In recent years concern has grown regarding nickel sulfide inclusions and related tempered glass breakage. The following discussion is meant to clarify the cause of nickel sulfide inclusions in glass and explain how these inclusions may cause tempered glass to break. Since these inclusions cannot be eliminated, the discussion addresses how to reduce the potential of nickel sulfide inclusion breakage through an alternative heat treating method and through statistical heat soaking. Methodology of the heat soaking program has been outlined in detail to provide the reader with sufficient information to evaluate the heat soaking program. There are occasions, however, when tempered glass is the only appropriate product.

In the manufacture of glass, there are approximately 50 identified chemical types present. Undissolved inclusions may be incorporated within the float glass. These are broadly categorized as "knots, dirt and stones ". Most of these are considered imperfections and do not influence the performance of the glass. The allowable size and minimum spacing of inclusions are defined in ASTM C 1036 for Flat Glass.

In cases where nickel-rich contaminants, such as stainless steel, are in the glass batch materials they may combine with sulphur to form nickel sulfide inclusions. Extremely minute powders or grains of nickel-rich materials will create these inclusions. They are impossible to totally eliminate and, while all glass has some present, are not considered a product defect.

Viracon tempers glass purchased from a number of base glass manufacturers. There is no known method by which Viracon can inspect the glass for nickel-sulfide inclusions or control their number. Glass producers use extensive batch control procedures to minimize the size and number of inclusion and Viracon must rely on the base glass manufacturers to control their process. As indicated above, this control may not totally eliminate the inclusions.

When glass is heat-treated, the nickel sulfide inclusions are modified into a form that grows or transforms with time and temperature. Once glass is installed and the nickel sulfide inclusions are solar heated, small cracks may develop from the inclusion. If these cracks penetrate the tension layer of fully tempered glass, the resulting release of energy will cause the glass to spontaneously break.

Viracons experience as a supplier of both tempered and heat strengthened glass has shown that heat strengthened glass is far less susceptible to spontaneous breakage due to nickel sulfide inclusions. Heat strengthened glass which has high surface compression levels may not be desirable, and limiting the surface compression range appears to be significantly important. Current data suggests that surface compression levels in the range of 3,500 psi to 7,500 psi may eliminate any potential for nickel sulfide related breakage.

For this reason, Viracon recommends heat strengthened glass for those applications which require additional thermal resistance and strength. Viracon's production specifications for heat strengthened glass limit surface compression levels to a range of 5,500 psi, ± 1500 psi. However, heat strengthened glass is not a substitute for tempered glass in those applica-tions requiring a safety glazing material. When considering the potential for spontaneous breakage, it is an alternative glazing material to fully tempered glass for spandrel applications, curtain walls and skylight glazing when considering the potential for spontaneous breakage.

The responsible specifier must weigh all considerations and decide whether the heat soaking process is appropriate. Viracon warrants that the sampling and processing is statistically valid. No warranty is made regarding the incidence of in-place breakage from inclusions of any type.

The level of sampling selected by Viracon results in a 95% probability that breakage of installed glass from nickel sulfide will not exceed 0.1%. The sampling is based on a theoretical random distribution of inclusions within the glass. Inclusions tend to occur in clusters over a short period of time and a run of glass that has a high incidence of inclusions would be discovered in the Viracon heat soaking process.

Sample sizes from the population of tempered glass are determined by the size of a lot and will vary. In cases where the testing results in any glass breakage, the number of panes tested is increased again. If there is a high incidence of inclusions, 100% of the tempered glass for a building may be tested. The net result, in all cases, is that the probability of breakage is less than indicated by the values above.