Searching for my dream column!

[greenhuntress]

I am analyzing a Pegylated molecule of ~120kD (total), and have searched for a column that will completely exclude, and not bind, my analyte. I wonder if anyone knows of a polymer-based column, with particle size 5-10 microns, pore size around 60 Angstroms...? My project also requires a column internal diameter of >5mm.

I have looked and looked, to no avail...Any of you geniuses know of my dream column?

[Uwe Neue]

You need a column designed for aqueous SEC. My recommendation is a column from the Ultrahydrogel family (Waters). The molecular weigth describes is at the very top end of the calibration curve of Ultrahydrogel 500, and Ultrahydrogel 120 and 250 will exclude it completely. Which one you pick depends on what you want to separate from your analyte. Standard column diameter is 7.8 mm and length is typically 300 mm. Precooked procedures exist.

[Bryan Evans]

So if I understand you correctly - you need to exclude a molecule
of 120 kDa? What is the mwt of the analyte that you do need to analyze?

What type of detection will you be using - UV, RI, ELSD, MS?

[greenhuntress]

I am analyzing the molecule by excluding it from everything else in solution. I am using a Fluorescent detector, and have found a ToSoh column with all my required traits, except that the internal diameter is so big that my peaks are TOO broad.

I was just hoping that someone knew of an SEC column with the aforementioned characteristics, with a small ID (like >5mm).

Thanks for your reply...

[Uwe Neue]

Smaller than 5 mm reads < 5 mm.

A column that will do the job for this purpose is Waters Biosuite 125. It has the right pore size and the right column dimensions. It is not a polymer-based packing, but it is a silica-based diol for SEC of proteins.

[greenhuntress]

Oops! duh...I used the wrong symbol twice (> instead of <).

Anyway, I appreciate all the replies, but I cannot use a silica-based column. My PEGylated compound sticks to silica. I have spoken to Waters, and they do not have an appropriate column for me.

I am presently settling for two (inferior) columns, and beginning to realize that my dream column may not exist!

[danko]

I am analyzing the molecule by excluding it from everything else in solution.

Forgive me for mentioning this, but an optimal method of analysis should not exclude the compound of interest. Many other compounds could co-elute at that time - as long as they have roughly the same or larger size, thus compromising the specificity of the method.
Also, broad peaks does not mean that the column diameter is too large – not necessarily anyway. Maybe you load too much of your analyte/sample? Maybe more than one compound elute at the same time (apropos). There are other possibilities, but you can try loading a smaller amount of your sample for a start.

Best Regards

[greenhuntress]

Thanks for your input. Everything else that is in solution is MUCH smaller than the analyte. The sample is the product of many purification steps, and the goal is simply to quantitate the pegylated molecule. Also, I am using orthogonal methods to investigate the analyte.

I concede that I am not as informed as I could be, but I am working with several very experienced colleagues who have analyzed similar compounds with the same method. I rely heavily on their input.

However, they, too, have tried to find a suitable column, to no avail.

I appreciate your input very much!

[Uwe Neue]

I do not understand why a diol column want work. PEG does not stick to diol. I assumes that the underlying structure is a protein. Some addition of buffer/salt around 100 mM will suppress the ionic interaction of the protein with residual silanols. It is actually at least that difficult to make a polymeric packing interaction-free.

[mbicking]

I don't know how far your search has gone but there are other aqueous columns available that might be useful.

Polymer Labs (www.polymerlabs.com) has the Aquagel series that is polymer based. It has a polyol surface which may be less sorptive. Polmer Standards Service (www.polymer.de) has a methacrylate copolymer for aqueous solutions.

I have sent application questions to both companies and have been generally satisfied with their responses. I have a preference for Polymer Labs, but that may be personal. Check out both if you haven't already.

I also want to echo the comments of Danko. It is generally not advisable to try to quantify anything that elutes in the void volume (or in this case, the exclusion volume). The peaks are not subject to the same level of reproducibility that you see in something that actually interacts with the stationary phase. I would expect precision to be somewhat worse.

[Uwe Neue]

Hmmh... Merlin, I don't know about your statement about precision in SEC or GPC. People use this technique for very careful characterizations of polymers. You have one less variable to worry about than in retention chromatography. So both of these arguments would speak for less variability with GPC/SEC than with retention chromatography. On the other hand, the retention space can be lower, but this is not an intrinsic feature of GPC/SEC, but a question of the details of the technique. For high-quality separations, people often use multiple columns in series. Of course, a completely excluded peak will coelute with a lot of different things, and this is what I believe Danko was saying.

[mbicking]

Uwe:
Actually, I was thinking mostly of the problem of interferences at the void/exclusion volume. I also might expect some (small?) differences in peak shape due to injection and concentration effects for these peaks. But I do not agree that GPC is inherently more precise than conventional LC. There is "retention" going on, it is just based on a different mechanism, and I know from my research many years ago that non-exclusion effects are far more common than most people realize, or want to admit. These all affect the accuracy and precision of the results. We just have fewer ways to control them in GPC. I don't consider it a bad technique, but I also don't think it is in the same league as a good reverse phase method. Fortunately, in most cases, it doesn't need to be. As you say, with high efficiency particles and multiple columns, some very good results can be obtained, by polymer chemistry standards.

[danko]

Of course, a completely excluded peak will coelute with a lot of different things, and this is what I believe Danko was saying.

Yes, that was my main point. And if one is dealing with a protein (besides the PEG) it is not a secret, that proteins are susceptible to aggregation. So any derived polymers will co-elute together with the monomer species if the analyte of interest is excluded.

Another point to be considered is the real reason for peak broadening. Which might as well be the load (either the analyte itself or the volume), but there are more possibilities.

Finally, I find it difficult to believe that the particle size (mentioned as one of the criteria for finding the “dream columnâ€

[mbicking]

The particle size will affect band spreading (vanDeemter "A" term) because the molecules have to flow through the packed bed.

I have worked with some of the small particle GPC columns, and the peak width is much better. So, it will help, but probably not as much as you would see with a reverse phase system.

[danko]

The particle size will affect band spreading (vanDeemter "A" term) because the molecules have to flow through the packed bed.

I agree that there would be some diffusion in different directions, but given the fact that no stationary phase interaction is expected (minimal retention) the broadening will be negligible.

I have worked with some of the small particle GPC columns, and the peak width is much better.

But you didn’t exclude the compound you analysed

Best Regards