News & Updates
CAUTION TO DATA USERS
“NJDEP-SRWM Low Level USEPA Method TO-15 March 2007"
Our repeated attempts to initiate constructive dialogue at various levels of the NJDEP Office of Site Remediation, the NJDEP Office of Quality Assurance and the National Environmental Laboratory Accrediting Program Board of Directors have proven fruitless. All we can do is address by way of caution data users of this method and regulatory bodies who may wish to incorporate it as a required reference method. Because the NJDEP version of EPA Method TO15 never underwent independent peer review, it suffers from several technical deficiencies which can directly affect data defensibility. To be brief I will only call your attention to the most import one……how a sample is taken out of the summa canister for analysis.
The New Jersey method mandates every sample be analyzed under receipt vacuum conditions (undiluted by makeup gas addition to bring the canister to positive pressure). However, many current TO15 laboratory analyzers (specifically the flow controllers used to accurately measure the volume of air withdrawn) are not rated by the manufacturer to operate under vacuum conditions. If there is insufficient pressure drop across the device, the device cannot operate in a controlled manner.
The uncontrolled state is recognizable by instrument inability to achieve set point and/or precipitous flow drop during sample withdrawal. Because of the logistics in how the quality control standards are introduced, the uncontrolled state of the instrument will not be caught by usual quality control measures. The degree to which the instrument is out of control will depend upon the vacuum in the canister and the volume of air removed for analysis. Low bias in the compromised data set will be the natural outcome. Data users may see duplicate results and co-located results with widely varying results.
When you click the button below you will see what the chemist in the lab sees when an aliquot of air is removed from a summa canister for analysis. The flow controller (rated for 10 to 100 mls/min with a 10 to 40 psi pressure drop from inlet to outlet) is preset to the desired volumetric flow and time interval. The digital read out is a NIST certified flow sensor that records actual flow produced. In the video we have set the device to 50 mls/minute for 10 minutes for a final sample volume of 500 mls. These are typical sample load conditions. The first video shows a standard EPA method TO15 pressure load maintaining control throughout the time duration. Due to web based limitations, we are only showing the first three minutes of sample load in each video. The second video shows a vacuum load with the canister at 5” Hg. Instrument inability to control flow to set point and instability in flow throughout the sampling period are obvious. When the same test is performed at 10”Hg the problems become more severe as portrayed in the third video. Inaccuracy in measurement can vary dramatically depending upon the vacuum present in the canister and desired load volume.
THREE VIDEOS MAY TAKE UP TO ONE MINUTE TO LOAD
|
Final Measured Flow Rate: 50 mL/min |
Instrument setpoint flow held throughout line flush and sample load periods. Instrument consistently in a controlled state. |
|
Final Measured Flow Rate: 34.2 mL/min |
Instrument flow controlled to set point throughout line flush period but drops precipitously during sample load period. Instrument is out of control during the sample load period. The lower flow rate means less volume of sample will be removed for analysis than expected. Degree of uncertainty in sample volume measurement may be as high as 35%. |
|
Final Measured Flow Rate: 24.8 mL/min |
Instrument flow not controlled to set point during both line flush and sample load periods. Instrument is out of control during the entire sampling process. Lower flow rates exhibited during the line flush period can lead to cross contamination between samples. Degree of uncertainty in sample volume measurement may be as high as 50%. |
Some labs incorporate a ‘time based correction’ in an attempt to approximate actual sample volume collected during the uncontrolled state. Incorporation of an approximation factor when determining initial sample volume severely compromises quantitative data generation. Consequently, data generated under the NJDEP SRWM Low Level USEPA Method TO15-March 2007 should be qualified as estimated and for screening purposes only unless the mandate for vacuum sample analysis is removed or until laboratory instruments rated under vacuum operating conditions by the manufacturer become available. In closing, the reporting limits established in the NJDEP ‘Low Level’ method are readily attainable using standard full scan EPA Method TO15 with slightly pressurized canisters and instruments operating in controlled states. The mandate for vacuum analysis of canisters negatively affects data accuracy, reproducibility and ultimate defensibility.
Linda L. Freeman
CEO
Editorial Commentary – August 13, 2007
“NJDEP-SRWM Low Level USEPA Method TO-15 March 2007"
In April of this year, Air Toxics Limited made a decision to delay accreditation under what was billed as a new and improved version of EPA method TO15 from the State of New Jersey and pursue various formal avenues available to us for dispute resolution. The essence of our complaint is the NJ Office of Data Quality appears to have independently drafted a very prescriptive compliance document specifying in great detail how air samples are to be analyzed. To our knowledge the method has not been reviewed by the scientific community nor promulgated by any regulatory community. Technical issues are present throughout the method and our attempts to engage in scientific discussion with the NJ-SRWM were declined in writing. We were informed the method was final even though the customary stages of scientific review had not been undertaken. We were informed we must comply with the criteria as written even though we know several are technically unfounded and may lead to compromised data. The method as it appears on the NJ-SRWM website:
- disallows the pressurization of samples necessary to improve data defensibility
- mandates vacuum analysis of the samples in direct contrast to how the accuracy and precision objectives of EPA Method TO15 were established
- mandates extra quality control check standards that substantially raise costs without providing significant additional value
- disallows data quality outliers in control standards making sample analysis statistically impossible
- disallows state of the art electronic data review and reporting critical to producing data of known and documented quality
Linda L. Freeman
CEO
August 13, 2007