Chromatography Forum

I have a waters column, XTerra MS C18, but I wasn't sure how to do column cleaning for it as the "care and use manual" provided by waters is really limited. I'm coming up with a column cleaning method in combination of a column storage method. So, any help will be very much appreciated.

Actually, I have been getting ghost peaks in my runs but the occurance wasn't very frequent, I find that hard to troubleshoot. So, I did suspect that it might be my column cleaning being insufficient. Though there might also be other factors affecting it like micro-bubbles, etc.. or samples get stucked in the sample loop.

Currently, I have been flushing it with water for 30mins -> 50%ACN50%water for 120mins -> 80%ACN20%water for 120mins. My column lasted <4 months before it went dead. That's less than 1300 injections within that few months. Or was it the column life span itself?

Being the column died so frequently, and also, having impurity peaks turning up in my blank runs, I think it might be the column. Please advise.

The recommended column storage method for nearly all RP columns from Waters is 100% acetonitrile.

Ghost peaks come rarely from the column. MUCH more often they are coming from the injector.

Common column life in RP is about 1000 injections and/or 3 months of use. This is a reasonable average over many different users.

Thanks uwe for your reply. I'm currently running gradient run. The problem is I have been getting an unknown peak presence in my run for mutiple injections around 6mins and the peak area remains constant. Do you think there is something wrong with my gradient run or perhaps, there is presence of micro-bubbles?

Mobile phase composition:
MP A = low molarity buffer phosphate 2:ACN (98:2)
MP B = low molarity buffer phosphate 2:ACN (78:22)

Information:
phosphate 2 with pka value of 7.2
pH: 7.2
column: waters xterra ms c18, 10cm x 4.6mm x 3.5um
injection volume: 10ul
flow rate: 1.5ml/min
temperature of column: 35 degC
detector: UV 230nm
solvent used: filtered with Empore SDB-XC filter, HPLC grade ACN
vials: agilent HPLC grade LC vials


Time MP A MP B
0 100 0
10 0 100
15 0 100
16 100 0
20 100 0

I've looked through my past data and got to know my senior did a little experiment using a restrictive column with injection mode and without injection mode.

The one without injection mode gave a very beautiful blank profile. It was concluded that the injector was at fault. Is there anyway, I can get more precise data to pinpoint it was the injector's fault?

Thanks alot for your help again!

Hi there Cactus,

since it is known that for some compounds the solubillity drops at low and/or high organic modifier, it is always a revealing experiment when you program your gradient run twice in a row, so without using the injector. If you see a (ghost) peak in the second gradient run, 9 out of 10 times you are experiencing a problem called column carry over and this can be solved by changing the gradient in such a way that the sollubillity of your compound(s) is not compromized.

I do not understand what you mean by restrictive column. A capillary tubing?

It could be the injector, it could also be your mobile phase. To test if it is related to the mobile phase, run your gradient after different equilibration times in the starting mobile phase. If the peak increases with increasing equilibration, the culprit is your solvent.

I do not understand what you mean by restrictive column. A capillary tubing?

My error, that was supposed to be written as a "restrictive capillary", it is a capillary tubing without packing materials in it.

run your gradient after different equilibration times

erm.. can I have an example of this?

A test was run without the injection, just the gradient run alone and it turned out to be ok. But when injection of blank (initial mobile phase), peaks started to appear. This was performed on several of our LCs, so that means, all our injectors need cleaning? I am stunned...

Thanks, Dries Vrielink. If I filtered my mobile phase using empore extraction disk, only particles smaller than 80Ã… will pass through. Is there a possibility that some particles start forming through the gradient run during the pump mixing? And sadly to say, I could not modify my gradient run as this is a registered method...

Still not clear.

When you compare the gradients with and without injection, and got no peak without injection and a peak with injection, did you do this experiment with the capillary tubing, or with the column?

If you did this with the capillary tubing, this does not mean that anything is wrong. It could be a dirty injector, but it also could be simply the injection solvent being of a different composition than the mixed mobile phase.

If you get a retained peak with a column, and the peak is the same on all instruments, we need to to the other troubleshooting path and see what is wrong with the mobile phase.

Experiment:

Restrictive capillary - with and without injection mode

Mobile phase composition:
MP A = low molarity buffer phosphate 2:ACN (98:2)
MP B = low molarity buffer phosphate 2:ACN (78:22)

wash solvent: 50:50 ACN:water
Run type: gradient

Time MP A MP B
0 100 0
10 0 100
15 0 100
16 100 0
20 100 0

If I'm running with capillary tubing, injected MP A for injection mode and run another using non-injection mode, for both gradient runs, peaks are present in the run having injection mode but a clean profile is what I got for one which does not have the injection mode. Does this imply that the injector is problematic?

When I have the column on, and doing multiple injections of the MP A, it shows presence of impurity peaks with consistent peak area.

So, I have shifted to using another system with a different column to run but I got the same result as before. Also with consistent peak area.

Thus, I thought it was the mobile phase having contaminants in it. There have been several factors I could think of.

1. contamination through mobile phase preparation
2. contaminants dissolved in organic solvent and are smaller in size than 80Ã… to be picked up by the filter paper <- is this possible?
3. problem with my gradient run
4. injector fouling
5. column contamination

There is one test that you can do to see where the mobile phase contamination is coming from. (This has been posted multiple times on this website; you can check for the results of others.)

Run the same gradient that you are running, but vary the time for which the column is equilibrated with the starting mobile phase (2x, 3x, 5x, 10x...). If the peaks in question increase with the equilibration time, they are due to contamination of the mobile phase. You then need to figure out, if it is mobile phase A, or B, or the filter, or the instrument...

Thank you very much, uwe. I'll try that and let you know hopefully next week.

Thanks, Uwe. I've still have a peak with constant peak area after multiple runs and it seems that it comes from the mobile phase. As currently, I'm lacking a new lot of phosphate salt, I'll have to stick to the old bottle till the new one arrives which I'm not so sure when.

One comment:

If you followed Uwe Neue's suggection and saw a constant size peak it means that the peak was not originates from your weak mobile phase.

Question: did you solve your problem? What is the end of the story?

Thanks for sharing,

Jim