NEW DEVELOPMENTS IN GLASS CONTAINERS* By WILLIAM R. PRINDLE Continental Can Compan, Inc., HazeL?It/as Glass Division, l•heeling, l •. l•a. GLASS tS AN interesting material to work with and to talk about because it has many unusual, even unique, characteristics. Glass is an ancient material (it has been called the first synthetic material) yet the composition of commercial glasses today are not vastly different from those used by the old Egyptians, Syrians and Romans. While you might inter- pret this as laziness and lack of imagination on the part of the glass in- dustry, it is really the result of economics and the laws of physical chemis- try. The raw materials that were used by the Phoenicians are still the cheapest and best to use. The same proportions of the raw materials still melt readily to form a viscous liquid that resists crystallization as it cools. The properties of glass are also rather unique. It is completely isotro- pic: its structure imparts no directional effects to its physical properties. The refractive index, the electrical conductivity, the strength, the thermal expansion and thermal conductivity are the same in all directions in the glass regardless of how the glass is manipulated or shaped. Glasses conduct electricity freely at high temperatures, yet are so com- pletely nonconducting at room temperature as to be used in making in- sulators. Glasses exhibit tremendous changes in viscosity with tempera- ture the viscosity of glass at 500øC. may be l•000,000,000,000 times as great as it is at 1300øC. Glass has no melting point or freezing point but exhibits a continuous change in all properties from room temperature up to very high tempera- tures. Glass fibers have been tested with breaking strengths of one million pounds per square inch, or much stronger than the strongest steels. Needless to say, the properties of glass have aroused the curiosity of clmmists and physicists and a considerable amount of fundamental re- search has been carried out on glass, particularly in recent years. This research has been carried out in government laboratories, at universities, at research institutes and by industry. Hazel-Atlas, Corning Glass Works, * Presented at the December 13, 1956, Meeting, New York City. 86

NEW DEVELOPMENTS IN GLASS CONTAINERS 87 Owens-Illinois, Bell Telephone Laboratories, Pittsburgh Plate Glass and General Electric are some of the companies that have contributed. Glass is a fine material in many ways it is inexpensive, odorless, tasteless, re- sistant to corrosion, impermeable to vapors, transparent and is capable of being formed in many shapes, etc., but it has the one disadvantage of not being as strong in service as we would like it to be. As a result much of the recent research on glass has been on strength. Just this last August the American Association for the Advancement of Science sponsored a one week Gordon Research Conference on glass strength to disseminate the latest ideas and to generate new ones. This concern about strength may sound strange in view of the comment above that strengths of one million pounds per square inch have been achieved. The extremely high strengths, however, have been achieved in the laboratory under special, carefully controlled conditions. In practical use the strength of glass is more of the order of 10,000 lb. per sq. in. The reason for the discrepancy between the high strength values obtained in the laboratory and the lower values observed in practice lies in the fact that glass is susceptible to stress concentration. Engineers call this "notch sensitivity." Glass is inherently very strong, but it may be seriously weakened by surface flaws, such as chips, pits, scratches and abrasions. These abrasions are the points where glass starts to break. Just as a notch cut in the edge of a piece of cloth before tearing it in two enables one to start the tear more easily, the scratches on glass act as notches that make it easier to break. Because glass does not flow plastically like most metals do, the stress at the head of a surface crack may build up to a very high value under load and cause the glass to fracture. There are various ways of dealing with these surface defects. One solution is to remove surface damage by fire polishing in which an intense flame played on the surface of the glass softens the glass and heals the flaws. Another method of flaw removal utilizes a hydrofluoric acid polishing solution to round off the defects so that they are no longer serious stress concentrations. Nei- ther of these methods has been used commercially because they are com- plicated and expensive. Another solution was to protect the surface. As a result, the idea of coatings to protect the glass from damage has been developed in the last few years. Glass is very strong as manufactured and has a surface that is very nearly free of scratches. Therefore, if one could keep the surface from being damaged and avoid the formation of stress concentrations the original strength of the container could be preserved. Coatings act like lubricants so that objects which might scratch the surface slide over it without digging in. Many materials have been tried as bottle coatings, but they must meet three criteria to be generally useful and acceptable: