The Collapse of Concrete Buildings: Dust Thou Art and Unto Dust Thou Shall Return.
We don’t know yet the causes of the recent collapse of the condo building in Surfside, Florida. But it is likely that the corrosion of the reinforced concrete was one of the main reasons that weakened the structure of the building. It is a subject that I described in one of the chapters of my book “Before the Collapse” (Springer 2019) that turned out to have been timely and, unfortunately, also prophetic. We may see many more of these collapses in the future
Extract from Chapter 3.1 of “Before the Collapse” (2019) by Ugo Bardi
In the late morning of August 14, 2018, I was busy writing this book when I happened to open my browser. There, I saw the images of the collapse of the Morandi bridge, in Genoa, almost in real time. It was a major disaster: the bridge used to carry more than 25 million vehicles per year and it was a vital commercial link between Italy and Southern France. When it collapsed, it not only took with it the lives of 43 people who were crossing it, but it was nothing less than a stroke for the Italian highway system, forcing the traffic from and to France to take a long detour. It will take years before a new bridge can be built and the economic damage has been incalculable.
How could it happen that the engineers who took care of the maintenance of the highway could not predict and contrast the collapse of such an important structure? Much was said in the debate that followed about incompetence or corruption. Perhaps the fact that maintenance of the highway was handed over to a profit-making company was a recipe for disaster: profit-maximizing may well have led to cutting corners in the maintenance tasks. But, on the whole, we have no proof that the company that managed the bridge was guilty of criminal negligence. Rather, the collapse of the Morandi bridge may be seen as another example of how the behavior of complex systems tends to take people by surprise.
Even in engineering, with all its emphasis on quantification, measurements, models, and knowledge, the phenomenon we call “collapse” or “fracture” remains something not completely mastered. If engineers knew exactly how to deal with fractures, nothing ever would break — but, unfortunately, a lot of things do, as we all know. We saw in a previous section how critical phenomena in a network can be initiated by small defects in the structure, it is the effect of cracks in real-world structures, according to the theory developed by Alan Griffith 100. The Morandi Bridge was a structure under tensile stress, sensible to the deadly mechanism of the Griffith failure.
The bridge went down during a heavy thunderstorm and that may have been the trigger that started the cascade of failures that doomed the bridge: one more case of the “Dynamic Crunch” phenomenon that leads to the Seneca Cliff. Somewhere, in one of the cables holding the deck, there had to be a weak point, a crack. Then, perhaps as an effect of a thunderbolt, or maybe of the wind, the cable snapped off. At that point, the other cables were suddenly under enhanced stress, and that generated a cascade of cable failures which, eventually caused a whole section of the bridge to crash down. You heard of the straw that broke the camel’s back, in this case we could speak of the lightning bolt that broke the bridge’s span. Complex systems not only often surprise you. Sometimes, they kill you.
But why was the Morandi Bridge so weakened? Just like many other bridges in Italy and Europe, it had been built using “pre-compressed concrete.” This is a material European engineers seem to like much more than their American colleagues who, on the contrary, tend to use naked steel cables and beams for their bridges. Pre-compressed concrete had more success in Europe because it was widely believed that concrete would protect the internal steel beams from corrosion and avoid the need for laborious maintenance work of painting and repainting required, instead, for steel bridges. But, over the years, it was discovered that steel corrodes even inside concrete, and that turns out to be a gigantic problem, not just for bridges.
In the case of the Morandi bridge in Genoa, the problem was known. The bridge had been opened in 1967 and, after more than 50 years of service, it needed plenty of attention and maintenance. Years before the collapse, engineers had noted that corrosion and the vibration stress caused by heavy traffic, had weakened the steel beams of the specific section that was to go down in 2018. A series of measurements carried out one year before the collapse had indicated that the steel in that section had lost 10% to 20% of its structural integrity. That was not considered to be dangerous enough to require closing the bridge to traffic, especially at the height of the busy summer season. After all, most buildings are built with a hefty safety margin with respect to their breakdown limit, typically at least 100%. But there was a plan to close the bridge for maintenance work in October 2018. Too late.
We see once more how the best plans of mice and men often go astray. The engineers who were working on the bridge may have made a typical mistake of linear thinking: they assumed that there is a certain proportionality between weakening and danger. In this case, they believed that a 20% weakening of the beams was not enough to cause the bridge to collapse. But that was an average, and complex systems may not care about averages: do you know the story of the statistician who drowned in a river of an average depth of 1.5 meters?
Bridges are just an example of the many engineered structures subject to collapsing under stress. The Griffith mechanism of crack propagation is typical of the fracture of structures under tensile stress, such as the beams of a suspension bridge, the beams of a roof, moving objects such as planes and ships, everyday objects such as bookshelves, and even the bones of living beings. These structures tend to go down rapidly, suddenly, and sometimes explosively, typical examples of Seneca Collapses. There also exists another category of engineered structures, those which must withstand only compression stresses: this is the case of pillars, walls, arcs, domes, and the legs of the chair you are sitting on. These structures can collapse, but are normally much safer than those under tension, because compression tends to close cracks instead of enlarging them, as tension does.
In ancient times, when reinforced concrete did not exist, buildings used to be made in such a way to avoid tensile stresses as much as possible. That was because the main construction material available in ancient times was stone, and stone just cannot take tensile stresses. So, stones can be used to build walls and buttresses, and also for bridges and roofs, provided that you arrange them carefully to form arcs and domes in order to make sure that all the elements are always under compression, never under tension.
But even compression structures have their limits. Ancient builders were perfectly aware that stone can crumble, even explode, when subjected to excessive stress. That generates a limit to the height of a building in stone: over a certain height, the stones at the base would burst out and bring the whole structure down. One of the arts that ancient builders needed to know was the capability of testing stones for their resistance to compression before using them, and they had developed sophisticated techniques to do just that. Maybe we are biased in our perception because what we see around us are only those ancient building which survived and arrived to our times, but it is true that many ancient buildings have survived the test of time beautifully, and are still around us after several centuries, even millennia.
Many Roman bridges are still standing and are used today. Another remarkable example of a building that survived from Roman times is the Pantheon temple, in Rome. It was built nearly 2,000 years ago and it still being used as a temple today, now a Catholic church. Gothic cathedrals built during the Middle Ages were also sturdy and resilient: there are only few examples of structural collapses caused by poor design. For instance, the Beauvais Cathedral, in France, built mainly during the 13th century, suffered lots of problems and some structural collapses, but it is still standing nowadays. Another example is the Pisa tower, in Italy, built during the 14th century. For centuries, it survived the bending caused by ground movements. During the 20th century, the bending had reached an angle of 5.5 degrees, bringing the tower to risk of collapse. Today, the tilt has been reduced to less than 4 degrees by acting on the foundations, and now the tower may well keep standing for more centuries in the future. Modern stone buildings are sometimes even more ambitious. The Washington Monument in Washington DC is an example of a building high enough (169 m) to be close to the limits of structural resistance of the stones at its basis. It was terminated in 1884 and seems to be still in good shape despite some cracks that it developed after an earthquake hit it in 2011.
But let us go back to the case of the Morandi bridge for a discussion on risk evaluation. I crossed that bridge by car several times in my life without ever even vaguely thinking that it was risky to do so. Probably, at least a billion vehicles safely crossed that bridge over its more than half a century of life, so the chance of seeing it collapse just when you were crossing it was abysmally low. Yet, it happened in 2018, and when a major bridge collapses, someone is bound to be crossing it. Obviously, it would have made no sense to avoid crossing the Morandi bridge, or any other concrete bridge, for fear that it could collapse. Yet, it makes perfect sense to consider the risk of collapse for a building that you use much more often than bridges: your home or the place where you work. Unfortunately, normally you have no idea of how well and carefully your home was built and maintained. Maybe all the standards were respected, maybe not and, in the second case, your life is at risk: the Seneca Collapse waiting for you could be rapid and deadly.
There are many cases when it was discovered, typically after the collapse of a structure, that the builders had saved money by reducing the amount of steel reinforcement for the concrete. Or maybe they had used poor quality sand; a typical trick to save money is to use sand taken from some beach. This sand is contaminated with sea salt and that favors the corrosion of the steel beams inside the concrete. In some cases, it is reported that instead of the standard steel beams, builders used wire mesh of the kind used for chicken coops.
Then, you have to consider that a building rarely remains untouched after it has been built. People open doors and windows in the walls, add more floors, remove walls or add them. They may also intervene in other damaging ways: for instance, everyone loves rooftop swimming pools, but they are heavy and may destabilize the whole structure of a building. These mongrel buildings may be very dangerous: one of the worse disasters in the history of architecture happened to a building that was modified and expanded without much respect for rules or for common sense. It is the case of the Rana Plaza collapse on April 24th, 2013 in Savar, a district of Bangladesh, when more than one thousand people died and more than 2,500 were injured. The owners had added four floors to the building without a permit (!!) and also placed the heavy machinery of a garment factory in those extra floors. Not only was the machinery heavy, but it also generated strong vibrations that further weakened the building. More than half of the victims were women workers of the factory, along with a number of their children who were in nursery facilities within the building. A good example of criminal negligence.
Building collapses are rare enough for the risk to be statistically low, so small that it is not normally listed in the various “Odds of Dying” tables that you can find on the Web. Yet, it is one of those risks for which you can take precautions and there is no reason for not doing so. If you live in a building made of reinforced concrete that is older than a couple of decades, you should check for the details that may indicate danger. In some cases, you can directly see the corrosion of the steel beams where the surrounding concrete has been eroded. Cracks in the walls are an evident symptom of troubles and it has been reported that the noise of a steel cable snapping open inside a concrete beam may be perceived as the noise of gunshots. In Europe, if you hear that kind of noise, you may reasonably think that there is something wrong with the structural integrity of the building you live in, but, of course, a different explanation may be much more likely in the US. By the way, the collapse of the Morandi bridge gave rise to noises that could be interpreted as explosions and — guess what! — that led some people to interpret the disaster as the result of a “controlled demolition” carried out by the evil “Zionist Illuminati” in analogy with the demolition theories proposed for the 2001 attack to the world trade center in New York. Human fantasy seems to have no limits in terms of crackpot theories.
Not seeing or hearing anything suspicious in a building does not necessarily mean it is safe. If it is older than 50 years, it would not be a bad idea to seek professional help to have it checked for its structural integrity. It is expensive, though, and not routinely done for private buildings. Stone buildings are normally safer and more durable than concrete ones; you have to be careful, though, because these buildings can crumble under the effect of lateral vibrations generated by earthquakes. Wooden houses are often said to be more resilient and safer than both concrete and stone buildings and that is probably true, within some limits. But take into account that wooden beams are susceptible to degradation, too: they may be attacked by termites and their presence may be difficult to detect because they eat away the interior of the wood before breaking through to the surface. In terms of structural safety, an Indian tepee or a Mongolian yurt would be the best choice for a place to live. Otherwise, you have just to accept that there are some risks in life.
In the end, the problem of concrete degradation is not with single buildings: it is a global problem that affects all the infrastructure built over the past century or so.
You see in the figure how cement production went through a burst of exponential growth from the 1920s all the way to a few years ago. Only in 2015 did the global production of concrete start to show signs of stabilizing and, probably, it will go down in the coming years. It means that our highways and our cities were built in a period of economic expansion and on the assumption that the needs for their maintenance would have been minimal, just as it had been for the previous generation of stone buildings. It turned out to be a wrong estimate.
In the future, we seriously risk an epidemics of infrastructure collapses if we do not allocate sufficient resources to the maintenance of their concrete elements. Otherwise, the result could be that a considerable fraction of the world’s buildings and roads will have to be sealed off and left to crumble. Worse, crossing a bridge or living in a skyscraper could come to be considered risky. It is already the situation you have in some poor countries. In Cuba, after the revolution of 1959, the government expropriated most buildings that had been owned by rich Cubans and foreigners, and distributed them among the poor. The problem is that these buildings had been erected using Portland cement made from beach sand contaminated with sea salt. Sea salt favors the corrosion of the steel beams — it is a very serious problem. It can be remedied, but it is expensive and requires sophisticated technologies that Cubans cannot afford today. The problems of old concrete buildings in poor countries do not seem to be related to a specific political ideology or government system. Puerto Rico is under the control of the American government but the problem of crumbling buildings seems to be the same as in Cuba, worsened in recent times by the Hurricane Maria that struck the island in 2017. Other areas with warm climates and close to the sea seem to be affected in the same way.
We lack worldwide statistical data for this kind of problems, but there seems to exist a “crumbling belt” of decaying buildings everywhere in tropical regions, especially near the sea, where higher temperatures and sea salt spread by the wind cause the steel beams of concrete building to corrode faster than in other regions of the world — incidentally, the Morandi Bridge was near the Mediterranean coast and it may well be that in that case too, sea salt had a role in the collapse. Add to that the fact that in many of these regions people are poor and unable to afford the costs involved in the remediation of these old buildings, and you have a big global problem: another Seneca Cliff awaiting.
In the end, the problem has to do with an old Biblical maxim: “dust thou art, and unto dust shalt thou return.” Applied to a concrete structure, it would sound more like, “sand thou art, and unto sand shalt thou return.” Concrete is nothing else than compacted sand, not unlike the sandcastles that children build on the beach. The substance that binds the sand in sandcastles is water, and when it evaporates the castle crumbles. In concrete, the binder is cement, and it is typically lime or calcium silicate. Of course, this kind of solid binder doesn’t evaporate and concrete lasts much longer than sandcastles, but not forever. So, what we are seeing today in Cuba and other poor tropical countries may be just an image of what our world will be in a not-so-remote future.