Skip to main content
Get your Wikispaces Classroom now:
the easiest way to manage your class.
Pages and Files
M Kevin Parfitt - Moderator
Building Failures Forum (Blog)
Bridge & Infrastructure
Building Failure Cases
Failures in the News
Fire Case Studies
Stadiums & Arenas
Systems & Failure Categories
Contribute a New Case
Comment on a Case
Journal of Performance
TCFE of ASCE
MatDL Failures Case Studies
Links - See Building Failures Forum
In Planning Stages:
Failures Photo Library
Penn State AE
About AE 537
Big Dig (I 90 Connector Tunnel)
Boston's Big Dig: 2006 Ceiling Panel Collapse and 2004 Slurry Wall Breach
Shaun M. Kreidel, Penn State, AE 537, 2010
The original Central Artery, Boston’s major interstate highway, later dubbed “the other Green Monster” (reference to the green monster of Fenway Park), was an elevated roadway that cut though the city, essentially creating a barrier between the downtown area and the waterfront. In addition to severing connection between the waterfront and downtown, this roadway displaced more than 20,000 residents and required the demolition of more than 1000 buildings. 
Due to the substantial increase in volume of motorized traffic in Boston over the years, the Central Artery soon exceeded its intended design vehicle capacity. At the time of its demolition, the daily capacity on the roadway was estimated as 170,000 vehicles, 100,000 more vehicles than designed for, which was a direct cause for an accident rate four times above the national average. Structurally, the elevated highway was compromised due to poor design construction. 
The idea to bury this stretch of highway and reengineer the traffic patterns was conceived in the 1980s in order to reconnect Boston with the harbor and to improve the traffic flow. Nearly three decades and 14.8 billion dollars later, the “Big Dig” would own claim as the most expensive highway project in the United States.
The project at completion consisted of 161 lane miles of highway along a 7 ½‑mile corridor, with 5 miles of tunnel, 6 interchanges, and 200 bridges .
Many construction achievements can be attributed to the big dig project, such as the use of jacking boxes, ground freezing, a massive underpinning endeavor of the old highway and the largest use of slurry wall construction in North America.
This wiki page; however, will concentrate its focus to the construction failures of the Big Dig including the cause of the 2004 slurry wall breach within the I-93 Tunnel, and the fatal ceiling panel collapse of 2006 in the I-90 Connector Tunnel.
Figure 1: Highway Map and Ceiling Panel Collapse Location (14)
I-90 Connector Tunnel Ceiling Collapse
Figure 2: Ceiling Panel Construction (13)
Because the scope of the Big Dig project relied heavily on the use of tunnels, the ventilation of fumes from automobile traffic became a major issue in design. The solution for the these tunnels came in the form of a suspended ceiling which created room for an exhaust duct between the ceiling and tunnel roof .
The connection between the steel turnbuckles and the tunnel roof of the I-90 Connector Tunnel was originally designed using cast in place channel inserts which were imbedded into the concrete. This method of construction was opted over an epoxy connection that was previously used in the ceiling construction of the Ted Williams Tunnel, which travels under Boston Harbor. The connections used in the Ted Williams Tunnel presented installation problems and a high failure rate. These failures were attributed to installation error and dampness within the tunnel and not related to a materials failure .
Due to improper placement of the cast in place channels within the I-90 Connector Tunnel, approximately 2,000 epoxy anchors, representing 10% of the total number of anchors in this section of tunnel, were used as a substitute. A 200 ft. section of the Connector Tunnel previously built beneath D Street, was not designed using the cast in place channel inserts as a connection to the tunnel roof, and solely relied on epoxy connecto
Figure 3: Section of Connector Tunnel (14)
rs for the drop ceiling construction. It was in this section of tunnel in which the fatal ceiling collapse occurred.
The concrete ceiling panels below D Street, installed by Modern Continental Construction Company, were suspended approximately 5.5 ft. from the tunnel roof by adjustable steel hangers as seen in Figures 2 and 3. The panels within the length of tunnel under D Street consisted of two rows of 12 ft. by 8 ft. by 4 in. concrete panels weighing 4,700 pounds each and one row of 6 ft. by 8 ft. by 4 in. concrete panels weighing 2,500 pounds each .
The ceiling module that experienced the failure consisted of 15 concrete panels, supported by 4 support beams positioned in the direction shown in the isometric depiction in Figure 4. Eight vertical hangers were suspended from the tunnel roof to support each beam. Diagonal hangers were present in various locations to prevent against horizontal movement. These support hangers were connected to the tunnel roof each by 8 in. stainles steel bolts held in place by an epoxy adhesive, shown in Figure 5. In total, 76 epoxy anchors would be used in this ceiling module.
The greatest load on any one anchor was determined by Gannet Flemming Engineers to be 2,600 pounds which was verified by a post-accident finite model analysis of the ceiling supports. Each anchor was specified and designed to support a service tensile load of 4,000 pounds . The epoxy was capable of resisting 6,350 pounds of force based on test data provided by Powers Fasteners, the epoxy supplier.
Figure 4: Isometric View of Ceiling Construction (14)
In 1999, a similar ceiling construction using epoxy imbedded anchors in the previously mentioned Ted Williams tunnel began to displace significantly. This deformation was thought to be an installation error rather than a materials failure .
In addition to the observed displacements of the ceiling bolts in the Ted Williams tunnel, epoxy anchor bolts within the I-90 Connector tunnel that were used as a substitute to the imbedded channel inserts also began to displace. In 2001 a failures report, filed by a Modern Construction Inspector indicated that:
Several anchors appear to be pulling away from the concrete. The
subject anchors were [previously] tested
to the revised value of 6350
lbs., all of which passed…. Reason for failure is unknown .
In response to this report, Modern Construction was instructed to remove and replace the compromised anchors. It was assumed that the anchor failure could be attributed to construction error. Even after the two major instances indicating anchor bolt displacement, a monitoring program was not instated.
Figure 5: Anchor Bolt Detail (13)
On July 10, 2006, approximately 26 tons of concrete ceiling panels and suspension hardware within the Interstate 90 Connector Tunnel below D Street, located in Figure 1, collapsed onto the roadway killing one and injuring another . After the ensuing investigation, it was determined that the reason for the collapse was directly linked to the inappropriate use of the epoxy anchor .
Epoxy is created from chemical reaction between an epoxy resin and a hardener. The crosslinking of these two components result in resistance of motion . The epoxy supplier, Power Fasteners Inc., offers both fast set and standard set products . Each of these products have a unique molecular arrangement causing them to perform differently.
The standard set epoxy is applicable for use in cases of long term tensile load . This product, however, was substituted with the fast set epoxy which possesses similar initial strength but displays very poor creep resistance over long term loading. In an 80 day laboratory test, presented in a NTSB (National Transportation Safety Board) meeting, it was verified that the creep levels were significantly higher in the long term loading of the fast set epoxy compared to that of the standard set. The fast set epoxy yielded large displacements at all loading levels while the standard set displayed minimal displacements throughout the testing .
Over a period of six years, the epoxy deformed and fractured causing the ceiling anchors to fail, ultimately resulting in a catastrophic collapse. F
Figure 6: Anchor Bolt Failure (13)
ollowing the failure and subsequent investigations, 1,100 of the epoxy anchors located in the I-90 Connector Tunnel yielded displacements similar in nature to the ones in Figure 6 . Within the section of tunnel beneath D Street following the accident, 161 of the 634 remaining epoxy anchors were found to have significant displacements . The ceiling panels have since been completely removed and not replaced within the D Street section of the Connector Tunnel. It was found that because of the relatively short length of the section and its proximity to the tunnel opening, ventilation was not necessary. In response to the epoxy failure, problematic anchors in other areas of the tunnels have been “doubled up” by mechanical undercut anchor, as part of a new, redundant support design .
Communication in this instance between the designer, supplier and project manager also proved to be fatal. Material limitations and applicability was not fully understood by all three parties. Complete understanding of the designed system would have also prevented this tragedy.
In a survey conducted by the FHWA (Federal Highway Administration) of tunnel finishes, epoxy anchors were found to have been rarely used in pure tension. When they do exist, these connections work in a system of lightweight panels. The survey also uncovered that most tunnels within the United States have continuous ceiling panels that extend into the tunnel walls, which act as a redundant system . Had the Connector Tunnel been designed with redundancy, the ceiling would not have collapsed upon failure of the anchors.
From NTSB investigations it was found that the inappropriate use of the epoxy could be linked to Gannett Fleming and
Bechtel/Parsons Brinckerhoff for not accounting for the creep in the design and specifications of the tunnel. The failure of Modern Continental Construction to continuously monitor the anchor bolts was determined as a contributing factor. It was also found that Powers Fasteners failed to supply the project with complete information regarding the suitability of the epoxy used .
In 2007 the Commonwealth accused Powers Fasteners of knowing the epoxy application in the tunnel and failing to differentiate the Fast-Set and Standard-Set epoxy to the project managers. It was found that Powers Fasteners had the knowledge and opportunity to prevent the ceiling panel collapse and was the only company charged with manslaughter .
A settlement was reached between Powers Fasteners and the victim's family for $6 million in 2007 . Additionally, $16 million has been paid by Powers to the state and city to settle the manslaughter charges .
I-93 Slurry Wall Breach
The design of the Big Dig tunnels highly relied on slurry wall construction which effectively allowed the city of Boston to dig itself up without shutting down . A similar technique had been used in the construction of the World Trade Center in New York. The Big Dig project, however, used the slurry wall technique on a much larger scale. It would own claim to the largest use of slurry construction in all of North America, creating a total of 26,980 linear feet of slurry wall .
In September 2004, after a year of tunnel operation, a breach occurred in one of the walls in one of the deepest areas in the I-93 tunnel, sending water gushing into and shutting down the tunnel and requiring authorities to initiate emergency procedures . This weak link in the slurry wall also marked the interface of two construction contracts awarded to Modern Continental Construction Co, Inc and the Framingham-based Perini/Kiewit/Cashman.
Figure 7: I-93 Tunnel Profile (Based On Figures From (4))
Slurry Wall Construction
Slurry wall construction relies on a viscous liquid usually containing bentonite clay that is pumped into an excavated area to support it before concrete is placed . These excavations were roughly 10 ft. long by 3.5 ft. wide, effectively forming a slurry wall panel. Steel H-piles were placed in the slurry at approximately a 5 ft. spacing, bearing on bedrock. Concrete was pumped in to replace the slurry, effectively creating composite walls that would transfer the shear created from the soil and water to the piles and bedrock . Figure 8 shows the process of the slurry wall construction.
Figure 8: Slurry Wall Construction (Based On Figures From (4))
Breach and Failure Description
In order to avoid the existing Red and Blue T-Subway Lines, the I-93 tunnel snakes its way vertically through the underground of Boston as seen in Figure 7. The varying elevation of the tunnel throughout its length creates a variety of loadings and soil conditions . The 2004 slurry wall breach occurred at the deepest part of this tunnel at an elevation of approximately 80 ft. below sea level.
After an investigation was conducted, it was found that construction errors and inadequate inspection during the slurry wall process occurred. As seen in Figure 8, the spaces between the flanges of the H-piles on the secondary panel side are closed off and filled with a granular material during construction of the primary panels. In the construction of the secondary piles, these end stops are removed and the contained material is removed, assuring a good contact between the steel pile and concrete .
In the case of the 2004 breach, it was found that an end stop and the contained fill of one of the secondary panels were not removed due to a concrete overflow of a primary panel. The failure to remove this end stop created a cold joint at the interface of the H-pile, preventing wall reinforcement from reaching the web of the H-pile, hindering the intended monolithic behavior of the panels. A cosmetic shotcrete patch covered the imperfections of the slurry wall created by construction errors . It was in this area that high pressure water found a path through a 12 in. by 12 in. hole and into the tunnel .
At the time of construction of the slurry wall panel, a field engineer had observed and recorded a failure in the construction of the panel; however, proper corrective measures were not taken. The 2004 slurry wall failure was completely attributed to construction error, not design flaws. Had the end stop been removed, and the reinforcing not been modified, this failure would not have occurred. Proper inspection follow up could have also prevented this breach.
To seal the breach, a reinforced steel plate was placed over the area in which the breach occurred to contain the hydrostatic forces. The steel plate has been reinforced with steel beams and encased in concrete.
The slurry wall breach has led to further investigations within the tunnel. Between October 2004 and March 2005, approximately 1,600 slurry wall panels have been inspected. Of these inspected panels, only one additional panel required major repair . Action has been taken against smaller leaks, using a grouting method discussed in Reference .
Overall Conclusions and Final Thoughts
Methods such as ground freezing, underpinning, and slurry wall construction were implemented on enormous scales. The overwhelming size of the project, and the amount of unknowns lead to an increase in cost from an estimated $2.6 billion to a total cost of $14.7 billion. Construction completion was also finished 8 years behind the initial completion date.
Time, cost, and project size contributed to the failures presented in this wiki page. Proper inspections and continuous monitoring of construction issues failed to occur that ultimately lead to both the ceiling collapse and the slurry wall breach.
Despite the construction failures, the Big Dig was wildly successful in rerouting and entire highway system below Boston without disrupting everyday city life at ground level. The Big Dig has reduced downtown rush hour travel time from 20 minutes to 3 minutes, and has effectively reconnected the downtown area with the waterfront. The positives that have stemmed from the Big Dig project will be felt by residents and visitors for generations to come; however, we must not forget and learn from the negatives that arose from the project as well.
1. Angelo, William (2007). “
Epoxy and Inspections Cited For Boston Tunnel Accident
.” ENR.com, July 2007.
This article describes and references the findings of the National Transportation Safety Board in their investigation of the Big Dig ceiling panel collapse.
2. Bechtel/Parsons Brinckerhoff (November 2005). “
Water Intrusion in the I-93 Tunnels: Causes and Cures
Bechtel/Parsons Brinckerhoff composed this document to describe the causes of the leaks in the I-93 tunnel and what is being done to suppress them. This document attempts to silence critics of the “Big Dig” project and defends the effectiveness of the waterproofing in the tunnels as compared to industry standards.
3. Central Artery/Tunnel Settlement Background (January 23, 2008). “
Background information to the Boston Central Artery/Tunnel Settlement with the Commonwealth of Massachusetts and the U.S. Attorney for Massachusetts
This document gives a detailed description of the failures associated with the “Big Dig” project. It presents the findings of the National Transportation Safety Board (NTSB) in regards to the ceiling panel collapse, and findings from investigations associated with the slurry wall breach in the I-93 tunnel and the numerous leaks that still riddle the tunnels today.
4. Christian, John T. (2007). “The Breach Problems in the Tunnels of the Boston Central Artery/ Tunnel Project.” Soils and Rocks, September-December 2007, pp. 163-167.
This article describes the large breach problem in the I-93 tunnel. It discusses the cause for the breach and the construction flaws that lead to the slurry wall failure.
5. Ezrin, Myer (2008). “
Boston’s Big Dig Fatal Epoxy Adhesive Failure
.” ANTEC 2008, pp. 2326-2329.
This article discusses the failure of the fatal ceiling panel failure in the I-90 connector tunnel. The epoxy used and overall design of the panel connection is examined. Alternatives to the heavy concrete panel design are given.
6. Gelinas, Nicole. “
Lessons of Boston’s Big Dig
.” The Manhattan Institute.
Gelinas addresses key events throughout the three decade long construction process. She discusses background and reasons for the Big Dig, the political scandal that arose from the project cost, and structural failures which were made public during construction.
7. Harper, Jack (2004). “
Report: Even More Big Dig Leaks Found
.” The Boston Globe.
This news report brings forth issues concerning leaking in the tunnels of the “Big Dig”. Harper claims that slurry walls, and an ineffective waterproofing membrane is responsible for the leaks. The South Station stretch of the tunnel, discussed in (Rodwell) is reported in this article as having nearly 700 leaks. A major leak in the Liberty Tunnel I-93 section is also discussed.
8. Murphy, Sean P. (2007). “
Leakage in Big Dig Tunnel Rises
.” The Boston Globe.
The leakage problem in the Thomas O’Neill Tunnel is examined. The reason for this failure is insufficient waterproofing at the joint of the wall and the roof of the tunnel. Statistics showing the amount of water entering the tunnel are presented, comparing year to year values.
9. Rodwell, Jason H. (2001). “Instrumentation Monitoring on Contract 9A4 of Big Dig in Boston.” Geotechnical News, Vol. 19, Issue 4, pp. 35-41.
This article discusses the excavation and construction process of the Central Artery/Tunnel around the area of South Station. Difficult ground conditions and new construction techniques such as ground freezing attributed to movements of the slurry walls. After the publication of this article, numerous leaks sprung within the tunnel that can be directly traced to the failure of the slurry walls.
10. Schultheisz, Carl. “
Epoxy Used in the D Street Portal
.” National Transportation Safety Board.
This presentation describe the epoxy materials at a molecular level and presents laboratory test data comparing fast set and standard set epoxy used in the connections of ceiling panels in the I-90 connector tunnel.
11. Sigmund, Peter (2007). “
Triumph, Tragedy Mark Boston’s Big Dig Project
.” Construction Equipment Guide, Story I.D. 8751.
This article gives a general overview of the planning of the Big Dig. It highlights the construction achievements and methods used in the completion of the project. Sigmund briefly touches on the numerous failures.
12. Smith, Michael W. “
Boston’s Big Dig: A Socio-Historical and Political Analysis of Malfeasance and Official Deviance
.” Saint Anselm College.
While a significant portion of this article is dedicated to the political corruption associated with the Big Dig project, Smith also discusses the ceiling panel collapse I-90 Connector Tunnel. Similar panels that were used in the Ted Williams Tunnel are examined. Lawsuits connected with this collapse are discussed.
13. Wander, Steven (2008). “
Tunnel of Terror
.” National Aeronautics and Space Administration: System Failure Case Studies, Volume 2 Issue 5, June 2008.
This article discusses the Big Dig Ceiling collapse and the multiple warning signs and steps that could have been taken to prevent the tragedy from occurring.
Highway Accident Report:
Ceiling Collapse in the Interstate 90 Connector Tunnel Boston, Massachusetts July 10, 2006
.” National Transportation Safety Board.
15. McNichol, D. (2005). "
The Leak at the End of the Tunnel
." Wentworth Magazine, Volume 47, Number 1, Spring 2005.
16. The Associated Press (2007). "
$6 Million Settlement in Big Dig Accident
." The New York Times, Dec. 26, 2007.
17. Goodnough, A. (2008). "
Settlement for Company Charged in Big Dig Death
." The New York Times, Dec. 17, 2008.
help on how to format text
Turn off "Getting Started"