Data Files Accompanying: “The Impact of pH and NaCl and CaCl2 Salts on the Speciation and Photochemistry of Pyruvic Acid in the Aqueous Phase”
Authors: Luo, Man. | Shemesh, Dorit. | Sullivan, Michael, N. | Alves, Michael, R. | Song, Meishi. | Gerber, R, Benny. | Grassian, Vicki, H.
Journal: JPCA
Contact: Grassian, Vicki, H., vhgrassian@ucsd.edu, Department of Chemistry & Biochemistry,Scripps Institution of Oceanography, Department of Nanoengineering, University of California, San Diego, La Jolla, CA 92093. 
Cite as: Luo, Man. | Shemesh, Dorit. | Sullivan, Michael, N. | Alves, Michael, R. | Song, Meishi. | Gerber, R, Benny. | Grassian, Vicki, H. (2020): Data from: “The Impact of pH and NaCl and CaCl2 Salts on the Speciation and Photochemistry of Pyruvic Acid in the Aqueous Phase”. In Center for Aerosol Impacts on Chemistry of the Environment (CAICE) Collection. UC San Diego Library Digital Collections. https://doi.org/10.6075/J04M932F
File Format: All data is held in comma delimited text files.
Folder Organization: There are experimental data folder and simulation data folder. 
In the experimental data folder, there are two folders, labeled as "main text" and "SI", for data from the main text and data in supporting information, respectively. Additionally, data from each figure have their own folder.   
In the computational data folder, there is one folder for Figure 4B.
Method: 
Experimental Methods and Materials.
Pyruvic acid (98%) was purchased from Sigma-Aldrich and distilled under reduced pressure to remove impurities. CaCl2 and NaCl salts were purchased from Fisher Scientific and baked at 200 °C overnight to remove organic contaminants. Aqueous solutions were prepared using Milli-Q water with an electric resistance of 18.2 MΩ. Pyruvic acid was prepared as 10 mM solutions in Milli-Q water, or as salt solutions of varying salt concentrations. Solution pH was determined using an Oakton 700 pH meter and was adjusted by either hydrochloric acid (1 N stock solution, Fisher Chemical) or sodium hydroxide (1 N stock solution, Fisher Chemical). 
Photolysis experiments were performed in an anaerobic environment by purging with nitrogen to displace dissolved oxygen and subsequently sealed with parafilm. We chose to perform our experiments under anaerobic conditions to eliminate the effect of dissolved oxygen on the photolysis of aqueous pyruvic acid, as it has been previously shown that the photolysis of aqueous pyruvic acid is highly sensitive to the concentration of dissolved oxygen and will deplete oxygen quickly if a certain sparger is not equipped.13,20,21 A nitrogen enriched environment has the benefit of representing the conditions similar to that of some coarse-mode aerosols (diameter > 14 μm) as this kind of aerosols may be oxygen-depleted due to the rapid pyruvic acid photolysis.13,20,21 All the photolysis experiments were performed at an unadjusted pH (2.3 ± 0.1). For all the experiments in pure water or NaCl/CaCl2 solutions, 70 ml of solution were made and irradiated with a 575 Watt metal halide arc lamp (Optical Energy Technologies INC.) for 5 hours, with 3.5 ml of sample extracted via syringe, wrapped in foil and refrigerated (at 4°C) at a series of time points of irradiation for NMR tests. The time between two samples ranged from 5 minutes to an hour. Before each irradiation, 3.5 ml was saved as the control. The solutions before and 5 hours after irradiation were stored in a freezer (-20°C) for the following mass spectrometry tests. The samples were allowed to reach room temperature before testing. The NMR tests on the photolysis samples were performed 1 to 8 hours after the samples were taken within the same day. The mass spectrometry tests were performed several days after the sample were taken. The output spectrum of the lamp used for the photolysis can be found in Figure S1. The lamp was not filtered as if has been found that the photochemical products of aqueous pyruvic acid observed using an unfiltered lamp with UV extended to about 220 nm are not significantly different from the photochemical products observed using a filtered lamp (λ > 300 nm).42 The rate of pyruvic acid photolysis was found to increase with the use of unfiltered lamp source due to the increased flux.42 The metal halide lamp also generates several high intensity peaks in the output spectrum (Figure S1), which could also increase the rate of photolysis caused by the increase in flux.
1H NMR experiments for pKa studies were performed using a 300 Bruker AVA NMR spectrometer with wet suppression of H2O. NMR samples were prepared with 10% (v/v) D2O (99.9%, Cambridge Isotope Laboratories, Inc.) for NMR field frequency lock. No correction was made for the deuterium isotope effect when measuring pH. Photo decay studies of pyruvic acid were performed using either a 300 Bruker AVA NMR spectrometer or a 500 JEOL ECA NMR spectrometer with wet suppression of H2O. The wet suppression methods have been used in many related studies under similar conditions.18,20,62 The quantitative 1H NMR measurements for pyruvic acid photolysis analysis were performed with a constant receiver gain. The 1H NMR spectra for pyruvic acid in water solution and 0.17 M CaCl2 solutions before photolysis, 1 hour after photolysis and 5 hours after photolysis are shown in Figure S2.
UV-Vis spectral information of each sample was obtained using a PerkinElmer Lambda 35 UV-Vis spectrometer in the wavelength range from 200 to 700 nm. As only the ketone form absorbs in the ultraviolet and visible regions in these experiments, the intensity of the peak arising from the n  π* transition is proportional to the concentration of the ketone form in the sample despite the slightly difference in peak maxima for the protonated and deprotonated ketone. The calibration curves for the UV-Vis intensity versus the ketone fraction in the sample can be seen in Figure S3. 
The samples before and 5 hours after irradiation were filtered with Bond Elut PPL solid phase extraction cartridges at pH 2 and then eluted with methanol. The eluted samples were then analyzed with negative-ion electrospray ionization mass spectrometry (ESI-MS) for detection of photolysis products. The ESI-MS spectra for pyruvic acid in water after irradiation without PPL extraction can be seen in Figure S4 (a) in order to compare with previous studies of pyruvic acid photolysis without PPL extraction, as the PPL may have a preference on adsorbing different types of organic molecules and therefore affect the ratio of the products analyzed with and without PPL extraction. There is no obvious difference found between Figure S4 (a) and previous studies.32–34 The ESI-MS spectra for pyruvic acid in water and other NaCl/CaCl2 solutions after irradiation with PPL extraction are shown in Figure S4 (b), (c), (d), (e).
Simulation Methods
Model system
The effect of the addition of Ca2+ ion on the absorption spectrum was modelled computationally. Pyruvic acid in aqueous solutions was simulated by embedding pyruvic acid in a small cluster of water molecules. The size of the water cluster needed to fulfill the following conditions: (1) The experimental data (see Results section) shows that at a solution pH of 4, the system is mainly composed of the anion ketone. (2) Since pyruvic acid underwent deprotonation, the additional H+ should also be present in the solution, i.e. the presence of H3O+ is necessary. The smallest cluster with includes the anionic ketone form and H3O+ has 6 additional water molecules. Below that number of water molecules, the anionic pyruvic acid undergoes spontaneous protonation to form pyruvic acid. Thus, a smaller water cluster doesn’t accurately represent pyruvic acid in aqueous solution. The effect of addition of Ca2+ ions was modeled by the same cluster and the addition of Ca2+ at the COO- group.
Vertical excitation energies
The model system with and without Ca2+ ion were optimized using the MP2 potential energy surface with the resolution-of-the-identity (RI) approximation63 for the evaluation of the electron-repulsion integrals implemented in Turbomole.64 The basis set SV(P)65 was successfully employed. Ten vertical excitation energies and their respective intensities were calculated using the ADC(2) method.66 ADC(2) has been used in previous studies, and has been proven to be accurate for spectrum calculations and for photochemical reaction dynamics of organic systems.67–71 A Lorentzian with a full width at half maximum of 5 nm was added in order to add finite temperature effects into the stick spectrum. The resulting orbitals can be seen in Figure S5. The excitation states, energies, orbital descriptions of the transitions as well as oscillator strengths and dipole moments can be seen from Table S1.