The information given by the total ion chromatogram is limited because of its complexity caused by co-elution and background noise. Instead, single ion chromatograms of different mass to charge ratios were extracted for isolating peaks ofindividual compounds, while avoiding co-elution and background noise. The stacked SIC of different thermal degradation compounds or aerosolized components from vaping aerosols of a) pure VEA, b) the mixture of VEA and extracted THC oil, c) extracted THC oil were shown in Figure 4.2. The black line represents the carbonyl/acids that can be generated from the thermal degradation of both VEA and extracted THC oil, the blue line represents the carbonyls/acids only from the thermal degradation of VEA, while the magenta line represents the thermal degradation carbonyls as well as cannabinoids only from the vaping aerosol of THC oil. Although the SICs were able to separate the co-eluting peaks with different m/z ratios, the isomers having identical m/z with similar structures and polarity are still not able to be separated or clearly identified. For example, multiple peaks are observed for C6H12O in the vaping aerosol of mixture of VEA and THC oil . While hexanal can be formed from terpenes, 97 4-methylpentanal could be formed from the thermal degradation VEA according to the proposed thermal degradation pathway in Scheme 4.1. However, since C6H12O is highly enhanced in the VEA aerosol , we assign the majority of this emission to 4-methylpentanal.Over 30 thermal degradation products from VEA were identified. All of the reported thermal degradation products from VEA are carbonyls/acids in this work, weed drying room consistent with other accounts. Riordan-Short et al. also identified several esters and alkanes with GC-MS .
While around 10 carbonyls and acids are identified both by us and Riordan-Short et al., carbonyls with VEA-specific structures have only been identified in this work. The lack of standard spectra for these VEA-derived compounds in GC-MS libraries may have prevented identification of peaks in the chromatograms of Riordan-Short et al.Moreover, some carbonyls identified by Riordan-Short et al. were not found in this work . The cause for discrepancy is unknown; however, we hypothesize it may be partially due to the difference in vaporization method .Many smaller thermal degradation carbonyls and acids appear to be formed by oxidation and bond cleavage of the aliphatic side-chain of VEA. The bond cleavage pathways for VEA degradation is proposed in Scheme 4.1. A proposed radical reaction mechanism is shown in Scheme 4.2. The thermal degradation reaction is initiated by H-abstraction by radicals such as OH, followed by the rapid reaction with O2 to form peroxy radicals . The peroxy radical can react with other RO2 to form carbonyls or alkoxy radicals. Alkoxy radicals may further react to form carbonyls , alcohols , and possibly alkenes . The primary thermaldegradation products may go through further oxidation steps and form more thermal degradation products . These RO2 –based mechanisms have been well studied and shown to be important in various chemical systems, like the atmosphere, biological redox, or fuel combustion. The relative peak intensity of carbonyls in Figure 4.2a support the proposed radical reaction mechanism in Scheme 4.2, since the most abundant peaks represented the formation of benzylic radical and tertiary radical formed in the first H-abstraction step which can be stabilized by the conjugation effect from benzene ring and positive hyper-conjugation effect from the adjacent C-H bonds. The proposed thermal degradation pathway is also supported by the detection of alkanes, including 2,6-dimethyl-1-heptene and 1-pristene, by Riordan-Short et al. and Mikheev et al., since these alkanes are generated in the proposed mechanism. Thus, our observations suggest that the C-C single bonds on the side-chain of VEA is easily oxidized and cleaved during the vaping process, which will cause the formation of a series of carbonyls that has VEA-specific structure, and also alkenes and alcohols.
These primary products may go through further thermal degradation process to generate secondary thermal degradation products like acids and dicarbonyls. Regarding products like duroquinone, durohydroquinone and ketene that have been identified previously by vaping or heating VEA we could not identified ketene as the it will form the same adduct molecular structure as acetic acid when reacting with 2,4-DNPH. We did not observe duroquinone for unknown reasons, possibly due to the difference in sample collection and methods of detection. Figure 4.2c shows the stacked SIC of vaping aerosol of THC oil. Besides thermal degradation carbonyl compounds, a large variety cannabinoids was also identified by HPLC-HRMS, since the phenolic hydroxyl group in cannabinoid structure is slightly acidic and can also be deprotonated in the negative mode of ESI. The thermal degradations products identified in the vaping aerosol of extracted THC oil may not only generated by THC, but can also from the thermal degradation of other cannabinoids, such as cannabinol , cannabidiol , cannabichromene , cannabigerol and corresponding acid , which have also been identified in the unvaped extracted THC oil. The mechanism of the production of carbonyls identified in the vaping aerosol of extracted THC oil may also involve the oxidation of the aliphatic side-chain followed by bond cleavage, since the main cannabinoids also have the side-chains with 5 carbons.Moreover, CBG may be the source of certain carbonyl products since it has a second side-chain with unsaturated bonds ; the specific mechanism is shown in Scheme 4.4. In contrast to VEA, the oxidation of CBG by OH proceeds through addition to the double bonds in the side chain instead of H-abstraction, consistent with the oxidation of other alkenes. The mechanism for the following steps are similar to the H-abstraction route.
The oxidation may also occur on the six member ring of cannabinoids such as THC can occur through pathways proposed in Scheme 4.3b and Scheme 4.5. For example, OH-initiated H-abstraction on THC can occur at the allylic site and OH-addition can occur at the endocyclic C=C, preferentially forming the tertiary alkyl radical. Then peroxy radical chemistry occurs through similar pathways as VEA, finally generating alcohols and potentially epoxides. Multiple SIC peaks are found at the m/z representing oxidized products of cannabinoids, suggesting a lot of different isomers exist. Our identification results are similar to those of Carbone et al., who utilized NMR for identification. Carbone et al. indicated peroxide products may also be formed during the oxidation process, a mechanism not shown in our schemes but would be consistent with RO2 chemistry. The oxidation products shown in Scheme 4.3b have the same number of carbons as THC; however, some thermal degradation products with different carbon numbers were also identified and are hard to trace back to precursor compounds. It is possible they may already exist in the original unvaped THC oil. Borille et al. found cannabinoid compounds or metabolites and 8 non-cannabinoid constituents in the extracts of cannabis plants by ESI-MS, with carbon number of cannabinoids range from C15 to C55. All molecular formulas of the THC oxidation products shown in Scheme 4.3b were also identified in cannabis extracts, suggesting that these components may already exist in the cannabis plant, and that oxidation from plant metabolism or during extraction couldhave occurred in addition to vaping. Moreover, the C19H28O3 has been identified as Cannabiglendol-C3 ; and there exist many possibilities for C23H34O4 ; C15H16O3/C15H18O3 had been identified as cannabispirenone/ cannabispiran. Some compounds in Table 4.1 still remains unidentified .Besides the oxidation products from vaping THC oil, for which the oxidation mechanism is described in Scheme 4.3, there remains unexplained formation pathway for the generation of some thermal degradation products . Couch et al. found the risk of exposure to VOC including diacetyl and 2,3-pentanedione during the decarboxylation and grinding process of dried cannabis material, but there is no clear mechanism given for their formation. The generation of these compounds may due to the thermal degradation of terpenes and terpenoids. Since there is still over 50% mass in the unvaped THC oil that remains uncharacterized, drying rack for weed it is likely that a portion of that mass are terpenes. Meehan-Atrash et al. identified degradation products from myrcene, limonene and linalool, including methacrolein, hydroxyacetone, methyl vinyl ketone. Tang et al. found 11 thermal degradation products from mixture of terpenoids, 7 of them are carbonyls including formaldehyde, acetaldehyde, acetone, acrolein, methacrolein, valeraldehyde and hexanal. These findings are consistent with the identification results in this work, illustrating that the extracted THC oil is a complex mixture, the complexity of which increases with thermal degradation chemistry. Further research on individual components is still needed for a better understanding on the whole picture of thermal degradation. For the mixture of THC oil and VEA, it is clear from the stacked SIC that the peaks shown in the chromatograph are mainly from aerosolization products of vaped THC oil instead of VEA. It is clear that the total signal from aerosolization products of the mixture is between that of vaping pure VEA and THC oil. Moreover, the oxidation of THC may also be suppressed by adding of VEA. While the signal ratio of cannabinoids in vaping aerosol of the mixture compared to unvaped THC oil is 0.34, the same ratio for oxidated cannabinoids in Scheme 4.3b is 0.22 .
THC was shown tohave a stronger tendency to degrade compared to VEA, since the boiling point for THC is 157 ˚C, while VEA start to decay at 240 ˚C without boiling. Table 4.2 shows the particle mass collected on the glass fiber filter at three temperatures and various e-liquid composition. It is clear that increasing temperature will increase the particle mass on the filter, which is consistent with expectations. However, the particle mass production is non-linearly suppressed with the addition of VEA compared to THC oil at the same temperature. The reason might be the formation of non-ideal solution with significant intermolecular interactions when VEA is added to the THC oil, as Lanzarotta et al. had found that hydrogen bonding exists between the molecules of VEA and THC. Given the fact that THC has a much higher aerosolization rate compared to VEA , the cartridge may be enriched in VEA since vaping continues until it is 100% VEA. In order to figure out the influence of VEA to the formation of carbonyls, it is informative to normalize the mass of carbonyls by the particle mass collected at the same temperature . While e-cigarette users who used nicotine products will self-titrate nicotine intake in daily use, there is also evidence that people who use higher potency cannabis for recreational purpose can titrate their THC dose. Figure 4.4 shows the normalized mass of 9 thermal degradation carbonyl compounds by particle mass produced from vaping VEA, THC oil and their mixture at 455 ± 10 °F . Within the C4 – C6 carbonyls shown in Figure 4.4, butyraldehyde, valeraldehyde, hexanal are thought to be from the thermal degradation of cannabinoids and terpenes , supported by Tang et al., while isobutyraldehyde, isovaleraldehyde and 4-methylpentanal are from the thermal degradation of VEA , supported by RiordanShort et al. Since some isomers can’t be separated in this work, we discuss the pair of isomers together. From the normalized carbonyl concentration, it is clear that certain carbonyls such as formaldehyde, hexanal/4-methylpentanal, glyoxal, diacetyl/3-oxobutanal are produced in much higher abundance from VEA compared to extracted THC oil. Although some products like formaldehyde can be produced from both VEA and THC, the production of formaldehyde from VEA is more favorable since it involves a tertiary radical intermediate in the first step , which is more stable than the secondary radicals formed from the side-chain of THC. The proposed chemistry is, thus, consistent with higher formaldehyde formation by VEA. The same explanation can also apply to the generation of 4-methylpentanal, which only comes from VEA and thus likely dominates the distribution of the isomer pair over hexanal. The formation of glyoxal, diacetyl and 3-oxobutanal from VEA likewise may be enhanced compared to THC due to higher stability of radical intermediates. Diacetyl is thought to be byproducts of cannabis plants,61 and there is no clear indication of formation of diacetyl from VEA . The formation of its isomer 3-oxobutanal can be expected from VEA, however. The corresponding SIC of diacetyl shows that multiple peaks exists in the vaping aerosol of extracted THC oil, but only one peak shown in the vaping aerosol of pure VEA, suggesting that cannabinoids and terpenes may generate multiple isomers which have the same m/z as diacetyl, but VEA propably generates only 3-oxobutanal.