Determination of macro- and micronutrients in plant leaves using high-resolution continuum source flame AAS

Complete text of publication follows. The determination of B, Ca, Cu, Fe, K, Mg, Mn, Mo, P, S and Zn is frequently required in foliar diagnosis. The determination of all these analytes by line source flame AAS is a difficult analytical task due to the following aspects: I. sample preparation is dependent on the analyte concentrations and respective linear calibration range, II. sulfur resonance lines lie in the vacuum-UV region demanding the use of other analytical technique, and III. sensitivities for B and P are poor. The high-resolution continuum source AAS is a promising technique to determine these elements because main and secondary lines are available for measurements, sensitivity and linear calibration range may be improved by measuring both line peak and wings, absorption measurements for diatomic molecules is possible (B. Welz et al., J. Braz. Chem. Soc., 14 (2003.) 220-229). Samples were dry-ashed and ashes were divided into 3 fractions to determine sequentially the following analyte groups: I. Cu, Fe, Mn, Zn, Ca, Mg, and K (fraction 1, diluted sample), II. B, P, and S (fraction 2, undiluted sample), and III. Mo (fraction 3, extraction into methyl isobutyl ketone). Measurements were performed at main lines for B (249.773 nm), Cu (324.754 nm), Fe (248.327 nm), Mn 279.482 nm), Mo (313.259 nm), P (213.618 nm), and Zn (213.857 nm), at secondary lines for Ca (239.856 nm) and K(404.414 nm), at the line wings of the secondary line for Mg (202.588 nm), and at the CS molecular band for S (258.055 nm). Using a sample flow-rate of 5 mL min-1, calibration curves for all elements were attained with regression coefficients better than 0.995. Accuracy and precision were evaluated using three CRM's (Peach, Spinach and Tomato leaves) and the results were in agreement with certified values at a 95% confidence level (paired t-test). The proposed method was applied to orange and sugarcane leaf samples. Recoveries were in the 82-112% interval range. Limits of detection were 0.3 mg L-1 B, 1.9 mg L-1 Ca, 7.3 μg L-1 Cu, 20.4 μg L-1 Fe, 1.5 mg L-1 K, 0.6 mg L-1 Mg, 4.2 μg L-1 Mn, 15.7 μg L-1 Mo, 29.2 mg L-1 P, 22.4 mg L-1 S, and 8.2 μg L-1 Zn. The authors are grateful to the financial support provided by the Fundacao de Amparo a Pesquisa do Estado de Sao Paulo (FAPESP, Process 06/56436-08).