongly overlapped
with dye band (the dye is commonly used in the analysis in large excess in relation to
IA) while for the batochromically shifted band overlapping with dye band is nearly
absent. An additional advantage is that the absorption of the dye in this region is
insignificant.
It is generally accepted that the wavelength, at which the difference between the
control and blank experiment is maximum, is taken as an analytical wavelength. The
biggest deviation in the difference spectrum is observed at the wavelength
corresponding to the maximum in the dye spectrum. This is not surprizing because this
band is absent in the spectrum of the IA (or spectrum of aggregated dye). Nevertheless,
at this wavelength the absorbance of the blank solution is very high that limits the
possibility of increasing the path length of cell and increases the influence of random
errors due to the loss of significance by subtracting two nearly equal numbers. Thus,
the gain in the molar absorptivity is offset by other negative factors.
It was found that the sensitivity of the analytical procedures based on the
formation of aggregates of dyes depends on two main factors. The first one is the value
of the molar absorptivity of the dye. As a rule, the higher the molar absorptivity of the
dye, the greater, that is naturally, the molar absorptivity of the ionic associate,
measured at the analytical wavelength. An inspection of the dependence of molar
absorptivity of IA on that for the dye proves that this is really the case. However, the
correlation coefficient is low (R
2
= 0.887). Therefore, we must take into account
another factor that determines the sensitivity of the method in question.
- 1424 -