Droplet activation measurement

The number concentration of activated particles was measured with a Cloud Condensation Nuclei Counter (CCNC, Droplet Measurement Technologies, CCN-100) for supersaturations (SS=RH-100%) between 0.17 and 1.1% with the setup described previously⁶¹. In parallel, the particle size distribution (15.1–399.5 nm) and total number concentration were measured with a SMPS system (TSI Model 3071 and CPC 3022A) and a water-based CPC (TSI, Model 3785), respectively. Before entering the instruments, the poly-dispersed aerosol was dried with a silica gel diffusion drier to RH <5% and then neutralized with a Kr-85 neutralizer (TSI Model 3077).

The activated fraction was calculated as the ratio of the number concentration of activated particles to the total particle number concentration. The calculated activated fraction was compared with the cumulative size distribution starting from the maximum diameter. The size, at which the cumulative particle size distribution was equal to the activated fraction, was defined as dry critical activation diameter (D_crit). It was assumed that the aerosol particles were internally mixed. This approach is applicable to the period when the nucleation already stopped i.e., we measured the CCN activity at the period when the conditions in the RC reached the steady state. Five different SS were set in the CCNC: the first step for 20 min, the others for 10 min each. To ensure stable conditions in the CCNC, only data from the last 6 min of each SS step were used to determine D_crit, which overlapped with three SMPS scans each lasting 2 min. For each set of SS, the D_crit values derived from three size scans were averaged. The contribution of multiple charged particles was corrected with the measured size distribution assuming a natural charge distribution. Ammonium sulfate aerosol was used to calibrate the SS of the CCNC based on data sets in the literature⁶² (OS1 data set therein).

From the D_crit and SS data, the hygroscopicity parameter κ was determined using the method in Petters and Kreidenweis²⁰. Petters and Kreidenweis²⁰ developed a theory to parameterize CCN activity data using κ based on Köhler theory⁶³. κ is a measure of the hygroscopicity of particles, which is defined in the following equation:

a_w, V_s and V_w are the water activity, volume of solute and volume of water in an activated droplet, respectively.

The following equation can be derived by using equation (1) in the original κ-Köhler theory.

S: saturation ratio, S=SS+1;

D_p: droplet diameter;

D_dry: dry particle diameter;

M_w: molecular weight of water;

σ_sol: surface tension of droplet solution;

ρ_w: density of water.

R: gas constant (8.314 J mol⁻¹ K⁻¹)

T: temperature.

σ_sol is assumed to be equal to that of water. Although organics can partition to the droplet surface, using the surface tension of water to calculate κ is a reasonable assumption for droplet at activation^64,65.

Comparing the κ definition in the κ-Köhler theory to other parameterizations of water activity (namely van't Hoff factor approach) yields²⁰,

where ρ_s and ρ_w are the density of solute and water, and M_s and M_w are the molecular weight of solute and water, respectively. i is the van't Hoff factor. It is the actual number of molecules or ions produced per solute molecule when a substance is dissolved^66,67,68. Since most organics do not dissociate, i is close to 1. The variability of density of most organics in SOA is small and can be assumed to be constant. equation (3) shows that κ is inversely proportional to the molecular weight of solute assuming other factors are relatively constant.

κ of SOA formed from the photooxidation of VOC emitted by unstressed trees and stressed trees were compared using t-test. κ of SOA formed at different plant temperature were also compared.