Late eluting peaks with SEC

[miro2009]

I'm working with a silica based SEC column (dimensions 7.8x300mm), and I see an unexplained peak at the end of each run, even with the injection of WFI USP grade from sealed ampoules which should be certainly free from any contaminants.

We run the mobile phase (phosphate buffer) at flow rate 0.5 ml/min, and we see this broad "bump" at around 40 min (equivalent to 20 ml, way beyond the column volume of 12 ml). We extended the run time to 80 min to ensure it's not something eluting late from the previous run. We also performed an HPLC systemand injector cleanup with isopropanol to exlude system contamination with the same obtained profile.

Does anyone have an idea what this "bump" is related to? What confuses us is that it elutes very late after the column volume by about 10 ml, and on 2 different columns on 2 different systems... or is it normal and could be disregarded?

Could it be the type of solution I use for rinsing the needle (line R) between injections (20% methanol prepared freshly) or is it highly unlikely that methanol would cause this problem?

To all those who have expericence with SEC columns, awaiting your opinions or suggestions.[/img]

[Gerhard Kratz]

That`s a very unusual phenomenon. Please let us know what kind of detector you have. If possible please do 1 injection with your buffer and after that one with water, without rinsing the needle with 20% methanol after each injection. And if you want please do a 3rd injection with your 20% Methanol rinsing solvent. If you use UV detection please let us know the wavelength.

[miro2009]

That`s a very unusual phenomenon. Please let us know what kind of detector you have. If possible please do 1 injection with your buffer and after that one with water, without rinsing the needle with 20% methanol after each injection. And if you want please do a 3rd injection with your 20% Methanol rinsing solvent. If you use UV detection please let us know the wavelength.

I'm using a UV detector, wavelength 215 nm. I'll try out rinsing the needle with water only instead of the methanol in line R, and repeat the water, buffer and methanol injections... I'll let you know the outcome though I'm not putting high hopes.

[miro2009]

That`s a very unusual phenomenon. Please let us know what kind of detector you have. If possible please do 1 injection with your buffer and after that one with water, without rinsing the needle with 20% methanol after each injection. And if you want please do a 3rd injection with your 20% Methanol rinsing solvent. If you use UV detection please let us know the wavelength.

I'm using a UV detector, wavelength 215 nm. I'll try out rinsing the needle with water only instead of the methanol in line R, and repeat the water, buffer and methanol injections... I'll let you know the outcome though I'm not putting high hopes.

Unfortunately, problem still not solved. Rinsing solution for needle was changed to water, injected water and buffer and still seeing that constant bump of almost similar intensities. Methanol injection showed nothing additional.
I'm very confused here and still don't know what could I be doing wrong??

[danko]

Would you mind posting a chromatogram, illustrating the problem?

Best Regards

[Alera]

It defintely sounds like it is a system peak, related to the injection step itself (not necessarily what is being injected). Have you tried a blank run (0 uL injection)? UV detection at 215 nm will pick up almost any disturbance in the mobile phase flow (refractive index change). It is possible that when the injector valve is switching, it causes the disturbance that you are observing. Depending on how long it takes to make an actual injection, this disturbance will have a longer "retention time" and might be closer to 20 mL that you are seeing. You might have to live with this peak if you are running at 215 nm.

[danko]

How would the lenght of the injection procedure/process influence the retention time?
40 min

[miro2009]

It defintely sounds like it is a system peak, related to the injection step itself (not necessarily what is being injected). Have you tried a blank run (0 uL injection)? UV detection at 215 nm will pick up almost any disturbance in the mobile phase flow (refractive index change). It is possible that when the injector valve is switching, it causes the disturbance that you are observing. Depending on how long it takes to make an actual injection, this disturbance will have a longer "retention time" and might be closer to 20 mL that you are seeing. You might have to live with this peak if you are running at 215 nm.

Thanks Alera for posting your opinion, it makes sense to me that it is system related because I also observe this on another "identical" system with another column, so it can't be that I have the same mistake/contamination/error on both systems.
Although the concept itself is new to me, you mean that the mobile phase flow is disturbed somehow during injection? Lets say the injection takes 5 seconds, why does it take 20 ml to appear? I feel that'a very long time! If I simply just turn off the pump for 5 seconds and back on again, ur right a baseline fluctuation occurs but usually within just 4-5 minutes at most (2-4 ml), or what do you think?
Please let me know your comments...

[Gerhard Kratz]

Miro, SEC is a "simple" separation mode, large molecules come out first, small molecules later. I also think that there is a system related problem, not a column problem. Please give us details about your system, autosampler, pump and if possible, a chromatogram. What temperature you have for the separation? Please don`t be confused, we will figure it out.

[Mattias]

All SEC columns have reversed-phase properties as well. Since you are running without any organic modifier, these properties are in full swing in your system.

My bet is that this is some hydrophobic junk coming from off your system. You can try to add 5-10% acetonitrile to your mobile phase, any hydrophobic peak should move then (or dissappear)

[Andy Alpert]

Matthias -
It is not true that all SEC columns have some reversed-phase properties. Carbohydrate-based materials (Sephadex; Suberose; etc.) do retain aromatic compounds. This involves interaction with the crosslinker used to connect the carbohydrate-based polymers; the higher the degree of crosslinking, the stronger the retention. Silica-based SEC materials do not exhibit this effect, and the dimers of phenylalanine and tryptophan elute earlier than the individual amino acids, as they should in SEC. See the following link for a detailed discussion: http://www.polylc.com/downloads/SEC_boo ... r_ver1.pdf
It is not possible to demonstrate any hydrophobic interaction with a silica-based material with a truly hydrophilic coating.

[Mattias]

I have only worked with carbohydrate and polymeric materials, so I was too fast to jump to conclusions.

It is however hard to explain peaks after the column volume if there are no other mechanisms than the size-exclusion effect.

[chris555]

It is not uncommon to see a late eluting peak even in the blank for SEC. We usually see a peak at ~45 minutes running a 0.5M NaCl/ 10 mM Sodium Phosphate bufffer mobile phase (55 minutes total run time, 0.25 ml/min flow rate, 215 nm detection, Tosoh TSK-Gel G3000 SWxl column, Waters Alliance). My guess is that the peak has to do with the MP.

[miro2009]

It is not uncommon to see a late eluting peak even in the blank for SEC. We usually see a peak at ~45 minutes running a 0.5M NaCl/ 10 mM Sodium Phosphate bufffer mobile phase (55 minutes total run time, 0.25 ml/min flow rate, 215 nm detection, Tosoh TSK-Gel G3000 SWxl column, Waters Alliance). My guess is that the peak has to do with the MP.

Dear Chris, but in your case this is normal, your peak is around the end of your column volume of ~12ml with flow 0.25ml/min (I'm guessing the dimensions of your column 7.8x300mm). In my case, it appears additionally after the peaks at end of my column volume, I wish there is some way I could attach the profile for everyone to see but can't find a link fo that.... is there a way to upload files on this forum?[/list]

[danko]

viewtopic.php?t=2617

Best Regards