Transcription
Examining the Challenges of Cannabis Sample Processing, Analysis and the Role of Reference Materials
Hello, welcome to our presentation: cannabis chemistry and chromatography, examining the challenges of cannabis sample processing, analysis, and the role of reference materials.
First, let's look at a typical cannabis or hemp workflow. You have, in the first step, sample selection. And then you take that sample, and you have to reduce the particle size, so sample homogenization. Then you go through a sample processing step. this could be an extraction, or a digestion, or some other processing step, like maybe a filtration. And then you do your instrumental analysis.
Let's take a look at the choices you make when you choose a particular part of a sample. The part you choose, and the results you're going to get from that piece of sample or that part of a sample, really depends on a lot of different external conditions. Like what part of a plant is it taken from, what are the weather conditions or the light effects of the plant that that sample is taken from, geographically where is it taking from. Is it taken from an area that has a higher water content, or are samples taken from a lower water content area, so there could be a lot of variation. It's important to know though that you need to have representative samples. Representative. samples represent the entire population size, so if you have like an outdoor grow you want it to
actually represent all the parts of that grow, all the light levels, all the geographical factors, all the water factors, and things like that. And it must be the statistically valid number of samples, and reasonable for the population. And is considered to be the smallest subset that you can have that retains all the properties of that specific population. Now this can be very difficult.
Think about an individual plant, think about where the variation could be for different target compounds. If you're looking at pesticides and they're being sprayed from overhead, then your buds that are overhead, your leaves, the fan leaves that are sticking out from the plant more and they get more coverage from a pesticide, are going to have higher pesticides. Ones that are more interior to the plant, maybe they don't get as much direct contact, they're going to have less pesticides. Think about things like the cannabinoids. Light has a lot to do with it, so areas of the plant that get more light,
maybe on the top, will have higher cannabinoid levels. If you're looking at things like metals or fiber, well that also depends on what part of the plant you're going to be looking at, are you going to be looking at the herd, or the seed, or the roots, because each one of them has a different chemical profile. So, there could be a lot of variation in one plant. If you're really concerned about the THC and the other cannabinoids, you can actually find them almost entirely throughout the plant, with the exception of some of the roots, you don't see some of the cannabinoids in the roots. But if you look at this little chart you can see that THC has been found in the seed, and the stem, and the leaves, some in the pollen, some in the flower. So, it could actually be found throughout the plant, with the highest concentrations obviously being in the flower. But you also have some high concentrations in the leaves as well. So, there is a lot of variability.
So, what's the difference between a representative sample and homogeneity? As we said representative samples reflect the population, that smallest subset that retains the properties of a population. Homogeneity is more on the batch or sample size, and this means to be a uniform and composition in character. If you think about a liquid, they have high homogeneity. They have a lot of homogeneity because they are very fluid, they can mix, so you get a very blended sample. Agricultural samples on the other hand, by their very nature tend to have low homogeneity. Imagine an apple orchard, no two apples are identical. But if you make the particle size smaller you, actually increase homogeneity. And the converse is true, the larger the particle size the less homogeneity there is.
Let's talk a little bit about sample homogeneity. Sample preparation is a critical first
step for any analytical process. And sometimes you're like well do I really need to get it down to a very fine particle size? I saw a study where they looked at the different uncertainty levels that are required by an analytical process and the different particle sizes. And what they found was if you have a large range for your uncertainty, like ten percent, and you have a five-millimeter particle size, that's the size of about a pencil eraser, you would need about 125 grams of sample to ensure that you have
homogeneity. But if you have a 1 percent uncertainty, you would need 12, 000 grams of sample in order to ensure that. If you take that size down to 0.5 millimeters or below,
you need a lot less sample. There's another reason too. If you look at the particle and if you're trying to extract it or digest it, if the particle is very large you may or may not get to the center of that particle, you might not be able to digest it, you might not
be able to extract it. But if you do have a smaller particle and you digest or you
extract that particle, you have a better chance of getting your solvent or your acid or whatever you're doing into that particle and being able to break it down for extraction or digestion.
Standards can be an important part of your sample processing. You can add standards before you measure a sample, before you do particle reduction, you can do it at a time where you segregate different pieces of a sample for further particle reduction. These are usually called spiking, recovery standards, or sometimes they call them also standard addition. And you can use them throughout your sample processing in order to track your process, and track how well your process is going to go or how efficient your process is.
Another important aspect for why you will reduce your particle size is your surface area. We talked a little bit about actually getting into the center of that particle. But how about surface area? Surface area is critical for you to extract or digest. So, if you have more surface area there's more area for interaction. So, if you have a five-millimeter particle, let's pretend that you have one particle. And I have numbers here for cubes and for
Spheres, and then the average for the total surface area. So, let's say that you have that one five-millimeter particle, your average surface area, let's say on this sort of amorphous, cubist sphere shape, is a 114 millimeters squared. But if you reduce that down to one millimeter, now suddenly that same weight of material is 125 particles and 571 millimeters squared of total surface area. If you go all the way down to a fine powder of 0.1 or below you've now increased your number of particles to 125,000 and you have almost 6,000 millimeters squared total surface area. So, you have a huge amount of surface area all with the same weight of material.
There are two basic types of standards or ways to use standards. You can use them as external standards, or they can be internal standards. External standards are outside of the sample, where internal standards are actually added to your sample. External standards can be either matrix matched or unmatched, meaning they're in a solvent or in an aqueous base, or you have internal standards by their very nature because they're added to a particular material or a particular sample, are matrix matched. External standards can be used for qualitative, quantitative and semi-quantitative work. And internal standards are often used for statistics, for correction of bias, to eliminate any
Variability, and for system monitoring. So, if you have an auto sampler that has a slight problem with its injections that there there's a lot more variability than you'd like, you can compensate with that with an internal standard. External standards tend to be duplicates, or analogs of your target. So, if you're using as a particular pesticide target,
like aldrin, then you would use aldrin as an external standard as well. But again, these do not compensate for any errors. So, if your instrument has a slight variation in its injections, then you're going to see a difference between your external standard and your sample. For your internal standard, you often use a labeled analog. So, that means that aldrin would be deuterated or you know C13 labeled. So, you would use a deuterated or C13 labeled analog of that aldrin, and then this is put into your sample internally and you could then compensate for if the injector had a lot more variability than you wanted.
There are several things you need to consider when choosing a standard. First, is it fit for purpose. Does it satisfy the requirements you have for the type of standard, the type of target, or the identity. You don't want to be using an aldrin standard for dieldrin analysis. So, are you matching the identity of what you're trying to target to the standards that you are using. Is it in the correct form? Well, here I have a plant material that I'm testing but the standard I am getting is in aqueous format, so is it in the right form. Is it in the right matrix? So, I have now a cannabis extract oil, but I have a standard in acetonitrile, so my matrixes don't match. Am I using the correct method for my standard? If I have a standard of LCMS pesticides, am I trying to use them on GCMS? And are they in the analytical range. I need percent level calculations for my analytes, but I have ppb standards, so am I using the right analytical range?
Then you have the role. What role is the standard going to play? Are you going to use it to produce qualitative data, or identity data, or quantitative data, do you actually need a numeric result from it. Or, are you doing quality data, are you doing methods development, or methods verification with this particular material?
Once your sample is homogenized and you've chosen your standards, then you move on to sample processing. So, your sample is all homogenized you have your particle reduction done and now you can also add standards, those additional spiking or recovery standards. You can do this before or after your extractions or digestions and then before your analysis you, of course, will have your internal standards added to your sample or your external standards that you run alongside your sample, for your like calibration curve.
When you're building your calibration curve, you really need to know some important pieces of information about your system. What's its limit of detection or its limit of quantitation? Where is its limit of linearity, and how do you find out your dynamic range? So, limit of detection is the lowest point at which you can detect that particular target and it's usually three standard deviations from the blank or three over three signal to noise. So, you can't calculate anything at this level, but this is the lowest level that you can see the sample at all. Then you have your limit of quantitation. This is the lowest point which you can actually quantitate or measure that particular target, and it often is 10 standard deviations from the blank, or greater between 10 to 20 signal to noise.
Limit of linearity is where you over saturate the detector. This is the point where your detector stops being linear, your results stop being linear, your detector is saturated, so this is the point where you actually have too much in your system. And then your dynamic range is that area from the limit of quantitation to the limit of linearity
where all of your working targets are best seen. So, this is where you do your
quantitation. And when you have your target, you want it to be in the middle of this range. And if you do your standards you want to make sure that you bracket your targets above, at, and below so that you have your calibration curve covering your range of possible targets.
There are a lot of standards available for cannabis, are needed for cannabis. You need internal external standards, you need blanks, calibration and tuning standards, standards for identification and quantification, also for method validation and verification. And these can be a range of things, your organic compounds like your cannabinoids, your terpenes, your flavonoids, different pesticides. You can have solvent analysis, nutrient analysis, and of course your toxic elements as well.
Now we're going to take a look at some of the analytical targets and how best to approach looking at them. Let's first take a bigger look at cannabis. Of course, we all know about the cannabinoids. But they are joined by other compounds, such as terpenes and flavonoids, in order to work within the plant and to work as a nutraceutical in cannabis products.
Let's start with what are flavonoids. Well, they provide color and defense to the plants. And they're composed of 15 carbons with two phenyl and a pyran ring. They are considered to be secondary metabolites, and therefore non-essential. They are produced by the ppm pathway, or the phenyl-propanoid pathway, and they are detected by the GABA receptors in the human body. Flavonoids can be grouped into many
different subgroups depending on their functionality. Some of them are very famous as antioxidants. We think about all the purples and the blues, and all of those colors. These are all different types of flavonoids. The major flavonoids are flavones or flavonols for cannabis, and they're responsible for a lot of the yellow colors, the oddball colors. And there are some unique flavonoids, especially the cannaflavins, that are particular only for cannabis plants. We said that they are manufactured in the phenol-propanoid pathway, or the ppa pathway. So, it starts with an amino acid and then it's converted into coenzymes until you get the different forms of the flavin. So, you've got the cannaflavin, the cannabisin D. So, you have all these flavonoids being produced in the
cannabis by the phenyl-propanoid pathway.
So, what are terpenes? They've gotten a lot of attention and a lot of press since cannabis became mainstream. But terpenes are the flavor and fragrance compounds for many different botanicals including cannabis. And they're built upon isoprene units, the C5H8, so they repeat over and over again, that's why a lot of terpenes have the same molecular weight. There are two types, there are the essential and the non-essential, so the essentials are required for the life of the plant, and they're formed by two metabolic pathways starting with a common compound, the isopentanol pyrophosphate, the ipp.
And the two pathways are the MVA and the MEP pathways. And they are received by the TRPV receptors. The way terpenes are grouped together are by their number of isoprene units. So, monoterpenes have two isoprene units, a sesquiterpene has three isoprene units, so something like limonene is actually a monoterpene, and humulene a sesquiterpene.
Different products have some terpenes in common. So, if you look at the products of wine, hops, which is a cousin of cannabis, they're in the same botanical grouping, and then you have cannabis, you can see they share some of the same terpenes. So, you have like linalool, that's the lavender kind of smell, they can be found in wine, hops, and a minor component in cannabis. Cannabis you see myrcene, again you can see that in in small parts in wine and hops, and in the majority in the hops and the cannabis. As I said, they are formed by two pathways, the MVA and the MEP pathways. And in this case, this starts with a pyruvate, and then a coenzyme comes in and you form the different terpenes; the monoterpenes in the MEP pathway, the triterpenes the sesquiterpenes and the MVA pathway. So, through a series of different interactions and reactions you get all the different forms of the cannabis terpenes.
And last, but not least, we have our cannabinoids. They interact with g-protein-coupled
Receptors. So, these are the receptors in the body that the cannabinoids interact with. Yhey have a carbon 20 or 20-carbon base, and they are split into about nine major groups. There are things called Phyto cannabinoids, that they are light responsive, and then you can also have cannabinoids that are created within or out of the cell. Or you can have synthetic cannabinoids. And they are formed by the polyketide pathway or the PKP pathway.
Here are the major groupings for all of the different types of cannabinoids. You can see there's a wide range of functionality. So, now you have the polyketide pathway, it's starting with fatty acids, and then using other coenzymes in order to react and create the different forms of the cannabinoids. Now in green you can see that these are some of the same components that were involved in the terpene synthesis. When you take a look at the different pathways, the cannabinoid pathway, the terpene pathway, and the flavonoid pathway, they actually are very interdependent upon one another. If you recognize the geranium phosphate that we found in the terpene pathway, you'll see it appear again in the cannabinoid pathway. The malonyl coenzyme you'll see that that actually becomes part of the synthesis of the fatty acids, the reaction within the cannabinoids, and within reactions of the different flavonoid pathways. So, they all become very interactive and they're all dependent upon one another.
In cannabis, chromatography can be very challenging. You have a lot of interrelated compounds. So, you have many different isomers, you have compounds that interconvert between each other due to heat, or pH, or exposure to light. So, there's a lot of rearrangement that goes on. You have similarities and things like your mass. So, if you're doing mass detection, you could be looking at the same fragments over and over and over again with very subtle differences to them. They can have very similar physical
Properties, like boiling points, or melting points. They can have repeated structures, so
they have many similar structures in common. And there could be a wide range of
functionalities, you can have ketones, and acids, and bases, and alcohols. So, there is a huge range of functionalities. There's also a wide range of solubilities and polarities which make them very difficult then sometimes to sort out or determine one method to do the most amount of compounds with. And then you of course have a very
sticky complex matrix that can interfere with a lot of your analysis making it
very difficult for you to decide on what type of analysis to do.
How you proceed with your analysis is really going to depend on functionality. Some functionality is better suited to GC, things like hydrocarbons and alkanes. If you're doing acids and alcohols, they're a little bit more deserving of LC, they're a little bit more amenable to LC. You can do acids on a GC, but you have to derivatize them. Now if you have very polar compounds, you're probably going to lean to an HPLC or an LCMS system where you have that ionization potential. If you have a good ionization potential, it works really well with LCMS. Then there's molecular weight. If you have less than 500 mass then yes, you can use either LC or GC. If you have more than 500 mass units, then or your molecular weight is more than 500, then you're going to more gravitate towards an LCMS. Then your volatility and your lability. How volatile or labile is your compound?
If it's prone to thermal degradation, you're going to want a cooler technique like an HPLC, where if it's very volatile you're going to want a low temp GC or an HPLC. And solubility, does it like polar solvents, then you're more than likely going to be doing reverse phase chromatography, like a reverse phase LC. Maybe some GC, if you're going to be using nonpolar solvents like methylene chloride, then you're going to be using GC or normal phase LC.
So, as you work with your cannabis samples, you're going to want to dissolve them in some sort of solvent in order to run your analytical processes. The grade of solvent depends on what type of testing you're going to do. Most of the time you want to keep to grades for chromatography like your GC, pesticide residue, your HPLC grade your LCMS grade because they are intended for chromatography use. You also want to match your solvents to your expected targets. If your compounds are polar you're going to want polar solvents, if your compounds are nonpolar then you're going to want no nonpolar solvents. So, polarity is reflected by the polarity index. Your polar solvents are the ones that are amenable to reverse phase HPLC, so they're going to be the ones that have the high polarity index, things like your water your acetonitrile, your methanol. And you want your polarity index to be in the same range of the pKa and the logP of your samples. So, these logP's and pKa’s show that we are in a higher polarity index so more polar, and then you're going to want to use more polar solvents. You also want to understand that as you mix your solvents your polarity index will change. So, if you're using 50:50 acetonitrile water in an HPLC, you're not going to get the polarity of water and you're not going to get the polarity of acetonitrile, you are going to get a blend of it both. So, depending on the percentage you will have to recalculate what your polarity index is for your solvent.
You also want to be aware of the miscibility of solvents, some solvents do not play well together. So, you don't want to be mixing water and an alkane, they're not going to mix well together, they're not going to mix it all together, so you want to be aware of solvent miscibility. You also might be needing some buffers, or modifiers, or additives. This will change the pKa, it will change the ph, mostly we use these buffers in HPLC and LCMS. So, for HPLC we use salts, acetates, phosphates, but things like that sometimes will create residue so for LCMS you use more volatile buffers, buffers that will aid in ionization without any residue. So, the buffers you want to use for LCMS are things like tfa, and formates, and acetates, and ammonia compounds. And you want these so that you can actually help to ionize your samples. For this you're going to really want your pH range to be about two units above or below the pKa of your sample. So, you can
see our pKa’s for our CBD and our THC are in that 10 to 11 range, so at the
minimum we want our buffers to be about two units, so the acetate, the ammonium,
the formats, and things like that, are at least two units above or below the pKa.
You're also going to want to choose columns for the types of compounds that you are trying to separate. So, if you want to do reversed phase you're going to be doing most of the blue, and if you're doing normal phase you'll be doing the red. If you your compounds are ionic then you're going to be running a C2 through a C18 column if you're doing reverse phase and you have neutral compounds that are hydrophobic then a phenyl, a biphenyl, a C1, a C30, or things like that. So, you want to match the column phase to the target of the analytes you'll be trying to separate.
Now let's look at chromatography and what type of information that can give to you.
So, when we're talking about writing with light, we're really talking about visible light in that small band between about 300 and something and 800, so that is the small visible light spectrum and we use that spectrum to determine what we monitor when we do UV-Vis on our HPLC because that gives us the targets. So, if your particular sample has a violet color that means that yellow green is absorbed and you should be looking in the 560 to 580 nanometer range. So, you can actually choose what nanometers or what ranges by the colors that you observe. You can also choose what range of
nanometers that you should be observing by the functionality. Different functionalities have different absorbances, so if it's a ketone it absorbs at 190. If it's a thiol it absorbs at 195 if it's an alkene 171. So, typical wavelengths for cannabis to monitor are 220, 228, 254 which is our benzene ring, 270, and 280.
You also want to be aware of the absorption wavelengths of the different common solvents. You don't want a solvent that is going to absorb at its max in the same range of where you want to be conducting your analysis. So, let's say you are using ethyl acetate, but you are looking for 254, well then that's going to be a problem because your UV cutoff for ethyl acetate is 255, so you're going to have a very high background from your solvent. Typical HPLC conditions for cannabis, well mostly we use a C18 or some form of a C18 column and this is reverse phase chromatography. A very typical start point is our standard column of 150 millimeters, 4.6 and 5 microns. So, we often start out with a gradient method of water and acetonitrile and you again you want to be LC or HPLC grade of acetonitrile and a typical preparation is a formic acid buffer of 0.01 to 0.5 percent and anywhere from 1 to 25 millimolar of ammonium formate. And again the wavelengths you're going to look at 220, 228, 254, 270, and 280. If you look at the most commonly examined cannabinoids, you have generally your early eluters these ones on on a CA column come out fairly early. These are things like your CBDA and your CBG,. Then you have your late eluters, your CBC, your THCA, things like that, and the ones that we often are very concerned with are D8, D9 and cbd, are more towards the center they tend to be a little bit more fully resolved. When you're doing method development for HLPC there is a method to the madness. You have to start thinking from the point of view of where are the majority of your compounds? And then that will give you an idea of where to start, what type of column to start, and then you can actually go through the decision tree and make decisions based on what you're trying to do. If you're trying to increase retention, decrease retention, increase selectivity, so there is a hierarchy to these decisions, and I've tried to outline them here in this chart.
You can also run cannabinoids on GCMS, and you want a mid-polar phase something like they call it like an XX35 so depending on the brand those first two letters change but it's usually a dash 35 and that means 35 percent diphenyl, 65 percent dimethyl polysiloxane and you want to derivatize your samples for your acids. So, you want to stabilize your acids by adding a derivatization agent will then stabilize your acid and they do not break apart in your injection port. And again, you want to use GCMS or pesticide grade solvents to dilute your samples.
Thank you if you have any questions please feel free to contact us on our website www.spex.com.
-final-2.png){kind=logo}