Sensitivity of EPA 325 to Field Handling Protocols
Heidi C. Hayes, Eurofins Air Toxics, 180 Blue Ravine Rd. Suite B, Folsom, CA 95630
J. Derek Reese, ExxonMobil Corporation, 22777 Springwoods Village Pkwy, Spring, TX 77389
As refineries prepare to implement the fenceline monitoring work practice required in the updated Refinery Risk and Technology Rule1, sample handling procedures are being developed and implemented for pre-compliance monitoring trials. With the goal of generating accurate and reliable measurements, stringent protocols detailing tube deployment, retrieval, and storage requirements are advisable. Highlighting the importance of sample handling, the quality control section of EPA 325A cites accidental contamination by the field sampler as a possible reason to eliminate an outlier in the data set2.
To quantify the sensitivity of EPA 325 to benzene contamination during tube handling, a series of tests was performed to quantify the impact of several likely field scenarios. The evaluated scenarios included field technicians engaging in cigarette smoking or fueling vehicle prior to sampling activities, idling work vehicle near sampling stations during tube deployment, application of insecticide to shelter prior to tube deployment, and exposure of sampled tubes to excessive temperatures during transit to the laboratory. Each experiment utilized both test samples and control samples to accurately determine the contribution of the field scenario.
The sources of benzene unrelated to fugitive emissions from refinery operations are both widespread and common. Benzene can be emitted from products and materials such as inks, paints, and adhesives and can be found in sources such as cigarette smoke, motor vehicle exhaust, and industrial solvents. Given the number of benzene emission sources, contaminating sample tubes during handling and storage becomes a concern in generating accurate and reliable fenceline monitoring benzene concentrations.
While the objective is to eliminate sources of benzene related to field activities, inadvertent lapses in protocols are likely to occur at some point during a monitoring program. Although EPA 325A describes the possibility of contamination during tube handling, it is unknown whether an outlier or exceedance can be partially or wholly attributed to sample handling missteps. Given that the Refinery Sector Rule requires root cause analysis and initial corrective action within 45 days after determining an exceedance of the action level, understanding the possible sources of benzene contamination during sample handling and their possible significance on reported benzene fenceline concentrations can facilitate the investigative process. This study focuses on several likely field handling scenarios with the potential to influence measured benzene concentrations.
Experiments were conducted to simulate sample handling under a specific field scenario, and each experiment utilized both test samples and control samples to quantify the impact of the field activity on the benzene concentration. As appropriate, field blanks were collected with each experiment. The field handling scenarios evaluated are summarized in Table 1.
Table 1. Scenarios tested to determine impacts to EPA 325 benzene concentrations
|Sample deployment and/or retrieval||Cigarette smoking||Field sampler smoking just prior to sample deployment and/or retrieval|
|Sample deployment and/or retrieval||Vehicle exhaust||Sample deployment and/or retrieval while vehicle is idling nearby|
|Sample deployment and/or retrieval||Fueling vehicle||Field sampler filling vehicle with gasoline just prior to sample deployment and/or
|Sample storage||Storing sample tubes at elevated temperatures||Sample tubes exposed to hot conditions for multiple days prior to analysis|
|Shelter preparation||Insecticide application||Shelter sprayed for insects immediately prior to sample deployment|
Sample Deployment/Retrieval Scenarios
To simulate the sample deployment and/or retrieval process, the long-term storage cap was removed from the inlet of a Carbopack™ X tube under the experimental conditions. The diffusion screen cap was then installed, and the sample tube was exposed to the conditions for a period of 3 minutes. After 3 minutes, the diffusion screen cap was removed and the long-term storage cap was replaced. Control samples were collected following the same procedure without exposure to the test conditions. Three minutes was selected as representative of the time it takes for a field technician to deploy or retrieve a tube in the field.
To determine the impact of smoking by field personnel on benzene concentrations, sample deployment was simulated in an area where a technician recently finished his cigarette break.
Additionally, deployment simulation was performed by the same technician away from the smoking area but immediately post-cigarette with residual smoke odors on his clothes. These deployment simulations were compared with control samples collected by a non-smoker away from the smoking area in a designated office area. Test samples and control samples were each collected in triplicate. A field blank was collected in the office area while the experiment was conducted.
Samples were analyzed by EPA 325B3 for benzene and results in units of nanograms are summarized in Table 2. All collected samples were less than half the laboratory reporting limit of 5 ng or below the method detection limit of 1.43 ng. The smoking area and post-cigarette samples reflected only a negligible increase in benzene, with detected concentrations near or below the field blank concentration.
Table 2. Benzene concentration in units of nanograms
|Sample #||Smoking Area||Post-cigarette||Control||Field Blank|
The effects of vehicle exhaust during sample deployment on benzene concentrations were determined by simulating deployment near the tail pipe of an idling work truck. The vehicle used in the test was a 2008 Chevy gasoline engine truck compliant with California emission standards. Control samples were initially collected within 3 feet of the vehicle’s tail pipe.
Immediately after the control samples were collected, the truck was started and five samples were collected within 3 feet of the tail pipe while the truck was idling. Two field blanks were collected during the test.
The benzene mass concentrations measured by EPA 325B are summarized in Table 3. All samples were less than the laboratory reporting limit of 5 ng. Comparison of the average benzene mass for the vehicle exhaust samples and the control and field blank samples showed a negligible increase in benzene for the samples deployed while the sampler’s truck was idling.
Table 3. Benzene concentration in units of nanograms
|Sample #||Vehicle Exhaust||Control||Field Blank|
The impact of a field technician fueling a vehicle immediately prior to handling sorbent tubes was evaluated by simulating deployment immediately after the technician pumped unleaded gasoline into a passenger vehicle. The field technician wore nitrile gloves while pumping gas and while directly handling the dispensing nozzle in contact with the fuel. The samples were collected inside the vehicle while wearing the same gloves used to pump the gas. This deployment simulation was compared with control samples collected inside the vehicle at the gas station prior to fueling the vehicle. A field blank was collected and remained in the vehicle during the test.
Samples were analyzed by EPA 325B for benzene, and results in units of nanograms are summarized in Table 4. Comparison of the average benzene mass for the post fueling samples and the control and field blank samples reflected a negligible increase in benzene for the samples deployed by the field sampler immediately after fueling a vehicle. However, the field blank collected was near the highest concentration measured in the post fueling sample set.
Table 4. Benzene concentration in units of nanograms
|Sample #||Post Fueling||Control||Field Blank|
Sample Storage Scenario – Elevated temperatures
To evaluate the impact on benzene concentrations when sample storage conditions exceed the recommended ambient temperature of 23°C, sample tubes were stored at 50 °C for a period of approximately 4 days. Concentrations were compared to samples stored at room temperature. Conditions of 50 °C were selected since these elevated temperatures can be quickly reached if samples are left inside a vehicle after sample collection.
To generate a representative matrix on the tubes, ambient air was sampled over a 14-day period near the parking lot of the laboratory facility. Two shelters were deployed side-by-side with about 8 feet horizontal distance separation. Five tubes were deployed in each of the shelters for a total of 10 samples.
After samples were retrieved, one sample was analyzed from each of the shelters to evaluate benzene concentration and precision between the shelters. (These two samples are listed as “Control” in Table 5.) With the sample tubes properly capped and housed in their storage vials as described in EPA 325, two samples from each shelter were placed in an oven at 50 °C for a period of approximately 4 days (91 hours). The remaining 2 samples from each shelter were stored capped in their storage vials at room temperature outside of the oven.
Samples were analyzed by EPA 325B for benzene, and results in units of µg/m3 are summarized in Table 5. Under these extreme storage conditions, the sample concentrations showed a slight decrease in benzene concentrations as compared to storage at ambient temperatures. A decrease of approximately 10% was observed in the small data set and the Relative Percent Difference (RPD) between the two average concentrations was less than 10% which is well within method specified lab and field precision.
Table 5. Benzene concentration in units of µg/m3
|Sample #||Shelter ID||50 deg C||Ambient T||Control|
Shelter preparation – Insecticide application
To evaluate the impact of spraying the EPA 325 shelter with insecticide on measured benzene concentrations, an abbreviated 2-hour sampling event was conducted immediately after the shelter was treated with RAID® Wasp & Hornet Killer. The shelter (PVC cap type) was removed from the fence and RAID® Wasp & Hornet Killer was sprayed inside the shelter. The spray dispensed as a liquid, and the entire shelter interior assembly was fully wetted with the product. The shelter was replaced on the fenceline and excess product drained from the assembly. Three sorbent tubes equipped with diffusive caps were clipped into the shelter immediately after shelter installation. The samplers were in place for a period of 2 hours in an effort to capture the initial vapors released from insecticide application. Prior to the spraying event, a set of control samples was collected in the same shelter and deployed for a 2-hour duration. Field blanks were not collected for this test.
The benzene mass concentrations measured by EPA 325B are summarized in Table 6. All collected samples were less than the laboratory reporting limit of 5 ng. Comparison of the average benzene mass for the post spray samples and the control indicated no increase in the benzene mass for the samples deployed over the 2-hour period immediately following insecticide application. As a note, the post spray analytical runs showed the presence of heavier molecular weight alkanes consistent with the compositional information on the product label which listed petroleum distillates as the major constituent. Over the period sampled, the hydrocarbons did not interfere with the identification or quantification of benzene in the samples.
Table 6. Benzene concentrations in units of nanograms
|Sample #||Post Spray||Control|
Sample Deployment/Retrieval Scenarios
The sample deployment and retrieval scenarios evaluated showed negligible effects on the benzene mass collected on the sorbent tube as compared to the control samples. Given the short period of time the sample tubes were exposed to these “non-ideal” handling conditions, incremental benzene mass collected from cigarette smoking or fueling a vehicle prior to sample deployment or idling the work vehicle during deployment was insignificant.
To increase the 14-day benzene fenceline concentration by 0.9 µg/m3 (1/10 of the 14-day action level), an additional 12 nanograms of benzene would need to be introduced to the sorbent tube during these handling activities. Assuming the tube is handled by the field technician for a period of 3 minutes during deployment and/or retrieval, a concentration increase of 0.9 µg/m3 means that the tube would require exposure to a benzene concentration of approximately 6000 µg/m3 or nearly 2 ppmv. Exposing the sample tubes to these high levels of benzene during routine sample handling is unlikely unless material or product containing high benzene was directly transferred to the diffusive cap or sorbent tube inlet. In this case, the material on the sampler could emit benzene vapors over the course of sample collection with the potential of accumulating benzene mass to appreciably increase the 14-day concentration. The field technician can easily avoid this outcome by wearing clean nitrile gloves during tube handling.
Sample Storage Scenario – Elevated temperatures
In addition to the tube handling scenarios, storing samples under non-ideal temperature conditions showed minor effects on the recovery of benzene. In cases where samples exceed the 23 °C storage specification, the impact to the measured concentration is expected to be within method error assuming that the sample tubes are properly sealed with long-term storage caps and stored in method specified non-emitting storage containers.
Shelter preparation – Insecticide application
The application of insecticide to the inside of the shelter prior to tube deployment had no impact on the benzene concentrations under the tested conditions. However, it is important to note that only one product was evaluated, and the product was applied to the shelter without sample tubes deployed. Direct application of products such as insecticides on exposed tubes can result in tube damage and/or unusable data. If material migrates to the tube’s sorbent bed, the sample will likely be unrecoverable due to gross contamination. As a result, spraying products near exposed tubes should be performed with caution and spraying the shelter directly when tubes are deployed should be avoided.
Furthermore, even when applying a product prior to sample deployment, the residual vapors from the product can result in high mass loading of non-target compounds potentially impacting laboratory instrumentation and sorbent tubes. It is also unclear whether the high mass loading of high molecular weight compounds negatively affects the benzene uptake rate due to competition of available adsorption sites on the sorbent tube. Based on these uncertainties, it is recommended that a test sample is collected in the field or by the laboratory prior to widespread application of a product near or in the shelter.
While every effort should be made to follow stringent sample handling and storage procedures to generate reliable data, the scenarios evaluated showed that EPA Method 325 is a robust method even when subjected to these non-ideal conditions. Outliers in the data set attributable to field handling missteps are expected to be minimal if basic protocols are followed. Additionally, careful evaluation and coordination with the laboratory are strongly recommended when considering application of a product near monitoring locations to determine potential impacts to benzene measurements.
- Petroleum Refinery Sector Risk and Technology Review and New Source Performance Standards; Final Rule, 40 CFR Parts 60 and 63; Federal Register, Vol. 80, No. 230; S. Environmental Protection Agency: Washington D.C., December 2015.
- Method 325A-Volatile Organic Compounds from Fugitive and Area Sources, Sampler Deployment and VOC Sample Collection; 40 CFR Appendix A to Part 63; Federal Register, Vol. 80, No. 230; U.S. Environmental Protection Agency: Washington D.C., December 2015.
- Method 325B-Volatile Organic Compounds from Fugitive and Area Sources, Sampler Preparation and Analysis; 40 CFR Appendix A to Part 63; Federal Register, Vol. 80, No. 230; U.S. Environmental Protection Agency: Washington D.C., December 2015.