A painstaking pursuit of better glass
For a fraction of a second following a bomb blast, window panes hundreds of feet away become sails in the blast’s pressure wave. Then they shatter into thousands of flying, razor-sharp daggers.
That’s why shattered glass is often among the most dangerous hazards in a terrorist bombing. During the 1995 Oklahoma City federal building attack, for instance, 200 injured people at locations other than the Murrah building at the time of the blast reported they were hurt by falling or flying glass.
A team of Sandia ceramics experts now is studying glass in a quest to develop windows that are stronger, that survive and fail when they’re supposed to, and that shatter into smaller pieces, like grains of sand rather than shards, when they break.
They also want to develop a standard set of methodologies to test samples of many different glass types and configurations so architects and builders have more information at hand when they select glass for a building.
The project is supported by Sandia’s Architectural Surety® program and funded through the Laboratory Directed Research and Development (LDRD) program, with past support from the US Air Force Blast Mitigation Action Group and Halliburton Energy Services, a Labs industrial partner.
"Glass is everywhere, and until now looks and energy efficiency have been everything," says project leader Jill Glass (1843). "Few building designers have considered the security and safety advantages better glass might offer."
Building Glass 101
Glass is a brittle material, she says, and its strength is highly sensitive to surface defects that often are too small to see. Once stress on the glass exceeds that required to activate a flaw, the pane fails, usually at the spot of the most severe defect.
"It’s like when your windshield has a tiny chip or crack in it — one day it’s fine, then the next day it goes zing," she says.
Every piece of glass is different, she adds. Its age, the way it is framed into a building, its position relative to the sun, or microscopic dings from wind-borne particles can significantly affect its strength.
That’s why building glass fails over a wide range of stresses.
A better understanding of glass behavior under dynamic loading conditions (sudden pressures, like those from a blast), and an ability to modify the fracture properties of glass, will ultimately bring stronger and safer windows, she says.
What we don’t know
In the lab, Jill and her team have subjected various glass types to dynamic and static (constant or slowly increasing pressures) loads to characterize their strengths and behaviors in blast and other environments.
They’ve studied when and where cracks begin, how large flaws must be to reduce a pane’s strength, what forces are at work within a pane when it fails, what size fragments are created when panes shatter, and how far and fast they fly in varying overpressure environments.
They’re looking into how water affects glass strength.
They’re examining behaviors of several variations of glass compositions, thicknesses, configurations, and engineered features, including glasses manufactured with strength-enhancing stresses stored within the material, such as automobile safety glass.
"What we’ve discovered is that there is a lot we don’t know about glass behaviors in a blast environment," she says. "Architects are making modifications and decisions about glasses based on little or no supporting scientific evidence that they are safety improvements." (See "Rushing headlong into the glass problem" right.)
An oxymoron, reliable glass
One needed improvement, says Jill, is glass that survives and fails very reliably.
"Reliable and glass haven’t often been used in the same sentence," she says.
Toward that end her team and researchers at Penn State University and the Universita’ di Trento in Italy are experimenting with a specialized family of glasses called Engineered Stress Profile (ESP) glass that offers high strength, reliability, and control of fragment size through modifications to the glass following production.
The team is refining the two-step ion exchange process used to make ESP glass to carefully manipulate the glass composition at the surface. The process creates glass with peak compressive stresses 10-20 microns under the surface rather than right at the surface, like automobile safety glass.
(A pane of auto safety glass incorporates compressive stresses at the pane’s surface that are counteracted by tensile stresses at the pane’s center. A surface crack or bending strain destabilizes these competing stresses, and the pane fails catastrophically. These tensile stresses cause the "dicing" behavior associated with the fracture of safety glass.)
This new processing trick results in a stress profile within the ESP glass that initially arrests penetration and growth of networks of surface cracks, then releases the stored compressive and tensile stresses when a critical stress value is reached.
The result is crack tolerance and failure at predictable, narrowly defined stress loads. (Safety glass offers control of fragment size but is highly sensitive to surface flaws and, thus, fails over a wide range of stress loads.) In addition, the surface crack density increases dramatically prior to failure, serving as a warning mechanism, says Jill.
Sandia, Penn State, and Halliburton have filed several patents on the processes to produce modified ESP glass and its applications.
"For the first time we have glass types that are both strong and dependable, that crack noncatastrophically, and that fracture into small fragments," she says.
Scientifically valid comparisons
Currently no standards exist for selection of blast-resistant glasses for nongovernment buildings, says Jill.
She has worked with the US Air Force’s Force Protection Battle Lab in San Antonio to compare the blast performances of large panels of ESP glass with those of conventional glasses under varying overpressure conditions.
The tests were conducted in a shock tube, providing carefully controlled test environments. Free-field blast tests at Sandia’s Explosives Components Facility are planned for this summer using smaller-scale glass samples.
"Even under nominally similar conditions, glasses perform differently from one day to the next," she says. "The goal is to come up with a set of protocols that allow engineers, manufacturers, researchers, and end users to make scientifically valid comparisons under rigidly controlled, standardized conditions."
"I believe the research being accomplished here and at Penn State with the ESP glass will revolutionize the application of glass as a structural material for buildings and infrastructures," says Rudy Matalucci (5862), Sandia Architectural Surety® program manager and chairman of the Engineered Glass Committee of the American Society of Civil Engineers’ Architectural Engineering Institute.
"I find a new excitement within the industry for using glass differently as a construction material now that the internal stress profile can be altered to meet a variety of strength requirements," he adds. "Many of the architects and engineers I talk with daily are encouraged to know that Sandia is investigating the basic performance of ESP glass."
And building glass is just one market for better glass, adds Jill.
"There are many other uses, including weapons and automobile applications, that would benefit from a glass you can trust," she says.