Maximum injection on HPLC column?

[Josh85]

Hi
can anyone tell me what the max inj volume should be as a % of column volume? C18 column.
Thanks

[Josh85]

I should add it's for analytical HPLC not prep.

[Noser222]

It depends if you use isocratic or gradient. Inject as much as you want until your peaks become distorted. There's a rule of thumb: Injection volume can be about 15% of the peak volume if you're injecting in a solvent that's the same strength as your mobile phase.

With a gradient, however, you can play other tricks. If you start out with a very weak mobile phase and focus your analytes at the head of the column, you can inject as much as a column volume or more.
I actually ran a method where I injected 1 mL on a 4.6 x 75 column.

[danko]

Hi chrom people,
I run very often into this misconception: The focus is on the column volume, rather than the diameter, when the load is considered. The fact is: The column diameter is the single most important parameter – stationary phase coverage, particle surface area etc. being equal.
If one imagines a very voluminous but hair- thin and very long column, it’s not difficult to imagine that the load must be minute, despite of the large volume. I know of purification people that didn’t take this fact into account and they just couldn’t upscale their purification process from pilot to a real life process. They had concluded that the process was non- scalable

Best Regards

[Josh85]

Thanks for replies so far. I want to do isocratic on 4.6mm id and 2.1mm id. I was told to inject 5x less on the 2.1 (ratio of cross-section areas) and was looking for a rule of thumb to give me an idea of what is genrally an OK injection volume using 5cm and 10cm long columns. Yes, I can do a series of experiments but for general education where would you start? Sorry, Darko, I don't totaaly understand your answer.

[danko]

Hi Josh,

The rule of 5 times less load on a 2.1 mm diameter column, compared to the load on a 4.6 mm column is OK.
My message concerned the diameter as the important dimension as opposed to the length. So, forget all about the length and load 5 times less the amount you used to load on the 4.6 mm column and you’ll be fine.
In terms of an absolute amount you can start with some 10 µg sample load and test some different loads around that value (on the 2.1 mm column).

Best Regards

[danko]

I forgot, you were mostly interested in the injection volume. Here, the ratio of 5 is valid as well. Otherwise the volume should be at most 1 % of the column volume, when solvent is different from the mobile phase and 10 % of the column volume when the solvent is the same as the mobile phase.

Best Regards

[Uwe Neue]

It may not be that terribly relevant for the question asked here, but you can inject a very large volume if (in RP) your sample has a solvent composition that contains much more water than the mobile phase. Assume that your analysis runs in 50/50 organic/water. You would usually make up the sample in the mobile phase, and inject. If you dilute it 1:1 with water, you can inject more than your injector can handle. This is one way to gain in injection volume and then in the sensitivity of the assay.

[Josh85]

Thanks for your comments. (Dancho, sorry for spelling your name wrong!). That's been very useful.
Uwe, I get it, like loading up a SPE cartridge to concentrate on top of the column and then elute.

[chromeleon]

so the loop has to be in inject position when introducing your sample?

apologies if I've misunderstood

[Karen01]

so the loop has to be in inject position when introducing your sample?

apologies if I've misunderstood

If you are doing manual injection then you put it to load to put sample in the loop. During load the mobile phase bypasses the loop and the loop is connected to the injection port and waste so you can fill it with sample. Inject put the loo in line and flushes the sample onto the column.

If you are injecting in a weaker solvent than the mobile phase and you are running isocratic then you can increase the column until your peaks distort either from starting to migrate OR overloading with analyte.

For isocratic the whole injection needs to be put on the column at once and so you sample loop and autosampler syringe (or manual syringe) needs to be large enough for what you want to inject. Many autosamplers allows you to change those things to increase injection volume

If you are using a gradient and a weak sample solvent and if your analyte does not migrate at the initial conditions you can do multiple injections and concentrate at the head of the column and THEN start the gradient... In that case you are only limited by column overloading.

HTH
- Karen