Trimyristine reproducibility struggle w/ELSD

[klugeelse]

Hello everybody,

I have enormous problems achieving reproducibilty - even over a time range of 8 hrs - with my HPLC-ELSD. Sample is trimyristine (saturated triglyceride with 3 C14-chains) in neat THF, concentration about 0,2mg/ml). The peak areas increase up to 30% within a day. I really don´t know what to do anymore and I have put in a decent amount of time to put it mildly.


HPLC-pump: Waters 515
autosampler: Waters 717 plus
detector: Alltech 3300 ELSD

Software: Clarity (Version?)


sample: trimyristine in THF, ca 0,2 mg/ml
column: ODS, 125mm*4mm; 5µm
eluentl: isocratic THF/ACN (45/55)
flow rate: 1,5 ml/min
injection volume: 10µl
run timet: 3 min

ELSD settings:
drift tube temperature: 40C
N2 flow: 1,6 l/min
gain: 1

I´ll give some more details that I think might be asked:

- the lab has an AC
- why 40C drift tube temperature? This is what the manufacturer advises for the given mobile phase, flow rate and sample concentration. Baseline looks awesome.
- The column is not heated, we don´t have an oven.
- All tests you can run are passed excellent according to the manual (i.e. gas flow, dirft tube temperature, clean optics.
- I run a hot cleaning (120C) for 1,5hrs; no changes.
- I took the detector apart and sonicated the nebulizer in MetOH/Water (50/50) . Didn´t change a thing. In the "light trap" I found some fluffy stuff that could be removed easily. Didn´t look like lipid at all.
- The peaks look good. Symmetrical, no fronting or tailing.
- Injection of neat THF or mobile phase don´t yield any peaks. Pure baseline.
-Autosampler works properly, verified by reproducible results with UV (substance for the test:cholesterylmyristate).
- All capillaries are steel, no peek involved.

What could be going on here?
The detector was purchased for this purpose and is also advertised for exactely this kind of application. I don´t want to do any derivatisation and neither do I have a GC.

I thank you very much for taking your time reading this and appreciate your input.

Best regards,

Klugeelse

[fandaga]

1. Do the peak areas look like they are leveling off on subsequent injections? Or do they keep going up? I imagine that eventually they would have to level off. Also, does the peak shape worsen?
2. My first thought is that the problem might be contamination of the nebulizer. Try sonicating it in a solvent that Trymyristin is more soluble in. It doesn't seem like it is soluble in 1:1 Water/ MeOH.
3. How soluble is this compound in the mobile phase? You might want to consider a washing routine at the end of each run.
4. How long have you let the ELSD warm-up before doing a run? I like to at least let the ELSD warm-up for 24 hours when I haven't used it in a while. When I'm using it a lot, I put it into standby with a low flow.

[klugeelse]

Hello Fandaga,

thank you very much for taking an interest!
Here are my answers to your questions:

1. Do the peak] areas look like they are leveling off on subsequent injections? Or do they keep going up? I imagine that eventually they would have to level off. Also, does the peak shape worsen?

When I started working on this I always did double injections (i.e. 2 subsequent runs of the same sample). Now I am doing 10 (!) subsequent runs. And it depends. On the same day, same sample, I had one sequence where it seemed to level off after the 7th injection. In an other sequence it looks like it doesn´t. I say "look likes" because even for the leveled off part I have a relative standard deviation of 3 percent which is not exactely brilliant.
Peak shapes do not worsen. Mostely the bigger area is due to a higher peak. Sometimes it is due to a slightely fatter peak.

2. My first thought is that the problem might be contamination of the nebulizer. Try sonicating it in a solvent that Trymyristin is more soluble in. It doesn't seem like it is soluble in 1:1 Water/ MeOH.

That makes sense, I´ll try THF. I have to say that taking the detector apart was my last nothing-to-lose-approach. I was afraid of destroying something so I stuck very close to the manual which advised MetOH.

3. How soluble is this compound in the mobile phase? You might want to consider a washing routine at the end of each run.

Well, to be sure not to get into solubility problems with the sample I dissolve it in neat THF rather than the mobile phase. Here is what puzzles me: when I inject neat THF solvent I don´t get any peak! Shouldn´t I get a detector response when washing off possible residues?

4. How long have you let the ELSD warm-up before doing a run? I like to at least let the ELSD warm-up for 24 hours when I haven't used it in a while. When I'm using it a lot, I put it into standby with a low flow.

OMG! I shut it down every night. When I turn it on it has a 5 minute "euqilibration" period to get the set temperature and gas flow. After that I connect it to the capillary (always hangs in a waste canister while flushing the column) and wait until I get a steady baseline with my mobile phase. All in all never more than an hour.


Best regards,

Klugeelse

[HW Mueller]

How about sample solvent evaporation?

[klugeelse]

How about sample solvent evaporation?

I´m sorry, I don´t understand this comment. What do you mean?

[HW Mueller]

THF is quite volatile, could it be that you are loosing solvent such that the sample gets more concentrated as time elapses?

[klugeelse]

THF is quite volatile, could it be that you are loosing solvent such that the sample gets more concentrated as time elapses?

Oh I see. No, I can exclude that. As for the flask I prepare the sample in, I watched the level over a whole day and it stays the same. The sample vials for the autosampler are with a snap fit plastic cap and should be at least good enough for 10 subsequent injections (i.e. overall time of 20 minutes)...

Thank you for your input.
Best regards klugeelse

[Chromatographer2010]

You could weigh the vials before and after injections to see if the injector is picking up the right amount during the day. Something like a leak in the syringe or an air bubble could cause the amount of sample picked up to change over the day. Also if the sample is slowly going into solution in the vial over the day it could increase your response.

Your point about the blanks not showing a peak does indicate that the extra peak area is not a contamination issue.

The other thing that could cause this is the nebulization or the sample transferance to the drift tube is getting more efficient over the day. Could be something with the temperature or nebulization gas flow or even a partial clog in the vent line. Sometime if you have condensation in the gas vent line it can clear over time and improve nebulization.

Also THF is not stable and forms peroxides over time especially after you take it out of the bottle. Might be some chemistry going on with the sample or the nebulization.