2019.05.29_CCO.p11_ChemoursCombinedComments

Scott Sivertsen – Review of submitted SOPs
Review of 11_NCDWR_Table 3+ SOP TestAmerica_05062019.pdf
This is a complete re-write of the SOP previously submitted for review and bears little resemblance to the procedures identified previously. Many of the prior comments are no longer
valid.
The use of second-source standards should be implemented. Without an independent source of standards, the quantitative accuracy of the data will not be able to be judged.
No reference to isotopically labeled reference standards is seen. Suggest that the use of isotopically labeled standards be implemented to verify acceptable injection performance. Having
at least one labeled compound may alert the analyst to injection anomalies.
Page 8, § 7.9 – The use of 100% water as a dilution solvent should be evaluated. The literature shows water to be inadequate in keeping PFAS in solution. Additional details should be
given on acceptable storage containers for long-term storage of the stock solution.
Page 10, § 9.2.1 – No mention is made of an LCS-type sample spiked at the minimum reporting level (MRL). Suggest addition of an MRL verification sample at a frequency of one-per-batch,
to verify method performance at the MRL.
Page 13, § 10.8.1.2 – No acceptance criteria for second-order calibration is stated. Specify criteria or disallow the use of the quadratic calibration model.
Page 13, § 10.8.1.3 – The stated %RSD at 50% is too high. Suggest that it be no higher than 20% and preferably 15%. (https://www.epa.gov/sites/production/files/2014-05/documents/calibration-guide-ref-final-oct2010.pdf)
Page 13, § 10.9 – The use of internal standards for calculations is referenced. The ICAL standard instructions in § 7.10 do not include internal standards. Clarify.
Page 15, § 11.2 – It appears that a 2.5-mL sample aliquot is being taken from a larger sample volume (250 mL?). Studies have shown that in the absence of sufficient organic solvent (typically
50% v/v) this practice will result in low bias due to surface adsorption of PFAS to sample containers. This is an unacceptable practice; the sample must be treated as a whole; aliquoting
is not permitted prior to addition of organic solvent. This comment also applies to matrix spike preparation.
Pages 17 – 18 – Suggest adding a qualitative product ion, where possible, to aid in compound identification. Lack of a confirmatory ion will increase chances of false positives.
Review of 11_NCDWR_Table 3+ SOP Eurofins_05062019.pdf
This is a complete re-write of the SOP previously submitted for review and bears little resemblance to the procedures identified previously. Many of the prior comments are no longer
valid.
The use of second-source standards should be implemented. Without an independent source of standards, the quantitative accuracy of the data will not be able to be judged.
Suggest that the use of isotopically labeled standards be implemented to verify acceptable injection performance. Having at least one labeled compound may alert the analyst to injection
anomalies.
Pages 9 – 10, Step 5; Calibration concentrations – concentrations appear to be incorrect. Verify dilutions/ICAL concentrations in the table.
Page 11, Step C., Native spiking solution – The use of 100% water as a dilution solvent should be evaluated. The literature shows water to be inadequate in keeping PFAS in solution.
There is no mention of what container the solution is transferred to for long-term storage following preparation.
Page 13, Procedure; Step A.2 – It appears that a 5-mL sample aliquot is being taken from a larger sample volume (250 mL?). Studies have shown that in the absence of sufficient organic
solvent (typically 50% v/v) this practice will result in low bias due to surface adsorption of PFAS to sample containers. This is an unacceptable practice; the sample must be treated
as a whole; aliquoting is not permitted prior to addition of organic solvent. This comment also applies to step B.2, matrix spike preparation.
Page 14, Step B.2 – The referenced attachment (Acquisition method) is not available: this is a critical element of the analysis. No evaluation of the procedures can be made.
Page 14, Step B.8 – Reference is made to relative retention times. No compounds have been identified as retention markers in the SOP; therefore, retention times are absolute, not relative.
Page 15, Quality Assurance/Quality Control – No mention is made of an LCS-type sample spiked at the minimum reporting level (MRL). Suggest addition of an MRL verification sample at a
frequency of one-per-batch, to verify method performance at the MRL.
Comments on additional Topics
Whether use of spiked field samples should be required (EPA ORD thoughts?)
Mark Strynar - My opinion is this is a needed addition to any analysis set to be conducted.  We did this for all of our work with NCDEQ in the past.  One must be able to demonstrate
two things with these assays: 1) blank samples are blank and 2) a spiked sample carried out in the field and measured along with collected samples at a later date is accurate.  This
is not the same as a matrix spike or a spiked sample made up on the day of analysis.  I have had some discussions with Chemours on this in the recent past.  I except a sample spiked
somewhere between 2x-10x the LOQ is an appropriate spiked level to aim for a trip spiked sample.
Require sample synthesized compounds to be submitted to NCDEQ and EPA R4/ORD as they are developed.
Scott Sivertsen - I think that NC should, if it can, require Chemours to fund the synthesis and purification of target compounds outside of their facilities. The standards which we received
would not be marketable as analytical standards on the open-market. Most (80%) lacked purity information, were demonstrated to contain interfering compounds and the solvent system which
they were distributed in (water) makes any stated concentration of those solutions questionable. Additionally, there are no second-source standards to verify primary standard concentration.
While Chemours gifting of those compounds is appreciated, they cannot be considered quantitative standards.
Mark Strynar -I believe it is crucial that Chemours share any synthesized standards with the appropriate labs as some of these standards are only available from them.  I think any standards
should be >90% pure, and if possible the standard could come with what other analytes are in any one standard. 
Allen Martin – It’s my opinion that we need high purity analytical standards that are prepared by an outside source. Ideally we would also have a second source for all analytical standards
to verify the concentrations of our primary solutions.
Require Q-TOF-MS data files and chromatograms to be submitted to NCDEQ with all results, which will be shared with EPA R4/ORD.
Mark Strynar - I believe in response comment T2-5 they agree to share data files with NCDEQ.
QQQ Acquisition Parameter – major concern in monitoring only one MRM for all of the compounds.
Mark Strynar - A modern day QQQ as I know Chemours is using will have no loss in sensitivity or selectivity by adding 2 MRMs for each analyte if they are available. Having 2 MRMs is
a needed parameter for any good analytical assay.
Allen Martin - I agree with Mark that when available 2 transitions should be monitored to generate reliable analytical data.