[Dwade]

While performing an automated colorimetric determination of nitrate/nitrite using hydrazine reduction (EPA Method 353.1 and Standard Methods 4500-NO3¯ H., 20th Edition, 1998) I occasionally run a sample that produces a negative peak, i.e. peak drops below baseline voltage. I have traditionally analyzed freshwater agricultural field runoff and/or groundwater samples but I recently analyzed a sample of seawater that also produced a peak below baseline. If I spike the sample with a control standard the peak will rise so I know that nitrate is still being measured but I do not understand what might be in the sample that interferes with the analysis. A spiked saline mixture (3.5% NaCl) behaves similarly to spiked deionized water so I know that the salinity of the seawater is not producing the unwanted effect. Ammonia analysis of the same seawater sample does not yield this negative peak.

Could anyone please offer some insight as to why the peaks drop below baseline during nitrate analysis?

Thanks.

[Dwade]

Problem solved. I needed to match the seawater matrix to my sample wash and control standards to eliminate differences of refractive index. Here's the citation:

EPA Method 353.4

Determination of Nitrate and Nitrite in Estuarine and Coastal Waters

by Gas Segmented Continuous Flow Colorimetric Analysis

 

http://www.epa.gov/nerlcwww/m353_4.pdf

 

Note: In nitrate and nitrite analysis absorbance of the _______________________________________

reagent water is higher than that of the LNSW. When

using reagent water as a wash solution, the change in

refractive index causes the absorbance of seawater to _______________________________________

become negative. To measure the absorbance due to

refractive index change in different salinity samples, Low

Nutrient Seawater must be used as a wash solution to

bring the baseline down.

[Borek]

Thank you for a follow up. Just because nobody didn't answered doesn't mean nobody read the original question.