Abstract:

The measurement errors of particulate carbon and nitrogen in water using three pretreatments were evaluated quantitatively, and impact factors were discussed. Next, an optimized sample collection, pretreatment, and measurement method was suggested. The results showed that the measurement errors of samples analyzed on whole filters were the lowest, and those of samples pretreated by scraping were the highest. The measurement errors of carbon were lower than those of nitrogen and were negatively related to the carbon content on the filter (r = –0.545, p = 0.000). When the carbon content on the filter was > 0.5 mg, the measurement error of the carbon was < 10%. When the carbon content on the filter reached 1 mg, the measurement error could be < 5%. In conclusion, the predominant factors influencing the measurement error were the carbon content, uniformity of the particulate, and sample residue on the filter. The results suggested that more complicated factors influenced the measurement of nitrogen content. An optimized method for the sample collection, pretreatment, and measurement was recommended, namely, a smaller glass microfiber filter (GF/F: diameter 25 mm) should be used to collect the samples, the whole filter should be analyzed without cutting and scraping pretreatment, and the carbon content of the particulate matter on the filter should be > 0.5 mg.

Key words: suspended particulate matter, pretreatment, carbon content, nitrogen content, element analyzer, measurement error