Active and passive FTIR monitoring of flare combustion efficiency

Active and Passive FTIR Monitoring of Flare Combustion Efficiency Industrial Monitor & Control Corporation History of Slow Learners  First passive FTIR program funded by US EPA in June 1983 at John Zink facility Tulsa, OK  TCEQ test in 2003 also at the John Zink facility in Tulsa  Marathon Texas City, TX September 2009  Ineos Addyston, Ohio October/November 2009  Shell Oil Houston, TX February/March 2010  Marathon Detroit, MI July 2010  TCEQ/UT Test John Zink, Tulsa ongoing Active Activeand andPassive Passive Open-Path Open-PathFTIR FTIR Monostatic FTIR Transceiver Monostatic FTIR Transceiver Active FTIR System - EPA Monitoring Program Retro Arrays For Active FTIR Retro Arrays For Active FTIR Passive Passive Open-Path Open-PathFTIR FTIR Passive FTIR Radiometer Passive FTIR Radiometer Passive Open-Path Signatures  Any hot gas emits infrared with exactly the same pattern that it absorbs it  Therefore species in emission spectra can be identified and quantitated in the same manner as they are in normal absorption spectroscopy However  The strength of emission is proportional to concentration as it is in absorption spectra but also to the temperature of the gas. Gas Emissions As A Function Of Temperature 1.4 300 C 500 C 700 C 1000 C Radiance (mw/cm2/str/cm-1) 1.2 1.0 0.8 0.6 0.4 0.2 0.0 -5000 -4000 -3000 -2000 Wavenumbers (cm-1) -1000 0 Hardware - Imacc Passive FTIR Radiometer Hardware - Passive FTIR At Flare Test Basic BasicElements Elementsof of Flare FlareRadiance RadianceMeasurements Measurements The Signal Observed •The FTIR Signal arises from Four elements: Background Radiance -Background Radiance -Flare Radiance -Atmospheric path Radiance and Transmission Flare Radiance Atmospheric Transmission & Radiance • The Total FTIR Signal is then: Rb * plume atm + Rp * atm + Ratm + Rinst • Measurements made: Robs = The observed plume radiance Mn = Instrument and ambient air background Mbb = Radiance of calibration source Mb = Radiance of the sky background Mir = An infrared source measurement giving air transmission Plume Signature and Efficiency • Combining the various measurements above, the plume transmission can be determined as:     plume       R obs  M n  L BB  air  P   M b M n L BB  air  P • The plume transmittance has in it the concentration of each gas in the plume • As a result, the flare efficiency can be obtained as: Eff  CO2 CO2  CO  HC  Soot Various Bands in the Plume Emission Spectrum Spectral Regions Fingerprint FingerprintRegion RegionVOCs VOCs CO CO HC HC&&CH4 CH4 CO2 CO2 Two-Color Infrared Detector CO2 and CO Emission Spectra CO2 CO2 CO CO Temperature Determination Slope = hc / kT Passive PassiveMeasurement Measurement Calibration Calibration&Validation &Validation Calibration and Validation • Calibration of the FTIR requires two types of measurements: • Radiance calibration using a "Black Body" Source • Gas concentration calibration using a "Hot Cell" source Imacc Calibration Source Calibration Cart for Passive FTIR Black Body Source on PFTIR Collimator Hot Cell for PFTIR Method Validation PFTIR Hot Cell Validation for CO2 90000 y = 0.9752x R2 = 0.9995 80000 Spike (ppm) 70000 60000 50000 40000 30000 20000 10000 0 0 20000 40000 60000 Quant (ppm) 80000 PFTIR Hot Cell Validation for CO 1000 y = 1.0563x R2 = 0.9805 900 800 Spike (ppm) 700 600 500 400 300 200 100 0 0 200 400 600 Quant (ppm) 800 1000 Stack StackComparisons Comparisons FCCU FCCUUnit Unit 20% 18% 16% 14% 12% 10% 8% 6% 4% 2% 0% -2% -4% -6% 15:30 7/8/2010 - FCCU Test CO2 & CO Results (Preliminary) 200 180 160 140 120 100 80 60 40 20 Trailer PFTIR CO2 (2000) Trailer PFTIR CO 15:40 15:50 16:00 16:10 Time (hh:mm) CEMS CO2 CEMS CO 16:20 16:30 0 -20 16:40 16:50 CO (ppm) CO2 (%) Marathon Petroleum Company Detroit Refinery - CP Flare PFTIR Flare Test - July 2010 20% 18% 16% 14% 12% 10% 8% 6% 4% 2% 0% -2% -4% -6% 15:30 7/8/2010 - FCCU Test CO2 & CO Results (Preliminary) 200 180 160 140 120 100 80 60 40 20 Trailer PFTIR CO2 (2000) Trailer PFTIR CO 15:40 15:50 16:00 16:10 Time (hh:mm) CEMS CO2 CEMS CO 16:20 16:30 0 -20 16:40 16:50 CO (ppm) CO2 (%) Marathon Petroleum Company Detroit Refinery - CP Flare PFTIR Flare Test - July 2010 Marathon Petroleum Company Detroit Refinery - CP Flare PFTIR Flare Test - July 2010 7/8/2010 - FCCU Test Temperature Results (Preliminary) 600 580 Temperature (°F) 560 540 520 500 Est. Stack Temp 480 Trailer PFTIR Temp 460 440 420 400 15:30 15:40 15:50 16:00 16:10 Time (hh:mm) 16:20 16:30 16:40 16:50 Marathon Petroleum Company Detroit Refinery - CP Flare PFTIR Flare Test - July 2010 7/8/2010 - FCCU Test Temperature Results (Preliminary) 600 580 Temperature (°F) 560 540 520 500 Est. Stack Temp 480 Trailer PFTIR Temp 460 440 420 400 15:30 15:40 15:50 16:00 16:10 Time (hh:mm) 16:20 16:30 16:40 16:50 FTIR Long Term Stability 1 0.98 EFF (%) Average value = 99.5% Standard Deviation = 0.11% 0.96 0.94 0.92 0.9 11:15 11:30 11:45 12:00 Time 12:15 12:30 PFTIR Validation • With the high precision and accuracy observed in the hot cell and stack data, one would expect high accuracy in the plume itself. •However: - It appears the plume is highly inhomogeneous even in terms of combustion efficiency -The CO/CO2 ratio varies with position in the plume producing a variable combustion efficiency - Low flow flares seem worse because they are not optically thick and the plume becomes "broken up" Flare Efficiency 1 Efficiency (%) 0.995 0.99 Moderate Over Steaming 0.985 Heavier Over Steaming 0.98 0.975 0.97 8:00 9:00 10:00 11:00 Time 12:00 13:00 IR Image of High Flow Flare IR Image of Low Flow Flare Combustion Efficiency (%) Marathon Petroleum Company Detroit Refinery - CP Flare PFTIR Flare Test - July 2010 A, B, & E Series CE vs. CZG NHV (Preliminary) 100% 98% 96% 94% 92% 90% 88% 86% 84% 82% 80% 78% 76% 74% 72% 70% 68% 66% A Series B Series E Series - 0.6 S/VG E Series - 1.0 S/VG 0 100 200 300 400 500 600 Combustion Zone Gas Net Heating Value (BTU/scf) 700 Where Are We • The In Maui methodology Appears solid: -We can reproduce concentrations from cell and stack tests accurately • Measurement procedures still require refinement to get representative data with plume inhomogeneity. This will require : - Longer temporal averaging - Larger fields of view for spatial averaging - Automation of the telescope pointing, to stay in proper temperature zone • An effort has been initiated to develop an EPA protocol for this approach Stay tuned for Scott Evans paper tomorrow showing more measurement results