Sample giving different profiles in SEC

[MestizoJoe]

I have injected the same sampe 10 times in SEC.

The sample is a cationic polymer, mw = 3-4 kDa.

The only part of the profile which changes is in the higher molecular weight region. The system has equilibrated overnight.

Why might I get these inconsistent peak profiles? the polymer is fully solubule in the diluent and mobile phase. The column is a yarra 2000, with the gaurd attached. Mobile phase is pH 3. Inj vol is 10.

See below how each injection gives a slightly higher area in the higher molecular weight regions and also pushes the average molecular weight higher and higher with each injection.

[danko]

You are dealing with unspecific retention – probably reversed phase or ion exchange based.
The decrease of it (i.e. increase areas etc.) is due to saturation of the sites that are involved in process.
If you keep injecting your sample you’ll see constant areas and retention times – eventually.

Best Regards

[MestizoJoe]

Not sure what you mean by unspecific retention. If retention times of the different polymer species vary the total area ought to be very similar, shouldn't they?

Reversed phase gave a very broad peaks, carry over, poor reproducibility, and poor linearity. I didn't try ion exchange due to lack of time.

This method seemed to work fine at first, with 0.6% RSD for n=5 injections and 0.9996 R^2 value for the linearity of standards from 50% - 150% of my nominal assay concentration. When I ran the method a second time for quantitation purposes I ran into this problem.



Notice how after about 7.3 minutes the profile look almost identical. Same retention time, and area.

I injected several times as suggested and area RSDs do get smaller but still I see each peak area getting larger. I think I will condition the system with 5 or 6 sample injections before running the assay.

[danko]

“Nonspecific” might be more correct than “unspecific” (?)
Anyway the term refers to the fact that stationary phases often bind/retain the analytes through more than one interaction. F.x. SEC stationary phase (in your case) separates according to the molecules’ sizes due to its pore size and nothing more (in a perfect world). But IRL it could attract molecules because of its and their polar or nonpolar properties, or by ion exchange mechanisms if the analyte and some of the stationary phase’ surface are oppositely charged.
In such cases some of the loaded substance/s will be retained and will thus remain on the column “for ever” unless a stronger eluent is applied. And because of the gradual saturation of the sites that act as described above, the low recovery (the permanent binding) will decrease and eventually cease occur. To avoid this kind of effects you can add some salt to the mobile phase, reducing the ion exchange effects, add some organic modifier (f. x. ACN or methanol) reducing the reversed phase effects etc.
Or as you’ve already done load the sample several times (potentially overload to reduce time consumption) until you observe constant area and retention times.
Finally it should be mentioned that not all of the compounds you load (f. x. If the sample contains more than one compound) will be equally affected by events described above, so you won’t always experience the same tendencies in relation to all of the compounds you’re loading.

Hope it helped

Best Regards

[MestizoJoe]

Thanks. It makes sense that some components of my polymer mixture might be irreversibly binding to the column. That would explain why only the larger molecules seem to be affected. And also why over time less of these bind, since those binding sites are being used up. I've been using ACN. I think I will switch to methanol which does a better job of dissolving this polymer.

[carls]

What is your mobile phase? Do the calibration standards show similar behavior?

[Coffo]

In some cases peaks at the high MW end can increase with time because the sample changes with time (i.e. from run to run). This can be due to sample aggregation (although you might expect a decrease in peak area for lower MW components if this is the case), or it can be due to high MW samples dissolving more slowly over time and therefore increasing in concentration.
However in this particular case I agree with Danko: it looks like there is some non-specific binding of the higher MW material. If you can't find conditions to prevent this binding (by changing mobile phase or perhaps brand of column) then passivating the column by applying numerous injections until profile is consistent from run to run is the only real solution.