VisualChem The interactive chemistry textbook
中文

Water by coulometric Karl Fischer titration

Check the titrator with a water standard, then measure a dry solvent holding only 25 ppm of water and a wet one at 0.5 %. Change the generator current and the drift and find out when the result can be trusted and when the drift must be subtracted.

Difficulty ●●○ ≈ 25 min

Aim Turn charge into water with Faraday’s law (10.71 C per milligram), check the recovery with a water standard, and explain why the drift correction matters most for small samples and slow titrations.
Charge converted by Faraday’s law · the cell is a deterministic model

A coulometric Karl Fischer titrator

Titration cell
499.9μg H₂Ogenerator0.99μAdouble Pt indicatorsample
Generator current (area = charge)
drift current end point 0204060time after injection / s020406080100current / mA
Double-platinum indicator current (dead stop)
0204060time after injection / s00.51indicator / μA

5.4605 C passed in 58.9 s → 509.8 μg of water, of which the drift supplied 10 μg/min × 58.9 s = 9.82 μg.

Result 499.9 μg, i.e. 999.9 ppm (true value 1000.0 ppm, error −0.01 %).

At the anode 2I⁻ → I₂ + 2e⁻; with methanol, SO₂ and a base present the iodine reacts 1 : 1 with water, so each milligram of water needs 2F/M(H₂O) = 10.71 C. The model assumes 100 % current efficiency and an instantaneous reaction; the controller throttles the current in proportion to how far the iodine excess is below its set point (set point equal to 20 μg of water, proportional band 10 μg), and the end point is declared when the current falls back to within “drift + 3 μg/min”. The indicator current is taken as proportional to the iodine excess (the limiting current at two platinum pins under a small voltage). These control settings are chosen by hand, not taken from any instrument, and the sample water contents are set for the exercise.

Procedure

  1. With a drift of 10 μg/min and a largest current of 100 mA, inject 0.50 g of the water standard (1.00 mg/g). Record the titration time, the charge and the result, and work out the recovery.
  2. Switch off “subtract the drift” and repeat; then lower the current to 20 mA and titrate once with and once without the drift correction.
  3. Switch to the dry solvent (25 ppm) and inject 2.0 g. Measure first at 100 mA with a drift of 10 μg/min; then set 20 mA and a drift of 20 μg/min and compare the results with and without the drift correction.
  4. Switch to the wet solvent (0.50 %), inject 0.50 g, and titrate at 20 mA and at 400 mA, comparing the times.

What to notice

  • Water standard: 5.461 C pass in 58.9 s, equivalent to 509.8 μg of water, of which 9.8 μg is drift; subtracting it gives 499.9 μg, a recovery of 99.99 %. Without the drift correction the result is 2.0 % high.
  • At 20 mA the same injection takes 307.5 s, during which 51.3 μg of water leaks in: with the correction the result is still 499.7 μg, without it 10.2 % too high.
  • The dry solvent holds only 50 μg of water: at 100 mA the end point comes after 9.9 s and gives 25.0 ppm; at 20 mA with a drift of 20 μg/min the titration takes 45.5 s, and without the drift correction the answer would be 32.4 ppm — almost 30 % high.
  • 0.50 g of the wet solvent holds 2500 μg of water and needs 26.8 C: 1483.6 s (almost 25 minutes) at 20 mA but only 68.6 s at 400 mA. Wet samples call for a large current, or for the volumetric method instead.

In a real lab

  • Karl Fischer reagents contain iodine, sulfur dioxide and an organic base, usually in methanol: they are toxic and flammable, so work in a ventilated place and collect the waste separately.
  • Inject with a dry syringe and take the sample mass as the difference in the syringe’s mass before and after injecting; push the needle through the septum and below the liquid surface before injecting.
  • Renew the desiccant in the cell’s drying tube regularly. A sudden rise in drift usually means a punctured septum, spent desiccant or exhausted reagent.

Related lessons

All experiments