Polysaccharide Chiral Stationary Phases for the Achiral and Chiral Separation of Cannabinoids

Abstract

Polysaccharide-based chiral stationary phases (CSPs) have been widely utilized in the pharmaceutical, agricultural, and natural product industries since their first-reported use and subsequent commercialization more than 50 years ago. Although they have been traditionally used for the separation of small drug molecules containing one or more chiral centers, their uses have recently grown to include achiral separations in emerging fields like the cannabis industry. The ability to separate and study individual cannabinoids is critical to understanding their impact in both medicinal and recreational applications. Furthermore, it is not difficult to envision a future where cannabinoids, particularly for medicinal use, are treated like pharmaceuticals—that is requiring rigorous purity testing, including the determination of chiral purity. While current methods of analysis are sufficient for the separation of achiral cannabinoid mixtures, some critical chiral pairs like cannabichromene cannot be separated fully. This is where the use of polysaccharide CSPs is and will continue to be important, as a chiral resolution will be needed to satisfy these potential requirements. This chapter will cover an introduction and evolution of polysaccharide CSPs, including a discussion on their unique separations mechanism, and review a number of the applications described in the literature of their uses for the achiral and chiral separation of cannabinoids.

1. Introduction

Polysaccharide-based chiral stationary phases (CSPs) have been reported in the literature for nearly 50 years as of the writing of this chapter. Hesse and Hagel made the first practical reports in 1973 using microcrystalline triacetylcellulose (MCTA) as a chiral separation medium, with a simple chiral model [1]. From this initial report, the applications have grown thanks to the advancements made by Prof. Yoshio Okamoto and many others, to include applications in the production of commercialized pharmaceuticals, polypeptides and biologics, natural products, and more recently, cannabis.

As a natural product, cannabis contains a wide range of compounds including, but not limited to, cannabinoids, terpenes, and other plant-based compounds [2]. These compounds typically exist as a single isomer as a requirement for further downstream processes. That is, many biological processes are enzymatically controlled, and require specific molecule confirmation for proper interaction and recognition. Therefore, biological systems have evolved to produce said single isomer that matches this confirmation. Common achiral phases like octadecylsilyl (ODS or C₁₈) and other non-polar analogs have therefore been successfully used for the separation and analysis of cannabis and cannabis-related products, as they are capable of separating achiral mixtures exclusively ([3, 4, 5, 6, 7] as examples). CSPs have been underutilized as a solution for the separation of such compounds and mixtures, as their cost and specialization have been seen as prohibitive or unnecessary. However, it is well established that polysaccharide CSPs are capable of performing both chiral and achiral separations, so they represent a unique opportunity for investigators to perform two types of separations at the same time. The nature of polysaccharide CSPs is unlike that of typical achiral phases. The polymeric structure of the CSPs, either cellulose or amylose-based, along with their functionalization with small molecule chiral selectors, creates an environment that can recognize the subtle structural differences that exist between enantiomers.

What exactly are enantiomers? The most effective way to envision this is to hold up one’s left and right hand – the hands are mirror images of each other (excluding the minor differences in jewelry, fingernail length, cuts/bruises, etc.), but are not superimposable. When you try to overlap them, there is clearly a difference in the structure, i.e. the geometry, of the hands. Compounds that are enantiomers are the same – they have the same combination of atoms or chemical groups connected (bonded) to a single atomic center (also referred to as a stereogenic center or chiral center – usually it is carbon, but can also be nitrogen, phosphorus, or sulfur). Enantiomers differ from each other in the configuration of said atoms or chemical groups around the chiral center. They can also arise from other elements of symmetry like a plane and/or axis where two distinct confirmations can exist. An example of the latter would be atropisomers. Atropisomers contain a rotatable single bond, but because of steric hindrance (a blockage caused by large/bulky groups), are locked into two distinct confirmations. These geometric differences are not exploitable by achiral SPs, but they are by polysaccharide CSPs.

This chapter will begin with a discussion on the mechanism by which polysaccharide CSPs are capable of separating achiral and chiral analytes. This is important to understand why CSPs are so effective in their function, and why they play an important role moving forward in the separation and analysis of cannabis and cannabinoids. This will be followed by a brief sharing of established and practical examples of CSP applications in a range of mature fields (pharmaceutical and agricultural for example). The chapter will conclude with numerous examples in the literature for the separation of cannabinoids on polysaccharide-based CSPs, under various mobile phase modes including normal phase and reversed phase high performance liquid chromatography (HPLC) and supercritical fluid chromatography (SFC).

2. A brief history of polysaccharide CSPs and their separation mechanism

Traditional achiral separations on widely available phases like ODS or silica are governed primarily by polarity. That is, the difference in polarity between the analytes (compounds) and the polarity of the stationary phase (SP). With a simple enough model, one can easily predict elution order based simply on the chemical structure (or polarity) of the analyte and the polarity of the SP. As a simple example, for the separation of phenol and toluene on a C₁₈ column with a mixture of acetronitrile/water, one would expect that phenol should elute first as it is more polar than toluene, which will be more strongly attracted/retained on the non-polar C₁₈ SP. The same modeling cannot be performed for chiral separations however, as enantiomers are equal in their polarity. As described above in the Introduction, enantiomers differ only by the geometry in which their atoms or functional groups are arranged around the chiral center. This geometric difference can only be exploited by a medium that can create an environment that facilitates chiral recognition, which is why CSPs are a critical tool for enantiomeric separations. A well-established (yet highly unpredictable) series of intermolecular interactions helps CSPs to distinguish these subtle differences to elicit a chiral separation.

2.1 Polysaccharide chiral stationary phase separation mechanism

At the core of polysaccharide CSPs are three components: the silica gel support material, the polysaccharide backbone (either cellulose or amylose), and the chiral selector (see Figures 1 and 2 for examples). The support material does not have too much of a role to play in the separation of enantiomers, but is important to provide CSPs with a rigidity and robustness to be used under high-pressure applications. The chiral selector and polysaccharide backbone are responsible for creating an environment that is able to distinguish the two enantiomers via a series of well-documented intermolecular (between two molecules) interactions that arise from it (see Table 1). When contained within in an enclosed system like a packed, chromatographic column, the potential combination of interactions is capable of producing a separation of the enantiomers. The chiral selector is key to creating these interactions - hydrogen bonding, π-π stacking, dipole forces, inclusion, and repulsion can exploit the subtle differences between the enantiomer geometries [8].

Figure 1.

Examples of structures of chiral selectors and names of coated polysaccharide-based CSPs.

Figure 2.

Examples of structures of chiral selectors and names of immobilized polysaccharide-based CSPs.

Type of interaction	Strength	Direction	Working distance
Hydrogen bonding	Very strong	Attractive	Long range
Steric hindrance	Weak to very strong	Repulsive	Short range
π-π Interaction	Strong	Attractive	Medium range
Dipole–dipole	Intermediate	Attractive	Short range

Table 1.

Several intermolecular forces known to occur between analyte and polysaccharide CSPs. Adapted from ref. [8].

Polysaccharide CSPs are unique in their ability to combine all of the above-mentioned differentiating interactions (Table 1) into a macromolecule that is capable of interaction with the racemic (chiral) mixture. The type and frequency of these interactions is highly dependent on several factors: (1) the type of polysaccharide backbone (e.g., cellulose or amylose), (2) the functionalization of the chiral selectors (e.g., carbamates, benzoates and their respective substituents), and (3) the combined 3D-structure created by supporting on silica. Furthermore, the solvation, swelling, or shrinking of the derivatized polymer backbone in the presence of certain solvents or additives plays an important role. Because of these factors, the interactions that take place on polysaccharide CSPs are much more unpredictable and a systematic screening becomes an essential tool for their effective application.

2.2 Development of polysaccharide CSPs for chiral separations and applications

After Hesse and Hagel published their first work using MCTA [1], the continued development of such phases lagged for more than a decade, because of structural and chromatographic inefficiencies. Professor Yoshio Okamoto in Japan made a breakthrough in 1984 by stabilizing the polysaccharide polymer (cellulose in that case), onto a solid silica gel support [9, 10, 11]. This allowed for HPLC or high pressure applications, and improved chromatographic efficiency. The first chiral selectors utilized were coated cellulose tribenzoate and coated cellulose triacetate, later commercialized by Daicel Corporation as CHIRALCEL OA and CHIRALCEL OB respectively [12, 13, 14].

In these early examples, simple models like trans-stilbene oxide and Troger’s base were used to demonstrate the chiral recognition of the new CSPs. The number of selectors continued growing to include coated cellulose tris(phenylcarbamate) (CHIRALCEL OC), coated cellulose tris(4-chlorophenylcarbamate) (CHIRALCEL OF), coated cellulose tris(4-methylphenylcarbamate) (CHIRALCEL OG), and coated cellulose tris(3,5-dimethylphenylcarbamate) (CHIRALCEL OD) [9, 10, 11, 15, 16, 17, 18, 19]. The exploration and study of amylose as a polysaccharide backbone was also critical, resulting in the development of coated amylose tris(3,5-dimethylphenylcarbamate) (CHIRALPAK AD), coated amylose tris((S)-α-methylbenzylcarbamate) (CHIRALPAK AS), coated amylose tris(5-chloro-2-methylphenylcarbamate) (CHIRALPAK AY), and coated amylose tris(3-chloro-4-methylphenylcarbamate) (CHIRALPAK AZ) [20, 21, 22] (see Figure 1 for full list of coated CSP selectors).

Further advancements in the production of the CSPs added robustness and increased solvent compatibility, via the incorporation of an immobilization step. This immobilization step both cross-links the polysaccharide backbone and bonds it to the silica gel surface, leading to the insolubilization of the polymer [23, 24, 25, 26, 27, 28, 29, 30, 31, 32]. This resulted in a new generation of immobilized CSPs, providing access to selectors that were previously not accessible and an expanded range of compatible solvents for expanded selectivity (see Figure 2 for full list of immobilized CSP selectors as of the time of this publication).

This diversification of selectors allowed for an expansion of selectivity that corresponded with a widening utilization in more application areas. β-blockers [33, 34, 35] and non-steroidal anti-inflammatory drugs (NSAIDs) [36, 37, 38] were two of the first classes of compounds to be screened for chiral recognition. Relevant examples included, but were not limited to, acebutolol and propranolol (β-blockers), ibuprofen and naproxen (NSAIDs). Other classes of compounds included proton-pump inhibitors like omeprazole [39, 40, 41], anti-histamines like cetirizine and meclizine [42, 43, 44], selective serotonin reuptake inhibitors (SSRIs) like sertraline and citalopram [45, 46, 47], and commercialized pharmaceuticals like Modafinil [48], Keppra [49], and Bicalutamide [50].

Agrochemicals have also become an important application area, as many pesticides, herbicides, and insecticides contain a chiral center. This application area historically received minimal attention, as there was no requirement to assess biological activity of these compounds, like there is/was for pharmaceuticals. However government regulations have changed over the last few decades, and polysaccharide CSPs have been critical for these analyses as well. There have been many papers published covering the separation of compounds like malathion, fipronil, metalaxyl, dichlorodiphenyltrichloroethane (DDT), bromuconazole, and etoxazole (as examples) [51, 52, 53]. The analysis of food has been an important application, as composition analysis is important for nutritional integrity and quality assurance. Many examples for the analysis and separation of flavanone, diketopiperizine, and naringenin-based compounds have been reported [54, 55, 56].

3. The separation of cannabinoids on polysaccharide CSPs

As mentioned in the introduction, CSPs have historically been overlooked for the analysis and separation of cannabinoids. This came primarily from the belief that cannabis did not contain any racemic pairs of compounds, or at least not any that were of particular interest. This has of course changed with the identification of cannabichromene (CBC) and cannabicyclol (CBL), as well as the rise of synthetic sources of cannabinoids, which have the potential to produce non-naturally occurring opposite enantiomers. This has also been affected by the understanding that polysaccharide CSPs are just as capable of separating achiral mixtures as they are chiral mixtures. CSPs were initially designed to exploit the subtle geometric differences that exist between enantiomers, but they are also capable of distinguishing between more pronounces achiral differences is structure.

3.1 High performance liquid chromatography (HPLC) separations

One of the earliest reports for the use of polysaccharide-based CSP for the separation of cannabinoids came from Levin et al. in 1993 [57]. This group used normal phase HPLC (defined as a mobile phase which contains a mixture of alkane [hexane or heptane] and alcohol [ethanol or isopropanol]), to achieve baseline resolution of several cannabinoid pairs, using CHIRALPAK AD. In 1994, Levin et al. published a paper using the methods developed in their original work, to explore the role of hydroxyl substitution (that is an oxygen with a hydrogen attached to it) and its effect on the chiral separation [58]. By acetylating (adding an acetyl group – carbon doubled-bonded to an oxygen, with a methyl group also attached to the carbon) the hydroxyl groups in the native cannabinoid structure, the resolution of most enantiomer pairs was decreased or lost entirely. In light of the discussion in Section 1 on the separations mechanism, this is not entirely surprising, although it is not often that a direct link between a structural feature and the chiral resolution can be made. A free hydroxyl group has a high potential for hydrogen bonding; given hydrogen bonding is one of the primary intermolecular interactions that takes place on column, the disruption of this interaction could be significant.

In 1995, Levin et al. continued their exploration of structural features of several cannabinoid pairs and the effects these had on their chiral separation [59]. Using CHIRALPAK AD again with normal phase HPLC, the group found some interesting results in particular with the enantiomeric pair of Δ⁶ THC. The pair was well-resolved using hexane-isopropanol as a mobile phase where the elution order was determined to be (+) Δ⁶ THC first followed by (−) Δ⁶ THC second. The addition of 1% by volume of ethanol was sufficient to reverse the elution order. While a reversal of elution order is not entirely uncommon in chiral separations, the identification of the reversal can be important for method development. For an impurity analysis, it is preferred to have the impurity that needs to be quantified elute first. This ensures the impurity, which is often at a low level, does not elute in the tail of the major peak, thus obscuring the level of detection (LOD) or level of quantification (LOQ). For preparative applications, it is preferred to have the target enantiomer elute first, as a higher purity can be achieved while maximizing the recovery.

Jumping back briefly to 1994, Yan et al. published the synthesis and chiral separation of two hexahydrocannabinol derivatives on CHIRALCEL OD [60]. The cannabinoids were derived from nabilone, which is a synthetic derivative of Δ⁹ THC. Much like previous reports, the separation was achieved with normal phase HPLC. In addition to the analytical method development and evaluation, the separation was also scaled to a preparative separation scale, allowing for the isolation and subsequent study of the effects of the individual cannabinoid isomers.

Thakar et al. published a paper in 2002, using CHIRALPAK AD and normal phase HPLC, for the separation of a pair of novel cannabinoid receptor ligands [61]. Using methods developed by Levin, the group was able to separate the two enantiomers and perform CB₁ and CB₂ receptor studies to demonstrate the effectiveness of one enantiomer over the other as a high-affinity ligand for potential therapeutic use.

Tarbox et al. presented a poster in 2009 at the Eastern Analytical Symposium on the separation of the isomers of Δ⁸ and Δ⁹ THC using again, CHIRALPAK AD [62]. The separation conditions were slightly modified from the previous reports, using instead ~96% by volume n-heptane with a mixture of methanol (~1%) and isopropanol (~3%). The significant decrease in mobile phase elution strength was required to achieve near-baseline resolution of the (+) Δ⁸ and (+) Δ⁹ THC isomers that eluted first and second respectively. Chiral Technologies later improved this same separation in 2018, which included the addition of the opposite (−) Δ⁸ and (−) Δ⁹ THC isomers [63]. In this application note, CHIRALPAK IF was used with normal phase HPLC to achieve baseline resolution of all four compounds. The elution order was determined to be (−) Δ⁸ THC first, followed by (+) Δ⁸ THC second, (+) Δ⁹ THC third, and (−) Δ⁹ THC fourth. The separation of the enantiomers of Δ⁹ THC was shared in the same year (2018) using coated amylose tris(3,5-dimethylphenylcarbamate) with normal phase HPLC [64].

Umstead published a paper in 2021 for the separation of several cannabinoids, including cannabicyclol, cannabichromene, Δ⁶, and Δ¹⁰ THC enantiomers [65]. There were several columns used for this work, including CHIRALPAK IB N-3, CHIRALPAK IG-3 (see Figure 3), CHIRALPAK IA-3, and CHIRALPAK IC-3. Normal phase HPLC was used including hexane-ethanol and hexane-isopropanol mobile phases ranging from 90–10 (v/v) to 98–2 (v/v) (see ref. [65] for full method details).

Figure 3.

Separation of cannabicyclol, Δ⁶, and Δ¹⁰ THC under normal phase conditions of hexane-ethanol = 95–5 (v/v) on CHIRALPAK IG-3 [65].

So far, only normal phase conditions have been reported, however aqueous mobile phases (containing water – also referred to as reversed-phase) have also been used for the separation of numerous cannabinoids. A particular advantage of using a reversed-phase (RP) mobile phase over normal phase is the MS compatibility, which assists in the analysis of complex cannabinoid mixtures. Onishi and Umstead published a paper in 2021 focused on the separation of a 10 cannabinoid mixture (which contained Tetrahydrocannabinolic Acid A (THCA-A), Cannabidiolic Acid (CBDA), delta-8 Tetrahydrocannabinol (Δ8-THC), Cannabidiol (CBD), (±)-Cannabichromene (CBC), Cannabinol (CBN), delta-9 Tetrahydrocannabinol (Δ9-THC), and Cannabigerol (CBG)) [66]. A particularly novel feature of this work was the use of ultra-high performance liquid chromatography (UHPLC) and Daicel Corporation’s sub-2 μm immobilized polysaccharide CSPs for the separation.

Figure 4 shows a comparison of Van Deemter plots for the performance of 5 μm, 3 μm, and sub-2 μm CHIRALPAK IA. A Van Deemter plot is a graphical representation of three competing terms that describe the chromatographic separation of an analyte by a chromatographic column. The A term (Eddy-diffusion), B term (diffusion coefficient), and C term (resistance to mass transfer) play different roles in the overall chromatographic separation efficiency. The A term is a constant, as it is assumes the pathway length through a packed particle is more or less the same (although the actual pathway is random). The B term is also more or less constant at functional chromatographic flow rates (although it sharply decreases at very low flow, significantly less than what you would use for a separation). The C term linearly increases from zero to infinity, with the slope being less shallow for smaller particles compared to larger particles (i.e. the plate height (H) decreases much less for small particles as flow rate increases). When combined, you see curves like in Figure 4.

Figure 4.

Van Deemter plot for different particles sizes of CHIRALPAK IA immobilized CSP showing column efficiency related to linear velocity (flow rate) [adapted from ref. 66].

The y-axis represents the theoretical plate height (H in μm), and the x-axis linear velocity (in mm/s). Intrinsically larger particle sizes like 5 and 3 μm (in green and red respectively) have a higher theoretical plate-height due to decreased packing efficiency when packing into a column i.e. the constant A term is larger for these particle sizes. However when the linear velocity is increased, they also lose efficiency more quickly than a smaller particle (due to the C term). For this reason, faster nominal flow rates can be achieved with the smaller particles, allowing for fast/ultra-fast separations with minimal loss of resolution, or the analysis of complex samples with higher resolution.

Circling back from the short tangent on chromatographic theory, Onishi and Umstead looked at both normal phase and reversed phase HPLC, and found a number of very efficient separations. For normal phase CHIRALPAK IB-U (Figure 5) and CHIRALPAK IH-U were found to be the best CSPs for the separation, which used n-hexane-isopropanol-ethanol-trifluoroacetic acid = 96-3-1-0.1 (v/v) as a mobile phase.

Figure 5.

10 cannabinoid mixture separation under normal phase conditions with CHIRALPAK IB-U [adapted from ref. 66].

For reversed phase, CHIRALPAK IG-U (Figure 6) and CHIRALPAK ID-U were found to be the best CSPs for the separation, which utilized water/acetonitrile/trifluoroacetic acid = 45-55-0.1 (v/v) or 55-45-0.1 (v/v) respectively.

Figure 6.

10 cannabinoid mixture separation under reversed phase conditions with CHIRALPAK IG-U [adapted from ref. 66].

De Luca et al. published a paper in 2022, which covered the screening on all available immobilized-type CSPs available from Daicel Corporation (at the time of publication). They found CHIRALPAK IC and IF to be very effective at the separation of a mixture containing cannabidiolic acid (CDBA), cannabidiol (CBD), tetrahydrocannabidiolic acid (THCA), CBC (racemic), and Δ⁹ THC [67]. Because CBC is a racemic cannabinoid, two peaks were observed (Figure 7), with good baseline resolution for all cannabinoids. The sample used for these separations was a true hemp extract (peak identification made by injection of prepared standards), so there are a number of other unidentified cannabinoids observed. For reference, the separation in blue was achieved with isocratic elution using 60% acetonitrile and the trace in red was achieved with isocratic elution using 70% acetonitrile.

Figure 7.

Reversed phase separation of CBDA, CBD, THC, CBC, and THCA with CHIRALPAK IC (in blue) and IF (in red). Adapted from ref. [67] with author permission.

Umstead published a paper in 2022 covering the separation of CBD enantiomers under both reversed phase and normal phase HPLC [68]. For reversed phase, CHIRALPAK IA and CHIRALPAK IG were found to be the most effective CSPs for separation, using water-acetonitrile = 45–55 (v/v) or 30–70 (v/v) respectively. For normal phase HPLC, IA and IG were again found to be very effective CSPs, with the addition of CHIRALPAK ID and CHIRALPAK IE yielding good baseline resolutions as well. For the normal phase HPLC separation on IG (which used hexane-ethanol = 95–5 (v/v)), the separation was also performed on the sub-2 μm version, CHIRALPAK IG-U. This resulted in a sub-15 second separation (Figure 8). Similarly, the reversed phase HPLC separation on IG was repeated on IG-U, resulting in a sub-20 sec separation.

Figure 8.

Separation of (+) and (−) CBD on CHIRALPAK IG-U with Hex-EtOH = 95–5 (v/v).

On a preparative scale, the separation of (+) and (−) Δ⁹ THC was patented by Gutman et al. in 2016 [69]. The group used CHIRALPAK AD with methods similar to what has already been described, but rather than high-pressure application, they used a flash chromatography or medium to low pressure chromatography setup.

3.2 Super-critical fluid chromatography (SFC) separations

The mechanism for chiral separation on polysaccharide CSPs is the same for SFC as it is for HPLC, i.e., a series of intermolecular interactions between chiral analyte and chiral selector. The main difference is the composition of the mobile phase. Rather than 100% organic solvent as is the case for HPLC, SFC uses super-critical carbon dioxide (CO₂) as its primary mobile phase component. There are numerous advantages to using SFC, including the reduction of waste and associated disposal cost, overall lower viscosity mobile phases, which allows for faster flow rates i.e. faster analyses, and the ability to use methanol as a modifier, which cannot be done under normal phase HPLC (miscibility of methanol and hexane is very poor).

Toyo’oka and Kikura-Hanajiri published a paper in 2015 on the SFC separation of several synthetic cannabinoids [70]. While the work contained mostly achiral separations, there was also a reported separation of enantiomers of cis and trans cannabicyclohexanol (CCH) on coated amylose tris(3,5-dimethylphenylcarbamate). The separation was achieved using methanol as a modifier, and a linear gradient from 10–55% over approximately 4 mins. This was also an MS-compatible method, which assisted in peak identification of other minor cannabinoids.

Runco et al. published an application note on the SFC separation of Δ⁸ and Δ⁹ THC using coated amylose tris(3,5-dimethylphenylcarbamate), coated cellulose tris(3,5-dimethylphenylcarbamate), and coated cellulose tris(3-chloro-4-methylphenylcarbamate) [71]. They used ethanol as a modifier and a gradient from 2 to 20% over 5 minutes to achieve baseline resolution on all three CSPs.

Breitenbach et al. published a paper in 2016 covering the SFC separation of synthetic cannabinoids originating from seized drugs [72]. This group also used the three coated CSPs utilized in ref. 71, but in this instance to separate unique cannabinoid JWH-018 and its nine positional isomers. Although not fully baseline resolved, coated cellulose tris(3-chloro-4-methylphenylcarbamate) with isopropanol as a modifier was able to resolve eight of the 10 cannabinoids baseline.

Denicola and Barendt presented a poster in 2018 that covered the analytical separation of a series of cannabinoid mixtures ranging from 9 to 16 cannabinoids, using CHIRALPAK IB N-5 and a methanol gradient from 11 to 14% [73]. Although some partial co-elution was observed, the use of peak deconvulsion software assisted in the baseline quantification of the more complex mixtures. The method was applied to a real hemp oil sample, demonstrating the effective quantification of THC to ensure compliance with the 2018 Farm Bill requirements of less than 3% THC in CBD containing products.

Later that year, Denicola and Barendt presented a second poster that focused on the preparative separation/removal of THC from the same hemp oil sample [74]. Using the method established in the previous poster, the authors showed the isolation of 1.2 kilograms of CBD/day was possible with this new method, which at the time, was about 1.5× more productive than the achiral C18 flash chromatography method that was being used.

4. Conclusions

Polysaccharide CSPs have a rich and storied history for the separation and analysis of chiral pharmaceuticals and agrochemicals, as well as important applications in the food and cosmetic industries. Although not traditionally used for achiral separations, their unique separations mechanism allows for the exploitation of small differences in energy and molecular geometry, meaning a broader range of applicability when compared to achiral SPs. As this awareness has grown, the applications in the field of cannabis separation and analysis have grown with it, particularly over the last decade. Their ability to separate diastereomers, structural isomers, and other positional isomers present in cannabis make them well suited for these applications.

As demonstrated in the chapter their ability to be used in a wide range of mobile phases makes them suitable for numerous applications, ranging from analytical and preparative scale, and with great flexibility in detection technique (mass-assisted or ultra-violet detection for instance). No doubt as the library of natural and synthetic cannabinoids continues to grow, the need for enantiomeric resolution will grow with it. As all application areas continue to expand, particularly for medicinal use, polysaccharide-based CSPs are and will be well suited to meet the needs for chiral purity testing.

Conflict of interest

The author declare no conflict of interest.