Chromatography Forum

Hi guys,

From many months ago I've a problem with extra peak in my chromatrogram by HPLC when I use a gradient eluition.

I enclose a blank chromatogram ( the sample is mobile phase A) acquire today. This is an example of my problem. It change from method to method.

The gradient is:

t____A____B____curve
3___100___0_____0
10__60___40_____1

Where A is the acqueous phase (2mM sodium heptane sulfonate and 0,1 % orto-phosporic acid) and B pure acetonitrile.

Have you experience about this?

Thanks!

Thanks you!

link: http://imageshack.us/photo/my-images/68 ... nexyr.jpg/

Which peak is the "extra peak" you are referencing? The blobs at 8.5 or 9.5 minutes, or the two peaks at 11.3 and 12.3? Or are they all "extra peaks"?

All are extra peaks

What stands curve "1" for? Linear change or stepwise?
What is your detection wavelength?
Flow rate?
Column?
Dwell volume? (gradient delay)

With heptane sulfonate, you're likely doing ion-pair chromatography.
Therefore you should stick to isocratic elution or at least try to keep the concentration
of the additive in both solvent constant.
Beside this, 2mM seems to be a bit too low for me.

If your substances don't need ion-pair reagent, switch to some generic buffer, e.g. NaH2PO4 or try only phosphoric acid for its own.

The broad region at 7-10 min could be the wash out of the ion pair reagent (if curve 1 is not a step),
whereas the signals at 11.1 / 12.3 min seem to be just nice peaks, related to some impurities from your solvent/reagent used for mobile phase A. (?)

How often do you refresh your solvent? Use of fresh bottles?

As Hollow suggested, gradients and ion-pair reagents are not a good combination. Even keeping the same concentration of IP reagent in both A and B will not eliminate the problem (it will help, however), because how much "sticks" to the stationary phase depends on the mobile phase composition. Contaminated mobile phase is easily checked for by running the "three blank gradient" test. There's a description in the "mini tutorial" I posted at the top of this section (viewtopic.php?f=1&t=19085).

What stands curve "1" for? Linear change or stepwise? Linear change
What is your detection wavelength? 254nm
Flow rate? 1.2 ml/min
Column? kinetex 75mm x 4,6 mm x 2.6 um
Dwell volume? (gradient delay) 2.5

With heptane sulfonate, you're likely doing ion-pair chromatography.
Therefore you should stick to isocratic elution or at least try to keep the concentration
of the additive in both solvent constant.
Beside this, 2mM seems to be a bit too low for me.

If your substances don't need ion-pair reagent, switch to some generic buffer, e.g. NaH2PO4 or try only phosphoric acid for its own.

The broad region at 7-10 min could be the wash out of the ion pair reagent (if curve 1 is not a step),
whereas the signals at 11.1 / 12.3 min seem to be just nice peaks, related to some impurities from your solvent/reagent used for mobile phase A. (?)

How often do you refresh your solvent? Use of fresh bottles?

Near your question I write my response.

The solvent that I'd use for this course was new, prepared at same day of injection.

The problem that you see is standard in all my gradient analysis. It's not only for ion-pair reagent. With all generic buffers I see extra peaks. Different from that you see now in my post, but ever
extra peaks.
I'd try to change all: Water, salt, organic solvent, bottle, column, ortophosporic acid. All that enter in my HPLC system! But nothing... this peak always eluite!
All HPLC system was revisionad also from customer technical. But it dosen't see malfunction.

I can't understand why I see extra peaks. I can't...

As Hollow suggested, gradients and ion-pair reagents are not a good combination. Even keeping the same concentration of IP reagent in both A and B will not eliminate the problem (it will help, however), because how much "sticks" to the stationary phase depends on the mobile phase composition. Contaminated mobile phase is easily checked for by running the "three blank gradient" test. There's a description in the "mini tutorial" I posted at the top of this section (viewtopic.php?f=1&t=19085).

Thanks Tom. I've see this post last week, but I can't understand what you say. I'm Italian and to understand english speak from english people it'm most difficult for me

I couldn't tell that from your written English, which is excellent!

A written version of the three-blank-gradient test is here:
http://www.lcresources.com/resources/TSWiz/hs400.htm

some additional tests I would do for trying to locate the source of the problem

- run the gradient without injection of anything (either 0µL injection or no inj at all)

- substitute the column with plain capillary (to get reasonable backpressure, e.g. 1m of 0.12mm)
-- make runs with and withouth injections of anything

- how does a gradient of plain water vs. acetonitril looks like?

? What system is it?
? How much is the injection volume?
-- as core shell particles, their loadability is drastically reduced. for your kinetex 2.6 µm, I estimate the loadability
would be that of 1.1 µm particles -> Equivalent to only 40% of your actual (column) volume

Hollow sorry but I can't well understand what you say about injection volume. I must reduce it? Now I inject 5ul.

My system is perkin elmer serie 200 (pump and autosampler) with 785A UV detector.

In past I've try all that you say. But peaks always eluite. They dosen't eluite when I've inject only water and acetonitrile, without other added. But This test is "useless". Without pH correction I don't reproduce analysis condition.

However I'll try again this test and I'll write what will happen.

thanks you very mich!

If your peaks (the expected ones) have good symmetries than you're probably ok, and 5 µl are not too much at all.

But if you encounter problems with peak symmetry, maybe try first to reduce the mass load by a) reducing inj volume b) dilute your sample.
Because you have a non-porous core within your particles the effective accessible surface is reduced compared to fully porous particles of the same dimension,
making it prone for overload effects if one doesn't take care of this.

I don't see the test as "useless" because it tells you that nothing seems to be wrong within the system nor with your water nor ACN.
So it is probable that there is something with the ion-pair salt and/or your phosphoric acid.
Have you tried to run the blank gradient with A=only the salt in water and A=only phosphoric acid in water?

And another question:
How do you prepare your eluent? Is there anything involved that could contaminate your mobile phase, e.g. pH-probe?

Again, the heptansulfonic acid salt is not just a buffer, but acting as ion-pair reagent.
It will stick to your C18 surface with it's C7-tailc turning your column "temporarly" into a cation exhanger by the sulfonat group.
The concentration of the equilibrium between C7 and C18 depends on temperature and the composition of the mobile phase.
If the ACN concentration gets higher, the C7-load will decrease and be washed out. That's what the humps probably are.
Therefore it should not be used with gradients.

Have you already tried to use an ordinary phosphate buffer (Na or K, e.g. 10mM) at the same pH? Will the separation work?

If you really need this ion-pair mechanism, switch to isocratic elution (if possible), try a dedicated SCX-column,
where the sulfonic acid groups are fixed to the surface or stationary phase, or maybe you could try a mixed-mode column (a certain user here will doubtlessly help you on this )


And yeah, I don't know the PE 200 so I can't give you any hardware specific help.

"Have you tried to run the blank gradient with A=only the salt in water and A=only phosphoric acid in water?" No. I don't try it. If I'll try this, what gradient curve I must use?

About ion-pair agent I'm still search a correct choice of it. If you want help me in my research you can write your opinion in my other post "how to improve purine peak shape?".

Thanks at all.

PS: at today I don't try nothing. I don't have much free time. I'll update as soon as possible

By by!

"Have you tried to run the blank gradient with A=only the salt in water and A=only phosphoric acid in water?"
No. I don't try it. If I'll try this, what gradient curve I must use?

About ion-pair agent I'm still search a correct choice of it.
If you want help me in my research you can write your opinion in my other post "how to improve purine peak shape?".

Thanks at all.

PS: at today I don't try nothing. I don't have much free time. I'll update as soon as possible

By by!

a) the idea of the blank gradients is to isolate the extra peaks either to the acid or the salt, so use the same gradient as always.

b) ion pair and gradient don't mix! so you could stop searching if you're not going to switch to isocratic mode.
Even then, ion pair is somehow difficult if you don't pay attention to the quality of the reagent or column equilibration.

c) I've read your other thread and support Danko's post on page 2.
What you should do is start over with the whole development and do a column screening first, or stick with what you have
and maybe some published methods, even if they are not perfect.
Chromatography is all about equilibria, so whatever you add or change is likely to also have an impact on your other compounds and
very quickly you can mess up everything that you've achieved so far.

d) I don't know your compounds well, but maybe HILIC is an option, as well as some mixed-mode or other columns designed for
polar compounds and compatible with 100% water like Waters Atlantis or others, or as suggested some designated SCX columns. There are also some non-silica column designed for basic compunds.

e) already played with column temperature?

f) get you some basic books like the one from Veronika Meyer "Practical HPLC" or the "Practical HPLC Method Development" form Snyder, Kirkland And Glajch

Hi Sir!

In june I was resolve extra peak problems with change of salt in mobile phase. I've change supplier and all was go ok.
But I say that with astonishment, because In may I've did try this way, that is the change of salt, but nothing were change. It's a mistake!

Then now, from a distance of to 2 months, I've reuse this salt ( the "perfect" lot) to prepare a new batch of mobile phase. I place before that for this 2 months I didn't open salt container. It were stock in my reagentary. I've lunch my HPLC method and surprise... All extra peaks are eluite again! My baseline is orrible! When I put in my run, with gradient mode, 5% of MeCN it happen a disaster. Extra peaks with a big absorbance value.

Someone can help me? I'm desperate

Than you so much!