I am using normal phase LC with Hypersil APS-2 NH2 column, 100% hexane mobile phase at 1mL/min isocratic to determine the content of a high boiling paraffinic oil in a resin/oil blend at about 70/30 ratio. RI detection. During the run, the pressure rises more than 10 bar over about 12 minutes then drops back to initial pressure. Analyte elutes at 2.5 minutes. This rise causes baseline drift up to the 12 min. then baseline drops back to zero. I have established that this is unique to a single LC system, I can't reproduce it on any of our other LC systems. Some sort of pump problem? Any ideas?

I am assuming the pressure rise only happens after an injection is made. That is, the pressure is not constantly cycling even if an injection is not made. In which case I would suspect a problem with the injector or with any guard frits mounted between the injector and column. Does the pressure rise occur if the injector valve is operated without making a sample injection? If you have an automatic injector, try making an injection with filtered hexane and see what the pressure does. I don't know what the significance of the 12 min drift is unless this is a time when the injector valve changes from the inject position to the load position.

Anonymous: Thank you very much for helping, you are correct in that the increase in pressure only occurs after an injection. The pressure increase does occur with a blank hexane injection. The LC is a HP 1090L with a DR5 pump system and autoinjector and I am pretty sure that the injector valve switches from the inject position to the load position immediately prior to an injection.

Assuming that the column used with the "problem" instrument is the same as the one used on the others and that there are no guard columns, in-line frits, etc unique to the "problem" instrument, I would first start looking at the injector.

I am not familiar with the HP system. We have a system consisting of separate modules so it is possible to replace parts of the module without altering the rest of the system. This may not be possible with your system. If it is possible, and if you have a manual injection valve, try replacing the auto-injector with the manual valve and see if the problem goes away. If it does, then the problem is in the auto-injector. If you do not have a manual valve, you could try switching auto-samplers from one instrument to another to try to isolate the problem.

If these tests are not feasible, try cleaning the injector valve and related tubing. Check to make sure the valve is assembled properly and that any alignment marks are in the proper position. It is possible to get a pressure increase if the valve is not opening completely. However, if the valve were not opening completely, I am still not sure why it would finally do so 12 min into the run unless something about the pressure build-up cuases it to do so. That does not happen with our system. Do you use this instrument for any other methods and do you have the same problem with it for the other methods? Is the operating pressure for these other methods greater than that for the "problem" method? Does the pressure continue to cycle over 12 min intervals or once it drops does it remain stable (try extending your run time, if necessary to check this)? Have you tried increasing your flow rate so you are operating above this pressure to see if you get any pressure increase after an injection when the initial pressure is above the point where the baseline drops? Hope some of this helps. Good luck!