An attempt is being made to analyze the forces between unit layers of a bentonite particle and between the flat surfaces of two particles from colloid chemical data and theory. The electric double-layer interaction, the van der Waals attraction, and the hydration energies are considered.

From experimental observations the general qualitative shape of the net potential curve of interaction may be derived.

In a quantitative discussion, first the long-range interaction is considered by applying the theory of Verwey and Overbeek in which the net interaction curve is found as the sum of the double-layer repulsion and van der Waals attraction. The fact that the layers do not spontaneously dissociate in solution despite the high charge density of the surfaces points to a considerable specific adsorption potential of the counter ions (the exchangeable ions) to the surface (≈0.2 ev). The consequences for ion exchange are briefly discussed.

Next, two possibilities for the status and position of the counter ions are considered: the cations are either unhydrated and located in or close to the holes in the tetrahedral sheet in six coordination with the oxygen six rings or they are partly or completely hydrated and with their hydration shell adsorbed on top of the oxygen sheet.

Then the short-range interaction of the unit layers is considered. In the case of bentonite there is the fortuitous situation that the net short-range potential curve for the layer interaction can be found experimentally from adsorption isotherms and x-ray diffraction. This net interaction energy is the sum of three contributing energies: the van der Waals attraction, the electrostatic interaction, and the hydration energies. Therefore, on computing the first two for either model mentioned above, the hydration term can be evaluated and its order of magnitude may decide between the two possibilities since ion hydration energy would be expected to be much higher than surface hydration energy.

For the model of unhydrated ions the van der Waals attraction is relatively small and partly compensated by the rather small electrostatic repulsion (repulsion of two finite layers of dipoles). Thus the net repulsion energy measured is practically the hydration energy in this case. This hydration energy figure is of the right order of magnitude if interpreted as hydrogen bonding of a Hendricks water layer to the oxygen surface.

For the model of the at least partly hydrated cations the van der Waals attraction is small, the same as in the first model, but the electrostatic interaction is now an appreciable attraction since the cations take up a midway position between the negative bentonite layers for at least the mono- and dilayer adsorption complex. This makes the computed hydration energy higher; this is readily explained by the availability of the higher ion hydration energy as compared with the surface hydration energy.

Thus the analysis shows both pictures to be acceptable within the limits with which the adsorption isotherms and therefore the net potential curve can be evaluated. More elaborate adsorption data at very low relative humidities would be required to obtain a better estimate of the energies involved in the removal of the last layer of water where the major portion of the ion hydration energy would enter the picture if ion hydration would indeed take place and be the primary cause of interlayer swelling of bentonite.

There is a correlation of bentonite hydration energy with the hydration energy of the counter ions present, but this correlation does not constitute positive evidence of ion hydration. A secondary effect of the ions on the formation of ideal Hendricks’ layers (and thus on their bonding energy to the clay surface), which also depends on ion size, is possible.

Other experimental and theoretical approaches to the problem are briefly mentioned and finally the differences between swelling and nonswelling clays are dealt with.