Preconcentration techniques - Determination of Anions in Natural and Treated Waters - page 751

The calibration graph obtained by the extraction method was linear for at least three
orders of magnitude of concentration above the detection limit. The detection limit,
defined as the concentration of phosphorus equivalent to three times the standard
deviation of the background signal (3) was 37µg L −1 phosphorus in the 5ml of extract,
hence, the detection limit for the original samples (500ml) was 0.37µg L −1 phosphorus.
Measurements of 5µg of phosphorus, extracted from 500ml portions of aqueous solution
gave a relative standard deviation of 2.1%.
More recently, Muyazaki et al. [72] have extended the detection limit of the
inductively coupled plasma emission spectrometric technique down to the sub µg L −1
level by measurements of the reduced molybdoanti-monylphosphoric acid, not at the
phosphorus P(I) 214.91nm line, as in the previous method, but at the more sensitive
molybdenum(II) 202.03nm or the antimony(I) 206.83nm lines. This method is simple,
sensitive and precise. Washing of the organic phase is not necessary because of the low
solubility of diisobutyl ketone in water.
Arsenic(V) caused serious interference but germanium, silicon and arsenic(II) did not
interfere in 100-fold amounts. Nitrite seriously interferes in spectrophotometric methods
for molybdenum blue. Molybdenum measurements by the inductively coupled plasma
method did not suffer from interference by up to 5mg of nitrite (2500-fold excess over
phosphorus). In the antimony measurements, nitrite was permissible up to 0.2mg (100-
fold amount). Above that amount, the interference decreased with increasing amount of
antimony in the molybdenum-antimony reagent, although the interference was not
completely suppressed. The addition of sulphamic acid decreased the interference from
nitrite. The difference in the interference mentioned above may be related to the
mechanism of the reduction of molyboantimonyphosphoric acid. No interference was
observed from peroxodisulphate. Pretreatment with peroxodisulphate as used for the
determination of total phosphorus in water, therefore, may be applicable to this method.
Bet-Pera et al. [73] have described an alternative phosphate preconcentration
procedure in which phosphate is converted to 12-molybdophosphoric acid which is then
extracted into isobutyl acetate. After evaporation of the organic solvent the complex is
dissolved in alkaline solution and after acidification molybdenum(VI) is reduced to
molybdenum(III) using a Jones reductor. The resulting molybdenum(III) is reoxidised
with iron(III) to molybdate and the resulting iron(II) is deter-mined

spectrophotometrically at 562nm as the iron(II) ferrozine complex.
Phosphates in non saline water samples were determined by this method in amounts as
low as 4µg L −1 in the final solution with a relative precision of 12% at 2α value.

15.1.8.2 Preconcentration on activated carbon

Hashitani et al. [74] preconcentrated phosphate in water using activated carbon loaded
with zirconium. Activated carbon (10g) was contacted with zirconyl nitrate solution (1g
zirconium at pH 1.6) at 25°C for 3 days. After filtration the material was air dried and
formed into a bed on a membrane filter, porosity 0.54µm. A 0.1-10L sample at pH 1.5
was passed through the bed. Phosphate was adsorbed quantitatively and instantly below
pH 8 and desorbed above pH 13.5. Pyrophosphate, tripoly-phosphate and metaphosphate
behaved as orthophosphate.