Chromatography Forum

Hello!

I'm a newbie to HPLC use and I need some serious help. I'm running caffeine stock solutions on the HPLC and not having much luck because I'm getting around 100 or more peaks each time with a 50 ppm caffeine stock solution injection.

Here's some information about my instrument/setup:
It's an Agilent 1100 series, 0.5 mL/min flow rate, 50/50 methanol water mobile phase, and the wavelength is set at 254 nm.

If anyone could help or direct me to readings to do or to another place that could help that would be fantastic. I feel so lost right now!

Thanks again!

What kind of test are you running- assay, related substances, content uniformity? The extra peaks could be down to a lot of things, but try to eliminate the simple things first- are you absolutely sure the wavelength is set correctly in the instrument method? Correct column type and length. Is the dilution and sample prep correct?
The column or detector could be contaminated also, can you send on a sample chromatogram to give us a better idea of what to suggest.

Sounds like dirt on the column or an air bubble or your lamp is dying. Your baseline should be 'flat' as a pancake.

Please run a blank (solvent only) with the same method and conditions and post the resulting chromatogram.

I agree with HPLC chemist. Maybe you have also air bubbles in the flow cell.
Try to purge your system with pure MeOH and check if your mixing valve is working fine! with 50:50% MeOH Water don't be shocked if you will get a negative peak when your coffein stock is containing only water. Pancake is not smooth enough, I would call it a Crepe. Use 75% MeOh and 25% water, wavelength is ok. Instead of MeOH you can use Wodka, should work fine.

Running caffeine at 0.5 flow I doubt that method is longer than 10 min, so I suspect that does are not 100 peaks but spikes? Is pressure stable? Also check the peak with and slit setting in detector. I would use 273 nm for caffeine with MeOH an water. Can you post picture of chromatogram? Also check that UV lamp is actually on and intensity test pass.

It would be *really* helpful to see a chromatogram. Instructions on how to embed chromatograms are here: http://chromforum.org/viewtopic.php?f=1&t=2617

Hi G.P.,

I agree with Tom and most of these other fellows above. I'll ask, is the backpressure reading out of the 1100 consistent, that is, is the backpressure steady during the caffeine sample injection(s)? Oops, just saw that someone had the same question...that question remains valid. This could also easily be a bubble in the UV cell...and Tom is correct. A picture of the chromatogram you see on your end will be priceless to us in our effort to help you.

What will always be true is that it is nearly impossible to "over-flush" an LC in order to clean out the injector, all connecting capillary tubing and the detector cell. If we think based on a chromatogram you post that the LC should be cleaned out before trying more injections of caffeine in earnest, I'm certain you will receive many recipes for cleaning a LC system. Best Wishes...please post a chromatogram when you can?

Oh, and never fear...seeing the IDs of these posters above me, we'll figure out what's going on and help you sort this out. These fellows above, they all know well what they are doing with LCs.

The "50" Peaks you see are likely noisy baseline peaks ( always check Y range, peak area) and your problem is you see no caffeine peak to distinguish from the baseline noise.
You should see caffeine at 254nm but Best at 273 nm
--- If you have no pressure your purge valve is open = no peaks
--- are using a C18 Column? inject directly with a union, the peak should fly out in 20 seconds in any solvent
---with c18 at 0.5 ml/ min , peak is about 2-5 min
--- this is easy fix if we can see chromatogram
---if above is ok, double inj volume
----overlay with blank diluent

Hi everyone,

First off, thank you so much for all the responses and I'm so sorry to be taking so long to get back to you all. I've posted a picture of my chromatogram.

http://imgur.com/a/okiPh

This was after running DI water through for 10 minutes on the same conditions as stated previously.

Hi GP1234,

My thanks, the photo says 1000 words at least!

To me, this looks like a dirty detector cell. As I recall, Gerhard and other make some suggestions for cleaning the HPLC system--you're likely using an Agilent, correct? I would remove the column and perform such a cleaning. Next, I would inject, say 1 microliter of the caffeine standard solution as suggested above--Without a Column attached...you'll get a nice-sized peak if all is well with the injector of the HPLC instrument. Then re-attach the column to the LC and try again.

My two cents' worth. Best wishes, and happy learning of HPLC. All of us started where you are at currently, so keep your spirits up. This will be resolved, so-to-speak...

Oh, I should note, when you clean the system with, say, 100% MeOH, this is a good idea, or 100% ACN. Water, it may not get the job done unless it is HOT, say 80 degrees Celsius. We want to remove organic contamination and clean the UV cell...and also ensure that bubbles in the cell are removed. Dirt attached to the inside of the UV cell catches air dissolved in the mobile phase Very Efficiently...sometimes it is necessary to place a backpressure source after the detector cell to help with this (can simply be a length of capillary tubing...and care is required for this to not do harm to the UV cell)...this is a lesson for another time, though.

Thank you so much! I will try to clean the column and see if that helps get a peak.

Hi again, GP1234,

It is Not the Column so much...Methinks it is the detector cell that needs the cleaning. If you purge the column with 100% MeOH that is good, please, though, Do Not Connect the Column to the Detector while purging the column.

It is a good idea (after John Dolan) to keep things separate and one-at-a-time whilst troubleshooting.

Separately, I'd also purge the LC (Without the Column Installed, Connect the Capillaries Using a Union Connector) with 100% MeOH...50:50 (v/v) MeOH/water, whatever Gerhard recommended above. Trust in Gerhard...he knows what he is doing.

Best Wishes!