Aqueous chemistry and its role in secondary organic aerosol (SOA) formation

Glyoxal has two carbons, two hydrogens, and two oxygens. It sounds almost trivial — the simplest aldehyde you could build with both carbonyl groups intact. But this molecule is quietly dissolving into cloud droplets and rainwater across the planet, and what it does once it gets there turns out to rewrite how we think about aerosol formation. Follow glyoxal into the water, and you follow it into a completely different theory of where fine particle pollution comes from. For decades, atmospheric scientists explained secondary organic aerosol, or SOA, through partitioning theory. The idea was straightforward: sunlight drives photochemical reactions that convert volatile organic compounds into lower-volatility products, and those products condense onto existing particles. Volatility, temperature, and the amount of organic mass in particles are the controls. It's a clean framework, and it works reasonably well in the lab. Lim and colleagues, however, point out that it fails to fully explain what we actually observe in the atmosphere. Chemical transport models underpredict the magnitude and distribution of measured organic aerosol. Aged ambient aerosol shows higher oxygen-to-carbon ratios than smog-chamber SOA ever does. In some locations, SOA tracers correlate more strongly with liquid water content than with organic mass. Each of those mismatches is a loose thread, and when you pull them all, they point to the same missing piece: the chemistry happening inside atmospheric water. The observational fingerprints are worth dwelling on. Water-soluble organic carbon accumulates in aerosol at levels that gas-phase partitioning alone can't explain. Oxalic acid — a tiny two-carbon diacid — appears persistently in ambient aerosol worldwide, and it's produced most efficiently not in the gas phase but in cloud water via aqueous chemistry. Hennigan and colleagues showed that SOA surrogates in some environments correlate more tightly with aerosol liquid water content than with anything else. Volkamer and colleagues measured SOA yields from glyoxal that tracked liquid water content almost linearly — approximately two percent yield at three micrograms per cubic meter of liquid water and seven percent at five micrograms per cubic meter. The clues in the ambient data all point the same direction: water in clouds, fogs, and wet aerosols is not a passive medium. It's a reactor. Glyoxal is the ideal molecule to probe that reactor. It's produced in the gas phase from photochemistry of both biogenic and anthropogenic precursors, at an estimated global source of forty-five teragrams per year. It's extraordinarily water-soluble — its effective Henry's law constant, the measure of how readily a gas dissolves into water, reaches roughly four hundred thousand moles per liter per atmosphere in pure water, and climbs orders of magnitude higher in salt solutions. In sodium sulfate, Ip and colleagues measured values as high as twenty-four million moles per liter per atmosphere. That enhanced uptake beyond what simple dissolution predicts is itself a clue that something chemical is happening once glyoxal enters the aqueous phase. The non-radical pathways come first, and they're real but limited. When glyoxal dissolves into acidic aerosol water, its carbonyl groups can be protonated, and a hydroxyl group on a neighboring hydrated glyoxal molecule attacks that carbonyl — a process called hemiacetal formation, the product of an alcohol attacking an aldehyde. Repeated steps build oligomers, particularly during droplet evaporation when concentrations rise. Acid catalysis also enables Fischer esterification and, in sulfate-rich aerosol, organosulfate formation. Liggio and colleagues measured reactive uptake coefficients for glyoxal on acidic seed particles in the range of two point two to seven point three times ten to the negative third, and notably, those values didn't change much across a pH range of zero point nine to five point three, suggesting that acidity is not the primary driver. Liquid water content matters more. The deeper problem with these non-radical routes is reversibility: several studies show that glyoxal oligomers can de-oligomerize under dilution or clean-air conditions. They form, but they can also fall apart. That limits how much persistent, refractory SOA they actually deliver. Radical chemistry is a different story entirely. When hydroxyl radicals, or OH, the atmosphere's primary oxidant, react with dissolved glyoxal, the products are largely irreversible, and the chemistry is faster. The mechanism Lim and colleagues describe begins with OH abstracting a hydrogen atom from hydrated glyoxal, generating an organic radical. That radical then faces a fork: it can react with dissolved molecular oxygen to form a peroxy radical that decomposes to glyoxylic acid, or two organic radicals can collide and couple directly, building a larger molecule. The competition between these routes — radical plus oxygen versus radical plus radical — depends entirely on concentration. Tan and colleagues ran the experiments that illuminate this branching. They dissolved glyoxal at three concentrations — thirty, three hundred, and three thousand micromolar — in water with hydrogen peroxide as an OH source, irradiated the samples, and tracked the products. At thirty micromolar, representative of cloud water, the aqueous chemistry model reproduced the measurements well. Oxalic acid was the dominant product, time series matched, and total organic carbon tracked as expected. That's the cloud story: dilute glyoxal, abundant oxygen, and OH chemistry cleanly converts a two-carbon precursor into the two-carbon diacid oxalic acid. At three thousand micromolar, the model broke down. Measured products increasingly deviated from predictions. Larger organic acids appeared — tartaric acid, malonic acid, compounds with six or more carbons — and something the original model hadn't accounted for was clearly happening. That something was radical-radical coupling: at high enough concentrations, organic radicals encounter each other before oxygen can intercept them, and they combine to build oligomers. The evidence from ultra-high-resolution mass spectrometry is striking. Lim and colleagues used Fourier-transform ion cyclotron resonance mass spectrometry, or FTICR-MS, a technique that identifies molecular formulas precisely by measuring mass to four to six decimal places, to characterize the products at high glyoxal concentration. Twenty-five of the thirty most intense peaks in the mass range of one hundred to five hundred atomic mass units could be explained by radical-radical reactions. A series of peaks incremented by exactly seventy-four mass units traced the fingerprint of stepwise oligomer growth — each step adding one dehydrated glyoxal radical unit. Critically, these high-mass products were absent in control experiments without OH, which rules out acid-catalyzed condensation or instrument artifacts as explanations. A new kinetic model that explicitly includes radical-radical coupling steps improves dramatically on earlier work. Both the old and new models handle the thirty-micromolar case fine — oxalic acid dominates, predictions match. But for three thousand micromolar, the earlier model overestimates the oxalic acid peak and predicts its decay too early, because it has no pathway for carbon to flow into larger products. Adding radical-radical reactions, using rate constants of one point three times ten to the ninth moles per liter per second for radical coupling versus one point zero times ten to the sixth for radical-oxygen reactions, brings the model into close agreement with measurements. Including a slow acid-catalyzed dehydration step also allows the model to reproduce malonic acid formation. The product distributions that emerge from this modeling are consequential. At cloud-relevant glyoxal concentrations, ranging from ten to the negative five to ten to the negative four molar, oxalic acid dominates, and the average oxygen-to-carbon ratio of products is around two. At aerosol-relevant concentrations of one to ten molar, oligomer yields exceed eighty percent, oxalic acid formation is negligible, and the modeled average oxygen-to-carbon ratio falls to approximately one. FTICR-MS data on the oligomers yield an average oxygen-to-carbon ratio of about one point two, consistent with that prediction. These are the humic-like substances — known in the field as HULIS — that show up in ambient aerosol measurements and whose origin has long been debated. Aqueous radical-radical chemistry in wet aerosols produces them. That concentration dependence has a direct atmospheric interpretation. Clouds run dilute chemistry, producing small organic acids like oxalic acid that stay in the particle phase after droplets evaporate. Wet aerosols run concentrated chemistry, producing large oligomeric HULIS with oxygen-to-carbon ratios near one that are effectively irreversible and highly oxygenated. Both products explain ambient observations that gas-phase models miss: the high oxygen-to-carbon ratios of aged aerosol, the correlation of SOA with liquid water, and the accumulation of water-soluble organic carbon beyond what partitioning theory predicts. For atmospheric models, the implication is direct. Current chemical transport models undercount SOA because they're missing an entire formation pathway. Aqueous-phase SOA forms from different precursors, through different chemistry, in different atmospheric regimes than gas-phase partitioning SOA. Including it increases predicted SOA mass and improves agreement with ambient measurements. Beyond glyoxal, Lim and colleagues note that methylglyoxal, other small carbonyls, alcohols, organic acids, and peroxides will undergo similar processing — the aqueous reactor is not specific to one molecule. Water in the atmosphere, it turns out, is where some of the most important organic chemistry happens. Glyoxal is just the simplest proof. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.