125 publications from this institution
Abstract Despite the low water to cementitious ratio used when preparing dry-cast reinforced concrete pipes, the concrete porosity of dry-cast reinforced concrete pipes has been found to range between 9 to 11 percent. These porosity values are higher than the porosity that is usually found on wet-cast concrete prepared with comparable low water to cementitious ratio. The high porosity could affect how chlorides are transported into dry-cast concrete. Recently, steel fibers have been suggested to reinforce dry-cast concrete pipes, replacing the traditional steel wire cage reinforcement. The corrosion resistance of steel fibers in this type of concrete is not known, nor the concrete durability to chlorides (marine) exposure. Steel fiber reinforced concrete pipes were prepared at a producers’ plant and cut segments were provided for this study. The goals of this investigation are: 1) characterize the transport properties of this concrete and 2) to assess a) whether the steel fibers are prone to corrosion, b) what is the chloride concentration that triggers corrosion, and c) to what depth into the concrete is corrosion observed. Rapid migration tests, modified rapid migration tests (e.g., longer exposure), and resistivity, porosity and sorptivity tests were used to characterize steel fiber reinforced concrete. Wet-cast concrete specimens with instrumented fibers were used to characterize the corrosion potential as the chloride builds-up around the fibers (via accelerated chloride transport). Selected specimens were terminated for forensic examination.
Two distinct binary blended concrete mixes were prepared for the study. The first mix involved a cement replacement of 50% slag, denoted as SL. The second mix incorporated a cement replacement of 20% fly ash, referred to as FA. No chlorides were added during the preparation of these concrete specimens. To accelerate chloride transport, electromigration was employed by placing specimens with varying reservoir lengths (ranging from 2.5 cm to 17.5 cm) on their top surfaces. These reservoirs were subsequently filled with a 10% NaCl solution. In this paper, corrosion propagation was monitored over a period of approximately 650 days using electrochemical measurements such as open circuit potential, linear polarization resistance (LPR), and electrochemical impedance spectroscopy (EIS). The evolution of rebar potential, polarization resistance, solution resistance, and corrosion current were analyzed to understand the corrosion behavior. This paper focuses on how the length of the solution reservoirs influences the corrosion-related parameters such as polarization resistance, solution resistance, rebar potential, and corrosion current. During the monitored propagation period, the corrosion current values (last 7 sets of readings) exhibited higher magnitudes for the embedded rebars in specimens made with SL mix in comparison to those made with FA mix. Corrosion current measurements likewise showed an increasing trend as the reservoir lengths increased. None of the specimens had any visible cracks or corroded products that could reach the concrete surface throughout the monitored period. The experimental results provide insights into the corrosion mechanisms and the effectiveness of accelerated corrosion techniques in simulating real-life conditions.
We describe recent computational and experimental studies on the corrosion properties of metallic coatings that can be tailored (tuned) to deliver up to three corrosion-inhibiting functions to an underlying substrate. Attributes are tuned by a selection of alloy compositions and nanostructures, ideally in alloy systems that offer flexibility of choice to optimize the corrosion-resisting properties. An amorphous Al-based coating is tuned for corrosion protection by on-demand release of ionic inhibitors to protect defects in the coating, by formation of an optimized barrier to local corrosion in Cl− containing environments, as well as by sacrificial cathodic prevention. Further progress in this field could lead to the design of the next generation of adaptive or tunable coatings that inhibit corrosion of underlying substrates.
Quantitative computational models have been developed to assist in predicting the course of corrosion-induced deterioration and the effect of corrosion protection models in reinforced concrete marine substructures. These models use the concept of a corrosion initiation stage (buildup of chloride ions until reaching a steel corrosion initiation threshold level) followed by a corrosion propagation stage (active corrosion leading to cracking and spalling of the concrete cover). Predictive models were applied to assess the reduction of corrosion rate by submerged and surface sacrificial anodes in marine piles. It was concluded that surface anodes located above water reduced corrosion significantly when the anode extended up to the top of the active steel zone. This predictive model was confirmed by parallel laboratory experiments with partially submerged piles and sacrificial anodes. Additional calculations applied the model to find the extent of cathodic protection feasible for partially submerged bridge footers with only submerged anodes. It was determined that useful protection may be obtained only when the concrete resistivity is very low. Modeling of the initiation stage of propagation resulted in the development of design derating factors to calculate the effect of diffusion geometries other than a flat wall (2- and 3-way corners, circular columns). Chloride ion binding by the concrete did not alter significantly the value of the derating factors, although it increased the time to corrosion initiation compared with a no-binding case. A final task of this work integrated the initiation and propagation phases of corrosion into a damage function model that predicts the amount of spalled area of a pile partially submerged in seawater as a function of service time. The quantitative formulation is suitable for selection of alternative corrosion protection strategies and incorporation in life cycle cost projection models.
The time-to-corrosion initiation (T i ) of reinforcement and the chloride threshold content (C th ) of a series of G109 concrete slabs were experimentally investigated. Results indicate that T i is increased by decreasing water-to-cement ratio or increasing cement alkalinity. The effects of a superplasticizer on initiation of corrosion of reinforcing steel bars (rebar) depend on concrete mix. Measurements reveal that chloride contents at both active (anodic) and passive (cathodic) sites increase linearly with the logarithm of T i regardless of concrete mix, whereas chloride contents measured at anodic sites are always greater than at cathodic sites. Defects at or nearby corrosion anodic spots could facilitate chloride accumulation: rebar corrosion would occur earlier than if no defect were present. Further analysis reveals that the weakest-link theory is applicable to predict the probability of initiation of rebar corrosion, as indicated by the Weibull distributions of T i and C th .
A computational code that was originally designed to model crevice corrosion was extended to multifunctional coatings on Al alloys exposed to thin layers of electrolytes. The model is able to calculate the transient distributions of potential, current density, and all chemical species concentration, enabling the dynamic simulation of inhibitor release, inhibitor transport, and sacrificial cathodic protection. The model has been applied to both inhibitor release from and aggressive anion capture by hydrotalcites (HTs) pigments in epoxy primer coatings applied to AA2024-T3. Computational studies were carried out to investigate the effects of HT/vanadate (HT/V) epoxy coating system parameters including scratch size, inhibitor release rate, Cl− gettering rate (GR), cathodic kinetics on the bare AA2024-T3, and solution layer thickness on system performance. The analyses of the computational results have quantified the important factors controlling successful corrosion inhibition by inhibitor release from coatings. The pH-dependence of the steady state inhibitor release rate was found to be the most important parameter controlling system performance. Cl− gettering can also reduce the aggressiveness of solution at long times, especially when considered in conjunction with inhibitor release. However, the ion exchange capacity required poses stiff design challenges involving the loading of the ion exchanger into the resin and the service conditions. The effectiveness of inhibition decreased significantly for the larger scratch sizes. The cathodic kinetics within the scratch play an important role in determining the ability of a given inhibitor to function effectively. When the scratch is the cathode in the galvanic couple with the substrate under the coating, inhibition was more effective. For the conditions simulated here, the net effect of a decreased solution layer thickness is to increase the protection ability of the system. The increase in the inhibitor concentration overcomes the decrease in the pH at the anode.
Abstract Flexible Fillers are specified by the Florida Department of Transportation instead of cementitious grout for post-tension tendons in bridges. The tendons are considered unbounded when a flexible filler is used. Several years ago, the anchorage of a bridge in the UK was not properly sealed and that allowed corrosion of the steel strand even in the presence of flexible filler surrounding the steel strands. There are concerns that voids could be present if the duct is not completely dry or if the filler is not pumped properly or due to temperature cycles some of the filler leaches out. If the anchorage is not properly sealed, then moisture and potentially rainwater with salts could penetrate. The current project aims to understand the corrosion initiation and propagation in cases in which water with salts or high moisture penetrates the duct during service. Samples prepared as part of previous project in which polycarbonate tube was used were drilled to create voids and different NaCl solution concentrations injected. This paper describes the findings after at least 5 months of environmental exposure.
Reproducible measurements of magnetic susceptibility χ m of laboratory and field extracted concrete core samples were achieved with simple instrumentation. There was a nearly linear relationship between χ m and the mass of fly ash per unit volume, or its volume fraction. The magnetic response of a given FA was not significantly affected by the process of curing and subsequent evolution of the concrete over two years, or by carbonation of the concrete. Field extracted concrete cores exhibited a wide range of χ m values. The group of specimens with the highest values of χ m also had the lowest chloride ion diffusivity, consistent with the presence of admixed FA. Conversely, specimens with nil magnetic response included those from concrete with the highest chloride diffusivity. The magnetic measurements provided reasonable order-of-magnitude indications of FA presence in field extracted cores. However, precise determination of FA content from magnetic measurements of field cores does not appear feasible in the absence of additional information.
Abstract Simulated deck slab samples cast with six rebars, three in the top row and three on the second row, are being monitored after 19 years of exposure. The rebars are different corrosion resistant reinforcement: 2304, 304, 316, two clad rebars (316 with carbon steel core) and an alloy with 12% Cr. The samples were exposed outdoors to 15% NaCl by wt% for at least 10 years (weekly wet and dry cycles). The concrete cover over the first row in most samples was one inch (2.53 cm). A variety of geometry and rebar surface preparation took place. Rebar potential and electrochemical methods were used to characterize current conditions. Most rebars appear not to be corroding. A couple of samples, containing clad rebar type 2 (two rebars were attached to create a crevice and the rebar ends were not coated), show cracks on the concrete reservoir surface due to corrosion of the rebar carbon steel core. Samples with 12% Cr also show corrosion signs; in this case an artificial crack reduced the concrete cover and two rebars were attached to each other to create a crevice. There are a few other samples with rebars that have no cracks and appear to be corroding based on corrosion rates. The rebars in these samples were either pickled or wire brushed with a carbon steel metal brush (this likely caused small crevices).
Surface resistivity profiles as a function of elevation are being measured on recent and mature marine reinforced concrete substructures. In a later phase of the project the surface resistivity values measured will be correlated with chloride diffusion coefficients measured from the same structures. Previous lab work suggests that a good correlation exists on saturated concrete between these two parameters. The objective of this study is to assess whether a similar correlation can be obtained from field surface resistivity readings. Preliminary results suggest that a conditioning method needs to be applied on-site to approximate water saturation conditions at the elevations of interest.
No abstract is provided for this article.
A set of partially immersed reinforced concrete columns with mature corrosion patterns was used to demonstrate that corrosion macrocell currents provided “cathodic prevention” to parts of the steel assembly that remained passive even though it was exposed to high chloride concentrations (up to ∼4.7% of the cement weight) by the end of nine years of testing. The potential of that steel increased following macrocell decoupling, and instances of depassivation were observed 15 days to 75 days after decoupling. The results are contrasted with previous observations of potential dependence of the chloride corrosion threshold in concrete, and an approximate functional relationship between the threshold value and the potential is proposed.
An alloy coating has been developed for an AA2024-T3 substrate that can serve as barrier, sacrificial anode, and reservoir to supply soluble inhibitor ions to protect any defect sites. In this paper, the chemical throwing power of such an metallic coating under thin electrolyte films representative of atmospheric conditions is modeled. The geometry is that of an surface with the presence of a scratch simulating exposed AA2024-T3. The model calculates the time necessary to accumulate and inhibitors over the scratch when released from the coating under different conditions. The model factors in the pH-dependent passive dissolution rate of an alloy to define the inhibitor release flux. Transport by both electromigration and diffusion are considered together. The effects of scratch size, initial pH, chloride concentration, and electrochemical kinetics of the material involved were studied. Sufficient accumulation of the released inhibitor (i.e., the concentration surpassed the critical inhibitor concentration over AA2024-T3 scratches) was achieved within a few hours (e.g., for scratches of ) when the initial solution pH was 6 and the coating was adjacent to the AA2024-T3.
An Al-Co-Ce alloy system has been developed with three important protection abilities when deployed as a metal coating over high-strength aerospace alloys such as AA2024-T351 (UNS A92024). These alloy coatings can serve as a local corrosion barrier with tunable anodic properties, act as a sacrificial anode (i.e., its open-circuit potential [OCP] is ca. 20 mV to 300 mV below that of AA2024-T351), and supply soluble ions that function as corrosion inhibitors. These three functions are not mutually exclusive. However, optimization of each may not be achieved in the same way. The Al-Co-Ce alloy system has demonstrated an excellent inherent resistance to corrosion due to its structural amorphicity and chemical composition, with electrochemical tunability based on solid solution Co and Ce composition. For example, the pitting potential (Epit) is more noble than the Epit of AA2024-T351 under acid, neutral, and alkaline conditions with pitting and repassivation potentials optimized by Co content. The gal...
In a previous paper (Part I), the utility of Zn-submerged bulk anodes (ZnSBA) for providing cathodic protection to the above-waterline portion of a marine-reinforced concrete bridge substructure was investigated. It was determined that polarizations greater than 100 mV reached to the top of cast-in place footers (about one meter above mean tide) but decayed sharply in the columns above this. As such, it was concluded that these anodes alone did provide some protection in the splash zone, albeit not as much as when combined with either zinc thermal-spray or zincjackets. In the present investigation, similar field trials were performed using Mg-submerged bulk anodes (MgSBA), which were considered to be potentially more effective in providing protection to the above-waterline zone because of greater driving voltage. Results of these trials indicate that 100 mV depolarization was achieved in some cases to as high as 2.5 m above mean water. In addition, the substructure footer and column with MgSBA wa...
No abstract is provided for this article.
The ability of a SS316L surface wetted with a thin electrolyte layer to serve as an effective cathode for an active localized corrosion site was studied computationally. The dependence of the total net cathodic current, I{sub net}, supplied at the repassivation potential E{sub rp} (of the anodic crevice) on relevant physical parameters including water layer thickness (WL), chloride concentration ([Cl{sup -}]) and length of cathode (Lc) were investigated using a three-level, full factorial design. The effects of kinetic parameters including the exchange current density (i{sub o,c}) and Tafel slope ({beta}{sub c}) of oxygen reduction, the anodic passive current density (i{sub p}) (on the cathodic surface), and E{sub rp} were studied as well using three-level full factorial designs of [Cl{sup -}] and Lc with a fixed WL of 25 {micro}m. The study found that all the three parameters WL, [Cl{sup -}] and Lc as well as the interactions of Lc x WL and Lc x [Cl{sup -}] had significant impact on I{sub net}. A five-factor regression equation was obtained which fits the computation results reasonably well, but demonstrated that interactions are more complicated than can be explained with a simple linear model. Significant effects on I{sub net} were found upon varying either i{sub o,c}, {beta}{sub c}, or E{sub rp}, whereas i{sub p} in the studied range was found to have little impact. It was observed that I{sub net} asymptotically approached maximum values (I{sub max}) when Lc increased to critical minimum values. I{sub max} can be used to determine the stability of coupled localized corrosion and the critical Lc provides important information for experimental design and corrosion protection.
A literature review was conducted with the goal of identifying alternative, low-cost, corrosion-resistant steel reinforcement materials. The most promising alternate reinforcing materials ...
