Anions in potable waters - Determination of Anions in Natural and Treated Waters - page 487

Hg >500 >500 100 1000

PO 4 >500 >500 >500 >500

Source: Reproduced with permission from Marcel Dekker Inc [16]

The method employed in the determination of fluoride was that of direct potentiometry
using a calibration curve constructed from standard solutions. Subsequent aliquots of the
same standard or sample solution were analysed until two EMF readings within 0.5mV
were obtained. A fresh calibration curve was constructed for each batch of buffer and
each temperature of analysis.
After insertion of the electrode into the solution, the system was allowed to achieve
equilibrium for 5min when the EMF readings was taken to the nearest 0.1mV. Further
EMF readings were taken at 1min intervals until consecutive EMF readings did not differ
by more than 0.5mV. This is considered to be the equlibriuni EMF.
Table 7.5 shows interference limits for various ions in the determination of fluoride
using the tri-ammonium citrate (TISAB III) buffer. Only aluminium, magnesium and
borate give cause for concern.
Of all the buffer masking agent combinations examined, triammonium citrate buffer,
employing CDTA and tri-ammonium citrate as complexing agents, was most efficient in
terms of masking ability. At 1mg L −1 fluoride, aluminium and magnesium up to 100mg
L −1 and 1000mg L −1 respectively were tolerated for 95% fluoride recovery. This is
greatly superior to TISAB II which, at the same fluoride concentration and recovery
level, tolerated only 2.7mg L −1 aluminium, 400mg L −1 magnesium, and, additionally,
1000mg L −1 calcium which is not a problem with the tri-ammonium citrate buffer.
The best temperature of analysis (20±0.1°C or 50±0.1°C) for maximum decomplexing
was dependent upon the species involved and the buffer employed. This prevents the
development of any simple method which would exploit the optimum decomplexing
conditions for each individual species. However, while the maximum possible
decomplexing may not be achieved in all cases, samples analysed at 20°C, 24h following
buffer addition, will, for each species and each buffer, at 1mg L −1 fluoride be beneficial
in terms of an increased fluoride recovery over that obtained at 20°C soon after buffer
addition.
Classic ion selection electrode methods for fluoride suffer from the disadvantages of
excessive electrode drift, long response time and inadequate precision.
Ballcizo [33] carried out direct potentiometric determinations of fluoride in potable
water. This was carried out with the aid of tiron (pyrocatechol 3,5-disodium
disulphonate) as a decomplexing agent. This substance releases fluoride from a wide
range of complex fluoride ions (eg aluminium, iron, titanyl, magnesium, silica and boric
acid) and thus permits the determination of total fluoride content, even in the presence of
appreciable amounts of aluminium. As the fluoride electrode is also highly sensitive to
hydroxyl ions the pH value must be kept at or slightly below neutral, but not so low as to
allow the formation of hydrochloric acid. The buffering action of tiron is also sufficient
to enable the use of any additional buffers such as those for ionic strength adjustment, to
be dispensed with.