Studies in Ancient Glasses and Glassmaking Processes. Part IV. The Chemical Composition of Ancient Glasses
JSGT 1956 V40 T162-T186
Chemical analyses, qualitative and quantitative, of ancient glasses carried out during the past 150 years have been assembled and critically examined. The specimens analysed range over a period of about 3300 years and had their origins in Egypt (from 1500 B.C. to A.D. 1400), Babylonia, Assyria, Crete, China, the Roman Empire (mainly German sites), India, and Mesopotamia; and in Germany, France, Russia, and England in medieval and Renaissance times. The results of modern spectrographic analyses, though much less helpful than has been claimed, have also been used for reference. The ancient glasses are characterised by complexity of composition, ten to twelve ascertained constituents being present in many eighteenth Dynasty Egyptian and Babylonian and Assyrian glasses, and nineteen in a blue glass of the fifteenth century A.D. from the church of St. Quentin, France. In about 300 chemical analyses, the glasses which contain more than 0·5% PbO constitute a little more than one-seventh of the total, and even this fraction most probably exaggerates greatly the proportion of lead glasses among the glass products of the 3300 years under review. The complete spectrographic analyses known include such a large number of those of lead (and barium) pre-Han and Han Chinese glass beads and other mainly small objects that an assessment of the relative general occurrence of non-lead and lead glasses cannot properly be made. (See bibliography.) The prevailing types of ancient glasses contain silica, alumina, iron oxide, lime, magnesia, soda, and potash. With these are usually associated P2O5, SO3, and Cl, and, in coloured glasses, one or more of the oxides Mn2O3, CuO, Cu2O, or CoO. The CaO content frequently reached 7 to 12 per cent in eighteenth Dynasty Egyptian and Babylonian and Assyrian glasses, and MgO 2 to 5%. In the medieval period in Western Europe (England, France, Germany), when the alkaline salts for glassmaking were derived from the ashes of beechwood or oak, the lime content of some window glasses exceeded 20% and the magnesia more than 5 per cent. During this same medieval period the dominant alkali was K2O; but outside Western Europe, and in all other periods everywhere, soda dominated potash in the ratio seldom less than three to one and in some cases exceeding ten to one. Totals for the two alkalis range from below 10% to about 30%. The compositions of some of the eighteenth Dynasty Egyptian, Assyrian, and other glasses would entitle them on modern standards to be regarded as of reasonable durability. Out of 180 analyses used as basis, only in 40 did the Fe2O3 content fall below 0·5%, and only two or three glasses would reach the modern requirement for reasonable freedom from colour. The use of iron oxide to make very dark or black glasses is seen in examples of second-first centuries B.C. and later dates quoted 1n the tables in the text. Manganese appears in the majority of the analyses, even in glasses from the earliest times, two from Nippur, dated 250 B.C. containing as much as 5·29% and 4·37% respectively, and examples from Egypt of 1500 B.c. exceeding 2 per cent. The chief use of the oxide was as a colouring agent. Blue glasses in various shades represent the earliest colours employed and copper oxide is present in varying amounts from a small fraction of 1% up to two or more percent. Cu2O in amounts up to as much as 12% is a constituent of the sealing-wax red glasses. Cobalt oxide is less frequently employed than copper. It is present, however, in occasional examples over the whole 3000 year period. Geilmann has recently found that nickel oxide may be associated with cobalt, varying amounts from 0·03 to 0·13% being found in coloured French window glasses. Antimony oxide is present in a small number of the specimens ranging from 1400 B.C. onwards; but arsenic is entirely absent from most ancient glasses; only in four cases in which it was detected did the amount exceed 0·01%.A bibliography of references to analyses of ancient glasses is appended.
W. E. S. Turner
