Gas chromatography and liquid chromatography coupled with different detectors including mass spectrometry, UV-Vis, and others are usually applied for the characterization of components in e-cigarette aerosol. Besides chromatography, nuclear magnetic resonance has also been applied to the characterization of e-cigarette thermal degradation products. Jensen et al.74 have published a library of 1H NMR spectra for many thermal degradation products in ecigarette aerosol that are derived from PG and VG. These technologies are also applied for thecharacterization of including volatile organic compounds , carbonyls , nicotine and tobacco specific nitrosamines , polycyclic aromatic hydrocarbons , heavy metals , and flavoring compounds. VOCs are normally detected by GC-FID, GC-MS and related instruments. Lee et al.99 detected VOCs including ethanol, acetonitrile, isopropyl alcohol, benzene and toluene using the GC method from National Institute for Occupational Safety and Health that employ evacuated canisters lined with fused silica for sample collection. Carbonyl compounds are usually derivatized by 2,4 – dinitrophenylhydrazine in solution or on silica gel cartridges impregnated by 2,4 – DNPH, followed by analysis on HPLC-UV or HPLC-MS. Nicotine, TSNAs, and PAHs are usually captured on filter pads and analyzed by GC-MS. Trace metals are usually captured on quartz filters and analyzed by ICP-MS. The e-cigarette aerosol includes a liquid-like particle phase and a gas phase. Most e-cigarette emissions are semivolatile, pots for cannabis plants which can partition to both the gas and particle phases depending on different environmental conditions and chemical properties .
Pankow et al. predicted that the phase distribution of various components in ecigarette aerosol are related to the mass concentration of particles , the composition of particles, the vapor pressure of the chemical, and the ambient temperature. For example, formaldehyde can be found mainly in the gas phase even at the highest level of total particular level of e-cigarette aerosol, while formaldehyde hemiacetals partition into the particle phase, even at the lowest total particulate matter levels.Although simple thermal degradation products such as formaldehyde, acetaldehyde, and acrolein are commonly quantified in the literature, the reported levels of specific components significantly vary in different publications. This might due to the diversity and complexity of ecigarette products, as devices from different brands and design generations vary in power settings, coil type and puff topography . Another potential reason is that there is no standard sampling and analytical methods for the target analytes in e-cigarettes aerosols. For example, Eddingsaas et al. found that the use of limited collection methods will not identify all aerosol components. This suggests a more comprehensive approach, using a combination of techniques, is needed for the collection of e-cigarette aerosols. Moreover, thousands of e-liquid formulations with different PG:VG ratios and different flavoring compounds introduce further challenges to a unified understanding of the e-cigarette chemistry in the literature. This suggests a need to isolate key e-liquid ingredients to study their fundamental chemistry, instead of sampling from commercial e-liquid blends that are proprietary in composition, in order to have a more predictive understanding of thermal degradation in e-cigarette vessels.
Since its introduction to the United States in 2007, the electronic cigarette market has expanded significantly. The prevalence of e-cigarette use was 3.2% for adults and 7.6% for young adults in 2018. The prevalence of e-cigarette use among high school students increased from 1.5% in 2011 to 27.5% in 2019, eclipsing conventional cigarettes among youth. With the growing population of e-cigarette users, the evidence that e-cigarette use is related to higher frequency of cigarette smoking, and the lack of historical governmental regulation, there is a significant need to fill existing data gaps on chemistry, toxicology, and clinical/behavioral patterns to inform on e-cigarette consumer safety and risk. E-cigarettes have been suggested as a reduced harm alternative to traditional tobacco-based products due to the reduced presence of well-studied toxicants formed during tobacco combustion. However, the use of e-cigarette may have its own risk, such as electronic cigarette or vaping-associated lung injury , respiratory function impairment, inhalation of carcinogenic carbonyls, and changes in gene expression. Furthermore, as e-cigarette emissions are not completely inhaled, there is potential for bystander or secondary exposure to non-users from the exhaled aerosol to the environment. Recent works have provided insights into how e-cigarette components and emissions affect indoor air quality and exposure pathways. Yet to date, there remain majorgaps in our knowledge of a complete chemical profile generated from the vaping process, as well as detailed mechanisms producing those chemicals. Moreover, the astonishing variety of ecigarette products and innumerable flavors available on the market, combined with the fast pace of product alterations due to the steady increase in e-cigarette popularity, present significant challenges in e-cigarette research and the estimation of user risk. The thermal degradation of propylene glycol and vegetable glycerin , the primary components of e-liquid, can generate complex chemical products through a series of reactions. Laino et al. showed that the thermal degradation of VG can form formaldehyde, acetaldehyde, acrolein by dehydration via the formation of glycidol, while PG can generate propionaldehyde and acetone via the intermediate formation of propylene oxide.Diaz et al. suggested PG could also participate in a heat-induced radical-mediated degradation pathway, initiated by O2 insertion to C-H bonds to generate the OH radical that further propagate the radical chain, forming at least five degradation products.
The radical-mediated pathway of VG has also been proposed by other researchers, and at least seven thermal degradation products have been observed in the process. Some degradation products can react further to form simple carbonyls, and accretion reactions between carbon-centered radicals or stable products can further complicate the chemistry of e-cigarette aerosols.The fragmentation of aliphatic alcohols tend to produce compounds that have a carbonyl moeity; However, since PG and VG are polyols, their degradation will also result in carbonyls functionalized with hydroxyl groups in addition to the simple types. Organic acid formation may also occur to an extent, possibly as a carbonyl oxidation process. Some thermal degradation products have well-documented toxicity to humans , while others have suspected toxicity . In addition to thermal degradation products, hundreds of flavoring ingredients may be added to e-liquids and vaporized in e-cigarette aerosol, which can potentially lead to adverse health impacts. Jensen et al. identified the largest variety of thermal degradation products to date from aerosolized e-liquid using Nuclear Magnetic Resonance ; however, the data are not quantitative in that work. Since most compounds in e-cigarettes have a carbonyl moiety, quantification is conventionally done by derivatizing with 2,4-dinitrophenylhydrazine to produce hydrazone adducts , followed by analysis with liquid-chromatography or gas chromatography using authentic carbonyl-DNPH standards for calibration of chromatographic peak areas. Even so, authentic carbonyl-DNPH standards are not available for many complex products. Synthesis of carbonyl-DNPH standards may be done; however, the process to synthesize, purify, indoor cannabis grow system and purity-check is laborious, requires specialty equipment, and requires reasonable syntheticchemistry skills. Synthesis of DNPH hydrazones of multi-carbonyls require additional purification steps to isolate the mono- and multi-hydrazones. In addition, some carbonyls are not commercially available as starting material, requiring their own separate synthesis. Thus, an approach to quantify without chemical standards is an attractive alternative. Furthermore, spectroscopic chromatography methods that rely on retention time and UV-visible absorbance spectra may be limited by co-elution or indistinctive spectra, even when utilizing authentic chemical standards. The coupling between chromatography and high-resolution mass spectrometry is a powerful tool for chemical identification, as it removes the co-elution limitation by enabling molecular formula assignments from exact mass. The goals of this work are twofold: use high mass resolving power coupled to chromatography to better identify DNPH hydrazones of functionalized and simple carbonyls and acids, and develop a method to quantify e-cigarette chemical products for which analytical standard are unavailable.First-generation disposable e-cigarettes from blu® , a popular e-cigarette brand,144 with “Classic Tobacco” e-liquid cartridges were used for this study. The blu® e-cigarettes are comprised of a rechargeable battery with a capacity of 140 mAh, an atomizer with coil resistance of 3.5 ohm, and a disposable, non-refillable e-liquid cartridge with proprietary ingredients. Batteries were charged after every 20 min of usage and the e-liquid cartridge was replaced after 400 puffs. A TE-2B smoking machine was used to generate e-cigarette aerosol for the analysis. The apparatus puffed two e-cigarettes, in alternating turns, at a frequency of 8 puffs/min for a 2 second puff duration. The average flow rate was 2.3 L/min and the puff volume was 77 mL, quantified by a primary flow calibrator . Ecigarette aerosol samples were collected through 2,4-dinitrophenylhydrazine cartridges for carbonyls/acids and 47 mm Polytetrafluoroethylene filters for nicotine. Atotal of 200 puffs were collected for each analysis, which is within the linear dynamic range of the analysis . The emission profile was stable, within the uncertainty of the analysis, for the first and second 200-puff collection of each cartridge. After collection, DNPH cartridges were extracted with 2 mL acetonitrile into 1.5 mL auto-sampler vials . Consecutive extractions of DNPH cartridges for 40-, 80-, and 200-puff samples confirmed that >97% of both DNPH and its hydrazones were extracted after the first 2 mL volume.
The samples were diluted using LC-MS acetonitrile to the desired concentrations for directinfusion HRMS and MSn analyses . Extracts were used for HPLC-HRMS analyses without dilution. All samples were promptly analyzed after preparation; sample collection and analyses were performed in triplicate.Diluted carbonyl-DNPH extracts were analyzed for molecular composition using a linear-trapquadrupole Orbitrap mass spectrometer at a mass resolving power of ~ 60,000 m/Δm at m/z 400. The extracts were directly infused into a capillary nano-electrospray ion source and the spectra taken in the negative ion mode. An external mass calibration was performed using the ESI-L tuning mix immediately prior to the MS analysis, such that the mass accuracy was adjusted to be approximately 1 ppm for standard compounds. Insights into molecular structure were obtained using collision induced dissociation multistage tandem mass spectrometry in the LTQ-Orbitrap. CID energy was tuned for each mass, such that the precursor ion has 10 – 20% normalized abundance. Thermo Xcalibur software was used for data processing.DNPH hydrazones were quantified by HPLC coupled to the same LTQ-Orbitrap in 2.2 with an electrospray ionization source, operating in the negative ion mode at a mass range of m/z 150 – 500 to cover the mass range of carbonyl-DNPH and dicarbonyl-2 adducts observed in this work. Separation by HPLC was performed using a C18 column end-capped with dimethyln-octadecyl silane and a mobile phase of LC-MS grade water with 0.1% formic acid and acetonitrile . The analytes were eluted over the course of 37 minutes at 0.27 mL/min with the following gradient program: 40% B , 50% B , 60% B , 100% B , 40% B . After separation by chromatography, single ion chromatography for the accurate m/z of DNPH adducts that were identified by the methods in 2.2.2 was used for quantification. A carbonyl-DNPH standard solution , comprised of 13 carbonyl-DNPH analytes was used to obtain the concentration standard curves for calculating the concentration of formaldehyde, acetaldehyde, acetone, acrolein and propionaldehyde in e-cigarette aerosol . From application of the standard curves and propagating the remaining errors of the analysis , the ±1σ uncertainty for calibrated compounds is 10%-20%. The concentration of the remaining carbonyls and organicacids are calculated by their SIC peak areas and the calculated sensitivities to the ESI negative ion mode in 2.2.4. The concentration of nicotine was also measured by the same method using the positive ion mode.The chemical structures of the DNPH hydrazones affect their deprotonation efficiency in the ESI negative ion mode, and thus, their calibration sensitivity in HPLC-HRMS. The Gibbs free energy change of the deprotonation reaction of carbonyl-DNPH compounds that occurs in the ESI negative ion mode was calculated by Gaussian 09 in both the gas phase and solution phase. The structural geometry optimization and frequency calculation was performed by density functional theory using the M06-2X functional and 6-31g+ basisset, which has been recommended for the study of main-group thermochemistry in recent years. 145-147 First, the ΔGd values for the 13 carbonyl-DNPH compounds in the analytical standard mixture were calculated to obtain a relationship to their measured ESI sensitivities. The relationship between the ΔGd and ESI sensitivities was then extended to calculate the relative theoretical ESI sensitivities for the DNPH hydrazones for which commercial standards are not available. Calculated sensitivities were then used to estimate the concentrations of carbonyl-DNPH hydrazones in e-cigarette aerosol extractions with the method described in 2.2.3.