Alen V. Gusa

One of the major environmental concerns with the recovery of unconventional gas resource from Marcellus Shale is the presence of Naturally Occurring Radioactive Material (NORM) in produced water. Ra-226 is the major component of NORM with a half-life of 1,600 years that is present at concentration as high as several thousand pCi/L. Most of the studies on NORM management are focused on above-ground scenarios. The main focus of this study was on functionalizing the proppant (i.e., quartz sand) that is used in hydraulic fracturing to prevent the closure of induced fractures formed during this process and allow release of natural gas so that it can also sequester NORM from the produced water before it reaches the surface. Five different sulfates and carbonates were tested for their ability to capture Ra-226 from aqueous solution and celestite (SrSO4) was identified as the best choice because of its affinity for Ra-226 sequestration even in the presence of very high total dissolved solids that are characteristic of Marcellus Shale produced water. Among possible ways of coating the proppant with celestite, precipitating celestite directly on the sand surface was found to be the best option as it provided uniform distribution of celestite and high uptake of Ra-226. Although quartz sand can adsorb some radium from the solution due to electrostatic interactions, adding a small amount of celestite on the sand surface (20-30 mg/g) increased radium removal from the solution containing 5,000 pCi/L of Ra-226 to more than 80% in dilute solution and to more than 50% in high salinity solution even in the presence of very high concentrations of competing divalent cations. The results of this study indicate the potential of coated proppant to sequester NORM in the subsurface and prevent adverse environmental impacts when radiogenic produced water is brought to the surface.

Accurate and precise analyses of oil and gas (O&G) wastewaters and solids (e.g., sediments and sludge) are important for the regulatory monitoring of O&G development and tracing potential O&G contamination in the environment. In this study, 15 laboratories participated in an inter-laboratory comparison on the chemical characterization of three O&G wastewaters from the Appalachian Basin and four solids impacted by O&G development, with the goal of evaluating the quality of data and the accuracy of measurements for various analytes of concern. Using a variety of different methods, analytes in the wastewaters with high concentrations (i.e., >5 mg L⁻¹) were easily detectable with relatively high accuracy, often within ±10% of the most probable value (MPV). In contrast, often less than 7 of the 15 labs were able to report detectable trace metal(loid) concentrations (i.e., Cr, Ni, Cu, Zn, As, and Pb) with accuracies of approximately ±40%. Despite most labs using inductively coupled plasma mass spectrometry (ICP-MS) with low instrument detection capabilities for trace metal analyses, large dilution factors during sample preparation and low trace metal concentrations in the wastewaters limited the number of quantifiable determinations and likely influenced analytical accuracy. In contrast, all the labs measuring Ra in the wastewaters were able to report detectable concentrations using a variety of methods including gamma spectroscopy and wet chemical approaches following Environmental Protection Agency (EPA) standard methods. However, the reported radium activities were often greater than ±30% different to the MPV possibly due to calibration inconsistencies among labs, radon leakage, or failing to correct for self-attenuation. Reported radium activities in solid materials had less variability (±20% from MPV) but accuracy could likely be improved by using certified radium standards and accounting for self-attenuation that results from matrix interferences or a density difference between the calibration standard and the unknown sample. This inter-laboratory comparison illustrates that numerous methods can be used to measure major cation, minor cation, and anion concentrations in O&G wastewaters with relatively high accuracy while trace metal(loid) and radioactivity analyses in liquids may often be over ±20% different from the MPV.