montelukast

[zompa]

Hi everybody,

I'm developing a hplc method for Montelukast. Can anybody give me some advice? At the moment I'm getting a nice impurity profile by using a luna hexyl-phenyl column and a gradient program. Yet the main peak comes out at a very high Rt. I'd like to develop a shorter run time method with a good resolution. Maybe someone can suggest a different stationary phase.
Any advice is very much appreciated.
Thanks a lot.

[tom jupille]

It would help to have more details (size of column, packing particle size, flow rate, gradient time, etc.)

[zompa]

The mobile phase I'm trying at the moment is
A) buffer:CH3CN (80:20)
B) CH3CN:buffer (80:20).
Buffer is 0.025N NaH2PO4 pH= 3.7.

I'm working out the gradient program in order to separate some impurities.
Anyway, starting from 50:50 and going up to 90% B in 50 something minutes.

Thanks for any advice.

[tom jupille]

That's a start. Now, how about the column dimensions and packing particle size?

And . . . what is the retention time of the first peak of interest and the last peak of interest?

[zompa]

yes, tom

the column is a hexyl-phenyl, 25 cm x 4.6, 5u.
Montelukast rt is about 45 mins. I like the profile but I have been requested to shorten the run length.
Maybe the stationary phase can be changed in order fulfil the request.
What you think.
Thanks for your help.
I also put on a topic on Ibandronate about 195 wavelength. Could you please have a look at it and let me know what you think?
Thanks again.

[serap g]

Hi Zompa

We studied montelukast..
Mobile phase 0.025M sodium dihydrogene phosphate pH:3.7 / Acetonitrile (20/80)
Column C18 150mm x 4.6 mm 5 um

Column temperature 40 degree centigrade and 345nm...

[zompa]

Thank you serap guven,

Really very kind of you...
I tried isocratic indeed but actually there are some impurities I couldn't separate from the main peak. This is why I started to study a gradient program and moved towards a different packing from c18.
I haven't studied the influence of temperature. Is it a very critic parameter in your opinion? The wavelength I have been detecting so far is 225 but will try 345 soon. If you have any more suggestions they will be highly appreciated.

Thanks again.

[tom jupille]

Several possibilities:

1. Depending on the retention time of your early peaks, you may be able to start at a higher %B (shorten the gradient time to keep the same steepness).

2. If you are already as high as you can go in initial %B, check with Phenomenex to see if the same packing is available in a 100-mm column with 3 or 3.5 μm packing. Keep the same flow rate but scale the gradient time to the column length.

3. Both gradient steepness and temperature will affect the spacing of the peaks. It is not unusual to see "sweet spots" (combinations of gradient time and temperature that give excellent results) with very fast run times. Most of the commercially available chromatography modeling programs (DryLab, ChromSword, ACD, etc.) have the ability to take experimental results from a 2 x 2 experimental matrix (2 gradient times x 2 temperatures) and map out the effects.

I no longer have any business connection to the DryLab program, but if you want to pursue that, you can contact Imre Molnar for support:
http://www.molnar-institut.com/

[zompa]

Thanks a lot Tom.

Actually I have tried to start with a higher %B but this has caused too impurities at around 15 mins to coalesce. And I don't want it to happen.
I'll talk to phenomenex about the column. This is the brand I use more.
Did you have a look at the Ibandronate ? I'd like to know your opinion if it doesn't bother you.
Thanks a lot for replying.

Zompa.

[HW Mueller]

Just wonder how an analyte of this type can be resolved better by a gradient rather than by an isocratic method?

[zompa]

actually it works..
there are some impurities that aren't resolved in isocratic from the main peak...
someone can help about Ibandronate detection at 195nm..
thank for being interested..

[tom jupille]

Just wonder how an analyte of this type can be resolved better by a gradient rather than by an isocratic method?

Usually it can't. In most cases, a gradient is used to accommodate a range of k' values that is too large to fit into a single isocratic run. I think that't the situation here.

[tom jupille]

Did you have a look at the Ibandronate ? I'd like to know your opinion if it doesn't bother you.

I've been following the thread, but I haven't posted because I really can't add anything to what's been said.