Testing cannabis

[thedude]

i bought an agilent 5890 series II GC for testing medical cannabis samples. when i run the machine with any method, i get i large peak followed by one medium peak followed by one very small hump. this keep continually repeating every minute as long as a method is running. I am totally new to gc, so am at somewhat of a loss. can anyone give me some idea what might be causing this problem and what i can do to correct it. i just want to get the machine running with a good baseline. not worried about testing ATM.
i am using FID with db5 cap column.
nitrogen carrier gas, hydrogen and air.

[Chandresh K. Soni]

first please provide the chromatogramm,
please for cannabis choose the column DB-XLB it is more better then HP-5,
also the colum dia and film thickness is very important.

[AICMM]

thedude,

You are looking at a test plot rather than the detector signal. In the signal part of the method you should be looking at sig1 or sig2 for the active detector and not test plot.

Best regards,

AICMM

[DSP007]

thedude,

You are looking at a test plot rather than the detector signal. In the signal part of the method you should be looking at sig1 or sig2 for the active detector and not test plot.

Best regards,

AICMM

Hi all. Too lazy to translate myself, komputer-iron, let it work

Perhaps worse, if a person has never worked with a chromatograph.

He sees a large peak of the solvent and impurities to the solvent
A desired peak can not see because there are not looking. (in first min)

1) Use a slow gradient from 100 C. The first minutes of cut off - there is a solvent. Cannabidiol fly at 200 C.
2) Carefully annealing column (325С reverse gas flow - revese column ) to remove the dirt in the column.
3) Put more of cannabis, to see exactly rush.

[Peter Apps]

thedude,

You are looking at a test plot rather than the detector signal. In the signal part of the method you should be looking at sig1 or sig2 for the active detector and not test plot.

Best regards,

AICMM

Hi all. Too lazy to translate myself, komputer-iron, let it work

Perhaps worse, if a person has never worked with a chromatograph.

He sees a large peak of the solvent and impurities to the solvent
A desired peak can not see because there are not looking. (in first min)

1) Use a slow gradient from 100 C. The first minutes of cut off - there is a solvent. Cannabidiol fly at 200 C.
2) Carefully annealing column (325С reverse gas flow - revese column ) to remove the dirt in the column.
3) Put more of cannabis, to see exactly rush.

If this was solvent and impurity peaks it would not keep repeating through the run. AICMM is right - this is the HP Agilent test signal.

Peter

[DSP007]

i bought an agilent 5890 series II GC for testing medical cannabis samples. when i run the machine with any method, i get i large peak followed by one medium peak followed by one very small hump. this keep continually repeating every minute as long as a method is running. I am totally new to gc, so am at somewhat of a loss. can anyone give me some idea what might be causing this problem and what i can do to correct it. i just want to get the machine running with a good baseline. not worried about testing ATM.
i am using FID with db5 cap column. nitrogen carrier gas, hydrogen and air.

I meant it.
Consider the peaks of impurities to the solvent for the "interesting" - a problem for many beginners. "Oh, got out again! " All through this pass. Skill comes with experience.
Still waiting. I hope he only examines hemp . And he will come back to us.

[Consumer Products Guy]

There are push buttons on the front right of the unit. Choose the signal, like front injector on or FID A. Sounds like you have a non-computerized system, or you could set all this through the software. Like start with def_gc.M

[thedude]

thank you for the replies. i am off sunday monday. that is why the lack of replies from me.
i am the founder/director of a medical cannabis dipensary in british columbia canada, so no it will not be only hemp being tested.
i am new to all of this and the learning curve is like a sheer rock face. i am eager to get a foothold and start climbing though.
will get on it tomorrow morning straight away.
any and all help is GREATLY appreciated

[thedude]

You are looking at a test plot rather than the detector signal. In the signal part of the method you should be looking at sig1 or sig2 for the active detector and not test plot.

that was the problem. thanks.
more questions to follow

[carl.nott]

Opportunity for anecdote and perhaps useful information:

While running semivolatile GC/MS at a previous job I was given a small sample of a green material that was submerged in a vial of DCM and told to analyze it by GC/MS and provide a Tentatively Identified Compound report. I did so (5890 GC, 5972 MS, HP-5MS column, standard EPA-8270 run parameters) and the TIC report came up with THC. The green material was sampled from the floor of our courier vehicle. Apparently our delivery driver spilled their THC applicator in the vehicle and management wanted to confirm the suspected content of said green material.

[Jumpshooter]

Dude:
First thing you need to get in order to qualify and/or quantify THC is some Certified Reference Material. Purchase it from USP (United States Pharmacopeia). See usp.org
This CRM is what you will use as a "Standard" to properly identify the "retention time" of THC. So that when you extract your buds with Methanol or Ethanol you will get numerous peaks; however, only the peak that is a match for the retention time of your USP Standard is "THC".
step 1: prepare a 1:100 dilution of your USP Standard
step 2: inject on GC and observe the peaks.
step 3: extract your bud using ~10 volumes of ethanol per gram of bud, shake, centrifuge, pull the green top layer and dilute that 1:5 with ethanol.
step 4: inject in GC and observe the chromatogram for peaks that match your CRM for retention time.

[MichaelStuart]

Hello,

I am new to the forum and new to Chromatography. I just picked up a 5890 Series ii with FID/ECD for quantifying potency and pesticide residue respectively. I'm using manual injection, and am unsure what kind of deviation I should expect in my results based on sample introduction technique. I think I'm being precise and still seem to vary wildly enough to not trust my quantification (ruling out of course paralaxing errors, bubbles, etc).

On my reference sample I can shoot it twice and have <1% deviation, then go on to shooting samples and my results seem to swing, more wildly as the day moves on.

Would these results still skew if I got an auto-sampler?

Any tips on Method, or injection technique will be appreciated.

Thanks ChromForum!

Mike

[Peter Apps]

Autosamplers usually do produce more repeatable injections than manual, but if you can consistently get 1% difference between two standards your manual technique is probably not too shabby.

When injecting manually do not try to emulate the fast, cold needle injections that are the default on a certain well known brand of instruments. It is impossible for a human to do these repeatably, and they are often not the best anyway. My favoured technique is to fill the syringe to volume with sample (i.e. no air bubbles, no solvent sandwiches or other fancy stuff), push the needle through the septum smoothly, count to three, depress the plunger, count to three, pull out. This gives me usable repeatability.

If you are seeing a trend of increasing deviations through the day I suspect that crud is building up in the inlet or column. What is your schedule for inlet maintenance and column bake-outs ?

Peter

[MichaelStuart]

Hi Peter,

I have been seeing a trend of increasing deviation as the day moves on. I planned to service the inlet after about 100 shoots as recommended by the fellow who installed the machine. I suspected some of the noise I was seeing was perhaps due to me coring the septa early on when trying to get my technique down. I switched the column to my back injector, and that defiantly cleaned things up a bit.

I will certainly try the 3 sec, 3 sec maneuver. I like that it would probably give enough time to vaporize everything out of the barrel, every time. I wouldn't need to worry as much about that throwing off my numbers by more getting vaporized one run to the next by trying to just pull it out immediately after shooting it.

I did bake out the column (290C for 5 mins or so) to see if it made a difference, and it was inconclusive due to my lack of a real data point.

Thanks so much for taking the time to respond to my post. I look forward to trying your technique.

Warm regards,

Mike

[MichaelStuart]

Hello cannabis testers,

I figured I would post to help me in setting my expectations in regards to testing cannabis on GC/FID. I am very interested to know what your experiences are with reproducibility, and anything you may have done to help dial it in. I see so many areas along the way that have been contributing to my problem space, and wanted to make sure I was setting reasonable expectations first and foremost. Thanks ahead of time for any contributions on the subject.

Mike