Improve Resolution

[Melissa]

Hello everybody!

I'm new in this field... so I really need your help...

How can I do to improve resolution? I need some tips for improving the resolution in my run.

Thanks a lot for your help!

Melissa

[JGK]

Melissa,

In order to make sensible assessments on possible courses of a ction we require more information.

Chromatographic conditions and a typical chromatogram would help greatly and avoid a lot of useless suggestions.

The more information you supply, the better assessment we can make for you.

[Melissa]

Well

I’m analyzing sulfonamides (Sulfapyridine-SPY, Sulfadiazine-SDZ, Sulfathiazole-STZ, Sulfametazine-SMZ, Sulfadimethoxine-SDM and Sulfaquinoxaline-SQX) in a gradient mode with acetic acid (2%) and acetonitrile, but I also have a flow gradient. The column is a Zorbax Eclipse XDB C18 150mm x 4.6 mm x 5um. Column temperature is 55°C. Fluorescence detector with Em: 495 nm and Ex: 405nm. Injection volume: 20uL.
I start with 23% of Acetonitrile and 77% of Acetic Acid (2%) hold for 15 min (1.7mL/min). Then 35% ACN and 65% Acetic Acid hold for 2 min (1.5mL/min) and then 40% ACN and 60% Acetic Acid hold for 8 min (1.2mL/min).
I separate four of the six sulfonamides but sulfapyridine and sulfathiazole elute almost together. This is the order of elution: SDZ (15.380min), SPY(16.443), STZ(16.996), SMZ (19.106), SDM(24.772), SQX(25.635).
I tried to reduce the flow in order to increase resolution but it didn’t work. Then I reduce the injection volume, then decrease column temperature, but the resolution get worst.

I'm really new in this field, and I don't know what else I can do.
Thanks for your help.

[bisnettrj2]

You're running that column pretty hot, and when you do that with acid you can shorten a column's lifetime. Agilent claims the XDB is stable at low pH/high temp, but it's usually a safer bet to run a little cooler than the max (esp w/ acid). Just an FYI.

I'd like to clarify your gradient conditions. From what you posted, it seems like your gradient is:

t0 - 77:23 2% HOAc:MeCN 1.7 mL/min
15 - 77:23 2% HOAc:MeCN 1.7 mL/min
15 - 65:35 2% HOAc:MeCN 1.5 mL/min
17 - 65:35 2% HOAc:MeCN 1.5 mL/min
17 - 60:40 2% HOAc:MeCN 1.2 mL/min
25 - 60:40 2% HOAc:MeCN 1.2 mL/min

I'm confused by this - most gradient are just that - gradients. It seems you have "steps" where you immediately change from one solvent composition to another. Unless I'm reading your description wrong and your gradient is more like this:

t0 - 77:23 2% HOAc:MeCN 1.7 mL/min
15 - 77:23 2% HOAc:MeCN 1.7 mL/min
17 - 65:35 2% HOAc:MeCN 1.5 mL/min
25 - 60:40 2% HOAc:MeCN 1.2 mL/min



Could you clarify your gradient parameters?

What is your injection solvent?

Did you inherit this analysis, or is it something you've developed?

You have so many variables in that analysis that it will be hard to pin down which will be the most effective in order to simplify your analysis and get the resolution you desire.

Personally, I would try running isocratically at 1.5 mL/min, 40 C, and 65:35 2%HOAc:MeCN, and adjust one parameter at a time (probably mobile phase concentrations first) to influence the separation.

From what I've learned here and elsewhere, mobile phase constituents and pH are probably going to be the biggest parameters to adjust. You may want to make a low pH buffer (maybe a formic acid:ammonium formate buffer around pH 3.5?) to isolate the effects of minor pH changes.

You may also want to try methanol instead of acetonitrile, but beware that you will see a higher backpressure with methanol than acetonitrile.

[lmh]

and further to bisnettjr2's post, if you are going to run a gradient rather than go with his isocratic approach, there's a lot to be said for keeping gradients very simple in the first instance, a single linear ramp. You can fiddle with the exact slope at different points later if need be, but very often this is with a view to speeding up the run rather than improving resolution (i.e. you are ramping quickly over regions where there are few relevant peaks). I am no expert on method development, but I do find that misguided attempts to improve resolution by flattening gradients can, without care, end up generating stretched chromatograms that are twice as long with peaks twice as broad...

[Melissa]

This analysis is half developed and half inherited...
My gradient is more like the second one. My description is not clear enough...sorry.

When you ask me about the injection solvent, do you mean in which is the sample dissolved?... If it is that, I use a buffer of sodium acetate 1M pH 3.5

I tried running isocratically but all the analytes elute almost together.

I'll try to start again chaging the parameters little by little... Maybe I find the right one... :?

Thanks for your help....

[bisnettrj2]

What isocratic conditions have you tried? Have you tried different organic constituents of your mobile phase (methanol, THF)?

[Melissa]

Well... actually I haven't tried another solvent... I only have used acetonitrile and acetic acid. I'll try first with methanol because I don't have THF in my lab.

My internal standard is sulfapyridine (SPY), what do you think about changing the internal standard in order to avoid coelution of SPY and STZ?

[Bryan Evans]

Below is HT analysis of sulfonamides on Unison UK-C18:
http://www.imtaktusa.com/site_media/fil ... TI258E.pdf

You can try similar conditions (just scale up the method accordingly)

[Melissa]

Thanks Bryan

[bisnettrj2]

Well, I think you'll want to stick with something similar to your analytes, but not something you'd expect to find in the samples you're looking for, so you could switch your IS to something that isn't co-eluting. Since you only have issues with the IS, you could swap it for something else that is still similar to what you're analyzing. But I really think you should work on a more straightforward analysis scheme in any case, as it will make it simpler to scale this analysis between different columns (shorter with smaller particles, or larger ID for prep or semi-prep) or transfer between systems.

To scale Bryan's column and gradient paramters to paramters that fit your column, you can use the following site for a decent approximation to get you started:

http://www.hplctransfer.com/GradientMethod.aspx

[Melissa]

bisnettrj2 Thank you very much!!!!!!!!!!!!!!!! Wao!

I'll work on a simple analysis scheme, and if I don't get anything I'll try to change the IS. I realize that my gradient is very complex, maybe I can get a shorter run with the resolution I want if I work a little bit on it

Thanks!!!

[bisnettrj2]

I found another application note on sulfonamides. Follow the link to download the application notes from YMC. There are a few different columns shown for sulfonamides.