HPS masthead
What's New?
Report - Medical Radiation Exposure Increases Seven-Fold Since 1980s
March Journal
March Newsletter
Upcoming Events
54th Annual Meeting of the Health Physics Society
12-16 July 2009
Minneapolis, Minnesota
2009 Professional Development School
16-18 July 2009
Minneapolis, Minnesota
07 March 2009

Answer to Question #629 Submitted to "Ask the Experts"

Category: Environmental and Background Radiation — Building and Construction Material

The following question was answered by an expert in the appropriate field:

Are there some modern technologies to process phosphogypsum (generated after fertilizer production) to make useful products without health hazard effects?
I assume from the wording of the question that the interest is in applying technology to existing phosphogypsum (PG) and not producing radiologically "cleaner" phosphogypsum. Also, it is assumed that the PG is produced via the wet phosphoric acid process. This subject was recently addressed with an international perspective in the journal Phosphorus & Potassium (Karakunen and Vermeulen 2000). The authors believed it necessary to reassess the radiological hazards of phosphate products and PG following the issue of new safety standards against dangers from ionizing radiation (96/29/Eurotom). The activity concentrations in many PG stacks exceed the limits and cannot be used as raw material in the construction industry, but the radioactivity levels in PG produced from igneous phosphate rocks is below the new limit. Uranium has normally been enriched by seawater in all marine-deposited phosphates, which results in higher activity concentrations than igneous deposits. Radium-226 activity concentrations in PG stacks produced by the wet phosphoric acid process fall within the range of 10 to 1,300 Bq/kg with daughter nuclides also in that range. The radon emanation factor is typically about 0.2. In 1992, the EPA ruled that PG intended for most applications, including agricultural and construction use, must have a certified average 226Ra concentration of no greater than 370 Bq/kg. Consequently, most of the PG produced in the United States is prohibited from use and is stockpiled in stacks. In theory, the 226Ra concentration of 370 Bq/kg yields an effective dose of about 1 mSv/y through the significant pathways. This effective dose limit is estimated using an index based on not only the 226Ra activity concentration, but also those of 232Th and 40K as well. Typical PG from many parts of the world will exceed this index, as do several building materials of nonphosphatic origin. In Florida alone, more than 900 million tons of PG are stacked in more than 25 stacks. Thirty million new tons of PG are produced each year. Most of that PG is restricted by the EPA for any application. Many products could be made from phosphogypsum that is under the EPA radioactivity limit or if the basis for that limit is reviewed and revised. Some examples are soil amendments (for the sulfur), vitrification for glass and ceramic products (roofing tiles, etc.), roadbeds, landfill cover, and oyster cultch.

Soil Amendments

FIPR has funded studies dealing with the agricultural use of soil amendments. These have primarily dealt with land application of PG. PG is a by-product of the production of phosphoric acid. Chemical processes result in a segregation of radionuclides such that uranium stays with the phosphoric acid product and radium goes with the PG. The majority of the uranium is not present, but the radium and other progeny will remain in disequilibrium status. Potential issues of concern are ingestion of foods after uptake of radioactivity (addressed previously), inhalation of dust bearing radioactivity during broadcast of the amendment, increased external exposure rates from radioactivity on the ground, groundwater intrusion, and increased radon potential from the soil. FIPR Publication No. 05-041-124 (Johnson et. al 1996) evaluated risk factors and estimates used in the EPA's issuance of a final rule under NESHAPS (40 CFR, Part 61). One situation evaluated was application of PG (containing 1110 Bq/kg 226Ra) as a soil conditioner or fertilizer. PG was assumed to have been applied to soil every two years for 100 years after an initial application of twice the biennial application rate. Six scenarios were considered with application rates ranging from 664 to 4000 kg of PG per acre and tillage depths from 22 to 46 cm. After the 100-year period, 226Ra concentrations in the soil were calculated to be from 25.53 to 115.44 Bq/kg. The risks (for fatal cancers) from direct gamma exposure to an agricultural worker for the application of PG with a radon concentration of 962 Bq/kg ranged from 1.4 x 10-6 to 6.4 x 10-6. The risk from dust inhalation ranged from 5.8 x 10-9 to 9.0 x 10 -9. An initial field study of the application of PG at relatively low rates (up to 4 Mg PG ha-1) for three years showed no statistically detectable increases in radionuclides in soils or in groundwater, or in airborne radon and gamma radiation measured 1 meter above the plots. A subsequent FIPR-funded study (Publication No. 05-038-141) developed data to assist in the comprehensive assessment of the environmental impacts of PG application at higher rates (up to 20 Mg PG ha-1) to an established bahiagrass pasture primarily in terms of the radionuclides and secondarily of the heavy metal impurities in PG. The specific objectives for the radiological aspects of the study include collecting data to support assessment of three potential exposure routes, namely:

  • Airborne radon progeny inhalation: by evaluating soil 226Ra (the radon production source), soil surface radon flux, and ambient airborne radon;
  • External gamma-radiation exposure: by direct measurement; and
  • Radionuclides in the water and food pathway (ingestion): by measuring 226Ra and its two long-lived decay products, 210Pb and 210Po, in soil, water, and forage.

For an annual application rate of 0.4 Mg ha-1 (0.18 tons/acre) over 100 years the results were:

  • Airborne radon progeny inhalation: the projected radon flux from amended soil was about 40 percent of the mean value for undisturbed nonphosphate mineralized land in Florida. The PG-attributable contribution to indoor radon (potential house construction on the treated plots) was 0.02 to 0.2 pCi (7.4 x 10-4 to 7.4 x 10-3 Bq) per liter of air (not very significant). It was estimated that a cumulative treatment up to 40 Mg ha-1 would contribute less than 0.1 pCi (0.37 Bq) L-1 to atmospheric radon over the field.
  • External gamma-radiation exposure: the radionuclides weathered, were removed with harvests, and/or penetrated into the soil to the extent that no contribution to gamma- radiation could be detected after the first year. For the assessment scenario, the PG contribution to the annual external gamma-radiation dose for an individual present 100 percent of the time over treated land was projected to be 0.028 mSv. This is well below the limit of 1 mSv above background that is recommended for members of the general public and is consistent with recommendations that doses from individual sources be limited to a fraction of the limit for all sources combined. The annual dose contribution would actually be less than 0.028 mSv as a result of the shielding provided by floor slabs during the time spent indoors and the fact that occupancy is usually less than 100 percent of time spent at home.
  • Ingestion pathway: The radionuclides contained in PG appear to have limited mobility in surface and groundwater during the first two years after PG application to the soil. However, it is possible that they may be gradually mobilized and appear in the groundwater at a later time. Levels of 226Ra after 100 years of PG use were projected to be on the order of 2 to 3 pCi (0.074 to 0.111 Bq) L-1. However, these concentrations are below the current drinking water standard of 5 pCi (0.185 Bq) L-1 and well below the proposed standard of 20 pCi (0.74 Bq) L-1. Levels of 210Pb were projected to be similar to baseline runoff and shallow groundwater levels of <1 pCi (<0.037 Bq) L-1 guideline of 1 pCi (0.037 Bq) L-1 of 210Pb in water requires further evaluation since existing background levels may already correspond to a significant fraction of the dose limit in proposed drinking water standards. The projected total concentration (background plus additions) of 210Po in runoff resulting from the 100-year practice is 35 percent of the 15 pCi (0.555 Bq) L-1 standard for gross alpha activity in drinking water. The 210Po projected concentration in shallow groundwater is similar to background and only a small fraction of the drinking water standard. For the food pathway, there are no standards for concentrations of radionuclides in forages. It does not appear that PG-attributable doses to humans through the ingestion of animal products traceable to the treated forage would be outside the range of variation in the normal diet. However, further assessment is suggested and would require projections of radionuclide intake by animals, transfer to foods of animal origin, and potential ingestion by humans and then evaluation of the resulting radiation dose.
  • Vitrification

    Glass and glass-ceramics can be readily manufactured from PG and tailings sand. These products inhibit the emanation of radon from the product, thus eliminating this potential health risk. Products range from glass—ceramic floor, wall and roof tile; synthetic wollastonite fibers used in ceramics and paints and as a nontoxic substitute for asbestos; synthetic stone for building facades; abrasives; flat "privacy" glass; and container glass for selected beverages and agricultural products. The risk assessment for tiles assumed a composition of CaO/SiO2 with a weight ratio of 25/75, and activity concentration of 884.3 Bq/kg 226Ra, density of 2.6 g cm-3 and radon emanation of 0.0058 pCi (2.146E-04) m-2 s- 1. The analysis showed that radon dose from the tiles for even a house completely tiled with PG-derived tile emanated radon at a rate of only 10 percent of normal construction materials. Even under conservative conditions, the indoor radon concentration was estimated to be 0.0187 pCi (6.919 x 10-4 Bq) L-1. That concentration results in an annual dose of 0.014 mSv under conditions of 18 h d-1 occupancy. The external dose from gamma emissions was estimated to be less than 0.5 mSv per year for the maximally exposed individual (large areas of tile use on floor and roof and extended exposure durations). More typical usage of floor tile was less than 0.1 mSv per year.


    Another study estimated that continual exposure in a 5m×5m×3m room in which the walls and ceiling are lined with 1 cm thick PG plasterboard containing 399.6 Bq kg-1 of 226Ra would lead to a dose of 0.13 mSv/y.

    Road Bed

    PG can be used as a binder for base course mixtures. PG mixtures are easier to work with than clay mixtures, and the operation cost, including equipment time, for PG roads is lower than that of clay roads. Rain during construction does not cause excessive delays because the compacted mixture does not absorb water to any great extent. Shrinkage cracks, frequently occurring in clay roads, are greatly reduced. The stability of compacted PG mixtures is superior to that of clay mixtures. In studies of experimental roads it was concluded that gamma levels from the roadbeds do not yield doses that approach the limit of 1 mSv for a member of the public. There are no significant effects of radon levels from the roads. Furthermore, investigation of groundwater and soil has shown no significant increase in 226Ra levels due to the presence of the roadbeds.

    Landfill Cover

    Bench and pilot scale testing indicates that PG used as a cover material speeds the degradation of waste materials and extends the useful life of landfills.

    Oyster Cultch

    It has been proposed that phosphogypsum could be used for marine applications such as oyster cultch or artificial reefs. Research has focused on finding a PG:fly ash:cement mixture that is stable in seawater. Dynamic leaching behaviors and diffusion coefficients must be understood to determine the potential long-term effects of stabilized PG on the surrounding environments. Bioaccumulation studies need to be conducted for toxic metals and radium. Multiple studies of different composites containing PG concluded that there is very little evidence of leaching or bioaccumulation of toxic metals and 226Ra. Reference: Karkunen, J.; Vermeulen, S. Natural radioactivity of phosphates and phosphogypsum. September/October 2000. Phosphorus & Potassium. Fertilizer International No. 378. pp 75-81. Brian Birky, Ph.D.

Answer posted on 9 February 2001. The information and material posted on this Web site is intended as general reference information only. Specific facts and circumstances may alter the concepts and applications of materials and information described herein. The information provided is not a substitute for professional advice and should not be relied upon in the absence of such professional advice specific to whatever facts and circumstances are presented in any given situation. Answers are correct at the time they are posted on the Web site. Be advised that over time, some requirements could change, new data could be made available, or Internet links could change. For answers that have been posted for several months or longer, please check the current status of the posted information prior to using the responses for specific applications.
Home Affiliates Ask the Experts Radiation Terms Employment Meetings