FTIR For Stack and CEM
- 1. Recent Developments in FTIR for Stack Emissions and CEM Monitoring in the Power Generation Industry Sylvie Bosch-Charpenay MKS Instruments On-Line Product Group
- 2. New Regulations New Standards for Combustion Engines – Standards of Performance for Stationary Spark Ignition Internal Combustion Engines : 40 CFR Part 60 subpart JJJJ. Gases to be measured: NOx, CO, THC – National Emission Standards for Hazardous Air Pollutants (NESHAP) for Reciprocating Internal Combustion Engines (RICE): 40 CFR Part 63 subpart ZZZZ. Gases to be measured: NOx, CO, THC, formaldehyde Hazardous Air Pollutants – Formaldehyde (year 2013, test on annual basis) – Future speciation of individual HCs (NMHC=non-methane HC’s, methane, ethane) – Methanol, etc… 2
- 3. Emission Monitoring Standard methods – NOx : Chemiluminescence (CLD) – lower accuracy in high NO2 – CO, CO2 : Non-Dispersive Infrared (NDIR) – separate analyzer for each gas – THC : Field Ionization Detectors (FID) – provides a single number (no speciation) – Electrochemical sensors – separate analyzer for each gas FTIR provides measurements of many gases simultaneously – CO, CO2, NO, NO2, N2O, NH3, CH4, HCl, HF, ethane, ethylene, propylene, formaldehyde, H2O, etc… 3
- 4. FTIR Advantages FTIR is cost-effective if more than 4 gases need to be measured FTIR requires minimum calibration and so reduces costs Can be easily (and inexpensively) shipped on-site, instead of deploying an entire vehicle FTIR is best method to measure formaldehyde 4
- 5. Infrared (IR) Spectroscopy Based on IR light absorption – Energy (IR radiation) heats molecule - vibrations and rotations – The pattern and intensity of the spectrum provides all the information about gas (type and concentration) H2O Spectrum 5
- 6. FTIR Provides Real-Time Analysis of Multiple Species 6 Averaging 15 sec to 1 minute per point
- 7. Measurements Requirements EPA Methods DIN EN 15277-3 3A, 4, 7E, 10 Sensitivity (=short <2% of cal span <2% of cal span term repeatability) Accuracy (= <2% of cal span <2% of cal span calibration error) Interferences <2.5% of cal span. Tested <4% of cal span. Tested (=cross-sensitivity) once. annually. Drift <3% of cal span <2% of cal span System Response _ Typically less than 200 sec Time cal span = calibration span = upper limit of calibration range 7
- 8. How Can FTIR Meet the Measurement Requirements? Sensitivity usually not an issue (long acquisition times OK) Accurate FTIR instrument needed Optimization of Analysis Method (minimal effect of interferences) Drift usually not an issue (background in N2 taken prior to testing) Optimization of Sampling System (response time, effective transport of “sticky” species) 8
- 9. Instrument Spectral Accuracy Requirements Instrument Accuracy Optimization – Spectra linearity Accurate absorbance in the whole range of absorbance level – Resolution Instrument has same resolution (i.e., line width) as the calibration spectra – Line position Instrument spectra are “lining up” exactly with the calibration spectra Validation – Standard historical approach is to run cal cylinder and create instrument-specific calibration – New, better approach is to use transferrable calibrations (possible because of excellent instrument to instrument matching) and verify cal cylinder 9
- 10. Accuracy Validation: Correct Resolution, Line Position and Absorbance Level Comparison of calibration (yellow) to sample (white) for CO 10 => Excellent overlap of calibration and sample spectra
- 11. Accurate Multi-Point Calibrations 11 Multiple frequencies and calibration levels
- 12. Tuning of Analysis Method to Minimize Interferences For minimal interferences, optimized analysis range and masking (“picket fencing”) Correction factors included to compensate for matrix effects (NO, CO) for best accuracy Custom water calibration may be needed (but only for very low calibration spans) “Canned methods” should be made available by manufacturer Additional components can be easily added 12
- 13. Ability to Measure NO between Water Peaks Sample = 150 ppm NO in 35% H2O Top: sample (white) and water spectrum (red) Bottom: sample minus water (white) and NO calibration (green). Grey 13 areas are “picket-fenced” regions which are not used in the analysis
- 14. No Interference of Water High Sensitivity H2O steps up to 40% Low detection limits No artificial bias even in very high water (up to 40%) Note: The HCl and HF sharp decaying peaks are real and represent small amounts accumulated in transfer lines.14 Other sharp positive and negative peaks are short duration artifacts due to fast water levels changes
- 15. Typical Achievable Measurement Ranges Species Ranges in mg/m3 NO 0-30, 0-200, 0-400, 0-1500 NO2 0-50, 0-100, 0-1000 N2O 0-50, 0-100, 0-500 NH3 0-10, 0-75 CO 0-75, 0-150, 0-1500 HF 0-5, 0-10 HCl 0-15, 0-90, 0-200 SO2 0-75, 0-300, 0-2000 CO2 0-25% H2O 0-40% CH4 0-15, 0-50 15
- 16. Optimization of FTIR Sampling System Heated probe with filtering – Metal or Glass – Stainless steel filter required for “sticky compounds” HF, HCl, NO2, NH3 – <0.1 um recommended (must keep particulate low) Heated sampling line – SS (not Teflon) recommended for most applications – As short a length as possible – Maintain temp – 191 C (very important, no cold spot!) – Maintain pressure – 1.0 Atm (+/- 5% recommended) Sampling pump before or after FTIR Gas Analyzer – Before: Be careful about contamination or sample loss Additional Filtering Possible – After: Be careful not to let pressure go too low 16
- 17. Mechanical & Equilibration Response Times 17 “Sticky” compounds are the same when H2O is present
- 18. Formaldehyde Easy to Measure H2O sample Formaldehyde Sample with 5 ppm formaldehyde and 5% water (broad peaks on right correspond to diesel, also measured) 18
- 19. Formaldehyde Testing In Stationary Combustion Turbines DL = 200-300 ppb under typical conditions DL as low as 30 ppb under optimized conditions 19
- 20. Q&A Information ASTM Method 6348 -03 Standard Test Method for Determination of Gaseous Compounds by Extractive Direct Interface Fourier Transform Infrared Spectroscopy http://www.epa.gov/ttn/atw/rice/fr05mr09.pdf 20