History of Synthetic Gemstones




Probably the first experiment that indicated clearly the possibility of preparing gemstone synthetically occurred in 1819. Dr. E.D.Clarke, experimenting with the newly discovered blowpipe, placed two rubies on charcoal and melted them into one bead. However, at that early date did not occur to him that the resulting bead might be crystalline; he assumed it was glass. In 1837, Gaudin, the French chemist, reported the successful crystallization of alumina. He evaporated a saturated solution of alum and potassium sulphate containing a trace of potassium dichromate. Gaudin used the oxy-hydrogen blowpipe for melting the residue to drive off sulphur-dioxide gas and leave the alumina. The small amount of potassium dichromate imparted the color to the alumina, which crystallized as it cooled. Gaudin had come to the erroneous conclusion that he had produced a ruby, glass by the solidification of the molten material, and that the cracking, which he saw represented the devitrification of the glass rather than ruby synthesis. Later experimenters used different methods to crystallize the alumling. In every-case in which rubies were produced, they were too small, thin and laminated to be of value as gemstones.

The name of de Senarmont (1850) is remembered in connection with this early experimental work, but little definite information regarding its success can be obtained. His general method, used also by Elsner, Sainte-Claire Devile, Caron and de Bray, was to form a molten mixture of salts in which alumina was soluble. Ebelman, in 1851, followed the earlier methods in producing small white sapphires by fusing alumina in boric acid. In 1877, Fremy and Hautefeuille crystallized aluminum Oxide,
but of a molten bath in sizes large enough to yield small cut stones. First, they fused aluminum oxide and lead oxide, forming a lead aluminate; this was kept molten for a long period in a fire-clay crucible. The silica in the crucible combined with the lead to form lead silicate, and the alumina was liberated to crystallize as white sapphire. By adding a chromium compound, small rubies were obtained.

Fremy, working now with Verneuil crystallized synthetic ruby by a red heat reaction between barium fluoride and alumina containing a small amount of chromium. This process produced rhombohedral ruby crystals weighing about one-third caret. Although stones cut from these small crystals were tiny apparently they were the first synthetics to be set in jewelry. They used several methods, some utilizing sand crucibles of alumina and potassium carbonate mixed with fine charcoal and packed around a core composed of alumina and calcium fluoride. This was heated gently for an eight-day period. Shortly after 1900, the brilliant Verneuil carried the work to a triumphant conclusion, making rubies of many carats and of great beauty and perfection. A careful search of the literature shows that Verneuil had actually begun work on the fusion growth of ruby sixteen years before his first public announcement in 1902, and had solved all the essential problems in six years. We do not know the reason for this long delay in publication. It appears to have been the only time in his career that Verneuil used sealed notes to establish priority of invention. His flame fusion process is described in detail in the next section of this assignment. Unfortunately, Fremy, the senior investigator, died before his young collaborator's brilliant success was achieved.

Although colorless sapphire and ruby were made readily by Verneuil, blue sapphires did not yield to this method of production immediately. Unlike genuine ruby, the color of a sapphire does not return after cooling. Various experiments were made with a lower oxide of chromium to impart a blue color, but without success. Cobalt oxide, which is used in the manufacture of enameled porcelain, was wholly unsuccessful, for it was not retained in the sapphire. Calcium oxide also was used and produced stones, but the result was not fully satisfactory. For a time, Verneuil seemed to have been discouraged; however, he returned to the problem in 1909, in collaboration with I.H. Levin and with the financial assistance of A.A.Heller and the Societe Hellerite. Previous to the synthesizing of sapphire, iron oxide was the only coloring agent known to be present in natural sapphire. A series of experiments resulted in the discovery that titanium oxide was also necessary to produce a color resembling that of natural blue sapphire. Verneuil believed that ferrous iron was produced in the reducing part of the flame, and that this was subsequently oxidized to ferric iron by the titanium which accordingly achieved a lower valence state which produced the color. Interestingly enough it is only very recently that the mechanism producing this color has been fully explained. The titanium is largely lost in the process, but a blue color is imparted to the alumina.

Now that gem-mineral synthesis in general and ruby in particular have been put into proper perspective, it is appropriate to consider the use and meaning of the term "reconstructed." Reconstructed is a term that was long associated with the early synthesis of ruby and often used in the jewelry industry. The definition of reconstructed was in effect the sintering of fragments of natural ruby to form a mass from which gemstones could be cut. In other words, the fragments of natural ruby were melted at the edges and pressed together to form a larger mass from which a reproduction of natural ruby could be made. It seems apparent in the light of present knowledge that this never occurred. No such stone has ever been reported. A second thought along these lines was that perhaps the fragments of natural ruby were melted and then fuse to form the reconstructed stone. Efforts on the part of Kurt Nassau, Ph.D. of Bell Telephone Laboratories and Robert Crowning shield of the GLA staff showed this to be an impossibility.

The early so called Geneva rubies first appeared in Europe in the 1880's. In 1882 a great number of the so called Geneva rubies appeared on the market, however, information available in the publications so that era indicated that the scientists were aware that they were characterized by the presence of spherical gas bubbles, an inclusion never encountered in natural rubies. The discovery that they were manufactured created consternation throughout the gem trade, perhaps more than any more recent development, because rubies were very popular and this was the first substitute that duplicated the properties of the natural stone. They were said to have been made by a priest in the Swiss village. Later, the name Wyse was mentioned; but whether the priest and Wyse were the same person is not clear.

In 1890, synthetic rubies, apparently from another source, began to appear on the market. They, too, were called reconstructed at the time. These were supposedly the product of a Swiss engineer or chemist named Michaud who claimed to be making them from small chips of genuine rubies. It seems quite likely that the first developed in 1902 had actually been made many years before that date, because there is no present method of distinguishing surely between early Verneuil flame-fusion synthetic rubies and the early so called reconstructed rubies that have been encountered since that time. Prices for the so called reconstructed stones fall very rapidly. The price, which began at $200 to $320 a carat , decreased until in 1904 it was only $100 to $250 a carat. Since that time, the rough boules, of course, have been available at much lower prices than that.

The work of Dr. Nassau at Bell Telephone Laboratories in New Jersey has shown that when natural ruby fragments are heated to a high temperature they lose their color and the resulting mass of material has a gray, almost opaque appearance. Since nothing with a character suggesting that fragments were melted on the edge and pressed together has ever appeared to our knowledge, we believe that reconstructed rubies never were made and that the early products of the flame-fusion process were called reconstructed. Thus, we can see no basis for the use of the term reconstructed as applied to ruby.