PS-19: Understanding the Six Core Tests for Continuous Emission Monitoring Systems (CEMS)

Facilities such as commercial sterilizers regulated by 40 CFR 63 Subpart O will soon become subject to Performance Specification 19 (PS-19) requirements, as outlined in Appendix B of 40 CFR 60.

PS-19 defines several required procedures for installing and certifying a CEMS, while Part 60, Appendix F, is expected to include Quality Assurance Procedure 7, which will stipulate a framework for maintaining CEMS compliance once certification has been achieved.

What is PS-19?

Arguably the most detailed and quality assurance-focused performance specification introduced so far, PS-19 is comprised of six primary tests. The guidance also necessitates the use of a time-shared system.

Source: Picarro

It should also be noted that:

1. The Interference Test will permit a passing response in cases where the sum of the interference responses is no more than 2.5% of the calibration span or ±3.0% of the equivalent EtO concentration employed.
2. Should the average RM emission level be less than 50% of the EtO concentration or emission standard, the emission standard may be utilized in the denominator of RA calculation. Results must also be ≤15%.
3. The source firm performing the Relative Accuracy Test Audit will generally use OTM-47 for the initial testing.

PS-19 Requirements

PS-19 begins with requirements similar to those of most CEMS. Certified reference gases must be used, all of which should meet the EPA Traceability Protocol for Assay and Certification of Gaseous Calibration Standards.

PS-19 deals with low-concentration gases; however, protocol gases may be diluted to meet the required values, as described in EPA Method 205.

To ensure the measurement point provides accurate measurements while remaining free of stratification, the CEMS must be fitted at a location at least (2.0) equivalent duct diameters downstream of a control device or flow disturbance. Its location must also be (0.5) equivalent duct diameters from the exhaust or any other control device.

PS-19 also requires a means of measuring the volumetric flow rate. If the EtO measurements require correction to account for moisture, moisture content should also be measured.

The flow monitor should be situated according to the above criteria, as well as being no more than (0.5) duct diameters away from the CEMS probe.

Where monitoring of moisture content is required, the system should either feature a continuous moisture sensor and oxygen analyzer able to measure oxygen on a wet and dry basis or include an optical measurement system that has been validated according to EPA Method 301.

It is a requirement that the CEMS system is capable of acquiring at least one valid measurement per measurement point, and this must occur at least once every 15 minutes. Time-sharing is a viable approach when meeting this requirement; for example, a system can monitor multiple measurement points with a single analyzer.

Once the system is fitted, set up, and calibrated appropriately, and these requirements are met, the facility can commence the Performance Specification Test Procedures. Below is an exploration of each of these six tests.

Interference Test

Interferences are common with many EtO measurement technologies, particularly those working with sources prone to gas matrix fluctuations, such as combustion sources.

Any gas other than the target gas can be used as an interference gas in EtO measurement applications if it leads the detection technology to potentially misreport the actual EtO levels.

The EPA requires that the CEMS manufacturer verify that their system can provide accurate EtO measurements even while exposed to select interference gases.

In a laboratory setting, the manufacturer must perform the interference test with a mixture of an EtO reference gas concentration at approximately 10 times the detection limit.

The following interferents may be introduced, either individually or as a blend:

• Methane (15 ppm - 25 ppm)
• Carbon Dioxide (0.8% - 1.2%)
• Water (4% - 6%)
• Nitrogen (as the balance gas)

The responses from interference testing results must not be more than 2.5% of the calibration span or 3.0% of the equivalent EtO concentration used to pass the test. The test is also considered passed in cases where the results do not exceed 10 times the detection limit or 30 ppbv.

Picarro CEMS Interference Test Results

Picarro undertook a series of interference tests on its CEMS system using a blend of CO₂, CH₄, and H₂O in N₂ with 50 ppbv of EtO (Figure 1).

Interference changes were negligible in this case, with average results of just ~1.3 ppbv drift. This result represents extremely precise readings, significantly reducing the risk of overreporting emissions where other gases are present within matrices.

Figure 1. Interference Gas Test. Image Credit: Picarro

Level of Detection (LOD)

Ascertaining the system’s LOD is essential to ensuring that EtO can be detected above a background in a gas matrix. The interference gases discussed earlier are also employed when conducting this test.

This test is typically performed in a laboratory environment. A blend of the interference gases and an EtO concentration (no more than 10 times the estimated limit of detection) is injected into the analyzer’s inlet for 15 minutes. Results are recorded and averaged, and the test is repeated seven times.

The LOD is determined by calculating 3x the standard deviation of the average measurement. However, the LOD must not exceed 20% of the applicable emission limit.

The controlled environment LOD must then be verified. This is achieved by performing dynamic spikes through the installed CEMS system to calculate a site-specific standard addition detection level. This level must also be <20% of the applicable emission limit.

Picarro CEMS LOD Results

To demonstrate compliance with PS-19, Picarro undertook a series of LOD tests with 6 ppbv of EtO and 1 ppbv of EtO (Figure 2). Results revealed a LOD of <0.2 ppb for each series of testing.

These results mean that clients can be assured they are consistently recording accurate emissions, even when their process is reporting single digit ppbv concentrations.

Figure 2. LOD Test Results – Picarro CEMS. Image Credit: Picarro

Measurement Error Test

The Measurement Error Test validates linearity by sending reference gases to the CEMS probe and through the whole system.

This involves introducing a zero gas and three upscale EtO reference gases to the probe non-consecutively, with gases flowing until a stable response is achieved. In this instance, a stable response would require two measurements to be collected and concentrations found to be within 1% of the instrument's span or 5 ppbv.

This process is repeated three times per gas for a total of 12 measurements. Each result must be within 5% of the instrument’s span, or there must be an absolute difference of 10 ppbv at the low-, mid-, and high-level concentrations.

Picarro CEMS Measurement Error Test Results

Picarro performed measurement error testing by recording the results of step changes across different spans over several days. Figure 3 features several randomized results from this specific test (100 ppbv span). These results show that Cavity Ring Down Spectroscopy (CRDS) is extremely precise when measuring EtO across an instrument's span.

Figure 3. Partial Measurement Error Test Table. Image Credit: Picarro

Response Time Determination

The response time of the CEMS system must be calculated while performing the Measurement Error Test.

A stable zero reading must be achieved for 30 seconds, and the time that this was acquired (hh:mm:ss) before introducing the high-level upscale gas must be recorded. The time the upscale gas achieves 95% of the final stable value is then recorded. Zero gas is then introduced, and the time that a stable value is reached is recorded again.

This measurement process is performed three times to enable the recording and calculation of the mean upscale and downscale response times.

Response time is an essential metric for systems utilizing a time-shared, multi-stack CEMS.

Each measurement point must be sampled for at least 3x the measured response time, but the total duration of measurement across all measurement points in a time-shared system should not exceed 15 minutes. This constraint limits the number of measurement points a given system can accommodate.

The sample line length and the specific sample pump used by the CEMS system are primary limiting factors in response time tests.

Picarro mitigates these limitations through the use of a robust sample pump able to accommodate comfortable time-sharing capabilities of three measurement points, even when working with sample lines that are longer than the average.

Calibration Drift Test (7-Day Drift Test)

The 7-day drift—or calibration drift—test must be regularly performed throughout the CEMS’s lifetime. This test also establishes the system’s initial stability for certification purposes.

In this instance, the system is defined as every component from the probe to the analyzer. The test requires the system to be challenged with zero gas and span gas at 24-hour intervals.

To maintain quality and meet the PS-19 requirements, the CEMS must achieve a <5% drift from span for the zero and span gas for seven consecutive days. This test would also be passed if the zero and span values only deviate by 10 ppbv or less.

Seven-Day Drift Assessment of Picarro

Picarro has undertaken several calibration drift assessments across a wide range of sites. The dataset in Figure 4 represents a completed seven-day drift assessment for an instrument spanning 200 ppbv, undertaken on behalf of a confidential client. The results presented here are consistently precise.

Figure 4. Example Seven-Day Drift Assessment. Image Credit: Picarro

Relative Accuracy Test Audit (RATA)

The final test involves CEMS measurements being compared against a Reference Method (RM), then displayed and noted in terms of the emission standard. PS-19 lists Method 320 as an approved RM, but as the only method specific to EtO, OTM-47 can be alternatively approved.

OTM-47 leverages the power of CRDS to provide sub-ppbv measurements in two-second intervals while maintaining minimal issues with interference. Facilities working with low levels of EtO are advised to request OTM-47 to ensure that all measured data is accurately represented.

A typical RATA comprises at least nine 21-minute runs and must be conducted by a certified source testing firm. Each run takes pollutant, diluent, and flow measurements (if applicable to the emission standard), with overall Relative Accuracy (RA) calculated after the final run based on the results of the CEMS and RM.

The RA must be ≤20% of the reference method if this is to be used in the denominator of the RA calculation. Where certain criteria are met, alternative specifications are also available for calculating RA with the emission standard in the denominator.

Picarro RATA Solution

Picarro was instrumental in developing OTM-47, the only test specific for EtO. OTM-47 has since been employed throughout the United States and its territories, and remains the best solution for performance testing at low detection levels.

However, it is important to note that Picarro does not own the rights to performance testing results.

Figure 5. Picarro mobile OTM-47 solution. Image Credit: Picarro

PS-19 Testing Results: Picarro CRDS Versus Alternative Technology

Before the finalization of Subpart O and PS-19, Picarro and the US EPA worked together to conduct vigorous PS-19 testing using CRDS.

Figure 6 displays Picarro’s internal results for the required tests versus those published by a competitor.

Image Credit: Picarro

Maintaining CEMS Compliance

After every quality assurance test has been completed, the facility must send the results to the applicable administrator, who will evaluate these and provide approval on the certification of the CEMS if this has been achieved.

Once approved, the facility is required to maintain its CEMS with regular preventative maintenance according to Quality Assurance Procedure 7.

Each facility must also maintain a quality assurance plan that outlines and provides instructions on the functions necessary to maintain a functioning CEMS. The development and maintenance of this plan can be challenging, with even power generation facilities possessing years of CEMS experience opting to delegate the creation or updating of these plans to specialized consulting firms.

To help address this widespread difficulty, Picarro provides every customer with a site-specific plan as part of their purchase, which includes the required components and any pertinent additional details.

Simple PS-19 Compliance with Picarro

Picarro leverages its decades of experience, a suite of robust measurement tools, and managed services expertise to simplify compliance with the new CEMS rules.

The company’s EtO system is designed to be seamlessly integrated into any facility, and its Environmental Systems (ESYS) team is always on hand to offer guidance and support.

Picarro’s systems employ a robust Data Acquisition System configured to enable easy reporting to the Electronic Reporting Tool (ERT). Each system is supplied in a robust, portable cabinet factory equipped with remote Cylinder Gas Audits, automated Above-Span Calibration checks, and remote access for expert CEMS support where required.

The company’s environmental experts have designed and created Quality Assurance documents that comply with federal regulations and operating permits. The team is available to support clients with any reporting needs.

A team of trained field technicians is also on hand, ensuring that all preventative maintenance and emergency support is completed efficiently and in a timely manner. This team also schedules annual RATAs, communicating with a system’s administrator and explaining why OTM-47 is the best option for performing the RATA.

Picarro systems offer a wide range of benefits, including:

• Rapid response times allow timeshare systems to be used, reducing overall costs. One analyzer can accommodate up to three separate measurement points.
• Picarro systems offer the lowest-in-class detection limit, meaning there is no need to overreport emissions or wait 15 minutes to generate a low response. These systems typically generate accurate results on two-second averages.
• End-to-end emissions monitoring solutions are provided as standard, with Picarro’s team installing each CEMS and providing operational support throughout a system’s working life.

Acknowledgments

Produced from materials originally authored by Sean Cronin from Picarro, Inc.

This information has been sourced, reviewed and adapted from materials provided by Picarro.

For more information on this source, please visit Picarro.