Objectives To optimize the detection of lead in urine by comparing different matrix modifiers to reduce the effect of matrix modifier on measurement results.

Methods Acidified urine samples and matrix modifier were mixed and using atomic absorption spectrophotometry (AAS) for the detection.

Results The linearity was good in the range of (0~200) μg/L. The r was more than 0.99 when measured by peak area or by peak height. The limit of detection (LOD) was 1.00 μg/L and the limit of quantitation (LOQ) was 3.00 μg/L. The recovery was in the range of 73.0%~86.0% and 73.0%~78.0% for peak height and peak area when ammonium dihydrogen phosphate was used as a matrix modifier; and the recovery was in the range of 97.0%~98.0% and 97.0%~103% for peak area and peak height when palladium chloride was used as matrix modifier. The RSD was in the range of 3.20%~14.0% and 4.60%~18.0% for peak height and peak area when ammonium dihydrogen phosphate was used as a matrix modifier; and the RSD was in the range of 1.00%~3.90% for peak area when palladium chloride was used as a matrix modifier.

Conclusions The quantitative result calculated by peak area was lower than that by peak height when ammonium dihydrogen phosphate was used as matrix modifier, but no limitation was observed on the quantitative result of palladium chloride matrix modifier treated samples. The service life of graphite tube was longer for palladium chloride matrix modifier and the recovery and precision of sample results treated by palladium chloride matrix modifier were better than those treated by ammonium dihydrogen phosphate matrix modifier. All technical indicators met the requirement of the Guidelines for the Establishment of Occupational Health Standards Part 5:Determination of Chemical Substances in Biological Materials (GBZ/T 210.5-2008). Therefore, palladium chloride was better than ammonium dihydrogen phosphate used as matrix modifier in the detection of lead in urine samples. When the concentration of lead in urine sample was out of the linear range, standard series and the sample treatment method should be the same, so as to avoid the deviation of results caused by matrix effect.