I have the ability to spot errors in many science based fictions, working as a chemist one learns some things.
I recently saw one, in CSI (Trilogy). A body part had dirt on it which was shown by ICP to contain the ‘big bad P‘ (plutonium) so then they thought that it must be from near to an atom bomb testing area (near Las Vegas). Then an actor said that the half life of that plutonium was about 90 years, I think that they said 90 years.
Lets break down what happened.
1. It is possible to use ICP (which is shorthand for ICPOES) to detect transuranium compounds, but I shudder to think how much activity there would be in a solution which is 1 ppm of Pu. 1 ppm is in the range which ICPOES can measure with ease.
The sensitivity of ICP-OES is quite low. If we assume it is Pu-239 then to make a 7 ml sample which is about 1 ppm we would need about 20 micrograms of Pu. I am assuming that we would lose about half of the Pu as we digest down, dilute and transfer the samples around. I think that 20 or even 10 micrograms of Pu on a person’s foot would be a lot for an ‘environmental sample’. 10 micrograms of Pu would be 42 nano moles. This is 2.52 x 10^16 atoms
Half life of Pu-239 is 24100 years, so half-life is 7.605 x 10^11 seconds, so lambda is 1.315 x 10^-12 s-1.
As Activity = number of atoms * decay constant (lambda)
Then the activity would be about 33 kBq. I think that level of alpha contamination on one’s foot would be a total nightmare ! If I have got the maths wrong please correct me.
If we go towards the limit of detection, and assume that the machine can measure down to 0.01 ppm then the amount of Pu on the foot will be reduced by a factor of 100, but 330 Bq is still a lot to find on a foot belonging to a member of the general public. (update at the end of Feb 2010)
2. OK we have changed machine to a typical ICPMS (one which can measure micrograms per litre) which is about 1000 times more sensitive, so now the dead body has a less nightmarish 33 Bq on it. Still a lot of plutonium to be walking around with on your foot. I guess that even if a hand foot checking machine is only 20% efficient for Pu-239 then you would get a count rate of about 6 or 7 counts per second. An alpha count rate of 6 per second would make me call for help, then take off my shoes with care, put them in a plastic bag before wishing that I could levitate and thus have avoided the contamination.
3. OK we go to a super trace level ICPMS machine and we take extra care to make sure that the uranium hydride peak will not blot out the peak at 239. I know that at high concentrations Pu-239 can be measured by ICPMS even with a little uranium present but when the uranium concentration is much higher than the Pu concentration then the UH peak will become more important. I suspect that many sand samples will contain uranium.
To deal with this problem this we could either do some chemical separations or use a more advanced ICPMS machine. We might then be able to measure Pu are very low levels. One problem with a chemical separation is that we might lose some of the Pu during the separation. I recall reading once a method for a determination of Pu in rainwater (M.P. Rubio Montero and A. Martin Sanchez, Applied Radiation and Isotopes, 2001, 55, 97-102) and to go from a rainwater sample to the small volume of sulphuric acid used to electroplate the Pu onto stainless steel disks was not a simple easy matter. For rainwater these Spanish workers did claim a chemical recovery of between 60 and 91 %, I trust their claims as they did add a Pu-242 internal standard at the start of their experiment. My worry is that people in the fictional CSI lab will never normally expect Pu, so they will be unlikely to add a known amount of Pu-242 to the sample as an internal standard, so it will be very hard to know what the chemical recovery of Pu will be in their case. (Up date at the end of Feb 2010)
4. There is a problem, unless we were looking for Pu we would be unlikely to take the special measures needed find it in a random sample from a crime when the body is found nowhere near a nuclear site.
5. I saw the CSI actor boiling up the sample in a glass beaker, the problem is that PuO2 tends to be insoluble in nitric acid. It is likely that the Pu in an environmental sample will be PuO2, soluble Pu only tends to be found in fuel cycle buildings such as THORP. To get PuO2 to dissolve it is likely that you will need to use HF.
6. HF digests down glass to make SiF6 dianions, so you have to use a polyethylene or teflon beaker, the beaker the actor had looked like glass. Also he was not working in a fume hood, I think HF digestions should be done in the fume hood. If the CSI men are doing HF digestions on the open bench then maybe they should ask the police to arrest themselves for breaking health and safety law !
7. If the half life of the Pu was circa 90 years then it is likely to be Pu-238. I would expect this to be found in a used UO2 fuel (even more in a used MOX fuel). Unless the bomb was using a very dirty reactor grade Pu then the Pu-238 content is likely to be very low. If the dead girl had Pu-238 on her foot then I would expect her to have got it somewhere other than near to a former atom bomb testing site. If you look at one paper in which a mineral sample taken from the site of the first ever atom bomb test had its plutonium isotope signature measured (P.P. Parekh, T.M. Semkow, M.A. Torres, D.K. Haines, J.M. Cooper, P.M. Rosenberg and M.E. Kitto, Journal of Environmental Radioactivity, 2006, 85, 103-120) you will see that Pu-238 is not the major plutonium isotope.
8. I would expect that if Pu-238 is detected with ICPMS then the alpha activity will be rather high again but lets forget about that error for a moment. Rather than using ICPMS it will be more easy to use with liquid scintillation counting or an alpha spectrometer to measure this isotope.
9. If they want to detect the activity from a nuclear testing site then I would suspect that some fission and activation products would be more easy to detect than the Pu. I would start my hunt by getting a gamma spectrum for the sample. If I saw Np-239, I-131, Tc-99m, Ru-106, Cs-137 or Eu-152 then I would suspect nuclear fission has occurred at a place. For a bomb testing site which has not been used for decades I would concentrate my search on the long lived fission and activation products such as Cs-137 and Eu-152.
So the TV show seems to have got it very wrong.