Unexplained peaks

[laudrup]

Hi,

I am fairly new to HPLC. I am using a brownlee aquapore column and guard column with C4 packing. I am using it to analyse wine proteins with buffers:

A: 10% ACN / 90% Milli Q water / 0.1% TFA
B: 90% Milli Q water / 10% ACN / 0.1% TFA

I am using this in an agilent 1100 machine and detecting peaks at 220 nm.

The gradient runs from 0 to 60% B over a period of 20 minutes (the gradient varies over this period).

I was getting some weird peaks coming off at about 12 minutes so I tried runnning some milli Q water samples and got the same peaks. I also tried running blanks and got the same peaks.

I have made up new buffers but that hasn't fixed things. I used a new bottle of ACN. We are right at the bottom of the bottle of TFA and we didn't have any more so I used TFA from the same bottle.

I cleaned the column with 100% B for an hour and when this didn't work I replaced the guard column and then the actual column when this didn't work either.

I don't know why we are suddenly having problems. I ran 100 similar very samples just 2 days earlier and had no problems.

Would greatly appreciate any help.

Cheers,
Laudrup

[WK]

Laudrup,
Looking at your description of mobile phase you are running isocratically.
I think you mean B is 90%MeCN/10%water with (0.1%TFA) ???

After checking that out you could clean the column with 100% MeCN after a few samples to see if anything else elutes.
However you did say that you replaced the column. Did the extra peaks elute in the first run? Are they broad or sharp?
Can you explain what you mean when you say the gradient is varying - is your pump/mixer not consistent?
This may help anyone looking at this posting
WK

[Kostas Petritis]

Laudrup,

Try at least to run the same (blank) gradient with mobile phases which does not contain TFA. This will eliminate the possibility of the TFA being the cause. You may have problems like this from old or polluted TFA.

Otherwise you can use the same mobile phases and increase the equilibration time. If the intensity of the peak incrases is again the TFA...

Another possibility can be polluted water... but my first bet would be the TFA

[DR]

Set up for some blank injections. If the strange peak is larger when you allow the system to run at starting settings for a longer period of time (like before the 1st injection) than it is for subsequent injections, you have to clean up your water. Use a C-18 Empore extraction disk to filter your water before A phase prep or put a small analytical C-18 column between your A pump outlet and mixing chamber to clean the A phase on the fly. The column method may not work quite as well and you do have to hit it with a high % organic phase now and then to clean it out.

[khimik]

I think DR is on the right track. Your mystery peaks may be mobile-phase impurities. A quick way to determine this for sure is to run your gradient program without an injection. If the peaks are still there then bingo, they are contaminants in the mobile phase(s). You can do this on an 1100 by setting the number of injections per vial to "OFF." TO do multiple cycles, program a vial range, e.g., "vials 1 to 3, # inj = 'OFF'" will give you three gradient cycles with nothing injected.

I agree with the other folks about old TFA being a common bad player. Definitely has caused me trouble before, that goes away when I make fresh mobile phases from a new bottle of TFA.

Another helpful trick for improving a gradient baseline is to equilibrate your column first with the high end of your gradient program, then re-equilibrate with initial conditions before your first injection for the same amount of time that it would normally re-equilibrate during a sequence, or maybe a little longer.