September 2000 Environmental Technology Verification Report Opsis Inc. AR-500 Ultraviolet Open-Path Monitor Prepared by Baiteiie Putting Technology To Work Battel le Under a cooperative agreement with «>EPA U.S. Environmental Protection Agency ElV ElV ElV ------- September 2000 Environmental Technology Verification Report ETV Advanced Monitoring Systems Center Opsis Inc. AR-500 Ultraviolet Open-Path Monitor By Jeffrey Myers Thomas Kelly Charles Lawrie Karen Riggs Battelle Columbus, Ohio 43201 ------- Notice The U.S. Environmental Protection Agency (EPA), through its Office of Research and Development, has financially supported and collaborated in the extramural program described here. This document has been peer reviewed by the Agency and recommended for public release. Mention of trade names or commercial products does not constitute endorsement or recommendation by the EPA for use. 11 ------- Foreword The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the Nation's air, water, and land resources. Under a mandate of national environmental laws, the Agency strives to formulate and implement actions leading to a compatible balance between human activities and the ability of natural systems to support and nurture life. To meet this mandate, the EPA's Office of Research and Development (ORD) provides data and science support that can be used to solve environmental problems and to build the scientific knowledge base needed to manage our ecological resources wisely, to understand how pollutants affect our health, and to prevent or reduce environmental risks. The Environmental Technology Verification (ETV) Program has been established by the EPA to verify the performance characteristics of innovative environmental technology across all media and to report this objective information to permitters, buyers, and users of the technology, thus substantially accelerating the entrance of new environmental technologies into the marketplace. Verification Organizations oversee and report verification activities based on testing and Quality Assurance protocols developed with input from major stakeholders and customer groups associated with the technology area. At present, there are 12 environmental technology areas covered by ETV. Information about each of the environmental technology areas covered by ETV can be found on the Internet at http://www.epa.gov/etv.htm. Effective verifications of monitoring technologies are needed to assess environmental quality and to supply cost and performance data to select the most appropriate technology for that assess- ment. In 1997, through a competitive cooperative agreement, Battelle was awarded EPA funding and support to plan, coordinate, and conduct such verification tests for "Advanced Monitoring Systems for Air, Water, and Soil" and report the results to the community at large. Information concerning this specific environmental technology area can be found on the Internet at http://www.epa.gov/etv/07/07_main.htm. 111 ------- Acknowledgments The authors wish to acknowledge the support of all those who helped plan and conduct the verification test, analyze the data, and prepare this report. In particular we would like to thank Brian Canterbury and Paul Webb of Battelle. We also acknowledge the participation of Carl Kamme and Paul Stenberg of Opsis in this verification test. iv ------- Contents Notice ii Foreword iii Acknowledgments iv List of Abbreviations ix 1. Background 1 2. Technology Description 2 3. Test Design and Procedures 4 3.1 Introduction 4 3.2 Test Design 6 3.3 Experimental Apparatus and Materials 8 3.3.1 Standard Gases 8 3.3.2 Dilution Gas 8 3.3.3 Gas Dilution System 8 3.3.4 Gas Cell 8 3.3.5 Temperature Sensor 8 3.3.6 Ozone Sensor 8 3.3.7 NO/NH3 Monitor 9 3.3.8 Benzene Measurement 9 3.4 Test Parameters 9 3.4.1 Minimum Detection Limit 9 3.4.2 Linearity 9 3.4.3 Accuracy 10 3.4.4 Precision 10 3.4.5 Interferences 10 4. Quality Assurance/Quality Control 12 4.1 Data Review and Validation 12 4.2 Changes from the Test/QA Plan 12 4.3 Calibration 14 4.3.1 Gas Dilution System 14 4.3.2 Temperature Sensor 14 v ------- 4.3.3 Ozone Sensor 14 4.3.4 NO/NH3 Monitor 14 4.3.5 Benzene Measurement 15 4.4 Data Collection 15 4.5 Assessments and Audits 16 4.5.1 Technical Systems Audit 16 4.5.2 Performance Evaluation Audit 16 4.5.3 Data Quality Audit 17 5. Statistical Methods 18 5.1 Minimum Detection Limit 18 5.2 Linearity 18 5.3 Accuracy 18 5.4 Precision 19 5.5 Interferences 19 6. Test Results 20 6.1 Minimum Detection Limit 20 6.2 Linearity 22 6.2.1 Source Strength Linearity 22 6.2.2 Concentration Linearity 23 6.3 Accuracy 26 6.4 Precision 26 6.5 Interferences 28 6.6 Other Factors 29 6.6.1 Costs 29 6.6.2 Data Completeness 29 7. Performance Summary 30 8. References 31 Appendix A: Data Recording Sheet A-l vi ------- Figures Figure 2-1. Opsis AR-500 Optical Open-Path Monitor 2 Figure 3-1. Test Site at West Jefferson Facility 6 Figure 3-2. Optical Open-Path Monitor Setup 7 Figure 6-1. Source Strength Linearity Plot of the AR-500 23 Figure 6-2. Concentration Linearity Plot of the AR-500 Challenged with NO 24 Figure 6-3. Concentration Linearity Plot of the AR-500 Challenged with Benzene 25 Figure 6-4. Concentration Linearity Plot of the AR-500 Challenged with Ammonia 25 Tables Table 3-1. Target Gases and Concentrations for Testing the AR-500 4 Table 3-2. Optical Open-Path Monitor Verification: Measurement Order for Each Target Gas 5 Table 4-1. Summary of Data Recording Process for the AR-500 Verification Test 15 Table 4-2. Summary of Performance Evaluation Audit Procedures 17 Table 6-1. MDL Data for the AR-500 21 Table 6-2. Minimum Detection Limits of the AR-500 22 Table 6-3. Source Strength Linearity of the AR-500 22 Table 6-4. Concentration Linearity Data for the AR-500 24 Table 6-5. Results of Accuracy Tests for the AR-500 26 Table 6-6. Data from Precision Tests on the AR-500 27 Table 6-7. Results of Precision Tests on the AR-500 27 vii ------- Table 6-8. Concentration Data from Interference Tests on the AR-500 28 Table 6-9. MDL Data from Interference Tests on the AR-500 28 viii ------- List of Abbreviations AMS Advanced Monitoring Systems CEM continuous emission monitor cm centimeter CO carbon monoxide co2 carbon dioxide DOAS differential optical absorption spectroscopy EPA U.S. Environmental Protection Agency ETV Environmental Technology Verification GC/FID gas chromatography/flame ionization detection GC/MS gas chromatography/mass spectroscopy HCHO formaldehyde HF hydrogen fluoride Hg° elemental mercury hno2 nitrous acid kg kilogram lb pound m meter MDL minimum detection limit nh3 ammonia NIST National Institute of Standards and Technology NO nitric oxide no2 nitrogen dioxide NOx nitrogen oxides (= NO + N02) o2 oxygen 03 ozone PPb parts per billion ppb*m parts per billion meters ppbv parts per billion by volume ppm parts per million ppm*m parts per million meters QA quality assurance QC quality control QMP Quality Management Plan RH relative humidity RSD relative standard deviation ix ------- sulfur dioxide technical systems audit ------- Chapter 1 Background The U.S. Environmental Protection Agency (EPA) has created the Environmental Technology Verification (ETV) Program to facilitate the deployment of innovative environmental tech- nologies through performance verification and dissemination of information. The goal of the ETV Program is to further environmental protection by substantially accelerating the acceptance and use of improved and cost-effective technologies. ETV seeks to achieve this goal by provid- ing high quality, peer-reviewed data on technology performance to those involved in the design, distribution, permitting, purchase, and use of environmental technologies. ETV works in partnership with recognized testing organizations; with stakeholder groups consisting of regulators, buyers and vendor organizations; and with the full participation of individual technology developers. The program evaluates the performance of innovative tech- nologies by developing test plans that are responsive to the needs of stakeholders, conducting field or laboratory tests (as appropriate), collecting and analyzing data, and preparing peer- reviewed reports. All evaluations are conducted in accordance with rigorous quality assurance protocols to ensure that data of known and adequate quality are generated and that the results are defensible. The EPA's National Exposure Research Laboratory and its verification organization partner, Battelle, operate the Advanced Monitoring Systems (AMS) Center under ETV. The AMS Center recently evaluated the performance of optical open-path monitors for use in ambient air or fence line measurements. This verification report presents the procedures and results of the verification test for the Opsis Inc. (Opsis) AR-500 ultraviolet (UV) optical open-path monitor. 1 ------- Chapter 2 Technology Description The objective of the ETV A VIS Center is to verify the performance characteristics of environ- mental monitoring technologies for air, water, and soil. This verification report provides results for the verification testing of the AR-500. The following description of the AR-500 is based on information provided by the vendor. The AR-500 ultraviolet differential optical absorption spectroscopy (CV DO AS) system uses a broad-band Xenon light-source that projects a narrow beam of light across a monitoring path ranging from 1 to 1,000 meters in length. The receiver telescope focuses the light into a quartz fiber optic cable that connects to the DOAS analyzer. The AR-500 is a compact, tunable, and fast-scanning spectrometer that measures spectra in the wavelength regions of interest. The system can provide path-averaged measurements, from the light source to the receiver, of, e.g., S02, NO, NO-,, NH3, 03, benzene, toluene, p-, m and o-xylene, styrene, HN02, HCHO, Hg°, and hydrogen fluoride (HF). The AR-500 is designated by the U.S. EPA as an Equivalent Method for measuring the criteria pollutants S02, NO,, and 03 in ambient air. The AR-500 evaluated in this verification test was bi-static, with separate emitters and receivers and a light beam that passed through the gas volume once. Figure 2-1. Opsis AR-500 Optical Open-Path Monitor From the AR-500 monitor, the results are transferred to a data collection system for presentation and reporting. The Opsis EnviMan software suite (Windows™ 95, 98, NT, 2000), provides the necessary functions for data analysis, presentations, and reporting. 2 ------- The AR-500 is designed for continuous operation and is used in a variety of applications, including ambient air quality measurements; fence-line measurements at industrial plants and airports; street-level monitoring and tunnel measurements; and industrial continuous emission monitoring (CEM) and process applications at power plants, incinerators, cement plants, and aluminum smelters. The AR-500 uses the Opsis ER-150 emitter/receiver unit for the monitoring path. Two tempera- ture signals are logged through the signal unit: the temperature of the calibration cell and the ambient air temperature. The temperature values are used to normalize data, which are stored in the analyzer and can be extracted directly from the analyzer in ASCII format. Data also are available on a separate computer that connects to the system. The AR-500 measures 60 x 44 x 26.6 cm (23.6 x 17.3 x 10.5 inches). It weighs (including the case) approximately 50 kg (110 lb). 3 ------- Chapter 3 Test Design and Procedures 3.1 Introduction This verification test was conducted according to procedures specified in the Test/QA Plan for Verification of Optical Open-Path Monitors ,(1) The test was designed to challenge the AR-500 in a manner similar to that which would be experienced in field operations and was modeled after Compendium Method TO-16.(2) The monitor was challenged using an optically transparent gas cell filled with known concentrations of a target gas. The gas cell was inserted into the optical path of the monitor during operation under field conditions, simulating a condition where the target gas would be present in the ambient air. The gas cell was used to challenge the monitor in a controlled and uniform manner. The monitor was challenged with three target gases at known concentrations, and the measure- ment result was compared to the known concentration of the target gas. The gases and concentra- tions used for testing the AR-500 are shown in Table 3-1. The verification was conducted by measuring the gases in a fixed sequence over three days. The sequence of activities for testing the monitor for a single gas is shown in Table 3-2. Table 3-1. Target Gases and Concentrations for Testing the AR-500 Concentration Target Gas Concentration Gas Cell Concentration3 Gas Level (ppm*m) (PPm) cl 3 60 Ammonia c2 6 120 c3 10 200 c4 20 400 cl 2 40 NO c2 5 100 c3 10 200 c4 15 300 cl 2 40 Benzene c2 3 60 c3 5 100 c4 10 200 aLength of gas cell = 4.98 cm 4 ------- Table 3-2. Optical Open-Path Monitor Verification: Measurement Order for Each Target Gas Times (min.) Meas. Gas Cell # of Path Length Verification Parameter # Cone. Activity Measurements Integrate Equilibrate (m) Calculated n2 Change gas & stabilize 10 100 1 n2° Collect spectra 25 1 100 Accuracy, Concentration linearity, MDL cl Change gas & stabilize 10 100 2 cl Collect spectra 5 1 100 Accuracy, Concentration linearity 3 cl Collect spectra - ND 1 5 1 100 Source strength linearity 4 cl Collect spectra - ND 2 5 1 100 Source strength linearity 5 cl Collect spectra - ND 3 5 1 100 Source strength linearity n2 Change gas & stabilize 10 100 6 n2 Collect spectra 5 1 100 Accuracy, Concentration linearity c2 Change gas & stabilize 10 100 7 c2 Collect spectra 5 1 100 Accuracy, Concentration linearity, Interference effect (Int.) n2 Change gas & stabilize 10 100 8 n2 Collect spectra 5 1 100 Accuracy, Concentration linearity c3 Change gas & stabilize 10 100 9 c3 Collect spectra 5 1 100 Accuracy, Concentration linearity 10 c3 Collect spectra - ND 1 5 1 100 Source strength linearity 11 c3 Collect spectra - ND 2 5 1 100 Source strength linearity 12 c3 Collect spectra - ND 3 5 1 100 Source strength linearity n2 Change gas & stabilize 10 100 13 n2 Collect spectra 5 1 100 Accuracy, Concentration linearity c4 Change gas & stabilize 10 100 14 c4 Collect spectra 25 1 100 Accuracy, Concentration linearity, Precision 14b n2 Collect spectra 5 100 Accuracy, Concentration linearity n2 Change gas & stabilize 10 100 15 n2 Collect spectra 25 5 100 Concentration linearity, MDL Change to Path length 2 20 250 16 n2 Collect spectra 5 5 250 Int. c2 Change gas & stabilize 10 250 17 c2 Collect spectra 5 5 250 Int., Accuracy, Concentration linearity n2 Change gas & stabilize 10 250 18 n2 Collect spectra 5 5 250 Int., Accuracy, Concentration linearity Change to Path length 3 20 optimumb 19 n2 Collect spectra 5 1 optimum Int., Accuracy, Concentration linearity c2 Change gas & stabilize 10 optimum 20 c2 Collect spectra 5 1 optimum Int., Accuracy, Concentration linearity n2 Change gas & stabilize 10 optimum 21 N, Collect spectra 25 1 optimum Int., MDL a See Table 3-1 for values of cl-c4 for the three target gases. b Vendor optimum of 250 meters was selected. ------- 3.2 Test Design The verification test was performed near West Jefferson, Ohio, at an outdoor testing area belonging to Battelle, between April 11 and April 16, 2000. This location provided sufficient length and a direct line of sight for each of the two path lengths used during the test, and provided an area that was away from major chemical sources that might affect the testing. The AR-500 receiver was mounted on top of an 8-foot-tall, concrete block column near the edge of a lightly traveled road and pointed toward the AR- 500 source, which was on top of another concrete block column located along the road at a distance of 100 meters. Another column was located at 250 meters from the receiver, and the source was located on top of this second column for the measurements that required a longer path length. The power supply, the computer, and the optical bench were located inside a temperature-controlled trailer near the receiver. The open space in the foreground of Figure 3-1 shows the test site at Battelle's West Jefferson facility. The testing area was near the edge of several farm fields. It also was located near a set of train tracks, and periodically trains passing by affected the NO measurements. In those cases, the testing was suspended until the train passed. Occasionally vehicles traveled along the road next to the test site. Testing was not suspended when vehicles passed, which may have contributed to background levels of NO. Figure 3-1. Test Site at West Jefferson Facility 6 ------- The AR-500 was challenged with the target gases shown in Table 3-1 at known concentrations, and the AR-500 measurement was compared to the known concentration of the target gas. For each target gas, the monitor was set up as if it were operating in the field, except that an optically transparent gas cell was placed in the light beam's path (see Figure 3-2). National Institute of Standards and Technology (NIST) traceable or commercially certified standard gases, a calibrated gas diluter, and a supply of certified high-purity dilution gas were used to supply the target gases to the gas cell. TARGET GAS CELL GAS CELL OPTICAL PATH LIGHT SOURCE FIBER OPTIC CABLE I OPTICAL BENCH & PC GAS DILUTION SYSTEM TO VENT TARGET GAS OR GASES Figure 3-2. Optical Open-Path Monitor Setup DILUTION GAS Target gases were measured at different path lengths, integration times, source intensities, and numbers of replicate measurements to assess Minimum detection limit (MDL) Source strength linearity Concentration linearity Accuracy Precision Sensitivity to atmospheric interferences. The test procedures shown in Table 3-2 were nested, in that each measurement was used to evaluate more than one of the above parameters. In Table 3-2, N2 in the gas cell concentration column denotes a period of cell flushing with high-purity nitrogen. The denotations cl, c2, c3, 7 ------- and c4 refer to the concentrations shown in Table 3-1. The last column shows the parameters to be calculated with the data from that measurement. 3.3 Experimental Apparatus and Materials 3.3.1 Standard Gases The standard gases diluted to produce target gas levels for the verification testing were NIST traceable gases or commercially certified gases. The gases were obtained in concentrations appropriate for dilution to the concentrations required for the test. 3.3.2 Dilution Gas The dilution gas was ultra-high-purity nitrogen obtained from commercial suppliers. 3.3.3 Gas Dilution System The dilution system used to generate known concentrations of the target gases was an Environics 2020 (Serial No. 2428). This system had mass flow capabilities with an accuracy of approxi- mately ±1%. The dilution system accepted a flow of compressed gas standard and could be diluted with high-purity nitrogen or air. It was capable of performing dilution ratios from 1:1 to at least 100:1. 3.3.4 Gas Cell A vendor-provided quartz gas cell 4.98 centimeters in length was integrated into the end of the receiver. This cell had two 1/4-inch tube fittings that allowed the target gas to flow through. 3.3.5 Temperature Sensor A thermocouple with a commercial digital temperature readout was used to monitor ambient air and test cell temperatures. This sensor was operated in accordance with the manufacturer's instructions and was calibrated against a certified temperature measurement standard within the 12 months preceding the verification test. 3.3.6 Ozone Sensor The sensor used to determine ozone in ambient air was a commercial UV absorption monitor (ThermoEnvironmental Model 49) designated by U.S. EPA as an Equivalent Method for this measurement. The UV absorption method is preferred for this application over the Reference Method (which is based on ethylene chemiluminescence) because the UV method is inherently calibrated and requires no reagent gases or calibration standards. The sensor was operated in accordance with the manufacturer's instructions. 8 ------- 3.3.7 NO/NH3 Monitor A chemiluminescent nitrogen oxides monitor [Advanced Pollution Instrumentation (API) Model 200, Serial No. 142] was used with a high-temperature ammonia converter (API Model 1000, Serial No. 100-233-120F-120H) to monitor the NO and NH3 concentrations supplied to the optical cell for verification testing. This monitor sampled gas immediately downstream of the optical cell to confirm the NO or NH3 concentrations prepared by dilution of high-concentration NO or ammonia standards. The API monitor was calibrated with a NIST-traceable commercial standard cylinder of NO in nitrogen. The conversion efficiency for NH3 was checked by comparing the calibration slope for NO with that found in calibrations with NH3. All NH3 measurements were corrected for the NH3 conversion efficiency, which was generally greater than 95%. 3.3.8 Benzene Measurement Benzene concentrations provided to the optical cell were checked by collecting a sample at the exit of the cell using pre-cleaned Summa® stainless steel air sampling canisters. The collected sample was then analyzed for benzene by gas chromatography with flame ionization detection (GC-FID), according to a method based on EPA Method 18. This method used certified commercial standards of propane in air for calibration. 3.4 Test Parameters 3.4.1 Minimum Detection Limit The MDL was calculated for each target gas by supplying pure nitrogen to the test cell in the optical path of the monitor and taking a series of 25 measurements using integration times of 1 and 5 minutes. The MDL was defined as two times the standard deviation of the calculated target gas concentrations. The sequence of measurements was conducted at both integration times, twice at a 100-meter path length and once at a 250-meter path length. 3.4.2 Linearity Two types of linearity were investigated during this verification: source strength and concentra- tion. Source strength linearity was investigated by measuring the effects of reducing the source intensity on the monitor's performance. In the field, light signal levels can be attenuated by mist, rain, snow, or dirty optical components. As a constant concentration of target gas was introduced into the gas cell, the light intensity of the source was reduced by placing an aluminum wire mesh in the path of the light to determine how the monitor's measurements were affected by an attenuated light source. Three aluminum wire screens of various meshes were placed in the beam path. These screens were approximately 1 foot square and had a mesh spacing of approximately Vi, Vi, and 1 inch. At each of these attenuation levels, a measurement was made, and the monitor analyzed for the target gas. The test was performed at two concentrations (2 ppm*m and 10 ppm*m) using NO. 9 ------- Concentration linearity was investigated by challenging the AR-500 with each target gas at the concentrations shown in Table 3-1, while the path length and integration time were kept constant. At each concentration, the monitor response was recorded and its linearity evaluated by com- paring the recorded response with the input target gas concentration. 3.4.3 Accuracy Accuracy of the monitor relative to the gas standards was verified by introducing known concentrations of the target gas into the cell. The gas cell was first flushed with at least five cell volumes of nitrogen, and five zero measurements were recorded. The target gas was then intro- duced into the cell and, after flushing with at least five cell volumes, five measurements of the target gas were obtained. The cell was again flushed with at least five cell volumes of nitrogen, and five more zero measurements were recorded. The concentration of the target gas was the average value with the target gas in the cell, minus the average of the zero measurements. The accuracy was evaluated at concentrations denoted as cl through c4, using an integration time of 1 minute. The accuracy was then evaluated at concentration c2 using a longer integration time, and then again at concentration c2 during the interference measurements (Table 3-2). The percent relative accuracy for an experimental condition is the absolute value of the difference between the average monitor response and the reference monitor response, divided by the reference monitor response, times 100 (see Section 5.3). 3.4.4 Precision The procedure for determining precision was very similar to the procedure for determining accuracy. The gas cell was flushed with at least five cell volumes of nitrogen. The target gas was then introduced into the cell and, after flushing with at least five cell volumes, 25 measurements of the target gas were obtained. The relative standard deviation (RSD) of this set of measure- ments was the precision at the target gas concentration. Precision was evaluated by this procedure at one concentration of the target gas (see Table 3-2). 3.4.5 Interferences The effects of interfering gases were established by supplying the gas cell with a target gas and varying the distance (i.e., the path length) between the source and detector of the monitor. For the UV measurement of the target gases, the main interferences in ambient air are 02 and 03, and changing the path length effectively changed the amount of interferants in the light path for the measurement. The purpose of the interference measurements was to determine the effects that the ambient atmospheric gases have on the accuracy and MDL of the AR-500. These tests were performed using two different integration times to determine the effect of integration time on the monitor's ability to perform measurements with interfering gases in the light path. To determine the effect of the interferences, the path length was first set to 100 meters. Then, the gas cell was supplied with nitrogen and, after flushing with at least five cell volumes, five measurements were recorded. Next, the target gas was introduced into the cell; and, after 10 ------- similarly flushing the cell, five measurements were recorded. Finally, the cell was flushed again and five more measurements were recorded. Atmospheric concentrations of 02 and 03 were recorded at the beginning and the end of these measurements. The path length was then set to 250 meters, which was the length that Opsis chose as optimum (i.e., the path length that theoretically yields the best signal-to-noise ratio), and the entire measurement procedure was repeated. The sensitivity of the monitor to the interferant was calculated by comparing the results at different path lengths (i.e., different ppm*m levels of 02 and 03). 11 ------- Chapter 4 Quality Assurance/Quality Control Quality assurance/quality control (QA/QC) procedures were performed in accordance with the quality management plan (QMP) for the AMS Center® and the test/QA plan(1) for this verification test. 4.1 Data Review and Validation Test data were reviewed by the Verification Testing Coordinator and disclosed to the Verification Testing Leader. The Verification Testing Coordinator reviewed the raw data and the data sheets that were generated each day. Laboratory record notebook entries also were signed and dated. 4.2 Changes from the Test/QA Plan Two types of changes from the test/QA plan could occur: planned changes to improve the test procedures for a specific vendor (amendments) and changes that occurred unexpectedly (deviations). Deviations from the test/QA plan were as follows: ¦ The test/QA plan called for a one-over-one data review within two weeks of generating the data. While the entire data set was reviewed within this two-week period, documentation of this task was not completed. Although this task was documented after the two-week period, no reduction in the quality of the data occurred. ¦ The thermocouple used in the verification test to monitor ambient air temperatures had not been calibrated within the previous six months, as specified in the test/QA plan. The thermocouple had been calibrated within one year, however, and was still within its calibration certification period. In addition, the thermocouple temperature measurement agreed with the mercury bulb thermometer temperature measurement during the performance audit. ¦ The test/QA plan calls for a performance evaluation audit of the NO measurement using a calibration standard obtained from an independent supplier. Instead, a separate NO standard obtained from the same manufacturer was used for the PE audit. 12 ------- ¦ An additional measurement was recorded because trains and trucks in the area caused varying background levels of NO. The additional background measurement was taken to help overcome this problem of fluctuations in the background NO levels. ¦ The test/QA plan called for acid rain CEM zero nitrogen to be used to flush the cell and as dilution gas. Instead, ultra-high-purity nitrogen was used. Deviation reports have been filed for each deviation. Before the verification test began, several planned amendments were made to the original test/QA plan to improve the quality or efficiency of the test. These procedural changes were implemented and, in each case, either increased the quality of the collected data set or removed inefficiencies in the test, ultimately resulting in a reduced test duration. A brief summary of these amendments is provided below: ¦ MDL was determined using twice the standard deviation, as described in section 3.4.1. The test/QA plan inadvertently called for the MDL to be determined by two different methods. The correct method was chosen and used during the verification test. ¦ The benzene analysis procedure was changed from that specified in the test/QA plan. The test/QA plan specified using Method 18, which is designed to determine the hydrocarbon emissions from combustion or other source facilities. This method broadly describes an analysis procedure, but does not specify how the analysis is to be done and calls for the use of Tedlar bags rather than Summa® canisters. Instead of as described in the test/QA plan, the analysis was done according to Battelle's GC/FID/MS analysis procedure for canister samples. ¦ The long and the short path lengths in the test/QA plan, which were specified as 100 and 400 meters, were changed to meet the specific technology requirements of the monitor tested. ¦ The order of testing in the test/QA plan was changed. The test order was originally developed to maximize the efficiency of the test procedure. Several improvements were made to the test matrix to further improve its efficiency. For example, instead of conducting all of the measurements for one gas then changing to the next gas, all of the short path measurements were conducted before moving to the long path. This was done because changing the path length was more time consuming than changing the target gas. ¦ One additional test was added to complete the data set collected. Originally, the test/QA plan lacked a nitrogen flush after measurement 14, under the same conditions as measure- ment 14. This additional measurement was added to the test matrix as measurement #14b (see Table 3-2). ¦ The test/QA plan specified that source strength linearity would be tested for each of the gases. The original intent was to conduct this test for one gas only. The source strength linearity test thus was conducted only for a single gas. 13 ------- ¦ The original test/QA plan specified that the ambient oxygen concentration be monitored by an oxygen analyzer. Instead, the ambient oxygen concentrations were assumed to be 20.9%. ¦ Although monitoring CO was part of the test/QA plan, it was decided that CO measure- ments would not add any useful information to the verification. No CO monitoring was conducted. ¦ The test/QA plan called for determining ammonia converter efficiency by placing two converters in series with the NO monitor. Instead, conversion efficiency was calculated by comparing NO and NH3 calibration curves. Amendments required the approval of Battelle's Verification Testing Leader and Center Manager. A planned deviation form was used for documentation and approval of all amendments. Neither the deviations nor the amendments had a significant impact on the test results used to verify the performance of the optical open-path monitors. 4.3 Calibration 4.3.1 Gas Dilution System Mass flow controllers in the Environics gas dilution system were calibrated prior to the start of the verification test by means of a soap bubble flow meter. Corrections were applied to the bubble meter data for pressure, temperature, and water vapor content. 4.3.2 Temperature Sensor The thermocouple was calibrated by comparing it to a certified standard in September 1999. This instrument has a one-year calibration period, and so was still within its calibration interval. 4.3.3 Ozone Sensor The UV absorption method of ozone measurement is inherently calibrated, relying as it does on the accurately determined absorption coefficient of ozone. As a result, routine calibration of the ozone monitor is not needed. However, the monitor was operated according to the manufac- turer's directions, with careful attention to the diagnostic indicators that assure proper operation. 4.3.4 NO/NH3 Monitor The NO/NH3 monitor was calibrated with both NO and NH3 standards. The NO standard was a Certified Master Class Calibration Standard of 6,960 ppm NO in nitrogen, of ±1% analytical uncertainty (Scott Specialty Gases, Cylinder No. K026227). The NH3 standard was also a Certified Master Class Calibration Standard, of 494 ppm NH3 in air, of ± 2% analytical 14 ------- uncertainty (Scott, Cylinder No. ALM 005256). The ratio of the slopes of the NH3 and NO calibration curves established the NH3 conversion efficiency. A performance evaluation audit was also conducted once during the test, in which the API monitor's response was tested with a different NO standard. For that audit, the comparison standard used was a NIST-traceable EPA Protocol Gas of 3,925 ppm NO in nitrogen, with ± 1% analytical uncertainty (Scott, Cylinder No. ALM 057210). 4.3.5 Benzene Measurement The GC/FID measurement for benzene was calibrated using two standard gases. One was an EPA Protocol Gas of of 32.73 ppm propane in air, with analytical uncertainty of ± 2% (Cylinder No. AAL 20803, Scott Specialty Gases). The other was a Certified Working Class Calibration Standard of 340 ppm propane in air, with ± 5% analytical uncertainty (Cylinder No. ALM 025084, also from Scott). 4.4 Data Collection Data acquisition was performed primarily by Battelle and the vendor. Table 4-1 summarizes the type of data recorded (see also Appendix A); where, how often, and by whom the recording was made; and the disposition or subsequent processing of the data. Test records were then converted to Excel spreadsheet files. Table 4-1. Summary of Data Recording Process for the AR-500 Verification Test Recorded Where Data Recorded By Recorded When Recorded Disposition of Data Dates, Times, Test Battelle Data Sheet Start of each test, Used to compile result, Events whenever testing manually entered into conditions changed spreadsheet as necessary Test Parameters Battelle Data Sheet Every hour during Transferred to (temp., RH, etc.) testing spreadsheet Interference Gas Battelle Data Sheet Before and after Transferred to Concentrations each measurement spreadsheet of target gas Target Gas Battelle Data Sheet At specified time Transferred to Concentrations during each test spreadsheet Optical Open-Path Battelle Data Sheet At specified time Transferred to Monitor Readings during each test spreadsheet 15 ------- 4.5 Assessments and Audits 4.5.1 Technical Systems Audit A technical systems audit (TSA) was conducted on April 13 and 14 for the open-path monitor verification test conducted in early 2000. The TSA was performed by Battelle's Quality Manager as specified in the AMS Center QMP. The TSA ensures that the verification test is conducted according to the test/QA plan(1) and that all activities associated with the test are in compliance with the AMS QMP.(3) Specifically, the calibration sources and methods used were reviewed and compared with test procedures in the test/QA plan. Equipment calibration records and gas certificates of analysis were reviewed. The conduct of the testing was observed, and the results were assessed. All findings noted during the TSA on the above dates were documented and submitted to the Verification Testing Coordinator for correction. The corrections were documented by the Verification Testing Coordinator and reviewed by Battelle's Quality Manager, Verification Testing Leader, and Center Manager. None of the findings adversely affected the quality or outcome of this verification test, and all were resolved to the satisfaction of the Battelle Quality Manager. The records concerning the TSA are permanently stored with the Battelle Quality Manager. In addition to the internal TSA performed by Battelle's Quality Manager, an external TSA was conducted by EPA on April 14, 2000. The TSA conducted by EPA included all the components listed in the first paragraph of this section. A single finding was noted in the external TSA, which was documented in a report to the Battelle Center Manager for review. A response and corrective action were prepared and returned to EPA. The finding did not adversely affect the quality or outcome of this verification test. 4.5.2 Performance Evaluation Audit A performance evaluation audit was conducted to assess the quality of the measurements made in the verification test. This audit addressed only those measurements made by Battelle in con- ducting the verification test. The performance audit procedures (Table 4-2) were performed by the technical staff responsible for the measurements. Battelle's Quality Manager was present to assess the results. The performance evaluation audit was conducted by comparing test measurements to independent measurements or standards. Each of the required procedures for the performance evaluation audit was conducted during the testing period in accordance with the direction specified in the test/QA plan, except for the deviation concerning the NO performance evaluation, listed in Section 4.2. The results from the performance evaluation are shown in Table 4-2. The temperature measurement agreed to within 0.4 °C and the ozone to within 3 ppb. The monitor used for NO/NH3 determination agreed with the performance evaluation standard within 4%, at a concentration of 75 ppm. 16 ------- Table 4-2. Summary of Performance Evaluation Audit Procedures Measurement Audited Audit Procedure Expected Actual Difference Temperature Compare to independent temperature measurement (Hg thermometer) 10°C 9.6°C 0.4°C Ozone Compare to independent ozone measurement 16.2 ppb 19 ppb 17.3% NO/NH3 Compare using another NO standard from the same supplier 75 ppm 72 ppm -4.0% Benzene Compare to results of gas chromatographic analysis of canister sample 0 ppm 40 ppm 60 ppm 200 ppm 0 ppm 37 ppm 56 ppm 168 ppm 0% -7.5% -6.7% -16.0% The benzene concentrations were audited by independent analysis of the test gas mixture supplied to the optical cell during verification testing. The results of the performance audit for the benzene concentrations were within 10% (except one canister, which was within 16%) of the expected concentrations, which met the test/QA plan criterion. 4.5.3 Data Quality Audit Battelle's Quality Manager audited at least 10% of the verification data acquired in the verifi- cation test. The Quality Manager traced the data from initial acquisition, through reduction and statistical comparisons, to final reporting. All calculations performed on the data undergoing audit were checked. 17 ------- Chapter 5 Statistical Methods The following statistical methods were used to reduce and generate results for the performance factors. 5.1 Minimum Detection Limit The MDL is defined as the smallest concentration at which the monitor's expected response exceeds the calibration curve at the background reading by two times the standard deviation (g0) of the monitor's background reading. MDL = 2a O 5.2 Linearity Both concentration and source strength linearity were assessed by linear regression with the certified gas concentration as independent variable and the monitor's response as dependent variable. Linearity was assessed in terms of the slope, intercept, and correlation coefficient of the linear regression. y = mx + b where y is the response of the monitor to a target gas, x is the concentration of the target gas in the optical cell, m is the slope of the linear regression curve, and b is the zero offset. 5.3 Accuracy The relative accuracy (A) of the monitor with respect to the target gas was assessed by R-T A= - x 100 R where the bars indicate the mean of the reference (R) values and monitor (T) results. 18 ------- 5.4 Precision Precision was reported in terms of the percent RSD of a group of similar measurements. For a set of measurements given by T\, T2, Tn, the standard deviation (a) of these measurements is O 1 n -1 1(T„-T) k= 1 1/2 where T is the average of the monitor's readings. The RSD is calculated from RSD x 100 and is a measure of the measurement uncertainty relative to the absolute value of the measurement. This parameter was determined at one concentration per gas. 5.5 Interferences The extent to which interferences affected MDL and accuracy was calculated in terms of sensitivity of the monitor to the interferant species, relative to its sensitivity to the target gas, at a fixed path length and integration time. The relative sensitivity is calculated as the ratio of the observed response of the monitor to the actual concentration of the interferant. For example, a monitor that indicates 26 ppm of cyclohexane in air with an interference concentration of 100 ppm of C02 indicates 30 ppm of cyclohexane when the C02 concentration is changed to 200 ppm. This would result in an interference effect of (30 ppm - 26 ppm)cyclohexane/(200 ppm - 100 ppm) C02 = 0.04, or 4% relative sensitivity. 19 ------- Chapter 6 Test Results The results of the verification test of the AR-500 are presented in this section, based upon the statistical methods described in Chapter 5. The monitor was challenged with nitric oxide (NO), benzene, and ammonia over path lengths of 100 to 250 meters. These gases were chosen because they are typical gases that this monitor would be used to detect in the field. Test parameters included MDL, linearity, accuracy, precision, and the effects of atmospheric interferants on concentration measurements. In many cases, verification results are based on comparing the test cell concentration of target gas calculated from the AR-500's open-path measurement to the actual gas cell concentration. In addition, where appropriate, the path-average concentrations are noted. The AR-500 reports concentration averages over the entire path length being monitored. In this report, a measured concentration of 1.5 ppb means that the path average concentration is 1.5 ppb over the entire 100 or 250 meters, depending upon the stated path length. The path- average concentration is determined by multiplying the gas cell concentration by the gas cell length and then dividing by the path length used during the given measurement. While the measurements conducted during this verification test were done in as controlled a way as possible, several uncontrollable factors should be pointed out before the results are presented. There may have been sources of the target gases near the test site that could have affected the monitor's response during these measurements such as trains, highway traffic, and local vehicular traffic. Attempts were made to suspend testing during obvious periods of source activity; however, not all potential sources of the target gases could be eliminated. 6.1 Minimum Detection Limit The MDL was calculated from measurements in which there were no target gases in the gas cell, but the monitor analyzed the absorption spectra for the presence of a target gas. The data used to determine the MDL were obtained under several experimental conditions, including different path lengths and integration times, as shown in Table 6-1. Table 6-2 shows the results of the MDL calculations, in terms of both the path-average MDL value in ppb and the integrated MDL value in ppb*m over the total path. The results in Table 6-2 show that the AR-500 has an MDL of between 0.9 and 1.4 ppb for NO, 0.4 and 1.5 ppb for benzene, and 2.8 and 5.8 ppb for ammonia, at the path lengths and integration times tested. Changing the path lengths between 100 and 250 meters and changing the integration 20 ------- Table 6-1. MDL Data for the AR-500 NO Benzene Ammonia Path Length (m) Path Length (m) Path Length (m) 100 250 100 100 250 100 100 250 100 Measure- Integration Time (min) Integration Time (min) Integration Time (min) ment 1 1 5 1 15 115 Number Concentration (ppb) 1 1.41 2.50 2.40 2.70 0.30 -0.30 -5.70 1.41 -0.40 2 1.43 1.60 4.28 2.70 0.20 0.30 -2.30 1.50 0.00 3 1.06 1.80 3.01 2.80 0.10 0.30 -6.30 1.00 -0.40 4 1.82 2.50 2.70 2.60 -0.10 0.10 -2.50 2.70 -0.50 5 0.59 1.50 2.35 2.70 0.00 0.30 -1.80 0.20 -0.50 6 1.94 2.00 2.79 2.90 0.30 0.30 1.20 2.80 -1.10 7 1.75 1.70 2.13 2.80 0.40 0.50 -3.50 3.90 2.50 8 1.68 1.00 5.63 2.90 0.60 0.30 -9.70 -0.10 -1.10 9 2.01 1.90 2.68 2.50 0.70 0.00 -1.60 1.30 -2.40 10 1.29 2.30 2.55 3.00 0.00 0.10 -2.90 0.20 -2.00 11 0.98 2.00 2.88 2.60 -0.10 0.00 -0.90 -0.20 0.60 12 1.73 1.80 2.47 2.50 -0.20 0.30 -1.00 0.20 2.60 13 1.39 1.90 2.89 3.40 0.10 0.30 0.40 1.20 -0.20 14 1.78 1.50 3.21 3.00 -0.10 0.30 -1.60 4.20 -0.60 15 1.23 1.30 2.82 -0.70 -0.20 0.50 -4.10 1.60 0.60 16 0.93 1.50 2.56 2.80 0.00 0.60 1.40 0.30 1.40 17 0.16 2.40 3.17 2.60 0.00 0.20 0.90 1.10 0.90 18 1.42 2.40 3.28 3.30 -0.20 0.00 0.00 -0.80 1.70 19 1.71 2.90 2.78 3.00 0.10 0.30 -0.30 -0.90 -1.60 20 0.51 2.50 2.64 2.50 -0.40 0.30 1.20 2.10 -0.80 21 1.07 2.00 2.64 2.10 0.10 0.30 2.60 3.50 -2.60 22 0.87 2.20 2.40 2.50 0.00 -0.10 -0.10 0.40 -2.60 23 1.06 1.60 2.39 2.70 0.00 -0.10 0.70 1.20 -0.30 24 0.43 1.50 2.61 2.40 0.10 0.10 2.50 2.60 1.90 25 0.76 2.10 2.59 3.30 0.20 0.00 -1.20 2.60 0.40 times between 1 and 5 minutes had little consistent effect on the MDL. For two of the three target gases, the MDL is lowest at the 250-meter path length, which is consistent with the vendor's claim of a better signal-to-noise ratio at longer path length. 21 ------- Table 6-2. Minimum Detection Limits of the AR-500 Path Length Integration MDL MDL Target Gas (m) Time (min) (ppb) (DDb*m) NO 100 1 1.01 100 NO 250 1 0.91 225 NO 100 1.42 140 Benzene 100 1 1.51 150 Benzene 250 1 0.50 125 Benzene 100 0.42 42.2 Ammonia 100 1 5.8 580 Ammonia 250 1 2.8 700 Ammonia 100 5 3.1 310 6.2 Linearity 6.2.1 Source Strength Linearity Table 6-3 shows the results from this evaluation of source strength linearity, and Figure 6-1 shows a plot of the effect that the light signal level has on the monitor's measurements. In Table 6-3, the relative signal power is the measure of light attenuation during that measurement. Table 6-3. Source Strength Linearity of the AR-500 Relative Signal Power NO Concentration (PPb) Monitor Response (ppb) 1.00 20 19.5 0.81 20 21.1 0.67 20 20.4 0.40 20 19.6 1.00 100 103 0.81 100 106 0.67 100 105 0.40 100 106 22 ------- X X X x y = -3.047k + 107.12 R2 = 0.4938 ~ Rath Average NO conc. (PPb) ¦ Instrument Response (ppb) Rath Average NO conc. (PPb) —X— Instrument Response (ppb) Linear (Instrurrent Response (ppb)) - - - Linear (Instrurrent Response (ppb)) y = 0.2619x+19.945 R2 = 0.0078 — - ¦ — ¦ ¦ — ¦ 0.00 0.20 0.40 0.60 0.80 1.00 1.20 Relative signal strength (Arb. Units) Figure 6-1. Source Strength Linearity Plot of the AR-500 For example, a relative signal power of 0.81 means that the light level for that test is 81% of what the light level is during normal operating conditions. The NO concentration is the con- centration of gas being delivered to the gas cell during the measurement, and the monitor response is the resulting reading from the AR-500. The source strength results show that there is little degradation in monitor performance during conditions of declining source strength. The maximum differences between AR-500 response and the NO concentration were 1.1 ppb at 20 ppb NO and 6 ppb at 100 ppb NO. The data do not indicate any consistent effect of source strength on NO measurement, with source reductions of up to 60%. In addition, the coefficients of determination (r2) of 0.0078 and 0.4938, shown in Figure 6-1, indicate that reducing the source strength had little effect on the monitor's response over the range tested. 6.2.2 Concentration Linearity Table 6-4 and Figures 6-2 through 6-4 show the path-average results of the evaluation of concentration linearity. The regression analysis results are shown on the individual figures. The concentration linearity results show that the AR-500 has a linear response over the concentration ranges tested. The monitor response as given by the slope of the linear regression line is 1.02 for NO, with an r2 value of 0.9994; a slope of 0.95 for benzene, with an r2 value of 0.9992; and a slope of 1.11 for ammonia, with an r2 value of 0.9997. 23 ------- Table 6-4. Concentration Linearity Data for the AR-500 Target Gas Target Gas Concentration (PPb) Monitor Response (PPb) NO 20 19.5 NO 20 22.7 NO 50 51.6 NO 20 23.4 NO 100 103 NO 150 155 Benzene 20 18.9 Benzene 12 10.4 Benzene 30 29.4 Benzene 12 10.4 Benzene 50 48.3 Benzene 100 93.9 Ammonia 30 33.1 Ammonia 24 25.4 Ammonia 60 66.7 Ammonia 24 23.2 Ammonia 100 110 Ammonia 200 221 Concentration Linearity - NO 200 y = 1.02x + 1.28 o. 150 Q. r2 = 0.9994 a) g 100 o a> 50 0£ 0 50 100 150 200 Input Concentration (ppb) Figure 6-2. Concentration Linearity Plot of the AR-500 Challenged with NO 24 ------- Concentration Linearity - Benzene 100 c o y = 0.95x - 0.22 ^ 80 r2 = 0.9992 ~ 60 0 20 40 60 80 Input Concentration (ppb) 100 120 Figure 6-3. Concentration Linearity Plot of the AR-500 Challenged with Benzene Concentration Linearity - Ammonia Q. >- s 3 a) c g £ o S Q. 0) 0) ai 250 200 150 100 50 0 y= 1.11x-1.47 r2 = 0.9997 50 100 150 200 250 Input Concentration (ppb) Figure 6-4. Concentration Linearity Plot of the AR-500 Challgenged with Ammonia 25 ------- 6.3 Accuracy The accuracy of the AR-500 was evaluated at each target gas concentration introduced into the cell. These concentrations were introduced at the path lengths and integration times shown in Table 6-5. The accuracy results compare the monitor response with the target gas concentration as delivered by the Environics 2020 diluter. The AR-500's relative accuracy ranges from 2.7 to 17% for NO, from 2.1 to 14% for benzene, and from 3.3 to 11% for ammonia. Integration time had little effect on accuracy of the AR-500. The longer path length improved accuracy for ammonia, but had the opposite effect for NO and benzene. Table 6-5. Results of Accuracy Tests for the AR-500 Expected Integration Monitor Relative Concentration Path Length Time Response Accuracy Target Gas (PPb) (m) (min) (PPb) (%) NO 20 100 1 19.5 2.7 NO 20 250 1 22.7 14 NO 50 100 1 51.6 3.2 NO 20 250 23.4 17 NO 100 100 1 103 3.4 NO 150 100 1 155 3.0 Benzene 20 100 1 18.9 5.8 Benzene 12 250 1 10.4 14 Benzene 30 100 1 29.4 2.1 Benzene 12 250 10.4 14 Benzene 50 100 1 48.3 3.4 Benzene 100 100 1 93.9 6.1 Ammonia 30 100 1 33.1 10 Ammonia 24 250 1 25.4 5.6 Ammonia 60 100 1 66.7 11 Ammonia 24 250 23.2 3.3 Ammonia 100 100 1 110 9.5 Ammonia 200 100 1 221 10 6.4 Precision Precision data were collected during measurement #14 (see Table 3-2) using an integration time of 1 minute and a path length of 100 meters. The target gas was introduced into the gas cell at a fixed concentration, and 25 successive analyses were made for the target gas. The data from these measurements are found in Table 6-6, and the results are shown in Table 6-7. In both tables, the data are shown in terms of the path-average concentration of the target gas. Table 6-7 shows precision of about 0.5% RSD for NO, 0.6% RSD for benzene, and 1.5% RSD for ammonia. 26 ------- Table 6-6. Data from Precision Tests on the AR-500 Target Gas NO Benzene Ammonia Analysis (PPb) (PPb) (PPb) 1 168 96.7 219 2 168 96.9 221 3 166 97.6 220 4 167 97.4 221 5 167 97.2 220 6 167 96.6 214 7 166 95.8 221 8 167 96.1 228 9 168 96.2 220 10 168 96.6 222 11 169 96.8 225 12 167 96.2 220 13 167 97.0 219 14 167 96.4 221 15 168 96.4 223 16 168 96.2 219 17 168 96.2 223 18 169 96.2 227 19 167 95.8 231 20 166 96.2 219 21 168 95.6 219 22 167 97.2 221 23 168 96.8 223 24 168 97.5 224 25 168 97.0 221 Table 6-7. Results of Precision Tests on the AR-500 a Gas Cell AR-500 Concentration Average Standard Relative Standard Target Gas (PPb) (PPb) Deviation (ppb) Deviation (%) NO 150 167 0.77 0.46 Benzene 100 96.8 0.55 0.57 Ammonia 200 222 3.40 1.53 integration time = 1 minute, path length = 100 meters. 27 ------- 6.5 Interferences Interference tests of the AR-500 evaluated the effects that the common atmospheric interferants 02 and 03 have on the monitor's ability to determine the concentration of the target gases and on the MDL for the target gases. Tables 6-8 and 6-9 show the data used to determine the inter- ference effects of ozone and oxygen on the concentration and MDL. Table 6-8. Concentration Data from Interference Tests on the AR-500 Gas Cell Path Concen- Concentration Concentration Concentration Relative Target Length tration of Oxygen of Ozone of Target Gas Accuracy Gas (in) (PPb) (%*m) (ppb*m) (PPb) (%) NO 250 20 5225 500 22.7 14 NO 100 20 2090 3440 19.5 2.5 NO 250 20 5225 1075 23.4 17 Benzene 250 12 5225 6725 10.4 13 Benzene 100 30 2090 1270 29.4 2.0 Benzene 250 12 5225 7100 10.4 13 Ammonia 250 30 5225 4575 33.1 10 Ammonia 100 24 2090 1900 25.4 5.8 Ammonia 250 24 5225 4450 23.2 3.3 Table 6-9. MDL Data from Interference Tests on the AR-500 Path Concentration Concentration Target Length of Oxygen of Ozone MDL Gas (111) (%*m) (ppb*m) (PPb) NO 250 5225 3400 0.91 NO 100 2090 3290 1.01 Benzene 250 5225 7325 0.50 Benzene 100 2090 530 1.51 Ammonia 250 5225 5900 2.81 Ammonia 100 2090 1150 5.75 Both ozone and oxygen have absorption features in the same spectral region that the AR-500 uses to analyze for the target compounds. Because the concentration of these two potential interferants is usually much greater than the concentration of the compounds of interest, the presence of these compounds can make analyzing for the target compounds difficult. The AR- 500 uses various methods to deal with these interferants, and this test evaluated the effectiveness of these methods. 28 ------- Changing the total number of ozone and oxygen molecules in the path length had little effect on the monitor's ability to accurately calculate the concentrations of the target gas. The best accuracy for benzene was found with the lowest 02 and 03 levels, but this was not clearly the case for the other gases. Overall, no consistent effect on relative accuracy could be inferred. During these measurements, the ozone concentration in the path changed from 500 to 7100 ppb*m, and the oxygen concentration varied from 2090 to 5225 %*m. These results did not permit calculation of relative sensitivity, as described in Section 5.5. Instead, a comparison of the measured concentrations was made to the input concentrations. Likewise, changing the total number of ozone and oxygen molecules in the path length had little effect on the monitor's MDL for the target gas. The MDL varied from 0.91 to 1.01 ppb for NO, from 0.50 to 1.51 ppb for benzene, and from 2.81 to 5.75 ppb for ammonia; while the ozone concentration in the path changed from approximately 530 to 7325 ppb*m, and the oxygen concentration varied from approximately 2090 to 5225%*m. 6.6 Other Factors 6.6.1 Costs The cost of the AR-500, as tested, was not available from the vendor. Costs for the AR-500 depend on the specific application and are established in discussion with the vendor. 6.6.2 Data Completeness All portions of the verification test were completed, and all data that were to be recorded were successfully acquired. Thus, data completeness was 100%. 29 ------- Chapter 7 Performance Summary The AR-500 detection limits ranged between 0.9 and 1.4 ppb for NO, between 0.4 and 1.5 ppb for benzene, and between 2.8 and 5.8 ppb for ammonia. While the variation in detection limits could be due to the changes in path length and integration time, there was no consistent trend. That is, longer integration times did not, in general, lead to lower detection limits, nor did the longer path lengths. The tests of the effects of source strength on the measurement capability of the monitor showed that there was little to no degradation of monitor performance, with reductions in source strength of up to 60%. Coefficients of determination at two different test concentrations were low, indicating that reducing the source strength had little effect on the monitor's response over the range tested. The concentration linearity results showed that the AR-500 had a slope of 1.02 and an r2 value of 0.9994 for NO over a range of 20 to 150 ppb; a slope of 0.95 and an r2 value of 0.9992 for benzene over a range of 12 to 100 ppb; and a slope of 1.11 and an r2 value of 0.9997 for ammonia over a range of 24 to 200 ppb. Percent relative accuracy was evaluated over the same ranges of concentration noted above for concentration linearity testing. Relative accuracy over these ranges was 2.7% to 17% for NO, 2.1 to 14% for benzene, and 3.3 to 11% for ammonia. The monitor performed about equally well at long and short integration times and at long and short path lengths. Precision results showed that the AR-500 had an RSD of 0.46% for NO at a concentration of 150 ppb, an RSD of 0.57% for benzene at a concentration of 100 ppb, and an RSD of 1.53% for ammonia at a concentration of 200 ppb. This RSD was calculated at one experimental condition using a path length of 100 meters and an integration time of 1 minute. Analysis of the effects of the interferences of oxygen and ozone on the measurement ability of the AR-500 showed that neither the accuracy nor the MDLs for the target gases were affected in a consistent way by the oxygen and ozone in the light path. Variations in MDL and accuracy were similar to those found during the other measurements made under normal operating conditions. 30 ------- Chapter 8 References 1. Test/QA Plan for Verification of Optical Open-Path Monitors, Battelle, Columbus, Ohio, October 28, 1999. 2. Compendium Method TO-16 Long-Path Open-Path Fourier Transform Infrared Monitoring of Atmospheric Gases, EPA-625/R-96/010b, U.S. Environmental Protection Agency, Cincinnati, Ohio, January 1997. 3. Quality Management Plan (QMP) for the E TV Advanced Monitoring Systems Pilot, U.S. EPA Environmental Technology Verification Program, Battelle, Columbus, Ohio, September 1998. 31 ------- Appendix A Data Recording Sheet A-l ------- Sample Gas: Date: Operator: Reviewed bv: Measurement # Dell Temp (F) \mbient 02 Concentrations (ppb) \mbient C02 Concentrations (ppb) \mbient RH (%) \mbient 03 Concentrations (ppb) \mbient Temp (F) Integration Time r'ath Length Concentration in Cell Dell Length Time of Measurement ------- |