Executive Summary
Phosphogypsum, a waste by-product derived from the wet process production of phosphoric acid, represents one of the most serious problems facing the phosphate industry today. This by-product gypsum precipitates during the reaction of sulfuric acid with phosphate rock and is stored at a rate of about 30 million tons per year on several stacks in central and northern Florida. The main problem associated with this material concerns the relatively high levels of natural uranium-series radionuclides and other impurities which could impact the environment and which makes its commercial use impossible. Our general approach to this problem was to start the task of detailing exactly where and how radionuclides are hosted within the material. In this way, it is hoped that ultimately one may develop purification schemes for this waste material.
Our experimental approaches for characterizing the radiochemistry and microbiology of Florida phosphogypsum has been directed along four specific lines of research. One component of this study involved detailed analyses of the same radionuclides in the phosphate ore rocks and the phosphogypsum. Another component includes preliminary investigations of radionuclides in shallow groundwaters collected from a limited number of monitor wells adjacent to gypsum stacks. The majority of research in the third component of this study involves dissolution and leaching studies of the phosphogypsum and developing techniques to isolate and concentrate specific radionuclides from the phosphogypsum matrix. The ultimate goal of this line of research was to identify the actual sites occupied by individual nuclides and how they are bound in the phosphogypsum. The final research component concerned microbiological studies which endeavored to identify and culture bacteria that either show the ability to release radionuclides from phosphogypsum or have promise of scavenging radionuclides from fluids that are associated with the material. The microbiological research was focused on microbes which metabolize Po and may play a role in the solubility and mobility of Po (and perhaps other radionuclides) in solutions circulating in and through gypsum stacks.
This Final Report presents data and discussion of these efforts which have been made since the study was initiated in 1992. The report is organized to show how these research topics were addressed: Chapter 1 presents general radiological data for phosphate ore rock, phosphogypsum that has been stored on gypsum stacks for various times, and water samples from monitor wells adjacent to phosphogypsum storage areas in Florida. The primary objective of the research described in this chapter was to radiochemically characterize and provide a comprehensive data base of uranium-series nuclides in Florida phosphogypsum and ore rocks. By evaluating all significant nuclides, including 226Ra, 210Pb, and 210Po in pairs of ore rock feed materials and phosphogypsum, sufficient data were obtained to describe fractionations between the various members of the 238U decay chain during processing. A thorough sampling of phosphogypsum of various ages stored in Florida has also been completed in order to assess whether some radioelements migrate preferentially to others during storage.
The actual location of uranium-series radionuclides within phosphogypsum was investigated in more detail in Chapter 2 by several different approaches. The hope here was that if the actual sites within the phosphogypsum that host radionuclides could be identified, this may represent a first step towards radiochemical purification of this material. In this way, a waste material that may potentially contaminate the environment could ultimately be put to some good use. The material presented in Chapter 2 is an extension of the broad-based radiochemical research presented in Chapter 1. We describe methods that have been investigated for fractionating, concentrating, and leaching radionuclides from Florida phosphogypsum by both physical and chemical methods. Our results show that the behavior of radium, and other radionuclides, is often sample dependent, i.e., there are significant differences in the solubility of radionuclides for different samples. In general, about 10-50% of the radium in Florida phosphogypsum is water soluble, although there are some data which suggest that the radium is actually associated with extremely fine-grain particles, perhaps colloids. The remaining, water-insoluble radionuclides are associated with the major elements Al, Fe, and P and the minor elements Ba, Sr, and rare-earth elements (as Ce and La). Our data suggest that the radionuclides appear to be hosted in an aluminum-rich phase, probably an aluminum phosphate resembling the mineral crandallite. The ultimate controls which govern the observed differences between phosphogypsum samples still eludes us although substantial progress has been made in detailing the mechanisms which influence release and migration of radionuclides from phosphogypsum.
Chapter 3 concerns investigations related to the elevated concentrations of 210Po, the last radioactive member of the 238U decay-series, which have been reported in a number of shallow wells from the Central Florida Phosphate District. Although the exact source is uncertain, the 210Po probably originates either from the naturally-occurring phosphate rock of the area or from phosphogypsum. We assessed the potential of a bacterial isolate to remove and incorporate dissolved polonium from solution by conducting comparative radiotracer experiments using 35SO4 and 208Po. Since the observed chemical concentration of Po in these wells is too low to serve in any direct metabolic function, it was suspected that it might be co-metabolized with its chemical analog sulfur. Our experiments were designed to (1) measure the rate of isotope uptake as a function of bacterial growth; and (2) fractionate the bacteria into various cellular components to determine how polonium and sulfur were partitioned within the cell. Results indicated that while the initial uptake mechanisms for SO4 and Po differ, once associated with the bacterial cells, polonium is dispersed between the cell walls, cytoplasm, and protein in a manner similar to sulfur. The uptake rate of polonium is sufficiently rapid that the potential exists for development of a bioremediation scheme for removal of polonium (and perhaps other contaminating ions) from process waters and other aqueous solutions.
The research reported in Chapter 4 is an extension of the uptake experiments presented in the previous chapter. In this section, we investigated the more environmentally significant bacterial release of polonium from phosphogypsum. Because of the chemical similarity of Po to sulfur (both occupy the same column in the periodic table) studies were initiated to determine whether bacteria, particularly those species active in sulfur cycling, could account for the selective solubilization and mobilization of Po. This chapter reports on experiments involving interaction of bacteria with this waste gypsum. Bacteria were isolated from gypsum that were capable of mediating Po release in column experiments when fed a growth medium. Sulfate-reducing bacteria were particularly effective at mediating Po release provided the sulfide levels did not rise above 10 µM, in which case Po was apparently coprecipitated as a metal sulfide. Thus, whether microbes release or take up polonium in this system depends upon the prevailing environment (oxic or reducing) which will ultimately dictate which bacteria are present.
William C. Burnett, James B. Cowart, Paul LaRock, and Carter D. Hull - Florida State University