Heteregenous oxidation of aerosol organic matter

Although to a large extent the initial steps in reaction mechanism between OH radical and organic aerosol is known,

it is the later steps in the mechanism that are presently debated by the scientific community. Understanding the details of this mechanism is important if we intend to constrain the sources and sinks of particulate organic matter in the Figure 1 shows the proposed reaction mechanism for alkane oxidation by OH. First step in his reaction is the abstraction of H-atom from the organic molecule forming an alkyl radical. The second step is addition of O2 to form an organic peroxy radical (RO2). Under low NOx conditions, RO2 can either self-react or react with HO2 to form multiple compounds. Fate of this RO2 radical is what is being debated. Does it react by pathway 1 to form peroxides or does it self react by pathway 2 to form stable carbonyl and alcohol products or does it selfreact to form an alkoxy (RO) radical? It is the formation and subsequent decomposition of RO radical by pathway 4 which leads to mass loss by volatilization of small volatile organic compounds (VOCs).

Figure 1: Reaction mechanism for heterogeneous reaction
of OH with an alkane (from George et al. ACP 2007)

Experiments

            We used palmitic acid as a proxy for aerosol organic matter. Being an saturated acid it mainly reacts with OH. It is homogeneously nucleated in a heated glass tube while flowing N2 thought it. To this aerosol stream, Ozone (O3), wet N2 and O2 (to simulate atmopsheric conditions) are added. This sample mixture is irradiated with 254nm lamp in a quartz photocell. OH generated by O3 photolysis oxidizes PA aerosol. The reacted aerosol are analyzed for size/volume change using a differential mobility analyzer (DMA), and changes composition using heated inlet CIMS (read McNeill et al. JPC 2006) for more on aerosol CIMS set-up).

Results
            To understand and explain the experimental data (symbols in Figure 2), we developed a simple model which has 3 main processes:
                1) reaction of PA with OH on aerosol surface
                2) reaction of HO2 and self reaction of RO2
                3) a surface renewal process which exposes fresh PA to the surface.
In-addition we allow the particle size to change with oxidation. The 4 main parameters that are adjusted to fit the data to model at the uptake coefficient for OH, uptake coefficient for HO2, surface renewal rate, and the factor linking changing PA mass to aerosol radius. Figure 2 shows the loss of palmitic acid for four different volume weighted mean radii with changing OH concentrations.

There are few interesting things to notice

Loss of PA is nonlinear with OH exposure (loss rate is decreasing with increasing OH). This is due to the fact that as OH concentrations increases PA at the surface is harder to find and hence the rate of reaction decreases
Loss is greater for smaller size aerosol than the larger ones. This is because for smaller size aerosol a larger fraction of the mass is at the surface compared to the bulk and hence is easily accessible for reaction with OH.
Model fits the data very well with the same set of parameters for all the sizes.
Also noticable is that after 6 days of processing for a 90nm radius aerosol, the maximum volume loss is only 30% with the loss rate decreasing.

Figure 2: Loss of PA as a function of OH exposure for 4 different aerosol sizes

Change in PA aerosol size with OH exposure

Figure 3 shows the change in aerosol size as a function of OH exposure for a 160nm and a 90nm aerosol. Symbols are experimental data and solid lines are model fits. Again, the model reproduces data very well with OH uptake coefficient of ~ 1. Also, we see the same non-linear decrease in size with OH. To understand compositional changes with oxidation, we looked at mass spectrum from CIMS and observed production of low-molecular weight products, mainly in gas-phase. This indicates:

       1) PA mass loss by OH oxidation and production of low-molecular weight             products confirming production of RO and its subsequent
            decomposition through reaction pathway 4 (see Figure 1)
       2) Oxidation cannot be a major loss process, as suggested by Molina et al.,             for organic aerosol as the observed loss is only 30% even after 6 days of             processing.

Figure 3: Change in PA aerosol size as a function of OH exposure

We also hypothesis that reaction pathway 1 is possible in this system (formation of peroxides and their photolysis producing an RO radical). But, unfortunately our current set-up does not allow for furthur investigation of this pathway.

If you are interested in reading more about the experiments and model you can download our recent paper.
McNeill, V. F., Yatavelli, R. L. N., Thornton, J. A., Stipe, C. B., and Landgrebe, O.(2008), The heterogeneous OH oxidation of Palmitic Acid in single component and internally mixed aerosol particles: vaporization and the role of particle phase , Atms. Chem. Phys., 8, 5465-5476

Reaction Mechanism