setting up a new Method and need help

[mikejungle]

Hey guys, i'm new to the forum and i would think fairly new to HPLC.

I've learned a lot within the past year and i've fixed several problems we've had with our agilent 1100 + sedex elsd (figured out our degasser was on the fritz so i took it apart and cleaned the pump, found solvent backing into the nebulizer and causing baseline issues, random chemstation issues...).

Now, we have a new agilent 1260 + agilent 380 elsd and i'm having issues setting it up and crossing over the same method from the 1100's.

We analyze diacylglycerols in crude and degummed "vegetable" oils (e.g. soybean), using an AOCS method.

Here's what i'm seeing:

1. The peaks kind of look like condoms; instead of a nice parabolic signal, the peak bulges in the middle and tapers to a point...what is this?

2. I'm finding split peaks and asymmetry in some of our samples. I know one reason may be because of overloading: with the lower part of our standard curve, the signal is really weak, so I was told to try a 50 uL injection (our normal on the 1100 is 16 uL).

3. Some of the peaks are not smooth, but wavy. Wavy on the way up and wavy on the way down. Our analytical chemist says that it's a smoothing setting, and that we shouldn't implement smoothing, because it'll compromise the integrity of the data, but i've never seen anything but smooth peaks, so i'm not sure what to think here.

If i've left anything out, please point them out and let me know what information i can provide.

[Wheels1978]

Hello,

We're using Agilent 1200 + 385 ELSD (ex Varian) here.

1. The peaks kind of look like condoms; instead of a nice parabolic signal, the peak bulges in the middle and tapers to a point...what is this?
I'm guessing this is a temperature issue. Not sure, but try a higher nebulizer temp.

2. I'm finding split peaks and asymmetry in some of our samples. I know one reason may be because of overloading: with the lower part of our standard curve, the signal is really weak, so I was told to try a 50 uL injection (our normal on the 1100 is 16 uL).
Chromatography issue maybe? Is your injection solvent the same as the mobile phase?When you use a larger injection volume it can't be neglected compared to mobile phase flow.

3. Some of the peaks are not smooth, but wavy. Wavy on the way up and wavy on the way down. Our analytical chemist says that it's a smoothing setting, and that we shouldn't implement smoothing, because it'll compromise the integrity of the data, but i've never seen anything but smooth peaks, so i'm not sure what to think here.
Data smoothing indeed. More smoothing means less data, but if you don't smooth, you can't integrate properly. Have a look at the data rate.

Good luck!

Regards,

Patrick

[mikejungle]

thanks very much for the reply. I'll look into these things and see what happens.