Here's a random thought that I don't as yet have a satisfactory solution: Let's assume we have a copper column resting on an insulated table. Copper is an electric conductor, which means that some electrons within the copper column move freely throughout the column, subject to electromagnetic repulsion between other electrons, gravity, and being confined to be within the block. If we assume a square column that's 10 cm to a side, and 1 m in height, what would be a good estimate (order of magnitude) for the voltage difference between the top and bottom of the column? The atoms of copper have more or less fixed locations, while the conduction electrons can flow freely. Because of gravity, I would assume that the electron density would be larger at the bottom of the column than on top, and that this charge imbalance would amount to a voltage difference. This voltage must be pretty small, otherwise the peaks of metal towers would have a significant positive charge which would drastically affect the air around them not unlike an ionic breeze machine. Any ideas?