Particles composed of secondary organic material (SOM) are abundant in the lower troposphere. The viscosity of these particles is a fundamental property that is presently poorly quantified yet required for accurate modeling of their formation, growth, evaporation, and environmental impacts. Using two unique techniques, namely a "bead-mobility" technique and a "poke-flow" technique, in conjunction with simulations of fluid flow, the viscosity of the water-soluble component of SOM produced by α-pinene ozonolysis is quantified for 20-to 50-μm particles at 293-295 K. The viscosity is comparable to that of honey at 90% relative humidity (RH), similar to that of peanut butter at 70% RH, and at least as viscous as bitumen at ≤30% RH, implying that the studied SOM ranges from liquid to semisolid or solid across the range of atmospheric RH. These data combined with simple calculations or previous modeling studies are used to show the following: (i) the growth of SOM by the exchange of organic molecules between gas and particle may be confined to the surface region of the particles for RH ≤ 30%; (ii) at ≤30% RH, the particle-mass concentrations of semivolatile and low-volatility organic compounds may be overpredicted by an order of magnitude if instantaneous equilibrium partitioning is assumed in the bulk of SOM particles; and (iii) the diffusivity of semireactive atmospheric oxidants such as ozone may decrease by two to five orders of magnitude for a drop in RH from 90% to 30%. These findings have possible consequences for predictions of air quality, visibility, and climate.aerosol | physical properties | secondary organic aerosol B iological sources (e.g., vegetation) and anthropogenic sources (e.g., transportation) emit copious quantities of volatile organic compounds, such as α-pinene and aromatic hydrocarbons, among others (1, 2). In the atmosphere, a complex series of chemical reactions oxidizes these volatile compounds to form semivolatile organic compounds (SVOCs) that condense to the particle phase (1, 2). This secondary organic material (SOM) constituting the particles is estimated to contribute typically 30-70% to the mass concentration of suspended submicron particles in most regions of the atmosphere (1). These particles can influence climate by scattering and absorbing solar radiation (direct climate effect) and by serving as nuclei for cloud formation (indirect climate effect), among other mechanisms (3). They can also influence air quality and health (4-6).Recently, molecular diffusion within SOM particles has become an area of intense scientific interest. Diffusion rates within particles can influence the mechanism and rates of growth of SOM particles (Fig. 1A) (7,8) and influence reactions of oxidants within the SOM particles (Fig. 1B) (9). As a result, quantitative modeling of the environmental impacts of SOM particles can depend on molecular diffusion within the particles (see, for example, Fig. 1C). Shiraiwa and Seinfeld (10) have shown that predictions of the mass concentration of SOM particles, a ke...

[1] This study investigated long-term chemical aging of model biogenic secondary organic aerosol (SOA) prepared from the ozonolysis of terpenes. Techniques including electrospray ionization mass spectrometry (ESI-MS), UV-visible spectroscopy, Fourier transform infrared (FTIR) spectroscopy, NMR, and three-dimensional fluorescence were used to probe the changes in chemical composition of SOA collected by impaction on substrates and also of aqueous extracts of SOA. The addition of ammonium ions or amino acids to limonene SOA reproducibly produced orange-colored species that strongly absorbed visible radiation and fluoresced at UV and visible wavelengths. Simultaneous addition of H 2 SO 4 to the SOA aqueous extracts inhibited this color transformation. These observations suggest the existence of aging processes leading to heavily conjugated molecules containing organic nitrogen. The presence of nitrogen in the chromophores was confirmed by the dependence of the absorption and fluorescence spectra on the amino acids added. In contrast to the strong change in the absorption and fluorescence spectra, there was no significant change in the ESI-MS, FTIR, and NMR spectra, suggesting that the chromophores were minor species in the aged SOA. Aqueous extracts of aged limonene + NH 4 + SOA were characterized by an effective base-e absorption coefficient of $3 L g À1 cm À1 at 500 nm. Assuming particulate matter concentrations typical of polluted rural air gives an upper limit of 0.2 M m À1 for the aerosol absorption coefficient due to the aged limonene oxidation products. Biogenic SOA can therefore become weakly absorbing if they undergo aging in the presence of NH 4 + -containing aerosol.

This work presents a systematic investigation of the molecular level composition and the extent of aqueous photochemical processing in different types of secondary organic aerosol (SOA) from biogenic and anthropogenic precursors including α-pinene, β-pinene, β-myrcene, Dlimonene, α-humulene, 1,3,5-trimethylbenzene, and guaiacol, oxidized by ozone (to simulate a remote atmosphere) or by OH in the presence of NO x (to simulate an urban atmosphere). Chamber-and flow-tube-generated SOA samples were collected, extracted in a methanol/water solution, and photolyzed for 1 h under identical irradiation conditions. In these experiments, the irradiation was equivalent to about 3−8 h of exposure to the sun in its zenith. The molecular level composition of the dissolved SOA was probed before and after photolysis with direct-infusion electrospray ionization high-resolution mass spectrometry (ESI-HR-MS). The mass spectra of unphotolyzed SOA generated by ozone oxidation of monoterpenes showed qualitatively similar features and contained largely overlapping subsets of identified compounds. The mass spectra of OH/NO x -generated SOA had more unique visual appearance and indicated a lower extent of product overlap. Furthermore, the fraction of nitrogen-containing species (organonitrates and nitroaromatics) was highly sensitive to the SOA precursor. These observations suggest that attribution of high-resolution mass spectra in field SOA samples to specific SOA precursors should be more straightforward under OH/NO x oxidation conditions compared to the ozone-driven oxidation. Comparison of the SOA constituents before and after photolysis showed the tendency to reduce the average number of atoms in the SOA compounds without a significant effect on the overall O/C and H/C ratios. SOA prepared by OH/NO x photooxidation of 1,3,5-trimethylbenzene and guaiacol were more resilient to photolysis despite being the most light-absorbing. The composition of SOA prepared by ozonolysis of monoterpenes changed more significantly as a result of the photolysis. The results indicate that aqueous photolysis of dissolved SOA compounds in cloud/fog water can occur in various types of SOA, and on atmospherically relevant time scales. However, the extent of the photolysis-driven change in molecular composition depends on the specific type of SOA.

The effect of relative humidity (RH) on the rebound of particles composed of isoprene, α-pinene, and toluene secondary organic materials (SOMs) was studied. A three-arm impaction apparatus was used to study rebound from 5 to 95% RH at 298 K. Calibration experiments using sucrose particles of variable but known viscosities showed that the transition from rebounding to adhering particles occurred for a change in viscosity from 100 to 1 Pa s, corresponding to a transition from semisolid to liquid material. The experimentally determined rebound fractions of the studied SOMs were compared with results from a model of the rebound processes of hard particles, taking into account the particle kinetic energy, van der Waals forces, and RH-dependent capillary forces. For low RH values, the hard-particle model explained the diameter-dependent rebound behavior for all studied SOMs. For elevated RH, however, the experimental observations deviated from the model predictions. On the basis of the calibration experiments using sucrose particles as well as a comparison between the observations and the predictions of the hard-particle model, the interpretation is made that a semisolid-to-liquid transition occurred at elevated RH. Material softening, increased adhesion, or a combination of the two implied the action of additional modes of energy relaxation that were not included in the hard-particle model. The RH threshold for the semisolid-to-liquid phase transition was 40% RH for isoprene SOM, 70% for toluene SOM, and 70% for α-pinene SOM. A correlation between the rebound fraction and the hygroscopic growth factor G was demonstrated, implying that absorbed water volume was a dominant governing factor of the semisolid-to-liquid transition for the studied classes of SOM. Simple heuristic rules based on G of 1.15 for the semisolid-to-liquid phase transition could be used for prognostication of the SOM phase in modeling applications at 298 K. With respect to atmospheric processes, the findings of this study suggest that both the chemical composition and the RH influence the phase state of organic particles. The findings can explain reports of solid organic particles for terpene-dominant conditions of a boreal forest at low RH compared to reports of liquid organic particles for isoprene-dominant tropical forests at high RH.

No abstract