Labile NO y Detection by Thermal Decomposition/Laser Induced Fluorescence

1. Labile NOy Detection by Thermal Decomposition/Laser Induced Fluorescence Douglas A. Day, Paul Wooldridge, Michael Dillon, Ronald C. Cohen Departments of Chemistry and of Geology and Geophysics University of California, Berkeley This work was supported by the NASA Instrument Incubator program Contract NAS1-99053

2. We describe a prototype designed to obtain in situ observations of the concentrations of the major NOy reservoirs. Thermal dissociation (TD) is used to fragment labile NOyspecies into NO2 and a companion radical. The NO2 fragment is then detected by laser-induced fluorescence (LIF).The LIF instrument developed in our group is accurate (~5%), sensitive (~10ppt/10seconds), precise and free from interferences. It is lightweight (as little as 200 lbs), compact, autonomous for days to months, and capable of making measurements from ground-based and airborne platforms. The TD section is a second channel operating in parallel with a channel detecting ambient NO2. In the TD channel, ambient air flows rapidly (15ms residence time) through a quartz tube heated by a nichrome wire. NOy species that fragment to NO2 are then detected by LIF. Calculations and laboratory experiments show the instrument can easily distinguish between signals due to N2O5, peroxynitrates (RO2NO2), and alkyl nitrates and hydroxy nitrates (RONO2), and HNO3 by varying the temperature in the dissociation region. A laboratory intercomparison between a NOy chemiluminescence detector and the TD/LIF technique showed excellent agreement for samples of n-propylnitrate. Details of the laboratory tests and of an ambient field trial will be presented.

3. hn NO3 HONO N2O5 NO2 hn O3 hn H2Ol OH hn OH NO2 HNO3 NO RO2, HO2, O3 hn , OH Emissions Deposition RONO2 ROONO2 NOy Photochemistry

4. Why a new technique? • N2O5 and HNO4 have never been measured in situ. N2O5 is an important path for loss of NOx through heterogeneous conversion to HNO3 in both the stratosphere and troposphere. • The extent to which NOx emission can lead to ozone production downwind of sources is affected by the partitioning of NOx between soluble and insoluble reservoir species. • Missing NOy: Often found that the sum of individually speciated NOy (NO, NO2, PAN, HNO3) do not add up to the total NOy measured separately. Missing NOy is often correlated with degree of processing suggesting importance of a slew of organic nitrates: alkyl nitrates, hydroxy nitrates, other peroxynitrates’s. • Important to measure speciated NOy in order to assess anthropogenic vs biogenic contribution to nitrate formation. • Existing techniques for ROxNO2 all use chromatography which is slow and may suffer from surface losses. • HNO3 is a difficult molecule to measure accurately due to inlet losses.

5. 2B2 Cooled PMT NO2* Filters to block laser scatter White Cell To Pump 585 nm Sample Inlet l>700 nm imaged on to PMT 2A1 Power Monitor Etalon Io Nd:YAG NO2 Reference I 585 nm Tunable Dye Laser 532 nm LIF detection of NO2 Using a pulsed, tunable dye laser, a specific rovibronic transition of NO2 is excited in a multipass white cell and the fluorescence is detected after physical and time-gated filtering using a PMT

6. offline online The NO2 Spectrum: Specificity 0.2nm @ 585nm Absorption Fluorescence The absorption of NO2 is monitored in a reference cell (red). After tuning to the peak, the dye laser continually dithers on and off of a chosen spectral feature to demonstrate an interference-free fluorescence signal (black).

7. Current Status LIF NO2: 1999 • Direct and spectroscopically specific • accurate: 5%, 1s • sensitive: 10ppt/10sec, S/N=2 • detection limit 1ppt • continuous and autonomous operation for more than a month • 450lbs 3kW Details are available in: Thornton, Wooldridge and Cohen, Analytical Chemistry, in press, 2000

8. Thermal Dissociation 1.0 0.8 0.6 yield 2 NO 0.4 N O 2 5 PAN 0.2 HNO 4 ClONO 2 n-propylnitrate 0.0 HNO 3 100 225 350 475 600 Temperature (C) Thermal Dissociation-LIF XNO2 + heat  X + NO2 The predicted NO2 yield vs. temperature curves for the different classes of nitrates compounds is unique T< 50C NO2 (1) T~ 130C N2O5+RO2NO2 +(1) (2) T~ 300C RONO2 +(2) (3) T~ 550C HNO3+(3) (4) The sampled air passes immediately through a heated quartz tube where, depending up the set temperature, only certain classes of compounds are pyrolyzed producing NO2 and a sister radical. The sample is quickly transported to the detection axis via PFA tubing where the NO2 is measure by laser induced fluorescence.

9. 1.0 0.9 [Nitric Acid] 0.8 0.7 0.6 0.5 [Alkyl- and hydroxy nitrates] 0.4 0.3 0.2 [peroxynitrates] 0.1 0.0 100 150 200 250 300 350 400 450 500 550 600 Temperature (C) NO2 yield occurs in three easily resolved steps

10. 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 30 130 230 330 430 N2O5 N O Thermal Dissociation A simple extremely well calibrated test: a) Measure NO2 b) Add O3 NO2+O3 NO3 +O2 NO2+NO3 N2O5 c) Add heat N2O5+heat NO2+NO3 d) One half the original N2O5 is recovered as NO2 2 5 NO relative to O =0 2 3 model (T shifted to match initial slope) 0 ] 2 ] / [NO Initial Concentration: 2 [NO NO 5ppm 2 O 15 ppm 3 Pressure 65 Torr 8% O 2 15 msec residence Temperature(C)

11. 24 21 18 15 12 9 6 3 0 30 80 130 280 330 380 430 480 530 n-propylnitrate Thermal Dissociation of n-Propylnitrate The thermal spectrum of n-propylnitrate is observed as predicted from theory. The vapor pressure of NPN is not know well enough to use it as a calibration standard. (ppb) 2 NO Temperature (C)

12. 600 500 400 300 200 100 0 250 30 80 130 180 230 280 330 380 430 480 530 HNO3 and C2H5NO3 Thermal Dissociation of Ethylnitrate and Nitric Acid Individual compounds can be detected within mixtures. In practice we have 2 or 3 channels set at different temperatures to record, for example, NO2, RO2NO2 and HNO3. Detected (ppb) 2 NO T measured remote from chemistry Temperature (C)

13. Intercomparison with an NOy sensor Two completely independent techniques give the same result over a decade of concentration.

14. Field Trial at Blodgett Forest • Site is a ponderosa pine plantation in the foothills of the Sierra Nevada mountains downwind from the polluted Sacramento Valley • Prevalent biogenic emissions mixed with a NOy rich urban plume are ideal for investigating the processes that lead to formation of organic nitrates. • At Blodgett Forest, peroxynitrates typically composed ¼ of the NOy and the alkyl nitrate + hydroxy nitrate group was also about ¼.

15. 3.0 2.5 2.0 HNO 3 1.5 NOy - NOx (ppb) Alkyl + Hydroxy Nitrates 1.0 0.5 PAN's 0.0 0 100 200 300 400 500 600 700 Oven Temperature (Celcius) 990927 Thermal Spectrum: 990927 This scan was obtained in the early evening in the peak of the urban plume at Blodgett Forest at the point indicated in the slide below.

16. 990927 130 degrC, PAN's 300 degrC, PAN's + Alkyl+hydroxy 600 degrC, PAN's + Alkyl+hydroxy + HNO 3 2.6 2.4 2.2 2.0 1.8 1.6 1.4 NOy - NOx (ppb) Temperature Scan 1.2 1.0 0.8 0.6 0.4 0.2 269.4 269.5 269.6 269.7 269.8 269.9 270.0 270.1 270.2 270.3 Julian Day

17. 4 3 NOy - NOx (ppb) Little or no Alkyl+Hydroxy step HNO 2 3 PAN's 1 Oven Temperature (Celcius) 0 0 100 200 300 400 500 600 991005 Thermal Spectrum: 991005 This scan was obtained in the early evening in the peak of the urban plume at Blodgett Forest at the point indicated in the slide below.

18. 991005 130 degrC, PAN's 300 degrC, PAN's + Alkyl+hydroxy 600 degrC, PAN's + Alkyl+hydroxy + HNO 3 4 Temperature Scan 3 2 NOy - NOx (ppb) 1 0 277.5 277.6 277.7 277.8 277.9 278.0 278.1 278.2 278.3 Julian Day

19. 3 2 ppb 2 NO 1 0 300 301 302 303 304 305 130 degrC, PAN's 300 degrC, PAN's + Alkyl+hydroxy 600 degrC, PAN's + Alkyl+hydroxy + HNO 3 2.0 1.6 (ppb) 1.2 2 - NO 0.8 * y NO 0.4 0.0 300 301 302 303 304 305

20. 2.0 PAN's Alkyl+hydroxy 1.6 HNO 3 1.2 ppb 0.8 0.4 0.0 300 301 302 303 304 305 Julian Day A few day sequence of NO2 (top panel), the sum of NOy compounds observed at different inlet temperatures, with the NO2 subtracted (middle panel) and individual classes of NOy compounds obtained by differencing signals obtained at characteristic temperatures.

21. Future Work and Applications TD/LIF Further analysis of the 1999 Blodgett campaign data addressing ozone production, sources of organic nitrates, and modeling the history and processing of the air sampled at the site. NO2, organic nitrates, and HNO3 will be measured at Blodgett Forest for the entire spring-summer-fall season in 2000. The dual cell technique has been incorporated in an airborne instrument to measure N2O5, HNO4 and PAN during TOPSE flights over the arctic winter/spring 2000. Fluxes of organic nitrates and HNO3 will be investigated at Blodgett Forest.