VFA Retention Time Problems

[Frosty]

Hello,
I'm quite new to HPLC so excuse me if I ask some silly questions!
I'm currently running some short chain VFA samples using an isocratic method and have started to experience some changes in the retention times. The retention times for my standards and samples are increasing with ones later in the chromatogram increasing more than earlier ones. The early peaks are coming off about 30 seconds later than before but ones towards the end are up to 2 minutes later. I am using 5mM sulphuric acid containing 0.1mM sodium azide as the mobile phase. The flow rate is 0.5ml/min and it's running at room temperature. I thought the column may have got a bit dirty so I've done a few injections of acetonitrile to clean it but the retention times are still varying. Unfortunately I don't have a colum heater so temperature around the column does vary. Would this cause the retention times to change in this way?

Thanks,
Frosty

[Bill Tindall]

you didn't say what column you were using. I will guess that it is ion exclusion, a sulfonated cross linked styrene like a Biorad 87H. When posting questions provide details like column, detection mode, what is being analyzed, etc. What detection mode are you using. UV at 205 nm works very well for the smaller acids. conductivity is less sensitive and results in nonlinear calibration curves.

There are lots of reasons a retention time may increase over time. If the increase is not casuing a problem with quantitation or resolution, don't worry about it.

Possible reasons- temperature of room decreased. Pump flow rate decreased. Could indicate a leak, or check valve or seal failure and these need to be fixed.
Some kinds of (older) pumps will run more reliably at higher flow rate.

These items do not relate to the increased retention time, but you don't need to add the azide and a flow rate of 0.5 mL/min is probably slower than necessary. Use as high a flow as the column will tolerate without degrading resolution or getting too close to upper pressure limit for column. Increased flow rate will get the analysis done faster and the pump may work better.

Let us know if the problem isn't fixed.

[Frosty]

Hello,
Thanks for your reply. I am using a Supelcogel H column (25cm x 6.4 mm) with a guard column and a UV detector set at 210nm. The HPLC itself is a rather old Shimadzu LC6A (with autosampler etc). The VFA standards have been made up in water and the actual samples are rumen fluid (it's been centrifuged/filtered). I have been running standards (and a few samples) on and off for the past 2-3 months and haven't noticed more than a slight drift in the retention time. Unfortunately in the past few days it has shifted by up to 5 minutes and the peaks have got broader and asymmetric. Yesteday I flushed the column with methanol for about 45 minutes and left it to equilibrate overnight (back to 5mM Sulphuric). This morning I ran a standard at 0.7ml/mn and the solvent peak is now coming off at around 7 minutes (compared to around 4 minutes).

I know there is a slight leak around the inlet (not sure what you call it!) but it's always had this and it hasn't affected it before. It's about to be sent off to be fixed and serviced, but I was trying to get a last run of samples done before it went. Maybe I should just send it off and see if things settle down when the machine returns.

Thanks agian for your help.
Frosty

[Bill Tindall]

Ah, rumen fluid. Brings back memories of cows with port holes in their sides and similar gross samples. I get reminded every time my sheep burp. What do you need to determine-acetic, propionic(C3), C4 isomers , C5 isomers?

Must be all the experts are on vacation. I can't believe I'm left alone to deal with this problem.......

I expect you have at least two problems. The increased retention time sounds like a pump problem. You can let the pump outlet run into a graduate cylinder for a while and measure the flow that way. If you have a pressure read-out see if it is pulsing which would indicate a leaky check valve. If liquid is running out somewhere a leak would cause the problem and it could also be piston seals in which case liquid will run out the back of the piston. All these are things you should be able to deal with yourself. Read the manual and have at it.

The asymetric peaks sounds like a column problem. These are nasty samples and I would not expect long column life. Yes, you have a gurad column, but how do you know when it needs to be changed? And if it is not changed soon enough then the crap will break through and mess up the main column. Take the guard off and run a standard on just the main column and see if peak shape improves. If you don't need to resolve C4&C5 isomers you may be able to do the analysis on just a guard column and then throw it away when it gets gunked up.

This separation will run on a reversed phase column as well. They can be cheaper and easier to clean but you will need acetonitrile to get the C4 and C5 isomers off.

You can also extract into isopropyl acetate or ethyl acetate and do by gc.

[Frosty]

Hello,
Thanks for your suggestions, I will try them out.
The HPLC has been away being serviced as it had developed a bad leak!
Hopefully it will be Ok when it gets back and I can get back to work on it.
I've just ordered some new guard columns so I can try using them alone.
Thanks again,
Frosty