Characterization of Miller-Similar Silica Sands for Laboratory Hydrologic Studies

The use of well-characterized porous media can simplify and improve the efficiency of laboratory subsurface flow and transport experiments. The objective of this study was to present a comprehensive set of hydrologically relevant properties for a unique set of commercially available silica sands. Features of sands selected for characterization included high sphericity, high batch-to-batch consistency, Millersimilarity, and availability in large quantities. Samples of four different sand grades (12/20, 20/30,30/40, and 40/50 sieve sizes) were characterized for physical properties, chemical composition, water retention, three-phase air-non-aqueous-phase liquid (NAPL)-water saturationpressure relationships for water and a model NAPL, Soltrol 220, and saturated and unsaturated hydraulic conductivity. Properties common to all sand grades included high chemical purity and low organic matter content. Water retention curves featured well-defined air entry pressures and the Miller-similarity of the media was demonstrated for both static and dynamic properties. During water retention measurements, we determined that the common assumption of a uniform vertical water content distribution in retention cells can result in significant errors in uniform porous media. A numerical correction procedure was developed and successfully applied to correct fitted water retention curve parameters, illustrating that potential errors of up to 70% in volumetric water content are made without proper analysis. The characterization data for the four sand grades presented here should facilitate their use in a wide range of laboratory flow and transport studies. F FLOW and solute transport are critical processes related to contamination, remediation, and conservation of soil and groundwater resources. Countless experiments are being conducted to study the movement of single fluid phases, i.e., water, NAPLs, and gases, multiple fluid phases, and contaminants in subsurface environments. Fluid flow and solute transport problems frequently involve complex physical, chemical, and biological processes. Laboratory experiments in physical models, chambers, and columns are often conducted to limit scientific studies to the investigation of specific phenomena. The selection of porous media is an important initial step for conducting laboratory subsurface hydrologic experiments. Availability, sometimes in large quantities, low cost, and well-defined physical, chemical, and biological properties are desired features of porous media. In addition, the porous medium should be representative of the natural environment. To interpret experimental results, knowledge of basic hydrologic properties is required. Obtaining these properties often requires specialized equipment and techniques that can be time consuming and expensive. For many applications, the use of previously characterized media may be desirable. However, for previously characM.H. Schroth and J.D. Istok, Dep. of Civil Engineering, and S.J. Ahearn and J.S. Selker, Dep. of Bioresource Engineering, Oregon State Univ., Corvallis, OR 97331. Received 11 Sept. 1995. "Corresponding author (schrothm® ucs. orst. edu). Published in Soil Sci. Soc. Am. J. 60:1331-1339 (1996). terized media to be reliable, it must possess high batch-tobatch consistency and uniformity in the desired properties. Therefore, commercial processing is usually required to produce this level of consistency and uniformity. The primary objective of this study was to present a comprehensive set of hydrologically relevant properties for commercially available silica sands so that the media could be employed by the broad scientific community. In addition, we wanted to verify that the four grades of sand characterized were hydrodynamically similar in the sense of Miller and Miller (1956). Finally, we wanted to develop a more accurate approach to interpreting water retention data obtained in retention cells for coarse, uniform-textured porous media. MATERIALS AND METHODS Porous Media and Liquids The porous media selected for characterization were four grades of silica sand (12/20, 20/30, 30/40, and 40/50 sieve sizes), available in large quantities from Unimin Corporation' (Le Sueur, MN), under the trade name Accusand. Accusand grades were obtained prewashed (with water) and presieved by the manufacturer. Prior to conducting characterization experiments, all Accusand grades were rinsed with distilled water to remove fine dusts. The Accusand grades were then oven dried at 50°C. No further treatment was employed. Liquids used for the characterization of Accusand grades were distilled water and Soltrol 220, hereafter referred to as Soltrol, a light NAPL manufactured by Phillips Petroleum Company (Bartlesville, OK). Soltrol is a mixture of branched da to C,7 alkanes with a specific gravity of 0.81. Its components can be found in a variety of petrochemical products, e.g., diesel fuel. Negligible water solubility, very low volatility at room temperature, as well as a low health hazard are reasons for the extensive use of Soltrol as a model NAPL in multiphase flow studies (e.g., Gary et al., 1994; Lenhard, 1992). Physical and Chemical Characterization Sieve analyses and determination of sphericity for four Accusand grades were performed by the manufacturer using standard methods (ASTM, 1987; American Petroleum Institute, 1986). Particle densities were measured in our laboratory by He pycnometry (Ayral et al., 1992) using an ACCUPYC Model 1330 pycnometer (Micromeritics Instrument Corp., Norcross, GA). The cation-exchange capacities for the four Accusand grades were determined by extraction (Page et al., 1982) and ICP/ AAS (U.S. Environmental Protection Agency, 1986). Total Fe, total Cd, total Cu, total Pb, total Mn, and total Zn contents were determined by HNO3 digestion and ICP/AAS (U.S. Environmental Protection Agency, 1986). Iron oxide (Fe2O3) content was determined by extraction (Page et al., 1982) and ICP/ 1 Reference to trade names or companies is made for information purposes only and does not imply endorsement by Oregon State University or the US-DOE. ________________ _____________ Abbreviations: NAPL, non-aqueous-phase liquid; ICP/AAS, inductively coupled plasma atomic absorption spectrophotometry; S-P, saturationpressure.