Necessary equipment for performing Method 30B
What is Method 30B?
Method 30B is a procedure for measuring total vapor phase mercury (Hg) emissions from coal-fired combustion sources using sorbent trap sampling and an extractive or thermal analytical technique. This method is only intended for use only under relatively low particulate conditions (e.g., sampling after all pollution control devices). Quality assurance and quality control requirements are included to assure that you, the tester, collect data of known and acceptable quality for each testing program. This method does not completely describe all equipment, supplies, and sampling and analytical procedures you will need, but instead refers to other test methods for some of the details. Therefore, to obtain reliable results, you should also have a thorough knowledge of these additional methods which are found in Appendices A-1 and A-3 to this part:
(a) Method 1—Sample and Velocity Traverses for Stationary Sources.
(b) Method 4—Determination of Moisture Content in Stack Gases.
(c) Method 5—Determination of Particulate Matter Emissions from Stationary Sources.
What does this method determine? This method is designed to measure the mass concentration of total vapor phase Hg in flue gas, including elemental Hg (Hg0) and oxidized forms of Hg (Hg+2), in micrograms per dry standard cubic meter (μg/dscm).
|Analyte||CAS No.||Analytical Range and Sensitivity|
|Elemental Hg (Hg0)||7439-97-6||Typically 0.1 μg/dscm to > 50 μg/dscm|
|Oxidized Hg (Hg+2)||-----||(Same)|
When is this method required? Method 30B is a reference method for relative accuracy test audits (RATAs) of vapor phase Hg CEMS and sorbent trap monitoring systems installed at coal-fired boilers and is also appropriate for Hg emissions testing at such boilers. It is intended for use only under relatively low particulate conditions (i.e., sampling after all pollution control devices); in cases where significant amounts of particle-bound Hg may be present, an isokinetic sampling method for Hg should be used. Method 30B may also be specified by New Source Performance Standards (NSPS), National Emission Standards for Hazardous Air Pollutants (NESHAP), emissions trading programs, State Implementation Plans (SIPs), and operating permits that require measurement of Hg concentrations in stationary source emissions, either to determine compliance with an applicable emission standard or limit, or to conduct RATAs of Hg CEMS and sorbent trap monitoring systems.
How good must my collected data be? Method 30B has been designed to provide data of high and known quality for Hg emissions testing and for RATA testing of Hg monitoring systems, including CEMS and sorbent trap monitors. In these and other applications, the principal objective is to ensure the accuracy of the data at the actual emissions levels and in the actual emissions matrix encountered. To meet this objective, NIST-traceable calibration standards must be used and method performance tests are required.
Known volumes of flue gas are extracted from a stack or duct through paired, in- stack sorbent media traps at an appropriate flow rate. Collection of mercury on the sorbent media in the stack mitigates potential loss of mercury during transport through a Probe/sample line. For each test run, paired train sampling is required to determine measurement precision and verify acceptability of the measured emissions data. A field recovery test which assesses recovery of an elemental Hg spike to determine measurement bias is also used to verify data acceptability. The sorbent traps are recovered from the sampling system, prepared for analysis as needed, and analyzed by any suitable determinative technique that can meet the performance criteria.
3.1 Analytical System is the combined equipment and apparatus used to perform sample analyses. This includes any associated sample preparation apparatus e.g., digestion equipment, spiking systems, reduction devices, etc., as well as analytical instrumentation such as UV AA and UV AF cold vapor analyzers.
3.2 calibration Standards are the Hg containing solutions prepared from NIST traceable standards and are used to directly calibrate analytical systems.
3.3 Independent calibration Standard is a NIST traceable standard obtained from a source or supplier independent of that for the calibration standards and is used to confirm the integrity of the calibration standards used.
3.4 Method Detection Limit (MDL) is the lowest mass of Hg greater than zero that can be estimated and reported by your candidate analytical technique. The MDL is statistically derived from replicate low level measurements near your analytical instrument’s detection level.
3.5 NIST means the National Institute of Standards and Technology, located in Gaithersburg, Maryland.
3.6 Run means a series of gas samples taken successively from the stack or duct. A test normally consists of a specific number of runs.
3.7 sorbent Trap means a cartridge or sleeve containing a sorbent media (typically activated carbon treated with iodine or some other halogen) with multiple sections separated by an inert material such as glass wool. These sorbent traps are optimized for the quantitative capture of elemental and oxidized forms of Hg and can be analyzed by multiple techniques.
3.8 Test refers to the series of runs required by the applicable regulation.
3.9 Thermal Analysis means an analytical technique where the contents of the sorbent traps are analyzed using a thermal technique (desorption or combustion) to release the captured Hg in a detectable form for quantification.
3.10 Wet Analysis means an analytical technique where the contents of the sorbent tube are first leached or digested to quantitatively transfer the captured Hg to liquid solution for subsequent analysis.
Interferences may result from the sorbent trap material used as well as from the measurement environment itself. The iodine present on some sorbent traps may impart a negative measurement bias. High levels of sulfur trioxide (SO3) are also suspected to compromise the performance of sorbent trap Hg capture. These, and other, potential interferences are assessed by performing the analytical matrix interference, Hg0 and HgCl2 analytical bias and field recovery tests.
What safety measures should I consider when using this method? This method may require you to work with hazardous materials and in hazardous conditions. You are encouraged to establish safety procedures before using the method. Among other precautions, you should become familiar with the safety recommendations in the gas analyzer user’s manual. Occupational Safety and Health Administration (OSHA) regulations concerning use of compressed gas cylinders and noxious gases may apply.
Prior to applying these procedures/specifications in the field, the potential hazards at the test site should be considered; advance coordination with the site is critical to understand the conditions and applicable safety policies. At a minimum, portions of the sampling system will be hot, requiring appropriate gloves, long sleeves, and caution in handling this equipment.
Policies should be in place to minimize risk of chemical exposure and to properly handle waste disposal in the laboratory. Personnel shall wear appropriate laboratory attire according to a Chemical Hygiene Plan established by the laboratory.
The toxicity and carcinogenicity of any reagents used must be considered. Depending upon the sampling and analytical technologies selected, this measurement may involve hazardous materials, operations, and equipment and this method does not address all of the safety problems associated with implementing this approach. It is the responsibility of the user to establish appropriate safety and health practices and determine the applicable regulatory limitations prior to performance. Any chemical should be regarded as a potential health hazard and exposure to these compounds should be minimized. Chemists should refer to the Material Safety Data Sheet (MSDS) for each chemical used.
Any wastes generated by this procedure must be disposed of according to a hazardous materials management plan that details and tracks various waste streams and disposal procedures.
The following list is presented as an example of key equipment and supplies likely required to measure vapor-phase Hg using a sorbent trap sampling system. It is recognized that additional equipment and supplies may be needed. Collection of paired samples is required.
A typical sorbent trap sampling system is shown in Figure 30B-1 in Section 17.0. The sorbent trap sampling system shall include the following components:
The sorbent media used to collect Hg must be configured in a trap with at least two distinct segments or sections, connected in series, that are amenable to separate analyses. Section 1 is designated for primary capture of gaseous Hg. Section 2 is designated as a backup section for determination of vapor phase Hg breakthrough. Each sorbent trap must be inscribed or otherwise permanently marked with a unique identification number, for tracking purposes. The sorbent media may be any collection material (e.g., carbon, chemically-treated filter, etc.) capable of quantitatively capturing and recovering for subsequent analysis, all gaseous forms of Hg in the emissions from the intended application. Selection of the sorbent media shall be based on the material’s ability to achieve the performance criteria contained in this method as well as the sorbent’s vapor phase Hg capture efficiency for the emissions matrix and the expected sampling duration at the test site. The sorbent media must be obtained from a source that can demonstrate their quality assurance and quality control (see Section 7.2). The paired sorbent traps are supported on a Probe (or Probes) and inserted directly into the flue gas stream.
Each Probe assembly shall have a leak-free attachment to the sorbent trap(s). Each sorbent trap must be mounted at the entrance of or within the Probe such that the gas sampled enters the trap directly. Each Probe/sorbent trap assembly must be heated to a temperature sufficient to prevent liquid condensation in the sorbent trap(s). Auxiliary heating is required only where the stack temperature is too low to prevent condensation. Use a calibrated thermocouple to monitor the stack temperature. A single Probe capable of operating the paired sorbent traps may be used. Alternatively, individual Probe/sorbent trap assemblies may be used, provided that the individual sorbent traps are co-located to ensure representative Hg monitoring.
A moisture removal device or system shall be used to remove water vapor from the gas stream prior to entering dry gas flow metering devices.
Use a leak-tight, vacuum pump capable of operating within the system’s flow range.
A gas flow meter (such as a dry gas meter, thermal mass flow meter, or other suitable measurement device) shall be used to determine the total sample volume on a dry basis, in units of standard cubic meters. The meter must be sufficiently accurate to measure the total sample volume to within 2 percent and must be calibrated at selected flow rates across the range of sample flow rates at which the sampling train will be operated. The gas flow meter shall be equipped with any necessary auxiliary measurement devices (e.g., temperature sensors, pressure measurement devices) needed to correct the sample volume to standard conditions.
Use a flow rate indicator and controller for maintaining necessary sampling flow rates.
Same as Section 184.108.40.206 of Method 5 in Appendix A- 3 to this part.
Same as Section 6.1.2 of Method 5 in Appendix A-3 to this part.
Device for recording associated and necessary ancillary information (e.g., temperatures, pressures, flow, time, etc.).
A known mass of gaseous Hg0 must be either present on or spiked onto the first section of sorbent traps in order to perform the Hg0 and HgCl2 analytical bias test and the field recovery study. Any approach capable of quantitatively delivering known masses of Hg0 onto sorbent traps is acceptable. Several spiking technologies or devices are available to meet this objective. Their practicality is a function of Hg mass spike levels. For low levels, NIST-certified or NIST-traceable gas generators or tanks may be suitable. An alternative system, capable of delivering almost any mass required, makes use of NIST-certified or NIST-traceable Hg salt solutions (e.g., HgCl2, Hg(NO3)2). With this system, an aliquot of known volume and concentration is added to a reaction vessel containing a reducing agent (e.g., stannous chloride); the Hg salt solution is reduced to Hg0 and purged onto the sorbent trap using an impinger sparging system. When available, information on example spiking systems will be posted at http://www.epa.gov/ttn/emc.
Any analytical system capable of quantitatively recovering and quantifying total Hg from the sorbent media selected is acceptable provided that the analysis can meet the performance criteria described in this method. Example recovery techniques include acid leaching, digestion, and thermal desorption/direct combustion. Example analytical techniques include, but are not limited to, ultraviolet atomic fluorescence (UV AF), ultraviolet atomic absorption (UV AA) with and without gold trapping, and X-ray fluorescence (XRF) analysis.
If correction of the measured Hg emissions for moisture is required (see Section 220.127.116.11), either Method 4 in Appendix A-3 to this part or other moisture measurement methods approved by the Administrator will be needed to measure stack gas moisture content.
Only NIST-certified or NIST-traceable calibration standards, standard reference materials, and reagents shall be used for the tests and procedures required by this method.
The sorbent trap media shall be prepared such that the material used for testing is of known and acceptable quality. sorbent supplier quality assurance/quality control measures to ensure appropriate and consistent performance such as sorptive capacity, uniformity of preparation treatments, and background levels shall be considered.
This section presents the sample collection and handling procedures along with the pretest and on-site performance tests required by this method. Since you may choose different options to comply with certain performance criteria, each test report must identify the specific options selected and document the results with respect to the performance criteria of this method.
What sampling site and sampling points do I select? Same as Section 8.1 of Method 30A of this appendix.
What performance criteria must my measurement system meet? The following laboratory and field procedures and associated criteria of this section are designed to ensure (1) selection of a sorbent and analytical technique combination capable of quantitative collection and analysis of gaseous Hg, (2) collection of an adequate amount of Hg on each sorbent trap during field tests, and (3) adequate performance of the method for each test program: The primary objectives of these performance tests are to characterize and verify the performance of your intended analytical system and associated sampling and analytical procedures, and to define the minimum amount of Hg (as the sample collection target) that can be quantified reliably.
(a) Analytical Matrix Interference Test;
(b) Determination of Minimum Sample Mass;
(c) Hg0 and HgCl2 Analytical Bias Test;
(d) Determination of Nominal Sample Volume;
(e) Field Recovery Test.
(a) The analytical matrix interference test is a laboratory procedure. It is required only if you elect to use a liquid digestion analytical approach and needs to be performed only once for each sorbent material used. The purpose of the test is to verify the presence or absence of known and potential analytical matrix interferences, including the potential negative bias associated with iodine common to many sorbent trap materials. The analytical matrix interference test determines the minimum dilution (if any) necessary to mitigate matrix effects on the sample digestate solutions.
(b) The result of the analytical matrix interference test, i.e., the minimum sample dilution required (if any) for all sample analyses, is used to establish the minimum sample mass needed for the Hg0 and HgCl2 analytical bias test and to determine the nominal sample volume for a test run. The analytical matrix interference test is sorbent material-specific and shall be performed for each sorbent material you intend to use for field sampling and analysis. The test shall be performed using a mass of sorbent material comparable to the sorbent mass typically used in the first section of the trap for sampling. Similar sorbent materials from different sources of supply are considered to be different materials and must be tested individually. You must conduct the analytical matrix interference test for each sorbent material prior to the analysis of field samples.
Digest and prepare for analysis a representative mass of sorbent material (unsampled) according to your intended laboratory techniques for field samples. Analyze the digestate according to your intended analytical conditions at the least diluted level you intend to use for sample analysis (e.g., undiluted, 1 in 10 dilution, etc.). Determine the Hg concentration of the undiluted digestate solution. Prepare a series of solutions with a fixed final volume containing graduated aliquots of the sample digestate and, a fixed aliquot of a calibration standard (with the balance being Hg-free reagent or H2O) to establish solutions of varied digestate dilution ratio (e.g., 1:2, 1:5, 1:10, 1:100, etc.---see example in Section 18.104.22.168, below). One of these solutions should contain only the aliquot of the calibration standard in Hg-free reagent or H2O. This will result in a series of solutions where the amount of Hg is held relatively constant and only the volume of digestate diluted is varied. Analyze each of these solutions following intended sample analytical procedures and conditions, determining the concentration for each solution.
Compare the measured concentration of each solution containing digestate to the measured concentration of the digestate-free solution. The lowest dilution ratio of any solution having a Hg concentration within ±5 percent of the digestate-free solution is the minimum dilution ratio required for analysis of all samples. If you desire to measure the digestate without dilution, the + 5 percent criterion must be met at a dilution ratio of at least 9:10 (i.e., ≥ 90% digestate).
An example analytical matrix interference test is presented below. Additional information on the conduct of the analytical matrix interference test will be posted at http://www.epa.gov/ttn/emc. Determine the most sensitive working range for the analyzer to be used. This will be a narrow range of concentrations. Digest and prepare for analysis a representative mass of sorbent material (unsampled) according to your intended laboratory techniques for sample preparation and analysis. Prepare a calibration curve for the most sensitive analytical region, e.g., 0.0, 0.5, 1.0, 3.0, 5.0, 10 ppb. Using the highest calibration standard, e.g.., 10.0 ppb, prepare a series of solutions by adding successively smaller increments of the digestate to a fixed volume of the calibration standard and bringing each solution to a final fixed volume with mercury-free deionized water (diH2O). To 2.0 ml of the calibration standard add 18.0, 10.0, 4.0, 2.0, 1.0, 0.2, and 0.0 ml of the digestate. Bring the final volume of each solution to a total volume of 20 ml by adding 0.0, 8.0, 14.0, 16.0, 17.0, 17.8, and 18.0 ml of diH2O. This will yield solutions with dilution ratios of 9:10, 1:2, 1:5, 1:10, 1:20, 1:100, and 0:10, respectively. Determine the Hg concentration of each solution. The dilution ratio of any solution having a concentration that is within ±5 percent of the concentration of the solution containing 0.0 ml of digestate is an acceptable dilution ratio for analyzing field samples. If more than one solution meets this criterion, the one with the lowest dilution ratio is the minimum dilution required for analysis of field samples. If the 9:10 dilution meets this criterion, then no sample dilution is required.
The minimum mass of Hg that must be collected per sample must be determined. This information is necessary in order to effectively perform the Hg0 and HgCl2 Analytical Bias Test, to estimate target sample volumes/sample times for test runs, and to ensure the quality of the measurements. The determination of minimum sample mass is a direct function of analytical technique, measurement sensitivity, dilutions, etc. This determination is required for all analytical techniques. Based on the analytical approach you employ, you should determine the most sensitive calibration range. Based on a calibration point within that range, you must consider all sample treatments (e.g., dilutions) to determine the mass of sample that needs to be collected to ensure that all sample analyses fall within your calibration curve.
Based on your instrument’s sensitivity and linearity, determine the calibration concentrations or masses that make up a representative low level calibration range. Verify that you are able to meet the multipoint calibration performance criteria in section 11.0 of this method. Select a calibration concentration or mass that is no less than 2 times the lowest concentration or mass in your calibration curve. The lowest point in your calibration curve must be at least 5, and preferably10, times the Method Detection Limit (MDL), which is the minimum amount of the analyte that can be detected and reported. The MDL must be determined at least once for the analytical system using an MDL study such as that found in section 17.0 of the proposed amendments to EPA Method 301 (69 FR 76642, 12/22/2004).
Note to Section 22.214.171.124: While it might be desirable to base the minimum calibration concentration or mass on the lowest point in the calibration curve, selecting a higher concentration or mass is necessary to ensure that all analyses of the field samples will fall within the calibration curve. Therefore, it is strongly recommended that you select a minimum calibration concentration or mass that is sufficiently above the lowest point of the calibration curve (see examples in sections 126.96.36.199.1 and 188.8.131.52.2 below).
Based on your minimum calibration concentration or mass and other sample treatments including, but not limited to, final digestate volume and minimum sample dilution, determine the minimum sample mass. Consideration should also be given to the Hg levels expected to be measured in Section 2 of the sorbent traps and to the breakthrough criteria presented in Table 9-1.
184.108.40.206.1 Example Determination of Minimum Sample Mass for Thermal Desorption Analysis. A thermal analysis system has been calibrated at five Hg mass levels: 10 ng, 20 ng, 50 ng, 100 ng, 200 ng, and shown to meet the calibration performance criteria in this method. Based on 2 times the lowest point in the calibration curve, 20 ng is selected as the minimum calibration mass. Because the entire sample is analyzed and there are no dilutions involved, the minimum sample mass is also 20 ng. Note: In this example, if the typical background (blank) Hg levels in section 2 were relatively high (e.g., 3 to 5 ng), a sample mass of 20 ng might not have been sufficient to ensure that the breakthrough criteria in Table 9-1 would be met, thereby necessitating the use of a higher point on the calibration curve (e.g., 50 ng) as the minimum calibration and sample mass.
220.127.116.11.2 Example Determination of Minimum Sample Mass for Acid Leachate/Digestate Analysis. A cold vapor analysis system has been calibrated at four Hg concentration levels: 2 ng/L, 5 ng, 10 ng/L, 20 ng/L, and shown to meet the calibration performance criteria in this method. Based on 2 times the lowest point in the calibration curve, 4 ng/L was selected as the minimum calibration concentration. The final sample volume of a digestate is nominally 50 ml (0.05 L) and the minimum dilution necessary was determined to be 1:100 by the Analytical Matrix Interference Test of Section 8.2.1. The following calculation would be used to determine the minimum sample mass.
Minimum sample mass = (4 ng/L) X (0.05 L) X (100) = 20 ng Note: In this example, if the typical background (blank) Hg levels in section 2 were relatively high (e.g., 3 to 5 ng), a sample mass of 20 ng might not have been sufficient to ensure that the breakthrough criterion in Table 9-1 would be met, thereby necessitating the use of a higher point on the calibration curve (e.g., 10 ng/L) as the minimum calibration concentration.
Before analyzing any field samples, the laboratory must demonstrate the ability to recover and accurately quantify Hg0 and HgCl2 from the chosen sorbent media by performing the following analytical bias test for sorbent traps spiked with Hg0 and HgCl2. The analytical bias test is performed at a minimum of two distinct sorbent trap Hg loadings that will: (1) represent the lower and upper bound of sample Hg loadings for application of the analytical technique to the field samples, and (2) be used for data validation.
Determine the lower and upper bound mass loadings. The minimum sample mass established in Section 18.104.22.168 can be used for the lower bound Hg mass loading although lower Hg loading levels are acceptable. The upper bound Hg loading level should be an estimate of the greatest mass loading that may result as a function of stack concentration and volume sampled. As previously noted, this test defines the bounds that actual field samples must be within in order to be valid.
22.214.171.124.1 Hg0 Analytical Bias Test. Analyze the front section of three sorbent traps containing Hg0 at the lower bound mass loading level and the front section of three sorbent traps containing Hg0 at the upper bound mass loading level. In other words, analyze each mass loading level in triplicate. You may refer to Section 6.2 for spiking guidance. Prepare and analyze each spiked trap, using the same techniques that will be used to prepare and analyze the field samples. The average recovery for the three traps at each mass loading level must be between 90 and 110 percent. If multiple types of sorbent media are to be analyzed, a separate analytical bias test is required for each sorbent material.
126.96.36.199.2 HgCl2 Analytical Bias Test. Analyze the front section of three sorbent traps containing HgCl2 at the lower bound mass loading level and the front section of three traps containing Hg Cl2 at the upper bound mass loading level. HgCl2 can be spiked as a gas, or as a liquid solution containing HgCl2. However the liquid volume spiked must be <100 μL. Prepare and analyze each spiked trap, using the techniques that will be used to prepare and analyze the field samples. The average recovery for three traps at each spike concentration must be between 90 and 110 percent. Again, if multiple types of sorbent media are to be analyzed, a separate analytical bias test is required for each sorbent material.
The target sample volume is an estimate of the sample volume needed to ensure that valid emissions data are collected (i.e., that sample mass Hg loadings fall within the analytical calibration curve and are within the upper and lower bounds set by the analytical bias tests). The target sample volume and minimum sample mass can also be determined by performing a diagnostic test run prior to initiation of formal testing. Example: If the minimum sample mass is 50 ng and the concentration of mercury in the stack gas is estimated to be 2 μg/m3 (ng/L) then the following calculation would be used to determine the target sample volume:
Target Sample Volume = (50 ng)/(2 ng/L) = 25 L
Note: For the purposes of relative accuracy testing of Hg monitoring systems under part 75 of this chapter and Performance Specification 12A in appendix B to this part, when the stack gas Hg concentration is expected to be very low (<0.5 μg/dscm) you may estimate the Hg concentration at 0.5 μg/dscm.
Sample run time will be a function of minimum sample mass (see Section 8.2.2), target sample volume and nominal equipment sample flow rate. The minimum sample run time for conducting relative accuracy test audits of Hg monitoring systems is 30 minutes and for emissions testing to characterize an emission source is 1 hour. The target sample run time can be calculated using the following example. Example: If the target sample volume has been determined to be 25 L, then the following formula would be used to determine the sampling time necessary to acquire 25 L of gas when sampling at a rate of 0.4 L/min.
Sampling time (min) = 25 L / 0.4 L/min = 63 minutes
The field recovery test provides a test program- specific verification of the performance of the combined sampling and analytical approach. Three sets of paired samples, one of each pair which is spiked with a known level of Hg, are collected and analyzed and the average recovery of the spiked samples is used to verify performance of the measurement system under field conditions during that test program. The conduct of this test requires an estimate or confirmation of the stack Hg concentrations at the time of testing.
Determine the sorbent trap spiking level for the field recovery test using estimates of the stack Hg concentration, the target sample flow rate, and the planned sample duration. First, determine the Hg mass
expected to be collected in section 1 of the sorbent trap. The pre-sampling spike must be within 50 to150 percent of this expected mass. Example calculation: For an expected stack Hg concentration of 5 ug/m3 (ng/L) a target sample rate of 0.40 liters/min, and a sample duration of 1 hour:
(0.40 L/min)*(60 min)*(5ng/L) = 120 ng
A Hg spike of 60 to 180 ng (50-150% of 120 ng) would be appropriate.
Set up two identical sampling trains. One of the sampling trains shall be designated the spiked train and the other the unspiked train. Spike Hg0 onto the front section of the sorbent trap in the spiked train before sampling. The mass of Hg spiked shall be 50 to 150 percent of the mass expected to be collected with the unspiked train. Sample the stack gas with the two trains simultaneously using the same procedures as for the field samples (see Section 8.3). The total sample volume must be within + 20 percent of the target sample volume for the field sample test runs. Analyze the sorbent traps from the two trains utilizing the same analytical procedures and instrumentation as for the field samples (see Section 11.0). Determine the fraction of spiked Hg recovered (R) using the equations in Section 12.7. Repeat this procedure for a total of three runs. Report the individual R values in the test report; the average of the three R values must be between 85 and 115 percent.
Note to section 188.8.131.52: It is acceptable to perform the field recovery test concurrent with actual test runs (e.g., through the use of a quad Probe). It is also acceptable to use the field recovery test runs as test runs for emissions testing or for the RATA of a Hg monitoring system under part 75 of this chapter and Performance Specification 12A in appendix B to this part, if certain conditions are met. To determine whether a particular field recovery test run may be used as a RATA run,, subtract the mass of the Hg0 spike from the total Hg mass collected in sections 1 and 2 of the spiked trap. The difference represents the mass of Hg in the stack gas sample. Divide this mass by the sample volume to obtain the Hg concentration in the effluent gas stream, as measured with the spiked trap. Compare this concentration to the corresponding Hg concentration measured with the unspiked trap. If the paired trains meet the relative deviation and other applicable data validation criteria in Table 9-1, then the average of the two Hg concentrations may be used as an emissions test run value or as the reference method value for a RATA run.
This section describes the procedures and criteria for collecting the field samples for analysis. As noted in Section 8.2.6, the field recovery test samples are also collected using these procedures.
Perform a leak check of the sampling system with the sorbent traps in place. For each of the paired sampling trains, draw a vacuum in the train, and adjust the vacuum to ~15” Hg; and, using the gas flow meter, determine leak rate. The leak rate for an individual train must not exceed 4 percent of the target sampling rate. Once the leak check passes this criterion, carefully release the vacuum in the sample train, then seal the sorbent trap inlet until the Probe is ready for insertion into the stack or duct.
Determine or measure the flue gas measurement environment characteristics (gas temperature, static pressure, gas velocity, stack moisture, etc.) in order to determine ancillary requirements such as Probe heating requirements (if any), initial sampling rate, moisture management, etc.
184.108.40.206 Remove the plug from the end of each sorbent trap and store each plug in a clean sorbent trap storage container. Remove the stack or duct port cap and insert the Probe(s). Secure the Probe(s) and ensure that no leakage occurs between the duct and environment.
220.127.116.11 Record initial data including the sorbent trap ID, date, and the run start time.
18.104.22.168 Record the initial gas flow meter reading, stack temperature, meter temperatures (if needed), and any other appropriate information, before beginning sampling. Begin sampling and target a sampling flow rate similar to that for the field recovery test. Then, at regular intervals (≤ 5 minutes) during the sampling period, record the date and time, the sample flow rate, the gas meter reading, the stack temperature, the flow meter temperatures (if using a dry gas meter), temperatures of heated equipment such as the vacuum lines and the Probes (if heated), and the sampling system vacuum readings. Adjust the sampling flow rate as necessary to maintain the initial sample flow rate. Ensure that the total volume sampled for each run is within 20 percent of the total volume sampled for the field recovery test.
22.214.171.124 Data Recording. Obtain and record any essential operating data for the facility during the test period, e.g., the barometric pressure must be obtained for correcting sample volume to standard conditions when using a dry gas meter. At the end of the data collection period, record the final gas flow meter reading and the final values of all other essential parameters.
126.96.36.199 Post-Test Leak Check. When sampling is completed, turn off the sample pump, remove the Probe(s) with sorbent traps from the port, and carefully seal the end of each sorbent trap. Perform another leak check of each sampling train with the sorbent trap in place, at the maximum vacuum reached during the sampling period. Record the leakage rates and vacuums. The leakage rate for each train must not exceed 4 percent of the average sampling rate for the data collection period. Following each leak check, carefully release the vacuum in the sample train.
188.8.131.52 Sample Recovery. Recover each sampled sorbent trap by removing it from the Probe and sealing both ends. Wipe any deposited material from the outside of the sorbent trap. Place the sorbent trap into an appropriate sample storage container and store/preserve in appropriate manner (see Section 184.108.40.206).
220.127.116.11 Stack Gas Moisture Determination. If the moisture basis of the measurements made with this method (dry) is different from the moisture basis of either: (1) the applicable emission limit; or (2) a Hg CEMS being evaluated for relative accuracy, you must determine the moisture content of the flue gas and correct for moisture using Method 4 in appendix A-3 to this part. If correction of the measured Hg concentrations for moisture is required, at least one Method 4 moisture determination shall be made during each test run.
18.104.22.168 Sample Handling, Preservation, Storage, and Transport. While the performance criteria of this approach provide for verification of appropriate sample handling, it is still important that the user consider, determine, and plan for suitable sample preservation, storage, transport, and holding times for these measurements. Therefore, procedures in ASTM WK223 “Guide for Packaging and Shipping Environmental Samples for Laboratory Analysis” shall be followed for all samples, where appropriate. To avoid Hg contamination of the samples, special attention should be paid to cleanliness during transport, field handling, sampling, recovery, and laboratory analysis, as well as during preparation of the sorbent cartridges. Collection and analysis of blank samples (e.g., reagent, sorbent, field, etc.,) is useful in verifying the absence or source of contaminant Hg.
22.214.171.124 Sample Custody. Proper procedures and documentation for sample chain of custody are critical to ensuring data integrity. The chain of custody procedures in ASTM D4840-99 “Standard Guide for Sampling Chain-of-Custody Procedures” shall be followed for all samples (including field samples and blanks).
Table 9-1 summarizes the QA/QC performance criteria that are used to validate the Hg emissions data from Method 30B sorbent trap measurement systems.
Table 9-1. Quality Assurance/Quality Control Criteria for Method 30B
|QA/QC Test or Specification||Acceptance Criteria||Frequency||Consequences if Not Met|
|Gas flow meter calibration (At 3 settings or points)||calibration factor (Yi ) at each flow rate must be within ± 2% of the average value (Y)||Prior to initial use and when post-test check is not within ± 5% of Y||Recalibrate at 3 points until the acceptance criteria are met|
|Gas flow meter post-test calibration check (Single-point)||calibration factor (Yi ) must be within ± 5% of the Y value from the most recent 3-point calibration||After each field test. For mass flow meters, must be done on- site, using stack gas||Recalibrate gas flow meter at 3 points to determine a new value of Y. For mass flow meters, must be done on-site, using stack gas. Apply the new Y value to the field test data|
|temperature sensor calibration||Absolute temperature measured by sensor within ± 1.5% of a reference sensor||Prior to initial use and before each test thereafter||Recalibrate; sensor may not be used until specification is met|
|barometer calibration||Absolute pressure measured by instrument within ±10mmHg of reading with a mercury barometer||Prior to initial use and before each test thereafter||Recalibrate; instrument may not be used until specification is met|
|Pre-test leak check||< 4% of target sampling rate||Prior to sampling||Sampling shall not commence until the leak check is passed|
|Post-test leak check||< 4% of average sampling rate||After sampling||Sample invalidated*|
|Analytical matrix interference test (wet chemical analysis, only)||Establish minimum dilution (if any) needed to eliminate sorbent matrix interferences||Prior to analyzing any field samples; repeat for each type of sorbent used||Field sample results not validated|
|Analytical bias test||Average recovery between 90% and 110% for Hg0 and HgCl2 at each of the 2 spike concentration levels||Prior to analyzing field samples and prior to use of new sorbent media||Field samples shall not be analyzed until the percent recovery criteria has been met|
|Multipoint analyzer calibration||Each analyzer reading within ± 10% of true value and r2 ≥ 0.99||On the day of analysis, before analyzing any samples||Recalibrate until successful|
|Analysis of independent calibration standard||Within ± 10% of true value||Following daily calibration, prior to analyzing field samples||Recalibrate and repeat independent standard analysis until successful|
|Analysis of continuing calibration verification standard (CCVS)||Within ± 10% of true value||Following daily calibration, after analyzing ≤ 10 field samples, and at end of each set of analyses||Recalibrate and repeat independent standard analysis, reanalyze samples until successful, if possible; for destructive techniques, samples invalidated|
|Test run total sample volume||Within ± 20% of total volume sampled during field recovery test||Each individual sample||Sample invalidated|
|sorbent trap section 2 breakthrough||< 10% of section 1 Hg mass for Hg concentrations > 1 μg/dscm;
≤ 20% of section 1 Hg mass for Hg concentrations ≤ 1 μg/dscm
|Every sample||Sample invalidated*|
|Paired sorbent trap agreement||≤ 10% Relative Deviation (RD) mass for Hg concentrations > 1 μg/dscm;
≤ 20% RD or < 0.2 μg/dscm absolute difference for Hg concentrations ≤ 1 μg/dscm
|Every run||Run invalidated*|
|Sample analysis||Within valid calibration range (within calibration curve)||All Section 1 samples where stack Hg concentration is ≥ 0.5 μg/dscm||Reanalyze at more concentrated level if possible, samples invalidated if not within calibrated range|
|Sample analysis||Within bounds of Hg0 and HgCl2 Analytical Bias Test||All Section 1 samples where stack Hg concentration is ≥ 0.5 μg/dscm||Expand bounds of Hg0 and HgCl2 Analytical Bias Test; if not successful, samples invalidated|
|Field recovery test||Average recovery between 85% and 115% for Hg0||Once per field test||Field sample runs not validated without successful field recovery test|
* And data from the pair of sorbent traps are also invalidated. 26
10.1 Only NIST-certified and NIST-traceable calibration standards (i.e., calibration gases, solutions, etc.) shall be used for the spiking and analytical procedures in this method.
The manufacturer or equipment supplier of the gas flow meter should perform all necessary set-up, testing, programming, etc., and should provide the end user with any necessary instructions, to ensure that the meter will give an accurate readout of dry gas volume in standard cubic meters for this method.
Prior to its initial use, a calibration of the gas flow meter shall be performed. The initial calibration may be done by the manufacturer, by the equipment supplier, or by the end user. If the flow meter is volumetric in nature (e.g., a dry gas meter), the manufacturer or end user may perform a direct volumetric calibration using any gas. For a mass flow meter, the manufacturer, equipment supplier, or end user may calibrate the meter using either: (1) a bottled gas mixture containing 12 ±0.5% CO2, 7 ±0.5% O2, and balance N2 (when this method is applied to coal-fired boilers); (2) a bottled gas mixture containing CO2, O2, and N2 in proportions representative of the expected stack gas composition; or (3) the actual stack gas.
Determine an average calibration factor (Y) for the gas flow meter by calibrating it at three sample flow rate settings covering the range of sample flow rates at which the sampling system will be operated. You may either follow the procedures in section 10.3.1 of Method 5 in appendix A-3 to this part or in section 16 of Method 5 in appendix A-3 to this part. If a dry gas meter is being calibrated, use at least five revolutions of the meter at each flow rate.
Alternatively, you may perform the initial calibration of the gas flow meter using a reference gas flow meter (RGFM). The RGFM may be: (1) a wet test meter calibrated according to section 10.3.1 of Method 5 in appendix A-3 to this part; (2) a gas flow metering device calibrated at multiple flow rates using the procedures in section 16 of Method 5 in appendix A–3 to this part; or (3) a NIST-traceable calibration device capable of measuring volumetric flow to an accuracy of 1 percent. To calibrate the gas flow meter using the RGFM, proceed as follows: While the Method 30B sampling system is sampling the actual stack gas or a compressed gas mixture that simulates the stack gas composition (as applicable), connect the RGFM to the discharge of the system. Care should be taken to minimize the dead volume between the gas flow meter being tested and the RGFM. Concurrently measure dry stack gas volume with the RGFM and the flow meter being calibrated for at least 10 minutes at each of three flow rates covering the typical range of operation of the sampling system. For each set of concurrent measurements, record the total sample volume, in units of dry standard cubic meters (dscm), measured by the RGFM and the gas flow meter being tested.
Calculate an individual calibration factor Yi at each tested flow rate from section 10.2.2.1 or 10.2.2.2 of this method (as applicable) by taking the ratio of the reference sample volume to the sample volume recorded by the gas flow meter. Average the three Yi values, to determine Y, the calibration factor for the flow meter. Each of the three individual values of Yi must be within ± 0.02 of Y. Except as otherwise provided in sections 10.2.2.4 and 10.2.2.5 of this method, use the average Y value from the initial 3-point calibration to adjust subsequent gas volume measurements made with the gas flow meter.
For a mass flow meter, if the most recent 3-point calibration of the flow meter was performed using a compressed gas mixture, you may want to conduct the following on-site calibration check prior to testing, to ensure that the flow meter will accurately measure the volume of the stack gas: While sampling stack gas, check the calibration of the flow meter at one intermediate flow rate setting representative of normal operation of the sampling system. If the pretest calibration check shows that the value of Yi, the calibration factor at the tested flow rate, differs from the current value of Y by more than 5 percent, perform a full 3-point recalibration of the meter using stack gas to determine a new value of Y, and (except as otherwise provided in section 10.2.2.5 of this method) apply the new Y value to the data recorded during the field test.
Check the calibration of the gas flow meter following each field test at one intermediate flow rate setting, either at, or in close proximity to, the average sample flow rate during the field test. For dry gas meters, ensure at least three revolutions of the meter during the calibration check. For mass flow meters, this check must be performed before leaving the test site, while sampling stack gas. If a one-point calibration check shows that the value of Yi at the tested flow rate differs by more than 5 percent from the current value of Y, repeat the full 3-point calibration procedure to determine a new value of Y, and apply the new Y value to the gas volume measurements made with the gas flow meter during the field test that was just completed. For mass flow meters, perform the 3-point recalibration while sampling stack gas.
Use the procedures and criteria in Section 10.3 of Method 2 in Appendix A-1 to this part to calibrate in-stack temperature sensors and thermocouples. Dial thermometers shall be calibrated against mercury-in-glass thermometers. calibrations must be performed prior to initial use and before each field test thereafter. At each calibration point, the absolute temperature measured by the temperature sensor must agree to within + 1.5 percent of the temperature measured with the reference sensor, otherwise the sensor may not continue to be used.
Calibrate against a mercury barometer as per Section 10.6 of Method 5 in appendix A-3 to this part. calibration must be performed prior to initial use and before each test program, and the absolute pressure measured by the barometer must agree to within + 10 mm Hg of the pressure measured by the mercury barometer, otherwise the barometer may not continue to be used.
Calibrate all other sensors and gauges according to the procedures specified by the instrument manufacturer(s).
See Section 11.1 of this method.
The analysis of Hg in the field and quality control samples may be conducted using any instrument or technology capable of quantifying total Hg from the sorbent media and meeting the performance criteria in this method. Because multiple analytical approaches, equipment and techniques are appropriate for the analysis of sorbent traps, it is not possible to provide detailed, technique-specific analytical procedures. As they become available, detailed procedures for a variety of candidate analytical approaches will be posted at http://www.epa.gov/ttn/emc.
Perform a multipoint calibration of the analyzer at three or more upscale points over the desired quantitative range (multiple calibration ranges shall be calibrated, if necessary). The field samples analyzed must fall within a calibrated, quantitative range and meet the performance criteria specified below. For samples suitable for aliquotting, a series of dilutions may be needed to ensure that the samples fall within a calibrated range. However, for sorbent media samples consumed during analysis (e.g., when using thermal desorption techniques), extra care must be taken to ensure that the analytical system is appropriately calibrated prior to sample analysis. The calibration curve range(s) should be determined such that the levels of Hg mass expected to be collected and measured will fall within the calibrated range. The calibration curve may be generated by directly introducing standard solutions into the analyzer or by spiking the standards onto the sorbent media and then introducing into the analyzer after preparing the sorbent/standard according to the particular analytical technique. For each calibration curve, the value of the square of the linear correlation coefficient, i.e., r2, must be > 0.99, and the analyzer response must be within + 10 percent of the reference value at each upscale calibration point. calibrations must be performed on the day of the analysis, before analyzing any of the samples. Following calibration, an independent standard shall be analyzed. The measured value of the independently prepared standard must be within + 10 percent of the expected value.
Carefully separate the sections of each sorbent trap. Combine for analysis all materials associated with each section; any supporting substrate that the sample gas passes through prior to entering a media section (e.g., glass wool separators, acid gas traps, etc.) must be analyzed with that segment.
Analyze the sorbent trap samples following the same procedures that were used for conducting the Hg0 and HgCl2 analytical bias tests. The individual sections of the sorbent trap and their respective components must be analyzed separately (i.e., section 1 and its components, then section 2 and its components). All sorbent trap section 1 sample analyses must be within the calibrated range of the analytical system. For wet analyses, the sample can simply be diluted to fall within the calibrated range. However, for the destructive thermal analyses, samples that are not within the calibrated range cannot be re-analyzed. As a result, the sample cannot be validated, and another sample must be collected. It is strongly suggested that the analytical system be calibrated over multiple ranges so that thermally analyzed samples do fall within the calibrated range. The total mass of Hg measured in each sorbent trap section 1 must also fall within the lower and upper mass limits established during the initial Hg0 and HgCl2 analytical bias test. If a sample is analyzed and found to fall outside of these limits, it is acceptable for an additional Hg0 and HgCl2 analytical bias test to be performed that now includes this level. However, some samples (e.g., the mass collected in trap section 2 or the mass collected in trap section 1 when the stack gas concentration is <0.5 μg/m3), may have Hg levels so low that it may not be possible to quantify them in the analytical system’s calibrated range. Because a reliable estimate of these low-level Hg measurements is necessary to fully validate the emissions data, the MDL (see section 126.96.36.199 of this method) is used to establish the minimum amount that can be detected and reported. If the measured mass or concentration is below the lowest point in the calibration curve and above the MDL, the analyst must do the following: estimate the mass or concentration of the sample based on the analytical instrument response relative to an additional calibration standard at a concentration or mass between the MDL and the lowest point in the calibration curve. This is accomplished by establishing a response factor (e.g., area counts per Hg mass or concentration) and estimating the amount of Hg present in the sample based on the analytical response and this response factor.
Example: The analysis of a particular sample results in a measured mass above the MDL, but below the lowest point in the calibration curve which is 10 ng. An MDL of 1.3 ng Hg has been established by the MDL study. A calibration standard containing 5 ng of Hg is analyzed and gives an analytical response of 6,170 area counts, which equates to a response factor of 1,234 area counts/ng Hg. The analytical response for the sample is 4,840 area counts. Dividing the analytical response for the sample (4,840 area counts) by the response factor gives 3.9 ng Hg, which is the estimated mass of Hg in the sample.
After no more than 10 samples and at the end of each set of analyses, a continuing calibration verification standard must be analyzed. The measured value of the continuing calibration standard must be within + 10 percent of the expected value.
The analysis of blanks is optional. The analysis of blanks is useful to verify the absence of, or an acceptable level of, Hg contamination. Blank levels should be considered when quantifying low Hg levels and their potential contribution to meeting the sorbent trap section 2 breakthrough requirements; however, correcting sorbent trap results for blank levels is prohibited.
You must follow the procedures for calculation and data analysis listed in this section.
The terms used in the equations are defined as follows:
|Bws||=||Moisture content of sample gas as measured by Method 4, percent/100.|
|Ca||=||Concentration of Hg for the sample collection period, for sorbent trap “a” (μg/dscm).|
|Cb||=||Concentration of Hg for the sample collection period, for sorbent trap “b” (μg/dscm).|
|Cd||=||Hg concentration, dry basis (μg/dscm).|
|Crec||=||Concentration of spiked compound measured (μg/m3).|
|Cw||=||Hg concentration, wet basis (μg/m3).|
|m1||=||Mass of Hg measured on sorbent trap section 1 (μg).|
|m2||=||Mass of Hg measured on sorbent trap section 2 (μg).|
|mrecovered||=||Mass of spiked Hg recovered in Analytical Bias or Field Recovery Test (μg).|
|ms||=||Total mass of Hg measured on spiked trap in Field Recovery Test (μg).|
|mspiked||=||Mass of Hg spiked in Analytical Bias or Field Recovery Test (μg).|
|mu||=||Total mass of Hg measured on unspiked trap in Field Recovery Test (μg).|
|R||=||Percentage of spiked mass recovered (%).|
|RD||=||Relative deviation between the Hg concentrations from traps “a” and “b” (%).|
|vs||=||Volume of gas sampled, spiked trap in Field Recovery Test (dscm).|
|Vt||=||Total volume of dry gas metered during the collection period (dscm); for the purposes of this method, standard temperature and pressure are defined as 20 C and 760 mm Hg, respectively.|
|vu||=||Volume of gas sampled, unspiked trap in Field Recovery Test (dscm).|
Calculate the percent recovery of Hg0 and HgCl2 using Equation 30B-1.
Use Equation 30B-2 to calculate the percent breakthrough to the second section of the sorbent trap.
Calculate the Hg concentration measured with sorbent trap “a”, using Equation 30B-3.
For sorbent trap “b”, replace “Ca “ with “Cb “ in Equation 30B-3. Report the average concentration, i.e., 1⁄2 (Ca + Cb).
Use Equation 30B-4 if your measurements need to be corrected to a wet basis.
12.6 Calculation of Paired Trap Agreement.
Calculate the relative deviation (RD) between the Hg concentrations measured with the paired sorbent traps using Equation 30B-5.
Calculate the measured spike concentration using Equation 30B-6.
Then calculate the spiked Hg recovery, R, using Equation 30B-7.
How do I validate my data? Measurement data are validated using initial, one- time laboratory tests coupled with test program-specific tests and procedures. The analytical matrix interference test and the Hg0 and HgCl2 analytical bias test described in Section 8.2 are used to verify the appropriateness of the selected analytical approach(es) as well as define the valid working ranges for sample analysis. The field recovery test serves to verify the performance of the combined sampling and analysis as applied for each test program. Field test samples are validated by meeting the above requirements as well as meeting specific sampling requirements (i.e., leak checks, paired train agreement, total sample volume agreement with field recovery test samples) and analytical requirements (i.e., valid calibration curve, continuing calibration performance, sample results within calibration curve and bounds of Hg0 and HgCl2 analytical bias test). Complete data validation requirements are summarized in Table 9-1.
1. EPA Traceability Protocol for Qualification and Certification of Elemental Mercury Gas Generators, expected publication date December 2008, see www.epa.gov/ttn/emc.
2. EPA Traceability Protocol for Qualification and Certification of Oxidized Mercury Gas Generators, expected publication date December 2008, see www.epa.gov/ttn/emc.
3. EPA Traceability Protocol for Assay and Certification of Gaseous calibration Standards, expected revision publication date December 2008, see www.epa.gov/ttn/emc.
Figure 30B-1. Typical sorbent Trap Sampling System