The possibility of terrorism against our nation’s bridges is an ever-increasing threat in today’s society. Data collected by the Mineta Transportation Institute indicate that at least 53 terrorist attacks worldwide specifically targeted bridges between 1980 and 2006, and 60% of those attacks were bombings. Moreover, accidental collapses of bridges in the U.S. and terrorist attacks against bridges in Iraq illustrate the large economic and socio-economic consequences of catastrophic bridge failures. To help ensure the safety of bridges in the U.S. and protect the nation’s infrastructure, there is a need for design and detailing guidelines for blast-resistant highway bridges. To address this need, the National Cooperative Highway Research Program (NCHRP) funded NCHRP Project 12-72 to investigate highway bridges subjected to explosive loads. Because bridge columns are integral to nearly all bridges regardless of the superstructure type, and because the loss of a key column could compromise the integrity of most bridges, the research team elected to focus its effort on reinforced concrete columns. The main goals of the research were to:
• Investigate the response of concrete bridge columns subjected to blast loads,
• Develop blast-resistant design and detailing guidelines for highway bridge columns, and
• Develop analytical models of blast-load distribution and the resulting column response that are validated by experimental data.
The research program consisted of two different phases of experimental testing, complemented by computational and analytical modeling. Phase I of the experimental test program included small-scale blast tests on square and round non-responding columns. These tests were conducted by the Engineering Research and Development Center of the U.S. Army Corps of Engineers at their test site in Vicksburg, Mississippi, and they were designed to determine how shock waves interact with slender structural members so that variations in pressures and impulses with both time and position along the height of a column could be studied. Phase II of the experimental testing program included close-in blast tests on half-scale reinforced concrete columns. These tests were conducted at a remote test site with help from Protection Engineering Consultants (PEC) and the Southwest Research Institute (SwRI). The test matrix included ten half-scale columns with three designs: a base design that represented a national survey of current bridge column specifications, a seismic design that reflected the current seismic detailing practices, and a blast design that consisted of a very dense mesh of transverse reinforcement. Along with these three designs, the five main parameters that were varied during the Phase II tests were scaled standoff, column geometry, amount of transverse reinforcement, type of transverse reinforcement, and splice location. The ten columns were tested with charges placed at small standoffs, and these tests were designed to observe the failure mode (i.e., shear or flexure) for the different column designs. Each of the columns tested during the Phase II experimental program contained six strain gauges, and the strain data gathered during the tests were used to verify boundary conditions and blast-load distribution for each small standoff test. Three of the ten columns tested during the small standoff tests experienced significant shear deformation at the base, and the other seven columns displayed a combination of shear and flexural cracking. Six of the columns were re-tested using close-in or contact charges, and the objective of these local damage tests was to observe spall and breach patterns of blast-loaded concrete columns. Two of the six columns tested during the local damage tests experienced significant breach, and the other columns experienced spalling of concrete from their side and back faces. Computational and analytical research extended the knowledge gained from the experimental tests. This work included the use of simplified methods for predicting loads and response as well as detailed, nonlinear finite element analyses. The simplified models used widely available analysis procedures, including empirical models for predicting blast loads and single-degreeof-freedom models for computing response. The experimental and analytical data gathered during this research program provided a basis for developing detailing guidelines and a general design procedure for blast-resistant bridge columns. These guidelines consist of three design categories, each of which contains increasingly demanding design requirements that specify greater amounts of transverse reinforcement as the design threat increases. Design Category A is the least restrictive of the three, requiring no special consideration for design threats associated with blast loads. Design Category B requires seismic detailing with the exceptions that larger than currently specified embedment lengths on the hooks of transverse reinforcement be used and plastic hinge detailing be used over the entire column height to account for uncertainties associated with the location of potential threat scenarios. Design Category C is for cases involving the most severe threats. Columns in this category must have a transverse reinforcement ratio that is 50% greater than that required for current seismic detailing and even larger embedment lengths on the hooks of transverse reinforcement than those used for Design Category B. Design Category C also requires a designer to conduct a single-degree-of-freedom dynamic response analysis to ensure that the ductility ratio and support rotation are less than recommended limits. Test results indicate that measures other than structural hardening should be taken to help mitigate design threats with very small scaled standoffs corresponding to contact (and near-contact) charges because local damage failures such as breaching will begin to control response for these cases. All of the columns tested during the small standoff tests in this research fell into Design Category C, and the results of those tests, along with analytical work, provided the basis by which the limits used to define each of the design categories were established. The following general guidelines are recommended for increasing the blast resistance of reinforced concrete bridge columns:
• Column performance improves significantly as standoff increases, and one of the best methods to reduce risk to a bridge is to increase the standoff. As such, the design categories allow a designer to capitalize on the fact that increasing standoff through the use of bollards, security fences, or vehicle barriers is a safe, efficient, and cost-effective alternative to increasing the design category and detailing requirements.
• Experimental and analytical data show that circular columns experience less net load than square columns subjected to the same charge weight and standoff distance. Therefore, bridge columns should employ circular cross sections whenever possible to reduce the applied net load of any threat. It should be noted, however, that square columns can be made blast resistant to an acceptable level with proper detailing and sizing.
• The cross-section size of a reinforced concrete column plays an important role on column performance, as this parameter controls the onset of breaching failure. As such, 2 increasing cross-section size will generally improve column resistance to blast loads, and a minimum column diameter of 30 in. is recommended for columns subjected to close-in blasts.
• Continuous spiral reinforcement provides better confinement and produces better overall behavior than discrete hoops for small standoffs and close-in threats and is preferable at all times. If design or construction constraints prevent the use of spiral reinforcement, columns should use hoops with an embedment length larger than that typically used for standard (i.e., gravity-controlled) and seismic designs.
• Increasing the transverse reinforcement ratio improves column response to blast loads. Experimental observations show that the blast designs, which employed very dense meshes of transverse reinforcement, performed better than the seismic designs, which in turn performed better than the gravity-load-controlled designs. Accordingly, the minimum amount of transverse reinforcement for a Category C column is now specified to be 50% greater than that which is currently specified for seismic designs, and this transverse reinforcement should extend over the entire height of the column to resist various potential threat scenarios.
• If possible, splicing of longitudinal reinforcement should be avoided, and removing splices from regions where charges may come into contact with a column can help minimize localized blast damage.
Finally, bridge engineers should remember that, while these design and detailing guidelines will increase the blast resistance of a bridge column, it is not possible to design a column to resist all possible threats. Bridge owners should conduct a thorough risk analysis to determine the most probable threats and have their bridges designed accordingly. A certain level of risk must be accepted for the extreme threat scenarios considered as part of this research. Failure should be expected if a sufficiently large quantity of explosive is placed close enough to any bridge column, regardless of the design and detailing.