Retention of small amine

[Thunderbunny]

Hi all,

We are looking for a method to determine levels of 1-methyl-1H-pyrazol-3-amine (CAS 1904-31-0) in part of a pharmaceutical synthesis. Ideally we'd like to look at this, the preceding and next compounds in the route (MW's in range 400-500), as well as any impurities if possible. The compound is not retained on any reversed phase column we've tried - we've used a 10 column screen of differing selectivities. A Primesep C column gave no retention. A ZIC-HILIC gave retention of k'~1, although we had to use 97% THF to achieve this (10x4.6 5um, 0.5ml/min). Unfortunately the preceding compound gives zero retention on this system and the product has slight retention (k' probably <0.1).

Any ideas welcome.

Thanks

(the pyrazole GCs well on a CAM column, but the other compounds of interest will never fly on GC)

[Patrik Appelblad]

Dear Thunderbunny,

Could you please provide us with somewhat more experimental details, like matrix, sample diluent, injection volume, etc.

[SIELC_Tech]

Thunderbunny,

What was your mobile phase for Primesep C column? Below pH-3 column is not having any ion-exchange properties. You might need strongre column (Primesep 100)

Here are few methods for simple amines with K' well over 1. You can use any buffer or organic acid (ammonium formate, acetate, sulphuric acid, TFA, etc.)

http://www.sielc.com/compound_005.html (simple amines)

http://www.sielc.com/compound_137.html (simple polyamines)

http://www.sielc.com/compound_079.html (pyridine)

http://www.sielc.com/compound_104.html (ethanolamines)

These applications show general applicability of Primesep to retention of amines (no ion-pairing reagent). You compounds are retained by combination of reverse phase and ion-exchange mechanisms.

What is pKa value of you amne?

Regards,

Vlad

[SIELC_Tech]

Here is the link for tuning Primesep C column:

http://www.sielc.com/pdf/SIELC_PrimesepC_Column.pdf

You have enhanced retention at pH-4-7

[SIELC_Tech]

Thunderbunny,

I tried to find commercial source for your 1-methyl-1H-pyrazol-3-amine and Aldrich/ Sigma don't sell it. Do you know where I can buy it? If we can buy it we will develop method for you.

Also we are conducting seminars (4 days) in UK starting next week and may be you can talk to our technical guru in person about your compound and Primesep C column (or other). Please contact Hichrom limited for information about seminars (may be we already presenting in your company);

Hichrom
http://www.hichrom.co.uk
1 The Markham Centre
Station Road
Theale
Berkshire
RG7 4PE
UK
Phone +44 (0) 118 930 3660
Fax +44 (0) 118 932 3484
Email: sielc@hichrom.co.uk

[Thunderbunny]

Thanks for your replies -

Patrik - so far we have just been running the amine and the other two compounds individually at a concentration of ~0.2mg/ml made up in the mobile phase (-buffer) so either MeCN/H2O or THF/H2O. Injection volume has been 5uL.


Vlad - the only place we can find that stocks the material is Maybridge (UK) (www.maybridge.com). However we have plenty of the material. We do have a seminar arranged for next Wednesday at 12, so can discuss this application more then [including safety info]. And I'll probably be introduced as something other than Thunderbunny.

Martin

[SIELC_Tech]

What was pH of your mobile phase fro Primesep C runs? Do you have any data on pKa?

Vlad

P.S. If you would like to take this discussion out from CF my direct email is
vlad.orlovsky@sielc.com

[JA]

I hope this problem continues to be discussed on the forums, it looks quite interesting

I'm surprised a molecule that polar wouldn't retain by HILIC, we've seen more lipophilic species than that stick on a silica column simply with H2O or formate buffer with ACN. We don't have any of the ZIC-HILIC type in-house so I cannot make a comparison.

I wonder if it could prove difficult to find HILIC conditions to retain that little pyrazole and your bulkier API anyway, so perhaps ion exchange is an approach to consider assuming your API has basic functionality..

[SIELC_Tech]

Thunderbunny,

I found at least 5 companies making your compound but they are all outside US (Europe, Asia, Russia). I am going to order 3-amino-pyrazole which will behave the same way as 1-methyl-1H-pyrazol-3-amine. Hopefully we can retain it on one of Primesep columns. You can also try to rerun it with ACN-water=9/95 with 10 mmol of ammonium formate pH-5 or 6. You can also try 0.1% of acetic acid instead of ammonium formate. The goal is to achieve retention, we can work on a peak shape later.

[Uwe Neue]

I agree with JA. I do not see why this compound would not stick easily to a HILIC silica column, due to the dual effect of HILIC and the interaction of the acidic silanols with the basic analyte. It is worth giving it a shot. Use a silica HILIC column, equilibrate in 95% acetonitrile and 5% water, then run a gradient to 50% acetonitrile and 50% water. If you do not have a silica HILIC column, I would try in your shoes an old silica column that I have lying around. If you get retention, it is OK. If the peak shape is not good, you may need to buy a silica HILIC column.

[Thunderbunny]

We only tried the Primesep C column with the initial conditions suggested in Method Development Note 1 - 60% MeCn with 0.5% TFA. I've only got a predicted (ACD Labs) pKa : 4.04.

We're also waiting on an Atlantis Silica HILIC and Atlantis dC18 column - though I've heard that these don't have great inter-batch reproducibility. Can anyone confirm/refute this?

Thanks
Martin

[Uwe Neue]

The standard deviation of all compounds in the batch test chromatogram is under 2%. This does not look bad to me...

[JA]

Oh my god where did you hear that?

[SIELC_Tech]

Ja,

Our spec is 5% because we have more complicated interactions. We are testing batches on 6-9 various compounds (benzylamine, benzoic acid, benzonitrile, dextromethorphan, ephedrine, doxylamine, guafenesin and guiacol). After batch pases the test column are tested only on three compounds. These test ensure reproducibility in a wide range of compounds. Of cause you might be able to find compound which produces higher variation. In this case manufacturers reserve lots for particular customers.

[HW Mueller]

Ah, Sielc_Tech, now you gave me a different way to word what I formerly described via not being able to get optimum single type of interaction with mixed up mode:
You would say that you can not switch off this or that mode completely?