1.2.3 go… Looking for feedback for preferred systems to use for in house analytics. 1. GC/FID 2. GC/ MS 3.HPLC. I know hplc seems to be the industry standard and most popular but what’s the justification behind that? I’m looking at Agilent systems across the board. Trying to determine best piece of analytic equipment for us as a hemp lab. Priorities are: 1. Accuracy of analytics 2. Initial investment and depreciation value 3. Modularity of equipment.
I’m going the GCMS route right now. I’m skipping all that “retained value” by jumping to the bottom of the depreciation curve (just bought a cp3800/saturn 2000 for $4500). Lets you do cannabinoids, res solvent, and terpenes in one system, gives you the benefit of confirming via mass spectra. On the other hand, generally more complex and certainly less common for this application.
6 Likes
5 Likes
I totally respect the survey at 43% hplc but why? Is it possibly a trend? Other than easily detected acid forms of cannabinoids what’s the advantage?
1 Like
Manufacturer support; general principles for separating nonvolatile analytes (especially labile ones); generally easier automation; simpler detection mechanism
Just some guesses, like I said I’m going the GC route
2 Likes
Liquid Chromatography for me - has been easier on sample preparation. I’ve done TLC, HPLC, GCFID, and GCMS for this work. @SidViscous is right - the equipment on the GCMS side of things will hold its value longer and can be used to do more types of tests than a standard HPLC.
The HPLC will be a faster run (think 10-15 minutes a sample) with the ramp on the GCMS taking longer than that for good separation. GC temperatures are a lot higher, so you won’t be able to see Acidic forms as easily (they are still there, but will also degrade and convert on the column at higher temperatures in the GC).
Support for an HPLC is A LOT easier than on a GCMS. The DAD basically drives itself and you don’t really have to know anything at all about the detection method to make it work. MS is more complicated but you can do more things with it for sure.
Changing a column out on an HPLC is minutes, seconds even. Changing one out on a GC is a good 15 minutes of your life, so you don’t screw anything up.
The electricity needed for your GC, plus the carrier gas and vacuum also means you have to have more space, more utilities, and a good supply of helium.
You don’t need those things on the HPLC - it runs with easily available materials, doesn’t need hardly any electricity or other supporting gasses/utilities.
Probably there are other reasons people like HPLC or GCMS better for one reason or another.
I’m down to chat more about it, if you have other questions. Both instruments will get you your answers - but will you be able to support an HPLC faster and cheaper without knowing too much about how the instrument is supposed to work.
Let us know which one you decide on and why. <3
10 Likes
My understanding is that terpenes don’t work very well with mass spec because many are isomers that will have the same fragmentation patterns. Based on that, gc-fid is preferred for terpenes
Hplc is non-destructive sample analysis and for cannabinoids that means you can measure the carboxylic acid forms directly without a derivative step. This is also why, when coupled with a QQQ, hplc is used to separate mixtures with up to 60 pesticides.
One advantage of gc is the mobile phase is some carrier gas like he or n2 instead of organic solvents like hplc so less of a waste issue there.
Changinng a column on a gc is more involved but I do know of the intuvo 9000 from Agilent which has a nee feature that makes column swaping take a few moments.
2 Likes
With GC-FID you only get a retention time. With GC-MS you get a retention time and mass spectrum. So if your GC system is able to separate all the terpenes to give individual retention times (baseline resolution) then it doesn’t matter if the detector is FID or MS. The one advantage of MS/MS (or even MS with SIM) is that you don’t have to fully baseline resolve all the compounds (but with terpenes you usually have to have baseline resolution due to them generally having very similar fragmentation patterns, which you pointed out).
The SRI 310MM GC is $10810 brand new with a two year warranty. It tests for cannabinoids, terps and residual solvents. Cost per analysis is about 15cents. Electricity consumption is about 400 watts or about 4 cents per hour. Compared to HPLC that is 100 times lower operating cost. Its easy and cheap ( 10 extra cents ) to derivitize the sample if you really want to differentiate between the acidic and neutral cannabinoids but in many cases its not really important to do that and it complicates the analysis because then you have to calibrate twice as many peaks ( THCA plus d9THC etc ). GCMS is wonderful technology but overkill for this test. Just turning it on takes all day.
Hugh
SRI
7 Likes
One more point is that it really only takes about 60 seconds to change the GC column if you have done it once before. With the SRI 310MM GC there are two columns permanently installed, one for cannabinoids and terps and the 2nd for residual solvents. You seldom have to change columns.
Hugh
SRI
6 Likes
HPLC-DAD. You can easily distinguish acids from neutrals by their UV spectrum. Without that capability edibles would be hard. You could of course derivatize acids and stabilize them so you can measure them with GC-FID, but that’s a pain in the ass. In addition, if you want to do tinctures and such HPLC is a much better choice. Injecting MCT or olive oil etc into a GC of any sort is just asking for problems.
GC-MS is much less reliable and less reproducible (ion suppression etc). Cost per sample would be much higher once you factor in disposables, maintenance and education level of the operator. My hat off to anyone who buys a $5K GC-MS on ebay and actually manages to get good data out of it.
2 Likes
Get you hair ready to show the world
Lol I hope. We’ll see how it goes
4 Likes
I’ve used an SRI GC for a couple years. The customer service is great, and for most standard testing it’s (likely?) The easiest and most straightforward system. Coupled with the training and support they offer it’s probably the best bang for buck (other than refurbished). It’s certainly not the best for acidic cannabinoids, but it’s got great separation for majorcannabinoids without needing a ton of knowledge or fine tuning.
6 Likes
I’ve found that the issue with using a GC/MS rather than HPLC is the instrument to instrument variation can be shockingly large even negating sample prep differences. Pretty much all analytical labs use HPLC because of the reasons that @swell mentioned. As a result, you are going to find that while your results are technically correct you’ll have some discrepancy in regards to potencies that will be reported on your official COAs from an external lab.
Analytical chemistry is like that, unfortunately. The question is not “is the call correct - are you sure this is the correct potency?” the question is “can you defend this call?” and since these analytical testing labs can defend their call due to their HPLC method, you’ll find your results do not always match and you nor the lab is wrong. That can be a pain when your GC/MS is saying the distillate is 91% and the analytical lab is reporting 84% via HPLC. (And unfortunately, making the point that the mass spec is a stronger detector won’t matter).
That’s sometimes hard for people (not necessarily you- I’m sure you got this) to wrap their heads around.
3 Likes
Are you sure about that? You yourself just encountered a major co-eluting peak situation on the SRI…
3 Likes
@SWELL I think fresh solvent and then the associated disposal costs need to be factored in to the $ per run. I would also argue that GC is a more simple process than HPLC, but maybe that is just my brain.
Edit: Brief search of ion suppression in chromatography → MS, it seems to be more related to LC? Solvent matrix effects.
Haha fair. The d10 and CBD was the first compounds I’ve seen co-eluting so far. I was going to give SRI a call to see if they had better parameters or a different column 
Reproducibility will always be less with MS. Just inherent to the technique.
You have more steps: Ionization > separation > detection.
FID and UV are much simpler and just measure an electrical signal.
MS is great for trace analysis and identification because it has extremely high selectivity, but reproducibility is always less then FID or UV.
Check out Agilent 1220 Autosampler + Cann-ID Software & Lab Integration- Call Fo – Ionization Labs
The company offers an HPLC subscription/turnkey solution for potency testing.
If you have the standard 15MXT500 column, then try the 15MXT35 column. I was able to get a 15MXT35 from sri that should help with the co-eluding peaks near cbc/cbd/d10. Tbh though, I have yet to run it due to time constraints. Happy to make that a higher priority if you are in need of a quick answer, lmk.
2 Likes