Limited Role of Malonic Acid in Sulfuric Acid–Dimethylamine New Particle Formation

Aerosols play an important role in climate and air quality; however, the mechanisms behind aerosol particle formation in the atmosphere are poorly understood. Studies have identified sulfuric acid, water, oxidized organics, and ammonia/amines as key precursors for forming aerosol particles in the atmosphere. Theoretical and experimental investigations have indicated that other species, such as organic acids, may be involved in atmospheric nucleation and growth of freshly formed aerosol particles. Organic acids, such as dicarboxylic acids, which are abundant in the atmosphere, have been measured in ultrafine aerosol particles. These observations suggest that organic acids may contribute to new particle formation in the atmosphere but their role remains ambiguous. This study examines how malonic acid interacts with sulfuric acid and dimethylamine to form new particles at warm boundary layer conditions using experimental observations from a laminar flow reactor and quantum chemical calculations coupled with cluster dynamics simulations. Observations reveal that malonic acid does not contribute to the initial steps (formation of <1 nm diameter particle) of nucleation with sulfuric acid-dimethylamine. In addition, malonic acid was found to not participate in the subsequent growth of the freshly nucleated 1 nm particles from sulfuric acid-dimethylamine reactions to diameters of 2 nm.

(a) shows the malonic acid baselines with the PCC taken for 5 days over a period of 6 weeks. These baseline measurements relate the measured [MaA2] detected at 207 m/z, 267 m/z, and 327 m/z, to [MaA1] detected at 103 m/z by the PCC using acetate as the chemical ionization reagent ion. The malonic acid baselines show that [MaA1] varied slightly across days. This is because the malonic acid reservoir had to be refilled often as the solution readily vaporized over the course of 3-4 days. Refilling the malonic acid reservoir during these few days led to slight differences in [MaA1] for the same nitrogen flow rates over the reservoir. Nonetheless, the slope of [MaA2] vs. [MaA1] remained constant. A constant slope between the sulfuric acid dimer vs. acid monomer has previously been used to indicate the absence of stabilizing bases which generally enhances the acid dimer formation. 1-3   Figure S1(b) shows the sulfuric acid dimer concentration ([SA2]) versus sulfuric acid monomer concentration ([SA1]) taken on the same days as the malonic acid baselines shown in figure S1(a). The sulfuric acid baseline obtained by changing the injection flowrate of sulfuric acid between 10-100 sccm relies on consistent sulfuric acid monomer (97 m/z for acetate chemical ionization) and dimer (195 m/z) concentrations with no base added to the flow reactor. The total reactor flow rate did not change significantly during the sulfuric acid baseline since this small sulfuric acid flow was injected into the 4.5 sLpm flow of the reactor. Generally, the sulfuric acid baselines are consistent across the different days but deviate slightly at the lower sulfuric acid concentration. This is because it is difficult to obtain repeatable sulfuric acid concentrations in the reactor at the lower injection flow rates since the flow rates at the lower range of the mass flow controller are not very repeatable. The presence of base contaminants in the reactor would have led to deviations in measured [N2] was not observed in figure S1(b).

Section 2: Flow reactor baseline conditions with the vWCPC
Daily baseline measurements were also taken with the versatile water condensation particle counter (vWCPC, TSI 3789), 4 which has a 50% detection efficiency for 1 nm ions (geometric diameter). The vWCPC sampled from one of the side ports of the reactor as shown in Figure 1.
The reactor and the vWCPC were operated such that the total flowrate in the reactor was kept constant for all the baseline measurements taken over 3 days as shown in Figure S2. The vWCPC baselines rely on monitoring the concentration of 1nm particles that were present in the reactor in the absence of added stabilizing bases like DMA. For a given sulfuric acid concentration, the malonic acid flowrate injected into the reactor was varied to observe the dependence of background particle concentrations on both sulfuric acid and malonic acid. As shown in Figure S2, the concentration of 1 nm particles generally does not vary with malonic acid flowrate injected into the reactor. However, at higher sulfuric acid concentrations, [SA1] = 8 x 10 7 cm -3 and [SA1] = 3 x 10 8 cm -3 , there is a small increase in the concentration of 1 nm particles at higher malonic acid concentrations. It is possible that at higher sulfuric acid injection flow rates, small amounts of base contaminants in the reactor could react with sulfuric acid to form 1 nm particle. Increasing the concentration of sulfuric acid injected into the reactor led to an increase in measured 1 nm particle concentrations. It is likely that sulfuric acid nucleated with residue DMA or other contaminants in the wick of the vWCPC to form these particles since background [DMA] in the reactor was less than 1 pptv.
Where, zi is the factor describing the mass transmission efficiency discrimination of the measured HSO4relative to the acetate reagent ion and tCI is the chemical ionization reaction time. In this study, the mass-dependent transmission efficiencies were obtained from Henritizi et al. 6 for a timeof-flight chemical ionization mass spectrometer with a similar mass filter as the PCC. 1 , the forward rate constant for reaction of acetate with sulfuric acid was previously measured to be 4.6 x 10 −9 cm 3 s −1 . 5 Sreagent for acetate is the sum of the signals at 59 m/z (CH 3 CO 2 − ), 77 m/z (H 2 O • CH 3 CO 2 − ), and 119 m/z (CH 3 CO 2 H • CH 3 CO 2 − ).

Equation 1
can also be used to solve for the malonic acid monomer concentration ([MaA1]) using the ratio of the corresponding malonate ion signal (C3H2O4 -) to the sum of the acetate reagent ion signals. Suppose malonic acid is ionized by acetate at the same rate constant as sulfuric acid assumed in this study ( 1 = 4.6 x 10 −9 cm 3 s −1 ). In that case, the malonic acid concentrations For the sulfuric acid malonic measurements, the tCI was ~ 25 ms based on a 4.8 cm plate distance between the source and PCC inlet orifice, 84 V/cm electric field strength, and an acetate ion mobility of 2.32 cm 2 / s•V. 7 The cluster thermodynamic properties used for ACDC simulations are presented in Table S1. The structure of the MaA1•SA1 heterodimer cluster is shown in Figure S3. Other properties as well as other structures for the several lowest binding free energy configurations of each cluster can be downloaded from the GitHub repository (under Kubecka and Fomete folders):

Section 4: The cluster structures and thermodynamic properties
https://github.com/elmjonas/ACDB/database_v2/DLPNO_vnw16/ Figure S3: The structure of the MaA1•SA1 heterodimer showing the ability of hydrogen bonding between the two compounds. Figure S4 show the concentrations of the MaA-SA dimer, the SA dimer, and all clusters containing two SA as a function of malonic acid concentration after the 10 s ACDC simulations. Figure S5 shows the particle formation rates at the end of the simulation. Clearly, the theoretical calculations and modelling do also suggest that MaA does not affect the NPF of SA-DMA system.