THE DETERMINATION OF CR, CU AND MN IN SEAWATER WITH TRANSVERSELY HEATED GRAPHITE-FURNACE ATOMIC-ABSORPTION SPECTROMETRY

The atomization signals of Cr, Cu and Mn in the presence of seawater s alts have been investigated using a new transverse heated atomic absor ption spectrometer and longitudinal Zeeman-effect background correctio n system. In unmodified seawater, the vaporization of the chloride mat rix at a relatively low background absorption signal level may induce errors of correction which cannot be neglected at low metal concentrat ion levels. Even without charring, seawater salts do not interfere wit h the recovery of Cr; however, due to temporal overlap of analytes and chloride vapours, seawater salts interfere severely with the recovery of Cu and Mn. For diluted seawater solutions (S < 3.5 parts per thous and) a charring step at about 500 degrees C almost suppresses the chlo ride chemical interference effect in the case of Cu and Mn, Mn being s ufficiently stabilized through MgCl2 hydrolysis to permit the eliminat ion of NaCl at 1000 degrees C before atomization. However, for more sa line solutions, due to mass effect the hydrolysis of MgCl2 is not comp lete; the recovery of Mn is lowered to about 80% and the recovery of C u is less than 20%, In the presence of nitric acid (more than 0.15 M) or oxalic acid (more than 0.1 M) which hydrolyses MgCl2 to MgO and sta bilizes Cr, Cu and Mn in their oxide forms, the recovery for Cr, Cu an d Mn is good and a charring step at about 1150 degrees C permits the e limination of the major part of the matrix (Na species) before atomiza tion at 2150 degrees C; these conditions are almost optimal for the si multaneous determination of Cr, Cu and Mn. The detection limits obtain ed for 20 mu l seawater 0.15 M in HNO3 or 0.1 M in oxalic acid are res pectively 0.042 mu g 1(-1) for Cr, 0.075 mu g 1(-1) for Cu and 0.026 m u g 1(-1) for Mn (3 sigma).