Chromatography Forum

Hi
I have problems with my dection limit using a RI detector and I consider swiching to ELSD. Do somebody know approximately how much I can gain using ELSD? Two of the compounds are phosphate and phosphite![/list]

Anders,

You may have a problem with your analytes here, as I believe one of the pre-requisites to using the ELSD is that compounds should be volatile.

You might use ELSD for non-volatile compounds, but in this case you need to form non-volatile salt. Volatile amines can be analyzed with TFA in the mobile phase. In case of TFA amine forms a salt. See methods below:

http://hplcmethods.com/compound_005.php (ammonia, methylamine, etc.)

Check the list below, almost all our methods were done with ELSD. It is very good technique for non-UV compounds, although detection limit is around 20 ppm (?):

http://hplcmethods.com/Applications_By_Compound.php

It suppose to read as "You might use ELSD for volatile compounds"

Evaporative Light Scattering Detection requires a volatile mobile phase and detects non-volatile (or only slightly volatile) solutes by the dust they leave upon evaporation of the mobile phase. The more volatile the analyte, the worse it is for ELSD.

To be honest, ELSD is probably a poor choice for this application, and making it work would be more an academic exercise than a practical analytical tool. Any conductivity detector will out-perform both ELSD and RI for detection of ions.

Phosphate and phosphite are in my opinion non-volatile compounds which are the criteria using ELS as detection method. If I then use a volatile mobile phase I can´t see the problem detecting these two compounds by the ELS method??

There is no obvious reason why you can't make ELSD work. Neither is there any obvious reason to expend the effort. What will ELSD do for you?

Anders,

Your limits of detection with a modern ELSD will be about 1 ppm assuming you do a good job selecting appropriate solvents and buffers for your LC method. I do not know how much you obtain with your RI now so I do not know if this is an improvement against your old method (but in general ELSD is more sensitive than RI.

Kostas,

Thank you for a concrete answer.