Sulfate aerosols are ubiquitious in Earth’s atmosphere, driving air quality and climate impacts. The liquid water of sulfate aerosols can reach high ionic strength with low hydration. This special environment was recently shown to promote unexpected photochemical reactivity of the sulfate ions (HSO₄^–, SO₄^2–) under tropospheric radiation, potentially through the formation of sulfur-centered free radicals. Yet, knowledge of the SO₄^2– photochemistry is still insufficiently constrained, which hinders fundamental understanding and incorporation into computational models. This work provides direct spectroscopic evidence that sulfate solutions absorb surface-level sunlight and eject electrons, forming reactive sulfate radicals (SO₄^•–) in a concentration-dependent manner on a transient (< microsecond) timescale. Quantification of radical production rates at long timescales (hours) was performed via radical trapping coupled with liquid chromatography high-resolution mass spectrometry analyses. Once formed, SO₄^•– accelerate the oxidative processing of organic aerosol constituents, reshaping aerosol composition, size, and properties. We show that compounds emitted from biomass fires photosensitize the SO₄^•– production from sulfate up to a factor of 5, with rates comparable to hydroxyl radical production. This photosensitized chemistry alters the aerosol direct and indirect climate effects through formation of light-absorbing brown carbon (BrC) and cloud-activating organosulfates, respectively. In particular, the formation of secondary BrC transforms the negative radiative forcing of sulfate aerosols to positive.  These findings have important implications for aerosol oxidant budgets, sulfur cycle reactions, and model representations of aerosol-chemistry-climate interactions.

Materials

Ammonium sulfate (≥99%, molecular biology grade), 2-methyl-3-buten-2-ol (≥98%), 3,4-dimethoxybenzaldehyde (≥99%), benzaldehyde (≥99%), 4-nitrocatechol, (≥97%), p-benzoquinone (≥99%), sodium sulfate (≥99%), were obtained from Sigma Aldrich. Reagent grade sodium dihydrogen phosphate (≥99%), and sodium monohydrogen phosphate (≥99%) were obtained from Macklin. All purchased chemicals were used without further purification. All solutions were made using Optima LC-MS (Liquid Chromatography-Mass Spectrometry) grade water from Fisher-Scientific.

Bulk Aqueous Experiments Photochemical reactions of bulk solutions were performed in a custom photochemical enclosure made of UV-reflective aluminum (Anomet Products Inc., Shrewsbury, MA) sheets and equipped with a UVB broadband fluorescent light (SolArc Systems Inc., Minesing, ON, Canada) with peak wavelength emission at 310 nm. A 4 mL solution of the organic compounds of interest at desired organic concentrations was prepared together with 3.7 M AS in LC-MS water. The aqueous solution was transferred to a 3.5 mL capped quartz cuvette with stopper (Thor Labs Inc., Newton, NJ) and placed in the photochemistry chamber to irradiate for twelve hours. Aliquots of 100 μL volume were taken every 2 hours for the chemical analysis procedure. Experiments were repeated outside on the rooftop of a four-story building in Davis, California, U.S in June July and August. Solar radiation data for ambient light exposure was obtained from the KCADAV116 weather station (38.54 °N, 121.73 °W) for each day.

 Analytical Measurements

HPLC coupled to High Resolution Mass Spectrometry (HPLC-HRMS) was used to quantify 3-methyl-2-buten-1-ol-organosulfates (MBO-OS) in bulk aqueous photochemical experiments. Analyses were performed on an Agilent 1100 HPLC coupled to a dual-cell linear ion trap-Orbitrap (Velos Pro-Orbitrap) mass spectrometer (ThermoCorp., Waltham, MA) at a mass resolving power of 30,000 m/Δm at m/z 400. Analysis of organosulfate products were preformed isocratically performed with a Luna Omega Polar C18 column (2.1 × 100 mm, 3 μm, 100 Å) at flow rate of 0.27 mL/min, column temperature of 30 ºC, and eluent mixture 40:60 liquid chromatography grade acetonitrile (MeCN, Fisher Scientific) and water with 0.1% formic acid (Fisher Scientific). For solutions with ammonium sulfate, 100 μL aliquot was mixed with 700 μL MeCN, filtered with 1500 μL MeCN, and transferred to a vial containing 3 mL LC-MS grade MeOH for rotary evaporation. The dried product was reconstituted in 100 μL MeCN with 1uM propyl sulfate (used as a surrogate standard). ⁸⁸ Controls were processed identically.

The kinetic decays of the triplet carbon precursors during experiments were measured using 100 µL aliquots extracted in 800 µL toluene or MeCN and analyzed with gas chromatography mass spectrometry (GC-MS) on an Agilent 6890 GC and 5973 MS detector using a HP-5MS column (30 m × 250 µm × 0.25 µm) and the following temperature program: 40 ºC (1 min hold), 10 ºC/min until 200 ºC (3 min hold). The analytes were identified by their electron ionization fragmentation at 70 eV, and the spectra were compared to those from the GC-MS library from the National Institute of Standards and Technology.

UV-vis spectroscopy was used to characterize the optical light absorption of reagents and brown carbon products, and for qualitative analyses of some reactions (in addition to HPLC-HRMS). Spectra in quartz cuvettes were obtained on a Shimadzu UV-1800 UV Vis spectrometer. Aliquots were collected every 4 hours for 12 hours, diluted in MeCN, to optimize brown carbon extraction.

Transient absorption spectroscopy utilized a custom-built Laser Flash Photolysis platform, which has been described previously.^{6, 89-90} Briefly, the third harmonic (355 nm, fwhm = 14 ns, pulse energy = 18 mJ) oscillation of a Nd:YAG Q-switched laser was used for the excitation of the working solutions. The flow rate of working solutions through the quartz cell (dimension of 10 × 10 × 10 mm) were controlled at 24 mL/min by a peristaltic pump (BT100–1L, Longer Pump). Transient signals were monitored by absorbing xenon lamp (Hamamatsu L2174) light and passed through a monochromator, with photomultiplier output recorded by an oscilloscope (Tektronix MD03052). Solutions of DMB and p-BQ were kept at pH 5.0 with a phosphate buffer solution. All solutions were deoxygenated with pure nitrogen gas for 15 minutes. A total of 512 laser shots were averaged, and all experiments were conducted in triplicate to ensure reproducibility.