Dairy product protein testing procedures define the standardized workflow for verifying protein content in milk and milk products, supporting regulatory compliance, accurate nutrition labeling, and fair commercial valuation. This technical guide outlines the applicable ISO and GB standards, sample handling requirements, reference methods such as Kjeldahl and Dumas combustion, and the quality control practices expected in accredited third-party laboratories.
Technical Overview
Protein is the principal compositional and economic indicator of dairy products. It determines nutritional value, influences yield in cheese and concentrate manufacturing, and serves as the basis for raw milk payment systems in most dairy-producing regions. Because dairy proteins are almost always quantified indirectly through nitrogen measurement, protein testing procedures must strictly control every step from sampling to calculation to ensure valid results.
The internationally harmonized framework for dairy protein determination is built on the ISO 8968 (IDF 20) series for nitrogen determination by the Kjeldahl principle, ISO 14891 (IDF 185) for the Dumas combustion method, and ISO 9622 (IDF 141) for rapid mid-infrared analysis. In China, the corresponding national standard is GB 5009.5, applied together with product standards such as GB 19301 for raw milk, GB 25190 for sterilized milk, and GB 19644 for milk powder. All nitrogen results are converted to protein using the dairy-specific factor of 6.38, derived from the average nitrogen content of milk protein of approximately 15.67 percent.
A complete testing procedure comprises sample acceptance and registration, conditioning and homogenization, determination of total nitrogen, optional fractionation of non-protein nitrogen, calculation and rounding, and reporting with associated quality control data such as blanks, duplicates, and certified reference material recoveries.
Testing Samples
Liquid Milk Products
This category includes raw milk, pasteurized milk, UHT and in-container sterilized milk, reconstituted milk, flavored milk beverages, and liquid whey. Before subsampling, milk should be conditioned to approximately 38 degrees Celsius with gentle inversion to redisperse separated fat, then cooled to room temperature while continuing gentle mixing, as described in ISO 707 for sampling of milk and milk products. Raw milk samples intended for delayed analysis are stored at 2 to 6 degrees Celsius and may be preserved with potassium dichromate where permitted by regulation.
Fermented Dairy Products
Set or stirred yogurt, drinking yogurt, kefir, and cultured buttermilk require thorough homogenization before testing because protein is distributed between the coagulum and the serum phase. Samples are mixed mechanically until uniform, with care taken to avoid whipping in air, and analyzed promptly to prevent continued acidification from altering protein solubility behavior during fractionation tests.
Milk Powders and Protein Ingredients
Skimmed and whole milk powder, whey protein concentrate, milk protein concentrate, lactose-free powders, and infant formula bases are hygroscopic and must be opened and handled in a low-humidity environment. Powders are homogenized by repeated inversion and rolling, and test portions are weighed rapidly using sealed vessels to prevent moisture uptake that would bias results on an as-is basis. Moisture and water activity are frequently determined in parallel so that protein can also be reported on a dry matter basis.
Cheese and High-Fat Products
Natural cheese, processed cheese, cream, and butterfat-rich products are grated, comminuted, or warmed slightly below 40 degrees Celsius to obtain a homogeneous mass. The digestion stage must be extended with additional acid and catalyst to ensure complete decomposition of the high fat and matrix content typical of these products.
Testing Items
Protein testing of dairy products typically covers the following measurable items, each with a distinct definition and field of application:
- Total nitrogen and crude protein: the classical protein value reported on product labels and certificates of analysis.
- True protein: total nitrogen minus non-protein nitrogen, representing only peptide and protein-bound nitrogen.
- Non-protein nitrogen (NPN): low-molecular-weight nitrogen compounds such as urea, amino acids, and peptides, determined after precipitation of proteins.
- Casein nitrogen and whey protein nitrogen: fractionated values used to characterize protein composition and heat treatment history.
- Whey protein nitrogen index (WPNI): used for the heat classification of skimmed milk powder.
- Protein on an as-is and dry matter basis: both values are reported for powders to allow specification comparisons.
Testing Methods
Kjeldahl Method (Reference Procedure)
The Kjeldahl method specified in ISO 8968-1 (IDF 20-1) and Method I of GB 5009.5 remains the legal and contractual reference for dairy protein. A weighed or pipetted test portion is digested with concentrated sulfuric acid in the presence of potassium sulfate and a copper-based catalyst at 380 to 420 degrees Celsius until the solution clarifies and all organic nitrogen is converted to ammonium sulfate. After cooling, the digest is alkalized with sodium hydroxide and steam distilled, and the liberated ammonia is trapped in a boric acid receiving solution. The captured ammonia is titrated with standardized hydrochloric or sulfuric acid using an indicator or potentiometric endpoint. Protein content is calculated as protein (g/100 g) = (V1 minus V0) multiplied by acid molarity, multiplied by 0.014, multiplied by 6.38, multiplied by 100, divided by the sample mass, where V1 and V0 are the titration volumes of the sample and reagent blank.
Dumas Combustion Method
ISO 14891 (IDF 185), ISO 8968-5 (IDF 20-5), and the combustion method of GB 5009.5 describe the high-temperature combustion alternative. The test portion is combusted at approximately 950 degrees Celsius in pure oxygen, combustion gases are cleaned and reduced so that all nitrogen oxides are converted to molecular nitrogen, and the nitrogen is quantified by gas chromatographic separation with thermal conductivity detection. The method uses no concentrated acids, requires only a few minutes per determination, and delivers results statistically equivalent to Kjeldahl when instruments are properly calibrated with certified reference materials.
Mid-Infrared Spectroscopy (Rapid Screening)
ISO 9622 (IDF 141) governs routine milk analysis by mid-infrared instruments used in dairy plants and milk collection centers. Protein is measured through absorbance of the peptide bond region around 6.46 micrometers, with correction filters for fat and lactose interferences. Instruments must be calibrated and periodically standardized against the Kjeldahl or Dumas reference methods, and daily performance is verified with control milk samples and inter-instrument comparison loops.
Nitrogen Fractionation Procedures
True protein and NPN are determined according to ISO 8968-4 (IDF 20-4) by precipitating proteins with trichloroacetic acid at a specified final concentration, filtering, and measuring nitrogen in the protein-free filtrate by Kjeldahl or Dumas. Casein nitrogen is determined after isoelectric precipitation at pH 4.6 following ISO 17997 (IDF 29), while the whey protein nitrogen index of milk powders is assessed using standard sedimentation and staining procedures referenced in ISO 29663. These fractionation tests are central to detecting nitrogen adulteration, because added non-protein nitrogen inflates the crude protein value while leaving true protein unchanged.
Quality Control and Validation
Each analytical batch must include a reagent blank, duplicate determinations agreeing within the repeatability limits of the applicable standard, and a certified reference material or control sample with known nitrogen content. Titration acids are standardized against primary standards, balances and pipettes are calibrated on a documented schedule, and laboratories participate in proficiency testing schemes to demonstrate ongoing competence. Uncertainty budgets typically report expanded uncertainties of 1 to 3 percent relative for protein content in liquid milk.
Applications
Dairy protein testing procedures are applied across the entire value chain. Raw milk intake and payment systems rely on rapid infrared screening calibrated to reference methods, while processed product release testing uses Kjeldahl or Dumas values for certificates of analysis. Nutrition labeling under regulations such as GB 28050 and the European Union food information regulation requires verified protein figures for front-of-pack and back-of-pack declarations.
Regulatory verification compares measured protein against product standards, for example minimum protein levels of 2.8 g/100 g for raw milk under GB 19301 and 2.9 g/100 g for heat-treated liquid milk under GB 25190, as well as the compositional ranges specified for infant formula in GB 10765. Fractionation and targeted screening support food fraud surveillance for nitrogen-rich adulterants such as urea and melamine, the latter controlled through dedicated chromatographic methods including GB/T 22388. Additional applications include heat classification of milk powders through WPNI, verification of protein content claims in fortified and protein-enriched beverages, and contractual settlement in international dairy trade where ISO or GB methods are cited in purchase specifications.
FAQ
Why is the conversion factor 6.38 used for dairy instead of 6.25?
Milk proteins contain on average 15.67 percent nitrogen, corresponding to a factor of 100 divided by 15.67, which equals 6.38. The general food factor of 6.25 assumes 16 percent nitrogen and would understate dairy protein by roughly 2 percent, so 6.38 is mandated in ISO 8968, GB 5009.5, and contractual dairy specifications.
What is the difference between crude protein and true protein?
Crude protein is total nitrogen multiplied by 6.38 and includes all nitrogenous compounds. True protein excludes non-protein nitrogen such as urea and free amino acids and is obtained by subtracting NPN determined after protein precipitation. Regulatory limits and labeling are generally based on crude protein, while the difference between the two values is a key indicator of adulteration or abnormal milk.
When should a laboratory choose Dumas over Kjeldahl?
Dumas is preferred for high sample throughput, minimal chemical handling, and fast turnaround, while Kjeldahl remains the statutory reference in many jurisdictions and is required where contracts or regulations explicitly name it. Both methods must be validated as equivalent through comparative studies against certified reference materials when used interchangeably.
How should liquid milk samples be stored before protein analysis?
Samples are held at 2 to 6 degrees Celsius and analyzed as soon as practicable, generally within 24 to 48 hours for unpreserved raw milk. Before testing, milk is warmed to about 38 degrees Celsius and mixed gently to re-emulsify fat, then equilibrated to room temperature so that the pipetted test portion is fully representative.
Can infrared analyzers replace reference methods entirely?
No. Infrared instruments deliver excellent precision and speed for routine screening but are indirect methods whose accuracy depends entirely on calibration transfer from Kjeldahl or Dumas reference values. Accredited testing programs combine daily infrared screening with periodic verification by reference methods on representative samples.
