U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
|This report is an archived publication and may contain dated technical, contact, and link information|
Publication Number: FHWA-RD-02-086
Date: August 2006
The four bridge deck overlay sites were visited by contractor personnel for the first time in 1994. During the visits, the test areas were laid out, corrosion testing was performed, samples were taken, and the overall condition of the overlays was determined. In most cases, these visits were used to establish a detailed initial condition of the structures, as no distress maps of the overlays were included in the original SHRP report, and only a few specific observations of the test sites were included. The primary testing performed and reported under SHRP C-206 included direct tension bond strength testing, bond shear strength, elastic modulus, compressive strength, split-cylinder tensile strength, scaling resistance, American Association of State Highway and Transportation Officials (AASHTO) T 277 "coulomb," and concrete temperature monitoring.
The work performed during this initial inspection included detailed visual observation and distress mapping, delamination surveying, half-cell potential surveying, linear polarization testing, chloride sampling, and shear bond testing. Available information was gathered from the State highway agencies. In the second, third, and fifth years only the detailed visual observation and distress mapping, delamination surveying, and chloride sampling were performed. In the fourth year (1997) the work performed included all of the testing performed during the initial inspection.
The distress mapping was performed in general accordance with the SHRP Long-Term Pavement Performance (LTPP) Distress Identification Manual, with modification as required to adapt the pavement-oriented manual to use for a bridge. The SHRP LTPP manual is oriented primarily towards structural distress of pavements cast on grade and does not address certain problems related to bridges. For example, the LTPP manual requires a listing of "corner breaks," "lane-to-shoulder dropoff," and "joint spalls," none of which would occur in a bridge deck. Also, bridge decks frequently exhibit frequent narrow nonstructural transverse or longitudinal cracking. Although this is covered by the LTPP manual, it is far more important on a bridge deck than on a jointed pavement.
The delamination surveys were conducted with a chain drag supplemented by manual rodding or hammering as described in the American Society for Testing and Materials (ASTM) D 4580, Standard Practice for Measuring Delamination in Concrete Bridge Decks by Sounding. In areas where the chains or rodding indicated delaminated areas, hand hammering was used to determine the extent of the delamination.
The half-cell potential survey was performed as described in ASTM C 876, Standard Test Method for Half-Cell Potentials of Uncoated Reinforcing Steel in Concrete, using a copper/copper sulfate reference cell and an automatic data recorder. In the test, the voltage difference between the reinforcing steel and the reference electrode is measured using a voltmeter. The measured voltage difference changes with the changing corrosion state of the reinforcing steel, allowing it to be used as an indicator of the corrosion state of the reinforcement. The test is performed by connecting the positive terminal of a high-impedance voltmeter to the reinforcing steel and the negative terminal to the reference cell. On the bridges, all of the connections to the reinforcing steel were made by coring through the concrete to the upper mat of reinforcing steel, removing the concrete around the bar with a cold chisel, and attaching to the reinforcing bar with locking pliers, an electrical clamp or by drilling and tapping into the steel. Once the connections were made, the readings were taken by holding the junction sponge of the reference cell in contact with the concrete and monitoring the observed voltage until it stabilized, at which point the voltage was recorded. At the bridges, all of the readings stabilized quickly and were reproducible within the ASTM-required 10 millivolts (mV) when retested. The high stability of the readings was probably due to the relatively moist condition of the concrete in the bridges, as stability is typically related to concrete moisture content.
The linear polarization testing was performed using a computer-controlled potentiostat using a surface electrode assembly patterned after that used by a "3LP" commercial device. The polarization resistance technique for the determination of corrosion rates has been used by corrosion engineers for a relatively long period of time. However, it has only recently been used for more than extremely limited work in reinforced concrete. Polarization resistance uses simplified electrochemical corrosion theory to estimate corrosion rates of metals in corroding systems, using the electrical properties of the system to determine the rate of the corrosion. This is accomplished by forcing the area under test to deviate slightly (20 mV) from its equilibrium corrosion potential while measuring the electrical current required to make the change in potential. Because the specimen potential and the corrosion current are approximately linear over the small potential range measured, the measured changes in potential (ΔE) and applied current (Δiapplied) and the Tafel slopes (βa and βc) can be used to determine the corrosion rate (icorr) of the system using the following equation:
The technique is limited by the accuracy of the test data, and their interpretation. In addition, the characteristic Tafel slopes of the anodic and cathodic portions of the specimen's current versus potential relationship, βa and βc , must be assumed. Although the equation is relatively insensitive to the β values, and a typical assumption of 0.12 volt (V)/decade is reasonable for most steel-based systems, the resulting estimates of the corrosion current, icorr can be incorrect by up to a factor of 3 or 4. Compensation is used to account for the voltage drop due to the concrete resistance ("IR compensation," or current (I) times resistance (R)), but it is not always entirely effective. The greatest limitation of the technique is its poor performance in high-resistance media, such as dry concrete. Under those conditions, the voltage output of the potentiostat required to penetrate the concrete can overshadow the voltage required to completely polarize the metal under test, and the IR compensation may not be able to sufficiently correct for the voltage drop between the embedded metal under test and the counter and reference electrodes on the concrete surface.
Practically, the PR test is performed by first making a connection to the reinforcing steel and monitoring its potential while an external voltage is applied via a counter-electrode on the concrete surface, as shown schematically in figure 1. The test is performed by changing the voltage applied between the counter electrode and the reinforcing bars to produce the 20 mV potential shift (ΔE) between reinforcing steel and the reference cell described previously. The current (iapplied) flowing between the counter electrode and the reinforcing steel is measured and used to compute the corrosion rate using the equation shown above. In reinforced concrete, some question arises regarding the normalization of the estimated currents to a corroding area.
Figure 1. Schematic drawing of polarization resistance test setup.
For this work, the polarized area was assumed to be the surface area of the upper half of the bar immediately below the counter electrode. The PR scans were performed using IR compensation at each step to determine and correct for the potential drop due to the concrete resistance. The PR scans started at a potential 10 mV below the equilibrium rest potential of the area under test and continued to a potential 10 mV above the equilibrium potential. Following the testing, the polarization resistances were determined by fitting a line to the recorded potential versus current plots. Using the polarization resistances, the corrosion rates were estimated by assuming Tafel slopes of 0.12, assuming that the upper bars were size number 4, and assuming that the upper half of the reinforcing bar directly under the counter electrode was polarized during the testing.
The chloride sampling was accomplished using cores instead of drilled powder samples. The contractor has found removing cores, followed by slicing and grinding the slices, to be a more accurate method for determining concrete chloride contents than removing powder samples using a rotary impact drill. Once the samples were cut and ground, they were analyzed by either an acid- or water-soluble chloride extraction followed by a potentiometric titration.
The shear bond testing was performed using a shear bond guillotine. All of the cores were bedded in a plaster-based material before testing to allow for uniform load applications.
The four bridge deck overlay sites were revisited for the annual inspections, as described in Task E, Field Evaluation of SHRP C-206 Test Sites of Project DTFH61-94-R-00009. The visits were usually in the fall. The cumulative results from the 5 years of inspections are detailed in the following sections for each individual site.