Sample preservation prior to analysis - Determination of Anions in Natural and Treated Waters - page 779

and more iodine added if necessary.
Place the bottles in a well ventilated fume cupboard and remove the screw caps. Expel
iodine vapour by squeezing the bottle gently. Rinse the bottle with a little ethanol and
drain.
Wash the bottle by repeated filling with water at daily intervals until the iodine content
of the water is reduced to the extent that the absorbance of the water at 680nm is not
greater than that of water stored in untreated bottles.
When not in use store the bottles in the dark filled with water.
Where the major requirement is to measure 'dissolved reactive phosphate' in samples
containing less than 25µg L −1 P either:
(a) filter the sample 'on site' through a glass fibre paper, pre-washed with about 500ml of
water [15], into a plastic bottle and analyse within 3h of sampling; or if this is not
possible—
(b) filter the sample through a pre-washed glass fibre paper [15] within 3h of sampling
into an iodised bottle and store in the dark at, or close to, 4°C
NOTE. If the filtration pressure is too high or if the sample is allowed to freeze,
changes of dissolved phosphate concentration may result from biological cell rupture
and the release of phosphate into the solution.
'On site' filtration may be mandatory depending upon which forms of phosphate are to be
measured. In this respect it should be noted that the insoluble phosphate content of an
unfiltered sample can increase due to the continuation of zooplankton activity to give
deposits containing phosphate in the sample.
In addition to filtration discussed above it may be necessary to consider whether the
special requirements of freshwaters with an oligotrophic nature indicate the need for
digestion of low levels of 'total phosphorus'.
Rossin and Lester [16] evaluated some of the methods available for the preservation of
condensed phosphates in samples of domestic waste waters, ie the prevention of the
hydrolysis of condensed phosphate to orthophosphate. They showed that mercuric
chloride at 40mg L −1 at normal pH is the most effective preservative examined for
samples stored at 20°C for 6 days (Table 18.3). Samples should be filtered as soon as
possible after they have been taken. Mercuric chloride at high pH was not a satisfactory
preservative. Addition of sodium hydroxide to raise the pH increases phosphatase activity
but lowers the rate of spontaneous hydrolysis and, on balance, is not a satisfactory
treatment. Formaldehyde at elevated pH was less efficient than mercuric chloride at
normal pH.
The samples treated with mercuric chloride at normal pH presented the smallest
difference from the mean with a range between −0.07 and +0.06. When formaldehyde
was used at high pH, the range was between −0.16 and 0.00. The samples containing
mercuric chloride at pH 10 showed a range between −0.15 and +0.21. A variation
between −0.21 and +0.35 was found when pH was increased in the filtered samples. The
use of formal-dehyde at normal pH resulted in a variation between −0.25 and +0.34.