Letter to W Bair, Et Al, Subject: Report on the Resuspension Study at Bikini (Encl 1: Comparison of Dose Methodology for Bone Marrow & Endoseal Cells for Surface Deposited Transuranics)

Page 1

Biomedical& Environmental Sciences Divisions Lawrence Livermore National Laboratory Ih. Wi-Uiam BaA 6 Rog Thomph on Enclosed is the report on the resuspension study at Bikini. You will note that Table 7 on page 19 of this report is the table that was included in the Bikini dose assessment report UCRL-53225. The problem with the units in Table 1, page 8 of UCRL-53225 is that the micro (p) symbol was omitted from the units of both the dust aerosol column and the soil Pu activity column. Thus, the units for the dust aerosol should be pg/m3 not g/m3. Similarly, the units for the soil Pu activity should be aCi/pg (which is proportional to pCi/g) and not aCi/g. This was done to make a direct calculation possible for pulmonary deposition in aCi/h; sorry for the I have also enclosed (Enclosure 1) a comparison of the dose methodology for bone marrow and endosteal cells for surface deposited transuranics for the Spiers and ICRP models. The values we use for the parameters in the lung model are those listed in ICRP 30, Part 3, and attached as enclosure 2. Terrestrial & Atmospheric Sciences

Page 2

Ratio of Bone Marrow to Endosteal Cell Dose E. Comparison of endosteal cell to red marrow dose ratio for the various Absolute Magnitude of the Bone Marrow Dose Conversion Our Method of Calculating the Marrow and Surface Cell Dose for Surface Areas of Difference Between Methods

Page 3

Comparison of Spiers and ICRP Models for Dose from Pu to Red Marrow and 1. Ratio of Bone Marrow to Endosteal Cell Dose In Spiers method the dose is calculated to a tissue filled cavity and is referred to as Do. The dose in bone, DB, is then related to DO by the ratio of the mass stopping powers of bone and tissue. From geometry consideration and deposition and energy loss characteristic for alpha emitters, the doses to red marrow, DM, and endosteal cells, DS, may be calculated by applying appropriate factors to DB. The factors for Thus, the ratio between the red marrow dose and endosteal cell dose is: B. ICRP (ICRP 30, Supplement to Part 1) From ICRP-30, Supplement to Part 1 page 414 and 415, are listed the committed Dose Equivalent (C.D.E. ) in target organs or tissues per intake of unit activity. The ratio of these C.D.E. for bone surface cells and bone marrow provides the necessary comparison for the ICRP model to Spiers model. Thus, for f, = 1.0 x 10-4, the red marrow CDE is 1.6 x 10-7 Sv/Bq and the bone surface CDE is 2.1 x 10-6 Sv/Ba. The ratio between the red marrow dose and endosteal cell dose is: 2.1 x 10 -6 =,3

Page 4

From NUREG/CR - 0150 Vol. 2, 1979 Estimates of Internal Dose Equivalent to 22 Target Organs for Radionuclides Occurring in Routine Releases from Nuclear This analysis is essentially that of ICRP. It is based on the same data from Marshall, Thorne, ICRP etc. In this report they 9ive “S-Factors” for various target organs in rem/pCi-d. For a Quality factor of 20 the “S Factor” for red marrow is 0.90 rem/pCi-d and for endosteal cells in 12.58 rem/UCi-d. The ratio of endosteal cells to red marrow is, therefore, In the 1982 UNSCEAR report the committed dose per unit intake of Pu is listed as 0.008 pGy/Ba for red marrow and 0.1 pGy/Bq for bone livin9 cells. Thus the ratio of endosteal cells to red marrow is E. Comparison of the endosteal cell to red marrow dose ratio for the The ratio of bone surface to bone marrow dose from Spiers is 12, from ICRP 13, from ORNL 14 and from UNCSEAR 12.5. Thus, from these comparisons, there is very little difference in the ratio of endosteal cell to bone marrow dose for the various options one could select for the calculation. 2. Absolute Magnitude of the Bone Marrow Dose Conversion. Spiers lists the dose conversion for bone marrow in reference 1 as 16.9 rad/y-pCi for a lpCi skeletal content endosteal surface 16m2). This converts to

Page 5

In ICRP 30, Supplement to equivalent dose rate is 315 rem/pC page 414, the committed dose for bone marrow is given as Based on these numbers the The value for bone marrow for a QF=20 of 0.90 rem/pCi-d is The UNSCEAR listed value of 0.008pGy/Bq is identical to the ICRp value of 1.6 x 107 Sv/Bq which I have already shown is equivalent to Spiers dose conversion for bone marrow of 338 rem/pCi-y is 7% higher than ICRP value of 315 rem/BCi-y. When these values for Spiers and ICRP are multiplied by the endosteal to marrow ratio of 12 and 13 respectively the resulting endosteal dose for ICRP numbers is 1% higher than that of Spiers. Thus, the two methods appear to be quite comparable for estimating the bone marrow and endosteal cell dose for Pu. 4. Our Method of Calculating the Marrow and Surface Cell Dose for Surface Starting with a pCi/d intake derived from the Pu concentration in foods times an average dietary intake for each food item, we calculate, using a gut transfer factor of 10-4 and 45% deposition in bone, the pCi/g of Pu in bone. This number is then multiplied by 338 rem/uCi-y to develop the bone marrow dose. The endosteal cell dose is then calculated by multiplying the bone marrow dose by 12, the ratio of endosteal to marrow dose.

Page 6

5. Areas of Difference between the Methods ICRP uses 12m2 as the surface area for endosteal tissue. This is equally divided between trabecular and cortical bone and the absorbed fraction is assumed to be the same in each, (ICRP 30, Part 1.) Spiers uses 16m2 for the surface area of endosteal tissue. I have not been able to determine what the distribution is between trabecular and cortical bone (ref. 2 and 3). The dose factor for trabecular endosteal cells is 3.11 while that for cortical endosteal cells is 50% higher at 4.7. This could reflect a difference in the distribution of the endosteal mass between trabecular and cortical bone. Spiers was the absorbed fraction for Pu rather than an average value for alpha emitters as used by ICRP and this would make some difference. Spiers does make the point that cortical endosteal tissue may not be at as high a risk per unit dose as the trabecular endosteal tissue. I think that the dose rates Spiers lists in reference 1 of 129 rad/UCi-y for trabecular endosteal cells and 193 rad/~Ci-y for cortical endosteal cells (compared with 16.9 rad/~Ci-y for bone marrow) are wrong. Based on these numbers, the trabecular endosteal cell ratio to bone marrow is 7.6 while that for cortical endosteal cells is 11.4. This is inconsistent with his published dose factor ratios of 0.26 for marrow, 3.11 for trabecular surface cells and 4.7 for cortical surface cells which give ratios of surface cells to marrow of 12 and 18 respectively. However, as I mentioned before, we use a dose conversion of 338 rem/uCi-y to develop the bone marrow dose and multiply by a factor of 12 to get the surface cell dose.

Page 7

F.W. Spiers and Janet Vaughn, “Hazards of plutonium with special reference to the skeleton.” Nature Vol. 259, 531-534 (1976). the radiosensitive tissues in bone irradiated by surface deposited radionuclides.” Phys. Med. Biol. Vol. 223, 481-494 (1978). and animals” in The Health Effects of Plutonium and Radium, ed. by Webster Jee, J.W. Press, Salt Lake City, Utah (1976).

Page 9

with an AMAD Fig. 5.2. Mathematical model used 10 describe clearance from the respiratory system. The values for the removal half-times. 7,-1 and compartmental fractions, FL are given in the tabular portion of the figure for each of the three classe< of retained materials. The values gi~en for ON-P, DT-B and Dp (left column) are the regional deposi!lons for an aerosol with an AMAD of I pm. The schematic drawing identifies the various Page 76 of Part 1 and pages 54 and 55ofitsSupplement:As a consequenceoftheCommission’s decisionto reduce its recommended dose equivalent limit for the lens of the eye from 0.3 Sv to 0.15 Sv in a year, values of DAC for 81Kr and ‘3mKr are amended as follows. Radionuclide cloud 1000 m’ room 500 m’ room 100 m3 room Lens Lens Lens Page 78, The values of ALI and DAC for 80Sr are incorrect because no allowance was made-for *ORb the daughter of 60Sr. The correctedthe contribution to committed dose equivalent of , dosimetric data are given in Supplement B of Part 3. The correct values of ALI and DAC are

Page 10

This is an informal report intended primarily for internal or limited external distribution. llre . opinions and conclusions stated are tfroseof the author and mayor may not be those of the ““

Page 11

RESUSPENSION STUDIES AT BIKINI ATOLL Joseph H. Shinn, Donald N. Homan, and William L. Robison Environmental Sciences Division Lawrence Livermore Laboratory The following experiments were conducted on Bikini Atoll to provide key parameters for an assessment of

Page 12

aeroso size, and total radioactivity per unit mass. Although necessary, these T parameters are insufficient to describe the fraction of suspended within the respirable size range and how either the concentration or the respirable fraction vary with surface conditions and local Our investigations recently conducted on Bikini and Enewetak Atolls in the . Marshall Islands-provided new data which have implications not only for the local ““ dose assessment concerning rehabitation of those sites, but which are important ‘\ forunderstanding low-level inhalation exposure totoxic radionuclides in general. “ This study conducted on Bikini Island in May 1978 provided a more complete . ‘- set of data, following our preliminary studies on Engebi Island of Enewetak Atoll in February 1977. The Bikini Island study utilized extensive soil sampling and in situ gamma spectroscopy to determine isotope concentrations in soil and vegeGtm Also various air sampling devices were used to determine particle size distribution and mass loading, and micrometeorol ogical techniques were used to determine aerosol fluxes. Subsequent wet chemistry analysis provided radionuclide and elemental concentrations in collected aerosols. Four simultaneous experiments were conducted: (1) a characterization of the normal (background) suspended aerosols and the contributions from sea spray off the windward beach leeward across the island, (2) a study of resuspension of radionuclides from a field purposely laid bare by bulldozers as a worst-case condition, (3) a study of resuspension of radioactive particles by vehicular and foot traffic, and (4) a study of personal inhalation exposure using small dosimeters carried by volunteers during their daily routines. c“Less complete studies similar to (1) and (2) had been performed ‘::: previously on Engebi Island at Enewetak Atoll and background studies similar to (1) were performed later on Eneu Island’at Bikini Atoll. Soil and vegetation samples were collected for analysis of radionuclide isotope dilution and alpha spectrometry and ‘oSr concentrations by yttrium-90 separation and beta counting. These analyses were performed by LFE Cor oration. Gamma spectroscopy using Ge(Li) diode detector was used to determine 13 Cs concentrations. Also, because the ratios (241Am)/(23g+2q0Pu) and (238Pu)/ (23g+240Pu~ are constant on Bikini, it was possible to estimate plutonium soil concentrations by measurinq 241Am soil concentrations usinq a qamma spectroscopy The detector was mounted facing downward on a tripod so that the volume of soil integrated was contained in a circle of probable detection of nominally 3 m radius and 5 cm depth. Because the nuclear events causing the original contamination of Bikini Island were far removed, the fallout was relatively evenly dispersed across Impingers were used to collect soluble sea spray aerosols in a 250 ml distilled water trap similar to the method of HSU and Whelan (2)., Air flow rates were 0.36 m3 h-l (6 1 rein-l) through the water trap and measured amounts of water were added each day to replace evaporated water (nominally 40 cm3 day-]). Impingers

Page 13

were set at four tower locations along a 60 m wide clearing from the windward beach inland, spaced at 3, 26, 52, and 102 m from the high-tide waterline and at 1 m and 4 m above the ground. Elemental analysis on the remaining water was The major particle collection system was an array of 14 staridard-filter, high-volume air samplers (HV) and two cascade impactors both using Gelman type AE glass fiber filters, . F1OW rates for HV were monitored at a pressure tap on the fan; discharge was nominally 100 m3 h-l (60 cfm). The lapsed time of filter operation was recorded for each HV and cascade impactor by counting pulses from a crystal-controlled clock activated by a pressure sensitive switch. Cascade located on a line from the windward beach inland at 5, 70, and 158 m from the high-tide waterline; the latter two were beneath a coconut grove canopy. One HV was located to the lee of a road at the traffic study site, and ten HV and two cascade impactors were placed in a square array nominally 10 m apart in the middle of a field (one hectare area) cleared by bulldozing. Special chemistry methods (LFE Corporation) were employed on the filters to determine the concentration of stable elements. See map, Figure 1, for locations of instruments. In addition to the HV, three non-standard ultra-high volume air samplers (UHV) were used having air inlets at 1.5 m height. Flow-rates, nominally 2550 m3 h-l (1500 cfm), were monitored both by a pitot tube pressure tap on the fan discharge and by a modified anemometer transducer measuring total discharge in the fan outlet which is ducted 4 m downwind before discharge. Filters were Luftfilter). One UHV was placed in the coconut grove 370 m from the windward high-tide waterline, and two UHV were operated in the cleared field at the downwind edge of the HV array. The UHV provided the advantage of detection of suspended radioactive aerosols at extremely low levels (e.g., worldwide background) in a matter of a few hours run time. (Locations of UHV are shown on Figure 1). Personal dosimeters (Model S, Monitaire, Mine Safety Appliances Co.) were used to dete~mine exposure rates of individual persons to suspended particles. The personal dosimeters (PD) are small, belt-mounted pumps with a hose connection to a cyclone particle discriminator and filter holder suspended by a chain worn around the neck. Flow rates are 0.10 ms h-l (0.06 cfm) and filters were porous-type membrane filters (37 mm diameter, 1 ~m pore diameter, Nuclepore Corporation). It was found that blank membrane filters inexplicably gained weight with time but that standard deviations within 10% of mass could be achieved where filter blanks from the same lot were monitored for weight gains over the time period of the experiment. The membrane filters were used as a substrate for a scanning electron microscope (SEM) study of particle characteristics. The SEM operated Particle Characterization Facility has a large chamber for specimens has a resolution of 0.015 um. In the SEM tiiount,microprobe chemistry aerosol particles on the membrane filters was accomplished by X-ray f“ with a resolution of 160 ev, which provided quantitation of particles elements with atomic numbers equal or g}eater than sodium.

Page 15

(.-,,. Light …