|Abstract or Summary
- A multielement spectrometer has been developed for the simultaneous determination of four elements by atomic absorption (AA) or atomic fluorescence (AF). The spectrometer is based on a multiple exit slit monochromator and uses a mirrored funnel to direct light from the various exit slits to a single photomultiplier. There are no moving parts during operation and a single element hollow cathode (HC) lamp is used for each element. The HC beams are combined for AA by beam splitters and directed along the optical axis through the atomizer into the monochromator entrance slit. HC lamp pulsing and data acquisition are implemented by a time multiplex approach. The HC lamp pulsing is controlled by simple digital circuitry which sequentially pulses each HC lamp and digitally integrates HC and background signals for each element at a 100 Hz rate for up to 256 cycles. The system presently allows the determination of any four of 10 possible elements for which slits have been made. With the present configuration a 200 nm spectral window can be utilized allowing the determination of any 4 of 27 elements in an air /C₂H₂ flame or of 24 in a N₂0/C₂H₂ flame if the slits are manufactured. Some spectral overlap problems do exist for AA; however, they can be minimized with the use of appropriate filters. Because beam splitters and sometimes filters are used, the HC intensities measured are generally lower than in single element work. This causes signal shot noise and background emission noise to be more important but usually does not critically affect performance. For flame AA under favorable multielement conditions the following detection limits (in ppb) were obtained: Au, 50; Cd, 8; Co, 300; Cu, 5; Fe, 60; Mg, 2; Mn, 5; Ni, 50; Pb, 50 and Zn, 20. These detection limits are all within a factor of 2 of those obtained on a good single element AA with a similar spectral bandpass and integration time. Under the least favorable multielement conditions only for Cu, Co and Zn did the detection limits differ by more than a factor of 2.5 from those obtained on the single element AA. The relative measurement precision, σ[subscript A} /A, at moderate absorbances was typically 1% or better and equivalent to that obtainable on a good single element AA. An air /C₂H₂ flame operated with an oxidizing stoichiometry was found to be an excellent compromise atomizer for multielement flame analysis. The system was briefly tested with a Varian model 63 carbon rod atomizer and interfaced to a computer for data acquisition and handling. Cd, Mn and Pb were determined in tap water and a H₂ continuum HC was used to monitor for flame background absorption. Compromise multielement detection limits of 0. 2, 2 and 8 ppb were obtained for Cd, Mn and Pb, respectively. Signal shot noise was limiting and the detection limits of Pb and Mn could be easily improved by a change in the present software control of the analog to digital converter. Relative measurement precisions of about 7% were obtained with hand injection of 5 μl solutions with transmittances of about 50%. The system is not well suited to background correction. Multielement AF was explored briefly using wider slits and an unsheathed air /H₂ flame. The AF performance for certain elements was very dependent on flame stoichiometry and the major noise source was flame background noise. Compromise multielement detection limits for a tap water analysis were determined to be 90, 50, 100 and 200 ppb for Cd, Cu, Mg and Zn, respectively. Relative measurement precisions of better than 1% were obtained for all elements at higher concentrations.