Optimization of atomic fluorescence spectrometry for determination of arsenic in urine

Background The digestion temperature and the performance parameters of electric heating plate are not clearly defined in the health industry standards for the detection of arsenic in urine in China, which leads to great differences in the pretreatment methods of urinary arsenic in different laboratories.

Objective This study analyzes the causes for reporting lower detection results of arsenic in urine by wet digestion-atomic fluorescence spectrometry than by ICP-MS, and optimizes the standard method.

Methods According to WS/T 474-2015 Determination of arsenic in urine by hydride generation atomic fluorescence spectrometry, the standard method (150℃ digestion, 220℃ digestion) and the gradient heating method (150℃ digestion for 1 h, 220℃ digestion for 1 h, 320℃ digestion for 30 min, till white fumes were evolved) were used for sample preparation. The determination results of total arsenic, inorganic arsenic (arsenite and arsenate), and organic arsenic (arsenic betaine and arsenic choline) in urinary samples (from healthy people with high arsenic exposure, patients taking arsenic trioxide, and patients with arsenic poisoning) by atomic fluorescence spectrometry and ICP-MS were compared. The wet digestion-atomic fluorescence spectrometry was optimized for the determination of arsenic in urine, and its performance indicators such as linearity, detection limit, lower limit of quantification, precision, accuracy, and anti-interference ability were verified.

Results The recoveries of 150℃ and 220℃ wet digestion-atomic fluorescence spectrometry were lower than 90%, while the results of wet digestion-fluorescence spectrometry with gradient heating to 320℃ were close to those of ICP-MS, ranging from 97.92% to 101.16%. After wet digestion at 150℃ and 220℃, arsenic betaine and arsenic choline were hardly detected, and the recovery rates were extremely low; after gradient heating to 320℃, the recovery rates were 97.59% and 99.96%. By combining gradient heating and wet digestion-atomic fluorescence method, when the urinary arsenic concentration was 5.0-40.0 μg·L^-1, the linear relationship was good, the correlation coefficient was 0.999 7, the detection limit was 0.06 μg·L^-1, the lower limit of quantification was 0.18 μg·L^-1, the relative standard deviation (RSD) was 0.95%-3.24%, and the recoveries were 95.95%-103.10%. Lead, zinc, cadmium, copper, selenium, mercury, antimony, iron, aluminum, and manganese at 200 times the concentration of arsenic caused a deviation of < ±10% in the measurement results.

Conclusion The standard method can not effectively remove arsenic betaine and arsenic choline due to the low digestion temperature. In this study, the optimized wet digestion-atomic fluorescence spectrometry can effectively remove organic arsenic in urine, and the performance indicators are validated; therefore, the improved method is suitable for the determination of total arsenic in urine.