Bromide was added to the NADP analyte list in 2009 and monitoring of Br levels in NTN and AIRMoN samples continues to date (2018).  However, routine measurement of bromide by Ion Chromatography (IC) is challenging and to date nearly 80% of the precipitation samples examined exhibit bromide levels below IC quantification limits (<3-4 µg/L).  Prompted by some recent precipitation measurements of bromide that were hard to reconcile geographically or geochemically, the CAL at WSLH initiated a critical evaluation of the NADP bromide IC methods to define the veracity of the bromide data.

A selection of 32 recent NTN samples with IC-measured bromide levels ranging from below detection to over 150 µg/L, were re-evaluated for bromine concentrations using magnetic-sector ICPMS at both Br isotopes (79 and 81). The SF-ICPMS method provides greatly enhanced detection (LOD of 0.05 µg/L) AND specificity. The SF-ICPMS data indicated only very low levels (3.0 ± 1.9 µg/L) of Br, suggesting that the IC data were biased and subject to an interfering species.  The most likely candidates for interference under the conditions of the anion-IC run are organic acids and of the common organic acids, oxalate was the most likely species.  We have measured oxalate in precipitation and snow samples at levels exceeding the apparent bromide concentrations and the published literature documents levels of oxalate in precipitation in the range of 1-50 µg/L.   We confirmed the problem by running both oxalate standards and NTN samples spiked with oxalate and demonstrated that oxalate and Br have nearly identical retention times under the conditions of the NADP IC protocol used by the CAL at both WSLH and ISWS - thus oxalate would be miss-quantified as Br.  We then initiated a method development program to identify IC run conditions that would enable baseline separation of oxalate and Br, but not compromise the chromatography of the primary NADP analytes (Cl, nitrate, sulfate).  Though still a work in progress, we are now able to isolate bromide from oxalate and preliminary data indicate only low (essentially below LOD) levels of bromide with oxalate accounting for nearly all of the “apparent” Br measured with the old method.