Evaluación de la Seguridad Alimentaria y Radiológica en la Aplicación del Fosfoyeso Como Enmienda de Suelos Agrícolas en las Marismas del Guadalquivir

Abstract

The main goal of this Project has been the radiological and alimentary assessment of the agriculture use of phosphogypsum (PG), a by-product of the fertilizer industry, as Ca-amendment in reclaimed salt-marsh soils in SW Spain. The work-plan included: PG characterization (concentrations of radionuclide and heavy metals, 222Rn exhalation from PG stacks), experimental field studies attending to the effect of PG in agriculture soils and drainage waters, and studies on soil-to-plant transfer of radionuclides and heavy metals (through both, field and greenhouse studies). PG characterization (PG stacks) FERTIBERIA, a fertilizer factory located in Huelva (SW Spain) owns a non-active PG stack lying in the right bank of the Tinto River. From this PG stack, of about 1 km2 and 8-10 m deep, PG could be extracted for agriculture uses. PG has being traditionally used since late 1970’s as a Ca-amendment for the reclaimed soils from the salt-marsh area of the Guadalquivir River. 222Rn exhalation in this non-active stack has been measured following the USEPA method 115 (adapted for using cylindrical charcoal canisters). Experimental studies conducted in collaboration with the Institute of Biophysics from the University of Salzburg, demonstrated the repeatability and accuracy of the charcoal canister method for the commented determinations. The accuracy was demonstrated against the exhalation chamber method. Finally, experiments with PG columns of different heights demonstrated that the major contribution to Rn exhalation comes from the 0-50 cm horizon. A total of 162 measurements of 222Rn exhalation have been carried out, corresponding to 49 sampling points distributed through three different regions in the stack (1, compacted and dry top surface area; 2, loose –tilled- dry top surface and 3, sides). Three charcoal canisters were placed at each sampling point, covering an area of ~1 m2 . 222Rn exhalation values ( 510 ± 470, 250 ± 270 y 360 ± 170 Bq h-1 m-2 for regions 1,2, y 3, respectively) show high dispersion, but their mean values were under the USEPA limit of 2664 Bq h-1 m-2. Charcoal canister placed over big cracks and freshly removed surfaces did not reveal any significant change in 222Rn exhalation rates. The study of the normalized exhalation rates at different spatial scales (0.1, 0.75, 60 and 200 m) shows almost uniform standard deviations of ~0.5. The major contribution to this variability in the scale 0.1-1.0 m has to be attributable to the variability, at the same spatial scale, of the micro and meso-structure of pores and cracks. For larger spatial scales, changes in 226Ra concentration and percentage of humidity, contribute to the variability in the 222Rn exhalation rates. PG samples were taken at surface level (0-20 cm) in 20 sampling points, and at three different depths (0-30 cm, 30-60 cm and 60-90 cm) in another 20 sampling points. 226Ra was measured by liquid scintillation, providing activity concentrations of 720 ± 260 Bq kg-1 and 690 ± 180 Bq kg-1(dry weight), for regions 1 and 2, respectively. These values are over the threshold level of 370 Bq kg-1 established by the USEPA to allow the agriculture use of PG. 210Po activity concentrations (measured by alpha spectrometry) of 660 ± 160 Bq kg-1 y 575 ± 95 Bq kg-1 were determined for regions 1 and 2, respectively. U-isotopes were measured by alpha spectrometry, obtaining 238U activity concentrations of 170 ± 110 Bq kg-1 for region 1, and 160 ± 80 Bq kg-1 for region 2. The isotopic ratio 238U / 234U was 0.97 ± 0.03, as expected from secular equilibrium. 210Pb concentrations were measured by gamma spectrometry in few samples, and their values were closed to the corresponding values for 210Po activity concentrations. The above activity concentration for 226Ra and its daughters (and in less extend U-isotopes), are more than one order of magnitude higher that those found in agriculture soils from the area of Las Marismas of Lebrija. Multi-elemental analysis of the PG samples was carried out by ICP-MS technique following USEPA 200.8 method. PG samples, after oven-dried, disaggregated and sieved, were acid digested with supra-pure nitric acid in a microwave-oven (pseudo-total recovery). Cd concentrations were 2.0 g/g, one order of magnitude higher than those found in the agriculture soils from Lebrija. Thus, the Cd and the radionuclides from the 238U series will be the elements of major concern in the agriculture use of PG from FERTIBERIA stack. For the rest of the elements considered in the USEPA 200.8 method, concentrations were not too much higher than the corresponding ones for soils (Hg was not measured). Thus, taking into account the important dilution of PG when is applied as soil amendment (homogenised in the 0-30 cm soil horizon), the input of these elements has to be considered as quantitatively negligible. Nevertheless, all these elements has been considered in the present study, since the scientific literature reports some effects of the PG in the mobility and bioavailability of some metals in soils.