Determination of boron in blood, urine and bone by electrothermal atomic absorption spectrometry using zirconium and citric acid as modifiers

A comparative study of various potential chemical modifiers (Au, Ba, Be, Ca, Cr, Ir, La, Lu, Mg, Ni, Pd, Pt, Rh, Ru, Sr, V, W, and Zr), and different 'coating' treatments (Zr, W, and W + Rh) of the pyrolytic graphite platform of a longitudinally heated graphite tube atomizer for thermal stabilization and determination of boron was undertaken. The use of Au, Ba, Be, Cr, It, Pt, Rh, Ru, Sr and V as modifiers, and of W + Ph coating produced erratic, and noisy signals, while the addition of La, Ni and Pd as modifiers, and the W coating had positive effects, but with too high background absorption signals, rendering their use unsuitable for boron determination even in aqueous solutions. The atomic absorption signal for boron was increased and stabilized when the platform was coated with Zr, and by the addition of Ca, Mg, Lu, W or Zr as modifiers. Only the addition of 10 mug of Zr as a modifier onto Zr-treated platforms allowed the use of a higher pyrolysis temperature without analyte losses. The memory effect was minimized by incorporating a cleaning step with 10 mul of 50 g l(-1) NH4F HF after every three boron measurements. The addition of 10 mul of 15 g l(-1) citric acid together with Zr onto Zr-treated platforms significantly improved the characteristic mass to m(o) = 282 pg, which is adequate for biological samples such as urine and bone, although the sensitivity was still inadequate for the determination of boron in blood of subjects without supplementary diet. Under optimized conditions, the detection limit (3 sigma) was 60 mug l(-1). The amount of boron found in whole blood, urine and femur head samples from patients with osteoporosis was in agreement with values previously reported in the literature. (C) 2001 Elsevier Science B.V. All rights reserved.