Dataset: Concentration and δ15N of amino acids in size-fractionated particles from the eastern tropical North Pacific ODZ in December 2020

Two particle fractions were collected at five stations in the eastern tropical North Pacific region. At each station, seawater was filtered sequentially through a 53-micrometer (µm) Nitex mesh and then a pre-combusted 142-millimeter (mm) diameter glass fiber filter (GF-75: 0.3 µm retention size; or GF/F: 0.7 µm nominal retention size) with the McLane pump in situ. These filters were promptly frozen upon retrieval.

Each 53 µm Nitex mesh was immersed in approximately 100 milliliters (ml) of MilliQ water and subjected to a 5-minute sonication. The solution was filtered onto pre-combusted 47-mm glass fiber filters (GF/F, 0.7 µm pore size), which retain the > 53 µm particle fraction.

The GF-75 and GF/F filters (0.3 or 0.7 – 53 µm fractions) were subsampled, packed in tin capsules, and analyzed for concentration and δ15N of bulk materials by an elemental analyzer isotope-ratio mass spectrometer at the UC Davis Stable Isotope Facility. Bulk δ15N analysis was not performed for the > 53 µm fraction due to the limited amount of collected material.

The sample pre-treatment procedures for δ15N-amino acids were based on the method detailed in Zhang et al. (2021). The glass fiber filters underwent hydrolysis with 6N HCl for 22 hours at 110 degrees Celsius (°C). Hydrophobic impurities were eliminated from the hydrolysates through liquid-liquid extraction using n-hexane/dichloromethane (6:5, v/v), followed by evaporation to dryness in a vacuum evaporator (RapidVap, Labconco). The samples were subsequently redissolved in 0.05N HCl and further purified via cation-exchange resin, following procedures adapted from Takano et al. (2010) to remove metal ions and salts. The purified amino acids in the samples were completely dried under vacuum.

Amino acids in the samples were separated and collected as individual fractions using an ICS-5000+ Ion-exchange chromatography system with the instrumental method adapted from Zhang et al. (2021). For each sample, Phe, Glu, and their corresponding IC procedural blanks were collected from 1 to 3 replicate injections. The IC-collected fractions were sequentially converted to NO2- and N2O through a two-step process involving hypochlorite oxidation and azide reduction, as described in McIlvin and Altabet (2005); Zhang et al. (2007); Zhang and Altabet (2008); and Zhang et al. (2021). δ15N-N2O was determined with a GV IsoPrime IRMS.