氨氮是衡量地表水质量状况的重要指标之一。 在现行的纳氏试剂分光光度法（HJ535—2009）检测氨氮含量过程中, 水样预处理（絮凝沉淀法）、 剩余浊度与仪器基线漂移等因素给检测过程带来不确定影响。 据此, 提出一种双波长（420和650 nm）光谱的改进方法, 通过扣除由剩余浊度与仪器基线漂移引入的吸光度, 以消除检测干扰。 先确定水样在420 nm处的浊度修正系数k, 后以420 nm处的吸光度减去650 nm处吸光度与k值的乘积, 得到氨氮显色反应的净吸收值, 从而实现对氨氮含量的准确定量检测, 并对该方法的精确度与准确性进行评价。 结果表明： 在单波长法中, 仅由絮凝沉淀法过滤过程引入的相对偏差可达8.67%, 而双波长法无絮凝沉淀步骤, 不受该偏差影响。 双波长法5次重复实验的精度标准差低至1.58%, 回收率在98.5~103%之间, 因此更为准确、 可靠。 与现行的纳氏试剂分光光度法相比, 该方法既省略了水样预处理的步骤, 又扣除了剩余浊度的干扰, 显著提高了实验的效率, 特别适用于大批量地表水样中氨氮含量的快速检测。

Ammonia nitrogen is an important indicator to measure the quality of surface water. In the process of detecting ammonia nitrogen content with Nessler’s reagent spectrophotometry method (HJ535—2009), water sample pretreatment (flocculation method), residual turbidity and instrument baseline drifting will bring uncertainty to the process. Accordingly, a spectroscopic method based on dual-wavelength (420 and 650 nm) measurement is proposed, so as to subtract the absorbance of residual turbidity and instrument baseline-drifting to eliminate such interference. We first figure out k, the turbidity correction coefficient of the water sample at the wavelength 420 nm, then divide the product of absorbance of 650 and k by the absorbance of 420, and finally obtain the net absorption of ammonia nitrogen chromogenic reaction. Thus we can make accurate quantitative detection of ammonia nitrogen content and evaluate the precision and accuracy of dual-wavelength spectrometry. The results shows that, with single-wavelength method, the relative deviation can be up to 8.67% caused by the filtration process of flocculation and sedimentation, while the dual-wavelength method would not be affected by the deviation, since dual-wavelength method includes no filtration process. The accuracy standard deviation of dual-wavelength method could be as low as 1.58%, and the recovery is between 98.5% and 103%, which shows that the method is more accurate and reliable. Compared with the current Nessler’s reagent spectrophotometric method, the present method not only omits the process of water sample pretreatment, but also avoids the interference of residual turbidity, which could significantly improve the efficiency of the experiment. Therefore, it is more suitable for rapid determination of ammonia nitrogen in large quantities of surface water samples.