identifying secondary peaks

[jackie1016]

hi,

i just want to ask how can i clearly identify a secondary peak in my gradient run?

i am currently doing the assay and impurity test for carbimazole tablets following the procedure as written in the british pharmacopoeia 2009.

my chromatographic comdition are as follows:

i am using waters hplc 2695

column: i have tried to use phenomenex Luna 250 x 4.6 5um, synergi FP 250 x 4.6 4 um. the BP 2009 required a column packed with base deactivated octadecylsilyl sical gel 150 mm

flow rate : 1.5 ml/min i have increased the flow rate because i also increased the length of the column..

the MPA : 5% ACN
MPB :20% ACN

time MPA MPB
0 100 0
10.0 100 0
10.1 0 100
36 0 100
36.1 100 0
46 100 0

i equilibrated the column for 1 hour on MP A
and 15 mins on MPB

i am required to inject
3 standard solutions

1. 0.0025% w/v carbimazole in 5% ACN
2. 0.0005% w/v thiamazole in 5% ACN
3. 0.01% w/v carbimazole and 0.0005% w/v thiamazole in 5% ACN

now here is my problem :

i inject the MPA first the 5% ACN for this is the one i used as a solvent of my samples...

the chromatograph looks clear to me except some baseline change at 14 mins...

when i injected the standard 1 the carbimazole standard i has no peak at the RT of thiamazole but it has a visible peak at 16 mins. the carbimazole peak eluted at around 19 mins.. is this a secondary peak? i am using my pure carbimazole standard here..

then i try to inject standard 2 the thiamazole standard which the standard for the impurity test.
it also have a clear peak at 16 mins, i really dont know where this peak is coming from. the thiamazole eluted at 5 mins..

then for my standard 3 which is for my system suitability, there are 3 peaks, the carbimazole peak ,the thiamazole peak which i think the area is 20x more the area of the standard 2 (Note : the concentration of my standard 2 is the same as this conc.) and my unknown peak at 16 mins..

can the peak at 16 mins be considered as my secondary peaks? it is realy big. both of my standard will fail the impurity test if i consider this as the secondary peak... i dont know how to justify this one..

and also my sample also have this peak at 16 mins..

please help

i have testing my sample for 3 weeks now..
i dont know if i will consider this unkown peak as a secondary peak.... or there is something wrong in the procedure that i do...


thank you so much ....

[tom jupille]

I suspect you may be misidentifying peaks. Could you post a chromatogram (preferably of the mixed standard #3) so we can see what's going on?

Instructions for linking to chromatograms are here:
viewtopic.php?f=1&t=2617

[jackie1016]

ok tom.

i will post it immediately
i'll just get my chromatograms

thanks

[jackie1016]

>here is the chromatogram for the solution 3 which contains both thiamazole and carbimazole:



>this image is for the solution that contains only thiamazole at 0.0005%w/v.



> this is my working standard solution that contains carbimazole one at 0.005% w/v



> this one is when i try to inject 0.4mg/ml of carbimazole standard
i just run at a different time so it juts eluted fast but the unknown peak is stll there, just before the peak of what i suspect as carbimazole..



and lastly, this my Mobile phase A/ solvent 5% acetonitrile..




thanks

[carls]

There seem to be several issues:
1) The run to run reproducibility of the gradient appears to be very poor.
Why do the gradient artifact peaks change retention time from run to run? 10min equilbration @ 1.5mL/min should be sufficient for a 250x4.6mm column so that's likely not an explanation.
2) The peak area for thiamazole is vastly different between solution #3 and solution #2 despite both having the same concentration. Any idea why?
3) Why is the retention time so different for the 0.4mg/mL standard? "i just run at a different time so it juts eluted fast but the unknown peak is stll there, just before the peak of what i suspect as carbimazole.." not sure what this means. Did you change the gradient program?
4) Is there an error in your description of the standards or chrmatograms? "this is my working standard solution that contains carbimazole one at 0.005% w/v" + "1. 0.0025% w/v carbimazole in 5% ACN" - is solution one 0.0025% or 0.005%?
Solution 3 contains "0.01% w/v carbimazole and 0.0005% w/v thiamazole in 5% ACN" or 2-4x times more than solution one but the peak height is close to that for solution 1

[jackie1016]

There seem to be several issues:
1) The run to run reproducibility of the gradient appears to be very poor.
Why do the gradient artifact peaks change retention time from run to run? 10min equilbration @ 1.5mL/min should be sufficient for a 250x4.6mm column so that's likely not an explanation.
2) The peak area for thiamazole is vastly different between solution #3 and solution #2 despite both having the same concentration. Any idea why?



> i really dont know why, i have equilibrated the new column for hours im now using the phenomenex Synergi Fusion-RP,..

>i have prepared the solution 2 and 3 carefully by this procedure:

thiamazole standard stock : 5 mg thiamazole to 50 ml with 5% ACN
i dilute 5 ml of the thiamazole standard stock solution to 100 ml by 5% ACN
the final conc : 0.005 mg/ml

for the solution 3: 20 mg of carbimazole in 100 ml and i take 10 ml of the solution and put in another 20 ml then i add 1 ml of the thiamazole standard stock solution.

final conc : 0.1 mg/ml carbimazole and 0.005 mg/ml thimazole,

but when i run the chromatogram, i keep getting really big are of the thiamazole in the sst.



this is how much they differ. i really dont know why.. is it because my standard have already degraded to thiamazole that's why in the sst it really has a big peak of thiamzole? and the unknown peak is really a product of my carbimazole standard?

3) Why is the retention time so different for the 0.4mg/mL standard? "i just run at a different time so it juts eluted fast but the unknown peak is stll there, just before the peak of what i suspect as carbimazole.." not sure what this means. Did you change the gradient program?

> about this one, yes, i have change the gradient program when i run this 1 because i can get the resolution required.. but there's still the unknown peak...


4) Is there an error in your description of the standards or chrmatograms? "this is my working standard solution that contains carbimazole one at 0.005% w/v" + "1. 0.0025% w/v carbimazole in 5% ACN" - is solution one 0.0025% or 0.005%?
Solution 3 contains "0.01% w/v carbimazole and 0.0005% w/v thiamazole in 5% ACN" or 2-4x times more than solution one but the peak height is close to that for solution 1

yes! im sorry 1 in not included.. this is my working standard solution that contains carbimazole at 0.005% w/v..

and this is the chromatogram of my solution 1: 0.0025% w/v carbimazole




the carbimazole in solution 3 is supposed to be 4x larger than this one...


im a really confused..

i dont know what i keep on doing wrong..

[bisnettrj2]

Can you confirm a few details?

1. If you run several injections of mobile phase A, do they all look the same? The same goes for replicate injections of each individual analyte - do multiple injections look alike? Same peaks, same retention times, etc?
2. How "pure" is your carbimazole? How do you know it is pure?
3. If you inject just your carbimazole 0.2mg/mL stock standard, do you see a thiamazole peak? Do you see the unknown peak at 16 minutes? If you make several injections from the same vial of the 0.2mg/mL standard over an extended period of time, do you see an increase in the unknown at 16 minutes and the thiamazole peak, and a decrease in carbimazole?
4. Are you filtering your standards at all?
5. You gave your gradient conditions, and then made this statement:

i equilibrated the column for 1 hour on MP A and 15 mins on MPB

and this one:

i really dont know why, i have equilibrated the new column for hours

What do you mean by "equilibrated"? Are you running multiple blank injections, or just letting the column run at 100%A?

EDIT - just an observation, but to me, only the 0.0025% w/v carbimazole injection and the mobile phase A injection look similar - all the peaks from the mobile phase injection are in the carbimazole injection, and the carbimazole injection has the two extra peaks. All the other injections have something else going on around 14-16 minutes.

As for your method, if you can change it, I would probably opt for a short (maybe 2 minute?) hold at 100%A, then do a linear gradient to 100%B over 10 minutes or so, hold for maybe 5 minutes at 100%B, then re-equilibrate at 100%A for 10 minutes before the next injection. Personally, I'd like a stronger B mobile phase (maybe 80% acetonitrile, instead of your 20% acetonitrile) to make sure my column is washed before each subsequent run. To me it looks like you have some late-eluting crud coming out at 30-33 minutes, when you've been at 100% B (20% ACN) for 22 minutes, but it looks like your analytes are done eluting by about 19 minutes. That's a lot of wasted time in your run.

[HPLCaddict]

Apart from all the tipps already presented:
Are you bound to the BP or could you switch to EP instead? The EP monography contains an isocratic method for related substances determinations which, I'd suppose, would cause MUCH less trouble...there's a whole lot of junk coming out when your step gradient hits the column.

[tom jupille]

This is a good example of why I *hate* "step gradients". Aside from all of the other issues that have been mentioned, there can be a wide variability in when the step actually gets to the column (depending on the dwell volume of the system) and in how steep the step actually is (depending on the mixing characteristics).

By changing the column length and flow rate, you have already modified the BP method, so you will have to revalidate in any case. HPLCAddict's suggestion about adopting the EP (isocratic) method sounds like a good bet. If that's impractical, then you really need to go back a few steps. If this were my problem, I'd do something very similar to bisnettrj2's suggestion (without the initial isocratic hold): run your mixed standard with a linear gradient starting with your existing "A" solvent (5% ACN) and running the ACN concentration up to at least 80%. 30 minutes should be about right (incidentally, you could cut the time in half with a 150 mm column at 2 mL/min). As a "rule of thumb", if the first and last peak are less than about 10 minutes apart on that 30 minute gradient, then you should be able to fit them into a single isocratic run. A reasonable first guess for %B in that case would be whatever the %B is at the halfway point between the first and last peaks, but that would probably require some tweaking.

If the first and last peaks are more than about 10 minutes apart, you will have to use a gradient. In my experience, linear gradients are *much* better behaved than step gradients because they avoid the abrupt dis-equilibration of the system that's causing all of your baseline grief.

[jackie1016]

@ HPLCaddict : i can use other Pharmacopeia including the EP but as of now my resources are limited., we only have the USP and BP. the USP doesn't have carbimazole tablets monograph . i have search the internet for the carbimazole EP monograph i can only find the monograph for the carbimazole API and it uses the isocratic method. i would really really want to try the EP method for the impurity test.... thanks for the suggestion i will try to find the EP monograph for carbimazole tablets

@ tom : hi, me too i really dont want to run in gradient and actually this only my 2 nd tym to run in gradient that why it is very difficult for me to identify which is which and to adjust the concentration of the mobile phase i have adjusted the chromatographic conditions according to what the BP allows., we can adjust the column for as much as 70% of the column length and the flow rate at 50% so i think it is still allowed. but i still have so many problems with the method i'll try to find first the EP monograph and if i cant find one i will try all your suggestions and validate the whole process... thank you so much!

thanks everyone!!!

[HPLCaddict]

I'm afraid EP only contains API methods, no tablets .

[bisnettrj2]

Since I'm relatively sure many of us don't have access to the British Pharmacoepia, can you post the original method's column and instrument parameters (flow rate, column temperature, gradient parameters, detection wavelength, etc), along with any changes you've made to the method to accomodate your different columns? Perhaps there is a simple error of implementation that has happened, and we may be able to identify the problem with all of the information. Also, is there an example provided in the Pharmacoepia of the chromatography developed in the original method? It might be useful to compare the original work with what you're seeing.

[jackie1016]

hello everyone!

i manage to find another pharmacopeia from a friend she said that this one is from the EP. the procedure is almost the same as the the BP 2009 which i used on the previous run.. the only difference is that the mobile phase A is 95% ACN (BP 2009 uses 5%ACN as mobile phase A) mobile phase b is still 20% ACN. the solvent that is used in this new procedure is 20% ACN (BP 2009 uses 5% ACN)

i have tried this procedure today, this is the original method coming from the new pharmacopeia:

COLUMN: stainless steel column (15 cm x 3.9mm) packed with based deactivated octadecylsilyl silaca gel for chromatography (5um)
> linear gradient elution using the following mobile phase
> mobile phase A : 95% acetonitrile
mobile phase B : 20% acetonitrile
>flow rate of 1.0 ml/min
>detection: 254 nm
>temperature: ambient
>injection volume: 20 ul

Time Mobile Phase A Mobile Phase B

0 100 0

4.5 100 0

4.6 0 100

30 0 100

30.1 100 0

40 100 0


this is the one that i used in my run:

COLUMN: synergi fusion FP 250 x 4.6 mm, 4 um
> linear gradient elution using the following mobile phase
> mobile phase A : 95% acetonitrile
mobile phase B : 20% acetonitrile
>flow rate of 1.5 ml/min
>detection: 254 nm
>temperature: ambient
>injection volume: 20 ul

Time Mobile Phase A Mobile Phase B

0 100 0

10 100 0

10.1 0 100

36 0 100

36.1 100 0

46 100 0

this procedure manage to eliminate the unknown peak from the previous run, but there is one problem. the peak form thiamazole was not a clear peak.



then, it tried to use the same column with this gradient program (same as the pharmacopeia running at 1.5 ml/min

Time Mobile Phase A Mobile Phase B

0 100 0

4.5 100 0

4.6 0 100

30 0 100

30.1 100 0

40 100 0

then i got this



i cant seem to get the peak of thiamazole.

i tried to run on isocratic at 1.0 ml/min using 95% ACN
i inject thiamazole standard 0.005mg/ml in 20% acn 20 ul

i got this,





i dont know why i cant get the thiamazole peak,

can anyone help me about this..

thanks you so much..

[jackie1016]

[quote="bisnettrj2"] is there an example provided in the Pharmacoepia of the chromatography developed in the original method?
unfortunately there is none.

[jackie1016]

this the closest shape of thiamazole that i can get...