Thursday, September 25, 2008

RADON FLUX – some thoughts by Phil Jenkins

RADON FLUX – some thoughts by Phil Jenkins

I have seen a few references on the listserv to sealing something to a surface (e.g., a
granite countertop), such as a paper plate or bowl, and placing a radon-measuring device,
such as a liquid scintillation charcoal device, in the volume. This kind of measurement
might give an INDICATION that there is radon coming from the surface; however, the
value of the measurement likely has limited usefulness beyond showing that there is
some RADIUM (not uranium, although that might be there, too) in the material. The
measurement would be very difficult to interpret in terms of how much radium is in the
material or what the impact might, or might not, be of the radon coming from the surface.
The reasons for this hopefully will be made clear here.

What should be of interest is the value of the RADON FLUX coming from the surface,
so that is the topic I’m tackling here first. The radon flux can be used, along with a host
of other information (such as the volume of the room or house, the ventilation rate, etc.),
to model the effect that the material might have on the radon concentration in the room or
house. The radon flux can be measured in a variety of ways. Here I describe briefly a
method that was used years ago, and could still be useful today. I had a reference for this
at one time, but that was twenty or more years ago. (I thought there was a procedure in
HASL-300, but I haven’t been able to find it so far. There is within EPA Method 115
some description of measuring radon flux from uranium mill tailings that might be
helpful.)

Using Charcoal Canisters to Measure Radon Flux

The first use of charcoal canisters for measurement of radon that I can recall seeing was
for measuring the radon flux from a surface. The surface could be something like
concrete or soil or, now granite. The charcoal canister was sealed to the surface and left
in place for something like three days. The following assumptions are made: 1) all of the
radon leaving the surface area sealed by the canister is adsorbed by the charcoal in the
canister, 2) having a closed canister sealed to the surface does not affect the flux itself, 3)
there is no leakage of radon from around the seal and 4) the radon flux is constant with
time.

The collected activity of radon as a function of time can be described by something like
the following (this is off the top of my head, folks, so don’t quote me, but I think it’s
correct):

Q(t) = F A [1 – exp(-λt)] / λ (1)

where Q(t) = activity of radon as a function of time captured from surface (pCi)
F = radon flux from surface (pCi m-2 s-1)
A = area of surface from which radon is collected (m2)
λ = decay constant of radon (s-1)
t = time (s)

(If you wish, you may substitute Bq for pCi in the above expression, but PLEASE bear in
mind that the conversion factor for pCi to Bq is NOT 37, that’s the conversion from pCi/l
to Bq/m3. The conversion from pCi to Bq is 0.037, in other words 1 pCi = 0.037 Bq, or
inversely 1 Bq = 27 pCi/l……just remember that a Bq is a lot MORE radon than a pCi,
not the other way around.)

Note that if radon did not decay, then the expression would merely be:

Q(t) = F A t (2)

and the radon activity would continue to increase in the canister linearly over time. But,
radon decays during the collection period, so in this case t is replaced by

[1 – exp(-λt)] / λ

which I like to call the Effective Sampling Time (EST, not to be confused with Eastern
Standard Time). Note that EST has the unit of time, just like t. The collected activity of
radon actually does increase rather linearly at first, but soon deviates from linearity,
tending toward a maximum value after several half-lives of radon. This maximum value
corresponds to a time when t is large enough that [1 – exp(-λt)] is essentially equal to
one, so the maximum collected activity of radon is

Qmax = F A / λ (3)


Also, note that one-half the maximum value will be reached after one half-life of radon,
¾ of the maximum will be reached after two half-lives, and so on.

Most charcoal labs should be able to analyze the canister and solve for the value of Q at
the end of the exposure period. This means correcting for decay to the END of the
exposure period, not the middle of the exposure period. I have always done that when
analyzing charcoal canisters, as I described in a publication back in 1991 (Jenkins, P.H.,
Equations for calculating radon concentration using charcoal canisters, Health Physics,
61, 131-136, 1991, if you’re interested, but in this article I believe that I showed a model
that I was using at that time to solve for the calibration factor, or effective sampling rate,
but about 15 years ago I switched to a much better model). With the method that I use,
the value of Q is always an intermediate step in the calculations. This requires, however,
that the counting efficiency be determined using a standard charcoal canister containing a
known quantity of 226Ra, so if the laboratory does not have such a standard it cannot
solve for Q.

So, if Q is measured, then Equation 1 can be solved for the value of F, the radon flux.
Again, knowing F, the effect on the radon concentration can be modeled if all the other
factors are known.


Big Louie

There are, of course, other ways to measure the radon flux. A different charcoal
technique uses what is called a Big Louie. This is a large charcoal container that has a
hole in it so it can “breathe” with changes in pressure. The same assumptions listed
above apply here except for the fact that the canister is not sealed from the ambient air.
An additional assumption is that radon is not lost through the hole in the canister. The
same equations shown above apply here. The advantage of the Big Louie is that it covers
a larger area of the surface.


Electret Ion Chambers for Measuring Radon Flux

A truly integrating device, such as an electret ion chamber, can also be used in the
volume that is sealed to the surface to measure the radon concentration in that volume. In
this case, the concentration as a function of time would merely be

C(t) = Q(t) / V (4)

where C(t) = concentration (pCi/l) as a function of time
Q = activity of radon (pCi) as a function of time, from Equation 1
V = volume (l)

This function can be integrated over time to find an expression for the average value of
C, as this is what should be measured by the electret ion chamber. So, if C and V are
known, Q can be calculated, and using Equation 1 F can also be calculated. This would
require that the gamma-ray exposure be measured accurately and taken into account.


Charcoal Devices Measuring Radon Concentration in Sealed Volume

So, what about using a charcoal device, like a liquid scintillation device, inside a volume
sealed to the surface? If one can assume that ALL of the radon that leaves the surface, in
the area confined by the bowl or whatever, is adsorbed by the charcoal device, then one
could use the approach described above to determine the radon flux, but not the normal
calculations that would be used to calculate radon concentration in air. Further, the
method described above assumes that the volume of the container sealed to the surface is
mostly occupied by charcoal and therefore it is reasonable to assume that all of the radon
is adsorbed in the charcoal. In a situation where the volume of charcoal is small
compared with the volume sealed to the surface, this assumption is likely not valid.
Therefore, the charcoal device likely comes into some state of equilibrium with the
surrounding air, just like it does in a room. However, charcoal devices that are used to
measure the radon concentration in a room are calibrated assuming that the radon
concentration remains relatively constant. In this case, the radon concentration is
constantly increasing. So, unless the charcoal device is calibrated for this type of
measurement, it will not give an accurate measure of the average concentration of radon
in the volume. Instead, it will likely produce a measurement that is larger than the
average (charcoal devices are not integrators, but are equilibrium devices). Such
measurements might be useful in a qualitative or relative sense in that one could say that
Surface A produced a measurement that was 10 times greater than Surface B, for
example. But such measurements would likely be questionable for determining the radon
flux quantitatively.


What does it mean?

Okay, so let’s assume that you have a good measurement of the radon flux, now what?
As mentioned above, IF you know a lot of other information you can model what the
effect of this flux might be on the radon concentration in the air. However, some nagging
questions might still remain. For example, how representative is that measurement of
radon flux for the entire surface of the material? Was the flux measured on a hot spot, or
were several measurements made over the entire area? Is it really valid to apply the
measured flux value to the entire surface of the material?

Let me relate a story from the early 1980’s. I was part of the “Radon Group” at DOE’s
Mound Facility. We were charged with trying to quantify the radon flux from the tops of
two concrete tanks containing a material with an extremely elevated concentration of
226
Ra. We sealed four-inch canisters to the concrete surface, but we also realized that in
some places there were obvious cracks in the concrete; our equivalent of “hot spots.” It
turned out that the radon flux that we measured varied by six orders of magnitude (a
factor of one million) from the lowest to highest measurements. Our conclusion was that
it was too heterogeneous to enable us to quantify the radon flux over the entire surface.
Now, I would be surprised if anyone found such a variation over the surface of a granite
countertop, but a variation of a factor of 10 or 100 might be possible. So, it is necessary
to characterize the radon flux over the entire surface of the material.


Bottom Line

Radon flux measurements may be interesting, and may play a vital role in research
projects, but the bottom line here, in situations where the material is already installed in a
dwelling….wouldn’t it just be easier and more appropriate to measure the concentration
of radon in the air, using the usual protocols, except for perhaps measuring in several
other rooms, including the room containing the material?

One more point: I’ve seen mention on the listserv measurements like 5.1 pCi/l in other
rooms and 5.7 pCi/l in the kitchen, therefore seeming to prove that the countertop
increased the radon concentration. BUT, please bear in mind that there are uncertainties
in ALL of our measurements. Unless you put an error bar around the number, it is
meaningless to compare one measurement with another. If one can truly say that these
measurements were something like 5.1 ± 0.1 pCi/l and 5.7 ± 0.1 pCi/l at the 2 sigma
confidence level, then okay, they are probably different. But, more realistically, what if
they are 5.1 ± 0.5 pCi/l and 5.7 ± 0.5 pCi/l at the 2 sigma level; now they are clearly NOT
different. We have to start taking into consideration the Minimum Detectable
Concentration and the Uncertainty in our measurements in order for them to be
meaningful.

No comments: