User experiences of fused-core particle columns?

[Jam]

Dear All,

What kind of experiences do you have analyzing pharmaceuticals (small molecules and peptides and proteins) with columns filled with fused-core particles (for example Halo and Zorbax Poroshell). Regarding the few articles I have read the analysis can be speed-up significantly (especially with larger molecules) using the conventional HPLC (the backpressure is not an issue). I wonder if the resolution power is equivalent compared to the fully porous particles...?

Thanks of any input beforehand!

JAM

[mbicking]

Fused core particles have been around for many years. They were originally called "pellicular" packings. The recent interest is due to the development of the smaller, 2.7 um fused core particles. (Poroshell is an older, 5 um material.)

The big theoretical advantage is the fact that the entire separating surface is in a thin layer on the outside of the surface. This means that molecules do not have to diffuse as far into the pores. The result is reduced mass transfer terms in the van Deemter equation. So, the efficiency is better and there is no minimum in the van Deemter plot; you can operate at higher flows without losing efficiency. (This is the quick explanation; the theory guys could spend days on this.)

The short diffusion distances means that these particles are particularly useful for large molecules (proteins, peptides) that normally do not show very good efficiency on standard particles (because they diffuse so slowly). But even small molecules show good efficiency. There are reports of reduced plate heights (=plate height/particle diameter) of 1.5, while conventional particles are typically 2.0 - 2.5. This means you can get better efficiency per unit column length. These columns compare favorably with the UPLC particles (1.7 um), but operate at lower pressures.

But there is always a downside. Primarily, the sample capacity is lower, since you have less surface area available.

I will be doing an evaluation on some large proteins in mid-September. Write to me later and I can share my general experiences.

[danko]

Hi Merlin,

Would you be kind to share the output of your upcoming evaluation with the rest of this forum.
It could be a natural continuation of this thread, for instance.
I’ve used – some time ago - the Zorbax–Poroshell column on a ca. 20 KDa protein with quite a success (where other columns didn’t do the job as well). But especially if you include the Hallo column type in your investigation I’m sure it will be interesting to hear a word from you.

Best Regards

[mbicking]

The project actually involves a monoclonal antibody separation, so these are quite large molecules. The initial work will be using Poroshell. Unfortunately, the fused core materials are not yet available with a large pore size.

I will share what I can. At this time I don't know how much will be confidential.

[Bryan Evans]

Hi Merlin -

Will this be determining homogenity of intact antibody?
Or will you be reducing the antibodies followed by separation of
light and heavy chains?

[mbicking]

Bryan:
Perhaps I have discussed this too much already. I will share as much information as I can, but I do not know if that will be some, all, or none. Please understand that it is not my decision.

[adam]

This is an interesting thread. I have recently been doing some reading about whether fused core particles really have any benefit for small molecules. Some of the earlier papers claimed the benefits were only for large molecules, but more recent papers (though not many of them) claim benefits for small molecules as well.

Does anyone have any experiences to offer where you did a head to head comparison, with standard 3 um porous particles.

[mbicking]

I hope to perform some similar experiments this fall, with final results in a Pittcon poster. I'll post some early results if possible.

[Kostas Petritis]

Merlin,

Are you also going to investigate loading capacity and maybe dynamic range of measurements (and how bad things become once we exceed them)? It would be most useful to include these aspects as well in your study.

I find also interesting the debate on who developed what and when:

http://www.agilent.com/about/newsroom/p ... 08006.html

[mbicking]

I have some flexibility in what I do, and you make some interesting suggestions. I will keep them in mind, and if I can always try to sneak them in if necessary.

[jzt]

I've had success using the fused-core column for impurity analysis of small molecule pharmaceutical compound. The drug substance has 3 aromatic rings, and 2 cholo atoms on different rings. One Cl is switched from position 2 to 5, resulting in a regio-isomer that’s very similar to the parent compound.

We screened more than a dozen different types of columns, with acetonitrile or methanol as the mobile phase B. This Cl-isomer could only be separated with methanol and C18 column. The impurity peak eluted at the tail of the main peak. Temperature and pH didn’t do much.

We compared Zorbax SB-C18 (1.8 micron) with Halo C18 (2.7 micron), both 50x4.6 mm, flow rate at 1 mL/min. back pressure on the Zorbax was 150 bar with resolution 2.7, while the Halo only gave 84 bar with resolution 3.2. the resolution here is a for a different critical pair, the Cl-isomer couldn’t be baseline separated with such short columns. We also compared Luna C18 (3 micron), which resulted in slightly lower back pressure but worse resolution than Zorbax. In both comparisons, peaks came out slightly earlier on the Halo column, the elution order remained the same. Halo just had the extra efficiency we needed without getting over the pressure limit of our instruments (400 bar).

In the end, we used the 150 mm long Halo C18 column to resolve the Cl-isomer (resolution 2.3 to 2.7 on different instruments/columns). In order to achieve desired sensitivity for impurities (LLOQ 0.05% of the main peak), the injection volume was 25 uL at 0.3 mg/mL. This resulted in a slightly fronting peak (UPS tailing factor 0.9). This may be an indication of the reduced sample capacity of the fused core particles. This method has been validated, used for GMP lot release and stability testing, and transferred to a contract manufacturer.

Overall, I am happy with our experience with the Halo C18 column (also sold as Ascentis Express C18).

[danko]

Hi jzt,

Thanks for data, or should I say the memo.
I think it’s a very good examination – you have performed.
If I may comment a single matter, I would like to address the fronting you observed after increasing the injection volume. My first anticipation would’ve been that the cause was volume overload. Because 7.5 μg load on a 4.6 mm column shouldn’t be an issue. Maybe you see a “strong solventâ€

[Tom Waeghe]

Regarding the smaller pore size HALO columns (90 Angstroms) any advantage for small molecules due to shorter diffusion distances is not that significant, until you get to analyte MWs between 600 and 1000 Da. The Van Deemter curves for these smaller particle size, superfically porous columns are similar in shape to those for sub-two-micron particle size columns, with even lower minima for reduced plate heights versus STM columns (1.3-1.5 vs. 1.5-3 for STM columns, depending on brand and conditions).

The HALO Fused-Core columns can deliver virtually the same performance as STM columns in half the time using the same column length at twice the flow rate at the same backpressure, the same performance at the same flow rate at half the pressure, or twice the efficiency with a column twice as long at the same pressure as STM column. The higher efficiency of the fused-core columns comes primarily from the extremely narrow particle size distribution, leading to very stable, well-packed beds with a much-reduced eddy diffusion contribution (A term) to H and h.

[mbicking]

Tom:
You may recall we met at Pittcon a few years ago (or maybe not). But I see a different affiliation now. You wouldn't, by any chance, be involved with the legal proceedings noted earlier, would you?

[Uwe Neue]

I had wanted to stay out of this, but if the vendor of Halo is presenting half-baked half-truths, I think I need to step in and correct the statements.

The Halo particles and Halo columns are rather good - for 2.5 micron particles. They do not give a lower backpressure than other 2.5 micron particles, but the performance at the minimum of the van Deemter curve is rather good.

But this is where it ends. All well-packed sub-2-micron columns beat a Halo column of the same length. A 5 cm Halo column has a maximum plate count of about 8000 to 9000 plates under fixed test conditions, while three competitors' sub-2-micron particles have between 11000 and 12000 plates at the van-Deemter optimum under the same test conditions. We need to note that this comparison was done on an Acquity system with low bandspreading. Of course, if you try to do such a comparison on a standard HPLC system, you won't be able to tell the difference, due the much larger extra-column effects. Thus the claim that Halo columns give the same performance as a sub-2-micron column of the same length is plainly false, possibly due to incompetent measurements. Also, a comparison of double the column length of Halo to a shorter sub-2-micron column is false, since the UPLC instrument allows you to run longer columns. Halo columns on an HPLC are not a substitute for sub-2-micron particles on a UPLC instrument.

Note that I am talking about real plates, and not "reduced plate heights" as the Halo marketing literature uses.

Of course, I do not deny that in application comparisons, where resolution counts, a Halo column can sometimes be more successful. But this has nothing to do with intrinsic performance.