Anions in surface, ground, mineral, borehole and pure waters - Determination of Anions in Natural and Treated Waters - page 462

Butler Sandstone 4.5 4.6

Lincolnshire Sandstone 7.9 7.9

Lincolnshire Sandstone 5.8 5.5

River Thames, Chiswick 11.0 10.9

Source: Reproduced with permission from the Royal Society of Chemistry [28]

6.3.5 Sulphate

6.3.5.1 Titration method

Pagenkopf et al. [29] have described a procedure for the determination of sulphate in
mineral waters which uses back titration of excess barium with a standard solution of
sulphate using nitrosulphonazo(III) indicator. The method was developed for the analysis
of water samples for the mining areas of the northern Great Plains of USA which are
often high in sulphate. Calcium and magnesium must be removed but phosphate,
chloride, nitrate and fluoride do not interfere in amounts up to 200mg L −1 . Ten samples
h −1 can be analysed.
R eagents. As follows.
B arium chloride, 10 −3 mol L −1 . Nitrosulphanazo(III) (2,7-bis (4 ′ -nitro-2 ′
sulphophenylazo), 8-dihydroxy naphthalene-3, 6-disulphonic acid, sodium salt), 10 −4 mol
L −1 . Acetone.
A 10.00ml aliquot of the sample is mixed with 10g of Dowex 50-X8 ion exchange
resin. A 5.00ml aliquot of the supernatant liquid (filter or decant, avoiding fines) was
transferred to a beaker, and 5.00ml of 10 −3 mole L −1 barium chloride solution added and
the pH adjusted to pH 1.90. At this point, 1.0ml of the indicator solution and 10ml of
acetone are added and the resulting mixture titrated to a feint pink in the blue coloured
solution with the standard sulphate solution. A blank is obtained by mixing 5.00ml of
distilled water with 5.00ml of barium chloride solution, adjusting the pH, adding the
indicator and acetone, and titrating with standard sulphate.
The use of a discontinuous titration is convenient for end-point detection when more
than one sample is being titrated. The use of a fluorescent lamp and comparison with a
solution that has been titrated to the equivalence point are helpful in assigning the end-
point. The sample should have a sulphate concentration range of 20-90mg L −1 for best

results, so dilution of a mineral water sample may be necessary.
Analysis of sodium sulphate solutions which had a sulphate concentration range of 10-
140mg L −1 by this method exhibited a positive bias of 2.42% with a standard deviation of
4.38%. The positive bias results from a slightly premature assignment of the end-point in
the titrations, and probably reflects the magnitude of the colour differences that the eye
can detect.
Standard addition studies were made to evaluate the method and gave values listed in
Table 6.5. The least squares correlation between sulphate observed and added is 0.997,
with an interception value of 164mg L −1 which is 1% above the expected amount. The