novice questions: why is my chromatogram periodic?

[glenchung]

Hello, I work at a high school and we have acquired an HP 5890 Series II GC.
I am running a sample of hexanes through a Restek capillary column, and I get a chromatogram that has one large peak and one tiny peak, but these two repeat very regularly over and over again throughout the chromatogram. What have I done wrong, please?

Also, most methods on this machine call for a 0.5 uL or 1 uL sample injection. I only have a 1-mL syringe, so what happens if I inject 100 uL instead of 1 uL of sample?

Here is my chromatogram, in a Dropbox folder: https://www.dropbox.com/sh/dsv92sl7d1lrg5q/emkR4C5rpY

Thank you for your time!

[Steve Reimer]

Not only are your peaks periodic, they appear identical so they are so probably an artifact.
As far as injecting 100 uL, on the first GC I every used my lab partner and I were trying to get a chromatogram of gasoline on an early HP with an FID (this was in the 70's). Every time we did an injection the flame went out. Or maybe I should say, every time my partner injected 10 uL of standard the flame went out.
Answer, you must use a proper syringe to have any hope of getting usable data. 10 uL, or smaller syringes shouldn't be hard to come by and you will need one to reproducibly inject 1 uL. More than 2 uL of solvent will overfill the inlet when it vaporizes.

[glenchung]

Thanks Steve!
-You say the peaks are probably just artifacts. Does this mean these are just noise, and that my sample (hexane) is not really shown at all on the chromatogram?
-I had injected 100 uL instead of 1 uL, so I overfilled the injection port. How do I fix this? The injection port had been set at 250C, and the sample was hexane, so I assume I do not need to take any action to empty out the injection port. Is this correct? The extra sample will just vaporize and go away? Or do I need to run carrier gas through the column for a few minutes in order to force the extra sample to go away?

Thank you again for your kind help.

[LiVD]

Your peaks are definitely not "just noise". Noise is much more random in a sort of uniform way (like grass).

Your GC seems to be doing something exactly 1.5 seconds apart. I don't know your equipment, but might it be possible it's trying to light its FID?

[Peter Apps]

What you see is a test signal generated by the GC electronics - this was a feature of the 5890 GCs. It is so long since I used one that I cannot remember how to turn it on and off though. There are current 5890 used on the forum who will be able to advise.

Sample injections to a GC need special glass microsyringes, do not use an ordinary plastic hypodermic syringe, no matter what the volume. The maximum volume you should inject is 5 ul, and nearly always you will inject 1 ul, so you need a 10 ul syringe.

Hexanes are quite difficult to separate on a routine GC setup - they call elute quickly and very close together. Some petrol (gasoline) diluted with hexane at about 1% will give an easier sample. What is the length, diameter and stationary phase of the column, and what kind of detector do you have ?, probably an FID.

Peter

[rb6banjo]

Do you have a chemstation to control the instrument or are you doing it from the keypad?

[glenchung]

We have a 30m simple RTX-5 column, fused silica, 0.53 id.
We use ChemStation.
How should I turn off the periodic signal?
Thanks!

[glenchung]

The machine has FID and ECD. I light the FID myself, by opening the air and H2 flow knobs on the GC and pressing the ignition button on the GC.
Thanks!

[muGC]

Peter is right, you are seeing the test signal. On page 2 of your pdf file, the signal type listed on the chromatogram is TST1A. Instead you want it to be FID1A (if the FID is your front detector).

On ChemStation, go to "Method & Run Control", select menu "Instrument", then select "Edit GC Parameters" (depending on your ChemStation version this may be listed differently). Then click on the "Signal" tab of the menu that pops up. Change Signal 1 from Test to Det and choose the front detector. Select an appropriate Data Rate and make sure you have the "Save Data" box checked! You probably also want to change your online signal from Test to the front detector, this is on the View menu.

[glenchung]

Thanks Peter and muGC.

We use ChemStation Rev. A.10.02 [1757] from the year 2003.

I have now set Instrument --> Signal --> Signal 1, Changed from TestPlot to Det B; Peak width 0.053 min.; Rate 5.0 Hz.
In this list, I only changed from TestPlot to Det B. There are two other Det B settings which I do not recognized, so I chose the setting that was just "Det B". The other two are "Det B--Comp1" and "Det B--Comp2". Det B is our FID, but what are Comp1 and Comp2 ?

I will try my test run again, and I hope to see only one peak for hexanes sample, since another forum member suggested my setup may not be sensitive enough to separate out the different hexane isomers.

But I am still going to use my 1000-uL syringe, trying hard to put in just 1 uL. We have a few of these 1000-uL syringes; would somebody like to trade one 1-uL syringe for a couple 1000-uL syringes?
glen

[muGC]

Comp is column compensation, it correct baseline rising from column bleed during temp programming. I don't use it, nor do any of my colleagues. Just use DetB.

You need a 10 uL or 5 uL syringe. 1 uL is too small.

[glenchung]

Thanks! I will get a 5 or 10 uL syringe.

Some protocols say 1 uL injection. Is my apparatus not good for such a small amount, and should I be injecting 5 uL every time?

[glenchung]

Yes! I have a huge, broad peak at about 1 minute of retention time, and no other peaks!
Thanks very much again muGC, Peter and others!

Now I need to acquire a smaller syringe.

[muGC]

In my syringe sizes I was assuming you are using an autosampler. If you're using an autosampler, ChemStation software will only allow you to inject a max of half the syringe volume. Minimum volume injection (for 5890/6890) is 2% of the syringe volume unless you have a nanoliter adapter. If you're manually injecting by hand, then its up to you and a 1 uL syringe would be fine.

Sounds like your flow rate is too fast if you're hexane RT is 1 minute for a 30 meter column.

[Peter Apps]

Yes! I have a huge, broad peak at about 1 minute of retention time, and no other peaks!
Thanks very much again muGC, Peter and others!

Now I need to acquire a smaller syringe.

Good progress. while you get a smaller syringe you can "inject" small quantities with your 1 ml syringe as follows. Suck up some of your sample, then push the plunger all the way back to zero. This leaves the needle filled with sample. Then quickly insert the needle through the inlet septum and immediately withdraw it. The slight temperature rise will expand a little bit of the sample out of the tip of the needle, which will vapourize quickly to give nice sharp starting bands for your chromatography. It is not repeatable, but it will do until you can get a proper micro-syringe.

Good luck

Peter