New to GC-FID

Hello,

I’m new to GC machine and I have some issues. We have GC-FID 7890A and using ChemStation G1701EA E.02.01.1177.

Method:

Method: inlet: 250°C, Split: 1:50 with a column: HP-Plot/Q (19095P-QO4) 30m x 0.530mm

Helium is carrier gas

Oven starts at 100°C with a hold time of 0.5min, goes to 200°C in 3min. I also tried isothermal at 150°C

Detector is at 350°C

Sample: 0.1 – 1 µl methanol (tried mixing with heptane as well )

 

First, the GC software (which I believe only control the GC machine) shows me an unusual plot whenever I run a test (Test plot) as shown in image_1. I see those almost consistent peaks all the time whenever I run a sample (in this case methanol) and even when I have no sample!!. I don’t think those peaks are related to the retention time of my sample at all.

When I tried to look at the data using “Data Analysis” software, I was able to see the response vs retention time, but again I also have a problem here. Every time I inject the sample, I’m only seeing noise. Today, I was luck to inject methanol ~1 µl using a syringe and got a peak at retention time ~ 1.2 mint as shown in image_3. After that, every time I inject the sample, I’m only getting noise (image_4).

I tried to change some of the parameters as follows:

Sample volume : 1 -5 µl (tried ethanol, methanol)

Injection temperature: 250 °C split mode (varied split ratio from 2:1 all the way to 50:1)

Overn temperature: 50 – 150 °C isothermal (tried to vary temperature from 50 to 150 °C )

Detector temperature: 350 °C (kept constant )

I don’t understand why I got one good data set (image_3) after that regardless of what I did, all what I’m getting is just noise.  If you could provide some guidance, I would really apricate it.

Parents
  • Glad to have a person new to GC instruments !   --- machines perform work, instruments perform measurements.  That software is MSD Chemstation as used with GC/MS systems.

    The test plot is the FID test plot signal with a big peak, the second peak is 10% of that peak, and the third peak is 10% of the second peak, repeating.   The injection volume is important as it is possible to inject too much which when vaporized becomes too large a volume and can cause problems. Typically inject only 1 µL.  The injection port pressure should always be >5 psi. The injection port total flow should always be 20 ml/min or greater.

    In the acquisition software, select Method, Edit Entire Method.  The software will step through every screen with all of the parameters used for acquisition. 

    When you get to the GC parameters, select Signal and the screen should look somewhat like this. This is from the current GC/MS MassHunter Acquisition software but the GC page looks very much the same.  The first line shows "Test Plot" -- in the red box. You don't want that. Click on the arrow to the right, the drop down will show Front Signal (FID), so select that, and make sure that the data rate is set to 20 Hz and that the Save checkbox is checked.  So your Signal 1 should look like this example's Signal 2 in the green box.

    From your screen capture it looks like your system has two injection ports, Front and Back. Is the column installed to the Back injection Port?  Make sure that the column configuration is set to match the physical column configuration.  I would suggest 2 ml/min of column flow and 20:1 split ratio as a place to start to see a solvent peak.  When you've stepped through all of the method screens, save the method with a memorable name - that has less than 40 total characters and no special ones no dollar signs, pound signs, etc.

    And try it.  Please share a screen capture, too. 

    On the GC you should be able to see the FID signal is not the test plot. When the flame lights the signal should go from 0.00 to something higher than 2.0 pA (Pico Amps of FID signal).  When the solvent peak elutes the signal should go quite high - maybe some hundreds of thousands of pA.

  • Thank­ you for the reply. I made sure that the injected sample (methanol in this case) is 1 µL. The injection port pressure and flow rate automatically adjust themselves whenever I changed the split ratio to 20:1and column flow rate of 2 ml/min.

    As you can see from the image. I checked the FID signal and set it to 20 Hz and saved it.

    Right now, the column is configured using back inlet and front detector (the back inlet and front detector are connected to the column).

    As you can see now, the signal I’m getting from the FID is just noise and the analysis software shows nothing. I did try to play with the “attn” to see if that will effect my baseline, run multiple tests, but still getting same issue.

    P.s. sorry for the image quality, I couldn’t screenshot it.

  • Is the FID lit? What is the signal reading on the front of the GC?  If you hold a mirror --- or a polished tool or something - an inch or so over the FID for a few seconds is there condensation?  Don't melt whatever you are using....

    And now an easy test.  Start a run. While looking at the signal, carefully insert a small screwdriver straight down into the FID exhaust until it touches the top of the jet and the side of the collector.  The signal should go crazy, as high as 830,000 pA.  That will prove that there is signal from the FID electronics to the software.

    If that test works, then turn off the detector gases and cool off the detector, remove the column, and put the end of the column into a vial or beaker of water or a very clean solvent. There should be bubbles out the end.  It's hard to guess what 2 ml/min looks like...  but turn the column flow up to 15 ml/min and see if it changes.

    2 ml/min is very low for a 530µM column.  Typical is 5 to 15 ml/min or so.  Also, 15 ml/min for Gas Saver is too low. The injection port needs 20 ml/min minimum for decent pressure/flow control.

  • - The FID is lit. The signal reading is 1.5.
    - I did hold a mirror when the FID is lit, and I do see condensation on the mirror.
    - I tried to insert a small metal rod, while running a test, the signal gives me a really high value (above 700,000), which could indicates that the FID is working.
    - I adjusted the flow rate to 8 ml/min, and gas sver mode to 20 ml/min.
    - I disconnected the column from the detector, and I put the end of it into a beaker of methanol, and I do see a lot of bubbles. which means the carrier gas is flowing the column.

    - I still do not get any reading from the FID when I inject sample. Could this be the column itself? I'm very confused at this moment... 

    your help will be appreciated. 

  • Signal reading 1.5.  There is a GC setting -- lit offset -- if the signal falls below that, the FID thinks the flame is not lit and will try to relight it. If that is set to 0.00, the GC never knows if the flame is lit or not.  Since your reading is 1.5 when you know the flame is lit because of the condensation test, make sure that the lit offset is set to something like 1.2. That way if the flame goes out, the GC will know and try to relight it.  The default is 2.0 and your baseline signal is lower than that so you will need to change that setpoint from the GC front panel.

    So the FID electronics are working because of the high signal with the metal rod/screwdriver test. Did you see signal on the computer while trying to do a run?

    The column has flow and the FID is lit and the electronics are working.  Have you tried a different syringe?  Does it squirt out sample onto a piece of paper? 1 µL is a tiny droplet for sure.

  • yes I did see a signal on the Computer software when I tried to run it while inseting the metal rod.
    The syringe works fine, I tested it, and whatever sample I put in the syringe comes out when I push it.

  • but I still don't see any signal whenever i inject a sample. 

  • I found the problem. It turns out that the Methanizer was clogged. The other end of my column was not connected to the FID directly. It's connected to a tube that goes to the methanzer first, then from the methanizer, it goes to the FID. I bypassed that and connected the column directly to the FID and it worked. now, I see that my peak is tailing, so I have to find where the issue. but any idea why I see solid white particles on the septum and the liner. I did clean the liner before, but once In a while I still see those particles ?

Reply
  • I found the problem. It turns out that the Methanizer was clogged. The other end of my column was not connected to the FID directly. It's connected to a tube that goes to the methanzer first, then from the methanizer, it goes to the FID. I bypassed that and connected the column directly to the FID and it worked. now, I see that my peak is tailing, so I have to find where the issue. but any idea why I see solid white particles on the septum and the liner. I did clean the liner before, but once In a while I still see those particles ?

Children
  • Particulates in the liner.  Most often they are reddish brown the same color as the septum. Those happen when the septum retaining nut is overtightened and the needle tears the septum instead of just piercing it as can be seen in the pictures below.

    If the column is not a Porous Layer Open Tubular (PLOT) column, the particles are from the sample itself.  That stuff could be contributing to the solvent peak tail.  You should perform all of the normal user maintenance of the injection port including replacing the septum, liner, liner O ring, and the gold seal at the bottom of the injection port as described in the hardware manuals.

    and also use a solvent dipped cotton tipped swab to remove any residue here, up inside the retainer weldment.

  • Well I did not see the reddish/brown particles. What I’m seeing is white particles all the time. I have PLOT column. Does that make a difference ?

  • Porous Layer Open Tubular columns - have a layer of particles attached to the inside of the fused silica column tubing.  They come loose for a variety of reasons and that is what you are seeing. 

    You must....must... Set the inlet pressure to 0.00 and let it reduce all the way to zero before changing the septum or opening up to change the liner.  If you don't, the pressure in the column will be released upwards when the septum nut or liner nut is loosened up and any loose plot particles will be blown up into the liner.

Was this helpful?