WHAT IS THE APPROPRIATE VALUE OF DEPOSITION VELOCITY FOR USE IN PREDICTIVE MODELS FOR EMISSIONS FROM WELL-ENGINEERED FACILITIES? (1994)

M J Fulker(1), G Shaw(2) & E.I.Evans(2)

(1) Westlakes Scientific Consulting, Moor Row, Cumbria, CA24 3LN
(2) T. H. Huxley School, Imperial College Centre for Environmental Technology Silwood Park, Ascot, Berkshire, SL5 7TE

ABSTRACT

The dry deposition velocity for aerosol particles is an important parameter in modelling the effects of releases of radionuclides to the atmosphere. Controlled laboratory conditions are conducive to the measurement of deposition velocity with high precision, but such measurements may not be applicable to real field conditions. Since particle deposition velocities are highly size dependent, reliable field measurements depend on matching the particle sizes of samples of particles in air and in the deposition flux, and in selecting an aerosol size range that is appropriate to the purpose. Modern, well-filtered stacks only release aerosols with sizes of less than a few um under normal conditions. For such particles, it is concluded that a dry deposition velocity in the range 1*10-3 to 5*10-3 ms-1 would be appropriate.

INTRODUCTION

Emissions of radioactive aerosols to the atmosphere from nuclear reactors and reprocessing plant are required to comply with very stringent discharge limits which are set in the UK by the national regulatory authorities. The assessment of the radiological consequences forms part of the process of reviewing the acceptability of discharge authorisations, and this may be approached by use of models such as PC-CREAM(1). British Nuclear Fuels (BNFL) uses its own AEROS model for the assessment of the dose consequences of routine releases to the atmosphere. In both these models, the effect of deposition on concentrations in food depends on the estimation of dry and wet deposition to surfaces and crops. The deposition of chemically active gases may also be described by a deposition velocity, but this paper is concerned specifically with the estimation of the flux of radioactive particles by processes of dry deposition. The deposition velocity, variously referred to as Vd (2,3) or Vg (4,1) is the flux of activity deposited per unit area per second, per unit air concentration.

Vg = flux to surface, per unit area per second (1)
air concentration, per unit volume

Values of Vg have been reviewed Slinn(3) and Semel(2). Many of the parameters which affect Vg vary widely under field conditions. Laboratory based values of Vg under more controlled conditions are more consistent, but should be applied with caution to the more variable situations appropriate for predictive modelling purposes.

METHODS AND RESULTS

Previous determinations of the deposition velocity under field conditions near the BNFL Sellafield reprocessing plant, using the measurement of radioactivity in air and deposition samples(5), have shown values of Vg typically in excess of 1*10-1, much higher than the value of 1*10-3 m s-1 which is assumed for small aerosol particles in the AEROS model. These high values were due to two factors: (i) the contribution to the deposition flux from large particles of resuspended soil, and (ii) the relatively poor collection efficiency of the high flow-rate air sampler for the larger particles above a few mm diameter, which contribute significantly to the deposition flux. Although these large, resuspended, particles may have absorbed radioactivity from earlier, historical, discharges from Sellafield, they are unlikely to have resulted directly from the plume from a modern, well-filtered stack fitted with HEPA (high efficiency particulate air) filters, and should not therefore be included in assessments of the ‘first pass’ of the plume.

In the present work, particles depositing by dry deposition were collected on large ‘frisbee’ collectors, with a depth to diameter ratio of 1.5, mounted 1 m above ground and sheltered from the rain by a rain-hood 1.3 m above the frisbee collection surface. The airborne particles were sampled using two types of high flow-rate air sampler with different particle size collection efficiencies, mounted 1 m above the ground level(6). A critical flow sampler, designed to cut off entry of particles above 10 um AMAD (Wedding PM10 sampler), was used to collect the smaller particles. An impaction rod sampler designed by AEA Technology (AEAT) used rods of different sizes to collect larger particles in three size ranges. This sampler had a wide entry, 0.4 m square, through which air was forced past the impaction rods at 5 m s-1 by a fan. The whole assembly was mounted to pivot into the wind to minimise the effects of changing wind directions on the collection efficiency. One of the impaction rods collected particles greater than 11 um diameter. Using these two samplers, particles in the size ranges <10 um and > 11 um were collected and measured by radiochemical methods for 137Cs, 239Pu + 240Pu and 238Pu from the collected mass.

The PM10 sampler would appear to be suitable for collection of the small particles from a well-filtered source, whereas the rod impactor, collecting particles >11 um collected the larger particles, including resuspended soil. In determining the deposition velocity in the field, the problem remains that the deposition collector is not size selective. The sizes and masses of the deposited particles can however be classified, from measurements of the projected area of large numbers of deposited particles, using an automated Scanning Electron Microscope (SEM). The dry deposition velocities can then be estimated for particle size ranges matching the particle size collection characteristics of the two air samples used.

Table 1 shows a comparison of deposition velocities determined from measurements of 137Cs and 239Pu+240Pu radioactivities in the deposition samples and air samples, with those based on mass, using the SEM classification of particle sizes in the deposition samples. For the radioactivity based measurements, the measured Vg is often much higher than the value of 1*10-3 ms-1 that would correspond to aerosol particles of a few um or less. This is partly because the particle sizes collected by the air samplers (either <10 um for the PM10 sampler, or <11 um for the rod impaction sampler) do not match the wider range of particle sizes collected by the deposition sampler. This can be seen in Table 1, in which high Vg values are shown for 137Cs and 239Pu+240Pu using the aerosol sample of particles of less than 10 um. Even when the size range of airborne particles is extended by adding the air concentrations for <10 um and >11 um, the value of Vg based on these radioactive measurements is still well in excess of 1x10-3 ms-1 because large particles of resuspended soil are present.