Hi,

I'm developing a new HPLC method for a new compound that have an amine group.

I'm working with a Kromasil C-18 (15x4.6) with a buffer (100 mM phosphate and 2 ml of Trhietylamine (TEA)) adjusted a pH 3.0 with phosphoric acid.

Mobil phase 95 buffer and 5 % acetonitrile.

I have good resolution with 6 potential impurities but the tailing factor for the main peak is very poor (2.7 !!!!)

I've changed the TEA for Diethylamine, Octylamine, but nothing has improved the method.

A change in the pH is not a good solution due to 2 potential impurities coeluated....

Does anyone has any good idea to improve the tailing factor

Many thanks in advance and sorry for my english expression

Basil

Add a gradient flow... lower injection vol...

Dear Basil
1)What is your injection volume?
Try to reduce it.

2)What the %C of your Kromasil column?
Try to use another C-18 column with %C at 23-30%

3) In Amines analysis the ideal TEA concentration is 30 mM, so use 4-5 mL of TEA on your mobile phase, with same pH, ok?!

4) What is your retenton time? Maybe you reduce your rate (mL/min)?


Happyness,


deST

There are probably better choices than Kromasil, like an ACE C-18 (mac-mod), Inertsil-ODS3, or Luna. Check out the C of A for NIST SRM 870, and also mac-mod's comparison guide to HPLC columns, or the Alltech column selection guide. Lots of good info - Kromasil is mentioned in all three. Also, a polar-embedded or polar end-cap might be more appropriate for such a high aqueous mobile phase. You might be walking on the de-wetting tightrope. Keep us informed. Agree that injection volume can often have drastic effect on tailing if not optimized. Do you have TEA (etc.) in your diluent?

Hi Anon 125

I agree with Anon 0206.

deST

Hi Anonym 0125;

1)Are there your "new" compound other polar groups, as acid group?

2) You can optimaze your tailing factor:

2.1)Using TEA at 30-40 mM in your mobile phase;

2.2)Improving the flow;

2.3) Improving the column temperature;

2.4) Reduce the internal volume of your system.


Happyness,

deST

Basil, please tell us also how much you are injecting (or what the AUFS of the main peak is, if you are using a UV detector). It is not impossible that the issue is simply overload.

First: Thanks for all your comments !!

Sample concentration: 2 mg/ml (diluted with fase A (buffer))
Injection volume: 10 µl (I have tried 5 µl at a 4 mg/ml but I have not better tailing factor)
Retention time for the main peak : 7 min
Time for all related substances: 12 minutes
Flow: 1.0 ml/min
Column Oven: 40 ºC

Chemcial sructure: The compound is an Hydrocloride. There is not acidic groups in the chemical. There are another amines but all are methylated or ethylated.

I've also tryied with Hypersil C-18 BDS 3µ (100x 4.6) and Symmetry C-8 (25cm x 4.6). All with no better results.

I'm waitting for a Zorbax-SB C-18 that should be supplied last week but .....

Maybe if we change the amine option (TEA) to use in the mobil phase to improve the chromatographic system ?

Thanks again

Basil

Sorry Uwe,

Answering your comments !!

Peak height: 800 mAU
Width (min): 0.42

If we try to reduce the sample amount the peak aspect is the same. I'm sure that there is not an overload question.

thanks agian

Basil

ANON 0206 here,
2 mg/mL is a very high conc., in my experience, esp. w/ a 10 uL inj volume. Are you sure you can't drop the conc? An 800 mV peak should give you enough sensitivity for related compound peaks to play with. No suprise that Symmetry column tailed like crazy. Expect that Zorbax will be worse than Symmetry. NIST SRM 870 certificate predicts that the Hypersil would also tail much more than other columns that have been suggested. You may keep throwing columns at this or check the available info on tailing and pick a more suitable column.
Here's a shot in the dark: We had a 40C method that when run on systems that pre-heat mobile phase effectively (Agilent 1100) the tailing factor went up relative to non-preheating w/ same system. Yet the TAIL of the peak didn't change at all, only the sharpness of the FRONT. In this case, the higher-tailing peak was actually much sharper, yet that tailing factor number was worse! Could this be your case? Unfortunately, we had to revise method to get poorer peak performance just to get the "better" (meaningless in this case) tailing factor.

Hi ANON 0206 !!!!! Thanks for your comments !!

What is NIST SRM 870 ???. Is it possible to check this predictions with Internet (what WWW ??)
mac-mod's comparison guide to HPLC columns ?? What's that ??

Related with amount or concentration again... I should detect an impurity with a 0.05 % related area..(ICH methods !)

If I had better Tailing Factor I would try to reduce sample amount.

Just to check I have tried 0.5 mg/ml (injecting 10 µl). The peak Area is 8000 [mAUxs] with an Agilent 1100. Symmetry 0.51 and USP Tailing 1.8 Efficiency plates (tangent method 5000)(better but not perfect to be validated and used in CQ Lab !!!)

Basil

Hi Basil, i work with a kromasil and also with basic compounds and i have no problems of tailing, so, the question is , are you adding TEA to organic and aquose phase , or only to aquose phase?

Be careful with pH and TEA...

http://www.alltechweb.com/productinfo/Technical/Electroblotter/hplcchrom/SelGde1.asp for Alltech column comparison- Lots of good info

Do a google search for "nist srm 870" and you should find the Certificate of Analysis for the NIST column performance test mixture, which they have used on many of the most popular columns. Nice and unbiased, since this is not a manufacturer paper. (ACE C-18 best by far)

I don't know if Mac-Mod has thier column comparison online, but search their site and at least you can request one. I'm sure they'd love to send it since the "ACE C18" they sell performs extremely well on their tests.

Now, I know these each use only one set of conditions and test probes etc. , but they are a great place to start. No, I don't work for Mac-Mod, but I've been pleased w/ the ACE so far.

Hi Anon 0748

I'm only adding TEa to the buffer (aquose phase) but as I work in an isocratic mode I would not expect any special problem.

Problems with TEA and pH ?

Anon0813:

Thanks. I will try the www.

Only one point:

ACE C-18 is possible to find in Europe ? This column was not related in Pharmacopeial Forum Volume 26, Number 1 page 273 (Feb 2000) "HPLC packings used in the USP-NF"


Basil

Basil,
The ACE is indeed a European column. Here goes:
Advanced Chromatography Technologies (ACT)
Aberdeen Science and Technology Park
Balgownie Technology Centre
Bridge of Don, Aberdeen
AB22 8GW, Scotland
Tel: +44 (0) 1224 704554
Fax: +44 (0) 1224 826622
Hope that makes sense to you- It doesn't to me!

Hi Basil,

I would not overlook Uwe's comment about overload. I have seen hydrophillic amines overload a column a very low on column amounts, much less than you are injecting.

If you dilute your sample by half and inject the same amount and see a greater retention time then your column is overloaded and no amount of additives is going to improve your peak shape. Just changing brand of column will probably not help an overloaded separation either.

If your column is overloaded I would try the separation on a base stable column like the Zorbax Extend or the Water's Xterra at a pH>8.0.

I think the organic fraction of binary mobile phases is adsorbed to the C18. Then your analyte first has to partition into this adsorbed layer of organic then it can adsorb to the C18. Anyway, acetonitrile is an aprotic solvent and is poor is solvating charge. By shifting to a high pH you can protonate the amines and your analyte will be much more soluble in ACN. Of course you will then have to redevelop for your impurities. At a high pH your analyte will be retained longer and might need a higher %organic to be eluted.

Good luck and let us know how it turns out.

P.S. what is the plate count for your current separation?

A general tailing question, maybe for Uwe- Have you seen a trend that more-retentve HPLC columns can withstand more sample loading before overload occurs, all things being equal (they never are, I know)

I am convinced that this is column overload. You need to inject at least 10 times less, even for a neutral compound. For a charged basic compound, this is the nature of the beast and you can buy the best column in the world, and it won't get much better.
If you can't inject that much less due to sensitivity of the impurities, my recommendation would be to just live with it. If my suspicion is correct, your best alternative would be to redevelop the method under conditions where your compound is not charged. Since it is base, this may mean at least at neutral or even better at alkaline pH. This is work, and I would think about if this is worth the effort.

Another option is to switch to another retention mode. Since all of the solutes are basic, it is likely that they can be separated via SCX without symmetry problems. Generally, potassium or better still cesium based eluent systems provide good chromatographic efficiency and symmetry.

Thanks to all.

Today is the last day just before holidays,I will try to rest and get new forces to work again on this product.

When I get the best solutions I will tell you. I will consider Uwe and Tom coments about sample amount

Bye

Basil