PROJECT NUMBER: S-280 (Formerly S-207)

TITLE: Mineralogical Controls on Colloid Dispersion and Solid-Phase Speciation of Soil Contaminants

DURATION: October 1, 1997 to September 30, 2002

STATEMENT OF THE PROBLEM

Widespread soil contaminant transport from agricultural and industrial operations to surface waters and ground waters is a major environmental concern for water quality deterioration. Soil contaminants such as heavy metals must be managed indefinitely because they are not degraded in soils, they are typically not converted to volatile forms, and their biouptake is low compared with macro-nutrients. Alexander (1995) suggested that the slow kinetics of sorption and desorption in soils causes a progressive sequestering of soil contaminants into forms that diminish their risk and toxicity. The complexity of the soil matrix limits our understanding of the basic mechanisms involved in natural attenuation of contaminant bioavailability and mobility in soils. Yet the long-term impacts of heavy metals and persistent organic contaminants in the environment depend on their potential to be mobilized either in a dissolved form or associated with mobile soil colloids. To complicate matters, there are 17,000 soils classified in the U.S. (USDA, 1994), making it impractical to quantify the behavior of contaminants on a per-site basis. By evaluating the soil mineralogical and chemical properties that are important for mobilization of soil colloids, and by determining the nature of association between selected soil contaminants and different mineralogical and organic fractions of soils, the proposed research will provide mechanistic insights that will be useful for managing the long-term impacts of persistent soil contaminants.

JUSTIFICATION

Pollution control and abatement is an enormous problem in the United States, with total projected costs for all pollution control activities between 1990 and 2000 estimated at $100 billion to $160-billion per year (US-EPA, 1990). Land and water pollution control costs represent about 70% of this total, with estimated annual costs of water quality programs totaling $57.6 billion and land pollution control costs (including superfund activities) amounting to $38.1 billion (US-EPA, 1990). Furthermore, 1992 estimates indicated that there were more than 25,000 possible contaminated waste sites in the U.S, with an additional 367,000 sites of municipal and on-site industrial landfills, underground storage tanks, and abandoned mines requiring further evaluation (Reed et al., 1992). Since public awareness increased more than 30 years ago, government regulation and compliance has largely been driven by the knowledge of the existence of contaminants or hazardous materials and the (sometimes perceived) public health risk. Pressure to clean up or contain all perceived hazards is a major factor affecting total costs. Many of the proposed or planned operations for remediation and containment are expensive and controversial, and it is not clear that these are the most effective ways to reduce the health risk or environmental impact, or even whether they will work at all over reasonable time scales. This controversy is due, in large part, to the fact that while the existence of hazardous elements and compounds is known, in many cases critical information such as the chemical state (speciation) of the elements, the state or nature of the environment in which they occur, and the processes affecting their interactions - including exposure pathways and mechanisms of transport - are not well understood.

Many of the problems with containment and remediation of environmental contaminants exist at DOE, DOD, USDA, and other Federal facilities, and at numerous sites that are the responsibility of state and local governments, private companies and individuals. Nonpoint source pollution from agricultural lands has also been recognized as a problem with regard to water quality deterioration, but the environmental costs of nonpoint source pollution control are more difficult to document. Examples of nonpoint source agricultural pollution include discharge of nutrients, particularly nitrogen, from land receiving animal waste, and buildup of heavy metals to potentially toxic levels on land used for municipal waste disposal (Chaney, 1994; Chaney and Ryan, 1993). Although nitrogen and phosphorus are priority nutrients of concern with animal waste disposal, it is also recognized that high levels of copper and zinc associated with the waste may pose a long-term environmental threat (Anderson et al., 1991). In addition, residues of persistent organic contaminants can remain in the soil long after their use is discontinued. For example, a recent study showed that 10% to 28% of DDT applied to soil plots in the western U.S. remained as parent or metabolite compounds after 23 years (Spencer et al., 1996). With heightened public awareness and concern for environmental contamination problems and prohibitive costs of remediation or prevention measures, there is a real need for research-driven solutions to environmental problems. Because soils are one of the principal media regulating contaminant movement and bioavailability, basic studies on contaminant-soil interactions are well justified.

Geochemical transformations of heavy metal contaminants in soils are poorly understood. The formation of unique mineral phases and adsorbed chemical species of contaminants in different soil environments is critically important for understanding the long-term stability and fate of the contaminants. *1* The proposed project will address two aspects of contaminant fate in soils: (i) soil properties affecting dispersibility of colloids and (ii) the nature of association of heavy metal contaminants with dispersible colloids and other soil constituents.

There is considerable evidence that heavy metals and certain organic compounds can be more rapidly transported either with eroded soil or through the soil matrix when sorbed onto mobile, colloid-sized particles. The mineralogy of mobile colloids in a lysimeter study on an Ultisol were found to be enriched in certain minerals (e.g., iron oxides and gibbsite at low water flow rates and quartz kaolinite, and hydroxyinterlayered vermiculite at high flow rates); the mobile colloids also were enriched in organic matter (Kaplan et al., 1993). Thus mineralogy and organic matter characteristics influence dispersibility (and mobility) and the nature of the dispersed particles. The mechanisms of preferential mineral dispersion and the role of organic matter and other chemical properties of the soil are not understood, yet influence soil erosion, colloid-mediated transport, and pedogenic processes such as development of an argillic horizon. In the previous project, a multiple linear regression model indicated that much of the variability in WDC between southern region soils could be accounted by several mineralogical and chemical properties, including CBD-Fe and Al, suspension pH, deviation of pH from the point of zero salt effect (a surface chemical property), and CEC.

The first research objective in the project will evaluate how water-dispersible clay (WDC) and DOM dispersion are affected by selected key identified in our current project to correlate with WDC. The method for obtaining WDC used in the past project and the proposed project has been correlated with field data on dispersion of soil clay sue to raindrop impact (Miller and Baharuddin, 1986). Because dispersion is a precursor to both erosion and transport of colloids through the soil matrix, the WDC method provides an operational approach to systematically studying a wide range of soils in the context of this regional project. Therefore, our results could potentially be correlated with projects involving soil erosion and sediment control, in addition to providing a more fundamental understanding of mechanisms of colloid mediated transport.

This project focuses on the mineralogical and chemical processes that affect the physical behavior of colloidal mineral and organic particles in porous media and aqueous suspensions. The actual transport of colloidal particles in soil systems also depends on a number of physical processes such as raindrop impact, water movement, soil pore-size distribution, and macroaggregate stability. Although our proposed project does not explicitly address these physical processes (except through correlation with the WDC procedure), the chemical and mineralogical aspects of soil colloids are known to exert a strong influence on the physical properties of soils. Thus, our focus on basic mechanisms of colloidal dispersion and contaminant binding in a range of soils will provide a valuable research foundation for more complex studies that address interactions between physical and mineralogical or chemical processes. By designing the proposed research around a predictive statistical model developed in the past project, we can determine how well we understand the fundamental processes, and identify gaps in our knowledge that can be addressed in the proposed project. In contrast to the past project, the proposed project will be more narrowly focused on regional soils and soil horizons that were found in the previous project to be more dispersible (the surface horizons of Ultisols and Alfisols). Also, we will narrow the analyses to several key soil properties found to be statistically significant with regard to colloid dispersibility. A statistical research model developed using a comprehensive set of soils from the southern region can potentially be used to improve the predictive capabilities of existing management models (e.g., CREAMS and GLEAMS) for soil erosion and crusting and contaminant transport in soils. By tying our results to spatial tools such as the Mineralogical Map of the Southern States (developed as part of a past regional mineralogy project) or a GIS database, we may help in identifying land areas of particular concern (hotspots) for erosion and contaminant transport. Land management approaches can then address physical, mineralogical, and chemical effects on erosion and contaminant transport.

To better understand the potential for soil contaminants to move along with dispersed colloids or to be sequestered into immobile forms in soils, the second objective will quantify the amounts of contaminant associated with mobile colloids and certain mineral fractions in a number of contaminated soils from the southern region. Because pedogenic processes produce complex associations of various mineral phases and organic matter that are unique to soils, the task of identifying the nature of contaminant associations on a microscale is a daunting task. However, by pooling resources and expertise from the various participants' labs, several soil-component separation techniques will be used to help determine trends for contaminant-soil associations. Identifying the types of minerals that are both mobile and tend to have associated contaminants will lead to a better understanding of contaminant solubility and bioavailability, and colloid-mediated transport. This information is useful for managing contaminated soils, determining acceptable loading levels, predicting long-term contaminant fate, and developing successful soil remediation approaches.

A pooling of regional resources and expertise is essential to the success of the research. Soils selected by participants from throughout the southern region will have a wide range of properties and would not be readily available to any given individual working in the group. Furthermore, selecting and sampling contaminated sites is more expedient when regional project members work within his or her own state. To advance our understanding of the complexity of contaminant-mineral associations in soils and its relationship to colloid dispersion and contaminant transport, the soils will be studied using the range of expertise and specialized techniques contributed by project participants. For example, expertise on colloid dispersion and mineralogical characterization will be contributed by various participants as in the previous project (SREL, GA, KY, VA, FL, PR). Techniques for fractionating and characterizing mineral-contaminant associations will include water dispersion (GA), high-gradient magnetic separation (TX), density fractionation (TN), and organic matter fractionation (MS). Applying these techniques on a variety of soils will yield a substantial amount of information about properties of soil contaminants and pedogenic processes affecting their long-term fate.

OBJECTIVES

To evaluate the capability of a colloid dispersion model to predict water-dispersible colloid content based on quantifiable mineralogical properties of southern region soils.

2. To determine the nature of contaminant-mineral associations in selected soils as influenced by mineralogy and pedogenic properties and processes, and in relation to water-dispersible colloids.