Chromatography Forum

I have an issue that I have never seen in 25 years of GC work. I am modifiying a method for triclosan in liquid soaps. The original method calls for a 10:1 split ratio. The correlation coefficient for a series of standards bracketing the expected analyte concentration is, at best, 0.95 (unacceptable). When I changed to splitless injection to try and improve the curve, the peak areas of the analyte and the solvent remained the same as they were for the split injection runs?!? This happens with both injectors. The instrument is an Agilent 6890 with two capillary injectors (EPC), FID detection, DB-5 column (15m x 0.53 mm, 1.0 um film). Has anyone else seen this kind of behavior?

I have a nice bridge I would like to sell you in NYC.

Now are you telling us that with an injector that is working correctly a 10:1 injection gives the same peak areas as a splitless injection?

By the way, did I mention I have a nice bridge to sell?

Sometimes I wish Agilent had never changed their split injector from the classical design when you KNEW a split was actually happening and when one wasn't happening.

I hope you discover your error, be it in software setpionts or in hardware pneumatics, but something is not right with your world of GC. I am guessing that your pneumatics are fixed in split mode and they are refusing to change to give you a splitless injection.

Or when you changed the method to run splitless the change did not stay in place.

Good luck in finding your gremlin.

Rod

Might be worth checking purge off time on splitless and gas saver times on both - just be sure no unreasonable value has crept in.

One of the most common problems on this forum with split-splitless inlets is the split lines and vent filters of Agilents getting blocked. When this happens you get the bypass flow out through the split vent outlet, but you are really doing a splitless injection. So your split injections were really splitless, and setting up for a real splitless inection then makes no difference. Do you see any change in peak area if you run splitless but change the split ratio ? - try 5:1, 20:1 and 50:1 so that any differences should be obvious. With such a short wide bore column your inlet pressure must be very low, and this will make the system more susceptible to restrictions in the split lines.

Peter

I have an issue that I have never seen in 25 years of GC work. I am modifiying a method for triclosan in liquid soaps.

Well, I "gots" tons of experience assaying triclosan. In the 1980s to early 1990s we assayed triclosan in liquid soaps by capillary GC after trimethylsilyl derivatization; that made the triclosan more stable in the inlet. But for cGMP reasons we switched decades ago to HPLC, reverse-phase, with UV detector, and cGMP-validated that.

Can you hear the splitless solenoid actuating when you use the splitless mode? If it's inoperative, then will just remain in "split".

We also determine triclosan by HPLC. For us, GC for this is a thing of the past.

I'd check the actual flows through the vents. I'd guess (as earlier posters have) that you aren't getting what you think.

Peter's advise is solid as always.

I fixed a coworker's Agilent GC last week - he had lost all his signal in split mode - splitless looked almost normal. He had already changed out his split line and filter. It turned out the split arm of the inlet was clogged with all sorts of stuff - likely it had enough flow to let the EPC function, but otherwise the inlet would not work correctly.

I am still puzzled - if anything I would have thought the run would have looked like a sloppy splitless run with the line mostly clogged - perhaps his injection was all blowing out the septum purge instead?

Tim