Column Packing: Constant Pressure vs. Constant Flow Approach

[M Farooq]

Would the forum readers like to share their experiences with the HPLC column packing procedures on phenomenological basis? I know it is considered a closely guarded secret but there is no harm in discussing the phenomena that take place during the column packing process - for the sake of science. I am interested in knowing the difference between constant pressure (or slam packing) vs. constant flow approaches. What is your criterion between choosing the two approaches lets say for silica phases?

The disadvantage associated with slam packing seems to be the fact that the bed is heterogeneous along the entire length of the column. The bottom part is essentially crushed or too tight and top part is essentially loose because of the pressure difference. Perhaps the middle part is more homogenous. Constant flow procedure would avoid the problem but it does not seem to be very popular (in the literature).

Thanks.

Farooq

[Andy Alpert]

Hi, Farooq,

Sometime around 1985-1995 (?) someone did an experiment to explore your speculation. They stacked 5 empty column segments in series and then packed them all as a continuous bed. The segments were then separated and their performance in chromatography assessed. Per your conjecture, the middle segments performed best. I expect that you could track down this paper in Journal of Chromatography with a little effort.

Andy Alpert

[M Farooq]

Hello Dr. Alpert,

Are you talking about this one "Evaluation of the Uniformity of Analytical-Size Chromatography Columns Prepared by the Downward Packing of Particulate Slurries" (2002) by Shalliker & Guichon? It was published in Anal. Chem but not in J. Chrom. They make an important comment "In all cases, the most poorly packed region of the bed was the column inlet section, irrespective of the bed length"

This is exactly what has been annoying me. The top part appears loose whether it is a polymer or silica. Constant flow approach might circumvent this but there is no constant flow packing paper.

What is your view/ experience about constant flow vs. constant pressure approach? I am interested in knowing the criterion for choosing one or the other.

Thanks.

[Andy Alpert]

Yes, that was it. Quick searching on your part.

RE: My views on constant pressure vs. constant flow for column packing: Your initial posting was correct; this is a closely-guarded secret by all HPLC column manufacturers. It takes considerable time and expense to climb the learning curve of packing different chromatography materials optimally. If this information were then presented publically, then the main beneficiaries would be our would-be competitors, who would thereby save all the trouble and expense involved.

[DJ]

I think Tom said he knew a colleague who said how to make a small fortune buying competitor's small particle material and selling self-packed columns: that was to start with a large fortune and work your way down.

[Gerhard Kratz]

In addition to Andy I would say:
In the states you have I hope not only 1 chief with 1,2 or 3 stars on his restaurant. You can get some recipes from that chief, with all ingrediences on it, but when you dry to cook the same dish at home it will 1. look different and 2. taste different. A good chief will never give out all the secrets.
There is a chief, so far I remember his restaurant is in Los Angeles, who is boiling lobsters not in hot water but in hot vinegar to keep the meat white like fresh snow. I saw this on TV but I'm sure that even I would get a fresh lobster here in the black forest and I would boil the lobster in vinegar it will be completely different. Also quality of the lobster is for sure different.
And a new prepacked column is cheaper than a dinner in this restaurant in LA.

[Vlad Orlovsky]

Since I am not afraid of competition
We pack columns with Haskel air-driven pump at constant pressure. This pressure depends on few things but the main one is the mechanical stability of the silica gel (or other particles). A typical packing pressure is 400-500 bar, unless you pack particles with big pores, then it is dropped to 300 bar, or even 250. Haskel pumps, in our packing stations, have a ratio 1:30 to 1:120 and our compressor is set 120 psi (10 bar) (so in theory you can create about 10,000 psi). if you don't have an air-driven pump, then you can use a regular HPLC pump which can deliver/work with 400-500 psi at certain flow rate (we packed our first columns with Knauer high pressure pump). The key top successful packing is to find right slurry (solvent in which silica is suspended) and pusher. Simple sedimentation studies can give you a rough idea what to use. After you pack couple of hundred columns you will know how to modify conditions to troubleshoot low plate count and poor symmetry.

Let me know if you have other questions. Good luck.

P.S. The fact that you don't have homogeneous bed tells me that you choice of slurry and pusher are not correct and that you are not passing enough pusher through your column.

[M Farooq]

Dear Vlad, Thanks for your suggestions and being open to scientific discussions. I am certainly not looking for recipes since I have been packing columns for the last 6 years with excellent efficiency with similar Haskel systems. Packing recipes are always useless and those are not transferable from one particle to another, that is why the scientific principles behind packing are more useful and it can certainly be discussed without revealing the actual procedures.

The colloidal properties of slurry are certainly important. If we consider packing a column with stationary phase as a high pressure filtration process, chemical engineering books treat constant pressure filtration and constant flow filtration differently.

(1) So in your opinion, do you see any benefit of constant pressure vs. constant flow approach?

(2) In the constant pressure approach a sudden jolt (so called slam packing) is better or a pressure gradient is better, in general?

In my experience, constant flow pump gives a very different bed morphology which is highly flow rate dependent with little correlation with efficiency.

Thanks.