THE ADAPTATION OF THE DICHROMATE DIGESTION METHOD FOR TOTAL MERCURY DETERMINATION BY COLD-VAPOR ATOMIC-ABSORPTION SPECTROMETRY TO THE ANALYSIS OF SOILS, SEDIMENTS AND SLUDGES

Previous research demonstrated the overall validity of a two-component wet digestion procedure based on the action of K2Cr2O7+H2SO4 (50%, v/v) at boiling temperature (180 degrees C) with vapors refluxing in an air condenser in the treatment of biological samples to be analyzed for total mercury by cold-vapor atomic absorption spectrometry (CVAAS). This study explored the ability of this procedure to cope with soils, sediments and sewage sludges. After demonstrating that cinnabar (HgS) is completely dissolved under the experimental conditions employed, a set of nine standard reference materials having a mercury content ranging from 0.0568 to 9.49 mu g g(-1) was analyzed. One difficulty was posed by chloride containing samples and was due to the formation of chlorine, a strong inhibitor of the spectrometric signal. Evidence is given that this effect is ascribable to the re-oxidation of elemental mercury by the chlorine present inside the reduction-aeration system. The elimination of chlorine by simply running the digestion without a condenser proved effective in some cases. However, when the samples had a high content of organic matter or carbonates the absence of the condenser caused mercury losses. A study employing mercuric nitrate and methylmercury chloride showed that the escape of mercury from the digestion tube not surmounted by the condenser is related to the rapid release of CO2 when organic matter and carbonates react with the digestion mixture at its boiling point; high temperature alone is not sufficient for mercury to be lost. Therefore, the digestion procedure was duly reconsidered and split in two parts: (a) an initial 60 min period with the condenser in place during which the destruction of all organic matter and carbonates takes place followed by (b) a 30 min span without condenser to allow chlorine and other potential interfering vapors and gases to leave the digest. The modified procedure was applied to the same reference materials as above and additionally to a biological standard of mussel tissue. The analytical results showed that total mercury could be determined in all samples with consistently good precision and accuracy (R.S.D.% range = 1.2 - 4.5; error %range = -3.8 to +8.7).