Why Ammonia Nitrogen Can Test Higher Than Total Nitrogen: Causes and Fixes

Total nitrogen and ammonia nitrogen are routinely measured by two different methods: alkaline potassium persulfate digestion with ultraviolet spectrophotometry (HJ 636-2012) for total nitrogen, and Nessler's reagent spectrophotometry (HJ 535-2009) for ammonia nitrogen. Total nitrogen is defined as the sum of all forms of inorganic and organic nitrogen in the water — nitrate, nitrite and ammonium as inorganic nitrogen, plus organic nitrogen in proteins, amino acids and organic amines — expressed as milligrams of nitrogen per litre. Ammonia nitrogen is the nitrogen present as free ammonia (NH3) and ammonium ion (NH4+). Both are used to express the degree of nutrient pollution of a water body.

In theory total nitrogen equals ammonia nitrogen plus organic nitrogen plus nitrate nitrogen. In practice the measured result often falls short of that, and in some samples ammonia nitrogen comes out equal to or greater than total nitrogen. Why does the result invert, and at which step did it go wrong?

1. Sample Storage

Nitrogen compounds in a sample change continuously, so a water sample must be analysed immediately after collection, or stored at below 4 °C for no more than 24 hours. Where longer storage is unavoidable, add 0.5 mL of sulfuric acid (1.84 g/mL) to 1000 mL of sample to bring the pH below 2, and measure as soon as possible. When subsampling, allow for the risk of cross-contamination with ambient air: seal the sample promptly after collection, and avoid the errors caused by light-induced temperature change and by the laboratory environment.

2. Laboratory Environment

Both determinations must be carried out in an ammonia-free laboratory. The room must contain no petroleum products or other nitrogen compounds, and total nitrogen must never be analysed in a laboratory where ammonia nitrogen and other nitrogen determinations are performed. Reagents and glassware must be stored separately, kept dry and clean, and protected from cross-contamination.

3. Reagent Selection and Preparation

Every solution depends on the water used to make it, starting with ammonia-free water. Ammonia or ammonium salts from the air dissolve in the water during preparation, contaminating the reagent water. The resulting blank is very hard to eliminate and inflates the total nitrogen blank in particular, so that total nitrogen reads low against the true value. Once prepared, ammonia-free pure water must be stored properly and, as far as possible, made fresh for each use.

Total nitrogen is determined by alkaline potassium persulfate digestion, a method that is simple in procedure and undemanding in instrumentation but demanding on blank absorbance. The main influence on the blank is the quality of the potassium persulfate. When preparing the alkaline potassium persulfate solution, dissolve the sodium hydroxide first and then the potassium persulfate. Persulfate dissolves very slowly, so a water bath may be used, with the temperature held between 55 and 60 °C. Once the persulfate solution is fully dissolved and cooled to room temperature, add it slowly to the sodium hydroxide solution while stirring, so that the heat of dilution cannot raise the local temperature far enough to decompose the persulfate. Poor-quality persulfate gives a high blank and can seriously distort the sample result, so keep the bath at 50 to 60 °C during preparation; above that range the persulfate decomposes and loses its effectiveness.

4. Digestion and Colour Development Time

Nessler's reagent determination of ammonia nitrogen involves nothing complicated — 10 minutes of colour development, with almost no loss of ammonia nitrogen during the determination. Total nitrogen is different: the atomic oxygen released converts nitrogen compounds to nitrate during 30 minutes of digestion at 120 to 124 °C, and total nitrogen is then calculated from the absorbance at 220 nm and 275 nm. In practice, problems with autoclave temperature or with the reagent can leave the persulfate digestion incomplete, and a colorimetric tube whose ground-glass stopper is not fully sealed allows ammonia to escape and the sample to evaporate during digestion. When the tube is then made up to the mark with pure water, the sample volume has increased and the total nitrogen content reads low — which is how total nitrogen ends up less than or equal to ammonia nitrogen.

5. Colour and Turbidity of the Sample

If the sample is not treated, or coagulation and sedimentation are incomplete, the ammonia nitrogen result reads high. In the total nitrogen determination the same colour and turbidity are removed by 40 minutes of high-temperature digestion, after which a white precipitate is visible at the bottom of the tube and the sample is colourless and clear. Where coagulation and sedimentation cannot remove colour and settle the sample completely before the ammonia nitrogen determination, use distillation instead; that removes the positive bias in the ammonia nitrogen result more reliably.

6. Ammonium in the Potassium Sodium Tartrate

When preparing the reagents for ammonia nitrogen, watch the ammonium content of the potassium sodium tartrate. A high ammonium content biases the ammonia nitrogen result upwards, so that it exceeds total nitrogen. Heating and boiling the tartrate alone does not remove the ammonia: add a small quantity of sodium hydroxide solution as well, boil off about 25% of the volume, then cool and dilute back to the original volume with ammonia-free water. That reduces the ammonium salt in the tartrate solution effectively and makes the result more accurate.

In Practice

1. Analyse total nitrogen and ammonia nitrogen on the sample as soon as it arrives, and have two analysts run the two determinations in parallel — that removes most of the error introduced by analysing at different times.

2. Control the quality of the reagents for total nitrogen and the whole preparation procedure strictly, so that incomplete digestion cannot bias the result.

3. Deal with colour, turbidity and the other factors that bias ammonia nitrogen high before running the ammonia nitrogen determination.

When ammonia nitrogen still reads higher than total nitrogen after all of this, the sample itself is the place to look: a genuinely unusual nitrogen balance is possible in industrial discharges, and the two results should then be confirmed against a distilled sample and a spiked recovery test rather than assumed to be an analytical error.