Sources of error in bulk polymer densities determined by the gradient column technique

Sources of error in the determination of the density of bulk‐crystallized polyethylene in gradient columns have been reconsidered. Preferential absorption of any one column liquid can lead to gross differences between the apparent polymer densities as determined in different columns. It is therefore important to select column liquids which have about the same swelling power for the polymer. Samples of very high specific surface should be avoided, though the threshold for significant errors of this kind is unlikely to be crossed in practice. Finally, water should not be used as a column liquid in connection with polyethylene measurements because water‐rich mixtures may not wet the polymer completely. The principles emerging from this work are likely to be of general application.