[Donlupedron]

Probe looks at possible problems with handling epoxy
By Scott Allen and Sean P. Murphy, Globe Staff | July 14, 2006

Investigators are focusing on the possibility that the failure of epoxy caused the ceiling collapse in the Interstate 90 connector tunnel Monday night.

Glue on bolts removed from the tunnel roof near the accident site was brittle and cracked, instead of having the consistency of smooth glass it should have had, according to a source briefed on the state investigation but not authorized to speak publicly. The decayed appearance may be a sign of premature aging.

Investigators are looking at problems Big Dig crews might have had in handling the epoxy, which requires a precise series of steps to get maximum holding power, according to the source.

The industrial-strength epoxy that was supposed to hold up the concrete ceiling is no ordinary glue. Used properly, it's actually stronger than concrete by some measures, and it's widely used in construction all over the world.

Typically, the epoxy comes in two cylinders, one containing resin and the other hardener, that must be mixed to create the glue on the spot. Workers installing the ceiling in the connector tunnel would have drilled holes into the concrete roof of the tunnel, squirted the epoxy ingredients inside, and then inserted a bolt. Once the epoxy had hardened and workers had tested it, they could suspend steel hangers from the bolt assembly, creating the frame that holds up the tunnel's drop ceiling.

In the past, ceiling bolts in the connector and in other Big Dig tunnels have failed safety tests because construction workers failed to mix the two epoxy ingredients correctly, didn't let it harden long enough, or failed to properly clean the surfaces being glued together, according to state reports.

In addition, moisture on the bolts or in drill holes could have weakened the epoxy, engineers said. The connector tunnel had water leaks in the roof in 2000, when the contractors were paid an additional $10,000 for cleanup, according to Big Dig documents. However, it is not clear whether the 2000 leaks were in the area of the accident.

``It was either a progressive decline in the strength of the epoxy or more stress was progressively put on the bolt fixture assembly than planned for," said Jeffrey C. Evans, head of the department of engineering at Bucknell University in Pennsylvania.

If manufactured and applied properly, industrial epoxies are strong enough to bear heavy loads for generations. A 1996 study by engineers at the University of Johannesburg in South Africa concluded that epoxy can hold a metal rod several times better than concrete. An official at Hilti USA of Oklahoma, a leading maker of industrial adhesives, said some adhesive anchors are designed to hold more than 100,000 pounds each, a far greater load than the 5,000- to 6,000-pound concrete panels in the connector ceiling.

The bolt-and-epoxy system for fastening the connector ceiling relied heavily on the strength of the glue because the bolt isn't screwed tightly into the concrete. The hole is deliberately larger than the bolt to allow room for the epoxy, which adheres to the concrete and the threads of the bolt.

The specifications for the kind of epoxy used in the tunnel ceiling would require a worker to use a masonry drill to leave a rough surface in the hole for better adherence, and the bolt would need to be free of grease. The worker would also need to clean out dust from the drill hole with a wire brush and high-powered air hose, acording to Paul J. Malinaric, a semi-retired physicist in Groton who worked for numerous industrial companies.

``It's a very exact protocol that has to be followed," he said.

In the past, crews constructing the drop ceiling in the nearby Ted Williams Tunnel sometimes fell short in following protocols, resulting in bolts that didn't hold strongly. In 1994, eight out of 50 bolts failed strength tests because of ``improper mixing of the epoxy," according to a 1998 report by the Massachusetts Inspector General.

Later in 1994, five out of 58 bolts in the Ted Williams Tunnel tested failed because crews didn't allow the epoxy to harden long enough, didn't use enough of the material, and didn't properly clean the drill holes.

As the ceiling was being installed in the connector tunnel, at least five bolts failed pull tests in 1999 to see if they were securely fastened, according to the attorney general's office. Investigators looking through construction logs this week found no indication the drill holes had been cleaned out by air hoses as required, and the source briefed on the state investigation said there is evidence workers improperly used water as a cleaner.

Gareth Cook of the Globe staff contributed to this report.