Sample evaporation and its impact on the operating performance of an automated selective-access analytical system

In a series of gravimetric and analytical experiments I measured how much analytical error is ascribable to sample evaporation during routine operation of the Dimension analytical system. This error, a time-dependent function, is related to the volume of liquid placed in either a 0.5- or 2.0-mL cup and whether the cup is capped. For the 0.5-mL cups, analytical error at a 4-h residence time ranged from 12% to 16% (uncovered) to 8% to 13% (covered), whereas errors for the 2.0-mL cup ranged from 6% to 10% (uncovered) to 2% to 5% (covered). To minimize evaporative loss, I recommend (a) protecting the surface areas of the samples, (b) using larger samples, and (c) minimizing the time that a sample is in its cup before and during analysis.