Bridging the gap between positive and null outcomes in controversial microdroplet oxidation chemistry experiments

Recent observations of unexpected oxidation reactions occurring in the presence of water microdroplets that otherwise do not occur in bulk water challenge our understanding of interfacial water chemistry. If such processes are occurring generally in water-based droplets suspended in air, there are remarkable implications for atmospheric chemistry, sustainability, and airborne disease transmission, among others. These observations, however, remain controversial. Some investigators have been unable to replicate previous reports and the underlying mechanism of oxidant generation remains highly speculative. Given the implications, it is imperative to establish whether this process is occurring generally or an experimental artifact to specific experiments. Here, we report on our efforts to replicate two surprising observations from the literature: Spray-based degradation of perfluorooctanoic acid (PFOA), a forever chemical, and oxidation of gaseous volatile organic carbon (VOC) compounds (e.g., methane). In the case of PFOA degradation, we readily validated spray-based PFOA breakdown, observing 3X more degradation than previously reported. In the case of VOC oxidation, however, we could not replicate previous reports without deviating from published experimental design and detection methods. Observing oxidation of methane and ethane necessitated the use of an ultra-trace detection scheme, ultimately detecting oxidation products at concentrations that are 5-6 orders of magnitude less than previously reported. Ionic and non-ionic solutes in the microdroplets promoted more oxidation, likely by maintaining droplet-phase water. Our observations suggest that microdroplet oxidation chemistry may be a general phenomenon in sprays, although with dramatic variability between experiments. High experimental variability with oxidation product concentrations below typical limits of detection could explain why this field has remained so controversial. To move this field forward to a predictive understanding, greater reproducibility in the literature is essential. We make multiple recommendations to minimize experimental variability and enhance reproducibility to ultimately move towards a mechanistic description and real-world impacts.