BY PHILLIP WEARNE AND JOHN KELLY
It has been more than a year since the attacks on the World Trade Center (WTC), yet a proper forensic engineering and fire investigation to discover how the WTC towers failed following the impact of the two Boeing 767 aircraft has only just begun. Yes, you read that correctly. The biggest structural and fire-induced building failures in history, which resulted in the biggest single loss of life on American soil since the Civil War (including 343 firefighters), has yet to be investigated.
The Building Performance Study (BPS), published by the Federal Emergency Management Agency (FEMA) and the American Society of Civil Engineers (ASCE), was a start; but, as many of its own team members were the first to admit, with limited time, money, and personnel, it was little more than that. As anyone who has followed two sets of congressional hearings, the press reports, and the lobbying efforts of the victims’ relatives, in particular those of the Skyscraper Safety Campaign, knows, the BPS furnished us with far more questions than answers.
Somewhat generally, the BPS recommended areas for future investigation and research, a task now passed to the National Institute of Standards and Technology (NIST) in Gaithersburg, Maryland. NIST unveiled an outline blueprint for its proposed study at public hearings in New York on June 24. In the words of one fire engineer, it was like motherhood and apple pie: “There’s nothing you could be against in it, but little to show what exactly they’re for.”
The final investigation plan, published on August 21, contains little more. “What specific questions they are going to try and answer, and how, still concern me,” says one member of the BPS team. “After all, it’s not as if we don’t know what we need to know now.”
Senior NIST officials insist their intent is “fact finding” and “analysis of the facts,” not “finger pointing” and “accusation.” In reality, the two may not be mutually exclusive. There is already enough information in the public domain—the result of court cases, media scrutiny, and leaks—to suggest that on issues of design, inspection, and maintenance, there may be more than enough blame to go around. The issue is not fact finding or even analysis and blame for its own sake; it is simply what lessons can be learned and, more urgently, how to ensure that they are applied.
The design, structure, maintenance, and emergency procedures in our buildings must now account for the possibility of a terrorist attack. This lesson was not sufficiently applied after the February 1993 attack on the WTC, or even after the April 1995 attack on the Alfred P. Murrah Building in Oklahoma City. It must surely be attended to now. We need to get as serious about this as many European states have and regard every government building, every prestige building, and perhaps every office or retail building as a potential terrorist target.
But other vital lessons need to be learned. Although the initial cause may have been exceptional and unprecedented—simultaneous suicide strikes by hijacked passenger aircraft—the fires resulting from them may not have been. The fires that raged in the WTC towers on September 11 exposed weaknesses that any severe fire resulting from more mundane causes might also have uncovered. Worse, many of the WTC design features that need the most scrutiny are the norm in tens of thousands of other buildings. The open-web floor trusses, the connections, the fireproofing, the shaft-wall partition system of the central core, and the limited means of egress in these buildings are replicated in other structures throughout the land and, indeed, the world.
Making things safer by retrofitting, rehabbing, enhancing, redesigning, rethinking, and integrating fire engineering in the design process is certainly going to cost money. But after September 11, everything else, from the war to the economy, is costing money. Why should construction and fire engineering be any different?
Any improvements must be based on precisely the kind of scientific research and investigation that has yet to take place. Listed below are just some of the questions that must be answered. They cover both elements of NIST’s mandate on this disaster—investigation and research. They are just some of what might be termed the “specs,” the “specifics,” not the generalities that too many of this disaster’s assessors and investigating experts have hidden behind to date.
•Where is the basic information? It took the BPS team four months to persuade the Port Authority to hand over the structural drawings. Yet the most basic information that any forensic structural or fire engineer requires—the “as-built” specifications—has not, as of this writing, been seen by anyone charged with this investigation. Other documentation still to be secured by NIST’s experts includes recent maintenance records, the fuel-loading surveys, and the late-1960s impact momentum study that assessed the buildings’ likely response to an impact from a large passenger jet. This documentation’s absence illustrates the necessity for the legal subpoena power that Congress granted NIST in establishing the National Construction Safety Team to investigate future structural failures.
•The impact of the impact. It is accepted that, after the impact, the fires that raged in the WTC towers heated the stressed steel external and core columns that then had to carry redistributed loads resulting from the severing of up to two-thirds of the external columns and an unknown number of the core columns. But how much did the extra stress on the steel reduce its fire resistance? Steel “plasticates”or loses its ability to carry loads in proportion to the temperature to which it is heated. Did the fires in these buildings not have to reach such exceptional temperatures or heat release rates to cause such sudden and spectacular collapses because the steel was also so stressed? If so, should much greater fire safety margins be built into protected steel structures for cases where blast or impact is followed by fire?
•The temperature, extent, and range of the fires. Just how hot and powerful were these fires? Initial estimates suggested that the fires were exceptional, with temperature guesses ranging up to 2,000°F. But modeling and analysis produced for lawyers—not, please note, federal investigators—suggests these were not exceptionally fierce fires. More modeling work on the ventilation and the fuel loads should come up with some more definitive answers. If only the cause but not the effect of these fires was exceptional, there are huge, vital lessons to learn here. Huge, open-plan floor spaces at high-rise or even super high-rise height may be much more vulnerable to fire than planners, code writers, or even fire chiefs currently realize, particularly when suppression systems, such as sprinklers, do not work.
•How inadequate was the fireproofing? The inadequacy of the fireproofing in the WTC is no longer disputed. Inspections by experts throughout the 1990s, including some by the Port Authority of New York and New Jersey, the owners of the WTC, show that fireproofing on the floor trusses was thin and nonexistent in places and was peeling away in room-high sheets from the core columns. The key question here is, What effect did this have in inducing the collapse? In 1995, the Port Authority began a fireproofing rehabilitation program, in particular spraying the floor trusses with 11/2 inches of CAFCO Blazeshield D/CF fireproofing where previously layers of just 1/2- or 3/4-inch had been considered sufficient to secure the three-hour fire rating needed to conform to the Class 1A building category provisions of the New York City building and fire code. All the floor trusses of the impact floors in the North Tower had been rehabbed with thicker, new fireproofing by September 11, whereas only one impact floor, the 78th floor, had been rehabbed in the South Tower. Was this one of the factors that enabled the North Tower to remain standing for nearly double the time of the South?
• Where are the fireproofing test results? The Port Authority says it based the rehabbing of the fireproofing on the floor trusses with 11/2 inches of coating on an Underwriters Laboratories test, specifically G805, listed in UL’s Fire Resistance Directory. The test was first listed in 1992. If 11/2 inches of CAFCO Blazeshield D/CF fireproofing was necessary to ensure three-hour fireproofing in 1992, why was only 1/2 inch of the same fireproofing material deemed necessary to secure the same fire rating on the same floor trusses in 1970 during construction? Where are the original fireproofing tests for the open-web floor trusses that justified such a thin coating of CAFCO Blazeshield D/CF fireproofing material? An equally urgent question, Where are the deflection/cohesion/adhesion tests for the CAFCO Blazeshield D/CF fireproofing? When first used in the WTC in 1970, it was a very new product, a last-minute nonasbestos substitute for its asbestos-based precursor, CAFCO Blazeshield D. Were such tests ever done? If not, why not?
• How much of the fireproofing survived impact? Some fire engineers have argued that the CAFCO Blazeshield D/CF fireproofing, a mineral fiber-based product, would not have survived the impact of the aircraft, so whether it was in place or was adequate before impact is irrelevant. Only deflection/cohesion/adhesion tests of the fireproofing on models of the trusses and columns will give us the answers as to how much fireproofing might have survived in the WTC towers and where. Such studies have huge import. Fire engineers, firefighters, architects, and structural engineers need to know what blast or impact forces—from a bomb, a gas explosion, an earthquake, or an aircraft—different types of fireproofing can withstand. Obviously, fireproofing only stands a chance of protecting steel or means of egress if it is in place when a fire takes hold.
•Column and beam vs. tube. The WTC towers were what is known as tube structures—closely spaced lattice-like columns of steel made up the external walls carrying most of the wind and gravity loads. The WTC towers were undoubtedly strong on the outside, but were they weak and particularly vulnerable inside? Once the aircraft had sliced through the external columns, were the buildings doomed? Would a traditional column-and-beam structure—with more widely spaced steel columns making up the outside walls with steel flange beams supporting the floors throughout the structure to give the building the traditional steel “skeleton”—have withstood such multifloor fires better as the experiences with high-rise fires in Philadelphia’s One Meridian Plaza and Los Angeles’ First Interstate Bank seem to indicate? And if it would have, how would a column-and-beam structure have reacted to the initial impact of the aircraft? Tube structures can be built with much less steel than traditional column-and-beam structures, which is one of their main attractions. But is it too little “body” or mass, particularly with so little concrete, masonry, or cement, to stand up to the committed terrorist?
• Trusses and their connections, primary members. The open-web trusses that were the basis for the floors in the WTC towers spanned distances of up to 60 feet in connecting the external columns to those of the core. Were the trusses or their connections (standard high-tensile field bolts, attached to relatively narrow seat angles with plates welded on top) strong enough to withstand the catenary bending action and the shear on the bolts induced by relatively modest fires, let alone those raging on September 11? Moreover, did the function of the floor trusses in bracing the core and external columns to keep them the requisite distance apart, in equilibrium, make the trusses primary, rather than secondary, members in this design? If so, even without the fires and weakening effect of impact on September 11, how many sets of floor trusses could you have removed from this design without the towers going into terminal and very rapid progressive collapse, as they did that morning?
•Shaft-wall partition system central core. The vast majority of the victims of the WTC attacks on September 11 died because they had no means of escape. They were trapped above the impact floors with the buildings’ umbilical cord—the elevators and three sets of stairs—severed below them. The WTC towers included a shaft-wall partition system central core—two layers of gypsum board attached to widely spaced core columns—20 feet off center. The only cement and concrete used in the central core was in the stairs themselves. Would a more traditional central core, with masonry, cement, and concrete as its main components, have survived the impact and fires better? All these elements are known for their blast-, impact-, and fire-resistance properties. Should central cores in high-rise buildings possess the rigidity and durability the name implies?
• Defend in place—surely it is dead. The towers each had three stairways, only one of which, at 54 inches, was considered sufficiently wide to comfortably accommodate two-abreast flows of evacuees. Where firefighters were ascending, there was one flow of evacuees descending and a counterflow of emergency personnel. No one knows how long it would have taken to completely evacuate the towers with such limited stair capacity when fully occupied—six or seven hours is one guess (based on the mass evacuation of February 1993 following the bomb attack). The sorry truth is that the towers were never designed for mass, simultaneous evacuation, yet each time there was a major incident—in 1975 (fire), in 1993 (bomb), and again in 2001 (fire)—that was precisely what was necessary. Even if “defend in place” is an option for fire and other emergency personnel in some instances, if the buildings’ occupants have no confidence in it or will not adhere to it, it must be scrapped as an emergency procedure. All high-rise buildings must have adequate means of egress for mass, simultaneous evacuation by all occupants in a set period of time. And in extreme circumstances, such as September 11, occupants do not have six or seven hours. Had the aircraft hit those buildings just one hour later, when the buildings would have been fully occupied, the death toll would have been much, much higher, perhaps between 10,000 and 15,000. The reason is simple: Their capacity to evacuate would have come nowhere near meeting the numbers needed to get out in the time allowed.
• Built-in redundancy to reduce/eliminate the threat of progressive collapse. The WTC towers’ collapse was the most extreme and deadly example of progressive collapse ever seen. The failure of one or more floors overloaded the floor or floors below, causing a chain reaction that took only 10 to 12 seconds to complete. Was sufficient redundancy built into the towers’ design? The external walls had sufficient redundancy—the ability to redistribute loads to other structural members to prevent failure—as the initial impact proved. But what about inside? What about the open-web floor trusses and core columns? Could the floor trusses and their connections have been more redundant with connections fastening them to the columns at the lower as well as upper chords? Could the core columns have been cross-braced with steel members for additional stability? What difference might this have made, especially for those occupants and rescue personnel trapped in the South Tower? Although it collapsed in less than an hour, one stairwell remained intact and at least two firefighters reached the 78th floor sky lobby. Here, at least, more time could have made all the difference.
PHILLIP WEARNE and JOHN KELLY produced the one-hour documentary “World Trade Center: Anatomy of a Collapse,” originally shown on Channel Four in Britain and The Learning Channel in North America earlier this year. An updated version of their film was broadcast on The Learning Channel to mark the first anniversary of the disaster on September 11, 2002.