The number of bases affect the fusibility of a clay. For instance, mixtures of lime and silica, potash and silica, soda and silica and alumina and silica can be made that will fuse at the same temperature; a combination of any two of these mixtures will fuse at a lower temperature; a combination of any three at a still lower temperature, and so on. If the fusibility is affected and lowered by the number of bases, the burning qualities at heats below fusion or vitrifaction must be affected, because any given heat must be nearer the vitrify-ing point of the compound than of the simple body, consequently the compound body must be closer or less porous.
From the short consideration given above to the chemical elements found in clays and their effect on burning qualities, we find that we can divide the component parts of burned clay into two divisions, one division containing the weakening elements, the other the binding elements, the binding elements being the silicates of alumina, iron, soda, potash and magnesia, and the weakening element the free silica, with lime, sometimes in one division, sometimes in the other, depending upon the heat at which the burning was finished.
I do not mean to claim that a clay composed of pure silicate of alumina is necessarily stronger than one containing an excess of silica, but do claim that at the same heat and under the same conditions of firing, the first will be stronger than the second. In other words, silica, added to a pure clay, will weaken it; silica, added to an impure clay, may weaken or may strengthen. All depends upon the proportion of silica and fluxes in the original clay. If the clay already contains enough silica to satisfy and saturate the fluxes, additional silica will weaken; if not, it may strengthen, but further additions of silica beyond the point of the satisfaction of fluxes will weaken.
In illustration of the action of lime in pure clays, I would like to cite one experiment out of many. This particular experiment I consider very conclusive. I had been experimenting for some time on hard porcelain bodies without success, and thought I would go right back to first principles and investigate the action of the different fluxes. I made one mixture of clay, flint and felspar and another mixture of the same clay, flint and felspar, adding quite a percentage of lime in the form of paris white. Each mix was made into a number of small thin cups and placed in a small experimental kiln and burned. When I opened the kiln I found the felspar pieces quite well vitrified and translucent, the transmitted light quite yellow. The pieces with the lime in were porous, very slightly shrunken, and so soft as to be easily scratched with a knife point I decided to try the same pieces again at a little higher heat. I did so, drawing my trial pieces at regular intervals. I approached the proper heat, and all my pieces could be seen in the kiln when I drew trial piece. In about an hour I opened the kiln again to draw a trial, when I was very much surprised to notice that some of my little cups had disappeared. I thought it about time to discontinue further firing on that kiln. I did so, and fully expected to find next day, when I emptied kiln, that my felspar pieces had melted, but I was very much mistaken. The felspar pieces were all there, and in about the same condition as when first fired - a little more thoroughly vitrified, slightly more translucent and the transmitted light a little whiter. The lime pieces were little puddles of glass on the kiln bottom. I think this proves quite conclusively that lime changes suddenly from a resistant to a flux.
In regard to action of silica, will cite another experiment, also due to the hard porcelain experiments. I used for saggers in some of my experiments a regular body clay we were using. This clay was a mixture of a white burning ball clay, china clay, flint and felspar. These saggers, which, by the way, were little things, not larger than a teacup, did not stand my heat very well, and would get crooked, and sometimes blister, so I thought I would fix them by adding flint or silica. I took some slip of the mixture above, that saggers had been made of, and added, without weighing, sufficient silica to make this slip stiff enough for a potter to work as clay. I had some saggers made and used them to hold my little experimental cups. They were failures as saggers, but they gave me the key to my whole trouble with the porcelain. When subjected to porcelain heat these saggers vitrified throughout, glazed themselves beautifully, and were a beautiful clear white in color. From the fact of their glazing and warping, not a slight warp, but the sides doubled over until they almost touched the bottom, I conclude that had the heat been continued long enough they would have melted. The clay, without silica added, never showed any signs of melting. Both these facts— that is, in regard to lime and silica - are too well known in glaze making to need comment. All glaze makers know that lime added to a low heat glaze makes it more refractory, and lime added to a high heat glaze makes it less refractory. Silica added to glaze sometimes makes it more and sometimes less refractory.
Some years ago I came across the statement, vouched for by an eminent scientist, whose name I have forgotten, that all solid silicates produced by fire expand an equal amount between zero and their melting point, and therefore, given the rate of expansion, the temperature at which fusion will take place can be calculated. It is this law which I consider is the law governing crazing and upon which my theories are founded.
Whether this law is absolutely correct or not through all possible combinations makes little difference for the purposes of this article, so long as it be approximately correct. If the law be true, a silicate which melts at 1,500 degrees will expand on being heated and contract on being cooled just twice as much per degree as a silicate which melts at 3,000 degrees.