Anions in potable waters - Determination of Anions in Natural and Treated Waters - page 485

medium. The distillate is treated with Alizarin Red S (C.I. Mordant Red 3) solution,
buffered at pH 2.5-3.5, and then allowed to react with a known volume of standard
thorium nitrate solution. After 30min, the colour is compared with those of a series of
standards covering the range 0.0-0.1mg L −1 fluoride.
One method is based on the bleaching effect of fluoride on the colour of the zirconium
erichrome organic R complex [21]. This method is applicable over the range 2-25µg
fluoride. Sulphate does not interfere in the determination. Lanthanum-alizarin complexan
has been used to determine fluoride [22]. The main interfering ions are removed from the
sample with a strongly acidic cation exchanger. The reagent (a buffered pH 4.3 mixture
of 3-amino-methylalizarin-NN-diacetic acid and lanthanum nitrate with added acetone) is
added, and the red chelate is allowed to develop in the dark for about 20min and the
extinction determined against a reagent blank at 620nm. If very high sulphate and
chloride concentrations are present in the sample, the results must be corrected by
reference to a calibration graph. Water samples containing aluminium (greater than
0.1mg L −1 ) must be adjusted to pH 9 with potassium hydroxide solution before analysis,
to ensure quantitative ion exchange.
Wierzbicki and Pawlita observed that free chlorine [23] and ozone [24] interfere in the
determination of fluoride by the zirconium-Alizarin Red-S-spectrophotometric method.
Chlorine partially decolorises the lake formed; this is prevented by the addition of sodium
thiosulphate in an equivalent amount to the concentration of free chlorine present. Ozone
interferes at concentrations as low as 0.1mg L −1 and its effect can also be overcome by
the addition of a slight excess of sodium thiosulphate.
Mehra and Lambert [25] used a solid ion association reagent comprising Buffalo Black
and Brilliant Green dyes supported on activated alumina. When this solid is added to the
sample, Brilliant Green reacts with fluoride releasing an equivalent amount of Buffalo
Black for spectrophotometric determination at 618nm. The test solution is passed through
a column of the reagent and any Brilliant Green released is selectively decomposed by
adding solid sodium bisulphite to the percolate before measurement of the extinction. The
calibration graph is rectilinear for up to 10mg L −1 fluoride and the detection limit is 1µg.
Interfering ions, especially sulphate and bicarbonate must first be removed by adsorption
on Amberlite IRA 400 (NO 3 − form) and preferential elution of fluoride with 0.1M
sodium nitrate; this treatment also serves to concentrate fluoride.
Lei et al. [26] determined fluoride in potable water by adding a lanthanum-alizarin
complex to the sample. The ternary fluorolanthanum-alizarin complex formed is
extracted into isoamyl alcohol containing N,N -diethylaniline and is mixed with carbon
disulphide and acetone to form a solution with a refractive index of 1.57 and the complex
is determined with a total-reflection long capillary cell colorimeter. The detection limit
for fluoride was 10ng L −1 and recoveries were in the range 93-104%.
Cherian and Gupta [27] have described a method for determining fluoride which
involves decomposition of the coloured thoriumchromotrope 2R complex to form the
more stable fluoride-thorium complex which is measured spectrophotometrically at
570µm.
Capitan-Vallvey et al. [28] reacted fluoride with zirconium and xylenol orange to
produce a charged complex which was preconcentrated in an ion exchange resin (Duvex
1-X8). Absorbance was measured directly at 580 and 750µm. The detection limit was