Method for avoiding the interference of formamide with the Karl Fischer titration

Abstract

The Karl Fischer titration is based on a specific chemical reaction. Several measures exist to make all the water of insoluble samples accessible for the chemical reactants. The most efficient are the titration at elevated temperatures, the use of a homogenizer in the titration vessel and the modification of the polarity of the working medium (essentially methanol) by the addition of appropriate solvents like chloroform or formamide. It is known however that formamide interferes with the Karl Fischer reaction and so causes more or less false results. This effect increases with higher temperatures. A method is therefore presented to avoid this interference, even when working at the boiling point of the working medium. It takes advantage of the fact that the side reaction has a practically constant velocity, at least as long as usual titrations last. Thus, a constant additional consumption of Karl Fischer reagent is observed. This can be accounted for by measuring this effect before the start of the determination and by deducting the additional reagent consumption, which is proportional to the duration of the titration, from the totally added volume. With certain modern titrators this can even be carried out automatically. They can continuously measure the so-called drift, the titration rate necessary to keep the titration cell dry, and have the capability to use this drift as stop criterion for the titration. This means that the analysis is terminated when the drift existing before the titration is reached again. The additional consumption of reagent, to be deducted from the total volume, can (automatically) be calculated from the drift rate and the titration time. The proposed procedure allows the use of formamide as additional solvent, even at high temperatures, in order to shorten determination times considerably. It avoids false results due to the interference, which has so far prevented its use when exact results were desired and when the duration of the analysis was long.