On the analysis of mercuric nitrate in flue gas by GC-MS

Recent research has demonstrated that in a simulated flue gas stream containing NO(2) and SO(2) elemental mercury is initially captured on a carbon or manganese oxide sorbent. After approximately an hour, however, mercury breaks through relatively rapidly, and the volatile form of mercury emitted is an oxidized species. The volatile mercury species emitted from a granular MnO(2) sorbent was trapped in an impinger containing cold acetonitrile. Subsequent evaporation of 95% of the acetonitrile in a Kuderna-Danish apparatus and gas chromatography (GC) of the concentrate resulted in a single mercury-containing GC peak at 5.5 min; the retention time and mass spectrum of this compound matched exactly those of a standard mercury(II) nitrate hydrate, Hg(NO(3))(2).H(2)O dissolved in acetonitrile. The volatile mercury component analyzed from injection of this standard solution was shown to be a form of methylmercury that is produced in the GC column by reaction of the highly reactive mercury nitrate with the methylsiloxane GC phase. Because the on-column derivatization reaction seems to be unique to mercury nitrate, the GC-MS (gas chromatography-mass spectroscopic) analysis provides strong evidence for identification of the trapped oxidized mercury species as mercury nitrate although, because the nitrate becomes detached from the mercury atom in the on-column reaction, the identity is not proven.