Even in the cutting edge world of skyscraper design, the Burj Khalifa stands apart. Officially opening in 2010, it remains the world's tallest building by a long way, and represents an amazing engineering achievement. To get the inside story on its design and construction, along with the unique challenges the project posed, we spoke with the Burj Khalifa's structural engineer, William F. Baker, a structural and civil engineering partner at Skidmore, Owings & Merrill.
Many readers will be familiar with the name Skidmore, Owings & Merrill (SOM), or if not, its varied projects. The firm was founded in Chicago in the 1930s and is responsible for many of the world's most recognizable buildings. These include Chicago's John Hancock Center and Willis Tower (aka Sears Tower), NYC's One World Trade Center, and, of course, Dubai's Burj Khalifa.
In addition to being the world's tallest skyscraper, the Burj Khalifa is the tallest free-standing structure in the world. It rises to a height of 829.8 m (2,723 ft)-tall, which is equal to around 11 Boeing 747-8 airplanes stacked tail to tip, or roughly 2.5 Eiffel Towers. The interior is a vertical city of sorts, with a total floorspace of 334,000 sq m (3,595,146 sq ft), which is over twice the floorspace of China's Forbidden City.
William F. Baker has been at SOM since 1981 and has led its structural engineering practice for over two decades. Best known for his work developing the Burj Khalifa's unique "buttressed core" structural system, he's a leader in his field and his awards and accolades are too many to list here.
We spoke with Mr Baker on the phone, and chatted about the Burj and his previous work leading up to it.
New Atlas: Please tell us a little about your background and how you ended up at Skidmore, Owings & Merrill.
I grew up in the Midwest United States, in a small town of about 10,000 people. I think the tallest building in town was probably two stories. I took an aptitude test in high school that suggested I should be an engineer and I went home and asked my mother what an engineer was.
Looking at the various engineerings, I decided I should take a shot at being a structural engineer. It actually turned out that both my grandfathers, who had passed away before I was born, were structural engineers so I guess it was slightly genetic.
So anyway, I went to the local university for a while and then I worked not in architecture or structural engineering, but for an oil company. Then I decided I really wanted to take a shot at structural engineering so I went to the University of Illinois in Urbana-Champaign, which is one of the top engineering schools in the country and it turns out that's where a lot of the people from my firm had gone to school. So there's kind of like, you might say, a mini-pipeline from the university to SOM. So I got put in the pipeline and I ended up here.
It was here that I learned about tall buildings and the design of them. At that point the firm had done the John Hancock Tower, it had done the Sears Tower [pictured below]. In fact, my first project as a project engineer was a renovation of the Sears Tower. Through the years I've worked on many, many projects, many tall projects, some not so tall.
By your accent are you in London? I've done a series of buildings in London, not necessarily so tall. One I'm very proud of is called Exchange House at Liverpool Street Station, it's a building that's both a bridge and a building, it spans over a rail yard at Liverpool Street. Anyway, I work at all scales, I also occasionally do single-family homes. So, from one story to the Burj, I find all that stuff interesting.
NA: What does your role in a typical project involve?
As you get more and more senior, you tend to get more involved in the front end part of the projects, when you develop the concepts. I'm most heavily involved in a project when it comes to the door.
SOM is pretty unique in the fact that we're architects and engineers and urban planners, and interior designers, civil engineers, sustainability engineers, all in one practice. So, when a client comes to the firm or there's a competition we're going after, we all see it at the same time and throw in ideas, either from our discipline or, we're not too restrictive and can throw in ideas from other disciplines too, if you want.
My heaviest involvement is at the very beginning, trying to get the right concept and early design, getting things into proportion and understanding issues of the project on-site and the climate, all those things. I'm involved all the way through. So, I visit the site during construction, but I spend less and less time of my time on it, other than overseeing the work of my team.
NA: Tell us about some of your past work leading up to the Burj Khalifa, obviously that kind of job doesn't just come from nowhere.
Well, we've done a whole series of tall buildings throughout my entire career now, but it was interesting, a few years before the Burj we had done a tower in Seoul, South Korea, called Tower Palace Three, and it was a residential building and it had three wings [like the Burj Khalifa]. It had a different structural system but the same global geometry – it doesn't look anything like the Burj – and some attributes that were the same as the Burj.
The initial design that we tested in the wind tunnel and did our initial engineering on was about 92 stories, and the building was behaving very, very well. Eventually because of complaints from the neighbors and so on, the tallest of three wings was pushed down to about 72 stories. But we'd made it work for the 92 stories. So from studying that building we got a lot of ideas about how we could go very tall. That was a very important precursor to the Burj.
Then when the Burj came in, it was a competition, and so I flew to New York with a couple of my partners and one of our senior managers and we met with these clients from the Middle East representing a developer called Eemar and so we talked about our experience, stuff like that.
We were having dinner across the East River in Brooklyn, looking back to the skyline in Manhattan and talked about, to the clients, how will you choose? And our suggestion, which they took up, was just do a two week idea competition from various tall building architects and in the end our scheme was selected – though it changed quite a bit from our initial proposal to the final building.
NA: Before the Burj, the world's tallest skyscrapers were being built with relatively minor height increases. It took around a quarter century for the Sears/Willis Tower to be surpassed – and even then it wasn't by much. Then you come along make this huge leap with the Burj, how was that possible?
Yeah that is true, y'know up until that point the tallest building was the Taipei 101 and each of these buildings were just, a lot of the time, poking up their spire to get a little taller, so it was very small incremental increases.
In fact if you go back to the Empire State building in 1929/1930 to the Taipei 101, it really wasn't a very steep hill, the increase in height was fairly modest for every building that was the record holder. Then we did the Burj Khalifa and I think we grew by 60 percent taller than our predecessor and a lot of that was because we did a lot of our design in the wind tunnel.
We kept going back and forth, it was a very collaborative process with structural engineers, architects, interior designers, M&E [Mechanical and Electrical] Engineers, vertical transportation [elevator] consultants, wind engineering consultants.
So we did a great many studies on how we could shape the building. The single most important structural parameter in a tall building is the architecture, by that I mean the shape of the building, because these tall buildings, particularly in that region, are controlled by the wind and depending on the shape of the building, you can have a huge difference.
In fact, in the last couple of years we actually built our own in-house wind tunnel that we used for conceptual designs. So we'll go down there with half a dozen different variations on a shape to see what's better and then eventually we'll go to a commercial wind tunnel to get the proper design forces, because we can't do the climatic and probabilistic calculations as well as they can.
But you'd be shocked at how some small changes in the shape of a building design can make a dramatic increase or decrease in the forces on on the building. And you can see that on a lot of projects that it's quite clear if the team understood or didn't understand that issue when they shaped the tower.
NA: That seems kind of similar to the way car designers will play with minor shape changes to make the vehicle more aerodynamic.
Yeah, in fact one of the wind tunnels we often use is near Heathrow Airport, I think they do a lot of testing for McLaren or somebody, they have like a locked room that they don't even have a key to, y'know? [laughing] Where McLaren goes in and tests very small refinements to the shape.
Though it's quite a different problem, when you have an automobile or airplane it's a different flow regime. A tall building is more about turbulent flow, we have something called a boundary layer, which is the wind near the surface is much slower as you get high up in the atmosphere. So, the higher you are the faster the wind. And the wind coming at you is fairly turbulent.
So we actually do our testing on what are called boundary layer wind tunnels, which try to replicate that surface boundary of the wind. That's what our wind tunnel here at the office here is, a boundary wind layer tunnel.
NA: The Burj Khalifa is a fairly unusual-looking building. Can you explain the basic idea behind its design and structural system, in layman's terms?
We kind of started with the system that we had on Tower Palace Three but determined after not too long that it really wasn't appropriate for this particular use, what we'd call in the US program, or in the UK you'd probably call a brief. It really wasn't working.
Anyway, so we kept on revising the systems. There are certain issues that you're looking at: you want it to be stiff, you don't want it to twist, you want it to be very stiff torsionally, it has to work with the interiors. We did a lot of work with the interior architects on how our structure interfaced with the interior layout and worked with the architects on how it interfaced with the shape of the building, stuff like that.
We set ourselves certain rules because it's such a big project. Rule number one was stay on module: to have a rigorous grid to the building, and avoid transfers. Those kind of rules help keep the project under control.
After a while, the system morphed and at a certain point we said, okay what is it we have, what's the essence of the system? And we gave it a name and called it a buttressed core.
One of the things I've found very helpful in my practice is at some point when the design is starting to solidify, when the idea is starting to gel, is to describe it in words. You're a writer and you know that if you have a jumble of words, too many words, you're not there yet, it needs to be edited. So, if it takes a lot of words to describe your system, you may not be there yet, it's too complicated.
And if you can distill it down to the essence, a lot of times you find you can sometimes get to a noun plus an adjective. If the idea is that clear, you can then explain the idea to the rest of your team, you can explain it to all your colleagues on the design side, explain it to the contractor, the owner, and so it also helps to resolve conflicts.
Tall buildings are amazingly complicated and you've got a lot of stuff going on with these buildings and there are always conflicts, so how do you decide who gets the right of way, which system is dominant. So it helps to kind of define a hierarchy, what is important.
Now, we call it a buttressed core. Around the elevators and center of the building, we have this kind of irregular hexagon and that acts like an axle, it keeps the building from twisting and that's very, very important – you don't want a building that's twisting in the wind. Besides some not-so-nice structural behavior, you'd also have this unusual phenomenon where you'd look out the window and the horizon's moving side to side and you really don't want to do that.
So we put this concrete core around the center elevators and stairs, so that took out the twist but it was much too slender to go to great heights, so we decided to create buttresses, like you see in a cathedral. We buttressed down each of the three wings in order to stiffen, to give more structural width to the building.
It's like if someone is pushing on you sideways and your feet are together, you'll instinctively spread your feet apart to stabilize yourself and that's what this is doing.
So, growing out of each of the three wings, it almost looks like an I-beam [a kind of girder] or half of an I-beam or something. You have a double web going down either side of the corridor and then you have cross walls, we call them flanges of beams. A beam looks like an I-shape, so we call the top bar and the bottom bar the flanges, and we call the vertical bar the web, and so we have a double web going down the corridor then in between the units we have these flanges.
Going back to the wind engineering, we did a tremendous amount of study on the wind and so one of the things you do to minimize wind force is to shape the architecture. We shaped it in such a way that the columns on the top would sit on the walls below so that there is no transfer.
So the wind tunnel is very important, in fact right after we finished the competition, we went immediately into the wind tunnel with our competition scheme and the building did not behave very well. The forces were quite large and the motions were quite large, and so we went through a fairly dramatic redesign. Aesthetically you wouldn't say it was a lot different but structurally it was a lot different.
So just by tuning the shape, actually rotating the building sixty degrees, stuff like that. So as we're doing that, the forces drop dramatically, the motions dropped dramatically. Quite frankly, if we hadn't had such a bad first behavior, we may not have changed it so much and been able to go so high. It's kind of an odd thing if you think about it, if it had just kind of worked we might have stuck where we were.
Because our first model in the wind tunnel was about 518 m (1,699 ft), slightly taller than Taipei 101, and during the design phase it grew from 518 to 828 m (2,716 ft), so it grew by 310 m, which is slightly bigger than the Eiffel Tower. Roughly, it grew by 1,000 ft during the design phase, and a lot of that growth happened after the foundations were laid. We were able to continuously refine the shape a little bit and reduce the forces.
NA: I imagine a project like this must be a collaborative process, with lots of different concerns and ideas being passed back and forth between architect and engineer, the construction team and others, is that correct?
Oh yes, tremendously collaborative. We all sit together in one office, well the consultants aren't here, the vertical transportation and wind people aren't here, but the rest of us are here. So we have almost constant meetings, for instance like on the core, the architects would propose a core [the Burj Khalifa's chief architect was Adrian Smith] and we'd say well that doesn't work for us and we'd counter-propose or they'd say that doesn't work for me and counter-propose.
So we'd go back and forth and finally we came up with something that worked for both of us. These things are Swiss watches: it takes a long time to get them efficient and effective.
NA: I was surprised at how efficient the Burj Khalifa is. I read the amount of steel used is proportionally less than the Empire State building. This seems deliberate; is efficiency something that drives your work?
That's not that fair a comparison because the Burj is mostly a concrete building whereas the Empire State is mostly steel, so it's a little apple and oranges, but it is a very efficient tower.
It's efficient in several ways: as far as material quantities it's fairly efficient, as far as constructibility it's very efficient, because we designed it based on how we assumed the contractor might want to build it.
So we kind of designed the building around what we knew about formwork systems, what you'd call [in the UK] shuttering systems, so that it could be built like a vertical factory and so that's how it was done.
In any project, particularly a very tall building, you might say time is your enemy; the longer it takes to build a project, the more things can happen: you can have political change, you can have economic change – all kinds of things can happen. Critical people can get ill, stuff like that. So it's very important with something of this scale that you think a lot about construction and think about how you can build this thing quickly and efficiently.
When I was working on the Burj it was kind of a hot time on tall buildings and there were a lot of proposals for buildings of the same height, similar height, or even higher, and they all got cancelled except for the Burj. So I did a little study, why do these buildings not get built? I think a lot of it was because they did not understand the issues of scale or the issues of constructibility – something like this you really have to have efficiency of construction.
NA: I've been looking at some of the old documents on the Burj Khalifa and read about mixing ice in the concrete because of high temperatures, clever ideas like that. What other notable challenges did this project pose that you had to overcome?
The successful bidder was a consortium of Samsung Construction, they're known as electronics of course but they also have an important construction firm, with BESIX, a Belgian firm, and Arabtec, an Emirati firm and they were the successful guys and they were all very, very good. They would anticipate a lot of problems before they became problems. The ice in Dubai is something that was developed before we got there.
The ground water in that part of the world is very corrosive. It has more salt and more sulphates than seawater, and the salts try to eat the rebar and the sulphate tries to eat the concrete, and so the solution to that is to have very high quality concrete that is very dense. So before we had gotten there, it improved while we were there but Dubai already had pretty high quality concrete systems including putting shaved ice into the mix. Because, if during the curing if the concrete gets too hot you can literally boil it and that's not good, what you end up with is not what you intended.
Another one which I thought was quite clever that Samsung addressed was the issue of pumping the concrete, say 600 m (roughly 2,000 ft) in the air. So I was scratching my head wondering how they'd test that. So what they did was, in the engineering of hydraulics of pipelines and stuff like that, every time a pipeline has a bend or curve in it, a 180-degree turn, you get a pressure loss and that's calculable and so what they did was they ran pipes back and forth across the desert floor there with enough bends in it that they got the equivalent pressure loss of pumping concrete 2,000 ft in the air.
So they went through a whole different series of concrete mix designs to see how they would modify the mix so that it could pump that high. Then they used a good German pump from somewhere near Stuttgart called a Putzmeister, which I think is a great name for a pump.
So the Putzmeister, it's like a piston pump, could actually pump the concrete 600 m in the air. So, you have this Putzmeister pump and then they tuned the concrete mix so they could verify that they could actually do it – and they did it.
NA: How big an issue does the wind pose for a building this size? What about lightning too, could it pose a problem for the Burj Khalifa in a bad storm?
We looked at very, very rare windstorms and the building's fine, there's no strength problem even in wind storms that happen well over a thousand years, multiple thousand years. We don't know what kind of storm that is, but we statistically extrapolate the recorded winds to very rare dates and it's fine. Even under normal storms it's a very quiet building. The building's instrumented and the data we get, the feedback we get is that it's very very quiet, so it doesn't move a lot.
Now, it gets hit by lightning all the time. Sometimes you get these storms coming in and every five minutes it gets hit by lightning. It's kind of like a lightning rod for the city of Dubai.
We knew it was going to be hit by lightning, so it is essentially a Faraday Cage. Michael Faraday was this 19th Century researcher in electricity and stuff like that. With a Faraday Cage, if you have a metal cage around an object that is grounded and the metal cage is like the cladding, the metal cladding around the building, and it's grounded, the lightning will not go inside the object.
Most of the lightning strikes I've seen have been onto the spire and I talked to the operations people over there and there's really no damage. We tied the exterior of the building into the rebar cages that are in the concrete and then it goes all the way down into the foundations and I think we have stainless steel bars that are down into the soil to dissipate the lightning. Usually we use copper grounding but the groundwater is so corrosive that we had to go with stainless steel.
NA: I visited Rome a while back and was struck by how well some of those ancient buildings have lasted over time and wondered about the lifespan of modern buildings. Do you think, all being well and assuming a decent level of maintenance, the Burj Khalifa could last so long?
Yeah, compared to other structures it could last indefinitely if you could keep the water out and maintain it. Bridges can get worn out because of what's called fatigue. They're loaded, like the trucks are heavy and the steel gets worked back and forth. Like, if you take a paperclip and work it back and forth eventually the paperclip will break. And then if you're in a high seismic area, the longer you wait the more likely you are to get hit by a big earthquake and you might have some serious damage to the building, but Dubai is fairly low on seismicity.
There's seismicity across the gulf over in Iraq but by the time the earthquake gets to there you might feel it in the building but there are really no structural issues. And one of the issues is the groundwater is so corrosive, but we did some very important protection of the foundations to give them a long life and we actually have a monitoring station inside the building to monitor if there's any corrosion in the foundations and there are ways to address that if it happens in a few hundred years or something.
NA: The Burj Khalifa is approaching a kilometer tall and it doesn't take too much imagination to see the kilometer mark being passed. Do you think it would be feasible to work on a mile-high building in the future?
Yeah definitely. I don't think it's a structural problem. I think I would use a different structural system to the Burj. I think the Burj's buttress core system runs out of gas at about 1.2 kilometers (0.75 miles). Above there and I'd probably come up with a different system and try out some different ideas. But the issue is probably more vertical transportation and the pressure change.
You actually go through different climates. Even on the Burj you can notice it quite a bit. During construction we would go up on the outside of the building, on the man hoist, and in the middle of the summer it would be unbelievably hot down on the ground and when you get up to like level 50 you could tell the difference, get to level 160 and it was pleasant, the air is cooler, there's less humidity, there's less dust in the air and so it was quite amazing, but you do go through a pressure change.
Denver is the mile high city, so its kinda like if you took an elevator from New Orleans to Denver. So we had to figure out how to deal with the pressure change. Which is true in all tall buildings. The technical issue is not structural, it's probably more vertical transportation and how you deal with the pressure differences.
NA: As a follow-up to that question, are we nearing the practical limit of how tall skyscrapers can be built?
I think we're probably reasonably close to the practical limit, other than a few outliers. If you look at the economic advantage that the UAE, or the city of Dubai, the developer Emaar got from the Burj Khalifa, it's huge. Probably the best investment they could have made. A huge amount of tourists, traffic, the building itself was sold during construction. All the properties adjacent to it went way up in value, any building that has a view – it's actually called the Burj effect – any property that you can actually see the Burj from is much more valuable than any other views.
It's like if you imagine you were somehow a developer in Manhattan and you controlled all the views of the Empire State Building or the Chrysler building can you imagine how much money you'd make? So it was a huge success for them so one could see where one might do a Burj-like investment, but except for that kind of special case, I think we're pretty close to the practical limits of a building that's not for that special purpose.
NA: With your experience, you have excellent insight into building and engineering projects. What building, skyscraper or otherwise, that you didn't work on personally do you admire as an engineer?
Actually one of my favorite buildings of all time is the John Hancock here in Chicago [pictured below], I just love it. It was by Skidmore, but was before my time. The structure's expressed as the symbol of Chicago. A few years ago, each of the states in the USA would have their own quarter, Wyoming would have a quarter, etc. On the Illinois quarter, they did this a few times, but on the backside, you could recognize two images: one was Abraham Lincoln, the other was the John Hancock.
Even though it's not as tall as the Sears Building, it's probably the most loved building here in Chicago, and the shape, the structure and also the use of it. There's a nice restaurant near the top, an observation deck near the top, it has a lot of residential, and the bottom part is retail. It's really a city in itself, it's really quite a unique building.