Ammonia Nitrogen Measurement: Picking Between Nessler, Salicylate and Electrode Methods

What is actually being measured

Combined nitrogen existing as free ammonia (NH3) and ammonium ions (NH4+) is called ammonia nitrogen. It is a nutrient in water bodies that drives eutrophication, a major oxygen-consuming pollutant, and toxic to fish and some aquatic organisms. The form that does the damage is free ammonia, tens of times more toxic than ammonium salts and more toxic still as alkalinity increases. Toxicity tracks both pH and water temperature: generally the higher the pH and the warmer the water, the stronger the effect, and the harm to fish resembles that of nitrite.

The harm divides into acute and chronic. Chronic exposure reduces feeding, slows growth, damages tissue and impairs oxygen transport between tissues; acute exposure produces hyperactivity, loss of balance, convulsions and death. There is also a drinking-water dimension: ammonia nitrogen can convert to nitrite, and nitrite combining with proteins forms nitrosamines, a strong carcinogen.

Pretreatment first: colour, turbidity and interference

Samples with colour or turbidity, or containing other interfering substances, distort the determination, so appropriate pretreatment is required. For relatively clean water use the flocculation-sedimentation method; for heavily polluted water or industrial wastewater, distil.

Flocculation-sedimentation works by adding an appropriate amount of zinc sulfate to the water sample, then sodium hydroxide to make it alkaline, generating a zinc hydroxide precipitate that is filtered out along with colour and turbidity. In practice, take 100 mL of water sample into a stoppered graduated cylinder or colourimetric tube, add 1 mL of 10% zinc sulfate solution and 0.1-0.2 mL of 25% sodium hydroxide solution, adjust the pH to about 10.5 and mix well. Let it stand to precipitate, then filter through medium-speed filter paper that has been fully washed with ammonia-free water, discarding the first 20 mL of filtrate.

Distillation starts by adjusting the water sample to pH 6.0-7.4 and adding magnesium oxide to make it slightly alkaline. Alternatively add pH=9.5 Na4B4O7-NaOH buffer to make it weakly alkaline, though too high a pH promotes hydrolysis of organic nitrogen and gives high results. The distilled ammonia is absorbed in sulfuric acid or boric acid solution: boric acid for the Nessler colourimetric or acid titration finish, sulfuric acid (H2SO4) for the salicylic acid-hypochlorous acid finish. Condition the apparatus first by distilling water with light magnesium oxide until the distillate is ammonia-free. Then take 250 mL of water sample, or an aliquot made up to 250 mL so ammonia nitrogen does not exceed 2.5 mg, add a few drops of bromothymol blue indicator, adjust to about pH=7 with sodium hydroxide or hydrochloric acid, add 0.25 g of light magnesium oxide and a few glass beads, connect the nitrogen bulb and condenser with the conduit tip below the surface of the absorption liquid, and distil until the distillate reaches 200 mL. Make up to 250 mL. If the sample contains residual chlorine, add 0.35% sodium thiosulfate solution; each 0.5 mL removes 0.25 mg of residual chlorine.

Nessler’s reagent colourimetry

An alkaline solution of mercuric iodide and potassium iodide reacts with ammonia to form a light reddish-brown colloidal compound with strong absorption over a wide wavelength range; the measurement wavelength is usually somewhere in the 410-425 nm band, and 420 nm is the common working wavelength.

Reagent preparation drives sensitivity more than anything else, because the ratio of mercuric iodide to potassium iodide in Nessler’s reagent has a significant effect on the colour reaction, and any precipitate formed after standing should be removed.

Calibration: pipette 0, 0.50, 1.00, 3.00, 5.00, 7.00 and 10.0 mL of ammonium standard working solution into 50 mL colourimetric tubes, make up to the mark, add 1.0 mL of potassium sodium tartrate solution and mix, then add 1.5 mL of Nessler’s reagent and mix again. After standing for 10 min, measure absorbance at 420 nm in a 20 mm path-length cuvette against water. Subtract the absorbance of the zero-concentration blank and plot ammonia nitrogen content against corrected absorbance. Supporting reagents are straightforward: 10% zinc sulfate, 25% sodium hydroxide stored in polyethylene, potassium sodium tartrate (KNaC4H4O6·4H2O) boiled to remove ammonia, and an ammonium stock made by weighing 3.819 g of ammonium chloride (NH4Cl) dried at 100°C, which contains 1.00 mg of ammonia nitrogen per mL.

Performance: the minimum detection concentration is 0.025 mg/L by photometry with an upper limit of 2 mg/L; using the visual colourimetric method the minimum detectable concentration is 0.02 mg/L. With appropriate pretreatment the method applies to surface water, groundwater, industrial wastewater and domestic sewage. Interferences are numerous: aliphatic amines, aromatic amines, aldehydes, acetone, alcohols and organic chloramines, plus inorganic iron, manganese, magnesium and sulfur, all produce colour or turbidity, and the sample’s own colour and turbidity affect colourimetry. Flocculation-sedimentation filtration or distillation handles them, easily volatile reducing interferents can be removed by heating under acidic conditions, and metal ion interference by an appropriate masking agent. Two housekeeping points matter: filter paper often carries trace ammonium salts, so wash it with ammonia-free water, and keep glassware away from ammonia in laboratory air.

Inter-laboratory checks are reassuring. Three laboratories analysing spiked water samples containing 1.14-1.16 mg/L ammonia nitrogen produced single-laboratory relative standard deviation not exceeding 9.5%, with spike recovery of 95-104%. Four laboratories on samples containing 1.81-3.06 mg/L gave relative standard deviation within 4.4% and recovery of 94-96%.

Salicylate-hypochlorite colourimetry

In the presence of sodium nitroprusside, ammonium reacts with salicylate and hypochlorite ions to form a blue compound with maximum absorption at 697 nm. The minimum detection concentration of this method is 0.01 mg/L with an upper limit of 1 mg/L, and it applies to drinking water, domestic sewage and most industrial wastewaters.

Reagents: weigh 50 g of salicylic acid (C7H6O3), add 100 mL of water, then 160 mL of 2 mol/L sodium hydroxide and stir until completely dissolved; separately dissolve 50 g of potassium sodium tartrate, combine, make up to 1000 mL and store in a brown glass bottle, where it is stable for at least a month. If the salicylic acid does not fully dissolve, add a few more mL of sodium hydroxide; the final pH should be 6.0-6.5. Standardise a sodium hypochlorite solution and dilute it with sodium hydroxide to 0.35% (m/V) available chlorine and 0.75 mol/L free alkali calculated as NaOH; it keeps for a week in a brown dropper bottle. Sodium nitroprusside Na2[Fe(CN)6NO]·2H2O is weighed at 0.1 g into a 10 mL stoppered colourimetric tube and made up fresh.

Standards: 3.819 g of ammonium chloride (NH4Cl) dried at 100°C, made to 1000 mL, contains 1.00 mg of ammonia nitrogen per mL; 10.00 mL of that diluted to 100 mL gives 0.10 mg per mL; 10.00 mL of the intermediate diluted to 1000 mL gives 1.00 µg per mL and should be prepared immediately before use. Calibration: pipette 0, 1.00, 2.00, 4.00, 6.00 and 8.00 mL of working solution into 10 mL tubes, dilute to 8 mL, add 1.00 mL of colour-developing solution and 2 drops of sodium nitroprusside, mix, add 2 drops of sodium hypochlorite, dilute to the mark and mix thoroughly. After standing for 1 h, measure absorbance at 697 nm in a 10 mm cuvette against water. Note that when the sample has been distilled, sulfuric acid should be the absorption liquid and sodium hydroxide must be added before colour development to neutralise it.

The electrode method and the titration method

The ammonia gas-sensing electrode is a composite electrode with a pH glass electrode as indicator and a silver-silver chloride electrode as reference, held in a plastic tube containing 0.1 mol/L ammonium chloride internal filling solution. Near the sensitive membrane a hydrophobic semipermeable film separates the internal electrolyte from the external test solution, leaving a very thin liquid film between membrane and electrode. Adding strong alkali to raise the sample above pH 11 converts ammonium to ammonia, which diffuses through the membrane while water and other ions cannot, shifting the ammonium chloride film equilibrium and changing hydrogen ion concentration. At constant ionic strength the measured electromotive force is linear in the logarithm of ammonia nitrogen concentration.

Colour and turbidity have no effect and the sample need not be pre-distilled, but the temperatures of standard and sample should match and total dissolved substances should be approximately equal. Minimum detection concentration is 0.03 mg/L and the upper determination limit is 1400 mg/L. Calibration: pipette 10.00 mL of ammonium standard solutions at 0.1, 1.0, 10, 100, 1000 mg/L into 25 mL beakers, immerse the electrode, add 1.0 mL of sodium hydroxide-Na2-EDTA solution and, while stirring, read the stable potential, taking the reading when the change does not exceed 1 mV within 1 min. Plot the E-log c curve on semi-logarithmic coordinates.

The titration method applies only to samples that have been distilled. Adjust to pH 6.0-7.4, add magnesium oxide to make it slightly alkaline, heat and distil, absorb the released ammonia in boric acid and titrate with standard acid using methyl red-methylene blue indicator. Standardise the acid against primary-standard-grade anhydrous sodium carbonate (Na2CO3) dried at 180°C for 2 h, weighing about 0.5 g to 0.0001 g, dissolving it in freshly boiled and cooled water and titrating aliquots against methyl orange to light orange-red. Concentration follows from the volume used. Be aware of one bias: anything that distils out under these conditions and reacts with acid during titration, volatile amines for instance, will push the result high.

Choosing a method

Nessler’s reagent colourimetry is simple to operate and sensitive, but calcium, magnesium and iron ions, sulfides, aldehydes and ketones, colour and turbidity all interfere and each needs its own pretreatment. The phenol-hypochlorite, or salicylate-hypochlorite, colourimetric method is sensitive and stable with the same interferences and the same remedies. The electrode method usually needs no pretreatment at all and has the advantage of a wide measurement range. When ammonia nitrogen content is high, distillation followed by acid titration remains a legitimate option. Decide on two things: what else is in the sample besides ammonia, and the concentration range you actually expect to see.