What really important in the column selection process is the estimation of how well particular column has been packed.
If column is showing efficiency lower than it should this means that packing bed is loose and you may expect it settled during the operation. Packing bed settling developing the void on the top of the column which act like a mixing chamber and the overall efficiency decreases dramatically.
The question is what efficiency we have to expect from particular column?
Major parameters affecting the column efficiency are:
All these parameters also affecting column backpressure. The wider the particles size distribution the lower your efficiency and the higher column backpressure. Irregular particles always show lower efficiency, but the majority of the packing materials on the market now are spherical.
Your main attention has to be on the first two parameters. Efficiency ideally is proportional to the column length, so is the backpressure. This means that you can not increase column length indefinitely. And you do not need to do this. In your separation you are actually after for the resolution, you are trying to achieve better resolution for your components of interest. And resolution is proportional to the , this means that the increase the column length and efficiency two times will only give you 1.4 times increase in resolution, but your backpressure will rise two times.
Let us discuss the table below. We will compare the efficiency of five different columns.
The specific efficiency is the number of the particles per one theoretical plate. In other words it means how many particles we need to perform one single separation step. It is obvious that the lower that number the better.
Columns 2, 3, and 4 packed with the particles of the same size but have different length, so we may expect proportional increase of the efficiency, ideally. In reality we did not get it, increase of the column length 2.5 times, from 10 cm to 25, gave us only 0.5 times increase of the efficiency. This is reflected in the rise of the specific efficiency value.
The reason for that is the packing procedure. Columns are usually packed by forcing the slurry (suspension of the particles in specific solvent) through the column tube with only bottom endfitting with the frit installed. Average applied pressure usually 10,000 psi (600 bar) and it kept constant.
At the beginning of the packing process slurry moves through the column very fast (constant pressure mode) and the flow is turbulent which ensure the formation of the dense uniform packing bed. While the process continues, the rising of the packing bed inside the column increase the resistance to the flow. Since we are using constant pressure, and we can not increase pressure any more, because porous silica particles are fragile, the flow rate decreases and it became laminar. The laminar flow will not form dense and uniform packing bed and the top of the column always packed slightly more lose than the bottom. The longer the column the more significant this effect.
The optimum packing efficiency is considered to be between 2 to 2.5 particles per one theoretical plate.
This means that we may expect the optimum efficiency for 3 m particles on the column of 10 cm and less, for 5 m particles it will be on the level of 15 cm, and 10 m particles could be used on 25 cm column.
Further increase of the column length will give you a proportional rise of the backpressure but you won't gain much of the efficiency.