Strange peaks using HPLC for ATP determination

[andrei1]

Dear all

I have some problems with ATP determination i'm using RP ion-pair with TBA, long story short... 3 weeks ago i had it figured out how to obtain a nice chromatogram of ATP standard, two weeks ago I've tried to do a standard curve and it all felled apart, first it started with peaks trailing, splitting, shouldering, the only difference is that the mobile phase it's the same but is new(I've remade it several times since then) Also i've having problems with the pump it keeps getting stuck(i was told that's because of the high concentration of phosphate buffer)

I''m running a gradient using Buffer A composed of 25mM Phosphate with 100 mg/l TBA pH 5 and Buffer B composed of 200 mM Phosphate with 100mg/l TBA and 10% ACN pH 4

Any suggestions on what I should do?
Here is a pic with the chromatogram

[danko]

First, both pH 4 (eluent B) and pH 5 (eluent A) are inappropriate for the phosphate buffer. And then Why on earth are A and B adjusted at different pH?
Finally, 10% ACN and in reversed phase chromatography tels me that you don't achieve good binding of your analyte to the stationary phase.
It seams to me you got it all wrong. If you need more help you'll have to share more information - i.e. chromatographic conditions.

Best Regards

[Vlad Orlovsky]

Here is alternative approach without ion-pairing reagent. Your ATP will behave similarity to CTP:
http://www.sielc.com/Compound-Cytidine- ... phate.html

[andrei1]

Hy guys, thanks for help, sorry for the late response. Here is a pic of my chromatographic conditions


Here is a link for the article from which i got my method, also i only have a column for RP
http://www19.zippyshare.com/v/99304865/file.html

[danko]

The mobile phase set/system resembles ion exchange chromatography conditions if there weren’t an ion pairing reagent. The column on the other hand is a plain RP. The pH is wrong and the small percentage acetonitrile in B tells me that the person that suggested these conditions have very limited knowledge of chromatography.
My suggestion to you would be as follows: Reduce the phosphate concentration in B to the same level as in A. Adjust the pH to 6.8 – 7 in both eluents. And optimize the TBA concentration by very small increments.
Best Regards

[andrei1]

The mobile phase set/system resembles ion exchange chromatography conditions if there weren’t an ion pairing reagent. The column on the other hand is a plain RP. The pH is wrong and the small percentage acetonitrile in B tells me that the person that suggested these conditions have very limited knowledge of chromatography.
My suggestion to you would be as follows: Reduce the phosphate concentration in B to the same level as in A. Adjust the pH to 6.8 – 7 in both eluents. And optimize the TBA concentration by very small increments.
Best Regards

How much ACN do you think it would be ok?

[danko]

You don't need to think about the ACN right away.
Do the other optimizations first and if the retention times increase more than you would like/accept then go for some more ACN - e.g. 20%

Best Regards

[andrei1]

Thanks for the fast response!

[lmh]

hmmm, it's always difficult to deal with a literature method that isn't quite working.

(1) The low percentage of acetonitrile in B. Actually this doesn't worry me. If a method only needs 10% ACN, what's the problem? If you want to pump a gradient from 0-10% accurately, you will get a better gradient by making 10% ACN by hand and pumping 0-100% than by using 100% ACN and pumping 0-10%. The only downsides are that you don't have a big reserve of organic available for washing undesirable strongly retained things from the column, and you need to mix solutions yourself.

(2) But not all C18 columns like running in 0% organic ("phase collapse"). I have no idea about the column this paper uses.

(3) pH, yes, 4.0 isn't a very sensible place for a phosphate buffer because the relevant pKa values are about 2, 6.8 (and 12.5), so 4 is too far from both the 2 and 6.8 pKa values of phosphate for the result to be a buffer. But if you are considering changing it, you have to think about what will happen to your analytes too. ATP has pKa values at 4, 4.1 and 6.5, so changing from 4 to 7 for the sake of the phosphate will affect the ATP fairly drastically. It will be much more ionised, which will increase its interactions with an ion pair reagent, and decrease what little chance it had of interacting directly with C18. On the other hand, running a column with the pH very close to the pKa of the analyte is usually not a good idea, because it means that minor changes in pH when preparing the buffer can lead to big changes in the chromatography. Also minor variations in pH caused by temperature differences between the middle of a column and the outside of the column can lead to different migration rates, leading to very broad peaks (I've heard... obviously I haven't been measuring the pH at different points in my columns!).

(4) Why on earth the change of pH and change of concentration of buffer? I'd guess that the original author may have been reasoning that an ion-pair method is actually a hybrid of reverse phase and ion-exchange, because the analyte is attached to the pair reagent by an ion-exchange mechanism, and the pair reagent is attached to the column by a reverse phase mechanism. Therefore, the thinking might be, it doesn't matter whether I elute my analyte-ion-pair combined thing off the column by increasing organic, or elute my analyte off the ion-pair reagent by increasing saltiness (or changing pH). I'm not 100% convinced, particularly on the pH, which sort of goes the wrong way. Maybe the saltiness was supposed to elute, and the pH was a mere accident of adding the same tetrabutyl ammonium to two different concentrations of the same phosphate "buffer"? I'd need to make it up to see.

[andrei1]

hmmm, it's always difficult to deal with a literature method that isn't quite working.

(1) The low percentage of acetonitrile in B. Actually this doesn't worry me. If a method only needs 10% ACN, what's the problem? If you want to pump a gradient from 0-10% accurately, you will get a better gradient by making 10% ACN by hand and pumping 0-100% than by using 100% ACN and pumping 0-10%. The only downsides are that you don't have a big reserve of organic available for washing undesirable strongly retained things from the column, and you need to mix solutions yourself.

(2) But not all C18 columns like running in 0% organic ("phase collapse"). I have no idea about the column this paper uses.

(3) pH, yes, 4.0 isn't a very sensible place for a phosphate buffer because the relevant pKa values are about 2, 6.8 (and 12.5), so 4 is too far from both the 2 and 6.8 pKa values of phosphate for the result to be a buffer. But if you are considering changing it, you have to think about what will happen to your analytes too. ATP has pKa values at 4, 4.1 and 6.5, so changing from 4 to 7 for the sake of the phosphate will affect the ATP fairly drastically. It will be much more ionised, which will increase its interactions with an ion pair reagent, and decrease what little chance it had of interacting directly with C18. On the other hand, running a column with the pH very close to the pKa of the analyte is usually not a good idea, because it means that minor changes in pH when preparing the buffer can lead to big changes in the chromatography. Also minor variations in pH caused by temperature differences between the middle of a column and the outside of the column can lead to different migration rates, leading to very broad peaks (I've heard... obviously I haven't been measuring the pH at different points in my columns!).

(4) Why on earth the change of pH and change of concentration of buffer? I'd guess that the original author may have been reasoning that an ion-pair method is actually a hybrid of reverse phase and ion-exchange, because the analyte is attached to the pair reagent by an ion-exchange mechanism, and the pair reagent is attached to the column by a reverse phase mechanism. Therefore, the thinking might be, it doesn't matter whether I elute my analyte-ion-pair combined thing off the column by increasing organic, or elute my analyte off the ion-pair reagent by increasing saltiness (or changing pH). I'm not 100% convinced, particularly on the pH, which sort of goes the wrong way. Maybe the saltiness was supposed to elute, and the pH was a mere accident of adding the same tetrabutyl ammonium to two different concentrations of the same phosphate "buffer"? I'd need to make it up to see.

Good points!

[andrei1]

I have one more question, how changing the tetrabutylammonium hydrogen sulfate with tetrabutylammonium bisulphate would affect a method of separation?

[danko]

it's the same stuff. Just two names

[andrei1]

it's the same stuff. Just two names

Thanks!