[alchmemist54656]

If I just run my HPLC without injecting a sample and just having the mobile phase go through the column should I get a reading from my device? I thought since the mobile phase doesn't interact with the column material and therefore it should not give a signal to the detector. However, when I just run the mobile phase I get different signals? Why is this so? The column is C18 and the mobile phase consists of Water, Acetonitril, and Triflouroacetic. I'm very new to HPLC and trying to just learn the basics.

[Borek]

What do you mean by "different signals"? What kind of detector? Do you run isocratic, or gradient elution?

[alchmemist54656]

Sorry to use the word "signals". I just mean the peaks in the graph. Shouldn't there be no peaks because nothing should be detected? It is a UV detector and i run an isocratic elution. I am very new to HPLC and chromotography in general and am trying to figure things out slowly. Do you recommend any books for me to pick up that will help me in the process?

[Borek]

Disclaimer: my experience with HPLC is limited and close to 30 years old, so you better wait for others. No, IMHO you should see no peaks, just a flat background signal.

[Arkcon]

Particularly for an isocratic run, I'd expect no peaks while "monitoring the baseline."  Is that what you're doing?  Try injecting mobile phase, and look at what you get.  The mobile phase may be old, or your column may have a buildup of impurities, or your system may benefit from a cleaning or other maintenance.  However, since this system is likely not yours, its really up to supervisors and other bosses as to what can be done, should be done, and whats worthwhile to do, and by whom.  Example:  how big are these peaks, are they negligible relative to what you're analyzing?  Do they occur in parts of the chromatogram far away from peaks of interest?  These are things you have to consider, before you start doing anything.  Or have you been assigned to "fix this", in which case, we might want to address the system stepwise.

[alchmemist54656]

I don't know what "monitoring the baseline" means. Can you please elaborate on that? At this point I don't want to inject a sample because I want to make sure the HPLC is working correct.

[Borek]

Monitoring the baseline basically means just what you did - run the instrument without a sample, to check if the baseline (background) is stable and doesn't drift in time.

[Igarrett]

Did you condition the column properly? Was the column stored properly?

You're right. You shouldn't see any major peaks but you may be reading contaminants or background noise. Condition the column for a period of time and try again.

[alchmemist54656]

I was wondering what wavelength should I set the spectrophotometer when monitoring the baseline of a mobile phase consisting of Water, Acetonitril, and Triflouroacetic. I know that water absorbs wavelengths below 190 nm but I am not able to find the respective UV absorption wavelengths for Acetonitril and Triflouroacetic,

[Arkcon]

Well, that's an easy one -- you should monitor at the wavelength your analysis requires.  You're correct, acetonitrile and other solvents absorb below about 220 nm, even water absorbs below 190 nm.  Not to worry, most UV spectrometer hardware isn't particularly efficient at very low or high wavelengths anyway.  You want to see if the system is good enough for your analysis -- it doesn't need to be way better than that.  That would be a waste of your time ... c.f. efficiency of spectrophotometer hardware at extreme wavelengths.

[Babcock_Hall]

The spectrum of TFA changes as the percentage of ACN changes, but this is a bigger problem when running gradients.  If you are attempting to detect peptides, this is often done around 215 nm.

[alchmemist54656]

So i just ran a sample of Paclitaxel and the graph came out very similar to the one in this link: http://www.chem.agilent.com/Library/applications/5988-7973EN.pdf . What I don't understand is why is there a peak before the main one? Shouldn't there be one peak for this sample? Someone told me the small peak before the big one is the "Column Dead Time." How would I go about making sure that this is the dead time? How would I determine the dead time?

[Arkcon]

OK, the first peak of an HPLC chromatogram is often the "void peak" -- it has other names.  This peak is "stuff" that didn't interact with the column at all, and just disturbed the UV detector.  You can verify this if your detector is a PDA, and you extract the spectrum   It won't have an absobance, it will just show noise or a baseline switch, that's a refractive index change caused by the pressure flux.

Often, you see the void peak when you inject a slug of mobile phase -- you shouldn't get a product peak in that case, unless you system is dirty with analyate.  You can also try to determine void volume for your system other ways.  You can inject a solution of potassium nitrate, because its an ionic compound it doesn't interact with reverse phase columns, and it has enough of a UV absorbance at reasonable UV wavelengths.  A trick of mine is to inject a slug of isopropanol, then monitor the pressure trace.  Alcohols are more viscous than water or acetonitrile, you can see when the slug of alcohol leaves the system by a sudden pressure drop.

Systems can have different void volumes depending on how much tubing they have, both internally and externally, so you'll want to know this information if you have to run on separate systems.