Jeremy,

Good afternoon! This is a good question, and one I get on a fairly frequent basis. The answer is: it depends. It depends on what your sample is, how 'clean' or how 'dirty' it is, how soluble it is in the diluent and mobile phase solvents. We do have some general guidelines you can follow, which are general approximations you can follow:

Luna C18(2) 250 x 4.6mm column... ~12 mg (5 mg analyte/g of sorbent)

Luna C18(2) 250 x 10mm... ~56 mg

Luna C18(2) 250 x 21.2mm... ~250mg

Luna C18(2) 250 x 30mm... 501 mg

As a guideline, you are safe injecting these amounts as a starting point. Once you have started to run your actual samples, you can start looking for the upper limit of your analysis and your column. Once your peaks staer to broaden, and you see other negative secondary column effects, it is likely you have reached the upper limits of loadability.

Is this helpful?

Paul

Hi Jeremy, I think there probably are rules of thumb, but it depends on your sample and separation as well. If you are performing an analytical separation, it is important not to overload the column as your response (peak area versus amount) will be affected. I think 1% of the total column volume is a good place to start for an analytical separation, and perhaps adjust it from there. Hope this helps, Dave

What are you looking for in the plasma? If you do not need protein, you should deproteinize the sample with organic solvents (Methanol, Acetonitrile, ...) in a ratio 1:1 or 1:2. Otherwise, the column will be clogged pretty soon. DA

Dear Jeremy,

in my knowledge there are not clear rules. It depends of your compounds: avoid coelutions, not saturated peaks. A good star is looking for works using your same column (or similar) and compare their injection volumes. But as I say when I am trying a column the guide is my chromatogram. David Gervais has pointed out a good answer.

Jeremy,

Good afternoon! This is a good question, and one I get on a fairly frequent basis. The answer is: it depends. It depends on what your sample is, how 'clean' or how 'dirty' it is, how soluble it is in the diluent and mobile phase solvents. We do have some general guidelines you can follow, which are general approximations you can follow:

Luna C18(2) 250 x 4.6mm column... ~12 mg (5 mg analyte/g of sorbent)

Luna C18(2) 250 x 10mm... ~56 mg

Luna C18(2) 250 x 21.2mm... ~250mg

Luna C18(2) 250 x 30mm... 501 mg

As a guideline, you are safe injecting these amounts as a starting point. Once you have started to run your actual samples, you can start looking for the upper limit of your analysis and your column. Once your peaks staer to broaden, and you see other negative secondary column effects, it is likely you have reached the upper limits of loadability.

Is this helpful?

Paul

Dear colleagues, thank you for your answers, which will be very helpful. You're right, there is no universal response to this question. Actually, we inject increasing amounts of digested raw plasma (digested proteins mix, only with a desalting step, no "purification"), until we reach the loss of linearity between signal and loaded amount, as pointed by David. But there is an another parameter to take in consideration : the use of electrospray source, very sensitive toward matrix effect. I think that the loss of linearity, implied by ESI saturation, is reached quicker than the "real" column saturation in case of very complex samples (see Bruins publication for some details). Unfortunately, this matrix effect is closely related to preparation protocol (how clean is my sample?), and method analysis (how is the quality of my separation?), as pointed by your comments.

However, the guidelines given by Paul are closed to our observed values, thank you for this confirmation.

Jeremy

The maximum quantity inject in a column depend on several parameters"

1) type of detection used .as this will affect the sensitivity issue.

2) dimension of the column i.e. length, internal diameter, particle size.

3) type of column used e.g. analytical, semipreparative or preparative.

The advise of injecting 1% of the total column volume is good start and then you can optimize the injection volume accordingly.

Jeremy, my experiences lead to a different conclusion than yours, even though the general guidelines given previously are technically correct.

First, failure to deplete metabolites from plasma can cause substantial problems with collecting good-quality fragmentation spectra, due to the abundance of precursors. I would also worry about column capacity considering this non-analyte background.

Second, column capacity should be considered within the larger context of the data collection workflow. Specifically, it's easy to have a quantity of peptide/protein which saturates neither the column nor the source, but causes carry-over in the next run. This could be from the injector, valves, or column, but the effect is the same: your "background" will increase and cause false positives or incorrect quantitation unless you perform really thorough inter-run cleaning. Such cleaning is not always feasible.

So my general conclusion is that the column is a concern before the source, but that's because of low-end performance (impaired via carry-over), not high-end linearity. In effect, you have to reduce analyte loading before you hit saturation in order to maintain low-end quantitation. This same observation has been confirmed via discussions with researchers working for MS vendors.

All experiments were performed using gas chromatograph with a mass selective detector (GC/EI-MS or GC/NICI-MS). For chromatographic separation was achieved using a DB-5HT capillary column (30 m × 0.320 mm I.D., 0.10 µm film thickness) from Agilent. Helium (purity 99.9996 %) carrier gas was flowed at a constant flow-rate of 1.0 - 3.0 mL/min after a pulsed flow injection in a split-less mode. The injection volume of sample was 1.0 µL.

Best regards,

PhD student Nerijus Karlonas

Department of Analytical and Environmental Chemistry, Vilnius University, Lithuania