Lesson 2 · 30 min

Kjeldahl nitrogen: from nitrogen to protein

How is the protein on a food label measured? Digestion, distillation and titration measure nitrogen, and a conversion factor turns it into protein. Use Jones’s 1931 data to see when 6.25 goes wrong, and why melamine could fool the method.

In this lesson

  • Describe the chemistry of the three Kjeldahl steps — digestion, distillation and titration
  • Calculate nitrogen content from a titre and a blank, and protein content with a conversion factor
  • Explain where the factor 6.25 comes from, its limits, and why nitrogen-rich adulterants fool the method

How is the “protein 3.4 g/100 mL” on a food label measured? There are thousands of different proteins, and weighing them one by one is impossible. In 1883 Johan Kjeldahl at the Carlsberg Laboratory in Copenhagen found a detour: measure not protein but nitrogen. Most of the nitrogen in food is in its protein, and proteins contain roughly the same proportion of nitrogen, about 16 % — so measure nitrogen, multiply by 100/16 = 6.25, and you have protein.

The method is simple, robust and easy to run in batches, and for well over a century it has been the standard in the analysis of food, feed, fertiliser and soil. It strings three basic operations together: concentrated sulfuric acid “burns” away the organic matter and turns the nitrogen into ammonium; alkali drives the ammonia out; and an acid–base titration counts the moles of ammonia.

Predict first

By N × 6.25, gelatin contains 91.4 % protein. Gelatin is almost pure protein, but it contains about 18 % nitrogen (more than the average 16 %). Roughly what is its real protein content?

Pick an answer and lock it in before you look. A wrong guess teaches more than no guess.

Digest, distil, titrate

Digestion: the sample is boiled with concentrated sulfuric acid, with potassium sulfate to raise the boiling point and a copper salt as catalyst. The organic matter chars black, its carbon and hydrogen are oxidised to CO₂ and H₂O, and the solution gradually clears; the nitrogen of amino and amide groups stays behind as NH₄⁺ (as ammonium hydrogen sulfate).

Distillation: after cooling, excess concentrated NaOH turns NH₄⁺ into NH₃, which is steam-distilled into a flask of boric acid. Boric acid is far too weak to titrate, but it catches every molecule of NH₃: NH₃ + B(OH)₃ + H₂O → NH₄⁺ + B(OH)₄⁻.

Titration: the B(OH)₄⁻ formed is a weak base about as strong as ammonia (pKb ≈ 4.8), strong enough to titrate accurately with standard hydrochloric acid; a methyl red–bromocresol green mixed indicator turns from green to grey-violet at the end point. One H⁺ per N. The reagents always carry a trace of nitrogen, so a blank without sample is run and its titre subtracted.

%N=(V−Vblank) cHCl×14.00710 m,%protein=%N×F\%\mathrm{N} = \frac{(V - V_\mathrm{blank})\,c_\mathrm{HCl}\times 14.007}{10\,m},\qquad \%\text{protein} = \%\mathrm{N}\times F
V in mL, c in mol/L, m in g; F is usually 6.25 — 6.38 for milk, 5.70–5.83 for wheat, 5.55 for gelatin.
VV
titre
VblankV_\mathrm{blank}
blank titre
FF
nitrogen-to-protein factor
Digest → distil into boric acid → titrate with 0.1000 M HCl · blank 0.10 mL

Kjeldahl nitrogen: how much protein is in milk

NXorganic→ΔHX2SOX4,CuNHX4X+\ce{N_{organic} ->[H2SO4, Cu][\Delta] NH4+} NHX3+B(OH)X3+HX2O→NHX4X++B(OH)X4X−\ce{NH3 + B(OH)3 + H2O -> NH4+ + B(OH)4-} B(OH)X4X−+HX+→B(OH)X3+HX2O\ce{B(OH)4- + H+ -> B(OH)3 + H2O}
Kjeldahl flask 05101520253035404550STIR50100150200
Burette reading 0.00mL
Receiver pH4.86

Kjeldahl flask: sample and conc. sulfuric acid

Digest, then distil, then titrate the borate in the receiver.

Each sample’s nitrogen is back-calculated from the protein listed as N × 6.25 in Table 6 of Jones (USDA Circular 183, 1931), itself from Atwater and Bryant’s food-composition tables; the conversion factors are from Table 5 of the same circular. Receiver: 50 mL of 4 % boric acid, about 100 mL after distillation; its pH uses the pKa of boric acid, 9.237, through the buffer equation, and is never more acidic near the end point than boric acid alone; the mixed-indicator colour is computed from the absorption spectra of methyl red and bromocresol green. The model assumes complete digestion and distillation, with one H⁺ consumed per NH₃.
Digest, distil, then titrate the receiver. Near the end point add single drops: the borate buffer’s pH drops steeply and the indicator turns from green to grey-violet. Switch to flour, peanuts and gelatin to see how 6.25 and the special factors differ.

Worked example

The protein in milk

After digestion and distillation, 5.000 g of milk takes 18.95 mL of 0.1000 mol/L HCl, with a blank of 0.10 mL. Find the nitrogen and protein contents (factor for milk 6.38).

Concept check

Why is the distilled ammonia led into boric acid rather than into a known amount of standard hydrochloric acid followed by a back-titration?

Try the calculation

1.000 g of flour takes 15.29 mL of 0.1000 mol/L HCl, with a blank of 0.10 mL. Calculate the protein content (%) with the factor for wheat flour, 5.70.

%

Key takeaways

  • Kjeldahl: digestion in sulfuric acid turns organic nitrogen into NH₄⁺; alkali drives NH₃ into boric acid; standard HCl titrates it, one H⁺ per N.
  • %N = (V − V_blank)·c·14.007/(10m); protein = %N × F, with F usually 6.25 (= 100/16) — 6.38 for milk, 5.70 for wheat flour, 5.55 for gelatin.
  • Kjeldahl measures total nitrogen: non-protein nitrogen and nitrogen-rich adulterants such as melamine are counted as protein.
Progress is stored only in this browser and never uploaded.

Data and references

  1. J. Kjeldahl, “Neue Methode zur Bestimmung des Stickstoffs in organischen Körpern”, Z. Anal. Chem. 22, 366–382 (1883).
  2. D. B. Jones, Factors for Converting Percentages of Nitrogen in Foods and Feeds into Percentages of Proteins, USDA Circular 183 (1931; revised 1941), Tables 5 and 6.
  3. D. C. Harris, Quantitative Chemical Analysis, 9th ed., W. H. Freeman (2016), ch. 10 (Kjeldahl nitrogen analysis).