Chromatography Forum

Hello there!


So, first of, I am a PhD candidate in a biochemistry lab and have been appointed the HPLC "god" of our lab. Which is funny, because my previous experience with this device is limited and the responsible postdoc(s) have left without leaving proper protocols or even manuals. I have already spent a decent amount of time in figuring the software (ChemStation) out, read A LOT about the principals and such, and have managed to not mess up our standard applications which are routinely used here. Of couse, in my first month of duty we had a major device failure (FLD lamp), but now everything is back in place and I have started to work with the device more and more.

For two weeks I have been trying to get advance a new protocol to detect two compounds via derivatization. The derivatization products are fluorescent, which is good, but the elution of the relevant peaks trouble me.

I am using a C18 column with 3,5 µm, 4.6x150 mm dimensions and have initially - to figure out when my compounds elute in general - run a long gradient over 20 min, with an initial 4 min 0% B @ 1 ml/min to exclude that they do not bind to the column. After I noticed that the compounds elute early, I have run an isocratic program with which I can very nicely see my peaks elute from the column.



Buffer A: 0.5% Acetic acid, pH 3.9
Buffer B: 100% MeOH
Injection volume: 50 µl

12% B for 12 min @ 1 ml / min
100% B for 2 min @ 1 ml / min
12% B for 2 min @ 1 ml / min

Peaks elute at 5 min and 10 min, respectively. The flow-through peak - small, but noticeable - is visible at 2.4 min.

Now to my problem: in several measurements I have noticed that the peaks can only be found in the flow-through, albeit similar sample preparations and a few times even for the sample when an aliquot of it was thawed and injected.
I am absolutely clueless how to attack this program. I have performed overnight runs with methanol, have injected column-cleaning reagent, yet the problem resurfaces and I am afraid that it will screw up my measurements when the method is applied to the regular context.

I'd be grateful for any help!

I'm not exactly a reversed phase expert, but my first suggestion would be a longer re-equilibration step at 12% B. By my calculations, you're running less than 1 column volume of your 12% B mobile phase after your 100% B wash step. I'd suggest upping that last 2 minute period to something 10 minutes at minimum. Otherwise, your retention may be hampered by residual methanol (mobile phase B) left in the column.

The first thing I would try is to make a longer equilibration. Make it very long, and then shorten it bit by bit until you find retention start to be irregular - then go back to the last reproducible setting

Scott and Carlo nailed it .

A good rule of thumb is that it takes about 10 column volumes to get complete wash-out of a column -- and then you have to add in the gradient delay volume ("dwell volume"). Your column volume should be somewhere around 1.5 mL (I suspect the 2.4 min peak is actually something retained) and a dwell volume of 1 or 2 mL would not be unusual, so 15 or 16 minutes would be a good initial time. You can then experiment to see how far you can cut it down.

Thank you all for your suggestions - I am implementing the changes as I am writing this, and will keep you posted!

Hello once again,


the problem has been solved, thanks to your suggestions. After the 100% methanol washing step, I have included an equilibration step of a total of 8 CV, using an increased flow rate to shorten overall method run time. This works like a charm, and in 60 runs the problem has not appeared once anymore.

Thanks for reporting back with your results. I wish more visitors to this forum would do the same. Firstly, it might help someone in the future with the same question. Less importantly it lets myself and the others who offered up a suggestion feel that satisfaction of knowing they were on the right track.

It's only fair after multiple people took the time to discuss my problem. As a person new to this field, a lot of my open questions are resolved through threads like this one, and the establishment - and also advancement of my method - procedes more from day to day.