High Noise HPLC-ELSD Analysis/ Spiked peaks

[Lilian Fiuza]

I am developing a method for the analysis of FAME by HPLC-ELSD. Initially, I was using heptane with ethyl acetate as the mobile phase. To improve the sensitivity of the method, I switched from heptane to hexane. However, the peaks became very noisy and spiked. I tried changing the brand of hexane, but it did not improve the results. Can any one help?

[vmu]

Try to filter the eluent through a membrane with a pore diameter of 0.1 micrometer.

[Consumer Products Guy]

I did use ELSD occasionally at my workplace, the only one to actually use it.

Not what you asked, but I had over 4 decades of experience with fatty acids, soaps, and fatty acid methyl esters, and I always used GC-FID. Do you not have such equipment available???

[Multidimensional]

ELSD are for advanced users only. These are very complex detectors to use and maintain requiring many years of advanced training (the sales people who sell them do not want customers to know this.No one would buy them). They are very difficult to use, optimize and service (and they require a lot of cleaning!). The problem that you are experiencing is because you did not re-develop your HPLC method to use the far more VOLATILE Hexane that you substituted for Heptane. The two solvents have completely different boiling points. To use an ELSD or CAD unit, the whole method must be developed for use with the detector. All of the gas and temperature settings on the ELSD/CAD must be optimized for the mobile phase. Any changes to flow rate or mobile phase composition will change the method. The hexane is boiling inside your detector right now because you set it up for HEPTANE. YOu will need to start over and optimize the detector for use with the new mobile phase. Start by following the linked technical article on how to do this and get the help of an experienced chromatographer.

"Evaporative Light-Scattering Detector (ELSD) Optimization Procedure. Varex ELSD IIA, Technical Note # 0024. Copyright 1990."; https://www.researchgate.net/publicatio ... ltext=true

The gas flow rate, gas temperature and atomization must all be optimized for use with the more volatile solution. Please be sure to optimize the signal based on the S/N ratio of the PEAK, NOT THE BASELINE! Many make the error of adjusting the settings to improve the appearance of the baseline (this is incorrect). We always optimize the signal output of an ELSD or CAD based on the actual S/N ratio of the peaks. This is because changes to gas flow and temperature will change the peak area/height and baseline, but not in a linear manner. Increased areas may also result un increased baseline noise (so no gain at all). Again, review the technical article for more information on how to use these detectors.

[DR]

Recommend GC w/ FID for that, and even then, your sample diluent and column choices are critically important depending on which FAMEs you are working on.

[Vlad Orlovsky]

Totally disagree with ELSD being the most complex detector. From my experience it is the easiest detector to operate and maintain. You need to have two separate systems - one for methods with TFA and another one for methods with ammonium buffer. Using dedicated systems prevents prolong wait time to wash the system to reduce noise. We have 6 ELSD from Sedere and this is by far the best detector for the task. We are also using air and not nitrogen. Ypu need to replace one bulb (easily accessible, which can be purchased for 20 dollars at any automotive store). You need to run this detector with water at 100*C from time to time and clean nubuluzer may be once a year.

Try to wash your system and «cook” ELSD at 100 *C overnight. Also replace all PEEK tubing and see if noise goes down.
I would also consider switching to GC if this is an option.