Chromatography Forum

Fellow chromatographers,

i've troubles optimising our analytical method : the peaks of all the target compounds are broadening (peak fronting) in function of time (see the picture below).



We have an Agilent 1100 LC/MS (MWD + Ion Trap) and as you can see, i use a gradient system with following settings @ a flow rate of 0.4 ml/min:

00.00min 85% H2O / 15% Acetonitrile + 0.1% formic acid
08.00min 60% H2O / 40% Acetonitrile + 0.1% formic acid
08.50min 0% H2O / 100% Acetonitrile + 0.1% formic acid
09.50min 0% H2O / 100% Acetonitrile + 0.1% formic acid
10.00min 85% H2O / 15% Acetonitrile + 0.1% formic acid
15.00min 85% H2O / 15% Acetonitrile + 0.1% formic acid

The column is a 2.1 mm x 20 mm C18, temperature controlled @ 35.0 C and with an injection volume of 3 µl. We measure @ 250 nm wavelength.

I started a sequence run with 1 standard at a concentration of 8 µg/ml of the highest target compound (RT 8 mins), and injected it to simulate a total analysis time of 71 hours. As you can see on the graph, the black chromatogram is the first injection and the red chromatogram is the last injection @ 71 hours.
The column supplier confirmed me that my LC-settings/parameters are acceptable for the column i use.
Increasing the temperature improved a bit the peak shape, but still not satisfactory (see graph). I was thinking to increase the equilibrating time to 7 mins instead of 5mins, but i wanted to hear what you guys were thinking about this. Can anyone help me to find a solution for this ?

Whenever all the peaks in a separation show the same problem, it's a good bet that the problem occurred when the peaks were still together (i.e., at or previous to the head of the column). It's hard to really tell, but it looks like most of the peaks are developing a shoulder on the back side (peaks at 4, 6, and 9.5 minutes, especially). Common causes are trapped air bubbles (intermittent problem), a void space at the head of the column (usually a gradual onset), or a partially-plugged inlet frit (usually an abrupt onset).

If you haven't already done it, go back and look at the results from the intermediate runs and see if the peak shape has been gradually getting worse (which is what I suspect from your post) or if the problem showed up suddenly. If it showed up suddenly, then backflushing the column may solve the problem (first, check with the supplier first to make sure the column *can* be backwashed!). If the onset was gradual, backflushing may still help, but is likely to be only a temporary fix. With that small column, it doesn't take much void volume to screw things up!

One other possibility is some void space at the inlet end fitting. This is more likely if you are using PEEK tubing or fittings (the tubing may have slippped a bit in the fitting).

Further to Tom's comments, if you are using an Agilent 1100 you should also be able to look at the pump pressure which would probably have been collected during the runs. I'm not familiar with their ion-trap software unfortunately.

If you have an air-bubble, you will probably see a very unstable pressure correlating with the poor runs. Obviously partially blocked inlet frits may show up as increased pressure.