A simple, fast, and reliable method was developed for the determination Of cadmium in urine specimens by graphite furnace atomic absorption spectrometry (GFAAS). The method involved dilution (1:1) of the specimens with a 4.0% HNO3, direct injection of a 10 mul aliquot of the corresponding solution into a hot transversely-heated graphite atomizer (110degreesC), and application of a fast atomization program (42 s) in which the conventional dry-pyrolysis sequence was substituted by a high-temperature (300 degreesC) drying step. The effect of the injection temperature (A), injection rate (B), pyrolysis' ramp (C) and hold (D) times over the analyte's integrated absorbance, peak-shape and repeatability of the measurements was evaluated by means Of a 2 4-1 fractional factorial design. All those individual variables. as well as their first-order interactions (AB, AC- and AD-type interactions) were found to exert a statistically significant effect (P < 0.05). The lack of a chemical modifier other than the nitric acid itself benefited the overall methodology by allowing low-temperature atomization (1200degreesC), enhanced atomic and background signals separation, and reduced blank values. A detection limit (3s, n = 20) of 0.06 mug l(-1) Cd, corresponding to 0.12 mug l(-1) Cd in the urine specimen, and a characteristic mass of 1.78 pg/0.0044 s were obtained under the optimized conditions. The standard calibration technique (SCT) was used for quantitation. The successful determination of cadmium in Seronorm(TM) Trace Elements Urine Batch No. 115 (Nycomed Pharma AS) and in four urine specimens from Volunteer donors (recoveries: 91.3-103.4%) attested to the robustness of the proposed method. (C) 2003 Published by Elsevier B.V.