Chromatography Forum

Hello

I am getting no response from the ELSD when I have the SEC column attatched to the system, regardless of what solvent I'm running through eg. water or methanol or acetonitrile. But I am getting some response when I attached another column and running a validated method. It seems to me that the problem is my SEC column, phase collapsed maybe? But what I don't understand is that twhy doesn't the ELSD give somekind of response (rather than zero) since there is mobil phase pumping into the ELSD.

Any advices/suggestions?

hi Vliu!
I have three questions befor I can have any idee of a solution for your problem:

1. What compound is it you inject (chemistry)?
2. What column are you using (packing?)
3. Are you sure the column doesnt work as ion exhange and therefor your compound sticks inside the column?

Did you consider your detector makes an Auto-zero at the injection? If you work in isocratic mode and your column is well equilibrate, depending on the amplification factor used by the detector you could be unable to see a drift.
Are you sure you use exactly the same setting than with the other columns?

Which mobile phase are you using for your SEC separation?

When you blow up the chromatogram do you see noise or is it really a flat line?

Check if lamp did not die, inject sample with out column and see if you have response, inject sodium chloride with out column and see if 0 response is due problem with detector, problem with sample or something else (temperature, nitrogen flow, etc.).
Sometimes you will have flat line if you are overloading detector (bleed from the column, mobile phase not compatible with ELSD, TFA in the channels while using ammonium formate or acetate, delivery from other channels containing non-volatile mobile phase due to faulty proportional valve)

Columns that bleed much can give indeed a flat baseline. We did a few years back a study on this and found quit some noise with polymeric columns.

To test this, you can start an blank injection without column. Switch off the pump during the run and attach the column. Do not give the detector an auto zero. Monitor the baseline for a while.

bisonium:

1. What compound is it you inject (chemistry)?
A: Surfactant, trying to measure both micelle and monomer form

2. What column are you using (packing?)
A: Size Exclusion column 2000

3. Are you sure the column doesnt work as ion exhange and therefor your compound sticks inside the column?[/quote]
A: Not sure.

bhuvfe:
When you blow up the chromatogram do you see noise or is it really a flat line?
A: Flat line, no spike at all.

Vlad Orlovsky:
Lamp reference energy is 200+ and has only used 1000 hours only. I do get a solvent front response when injecting water without column. Will try injecting sodium chloride.

I am using compressed air rather than Nitrogen, I've compared my results using both gases before with a validated method, and the results were comparable.

What can I do if column bleed is causing the flate basline on the ELSD? Does it mean I need a new column? I am using a guard column already...

I am using 50% acetonitrile with 0.01M ammonium acetate (I've tried other % of acetonitrile as well). I think I should mention that I did get some response using the Refractive detector but it is not sensitive enough for my purpose.

All lines are in MeOh/water.

Koen Hollebekkers:
How did you overcome column bleed?

So you are suggesting testing the baseline with the pump off ?

I am using compressed air rather than Nitrogen, I've compared my results using both gases before with a validated method, and the results were comparable.

Are you sure your air is completely oil-free? If not you could dirty your detector with oil drops.

I am using compressed air rather than Nitrogen, I've compared my results using both gases before with a validated method, and the results were comparable.

Are you sure your air is completely oil-free? If not you could dirty your detector with oil drops.

Yes, it's oil and particulate free as I have a filter in place before reaching my detector.

Using salt buffers with ELSDs is not advisable. As the ELSD evaporates off the solvent, it leaves the salt, which will cake the nebuliser. Granted, it has been a few years since I used ELSDs and I know they have changed, but I would think this principle has not changed.

As luck would have it, I have had to look into characterization of loaded liposomes recently, which is fairly similar to what it appears you may be doing with the micelles. For my case, average liposome size is 200 - 900 nm, which 90% of the distribution being in the range of 200 - 300 nm. Columns people seem to have luck with in the literature are TSK-G6000 colum in series with a TSK-G4000 or 5000 column.

I am not sure of the exact manufacturer of your column, but if the nomenclature is the same as the ones I referenced, then yours might be too small and the analyte is unable to pass.

Finally, check your temperatures to make sure you have a large enough difference between mobile phase/analyte and that the components of your ELSD are not too hot/cold. Sorry I can't be more exact, I know the recommended settings have changed since I last used ELSDs.

He's using 0.01M ammonium acetate. It's volatile and perfectly acceptable for ELSD (or MS).

In terms of column bleed - just purchase from a column vendor that has LC-ELSD data.

Using the ELSD required more maintenance work. Perhaps you can implement a "steam cleaning" procedure to clean out the junk in the nebulizer and evap tube?
We do it on our PL-ELSD by setting nebulizer at 90C, evap at 120C, and gas flow at 2.8 SLM. MP is degassed, HPLC grade water @1mL/min, and run this for 2 hrs. Then, we do the "dry-clean" by stopping the liquid flow, and keep the hi-temp for >30min.
This cleaning procedure may be described in your manual (check it out). It will help to restore the sensitivity, and reduce the baseline noise.

Also, are you developing a new method, or using an established method? If yours is new, perhaps your material is retained on the column (add more organic)? Or your concentration is too low? Try to inject more, or use higher conc.

Have fun with your ELSD!