Did you know that Pyrex glassware used in chemistry labs is different than Pyrex glassware used in kitchens? Pyrex glass used in chemistry experiments is made of borosilicate glass, whereas the Pyrex used when baking is made of soda lime glass. What’s the difference? Borosilicate glass is resistant to thermal shock, but soda lime glass is not. The video below dramatically shows the effect of thermal shock in a measuring cup made of Pyrex soda lime glass.
This difference can be important! How many times have you rinsed a very hot beaker with room temperature water when working in the lab? I have done so many times, knowing that borosilicate lab glassware is capable of withstanding large differences in temperature without cracking. I had always assumed that Pyrex glass was synonymous with borosilicate, thermal-shock resistant glass. This was indeed the case until the late 1990’s when Corning, Inc. sold rights to the Pyrex name to World Kitchens. After the sale, Corning continued to sell borosilicate glass for laboratory use under the Pyrex name. However, World Kitchens began to sell soda lime glass for kitchen use – also under the Pyrex name. World Kitchens made this change because soda lime glass is cheaper to manufacture and more resistant to breakage from mechanical stress than borosilicate glass. Imagine my surprise one day in the early 2000’s when I decided it would be safe to boil water in a Pyrex measuring cup directly on the stovetop...the measuring cup cracked, sending broken glass and hot water all over the stove and floor! In fact, reports of “exploding glassware” began to crop up as other folks began to unwittingly expose World Kitchens’ soda lime glass to extreme temperature differentials while working in the kitchen. Now Pyrex is simply a brand name that has nothing to do with the chemistry of the glassware being sold.
The difference in heat shock resistance between borosilicate glass and soda lime glass can be quantified using some simple relationships. Borosilicate glass has a low coefficient of thermal expansion (3 x 10-6 K-1), which means that borosilicate glass does not undergo sizeable expansion upon heating or contraction upon cooling. On the other hand, soda lime glass has a high coefficient of thermal expansion (9 x 10-6 K-1), which means that it will undergo sizeable expansion upon heating and contraction upon cooling. The following equation provides a simple quantitative relationship between thermal shock and coefficient of thermal expansion, . The linear elastic thermal stress, , on a glass is a measure of the stress the glass can withstand without shattering:
In Equation 1, is the temperature differential a glass object can experience without breaking, is the coefficient of thermal expansion of the glass and E is the elastic modulus of the glass. Most glasses – including both borosilicate and soda lime glasses – are capable of withstanding stresses of about 5000 psi. Borosilicate and soda lime glass have a similar elastic modulus: 9.1 x 106 psi for borosilicate glass and 10.2 x 106 psi for soda lime glass. Using the appropriate values for each glass type in Equation 1, we see that borosilicate glass can experience a 183 K difference in temperature without shattering. However, soda lime glass can only experience a 54 K difference in temperature before shattering. Thus, you can safely pour boiling water (373 K) in an ice-cold beaker (273 K) of borosilicate glass without breaking the beaker (= 100 K < 183 K). Try the same thing with a measuring cup of soda lime glass, and the measuring cup will likely break, because in this case = 100 K > 54 K. I wish someone would have told me all this a few years ago…
Thanks go to Kristen Lewis and Nathan Ford for assistance in filming and carrying out the experiment of pouring room temperature water on a very hot measuring cup.
To learn more about this topic, check out the following:
1. R.C. Brandt and R.I. Martens, “Shattering Glass Cookware” http://www3.nd.edu/~rroeder/ame60646/slides/glasscookware.pdf
2. Doris and Kenneth Kolb, “Glass – Sand + Imagination” http://pubs.acs.org/doi/abs/10.1021/ed077p812