«February 2014 Final United States Environmental Protection Agency Office of Solid Waste and Emergency Response Office of Resource Conservation and ...»
Infiltration Rate Water movement through cracked concrete is a complicated phenomenon that is difficult to model accurately. The actual amount of water that can pass through concrete is determined by a number of variables, including, but not limited to, the size, number, and orientation of cracks present on the road surface (Apul et al., 2002). To avoid compounding uncertainty by specifying ranges for each of these variables, the evaluation only considered a high-end scenario where infiltration through concrete was not a limiting factor. In this scenario, the amount of water that can pass into the ground water table is bounded by the infiltration rate of the soil underlying the concrete road, base, and sub-base. In practice, the use of drainage pipes beneath the roadway may reduce the amount of infiltrating water that can reach the ground water table. However, not all roadways include this design feature. Furthermore, without proper maintenance, these pipes can become clogged and ineffective at rerouting water. Therefore, the evaluation does not account for the potential effects of drainage pipes on infiltration rates in this evaluation. These assumptions likely overestimate releases and subsequent exposures, but the magnitude of this overestimation is unknown.
Smaller cracks may exist in the concrete matrix that can retain infiltrating water for longer durations than those modeled, and soils may retain water in contact with the concrete for some time after a precipitation event has ended. Both scenarios may allow greater mobilization and accumulation of COPCs than considered in this evaluation. However, in both cases, capillary forces will act against gravity and impede the flow of water, resulting in much of the water evaporating before it can migrate to the subsurface. Instead, the current evaluation focuses on high infiltration rates through the concrete.
The much larger volume of water assumed to pass through these larger and more numerous cracks, together with the high-end COPC concentrations used in the evaluation, will result in a higher mass flux of COPCs into the subsurface environment. Even if a smaller crack size exists that could result in higher mass fluxes than those considered in this evaluation, these cracks will propagate and expand, becoming larger with time. Thus, small cracks are less representative of a long-term, high-end leaching scenario.
These assumptions likely overestimate releases and subsequent exposures, but the magnitude of this overestimation is unknown.
Leaching Evaluation Assessment Framework Data The current evaluation used the LEAF Method 1313 data from Kosson et al. (2013) to bound the COPC concentrations that can be present in leachate. This sampling method requires the sample to be ground up prior to leaching. Grinding a material increases the surface area available for contact with the leachant. The higher surface area exposed to the leachant results in higher leachate concentrations for those COPCs than are likely to occur during use. The evaluation used LEAF Method 1315 data from Garrabrants et al. (2013) to calculate COPC concentrations in leachate passing over the concrete surface.
This sampling method retains the sample in monolith form. The concrete monolith is submerged in a tank containing deionized water for a set time, and then is transferred to another tank with fresh water.
Then, the resulting leachate in each tank is measured. Although the leachate from this method is more representative of water exposed to an encapsulated material, the contact time between the water and 5-13 concrete is much higher than would generally occur during a precipitation event. Therefore, the current application of both Method 1313 and 1315 data is likely to overestimate COPC releases, but the magnitude of this overestimation is unknown.
The Cementitious Materials Report (US EPA, 2012b) compared the pH-dependent leachate data with single pH data collected in other studies (Cheng et al., 2008 and Zhang et al., 2001) to determine whether the pH-dependent data had accurately captured the range of potential leaching behaviors. The single pH leaching tests were conducted with either the Synthetic Precipitation Leaching Procedure (SPLP, EPA Method 1312) or the Standard Test Method for Shake Extraction of Solid Waste with Water (ASTM D3987-85). The report did not consider data from the Toxicity Characteristic Leaching Procedure (TCLP, EPA Method 1311) because the pH of the leaching solution was below the lowest theoretical pore water pH of 7.0 for concrete. The comparison showed that the single extraction leaching tests produced results that were generally consistent with the pH-dependent leaching tests at the same pH. However, single extraction leaching tests do not provide an indication of the changes in material leaching with changes in pH that occur as a consequence of material aging. The few single extraction point measurements that deviated from this trend exhibited lower leachate concentrations. These results may be a consequence of partial carbonation that occurred during preparation and testing of the ground up samples (Garrabrants et al., 2004). Based on these results, US EPA (2012b) concluded that the LEAF leachate data agree well with other single-pH leachate tests. Therefore, reliance on LEAF data, rather than single-pH data, is likely to reduce the amount of uncertainty in the evaluation.
Additional Leach Test Parameters The current evaluation of leaching relies on data from Kosson et al. (2013) and Garrabrants et al.
(2013) to evaluate leaching from fly ash concrete. The primary focus of these studies was to evaluate the effects of pH, liquid to solid ratio, and physical form (e.g., ground, monolithic) on leaching from concrete. All of these parameters are known to have a major impact on the leaching of inorganic COPCs. There may be other factors, such as light and heavy fractionation; mineral phases; trace metal speciation; solution composition; and background electrolyte and ionic strength, which influence measured concentrations to some degree. However, these factors are not anticipated to be major sources
of uncertainty in the current evaluation because:
- The current leaching evaluation is based on use of the entire material that was subjected to leach testing, rather than some fraction of the material. Therefore, no distinction between light and heavy fractionation needed to be made for this evaluation.
- Because the COPCs identified in this evaluation are trace inorganics for which mineral phases are below typical instrument detection limits, COPC concentrations were measured directly through leaching tests rather than inferred from major mineral determinations.
- The speciation of certain COPCs, such as chromium, in leachate was considered as part of previous evaluations of CCR leaching (US EPA, 2009c). The potential effects of reducing conditions on COPC speciation in concrete leachate were not considered. However, this is not anticipated to be a source of uncertainty, as oxidizing conditions are anticipated to be prevalent around the uses of concrete discussed in this document. It is important to note that this is distinct from reducing 5-14 conditions that may occur in subsurface soils and ground water, which are addressed through fate and transport modeling.
- Other parameters were considered through previous evaluations of fly ash leaching because of the potential to affect COPC leaching on a case-by-case basis. However, none of these parameters had a consistently strong impact on leaching. Some of these less significant parameters were discussed in US EPA (2009c). Leachate concentrations for a number of minor analytes in the samples tested [e.g., dissolved organic and inorganic carbon, conductivity (which can be converted to ionic strength), copper, iron, manganese, nickel, silicon, sodium, zinc, and others] were evaluated, but not presented in US EPA (2009c). These data are available from the authors, and are included in Leach XS Lite. 11 A more extensive analysis of specific fly ash concrete samples may reveal individual cases where one of these additional parameters is important. However, for an evaluation that is intended to estimate a national bounding of releases from fly ash concrete, it is believed that the current focus on the parameters known to consistently have the greatest effect on leaching remains appropriate. Therefore, while the data used in this evaluation may over- or underestimate concrete leaching on a case-by-case basis, the overall magnitude of these uncertainties is expected to be small.
Concrete Aging As concrete ages, physical and chemical changes that occur in the concrete matrix may alter the rate at which the concrete releases COPCs. Carbonation is the primary mechanism that drives concrete aging. Carbonation is the reaction of carbon dioxide [CO2] with the various alkaline constituents in the concrete matrix. The most important of these reactions is with calcium hydroxide [Ca(OH)2], which ultimately generates calcium carbonate [CaCO3]. This shift from calcium hydroxide to calcium
carbonate can alter leaching in several ways:
- The first way that aging may alter leaching is through changes to the concrete pore water pH. An initial pH of roughly 12.4 is common for newly-poured concrete based on the dissolution chemistry of calcium hydroxide. Complete carbonation of the concrete matrix may result in a pH as low as 7.0, based on the dissolution chemistry of calcium carbonate (Garrabrants et al., 2004). Changes in this pH will alter the leaching behavior of constituents with pH-dependent solubility.
- The second way that aging may alter leaching is through changes to the composition of the concrete.
Carbonation of certain minerals in the concrete matrix may result in desorption of COPCs that otherwise would have remained bound within the concrete matrix (Garrabrants et al., 2004). Studies have shown that this desorption acts in concert with changing pH to alter leaching rates. Müllauer et al. (2012) demonstrated higher cumulative leaching of chromium from highly-carbonated ground concrete. Sanchez et al. (2002) demonstrated higher cumulative leaching of arsenic, but lower cumulative leaching of lead, from highly-carbonated ground concrete.
Leach XS Lite is a tool that allows users to evaluate and characterize the release of constituents based on comparisons
derived from leaching test results for a wide range of materials and waste types This tool is available on-line at:
- The final way that aging may alter leaching is through physical changes to the porosity of the concrete matrix. Van Gerven et al. (2006) compared the total porosity of concrete samples carbonated for 60 days with total porosity of similar samples with relatively little carbonation. The results showed a 12 percent reduction in the porosity of the carbonated samples, compared to the relatively uncarbonated samples. Van Gerven (2006) noted that this decrease in porosity resulted in decreased leaching for sodium and potassium. Sanchez et al. (2002) also noted that the permeability of an intact concrete matrix decreased with carbonation and, as a result, the retention of arsenic, chromium, and lead all increased (Lange, 1996 as cited in Sanchez et al., 2002).
The findings of Van Gerven (2006) and Lange (1996) (as cited in Sanchez et al., 2002) contrast with those of Sanchez et al. (2002), Garrabrants et al. (2004), and Müllauer et al. (2012), but do not contradict them. The latter three studies evaluated samples of ground concrete, which eliminated the physical concrete matrix and found higher leaching with increased carbonation, while Van Gerven et al. (2006) evaluated samples of intact concrete and found lower leaching with increased carbonation. Grounding concrete allows samples to become highly and uniformly carbonated by breaking down the dense concrete matrix. However, it is unlikely that such a high degree of carbonation will occur during the useful life of most intact concretes. This results in uncertainty regarding the ultimate effects of carbonation on leaching behavior of carbonated concrete, and how such effects impact releases of inorganic constituents. As a result, the current evaluation may underestimate or overestimate long-term leaching on a case-by-case basis, but the magnitude of this misestimate is unknown.
Fish Bioconcentration Factors US EPA (2003a) recommends the use of bioconcentration factors (BCFs) to assess exposure to inorganic metals. Therefore, this evaluation calculated the potential exposure from fish ingestion using BCFs for trophic level 3 (TL3) and trophic level 4 (TL4) fish (i.e., fish at the higher levels of the food chain where bioconcentration is greater) to estimate the transfer of pollutants from environmental media into fish. In the current evaluation, the evaluation used only BCFs from laboratory or field studies of TL3 and TL4 fish, rather than values estimated from physical or chemical properties [e.g., octanol-water partition coefficient (Kow)]. Aquatic BCFs are developed by dividing measured concentrations in aquatic biota by total surface water concentrations. There are several sources of uncertainty associated with the models used to estimate BCFs for aquatic biota.
One source of uncertainty is experimental error that may affect the true value of the BCF. Error may originate from the relatively short exposure timeframe used in a study compared to the lifetime of exposure in the field; some laboratory BCF studies may not have attained steady-state concentrations in the fish due to short exposure durations (Arnot and Gobas, 2006; and CalEPA, 2012). Other sources of error may be laboratory-prepared water with concentrations that are not representative of field conditions (e.g., constant and unrealistically high concentrations), the use of radio-labeled compounds without adequate correction for the parent signal, or the use of a less precise analytical method when analyzing samples (Arnot and Gobas, 2006; and CalEPA, 2012). These laboratory errors may bias the resulting measurements to be either high or low; the overall magnitude of this uncertainty is unknown.