I have been asked the same question a couple of times in recent weeks, so I thought that I would address it in a newsletter since it may be of even wider interest now. The recovery and processing of used refrigerants is somewhat difficult to monitor accurately so there is some confusion about methodology and accuracy of the various methods.
First, the list of the types of refrigerants that may be encountered is pretty lengthy and will depend heavily on the use for which the unit in question was initially designed. The focus seems to be more on the smaller salvaged units, so there would probably be more of a version of a CFC or HFC rather than ammonia in these cases. Given that Method 25-A is the usual screening method for something like this, the question then turns to how accurately are these compounds being measured. The question often just revolves around whether another method, such as Method 25, would report lower concentrations. The number of refrigerants makes it impossible to give a general yes or no answer, but for a general answer Method 25 will give a more accurate measurement due to the elimination of that variable response issue with inconsistent sample streams.
Unless the refrigerant stream composition is consistent there is no way to accurately determine a response factor for a complex mixture of compounds when the composition can change not only between test runs but during the same test run. I pulled up the relative response for three of the refrigerants for a portable FID to illustrate this problem. The comparison is based on the “actual concentration” of the test divided by the “measured concentration” based on a calibration with methane. Thus methane would have a 1.0 factor in this situation.
I am using the three refrigerants R10, R20, and R30 for this comparison just for ease of use single carbon molecules, but there are many more with just as wide of a range of responses.
R10 has a listed response of 25.9, which means that this particular compound would be reported at a much lower concentration than actually present based on a calibration with methane. This would represent a reported concentration of 3.86 ppm for every 100 ppm of R10.
R20 has a listed response of 3.5, which means this compound would be reported as a lower concentration than actually present but closer than the R10 would provide. This would represent a reported concentration of 28.8 ppm for every 100 ppm of R20.
R30 has a listed response of 1.4, which means this compound is nearly the same in response as methane. This would represent a reported concentration of 71.4 ppm for every 100 ppm of R30. While much closer in response, this shows how even a smaller variation can have an impact.
If these compounds were consistent in the percentage of the refrigerant stream, it would not be that difficult to develop a correction factor to show the actual concentration in the stream from the measured concentration as methane. The measured concentration by Method 25-A calibrated with methane is most probably going to be significantly lower than the actual concentration, but just how much lower depends on the constituents.
Of course, a problem arises when the components in the stream vary instead of having all three in consistent percentages. For example, if there is more R10 and less R30 or more R20 and less R10, the corrected reported concentration will be proportionally lower than the actual concentration. If there is less R10 and more R20 or R30, the corrected reported concentration will be proportionally higher than the actual concentration in the stream. This is why the Method 25 would be more accurate for the determination of the actual concentration than Method 25-A and the development of a response factor for the sample stream.
I have been told this may be a new focus for one or more of the EPA Regions given the local pollution issue along with the concern over the long term effect on the ozone layer worldwide. I have read of studies indicating there is a new production source in Asia for CFC refrigerants. This new production has been the source of fugitive emissions that is affecting the previous downward trend of the ambient concentration of those types of refrigerants.
Triangle Environmental Services, Inc.
P.O. Box 13294 122 US Hwy 70 E
Research Triangle Park, NC 27709 Hillsborough, NC 27278
(919) 361-2890 (800) 367-4862 Fax: (919) 361-3474