Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BHNRC) » Beltsville Human Nutrition Research Center » Food Composition and Methods Development Laboratory » Research » Publications at this Location » Publication #119322


item Harnly, James - Jim
item Schuetz, Marcus
item Murphy, James
item Glmutdinov, Albert

Submitted to: Journal of Analytical Atomic Spectrometry
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/20/2001
Publication Date: 10/18/2001
Citation: Harnly, J.M., Schuetz, M., Murphy, J.R., Glmutdinov, A. 2001. Spatially resolved continuum source graphite furnace atomic absorption spectrometry. Journal of Analytical Atomic Spectrometry. 16:1241-1252

Interpretive Summary: Analytical errors can be introduced into trace metal determinations made by atomic absorption spectrometry (AAS) with graphite furnace atomization because the atom population is not homogeneously distributed throughout the furnace. Using a new spectrometer developed at USDA, intensities transmitted through the furnace were detected a 2 dimensional array detector. Computing an absorbance for each pixel provides spatially resolved absorbances. Conventional AAS, sums intensities and computes only a single spatially integrated absorbance. Non-homogeneous atom distribution can introduce errors into the computed absorbance (photometric errors). Analytical errors can result if the sample matrix produces photometric errors in the sample that are different than those in the standards. This study investigated photometric and analytical errors for the determination of Al, Cr, Cu, and Pb. These results will impact future AAS design and produce more accurate analytical results.

Technical Abstract: A continuum source atomic absorption spectrometer consisting of a high resolution echelle spectrometer and a high frame rate, 2 dimensional charge coupled device (2D-CCD) was used to measure intensities as a function of height in the graphite furnace. Spatially resolved absorbances, ASR were obtained by computing an absorbance for each pixel. Spatially integrated absorbances, ASI, were computed by summing the intensities for all the pixels and computing a single absorbance. Photometric errors, caused by the uneven distribution of atoms as a function of height in the furnace, were determined by comparing ASR and ASI. Analytical errors were determined by comparing sample recoveries for ASR and ASI. Results were determined for Al, Cr, Cu, and Pb using standard reference material 1572, citrus leaves. The largest photometric error (2% to 14% of the integrated absorbance) was introduced by the dichotomy of absorbances created by the presence of a platform. Less photometric error (less than 2% of the integrated absorbance) was introduced by the absorbance gradient (decreasing absorbance with increasing height in the furnace) arising from interaction of the analyte with the furnace wall. Both sources of photometric error affected the sample and the standards equally. For this study, no analytical inaccuracy could be attributed to photometric errors introduced by the analyte matrix.