fast screening

[sergioanis]

hi everybody,

I'd like to know what people is using for fast screening methods with HP1100 or simmilar instrument.
I was asked to develop a general and fast method and so far I'm considering:

column: Onyx Monolithic C18, 3.0 x 100 mm
flow: 1.8 ml/min
solvent A: water (0.1% HCOOH)
solvent B: ACN (0.1% HCOOH)
gradient: 5 to 100% B in 3 min
pre-eq and wash: 0.5 min each

I'd appreciate if you can share your experiences.

thanks,
Sergio

[Molly]

Hi, Sergio:

Here is just my opinion on your method. Monolithic column is good for very fast separations (1-2 min) at high flow rate (5ml/min) and for relative simple separations. The peak shape is somewhat worse than typical packed columns and I found the narrow bore monolithic column tends to have fronting in peak shape. If you want to have more plates/peak capacity, I would suggest you try a typical C18 column (like Agilent SB-C18, Waters Symmetryshield RP18, or Luna C18) at 50X4.6 (3um) at 2ml/min and run the same 3min gradient. If you can modify your 1100 to do high throughput separation(you can buy kit from agilent so that extra column volume is minimized), then the new sub 2um column is a good choice for fast screening as well. If your method does not need to be MS compatible, 0.1% H3PO4 will provide better baseline and sensitivity at low detection wavelength. For MS friendly method, I saw more people use TFA than formic acid, but I do not know why. The re-equ volume has to be at least 1 column volume+ twice dwell volume, so 0.5min proposed might not be long enough. For a typical 5cm column, void volume is 0.5ml, and for Agilent 1100 with quaternary pump, dwell colume is 1.1mL (smaller for binary pump), so the total re-equ volume need to be 0.5+2X 1.1=2.7mL, at 2ml/min, you need to flush at lease 1.3min to fully equlibrate your column.

Thanks, Molly

[Koen Hollebekkers]

Without taking about peak shapes, it is possible to get high plate counts at high flow rates.

It has been shown by kinetic plots, that with UPLC you will get more plates than with monoliths. In addition, monoliths will be more effective when you couple them to get long columns with high plate numbers and low backpressure.

The thing you have to take in mind when validating monolithic methods, is that Merck has the patent of this product. If Merck decides to discontinue, you will have a problem.

Good luck!

[sergioanis]

hi Molly and Koen,

thanks for your replay. Let me explain a little more what my situation is:

I have only open access instruments that are used to follow reaction progression. So, you need fast separation, MS compatible method, you don't see many peaks (if you do, you have to rethink your synthesis), you don't care much about the shape of the peaks, you do need good reproducibility

I inherit one system that use Luna C18 5u, 2.0x30 mm. It has several gradients to cover for most of the possibilities. But, this system need to be improved.

I set up another system with the Onyx Monolithic 3.0x100 mm (trying to improve the resolution and time). For this diameter you should use 2.1 ml/min instead of 5. Since I have a MS (with ESI) I need to use a splitter. The MS and splitter create too much backpressure for the detector cell, for that reason I use 1.8 ml/min.

At 1.8 ml/min the solvent peak comes at 0.35 min. So, 0.5 min equilibration + 0.2 min head time (injection + wash syringe) gives more that 1 column volume. Consecutives runs have shown very good reproducibility.

It is true, 2 coupled monolithic will be better but that will increase the running time.

Phenomenex got a license from Merk to make monolithic columns. They sell it with the brand name Onyx. For what I know, they are developing more monolithic columns (different sizes and diameters). So, availability of the column is not a concern.

I do not know why people would use TFA, for what I understand it is not good for ESI signal.

UPLC is out of consideration because I can not buy new instruments.

I have another instrument to put together. For this one I am thinking on the new sub 2um from Phenomenex. I read the brochure and I am waiting for them to release the technical notes to know what changes I have to make to the LC (the vendor said that he will send it to me)

thanks again,
Sergio

[Bryan Evans]

Here is an example of a high-throughput LC-MS application using our Cadenza CD-C18 column:

http://www.imtakt.com/TecInfo/TI085E.pdf

[Klaus I.]

I think you should use a C18 1,8 µm Column with 30 x 4.6 mm and a Flowrate of 4 mL/min for your gradient.
Replace the 1100 Standard-Mixer with the Upchurch prep Filter and use a passiv flow splitter between UV-detector and msd.

Regards
Klaus

[Koen Hollebekkers]

Sergio,

That Phenomenex is distributing them does not make a difference. Imagine, for some reason they got a problem in production of monoliths and are not able to solve it. What will your alternative be? None! That's what I'm trying to say. For sure they can supply you with as much columns as you want at the moment. But true validated methods need a plan B.

TFA is often used because it is more acid than FA. In general this gives better peak shapes and/or alternated selectivity. In peptide analysis TFA can also be an ion paring agent. The drawback of this is that it isn't as volatile as FA and therefore you will have a reduced signal in MS

Nevertheless I think that monoliths are the ideal candidates for people that don't have the money for a high price segment UPLC system. Unless someone else have has a different argument, I think that putting sub 2um material in a normal HPLC is like putting a Ferrari engine in a Lada. This just not make sense. The advantage of these sub 2um can be found in the high pressure area. Not in the few extra plates gained at optimum flow rate

An alternative for you can also be High temperature LC, did you already considered this?

[sergioanis]

hi Klaus and Koen,

Klaus, did you actually used 4ml/min with a 1.8u column? It seems to me that the backpressure will be to much.

Koen, I set the termostate at 35ºC, I'm not sure to increase it because I don't know what is going into the column (nobody does stability at this stage)

May be you are right about sub 2u. What do you think of the new Luna 2.5u? It doesn't seem to me much different from what you can get with Luna 3u


Seems that there is not many options, so thanks to everybody for the advices.

regards,
Sergio

[Klaus I.]

Hi Koen,

I think Sub-2-um are in a normal HPLC are pretty usefull when you don't need the "extra"-Resolution and use short columns with 30 mm and a flowrate > 1 mL/min for a 2 mm ID. You have extra work to reduce the dead volume but then you will have a good alternative for buying a high price UPLC. Alternative you use 4.6 mm ID Columns.

A UPLC is specified to the pressure of 1000 bar only by a flowrate below 1 mL/min. Over 1 mL/min the specified pressure is 600 bar. A normal HPLC-System has 400 bar this gives a factor 1,5. You can combine this with a higher temperature.

[Uwe Neue]

For rapid analysis, you always need to work at a high flow rate. Commonly, you still get better results from smaller particle sizes, but if you play your cards right, you can get pretty decent results with a 2.5 micron particle. The issue is that you will nearly always reach the pressure limit.

For a 2 mm id. 2.5 micron column, and the short run times that are targeted here, you want to work around 1.5 mL/min.

I also need to say that the decrease in flow rate of the UPLC is gradual. At 1 mL/min the pressure limit is 1000 bar. At 2 mL/min the limit is 660 bar. For flow rates in between these two values, the pressure limit is also between these two values.

To those interested in how to set up fast separations, here are three references:
U. D. Neue, Y.-F. Cheng, Z. Lu, “Fast Gradient Separationsâ€

[Koen Hollebekkers]

Uwe,

I think you made a point here that many people do not see.

The main benefit of sub 2um particles is that you can work at high flow rates without significant loss in plates/resolution.

The increased plate number is an additional advantage.

So the question I have is the following; does a 600 bar system really exploit the benefits of a sub 2um column? The logic question after this; does it make sense to produce sub 2umcolumns with a maximum operation pressure of 600 bars?

I will look to the literature if it can enrich my knowledge

[Uwe Neue]

Koen,

In principle, you will be better off with a higher pressure. With more pressure, you can either use longer columns, if you care about plates, or you can get to a faster analysis. I just posted a link to a picture with a 10 second gradient separation executed at 9500 psi, which is about 650 bar.

If you are interested in a 2 to 3 minute separation, you are essentially in the same boat: the smaller the particle and the more pressure you have available, the better will be your separation. The trick is too expand the gradient (in volume units) over the target time frame, and for the short times that we are talking about here, you will always want to work at the maximum flow rate or pressure that your instrument can deliver. Higher pressure means better results.

This is one of these cases where more is better.