Is there a difference between Waters & Zorbax columns?

[Dr-mostafa_omar]

On the analysis of mesna raw material for related substances by an HPLC method from BP2007, i injected the test solution on a Zorbax C18 4.6x250mm column and i obtained following chromatogram:



as you see the peak three fused peaks are very strange and were unseperated by all changes i did (new mobile phase, decreasing the flow rate). the same chromatogram was given when i used another zorbax column (i did this to exclude the possibility of column problem)
But when i injected the same sample but using a Waters bondapak C18 column, this is what i got:



i could not find any possible reason for this except that there is a difference between zorbax and Waters columns that caused this.

Could anyone please tell me what would be the cause of this?

[grzesiek]

You shouldn't assume that all C18 columns are the same cos they're not

If the type of column was not mentioned in monograph you should contact person responsible for the monograph and ask exactly what type of column was used (manufacturer, type), I know it's not worth to try to find the one - waste of time

[Dr-mostafa_omar]

The problem is that if I use the areas in the chromatogram obtained from the Waters column, the result of the test will be conforming to specifications while with the Zorbax column, due to the appearance of two peaks (one before the mesna peak and one after it) the result will be that both impurity E and inivdual other impurities will not be conforming to specifications. this is the dilemma i'm in.

[krickos]

Hi

Well for me it is pretty clear that the Waters column should be used.

The BP monograph refers to the Ph Eur monograph, and in the history database the following description of used column is: Column : l = 0.25 m, diam. = 4.6 mm, 10 µm, C 18 Bondapack, Waters

Further more the zorbax column seems not to comply with the SST demand for resoution (System suitability Reference solution (e): esolution: minimum 3.0 between the peaks due to mesna and impurity C.).

Why this differance exist, others a properbly more suiteble to answers but as previously indicated do not expect all C18 columns to perform the same and given the nature/structure of Mesna (Sodium 2 sulphanyl ethanesulphonate) and the method one might not be entirely surprised.

[grzesiek]

"Well for me it is pretty clear that the Waters column should be used."

well it's not so clear to me... that this method should be used

first - using bondapack column i do not see impurity c at all so you cannot say that resolution is minimum 3.0
second - if synthesis route was changed there can be new impurities and so selectivity/specificity should be evaluated, so you should show that mesna peak is pure

[danko]

Guys,

How can one say: Just use the column that shows the greatest purity and be happy?
The “mainâ€

[krickos]

Hi Grzesiek

first - using bondapack column i do not see impurity c at all so you cannot say that resolution is minimum 3.0

Well I said it "seems" and secondly, the attached chromatogram does show the test solution, NOT the SST solution (Reference solution (e) so so impurity C might not be present in the test solution/sample.

second - if synthesis route was changed there can be new impurities and so selectivity/specificity should be evaluated, so you should show that mesna peak is pure

I agree, that is always something that have to be considered, but that was not the question or?


EDIT:
ohh now I see your point, however it is not clear how the SST looked at both columns

[grzesiek]

hi krickos

that was a general comment, I (wrongly?) assumed that this was solution which contained impurities to check if the method is good (suitable)

to danko

that's exactly the point - and purity can be evaluated different ways, for example ortogonal seperation can show that impurity profile is different which is the case here

to Dr-mostafa_omar

wraping it up - you are responsible for the analysis, and the problem is that you know that there are more impuritites than you can see using bondapack so if i were you i would not say that this analysis is ok, and i have seen it many times that pharmacopeial methods usually need changing or replacing

[Dr-mostafa_omar]

First of all, i would like to thank you all for these valuable contributions.
Secondly, i would like to inform you that upon the injection of the SST (reference solution (e)) both columns showed acceptable resolution between the mesna and impurity C and surprisingly the resolution in case of the Zorbax column was 4.81 while it was 3.67 in case of the Waters column.
As for me, i almost reached a conclusion that in this specific case the zorbax columns are more sensitive (may due to the smaller particle size of the stationary phase) but the three peaks appearing in the zorbax were unseparatable from each other by any means...when i decreased the flow rate to 0.5ml/min instead of 1.0ml/min i expected a better resolution between them but nothing changed at all...the same shape and even at the same heights...more surprisingly that when i increased the flow to 1.5ml/min also no change to the shape( if these we different compounds their resolution would have decreased).

[danko]

but the three peaks appearing in the zorbax were unseparatable from each other by any means...when i decreased the flow rate to 0.5ml/min instead of 1.0ml/min i expected a better resolution between them but nothing changed at all...the same shape and even at the same heights...more surprisingly that when i increased the flow to 1.5ml/min also no change to the shape( if these we different compounds their resolution would have decreased).

They are impurities alright – no doubt about it.
Incidentally you might like to try other optimization parameters than flow-rate, which seldom has dramatic influence on the separation (within reasonable limits).
Try to play around with pH, column temperature and organic modifier!

Best Regards

[Vlad Orlovsky]

I would inject something simple with different retentions (parabens, benzonitrile) or compounds from test mixture. You need to see if merged peaks is a result of column packing (which I don't believe).
What is your flow rate? Your retention is not that great and there is a chance that you have impurities which co-elute. You would suggest to run very low organic to see if peaks at the beginning of chromatograms separate.

[krickos]

They are impurities alright – no doubt about it.
Incidentally you might like to try other optimization parameters than flow-rate, which seldom has dramatic influence on the separation (within reasonable limits).
Try to play around with pH, column temperature and organic modifier!

Hi

first an apology to grzesiek, due to my perhaps rash wording.

Otherwise, are we rushing ahead here or?

Omar, if you overlay the test solution and SST chromatograms with the Waters column and the test solution do not completly overlap the main peak in the test solution where impurity C is not present, can one really disregard the column factor?

In other words if the "standard colum"/method shows no signs of impurity C in the test solution and the SST shows enough resolution, is additional impurities really the first "thought" or is "column issues" more likely when using zorbax column?

Might add that the SST solution is typically more diluted than the test solution, also naming of peaks seems weird in the zorbax chromatogram (Mesna a smaller riding peak?).

Can you inject Mesna from another source (ie BP, Ph Eur not your source) at the test solution concentration and see if the same/similar thing happens?

I am not ruling out other impurities in Omars "source"/sample but an open mind might be needed.

[grzesiek]

yes, i think i am

but having seen so many pharmaocpoeial methods don't work with new synthesis route i'm pretty sure that's the case

of course it would be nice to see SST as i dont have access to BP2007

[oscarBAL]

Dr-mostafa_omar. escuseme for the question but both columns are 10um particle size? I really do not know what BP2007 is so I could be speaking with out any base but could not you use 5 um columns? as far as I know for USP 50% decrease in particle size is allowed.

[Dr-mostafa_omar]

In answer for your comments, i would like to clarify the following:

Your retention is not that great and there is a chance that you have impurities which co-elute. You would suggest to run very low organic to see if peaks at the beginning of chromatograms separate.

I injected the diluent as a blank and no peaks appeared that could interfere with the peaks appearing in the test chromatogram.

Might add that the SST solution is typically more diluted than the test solution, also naming of peaks seems weird in the zorbax chromatogram (Mesna a smaller riding peak?).

Can you inject Mesna from another source (ie BP, Ph Eur not your source) at the test solution concentration and see if the same/similar thing happens?

i named the peak of Mesna as the smaller one because this is the same retention time of mesna peak in the SST. I prepared a test solution from Mesna USP standard (surprisingly there is no Council of europe standard for Mesna, they only supply standards for impurities C & D), the chromatogram of the test solution from USP standard showed the SAME tri-peaked shape.

Dr-mostafa_omar. escuseme for the question but both columns are 10um particle size? I really do not know what BP2007 is so I could be speaking with out any base but could not you use 5 um columns? as far as I know for USP 50% decrease in particle size is allowed.

Not bothe columsn have the same particle size. Waters column's particle size is 10u while zorbax is 5u.

it would be nice to see SST

This is the SST for the Zorbax column


and this is the SST when using the Waters column