Chromatography Forum

Hi there!

We're running a method on an Acquity H-Class system that is generally isocratic (at least for our peaks of interest) with 0.1%TFA in 91:9 Water:Acetonitrile. When we pre-mix the mobile phase so that the method just runs at 100% A, the baseline looks amazing. However, when we prep MPA as 0.1% TFA in Water and MPB as 0.1% TFA in Acetonitrile and run it as 91% A and 9%B, the baseline is unacceptable for what we need it for (low-level impurity detection/quantitation). We've run both (pre-mixed and instrument-mixed) on several Acquity systems in our lab and have seen the same results.

Is our 100uL mixer inadequate? Does anyone have any suggestions on improving the baseline and keeping the simplified MBA/MPB approach? The flow rate is set at 0.6mL/min - 225nm detection wavelength. See below for a screenshot of baselines (blank injections).

Is our 100uL mixer inadequate?

The short answer is "yes".

Off the top of my head, I don't know what options Waters offers in mixers for that system, but you can certainly find others from aftermarket suppliers (e.g. http://www.hplc-asi.com/mixers/). If you are only running isocratic methods on a system, you can go to a much larger volume to increase mixing efficiency at the minor cost of increased equilibration time when you change mobile phases. If you will be running gradient methods, you will have to trade off the better mixing versus increased dwell volume and gradient distortion.

If this is isocratic and blank injection , why the baseline going up after 5 minutes ?

If this is isocratic and blank injection , why the baseline going up after 5 minutes ?

Hi - It's not entirely isocratic .... just for most of it (i.e., the first 5 minutes, where all of our peaks of interest elute).

There's a gradient at the end to get some other junk off the column (the quality of the baseline for this portion of the analysis is not as critical).

Is our 100uL mixer inadequate?

The short answer is "yes".

Thanks, Tom. I'll have to look into some of those aftermarket mixers ... I was afraid the short answer would be "yes" on that question!