CMAQv5.0 POA aging

Heather Simon

The POA aging scheme described below (and results/evaluation) were published in Simon and Bhave (2012): +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

CMAQ simulates POA aging as a second order reaction between reduced primary organic carbon (POCR) and OH radicals. Here POCR is defined as the moles of carbon (from model species APOCI/J) minus the moles of oxygen in PNCOM (Equation (1)).

Equation (1): POCR=C-O

Modeling aging as a function of POCR relies on the assumption that each carbon atom in the ambient OM can only bond with a single oxygen atom. This also means that by design, POA that is already highly oxidized ages at slower rate and to a lesser extent than POA that is not highly oxidized.

Equation (2) is used to calculate C.

Equation (2): C = (APOCI/J)/12

Equations (3)-(9) are used to calculate O. We use the recent work by Heald et al . (2010) to derive a relationship between OM/OC and the hydrogen to oxygen ratio (H/O) in OM in order to calculate O. Note that although we derive the relationship of OM/OC to H/O in all OM, here we apply it only to POA since we are only aging the primary portion of the OM. Heald et al. (2010) map OM from several field and lab studies to a Van Krevelen diagram with H:C on the y-axis and O:C on the x-axis (both in molar units). They find that for varied types of OM, H:C and O:C have a linear relationship fitting to

Equation (3): O/C+H/C=2

In order to convert PNCOM into moles of O, we make two assumptions: 1) primary OM falls on the line in the Van Krevelen diagram described by Equation (3) and 2) NCOM is comprised completely of oxygen and hydrogen (contributions of other elements to NCOM mass are negligible) (Equation (4)).

Equation (4): OM=H*1+O*16+C*12

In Equation (4), OM is given in units of grams and H, O, and C are given in units of moles. If we divide Equation (4) by OC (units of grams or the equivalent 12*C) and combine with Equation (3), we can derive a relationship between OM/OC and O/C or H/C.

Equation (5): O/C= (OM/OC-14/12)*12/15

Equation (6): H/C=(44/12-OM/OC)*12/15

By dividing Equation (6) by equation (5) we derive a relationship between H/O and OM/OC:

Equation (7): H/O=(H/C)/(O/C)=((44/12-OM/OC))/((OM/OC-14/12))

Once we obtain H/O, a simple mass balance on NCOM can be used to determine moles of oxygen:

Equation (8a): H*1+O*16=O*(H/O)+O*16=NCOM

Equation (8b): O = NCOM/(16+(H/O))

However, limits must be set on OM/OC to avoid a negative H/O. Any OM with OM/OC <= 14/12, can be thought of as fully reduced, and we thus model the NCOM as containing no oxygen. Any OM with OM/OC >= 44/12 can be thought of as being fully oxidized, and therefore we model NCOM as containing pure oxygen (Equation (9)).

Equation (9a): if OM/OC <= 14/12; O = 0

Equation (9b): if OM/OC >= 44/12; O = NCOM/16

Equation (9c): if 14/12 <= OM/OC <= 44/12; O = NCOM/(16+(H/O))

Heald et al. (2010) also point out that the slope of their fitted line in the Van Krevelen diagram is -1, which equates to the conversion of a methyl group (CH3) to a carboxylic acid functional group (COOH). This should not be interpreted to mean that any particular oxidation reaction leads to the addition of a carboxylic acid, but instead that on average as the bulk OM ages it follows this trend. In keeping with this finding, we use a molecular weight of 15 g/mol for every new PNCOM molecule formed through aging, meaning that on average for every oxygen atom added to the mass, a hydrogen atom is lost. Therefore, in CMAQ, as POA ages it moves to the right and down on the line defined by equation (3). Thus CMAQ POA aging is modeled using Reaction (1) and Equation (10).

Reaction (1): POCR + OH → OH + PNCOM + POCO

Equation (10): [PNCOM]_(t+∆t)= [PNCOM]_t+15[POCR]x(1-e^{(-k_OH,eff [OH]∆t)} )

In Reaction 1, POCO represents oxidized POC to show that POC is not being destroyed in this reaction, it is just being converted from reducted to oxidized form.  However, POC is the transported primary carbon species in CMAQ; POCR and POCO are not transported species.  Note that OH is not consumed by Reaction (1). This is consistent with sesquiterpene reactions in CMAQ, which also do not consume oxidants so as not to affect the gas-phase chemical mechanisms which were developed without explicit links to particle formation and processing.

Based on studies by George et al. (2007), Weitkamp et al (2008), and Lambe et al. (2009), we choose an effective rate constant for Reaction 1 (K_OH,eff) equal to 0.25x10^-11 cm³molec^-1s^-1 for this study. We do not vary the effective rate constant with temperature in our model since we are not aware of any studies which evaluate its dependence on this variable.

Affected files: AERO_DATA.F, PRECURSOR_DATA.F, aero_subs.F, poaage.F (new), GC_cb05cl_ae6_aq_recon.nml, GC_cb05cl_ae6_aq_recon.csv, Makefile

References:

George, I. J.; Vlasenko, A.; Slowik, J. G. et al., Heterogeneous oxidation of saturated organic aerosols by hydroxyl radicals: uptake kinetics, condensed-phase products, and particle size change.

Heald, C. L.; Kroll, J. H.; Jimenez, J. L. et al., A simplified description of the evolution of organic aerosol composition in the atmosphere. Geophys. Res. Lett. 2010, 37.

Lambe, A. T.; Miracolo, M. A.; Hennigan, C. J. et al., Effective Rate Constants and Uptake Coefficients for the Reactions of Organic Molecular Markers (n-Alkanes, Hopanes, and Steranes) in Motor Oil and Diesel Primary Organic Aerosols with Hydroxyl Radicals. Environ. Sci. Technol. 2009, 43 (23), 8794-8800.

Simon, H., Bhave, P.V.; Simulating the degree of oxidation in atmospheric organic particles. Environ. Sci. Technol. 2012, 46, 331-339.

Weitkamp, E. A.; Lambe, A. T.; Donahue, N. M. et al., Laboratory Measurements of the Heterogeneous Oxidation of Condensed-Phase Organic Molecular Makers for Motor Vehicle Exhaust. Environ. Sci. Technol. 2008, 42 (21), 7950-7956.