1. Office of Research and Development National Risk Management Research Laboratory, Air Pollution Prevention and Control Division Photo image area measures 2” H x 6.93” W and can be masked by a collage strip of one, two or three images. The photo image area is located 3.19” from left and 3.81” from top of page. Each image used in collage should be reduced or cropped to a maximum of 2” high, stroked with a 1.5 pt white frame and positioned edge-to-edge with accompanying images. Importance of Temperature Calibration for the Sunset Laboratory Carbon Analyzer: NIOSH and IMPROVE Temperature Protocols Jelica Pavlovic, Oak Ridge Institute for Science and Education, Oak Ridge, TN and John S. Kinsey, U.S. Environmental Protection Agency, Office of Research and Development, National Risk Management Research Laboratory, Research Triangle Park, NC AAAR 31 st Annual Conference, Minneapolis, MN, October 9 th, 2012

2. Thermal – Optical Method Measure Total Carbon (TC), Elemental carbon (EC), and Organic Carbon (OC), and OC and EC sub-fractions Two widely used temperature protocols: –Interagency Monitoring of Protected Visual Environment (IMPROVE) – traditionally developed by Desert Research Institute (DRI), NV and analyzed on DRI Carbon Analyzer –National Institute of Occupational Safety and Health (NIOSH 5040) – developed by the NIOSH and analyzed on the Sunset Laboratory, Inc. Carbon Analyzer Two protocols differ in temperature and optical monitoring, and maximum carbon burn-off temperature 1 Dual Optics Sunset Lab Carbon Analyzer DRI Carbon Analyzer

3. Temperature biases (up to +50˚C) were found between set-point temperature and actual (filter) temperatures for Sunset Carbon Analyzer, and < 30˚C for the DRI Analyzer (Chow et al. (2005); Phuah et al. (2009)) No influence on IMPROVE OC and EC concentrations, but significant influence on carbon sub-fractions Problem: Tempilaq˚ liquid used for calibration ruins the sample oven No studies have been done before on the influence of temperature calibration on NIOSH carbon fractions and sub-fractions Sub-fractions are important since they were used in the past to distinguish between diesel and gasoline emissions, to characterize different source and combustion profiles, and to estimate the source contributions to suspended particulate matter. 2 Research Problem

4. Test the new non-destructive temperature calibration developed by Sunset Laboratory, Inc. under IMPROVE and NIOSH 5040 operating temperatures and measure the temperature differences; Correct temperature differences (offsets); Measure influence of temperature calibration on NIOSH and IMPROVE carbon fractions and sub-fractions; Measure difference between NIOSH 5040 and IMPROVE carbon fractions before and after temperature calibration 3 Goals of present study

5. 4 Normal Operation vs. Temperature calibration - During normal operation internal thermocouple is located ~ 2 cm downstream of the center of sample (quartz filter) - During the temperature calibration the tip of external calibration thermocouple is located 2 cm upstream of the internal (oven) thermocouple

6. Temperature Calibration 5

7. Temperature Calibration (TCAL) Results 6 Temperature bias before TCAL: ≤ 43˚C for T< 500˚C ≤ 75˚C for T< 890˚C Temperature bias after TCAL: ≤ 5˚C for T< 500˚C ≤ 9˚C for T< 890˚C

8. Sample Analysis 7 Comparisons Number of tests TC range (µg/cm 2 ) OC range (µg/cm 2 ) EC range (µg/cm 2 ) % OC% EC IMPROVE BEFORE vs. AFTER TCAL 14 1.81 - 18.00 1.01 – 6.23 0.73 – 15.81 10.9 – 59.240.8 – 89.1 NIOSH BEFORE vs. AFTER TCAL 15 2.05 – 17.98 1.23 – 6.01 0.82 – 15.43 14.1 – 62.937.1 – 85.9 IMPROVE vs. NIOSH BEFORE TCAL 22 1.80 – 29.30 0.99 – 12.58 0.81 – 22.82 10.9 – 62.937.1 – 89.1 IMPROVE vs. NIOSH AFTER TCAL 50 1.79 – 19.60 0.58 – 8.09 0.72 – 14.11 8.0 – 70.729.3 – 92.0 - Samples collected from aerosol generator – propane diffusion burner (5201 Mini-CAST Jing Ltd., Switzerland)

9. 8 IMPROVE Before vs. After TCAL ANALYSIS Average of ratios y/x ± STD Paired t-test P YX TC AFTERTC BEFORE 0.99 ± 0.04 0.281 EC AFTEREC BEFORE 0.97 ± 0.11 0.641 OC AFTEROC BEFORE0.97 ± 0.110.326 - No significant differences (P > 0.05) between IMPROVE TC, EC, and OC before and after temperature calibration

10. 9 IMPROVE Before vs. After TCAL OC1, OC3, OC4, EC1, and EC3 significantly changed after calibration (P < 0.05)

11. 10 NIOSH Before vs. After TCAL ANALYSIS Average of ratios y/x ± STD Paired t-test P YX TC AFTERTC BEFORE 0.97 ± 0.06 0.038 EC AFTEREC BEFORE 0.89 ± 0.06 < 0.001 OC AFTEROC BEFORE1.16 ± 0.17< 0.001 - Significant differences (P < 0.05) were found between NIOSH 5040 TC, EC, and OC before and after temperature calibration - EC was ~ 11% lower and OC ~ 16% higher than before calibration

12. 11 NIOSH Before vs. After TCAL OC1, OC4, EC1, EC2, EC3 and EC4 significantly changed after calibration (P < 0.50)

13. 12 IMPROVE vs. NIOSH Before TCAL ANALYSIS Average of ratios y/x ± STD Paired t-test P ComparisonsYX IMPROVE vs. NIOSH BEFORE TCAL TC IMPROVETC NIOSH 1.00 ± 0.04 0.751 EC IMPROVEEC NIOSH 1.10 ± 0.06 < 0.001 OC IMPROVEOC NIOSH 0.79 ± 0.11 < 0.001 IMPROVE vs. NIOSH AFTER TCAL TC IMPROVETC NIOSH 1.00 ± 0.06 0.170 EC IMPROVEEC NIOSH 1.17 ± 0.12 < 0.001 OC IMPROVEOC NIOSH0.68 ± 0.18< 0.001 - Before and after the calibration, TC is the same measured by both protocols. - After calibration, the IMPROVE EC is even higher than the NIOSH EC (17% vs. 10% before calibration) and the IMPROVE OC even lower than the NIOSH OC (32% vs. 21% before calibration).

14. Simple temperature calibration of the Sunset Carbon Analyzer fixed the substantial temperature offset between filter and setpoint temperatures; The calibration did not have significant influence on IMPROVE carbon concentrations, but significantly changed the NIOSH 5040 TC, OC, and EC concentrations; All thermal carbon sub-fractions were different after calibration for both protocols, and most of them statistically significant which is important for inter-comparison of studies focused on thermally-derived carbon fractions in source profiles for emission inventories and receptor source apportionment modeling studies; The calibration increased the already existing gap between the NIOSH and IMPROVE carbon results The study emphasizes the need for individual instrument calibration to avoid misinterpretation of the data in future 13 Conclusions

15. Oak Ridge Institute for Science and Education, Oak Ridge, TN and U.S. Environmental Protection Agency, Office of Research and Development, National Risk Management Research Laboratory, Research Triangle Park for supporting my Post-Doctoral Research at EPA Sunset Laboratory Inc. at Hillsborough, NC for assisting with the temperature calibration 14 Acknowledgements

16. 15 QUESTIONS???

17. References: Chow, J. C., Watson, J. G., Chen, L.W.A., Paredes-Miranda, G., Chang, M.C.O., Trimble, D., Fung, K. K., Zhang, H., Zhen Yu, J., 2005. Refining Temperature Measures in Thermal/Optical Carbon Analysis. Atmospheric Chemistry and Physics 5, 2961-2972. Phuah, C. H., Peterson, M. R., Richards, M. H., Turner, J. H., Dillner, A. M., 2009. A Temperature Calibration Procedure for the Sunset Laboratory Carbon Aerosol Analysis Lab Instrument. Aerosol Science and Technology 43, 1013-1021. 16