Triangle NoTES - February 2012

The differences between NonMethane (NMOC) and NonMethane NonEthane (NMNEOC) Organic Compound analyses and the potential for bias in the results. Natural gas supplies may include significant percentages of ethane as it is the second-largest component of natural gas, but may be removed during normal processing.

A recent discovery that a state other than California was excluding ethane from the NMOC in some cases, caused us to review the way we have reported results for this type of situation. The pros and cons of the different procedures were discussed to make sure the procedures we used were the most technically correct.

The main problem with NMNEOC analysis is the definition of NMNEOC varies depending on how the analysis is performed and what equipment is used for that analysis. This means the same source could have samples taken at the same time and analyzed using any of the possible procedures and get a wide variety of different results. I will discuss these differences, their potential magnitude, and reason for them for the three basic methodologies of which I am aware.

The first and obvious methodology would be that based on the South Coast Air Quality Management District Method 25.1 analytical system. This system does not specify a particular column set so anything may be used which can meet the performance criteria specified. Since the column in this system not only separates compounds, but also acts as a trap, the trapping efficiency can have an impact on what is and is not seen in the analysis. In the analysis of a tank and trap this variation is generally much less due to the majority of the sample being in the trap fraction. However, there can be multiple variations due to the analytical procedures. The first is assuming the column backflush is changed to allow ethane to elute rather than only carbon dioxide. This may allow other compounds to elute after the carbon dioxide but before the ethane, depending on the column sets used. This may generally create a small negative bias, but in some cases the bias could be significant. The second bias relates to the analysis of a sample taken without the dry ice condensate trap, such as the case for landfill gases. The alternative columns can exhibit a variable recovery for different compounds compared to the standard columns used in the USEPA reference methods. Thus, different columns can give different results, which was documented in a study by Research Triangle Institute for the USEPA in the late 1990s. The best comparison for a compound tested on one of the column sets showed a 20% negative bias in this study. We had seen the worst comparison in a round robin program where one column set reported a non-detect for a sample showing hundreds of ppm by another SCAQMD based column set and the Method 25 column set. This choice has the potential for some very significant negative biases when compared to the USEPA reference method.

The second methodology is based on the USEPA Method 25 analytical system, which would also be modified to backflush the column set after the elution of ethane instead of after the elution of carbon dioxide. This eliminates the comparison issues, but still can maintain a negative bias. The elution of ethane is preceded by the elution of ethene and ethylene, so these compounds must be added back into the total or the definition of the term “NonEthane” changed accordingly. The other concern with this approach is whether the modification of the backflush timing will result in compounds being lost in the peak created on the chromatogram by the valve change upset. If such a loss were to happen it probably would be noticed if the compound were in sufficient concentration, which would probably tend to limit the bias to a specific quantity. The percentage of that bias could be more significant in lower sample concentrations. Thus, this method would appear to be better than the previous method due to the lower potential impact.

The third and final methodology is the one we have generally used in the cases where components have needed to be subtracted for the NMOC concentration. We have done this process for ethane and ethylene, which is why we know the elution order of these compounds for the column sets. The analysis of the NMOC per the Method 25 analytical requirements is followed by a second analysis where the compound of interest is allowed to elute before the column is backflushed. This elution time is confirmed by the injection of a known standard of that compound. Only the compound peak is used to quantify the concentration in the sample using the standard propane response. This allows the compound to be reported in ppmC just as the reference method results are reported. Thus, the compound can be directly subtracted from the reported NMOC concentration with no further adjustments. This is a little more expensive than the previous method, but the potential for bias is all but eliminated and there is also a set of results directly following a standard reference method which can be more accurately compared to a wider range of sources and time periods if necessary.

As I indicated, we have used this method to determine the three C2 compounds for subtraction from the NMOC concentration without issue. I do not know how well this process would work for any other compounds, but it does get a little more problematic as the elution time extends. There is a secondary concern of sufficient extension time allowed for the backflush to be completed since there could be some effect since the components could eventually have time to break through from one column to the other. This could potentially cause some cross contamination, but at a time removed from the initial analysis. As a result of this concern, I would not recommend the attempt for any C3 or above compound.

Of course, the better option would be a standard reference method for NMNEOC analysis if this were to be considered as a replacement for NMOC analysis on a wider scale. Until such time, I would recommend the standard NMOC analysis with the subtraction modification as the current choice for the most accurate results for all cases, especially outside California.

Wayne Stollings

Triangle Environmental Services, Inc.

Wstollings@aol.com