The Dumas method is a combustion technique that measures total nitrogen in a sample by burning it at extremely high temperatures, then quantifying the nitrogen gas that remains. First developed in 1831 by French chemist Jean-Baptiste Dumas, it is one of the two primary ways laboratories determine protein content in food, animal feed, soil, and other organic materials. The other is the Kjeldahl method, which uses acid digestion instead of combustion.
Because proteins contain nitrogen in predictable proportions, measuring nitrogen gives you a reliable estimate of how much protein is present. This makes the Dumas method a cornerstone of food safety testing, agriculture, and nutrition labeling.
How the Dumas Method Works
The process starts by placing a small sample, typically between 50 milligrams and 2 grams, into a combustion tube heated to around 940°C. Oxygen is injected alongside the sample, triggering what’s called flash combustion. During this burst, temperatures can briefly spike to 1,800°C, which is hot enough to completely break down all organic matter into simple gases: carbon dioxide, water vapor, nitrogen oxides, and sulfur compounds.
These gases then pass through a series of cleanup stages. First, they flow over heated copper at about 700°C. The copper does two things: it converts nitrogen oxides into pure nitrogen gas and traps sulfur compounds along with any leftover oxygen. Next, the gas stream is dried and the carbon dioxide is removed by a chemical trap. What’s left is essentially pure nitrogen carried along in an inert stream of helium.
The final step is detection. A thermal conductivity detector (TCD) measures the nitrogen by comparing two gas flows: one of pure helium and one of helium mixed with the sample’s nitrogen. Because nitrogen and helium conduct heat differently, the detector can precisely measure how much nitrogen is present based on the difference between the two flows. The instrument converts this reading into a percentage of nitrogen in the original sample.
From Nitrogen to Protein
The Dumas method measures nitrogen, not protein directly. To get a protein estimate, laboratories multiply the nitrogen percentage by a conversion factor. This factor varies by food type because different proteins contain different proportions of nitrogen. Milk and whey products use a factor of 6.38, while most other foods use the general factor of 6.25. These are known as Jones factors, and choosing the right one matters for accurate labeling.
One important caveat: the method measures all nitrogen in a sample, not just nitrogen from protein. Nitrogen from other sources, such as nitrates in soil or non-protein nitrogen compounds in food, gets counted too. For most food testing this isn’t a significant problem, but it’s something analysts account for when working with unusual matrices.
What Gets Tested With This Method
The Dumas method is used across a remarkably wide range of materials. The FDA routinely uses it to test infant formulas (both milk-based and soy-based), medical foods, dietary supplements, and general food products. Reference materials that labs use for calibration include peanut butter, soy flour, fortified breakfast cereal, whey protein drink mix, dried cat food, blue corn flour, dried taro, and frozen turkey meals. Sample protein content in these materials ranges from about 2.5% to 75%.
Beyond food, the method is standard in agricultural labs for measuring total nitrogen in soil, which is a key indicator of soil fertility. It’s also used in animal feed testing and pharmaceutical applications where nitrogen content matters.
The formal standard most labs follow is AOAC 992.15 for Dumas combustion, which sits alongside the Kjeldahl standards (AOAC 981.10 and 928.08). Both methods are recognized by the International Organization for Standardization (ISO) and several other industry bodies including the American Oil Chemists’ Society and the Cereals & Grains Association.
Dumas vs. Kjeldahl
The Kjeldahl method, developed about 50 years after the Dumas method, dominated laboratory protein testing for most of the 20th century. It works by digesting a sample in concentrated sulfuric acid, converting organic nitrogen to ammonium, then measuring that ammonium through distillation and titration. It’s a wet chemistry approach that takes considerably longer per sample.
The Dumas method’s biggest advantage is speed. A single analysis typically takes three to five minutes, while Kjeldahl can take an hour or more. Dumas also avoids the use of concentrated acids and toxic catalysts, making it safer and producing less chemical waste. Modern automated Dumas instruments can run samples back to back with minimal hands-on time.
Accuracy is comparable between the two. FDA testing on infant formulas showed Dumas spike recoveries (a measure of how accurately the method captures known amounts of nitrogen) ranging from about 99% to 104%, with very low variation between repeated measurements. This level of precision, combined with the speed advantage, is why many labs have shifted to Dumas as their primary method while keeping Kjeldahl available for cross-validation.
The Kjeldahl method still has an edge in certain situations. It can handle larger sample sizes more easily, which helps when testing materials that aren’t uniform. Some newer Dumas instruments accept samples up to 3 grams, which narrows this gap. For samples with very low nitrogen content, the Dumas method has a typical detection limit of about 0.1 grams of nitrogen per kilogram of sample and a quantification limit of about 0.3 grams per kilogram, though this varies by instrument.
Sample Preparation
Because the Dumas method uses such small sample sizes, preparation matters. If you’re weighing out more than 1 gram, a standard sieved sample (particles smaller than 2 mm) is usually fine. But for samples under 1 gram, it’s recommended to mill the material down to 0.5 mm or smaller. This ensures the tiny portion you analyze actually represents the whole batch. Running samples in triplicate is standard quality assurance practice, and the FDA has found this approach highly useful for catching inconsistencies.
Samples can be tested in powder, liquid, or solid composite form. Liquid samples like ready-to-feed infant formula require slightly different handling than dry powders, but modern instruments accommodate both. The key requirement is that whatever goes into the combustion tube must be homogeneous enough that a few hundred milligrams tells you what’s in the whole product.