Interpreting Your CSIA Laboratory Report
For many of you, this is the first CSIA report you have received from Microseeps. Because this analysis is different than the concentration analyses that are typically obtained from a laboratory, this report is a bit unusual. This document is an attempt to help you better comprehend the report, and more importantly, how to use it. This interpretation will also focus on the report itself and how its content relates to your samples. It does not attempt to interpret the results in terms of site remediation; we have included resources at the end of this document that will assist in this.
Sample Results Area When chromatography (such as GC) is used to measure concentration, the instrument reports an area for each compound and the initial calibration is used to convert that area into a concentration. While the instrument used in CSIA can provide very sensitive and accurate measurements of concentration, the purpose of the instrument is to provide sensitive and accurate isotopic ratio measurements. When looking to measure concentration, SW846-8260B and EPA-624 are highly recommended methods. Microseeps’ CSIA quality control (QC) program has been modeled after those methods because they use a very good and widely recognized QC program, however, CSIA is not an alternative to those methods when measuring concentration. As such, the CSIA section of Microseeps’ laboratory report does not specify concentrations, but area, reported in volt∙seconds (Vs). Area is generally proportional to concentration and area values are provided in your CSIA report so that the two points can be assessed.
Signal Strength As with any instrument, there is some minimum threshold of signal required for a reliable isotopic ratio to be accurately measured. The PQL (Practical Quantitation Limit) is a systematically determined “reporting limit” that gives the minimum area necessary to produce an isotopic ratio measurement accurate to within ±0.5 per mil. The CSIA Standard Operating Procedure (SOP-AM24) gives the detailed procedure used for PQL determination. The measurement is somewhat less accurate, but still reliable, when the area is below the PQL. The range of this accuracy is very matrix dependent - our experience has shown that duplicates still match to within two or three per mil when the signal is at least 80% of the PQL, though it can be considered that the method detection limit (MDL) is 80% of the PQL. Any areas which lie between the MDL and PQL are flagged with a J. Areas that are less than the MDL are treated as “non-detects” and the area is reported as being less than the PQL and flagged with a “U”. The measured area is reported to the user so that they can see if the signal strength was sufficient and/or what flags should be put upon the data.
Dilution If the signal is too strong, an accurate measurement of the isotopic ratio can not be made. Therefore, the signal must be reduced through the recommended method of dilution. When measuring isotopic ratios it is very important to treat all masses the same; so-called “mass discrimination” occurs whenever anything is done that treats the masses unequally. While each of the techniques used for the reduction of signal strength can be performed without the introduction of mass discrimination, the most robust way to avoid it is by diluting samples that are too concentrated. For CSIA work, Microseeps always uses dilution to reduce signal strength for highly concentrated samples. The area is reported to the user so that they can assess whether a sample was properly diluted.
Co-elution In CSIA, compounds can only be separated as well as the gas chromatograph can separate them. A concentration measurement such as an SW846-8260 then further distinguishes by mass. However, in CSIA, all analytes are first combusted to carbon dioxide before entering the mass spectrometer. As such, the mass spectrometer does not contribute to the identification of analytes; therefore, if a co-contaminant elutes from the chromatographic column with a target analyte, the CSIA measurement of said analyte may not be entirely of a single compound, but may be corrupted by the presence of the co-eluting compound.
This can be detected through two techniques. The first is assessment of the peak shape of the raw IRMS signals for the individual masses; this is available directly from the IRMS. The second way to detect co-elution is by a comparison of the CSIA areas with the concentrations measured through a traditional GCMS analysis. Microseeps does use such a comparison, and it allows us to detect complete co-elutions in which the peaks have exactly the same retention time, so the co-elution does not affect the peak shape.
Co-elution may be very slight, or may be complete. Often, complete co-elution problems are discussed in the case narrative. However, in all cases the data user should realize that if co-elution is reported, the integrity of the data point is suspect.
Analysis A unique, sequential “analysis number” is automatically assigned to every analysis. This number, provided with each result, identifies the analysis that measured the reported result. The data-user can use it as the cross-reference to the surrogate table. From that, the data user can see what the area response of the surrogate was in that sample, what dilution was analyzed to get the reported result and what del was measured for the surrogate in that sample.
Del This is the ultimate CSIA result. It is linearly related to the isotopic ratio, but expressed in more convenient units. This is the result the data user needs the most. How it relates to the measured values is discussed in SOP-AM24, but other than for a one time data validation, that information is not required.
Batch Quality Control The blank, LCS_Lo and LCS_Hi are all QC samples. When the samples are run with each instrument batch, they have results particular to each analyte.
Blank The blank serves the typical role of proving that there is no contamination left from the previous run. While we are not measuring concentration, carry-over can still pose a problem because the del values of a particular analyte differs from sample to sample. If there is carry-over, the measured del will be a mixture of the current sample and the previous one. As such, the data user would like to see that for each target analyte in the blank there is a U, but for the surrogate in the blank there is a strong signal and the expected del is measured for that surrogate. (The absence of target analytes is proven by the U’s, and the surrogate response proves that if the target analytes were present, they would have been seen, so the U’s are valid.)
LCS_Lo and LCS_Hi To insure quality, Laboratory Control Samples (LCS) are constructed from laboratory grade DI water spiked with all of the analytes. The del values measured from these samples can be checked against the “LCS accepted value” and the results should lie within that range. These values were formed by repeated analysis of these constituents in laboratory derived samples, and details of that procedure are available in Microseeps’ SOP-AM24. The LCS samples serve multiple purposes:
Measure the del of each target analyte in a isotopically known sample. This ensures that the measurements are accurate at the concentration in the LCS. Measure the del in two samples that are isotopically identical but differ in concentration. This ensures there is no significant concentration dependence to the del measurements. Calculate the calibration factor for later use in diagnosing co-elution. The contents of the LCS’s are completely known, and there should be no co-elution in them. While this method is not intended to measure concentration, the area response, corrected for dilution, should be proportional to the concentration. That proportionality is used to insure that the peak area of a sample could all be attributed to the target analyte. If the peak area in a sample peak is larger than would be expected given the measured concentration, there is probably a co-elution. Co-elutions have been discussed in more detail above. Matrix Spikes and Laboratory Control Samples
Matrix spikes and matrix spike duplicates are useful tools for validating concentration results. While these tools are unfortunately not meaningful in CSIA, several things have been done to make up for this. Duplicate samples are analyzed more frequently, and rather than just a single LCS, two LCS’s are used.
To insure that the sample is properly analyzed - that is, purged, cryo-focused and injected, and then separated, combusted and measured - a surrogate is injected into every field sample and every QC sample during analysis. The surrogate is chosen to be something that would not be present in a field sample, but there is still the potential for corruption of the surrogate peak by interference present in the sample. Surrogate Acceptance Limits are given and the measured del should be within those limits, unless a co-elution is reported.
In the case narrative, any exceptions are discussed, as is their potential effect upon the reported data. Microseeps makes every effort to not only issue the most valid reports possible, but to also flag any suspect data that is reported. In no way is the case narrative intended to alarm the data user, nor is it intended to be a list of excuses for the laboratory. Rather, the case narrative is used to be a succinct description of the analytical project where the data validity is assessed and any potential detractions from that validity are explained in terms of their cause and their effect upon the reported data. This is presented to empower the data user to be more confident about their data.
This concludes the discussion of the report you have received from Microseeps. As mentioned previously, interpretation of what the del values mean for your site is another very important issue. While that is beyond the scope of this document, there are several excellent sources of information we would like to point out:
“Monitored Natural Attenuation of MTBE as a Risk Management Option at Leaking Underground Storage Tank Sites.” 2005. USEPA, EPA/600/R-04/1790
It is focused upon MTBE, but it covers a lot of vital fundamentals.
“Compound Specific Isotope Analysis: The Science, Technology and Selected Examples from the Literature with Application to Fuel Oxygenates and Chlorinated Solvents.” 2007. Available at http://www.microseeps.com/pdf/csia.pdf
This paper is a review of much of the available literature, starting from the fundamental basics and covering MTBE remediation, chlorinated solvent remediation, biodegradation and such other topics as CSIA and ISCO or CSIA and modeling.
The USEPA, in cooperation with the IAEA of the UN, is preparing “A Consensus Guide for Assessing Biodegradation and Source Identification with Compound Specific Isotope Analysis (CSIA).” That document is planned for release in October of 2008. It was prepared by a team of renowned experts and covers a multitude of issues in great depth.
We are confident that you will find the report you have received to be a very useful tool, and that CSIA can be a powerful part of your remediation work. Please feel free to contact us not only about work that has been done or is ongoing, but also work that is planned. We would love to discuss your project goals and potential ways we can fill them.