Oxygen Isotope Analysis of Nanomole Phosphate Using PO3- Fragment in ESI-Orbitrap-MS.

Abstract

The oxygen isotope composition of phosphate is a useful tool for studying biogeochemical phosphorus cycling. However, the current Ag3PO4 method is not only tedious in PO43- extraction and purification but also requires a large-sized sample at the micromole level, thereby limiting its application. Here, we present an approach to measuring the oxygen isotope composition, δ18O, of dissolved phosphate at the nanomole level using electrospray ionization Orbitrap mass spectrometry (ESI-Orbitrap-MS). We compared the reproducibility of δ18O measurements using the H2PO4- ions (m/z = 97 and 99 for H2P16O4- and H2P18O16O3-, respectively) and using the PO3- fragment ions (m/z = 79 and 81 for P16O3- and P18O16O2-, respectively) generated by source fragmentation and by higher-energy collisional dissociation, respectively. The results demonstrate that phosphate δ18O can be more reliably measured by the PO3- ions than by the H2PO4- ions. PO3- generated by source fragmentation at 40 V achieved the highest reproducibility for δ18O based on precision tests. Furthermore, the mass spectrum for a 50:50 μM mixed solution of phosphate and sulfate revealed that PO3- ions resulting from source fragmentation at 40 V are the predominant species in the Orbitrap analyzer. Notably, P16O3- ions (m/z: 79) are not interfered with by 32S16O3- (m/z: 80) ions. This is in contrast to the case for 1H2P16O4- ions, which share the same m/z value with 1H32S16O4- ions and exhibit much lower signal intensity than HSO4- ions. Using the PO3- fragment method and six phosphate standards with a wide range of δ18O values, we obtained a calibration line with a slope of 0.94 (R2 = 0.98). The overall uncertainty for ESI-Orbitrap-MS phosphate δ18O measurement was 0.8‰ (n = 30; 1 SD). With much room for improvement, the PO3- fragment method presents a better approach to measuring the phosphate oxygen isotope composition, applicable to nanomole sample sizes in a liquid phase.