Relative standard deviations for a group of 14 elements included in this study have been reported by the author to range between 5 and 20% (4). Precision and accuracy attainable by this technique are dependent to a large extent on the selection of appropriate internal standards. The results of moving plate studies provide a criterion for the choice of internal standards.

No abstract

No abstract

Mansfield et al. (7) have recently reviewed the historical development of electrodeless discharge lamps and have described the preparation and powering of these sources for atomic fluorescence flame spectrometry. By means of a statistical evaluation of data, they showed that source intensities were related directly to the lamp diameter, inert gas fill pressure, and form of element (metal or metal iodide), but were independent of the weight of the metal or metal iodide introduced into the lamp. They also found that the "A" type antenna with an open quartz jacket surrounding each lamp produced the largest atomic fluorescence signals for most elements as compared to the "C" type antenna, the Vi-wave (Evenson) cavity, and the 3/Vwave (Broida) cavity. In this study, increases of intensity and stability of lamps of some elements result if the lamps containing metal iodides also contain excess metal. In addition, many lamps produce more intense atomic fluorescence when operated in an evacuated jacket on the "A" antenna or when operated in a rectangular cavity.Dagnall, Thompson, and West (2-4) as well as Rains, Snelleman, and Menis (5) have also obtained more intense electrodeless discharge lamps for atomic absorption and fluorescence flame spectrometry, if excess metal is present when using metal iodide lamps. However, both groups have used the V4-wave (Evenson) type cavity for all of their measurements. EXPERIMENTALPreparation of Electrodeless Discharge Lamps. All electrodeless discharge lamps were prepared by using the vacuum system described by Mansfield et al. (1,6). The important characteristics of lamps for each element are given in Table I. The amount of material in the lamps was unimportant as found by Mansfield et al. (1) as long as the amount was small-i.e., too much material results in reduced transmittancy of the lamp walls and in diminished intensities. Only electrodeless discharge lamps which have been evaluated for atomic fluorescence flame spectrometry are listed in Table I. The characteristics of lamps of other elements will be published in future articles concerning the atomic fluorescence of selected groups of elements.

No abstract