COD Is a Method, Not a Substance: What Your Dichromate Result Does and Does Not Say

Chemical oxygen demand is probably the most quoted number in industrial water analysis and one of the least understood. It is not the concentration of a substance. It is the amount of oxidant consumed when a sample is treated with a strong oxidant under a defined set of conditions, expressed as milligrams of oxygen per litre. What it really reports is the reducing load in the water, and that load is usually - but not always - organic matter.

That distinction matters, because the number behaves differently depending on which oxidant was used, how long the digestion ran, and what else was in the sample.

Two oxidants, two different numbers

The oxidants in routine use are potassium dichromate and potassium permanganate, and results from the two are not interchangeable - the detection method has to be stated alongside the value for the number to mean anything. Each country sets monitoring standards so that results stay comparable.

The dichromate result is what is conventionally called chemical oxygen demand, COD. The permanganate result is called oxygen consumption, abbreviated OC, and also referred to as the permanganate index. Potassium permanganate oxidises less completely but is simpler, and is adequate when you only need a relative comparison of organic content between samples. Potassium dichromate oxidises at a high rate with good reproducibility and is the right choice for total organic load.

There is a further wrinkle. Nitrogen-free organic matter oxidises readily under permanganate, while nitrogen-containing organic matter resists it. So oxygen consumption suits natural water and easily oxidised general wastewater, and the dichromate method is the one to reach for with complex organic industrial effluent.

The classic dichromate method, step by step

The reference procedure is specific. In a sulfuric-acid medium, potassium dichromate is the oxidant, silver sulfate the catalyst, and mercuric sulfate the masking agent for chloride ions. The digestion liquid is held at 9 mol/L sulfuric acid acidity, brought to the boil, and the boiling point of 148°C ±2°C becomes the digestion temperature. Reflux with water cooling runs for 2 h. After the digest cools naturally, ferroin indicator is added and the residual dichromate is titrated with ammonium ferrous sulfate solution; COD is calculated from that consumption. The oxidising species is hexavalent chromium, which is why it is called the dichromate method.

China's GB11914 and the international standard ISO6060 both codify this, and it has become the universally recognised classic standard. High oxidation rate, good reproducibility, accurate and reliable - and genuinely awkward to run. The reflux apparatus takes up a lot of bench space, consumes water and electricity, uses large reagent volumes, and does not lend itself to batch work.

The rapid sealed-tube alternative

Most of the development work since has gone into keeping the chemistry and shrinking everything else. The principle is the same: dichromate oxidises the organic matter, hexavalent chromium becomes trivalent chromium, and COD is read from the absorbance of one or the other.

Representative methods include the U.S. EPA Method 0410.4 automated and manual colorimetric method, ASTM D1252-2000 for the closed-reflux titrimetric dichromate reactor, and the international standard ISO15705-2002 for the small-scale sealed-tube method. China uses the State Environmental Protection Administration's unified rapid sealed catalytic digestion method, spectrophotometry included.

Relative to the classic method, the rapid versions push digestion-system sulfuric-acid acidity from 9.0 mg/L to 10.2 mg/L, raise reaction temperature from 150°C to 165°C, and cut digestion time from 2 h down to 10-15 min.

The hardware, and why the dimensions matter

The rapid digestion spectrophotometric method puts a small measured sample and reagent into a sealed tube, heats it in a small constant-temperature bath, and reads the result photometrically. Tube specification is 16 mm outside diameter, length 100 mm-150 mm, wall thickness 1.0 mm-1.2 mm, with a screw-open mouth and screw-on sealing cap. It has to be acid-resistant, high-temperature-resistant, and pressure- and explosion-proof. One version is digestion-only, called a digestion tube; another doubles as a colorimetric tube and is called a digestion-colorimetric tube.

The digester uses an aluminium block as the heating element with evenly distributed holes of 16.1 mm diameter and 50 mm-100 mm depth, set to the digestion reaction temperature. Tube dimensions are chosen so the digest occupies the right proportion of tube volume. The tube sits partly in the hole - bottom held at a constant 165°C, top protruding into air and cooling naturally to roughly 85°C. That temperature difference is what drives internal reflux and keeps the digest from boiling out.

Reading is done at two wavelengths depending on range: at 600 nm samples of 100-1000 mg/L can be measured, and at 440 nm samples of 15-250 mg/L. Microwave digestion has also been tried to speed things up further, but microwave ovens vary too much in power to allow a unified time-and-power standard, and the units are expensive.

Reading the number

China's surface-water classes set COD ceilings at 15 mg/L for Class I and II water, 20 mg/L for Class III, 30 mg/L for Class IV and 40 mg/L for Class V. In circulating cooling water, once COD by the KMnO4 method passes 5 mg/L the water quality has already begun to deteriorate.

In demineralisation, boiler water and circulating-water service, lower is simply better and there is no unified limit. Organic matter fouls ion-exchange resin, anion resin especially, cutting exchange capacity. Pretreatment by coagulation, clarification and filtration removes roughly half of it, but the desalination system removes essentially none, so what survives enters the boiler with make-up water and pulls boiler-water pH down. It can also carry into the steam and condensate system, lowering pH there and corroding it.

What a high COD does and does not prove

A high value means the water carries a large load of reducing substances, mainly organic pollutants - potentially from pesticides, chemical plants or organic fertiliser. Untreated, much of that organic load adsorbs to bottom sediment and persists there, producing toxic effects on aquatic life for years. Kill the aquatic community and the river's ecosystem collapses. Organisms that survive accumulate toxins that are frequently carcinogenic, teratogenic and mutagenic, and anything eating them takes that load up. Irrigating with the water spreads the problem to crops.

But high COD is not automatically that story. It needs disaggregating: which organic compounds, what they actually do to water quality and ecology, whether they are harmful to people. If that analysis is not available, a cheap proxy works - re-measure after a few days. A large drop indicates the reducing substances were mostly readily degradable organic matter, which is a far milder problem. Persistence indicates the refractory fraction that will end up in sediment.

BOD, BOD5 and the ratio that tells you about treatability

Biochemical oxygen demand measures the same broad question biologically. Full biological consumption takes more than 20 days, which is impractical, so the convention is to take the oxygen consumed after five days - roughly 95% - and report it as BOD5.

The BOD/COD ratio is the useful output, because it indicates biodegradability. A stream with a high COD and a low BOD is carrying material that biology will not touch, and a biological plant sized on the COD number will underperform badly. That ratio, not either number alone, should drive process selection.

Practical advice

Report the method with the number - dichromate or permanganate, reflux or sealed tube - or the figure is not comparable across sites or over time. For complex industrial effluent use dichromate; for relative comparison of natural water, permanganate is sufficient and cheaper. Match the wavelength to the expected range rather than diluting blind. And treat COD as a screening parameter with real limits: it includes inorganic reducing substances such as nitrites, sulfides and ferrous salts, so a spike is not proof of an organic discharge until you have checked what else moved.

On the programme side, the lesson from emission-reduction work is unglamorous but consistent: treat the treatment plant, the sewer network, sludge handling and water reuse as one system rather than four projects, and weight operational effectiveness over construction. A plant that runs well beats a bigger plant that runs badly.