«R. SHANKAR NAIR R. Shankar Nair R. Shankar Nair, Ph.D., P.E., S.E. is a principal and senior vice president of Teng & Associates, Inc. in Chicago. In ...»
Indeed, an explosion that could take out the 2000 lb/ft column would likely destroy several of the W8 columns, making one-column redundancy inadequate to prevent collapse in that case. And yet, codes and standards that mandate redundancy do not distingu ish between the two situations;
they treat every column as equally likely to be destroyed.
In fact, since it is generally much easier to design for redundancy of a small and lightly-loaded column, redundancy requirements may have the unfortunate consequence of encouraging designs with many small (and vulnerable) columns rather than fewer larger columns. For safety against deliberate attacks (as opposed to random accidents), this may be a step in the wrong direction.
In this approach, susceptibility to progressive/disproportionate collapse is reduced by providing critical components that might be subject to attack with additiona l resistance to such attacks.
This requires some knowledge of the nature of potential attacks. And it is very difficult to codify in a simple and objective way.
Interconnection or Continuity
This is, strictly speaking, not a third approach separate from redundancy and local resistance, but a means of improving either redundancy or local resistance (or both). Studies of many recent building collapses have shown that the failure could have been avoided or at least reduced in scale, at fairly small additional cost, if structural components had been interconnected more effectively. This is the basis of the “structural integrity” requirements in the ACI 318 specification (ACI, 2002).
Approaches Used in Codes and Standards
The following tabulation shows which of these approaches to preventing disproportionate collapse are used in each of the five codes and standards discussed previously. Redundancy is the clear favorite, being the primary approach used in three of the five sources. [The rational threat-dependent analysis specified in the 2003 GSA PBS Facilities Standard could include any or all of the three design approaches.]
To illustrate the techniques for reducing susceptibility to disproportionate collapse, consider how redundancy, local resistance or interconnection might have been used to improve the performance of Ronan Point, the Murrah Building and WTC 1 and 2.
Case Outline -- Ronan Point Greater redundancy would have been difficult to build into the type of structure employed in the Ronan Point tower. Improved local resistance, in the form of greater strength of the precast concrete wall panels that blew out, precipitating the collapse, would not have helped; the panels would have blown out regardless of their strength. Better interconnection of structural components is the key for this structure. Stronger and more positive connections between the wall panels and the floors, with less reliance on friction due to weight to hold everything together, is likely to have greatly reduced the scale of the collapse of the Ronan Point building.
Case Outline – Murrah Building
The columns at the front face of this reinforced concrete building were at 20-ft centers on upper floors and 40-ft centers at ground level, with a transfer girder to make the transition. A
requirement for one-column redundancy would almost certainly have eliminated the transfer:
The smaller columns 20 ft apart would have extended down to the ground and the structure would have been designed to tolerate the loss of one of them. Would this have reduced the magnitude of the collapse on 19 April 1995? Probably not. The explosion would almost certainly have taken out several (at least five) of the small closely-spaced columns, easily overwhelming the one-column redundancy built into the design, leading to a collapse not significantly different from what actually occurred.
Improved local resistance, within plausible limits, would not have prevented destruction of the ground-floor column closest to the bomb. But improved ductility and shear capacity of the columns (possibly through the use of the kind of reinforcing steel details used in earthquake prone regions), and better interconnection and continuity throughout the building, could have prevented the loss of any of the other large ground-floor columns and could have limited the collapse to a 60- to 80-ft width of structure from the ground to the roof — a major disaster but much less than what actually happened. The conclusion, then, is that the performance of the Murrah building on 19 April 1995 would not have been improved by a requirement for redundancy in the design, but could have been improved by better interconnection and continuity throughout the structure and different reinforcing steel details in the columns.
Case Outline – WTC 1 and 2
The exterior frame of each WTC tower was already so highly redundant that greater redundancy would be hard to contemplate. The interior columns were not redundant, except for the limited redundancy created by the hat trusses. But the impact and fire damage were so pervasive that greater redundancy in the interior is not likely to have changed the outcome. Greater local resistance (in the strictly structural sense, fire protection may be a different issue) was not a practical proposition for these towers. Finally, notwithstanding early reports to the contrary, connection failures do not now appear to have contributed significantly to the disaster, so improved interconnection would not have been useful.
The conclusion that none of the typical means of preventing disproportionate collapse would have been useful for the WTC towers reinforces the idea that the collapse of these buildings was not disproportionate to begin with.
Application of Codes and Standards to the Cases Considered A tabulation showing which of the approaches to preventing disproportionate collapse were used in each of five selected codes and standards was presented earlier. That tabulation is expanded below to show whether use of those codes and standards in the design of Ronan Point, the Murrah Building and the WTC towers would plausibly have improved the performance of those structures. The results (see the box in the tabulation) indicate that use of current codes and useragency standards would not consistently provide assurance against the types of collapse that occurred in those buildings — not even against the clearly disproportionate collapse at Ronan Point or the “possibly disproportionate” collapse at the Murrah Building.
Progressive collapse is the collapse of all or a large part of a structure precipitated by damage or failure of a relatively small part of it. Prevention of progressive collapse is one of the unchallenged imperatives in structural engineering today. But in fact, a building’s susceptibility to progressive collapse should be of particular concern only if the collapse is also disproportionate. Indeed, the engineering imperative should be not the prevention of progressive collapse but the prevention of disproportionate collapse (be it progressive or not).
There are, in general, three approaches to designing structures to reduce their susceptibility to
• Redundancy or alternate load paths, where the structure is designed such that if any one component fails, alternate paths are available for the load in that component and a general collapse does not occur
• Local resistance, where susceptibility to progressive/disproportionate collapse is reduced by providing critical components that might be subject to attack with additional resistance to such attacks
• Interconnection or continuity, which is, strictly speaking, not a third approach separate from redundancy and local resistance, but a means of improving redundancy or local resistance or both The emphasis on redundancy over all alternatives in some recent codes and standards and useragency requirements may not lead to buildings that are less susceptible to disproportionate collapse as a result of deliberate attack.
The writer wishes to thank the staff of the American Institute of Steel Construction, especially Charlie Carter, for the considerable assistance they provided throughout the preparation of this work.
ACI (2002), Building Code requirements for Structural Concrete (ACI 318-02), American Concrete Institute, Farmington Hills, Michigan.
ASCE (2002), Minimum Design Loads for Buildings and Other Structures (SEI/ASCE 7-02), American Society of Civil Engineers, Washington, DC.
GSA (2000), Facilities Standards for the Public Buildings Service, P100-2000, General Services Administration.
GSA (2003a), Facilities Standards for the Public Buildings Service, P100-2003, General Services Administration.
GSA (2003b), Progressive Collapse Analysis and Design Guidelines for New Federal Office Buildings and Major Modernization Projects, General Services Administration.
Anatol Longinow, PhD, PE is an Adjunct Professor of Civil Engineering at Illinois Institute of Technology (IIT), Chicago, Illinois and at Valparaiso University (VU), Valparaiso, Indiana. Prior to his current affiliations he was a member of the firm of Wiss, Janney, Elstner Associates, Inc. (WJE) for 16 years. There he specialized in evaluating the effects of accidental explosions, vibrations, wind and seismic loadings on buildings. Before joining WJE, he was on the faculty of the Department of Civil Engineering at IIT and prior to that at VU. He spent 20 years with the IIT Research Institute (IITRI) where he specialized in the evaluation of the effects of nuclear weapons, HE weapons, wind and seismic loading on structures.
This article (a primer on blast effects) begins by identifying general military and most commonly used commercial explosives. The manner in which explosives release their energy is described and the primary blast parameters for explosions in air are identified. The emphasis is on blast from surface explosions. This is followed by a brief discussion of the interaction of the blast wave with building structures. Strength of buildings when subjected to blast effects of high yield (nuclear) explosions is quantified. This is followed by a brief discussion of internal explosions. The final topic is a brief presentation of results of studies dealing with casualties in buildings produced by external blast.
INTRODUCTIONThe purpose of this article is as a primer on the basic aspects of explosion phenomena as this applies to the effects of high explosives (military and commercial) and effects of accidental explosions from commercial substances such as natural gas, propane, and liquid fuels, etc., on structures and people.
Deliberate explosives come in two general categories, i.e., military and civilian or commercial. Military explosives include cased explosives such as bombs, mortar shells, bullets, etc., each designed for a specific form of delivery.
This category also includes uncased explosives such as various plastic explosives used for demolition and other functions. These are referred to as high explosives. Low explosives include such products as propellants.
Commercial explosives include such products as dynamite, TNT (trinitrotoluene) and Ammonium Nitrate among others. Ammonium Nitrate is an essential ingredient in nearly all commercial explosives. Its predominant use is in the form of AN prill, a small porous pellet with fuel oil. More than two million pounds of these mixtures, commonly referred to as ANFO (Ammonium Nitrate Fuel Oil), are consumed each year. They account for approximately 80% of the domestic commercial market.
ANFO products have found extensive use in a variety of blasting applications including surface mining of coal, metal mining, quarrying and construction. Their popularity has increased because of economy and convenience.
The most widely used ANFO product is oxygen balanced free-flowing mixture of about 94% ammonium nitrate prills and 6% No. 2 Diesel fuel oil.
Items which are capable of exploding, but whose primary function is not to act as explosives, include natural gas, propane, liquid fuels such as gasoline and many other chemicals. These are generally referred to as low explosives (Longinow, A., Alfawakhiri, F., 2003)
High explosives release their energy by a process called detonation, and low explosives, such as propellants, natural gas, propane, etc., by the process of rapid burning. The time required for the detonation of a quantity of high explosive is much less than that for the burning of a like amount of propellant. With high explosives, the rate of detonation is not markedly affected by the particle size; with propellants, grain size is all-important. The shattering effect of a high explosive detonation is great, that of low explosives much less so. These distinctions are not completely clear-cut, however.
A number of so-called low explosives can be made to detonate--even black powder, under great pressure, and proper conditions. The military (and terrorist) use to which high explosives are put depend on their great shattering power and their high rate of detonation.
Some high explosives, such as mercury fulminate for example, are very sensitive to heat and shock and can be easily detonated by a spark or other local application of heat. These types of explosives are used to initiate less sensitive explosives and are called primers. Other explosives less sensitive to heat and shock than primers are used as boosters, i.e., intermediates between the primer and the main body of the explosive. These are capable of being initiated by the former and of initiating the latter.
The quantities of these three types of explosives in a given weapon differ greatly. 1) A very small quantity of primer, usually less than one gram, is used; 2) the booster weight is ordinarily of the order of a pound to a few pounds; and 3) the bulk of the explosive content of a weapon, the insensitive part, may constitute over 99% of explosive.