In this week's editorial a recent and very well publicized
possible adhesive failure, Boston's Big Dig Project, is examined. Potential
reasons for failures are identified. As with most accidents there may
be a multiplicity of events that contributed to the problem.
On July 10, 2006, a motorist driving through Boston's Ted Williams tunnel
(known as the "Big Dig Project") was crushed and killed by several
tons of falling concrete ceiling tiles. The epoxy-and-bolt system used
to fasten the concrete slaps overhead failed, although the technique is
used extensively in construction applications. An investigation into the
cause of the failure is ongoing, and it is premature to draw conclusions
about this tragedy. However, published reports have indicated that there
are several suspicious areas that will be studied. It seems that most
serious accidents are the cause of multiple unexpected occurrences rather
than one event, and this case possibly falls into this class.
More than 1,100 of the tunnel's bolt support assemblies were installed
with epoxy. It appears that the $14.6 billion USD Big Dig project used
a limited number of bolts and as a result there was a very low margin
of safety. It is suspected that one bolt failure may have resulted in
additional stress on the remaining bolts that then precipitated other
failures in a chain reaction.
Some bolts from the ceiling wreckage have shown indications of very little
adhesive being applied. Thus, improper installation or errors in the mixing
of the epoxy before application is also being investigated, and this could
vindicate the tunnel's design. Investigators have subpoenaed construction
documents to determine whether contractors dismissed installation protocol.
Management is also being suggested as a culprit. The chief complaints
are1 that the state government lacked adequate supervision of private
contractors and2 that the same company was involved in both the design
and construction efforts - an arrangement that some observers say may
have complicated the oversight. Several ceiling bolts apparently failed
during construction of the tunnel in 1999, and although this was reported,
action may have been insufficient to identify and solve the problem.
Also, Big Dig managers and designers eliminated half the ceiling bolts
called for in the original ceiling design to save time. Ceiling bolts
in the area of the collapse were safety-tested at the time of construction
with a weight now regarded as too low, potentially allowing defective
bolts to pass.
Several months ago it us believed that contractors may have used the
wrong adhesive to install at least some bolts. Invoices showed that the
construction company received and apparently used at least one case of
fast setting epoxy to secure the ceiling bolts to the tunnel roof rather
than the standard epoxy, which the ceiling designers had specified. The
"fast-set" epoxy holds 25 percent less weight than standard
epoxy and is not recommended for suspending heavy objects overhead.
And then on July 10 of this year, Federal investigators apparently singled
out another possible cause - creep of the epoxy. The National Transportation
Safety Board indicated that the collapse could have possibly been avoided
if designers and construction crews had considered that the epoxy support
anchors when under a load of 4600 pounds each could slowly pull away from
After the accident, the tunnel ceiling was permanently removed in the
area of the accident, and elsewhere in the tunnel the ceiling was reinforced
with additional bolts and brackets. Investigations into the cause are
still underway and potential legal actions are being considered. However,
the multiplicity of possible causes for the Big Dig failure is indicative
of most adhesive failure settings. There are many reasons why adhesive
bonds fail. There are generally four factors that contribute to failure
of an adhesive bond.
- Selection of materials
- Joint design
- Processing (e.g., mixing, application)
- Service conditions
Close examination of failed adhesive joints can usually lead to an explanation
of why the bond failure occurred. The investigator's "tool box"
for determining the cause of adhesive failures generally includes both
inexpensive methods as well as state-of-the-art instrumentation. But,
perhaps, the most important tools to use in the investigation are simple
common sense and a predetermined, coordinated approach. The current article
in Omenxus4Adhesives, "The Simple Approach to Nondestructive Testing
and Failure Analysis", describes several inexpensive techniques that
can be used early to determine the cause of failure.
When a bond failure does occur, a wide variety of processes can be used
in a failure analysis investigation. Not all of these, however, are always
appropriate. Some are not relevant to a particular failure, and time and
money are associated with each element of the analysis. Probing questions
need to be asked to efficiently plan and develop the methodology required
to analyze the failure.
- "Bolt Failure at Big Dig: An Anomaly?", Christian Science
Monitor, July 20, 2006.
- "Big Dig Job May Have Used Wrong Epoxy", The Boston Globe,
May 3, 2007.
- "Epoxy Creep Blamed in Big Dig Death", Associated Press,
AP Online, July 11, 2007.
Should you have any comments or feedback, please contact
Edward M. Petrie