FINAL REPORT Environmental Chamber Experiments to ...aqrp.ceer.utexas.edu/projectinfo/10-042/10-042...

FINAL REPORT

Environmental Chamber Experiments to Evaluate NOx Sinks and Recycling in Atmospheric Chemical Mechanisms

AQRP Project 10-042

Prepared by:

Greg Yarwood, ENVIRON Gookyoung Heo, CE-CERT

William P.L. Carter, CE-CERT Gary Z. Whitten, SmogReyes

Center for Environmental Research and Technology

College of Engineering University of California

Riverside, California 92521

ENVIRON International 773 San Marin Drive

Suite 2115 Novato

California, 94998

SmogReyes PO Box 518

Point Reyes Station California, 94956

Prepared for:

Dr. Elena C. McDonald-Buller Texas Air Quality Research Program

The University of Texas at Austin

February 17, 2012

ENVIRON Project Number: 06-25699B1

ACKNOWLEDGMENT The preparation of this report is based on work supported by the State of Texas through the Air Quality Research Program administered by The University of Texas at Austin by means of a Grant from the Texas Commission on Environmental Quality.

i

TABLE OF CONTENTS ACKNOWLEDGMENT ........................................................................................................................II

EXECUTIVE SUMMARY ................................................................................................................... IV

1.0 INTRODUCTION ......................................................................................................................... 1

2.0 EXPERIMENTS PREFORMED AT EUPHORE ................................................................................ 2 Data and Chamber Characterization ........................................................................................ 2

3.0 EXPERIMENTS PERFORMED AT UCR ......................................................................................... 6 Design of NOx Sink Experiments ............................................................................................... 6 Design of NOx Source Experiments .......................................................................................... 8 Isoprene Experiments ............................................................................................................... 9 Control and Characterization Experiments ............................................................................... 9 Light Source Employed ........................................................................................................... 11 Instrumentation ...................................................................................................................... 13 Experiments Performed .......................................................................................................... 14 NOx Sink Experiments ............................................................................................................. 14 NOx Source Experiments ........................................................................................................ 19 Isoprene Experiments ............................................................................................................. 24 Control and Characterization Experiments ............................................................................. 25

4.0 IMPROVEMENTS TO CB6 ........................................................................................................ 29 Carbon Bond version 6 (CB6) .................................................................................................. 29 Mechanism Updates for CB6r1 ............................................................................................... 39 Simulations of EUPHORE Experiments with CB6r1 ................................................................. 43 Simulations of UCR Experiments with CB6r1 .......................................................................... 48

5.0 PERFORMANCE EVALUATION OF CB6R1 ................................................................................ 54 Hierarchical Approach to Mechanism Evaluation .................................................................. 54 Data Used to Evaluate CB6r1 .................................................................................................. 55 CB6r1 Evaluation Results ........................................................................................................ 59 Impact of Uncertain OH + NO2 Rate Constant ........................................................................ 62

6.0 CONCLUSIONS AND RECOMMENDATIONS ............................................................................. 67

7.0 REFERENCES ............................................................................................................................ 70

APPENDICES Appendix A: Experimental Methods for the EPA Chamber at UCR

Appendix B: Results of SAPRC-07 Evaluation

Appendix C: Summary of Chamber Experiments Performed for This Project

TABLES Table 1. Initial conditions (ppb) for 9 EHPHORE chamber experiments. ................................. 3

ii

Table 2. Numbers of experiments of different type performed using the EPA chamber at UCR. ....................................................................................................................... 14

Table 3. Summary of NOx sink experiments carried out for this project. .............................. 15 Table 4. Summary of NOx source and related chamber characterization experiments. ....... 21 Table 5. Summary of isoprene - NOx mechanism evaluation experiments and the single VOC

- NOx control and characterizaton experiments carried out for this project. ...... 26 Table 6. Listing of reactions and rate parameters for CB6. .................................................... 30 Table 7. Model species names for CB6. .................................................................................. 38 Table 8. Environmental chambers at UCR and TVA used for mechanism evaluation. ........... 56 Table 9. Summary of 194 chamber experiments for single test compounds and special

mixtures. ............................................................................................................... 57 Table 10. Summary of 145 surrogate mixture experiments. .................................................. 58 Table 11. Model errors for Max(O3), Max(D(O3-NO)) and the NOx crossover time. .............. 60 Table 12. Model errors for Max(O3), Max(D(O3-NO)) and the NOx crossover time with an

alternate OH + NO2 rate constant. ........................................................................ 63

FIGURES Figure 1. Environmental chamber simulation of EUPHORE CO experiment EU033104. ......... 4 Figure 2. Environmental chamber simulation of EUPHORE ethene - NOx experiment

EU100101 ................................................................................................................ 5 Figure 3. Spectrum of the light sources used with the UCR EPA environmental chamber, with

representative solar spectrum also shown for comparison. The relative intensities in the spectra are normalized to give the same NO2 photolysis rate. 12

Figure 4. Plots of selected results of NOx sink experiments with toluene. (Model calculations are shown for the base case.) ............................................................................... 16

Figure 5. Plots of selected results of NOx sink experiments with o-cresol and furan. (Model calculations are shown for the base case.) ........................................................... 17

Figure 6. Plots of selected results of NOx sink experiments with isoprene. (Model calculations are shown for the base case.) ........................................................... 18

Figure 7. Plots of selected results of NOx source experiments with the organic nitrates. .... 22 Figure 8. Plots of selected results of NOx source experiments with 2-nitrophenol. ............. 23 Figure 9. Plots of selected results of the acetaldehyde - H2O2 - NOx control experiments

carried out for this project. (Results of the model calculations are also shown.) 24 Figure 10. Plots of selected results of the isoprene - NOx experiments carried out for this

project. .................................................................................................................. 27 Figure 11. Plots of selected results of the ethene - NOx and propene - NOx control

experiments carried out for this project. (Results of the model calculations are also shown.) .......................................................................................................... 28

Figure 12. Model simulations with CB6r1 of EUPHORE experiment EU100201 with 2-butenedial and NOx. ............................................................................................. 44

Figure 13. Model simulations with CB6r1 of EUPHORE experiment EU100301 with 4-oxo-2-pentenal and NOx. ................................................................................................ 45

Figure 14. Model simulations with CB6r1 of EUPHORE experiment EU100401 with o-cresol, HONO and NOx. .................................................................................................... 46

iii

Figure 15. Model simulations with CB6r1 of EUPHORE experiment EU092501 with toluene and NOx................................................................................................................. 47

Figure 16. Model simulations with CB6r1 of UCR NOx-sink experiments with toluene, o-cresol, furan and isoprene added to a base mixture of ethene and NOx . .......... 50

Figure 17. Model simulations with CB6r1 of UCR NOx-source experiments for alkyl nitrates added to CH3CHO and H2O2 or CO and H2O2. ....................................................... 51

Figure 18. Model simulations with CB6r1 of UCR NOx-source experiments for 2-nitrophenol added to CH3CHO and H2O2 or CO and H2O2. ....................................................... 52

Figure 19. Model simulations with CB6r1 of UCR experiments for isoprene and NOx. ......... 53 Figure 20. Hierarchy of species for evaluating CB6 systematically. ....................................... 55 Figure 21. CB6 and CB6r1 model errors (%) for Max(O3). ...................................................... 60 Figure 22. CB6 and CB6r1 model errors (%) for Max(D(O3-NO)). .......................................... 61 Figure 23. CB6 and CB6r1 model errors (minutes) for NOx crossover time. .......................... 61 Figure 24. CB6 and CB6r1 model errors (%) for Max(O3) with an alternate OH + NO2 rate

constant. ............................................................................................................... 64 Figure 25. CB6 and CB6r1 model errors (%) for Max(D(O3-NO)) with an alternate OH + NO2

rate constant. ........................................................................................................ 65 Figure 26. CB6 and CB6r1 model errors (minutes) for NOx crossover time with an alternate

OH + NO2 rate constant. ....................................................................................... 65

iv

Executive Summary Formation of ground level ozone requires both nitrogen oxides (NOx) and volatile organic compounds (VOCs), and air quality management planning seeks a combination of NOx and VOC emission reductions that will most effectively reduce ozone. When VOCs undergo chemical reactions in the atmosphere, they can reduce the availability of NOx by converting it to compounds (e.g., nitric acid and organic nitrates) that react slowly, which we refer to as NOx-sink compounds. However, the NOx sink compounds may eventually react to return NOx back to the atmosphere (NOx recycling) and potentially cause additional O3 production in NOx-limited regions. It is important that this be properly taken into account when modeling regional ozone formation, particularly in multi-day episodes where transport to NOx-limited regions is important. Reactions of “NOx-sink” compounds that return the once sequestered NOx to an active form are referred to as “NOx-source” reactions.

The chemical reactions of VOCs with NOx can be characterized by environmental chamber experiments that expose controlled amounts of VOC and NOx to light and measure the products (e.g., ozone) that are formed. This project carried out new environmental chamber experiments to characterize NOx sinks and sources for several VOCs that are poorly understood. At the same time, data of chamber experiments performed in Europe were obtained and formatted for use in mechanism evaluation and development. The data obtained were used to improve the chemical reaction mechanisms that are used in the TCEQ’s State Implementation Plan (SIP) ozone modeling for control strategy development. The project benefit will be more accurate modeling of the ozone impacts of emission control strategies in Texas and elsewhere.

The importance of NOx sinks was demonstrated experimentally from the effect on O3 formation of adding representative compounds to alkene – NOx mixtures. Addition of toluene and isoprene reduced O3 formation from ethene - NOx and (in the case of toluene) propene - NOx mixtures, demonstrating the importance of NOx sinks for toluene and isoprene. Even larger O3 reductions, and therefore larger NOx sinks, were observed when o-cresol and furan were added to ethene - NOx mixtures, indicating that degradation products of aromatics play an important role in producing the NOx sink observed for toluene.

NOx recycling was demonstrated experimentally from representative alkyl nitrates and the representative aromatic oxidation product 2-nitrophenol in experiments where formation of NO2 and/or PAN was observed when these compounds were reacted in the absence of injected NOx. The amounts of NOx formed were substantially greater than can be attributed to any artifact caused by the chamber walls.

The experimental data obtained were used to test and improve the mechanisms for isoprene and aromatics in version 6 of the Carbon Bond mechanism (CB6). The revised mechanism is called CB6r1. CB6r1 performed better than CB6 in simulating experiments for toluene, xylenes and mixtures combining aromatics with other VOCs. However, mechanism performance for simulating VOC mixtures remains poorer when the mixture contains aromatics demonstrating that some aspects of the aromatics chemistry are still not fully understood. Simulations of experiments with 2-butenedial and 4-oxo-2-pentenal (degradation products of aromatics) performed poorly suggesting an area where the aromatics chemistry could be improved.

v

The experimental data obtained in this study strongly support the occurrence of NOx-recycling in the photolysis reactions of the NOx-source compounds isopropyl nitrate, isobutyl nitrate and 2-nitrophenol. Accordingly, CB6r1 includes NOx-recycling from photolysis of NTR (representing alkyl nitrates) and CRON (representing nitrocresols). The results of this study suggest that OH-reactions of isopropyl and isobutyl nitrate also lead to NOx recycling but the results are not conclusive because, under our experimental conditions, the isopropyl nitrate and isobutyl nitrate were mainly consumed by photolysis rather than OH reaction. CB6r1 tentatively includes NOx recycling from the reaction of OH with NTR (a change from CB6) but we recommend conducting photochemical model sensitivity tests to evaluate the impacts of this change. Additional experiments to test for the occurrence of NOx recycling from alkyl nitrates larger than isopropyl and isobutyl are needed because OH reaction is more important relative to photolysis and the nitrogen-containing products could be different for larger alkyl nitrates than for the compounds studied here.

In the course of developing CB6r1 several errors in CB6 were identified and corrected. The corrected CB6 mechanism is implemented in CAMx version 5.40 and documented in the User’s Guide available from www.camx.com. These mechanism corrections were to the products of reactions 82 (XO2 + RO2), 86 (XO2N + RO2) and 112 (ALDX photolysis).

The rate constant for the reaction of OH with NO2 is uncertain with currently recommended values varying by 25%. CB6r1 uses a rate constant at the middle of this range recommended by the NASA-JPL evaluation panel. To assess the importance of this uncertainty, CB6r1 was evaluated with alternate slower OH + NO2 rate constant (by Mollner et al. 2010) as a sensitivity test. The main conclusions were that (1) either the rate constants recommended by NASA-JPL or Mollner et al. (2010) could be used with CB6r1 and (2) the results of evaluating CB6r1 with alternate OH + NO2 rate constants should not be used to decide which value is preferable. At this time, the OH + NO2 rate constant recommended by the NASA-JPL evaluation panel is used in CB6r1.

We make the following recommendations for additional chamber experiments and other activities to support improvements in chemical mechanisms:

1. Chamber experiments are needed to quantify NOx recycling from alkyl nitrates larger than isopropyl nitrate and isobutyl nitrate (studied here). The experimental techniques developed to study NOx source compounds should be used again. Larger alkyl nitrates should be studied because, compared to smaller alkyl nitrates, reaction with OH will become more important than photolysis and the products of reaction with OH could be different for large alkyl nitrates.

2. Chamber experiments performed at the Euphore chamber proved to be useful for evaluating and improving toluene mechanisms. Euphore data for additional aromatic hydrocarbons, e.g. for benzene, xylenes and trimethylbenzenes, are available and should be obtained and used for mechanism evaluation/development.

3. Chamber experiments should be performed with unsaturated dicarbonyls that are degradation products of aromatic hydrocarbons, e.g., 2-butenedial and 4-oxo-2-pentenal. Experiments are needed at low to moderate initial NO concentrations for a range of VOC/NOx ratios. Experiments that include a tracer to measure OH-radical production would

http://www.camx.com/

vi

be valuable. NOx-sink experiments, using the methods developed in this study, would be valuable. It would be useful to measure separately the yields of PAN and total PAN-type compounds to quantify the amount of PAN-analogues formed and how quickly they decay.

4. Chamber data for acetylene and glyoxal are very limited but the available data suggest that the mechanisms are biased or incomplete. New experiments in the EPA chamber for acetylene and glyoxal would be useful to reduce uncertainty in their mechanisms. The glyoxal mechanism is especially important because it is a reactive product of aromatics and isoprene.

5. Chamber experiments with surrogate hydrocarbon mixtures are important to evaluate how well mechanisms simulate urban atmospheres. Experiments with mixtures also test interactions between the mechanisms for individual mixture components. New chamber experiments should be performed using surrogate hydrocarbon mixtures designed to reflect conditions relevant to Houston (urban emissions with chemical industries) and Dallas (urban emissions with oil and gas production).

6. Ethene, propene, 1,3-butadiene, 1-butene, isobutene, trans-2-butene, cis-2-butene are regulated in Texas as HRVOCs. Data available for mechanism evaluation are extensive for ethene and propene but limited or missing for the other HRVOCs. Chamber experiments are recommended for 1-butene, isobutene, trans-2-butene, cis-2-butene, and 1,3-butadiene in order to develop/evaluate more explicit mechanisms for each compound. Improved chemical mechanisms for HRVOCs will provide more accurate estimation of the impacts of HRVOC emissions on ozone formation in Houston and other regions.

7. The CB6r1 mechanism should be implemented in air quality models (i.e. CAMx and CMAQ) for further testing and evaluation under ambient conditions.

1

1.0 Introduction Ozone (O3) formation results from a series of complex atmospheric chemical reactions involving both volatile organic compounds (VOCs) and oxides of nitrogen (NOx) (NRC, 1991) and the ability to model the chemistry of O3 formation is needed to develop effective emission control strategies. Although VOCs are generally considered in terms of enhancing O3 formation when NOx is present, they are also important in affecting NOx levels due to their atmospheric reactions. Most VOCs react to at least some extent to remove NOx by forming NOx sink compounds such as organic nitrates, peroxy nitrates, or nitro compounds, which reduces the availability of NOx to form O3. Some VOCs, such as aromatics, have sufficiently strong NOx sinks that increasing their concentration reduces O3 under NOx-limited conditions. However, these NOx sink species may eventually react to return NOx back to the atmosphere (NOx recycling) potentially causing additional O3 production in NOx-limited regions. It is important that this be properly taken into account when modeling regional ozone formation, particularly in multi-day episodes where transport to NOx-limited regions is important.

The chemical mechanisms needed to predict effects of VOC and NOx controls on ozone formation have uncertainties, and their predictive capabilities need to be evaluated by comparing their predictions against results of environmental chamber experiments (Jeffries et al, 1992). Chambers have advantages of being able to control composition and environmental factors but the disadvantage of background or wall effects, which among other problems can require that higher than ambient concentrations to be employed. The EPA chamber at the University of California at Riverside (UCR) (Carter, 2002; Carter et al, 2005a,b,c) was designed to mitigate some of these disadvantages and has been used successfully to perform numerous experiments at near-ambient concentrations (Carter, 2004, 2010a and references therein). The European Commission developed an environmental chamber in Valencia, Spain, called the European Photo-Reactor (EUPHORE), where experiments in natural sunlight at near ambient concentrations can be performed.

Most of the currently available chamber experiments were designed to test the rate at which different VOCs promote ozone formation in the presence of NOx. Although such experiments can have some sensitivity to NOx-sink processes, they are not adequate to fully evaluate model predictions of NOx sinks or sources generated from atmospheric oxidation of VOCs. Single VOC experiments do not test how VOCs can reduce O3 formation due to NOx sinks, most mixture experiments do not test VOC mechanisms in the absence of uncertainties of mechanisms of other VOCs, and most of the previous experiments carried out for mechanism evaluation, including those at UCR and in the EUPHORE, are single day irradiations whose results are not significantly affected by NOx recycling processes from secondary reactions of VOC products.

The first objective of this project was to design and carry out environmental chamber experiments that focus on the effects of NOx sinks and sources in VOC atmospheric reactions. Then, the data obtained were used in combination with other data to improve the capabilities of current mechanisms to model NOx sink and NOx source processes and, therefore, to improve mechanism performance in simulating O3 formation under NOx-limited conditions. The NOx sink experiments focused on aromatic hydrocarbons, because they introduce strong NOx sinks, and isoprene, because it is abundant in the atmosphere. The NOx source experiments focused on simple alkyl nitrates, because they are prototypes for an abundant class of compounds, and nitrophenols because they are formed when aromatic hydrocarbons react in the presence of NOx. Data of several experiments performed using the EUPHORE chamber were obtained and processed for use in this project because they focus on important products of the atmospheric reactions of aromatic hydrocarbons.

2

2.0 Experiments Preformed at EUPHORE The chemical mechanisms used by TCEQ for ozone SIP modeling were developed using environmental chamber data from the Universities of California at Riverside (UCR) and North Carolina (UNC). The European Commission developed an environmental chamber in Valencia, Spain, called the European Photo-Reactor (EUPHORE). Experiments that are relevant to the characterization of NOx-sinks in the chemistry of aromatic hydrocarbons were obtained and analyzed to enable their use for mechanism development in this project. A total of 9 experiments were obtained, 2 chamber characterization experiments needed to construct a chamber wall mechanism and a method to calculate photolysis frequencies (j values) and 7 experiments related to aromatics (4 for toluene, 1 for each of o-cresol, 2-butenedial and 4-oxo-pentenal). These experiments are now part of the database of chamber experiments maintained by UCR and will be available for use in future projects.

Simulations of the 9 EUPHORE chamber experiments were carried out to produce useful information to design chamber experiments to evaluate sinks and sources of NOx (NO and NO2) in atmospheric chemical mechanisms of aromatics such as toluene.

Data and Chamber Characterization The EUPHORE chamber is a large outdoor environmental chamber with two reactors each with a volume of ~200 m3. Data are publicly available from the EUROCHAMP Data Server (http://eurochamp-database.es/) and data for 9 experiments listed in Table 1 were formatted for use with the SAPRC chamber simulation software (Carter, 2000 and 2010). For details of instruments used for the EUPHORE experiments, refer to Bloss et al. (2005a) and Zádor et al. (2006). Simulating chamber experiments requires good characterization of conditions including the light source and the effect of chamber walls on radical and NOx concentrations (Carter et al., 2005). We developed (1) a wall mechanism to describe chamber-dependent HONO and HCHO offgasing and (2) a method to calculate photolysis frequencies (j values) based on measured j(NO2) at the EUPHORE chamber.

One CO experiment, EU033104 (conducted on March 31, 2004) was used to evaluate two wall mechanisms for the EUPHORE chamber from Bloss et al. (2005a) and Zádor et al. (2006). This experiment is one of three experiments used by Zádor et al. (2006). The wall mechanism of Zádor et al. (2006) was constructed using measured NO, NO2, HCHO and HONO at the EUPHORE chamber and gave better fits to measured data (Figure 1) than the wall mechanism of Bloss et al. (2005) when used with the CB6 (Yarwood et al., 2010) and SAPRC10 mechanisms. The wall mechanism performed well in simulating the rate at which NOx and HCHO were produced and the resulting concentrations of HO2 and NO2. Based on these results, the wall mechanism of Zádor et al. (2006) was chosen as the wall mechanism for simulating other EUPHORE chamber experiments listed in Table 1.

http://eurochamp-database.es/

3

Table 1. Initial conditions (ppb) for 9 EHPHORE chamber experiments. Experiment ID EU033104 EU100101 EU091201 EU092401 EU092501 Test compound CO Ethene Toluene Toluene Toluene Date (mm/dd/yyyy) 3/31/2004 10/1/2001 9/12/2001 9/24/2001 9/25/2001 Test compound [a] 1.65E+05 [d] 613 511 514 512 NO 0 175 42 47 422 NO2 0 23 7.9 4.4 52 HONO [b] 0 0.5 0.1 0.1 1.5 O3 0 0 0 0.9 0 HCHO 0 0.5 0.8 0 0 HNO3 0 2.6 0.4 0 0 Glyoxal 0 0 0.3 0 0 CO 1.65E+05 [d] 424 267 197 301 Start time [c] 10:00 10:05 10:05 10:05 9:45 End time [c] 12:30 16:00 13:55 13:25 15:25 Experiment ID EU092701 EU100401 EU100201 EU100301 Test compound Toluene o-Cresol 2-Butenedial 4-Oxo-pentenal Date (mm/dd/yyyy) 9/27/2001 10/4/2001 10/2/2001 10/3/2001 Test compound 496 297 641 303 NO 122 23.5 82.3 162 NO2 21 22.6 36.4 23 HONOb 1.5 65 0.1 1.5 O3 0.6 0.1 0.7 0 HCHO 1.5 0.7 0.4 0 HNO3 1 0.7 0 1.2 Glyoxal 0 0 0.4 0.5 CO 352 384 424 390 Start time [c] 10:05 11:05 10:05 10:00 End time [c] 14:50 13:30 13:10 14:15

[a] References: Bloss et al. (2005a and 2005b), Zádor et al. (2005 and 2006) [b] Initial HONO estimated except for EU100401. [c] Times in format hh:mm GMT (= LST -1 hr) [d] Injected CO concentration was based on measured CO concentrations around 10:45 GMT. The SAPRC software can model photolysis based on spectral characterization of the chamber light source and measured j(NO2). However, only measured j(NO2) was available for the EUPHORE experiments. Thus, photolysis frequencies were calculated in two steps. First, j values were calculated for the location of the EUPHORE chamber assuming clear sky conditions. Second, those j values were scaled to match the measured j(NO2) for each experiment. One Ethene – NOx experiment was selected to evaluate this methodology because ethene degradation is very sensitive to photolysis rate of HCHO. Ethene, NO, CO and glyoxal were relatively well simulated (Figure 2) while HCHO, HNO3 and were less well simulated. Correctly simulating the decay of ethene indicates that the OH concentration is simulated well. The peak NO2 was over-predicted by both CB6 and SAPRC10, which led to over-predicted O3 by ~20-25% after 5 hours of irradiation. However, considering the uncertainties in the chamber characterization and run-to-run variability observed between different chamber experiments, this level of performance is acceptable and indicates that the wall mechanism and photolysis frequencies are reasonable.

4

(a) CO (b) O3

(c) NO (d) NO2

(e) HCHO (f) HO2

(g) OH

Figure 1. Environmental chamber simulation of EUPHORE CO experiment EU033104.

EU033104 (31 March 2004, CO - air)

0

20000

40000

60000

80000

100000

120000

140000

160000

180000

10:00 11:00 12:00 13:00

Time (GMT)

CO

(ppb

)

CO (Expt)CO (CB6)CO (SAPRC10)

CO injection

Light off Light onLight on

NO2 (~3.5 ppb) injection

EU033104 (31 March 2004, CO - air)

-5

0

5

10

15

20

25

30

10:00 11:00 12:00 13:00

Time (GMT)

O3

(ppb

)

O3 (Expt)O3 (CB6)O3 (SAPRC10)

CO injection Light off Light onLight on


EU033104 (31 March 2004, CO - air)

-0.5

0.0

0.5

1.0

1.5

2.0

10:00 11:00 12:00 13:00

Time (GMT)

NO

(ppb

)

NO (Expt)NO (CB6)NO (SAPRC10)



EU033104 (31 March 2004, CO - air)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

10:00 11:00 12:00 13:00

Time (GMT)

NO

2 (p

pb)

NO2 (Expt, ECO)NO2 (CB6)NO2 (SAPRC10)



EU033104 (31 March 2004, CO - air)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

10:00 11:00 12:00 13:00

Time (GMT)

HC

HO

, FO

RM

(ppb

)

HCHO (Expt, FTIR)FORM (CB6)HCHO (SAPRC10)

EU033104 (31 March 2004, CO - air)

-10.0

0.0

10.0

20.0

30.0

40.0

50.0

60.0

10:00:00 11:00:00 12:00:00 13:00:00

Time (GMT)

HO

2 (p

pt)

HO2 (Expt, LIF)HO2 (CB6)HO2 (SAPRC10)

CO injection Light off Light onLight onNO2 (~3.5 ppb) injection

EU033104 (31 March 2004, CO - air)

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

10:00 11:00 12:00 13:00

Time (GMT)

OH

(ppt

)

OH (Expt, LIF)OH (CB6l)OH (SAPRC10)


5

(a) Ethene (b) O3

(c) NO (d) NO2

(e) HCHO (f) HNO3

(g) Glyoxal (h) CO

Figure 2. Environmental chamber simulation of EUPHORE ethene - NOx experiment EU100101

EU100101 (1 October 2001, Ethene - NOx)

0

100

200

300

400

500

600

10:00 11:00 12:00 13:00 14:00 15:00

Time (GMT)

Ethe

ne, E

TH (p

pb)

Ethene (Expt, FTIR)

ETH (CB6)

Ethene (SAPRC10)


0

100

200

300

400

500

600

10:00 11:00 12:00 13:00 14:00 15:00

Time (GMT)

O3

(ppb

)

O3 (Expt, ML)O3 (CB6)O3 (SAPRC10)O3 (Expt, FTIR)


0

50

100

150

200

10:00 11:00 12:00 13:00 14:00 15:00

Time (GMT)

NO

(ppb

)

NO (Expt, ECO)NO (CB6)NO (SAPRC10)NO (Expt, API)


0

20

40

60

80

100

120

140

160

10:00 11:00 12:00 13:00 14:00 15:00

Time (GMT)

NO

2 (p

pb)

NO2 (Expt, DOAS)NO2 (CB6)NO2 (SAPRC10)NO2 (Expt, API)


0

50

100

150

200

250

10:00 11:00 12:00 13:00 14:00 15:00

Time (GMT)

HC

HO

, FO

RM

(ppb

)

HCHO (Expt, FTIR)

FORM (CB6)

HCHO (SAPRC10)


0

10

20

30

40

50

60

10:00 11:00 12:00 13:00 14:00 15:00

Time (GMT)

HN

O3

(ppb

)

HNO3 (Expt, FTIR)

HNO3 (CB6)

HNO3 (SAPRC10)


-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

10:00 11:00 12:00 13:00 14:00 15:00

Time (GMT)

Gly

oxal

, GLY

(ppb

)

Glyoxal (Expt, DOAS)

GLY (CB6)

GLY (SAPRC10)


0

100

200

300

400

500

600

700

10:00 11:00 12:00 13:00 14:00 15:00

Time (GMT)

CO

(ppb

)

CO (Expt)

CO (CB6)

CO (SAPRC10)

6

3.0 Experiments Performed at UCR NOx sinks and NOx sources inherent to the atmospheric reactions of selected VOCs were characterized in experiments performed using the EPA environmental chamber facility at UCR. This section describes the experimental design and documents the experiments performed. The experimental methods are described in Appendix A.

Design of NOx Sink Experiments The method used to evaluate model predictions of NOx sinks is to conduct experiments to assess the effects of adding various test compounds on O3 formation and NOx species concentrations when the test compound is added to a simple model "base case" VOC - NOx irradiation system that causes O3 formation, but where the maximum O3 formed is limited by the availability of NOx. The presence of NOx sink processes in the reactions of the test VOC (or its reactive oxidation products) reduces the amount of NOx available to promote O3 formation, and thus causes less O3 to be formed than is the case when the test VOC is absent. The extent to which this occurs, i.e., decrease in O3 by addition of the test compound, can be used to evaluate the ability of the chemical mechanism to predict the magnitude of these NOx sink processes. Measurements of nitrogen containing species (e.g.., NO and NO2 and PAN) during in the base case and added test compound experiments provide additional tests to the mechanisms' predictive capabilities in this regard.

As discussed below, the UCR EPA environmental chamber system employed in this study consisted of two dual reactors, where the same base case VOC and NOx mixture can be injected in equal quantities into both reactors, and where the test compound can be injected into one of the reactors, and both can be irradiated under the same temperature and light conditions. This is similar to, an "incremental reactivity" experiment that was first described by Carter and Atkinson, (1987), and has been employed extensively to evaluate mechanisms for many types of VOCs (e.g., Carter and Malkina, 2005, 2007; reports cited in Carter, 2011), but in this case the objective of the experiment, and therefore the optimum base case, is somewhat different. In the case of incremental reactivity experiments, the objective is to use a base case experiment that represents ambient VOC and NOx systems, though in some cases highly simplified representations are employed (e.g., Carter et. al., 1995a). In the case of this study, the objective is to provide as simplified and well understood a base case system that provides a sensitive test of the impact of NOx sinks on O3 formation. The ideal base case experiment should employ a single compound that is reactive towards ozone formation, whose mechanism is reasonably well established, and where there is a large database of existing experiments that can be routinely and successfully modeled using current chemical mechanisms.

The choice of VOC for the base experiments was based on the considerations discussed above and simple alkenes, namely ethene or propene, were selected. Other compounds were ruled out for the following reasons; alkanes, CO or simple alcohols are unsuitable because they lack sufficient O3 formation tendency; aromatics are unsuitable because of uncertainties in their mechanisms; formaldehyde is unsuitable because it is difficult to handle and monitor experimentally and lacks sufficient NOx sinks of its own (e.g., no PAN formation)to achieve O3 maxima in reasonable amounts of time; acetaldehyde is unsuitable because it does not promote sufficient O3 formation without other reactants present; and other compounds lack an established database of experiments evaluated using multiple mechanisms or have other issues

7

such as uncertain mechanisms. Ethene - NOx and propene - NOx experiments promote O3 formation readily and achieve O3 maxima due to NOx depletion in a reasonable amount of time if appropriate alkene/NOx ratios are employed, have a large database of experiments that have been used to evaluate existing mechanisms, and have sufficient internal radical sources to promote rapid O3 formation and, as a result, are not highly sensitive to chamber effects, at least in cleaner chambers with low background radical sources.

Model calculations were carried out to determine the optimum alkene and NOx reactant levels to obtain maximum sensitivity of O3 formation to added VOCs that produce NOx sinks at realistic atmospheric NOx levels, and to determine which of the two alkenes (i.e., ethene and propene) provide the most sensitive and easy-to-model base case experiment. Propene has the advantage of having a larger database of chamber experiments that generally are well modeled, while ethene has the advantage of O3 formation being somewhat more sensitive to the addition compounds with NOx sinks on maximum O3 formation. There are some inconsistencies in model simulations of ethene experiments in some of the older chambers (e.g., Carter, 2010a,b; Yarwood et al., 2010), though ethene - NOx experiments in the UCR EPA chamber at low NOx levels are reasonably well simulated (Carter, 2004; Yarwood et al., 2010). O3 formation in base case experiments is about equally sensitive to changes in initial reactant concentrations within their ranges of uncertainty. Because of these considerations, we decided to focus primarily on ethene - NOx experiments for the base case for the NOx sink study, though several experiments using the propene - NOx as the base case were also employed using toluene as the test compound.

It was determined that experiments with ~15 parts per billion (ppb) of NOx would be appropriate for the objectives of this study. This NOx level is low enough to be atmospherically relevant, but not so low that experiments would be affected by chamber background effects or analytical limitations. Model calculations indicated that base case experiments with ~1 part per million (ppm) ethene or ~0.3 ppm propene supplied the appropriate amount of O3 and reached NOx-limited conditions in a reasonable amount of time and were appropriately sensitive to the addition of reactive species with NOx sinks (NOx sink species).

Four compounds whose oxidation processes are known or expected to produce NOx sinks were chosen to study to address the objectives of this project as follows:

1. Toluene: Toluene was chosen as a representative aromatic hydrocarbon that is important in emissions and is known to have NOx sink characteristics (Calvert et al., 2002).

2. o-Cresol: Phenolic compounds such as cresols are reactive aromatic oxidation products whose subsequent reactions are believed to make significant contributions to the NOx sink characteristics of aromatic compounds. o-Cresol was chosen because it is the phenolic compound formed in the highest yield from toluene (Calvert et al., 2002).

3. Furan: Aromatic ring opening processes are believed to form compounds such as unsaturated 1,4-dicarbonyls, which are believed to be highly reactive and important contributions to the reactivities of aromatics (Calvert et al, 2002), but the role of their reactions in contributing to NOx sinks in aromatic photooxidations is not well known (Liu et al., 1999; Bloss et al., 2005; Hu and Kamens, 2007; Whitten et al., 2010). However, 1,4-dicarbonyls are not commercially available and are difficult to handle and analyze

8

experimentally. Furan is known to rapidly react to form 2-butene-1,4-dial, the simplest example of these compounds, in high yields (Bierbach et al., 1995; Gόmez Albarez et al., 2009; Berndt et al., 1997). Therefore, experiments with furan may provide a relatively straightforward method for testing the mechanisms for these compounds. Experiments with methyl furans were not attempted because they have lower yields of 1,4-dicarbonyls (Bierbach et al., 1995; Gόmez Albarez et al., 2009) and therefore the co-products would complicate mechanism evaluations.

4. Isoprene: Isoprene is an important biogenic compound that dominates emissions in many non-urban and some urban areas. Model calculations indicate that its reactions have NOx sinks as well, so experiments to test this are included as part of this project.

The amount of the test compound to be added in the test experiments was derived such that the addition was predicted to have a significant effect on O3 formation while not overwhelming O3 formation in the base case system, according to model calculations using the SAPRC-07 mechanism (Carter, 2010a).

Design of NOx Source Experiments The method used to evaluate NOx sources was to irradiate the NOx source compounds in the presence of radical initiators and without injected NOx, and to measure the amount of NOy species formed. The amount of O3 formed in these experiments also provides an indirect method for evaluating NOx sources because NOx is required for O3 formation and many moles of O3 are formed from each mole of NOx under low NOx conditions. However, directly measuring NOx species is the priority objective for these experiments because this provides the most unambiguous information for model evaluation in regard to NOx sources.

The presence of a radical initiator such as H2O2 is usually necessary to assure that a sufficient amount of the NOx source compound reacts during the course of the experiment to obtain a measurable amount of NOx formation. It is also critical that the NOx input results primarily in the formation of NOy species that can be monitored. In particular, formation of HNO3 must be minimized because it is difficult to reliably quantify this compound and because it is lost to a significant extent on the chamber walls or surfaces of sampling lines. The formation of unknown nitrogen-containing products also needs to be minimized.

Two approaches can be employed to minimize the formation of NOy species that cannot be readily quantified. One approach is to add a reactant such as acetaldehyde that forms peroxyacetyl radicals (CH3C(O)OO·) at sufficient levels that its reaction with NO2 to form PAN (CH3C(O)OONO2) is the primary sink for any NOx that formed. PAN can then be monitored if a suitable method is available. This approach has merit and was used for this project, but it has two drawbacks. The first is that the analytical methods available to this project were such that the analysis of PAN turned out to be somewhat more difficult and uncertain than the analysis of NO2. The second is that in the absence of NOx the acetyl peroxy radical reacts with HO2 to some extent to form O3, so O3 yields do not provide as sensitive a secondary test for NOx sources than would be the case if it were absent. However, this would be less of a problem if the NOx sources in the experiment were relatively large, which would result in the O3 formed from the NOx reactions being more important than those formed from acetyl peroxy radicals..

9

Because of this, we also carried out NOx source experiments designed so that NOx sources primarily result in the formation of NO2, with formation of HNO3 and other NOy species being minimized. Note that this requires methods to measure NO2 in the absence of other interferences such as occur with commercial NO-NOx analyzers that measure NO2 by first converting NO2 into NO and measuring the chemiluminescence of reaction NO + O3, since these analyzers respond to many NOy species such as organic nitrates as if they were NO2. Two independent methods (Cavity Enhanced Absorption Spectroscopy and Gas Chromatography - Luminol detection) were available to measure NO2 without such interferences, and these were used for this project. Model calculations indicate that CO - H2O2 experiments at appropriate levels should result in NO2 being the dominant form of NOx that is formed. In this chemical system, H2O2 acts as a radical source, and CO at a high concentration (e.g., 25 ppm) quickly converts OH into HO2 and limits HNO3 formation from reaction between OH and NO2.

Model calculations using the SAPRC-07 mechanism were used to design the experimental conditions for the acetaldehyde - H2O2 and CO - H2O2 so that the dominant NOx species in the NOx source experiments is either NO2 or PAN (NO2 captured in the form of PAN). The conditions chosen based on the results of these calculations were ~0.5-1 ppm acetaldehyde and ~0.5 ppm H2O2 for the experiments with acetaldehyde, and ~50-100 ppm CO and ~0.5 ppm H2O2 for the experiments with CO. The specific levels are not critical as long as they are in these approximate ranges.

Isoprene Experiments Isoprene is an important biogenic compound whose mass emissions dominate other VOCs on a continental scale (Guenther et al., 2006), and whose reactions and those of its oxidation products, can have significant effects on O3 formation and NOx recycling in multi-day episodes. A number of environmental chamber experiments with isoprene have been carried out and used to develop and evaluate the existing mechanisms for this compound (e.g., Paulson and Seinfeld, 1992; Carter and Atkinson, 1996, Carter, 2010a), but most have been carried out at relatively high reactant concentrations that are not representative of conditions where isoprene is actually emitted. Recently, Azzi et al., (2010) reported isoprene - NOx experiments in the CSIRO chamber where the O3 formation rates tended to be consistently underpredicted by mechanisms that performed well in simulating O3 in higher concentration experiments (Azzi et al, 2010; Carter, 2010b). However, the artificial light source employed at CSIRO (blacklight blue) has low UV intensity compared to sunlight or conventional blacklights and this may be contributing to the discrepancy. Therefore, there has been a need to obtain additional mechanism evaluation data for isoprene experiments in chambers with more realistic reactant concentrations and light sources. For this reason, we included in this project a few isoprene - NOx irradiations in the UCR EPA chamber at lower NOx levels using a blacklight source that gives a better representation of UV intensity compared to sunlight than the light source used by Azzi et al. (2010).

Control and Characterization Experiments In order for environmental chamber experiments to be suitable for mechanism evaluation, it is necessary to also conduct control and characterization experiments to assure validity of the data and to methodologies and to derive chamber dependent parameters needed for

10

modeling. The control and characterization experiments carried out for this project are as follows.

1. Ethene - NOx and propene - NOx irradiations were control experiments to evaluate the base case experiments for the NOx sink study, and to assure that equivalent results are obtained in simultaneous irradiations in each of the dual reactors if the same mixtures are irradiated. It is also important to assure that the base case experiment can be adequately be simulated by current mechanisms for them to be useful for evaluating mechanisms for the effects of adding the test compound. Propene - NOx experiments had the additional utility of providing data to evaluate the measurement method used for PAN, since pure PAN samples of known concentrations were not available for calibration purposes, modeling of previous propene experiments shows that current mechanisms can give reasonably good simulations of PAN.

2. Acetaldehyde - H2O2 - NOx irradiations were carried out as control experiments to evaluate the use of PAN measurements in the acetaldehyde - H2O2 - test VOC NOx source experiments to measure the amount of NOx produced. In these experiments the NOx was introduced in known quantities, and the measured PAN formed should agree with the initial NOx.

3. A biacetyl - CO - NOx irradiation was carried out to evaluate the use of this system in place of acetaldehyde + H2O2 as an alternative NOx source experiment based on PAN measurements. The results of this and the acetaldehyde - H2O2 - NOx control irradiations indicated that use of the acetaldehyde - H2O2 system is sufficient, so this alternative was not explored further.

4. CO - H2O2 irradiations were carried out to measure background NOx offgasing rates (NOx supplied by chamber-dependent processes such as wall reactions; e.g., refer to Carter et al, 2005c) in our chamber experiments. These experiments also serve as controls for the CO - H2O2 - VOC experiments used to measure NOx sources. Evaluating this CO - H2O2 chemical system under chamber conditions is critical to interpreting and modeling the results of the NOx source experiments, as well as a necessary input when modeling all chamber experiments. For these experiments, determining the NOx offgasing rate that gives the best fits to the O3 yields provides the most sensitive method to obtain the amount of NOx offgasing, though measurements in formation of NO2 or changes in total NOx provide more direct measures that are generally consistent with the rates that fit the O3 data.

5. Acetaldehyde - H2O2 irradiations were also carried out to measure background NOx offgasing rates and also served as controls for the acetaldehyde - H2O2 - VOC experiments to measure NOx sources. In this case, the amounts of PAN formed provide the measure of NOx offgasing. Modeling O3 formation is not useful in these cases because most of the O3 formed in these experiments is calculated to come from the reaction with HO2 of peroxyacetyl radicals formed in this system in the absence of NOx (IUPAC, 2006).

6. CO - NOx irradiations were carried out were carried out to measure the chamber-dependent radical source during the experiments carried out for this project, which is a chamber-dependent factor that must be taken into account when modeling environmental chamber experiments (Carter et al, 1982; Carter et al., 2005b).

11

7. NO2 actinometry experiments were carried out to measure the NO2 photolysis rates in our chamber experiment, which provide the primary measure of light intensity.

As discussed previously (Carter, 2004; Carter et al, 2005b), the background NOx offgasing and the chamber radical source being measured by some of these characterization experiments are believed to be a result of the same process, which is a light-induced offgasing of HONO from the chamber walls. HONO photolyzes rapidly to form OH radicals that can serve as the chamber radical source, and has been observed to be formed in other large chambers at rates comparable to the NOx offgasing and radical source rates observed in our chamber (Carter et. al., 2005b). In addition, NOx offgasing rates and the chamber radical source rates derived from the characterization results in this and other Teflon film reactors generally have similar magnitudes.

The results of the NO2 actinometry measurements are used to assign an NO2 photolysis rate to be used when modeling these experiments. This assigned NO2 photolysis rate, combined with the spectral distribution of the chamber lights taken previously (Carter et al, 1995b) and the absorption cross sections and quantum yields associated with the mechanism, can then be used for calculating the photolysis rates when modeling the chamber experiments.

Light Source Employed All the experiments for this project were carried out in an indoor environmental chamber because indoor chamber experiments are much more reproducible and more straightforward to characterize for modeling. They also allow for increased productivity because the ability to conduct experiments does not depend on the weather or the time of day. However, indoor chambers have the disadvantage of requiring use of an artificial light source, which necessarily have spectral differences from sunlight. Spectral differences can be taken into account for photolysis reactions with known cross sections and quantum yields. However, because cross sections and quantum yields are poorly characterized for some photolysis reactions it is important that the light source used in environmental chamber experiments approximately represent the spectrum of sunlight in the wavelength region affecting photolysis rates.

Indoor environmental chamber experiments used previously for mechanism evaluation purposes have employed both blacklights and arc lights to simulate sunlight, and generally consistent mechanism evaluation results have been obtained in both cases, at least for evaluations of the more recent versions of the SAPRC mechanisms (Carter, 2000, 2010a,b). Representative spectra of these two types of light sources are shown on Figure 3, where they are compared with a representative spectrum of sunlight. As shown on Figure 3, arc lights give the best representation of the solar spectrum throughout the entire wavelength range, but blacklights also give an adequate representation in the important UV spectral range that affects most photolysis reactions of non-aromatic species. Although evaluations of experiments with aromatics may be most affected when blacklights are used because they contain a number of photoreactive products with uncertain action spectra that may photolyze in the higher wavelength region where blacklights are deficient, evaluations of aromatic experiments using the SAPRC-07 mechanism show no significant differences in model performance with these two different types of light sources (Carter, 2010b).

12

Figure 3. Spectrum of the light sources used with the UCR EPA environmental chamber, with representative solar spectrum also shown for comparison. The relative intensities in the spectra are normalized to give the same NO2 photolysis rate.

If there is a choice, then clearly use of arc lights would be preferred for indoor chamber experiments. However, arc lights are extremely expensive to acquire and expensive and difficult to operate, while blacklights are relatively inexpensive and easy to operate, and are used in many atmospheric chemistry laboratories. For non-aromatic compounds or systems where the major photolysis reactions have reasonably well known action spectra, use of blacklights is considered sufficient. For this reason, most of the chamber experiments currently available in our database of chamber experiments for mechanism evaluation consist of experiments utilizing blacklight light sources (Carter, 2010a,b). However, the availability of arc-light experiments is an important and necessary complement to blacklight experiments, and experiments with different light sources provide a useful resource for evaluating how well mechanisms represent action spectra of the major photoreactive species.

The UCR EPA chamber employed in this study has both types of light sources, and their spectra are those shown on Figure 3. Although initially the chamber only had the arc light, because of the cost and difficulty in operating it we constructed banks of blacklights to be used for experiments where blacklights are judged to be sufficient. This was considered to be the case for most of the experiments carried out in this chamber, and the arc light system had not been operated since early 2008, and was in need of repairs at the time this project was initiated.

Arc lights were the preferred light source for this project and efforts were made to make the arc light operational. Unfortunately, one problem followed another in our attempts to repair this system and eventually we ran out of time and resources available to this project. Therefore, all experiments for this project were carried out using blacklights. While this was not what we had preferred, we believe the blacklight experiments we conducted addressed the objectives for this project. This is because we can correct for the differences when modeling the experiments, and because previous evaluations do not indicate strong light source effects. The

0.0

0.2

0.4

0.6

0.8

1.0

0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60Wavelength (nm)

Rel

ativ

e in

tens

ity(p

hoto

ns p

er w

avel

engt

h ra

nge) Arc Light

BlacklightsSolar Z=0

0.0

0.1

0.2

0.30 0.33

13

lower cost of operating the blacklights also permitted a larger number of useful mechanism evaluation experiments to be completed with the resources available.

Instrumentation Standard monitoring instrumentation was employed as used in previous EPA chamber experiments. The instrumentation used to obtain the primary data used for mechanism evaluation is briefly summarized below.

1. O3 was monitored using a commercial UV absorption ozone analyzer, calibrated using standard methods used for ambient air quality monitoring stations. This instrument is subject to positive interferences from certain oxidized aromatic species such as cresols and nitrophenols, and this needs to be taken into account when using O3 data to evaluate mechanisms for these compounds.

2. Organic reactants were monitored by gas chromatography with flame ionization detection (GC-FID). The relatively volatile VOCs such as ethene and toluene were sampled by online loop analysis and the GC-FID response factors were derived from GC span analysis using GC standards or calculated amounts of the VOCs injected into the reactors (e.g., for furan). The less volatile VOCs (specifically o-cresol and 2-nitrophenol) were analyzed by absorption onto cartridges and subsequent desorption onto the GC, and were calibrated by placing known solutions of the materials on the loops prior to analysis.

3. NO and NOy, defined as NO plus the sum of NO2 and other nitrogen-containing species converted to NO with a heated catalytic converter, was monitored by a commercial NO-NOx analyzer. The particular analyzer employed is an unusually sensitive one that we acquired from the California Air Resources Board (CARB) after we intercompared a number of such instruments for a previous CARB project. It is calibrated for NO using standard NO mixtures and

4. a dilution calibration system and the efficiency of the converter is checked by gas-phase titration (GPT) by adding known amounts of O3 to an excess of known amounts of NO in a calibration system. The limitations and utility of the NOy measurements are discussed further below.

5. NO2 was monitored by Cavity Enhanced Absorption Spectroscopy (CEAS), using an instrument that was interfaced to this chamber and operated as part of this project. The NO2 signal was calibrated using the same GPT method as used to test the converter efficiency of the NOy analysis discussed above. The utility and limitations of the data from this instrument are discussed further below.

6. NO2 was also monitored by a GC-Luminol system that is being developed by Fitz Aerometric Technologies and made available for use with this project. It operates by separating NO2 from PAN and other nitrogen-containing species using a GC column and detecting the species using a luminol-based wet chemical method. It also obtains data for peroxy acetyl nitrate (PAN) and can potentially obtain data for other PAN analogues if they can be separated from PAN on the GC column, though as presently operated it only reports concentrations of PAN. The NO2 signal is calibrated using the same GPT as used for the CEAS and the NO and NOy analysis. The PAN signal is calibrated by assuming it has the same response factor as NO2, which had been verified previously. The limitations and utility of this instrument for NO2 and PAN analysis are discussed further below.

14

7. Temperature was monitored by shielded and calibrated thermocouples inside the sampling lines.

8. Light intensity was qualitatively monitored using a light radiation sensor, and more quantitatively by conducting periodic NO2 actinometry experiments in conjunction with some of the experiments.

Experiments Performed A total of 33 dual reactor environmental chamber experiments were carried out in the UCR EPA chamber, of which 29 obtained data of potential utility for modeling, for either mechanism evaluation or chamber characterization. Because of the dual reactor design, each successful experiment provides data for two separate reactor irradiations, each of which can be treated as a separate experiment for modeling purposes. Modeling input and experimental output data were obtained for a total of 55 such reactor irradiations (runs), as summarized in Table 2.

Table 2. Numbers of experiments of different type performed using the EPA chamber at UCR. Number of Runs Type

20 NOx Sink with base case (10 test runs, each with a base case) 4 NOx sink without base case (test reactant accidentally injected into both reactors) 8 NOx Source (two methods) 3 Isoprene - NOx, with varied initial concentrations (one duplicate experiment excluded) 6 Ethene or Propene - NOx Control (two duplicate experiments excluded) 3 NOx source control (NOx added) 9 Background NOx characterization (two methods) 2 Radical source characterization

Tables below summarize the results of the various experiments considered useful for modeling and figures show the experimental data that can be used for comparison with model predictions. Model simulations are shown for the control and characterization runs because modeling these runs was used as part of the data processing, to assess run-to-run consistency, and derive appropriate assignments of characterization parameters and initial reactant concentrations for modeling input files. These calculations were done using the SAPRC-07 chemical mechanism as documented by Carter, 2010a.

NOx Sink Experiments NOx sink experiments were carried out for toluene, o-cresol, furan (a precursor to the aromatic fragmentation product 2-butene-1,4-dial), and isoprene, using ethene - NOx as the base case experiment in all cases, and also using propene - NOx as the base case for toluene.

Table 3 gives a summary of the conditions and results of these experiments, and concentration time plots are shown in Figure 4 for the toluene runs, Figure 5 for the runs with o-cresol or furan, and Figure 6 for the runs with isoprene. These figures also show the data from the base case experiment on the same plot, except for the two experiments (EPA1403 and EPA1444) where there are no base case data. In those cases, the initial reactant concentrations and results were very close on both reactors, and the results for the two reactors are also shown. Different symbols are used for the NO2 data obtained using the CEAS vs. the GC-Luminol instruments, to show the consistencies or differences between the data from these

15

instruments, and to indicate which was used where there is only one type of NO2 measurement. The figures also include SAPRC-07 model simulations of the base case experiments, where applicable.

Table 3. Summary of NOx sink experiments carried out for this project. Run [a]

Note [b]

Test VOC NOx Base [c] Hours [d] Maximum O3 [e] (ppb) (ppb) (ppm) Test Base Test Base Change

Propene - NOx used as Base Case Experiment

EPA1418 Toluene 301 16 0.30 6.5 4.0 112 139 -27 EPA1436 Toluene 391 14 0.43 5.5 3.0 109 129 -20 EPA1443 Toluene 407 18 0.31 8.0 5.5 120 149 -29

Ethene - NOx used as Base Case Experiment

EPA1401 1,2 Toluene 367 14 1.10 3.0 1.0 165 See Note 2 EPA1407 2 Toluene 344 14 0.84 5.5 3.5 153 >240 <-87 EPA1408 o-Cresol 98 16 1.00 5.5 6.0 116 291 -174 EPA1402 1 Furan 129 13 1.10 4.5 4.5 149 267 -118 EPA1448 2 Furan 130 15 0.88 7.0 4.5 149 >199 <-50 EPA1403 1,3 Furan 47 13 0.98 4.5 2.5 196 See Note 3 EPA1404 1,2 Isoprene 51 14 1.14 5.5 3.5 178 >264 <-86 EPA1446 1,2 Isoprene 50 14 0.92 6.0 3.0 171 >237 <-66 EPA1444 3 Isoprene 17 14 0.80 8.0 4.0 226 See Note 3 [a] The runs judged to have the highest quality for modeling for each type are indicated by underlines. [b] Notes for experiments

1 NOx injected in the form of NO2. (NOx injected in the form of NO for all other runs.) 2 Base case run ended too early to give true O3 maximum. In the case of run 1401, the run ended too early to indicate

any effect of the test compound on O3. In the other cases, the final O3 for the base case is given, which is a lower limit of the true O3 maximum.

3 Test compound (furan or isoprene) mistakenly injected in both reactors, so there is no control experiment.

[c] Amount of propene or ethene injected into both reactors.

[d] "Hours" refers to the number of hours for which O3 data are available, truncated to the nearest half hour. "Test" refers to the reactor with the added test VOC, which was always Reactor A, and "base" refers to the reactor with only ethene or propene and NOx, which was always Reactor B. Because of a persistent leak in Reactor B, the experiments usually ended earlier in that reactor.

[e] The maximum O3 concentration in the experiment if the experiment attained a "true" O3 maximum (i.e., was not still increasing when the run ended), and “Change” is Max O3 (Test) - Max O3 (Base). If a true maximum was not attained, as was the case in several base case experiments, then the Base maximum O3 is given as a lower limit of Max O3 and the change in Max O3 is given as a lower limit of the change in the maximum O3 (i.e., the actual change in Max O3 can be larger).

16

Irradiation Time (minutes)

Figure 4. Plots of selected results of NOx sink experiments with toluene. (Model calculations are shown for the base case.)

EPA1418 Toluene EPA1436 Toluene EPA1443 Toluene EPA1401 Toluene EPA1407 TolueneOzone (ppm)

Toluene (ppm)

Propene (ppm) Ethene (ppm)

NO2 (CEAS) (ppm)

PAN (ppm)

PM formed (µg/m3)

0.00

0.04

0.08

0.12

0.16

0 120 240 360

Test ExptBase Expt

0.0

0.1

0.2

0.3

0.4

0 120 240 360

No experimental data

0.0

0.1

0.2

0.3

0.4

0 120 240 360

0.000

0.005

0.010

0.015

0 120 240 360

0.000

0.002

0.004

0.006

0.008

0.010

0 120 240 360

0

1

2

3

0 120 240 360

0.00

0.05

0.10

0.15

0 120 240 360

Base Calc

0.0

0.1

0.2

0.3

0.4

0.5

0 120 240 360

0.0

0.1

0.2

0.3

0.4

0.5

0 120 240 360

0.000

0.005

0.010

0.015

0 120 240 360

0.000

0.002

0.004

0.006

0.008

0.010

0 120 240 360

0.0

0.4

0.8

1.2

0 120 240 360

0.00

0.04

0.08

0.12

0.16

0 120 240 360 480

0.0

0.1

0.2

0.3

0.4

0.5

0 120 240 360 480

0.0

0.1

0.2

0.3

0.4

0 120 240 360 480

0.000

0.005

0.010

0.015

0.020

0.025

0 120 240 360 480

0.000

0.004

0.008

0.012

0 120 240 360 480

0

1

2

3

4

5

0 120 240 360 480

0.00

0.04

0.08

0.12

0.16

0 60 120 180

0.0

0.1

0.2

0.3

0.4

0 60 120 180

0.0

0.4

0.8

1.2

0 60 120 180

0.000

0.004

0.008

0.012

0.016

0 60 120 180

0.000

0.001

0.002

0.003

0 60 120 180

0.0

0.2

0.4

0.6

0.8

0 60 120 180

0.00

0.05

0.10

0.15

0.20

0.25

0 120 240 360

0.0

0.1

0.2

0.3

0.4

0 120 240 360

0.0

0.2

0.4

0.6

0.8

1.0

0 120 240 360

0.000

0.004

0.008

0.012

0.016

0 120 240 360

0.0

0.4

0.8

1.2

1.6

0 120 240 360


17

Figure 5. Plots of selected results of NOx sink experiments with o-cresol and furan. (Model calculations are shown for the base case.)

EPA1408 o-Cresol EPA1402 Furan EPA1448 Furan EPA1403 Furan * Ozone (ppm)

o-Cresol (ppm) Furan (ppm)

Ethene (ppm)

NO2 (CEAS) (ppm) NO2 (Luminol) (ppm) NO2 (CEAS) (ppm)

PM formed (µg/m3)

Irradiation time (minutes)

0.0

0.1

0.2

0.3

0 120 240 360

Test ExptBase Expt

0.00

0.04

0.08

0.12

0 120 240 360

0.0

0.4

0.8

1.2

0 120 240 360

0.000

0.004

0.008

0.012

0 120 240 360

0

5

10

15

20

0 120 240 360

0.00

0.05

0.10

0.15

0.20

0.25

0 120 240

Base Calc

0.00

0.04

0.08

0.12

0.16

0 120 240

0.0

0.4

0.8

1.2

0 120 240

0.000

0.004

0.008

0.012

0 120 240

0.00

0.04

0.08

0.12

0 120 240

0.00

0.05

0.10

0.15

0.20

0.25

0 120 240

Test Expt B

* This run had furan in both

reactors

0.00

0.02

0.04

0.06

0 120 240

0.0

0.4

0.8

1.2

0 120 240

0.000

0.004

0.008

0.012

0.016

0 120 240

0.000

0.002

0.004

0.006

0 120 240

0.00

0.05

0.10

0.15

0.20

0.25

0 120 240 360 480

0.00

0.04

0.08

0.12

0.16

0 120 240 360 480

0.0

0.2

0.4

0.6

0.8

1.0

0 120 240 360 480

0.000

0.004

0.008

0.012

0 120 240 360 480

0.00

0.02

0.04

0.06

0 120 240 360 480

18

Figure 6. Plots of selected results of NOx sink experiments with isoprene. (Model calculations are shown for the base case.)

In all cases studied, it was found that the addition of the test compound reduced O3 formation, though in some cases the full effect could not be determined from the run data alone because the base case experiment ended before attaining a true ozone maximum. Except for toluene +

EPA1404 Isoprene EPA1446 Isoprene EPA1444 Isoprene * Ozone (ppm)

Isoprene (ppm)

Ethene (ppm)

NO2 (CEAS) (ppm) NO2 (ppm)

PAN (ppm)

PM formed (µg/m3)


0.00

0.05

0.10

0.15

0.20

0.25

0 120 240 360

Test ExptBase ExptBase Calc

0.00

0.02

0.04

0.06

0 120 240 360

No Side A Data

0.0

0.4

0.8

1.2

0 120 240 360

0.000

0.004

0.008

0.012

0.016

0 120 240 360

0.00

0.04

0.08

0.12

0 120 240 360

0.00

0.05

0.10

0.15

0.20

0.25

0 120 240 360

0.00

0.02

0.04

0.06

0 120 240 360

0.0

0.2

0.4

0.6

0.8

1.0

0 120 240 360

0.000

0.004

0.008

0.012

0.016

0 120 240 360

0.00

0.04

0.08

0.12

0.16

0 120 240 360

0.00

0.05

0.10

0.15

0.20

0 120 240 360 480

Test Expt B

* This run had isoprene in

both reactors

0.000

0.004

0.008

0.012

0.016

0.020

0 120 240 360 480

0.0

0.2

0.4

0.6

0.8

1.0

0 120 240 360 480

0.000

0.004

0.008

0.012

0 120 240 360 480

No Valid Data

0.00

0.02

0.04

0.06

0.08

0.10

0 120 240 360 480

0.000

0.001

0.002

0.003

0 120 240 360

No data No data

19

ethene run EPA1401, the base case run was probably sufficiently close to the true O3 maximum to provide useful data for evaluating the effects of NOx sinks on the maximum O3.

In terms of percentage change in maximum O3 relative to the base case ((Test – Base)/Base*100%) for the highest quality run, toluene caused a 19% reduction in the propene experiment (EPA1443) and a ≥36% reduction in the ethene experiment (EPA1407), and o-cresol, furan, and isoprene caused 60%, 44%, and ≥28%, reductions, respectively, in the ethene experiments (EPA1408, EPA1402 and EPA1446, respectively). The toluene results demonstrate a greater sensitivity of the ethene system, as expected based on the calculations done when designing the NOx sink experiments. Ranking these NOx-sink compounds in terms of their strength of reducing the maximum O3 is not straightforward because the chamber experiments were designed to maximize the possibility of successfully obtaining experimental evidence of decreases in the maximum O3 by these NOx sinks and the injected base and test VOC levels were adjusted for this purpose. Note that the initial concentrations of the test VOC (toluene, o-cresol, furan or isoprene in Table 3) were also designed to see measurable NOx sink effects in terms of measured ozone (and NO2 and PAN). The injected amount of the test VOC affects the magnitude of maximum O3 reduction although the effect may not be linear. For example, when we assume the maximum O3 of EPA1402 (Base) can be used in combination of the maximum O3 of EPA1403 (Test), addition of ~50 ppb furan resulted in 27% reduction in the maximum O3 for EPA1403, which is quite different from ~45% reduction shown by ~130 ppb furan addition for EPA1402. However, when we compare experimental data for experiments where ethene and NOx were injected in their base experiments, o-cresol showed a stronger NOx sink effect than toluene, furan and isoprene.

Most NOx sink experiments had data for NO2 that are useful for modeling, and the NO2 data from most of the base case experiments were generally consistent with model predictions. The CEAS NO2 data are judged to be somewhat less subject to interferences and have fewer experimental issues, so are preferred for modeling if data from both the CEAS and GC-Luminol instruments are available, and are therefore shown on the figures except for EPA1402, which did not have useable CEAS data. Runs 1401, 1404, 1418, and 1436 also had GC-Luminol data that were consistent with the CEAS data, and could probably be used just as well. The other experiments either did not have GC-Luminol NO2 data, or the data were judged to be affected by O3 interference and were rejected.

Only a few NOx sink experiments had data for PAN that may be useful for modeling, three with toluene (two with only final yields in the test reactor), and one with isoprene. The PAN formed in the propene - NOx base case experiment EPA1443B was about half model prediction, which is not as good a fit as expected based on modeling previous propene experiments. This should be considered when using the PAN data from these experiments for mechanism evaluation.

NOx Source Experiments NOx source experiments were carried out with isopropyl and isobutyl nitrates and 2-nitrophenol, in all three cases using both VOC - acetaldehyde - H2O2 and VOC - CO - H2O2 irradiations. Associated with these were several control experiments, in which the test compound was not added, to determine the amount of NOx that might be generated from chamber background effects in the absence of the NOx source VOC. The conditions and results of these experiments are summarized in Table 4, and concentration-time plots of the data from

20

the mechanism evaluation and control experiments are shown on Figure 7 through Figure 9. SAPRC-07 model calculations are shown in the figures showing the results of the control experiments on Figure 9.

Table 4 shows that the levels of NOx species measured in the NOx source experiments with the alkyl nitrate or nitrophenol test compounds were in all cases significantly higher than the measured NOx species in the characterization experiments where the test compounds was absent. The amounts of O3 formed in the experiments with the test compounds were also higher, though this is uncertain in the experiments with 2-nitrophenol because of nitrophenol interference with the O3 analysis. Therefore, these experiments were successful in showing that these compounds indeed release measurable amounts of NOx when they react, and background NOx offgasing effects should not significantly affect interpreting and modeling the results.

Figure 7 and Figure 8 show plots of the data for the alkyl nitrate and the 2-nitrophenol experiments, respectively. In the case of the 2-nitrophenol experiments shown on Figure 8, the O3 data are not suitable for quantitative mechanism evaluation because of the strong interference by 2-nitrophenol on the UV absorption O3 analysis. Correcting the O3 data by using the nitrophenol data and the O3 signal at the start of the experiment when it is due entirely to the interference is not recommended, since the nitrophenol oxidation products may also have interferences on the O3 analysis. However, the O3 signal decreases at the start of the acetaldehyde experiments at a rate comparable to the nitrophenol consumption rate, suggesting that nitrophenol oxidation products may cause smaller interferences than nitrophenol. The increase in O3 during the later stages of the acetaldehyde experiments, or throughout the CO experiments, suggests that O3 is indeed formed in the nitrophenol experiments, as would be expected since measurable NOx species are also formed.

All the alkyl nitrate and nitrophenol NOx sink experiments have NO2 and PAN data of potential utility for model evaluation, at least for the purpose of determining upper limits when measurable amounts are not expected to be formed. All of the acetaldehyde experiments had usable PAN data, and all the CO experiments had PAN data that showed that the PAN formed was at or below the detection limit, as expected. All of the CO experiments had usable CEAS NO2 data to serve as measures of NOx sources in these experiments. The NO2 as measured by CEAS was below the detection limit in the acetaldehyde experiments with the alkyl nitrates, as expected since most of the NOx should be converted to PAN in those experiments. However, the CEAS instrument was sufficiently sensitive to detect and quantify the NO2 in the acetaldehyde experiment with 2-nitrophenol, though as expected it was about an order of magnitude lower than the amount of PAN that was formed. No valid GC-Luminol data were obtained for the isopropyl nitrate or the 2-nitrophenol experiments, but they were available in the experiments with isobutyl nitrate, and were sufficiently sensitive to quantify NO2 in the acetaldehyde run, where again NO2 was about an order of magnitude less than the PAN formed. Thus, these NO2 data can serve as an additional means for mechanism evaluation when modeling at least some of the VOC - acetaldehyde - H2O2 experiments.

21

Table 4. Summary of NOx source and related chamber characterization experiments.

Run Test Compound Initial Conc (ppm)

Hours Final or Max Yield. (ppb)

Compound (ppm) H2O2 CO CH3CHO O3 NO2 PAN [b]NOy

VOC - CH3CHO - H2O2 Experiments (PAN data test NOx sources)

EPA1449A Isopropyl nitrate 0.50 0.51 0.49 7 77 ~6 EPA1439A Isobutyl nitrate 0.38 0.51 1.00 9 100 ~1 23 EPA1441A 2-Nitrophenol 0.109 0.51 0.53 5 <157 [a] ~3 60 EPA1442A 2-Nitrophenol 0.111 0.51 0.42 8 <246 [a] ~5 45 EPA1415B NO (control) 0.025 1.02 0.66 4 123 no data EPA1447A NO2 (control) 0.013 0.51 0.40 6 101 14

VOC - CO - H2O2 Experiments (NO2 data test NOx sources)

EPA1449B Isopropyl nitrate 0.51 0.51 24 5 248 10 [d] EPA1439B Isobutyl nitrate 0.37 0.51 141 3 115 ~5 [d] EPA1441B 2-Nitrophenol 0.109 0.51 127 3 <305 [a] 10 [d] EPA1442B 2-Nitrophenol 0.111 0.51 25 5 <338 [a] 10 [d]

CH3CHO - H2O2 Experiments (PAN and ∆[NOy] data test background NOx offgasing)

EPA1428A 0.51 0.34 8 6 no data ~0.6 EPA1428B 0.51 0.34 5 4 no data ~0.5 EPA1434A 0.51 0.52 8 10 no data ~0.8 EPA1434B 0.51 0.52 6 8 no data ~0.8 EPA1447B 0.51 0.40 3 3 ~0.6 ~0.4

CO - H2O2 Experiments (NO2 and ∆[NOy] data test background NOx offgasing)

EPA1429A 0.51 45 6 15 [c] ~0.6 EPA1429B 0.51 45 4 7 [c] ~0.4 EPA1431A 0.51 49 6 0 ~0.4 ~0.4 EPA1431B 0.51 49 4 6 ~0.4 ~0.4 [a] 2-Nitrophenol has a significant positive interference in the O3 analyzer. The O3 data therefore must be considered only

valid as upper limits. [b] Data are highly scattered in the 0 - 1 ppb range [c] Data too scattered to provide useful information other than NO2 being less than ~3 ppb. [d] Below or near the detection limit for PAN.

22

Figure 7. Plots of selected results of NOx source experiments with the organic nitrates.

Isopropyl Nitrate Isobutyl Nitrate EPA1449A (Aldehyde) EPA1449B (CO) EPA1439A (Aldehyde) EPA1439B (CO)

Ozone (ppm)

NO2 (ppm)

PAN (ppm)

Isopropyl Nitrate (ppm) Isobutyl Nitrate (ppm)

Acetaldehyde (ppm) Acetaldehyde (ppm)


0.00

0.02

0.04

0.06

0.08

0.10

0 120 240 360 480

0.000

0.001

0.002

0.003

0 120 240 360 480

0.000

0.005

0.010

0.015

0.020

0.025

0 120 240 360 480

0.0

0.1

0.2

0.3

0.4

0 120 240 360 480

0.0

0.4

0.8

1.2

0 120 240 360 480

0.00

0.04

0.08

0.12

0.16

0 60 120 180 240

0.000

0.002

0.004

0.006

0.008

0 60 120 180 240

0.0000

0.0001

0.0002

0.0003

0 60 120 180 240

0.0

0.1

0.2

0.3

0.4

0 60 120 180 240

0.00

0.02

0.04

0.06

0.08

0 120 240 360

Expt.

0.000

0.001

0.002

0.003

0 120 240 360

0.000

0.004

0.008

0.012

0 120 240 360

0.0

0.2

0.4

0.6

0 120 240 360

0.0

0.2

0.4

0.6

0 120 240 360

0.00

0.06

0.12

0.18

0.24

0 60 120 180 240 300

0.000

0.004

0.008

0.012

0 60 120 180 240 300

CEASLuminol

0.000

0.001

0.002

0.003

0 60 120 180 240 300

0.0

0.2

0.4

0.6

0 60 120 180 240 300

23

Figure 8. Plots of selected results of NOx source experiments with 2-nitrophenol.

Table 4 and Figure 9 show the data for the acetaldehyde - H2O2 - NOx control experiments, and results of SAPRC-07 model calculations are also shown on Figure 9. The data for EPA1415B (for details, refer to Appendix C) were not useful for evaluating the NOx source measurement because no valid PAN data were obtained due to inadequate sampling times, but the NO2 data from the CEAS and the GC-Luminol instrument were reasonably consistent with each other and model predictions. On the other hand, EPA1447A provided useful data to evaluate the use of

2-Nitrophenol - Acetaldehyde - H2O2 runs 2-Nitrophenol - CO - H2O2 runsEPA1441A EPA1442A EPA1441B EPA1442B

Ozone (ppm) (Note: ozone data are upper limits because of nitrophenol interference)

NO2 (CEAS) (ppm)

PAN (ppm)

2-Nitrophenol - Acetaldehyde - H2O2 runs (ppm)

Acetaldehyde (ppm)


0.00

0.05

0.10

0.15

0.20

0.25

0 60 120 180 240 300

Expt.

0.000

0.001

0.002

0.003

0.004

0.005

0 60 120 180 240 300

0.00

0.02

0.04

0.06

0.08

0 60 120 180 240 300

0.00

0.04

0.08

0.12

0 60 120 180 240 300

0.0

0.2

0.4

0.6

0 60 120 180 240 300

0.0

0.1

0.2

0.3

0.4

0 60 120 180

0.000

0.005

0.010

0.015

0 60 120 180

0.0000

0.0001

0.0002

0.0003

0.0004

0.0005

0 60 120 180

0.00

0.04

0.08

0.12

0 60 120 180

No data this reactor

0.0

0.1

0.2

0.3

0 120 240 360 480

0.000

0.002

0.004

0.006

0 120 240 360 480

0.00

0.01

0.02

0.03

0.04

0.05

0 120 240 360 480

0.00

0.04

0.08

0.12

0 120 240 360 480

0.0

0.1

0.2

0.3

0.4

0.5

0 120 240 360 480

0.0

0.1

0.2

0.3

0.4

0 60 120 180 240 300

0.000

0.005

0.010

0.015

0 60 120 180 240 300

0.0000

0.0001

0.0002

0.0003

0 60 120 180 240 300

0.00

0.04

0.08

0.12

0 60 120 180 240 300


24

PAN data to measure the amount of NOx present in these acetaldehyde - H2O2 experiments, and the amount of measured PAN agreed reasonably well with the amount of NOx injected. But the GC-Luminol measurements showed the PAN as having a maximum of about 14 ppb at hour 2 and declining to about 10 ppb by the end of the experiment, while the model predicted it should be constant from about hour 2 onwards. It is unclear without additional experiments whether this is an experimental or a model issue, but we suspect it is the former.

The results of the acetaldehyde - H2O2 and CO - H2O2 characterization experiments listed in Table 4 were used to derive the chamber effects parameters related to NOx offgasing. Results of model calculations using the SAPRC-07 mechanism with the chamber effects model assigned for these experiments are also shown on these figures. As discussed above, the NOx offgasing appears to be relatively minor compared to NOx formed from the VOC reactions in the NOx source experiments, but the chamber effects model still needs to have appropriate representations.

Figure 9. Plots of selected results of the acetaldehyde - H2O2 - NOx control experiments carried out for this project. (Results of the model calculations are also shown.)

Isoprene Experiments Three isoprene - NOx experiments were carried out during the course of this project, and the initial conditions and selected results are summarized on Table 5, and plots of data useful for mechanism evaluation are shown on Figure 10. These experiments employed varying isoprene and NOx levels and carried out at lower concentrations than other isoprene experiments for mechanism evaluation, except for the low-UV CSIRO experiments discussed above. Although the number of experiments is limited, they represent an important complement to the data set of CSIRO isoprene experiments with the low UV light source, and to the isoprene experiments from other chambers carried out at higher concentration ranges.

EPA1415B EPA1447AOzone (ppm) Acetaldehyde (ppm) Ozone (ppm) Acetaldehyde (ppm)

PAN (ppm) NO2 (ppm) PAN (ppm) NO2 (ppm)


0.00

0.04

0.08

0.12

0.16

0 60 120 180 240

DataCalc

0.0

0.2

0.4

0.6

0.8

0 60 120 180 240

0.000

0.004

0.008

0.012

0 60 120 180 240

CEASLuminol

0.00

0.01

0.02

0.03

0 60 120 180 240

No Data

0.00

0.04

0.08

0.12

0 120 240 3600.0

0.1

0.2

0.3

0.4

0.5

0 120 240 360

0.000

0.005

0.010

0.015

0 120 240 3600.000

0.004

0.008

0.012

0.016

0 120 240 360

25

All three of these experiments achieved true O3 maxima, though O3 formation occurred at different rates in all cases. Two of these experiments had NO2 data from both the CEAS and GC-Luminol instruments, and in those cases the two measurements were in reasonably good agreement. However, there were no data available for PAN or any of isoprene's oxidation products. Therefore these data will be useful for mechanism evaluation for model predictions of O3, NO, NO2, and isoprene consumption rates only.

Control and Characterization Experiments As part of this project, an ethene - NOx and several propene - NOx control experiments were also carried out, as well as a CO - NOx characterization experiment to evaluate background chamber radical sources. The initial conditions and O3 formed in these experiments are included in Table 5. Concentration - time plots of species useful for model evaluation are shown on Figure 11 for the ethene - NOx and propene - NOx experiments. Results of model calculations using the SAPRC-07 mechanism and the chamber effects model assigned for these experiments are also shown on these figures.

These runs had the same reactant concentrations in the dual reactors, so equivalent results should be obtained if the conditions in the two reactors are the same. Figure 11 shows that this is the case for the ethene and propene experiments, which means that the differences between the test and base results in the NOx sink experiments are not due to reactor differences. The SAPRC-07 mechanism gives very good fits to the data in the propene experiments shown on Figure 11, though it tended to somewhat underpredict O3 in the propene - NOx base case experiments for toluene as shown on Figure 4. The mechanism tended to somewhat underpredict O3 formation in the ethene experiment shown on Figure 11, and also underpredicted O3, though generally to a slightly lesser extent, in the base case ethene experiments shown in Figure 4 through Figure 6. This may be different for other mechanisms, but these results give an indication of the run-to-run variability in simulating the conditions of these experiments.

The CO - NOx characterization experiment was carried out to obtain information on the chamber-dependent radical source (Carter et al, 1982), which is needed to obtain chamber effects characterization parameters used when modeling these experiments.

26

Table 5. Summary of isoprene - NOx mechanism evaluation experiments and the single VOC - NOx control and characterizaton experiments carried out for this project.

Run Initial Conc (ppb)

Hours Maximum

NO NO2 VOC O3 (ppb)

Isoprene - NOx Experiments EPA1397A 24 0 250 7 163 EPA1405A 16 0 192 5 112 EPA1405B 6 0 200 3 47

Ethene - NOx Control Experiment

EPA1400A 0 13 909 4 276

Propene - NOx Control Experiments EPA1391A 0 14 311 4 144 EPA1395A 2 10 321 3 130 EPA1409A 14 0 328 8 138

CO - NOx Radical Source Characterization Experiment

EPA1456A 16 0 49 ppm 6 110 EPA1456B 15 0 50 ppm 4 70

27

Figure 10. Plots of selected results of the isoprene - NOx experiments carried out for this project.

EPA1397A EPA1405A EPA1405BOzone (ppm)

NO (ppm)

NO2 (ppm)

Isoprene (ppm)

PM Formed (µg/m3)


0.00

0.04

0.08

0.12

0.16

0 120 240 360 480

Expt.

0.00

0.01

0.02

0.03

0 120 240 360 480

0.000

0.005

0.010

0.015

0.020

0 120 240 360 480

No Data

0.0

0.1

0.2

0.3

0 120 240 360 480

0.0

0.5

1.0

1.5

2.0

2.5

0 120 240 360 480

0.00

0.04

0.08

0.12

0 60 120 180 240 300

0.000

0.005

0.010

0.015

0.020

0 60 120 180 240 300

0.000

0.004

0.008

0.012

0 60 120 180 240 300

0.00

0.05

0.10

0.15

0.20

0.25

0 60 120 180 240 300

0.0

0.1

0.2

0.3

0.4

0 60 120 180 240 300

0.00

0.01

0.02

0.03

0.04

0.05

0 60 120 180

0.000

0.002

0.004

0.006

0.008

0 60 120 180

0.000

0.001

0.002

0.003

0.004

0 60 120 180

CEASLuminol

0.00

0.05

0.10

0.15

0.20

0.25

0 60 120 180

0.000

0.004

0.008

0.012

0.016

0 60 120 180

28

Figure 11. Plots of selected results of the ethene - NOx and propene - NOx control experiments carried out for this project. (Results of the model calculations are also shown.)

Ethene PropeneEPA1400 EPA1391 EPA1395 EPA1409

Ozone (ppm)

NO (ppm)

NOy-NO (ppm) NO2 (ppm)

Ethene (ppm) Propene (ppm)

PAN (ppm)


0.0

0.1

0.2

0.3

0 60 120 180 240

Side ASide B

0.000

0.001

0.002

0.003

0.004

0.005

0 60 120 180 240

0.000

0.004

0.008

0.012

0.016

0 60 120 180 240

0.0

0.2

0.4

0.6

0.8

1.0

0 60 120 180 240

0.00

0.04

0.08

0.12

0.16

0 60 120 180 240

0.000

0.001

0.002

0.003

0.004

0 60 120 180 240

0.000

0.004

0.008

0.012

0.016

0 60 120 180 240

0.0

0.1

0.2

0.3

0.4

0 60 120 180 240

0.00

0.04

0.08

0.12

0.16

0 60 120 180

0.000

0.001

0.002

0.003

0.004

0 60 120 180

0.000

0.004

0.008

0.012

0 60 120 180

0.0

0.1

0.2

0.3

0.4

0 60 120 180

0.00

0.04

0.08

0.12

0.16

0 120 240 360 480

0.000

0.004

0.008

0.012

0.016

0 120 240 360 480

0.000

0.004

0.008

0.012

0 120 240 360 480

0.0

0.1

0.2

0.3

0.4

0 120 240 360 480

0.000

0.002

0.004

0.006

0.008

0.010

0 120 240 360 480

29

4.0 Improvements to CB6 The new chamber experiments conducted at UCR and the EUPHORE experiments retrieved for this project were used to improve the CB6 mechanism as described in this section. First, the CB6 mechanism is documented as the starting point for improvements. Then the mechanism improvements are explained. Finally the performance of the improved mechanism, called CB6r1, is shown for simulations of UCR and the EUPHORE experiments. The approach followed for reactions of aromatics was that mechanism improvements were developed using EUPHORE experiments with aromatic degradation products (2-butenedial, 4-oxo-2-pentenal and o-cresol) and evaluated using the NOx-sink experiments with toluene, o-cresol and furan conducted at UCR for this project.

Carbon Bond version 6 (CB6) CB6 is the 6th version of the Carbon Bond mechanism (Yarwood et al., 2010). The reactions included in CB6 are listed below in Table 6 and mechanism species are identified in Table 7. The core inorganic chemistry mechanism for CB6 is based on evaluated data from the IUPAC tropospheric chemistry panel as of January, 2010 (Atkinson et al., 2010). An exception is the rate constant for reaction of OH with NO2 which is from the 2006 NASA/JPL data evaluation (Sander et al., 2006). Atkinson et al. (2010) also is the primary source for photolysis data in CB6 with some data from Sander et al. (2006) and other sources for photolysis of some organic compounds. Compared to CB05 (Yarwood et al., 2005), several organic compounds that are long-lived and relatively abundant, namely propane, acetone, benzene and ethyne (acetylene), are added explicitly in CB6 so as to improve oxidant formation from these compounds as they are oxidized slowly at the regional scale. Alpha-dicarbonyl compounds (glyoxal and analogues), which form secondary organic aerosol (SOA) via aqueous-phase reactions (Carlton et al., 2007), are added in CB6 to improve support for SOA modeling. Precursors to alpha-dicarbonyls in CB6 are aromatics, alkenes and ethyne. CB6 includes several updates to peroxy radical chemistry that improve formation of hydrogen peroxide (H2O2) and therefore sulfate aerosol formation. The gas-phase reaction of dinitrogen pentoxide (N2O5) with water vapor is slower in CB6 which will reduce nighttime formation of nitric acid although heterogeneous reactions on aerosol surfaces may dominate nitric acid formation at night (Brown et al, 2006). When CB6 is used in atmospheric models the heterogeneous reaction between N2O5 and water vapor should be accounted for in the heterogeneous chemistry component of the model. There are changes to the organic chemistry for alkanes, alkenes, aromatics and oxygenates. The most extensive changes are for aromatics and isoprene. Chemistry updates for aromatics were based on the updated toluene mechanism (CB05-TU) developed by Whitten et al. (2009) extended to benzene and xylenes. The isoprene mechanism was revised based on several recently published studies (Paulot et al., 2010; Peeters et al., 2010).

CB6 was evaluated using 339 experiments from several chambers at the University of California at Riverside and the Tennessee Valley Authority (Yarwood et al., 2010). The performance of CB6 and CB05 in simulating chamber studies was comparable for alkanes, alkenes, alcohols and aldehydes with both CB6 and CB05 performing well and exhibiting 20% or less bias for maximum ozone. For species that were explicitly added in CB6 (ethyne, benzene and ketones), CB6 performed much better than CB05. For aromatics, CB6 improved upon CB05 by reducing under prediction bias in maximum ozone to about 10% for benzene, toluene and xylene. For isoprene, both CB05 and CB6 show little bias for maximum ozone (less than 5%) but CB6 tended to form

30

ozone too slowly. CB6 improved upon CB05 for simulating mixtures of VOCs. For mixtures without aromatics, both CB05 and CB6 showed minimal bias for maximum ozone. For mixtures including aromatics, both CB05 and CB6 under predicted maximum ozone but bias was reduced from about 30% for CB05 to about 20% for CB6.

Table 6. Listing of reactions and rate parameters for CB6.

Number Reactants and Products k298 Rate Parameters

Notes A Ea B 1 NO2 = NO + O Photolysis a 2 O + O2 + M = O3 + M 5.78E-34 5.68E-34 0.0 -2.60 a 3 O3 + NO = NO2 1.73E-14 1.40E-12 1310.0 0.00 a 4 O + NO + M = NO2 + M 1.01E-31 1.00E-31 0.0 -1.60 a 5 O + NO2 = NO 1.03E-11 5.50E-12 -188.0 0.00 a 6 O + NO2 = NO3 2.11E-12 Falloff, F=0.60 ,N=1.00 a k0 1.30E-31 0.0 -1.50 k∞ 2.30E-11 0.0 0.24 7 O + O3 = 7.96E-15 8.00E-12 2060.0 0.00 a 8 O3 = O Photolysis a 9 O3 = O1D Photolysis a 10 O1D + M = O + M 3.28E-11 2.23E-11 -115.0 0.00 a 11 O1D + H2O = 2 OH 2.14E-10 2.14E-10 a 12 O3 + OH = HO2 7.25E-14 1.70E-12 940.0 0.00 a 13 O3 + HO2 = OH 2.01E-15 2.03E-16 -693.0 4.57 a 14 OH + O = HO2 3.47E-11 2.40E-11 -110.0 0.00 a 15 HO2 + O = OH 5.73E-11 2.70E-11 -224.0 0.00 a 16 OH + OH = O 1.48E-12 6.20E-14 -945.0 2.60 a 17 OH + OH = H2O2 5.25E-12 Falloff, F=0.50 ,N=1.13 a k0 6.90E-31 0.0 -0.80 k∞ 2.60E-11 0.0 0.00 18 OH + HO2 = 1.11E-10 4.80E-11 -250.0 0.00 a 19 HO2 + HO2 = H2O2 2.90E-12 k = k1 + k2[M] a k1 2.20E-13 -600.0 0.00 k2 1.90E-33 -980.0 0.00 20 HO2 + HO2 + H2O = H2O2 6.53E-30 k = k1 + k2[M] a k1 3.08E-34 -2800.0 0.00 k2 2.66E-54 -3180.0 0.00 21 H2O2 = 2 OH Photolysis a 22 H2O2 + OH = HO2 1.70E-12 2.90E-12 160.0 0.00 a 23 H2O2 + O = OH + HO2 1.70E-15 1.40E-12 2000.0 0.00 a 24 NO + NO + O2 = 2 NO2 1.95E-38 3.30E-39 -530.0 0.00 a 25 HO2 + NO = OH + NO2 8.54E-12 3.45E-12 -270.0 0.00 a 26 NO2 + O3 = NO3 3.52E-17 1.40E-13 2470.0 0.00 a 27 NO3 = NO2 + O Photolysis b 28 NO3 = NO Photolysis b 29 NO3 + NO = 2 NO2 2.60E-11 1.80E-11 -110.0 0.00 a 30 NO3 + NO2 = NO + NO2 6.56E-16 4.50E-14 1260.0 0.00 b 31 NO3 + O = NO2 1.70E-11 1.70E-11 a 32 NO3 + OH = HO2 + NO2 2.00E-11 2.00E-11 a 33 NO3 + HO2 = OH + NO2 4.00E-12 4.00E-12 a 34 NO3 + O3 = NO2 1.00E-17 1.00E-17 c,k 35 NO3 + NO3 = 2 NO2 2.28E-16 8.50E-13 2450.0 0.00 b 36 NO3 + NO2 = N2O5 1.24E-12 Falloff, F=0.35 ,N=1.33 a

31


Notes A Ea B k0 3.60E-30 0.0 -4.10 k∞ 1.90E-12 0.0 0.20 37 N2O5 = NO3 + NO2 4.46E-02 Falloff, F=0.35 ,N=1.33 a k0 1.30E-03 11000.0 -3.50 k∞ 9.70E+14 11080.0 0.10 38 N2O5 = NO2 + NO3 Photolysis a 39 N2O5 + H2O = 2 HNO3 1.00E-22 1.00E-22 a 40 NO + OH = HONO 9.77E-12 Falloff, F=0.81 ,N=0.87 a k0 7.40E-31 0.0 -2.40 k∞ 3.30E-11 0.0 -0.30 41 NO + NO2 + H2O = 2 HONO 5.00E-40 5.00E-40 c,l 42 HONO + HONO = NO + NO2 1.00E-20 1.00E-20 c,m 43 HONO = NO + OH Photolysis a 44 HONO + OH = NO2 5.98E-12 2.50E-12 -260.0 0.00 a 45 NO2 + OH = HNO3 1.06E-11 Falloff, F=0.60 ,N=1.00 b k0 1.80E-30 0.0 -3.00 k∞ 2.80E-11 0.0 0.00 46 HNO3 + OH = NO3 1.54E-13 k = k1+k3M/(1+k3M/k2) a k1 2.40E-14 -460.0 0.00 k2 2.70E-17 -2199.0 0.00 k3 6.50E-34 -1335.0 0.00 47 HNO3 = OH + NO2 Photolysis a 48 HO2 + NO2 = PNA 1.38E-12 Falloff, F=0.60 ,N=1.00 a k0 1.80E-31 0.0 -3.20 k∞ 4.70E-12 0.0 0.00 49 PNA = HO2 + NO2 8.31E-02 Falloff, F=0.60 ,N=1.00 a k0 4.10E-05 10650.0 0.00 k∞ 4.80E+15 11170.0 0.00 50 PNA = 0.59 HO2 + 0.59 NO2 + 0.41 OH + 0.41 NO3 Photolysis a

51 PNA + OH = NO2 3.24E-12 3.20E-13 -690.0 0.00 a 52 SO2 + OH = SULF + HO2 8.12E-13 Falloff, F=0.53 ,N=1.10 a k0 4.50E-31 0.0 -3.90 k∞ 1.30E-12 0.0 -0.70 53 C2O3 + NO = NO2 + MEO2 + RO2 1.98E-11 7.50E-12 -290.0 0.00 a 54 C2O3 + NO2 = PAN 1.05E-11 Falloff, F=0.30 ,N=1.00 a k0 2.70E-28 0.0 -7.10 k∞ 1.20E-11 0.0 -0.90 55 PAN = NO2 + C2O3 3.31E-04 Falloff, F=0.30 ,N=1.00 a k0 4.90E-03 12100.0 0.00 k∞ 5.40E+16 13830.0 0.00 56 PAN = 0.6 NO2 + 0.6 C2O3 + 0.4 NO3 + 0.4 MEO2

+ 0.4 RO2 Photolysis a

57 C2O3 + HO2 = 0.41 PACD + 0.15 AACD + 0.15 O3 + 0.44 MEO2 + 0.44 RO2 + 0.44 OH

1.39E-11 5.20E-13 -980.0 0.00 a

58 C2O3 + RO2 = C2O3 1.30E-11 8.90E-13 -800.0 0.00 a 59 C2O3 + C2O3 = 2 MEO2 + 2 RO2 1.55E-11 2.90E-12 -500.0 0.00 a 60 C2O3 + CXO3 = MEO2 + ALD2 + XO2H + 2 RO2 1.55E-11 2.90E-12 -500.0 0.00 a

61 CXO3 + NO = NO2 + ALD2 + XO2H + RO2 2.10E-11 6.70E-12 -340.0 0.00 a 62 CXO3 + NO2 = PANX 1.16E-11 Falloff, F=0.30 ,N=1.00 a k0 3.00E-28 0.0 -7.10

32


Notes A Ea B k∞ 1.33E-11 0.0 -0.90 63 PANX = NO2 + CXO3 3.68E-04 Falloff, F=0.30 ,N=1.00 a k0 1.70E-03 11280.0 0.00 k∞ 8.30E+16 13940.0 0.00 64 PANX = 0.6 NO2 +0.6 CXO3 + 0.4 NO3 + 0.4 ALD2

+ 0.4 XO2H + 0.4 RO2 Photolysis a

65 CXO3 + HO2 = 0.41 PACD + 0.15 AACD + 0.15 O3 + 0.44 ALD2 + 0.44 XO2H + 0.44 RO2 + 0.44 OH

1.39E-11 5.20E-13 -980.0 0.00 a

66 CXO3 + RO2 = CXO3 1.30E-11 8.90E-13 -800.0 0.00 a 67 CXO3 + CXO3 = 2 ALD2 + 2 XO2H + 2 RO2 1.71E-11 3.20E-12 -500.0 0.00 a 68 RO2 + NO = NO 8.03E-12 2.40E-12 -360.0 0.00 a 69 RO2 + HO2 = HO2 7.03E-12 4.80E-13 -800.0 0.00 a 70 RO2 + RO2 = 3.48E-13 6.50E-14 -500.0 0.00 a 71 MEO2 + NO = FORM + HO2 + NO2 7.70E-12 2.30E-12 -360.0 0.00 a 72 MEO2 + HO2 = 0.9 MEPX + 0.1 FORM 5.21E-12 3.80E-13 -780.0 0.00 a 73 MEO2 + C2O3 = FORM + 0.9 HO2 + 0.9 MEO2 +

0.1 AACD + 0.9 RO2 1.07E-11 2.00E-12 -500.0 0.00 a

74 MEO2 + RO2 = 0.685 FORM + 0.315 MEOH + 0.37 HO2 + RO2

3.48E-13 k = kref*K a k(ref) ref = 70 K 1.00E+00 0.0 0.00 75 XO2H + NO = NO2 + HO2 9.04E-12 2.70E-12 -360.0 0.00 a 76 XO2H + HO2 = ROOH 9.96E-12 6.80E-13 -800.0 0.00 a 77 XO2H + C2O3 = 0.8 HO2 + 0.8 MEO2 + 0.2 AACD +

0.8 RO2 1.30E-11 k = kref*K a

k(ref) ref = 58 K 1.00E+00 0.0 0.00 78 XO2H + RO2 = 0.6 HO2 + RO2 3.48E-13 k = kref*K a k(ref) ref = 70 K 1.00E+00 0.0 0.00 79 XO2 + NO = NO2 9.04E-12 k = kref*K a k(ref) ref = 75 K 1.00E+00 0.0 0.00 80 XO2 + HO2 = ROOH 9.96E-12 k = kref*K a k(ref) ref = 76 K 1.00E+00 0.0 0.00 81 XO2 + C2O3 = 0.8 MEO2 + 0.2 AACD + 0.8 RO2 1.30E-11 k = kref*K a

k(ref) ref = 58 K 1.00E+00 0.0 0.00 82 XO2 + RO2 = 0.6 HO2 + RO2 3.48E-13 k = kref*K a k(ref) ref = 70 K 1.00E+00 0.0 0.00 83 XO2N + NO = NTR 9.04E-12 k = kref*K a k(ref) ref = 75 K 1.00E+00 0.0 0.00 84 XO2N + HO2 = ROOH 9.96E-12 k = kref*K a k(ref) ref = 76 K 1.00E+00 0.0 0.00 85 XO2N + C2O3 = 0.8 HO2 + 0.8 MEO2 + 0.2 AACD +

0.8 RO2 1.30E-11 k = kref*K a

k(ref) ref = 58 K 1.00E+00 0.0 0.00 86 XO2N + RO2 = 0.6 HO2 + RO2 3.48E-13 k = kref*K a

33


Notes A Ea B k(ref) ref = 70 K 1.00E+00 0.0 0.00 87 MEPX + OH = 0.6 MEO2 + 0.6 RO2 + 0.4 FORM +

0.4 OH 1.00E-11 5.30E-12 -190.0 0.00 a

88 MEPX = MEO2 + RO2 + OH Photolysis a 89 ROOH + OH = 0.54 XO2H + 0.06 XO2N + 0.6 RO2

+ 0.4 OH 6.05E-12 3.20E-12 -190.0 0.00 a

90 ROOH = HO2 + OH Photolysis a 91 NTR + OH = HNO3 + XO2H + RO2 8.10E-13 8.10E-13 a,c 92 NTR = NO2 + XO2H + RO2 Photolysis a,c 93 FACD + OH = HO2 4.50E-13 4.50E-13 a 94 AACD + OH = MEO2 + RO2 6.93E-13 4.00E-14 -850.0 0.00 a 95 PACD + OH = C2O3 6.93E-13 4.00E-14 -850.0 0.00 a 96 FORM + OH = HO2 + CO 8.49E-12 5.40E-12 -135.0 0.00 a 97 FORM = 2 HO2 + CO Photolysis a 98 FORM = CO + H2 Photolysis a 99 FORM + O = OH + HO2 + CO 1.58E-13 3.40E-11 1600.0 0.00 b 100 FORM + NO3 = HNO3 + HO2 + CO 5.50E-16 5.50E-16 a 101 FORM + HO2 = HCO3 7.90E-14 9.70E-15 -625.0 0.00 a 102 HCO3 = FORM + HO2 1.51E+02 2.40E+12 7000.0 0.00 a 103 HCO3 + NO = FACD + NO2 + HO2 5.60E-12 5.60E-12 a 104 HCO3 + HO2 = 0.5 MEPX + 0.5 FACD + 0.2 OH +

0.2 HO2 1.26E-11 5.60E-15 -2300.0 0.00 a

105 ALD2 + O = C2O3 + OH 4.49E-13 1.80E-11 1100.0 0.00 b 106 ALD2 + OH = C2O3 1.50E-11 4.70E-12 -345.0 0.00 a 107 ALD2 + NO3 = C2O3 + HNO3 2.73E-15 1.40E-12 1860.0 0.00 a 108 ALD2 = MEO2 + RO2 + CO + HO2 Photolysis a 109 ALDX + O = CXO3 + OH 7.02E-13 1.30E-11 870.0 0.00 c,n 110 ALDX + OH = CXO3 1.91E-11 4.90E-12 -405.0 0.00 a 111 ALDX + NO3 = CXO3 + HNO3 6.30E-15 6.30E-15 a 112 ALDX = MEO2 + RO2 + CO + HO2 Photolysis f 113 GLYD + OH = 0.2 GLY + 0.2 HO2 + 0.8 C2O3 8.00E-12 8.00E-12 a 114 GLYD = 0.74 FORM + 0.89 CO + 1.4 HO2 + 0.15

MEOH + 0.19 OH + 0.11 GLY + 0.11 XO2H + 0.11 RO2

Photolysis a,b,f

115 GLYD + NO3 = HNO3 + C2O3 2.73E-15 1.40E-12 1860.0 0.00 a 116 GLY + OH = 1.7 CO + 0.3 XO2 + 0.3 RO2 + HO2 9.70E-12 3.10E-12 -340.0 0.00 a 117 GLY = 2 HO2 + 2 CO Photolysis a,q 118 GLY + NO3 = HNO3 + CO + HO2 + XO2 + RO2 2.73E-15 1.40E-12 1860.0 0.00 a 119 MGLY = C2O3 + HO2 + CO Photolysis a 120 MGLY + NO3 = HNO3 + C2O3 + XO2 + RO2 2.73E-15 1.40E-12 1860.0 0.00 a 121 MGLY + OH = C2O3 + CO 1.31E-11 1.90E-12 -575.0 0.00 a 122 H2 + OH = HO2 6.70E-15 7.70E-12 2100.0 0.00 a 123 CO + OH = HO2 2.28E-13 k = k1 + k2[M] a k1 1.44E-13 0.0 0.00 k2 3.43E-33 0.0 0.00 124 CH4 + OH = MEO2 + RO2 6.37E-15 1.85E-12 1690.0 0.00 a 125 ETHA + OH = 0.991 ALD2 + 0.991 XO2H + 0.009

XO2N + RO2 2.41E-13 6.90E-12 1000.0 0.00 a

126 MEOH + OH = FORM + HO2 8.95E-13 2.85E-12 345.0 0.00 a 127 ETOH + OH = 0.95 ALD2 + 0.9 HO2 + 0.1 XO2H +

0.1 RO2 + 0.078 FORM + 0.011 GLYD 3.21E-12 3.00E-12 -20.0 0.00 a

34


Notes A Ea B 128 KET = 0.5 ALD2 + 0.5 C2O3 + 0.5 XO2H +0.5 CXO3

+ 0.5 MEO2 + RO2 - 2.5 PAR Photolysis a

129 ACET = 0.38 CO + 1.38 MEO2 + 1.38 RO2 + 0.62 C2O3

Photolysis a

130 ACET + OH = FORM + C2O3 + XO2 + RO2 1.76E-13 1.41E-12 620.6 0.00 a 131 PRPA + OH = 0.71 ACET + 0.26 ALDX + 0.26 PAR +

0.97 XO2H + 0.03 XO2N + RO2 1.07E-12 7.60E-12 585.0 0.00 a

132 PAR + OH = 0.11 ALDX + 0.76 ROR + 0.13 XO2N + 0.11 XO2H + 0.76 XO2 + RO2 - 0.11 PAR

8.10E-13 8.10E-13 c

133 ROR = 0.2 KET + 0.42 ACET + 0.74 ALD2 + 0.37 ALDX + 0.04 XO2N + 0.94 XO2H + 0.98 RO2 + 0.02 ROR - 2.7 PAR

2.15E+04 5.70E+12 5780.0 0.00 a,c

134 ROR + O2 = KET + HO2 3.78E+04 1.50E-14 200.0 0.00 a,c 135 ROR + NO2 = NTR 3.29E-11 8.60E-12 -400.0 0.00 a,c 136 ETHY + OH = 0.7 GLY + 0.7 OH + 0.3 FACD + 0.3

CO + 0.3 HO2 7.52E-13 Falloff, F=0.37 ,N=1.30 a

k0 5.00E-30 0.0 -1.50 k∞ 1.00E-12 0.0 0.00 137 ETH + O = FORM + HO2 + CO + 0.7 XO2H + 0.7

RO2 + 0.3 OH 7.29E-13 1.04E-11 792.0 0.00 c,o

138 ETH + OH = XO2H + RO2 + 1.56 FORM + 0.22 GLYD

7.84E-12 Falloff, F=0.48 ,N=1.15 a,g k0 8.60E-29 0.0 -3.10 k∞ 9.00E-12 0.0 -0.85 139 ETH + O3 = FORM + 0.51 CO + 0.16 HO2 + 0.16

OH + 0.37 FACD 1.58E-18 9.10E-15 2580.0 0.00 a,g

140 ETH + NO3 = 0.5 NO2 + 0.5 NTR + 0.5 XO2H + 0.5 XO2 + RO2 + 1.125 FORM

2.10E-16 3.30E-12 2880.0 0.00 a,g

141 OLE + O = 0.2 ALD2 + 0.3 ALDX + 0.1 HO2 + 0.2 XO2H + 0.2 CO + 0.2 FORM + 0.01 XO2N + 0.21 RO2 + 0.2 PAR + 0.1 OH

3.91E-12 1.00E-11 280.0 0.00 c,o

142 OLE + OH = 0.781 FORM + 0.488 ALD2 + 0.488 ALDX + 0.976 XO2H + 0.195 XO2 + 0.024 XO2N + 1.17 RO2 - 0.73 PAR

2.86E-11 Falloff, F=0.50 ,N=1.13 a,g k0 8.00E-27 0.0 -3.50 k∞ 3.00E-11 0.0 -1.00 143 OLE + O3 = 0.295 ALD2 + 0.555 FORM + 0.27

ALDX + 0.15 XO2H + 0.15 RO2 + 0.334 OH +0.08 HO2 + 0.378 CO + 0.075 GLY + 0.075 MGLY + 0.09 FACD + 0.13 AACD + 0.04 H2O2 - 0.79 PAR

1.00E-17 5.50E-15 1880.0 0.00 a,g

144 OLE + NO3 = 0.5 NO2 + 0.5 NTR + 0.48 XO2 + 0.48 XO2H + 0.04 XO2N + RO2 + 0.5 FORM + 0.25 ALD2 + 0.375 ALDX – PAR

9.54E-15 4.60E-13 1155.0 0.00 a,g

145 IOLE + O = 1.24 ALD2 + 0.66 ALDX + 0.1 XO2H + 0.1 RO2 + 0.1 CO + 0.1 PAR

2.30E-11 2.30E-11 c,o

146 IOLE + OH = 1.3 ALD2 + 0.7 ALDX + XO2H + RO2 5.99E-11 1.05E-11 -519.0 0.00 a,g 147 IOLE + O3 = 0.732 ALD2 + 0.442 ALDX + 0.128

FORM + 0.245 CO + 0.5 OH + 0.3 XO2H + 0.3 RO2 + 0.24 GLY + 0.06 MGLY + 0.29 PAR + 0.08 AACD + 0.08 H2O2

1.57E-16 4.70E-15 1013.0 0.00 a,g

148 IOLE + NO3 = 0.5 NO2 + 0.5 NTR + 0.48 XO2 + 0.48 XO2H + 0.04 XO2N + RO2 + 0.5 ALD2 + 0.625 ALDX + PAR

3.70E-13 3.70E-13 a,g

149 ISOP + OH = ISO2 + RO2 9.99E-11 2.70E-11 -390.0 0.00 a 150 ISO2 + NO = 0.117 INTR + 0.883 NO2 + 0.803

HO2 + 0.66 FORM + 0.66 ISPD + 0.08 XO2H + 0.08 RO2 + 0.05 IOLE + 0.042 GLYD + 0.115 PAR + 0.038 GLY + 0.042 MGLY + 0.093 OLE + 0.117 ALDX

8.13E-12 2.39E-12 -365.0 0.00 r,s

35


Notes A Ea B 151 ISO2 + HO2 = 0.88 ISPX + 0.12 OH + 0.12 HO2 +

0.12 FORM + 0.12 ISPD 7.78E-12 7.43E-13 -700.0 0.00 r,s

152 ISO2 + C2O3 = 0.709 HO2 + 0.583 FORM + 0.583 ISPD + 0.071 XO2H + 0.044 IOLE + 0.037 GLYD + 0.102 PAR + 0.034 GLY + 0.037 MGLY + 0.082 OLE + 0.103 ALDX + 0.8 MEO2 + 0.2 AACD + 0.871 RO2

1.30E-11 k = kref*K r,s k(ref) ref = 58 K 1.00E+00 0.0 0.00

153 ISO2 + RO2 = 0.803 HO2 + 0.66 FORM + 0.66 ISPD + 0.08 XO2H + 0.05 IOLE + 0.042 GLYD + 0.115 PAR + 0.038 GLY + 0.042 MGLY + 0.093 OLE + 0.117 ALDX + 1.08 RO2

3.48E-13 k = kref*K r,s k(ref) ref = 70 K 1.00E+00 0.0 0.00

154 ISO2 = 0.8 HO2 + 0.04 OH + 0.04 FORM + 0.8 ISPD 1.00E+00 1.00E+00 j,t 155 ISOP + O3 = 0.6 FORM + 0.65 ISPD + 0.15 ALDX +

0.2 CXO3 + 0.35 PAR + 0.266 OH + 0.2 XO2 + 0.2 RO2 + 0.066 HO2 + 0.066 CO

1.27E-17 1.03E-14 1995.0 0.00 c

156 ISOP + NO3 = 0.35 NO2 + 0.65 INTR + 0.64 XO2H + 0.33 XO2 + 0.03 XO2N + RO2 + 0.35 FORM + 0.35 ISPD

6.74E-13 3.03E-12 448.0 0.00 u

157 ISPD + OH = 0.095 XO2N + 0.379 XO2 + 0.318 XO2H + 0.792 RO2 + 0.843 PAR + 0.379 C2O3 + 0.209 CXO3 + 0.379 GLYD + 0.24 MGLY + 0.24 FORM + 0.067 OLE + 0.079 CO + 0.028 ALDX

3.38E-11 6.31E-12 -500.0 0.00 r,s

158 ISPD + O3 = 0.02 ALD2 + 0.15 FORM + 0.225 CO + 0.85 MGLY + 0.36 PAR + 0.114 C2O3 + 0.064 XO2H + 0.064 RO2 + 0.268 OH + 0.09 HO2

7.10E-18 4.17E-15 1900.0 0.00 c

159 ISPD + NO3 = 0.643 CO + 0.282 FORM + 0.357 ALDX + 1.282 PAR + 0.85 HO2 + 0.075 CXO3 + 0.075 XO2H + 0.075 RO2 + 0.85 NTR + 0.15 HNO3

1.00E-15 1.00E-15 c

160 ISPD = 0.333 CO + 0.067 ALD2 + 0.9 FORM + 0.832 PAR + 0.333 HO2 + 0.7 XO2H + 0.7 RO2 + 0.967 C2O3

Photolysis c,f

161 ISPX + OH = 0.904 EPOX + 0.933 OH + 0.067 ISO2 + 0.067 RO2 + 0.029 IOLE + 0.029 ALDX

7.77E-11 2.23E-11 -372.0 0.00 r,s

162 EPOX + OH = EPX2 + RO2 1.51E-11 5.78E-11 400.0 0.00 r,s 163 EPX2 + HO2 = 0.275 GLYD + 0.275 GLY + 0.275

MGLY + 1.125 OH + 0.825 HO2 + 0.375 FORM + 0.074 FACD + 0.251 CO + 2.175 PAR

7.78E-12 7.43E-13 -700.0 0.00 r,s

164 EPX2 + NO = 0.275 GLYD + 0.275 GLY + 0.275 MGLY + 0.125 OH + 0.825 HO2 + 0.375 FORM + NO2 + 0.251 CO + 2.175 PAR

8.13E-12 2.39E-12 -365.0 0.00 r,s

165 EPX2 + C2O3 = 0.22 GLYD + 0.22 GLY + 0.22 MGLY + 0.1 OH + 0.66 HO2 + 0.3 FORM + 0.2 CO + 1.74 PAR + 0.8 MEO2 + 0.2 AACD + 0.8 RO2

1.30E-11 k = kref*K a,r,s k(ref) ref = 58 K 1.00E+00 0.0 0.00 166 EPX2 + RO2 = 0.275 GLYD + 0.275 GLY + 0.275

MGLY + 0.125 OH + 0.825 HO2 + 0.375 FORM + 0.251 CO + 2.175 PAR + RO2

3.48E-13 k = kref*K a,r,s k(ref) ref = 70 K 1.00E+00 0.0 0.00 167 INTR + OH = 0.63 XO2 + 0.37 XO2H + RO2 + 0.444

NO2 + 0.185 NO3 + 0.104 INTR + 0.592 FORM + 0.331 GLYD + 0.185 FACD + 2.7 PAR + 0.098 OLE + 0.078 ALDX + 0.266 NTR

3.10E-11 3.10E-11 r,s

168 TERP + O = 0.15 ALDX + 5.12 PAR 3.60E-11 3.60E-11 c 169 TERP + OH = 0.75 XO2H + 0.5 XO2 + 0.25 XO2N +

1.5 RO2 + 0.28 FORM + 1.66 PAR + 0.47 ALDX 6.77E-11 1.50E-11 -449.0 0.00 c

170 TERP + O3 = 0.57 OH + 0.07 XO2H + 0.69 XO2 + 0.18 XO2N + 0.94 RO2 + 0.24 FORM + 0.001 CO + 7 PAR + 0.21 ALDX + 0.39 CXO3

7.63E-17 1.20E-15 821.0 0.00 c

171 TERP + NO3 = 0.47 NO2 + 0.28 XO2H + 0.75 XO2 + 0.25 XO2N + 1.28 RO2 + 0.47 ALDX + 0.53 NTR

6.66E-12 3.70E-12 -175.0 0.00 c

36


Notes A Ea B 172 BENZ + OH = 0.53 CRES + 0.352 BZO2 + 0.352 RO2

+ 0.118 OPEN + 0.118 OH + 0.53 HO2 1.22E-12 2.30E-12 190.0 0.00 a,d,e

173 BZO2 + NO = 0.918 NO2 + 0.082 NTR + 0.918 GLY + 0.918 OPEN + 0.918 HO2

9.04E-12 2.70E-12 -360.0 0.00 d,h

174 BZO2 + C2O3 = GLY + OPEN + HO2 + MEO2 + RO2 1.30E-11 k = kref*K a,d,h k(ref) ref = 58 K 1.00E+00 0.0 0.00 175 BZO2 + HO2 = 1.49E-11 1.90E-13 -1300.0 0.00 d 176 BZO2 + RO2 = GLY + OPEN + HO2 + RO2 3.48E-13 k = kref*K a,d,h k(ref) ref = 70 K 1.00E+00 0.0 0.00 177 TOL + OH = 0.18 CRES + 0.65 TO2 + 0.72 RO2 +

0.1 OPEN + 0.1 OH + 0.07 XO2H + 0.18 HO2 5.63E-12 1.80E-12 -340.0 0.00 a,d,e

178 TO2 + NO = 0.86 NO2 + 0.14 NTR + 0.417 GLY + 0.443 MGLY + 0.66 OPEN + 0.2 XOPN + 0.86 HO2

9.04E-12 2.70E-12 -360.0 0.00 d,h

179 TO2 + C2O3 = 0.48 GLY + 0.52 MGLY + 0.77 OPEN + 0.23 XOPN + HO2 + MEO2 + RO2

1.30E-11 k = kref*K a,d,h k(ref) ref = 58 K 1.00E+00 0.0 0.00 180 TO2 + HO2 = 1.49E-11 1.90E-13 -1300.0 0.00 d 181 TO2 + RO2 = 0.48 GLY + 0.52 MGLY + 0.77 OPEN +

0.23 XOPN + HO2 + RO2 3.48E-13 k = kref*K a,d,h

k(ref) ref = 70 K 1.00E+00 0.0 0.00 182 XYL + OH = 0.155 CRES + 0.544 XLO2 + 0.602 RO2

+ 0.244 XOPN + 0.244 OH + 0.058 XO2H + 0.155 HO2

1.85E-11 1.85E-11 d,e,p

183 XLO2 + NO = 0.86 NO2 + 0.14 NTR + 0.221 GLY + 0.675 MGLY + 0.3 OPEN + 0.56 XOPN + 0.86 HO2

9.04E-12 2.70E-12 -360.0 0.00 d,h

184 XLO2 + HO2 = 1.49E-11 1.90E-13 -1300.0 0.00 d 185 XLO2 + C2O3 = 0.26 GLY + 0.77 MGLY + 0.35

OPEN + 0.65 XOPN + HO2 + MEO2 + RO2 1.30E-11 k = kref*K a,d,h

k(ref) ref = 58 K 1.00E+00 0.0 0.00 186 XLO2 + RO2 = 0.26 GLY + 0.77 MGLY + 0.35 OPEN

+ 0.65 XOPN + HO2 + RO2 3.48E-13 k = kref*K a,d,h

k(ref) ref = 70 K 1.00E+00 0.0 0.00 187 CRES + OH = 0.06 CRO + 0.12 XO2H + HO2 + 0.13

OPEN + 0.732 CAT1 + 0.06 CO + 0.06 XO2N + 0.18 RO2 + 0.06 FORM

4.12E-11 1.70E-12 -950.0 0.00 d

188 CRES + NO3 = 0.3 CRO + HNO3 + 0.24 XO2 + 0.36 XO2H + 0.48 ALDX + 0.24 FORM + 0.24 MGLY + 0.12 OPEN + 0.1 XO2N + 0.7 RO2 + 0.24 CO

1.40E-11 1.40E-11 d

189 CRO + NO2 = CRON 2.10E-12 2.10E-12 d 190 CRO + HO2 = CRES 5.50E-12 5.50E-12 d 191 CRON + OH = CRNO 1.53E-12 1.53E-12 d 192 CRON + NO3 = CRNO + HNO3 3.80E-12 3.80E-12 d 193 CRNO + NO2 = 2 NTR 2.10E-12 2.10E-12 d 194 CRNO + O3 = CRN2 2.86E-13 2.86E-13 d 195 CRN2 + NO = CRNO + NO2 8.50E-12 2.54E-12 -360.0 0.00 d 196 CRN2 + HO2 = CRPX 1.88E-11 2.40E-13 -1300.0 0.00 d 197 CRPX = CRNO + OH Photolysis a,d 198 CRPX + OH = CRN2 3.59E-12 1.90E-12 -190.0 0.00 d 199 XOPN = CAO2 + 0.7 HO2 + 0.7 CO + 0.3 C2O3 +

RO2 Photolysis d,p

200 XOPN + OH = CAO2 + MGLY + XO2H + RO2 9.00E-11 9.00E-11 d,p

37


Notes A Ea B 201 XOPN + O3 = 1.2 MGLY + 0.5 OH + 0.6 C2O3 + 0.1

ALD2 + 0.5 CO + 0.3 XO2H + 0.3 RO2 2.02E-17 1.08E-16 500.0 0.00 d,p

202 XOPN + NO3 = 0.5 NO2 + 0.5 NTR + 0.45 XO2H + 0.45 XO2 + 0.1 XO2N + RO2 + 0.25 OPEN + 0.25 MGLY

3.00E-12 3.00E-12 d,p

203 OPEN = OPO3 + HO2 + CO Photolysis d,p 204 OPEN + OH = 0.6 OPO3 + 0.4 CAO2 + 0.4 RO2 4.40E-11 4.40E-11 d,p 205 OPEN + O3 = 1.4 GLY + 0.24 MGLY + 0.5 OH + 0.12

C2O3 + 0.08 FORM + 0.02 ALD2 + 1.98 CO + 0.56 HO2

1.01E-17 5.40E-17 500.0 0.00 d,p

206 OPEN + NO3 = OPO3 + HNO3 3.80E-12 3.80E-12 d,p 207 CAT1 + OH = CAO2 + RO2 7.00E-11 7.00E-11 d 208 CAT1 + NO3 = CRO + HNO3 1.70E-10 1.70E-10 d 209 CAO2 + NO = 0.86 NO2 + 0.14 NTR + 1.2 HO2 +

0.344 FORM + 0.344 CO 8.50E-12 2.54E-12 -360.0 0.00 d

210 CAO2 + HO2 = 1.88E-11 2.40E-13 -1300.0 0.00 d 211 CAO2 + C2O3 = HO2 + 0.4 GLY + MEO2 + RO2 1.30E-11 k = kref*K d

k(ref) ref = 58 K 1.00E+00 0.0 0.00 212 CAO2 + RO2 = HO2 + 0.4 GLY + RO2 3.48E-13 k = kref*K d k(ref) ref = 70 K 1.00E+00 0.0 0.00 213 OPO3 + NO = NO2 + XO2H + RO2 + ALDX 1.00E-11 1.00E-11 d 214 OPO3 + NO2 = OPAN 1.16E-11 k = kref*K d k(ref) ref = 62 K 1.00E+00 0.0 0.00 215 OPAN = OPO3 + NO2 9.92E-05 Falloff, F=0.30 ,N=1.00 d k0 4.60E-04 11280.0 0.00 k∞ 2.24E+16 13940.0 0.00 216 OPO3 + HO2 = 0.41 PACD + 0.15 AACD + 0.15 O3

+ 0.44 ALDX + 0.44 XO2H + 0.44 RO2 + 0.44 OH 1.39E-11 k = kref*K d

k(ref) ref = 57 K 1.00E+00 0.0 0.00 217 OPO3 + C2O3 = MEO2 + XO2 +ALDX + 2 RO2 1.55E-11 k = kref*K d k(ref) ref = 59 K 1.00E+00 0.0 0.00 218 OPO3 + RO2 = 0.8 XO2H + 0.8 RO2 + 0.8 ALDX +

0.2 AACD 1.30E-11 k = kref*K d

k(ref) ref = 58 K 1.00E+00 0.0 0.00

Table notes: k298 is the rate constant at 298 K and 1 atmosphere using units molecules cm-3 and s-1 See Table 7 for species names

a IUPAC: Atkinson et al., (2010) h Arey et al. (2009) o Cvetanovic (1987) b JPL: Sander et al., (2006) i Hu et al. (2007) p Calvert et al. (2002) c CB05: Yarwood et al (2005) j Archibald et al. (2010) q Feierabend et al. (2009) d CB05-TU: Whitten et al., 2010) k Hjorth et al. (1992) r Paulot et al. (2009a) e Bloss et al. (2005) l Kaiser and Wu (1977) s Paulot et al. (2009b) f SAPRC-99: Carter (2000) m Jeffries et al. (2002) t Peeters et al. (2009) g Calvert et al. (2000) n Herron (1988) u Perring et al. (2009)

38

Table 7. Model species names for CB6. Species Description AACD Acetic acid ACET Acetone ALD2 Acetaldehyde ALDX Propionaldehyde and higher aldehydes BENZ Benzene BZO2 Peroxy radical from OH addition to benzene C2O3 Acetylperoxy radical CAO2 Peroxy radical from aromatic degradation products CAT1 Methyl-catechols CH4 Methane CO Carbon monoxide CRES Cresols CRN2 Peroxy radical from nitro-cresol CRNO Alkoxy radical from nitro-cresols CRO Alkoxy radical from cresol CRON Nitro-cresols CRPX Nitro-cresol hydroperoxides CXO3 C3 and higher acylperoxy radicals EPOX Epoxide formed from ISPX reaction with OH EPX2 Peroxy radical from EPOX reaction with OH ETH Ethene ETHA Ethane ETHY Ethyne ETOH Ethanol FACD Formic acid FORM Formaldehyde GLY Glyoxal GLYD Glycolaldehyde H2O2 Hydrogen peroxide HCO3 Adduct from HO2 plus formaldehyde HNO3 Nitric acid HO2 Hydroperoxy radical HONO Nitrous acid INTR Organic nitrates from ISO2 reaction with NO IOLE Internal olefin carbon bond (R-C=C-R) ISO2 Peroxy radical from OH addition to isoprene ISOP Isoprene ISPD Isoprene product (lumped methacrolein, methyl vinyl ketone, etc.) ISPX Hydroperoxides from ISO2 reaction with HO2 KET Ketone carbon bond (C=O) M Air MEO2 Methylperoxy radical MEOH Methanol MEPX Methylhydroperoxide MGLY Methylglyoxal N2O5 Dinitrogen pentoxide NO Nitric oxide NO2 Nitrogen dioxide NO3 Nitrate radical NTR Organic nitrates

39

Species Description O Oxygen atom in the O3(P) electronic state O1D Oxygen atom in the O1(D) electronic state O2 Oxygen O3 Ozone OH Hydroxyl radical OLE Terminal olefin carbon bond (R-C=C) OPAN Peroxyacyl nitrate (PAN compound) from OPO3 OPEN Aromatic ring opening product (unsaturated dicarbonyl) OPO3 Peroxyacyl radical from OPEN PACD Peroxyacetic and higher peroxycarboxylic acids PAN Peroxyacetyl Nitrate PANX C3 and higher peroxyacyl nitrate PAR Paraffin carbon bond (C-C) PNA Peroxynitric acid PRPA Propane RO2 Operator to approximate total peroxy radical concentration ROOH Higher organic peroxide ROR Secondary alkoxy radical SO2 Sulfur dioxide SULF Sulfuric acid (gaseous) TERP Monoterpenes TO2 Peroxy radical from OH addition to TOL TOL Toluene and other monoalkyl aromatics XLO2 Peroxy radical from OH addition to XYL XO2 NO to NO2 conversion from alkylperoxy (RO2) radical XO2H NO to NO2 conversion (XO2) accompanied by HO2 production XO2N NO to organic nitrate conversion from alkylperoxy (RO2) radical XOPN Aromatic ring opening product (unsaturated dicarbonyl) XYL Xylene and other polyalkyl aromatics

Mechanism Updates for CB6r1 Potential updates to CB6 for CB6r1 were identified in several ways.

1. New information that had become available for some reactions, e.g., the OH + NO2 rate constant, was reviewed and evaluated.

2. The CB6 mechanism was reviewed and several corrections were identified. 3. The performance of CB6 in simulating EUPHORE experiments with several aromatic

degradation products (i.e., 2-butenedial, 4-oxo-2-pentenal, o-cresol) was evaluated and mechanism changes were developed to improve consistency between mechanism predictions and experiments. The EUPHORE experiments are particularly useful because they reported concentrations for several species including HCHO, glyoxal and HNO3.

Fourteen potential mechanism improvements were identified as discussed below.

40

1. OH + NO2 rate constant

CB6 NASA (2011)

Fall-off: 𝑘0 = 1.80 × 10−30 � 𝑇300�−3

; 𝑘∞ = 2.80 × 10−11;𝐹 = 0.6;𝑛 = 1

Mollner et al. (2010)

Fall-off: 𝑘0 = 1.48 × 10−30 � 𝑇300�−3

; 𝑘∞ = 2.58 × 10−11;𝐹 = 0.6;𝑛 = 1

In CB6, the OH + NO2 rate constant is set to the value recommended by the NASA evaluation panel in 2006 and 2011 (NASA, 2011) which is slower than the value recommended by the IUPAC evaluation panel in 2006 (IUPAC, 2006). Recently, Mollner et al. (2010) published new rate constant measurements, confirmed the previously suspected explanation for why the NASA and IUPAC recommendations differed (Donahue, 2011) and made a new recommendation for the rate constant (shown above). The NASA evaluation panel recently released a new rate constant evaluation (NASA, 2011) that made no change to the 2006 recommendation for the OH + NO2 rate constant, but did cite the work of Mollner et al. (2010). Therefore, CB6r1 retains the OH + NO2 rate constant recommended by NASA in 2006 and 2011. As discussed in Section 5 of this report, the sensitivity of CB6r1 to changing the OH + NO2 rate constant from the recommendation of NASA (2011) to Mollner et al. (2010) was evaluated as a sensitivity test.

Corrections to CB6

2. Correct products of reaction of CAO2 with NO (reaction 209)

Original CAO2 + NO = 0.86 NO2 + 0.14 NTR + 1.2 HO2 + 0.344 FORM + 0.344 CO

Revised CAO2 + NO = 0.86 NO2 + 0.14 NTR + 0.86 HO2 + 0.344 GLY

The species CAO2 was introduced in the CB05-TU mechanism before glyoxal was introduced as an explicit species in CB6 (GLY). In CB05-TU, glyoxal was represented as FORM + CO + HO2. The change shown above corrects an oversight in CB6 where the CB05-TU representation of glyoxal had not been changed to GLY. Note that this change became moot when the species CAO2 was completely eliminated by other changes discussed below.

3. Correct reactions 82 and 86

Original 82) XO2 + RO2 = 0.6 HO2 + RO2

86) XO2N + RO2 = 0.6 HO2 + RO2

Revised 82) XO2 + RO2 = RO2

86) XO2N + RO2 = RO2

CB6 reactions 82 and 86 included erroneous production of HO2 which was eliminated.

41

4. Correct reaction 112

Original ALDX = MEO2 + RO2 + CO + HO2

Revised ALDX = ALD2 + CO + HO2 + XO2H + RO2

CB6 reaction 112 (photolysis of higher aldehydes, ALDX) had the erroneous product MEO2 rather than acetaldehyde (ALD2) + HO2.

Changes to γ-dicarbonyls (e.g., 2-butenedial and 4-oxo-2-pentenal) from aromatics

5. Increase OPEN photolysis (reaction 203) from 0.028 * JNO2 to 0.08 * JNO2 based on simulations of EUPHORE chamber experiments with and NOx. The rate of OPEN photolysis is set as a ratio to JNO2 because the cross-sections are poorly characterized and the quantum yields unknown. OPEN represents all γ-dicarbonyls formed from toluene which includes both 2-butenedial and 4-oxo-2-pentenal.

6. Change product of XOPN photolysis from CAO2 to OPEN

Original XOPN = CAO2 + 0.7 HO2 + 0.7 CO + 0.3 C2O3 + RO2

Revised XOPN = OPEN + 0.7 HO2 + 0.7 CO + 0.3 C2O3 + RO2

The products of XOPN photolysis were changed to improve simulation of the EUPHORE experiment with 4-oxo-2-pentenal and NOx. The XOPN photolysis rate was unchanged at 0.05 * JNO2. The rate of XOPN photolysis is set as a ratio to JNO2 because the cross-sections are poorly characterized and the quantum yields unknown. XOPN represents all γ-dicarbonyls formed from toluene which includes 4-oxo-2-pentenal and unsaturated diketones.

7. Change reaction of OPEN with OH (reaction 204)

Original OPEN + OH = 0.6 OPO3 + 0.4 CAO2 + 0.4 RO2

Revised OPEN + OH = 0.6 OPO3 + 0.8 GLY + 0.4 XO2H + 0.4 RO2

This change is based on measurements of glyoxal formed in the EUPHORE chamber experiment with 2-butenedial and NOx.

8. Change reaction of XOPN with OH (reaction 200)

Original XOPN + OH = CAO2 + MGLY + XO2H + RO2

Revised XOPN + OH = OPEN + MGLY + XO2H + RO2

42

This change is based on simulating the EUPHORE chamber experiment with 4-oxo-2-pentenal and NOx.

9. Change rate constant for OPAN decomposition in reaction 215 to be the same as PANX decomposition in reaction 63

In CB6, the thermal decomposition of OPAN (the PAN-type compound formed from OPEN) was made slower than other PAN-type compounds (e.g., PANX) to account for a tendency for OPAN to partition into organic aerosol and be stabilized. However, this resulted in too much conversion of NO2 to OPAN when simulating the EUPHORE chamber experiment with 2-butenedial and NOx.

10. Change OH + TOL product OPEN to XOPN

Original TOL + OH = 0.18 CRES + 0.65 TO2 + 0.72 RO2 + 0.1 OPEN + 0.1 OH + 0.07 XO2H + 0.18 HO2

Revised TOL + OH = 0.18 CRES + 0.65 TO2 + 0.72 RO2 + 0.1 XOPN + 0.1 OH + 0.07 XO2H + 0.18 HO2

Prompt ring-opening following OH attachment to toluene leads to formation of a 7-carbon product that may be better represented by XOPN than OPEN.

Changes to Cresol Products

11. Change reaction of CAT1 with OH (reaction 207) and eliminate CAO2

Original CAT1 + OH = CAO2 + RO2 k = 7E-11

Revised CAT1 + OH = CRO k = 2E-10

CAT1 represents catechols that are formed in high yield from reaction of cresols with OH. The CB6 reactions for CAT1 are from CB05-TU and were based on Hu et al. (2007). The MCM v3.2 (http://mcm.leeds.ac.uk; Jenkin et al., 1997) has a faster rate constant and different products for OH reaction with CAT1. This change is expected to provide more NOx sink from cresols. This change, combined with changes 6-8, eliminated the species CAO2 and its reactions.

12. Add photolysis of nitrocresols

Add CRON = HONO + XOPN + XO2 + RO2 with J = 0.015 * JNO2

Nitrocresols are formed following reactions of cresol with NO3 or OH. Bejan et al. (2010) have reported that nitrophenols (represented by CRON in CB6) photolyze quite rapidly to form HONO with a photolysis rate within the range 0.007 to 0.0275 x JNO2. Since HONO also photolyzes to OH and NO, the implications are OH formation and recycling of nitrogen from an inactive (NOz) to and active (NOx) form. A similar reaction may be needed for dinitrocresols.

http://mcm.leeds.ac.uk/

43

13. Update to the products of glyoxal reactions with OH and NO3

Original 116) GLY + OH = 1.7 CO + 0.3 XO2 + 0.3 RO2 + HO2 118) GLY + NO3 = HNO3 + CO + HO2 + XO2 + RO2

Revised 116) GLY + OH = 1.8 CO + 0.2 XO2 + 0.2 RO2 + HO2 118) GLY + NO3 = HNO3 + 1.5 CO + 0.5 XO2 + 0.5 RO2 + HO2

The products of the OH and NO3 reactions of glyoxal were revised based on Setokuchi (2011).

14. Updates to the reactions of isoprene

There have been numerous recent publications (including Paulot et al. 2009; Peeters et al., 2009; Archibald et al, 2010; Crounse et al., 2011) on the atmospheric oxidation reactions of isoprene following the report by Lelieveld et al. (2008) of substantial OH radical production when isoprene is oxidized under low NOx conditions. The updates to isoprene reactions in CB6r1 are as follows:

1. CB6 included the 1,6-H-shift isomerization of the OH-isoprene-O2 adduct (ISO2) and in CB6r1 the products are revised to C5-hydroperoxyaldehydes (HPLD) which photolyze rapidly to produce OH radicals (Peeters et al., 2009; Archibald et al., 2010).

2. The unimolecular rate constant for isomerization of ISO2 in CB6r1 is set to the experimentally based value reported by Crounse et al. (2011).

3. The major reaction product of ISO2 with NO (ISPD) which represents methacrolein, methyl vinyl ketone (MVK) and other less abundant products was revised by condensing the reactions included in MCM v3.2 (http://mcm.leeds.ac.uk; Jenkin et al., 1997).

4. The photolysis data for ISPD were updated using IUPAC (2010) data for methacrolein and MVK.

Simulations of EUPHORE Experiments with CB6r1 The results of simulating EUPHORE experiments for 2-butenedial, 4-oxo-2-pentenal, o-cresol and toluene with CB6r1 are shown in Figure 12 through Figure 16, respectively. For 2-butenedial and 4-oxo-2-pentenal, the decay rates of OPEN and XOPN depend both upon their photolysis rates and reactions with OH and are simulated quite well (Figure 12 and Figure 13, respectively). The yields of HCHO, glyoxal and O3 are simulated quite well. The consumption of NO2 is too complete and this is an area where the mechanisms for OPEN and XOPN could be improved.

Figure 14 shows the CB6r1 simulation of the EUPHORE o-cresol experiment. This particular EUPHORE experiment is quite unusual because HONO is used to supply nearly 60 percent of the overall NOx. The photolysis of HONO thus supplies not only NO, but an equal amount of OH, thus strongly driving the HOx radical input to this system. The HONO decay is simulated well indicating that the OH input from HONO photolysis is correct. The consequent decay of o-cresol and production of HCHO are simulated well. The decay in NO2 and O3 rise are simulated

http://mcm.leeds.ac.uk/

44

well but with some overprediction of the amount of O3 formed. It would be useful to have additional experiments with o-cresol for mechanism evaluation.

The EUPHORE data provide four toluene experiments and the one with the lowest VOC/NOx ratio (EU092501) showed the poorest (too slow) simulated toluene decay rate with CB6. Figure 15 shows that CB6r1 performs well in simulating toluene-NOx experiment EU092501. This result confirms that mechanism revisions that improved performance for OPEN, XOPN and cresol also improved mechanism performance for toluene.

2-Butenedial O3

NO2 HCHO

Glyoxal

Figure 12. Model simulations with CB6r1 of EUPHORE experiment EU100201 with 2-butenedial and NOx.

45

4-Oxo-2-pentenal O3

NO2 HCHO

Glyoxal PAN

Figure 13. Model simulations with CB6r1 of EUPHORE experiment EU100301 with 4-oxo-2-pentenal and NOx.

46

o-Cresol O3

NO2 HCHO

HONO HNO3

Figure 14. Model simulations with CB6r1 of EUPHORE experiment EU100401 with o-cresol, HONO and NOx.

47

Toluene O3

NO NO2

HCHO HNO3

Figure 15. Model simulations with CB6r1 of EUPHORE experiment EU092501 with toluene and NOx.

48

Simulations of UCR Experiments with CB6r1 NOx Sink Experiments The results of NOx-sink experiments conducted with toluene, o-cresol, furan (a precursor to 2-butenedial) and isoprene are shown in Figure 16. Where multiple experiments are available, Figure 16 shows the experiment identified in Section 3 as being most useful for modeling. The effect on ozone of adding the test compound was simulated very well for o-cresol and isoprene and fairly well for toluene and furan. The change in ethene decay from adding the test compound indicates how the OH concentration was affected and was simulated very well for o-cresol and isoprene and fairly well for toluene and furan. There was little PAN data available to evaluate the model simulations but the effect on NO2 of adding the test compound was simulated well in all cases except furan. The simulation results for furan are poorer than for the other compounds indicating problems either in the reactions for furan or 2-butenedial. The simulation of a 2-butenedial experiment from the EUPHORE chamber, discussed above, also showed some performance problems but they were less than shown in Figure 16. Additional investigation is needed for furan/2-butenedial. These results suggest that CB6r1 is performing well in representing the strengths of the NOx-sinks present for toluene, o-cresol, and isoprene.

NOx Source Experiments Reactions of isopropyl nitrate (IPRN) and isobutyl nitrate (IBUN) were added to CB6 in order to simulate the NOx source experiments. The reactions with OH assumed that all of the nitrogen in the organic nitrate is released in a reactive form:

IPRN + OH --> ACET + NO2 with k = 6.2 x 10-13 exp(-230/T) cm-3 molecule s-1 IBUN + OH --> ACET + FORM + NO3 + 2 XO2 + 2 RO2 with k = 7.69 x 10-13 cm-3 molecule s-1 The photolysis reactions for IPRN and IBUN were implemented with the photolysis rate for NTR from CB6 and assuming that all of the nitrogen in the organic nitrate is released as NO2:

IPRN + hv --> ACET + HO2 + NO2 with J = JNTR IBUN + hv --> ALDX + 2 PAR + HO2 + NO2 with J = JNTR The NOx source experiment using 2-nitrophenol was simulated using nitrocresol (CRON) and with CRON photolysis added as described above:

CRON = HONO + XOPN + XO2 + RO2 with J = 0.015 * JNO2

Since all of these reactions assume that the nitrogen in the organic nitrate is released in a reactive form they represent an upper limit to the potential for NOx-recycling from organic nitrates.

The OH reaction with IPRN is slow (k = 2.9 x 10-13 cm-3 molecule s-1 at 298 K) such that virtually all of the IPRN decayed by photolysis in experiments EPA1449 A and B. This is especially the case for EPA1449B where the addition of CO scavenged most of the OH. The yield of NO2 (and O3) was simulated very well for EPA1449B supporting the proposed photolysis reaction for IPRN (Figure 17). In EPA1449A, the NO2 yield was underpredicted and the PAN yield was overpredicted indicating that the CH3COO radical concentration was overpredicted.

49

The OH reaction with IBUN (k = 7.7 x 10-13 cm-3 molecule s-1 at 298 K) is fast enough to compete with IBUN photolysis in experiment EPA1439A, but not in experiment EPA1439B where the addition of CO scavenged most of the OH. The yield of NO2 (and O3) was simulated very well for EPA1439B supporting the proposed photolysis reaction for IBUN (Figure 17). EPA1439A was the only experiment to yield information on the reaction of a simple alkyl nitrate with OH. Both NO2 and PAN are underpredicted in experiment EPA1439A. Since the reaction of OH with IBUN already 100% NOx recycling there is no way to raise the simulated yields of NO2 and PAN and the simulation of experiment EPA1439A supports 100% NOx recycling from the reaction of OH with IBUN. Additional experiments of improved design would be useful to confirm this conclusion.

Simulations of experiments with 2-nitrophenol (represented by CRON) are shown in Figure 18. The 2-nitrophenol decay is simulated well in experiments with data available. The final NO2 yields are underpredicted and the PAN yields are overpredicted indicating that the CH3COO radical concentration was overpredicted. This is not surprising because information is lacking for the organic products of 2-nitrophenol decay and they are represented simply as XOPN. O3 suffers from strong measurement interference (i.e., UV absorption) by 2-nitrophenol and even the apparent good agreement for the final O3 may be misleading. Overall, the simulation results are reasonable and support the proposed photolysis reaction for CRON but additional experiments would be useful.

Isoprene Experiments Several isoprene-NOx experiments were performed with low initial NO concentrations ([NO]0 = 6 to 25 ppb) to complement the isoprene NOx-sink experiments and supplement other isoprene-NOx experiments available from UCR that have higher initial NOx. Simulations of these experiments with CB6r1 are shown in Figure 19. Ozone concentrations are predicted well in all experiments but with some tendency to under-predict ozone at the highest [NO]0 of 25 ppb. The NO and NO2 concentrations are predicted well in all cases. The decay of isoprene in these experiments depends strongly on the OH concentration and is predicted well in all cases with only a small tendency to under predict the isoprene decay rate. There is a noticeable change in isoprene decay rate when the initial NO is depleted and CB6r1 captures this transition. These results suggest that CB6r1 is accurately predicting the OH concentration in isoprene experiments with low initial NO (6 to 25 ppb) and that the OH production from photolysis of C5-hydroperoxyaldehydes (HPLD) included in CB6r1 is consistent with the experimental data.

50

EPA 1407 Toluene EPA 1408 o-Cresol EPA 1402 Furan EPA 1446 Isoprene Ozone (ppm)

Ethene (ppm)

NO2 (ppm)

PAN (ppm)

Figure 16. Model simulations with CB6r1 of UCR NOx-sink experiments with toluene, o-cresol, furan and isoprene added to a base mixture of ethene and NOx .

51

EPA 1449A Isopropyl nitrate / CH3CHO

EPA 1449B Isopropyl nitrate / CO

EPA 1439A Isobutyl nitrate / CH3CHO

EPA 1439B Isobutyl nitrate / CO

Organic nitrate (ppm)

NO2 (ppm)

PAN (ppm)

Ozone (ppm)

Figure 17. Model simulations with CB6r1 of UCR NOx-source experiments for alkyl nitrates added to CH3CHO and H2O2 or CO and H2O2.

52

EPA 1441A 2-nitro phenol / CH3CHO

EPA 1442A 2-nitro phenol / CH3CHO

EPA 1449B 2-nitro phenol / CO

EPA 1442B 2-nitro phenol / CO

Organic nitrate (ppm)

NO2 (ppm)

PAN (ppm)

Ozone (ppm)

Figure 18. Model simulations with CB6r1 of UCR NOx-source experiments for 2-nitrophenol added to CH3CHO and H2O2 or CO and H2O2.

53

EPA1397A Isoprene [NO]0 = 25 ppb

EPA1405A Isoprene [NO]0 = 15 ppb

EPA1405B Isoprene [NO]0 = 6 ppb

Ozone (ppm)

Isoprene (ppm)

NO (ppm)

NO2 (ppm)

Figure 19. Model simulations with CB6r1 of UCR experiments for isoprene and NOx.

0.00

0.05

0.10

0.15

0.20

0 120 240 360 480

0.00

0.04

0.08

0.12

0.16

0 120 2400.00

0.02

0.04

0.06

0 60 120 180

0.0

0.1

0.2

0.3

0 120 240 360 480

0.00

0.05

0.10

0.15

0.20

0.25

0 120 2400.00

0.05

0.10

0.15

0.20

0.25

0 60 120 180

0.00

0.01

0.02

0.03

0 120 240 360 480

0.000

0.005

0.010

0.015

0.020

0 120 2400.000

0.002

0.004

0.006

0.008

0 60 120 180

0.000

0.005

0.010

0.015

0.020

0 120 240 360 480

0.000

0.004

0.008

0.012

0 120 2400.000

0.001

0.002

0.003

0.004

0 60 120 180

54

5.0 Performance Evaluation of CB6r1 The revised CB6 mechanism (CB6r1) was evaluated using the same data and procedures used previously to evaluate CB6 (Yarwood et al., 2010).

A singular challenge for mechanism development and evaluation is uncertainty in the rate constant for the reaction of OH with NO2 forming nitric acid (HNO3) which exerts a strong influence on mechanism predictions because it influences both radical concentrations and the lifetime of NOx. CB6 uses a rate constant for this reaction recommended by the NASA-JPL evaluation panel (NASA, 2011) but the IUPAC evaluation panel (IUPAC, 2006) recommends a rate constant that is 12% faster at room temperature and pressure. Recently, Mollner et al. (2010) performed new experimental measurements and derived a rate constant 13% slower than recommended by NASA at room temperature and pressure. To assess the importance of this uncertainty CB6r1 has been evaluated with the OH + NO2 recommended both by NASA (2011) and Mollner et al. (2010). At this time, and until new recommendations emerge from NASA and/or IUPAC, the OH + NO2 rate constant recommended by NASA (2011) is used in CB6r1.

Hierarchical Approach to Mechanism Evaluation A comprehensive atmospheric chemical mechanism, such as CB6, is comprised of numerous components that describe individual compounds such as carbon monoxide (CO), acetaldehyde or groups of compounds such as 1-alkenes (OLE). Some components depend upon others, for example when OLE reacts it forms acetaldehyde which in turn forms CO. As a result, interactions between mechanism components make it difficult to test each component in isolation and systematically evaluate the entire chemical mechanism while minimizing compensating errors between components. An approach to dealing with this challenge is the concept of hierarchical mechanism evaluation (Whitten, 1983) which was used in this project for evaluating CB6 as shown in Figure 20. For example, the CO chemistry in CB6 was evaluated first using CO-NOx chamber experiments. Second, FORM chemistry was evaluated using HCHO-NOx chamber experiments while building on the evaluated CO chemistry. The ALD2 and PAN chemistries were evaluated together using CH3CHO-NOx experiments. The chemistries of ALDX (a model species similar to ALD2 but for higher aldehydes (e.g., CH3CH2CHO (propanal))) and PANX (a species similar to PAN but formed from higher aldehydes) were indirectly evaluated by using chamber experiments of terminal and internal alkenes (OLE and IOLE) due to lack of suitable chamber data.

55

Figure 20. Hierarchy of species for evaluating CB6 systematically.

Data Used to Evaluate CB6r1 The CB6 mechanism was evaluated by simulating chamber experiments in which mixtures of VOC and NOx were irradiated to form ozone. A database of experiments compiled by UCR (http://www.cert.ucr.edu/~carter/SAPRC/SAPRCfiles.htm) containing about 2000 experiments was used (version of April 23, 2010). The database includes various types of chamber experimental data produced at UCR and the Tennessee Valley Authority (TVA). An overview of various environmental chambers at UCR and TVA is given in Table 8. There were 7 environmental chambers at UCR (EC, ETC, OTC (outdoor), DTC, XTC, CTC and EPA) and 1 chamber at TVA.

http://www.cert.ucr.edu/~carter/SAPRC/SAPRCfiles.htm

56

Table 8. Environmental chambers at UCR and TVA used for mechanism evaluation.

Chamber Chamber ID Reactor

type Reactor

volume (m3) Light source Relative humidity

Operation period

Indoor chamber Evacuable Chamber at UCR EC single ~5.8 xenon arc ~50% 1975-84 Ernie's Teflon Chamber at UCR

ETC single ~3.0 blacklight dry (< 5%) 1989-93

Dividable Teflon Chamber at UCR

DTC dual ~5.0 (X 2) blacklight dry (< 5%) 1993-99

Xenon arc Teflon Chamber at UCR

XTC single ~5.0 xenon arc dry (< 5%) 1993

CE-CERT Teflon Chamber at UCR

CTC (runs 11-82)

single ~5.0 xenon arc dry (< 5%) 1994-95

CE-CERT Teflon Chamber at UCR (rebuilt)

CTC (runs 82+)

dual ~2.5 (X 2) xenon arc dry (< 5%) 1995-99

UCR EPA chamber EPA dual ~90 (X 2) argon arc/ blacklight

dry (< 1%) 2003-present

TVA indoor chamber TVA single ~28 3 types including blacklight

~15% 1993-95

Outdoor chamber Outdoor Teflon Chamber at UCR

OTC dual ~20 (X 2) sunlight dry (< 5%) 1992-93

Screening criteria were applied to select experiments that minimize uncertainties due to high initial NOx concentration ([NOx]0), low production of O3 relative to the initial NO (Max(O3)/[NO]0), or the chamber light source (whether a blacklight light source was used or not). The criteria were applied as follows:

Criteria generally applied for selecting single compound - NOx experiments:

1. Max(O3)/[NO] 0 >= 1. 2. 10 ppb ≤ [NOx] 0 < 300 ppb 3. Exclude blacklight-used experiments when an aromatic compound (e.g., toluene) was

injected.

Criteria generally applied for selecting VOC mixture - NOx experiments:

1. Max(O3)/[NO] 0 >= 1. 2. 10 ppb ≤ [NOx] 0 < 200 ppb. 3. Exclude blacklight-used experiments when an aromatic compound was injected.

After applying these criteria, with some exceptions stated as notes in Table 9, 194 experiments of “single test compound – NOx” or “special mixture – NOx” and 145 experiments of “surrogate mixture – NOx” were selected from around 2000 experiments in the UCR chamber database (Table 9 and Table 10). Table 9 summarizes 194 chamber experiments used for testing single components of CB6 (e.g., CO) and Table 10 summarizes 145 surrogate-mixture experiments used for testing interactions of various components of CB6 and testing the performance of CB6 in simulating VOC mixtures.

57

Table 9. Summary of 194 chamber experiments for single test compounds and special mixtures.

Group CB6

Species Test compounda Experiment typeb

Number of experiments per

compound [total (blacklight)]c

Number of experiments per

CB6 species [ total (mixture)]d Note

1 CO CO single 33 33 2 FORM HCHO single 9 9 3 CH4 CH4 - 0 0 4 MEOH CH3OH IR 2 (2) 2 (2) 1 5 ETH ethene single 11 11 2 6 ALD2 CH3CHO single 8 8 7 ETOH C2H5OH IR 3 (3) 3 (3) 3 8 ACET CH3C(O)CH3 single 4 4 9 KET CH3C(O)C2H5 (MEKe) single 2 2 10 ETHA ethane IR 5 (5) 5 (5) 4 11 ALDX higher aldehydes - 0 0 5 12 PAR n-butane single 3 5 6 n-butane/2,3-dimethyl butane mixture 2 13 OLE propene single 47 48 1-butene single 1 14 IOLE trans-2-butene single 3 3 15 TOL toluene single 18 20 ethylbenzene single 2

16 XYL o-xylene (o-XYL) single 4 27

m-XYL single 15 p-XYL single 2

123-trimethyl benzene (123-TMB) single 2

124-TMB single 2 135-TMB single 2 17 ISOP isoprene single 6 6 18 TERP α-pinene single 1 2 7 β-pinene single 1 19 PRPA propane IR 2 (2) 2 (2) 8 20 BENZ benzene single 2 2

21 ETHY ethyne (acetylene) single 2 2

Total 194 (12) 194 (12) a A test compound was injected as a “single” test compound with NOx (and optionally with CO), injected excessively relative to

other co-injected compounds (e.g., in “Incremental Reactivity (IR)” style experiments), or injected with other related compounds (e.g., as an alkane “mixture”). “Opt” is “optional” PRPA, BENZ, ACTY and BOLE are optional (Opt) model species for propane, benzene, acetylene and branched olefins (e.g., isobutene), respectively.

b“Single”, “IR” and “mixture” mean “injected as a single test compound”, “injected in an IR style”, and “injected as a mixture with other closely related compounds (e.g., as an alkene mixture)”.

c Total number of blacklight experiments in the parentheses. d Total number of selected experiments for each CB species (e.g., ALD2, PAR and XYL) and total number of “test compound –

other VOCs – NOx” experiments in the parentheses. 1Only two blacklight/mixture type experiments were available for testing the MEOH chemistry of CB6. 2 22 blacklight experiments were also used to compare ETH performance for non-blacklight and blacklight experiments. 3 Only three blacklight/mixture type experiments were available for testing the ETOH chemistry of CB6. 4 Only five blacklight/mixture type experiments were available for testing the ETHA chemistry of CB6. 5 No experiment was available for specifically testing the ALDX chemistry of CB6. 1-Butene – NOx experiments can be indirectly

used.

58

6 For testing the PAR chemistry of CB6, two mixture-type experiments where n-butane and 2,3-dimethyl butane were both injected were used as well as 3 experiments where n-butane was injected in presence of NOx (NO and NO2).

7 Blacklight experiments were also used to provide supplementary information. 14 blacklight terpene – NOx experiments were used: α-pinene (4), β-pinene (1), 3-carene (3), d-limonene (3) and sabinene (3).

8 Only two blacklight/mixture type experiments were used for testing the PRPA (propane) chemistry of CB6.

Table 10. Summary of 145 surrogate mixture experiments.

Group Descriptiona Subgroupb VOC/NOx ratio and

initial [NOx]o Number of

experiments

Group 1 Incomplete surrogate without aromatics (Surg-NA) Surg-NA Vary

Variable VOC/NOx; [NOx]0 < 200 ppb 2

Group 2 Incomplete surrogate but with TOL or XYL - - Sub-total: 57

Surg-7 MIR2 Low VOC/NOx; [NOx]0 < 100 ppb 21

Surg-7 LN2 [NOx]0 < 100 ppb 26 ECsrg-7 [NOx]0 < 100 ppb 2 TVA srg-1 [NOx]0 ~ 50 ppb 8 Group 3 Full surrogate Sub-total: 86

Surg-8 MIR2 Low VOC/NOx; [NOx]o < 100 ppb 10

Surg-8 LN1 100 ppb < [NOx]0 < 200 ppb 19

Surg-8 LN2 [NOx]0 < 100 ppb 9

Surg-8 Vary Variable VOC/NOx; [NOx]0 < 200 ppb 43

TVA srg-2 [NOx]0 < 100 ppb 5 Total 145

a Surg-8, Surg-7, Surg-NA, TVAsrg-2 and ECsrg-7 experiments contains at least 7 different VOCs, at least one in each class (alkanes, alkenes, aromatics).

b Surg-8: 8-component VOC mixture (n-butane, n-octane, ethene, propene, trans-2-butene, toluene, m-xylene, HCHO) with NOx. Surg-7: Surg-8 without HCHO. Surg-NA: Surg-8 without aromatics (toluene, m-xylene) and HCHO. TVAsrg-1: mixtures of n-butane, 2-methyl butane, ethene, propene, toluene and HCHO. TVAsrg-2: complex mixtures of alkanes, alkenes and aromatics. ECsrg-7: EC chamber experiments using 7-component surrogate (n-butane, 2,3-dimethyl butane, ethene, propene, t-2-butene, toluene, m-xylene). MIR1, MIR2, LN1, LN2 and Vary are acronyms stating experimental conditions related to the VOC/NOx ratio and initial NOx level as follows: MIR1: Low VOC/NOx, MIR (maximum incremental reactivity)-like conditions. NOx 300-500 ppb. MIR2: Low VOC/NOx, MIR-like conditions, NOx < 100 ppb. LN1: Lower NOx conditions, NOx >100 ppb. LN2: Lower NOx, NOx < 50 ppb. Vary: Non-standard ROG/NOx. Conditions varied. Chamber simulations were performed using the SAPRC software that has been used for evaluating several mechanisms (Carter, 2010; Yarwood et al., 2010; Whitten et al., 2010). Performance metrics that were used for evaluating CB6 include the following: The maximum ozone concentration (Max(O3)), Maximum D(O3-NO) (i.e., maximum (([O3]-[NO])t - ([O3]-[NO])t0)), and NOx crossover time (i.e., the time when the NO2 concentration becomes equal to the NO concentration). Means and standard deviations of these metrics were used to characterize performance over multiple experiments, especially performance against surrogate mixture experiments. Positive values for Max(O3) and Maximum D(O3-NO) indicate a tendency to over predict formation of O3 and/or oxidation of NO. Positive values for NOx crossover time

59

indicate a late crossover prediction because formation of O3 and/or oxidation of NO initially proceeded too slowly.

Running chamber simulations requires a wall mechanism to characterize chamber-dependent effects such as radical sources and NOx offgasing. For CB6, wall mechanisms that were used for evaluating SAPRC-07 by William Carter at UCR were used based on two facts: (1) the rate constant for reaction OH + NO2 = HNO3, the most important radical sink under most chamber conditions, is the same in CB6 and SAPRC-07; (2) chamber simulation results for CO - NOx experiments were similar between CB6 and SAPRC-07. The CO - NOx chemical system is highly sensitive to chamber-dependent radical sources because CO does not photolyze or produce products that photolyze. In contrast, chemical systems (such as propene - NOx and surrogate mixture - NOx) that have significant internal radical sources (e.g., photolysis of aldehydes) are relatively insensitive to chamber-dependent radical sources.

CB6r1 Evaluation Results The performance of CB6 and CB6r1 in simulating 339 chamber experiments is given numerically in Table 11 and summarized graphically in Figure 21, Figure 22 and Figure 23 for the performance metrics Max(O3), Max(D(O3-NO)) and the NOx crossover time, respectively. Mean model errors for Max(O3) and Max(D(O3-NO)) were calculated as {(modeled - experimental)/experimental} and expressed as percentages. For the NOx crossover time mean model errors were calculated as (modeled - experimental) and expressed as minutes.. The VOC composition for surrogate mixture experiments is as follows: Surg-NA mixtures are incomplete surrogate mixtures without toluene, xylene or formaldehyde; Surg-Inc (Surg-incomplete) mixtures are incomplete surrogate mixtures containing at least one of toluene or xylene; Surg-Full mixtures are full surrogate mixtures that contain at least 8 different VOCs (n-butane, n-octane, ethene, propene, trans-2-butene, toluene, m-xylene, formaldehyde) with NOx. The Surg-NA, Surg-Inc and Surg-Full experiments are Group 1, Group 2 and Group 3 in Table 11.

60

Table 11. Model errors for Max(O3), Max(D(O3-NO)) and the NOx crossover time.

CB6 species Number of Expts.

Max(O3) [%] Max(D(O3-NO) [%] NOx crossover [min] CB6 a CB6r1 CB6 CB6r1 CB6 CB6r1

CO 33 11 (31) 10 (30) 6 (25) 5 (25) -10 (23) -9 (23) FORM 9 -5 (5) -5 (5) -4 (4) -4 (4) 0 (3) 0 (3) MEOH 2 -6 (10) -3 (9) -4 (6) -2 (6) 26 (1) 16 (0) ETH 11 -12 (16) -13 (17) -9 (10) -9 (10) 8 (18) 8 (18) ALD2 8 -6 (9) -6 (9) -4 (6) -4 (6) 2 (6) 2 (6) ETOH 3 9 (14) 21 (16) 3 (5) 7 (5) 17 (4) 8 (5) ACET 4 2 (8) 3 (8) 2 (7) 3 (6) 2 (4) 2 (4) KET 2 -12 (14) -13 (13) -8 (8) -9 (7) 18 (15) 18 (15) ETHA 5 10 (42) 13 (41) 3 (29) 5 (28) 11 (15) 6 (14) PAR 5 -20 (37) -27 (33) -15 (30) -21 (27) 35 (9) 36 (9) OLE 48 -7 (16) -9 (15) -6 (11) -7 (10) 6 (14) 7 (15) IOLE 3 14 (6) 10 (8) 11 (5) 7 (7) 7 (10) 9 (11) TOL 20 -11 (15) -1 (15) -10 (12) -2 (12) 22 (20) -1 (14) XYL 27 -9 (12) -6 (14) -6 (9) -4 (10) 16 (39) 5 (36) ISOP 6 1 (18) 20 (13) 0 (14) 13 (9) 6 (5) -2 (4) TERP 2 13 (37) 10 (39) 7 (22) 5 (24) -30 (74) -30 (74) PRPA 2 20 (24) 16 (23) 13 (16) 10 (15) 11 (7) 6 (6) BENZ 2 -10 (2) -12 (3) -6 (1) -7 (2) -15 (2) -39 (4) ETHY 2 -44 (26) -63 (24) -36 (21) -52 (19) 60 (8) 66 (9) Surg-NA 2 3 (16) 0 (18) 1 (13) -1 (15) 4 (7) 4 (7) Surg-inc 57 -15 (14) -11 (14) -14 (14) -11 (13) 12 (10) 8 (8) Surg-Full 86 -21 (12) -18 (11) -18 (10) -15 (10) 9 (13) 6 (11)

a Mean error with standard deviations given in parentheses.

Figure 21. CB6 and CB6r1 model errors (%) for Max(O3).

-100

-80

-60

-40

-20

0

20

40

60

80

100

Erro

r in

Max

(O3)

, %

CB6 CB6r1

61

Figure 22. CB6 and CB6r1 model errors (%) for Max(D(O3-NO)).

Figure 23. CB6 and CB6r1 model errors (minutes) for NOx crossover time.

The mechanism updates from CB6 to CB6r1 were mainly for the aromatic hydrocarbons (TOL, XYL and BENZ), isoprene (ISOP) and organic nitrates (NTR and CRON). In addition, there were

-100

-80

-60

-40

-20

0

20

40

60

80

100Er

ror i

n M

ax(D

(O3-

NO

)), %

CB6 CB6r1

-100

-80

-60

-40

-20

0

20

40

60

80

100

Erro

r in

NO

x cr

osso

ver t

imes

, min

utes

CB6 CB6r1

62

several mechanism corrections including corrections to the products of reactions 82 (XO2 + RO2), 86 (XO2N + RO2) and 112 (ALDX photolysis). Performance changes for several CB6 species that were not updated (e.g., MEOH, ETOH, ETHA, PRPA, PAR, OLE and IOLE) are attributable to these mechanism corrections. For the aromatics TOL and XYL, CB6r1 improved the model performance for all three metrics. CB6r1 also improved performance for surrogate mixtures that included aromatics (Surg-Inc and Surg-Full) although a tendency to under predict ozone production remains. For BENZ, CB6 tended to under predict ozone formation but was early on the NOx crossover time and CB6r1 degraded this performance. For ISOP, CB6 already performed very well and CB6r1 degraded performance by over predicting ozone production although the performance for NOx crossover time improved. However, we attach limited weight to these isoprene evaluation results from only 6 experiments at moderately high initial NOx and note that CB6r1 performed well in simulating experiments with low initial NOx as described in Section 4. The changes to NTR will have little impact on the mechanism evaluation because NTR reacts very slowly and is essentially a terminal product under the conditions of the evaluation experiments (Table 9 and Table 10). The changes to CRON (nitrocresols) are part of other changes to aromatics chemistry already discussed for TOL, XYL and BENZ.

There were changes in model performance for other species such as alcohols, alkanes, alkenes and ethyne. The changes for alcohols, alkanes and alkenes are attributed to correcting two errors in CB6 for the products of (1) the reaction of XO2 and XO2N with RO2 and (2) photolysis of ALDX, as discussed above. The performance of CB6r1 for PAR was poorer than CB6 but there are only 5 experiments for PAR (using n-butane and 2,3-dimethyl butane) and both CB6r1 and CB6 performed very well for mixtures of alkanes with alkenes (Surg-NA). Model performance for ethyne (ETHY) was degraded by changes to glyoxal chemistry, however the main uncertainty for ethyne (and glyoxal) is considered to be the quantum yields for glyoxal photolysis.

The overall evaluation of CB6r1 compared to CB6 shows some performance gains (aromatics, mixtures containing aromatics), some losses (isoprene, acetylene) and small changes for other species. As discussed above, we attach limited weight to the isoprene evaluation results from only 6 experiments at moderately high initial NOx because CB6r1 performed well in simulating experiments with low initial NOx. Therefore, CB6r1 should be preferred over CB6 (and both mechanisms should be preferred over CB05).

The changes in model performance with CB6r1 that are expected to be most important for photochemical modeling are (1) improved performance for ozone production from toluene and xylenes; (2) changes in ozone production from isoprene that improved performance at low initial NOx but degraded performance at higher initial NOx. Changing the products of the OH reaction of alkyl nitrates (NTR) to produce NO2 is expected to increase ozone production regionally and may be an important change in regions where ozone formation is NOx limited.

Impact of Uncertain OH + NO2 Rate Constant As discussed above, the rate constant for the reaction of OH + NO2 --> HNO3 is uncertain with three current recommended values varying by 25% at room temperature and pressure. CB6 uses a rate constant at the middle of this range as recommended by the NASA-JPL evaluation panel (NASA, 2011). The IUPAC evaluation panel (IUPAC, 2006) recommends a rate constant 12% faster whereas Mollner et al. (2010) recommend a rate constant 13% slower than the NASA-JPL recommended value at room temperature and pressure. To assess the importance of

63

this uncertainty, CB6r1 was evaluated with the OH + NO2 recommended by Mollner et al. (2010) as a sensitivity. At this time, and until new recommendations emerge from the NASA-JPL and/or IUPAC panels, the OH + NO2 rate constant recommended by NASA (2011) should be used in CB6r1.

The rate constant for the reaction OH + NO2 --> HNO3 exerts a strong influence on mechanism predictions because it influences both radical concentrations and the lifetime of NOx. A consequence for simulating chamber experiments is that the chamber wall mechanism (describing chamber sources of radicals and NOx) must be adjusted to work with the selected OH + NO2 rate constant (Mollner et al., 2010). The chamber wall mechanism was adjusted for the alternate OH + NO2 rate constant to minimize bias for CB6 in simulating CO experiments that are very sensitive to the chamber wall mechanism.

The performance of alternate CB6 and CB6r1mechanisms using the OH + NO2 rate constant of Mollner et al. (2010) in simulating 339 chamber experiments is given numerically in Table 12 and summarized graphically in Figure 24, Figure 25 and Figure 26. Overall, changing OH + NO2 rate constant did not substantially alter mechanism performance indicating that revising the chamber wall mechanism (as discussed above) compensated for the reduced radical sink provided by a slower OH + NO2 rate constant. The changes in bias for Max(O3) for surrogate mixture experiments were less than 4% in all cases (compare Table 12 and Table 11). These results cannot be used to decide which value for the OH + NO2 rate constant is more accurate. There were changes in CB6r1 performance for individual species, most notably CO, PAR and XYL. The change in performance for CO is because the chamber wall mechanism was adjusted for the alternate case (Table 12) but used as provided for the standard case (Table 11) resulting in reduced bias for the alternate case. The change in performance for PAR follows the change for CO because PAR (like CO) depends strongly upon the chamber wall mechanism. The change in performance for XYL contrasts with other highly reactive species that changed little (alkenes and aldehydyes) and suggests that it would be useful to investigate how the OH + NO2 reaction influences the XYL chemistry in CB6r1.

Table 12. Model errors for Max(O3), Max(D(O3-NO)) and the NOx crossover time with an alternate OH + NO2 rate constant.

CB6 species Number of Expts.

Max(O3) [%] Max(D(O3-NO) [%] NOx crossover [min] CB6 a CB6r1 CB6 CB6r1 CB6 CB6r1

CO 33 0 (26) -2 (25) -2 (22) -4 (21) -3 (22) 5 (22) FORM 9 -2 (6) -3 (6) -2 (5) -2 (5) -1 (3) -1 (3) MEOH 2 7 (6) -8 (10) 5 (4) -5 (6) 23 (1) 33 (1) ETH 11 -8 (15) -10 (16) -6 (9) -7 (10) 8 (17) 13 (19) ALD2 8 -6 (9) -6 (9) -4 (7) -5 (7) 2 (6) 3 (6) ETOH 3 33 (17) 25 (14) 12 (5) 9 (5) 14 (4) 23 (4) ACET 4 6 (7) 5 (7) 5 (6) 5 (6) 1 (4) 2 (4) KET 2 -12 (11) -14 (12) -8 (6) -10 (7) 18 (15) 19 (16) ETHA 5 15 (44) 6 (36) 6 (30) 1 (26) 10 (15) 17 (17) PAR 5 -31 (34) -39 (29) -25 (28) -32 (24) 46 (14) 53 (16) OLE 48 -6 (16) -8 (15) -5 (11) -7 (10) 5 (13) 10 (15) IOLE 3 18 (6) 14 (7) 14 (5) 10 (6) 5 (10) 11 (11) TOL 20 -10 (15) -3 (15) -9 (13) -4 (12) 23 (19) 9 (15) XYL 27 -7 (12) -23 (15) -5 (8) -15 (11) 16 (39) 44 (45) ISOP 6 2 (18) 22 (13) 1 (13) 14 (9) 7 (5) 1 (4)

64

TERP 2 14 (36) 11 (39) 8 (22) 6 (24) -30 (74) -29 (75) PRPA 2 33 (18) 14 (20) 21 (12) 9 (13) 11 (6) 19 (5) BENZ 2 -9 (2) -11 (3) -6 (1) -7 (2) -15 (2) -37 (4) ETHY 2 -41 (28) -65 (25) -33 (23) -55 (20) 65 (11) 83 (15) Surg-NA 2 5 (14) 3 (16) 3 (12) 1 (13) 3 (6) 4 (7) Surg-inc 57 -13 (14) -15 (14) -12 (14) -14 (14) 11 (9) 16 (11) Surg-Full 86 -19 (11) -21 (11) -16 (10) -18 (9) 8 (12) 11 (14)

a Mean error with standard deviations given in parentheses.

Figure 24. CB6 and CB6r1 model errors (%) for Max(O3) with an alternate OH + NO2 rate constant.

-100

-80

-60

-40

-20

0

20

40

60

80

100

Mod

el e

rror

s of

Max

(O3)

, %

CB6 CB6r1

65

Figure 25. CB6 and CB6r1 model errors (%) for Max(D(O3-NO)) with an alternate OH + NO2 rate constant.

Figure 26. CB6 and CB6r1 model errors (minutes) for NOx crossover time with an alternate OH + NO2 rate constant.

-100

-80

-60

-40

-20

0

20

40

60

80

100M

odel

err

ors

of M

ax(D

(O3-

NO

)), %

CB6 CB6r1

-120

-100

-80

-60

-40

-20

0

20

40

60

80

100

Mod

el e

rror

s of

NO

x cr

osso

ver t

imes

, min

CB6 CB6r1

66

The main conclusions from evaluating the sensitivity of mechanism performance to the OH + NO2 rate constant are that (1) either the OH + NO2 rate constant recommended by NASA (2011) or Mollner et al. (2010) can be used with CB6r1 and (2) the results of evaluating CB6r1 with alternate OH + NO2 rate constants cannot be used to decide which value is more accurate. At this time, and until new recommendations emerge from the NASA-JPL and/or IUPAC panels, the OH + NO2 rate constant recommended by NASA (2011) should be used in CB6r1.

67

6.0 Conclusions and Recommendations Environmental chamber experiments of novel design were performed in this study using the EPA chamber at UCR which is capable of performing experiments at low NOx concentrations of only a few ppb. The new experiments investigated NOx sinks in the atmospheric reactions of toluene, isoprene and aromatic oxidation products, and NOx sources (i.e., processes that recycle organic nitrates to NOx) involving reactions of representative oxidation products formed in the reactions of alkanes and aromatics. In addition, several isoprene - NOx experiments were performed with low initial NO concentrations (6 to 25 ppb) that are useful to test the performance of isoprene mechanisms at atmospherically-relevant NOx concentrations.

The importance of NOx sinks was demonstrated experimentally from the effect on O3 formation of adding representative compounds to alkene – NOx mixtures. Addition of toluene and isoprene reduced O3 formation from ethene - NOx and (in the case of toluene) propene - NOx mixtures, demonstrating the importance of NOx sinks for toluene and isoprene. Even larger O3 reductions, and therefore larger NOx sinks, were observed when o-cresol and furan (a precursor to 2-butenedial, which is a major ring-opening product of toluene) were added to ethene - NOx mixtures, indicating that these products play an important role in producing the NOx sink observed for toluene.

NOx recycling was demonstrated experimentally from representative alkyl nitrates and the representative aromatic oxidation product 2-nitrophenol in experiments where formation of NO2 and/or PAN was observed when these compounds were reacted in the absence of injected NOx. The amounts of NOx formed were substantially greater than can be attributed to any artifact of the chamber walls.

Several experiments that are relevant to the characterization of NOx sinks in the chemistry of aromatic hydrocarbons were obtained from the European Photo-Reactor (EUPHORE). A total of 9 experiments were obtained and analyzed including 2 chamber characterization experiments and 7 experiments related to aromatics (4 for toluene, 1 for each of o-cresol, 2-butenedial and 4-oxo-2-pentenal). These experiments are now part of the database of chamber experiments maintained by UCR and will be available for use in future projects.

The experimental data obtained were used to test and to improve the mechanisms for isoprene and aromatics in version 6 of the Carbon Bond mechanism (CB6). The revised mechanism is called CB6r1. CB6r1 performed better than CB6 in simulating experiments for toluene, xylenes and mixtures combining aromatics with other VOCs. However, mechanism performance for simulating VOC mixtures remains poorer when the mixtures contain aromatics demonstrating that some aspects of the aromatics chemistry are still not fully understood. Simulations of EUPHORE experiments with 2-butenedial and 4-oxo-2-pentenal (degradation products of aromatics) performed poorly suggesting an area where the aromatics chemistry could be improved.

The experimental data obtained in this study strongly support the occurrence of NOx-recycling in the photolysis reactions of the NOx-source compounds isopropyl nitrate, isobutyl nitrate and 2-nitrophenol. Accordingly, CB6r1 includes NOx-recycling from photolysis of NTR (representing alkyl nitrates) and CRON (representing nitrocresols). The results of this study suggest that OH-

68

reactions of isopropyl and isobutyl nitrate also lead to NOx recycling but the results are not conclusive because, under our experimental conditions, the isopropyl nitrate and isobutyl nitrate were mainly consumed by photolysis rather than OH reaction. CB6r1 tentatively includes NOx recycling from the reaction of OH with NTR (a change from CB6) but we recommend conducting photochemical model sensitivity tests to evaluate the impacts of this change. Additional experiments to test for the occurrence of NOx recycling from alkyl nitrates larger than isopropyl and isobutyl are needed because OH reaction is more important relative to photolysis and the nitrogen-containing products could be different for larger alkyl nitrates than for the compounds studied here.

In the course of developing CB6r1 several errors in CB6 were identified and corrected. The corrected CB6 mechanism is implemented in CAMx version 5.40 and documented in the User’s Guide available from www.camx.com. These mechanism corrections were to the products of reactions 82 (XO2 + RO2), 86 (XO2N + RO2) and 112 (ALDX photolysis).

The rate constant for the reaction of OH with NO2 is uncertain with three current recommended values varying by 25% at room temperature and pressure. CB6r1 uses a rate constant at the middle of this range as recommended by the NASA-JPL evaluation panel (NASA, 2011). To assess the importance of this uncertainty, CB6r1 was evaluated with a slower OH + NO2 rate constant as recommended by Mollner et al. (2010) as a sensitivity test. The main conclusions were that (1) either the rate constants recommended by NASA (2011) or Mollner et al. (2010) can be used with CB6r1 and (2) the results of evaluating CB6r1 with alternate OH + NO2 rate constants should not be used to decide which value is more accurate. At this time, and until new recommendations emerge from the NASA-JPL and/or IUPAC panels, the OH + NO2 rate constant recommended by NASA (2011) is recommended for use in CB6r1.

We make the following recommendations for additional chamber experiments and other activities to support improvements in chemical mechanisms:

1. Chamber experiments are needed to quantify NOx recycling from alkyl nitrates larger than isopropyl nitrate and isobutyl nitrate (studied here). The experimental techniques developed to study NOx source compounds should be used again. Larger alkyl nitrates should be studied because, compared to smaller alkyl nitrates, reaction with OH will become more important than photolysis and the products of reaction with OH could be different for large alkyl nitrates.

2. Chamber experiments performed at the Euphore chamber proved to be useful for evaluating and improving toluene mechanisms. Euphore data for additional aromatic hydrocarbons, e.g. for benzene, xylenes and trimethylbenzenes, are available and should be obtained and used for mechanism evaluation/development.

3. Chamber experiments should be performed with unsaturated dicarbonyls that are degradation products of aromatic hydrocarbons, e.g., 2-butenedial and 4-oxo-2-pentenal. Experiments are needed at low to moderate initial NO concentrations for a range of VOC/NOx ratios. Experiments that include a tracer to measure OH-radical production would be valuable. NOx-sink experiments, using the methods developed in this study, would be valuable. It would be useful to measure separately the yields of PAN and total PAN-type compounds to quantify the amount of PAN-analogues formed and how quickly they decay.

http://www.camx.com/

69

4. Chamber data for acetylene and glyoxal are very limited but the available data suggest that the mechanisms are biased or incomplete. New experiments in the EPA chamber for acetylene and glyoxal would be useful to reduce uncertainty in their mechanisms. The glyoxal mechanism is especially important because it is a reactive product of aromatics and isoprene.

5. Chamber experiments with surrogate hydrocarbon mixtures are important to evaluate how well mechanisms simulate urban atmospheres. Experiments with mixtures also test interactions between the mechanisms for individual mixture components. New chamber experiments should be performed using surrogate hydrocarbon mixtures designed to reflect conditions relevant to Houston (urban emissions with chemical industries) and Dallas (urban emissions with oil and gas production).

6. Ethene, propene, 1,3-butadiene, 1-butene, isobutene, trans-2-butene, cis-2-butene are regulated in Texas as HRVOCs. Data available for mechanism evaluation are extensive for ethene and propene but limited or missing for the other HRVOCs. Chamber experiments are recommended for 1-butene, isobutene, trans-2-butene, cis-2-butene, and 1,3-butadiene in order to develop/evaluate more explicit mechanisms for each compound. Improved chemical mechanisms for HRVOCs will provide more accurate estimation of the impacts of HRVOC emissions on ozone formation in Houston and other regions.

7. The CB6r1 mechanism should be implemented in air quality models (i.e. CAMx and CMAQ) for further testing and evaluation under ambient conditions.

70

7.0 References Archibald,A.T., Cooke,M.C., Utembe,S.R., Shallcross,D.E., Derwent,R.G., and Jenkin,M.E. (2010):

“Impacts of mechanistic changes on HOx formation and recycling in the oxidation of isoprene,” Atmos. Chem. Phys., 10, 8097-8118, doi:10.5194/acp-10-8097-2010.

Ball, S. M., J. M. Langridge, and R. L. Jones (2004): “Broadband cavity enhanced absorption spectroscopy using light emitting diodes,” Chemical Physics Letters 398, 68-74.

Behrentz, E., L. D. Sabin, A. M. Winer, D. R. Fitz, D. V. Pankratz, S. D. Colome, and S. A. Fruin (2005): “Relative importance of school bus-related microenvironments to children’s pollutant exposure,” Journal of the Air & Waste Management Association 55(10), 1418-1430.

Bertram, T. H., R. C. Cohen, W. J. Thorn III and P. M. Chu (2005): “Consistency of ozone and nitrogen oxides standards at tropospherically relevant mixing ratios,” Journal of the Air & Waste Management Association 55(10), 1473-1479.

Bejan, I., G.Villena, S. Springmann, H. Sonderfeld, T. Benter, I. Barnes, R. Koppmann, P. Wiesen, and J. Kleffmann (2010): “The Photolysis of Different Nitroaromatics: A Daytime Source of HONO,” Presented at the Atmospheric Chemical Mechanisms conference, Davis, CA. Available at: http://airquality.ucdavis.edu/pages/events/2010/acm/BEJAN.pdf

Bloss, C., Wagner, V., Bonzanini, A., Jenkin, M.E., Wirtz, K., Martin-Reviejo, M., Pilling, M.J., (2005a): Evaluation of detailed aromatic mechanisms (MCMv3 and MCMv3.1) against environmental chamber data. Atmospheric Chemistry and Physics 5, 623-639. (http://www.atmos-chem-phys.net/5/623/2005/acp-5-623-2005.pdf)

Bloss, C., Wagner, V., Jenkin, M.E., Volkamer, R., Bloss, W.J., Lee, J.D., Heard, D.E., Wirtz, K., Martin-Reviejo, M., Rea, G., Wenger, J.C., Pilling, M.J., (2005b): Developments of a detailed chemical mechanism (MCMv3.1) for the atmospheric oxidation of aromatic hydrocarbons. Atmospheric Chemistry and Physics 5, 641-664. (http://www.atmos-chem-phys.net/5/641/2005/acp-5-641-2005.pdf)

Burrows, J.P., A. Dehn, B. Deters, S. Himmelmann, A. Richter, S. Voigt and J. Orphal (1998): “Atmospheric remote-sensing reference data from GOME: Part I. Temperature-dependent absorption cross-sections of NO2 in the 231-794 range,” Journal of Quantitative Spectroscopy & Radiative Transfer 60(6): 1025-1031.

Calvert, J. G., Atkinson, R., Becker, K.H., Kamens, R.M., Seinfeld, J.H., Wallington, T.J., Yarwood, G., (2002): The Mechanisms of Atmospheric Oxidation of Aromatic Hydrocarbons, Oxford University Press, New York, 566p.

Carter, W. P. L., J. H. Seinfeld, D. R. Fitz, and G. S. Tonnesen (1999): “Development of a Next-Generation Environmental Chamber Facility for Chemical Mechanism and VOC Reactivity Evaluation,” Proposal to the U. S. Environmental Protection Agency, February 22. Available at http://www.cert.ucr.edu/~carter/epacham.

Carter, W. P. L. (2000): “Documentation of the SAPRC-99 Chemical Mechanism for VOC Reactivity Assessment,” Report to the California Air Resources Board, Contracts 92-329 and 95-308, May 8. Available at http://cert.ucr.edu/~carter/absts.htm#saprc99 and http://www.cert.ucr.edu/~carter/ reactdat.htm.

http://www.atmos-chem-phys.net/5/623/2005/acp-5-623-2005.pdf



71

Carter, W. P. L. (2002): “Development of a Next Generation Environmental Chamber Facility for Chemical Mechanism and VOC Reactivity Research,” Draft Research Plan and First Progress Report to the United States Environmental Protection Agency Cooperative Agreement CR 827331-01-0, January 3. Available at http://www.cert.ucr.edu/~carter/epacham.

Carter, W. P. L. (2004): "Evaluation of a Gas-Phase Atmospheric Reaction Mechanism for Low NOx Conditions,” Final Report to California Air Resources Board Contract No. 01-305, May 5. Available at http://www.cert.ucr.edu/~carter/absts.htm#lnoxrptf

Carter, W. P. L. (2010): "Development of the SAPRC-07 Chemical Mechanism and Updated Ozone Reactivity Scales," Revised Final report to the California Air Resources Board Contract No. 03-318. January 27. Available at www.cert.ucr.edu/~carter/SAPRC.

Carter, W. P. L. (2011): "Environmental Chamber Study of Ozone Impacts of Coatings VOCs," Final report to California Air Resources Board Contract No. 07-339, May 11. Available at http://www. cert.ucr.edu/~carter/absts.htm#coatrpt2

Carter, W. P. L., R. Atkinson, A. M. Winer, and J. N. Pitts, Jr. (1982): “Experimental Investigation of Chamber-Dependent Radical Sources,” Int. J. Chem. Kinet., 14, 1071.

Carter, W. P. L. and F. W. Lurmann (1991): “Evaluation of a Detailed Gas-Phase Atmospheric Reaction Mechanism using Environmental Chamber Data,” Atm. Environ. 25A, 2771-2806.

Carter, W. P. L., D. Luo, I. L. Malkina, and D. Fitz (1995): “The University of California, Riverside Environmental Chamber Data Base for Evaluating Oxidant Mechanism. Indoor Chamber Experiments through 1993,” Report submitted to the U. S. Environmental Protection Agency, EPA/AREAL, Research Triangle Park, NC, March 20.

Carter, W. P. L., D. R. Fitz, D. Cocker, III, I. L. Malkina, K. Bumiller, C. G. Sauer, J. T. Pisano, C. Bufalino, and Chen Song (2005a): "Development of a Next-Generation Environmental Chamber Facility for Chemical Mechanism and VOC Reactivity Research," Final Report to the United States Environmental Protection Agency Cooperative Agreement CR 827331-01-0. June 27. Available at http://www.cert.ucr.edu/~carter/absts.htm#chamrpt.

Carter, W. P. L., D. R. Cocker III, D. R. Fitz, I. L. Malkina, K. Bumiller, C. G. Sauer, J. T. Pisano, C. Bufalino, and C. Song (2005b): “A New Environmental Chamber for Evaluation of Gas-Phase Chemical Mechanisms and Secondary Aerosol Formation”, Atmos. Environ. 39, 7768-7788.

Carter, W. P. L., I. L. Malkina, D. R. Cocker III, and C. Song (2005c): “Environmental Chamber Studies of VOC Species in Architectural Coatings and Mobile Source Emissions,” Final Report to the South Coast Air Quality Management District Contract No. 03468, July 5. Available at http:// www.cert.ucr.edu/~carter/absts.htm#scaqcham.

Carter, W. P. L. and I. L. Malkina (2005): “Evaluation of Atmospheric Impacts of Selected Coatings VOC Emissions,” Final report to the California Air Resources Board Contract No. 00-333, March 21. Available at http://www.cert.ucr.edu/~carter/absts.htm#coatrpt.

Carter, W. P. L. and I. L. Malkina (2007): “Investigation of the Atmospheric Impacts of Selected Pesticides,” Final Report to the California Air Resources Board Contract 04-334, January 10. Available at http://www.cert.ucr.edu/~carter/absts.htm#pestrep.

72

Crounse, J. D., F. Paulot, H. G. Kjaergaard and P. O. Wennberg (2011): “Peroxy radical isomerization in the oxidation of isoprene,” Physical Chemistry Chemical Physics, 13 (28). ISSN 1463-9076.

Day, D. A., P. J. Wooldridge, M. B. Dillon, J. A. Thornton, and R. C. Cohen (2002): “A thermal dissociation laser-induced fluorescence instrument for in situ detection of NO2, peroxy nitrates, alkyl nitrates, and HNO3,” Journal of Geophysical Research 107, NO. D6, 4046, 10.1029/2001JD000779.

Donahue, N.M. (2011). Atmospheric chemistry: The reaction that wouldn't quit. Nature Chemistry, 3:9899, doi:10.1038/nchem.941

Engeln, R., G. Berden, R. Peeters, and G. Meijer (1998): “Cavity enhanced absorption and cavity enhanced magnetic rotation spectroscopy,” Review of Scientific Instruments 69(11), 3763-3769.

Fiedler, S. E., A. Hese, A. A. Ruth (2003): “Incoherent broad-band cavity-enhanced absorption spectroscopy,” Chemical Physics Letters 371, 284-294.

Fitz, D. R. (2004): Measurements of NOy, nitric acid, NO, and PAN in the South Coast Air Basin, Final Report, Prepared for California Air Resources Board and California Environmental Protection Agency, Research Division, Contract No. 01-318, May 2004. Available at http://www.arb.ca.gov/ research/apr/past/01-318.pdf.

Fitz, D. R., D. V. Pankratz, K. Bumiller, and M. R. Smith (2002): “Measurement of NO2 and PAN by gas chromatography with luminol detection,” Presented at the Symposium on Air Quality Measurement Methods and Technology – 2002, November 13-15, 2002, San Francisco, CA.

Fried, A, and J. Hodgeson (1982): “Laser photoacoustic detection of nitrogen dioxide in the gas-phase titration of nitric oxide with ozone,” Anal. Chem. 54 (2), 278–282.

Hu, D., Tolocka, M., Li, Q., Kamens, R.M. (2007): “A kinetic mechanism for predicting secondary organic aerosol formation from toluene oxidation in the presence of NOx and natural sunlight,” Atmospheric Environment 41, 6478–6496

Jenkin, M.E., Saunders, S.M., Pilling, M.J., 1997. The tropospheric degradation of volatile organic compounds: a protocol for mechanism development. Atmospheric Environment, 31, 81–104.

IUPAC (2010): “Evaluated Kinetic and Photochemical Data”. IUPAC Subcommittee on Gas Kinetic Data Evaluation for Atmospheric Chemistry. Web Version. Available at http://www.iupac-kinetic.ch.cam.ac.uk. Data sheets dated July 2010.

IUPAC (2006): “Evaluated Kinetic and Photochemical Data”. IUPAC Subcommittee on Gas Kinetic Data Evaluation for Atmospheric Chemistry. Web Version. Available at http://www.iupac-kinetic.ch.cam.ac.uk. Data sheets dated June, 2006.

Jenkin, M.E., Saunders, S.M., Pilling, M.J. (1997): “The tropospheric degradation of volatile organic compounds: a protocol for mechanism development”, Atmospheric Environment, 31, 81–104.

73

Hargrove, J. and J. Zhang (2008): “Measurements of NOx, acyl peroxynitrates, and NOy with automatic interference corrections using a NO2 analyzer and gas phase titration,” Review of Scientific Instruments 79, 046109.

Langridge, J. M., S. M. Ball, and R. L. Jones (2006): “A compact broadband cavity enhanced absorption spectrometer for detection of atmospheric NO2 using light emitting diodes,” Analyst 131, 916-922.

Lelieveld, J., T. M. Butler, J. N. Crowley, T. J. Dillon, H. Fischer, L. Ganzeveld, H. Harder, M. G. Lawrence, M. Martinez, D. Taraborrelli, J. Williams (2008): “Atmospheric Oxidation Capacity Sustained by a Tropical Forest,” Nature 452, 737 – 740, doi: 10.1038/nature06870.

Marley, N. A., J. S. Gaffney, R. V. White, L. Rodiguez-Cuadra, S. E. Herndon, E. Dunlea, R. M. Volkamer, L. T. Molina, and M. J. Molina (2004): “Fast gas chromatography with luminol chemiluminescence detection for the simultaneous determination of nitrogen dioxide and peroxyacetyl nitrate in the atmosphere,” Review of Scientific Instruments 75(11), 4595-4605.

Mollner, A.K., S.Valluvadasan, L. Feng, M.K. Sprague, M. Okumura, D.B. Milligan, W.J. Bloss, S.P. Sander, P.T. Martien, R.A. Harley, A.B. McCoy, W.P.L. Carter. (2010). Rate of Gas Phase Association of Hydroxyl Radical and Nitrogen Dioxide.Science, 300:646-649, doi:10.1126/science.1193030

NASA (2011): “Chemical Kinetics and Photochemical Data for Use in Stratospheric Modeling, Evaluation Number 16,” JPL Publication 10-6, Jet Propulsion Laboratory, Pasadena, California, July. Available at http://jpldataeval.jpl.nasa.gov/.

NASA (2006): “Chemical Kinetics and Photochemical Data for Use in Stratospheric Modeling, Evaluation Number 15,” JPL Publication 06-2, Jet Propulsion Laboratory, Pasadena, California, July. Available at http://jpldataeval.jpl.nasa.gov/.

Paul, J. B., L. Lapson, and J. G. Anderson (2001): “Ultrasensitive absorption spectroscopy with a high-finesse optical cavity and off-axis alignment,” Applied Optics 40(27), 4904-4910.

Paulot, F, J.D. Crounse, H.G. Kjaergaard, A. Kurten, J.M. St.Clair, J.H. Seinfeld, and P.O. Wennberg (2009): “Unexpected Epoxide Formation in the Gas-Phase Photooxidation of Isoprene,” Science, Vol. 325, pp 730-733.

Peeters, J., Nguyen, T.L., Vereecken, L. (2009): “HOx radical regeneration in the oxidation of isoprene,” Physical Chemistry and Chemical Physics 11, 5935-5939.

Pitts, J. N., Jr., E. Sanhueza, R. Atkinson, W. P. L. Carter, A. M. Winer, G. W. Harris, and C. N. Plum (1984): "An Investigation of the Dark Formation of Nitrous Acid in Environmental Chambers," Int. J. Chem. Kinet., 16, 919-939.

Rohrer, F., Bohn, B., Brauers, T., Brüning, D., Johnen, F. -J., Wahner, A. and Kleffmann, J. (2004): “Characterisation of the photolytic HONO-source in the atmosphere simulation chamber SAPHIR,” Atmos. Chem. and Physics Disc., 4, 7881-7915

Sabin, L. D., E. Behrentz, A. M. Winer, S. Jeong, D. R. Fitz, D. V. Pankratz, S. D. Colome, and S. A. Fruin (2004): “Characterizing the range of children’s air pollutant exposure during school

74

bus commutes,” Journal of Exposure Analysis and Environmental Epidemiology (2004), 1–11

Setokuchi, O. (2011): “Trajectory calculations of OH radical- and Cl atom-initiated reaction of glyoxal: atmospheric chemistry of the HC(O)CO radical,” Phys. Chem. Chem. Phys., 13, 6296-6304. DOI: 10.1039/C0CP01942A.

Singh, T, R. F. Sawyer, E. S. Starkman, and L. S. Caretto (1968): "Rapid continuous determination of nitric oxide concentration in exhaust gases," Journal of the Air Pollution Control Association 18(2), 102-105.

Tuazon, E. C., R. Atkinson, C. N. Plum, A. M. Winer, and J. N. Pitts, Jr. (1983): “The Reaction of Gas-Phase N2O5 with Water Vapor,” Geophys. Res. Lett. 10, 953-956.

Washenfelder, R. A., A. O. Langford, H. Fuchs, and S. S. Brown (2008): “Measurements of glyoxal using an incoherent broadband cavity enhanced absorption spectrometer,” Atmospheric Chemistry and Physics 8, 7779-7793.

Wendel, G. J., D. H. Stedman, C. A. Cantrell, and L. Damrauel (1983): “Luminol-based nitrogen-dioxide detector,” Analytical Chemistry 55(6), 937-940.

Whitten, G.Z., Heo, G., Kimura, Y., McDonald-Buller, E., Allen, D.T., Yarwood, G. (2010): A new condensed toluene mechanism for Carbon Bond: CB05-TU. Atmospheric Environment 44, 5346-5355.

Whitten, G. Z. (1983): “The chemistry of smog formation: A review of current knowledge,” Environmental International 9, 447-463.

Winer, A. M., J. W. Peters, J. P. Smith and J. N. Pitts, Jr. (1974): “Response of Commercial Chemiluminescent NO-NO2 Analyzers to Other Nitrogen-Containing Compounds,” Environ. Sci. Technol., 8, 1118.

Yarwood, G., Rao, S., Yocke, M., Whitten, G.Z. (2005): Updates to the Carbon Bond mechanism: CB05. Report to the U.S. Environmental Protection Agency, December 2005. (http://www.camx.com/publ/pdfs/CB05_Final_Report_120805.pdf)

Yarwood, G., Whitten, G.Z., Jung, J., Heo, G., Allen, D.T. (2010): Development, evaluation and testing of version 6 of the Carbon Bond chemical mechanism (CB6), Final report to the Texas Commission on Environmental Quality, Work Order No. 582-7-84005-FY10-26.

Zádor, J., Wagner, V., Wirtz, K., Pilling, M.J. (2005): Quantitative assessment of uncertainties for a model of tropospheric ethene oxidation using the European Photoreactor (EUPHORE). Atmospheric Environment 39, 2805-2817.

Zádor, J., Turányi, T., Wirtz, K., Pilling, M.J. (2006): "Measurement and investigation of chamber radical sources in the European Photoreactor (EUPHORE)," Journal of Atmospheric Chemistry 55, 147-166.

Zafonte, L., P. L. Rieger, and J. R. Holmes (1977): “Nitrogen Dioxide Photolysis in the Los Angeles Atmosphere,” Environ. Sci. Technol. 11, 483-487.

http://www.camx.com/publ/pdfs/CB05_Final_Report_120805.pdf

Appendix A

Experimental Methods for the EPA Chamber at UCR

A-1

Appendix A. Experimental Methods for the EPA Chamber at UCR All of the environmental chamber experiments for this project were carried out using the UCR EPA environmental chamber. This chamber was constructed under EPA funding to address the needs for an improved environmental chamber database for mechanism evaluation (Carter et al, 1999, Carter, 2002). The objectives, design, construction, and results of the initial evaluation of this chamber facility are described in more detail elsewhere (Carter et al, 1999; Carter, 2002, 2004; Carter et al, 2005a,b).

The chamber consists of ~85,000-liter dual reactors constructed of flexible 2 mil fluorinated ethylene propene (FEP) Teflon film, located inside a 16,000 cubic ft temperature-controlled “clean room” that is continuously flushed with purified air to minimize contaminants entering the reactors due to permeation or leaks. Two alternative light sources can be used, but as discussed above only blacklights were used for the experiments carried out for this project. Most of the analytical instrumentation was located in the laboratory on the first floor below the reactors on the second floor, but the instrumentation for measuring particle species and spectroscopic instrumentation employed with this study were located adjacent to the reactors. Dry purified air to flush the enclosure and fill the reactors is supplied by an AADCO air purification system. The air system also contains a humidification system, but that was not employed for this project. The temperature inside the reactor can be varied, but all the experiments for this project were carried out at 298±1 K. A diagram of the enclosure and reactors is shown in Figure 19.

Instrumentation Except as indicated standard monitoring instrumentation was employed as used in previous chamber experiments in our laboratory. The instrumentation used to obtain the primary data used for mechanism evaluation is briefly summarized below.

1. O3 was monitored using a commercial UV absorption ozone analyzer, calibrated using standard methods used for ambient air quality monitoring stations. This instrument is subject to positive interferences from certain oxidized aromatic species such as cresols and nitrophenols, and this needs to be taken into account when using O3 data to evaluate mechanisms for these compounds.

2. Organic reactants were monitored by gas chromatography with flame ionization detection (GC-FID). The relatively volatile VOCs such as ethene and toluene were sampled by online loop analysis and the GC-FID response factors were derived from GC span analysis using GC standards or calculated amounts of the VOCs injected into the reactors (e.g., for furan). The less volatile VOCs (specifically o-cresol and 2-nitrophenol) were analyzed by absorption onto cartridges and subsequent desorption onto the GC, and were calibrated by placing known solutions of the materials on the loops prior to analysis.

3. NO and NOy, defined as NO plus the sum of NO2 and other nitrogen-containing species converted to NO with a heated catalytic converter, was monitored by a commercial NO-NOx analyzer. The particular analyzer employed is an unusually sensitive one that we acquired from the California Air Resources Board (CARB) after we intercompared a number of such instruments for a previous CARB project. It is calibrated for NO using standard NO mixtures and

A-2

Figure 19. Schematic of the UCR EPA environmental chamber reactors and enclosure.

a dilution calibration system and the efficiency of the converter is checked by gas-phase titration (GPT) by adding known amounts of O3 to an excess of known amounts of NO in a calibration system. The limitations and utility of the NOy measurements are discussed further below.

4. NO2 was monitored by Cavity Enhanced Absorption Spectroscopy (CEAS), using an instrument that was interfaced to this chamber and operated as part of this project. The NO2 signal was calibrated using the same GPT method as used to test the converter efficiency of the NOy analysis discussed above. The utility and limitations of the data from this instrument are discussed further below.

5. NO2 was also monitored by a GC-Luminol system that is being developed by Fitz Aerometric Technologies and made available for use with this project. It operates by separating NO2 from PAN and other nitrogen-containing species using a GC column and detecting the species using a luminol-based wet chemical method. It also obtains data for peroxy acetyl nitrate (PAN) and can potentially obtain data for other PAN analogues if they can be separated from PAN on the GC column, though as presently operated it only reports concentrations of PAN. The NO2 signal is calibrated using the same GPT as used for the CEAS and the NO and NOy analysis. The PAN signal is calibrated by assuming it has the same response factor as NO2, which had been verified previously. The limitations and utility of this instrument for NO2 and PAN analysis are discussed further below.

6. Temperature was monitored by shielded and calibrated thermocouples inside the sampling lines.

7. Light intensity was qualitatively monitored using a light radiation sensor, and more quantitatively by conducting periodic NO2 actinometry experiments in conjunction with some of the experiments.

8. Numbers of particles in various size ranges, from which total particle number and volumes can be calculated, were monitored using a Scanning Mobility Particle Sizer (SMPS).

A-3

Separate sample lines were used for this instrument, which was located inside the enclosure to avoid artifacts due to changing temperatures.

The sampling lines leading from the reactors to the O3 and NOx species were on a switching system such that air from the two reactors, zero air, or calibrator output can alternately be sampled, under control of the data acquisition system. For normal experiments, the system alternatively sampled between the two reactors and zero air for five minute intervals. However, as during the course of the project it was found that longer sampling times were required to obtain valid measurements for PAN (discussed below), and for subsequent experiments where PAN data were important the reactor sampling times were increased to 30 minutes.

As indicated above, although obtaining data to evaluate models for particle formation is beyond the scope of the project as proposed, particle number and size data were obtained from a SEMS instrument during almost all of the mechanism evaluation experiments carried out for these project. The particle volume and number data are included in the measurement data files made available for mechanism evaluation, and particle volume data (expressed in mass units assuming a density of 1 gm/cm3) are included in the data tabulations and plots given in the results section. However, the utility of these data for modeling is not discussed further here.

Special considerations of systems used to monitor NOx species Because of their importance to this project, the various methods used to monitor NOx species are further discussed below.

The NOy or "uncorrected NO2" (NOy-NO) data obtained from the commercial NOx analyzer are generally not useful for mechanism evaluation other than determining initial NO2 concentrations before the irradiation begins. This is because the "NO2" analysis is known to respond quantitatively to PAN and other organic peroxy nitrates and organic nitrates, and semi-quantitatively HNO3 and other nitrogen species (Winer et al, 1974). There is also a response to aromatic nitro compounds and nitrophenols, but the response is probably also not quantitative. The instrument had a scrubber designed to remove HNO3, but the effectiveness of the scrubber was not evaluated during the period of this project. However, although the data from this instrument was useless in the NOx source experiments because the starting materials had a response on the NOy analysis, the instrument was found to have utility in determining or supporting other measurements of the total amount of NOx species formed in the background experiments.

The CEAS system for measuring NO2 was newly installed for this project. This system probably provides the most reliable and interference-free method for monitoring NO2, though it has two limitations. The first is that suffered from zero drift for many experiments, and the periodic zero samples are needed to properly zero the data during the experiments and the span operations. In some cases it was necessary to extrapolate the zeros in order to span the instrument, which leads to some uncertainty, though it is generally not large because the zero drift is small compared to the NO2 levels used in the span. On the other hand, the zero drift was not small compared to the NO2 readings during many of the experiments useful for mechanism evaluation. The other limitation is that as operated during this project the sensitivity of the instrument was not always as great as expected. In most cases, it could provide useful NO2 data

A-4

only if the NO2 levels were greater than ~3 ppb, making it useless for background NOx offgasing characterization.

The group that developed the CEAS system used for this project also delivered a dual-channel cavity ring-down spectroscopy (CRDS) system. This system was intended to be used to monitor NO2 and total peroxy acyl nitrates (PANs) and total organic nitrates using a thermal decomposition system to convert these compounds to NO2, which is then detected using a separate detector (Hargrove and Zhang, 2008). The difference between the NO2 in the detector with the heated inlet from the NO2 in the detected from unheated inlet would give total PANs if an appropriate heater temperature is used, and would give total PANs + organic nitrates if a higher appropriate temperature is used. Such data would clearly be useful for the purpose of this project, and therefore use of this instrument was included in this proposal.

Unfortunately, we were unable to get the CRDS system operational during the period of this project, so the instrument and its data are not described further in this report. However, we made an attempt to use the CEAS system to obtain total PANs and organic nitrate data at the end of a number of experiments in Reactor A by ramping the temperature in the inlet tube leading to the CEAS instrument. Although the NO2 signal did increase as expected, we were unable to obtain definitive data to verify that the responses were quantitative for total PANs or organic nitrates. Therefore, although temperature ramp data for the CEAS are available for some experiments conducted for this project, they have not been validated for use for model evaluation and are therefore not discussed further in this report. Therefore, no total PANs and nitrate data are available for use in mechanism evaluation from these experiments.

The GC-Luminol system, while providing a sensitive and usually interference-free measurement for NO2 and an easy-to-calibrate measurement for PAN, also has its limitations. The NO2 measurement is known to have an O3 interference, and an O3 scrubber is needed to remove the O3 in order to obtain valid NO2 data. This scrubber failed during the course of this project, resulting in the NO2 data not being valid during some of the later experiments. This is noted in the discussion of the results, and the affected data are excluded from the mechanism evaluation dataset. Of potentially greater concern is that luminol is known to respond to H2O2, which may not be retained on the GC column and therefore eluted with NO2, where it may cause an interference. However, there is no evidence of a significant H2O2 interference in the GC-Luminol NO2 or PAN data in the experiments where H2O2 was present, so these data were not rejected for these experiments. Either the GC column separates the H2O2 from both NO2 and PAN, or it is destroyed by the O3 scrubber or otherwise does not make it to the detector.

Another limitation of the NO2 analysis of the GC-luminol system is that it takes a long time for the NO2 readings to stabilize during GPT calibrations when our usual level of ~90 ppb is employed. Because of this, the O3 levels used in the GPT calibrations were reduced to ~25 ppb, which was low enough for the GC-Luminol NO2 measurements to stabilize relatively quickly. But this was not immediately realized, resulting in the spans for this instrument being uncertain for some of the initial experiments.

The PAN analysis had two limitations. The most significant for this project was that it requires at least 30 minutes to stabilize, which meant that PAN measurements made using our usual 5 minute sampling cycle were consistently low. Because of this, the GC-Luminol PAN data for a

A-5

number of our earlier experiments had to be rejected. Fortunately, for many of these runs we sampled for longer periods from the A reactor during the end of the experiments because of the temperature ramp studies done with the CEAS (discussed above), allowing valid PAN data to be obtained for the end of these experiments. For the later experiments, especially the acetaldehyde - H2O2 NOx source experiments where PAN data were particularly important, we used 30 minute sampling cycles that allowed for valid PAN data to be obtained.

The PAN analysis also has the limitation that it is specific for peroxyacetyl nitrate, and other PAN analogues may be completely separated from PAN in the GC system and not be detected, or be partially separated and cause an interference. This system is therefore not useful for obtaining total PANs data as would be the case for a thermal decomposition-NO2 system such as discussed above for the CEAS and CRDS systems. Such PAN analogues may be formed in the ethene and isoprene experiments carried out for this project. However, an examination of the GC traces during those experiments did not indicate any additional or partially resolved peaks that may be due to significant formation of other PAN analogues.

Materials The sources of the NO, CO, H2O2 and the other reagents used in this project except acetaldehyde, biacetyl, furan, isopropyl nitrate and isobutyl nitrate came from various commercial vendors as employed in previous projects at our laboratory. CO (Praxair, CP grade) was scrubbed with carbon charcoals before injection into the reactors to remove carbonyl -containing compounds produced by reaction of CO and the cylinder surface. NO2 was in-situ generated before injection by chemical conversion of NO inside small Pyrex bulbs with known volumes. H2O2 was purchased from Sigma-Aldrich as H2O2 solution in water (Sigma-Aldrich, 50 wt. % in H2O, stabilized, 516813) to use as a radical source. The concentration of H2O2 in the solution was measured so that the amounts of H2O2 injected into the chamber could be determined from the volume of solution used. Acetaldehyde (Sigma-Aldrich, ≥99.5% (ACS reagent), 402788-100ML), biacetyl (Aldrich, 97%, B85307-100ML), furan (Aldrich, >99%, 185922-100ML), isopropyl nitrate (Fluka, >99.0% (GC), 59640-500ML-F) and isobutyl nitrate (Aldrich, 96%, 472697-5ML) were purchased from Sigma-Aldrich. Ethene (Matheson), propene (Matheson), isoprene (Sigma-Aldrich), toluene (Sigma-Aldrich), o-cresol (Sigma-Aldrich) and 2-nitrophenol (Sigma-Aldrich) were in stock at the laboratory, and were not purchased during this project.

Procedures Before each experiment both reactors were thoroughly flushed with dry purified air and filled to their maximum volume prior to injection of any reactants. Because of the design of the chamber the reactor volumes were generally reproducible and known at the time of reactant injection, allowing the amounts of reactants injected to be calculated based on known amounts injected and the reactor volumes. This was necessary to determine the amounts of H2O2 injected and also used to calibrate the GC analysis of the relatively volatile reactant VOCs. Common reactants for both reactors were simultaneously injected into the two reactors and their contents were mixed by flushing from one reactor to the other to assure equal levels. The valve allowing mixing between the reactors was then closed and any reactants that are different in the two reactors are then injected and mixed. Once reactant levels have stabilized and samples were taken for analysis to determine initial reactant concentrations, the irradiations were begun by turning on the lights. Sampling was conducted alternatively or

A-6

simultaneously from both reactors during the irritation, depending on the measurement being made.

During the course of the experiment the volumes of the reactors decrease as air is withdrawn from the reactors due to sampling or lost because of leaks. The frameworks at the top of the reactors were lowered to maintain constant positive pressure so that the loss of air does not cause dilution of reactants and infiltration of contaminants is minimized. The location of the framework is monitored, and if it reaches the low limit level then the experiment is terminated and all the subsequent data from that reactor are discarded. The experiments were ended after about 8 hours or when both reactors were depleted, whichever happened first (usually the latter). Unfortunately, during the course of this project there was a persistent single large leak or multiple leaks in Reactor B that were never found, so the experiment on Side B always ended earlier than Side A, often after only 3-4 hours. Note that Side B was always used as the base case in the NOx sink experiments, so the experiments with the added test compound generally had data for the longer durations. But in most cases the duration of the Side B experiment was sufficient to obtain useful data for the objectives of this project.

After the experiment the chamber was emptied and filled several times and flushed with purified air sufficiently so that no detectable pollutants remained in the reactors.

Data Processing The data obtained during the experiments were collected using a data acquisition system interfaced to most of the instruments, and using separate computers to collect and process the GC-FID, CEAS and SMPS data. These data from these sources were then loaded into Excel spreadsheets for each experiment (called "run files") using various macros, the applicable span and zero corrections were applied, and the various measurements were associated to the individual reactors or calibration or zero states based on information provided by the data acquisition computers or the operators.

The data in the run files were then examined, plotted, and compared to model predictions for the purpose of quality assurance, assigning initial values for modeling, and determining which measurements are valid for model evaluations. In some cases the assigned initial concentrations were different from the measurement values, or measurement data may be insufficient to determine initial concentrations. In those cases, initial concentrations were determined by manual examination of plots of the data (aided in some cases by modeling), or from calculated amounts injected. These assigned initial concentrations were entered into an appropriate sheet in the run files, along with characterization assignments such as the NO2 photolysis rate or characterization set identifiers. Excel macros are then used to output model input (.INP) files containing the run-specific inputs needed for modeling these experiments, and measurement data (.GDT) files containing the measurement data of potential utility for mechanism evaluation.

In addition to the run-specific input (.INP) and output (.GDT) files, .SDR files giving the spectral distribution of the light source and .CHR files giving the chamber characterization input parameters applicable to groups of runs assumed to have the same chamber effects are also needed and included. In the case of these experiments, only a single spectral distribution file (ITCUSE.SDR) is included because only a single light source was used, and only a single characterization set is used for each reactor (EPA-9A.CHR for Reactor A and EPA-9B.CHR for Reactor B).

Appendix B

Results of SAPRC-07 Evaluation

B-1

Appendix B. Results of SAPRC-07 Evaluation Although evaluating and improving mechanisms other than CB6 is beyond the scope of this project, the new chamber data obtained should be equally useful for evaluating and improving other mechanisms. The SAPRC-07 mechanism is another reasonably up-to-date mechanism developed for urban and regional ozone modeling that has been used in simulations of regional ozone formation in Texas and elsewhere. In this Appendix, we discuss the performance of the SAPRC-07 mechanism (Carter, 2010) in simulating these data and the cases where the data indicate that improvements are needed for this mechanism as well. Although improving the SAPRC-07 mechanism is beyond the scope of this report, the data obtained for this project will be taken into account as part of ongoing or anticipated projects to update this mechanism.

Simulations of Control Experiments Figure B-1 and Figure B-2 show experimental and calculated concentration-time plots for selected species potentially useful for mechanism evaluation for two types of control experiments carried out for this project: (1) acetaldehyde - H2O2 - NOx, (2) ethene (or propene) - NOx. The simulations of these two types of experiments are discussed separately below.

For evaluating the use of acetaldehyde - H2O2 irradiations to quantify NOx release from NOx sourcesby capturing NOx in the form of PAN and measuring PAN , runs were carried out using known amounts of NOx (25 ppb NO for EPA1415B; 13 ppb NO2 for EPA1447A), and the experimental and calculated data for these runs are shown on Figure B-1. The SAPRC-07 mechanism gave good fits to the base mechanism for this system.

Figure B-1. Plots of selected results of the acetaldehyde - H2O2 - NOx control experiments carried out for this project. Results of SAPRC-07 model calculations are also shown.

Unfortunately, there were no useful PAN data for EPA1415B so that run was not useful for evaluating the use of PAN data for measuring NOx offgasing, but the model simulations of the

EPA1415B EPA1447AOzone (ppm) Acetaldehyde (ppm) Ozone (ppm) Acetaldehyde (ppm)

PAN (ppm) NO2 (ppm) PAN (ppm) NO2 (ppm)


0.00

0.04

0.08

0.12

0.16

0 60 120 180 240

DataCalc

0.45

0.50

0.55

0.60

0.65

0.70

0 60 120 180 240

0.000

0.004

0.008

0.012

0 60 120 180 240

CEASLuminol

0.00

0.01

0.02

0.03

0 60 120 180 240

No Data

0.00

0.04

0.08

0.12

0 120 240 3600.25

0.30

0.35

0.40

0 120 240 360

0.000

0.005

0.010

0.015

0 120 240 3600.000

0.004

0.008

0.012

0.016

0 120 240 360

B-2

final NO2 concentrations were in good agreement with the measurements using both the CEAS and the GC-Luminol measurements. There were useful PAN data for run EPA1447A, and the measured PAN levels were in reasonably good agreement with model predictions. However, the measured PAN tended to decline somewhat between the 2nd hour of the experiment and the end, while the model predicted that the PAN levels should be constant. It is unclear whether this is an experimental or a mechanism problem without data to assess the reproducibility of this result.

For evaluating the use of ethene - NOx and propene - NOx experiments to evaluate NOx removal by NOx sinks, we conducted ethene - NOx and propene - NOx control irradiations without any added test compound in either reactor. The experimental and calculated results for these experiments are shown on Figure B-2. The primary purpose of these experiments is to test the side equivalency of the dual reactors (side A and side B) when the same base case mixture is irradiated, and as shown on Figure B-2 good side equivalency was obtained. The SAPRC-07 mechanism gave reasonably good simulations of the measurement data useful for mechanism evaluation, though it tended to somewhat underpredict the maximum ozone yield in the ethene - NOx control experiments (e.g., see ozone plots for EPA1400 in Figure B-2). Good predictions of NO2 were obtained in the two propene experiments that had NO2 data, and reasonably good predictions of PAN were obtained in the one propene experiment that had useful PAN data (Figure B-2).

Overall, it can be concluded that this mechanism gives reasonably good simulations of these control experiments. This indicates that the data for similar experiments with added test compounds should be useful for evaluating the mechanisms of the test compound. If poor model performance is seen, it can be reasonably attributed to problems with the mechanisms for the test compounds, not with the mechanisms for the base case system.

Simulations of NOx Sink Experiments Toluene Toluene can be represented either explicitly or using the lumped ARO1 model species in the SAPRC-07 mechanism, but evaluations are most appropriately carried out using the most explicit representation. Figure B-3 shows the experimental and calculated results for the five NOx sink experiments carried out with toluene, where the calculations use the explicit representation for toluene. The first three on the left are propene - NOx experiments with added toluene on one side, while the other two are ethene - NOx with added toluene. The open rectangles are data from the base case experiments and the solid lines are the model simulations of these experiments. The filled diamonds are data from the added toluene experiments and the dashed lines are model simulations of the added toluene experiments.

The model tended to somewhat overpredict O3 in the three propene - NOx base case experiments but gave reasonably good simulations of O3 in the ethene - NOx base case experiments, though the ethene base case experiments were not conducted long enough to get "true" maximum O3 data because of reactor leakage problems. The model overpredicted PAN by about a factor of 2 in the one propene - NOx base case experiment (EPA1443) shown on Figure B-3 that had valid PAN data on the base case side. This is in contrast with the result for the one propene - NOx control experiment with PAN data (EPA1409) shown on Figure B-2,

B-3

where reasonably good model simulations of the PAN data were obtained. This suggests that there may be problems with the consistency with the PAN data, which should be taken into account when evaluating model performance in simulating PAN data in added test compound experiments.

Figure B-2. Plots of selected results of the ethene - NOx and propene - NOx control experiments carried out for this project. Results of SAPRC-07 model calculations are also shown.

Ethene Propene Propene PropeneEPA1400 EPA1391 EPA1395 EPA1409

Ozone (ppm)

NO (ppm)

NOy-NO (ppm) NO2 (ppm)

Ethene (ppm) Propene (ppm)

PAN (ppm)


0.0

0.1

0.2

0.3

0 60 120 180 240

Side ASide B

0.000

0.001

0.002

0.003

0.004

0.005

0 60 120 180 240

0.000

0.004

0.008

0.012

0.016

0 60 120 180 240

0.0

0.2

0.4

0.6

0.8

1.0

0 60 120 180 240

0.00

0.04

0.08

0.12

0.16

0 60 120 180 240

0.000

0.001

0.002

0.003

0.004

0 60 120 180 240

0.000

0.004

0.008

0.012

0.016

0 60 120 180 240

0.0

0.1

0.2

0.3

0.4

0 60 120 180 240

0.00

0.04

0.08

0.12

0.16

0 60 120 180

0.000

0.001

0.002

0.003

0.004

0 60 120 180

0.000

0.004

0.008

0.012

0 60 120 180

0.0

0.1

0.2

0.3

0.4

0 60 120 180

0.00

0.04

0.08

0.12

0.16

0 120 240 360 480

0.000

0.004

0.008

0.012

0.016

0 120 240 360 480

0.000

0.004

0.008

0.012

0 120 240 360 480

0.0

0.1

0.2

0.3

0.4

0 120 240 360 480

0.000

0.002

0.004

0.006

0.008

0.010

0 120 240 360 480

B-4

Figure B-3. Plots of selected results of NOx sink experiments with toluene. Results of SAPRC-07 model calculations are also shown.

Generally, the SAPRC-07 mechanism gave reasonably good simulations of the effects of toluene addition on O3 formation, predicting that toluene suppresses peak O3 in these experiments by approximately the amount observed. However, the actual amount of suppression might be slightly overpredicted by the SAPRC-07 mechanism. In the case of the propene + toluene experiments the model gives reasonably good simulations of peak O3 in the added toluene reactor, but since it somewhat overpredicts O3 in the base case experiment it is actually overpredicting the effect of toluene addition. It is more difficult to assess this in the case of the

EPA1418 Toluene EPA1436 Toluene EPA1443 Toluene EPA1401 Toluene EPA1407 Toluene

Ozone (ppm)

Toluene (ppm)

Propene (ppm) Ethene (ppm)

NO2 (CEAS) (ppm)

PAN (ppm)


0.00

0.04

0.08

0.12

0.16

0 120 240 360

0.0

0.1

0.2

0.3

0.4

0 120 240 360


0.0

0.1

0.2

0.3

0.4

0 120 240 360

0.000

0.005

0.010

0.015

0 120 240 360

0.000

0.002

0.004

0.006

0.008

0.010

0 120 240 360

0.00

0.05

0.10

0.15

0 120 240 360

0.0

0.1

0.2

0.3

0.4

0.5

0 120 240 360

0.0

0.1

0.2

0.3

0.4

0.5

0 120 240 360

0.000

0.005

0.010

0.015

0 120 240 360

0.000

0.002

0.004

0.006

0.008

0.010

0 120 240 360

0.00

0.04

0.08

0.12

0.16

0 120 240 360 480

0.0

0.1

0.2

0.3

0.4

0.5

0 120 240 360 480

0.0

0.1

0.2

0.3

0.4

0 120 240 360 480

0.000

0.005

0.010

0.015

0.020

0.025

0 120 240 360 480

0.000

0.004

0.008

0.012

0 120 240 360 480

0.00

0.04

0.08

0.12

0.16

0 60 120 180

0.0

0.1

0.2

0.3

0.4

0 60 120 180

0.0

0.4

0.8

1.2

0 60 120 180

0.000

0.004

0.008

0.012

0.016

0 60 120 180

0.000

0.001

0.002

0.003

0 60 120 180

0.00

0.05

0.10

0.15

0.20

0.25

0 120 240 360

0.0

0.1

0.2

0.3

0.4

0 120 240 360

0.0

0.2

0.4

0.6

0.8

1.0

0 120 240 360

0.000

0.004

0.008

0.012

0.016

0 120 240 360


Test Expt Base Expt Base Calc Test Calc

B-5

ethene experiments because the base case data cannot test the ability of the model to simulate the peak O3. The model tends to underpredict O3 in the ethene experiments with the added toluene, but Figure B-2 shows that it also tends to underpredict peak O3 in the ethene base case experiments as well.

The model gives reasonably good simulations of the NO2 data in both the base case and the added toluene experiments, indicating that it does not have large problems in simulating NOx data. The PAN levels in the propene experiments with added toluene are simulated to within the likely uncertainty of the data, and the PAN levels in the one ethene + toluene experiment with PAN data are reasonably well simulated.

o-Cresol The CRES model species is used to represent all phenolic compounds in the SAPRC-07 mechanism, but the mechanism actually used for it is based on data for o-cresol. Therefore, experiments with o-cresol are appropriate for evaluating SAPRC-07 without consideration of lumping issues. Of course, the issue about whether separate model species should be used for other phenolic compounds is not addressed by this evaluation.

The left hand plots on Figure B-4 show the experimental and calculated results for the ethene - NOx experiment with added o-cresol. The base case experiment is reasonably well simulated, but as with the ethene control experiment on Figure B-2 the SAPRC-07 mechanism somewhat underpredicts the maximum O3 yields. The SAPRC-07 mechanism significantly underpredicts the initial rate of cresol consumption and the initial rate of O3 formation in the added cresol experiment, but it gives a reasonably good simulation of the large effect of o-cresol in reducing both the peak O3 level and also the rate of ethene consumption. This indicates that although SAPRC-07 underpredicts the initial reactivity of o-cresol, it gives reasonably good simulations of NOx sinks in the cresol mechanisms, as well as their inhibiting effects on radical levels.

As part of a project for developing a mechanism for secondary organic aerosol (SOA) formation from aromatics, we conducted a number of cresol - NOx experiments at lower, and more atmospherically relevant, NOx levels than used in the cresol - NOx experiment that the current SAPRC-07 mechanism was adjusted to simulate. These experiments indicate much higher reactivity for cresols under lower NOx conditions than the earlier, high NOx experiment. When the cresol mechanism is adjusted to fit these new data, then much better simulations of O3 formation, cresol consumption, and NO2 consumption rates are obtained for the run shown on Figure B-4, while still having a reasonably good prediction of effects on peak O3 levels (Carter, unpublished results). Simulations of this experiment will be included among those used to evaluate this update to the SAPRC-07 cresol mechanism.

Furan As discussed in the main body of this report, NOx sink experiments with furan were conducted because the reaction of furan with OH is expected to form high yields of 2-butenedial, a representative highly photoreactive aromatic oxidation product. The SAPRC-07 mechanism uses the model species AFG1 and AFG2 to represent the highly photoreactive aromatic oxidation

B-6

Figure B-4. Plots of selected results of NOx sink experiments with o-cresol and furan. Results of SAPRC-07 model calculations are also shown.

products such as the 2-butenedial formed in the furan system. These two model species have the same mechanism except that AFG1 photolyzes to form radical products and AFG2 photolyzes to form stable products, and the ratios of the yields of the two products are adjusted to give overall radical formation quantum yields that best fit rates of NO oxidation and O3 formation in aromatic - NOx or furan - NOx experiments. Therefore the mechanism has the same ultimate products being formed from the reactions of these products formed from aromatic hydrocarbons as is the case for furan, but the overall quantum yields are somewhat different. This means that the ability of the mechanism to simulate NOx sinks in the furan system should have relevance to the performance of the aromatic mechanism in general.

The experimental and model calculation results for the three ethene - NOx experiments with added furan are shown on the middle and right side of Figure B-4. Only one base case experiment had useful data on the final O3 yield, but the results of the model simulations were

EPA1408 o-Cresol EPA1402 Furan EPA1448 Furan EPA1403 Furan * Ozone (ppm)

o-Cresol (ppm) Furan (ppm)

Ethene (ppm)

NO2 (CEAS) (ppm) NO2 (Luminol) (ppm) NO2 (CEAS) (ppm)


0.0

0.1

0.2

0.3

0 120 240 360

0.00

0.04

0.08

0.12

0 120 240 360

0.0

0.4

0.8

1.2

0 120 240 360

0.000

0.004

0.008

0.012

0 120 240 360

0.00

0.05

0.10

0.15

0.20

0.25

0 120 240

0.00

0.04

0.08

0.12

0.16

0 120 240

0.0

0.4

0.8

1.2

0 120 240

0.000

0.004

0.008

0.012

0 120 240

0.00

0.05

0.10

0.15

0.20

0.25

0 60 120 180 240

Test Expt B

* This run had furan in both

reactors

0.00

0.02

0.04

0.06

0 60 120 180 240

0.7

0.8

0.9

1.0

1.1

0 60 120 180 240

0.000

0.004

0.008

0.012

0.016

0 60 120 180 240

0.00

0.05

0.10

0.15

0.20

0.25

0 120 240 360

0.00

0.04

0.08

0.12

0.16

0 120 240 360

0.6

0.7

0.8

0.9

0 120 240 360

0.000

0.004

0.008

0.012

0 120 240 360

Test Expt

Base Expt

Base Calc

Test Calc

B-7

consistent with the simulations of the other ethene - NOx control experiments, where peak O3 was slightly underpredicted. In general, the SAPRC-07 mechanism gives fair simulations of O3 formation in the added furan experiments, though it does not simulate the rate of O3 consumption once the peak O3 level is formed, and significantly underpredicts the consumption rates for furan and also the consumption rates for ethene. The amount of inhibition of the peak O3 yields is also somewhat underpredicted, suggesting that SAPRC-07 may not have strong enough NOx sinks for these oxidation products.

These data will be potentially useful for changes to the mechanism regarding representation of the aromatic ring fragmentation products, but only if it is assumed that the same types of compounds are also formed in the furan system. This is assumed in the current SAPRC-07 mechanism, but given the uncertainties concerning these photoreactive unsaturated ring fragmentation products this assumption may not necessarily be correct.

Isoprene Figure B-5 shows experimental and calculated data for the two NOx sink experiments with isoprene. Both experiments use the ethene - NOx system as the base case, and, as with the other ethene experiments, the peak O3 is slightly underpredicted by SAPRC-07. The model gives good simulations of the effects of added isoprene on peak O3, time profiles for NO2, and ethene consumption rates. It also gives a reasonably good simulation of PAN in the one experiment with useful PAN data (EPA1446). Therefore, these results do not indicate problems with the SAPRC-07 isoprene mechanism.

Simulations of NOx Source Experiments Isopropyl and Isobutyl Nitrates Neither isopropyl nor isobutyl nitrates are represented explicitly in the SAPRC-07 mechanism; instead a lumped model species "RNO3" is used to represent all organic nitrates formed in the

Figure B-5. Plots of selected results of NOx sink experiments with isoprene. Results of SAPRC-07 model calculations are also shown.

EPA1404 Isoprene EPA1446 IsopreneOzone (ppm) Isoprene (ppm) Ozone (ppm) Isoprene (ppm)

Ethene (ppm) NO2 (ppm) Ethene (ppm) NO2 (ppm) PAN (ppm)


0.0

0.1

0.2

0.3

0 120 240 3600.00

0.02

0.04

0.06

0 120 240 360


0.0

0.4

0.8

1.2

0 120 240 3600.000

0.004

0.008

0.012

0.016

0 120 240 360

0.0

0.1

0.2

0.3

0 120 240 3600.00

0.02

0.04

0.06

0 120 240 360

0.0

0.2

0.4

0.6

0.8

1.0

0 120 240 3600.000

0.004

0.008

0.012

0.016

0 120 240 3600.000

0.001

0.002

0.003

0 120 240 360

Test Expt

Base Expt

Base Calc

Test Calc

B-8

photooxidation reactions of non-aromatic compounds. The mechanism for this lumped model species is derived by averaging the mechanisms derived for individual alkyl nitrates predicted to be formed from the reactions of the alkanes in the standard ambient reactive organic gas (ROG) surrogate mixture used as the basis for deriving the SAPRC-07 condensed mechanism (Carter, 2010), which is based on ambient measurements of Lonneman (1986). The specific organic nitrate compounds used to derive the RNO3 mechanism are CH3CH2CH(CH3)ONO2, CH3CH(OH)CH2-CH2CH2ONO2, CH3CH2CH(CH3)CH(CH3)-ONO2, H3CH2CH2CH2CH2CH(ONO2)CH2OH, CH3CH2C(CH3)(ONO2)CH2CH(CH3)CH3, and CH3CH2-CH2CH2CH2CH2CH2CH(ONO2)-CH2CH3, each weighed equally.

The mechanisms for these organic nitrate compounds, and therefore the mechanism for the RNO3 species used to represent them, were derived using the SAPRC mechanism generation and estimation system (Carter, 2000, 2010) based on various structure-reactivity estimation methods. Therefore, the validity of the SAPRC-07 representation of organic nitrates is based ultimately on the ability of the mechanism generation system to estimate appropriate mechanisms for these compounds. Because of this, we used this same mechanism estimation and generation system to derive mechanisms for the specific compounds studied experimentally for the purpose of using these data to evaluate the SAPRC-07 representation of organic nitrates. This allows for the chemical basis for the mechanism to be evaluated without consideration of the lumping issues involved. This is appropriate because isobutyl and isopropyl nitrates are not actually formed to significant extents in actual ambient atmospheric conditions.

The mechanisms derived for isopropyl nitrate (IC3-ONO2) and isobutyl nitrate (IC4-ONO2), in terms of SAPRC-07 lumped model species, are as follows:

Isopropyl nitrate:

IC3-ONO2 + OH = 0.188 NO2 + 0.812 RO2C + 0.188 ACET + 0.047 xNO2 + 0.764 xHO2 + 0.047 xHCHO + 0.047 xCCHO + 0.764 xRNO3 + 0.812 yROOH (kOH=4.21e-13 cm3 molec-1 s-1)

IC3-ONO2 + HV = NO2 + HO2 + ACET (Same rate as RNO3)

Isobutyl nitrate:

IC4-ONO2 + OH = 0.017 NO2 + 1.032 RO2C + 0.043 RO2XC + 0.043 zRNO3 + 0.017 RCHO + 0.554 xNO2 + 0.114 xHO2 + 0.272 xMEO2 + 0.646 xHCHO + 0.554 xACET + 0.026 xRNO3 + 0.983 yROOH + 0.403 XN (kOH=2.7e-12 cm3 molec-1 s-1)

IC4-ONO2 + HV = NO2 + 0.638 HO2 + 0.347 RO2C + 0.014 RO2XC + 0.014 zRNO3 + 0.362 HCHO + 0.638 RCHO + 0.347 xHO2 + 0.347 xACET + 0.362 yROOH (Same rate as RNO3)

See Carter (2010) for a description of SAPRC-07 model species and the absorption cross sections and quantum yields for organic nitrates, which are based on data for isopropyl nitrate. Note that "XN" formed from OH + isobutyl nitrate represents nitrate-containing organics that are estimated to react relatively slowly, and thus their formation does not represent a significant NOx source.

B-9

Measured and calculated concentrations of selected species in the NOx source experiments with these nitrates are shown on Figure B-6, where the SAPRC-07 calculations used the mechanisms derived for these specific compounds as discussed above. The model gave reasonably good fits to the data for these experiments, indicating that the methods used to estimate mechanisms for organic nitrates, and therefore the mechanisms used to derive the mechanism for the RNO3 lumped model species, performs reasonably well. There may be a modest bias towards underpredicting NOx source processes, as indicated by the slight underprediction of O3 in most of the experiments, the slight underprediction of PAN in the acetaldehyde experiments, and the tendency to underpredict at least some of the NO2 data. However, these discrepancies are relatively small considering the uncertainties in the estimation methods, and do not indicate clear systematic problems with the estimation methods and organic nitrate mechanisms that need to be addressed.

B-10

Figure B-6. Plots of selected results of NOx source experiments with the organic nitrates. Results of SAPRC-07 model calculations are also shown.

Nitrophenol The NPHE model species is used in the SAPRC-07 mechanism to represent nitrophenols and also other, unknown, products formed in the reactions of NO3 radicals with phenolic compounds and also from the reactions of phenoxy radicals with NO2, predicted to be their major atmospheric sink. It is uncertain how well 2-nitrophenol represents the actual set of

Isopropyl Nitrate Isobutyl Nitrate EPA1449A (Aldehyde) EPA1449B (CO) EPA1439A (Aldehyde) EPA1439B (CO)

Ozone (ppm)

NO2 (ppm)

PAN (ppm)

Isopropyl Nitrate (ppm) Isobutyl Nitrate (ppm)

Acetaldehyde (ppm) Acetaldehyde (ppm)


0.00

0.02

0.04

0.06

0.08

0.10

0 180 360 540

0.000

0.001

0.002

0.003

0 180 360 540

0.000

0.005

0.010

0.015

0.020

0.025

0 180 360 540

0.32

0.34

0.36

0.38

0 180 360 540

0.0

0.4

0.8

1.2

0 180 360 540

0.00

0.04

0.08

0.12

0.16

0 60 120 180 240

0.000

0.002

0.004

0.006

0.008

0 60 120 180 240

0.0000

0.0001

0.0002

0.0003

0 60 120 180 240

0.34

0.35

0.36

0.37

0.38

0 60 120 180 240

0.00

0.02

0.04

0.06

0.08

0 120 240 360

Expt.

Calc

0.000

0.001

0.002

0.003

0 120 240 360

0.000

0.004

0.008

0.012

0 120 240 360

0.48

0.49

0.50

0.51

0 120 240 360

0.0

0.2

0.4

0.6

0 120 240 360

0.00

0.06

0.12

0.18

0.24

0.30

0 60 120 180 240 300

0.000

0.002

0.004

0.006

0.008

0.010

0 60 120 180 240 300

CEASLuminol

0.000

0.001

0.002

0 60 120 180 240 300

0.48

0.49

0.50

0.51

0 60 120 180 240 300

B-11

compounds it is intended to represent, and good model performance in simulating nitrophenol experiments may not necessarily mean that the representation is appropriate when using the general aromatics mechanism.

Figure B-7. Plots of selected results of NOx source experiments with 2-nitrophenol. Results of SAPRC-07 model calculations are also shown.

Experimental and calculated data for the four NOx source experiments with 2-nitrophenol are shown on Figure B-7. Note O3 data are not shown because of a positive interference by nitrophenol on the UV ozone analysis instrument, but the results are not inconsistent with model predictions when that is taken into account. However, the NO2 data in all the experiments and the PAN data in the acetaldehyde experiments clearly indicate that the model is significantly underpredicting NOx release from NOx sources in the nitrophenol system. Although the model gives reasonably good simulations of the 2-nitrophenol consumption rates

2-Nitrophenol - Acetaldehyde - H2O2 runs 2-Nitrophenol - CO - H2O2 runsEPA1441A EPA1442A EPA1441B EPA1442B

NO2 (CEAS) (ppm)

PAN (ppm)

Nitrophenol (ppm)

Acetaldehyde (ppm)


0.000

0.001

0.002

0.003

0.004

0.005

0 120 240 360

0.00

0.02

0.04

0.06

0.08

0 120 240 360

0.00

0.04

0.08

0.12

0 120 240 360

0.0

0.2

0.4

0.6

0 120 240 360

0.000

0.005

0.010

0.015

0 60 120 180

0.0000

0.0001

0.0002

0.0003

0.0004

0.0005

0 60 120 180

0.00

0.04

0.08

0.12

0 60 120 180


0.000

0.002

0.004

0.006

0 120 240 360 480

0.00

0.01

0.02

0.03

0.04

0.05

0 120 240 360 480

0.00

0.04

0.08

0.12

0 120 240 360 480

0.0

0.1

0.2

0.3

0.4

0.5

0 120 240 360 480

0.000

0.005

0.010

0.015

0 120 240

0.0000

0.0001

0.0002

0.0003

0 120 240

0.00

0.04

0.08

0.12

0 120 240


Expt.

Calc

B-12

in the two experiments with valid 2-nitrophenol data, it significantly underpredicts the consumption rates for acetaldehyde in the experiments with this compound. This could be due to the underprediction of OH radicals in the model simulations, perhaps due to underprediction of overall NOx levels, and thus overprediction of termination due to peroxy + peroxy reactions.

Note that the relative contribution of reaction with OH to degradation of 2-nitropheno is expected to be smaller than that for acetaldehyde under experimental conditions for EPA1441A and EPA1442A. The major consumption reactions for the NPHE model species (used to represent 2-nitrophenol) are photolysis and reaction with OH radicals, with the photolysis reaction being predicted to be the major loss process under the conditions of these experiments. About 10% of the photolysis is represented as forming HONO but the remainder is represented as forming no active NOx species; i.e., it is assumed that NOx is released only 10% of the time when NPHE is photolyzed. The data clearly indicate that more NOx is released when 2-nitrophenol photolyzes, and probably the mechanism for the NPHE species should be modified accordingly. However, if this results in underprediction of NOx sinks in cresol - NOx and aromatic hydrocarbon - NOx experiments, then other NOx sink processes would have to be added, or a nitrophenol model species may not be a good representation for the actual set of products it is being used to represent.

Simulations of Isoprene - NOx Experiments Isoprene is represented explicitly in the SAPRC-07 mechanism, as is the case for most other mechanisms currently used in regional air quality models. Figure B-8 shows experimental and calculated data for the three isoprene - NOx experiments that were carried out for this project. As discussed in the main body of the report, the NOx is injected in these experiments entirely in the form of NO, and the default model assumes that no NO2 is initially present. However, the Carbon Bond mechanism was found to give better simulations of initiation times if some NO2 is also assumed to be present. This could be due to some of the NO being converted to NO2 during the injection process, due to the reaction of NO with O2 before the NO is diluted. Although the conversion should not be large because the NO is diluted with N2 prior to injection, but a conversion of up to 0.5% is not unreasonable. Because of this, model calculations assuming that 0.5% of the initial NOx is in the form of NO2 are also shown on Figure B-8.

The SAPRC-07 mechanism was found to reasonably simulate the initiation times in these experiments if 0.5% of the NOx is assumed to be in the form of NO2, though the underprediction of the initiation with no initial NO2 present is not large. The model tends to underpredict peak O3 levels in two out of the three experiments, despite the fact that it gives reasonably good simulations of the effects of isoprene on O3 formation in the NOx sink experiments, as shown in Figure B-5. The final NO2 levels in the experiments are simulated reasonably well. The model tends to underpredict isoprene consumption rates during the later stages of the experiments. This may be due to the model not having enough radical regeneration processes under low NOx conditions. More experiments would be useful to evaluate the consistency of these apparent biases in the model and to assess whether changes to the isoprene mechanism may be appropriate.

B-13

Figure B-8. Plots of selected results of the isoprene - NOx experiments carried out for this project. Results of SAPRC-07 model calculations are also shown.

Summary This Appendix presented results for (1) characterization of control (base-case) experiments, (2) evaluating impact of adding NOx-source generating compounds (toluene, o-cresol, furan (as in-situ 2-butenedial) and isoprene), (3) evaluating NOx release from NOx sources (isopropyl nitrate, isobutyl nitrate, 2-nitrophenol) and (4) evaluating isoprene mechanisms under relatively low-NOx conditions (i.e., [NOx]o < 30 ppb). In general, control experiments were reasonably simulated by SAPRC-07 (Figure B-1 and Figure B-2). Addition of toluene, o-cresol, furan or isoprene resulted in lower peak ozone formation due to removal of active NOx caused by addition of one of these NOx-sinks generating compounds (Figure B-3, Figure B-4 and Figure B-5). NO2 and/or PAN (NOx captured in the form of PAN) measured for the NOx-sources

EPA1397A EPA1405A EPA1405BOzone (ppm)

NO (ppm)

NO2 (ppm)

Isoprene (ppm)


0.00

0.04

0.08

0.12

0.16

0 120 240 360 480

0.00

0.01

0.02

0.03

0 120 240 360 480

0.000

0.005

0.010

0.015

0.020

0 120 240 360 480

No Data

0.0

0.1

0.2

0.3

0 120 240 360 480

0.00

0.04

0.08

0.12

0 60 120 180 240 300

0.000

0.005

0.010

0.015

0.020

0 60 120 180 240 300

0.000

0.004

0.008

0.012

0 60 120 180 240 300

0.00

0.05

0.10

0.15

0.20

0.25

0 60 120 180 240 300

0.00

0.01

0.02

0.03

0.04

0.05

0 60 120 180

0.000

0.002

0.004

0.006

0.008

0 60 120 180

0.000

0.001

0.002

0.003

0.004

0 60 120 180

CEASLuminol

0.00

0.05

0.10

0.15

0.20

0.25

0 60 120 180

Experimental

Calculated

Calc with 0.5%NO2

B-14

experiments showed that NOx can be recycled to various degrees from organic nitrates (e.g., isopropyl nitrate and isobutyl nitrate; Figure B-6) and organic nitrate formed from aromatics (e.g., 2-nitrophenol; Figure B-7) and these recycled NOx can contribute to ozone formation under low-NOx conditions. These experimental data for NOx sinks and sources demonstrated the importance of NOx removal and recycling processes in ozone formation chemistry. A limited amount of experimental data for isoprene under low-NOx conditions showed that the current isoprene scheme reasonably simulates ozone formation from isoprene under experimental conditions down to a range of 5 ppb ~ 30 ppb NOx but has difficulty in simulating radical formation from isoprene under low-NOx conditions (Figure B-8).

References Carter (2000): Carter, W.P.L., 2000. Documentation of the SAPRC-99 chemical mechanism for

VOC reactivity assessment, Report to the California Air Resources Board, Contracts 92-329 and 95-308. (http://www.engr.ucr.edu/~carter/absts.htm#saprc99)

Carter (2010): Carter, W.P.L., 2010. Development of the SAPRC-07 chemical mechanism and updated ozone reactivity scales, Final report to the California Air Resources Board, Contract No. 03-318, 06-408, and 07-730, Revised January 27, 2010. (http://www.cert.ucr.edu/~carter/SAPRC/)

Lonneman (1986): Comparison of 0600–0900 AM hydrocarbon compositions obtained from 29 cities. In: Proceedings of the 1986 APCA/U.S. EPA Symposium on Measurements of Toxic Air Pollutants, Raleigh, NC.

http://www.engr.ucr.edu/~carter/absts.htm#saprc99

http://www.cert.ucr.edu/~carter/SAPRC/

Appendix C

Summary of Chamber Experiments Performed for This Project

C-1

Table C-1. Summary of Chamber Experiments Performed for This Project.

Run ID Type Time Temp. Dilution Test Reactant Initial Concentrations (ppm)

(min) (K) (min-1) Compound (ppm) NO NO2 H2O2 CO Ethene Propene Acetald. Tracer

NOx Sink Experiments

EPA1401A Toluene - Ethene - NOx 207 298.8 - Toluene 0.367 0.000 0.013 1.10 0.073

EPA1401B Ethene - NOx 110 298.8 - 0.000 0.013 1.07 0.075

EPA1402A Furan - Ethene - NOx 288 297.3 5.0e-5 Furan 0.129 0.003 0.010 1.10 0.059

EPA1402B Ethene - NOx 288 297.3 7.0e-5 0.003 0.010 1.11 0.060

EPA1403A Furan - Ethene - NOx 289 297.4 - Furan 0.047 0.013 0.98 0.072

EPA1403B Furan - Ethene - NOx 289 297.4 - Furan 0.050 0.013 1.06 0.073

EPA1404A Isoprene - Ethene - NOx 343 298.2 - Isoprene 0.051 0.000 0.014 1.14 0.073

EPA1404B Ethene - NOx 223 298.2 - 0.000 0.014 1.14 0.073

EPA1407A Toluene - Ethene - NOx 363 298.8 2.0e-5 Toluene 0.344 0.014 0.000 0.84 0.050

EPA1407B Ethene - NOx 259 298.8 - 0.014 0.000 0.84 0.050

EPA1408A o-Cresol - Ethene - NOx 368 298.4 - o-Cresol 0.098 0.016 1.00 0.072

EPA1408B Ethene - NOx 368 298.4 - 0.015 1.01 0.072

EPA1418A Toluene - Propene - NOx 421 298.2 - Toluene 0.301 0.016 0.000 0.303 0.075

EPA1418B Propene - NOx 270 298.2 3.0e-5 0.015 0.000 0.303 0.075

C-2



EPA1436A Toluene - Propene - NOx 348 298.1 5.0e-5 Toluene 0.391 0.014 0.431 0.104

EPA1436B Propene - NOx 225 298.1 2.0e-5 0.014 0.443 0.106

EPA1443A Toluene - Propene - NOx 481 298.0 - Toluene 0.407 0.017 0.001 0.305

EPA1443B Propene - NOx 342 298.0 - 0.017 0.001 0.307

EPA1444A Isoprene - Ethene - NOx 509 298.1 - Isoprene 0.017 0.014 0.000 0.80

EPA1444B Isoprene - Ethene - NOx 252 298.1 - Isoprene 0.017 0.014 0.000 0.80

EPA1446A Isoprene - Ethene - NOx 369 298.5 - Isoprene 0.050 0.001 0.013 0.92

EPA1446B Ethene - NOx 214 298.5 - 0.001 0.014 0.92

EPA1448A Furan - Ethene - NOx 469 297.9 - Toluene 0.130 0.015 0.88

EPA1448B Ethene - NOx 195 297.9 - 0.015 0.88

NOx Source Experiments

EPA1439A Isobutyl nitrate - Acetald. - H2O2 545 298.8 1.0e-4 Isobutyl nitrate 0.375 0.507 1.001 0.058

EPA1439B Isobutyl nitrate - CO - H2O2 238 298.8 1.0e-4 Isobutyl nitrate 0.371 0.507 140.5 0.058

EPA1441A 2-Nitrophenol - Acetald. - H2O2 338 298.1 5.0e-5 2-Nitrophenol 0.109 0.000 0.507 0.531 0.124 *

EPA1441B 2-Nitrophenol - CO - H2O2 220 298.1 - 2-Nitrophenol 0.109 0.000 0.507 126.5 0.125 *

EPA1442A 2-Nitrophenol - Acetald. - H2O2 482 298.1 - 2-Nitrophenol 0.111 0.507 0.423

EPA1442B 2-Nitrophenol - CO - H2O2 342 298.1 - 2-Nitrophenol 0.111 0.507 25.3

C-3



EPA1449A Isopropyl nitrate - Acetald.-H2O2 447 298.0 - Isopropyl nitrate 0.500 0.507 0.520

EPA1449B Isopropyl nitrate - CO - H2O2 333 298.0 - Isopropyl nitrate 0.505 0.507 24.0

Isoprene Experiments

EPA1397A Isoprene - NOx 477 299.6 2.0e-5 Isoprene 0.250 0.024 0.078

EPA1405A Isoprene - NOx 326 299.0 - Isoprene 0.192 0.016 0.072

EPA1405B Isoprene - NOx 207 299.0 - Isoprene 0.200 0.006 0.075

Other Single VOC Experiments

EPA1456A CO - NOx 363 298.0 5.0e-5 0.016 0.000 49.2 0.079

EPA1456B CO - NOx 283 298.0 3.0e-5 0.015 0.000 50.4 0.076

EPA1400A Ethene - NOx 265 298.5 - 0.000 0.013 0.91 0.067

EPA1391A Propene - NOx 246 298.3 5.0e-5 0.000 0.014 0.311 0.072

EPA1391B Propene - NOx 227 298.3 4.0e-4 0.000 0.014 0.313 0.073

EPA1395A Propene - NOx 225 299.8 4.5e-4 0.002 0.010 0.321 0.069

EPA1395B Propene - NOx 157 299.8 - 0.002 0.011 0.321 0.069

EPA1409A Propene - NOx 489 298.5 - 0.014 0.328 0.058

C-4



Control Experiments

EPA1415B Acetaldehyde - H2O2 - NOx 257 298.6 - 0.024 0.001 1.015 0.665 0.068

EPA1447A Acetaldehyde - H2O2 - NOx 372 298.6 - 0.001 0.012 0.507 0.396

EPA1415A BIACETYL - CO - NOx 382 298.6 - Biacetyl 0.480 0.027 0.000 56.3 0.069

Background NOx Experiments

EPA1428A Acetaldehyde - H2O2 497 298.2 4.0e-5 0.507 0.340 0.055

EPA1428B Acetaldehyde - H2O2 309 298.2 5.0e-5 0.507 0.340 0.055

EPA1429A CO - H2O2 416 298.3 2.0e-5 0.507 45.4 0.070

EPA1429B CO - H2O2 266 298.3 1.0e-5 0.507 45.5 0.070

EPA1431A CO - H2O2 373 298.5 - 0.507 48.5 0.073

EPA1431B CO - H2O2 283 298.5 - 0.507 48.6 0.073

EPA1434A Acetaldehyde - H2O2 511 297.8 5.0e-5 0.507 0.520 0.071

EPA1434B Acetaldehyde - H2O2 418 297.8 1.0e-4 0.507 0.520 0.071

EPA1447B Acetaldehyde - H2O2 211 298.6 - 0.507 0.396

Notes:

* Tracer was perfluoro n-hexane (n-C6F14) for all experiments except for EPA1441, where perfluorobenzene (C6F6) was used

FINAL REPORT Environmental Chamber Experiments to ...aqrp.ceer.utexas.edu/projectinfo/10-042/10-042...

Documents

Transcript of FINAL REPORT Environmental Chamber Experiments to ...aqrp.ceer.utexas.edu/projectinfo/10-042/10-042...