I'm in way over my head - Cannabis testing lab setup

CannaBusiness
1

Hey everybody, I’ve been a longtime lurker and only decided to sign up and finally ask the pressing question today.

I owned a hemp company, of which I’ve made a glorious exit very recently. Part of the deal is that I stay on board for 6 months part-time to aid throughout the transition.

Here’s where it gets interesting. I was planning to outfit the business with an in-house lab as our company handled a high volume of products. It seemed that my pathway to do this was simple, but it’s much more complicated than I originally anticipated.

The original plan was this: HT2800T Headspace sampler + Agilent 6890 along with a 5975 MSD. From my research, it appears that this will check most of the boxes in regards to a “full panel” in-house test. Now, this setup appears to be great and all, but I didn’t realize that they aren’t as beautifully set up and “idiot-ready” as the newer Shimadzu or Agilent equipment is.
They want to test for Potency, Terpenes, Residual Solvents, Heavy Metals, Pesticides and microbials

With a $100k equipment budget, is there a solution that would be a better choice than the equipment combination mentioned earlier? We considered some of the Agilent equipment with the E-Methods and Shimadzu has cannabis-specific programs, which are relatively user-friendly. I’m open to suggestions! I’ll be on board for the next few months and I want to deliver on this.

Pat, the new owner suggested today that if the lab setup goes well, he’d love to get into 3rd party testing for other facilities. Not sure what’s involved with that regulatory-wise, but that’s not my problem. So equipment that could potentially port over to that side would be ideal.

They will be hiring a full-time lab technician(s) but they’re looking for user-friendly so that management can be trained on the equipment. My biggest roadblock is this isn’t my field, but I made one too many claims that I can handle it, something I deeply regret. The number of mix & match units are simply overwhelming.

Thanks for the help guys, you’ll save my ass on this one. Pat knows I’m underwater here, but I’d like to knock this out as my final task.

2 Likes
2

For potency testing, you will want HPLC, not GC. You can use GC for terpenes and residual solvents.

At $100k you do not have the budget for Heavy Metals and Pesticides, and I wouldn’t really bother, unless you are trying to start a brand new testing lab enterprise that is doing more than just in-house testing. If that interests you I know someone who has a brand new full setup, still in boxes for that. DM for price tag.

What state is the hemp lab in?

5 Likes
3

You can use the GC for potency, although the sample preparation won’t be as straightforward as cannabinoids by HPLC because the acidic cannabinoids will be converted during the injection/oven ramp and would require a derivitization step. You’ll probably have to use different columns and detectors dedicated for your different analyses: one for cannabinoids, one for terpenes, one for solvents, and one possibly for some pesticides (along with the right kind of detector).

On this budget unlikely you will be able to do all the tests mentioned (potency, solvents, terpenes, micro, pesticides). The mostly costly will be the pesticides. Most labs in the rec states are using LC-QQQ for test for 60-70 pesticides. You could do some of the pesticides on the MSD, but won’t be as comprehensive as on the LC QQQ. You will have to make a determination for what pesticide analytes are most important for you to test, determine what instrumentation could reliably perform validated testing for those analytes , and base your instrument acquisition based on that requirement.

Heavy metals are done on ICP-MS and probably won’t be within your budget.

Don’t forget the instruments are just the start. You’ll need ancillary equipment (balances, pipettes etc), reference standards, fridge for storing standards, software, consumables. It all quickly adds up.

8 Likes
4

Thank you, I would have to agree with this comment. The budget has been the number one roadblock thus far and the conversations have evolved from simply testing potency & terpenes to a full panel. With that, the budget has risen to it’s current value of $100k.

The business is New York based. One of the worst states to do business in, in general.

5

This was a very thorough comment, it’s greatly appreciated!

Originally we planned HPLC with an Agilent 1100 simply out of cost during the infancy of this laboratory testing pipe-dream. Once we realized the troubles with maintaining and integrating an 1100 we trained our focus elsewhere and landed on the GC system listed above. Only for the fact we’re reducing our equipment costs overall.

I don’t know how I missed the ICP-MS aspect of Heavy Metals. Thank you very much for bringing this up, it’s noted and we’ll absolutely be passing on this!

If you had a $100k budget and you wanted to hit as many “Targets”, what would your suggestion be?

We’d be buying second-hand from LabX or similar. I was very impressed with the Shimadzu systems and if I had my way we’d simply go for an LC2030c and a GCMS TQ8050, but the GCMS unit’s being quite out of our range. I’ve been waiting to hear back on several used/referb units, though. Who knows, I might get lucky there!

2 Likes
6

I can’t speak to Shimadzu as I have never operated one, but we have two Agilent 1100’s in our R&D lab and they aren’t really that hard to maintain and usually operate flawlessly with regular use. The only catch to the older systems is getting a computer with chemstation software on it.

But you can usually find a refurbished system with a computer with working software for around $25k. Get that and maybe a dual oven GC for about the same price and now you’ve got terpenes, solvents, and potency testing with $50k left over.

What you really need to budget for is an analytical chemist to run all of this and be able to troubleshoot as needed. That’s gonna cost you about $50-70k/yr.

2 Likes
7

Seriously, ???
You shouldn’t be even thinking of doing in house analytics as you propose. Your budget wouldn’t even cover the chemist’s salary
to set the lab up, get everything up and running/calibrated.

Why do you not decide…exactly what your hemp product is /
going to be: the one that makes you MONEY…
when all the production bills are paid…and you have your “profit” in hand,
then decide all the BS about whether you want to be in the “3rd party”
testing business.
right now you only need to have equipment necessary to make that product and sell it…the rest is 100K dream chasing 800K set up.

If your plan is to process HEMP…biomass…get out your pencil and pen.
worry about the GC-MS latter.

In NY state?

2 Likes
8

I was the QA manager of a full-service testing lab for a long time and would be happy to consult on this project. Everyone correctly observed, it is not possible to do all these functions on the listed budget, even buying used. The first thing to do would be to get in a meeting with whoever is making those budgeting decisions, make them recognize that reality, and then prioritize the few core functions you want to make sure you can do.

8 Likes
9

@srihugh1 could probably help you out

10

HPLC, and whore the rest out to other labs with larger budgets.

HPLC analysis takes time to learn or some uni.

If learning isn’t a possibility in time to meet your in house testing needs, consider lowering your LOQ standards and getting a Purpl or similar.

1 Like
11

Probably GC simply because you have the option to do residual solvents and terpenes, neither of which are feasible on HPLC. Of course you will need appropriate detectors and columns. For terpenes you can use FID. For solvents you could also use FID but I think a MS would probably be better. You can also do cannabinoids but the sample prep will be an added step because you wont be able to measure the acid cannabinoids directly. Depending on your MS configuration you could do some pesticides but honestly pesticides for in house is beyond the scope of most.

Honestly in the rec space people don’t seem all that concerned about cannabinoids and even for compliance testing it seems like a rubber stamp as I still see flower being reported with greater than 30% delta 9 so its a bit of a joke. The hempers seem more interested in cannabinoid content and HPLC really is the best way to test cannabinoids.

2 Likes
12

HS-GCMS is powerful as hell and a nice setup as far as consumables (vs LC)… well until we all run out of helium but that is for another day. It is the instrument I would pick if I could only have one. But you are going to want to prioritize the testing that is most important to the purposes of the business.

Honestly, a full pesticides panel (all state mandated pesticides) is out of reach. Maybe you could work out a limited panel of the worst offenders for your own QC, but this will never be able to do all the state mandated ones. Pestisides would require a specialized GC column.

Residual solvents are also very tricky to test all the mandated ones, but you could easily come up with your own panel for internal QC. Terpenes are actually pretty simple once you get the right method dialed in. Solvents and terpenes can be ran on the same column, maybe not on the same method but same column.

Potency is straightforward if you are ok looking at the sum of acids and neutrals, but you might need to do direct injection… so requiring another injector system but that is pretty cheap compared to HS and DI might just expand your pesticides options. Not sure the column required for potency.

Definitely metals is out of the question but that is an easy contract out to another lab. Would NY accept in-house testing for state mandated tests? Id think they would require 3rd party tests, or is this all for internal QC?

2 Likes
13

Purpl Scientific is a joke. We had a friend drop one off in our lab this summer and we tested it against our calibrated HPLC system. Basically Purpl is a random number generator. Some people swear by it, but that doesn’t mean it works. It’s a fancy paper weight that generates random numbers so you can haggle with biomass sellers in the field, and the guys selling it to you don’t know enough about the magic Purpl device to argue with you about it’s accuracy. That’s all it’s good for.

9 Likes
14

The Agilent 6890/5975 GCMS - with both a headspace HT2800T and standard autosampler 7683 injector can do Potency, terpenes, residual solvents, and pesticides. This thing is a beast and you can beat it for years and it will still show up to work for you. This is a machine for humans that actually understand chromatography… so the plan for management running something… It is user friendly - one of the first machines I set up. The software from Agilent is also easy to use, configure, and report from. For someone who has any clue about chemistry and computers.

Microbials you can do with an incubator and plates, get yourself a clean place or duplicate your plate blanks / negatives.

Heavy metals you are not going to be able to do on that budget. Its just not in the cards, even the older technology is going to cost you more.

Don’t forget the other things you will need - Ion is leading you the right direction.

Analytical balance
set of calibrated pipettes
volumetric glassware
hood to do the work in
vacuum pump for your GCMS
Solid weekly/monthly supply of solvents
monthly supply of GC gas (don’t skimp, it matters)
Centrifuge
Filters
autosampler tubes
pipette tips
media bottles
freezers
some kind of cannabis muncher to grind all the things to bits
etc.

@ionstorm is super correct - while you can do these tests on the same machine, the additional consumables needed to get there may make having more than one important. People were doing pesticide testing without triple quads for decades. So its possible - but your LOQ may not be where you want it to be. With just one machine you won’t be able to be commercial - because you won’t have enough capacity, IMO. -shrug- And the “plug and play” methods are not going to work for you - you’ll have to modify something to work for your equipment, which means longer time to get up and running.

Its possible. But…

This quote is so very true. You’re not going to get me in your budget. Probably not any of us that do this kind of work. So if you don’t have the skills on staff already - where is that going to come from? Paying the high prices for plug and play, single button GO instruments comes with a cost because you don’t have to spend time on methods or maybe even time on educated people.

Good luck - and please let us know what you decide to do. :smiley:

10 Likes
15

It would be more accurate to say people were able to test for some pesticides. You won’t get the whole list with GCMS alone. Some aren’t even able to be volatized at all.

5 Likes
16

Can u maybe list some super heavy ones @MagisterChemist? Interesting.

1 Like
17

Gcms can do around 118 pesticides. Some people only need to look for a couple. I was pretty specific about not using headspace.

He hasn’t even said which ones he is testing for yet. Does hemp in New York have a special list?

Some can only be tested on gcms as well. Some states only have a list of like 8 compounds, which can definitely be done with gcms.

1 Like
18

I checked. The New York list is super short. But you have to be a specially licensed lab anyway using their methods.

One of the methods uses gcms. The other 2 methods use lcms.

That’s for medical. I didnt find a special list for hemp or conversions.

1 Like
19

Spinosad, Abamectin, and Azadirachthin (found on NY list) are three that are completely undetectable by GC. Some others are detected but with extreme unreliability. I remember Naled and some others being in that category. That’s of the top of my head but i believe of the Oregon list, we determined something like a dozen of them were not amenable to GCMS at all.

5 Likes
20

@Studio42 hello friend! I just wanna share this to you, and maybe you can get an idea on how to set up from it:) https://www.sunwestgenetics.com/cannabis-greenhouse-sheltering-your-weed-plants/