Nanocluster Aerosols from Ozone–Human Chemistry Are Dominated by Squalene–Ozone Reactions

Nanocluster aerosols (NCAs, <3 nm particles) are associated with climate feedbacks and potentially with human health. Our recent study revealed NCA formation owing to the reaction of ozone with human surfaces. However, the underlying mechanisms driving NCA emissions remain unexplored. Squalene is the most abundant compound in human skin lipids that reacts with ozone, followed by unsaturated fatty acids. This study aims to examine the contribution of the squalene–ozone reaction to NCA formation and the influence of ozone and ammonia (NH3) levels. In a climate-controlled chamber, we painted squalene and 6-hexadecenoic acid (C16:1n6) on glass plates to facilitate their reactions with ozone. The squalene–ozone reaction was further investigated at different ozone levels (15 and 90 ppb) and NH3 levels (0 and 375 ppb). The results demonstrate that the ozonolysis of human skin lipid compounds contributes to NCA formation. With a typical squalene-C16:1n6 ratio found in human skin lipids (4:1), squalene generated 40 times more NCAs than did C16:1n6 and, thus, dominated NCA formation. More NCAs were generated with increased ozone levels, whereas increased NH3 levels were associated with the stronger generation of larger NCAs but fewer of the smallest ones. This study experimentally confirms that NCAs are primarily formed from squalene–ozone reactions in ozone–human chemistry.

. Summary of experimental conditions and associated NCA emission rates S2

Section S1. Climate chamber
The chamber was ventilated by filtered compressed air to ensure low background levels of particles (<100 #/cm 3 below 10 nm) and ozone (<1 ppb).The air was supplied through two side inlets and exhausted via two side outlets.The chamber air temperature was controlled at 24 °C by water bath controllers, while the relative humidity was maintained at 40 ± 5% by adjusting airflow passing through the gas wash bottle (Fig. S2).Ozone was generated by a Jelight 600 UV generator (Jelight Co. Inc., USA).In experiments with NH 3 , we injected NH 3 from a gas cylinder (10 ppm, purity >99.9%,Cabagas Inc., CH).A stainless-steel stand inside the chamber held the glass plate for conducting ozone-squalene and -fatty acid reactions.Two desk fans facing the chamber walls ensured air mixing.All surfaces were thoroughly cleaned by methanol and then distilled water prior to each experiment.

Section S2. Principle of NCA measurements
NCA levels within the chamber were measured using a Nano Condensation Nucleus Counter (Airmodus A11 nCNC System, Airmodus, Finland), which incorporates a Particle Size Magnifier (PSM A10) and a Condensation Particle Counter (CPC A20).The PSM serves to enlarge small particles to a size detectable by the CPC, through a mixing-type mechanism. 1is mixing ratio can be swiftly adjusted, resulting in varied smallest particles that can be amplified by the PSM.Continuous scanning of the mixing ratio enables the measurement of aerosol size distribution within the 1-4 nm range.
The PSM operated in scanning mode with the saturator flow scanning consistently at a constant rate, enabling the detection of aerosols (named activation size) ranging from 1.2 to 4 nm.A complete scan comprised two 2-minute intervals: first, the saturator flow increased from 0.1 to 1.3 L/min (up-scan), followed by a return to 0.1 L/min (down-scan).Typically, aerosol particle concentrations for each size bin were averaged over the two periods, resulting in a time resolution of 4 minutes.In instances where NCA concentrations exhibited substantial variations, a 2-minute resolution was employed to better capture the NCA dynamics.

Compared to alternative particle detection technologies like the Scanning Mobility Particle
Sizer (SMPS), this method minimizes losses of the smallest aerosols as it does not involve prior size selection or particle charging.][5]

Section S3. Contribution to daily indoor NCA levels
The primary goal of this study was to explore which chemical process predominantly generates NCA in ozone-human chemistry.Instead of attempting to fully replicate realistic indoor environment conditions, this study serves as a crucial stepping stone, providing insights that will inform and guide future research conducted in real indoor environments.][10] Furthermore, NCA deposition rates (7.0-7.9 h -1 ) inside the chamber closely approximated measurements or models observed in indoor environments. 113][14][15] Hence, based on the emission rates obtained from this study, we can expect a similar NCA level (10 3 -10 4 particles/cm 3 ) in typical indoor environments generated via ozone-human chemistry.Such a level is at the same order of magnitude of the limited measurements of indoor NCA concentrations without other prominent and temporary sources, 16 such as candle burning, cooking, and mopping. 3It is worth mentioning that the results in this study were obtained with the air mixing fans activated.Recent study has shown that the operation of mixing fans substantially decreases NCA formation and subsequent growth. 17erefore, in daily indoor environments without fan operations, we can expect a larger contribution of ozone-human chemistry to indoor NCA and ultrafine particle levels.On the other hand, our previous study demonstrated that increasing clothing coverage led to decreased NCA emissions. 5Hence, while presence of occupants' clothing is expected to mitigate the NCA levels, we anticipate a significant net increase in NCA levels in the absence of mixing fans.The implication is that ozone-human chemistry considerably contributes to daily build-up of indoor NCA levels.However, given the sensitivity and complexity of the gas-to-particle conversion processes and NCA measurements, 18,19 as well as the potential effects on NCA formation caused by coagulation and/or condensation onto existing larger particles indoors and indoor air movement, future investigations into squalene-ozone reaction in various indoor environments, including off-body squalene ozonolysis, are warranted. 20 addition to indoor environments, ozone-human chemistry may also contribute to outdoor NCA.However, it should be noted that the emission rates obtained in this study cannot be directly applied to outdoors, as the outdoor environment significantly differs from indoors, particularly in air movement and surface-to-volume ratio, which are key factors influencing the dynamics and fate of NCA. 17 Fig. S1.Schematic figure of the climate chamber setup Fig. S2.RH levels in experiments Fig. S3.Visualization of experiment procedures Fig. S4.NCA levels in experiments Fig. S5.An example of zero check of the instrument TableS1.Summary of experimental conditions and associated NCA emission rates Future outdoor field measurements can address the potential contribution of ozone-human chemistry to the atmospheric NCA.

Fig. S1 .
Fig. S1.Schematic figure of the climate chamber setup.The chamber air temperature was controlled by the temperature of the chamber surfaces though which water was circulating and whose temperature was controlled by two water bath controllers.The relative humidity was controlled by adjusting airflow passing the gas wash bottle using mass flow controllers (MFC).

Fig. S2 .
Fig. S2.Time-series plots of RH level in each experimental run, controlled at 40±5%.The dashed lines represent results from replicate experiments.

Fig. S3 .
Fig. S3.Experiment procedures for (A) comparing ozone reaction with squalene and C16:1n6; (B) investigating the impact of ozone level; and (C) investigating the impact of NH 3 level, where ozone was injected throughout the 6 h reaction.

Fig. S4 .
Fig. S4.Time-series plots of NCA concentration in each experimental run.The dashed lines represent results from replicate experiments.

Fig. S5 .
Fig. S5.An example of zero-check of the nCNC instrument during the ozone-squalene reaction.This zero check was performed during one test in Experiments A for squalene ozonolysis.A zero-check filter was connected between the chamber exhaust and the nCNC inlet after 3-h reaction.As seen, during the zero-check, the instrument background showed <10 particles/cm 3 , whereas the measured NCA went back to the previous level after the filter being disconnected.It indicates the low background of the instrument itself, and proves that the measured clusters were predominantly in particle phase.

Table S1 . Summary of experimental conditions and associated nanocluster aerosol (NCA) emission rates
. SS: steady state, Rep.: replicate