Summary
The subject of argon diffusion is reviewed in its different aspects, but with emphasis upon the mechanism of diffusion and with the object of formulating criteria of retentivity for the different minerals.
The numerous formulae which have been developed to relate the observed argon loss from a sample to the fundamental physical concept of activation energy make a number of assumptions which have not been sufficiently checked. The most simple model of diffusion assumes identical, highly symmetrical grains of a homogeneous, isotropic substance with regularly spaced diffusion sites of a single activation energy. Actual crystal grains may vary in size, shape and composition, are usually anisotropic, and, because the argon atom is too large to diffuse through a perfect lattice, diffusion proceeds through imperfections which may be irregularly spaced and of varying activation energies, and may vary among apparently similar crystals. The problem is further complicated because argon is formed in an excited state and recoils out of its lattice position when it goes into the ground state. Only some of these complications have been considered by experimenters.
The concept of activation energy has little practical use in this subject as its deduction from the experimental observations requires the largest number of assumptions, and because it is a very insensitive measure of argon loss.
A number of experimental methods have been used but are not equivalent because they make different sets of assumptions, some of which are not valid. Confusion has also resulted from use of incorrect formulae and from misinterpretation of results.
Loss of argon from minerals in laboratory experiments differs from natural loss in three ways: the mineral grains may have been damaged in extraction from the rock, the separated grains lose argon into a vacuum or gas atmosphere and not into the rest of the rock, and loss usually takes place at a much higher temperature. The evidence suggests these all may be important, but they have not been taken sufficiently into account.
It is concluded that no criteria of retentivity can yet be formulated, and that a great deal of work must yet be done before this can be achieved. However, the evidence suggests that, because it can only diffuse through major crystal imperfections, argon may be retained completely within small regions of a crystal which, as a whole, has lost argon. The separation of this argon and associated potassium may be possible using special techniques, so yielding a true age.
