Chromatography Forum

Hello,

I'd like to know about some kind of calculation that could eliminate salt interference from my HPLC data. I analysed seawater samples extracts in order to detect mycosporine-like amino acids in phytoplankton, but the main peak in my chromatograms was overlapped by the salt. At that time, I couldn't do any lab procedure to remove salt from the samples, but I did pass filtered seawater blancs through the column (different volumes were filtered: 200, 400 and 600 mL). As I knew that at the 360nm-scan there was no MAA (amino acid), I considered that the signal was all due to the salt, but at 310 and 330 nm, both the MAA and the salt were present. When I did a regression to try to estimate the contribution of salt in the MAA maximum wavelength (330 nm) and subtracted the salt integrated peak area from the total, I ended with some negative values. So I'll be very thankful if someone can pass me any reference about calculations that separate overlapping peaks.

Thank you,

Bruna

What about removing the chloride prior to injection, for example by passing through Alltech Ag Max-Clean cartridges?

Yes, it could have been a good solution, but the analyses were done a couple of years ago, and now I just have the data to deal with. Anyway, thanks for your reply.

Neither sodium nor chloride ions absorb in the UV, so if you were seeing any peak at all from your salt, it would have been due to a change in RI (UV detectors do respond somewhat to RI shifts), and therefore would be larger at longer wavelength (i.e., I'm not surprised that you got negative numbers). In the absence of spectral data, any peak deconvolution algorithm would have to be based on assumptions about peak shape (e.g., Gaussian peaks), and with a massive overload of salt, I doubt that such assumptions would be of much use.

Bottom line on all this is that I can't see any way to extract what you need from the data without having access to the original system and doing some additional measurements to calibrate the response to NaCl as a function of wavelength on that system.