Falling Down

An interview with W. Gene Corley, leader of an American Society of Civil Engineers investigation of the destruction of the World Trade Center

8 min read

For experience with assessing how structures perform in the wake of a terrorist attack. W. Gene Corley is hard to beat. He is senior vice president of Construction Technology Laboratories, Skokie, Ill., and was principal investigator and leader for a study sponsored by the American Society of Civil Engineers (ASCE) of the 1995 Murrah Federal Building bombing in Oklahoma City.

Now he has been called upon again by the ASCE, this time to lead its investigations into the 11 September collapse of buildings at the World Trade Center (WTC), in New York City, and the destruction of part of the Pentagon, in Washington, D.C., on the same day. As might have been expected, he has reassembled a good part of his Oklahoma team.

Actually, two teams are being put together by Corley and the ASCE’s Structural Engineering Institute. Each consists of seven to 10 experts, including structural engineers as well as experts in blast and fire protection. All are working pro bono for the time being, though the ASCE is paying out-of-pocket costs. The cost in terms of salary-related expenses will be upwards of US $500 000. The cost of data acquisition and computer simulations will add substantially to that figure.

Last week, Corley spoke with IEEE Spectrum’s Harry Goldstein about the investigation, how the Towers 1 and 2 collapsed, and what the future holds for high-rise construction.

When did the investigation start?

The day after the attack.

What does the data acquisition involve?

First, we will be looking for photos showing the condition of the building from the time it was first hit and later. Still shots will be the most useful because we need their quality to identify the impact damage to the steel columns on the exterior of the building. We also will be looking at the videos–those should show us some of the damage that was not picked up by the still shots–and at Tower 7, which also collapsed, and at buildings that were badly damaged but stood up.

We will also be looking for information about the aircraft and estimates of how fast they were going when they hit, how much fuel they had, that sort of thing, in order to analyze their impact on the buildings. Then we’ll determine where the fire spread to immediately and how it spread after that, and make calculations to model what happened to each tower and the other buildings, to see if we can predict the way and the time at which they collapsed.

Based upon that, we will examine some structural variables to see if any reasonable things could be done in the future to give more time before collapse. The ideal would be to have time enough to put the fire out. Once the fire’s out, the buildings could be stabilized.

If the fire could have been stopped, would the buildings still have collapsed?

The general belief is that the impact of the aircraft did not bring the buildings down, it was the fire that followed. So if the fire could have been put out, it is reasonable to believe that the buildings would not have collapsed.

And that’s because the steel columns became heated?

Yes. As steel heats up, it loses strength. Depending on the type of steel you have at a given location, you can determine what the loss of strength is for a given temperature. When the planes hit, they took out some of the structure, but the building redistributed the loads and carried its the weight. The surviving structural members had to carry more load, so the fire didn’t have to heat them up as much as it would have normally for them to collapse.

At what temperature does steel fail?

You have to look at it for each member in the building because there are different kinds of steel in different members. A rule of thumb is that if you heat it up to around 600 C, you’ll lose approximately half the strength of the member and eventually, you can no longer to carry the loads. That’s not necessarily what happened, 600 C isn’t necessarily the temperature that was reached. That’s something we will be determining.

What kind of computer simulations will you be running?

I don’t have the name of the computer program yet, but it’s been described to me as one that can model impact, fire, and structural performance. If we can’t get that program or it can’t do all that, we’ll just have to go back and forth among separate programs. We definitely have access to programs that can model a fire, and we can model the impact in another and the structure in a third. We can use output from one to feed the other and iterate until we get an answer. We haven’t gotten far enough to decide all of these things yet.

What’s the timetable for data acquisition and the modeling?

The data acquisition started virtually the same day we started forming the committee. We had a press conference the Thursday after the attack. We were really starting to collect information then. We asked people for photos and video clips to analyze the exterior damage. Then we have had Bill Baker, a structural engineer with Skidmore, Owings and Merrill, Chicago, on the site; he was helping the search and rescue, but also was making mental and some written notes on what he saw as they unscrambled the debris. The next step will be to get our team on site to collect the needed information, and also get them access to plans and specifications and other things that we need to model the structure.

What’s the process of collecting data at ground zero and what about the material already removed to the Fresh Kills landfill on Staten Island?

As far as on-site part of the operation is concerned, the pieces from the towers have marks on them that we will be using identify where they came from when we get ok to go in, which we expect any day. If we find a piece that’s being removed, we’ll have to take a photograph and attempt to see it later off-site. We are already seeking permission to look at the debris if it is indeed being kept at Fresh Kills and is being put in some order by the criminal investigators.

What are you looking for in the rubble to tell you what happened to the interior?

The same types of things the National Transportation Safety Board will be searching for. They will be looking more for the aircraft parts, but we’ll be looking for pieces of the building in the vicinity where the aircraft hit, and for the impact damage. We’ll also be looking to sample the parts of the structure involved in the fire, so we can look at the metallurgy of the steel and also look for any evidence of damage due to the fire.

How can you tell impact from an aircraft as opposed to impact from collapse?

The pieces of the building still have the marks used for its erection, so we can identify locations at least for the not badly burned pieces. From the photos we anticipate telling the precise impact locations, and which pieces of the building were hit by the aircraft and which were damaged further from the buildings’ collapse.

I’ve already verified that we can find the marks on the pieces of the building and presumably we will get access to the erection schedules or find someone who knows how to interpret them and identify the pieces. It appears that they have the floor numbers right on the piece. In some cases it’s white paint through a stencil, in other cases it’s chalk marks. The chalk marks are there very clearly. The thing will be to interpret what they mean.

This whole operation could take months?

I’m estimating that in 6—18 months, if the criminal investigation doesn’t put a hold on it, we can have a report ready. I have to emphasize that we give precedence to the criminal investigation–we don’t want to do anything to interfere with that.

Officially the report will be presented to the ASCE. There are several other organizations that are cooperating, including the American Institute of Steel Construction, American Concrete Institute, National Council of Structural Engineers Associations, and the National Fire Protection Association.

So you’re the main forensic team on this?

I’d like to avoid the word forensic, because some people might think that we’re going out to find things for lawsuits, but that’s not what we’re doing. We like to refer to it is as a Building Performance Assessment.

All of us think that the buildings did extremely well to take the impact of a fuel- loaded 767 and not collapse instantaneously. That was somewhat beyond the design criteria, which were for a smaller 707.

I’ve been told unofficially that the amount of fuel carried by the planes that hit the towers was beyond the capacity of the jets the buildings’ architects and engineers designed the buildings for, so it was probably not considered. But the buildings did what they were designed to do–one stood up for over an hour, the other almost an hour, and allowed 25 000 people to get out alive. What we’ll be looking for is what might be done differently in the future that would give even more time to evacuate and save even more lives.

Is there a realistic way to retrofit existing skyscrapers with collapse protection, or must you design it into new buildings and if so, how does that affect the cost?

We anticipate that not all buildings need to be designed for this kind of catastrophe. For those that might be looked at as a symbol that someone might want to destroy, we may be able to recommend additional things that would give them more resistance. But we don’t think it’s appropriate or necessary at all to do that to a vast majority of the building stock.

As far as retrofit is concerned, it’s too early at this point to say whether anything we come up with might be feasible as a retrofit. Usually retrofit is so expensive that it’s not practical.

Are there any lessons from Oklahoma City that apply to the World Trade Center or are they just too different?

They are completely different. Oklahoma City was a classic case of a progressive collapse that happens instantaneously, as soon as the first structural member is destroyed. There the bomb, as our report says, took out three of the four columns on the north side of the building. To protect against progressive collapse, you can’t design so that 75 percent of the vertical support for a section of the building is removed and the building still stands up. So if that happens due to an airplane impact or explosive, then our recommendations would not cover that.

But in our Oklahoma City report we did recommend ways to reduce the amount of the building that would be taken out by the blast, and have more redundancy in what remains. At the World Trade Center, part of the formula was in place. A large part of the structure was taken out, or at least badly damaged, and it did redistribute the loads and stand up for a while.

How common is the tubular construction of the WTC [thin steel columns spaced 1 meter apart around the outer walls, with further support inside interior from a central steel and concrete core]?

A lot of buildings have used it since the early 1970s. It was extremely popular at the time those buildings were built, and it’s certainly an attractive structural system to use. It performed really quite well under the impact.

Did it also aid in containing the collapse vertically?

I don’t think it would have made any significant difference for a building that tall. Most tall buildings have so little lateral load resistance [ability to resist side-to-side movement due to wind or earthquakes] that they can’t really tilt over. They may tilt a little, like one of the WTC towers did, but when they get just a few degrees over, they come straight down. Most buildings don’t fall over, they fall down.

Should we stop building skyscrapers?

Absolutely not. The skyscraper is the city. If we stop building skyscrapers, the terrorists have won. We have to build up. It’s absolutely impractical to spread everything out with four-story buildings. It creates unlivable cities. And in reality not a lot of skyscrapers are even going to be targets. If there are any changes to be made, they should apply only to those buildings that are likely to be targets.

I guess that would apply to any high rise they decide to put up at the WTC site?

That’s one thing that we can’t discount, that whatever goes there has the potential for being a target.

And, while this is speculation, that building might incorporate some of your team’s recommendations?

That’s too much speculation. We’ve got a whole lot of ideas, but we’ve got to find out if they make sense. The test is that if we come up with something and someone says, "OK, I have to have a certain safety against a 767 hitting a building," then is that a viable building when you build that much resistance into it? No building is successful if it’s too expensive for people to use.

What cost/benefit analysis will go into deciding to implement these ideas?

Whether or not we can go very far with that probably depends on whether we get some additional support for our effort. We’ve bitten off a big chunk already. We’ve asked people to do a lot free of charge while they also try to make a living, and so that will somewhat limit how much we can do.

But if we can get the resources to do it, we will calculate costs of the changes and do an estimate. We did do that for Oklahoma City [with support from FEMA [Federal Emergency Management Agency], it was not pro bono. We would hope that somebody would come in and provide us with the resources to do more than we plan to at the moment.

This article is for IEEE members only. Join IEEE to access our full archive.

Join the world’s largest professional organization devoted to engineering and applied sciences and get access to all of Spectrum’s articles, podcasts, and special reports. Learn more →

If you're already an IEEE member, please sign in to continue reading.

Membership includes:

  • Get unlimited access to IEEE Spectrum content
  • Follow your favorite topics to create a personalized feed of IEEE Spectrum content
  • Save Spectrum articles to read later
  • Network with other technology professionals
  • Establish a professional profile
  • Create a group to share and collaborate on projects
  • Discover IEEE events and activities
  • Join and participate in discussions