Detection of soil contamination, MDA, etc

["Brian Rees" <brees@lanl.gov>]

We had to clean soil contamination to <5 pCi/g Ra-226 at an UMTRA project 
and used the following method:

In a small (5m x 5m) area a number (I think about 15) of soil samples were 
taken, concentrating on the top 6" of soil, and the center of the section.  
An Eberline SPA-3 (2"x2" NaI) collimated with about 1/4" Pb was used to 
obtain a gamma reading (Bi-214 609 keV) in the center of the section.  The 
soil samples were analyzed for Ra content.  This process was repeated about 
10 times in different areas and concentrations to establish a correlation.  
This gamma count rate was then used to determine Ra content during 
excavation, with confirmatory soil samples taken when excavation was 
complete.   

We were able to readily demonstrate detection of levels <1 pCi/gm Ra-226.  
When asked for a MDA/MDC, we changed the thrust of the question by 
explaining that soil samples confirmed the efficacy of the technique at < 1 
pCi/gm.  Anyway, who cares about MDA/LLD/MDC when the regulatory limits are 
met at the end of the job?  Soil samples were the proof, and the LLD for 
them is fractions of pCi/gm in the lab.  In my OPINION, without soil 
sampling to establish a correlation, you're going to have a hard time 
getting approval of a competant regulator (hmm...).

This method worked pretty well, though if a large area with differing 
geology is excavated, K-40 or other interfering nuclides can ruin your day 
(or even month).  We didn't publish unfortunately, and it was for a private 
company, but if you think this is the way you want to go, contact me for 
further information.

The above information has no bearing on my employer (past or present).

Brian Rees
Los Alamos National Lab
505-667-9980
brees@lanl.gov