Can someone help me with this chromatogram

[Chaboes]

@tokesandtinkery

If I wasn’t clear, my intention would be to never contaminate the standard, and should always be transferred via disposable pipette from the standard vial to a clean vial; and the standard vial is to be kept in the freezer.

I just meant in terms of measurement variability. The more steps in a process such as weighing from one vial to another introduces a small amount on measurement variability (or uncertainty). This is basically why measuring out a large weight once is more accurate than the several small weights being added together to get to the large weight. This variability is based on operator error and instrument (pipette).

@tokesandtinkery

the 264mg ITSD figure doesn’t need to be specific and should be formulated for simplicity and accuracy.

Correct, just needs to be constant and fall into the happy range of your detector. I tend to make mine a number that makes calculations easy down the line.

@tokesandtinkery

all standards, including the terpenes, should be derivitized at the 5 calibration points with an equal amount of itsd per calibration point. (As best as possible anyway).

Kind of, yes for cannabinoids, not sure on terpenes. This may be a better question for @Dr_Jebril as I know they test terpenes on their GC.

Terpenes and cannabinoids could exist at different orders of magnitude within the sample. Lets say you are testing flower and concentrates with a typical potency range of 8% to 99%, and lets make the simple assumption that you made a method where 0ng/µL-100ng/µL can be translated to 0-100% potency in the sample. You calibration curve works well for the flower as you have calibration points at 0,10,25,50,75,and 100%. BUT, what if the terpenes are only present at 0-1% having a calibration curve which they lowest point in 10% is not very helpful. A good rule of thumb is you need to have a calibration point at or below your limit of quantification (LOQ), so it would be a bad quantification to try to say “this terpene is at 0.3%” with your lowest calibrator being at 10%. So you would want to design your cal curve for terpenes on a smaller scale at a lower range.

This is all method validation and development and you will learn as you go. A 5 point calibration curve is standard but often you need to have additional cal points near your lowest quantifiable values.

@tokesandtinkery

Am I correct in understanding I only need to do this this process of calibrating with standards when my internal standard solution changes (presuming I am imperfect in measuring/formulating)

This is based on your own internal QC measures, you could get relatively accurate results between IS solutions as long as you are making them consistent.

@tokesandtinkery

Can I run a cannabinoid and terpene test at the same time on the 35Sil MS column?

@Dr_Jebril

@tokesandtinkery

Anything obvious I’m missing, or any tricks that make these processes easier? I’m assuming the calibration will be the biggest thing to get correct for accurate results

This, and knowing your sample prep inside in out especially with a derivatization. Make sure to do a proper validation.

[SidViscous]

@Chaboes and @tokesandtinkery you guys are awesome and this discussion has been super helpful for me too! Great questions and great answers.

[tokesandtinkery]

By the time we’re all done we’ll have a nice little sop to open source

[SidViscous]

Yes we will! Hopefully something worthy of a pin in the testing category

[tokesandtinkery]

@kcalabs @iontrap @Dr_Jebril @bigbone

Any input you can share on what you use as calibration points for cannabinoids and terpenes [on FID and/or MS]?

[SidViscous]

Don’t forget @Cassin and @MagisterChemist!

[tokesandtinkery]

Also @thesk8nmidget! I’m sure I’m missing others!

[thesk8nmidget]

Do you mean our retention times for different cannabinoids on our methods? or the area were getting while calibrating?

Either of these would dependent on your column, your temp profile, the dilution amount in your method and even the current temp/climate in the room with your machine.

Here is a sample chromatogram from SRI using their MXT500 and MXT505 column.

[tokesandtinkery]

Sorry, to be specific; what ng/ul calibration points do you use? particularly interested in how it may differ for terpenes

[tokesandtinkery]

Hoping to get input as to how this looks for formulating the standards for 1ul injections.

All of the following are based on Restek standards (1000ug per cannabinoids, 2500ug per terpene).

1. Formulating the ISTD solution

• Combine 20mg methyl stearate with 1000ml acetone (784.58g)
  1ul = 20ng ISTD

2. Formulating the Dilute ESD cannabinoid calibration solution

• Combine 1ml (remove 1ul) acetone with 1ul Restek ESD cannabinoid standard
  1ul = 1000ng

3. Formulating the working ESD cannabinoid calibration solution

• Combine 2ul dilute ESD cannabinoid calibration solution with 8ul acetone
  1ul = 200ng

4. Formulating the working ESD cannabinoid calibration point solutions

• 100ng/ul: Add 1ul working ESD cannabinoid calibration solution to a vial.
• 75ng/ul: Add 1.5 ul working ESD cannabinoid calibration solution and combine with 0.5 ul acetone to a vial. (2ul = 300ng. 1ul of solution = 150ng)
• 50ng/ul: Add 1 ul working ESD cannabinoid calibration solution and combine with 1 ul acetone to a vial. (2ul = 200ng, 1ul of solution 100ng)
• 25ng/ul: Add 1 ul working ESD cannabinoid calibration solution and combine with 3 ul acetone to a vial. (4.0ul = 200ng, 1ul of solution = 50ng)
• 10ng/ul: Add 1 ul working ESD cannabinoid calibration solution and combine with 4 ul acetone to a vial. (5.0ul = 200ng. 1ul of solution = 20ng)

5. Formulating the working cannabinoid calibration standard:

For each calibration point, add 1ul working ESD cannabinoid calibration point solutions, add 1ul of ISTD solution

6. Formulating the 100ng test calibration solution

• Combine 2 ul terpene calibration with 25ml acetone (remove 2ul), 1ul = 200ng

7. Formulating the working ESD terpene calibration point solutions

• 100ng/ul: Add 1ul of 100ng test calibration solution to a vial.
• 80ng/ul: Add 4 ul of 100ng test calibration solution with 1ul of acetone to a vial. (5 ul = 800ng. 1ul = 160ng)
• 50ng/ul: Add 1 ul of 100ng test calibration solution with 1ul of acetone to a vial. (2ul = 200ng. 1ul = 100ng)
• 25ng/ul: Add 1 ul of 100ng test calibration solution with 3 ul of acetone to a vial. (4ul = 200ng, 1ul = 50ng)
• 12.5ng/ul: Add 1ul of 100ng test calibration solution with 7ul of acetone to a vial. (8ul = 200ng. 1ul = 25ng)

8. Formulating the working terpene calibration standards:

For each calibration point, add 1ul working ESD terpene calibration point solutions and 1ul of ISTD solution to a vial.

9. Derivitizing the working calibration standards:

For each working calibration standard (terpene and cannabinoids), add 2ul of MSTFA (clean syringe with methanol then acetone immediately after use)

[OBXtracts]

@tokesandtinkery

• Can I run a cannabinoid and terpene test at the same time on the 35Sil MS column?

This article specifically is in regards to testing terpenes and cannabanoids at the same time

molecules-25-05872-v2.pdf

Such a methodology is certainly very
detailed and useful when advanced studies have to be done, like in chemotaxonomy. For routine
analyses, however, such a setup is probably too complicated and expensive.

For your standards though… I’m basing this on my experience with SRI’s peaksimple specifically.

@tokesandtinkery

1. Formulating the ISTD solution

• Combine 20mg methyl stearate with 1000ml acetone (784.58g)
  1ul = 20ng ISTD

If following the crude “1gram to 1 gallon” for internal standard solution of methyl stearate, 20mg may be too little for 1L of solvent. ~200mg would be more ideal, but really depends on what your sensor/setup reads most consistently. I wouldn’t use such a low amount for internal standard. Any variation in baseline can significantly change the area count on such a small amount.

@tokesandtinkery

2. Formulating the Dilute ESD cannabinoid calibration solution

• Combine 1ml (remove 1ul) acetone with 1ul Restek ESD cannabinoid standard
  1ul = 1000ng

Restek triple calibration standard (1000ug/ml) is 1000ng/uL.

If 1 μL of the 1000 μg/mL stock solution is diluted with diluent to a final volume of exactly 1 mL, the resulting concentration of the diluted solution will be 1.00000 ng/μL.

Here’s a handy little calculator to help

I do a 4, sometimes 5 point calibration. This is for SRI stuff but the math should be the same, at lest for internal standard.

Here’s my procedure and calibration.

For internal standard solution- 1 gram to 1gallon

1000ng- 100uL Restek Standard, 100uL Internal standard solution
500ng-50uL Restek Standard, 50uL Solvent, 100uL Internal Standard Solution
250ng-25uL Restek Standard, 75uL Solvent, 100uL Internal Standard Solution
50ng-5uL Restek Standard, 95uL Solvent, 100uL Internal Standard Solution

Again this is for internal standard calibration, which is a representation instead of an exact feed.

You are telling the machine, “hey machine, when you see this X amount of area of internal standard solution, and Y area of analyte, it represents this specific X/Y ratio”

[Cassin]

@tokesandtinkery

Any input you can share on what you use as calibration points for cannabinoids and terpenes [on FID and/or MS]?

100ng/ul is seriously dilute! That’s only 1ppm. Are you always diluting everything you are testing to where your always below 1ppm? If not - you should consider that you will often have results that will be more than double your peak height - and those results should be disregarded and an additional dilution run.

I use 10 times my LOQ as my CCV for terpenes on MS as my calibration point. I have a 7 point calibration for those which has usually provided me solid linearity and consist pins. If I had all the time in the work I’d use point 3 and point 6 on my curve every time but I don’t so I stick with ~3.5 instead.

For cannabinoids I have a 3 stage dilution - each dilution level has its own calibration, and its a 13 point calibration curve (because minors are small and majors are not) so that I can make sure to see things accurately. I run the minors separately from the majors on the upper end (aka dilutions 1 and 2). I don’t make products that have minors that are above 10%, so starting them at dilution 3 works for me.

I don’t run FID (not because it doesn’t work, but because I don’t have the instrument) so I cannot speak to that. And I’ve found that I can run using this same thought process on for cannabinoids on HPLC, but not for terpenes. I also found that for terpenes - pushing headspace was helpful often. And when I did that, cannabinoids did not work as well. So there are other things to consider beyond detection method.

@tokesandtinkery

Sorry, to be specific; what ng/ul calibration points do you use? particularly interested in how it may differ for terpenes

I’ve used the same method for my starting point for calibration my whole career. I start with a 10x dilution of whatever is available as my reference standard. That’s my top point. Then I create at least 6 points below that (7 point cal curve) and each step down is 25% of the previous point. Then I run it and check for linearity - if I’m good, huzzah, I have my cal curve. If I’m not good - then I do that same process and either go 20% or 30% instead (depending on if things were high or low) and I add in extra points.

Also - with cannabinoids at this point I don’t worry about starting new. I go with the established AOAC method, which defines how calibration should occur. For terpenes - I actually use the FCC and USP for each of the specific substances (since they exist outside of cannabis) and go from there.

If I’m getting a mixture of standards - then I go with the method as provided by the standard supplier, its almost always compendial - sometimes its specific to their mixture. Sometimes this is publicly available, other times I just ask them for it. Its the method they are going to use for in-house quality control. No reason you shouldn’t be able to use it.

@tokesandtinkery

Hoping to get input as to how this looks for formulating the standards for 1ul injections.

My injection volumes are higher than yours. And I 100% agree with @OBXtracts that you’ll probably want more IS around, just to make sure its always noticed, regardless of subsequent dilutions.

I think your math looks okay - and it looks like you are sort of following my 25% of the previous dilution method. I personally start with more points - because you never know when you’ll have to drop one (or two) because of instrument issues beyond your immediate control.

I’m a bit crazy when it comes to calibration though - so probably 5 points is okay.

Let us know what your linearity looks like when you run this. I bet it will look just fine.

[srihugh1]

I think you might be over-complicating this.
Call for a quick phone call and I think I can simplify this.
Hugh
SRI
310-214-5092

[tokesandtinkery]

Attempting to replace the liner and injector parts. Broke the carrier gas line.

Guess today is a good day to take apart the parts machine…
Going to take the injectors and FIDs from this unit

and move to

My series ii

It would probably be less work to move the computer parts but the physical box of the parts unit is in bad shape

[tokesandtinkery]

Ughhhhhhh.

Going to buy some spare splitless injectors. Might as well…

[SidViscous]

If it makes you feel better, I think the ethernet card on my GC is bad…

Also, @Killa12345 do you think it’s probably time that @tokesandtinkery gets their own thread? Lots of great info here and I think others might skip based on the title alone

[tokesandtinkery]

@Killa12345 if you wouldn’t mind splitting this thread out. I’ve got a bit more to dump here shortly

[tokesandtinkery]

Day two down. Complete rebuild of the 5890 nearly finished, a handful of parts needed after this. What follows is a rough attempt at documenting the process of re installing the FID heater block and other pin outs.

This is the stock split port injector that came with my Series II, I used this as a template for cannibalizing pieces from the parts unit which came with purge packed injectors (whoops).

This is how it came out of the parts unit. I would end up disconnecting the regular an copper, from the other manifold in the picture, i would take the purge port (gas outlet port under the 2 black knobs to add to this one. I didn’t want to risk using the wrong part, so I took the part number for the missing regulator and popped that into eBay

This is the first time I’ve ever re terminated a block like this and I undoubtedly made some mistakes. Best as i can tell, the process that worked for me was to push the key into the pin, with some slight twisting and pressure it would pop out.

Here I am removing the existing wiring for the heater blocks from the purge packed injectors, starting with injector B, and moving to injector A.

I am unclear if the order of the wires matters here. I did not make any effort to track during this process.

Using the tool, push the pin from the new injector block into the terminal block.

Here I left the original connector block (white two wires on top) which connect to the temperature sensor in the oven, the second block is from the assembly with the two injectors. I will be removing the oven temperature wires and moving them to the other block while also removing the injector wires and keying in the new ones.

Not pictured: i added 2 additional wires for the injector b purge solenoid, and an additional 1 wire for the ignition for the detector b. You can see these wires, unterminated hanging off the right side.

The terminal connectors (dupont or some close knock off) i had did not fit exactly correctly into the existing terminal block. This lead to me having to plug the block in, then the subsequent individual wires to make contact with their pins.

Fingers crossed and a lot of hope that this works.

One of the things the parts unit had that my Series II did not have, was this pressure regulating manifold, it seemed like a good idea to add that in. Here the manifold is painstakingly bolted down via a 1/4 wrench. This was a shoulder killing process

Here the regulators are added back to the manifold and the gauges reconnected. This would eventually be disconnected for hooking up the gas lines.

Installation of the FID flow control blocks. Added the injector A manifold and solenoid, connecting the purge ports to the solenoid to help secure it in place.

Second injector manifold (incomplete) in place.

Here I have begun hooking up the gas lines. Made a few mistakes and had to re-do this only to come to a final conclusion I’ll have to replumb it anyway, as the copper I re-used was not configured for a dual port split port injector setup and my setup will require 2 additional splits off the carrier gas to the split injector purge inlets.

Closer view of the regulator, one line going to the back for the input and the other hooking into the flow control blocks.

Side view with the 3 gas lines all connected (still need to split the carrier)

At this point, nearly 4 hours later I was running out of steam and was less diligent about taking photos. At this point I began to hookup the injector purge and carrier lines for the injectors. At this time I also decided I should do the injector maintenance that started this whole process. Here is the replacement gold seal

One of the replacement injectors, I don’t have a connector part for this, another ebay part for 1/16 compression to 1/8 compression adapter. I’ve also ordered 2 additional injectors as spares in case this happens again.

Injectors in place, Septa’s still need to be replaced, mostly this is in place to protect the new glass liners I placed in there.

What a task this has been. I am grateful for it, along the way I have learned how to somewhat successfully reterminate terminal blocks. Additionally, the Series II came with 1 FID injector; this process will give me 2 injectors and detectors so I may have the rxi-624Sil and rxi-35Sil columns without having to swap them.

Unfortunately at the end of day two I am left with just as many parts to order as I was last week. These include:

1/16 to 1/8 adapter
1/8 copper fittings, tees and tubing, ratcheting tubing cutter

As I prepare for what will be required to run the columns, particularly looking at the chromatographs provided by Restek I started to question, how do I reliably test the flow which lead to two more ebay purchases:

Agilent ADM2000 and a part for the flow injector to hook the ADM up to.

I’ve got two more parts in the mail I’m waiting on, a solvent caddy (not entirely sure I needed it for my setup but figured for $1 why not) and a injector insulation cover.

I didn’t document, but I took my ultrasonic leak detector to each of the inlets and all of the fittings. Even with improper fittings in places I was able to get leak free at 120psi from the compressor.

Still to be done:

• reinstall the injector fittings
• reinstall injector covers
• reinstall detector fittings
• replace septa on injector inlets.
• perhaps use acetone and a kemwipe to clean the injector. I’m not sure yet
• reinstall the columns. Use microscope to verify column cut meets standard. Intending to install column with 4mm above ferrule
• with everything hooked up, flush with dry nitrogen
• hook up the 4 inch duct and blower fan to exhaust heat from the GC (considering it is in my closet). I think I will get a heat barrier to tape to the drywall behind it as additional protection
• hope it all fucking works when you turn it on next

[tokesandtinkery]

Day three of the rebuild…

Ebay to the rescue, the missing regulator (hopefully the last one) for the second split port injector.

Installed and in place. Time to start hooking up the gauges and everything else. I had to go digging in the parts machine to find enough suitable cables to get the job done

Purge port connected, regulator hooked up

Now attempting to hookup the purge port solenoid for the second injector

Not the prettiest, but we’re all hooked in! Next is to tackle adding the carrier gas lines to the injectors.

Definitely not using hardware store parts here… @Chaboes
Pulling out the ultrasonic leak detector to track down all the leaks.

Only needed to tighten up a few tee’s to get us leak free (Best as I can tell anyway)

Another ebay special, 1/16 to 1/8 compression union. Hopefully I don’t fuck up this injector…

NGL, this kills me just a little

So after getting everything hooked up, i plugged the unit in to see if it would pass self test… and… it didn’t.

“INJB TEMP RDG” was the fault. To troubleshoot i simply disconnected the injector 2 pins from the temp block and see if the error would go away. Sure enough it does. I added the injector pins back to the block, reversing the order (idk if that mattered) and fired it up.

BTW this message is a real asshole, because it goes “passed self test” to JK here’s the fuckups!

Injector A and B temp readings. I turned them on to see if they would heat up and all looks well.

At this point I am very hopeful I can be testing something soon.

Still to be done:

• install the injector insulation covers
• install the column connector pieces for the detectors.
• reinstall the columns. Use microscope to verify column cut meets standard. Intending to install column with 4mm above ferrule
• with everything hooked up, flush with dry nitrogen
• install the heat barrier behind the GC on the wall
• hook up the 4 inch duct and blower fan to exhaust heat from the GC (considering it is in my closet). I think I will get a heat barrier to tape to the drywall behind it as additional protection
• plug in the auto samplers and set it all up
• hope it all fucking works when you turn it on next

I’m waiting on a refrigerated air dryer (go figure the one I bought didnt work >.>) to add to the compressor lines before hooking everything up to the GC for a test.

Getting closer…

Finished getting all the panels back on, setting everything up as it was before, testing the auto samplers for fit and sure enough, it wont sit correctly. This injector gave be a great deal of trouble and finally broke.

Unfortunately the replacements I ordered hadn’t yet been shipped soooooo. FML

[cyclopath]

@SidViscous

I think the ethernet card on my GC is bad…

sourcing compatible ethernet cards (and drivers) was the biggest issue I had with the HP’s/Agilent’s in my former life as an instrumentation technician…at the time they were at least obtainable, just damn expensive.