Method Statement

TOTAL AMINES ANALYSIS

@(#) Uncontrolled Document amines.html Version 1.5 Last Revised on 11/20/02

The method of quantitation employed when determining total amines in waters or silica gel tubes is known as semi-quantitative. This involves the analysis, prior to the samples, Aa standard containing a known concentration of a nominated amine (diethylamine). Semi-quantitative analysis does not take into account the difference in detector response between different compounds. For example, 1ug injected of the diethylamine standard may produce a peak area of 2000 (arbitrary units) whereas a different amine present in the sample may only give a peak area of 1600 for 1ug injected. Quantitation on this basis would give a result of 0.8ug compared to the true value of 1ug. For quantitative analysis the standard contains the amine(s) known, or thought, to be in the sample and therefore there is no systematic error caused by the difference in detector response. However a specific amine(s) must be nominated for the analysis to be quantitative.

For GC/MS analysis there are two types of injection technique; splitless and split. In the splitless mode 100% of the injected aliquot is transferred into the GC/MS system whereas in split mode only, say, 10% of the injected aliquot enters the system. In normal organic analysis where the target compounds are dissolved in an organic solvent (ie. DCM, acetone) the splitless mode is used. This is because when the solvent is injected into the injection port it forms a fine, uniformly distributed ‘mist’ which is easily transferred into the system. However when water is injected (the solvent amines are soluble in) it does not behave as organic solvents but instead ‘globulizes’ in a non-uniform distribution and in the splitless mode not all of the ‘globules’ will transfer into the system. As a result two injections of the same solution will produce two different peak areas. In the split mode the gas flow arrangement allows the same % of the ‘globules’ to enter the system and therefore give a more reproducible result. However because only 1/10th of the injected aliquot enters the system the detection limit is roughly x10 higher than for a splitless injection.