Sample preservation prior to analysis - Determination of Anions in Natural and Treated Waters - page 774

oxidised nitrogen or 1mg L− 1 of nitrite.
In certain circumstances however, nitrite may undergo either reduction or oxidation
with some rapidity in certain types of sample. It may also be formed either by reduction
of nitrate or oxidation of ammonia. Stringent precautions are necessary when nitrite is to
be determined, when nitrate is present in significant concentrations or when nitrate alone
is to be determined in the presence of a significant concentration of nitrite.
Samples should be collected in glass bottles and must be analysed as soon as possible
after collection. Tests should be carried out on each par-ticular sample type to determine
the maximum acceptable delay between sampling and analysis. No reliable preservation
method is known.
When the nitrite concentration is insignificant, samples may be collected in glass or
hard polyethylene bottles and should be analysed as soon as possible. When delays in
analysis are inevitable, storage at 4°C may improve stability. Allow the samples to warm
to room temperature before analysis. Preservation may be affected by addition of
sufficient preservative solution (1% phenyl mercuric acetate in 20:80 v/v methanol:water)
to give an overall concentration of 0.1% v/v. Tests have shown that the nitrate
concentration of raw and potable water is thus stabilised for at least four weeks.
However, tests should be carried out to verify the stability of other types of sample.
If the samples are to be analysed using ultraviolet spectroscopy [21], this form of
preservation may be unacceptable because of the interference arising. However, addition
of exactly equivalent amounts of preservation solution to all standards and samples will
allow correction of the interference, although with an increase in the limit of detection.

Table 18.1 Results of stability tests for total oxidised nitrogen

Results in mg N/L

Change in sample concentration

after T/4 after T

Lab
No.

Time T
(hours)

Sample
concn

Stability
target

Mean (Max 95%
confidence
limit)

Mean (Max 95%
confidence
limit)

2 30 1.9 0.10 0.010 0.098 −0.053 −0.074

3 24 1.75 0.088 −0.013 −0.0998† 0.0 II 0.086

3 24 1.98 0.099 0.010 0.0408 −0.011 −0.094

6 24 4.82 0.24 −0.040 −0.064 −0.080 −0.102

7 48 8.4 0.42 - - −0.012 −0.126

8 48 5.2 0.26 0.013 0.064 −0.018 −0.057

8 48 1.1 0.055 0.037 0.062* 0.016 0.041

10 a

48 2.7 0.14 −0.015 −0.049 −0.035 −0.061