The AIA Trust home

Case Study

The Citicorp Center Design

In the early 1960s, Citicorp’s headquarters, located at 399 Park Avenue in Manhattan, became too small for the company’s ongoing growth and expansion. At the same time, St. Peter’s Lutheran Church, constructed in 1905 and in need of serious repair without the funds to do so, began to consider selling its valuable property on the northeast corner of Lexington Avenue and 54th Street. As fate would have it, the church was located across the street from the bank’s existing offices.

Citicorp spent five years and approximately $40 million to purchase all but one portion of the entire block.[4] The financial institution began planning to erect a new building on the church’s block. But there was one important condition of the church’s sale of its property to the bank: the bank would have to build a new church on the same corner of the block with no connection to the Citicorp building and no columns were permitted to pass through the new church to support the bank’s building. The conditions imposed presented a significant architectural and engineering obstacle to constructing a well-designed high-rise building on the site, and required a team equal to the challenge.

In the early 1970s, Citibank retained Hugh Stubbins to be the architect of the new building.[5] Stubbins had started his firm roughly 20 years earlier and was experienced in designing high-rise projects. In fact, it was during the construction of Stubbins’ first high-rise building—the State Street Bank Building in Boston—where his retention of, and relationship with, structural engineer William LeMessurier began.[6]

sketch of the tower
Like many stories involving design professionals, this one begins with a simple paper napkin. In a Greek restaurant close to his firm’s offices in Cambridge, LeMessurier pondered how to overcome the restriction of placing one of the columns in the traditional corner location. The proposed solution he sketched on the napkin was a chevron pattern of structural bracing to transfer loads to columns placed at the mid-span of the building’s exterior walls. LeMessurier’s idea permitted each corner of the building to cantilever out 72 feet, which allowed Stubbins to delicately nestle the new church building into the northwest corner of the property while maintaining the use of the entire lot for Citicorp. The columns would act as stilts raising the building 114 feet in the air.[7] In addition to allowing the preexisting church to remain in place, the innovative design also left considerable open space at the ground level that enabled the height of the building to rise above what would otherwise be permitted by the zoning laws.[8]

Although the 160-foot angled top was originally conceived to face west and provide for residential penthouse setbacks, the City refused to grant a permit for residential use.[9] As a result, the configuration was re-oriented to the south to maximize the potential for generating solar power. Unfortunately, due to the lack of technological advancement of solar collectors at that time, the solar plan was abandoned. However, the unique angular top had become a signature of the design, and it was retained as housing for most of the mechanical equipment.

The Citicorp Center, completed in 1977, reportedly cost an estimated $195 million. At approximately 25,000 tons, the steel superstructure is considerably lighter than other skyscrapers of similar size. Due to the relatively light weight, building sway was a serious concern that LeMessurier addressed in the design phase. At the cost of $1.5 million, a “tuned mass damper” was installed in the angled top, the first of its kind implemented in the United States. This 400-ton block of concrete, which floats on a film of oil, is computer controlled to absorb the energy of the swaying structure and to reduce the building’s wind-induced movement by as much as 50%.[10] At the time it was designed and installed, no one could have foreseen just how valuable the tuned mass damper might prove to be.

A Student’s Inquiry

In June 1978, Princeton University engineering student Diane Hartley wrote her undergraduate thesis, “Implications of a Major Office Complex: Scientific, Social and Symbolic Implications” on The Citicorp Center.

The Citicrop Center under construction
In examining the structural engineering of the project in the preparation of her thesis, Ms. Hartley spoke to a junior engineer at LeMessurier’s firm and requested the structure’s plans and engineering calculations. She received the documents, and set about performing her own calculations—which indicated that quartering winds produced significantly higher stresses than those produced by winds hitting just one face of the building. This structural behavior was different from a conventional building with columns placed at the four corners, where wind perpendicular to one face presents the worst loading case.[11] As she reviewed the structural design, Ms. Hartley believed that it was not sufficient to withstand the quartering winds. She questioned LeMessurier’s junior engineer about her findings, but was assured that the building was “more efficient” then her calculations suggested.[12]

Ms. Hartley submitted her thesis, which included both the inconsistency noted in her calculations, and the response from the junior engineer. She was not the only one to question the calculations. Her professor, David Billington, also questioned this inconsistency in his comments on Ms. Hartley’s thesis.[13] In deference to the reputation and expertise of LeMessurier’s firm, Ms. Hartley, an undergraduate student, accepted the response from the engineering firm.[14]

Years later, LeMessurier recounted that when he was first informed about the student’s inquiry, he thought that the story would provide an interesting classroom lecture for his structural engineering class at Harvard University. Others say that LeMessurier’s own dismissal of the significance of the quartering winds design condition encouraged him to revisit the issue. According to Robert McNamara, LeMessurier’s partner at the time, LeMessurier had in fact considered the quartering winds during the design phase,[15] but determined that the quartering winds did not produce as much stress on the building as the perpendicular winds. The result was that the impact of the quartering winds was not closely analyzed since they were not thought to be the controlling design condition.[16]

LeMessurier passed away in June 2007, and never clarified what portion of the exchange between Ms. Hartley and the junior engineer was relayed to him. But this much is certain, the inquiry was relayed to him quickly and the inquiry piqued his interest. This may have been caused by a recent discovery he had made about his own design.

The Approved Substitution

In May 1978, only a month before Ms. Hartley raised her inquiry, LeMessurier was discussing two new skyscrapers slated to be constructed in Pittsburgh, PA. The design of the two buildings envisioned wind braces similar to the chevron pattern utilized at the Citicorp Center. One of the potential bidders for the Pittsburgh project raised a concern that LeMessurier’s specifications required welded joints to the chevron braces instead of high-strength bolted joints—the welded joints were by comparison much more labor intensive and expensive. The bidder explained to LeMessurier that if bolted joints could not be substituted for welded joints, his firm might not be interested in moving forward.

It is unclear if LeMessurier knew, or even remembered that the request to bolt rather than weld the joints on the Citicorp Center had been approved by the design team. During the course of the Citicorp Center project, the steel contractor, Bethlehem Steel, proposed a substitution to the full penetration weld required in the construction documents. It offered Citicorp a $250,000 credit to use bolts instead of welds.[17] The substitution was accepted and the design team provided the design forces to Bethlehem Steel for Bethlehem’s use in designing the bolted connections. Bethlehem’s design was then reviewed and approved through the shop drawing process.

LeMessurier certainly knew of the substitution in June 1978 when he became aware of the student’s inquiry. LeMessurier gave the issue serious thought and chose to further investigate. LeMessurier performed some preliminary calculations on the effect of the quartering winds and uncovered what he called “very peculiar behavior.”[18] He found that in four chevrons in each tier, the diagonal wind stresses practically disappeared, while the stresses on the remaining four chevron braces nearly doubled.[19]

By itself, the doubling of the forces that the diagonals would have to resist was not a concern for the steel members.[20] Rather, the concern was that the bolts might not be strong enough to carry the imposed forces. LeMessurier later explained that “I didn’t go into a panic over it, but I was haunted by a hunch that it was something I’d better look into.”[21]

LeMessurier’s Investigation

On July 24, 1978, LeMessurier traveled to his New York City office to carefully evaluate the impact of the substitution on the bolted connections. He found that the diagonal braces had not been treated as columns as he believed they should have been, but were, instead, treated as trusses. The force of wind against the building creates tension in the members as they resist the wind’s efforts to topple the building. Since the building has weight, that weight (dead load) naturally counteracts the toppling force. Accordingly, the members need only account for the difference between the force resulting from the dead load (compressive) and the toppling (tensile) force. According to the American Institute of Steel Construction specifications in effect at the time, an engineer could only subtract three-quarters of the dead load force imposed in the columns.

However, in considering the diagonal bracing as truss members, LeMessurier’s office incorrectly subtracted the entire dead load force. This relatively subtle difference resulted in half the amount of bolts that should have been used at each connection, which is equal to the same amount of bolts that would have been used if the diagonals were considered columns.[22]

sketch of the tower
The fact that the capacity of the connections was under-designed became even more troubling when combined with the failure to take into account additional loads due to quartering winds. LeMessurier calculated that the quartering winds increased the member stresses by 40%. As a result of LeMessurier’s calculations, thoughts of a very realistic potential structural system failure and building collapse began to emerge.

On July 26, 1978, LeMessurier flew to Ontario, Canada to meet with Alan Davenport, a wind tunnel expert at the University of Western Ontario. Davenport modeled the building during the design phase and, at LeMessurier’s request, retrieved his earlier wind modeling files. Davenport confirmed LeMessurier’s fears and, worse, also opined that the increase in member stresses could likely climb higher than the theoretical calculations.[23]

At LeMessurier’s request, Davenport provided detailed loading information on all of the structural members. LeMessurier’s fears were confirmed: there was a major problem.

The Decision

Just two days later, on July 28, 1978, LeMessurier drove to his summer home located on a private island on Sebago Lake in Maine and spent the weekend analyzing Davenport’s information to determine the full extent of the problem. LeMessurier’s calculations revealed that 70 mph quartering winds would topple the building. According to the New York City weather records, the likelihood of a storm with 70 mph winds was one in every sixteen years. By taking the tuned mass damper into account, LeMessurier calculated that the probability of failure could be reduced to one in fifty-five years, but that assumed that the power to the building, which was needed to keep the tuned mass damper working, was maintained without disruption during such a storm.

LeMessurier evaluated his predicament. He would later recount that he figured he had three options: stay silent; commit suicide; or tell others of the problem. Despite articulating those options, other comments by LeMessurier made clear that he never seriously considered any option but the third.[24] He knew that disclosing the problem could lead to lawsuits, bankruptcy or possibly end his career, but LeMessurier believed his selfish worries were not enough to overcome his “social obligation.” Indeed, in a videotaped lecture given at the Massachusetts Institute of Technology, LeMessurier discussed the ethical dilemma he faced and said:

If you’ve got a license from the State and a certification from the University first and now you’re gonna [sic] use the license to hold yourself out as a professional, you have a responsibility beyond yourself, if you see something that is a social risk...good heavens this thing would kill thousands! You must do something. You must do something.[25]

After deciding to act, LeMessurier developed his plan to fix the connections that same weekend. Fortunately, the diagonally bracing connections that needed retrofitting were located between the floors and were relatively accessible.[26] LeMessurier’s plan not only included the work to the braces, but rather detailed the construction process so as to minimize the interruption to the employees in the building.[27] In a speech LeMessurier gave twenty years later, he explained that when he decided that he would inform Citicorp of the issue, he also needed to present a solution. “I had a scheme which I thought of before I opened my mouth. That’s terribly important. You don’t just cause havoc without having a solution.”[28]

A Fortuitous Chain of Extraordinary Events

On Monday morning, July 31, 1978, LeMessurier attempted to contact Hugh Stubbins, but was unable to do so. Instead, he spoke to Stubbins’ attorney (and his friend) and explained the issue. The attorney advised LeMessurier to contact his own insurance carrier before he told anyone else.[29] LeMessurier did as he was advised and contacted his insurer. The insurer appreciated the enormity of the situation and, the next morning, he met with lawyers appointed by his carrier in New York City. The team immediately made an important strategic decision by contacting Leslie Robertson, a structural engineer who had experience with both high-rise design and disaster management.[30] The lawyers met with LeMessurier and Robertson later that same day and, during the meeting, everyone agreed that LeMessurier and Stubbins should meet in person with Citicorp’s chairman to inform them of this discovery.

At this point, Stubbins had no reason to believe that anything was wrong with the project. LeMessurier flew back to Boston that evening and met Stubbins just as Stubbins arrived home from a trip. LeMessurier informed Stubbins of his discovery, and Stubbins agreed to fly with him to New York the next morning to meet with Citicorp.

On August 2, 1978, LeMessurier and Stubbins attempted to meet with Citicorp’s chairman Walter Wriston. They were unable to arrange a meeting with him and instead met with John S. Reed, who was then the executive vice-president.[31] Reed had an engineering background and had been involved in discussions during the project about installing the tuned mass damper.[32] LeMessurier explained the issue to Reed and how he believed it could be rectified. They discussed the estimated repair cost, and LeMessurier and Stubbins left the meeting with little feedback from Reed except that they should return to LeMessurier’s office and await further instructions.

Two hours later, John Reed and Walter Wriston arrived at LeMessurier’s office to discuss the predicament. LeMessurier recounted his sobering findings and the proposed solution to rectify the deficiency. According to LeMessurier, “Wriston was fantastic” and did not become adversarial. Rather, Wriston offered support in getting the remedial work done. Shortly after the meeting, LeMessurier’s team arranged for emergency generators to be installed to insure that the tuned mass damper would remain fully operational even if power to the building was interrupted.

The Remedial Work

On August 3, 1978, LeMessurier and four associates met with Robertson and two Citicorp employees that were charged with overseeing the repair work.[33] LeMessurier explained his proposed design, which included welding two inch thick by six feet long gusset plates to more than 200 bolted connections. Citicorp approved the proposed design provided that the necessary materials and labor could be secured immediately. After some coaxing, Robertson was able to persuade Karl Koch Erecting to undertake the work, even though they were initially hesitant to do so because of commitments on other projects. Fortunately, Koch had the material on hand and agreed to start work once LeMessurier’s office issued drawings.[34]

In addition to pushing the work forward as quickly as possible, LeMessurier and Robertson took several other critical steps to address the obvious safety concerns. First, they arranged to have the manufacturer of the tuned mass damper provide 24-hour service to insure no technical problems would cause an interruption in service.[35] Second, they placed strain gauges on several structural members so that the team could monitor the stresses imposed on the members. Finally, weather experts and forecasters were retained to provide weather and wind predictions four times each day.[36]

The Press (or Lack Thereof)

On August 7, 1978, LeMessurier’s office issued the drawings for the repair work. LeMessurier and the team understood that, with the release of the drawings, a new element was about to be introduced into the very delicate situation. Up until this point, Citicorp, the design team members, Davenport, Robertson, the lawyers and the contractors were the only ones that knew about the existence of the problem. However, the drawings would have to be submitted to the Department of Buildings for review and permitting before any work could get underway, and consequently, the group of those with knowledge was about to greatly expand.

In addition, that same day, Citicorp and Robertson met with the director of disaster services for the New York metropolitan area of the American Red Cross to discuss formulating an evacuation plan.[37] The Mayor’s Office of Emergency Management and the New York Police Department were included, and a plan was developed in case the predicted winds reached an unsafe level.[38] The evacuation plan included a ten-block radius, as the Red Cross estimated that 200,000 people could die if the structure collapsed.[39]

On the following day, August 8, 1978, Citicorp issued a bland, corporate jargon-filled press release that stated that the engineers had recommended strengthening certain connections of the wind bracing system. The Wall Street Journal quoted Henry DeFord, Citicorp’s senior vice president, as stating that “engineers have assured the bank that the building isn’t in any danger. The work is being done to anticipate the impossible that might happen.”[40]

sketch of the tower
Later that day, LeMessurier and the team, along with the contractor, met with the Acting Building Commissioner, Blaise Parascandola, and nine other senior New York City officials to explain the reasons for the repairs and the solution contained in the permit submission. LeMessurier told the city officials that his own office was responsible for the predicament, but that the repairs would fully remediate the problem. LeMessurier’s message was well-received by the City representatives. To the surprise of some, the City readily volunteered to do its part in assisting the effort. Concerned with a shortage of certified welders, the Department of Buildings offered one of the most trusted steel inspectors to provide testing and immediate certification of any welder that might be available to help with the repair work.[41]

The City’s response was remarkable in some ways, but was not unique. Interesting insight is gleaned from Arthur Nusbaum, the project manager for the contractor for the original construction and the repair work. He remarked, “[i]t started with a guy who stood up and said ‘I got a problem, I made the problem, let’s fix the problem.’ If you’re gonna kill a guy like LeMessurier, why should anybody ever talk?”[42] LeMessurier’s actions seemed to have commanded respect from all those involved.

Once the press got wind of the meeting with City officials, reporters gained a heightened level of interest in the proposed scope of work. However, Citicorp continued its efforts in downplaying the repairs by continuing to paint the news as mundane and not newsworthy. In a New York Daily News interview on August 8, DeFord stated, “As it is, the building could withstand a one-hundred year wind...We are a very cautious organization—we wear both belts and suspenders here. We don’t want people concerned, so we sent out a press release announcing the work.”[43] Likewise the Acting Commissioner of the Department of Buildings stated in the same article “of course it’s improbable, but there’s always the chance of winds up to 150 mph, which...could break the bolts.”[44] LeMessurier’s handling of the situation with the City officials seemed to have created a fair amount of good will that became invaluable, especially in light of statements by City officials completely removing any notion that an engineering error was the cause for the repairs.

Later that evening, LeMessurier found out that a New York Times reporter had been trying to reach him all afternoon.[45] LeMessurier became concerned that an investigative reporter was insistent on attempting to reach him, and reluctantly returned the reporter’s call that evening. Instead of having the dreaded conversation, LeMessurier was greeted by a recording indicating that a strike had just closed all of the newspapers in town.[46] Without newspapers, information dissemination in New York City skidded to a halt, leaving no one to question the repairs and the reasons for them.

The Weather Cooperates

Another fortuitous event came about a month later. August is hurricane season in New York. To retrofit the building in the most efficient way possible, LeMessurier prioritized every single connection, to insure that the welders were always working on the next most vulnerable connection. As welders raced to complete the welds, the temporary command center set up in Leslie Robertson’s office was receiving terrible news about Hurricane Ella which had formed south of Bermuda on August 30.[47] By September 1, Hurricane Ella’s winds reached 125 mph and was on a course to pass close to the Outer Banks of North Carolina. While many people on the East Coast made alternate plans for what seemed to be a ruined Labor Day weekend, very few fully appreciated the devastation that Hurricane Ella held in its sights. The dreaded evacuation plan was hours away from implementation.

Then, as if Hurricane Ella was responding positively to LeMessurier’s confession in much the same way as Citicorp, the City and the contractor had, it miraculously changed course and, unexpectedly, headed out to sea. With disaster avoided, welding continued and was eventually completed in October, just four months after receiving the student’s phone call. The building now stands as one of the safest skyscrapers in New York City, able to withstand a 700-year storm without the aid of the tuned mass damper.

The Silence is Broken

With the exception of those involved, very few knew the real reason for the fixes to the wind bracing in 1978. But in May 1995, Joe Morgenstern broke LeMessurier’s story in an engaging article published in The New Yorker. For almost 17 years, the story had remained hidden from the general public. Since then, many have viewed the actions of LeMessurier as nearly heroic, and many engineering schools and ethics educators now use LeMessurier’s story as an example of how to act ethically.[48]

The Impact

While LeMessurier’s initial estimate for the repair work was approximately $1 million, after the work was completed, he revised his estimate to be closer to $4.3 million. The contractor later reported that the construction work alone cost $8 million. While Citicorp never disclosed the actual cost of the remedial efforts, in some respects, it was not important. At the first ‘settlement meeting’ LeMessurier met with DeFord and Dexter, another Citicorp VP, without attorneys. While LeMessurier recounted that Citicorp pushed for a more significant contribution, at the second meeting with LeMessurier’s attorney, Citibank agreed to accept the $2 million as full payment and agreed to seek nothing from Stubbins’ firm.

‹‹ back to Overview | top of this page | on to next section ››