In recent years, heated tobacco products (HTPs) have emerged as a significant innovation in the tobacco and nicotine markets. Unlike conventional cigarettes that burn tobacco at high temperatures, HTPs are designed to heat tobacco to lower temperatures (typically below 350 °C), producing an inhalable aerosol without pyrolysis (1). This process reduces or avoids the formation of many harmful and potentially harmful constituents (HPHCs) associated with tobacco pyrolysis, offering an alternative experience that is closer to traditional smoking than e-cigarettes, but potentially with lower toxicant exposure (2).
The commercial success of products like ‘IQOS’ (Philip Morris International), ‘glo’ (British American Tobacco), and ‘Ploom’ (Japan Tobacco International) has driven rapid global market expansion of HTPs (3). These products appeal particularly to adult smokers seeking alternatives to cigarettes, contributing to a shift in consumer preferences and public health discussions. Regulatory authorities and scientific bodies have responded with increased research into their chemical emissions, health effects, and population-level impacts (4).
A distinctive sensory characteristic of HTPs is their relatively mild flavors, which mainly arise from the gentle thermal release of compounds naturally intrinsic to tobacco leaves. However, pure tobacco without added flavorings or extensive casing produces a flavor profile, often described as woody, slightly nutty, faintly leathery, and mildly sweet (5). Heating tobacco at controlled temperatures preserves these delicate notes, while minimizing the formation of harsh, burnt flavors typically associated with cigarette smoke (5, 6). Additionally, the use of humectants like glycerol can influence the aerosol’s mouthfeel, contributing to a smoother sensory experience (7, 8).
Despite the potential for reduced exposure, debates continue regarding the long-term health risks of HTPs. Some regulatory agencies, such as the U.S. Food and Drug Administration (FDA), have authorized the marketing of certain HTPs as modified risk tobacco products (MRTPs) (2), but cautioning that these products are not safe and that complete cessation remains the best choice for health (4). Overall, heated tobacco products represent an evolving landscape of nicotine delivery technology, reflecting both technological innovation and public health challenges in the 21st century.
The natural flavor of tobacco is derived from a complex group of volatile organic compounds synthesized through the plant’s endogenous biochemical pathways and subsequently transformed during curing and fermentation. Research has demonstrated that terpenoids, alkaloids, and phenolic compounds interact synergistically to establish the distinctive sensory profile of tobacco (9). These intrinsic molecules not only define the organoleptic properties of the product but also function as chemical markers for the evaluation of tobacco quality and authenticity. The elucidation of these interactions underscores the convergence of traditional agricultural practices and modern analytical methodologies in advancing the scientific understanding of tobacco flavor chemistry (9).
The flavor profile of tobacco is influenced by several key factors, including the tobacco variety and the curing process employed after harvest.
Tobacco varieties: Different tobacco varieties contribute distinct natural flavor characteristics due to genetic and biochemical variations: Virginia (flue-cured) tobacco possesses natural sweetness with honey-like and citrus undertones, primarily attributed to high sugar content (10), Burley tobacco exhibits a more earthy and nutty flavor with lower natural sweetness, reflecting its low sugar and high alkaloid content. Oriental tobacco is characterized by spicy, aromatic, and floral notes, owing to its smaller leaf size and higher concentrations of volatile aromatic compounds (9, 10).
The methods of curing tobacco leaves after harvest are different depending on the kind of tobacco, and significantly modify the chemical composition and, consequently, the flavor. Air curing is a process of slow drying in well-ventilated barns and produces mild, earthy, and low-sugar flavor profiles (10). Fire curing (exposure to open wood fires) introduces smoky, robust, and heavily aromatic flavors, enriching the tobacco with smoke-derived phenolic compounds (9).
Sun-curing under direct sunlight exposure yields lighter, aromatic flavors by promoting natural enzymatic reactions at low temperatures (9). Flue-curing process in controlled environments enhances sweetness and develops bright, flavorful notes, primarily by preserving natural sugars while reducing green leaf bitterness (10, 11).
These factors collectively determine the sensory characteristics of tobacco products, influencing their application in cigarettes, cigars, and pipe blends. The organoleptic impact of chemical composition and processing factors is most established in the context of smoked tobacco (e.g., cigarettes, cigars, pipes) (7, 11). The key aroma compounds in tobacco and their sensory properties are given in several publications (12, 14, 15).
Heated tobacco and smokeless products are studied with similar frameworks, but data is more recent and evolving. While many flavor-affecting principles are shared with smoked tobacco, HTPs rely more on: thermal degradation (not combustion), glycerol-mediated transport, and volatile precursor chemistry. This results in smoother, less harsh and more aroma-focused profiles, but also introduces more variability depending on device design and tobacco formulation (6, 7, 12, 13).
The moderate temperatures employed in HTPs help preserve subtle flavor notes, such as woody, nutty, floral, and slightly sweet aromas, which are often diminished by combustion. Although the flavor derived from pure tobacco is relatively restrained compared to flavored or additive-enhanced products, natural volatiles like alcohols, esters, aldehydes, and ketones play a central role in defining the aerosol’s aroma and taste. Moreover, microbial fermentation during curing further diversifies the flavor composition (9, 11).
Heated tobacco products (HTPs) operate at lower-than-burning temperatures (typically below 350 °C), causing tobacco flavor deliverance while avoiding the generation of combustion byproducts. Studies have demonstrated that the heat-not-burn process preserves a greater proportion of the natural volatiles compared to conventional smoking, allowing for a milder, less acrid flavor profile (9, 16). Characterization of aerosols from HTPs has revealed the presence of flavor-active compounds such as β-damascenone, megastigmatrienone, furans, and pyrazines, which contribute to floral, sweet, woody, and nutty notes (8, 16).
Several studies have documented volatile and semi-volatile flavor-related compounds generated under heat-not-burn conditions (6,7,8, 12, 16). These works provide specific compound lists based on the analytical methods employed in each study.
The aim of this study was to identify and characterize the volatile compounds using a thermal separation probe, which enables direct thermal release from the tobacco matrix without preliminary sample derivatization, extraction, or concentration. The composition of volatile and semi-volatile flavor compounds obtained through direct thermal detachment and volatilization from the tobacco matrix has not been previously reported. The study investigated which natural flavor compounds are released from different tobacco types when heated at temperatures typical of HTPs and how these compounds contribute to the flavor and aroma profiles.
Research ground tobaccos and tobacco from reference cigarettes were purchased from the Center for Tobacco Reference Products (17), University of KY, Lexington, USA. Research tobaccos included: RT2 Ground Flue-Cured Tobacco; RT3 Ground Oriental Tobacco; RT4 Ground Burley; RT9 DAC Ground Dark Air-Cured Tobacco. Each of those reference tobaccos represented a blend of different grades with each grade coming from any country in the world where they grow that tobacco (C. Fisher, personal communication).
The tobacco from reference cigarettes has a long history of use in the cigarette industry as quality control monitors, model cigarette systems, and analytical method development tools. The filler of 1R6F reference cigarettes contain the following materials: flue-cured 34%; Burley 24%; Oriental 12%; reconstituted 20% (by products of stemming); expanded flue-cured 7%; expanded Burley 3%; Glycerin ~1.7% dry weight basis, propylene glycol (PG) 1%, and Isosweet ~ 6.3%. Isosweet refers to a humectant (high fructose corn syrup) used in the specific blend of the 1R6F Kentucky reference cigarette. It is not a standard commercial product name on its own, but a specific measurement within a tobacco product formulation (17). The fillers of five cigarettes were grounded to 1 mm-size particles using a Thomas Scientific Wiley Mill (Swedesboro, NJ, USA) before further processing.
Samples were stored at −20 °C. They were conditioned at controlled temperature (22 ± 1 °C) and relative humidity (60 ± 5%) for at least 24 h prior to analysis to ensure moisture equilibrium. There was no extraction or derivatization of samples. Tobacco samples were heated under controlled temperatures simulating typical HTP operating temperatures (200–325 °C).
GC-MS were used to access each of the four research ground tobaccos and the filler blend from the 1R6F reference cigarettes for each of the three defined temperatures. Thermal separation probe (TSP) (Agilent, Santa Clara, CA, USA) is an adapter connected to the inlet of the gas chromatography (GC) column, into which a micro vial containing a sample is inserted. TSP applies precise temperature without combustion. TSP utilizes the GC multimode inlet to introduce the sample into the gas stream. The inlet split mode minimizes column and detector overload as well as detector contamination (18).
The inlet and the GC column have independent temperature programming. Only the compounds vaporized at a specific inlet temperature are carried by the helium gas into the column and the detector. Non-vaporized compounds with high boiling points remain inside the micro vials and can be discarded after each run (18).
No preliminary sample processing, extraction, purification, or derivatization was applied. Cured plant material was directly analyzed. The leaf material (5 mg) was placed into an ultra inert micro vial (Agilent Part No: 5190-3187) which was then inserted into a TSP adapter (Agilent Part No: G4382-6300). The TSP was introduced into the Multimode Inlet (MMI). This setup allows for the release of volatile and semi-volatile compounds from the leaf matrix in the TSP inlet before entering the GC column. This method minimizes cross-contamination between samples (18).
The following instrumentation was used in the experiments: Gas Chromatograph: Agilent 8890 GC System; Mass Spectrometer: Agilent 5977B GC MSD; Software: Agilent Mass Hunter (Agilent Technologies) for GC-MS (19).
The operational conditions were based on the application note for the analysis of alcohols (20). Agilent application notes are technical documents that describe practical use cases, methods, and performance evaluations for Agilent’s instruments. Gas chromatography was performed using an HP-5MS UI non-polar column (30 m × 0.25 mm internal diameter × 0.25 µm film thickness). Helium (99.9% purity) was used as the carrier gas at a constant flow rate of 1.2 mL·min−1, with the initial temperature set at 50 °C. The oven temperature program was as follows: an initial temperature of 50 °C held for 1 min, followed by a ramp of 3.4 °C·min−1 to 230 °C, where it was held for 6 min, resulting in a total run time of 59.94 min. The inlet was operated in split mode with a split ratio of 40:1. Three separate inlet temperatures (200 °C, 275 °C, and 325 °C) were applied for each tobacco sample. Separate experiments were carried out for each tobacco sample, with three replicates at 200 °C, three at 275 °C, and three at 325 °C. Mass spectrometry (MS) parameters were as follows: the ion source temperature was set to 230 °C, and the interface temperature was maintained within a range of 250 °C to 320 °C. Electron ionization (EI) was used as the ionization mode, with an electron energy of 70 eV. The quadrupole temperature was held at 150 °C. Mass spectra were acquired in scan mode over a mass range of 30.00 to 550.00 m/z, with a scan rate of 2.8 scans per second. An electron multiplier was used as the detector. The column HP5 UI is a nonpolar one. It retains compounds mainly based on boiling point. Highly volatile compounds elute within a few minutes. Moderately volatile compounds may elute in 10–30 min (e.g., fatty acid methyl esters, terpenes). Semi-volatile compounds (e.g., steroids, alkaloids) often elute after 30 min. However, the compounds are still volatile enough to be analyzed by GC-MS, meaning they have a boiling point low enough for thermal desorption and vaporization in the injector (21).
In the current study, a non-targeted analysis (NTA) of chemicals in tobacco was applied. NTA aims to identify a wide range of chemicals present, without specifically targeting a limited set. This contrasts with targeted analysis, where a small, predefined set of chemicals is the focus (22). For each temperature at least a hundred compounds were detected per run. Total Ion Chromatograms (TICs) were obtained in triplicate for each temperature (200 °C, 275 °C, or 325 °C). In order to chemically identify hundreds of compounds from each of the chromatographic runs we applied a rapid search of mass spectral library. This search presented Probability Based Matching (PBM) algorithm. The match score is a mathematical calculation, often based on a modified cosine similarity, that compares the presence and relative intensity of the ion fragments on the sample’s mass spectrum to those in the library’s known compound spectrum. The match score is expressed in percentage. An 80% match is considered good with moderate reliability, but it does not guarantee accuracy (23). Only compounds having a quality score of ≥80% were selected. Further identification of the selected compounds was carried out using the National Institute of Standards and Technology (NIST) Mass Spectral Library, version 17.1. The screening of compounds from the three repeats was performed using the R-project for statistical computing (24). The constituents having a quality score ≥80% in at least two out of three replicates were selected and processed further. This process was repeated for each of the three temperatures (200 °C, 275 °C, and 325 °C). The abundance of peaks was measured as signal intensity or abundance of the detected ions. There was a problem using obtainable external standards for confirmation of some compounds when added to the samples before TSP volatilization. The standards’ retention times and abundance were different, when applied separately, and when added to the tobacco powder. Due to the specificity of the method, the matrix effect may play a big role in volatilization. The presence of
The data were organized into tables showing the identified chemicals at each of the three temperatures (Supplemental Tables S1–S6). Additionally, the volatiles from studied samples were compared, and their organoleptic potential was estimated.
At 200 °C, 25 compounds with ≥80% match were observed; at 275 °C, there were 41, and at 325 °C there were 61 compounds, respectively. The screening of those compounds was conducted as described in the section ‘Data processing’. However, many of those compounds although natural products are not thought to possess organoleptic properties so they were excluded. After the screening, a total of 17 compounds were selected. Some were found at all temperatures and some were unique for only one temperature (Supplemental Material, Table S1). The numbers present relative average abundance (peak area × 106) and the standard deviation (× 106). The most abundant compound was neophytadiene, followed by acetic acid. As seen from Table S1, the standard deviation for most of the selected chemicals was below 20%, but some were above 30%, as for acetate. The later one gave a poorly shaped peak with variable retention, reflecting in high standard deviation. The average percentage standard deviations in non-targeted analysis in the range of 20–30% are suggested as potentially expected or acceptable in certain NTA contexts (30). Similar variations in the standard deviation were found in the other varieties of research tobacco and in the tobacco from reference cigarettes.
Compound screening, similar to that done for flue-cured tobacco was used for the Oriental tobacco. Twenty-one compounds were chosen (Supplemental Material, Table S2) according to the Data processing criteria. What is notable is a relatively high abundance of furfural; 2-furanmethanol; 2-furancarboxaldehyde, 5methyl-; and furaneol. These chemicals are products of Maillard reaction, where sugars interact with amino acids to form fragrant compounds. Oriental tobacco is unique in producing sugar esters on the surface of leaves. At higher temperatures sugar esters are degraded to acyl groups and sugar moieties that further are degraded to form reducing sugars (glucose, fructose, lactose, and maltose) involved in Maillard reaction (31, 32). Most of the chemicals were detected only at 275 °C and 325 °C.
After the screening of volatile and semi-volatile compounds, 16 flavor and fragrance compounds were chosen (Supplemental Material, Table S3) according to the ‘Data processing’ criteria. The general organoleptic characteristics of Burley tobacco are earthy and nutty flavors, with less sweetness in the taste. The highest abundances were found for neophytadiene, followed by 2,3′-Dipyridyl and phenol. Most of the compounds were detected at 325 °C. 2,3′-Dipyridyl and phenol together intensify smoky and bitter/harsh undertones, especially under thermal exposure (low-temperature pyrolysis), contributing to the drier, harsher end of the tobacco flavor spectrum (11, 33).
Dark air-cured tobacco has a deep, rich, and fermented organoleptic profile. It is dominated by: phenolics (smoky and leathery tones), nitrogen heterocycles (bitter, dry notes), sweet and spicy volatiles (furans and ketones), fermentation-related acids and esters (tangy, fruity, pungent notes) (11, 34).
In RT9 DAC Ground Dark Air-cured Tobacco leaves 15 organoleptic compounds were selected (Supplemental Material, Table S4) according to the data processing criteria. The most abundant compounds were neophytadiene and farnesol, followed by furfural and trans-geranylgeraniol. Neophytadiene itself is not highly fragrant but contributes to the overall flavor and sensory experience of tobacco and can act as a flavor enhancer (35). Farnesol is a natural compound with a delicate green floral scent and notes of lily of the valley. It is listed as one of the 599 additives in cigarettes in a 1994 report by top cigarette companies (36). In the dark air-cured research tobacco the farnesol is not an additive but naturally found. However, its natural emissions probably are not sufficient to give distinctive flagrance.
1R6F Reference cigarettes are established currently as major reference cigarettes. Their filler is a combination of flue-cured, Burley, and Oriental types of tobacco, as described in the Materials section. A total of 16 organoleptic compounds were found (Supplemental Material, Table S5) according to the data processing criteria. They prevailed at 275 °C and 325 °C. The most abundant compounds were furfural, 2-furanmethanol, and acetic acid, followed by neophytadiene and trans-geranylgeraniol. Trans-geranylgeraniol is described with woody and herbal aromas, with mild floral and citrus notes (37).
For each individual tobacco type, as well as for the filler of 1R6F reference cigarettes, the compounds that met the specified criteria were organized as follows: all volatile and semi-volatile compounds detected at the three temperatures were combined into a single column for each sample. The temperatures at which the compounds were released is shown in the respective supplementary tables. The five resulting columns were then aligned. Identical compounds across all samples were placed in the same row to allow direct comparison. The composition of volatile and semi-volatile compounds from all five reference samples was compared, and the organoleptic properties of each compound are shown in Table 1. In the last column of the table, the references to the flavor and/or fragrance were given. A total of 32 natural compounds with organoleptic properties were found according to the data processing criteria. The most ubiquitous volatiles, found in at least four of the five reference products were acetic acid; furfural; 2-Furancarboxaldehyde, 5-methyl-; megastigmatrienone; 2,3′-dipyridyl; neophytadiene; trans-geranylgeraniol; and thunbergol. Only four volatiles were common for all reference products.
Comparison of volatile and semi-volatile compounds in the five studied samples.
| CAS | Name | Flue-cured | Oriental (Turkish) | Burley | Dark air-cured | 1R6F filler | Organoleptic properties | Ref. numbers |
|---|---|---|---|---|---|---|---|---|
| 000064-19-7 | Acetic acid | + | + | — | + | + | flavor food additive | 26 |
| 000098-01-1 | Furfural | — | + | + | + | + | brown type flavor; aromatic odor reminiscent of almonds | 26 |
| 000098-00-0 | 2-Furanmethanol | — | + | — | — | + | burnt type flavor, bready type odor | 26 |
| 000620-02-0 | 2-Furancarboxaldehyde, 5-methyl- | + | + | + | + | + | fruity flavor, sweet, and caramel-like | 26 |
| 010230-62-3 | Furaneol | — | + | — | — | — | less sweet, caramel-like aroma | 31 |
| 000110-86-1 | Pyridine | — | + | — | — | + | fishy-type odor | 25 |
| 000765-87-7 | 1,2-Cyclohexanedione | — | + | — | — | — | nutty flavor and odor | 40 |
| 005989-27-5 | + | — | + | + | + | citrus flavor | 26 | |
| 000138-86-3 | Limonene | + | — | + | + | + | citrus flavor | 26 |
| 000765-70-8 | Cyclotene | — | + | + | — | — | rich caramel, burnt sugar, or maple syrup-like flavor | 25 |
| 000090-05-1 | Guiacol | — | + | — | — | — | woody flavor, phenolic odor | 40 |
| 010468-36-7 | Spiro[3.4]octan-5-one | — | + | — | — | — | woody, camphoraceous odor | 41 |
| 000118-71-8 | Maltol | + | — | — | — | — | a sweet aroma, of caramel and cotton candy-like | 42 |
| 000536-90-3 | Ethanone, 1-(3-methylphenyl)- | — | + | — | — | — | fruity, sweet, and floral aroma and fragrance | 26 |
| 028564-83-2 | 4H-Pyran-4-one, 2,3-dihydro-3,5-dihydroxy-6-methyl- | + | + | — | — | + | sweet, caramel-like notes | 26 |
| 019895-35-3 | 6-Ethyl-5,6-dihydro-2H-pyran-2-one | + | — | — | — | — | sweet, caramel-like, coconut, slightly creamy | 43 |
| 001073-96-7 | 5-Hydroxymaltol | + | — | — | — | — | sweet, caramel-like, slightly toasty aroma | 26 |
| 000067-47-0 | 5-Hydroxymethylfurfural | + | — | — | — | — | mildly sweet, caramel-like | 26 |
| 007786-61-0 | 2-Methoxy-4-vinylphenol | — | + | — | — | — | smoky, clove-like, spicy, and slightly sweet | 26 |
| 054868-48-3 | Solanone | — | — | + | — | — | warm, woody, and tobacco-like scent | 26 |
| 000108-95-2 | Phenol | + | — | + | + | — | smoky and tar-like aroma | 44 |
| 038818-55-2 | Megastigmatrienone | + | + | + | + | + | tobacco / insence aroma | 25 |
| 000581-50-0 | 2,3'-Dipyridyl | + | + | + | + | + | tobacco taste and flavor improving agent | 25 |
| 000504-96-1 | Neophytadiene | + | + | + | + | + | tobacco flavor enhancer; green, herbal, and mildly sweet aroma | 35 |
| 004674-50-4 | Nootkatone | — | — | — | — | + | scent and flavor of grapefruit | 26 |
| 025491-20-7 | Patchoulane | — | — | + | + | — | mild, earthy, and camphoraceous aroma | 26 |
| 000475-20-7 | Longifolene | — | + | + | + | + | woody, pine-like, resinous scent | 26 |
| 024034-73-9 | trans-Geranylgeraniol | + | + | + | + | + | floral, rose-like scent. | 26 |
| 034318-21-3 | Ionol, 3-oxo- | + | + | — | + | — | warm, woody, and floral aroma and flavor | 26 |
| 000150-86-7 | Phytol | — | + | — | + | — | grassy and green notes | 45 |
| 025269-17-4 | Thunbergol | + | + | + | + | + | characteristic, pleasant aroma | 43 |
| 004602-84-0 | Farnesol | — | — | + | + | + | sweet, floral, with a delicate lily-of-the-valley nuance | 26 |
The presence or absence of a chemical is indicated by “+” and “—”. Numbers in the right most column refer to reference numbers.
In order to compare their organoleptic profiles, the compounds were sorted in chemical classes contributing to the aroma and taste (Table 2). Table 2 shows the classifications: Pyranones/lactones prevailed in flue-cured leaves; furan derivatives and ketone/diketones prevailed in Oriental tobacco; diterpenes/diterpenoids prevailed in dark air-cured tobacco.
Classification of volatile compounds identified in research tobaccos and tobacco from 1R6F reference cigarettes.
| Chemical Class | Flue-cured | Oriental (Turkish) | Burley | Dark air-cured | 1R6F filler |
|---|---|---|---|---|---|
| Carboxylic acids | 1 | 1 | — | 1 | 1 |
| Furan derivatives | 2 | 4 | 2 | 2 | 3 |
| Pyranones / lactones | 4 | 1 | — | — | 1 |
| Hydroxy ketones | — | — | — | — | — |
| Ketones / diketones | — | 4 | 2 | — | 2 |
| Phenolics | 1 | 2 | 1 | 1 | 1 |
| Monoterpenes | 2 | — | 2 | 2 | 1 |
| Norisoprenoids | 2 | 2 | 1 | 2 | 1 |
| Sesquiterpenes | — | 1 | 2 | 1 | 1 |
| Sesquiterpenoid alcohols | 1 | 1 | 1 | 1 | 1 |
| Sesquiterpenoid diols | — | — | 1 | 1 | — |
| Diterpenes / diterpenoids | 2 | 3 | 2 | 3 | 2 |
| Bipyridine derivatives | 1 | 1 | 1 | 1 | 1 |
| Nitrogen heterocycles | — | 1 | — | — | 1 |
Numbers indicate the number of compounds identified in each chemical class. The absence of a compound is indicated by “—”.
The five tobacco samples exhibited distinct organoleptic profiles determined by the composition of compound classes (Table 3). The estimation of the natural aroma and the taste of the research tobaccos and tobacco from 1R6F reference cigarettes was made on the basis of the chemical’s composition and chemical class profiles, and with the help of Chat GPT4o. It is important to notice that the organoleptic properties were estimated solely on the basis of selected flavor compounds, not on the basis of all compounds found in the chromatograms.
Estimated natural flavor of research tobaccos and tobacco from 1R6F reference cigarettes.
| Sample | Aroma | Taste |
|---|---|---|
| Flue-cured | Rich, sweet, and fruity; caramel-like from high pyranones; citrus/pine from monoterpenes and norisoprenoids | Mild acidity; light phenolic character; slightly woody or nutty undertones |
| Oriental (Turkish) | Complex and layered; smoky-caramel and roasted from furans and ketones; spicy-dry from sesquiterpenes and diterpenes | Astringent and slightly bitter; phenolic and dry aftertaste; lingering spice |
| Burley | Subtle aroma; low in sweet lactones; earthy and woody from sesquiterpenes; minor smoky hints from furans | Noticeably bitter and dry; less acidic; mild nutty/burnt note from diketones |
| Dark air-cured | Bold and heavy; roasted and resinous from furans and diterpenoids; earthy-spicy from terpenes | Full-bodied and bitter; phenolic and leathery taste with slight smoky-charred notes |
| 1R6F Filler | Mixed aroma; moderately sweet and woody; pyranones add caramel note; dry herbal from diterpenes | Balanced but slightly bitter; dry and faintly astringent, likely due to phenolics and furans |
Flue-cured sample was characterized by a high number of pyranones/lactones (four) and balanced terpenoid profile, suggesting a sweet, creamy aroma and mild acidic taste. Oriental tobacco stood out with the highest number of ketones/diketones (four) and furans (four), indicating strong roasted, caramelized, and smoky aroma and taste characteristics. Burley tobacco sample was chemically simpler regarding aroma-related classes (no pyranones or hydroxy ketones), with only one sesquiterpenoid-diol and two ketones, indicating a light and less intense profile. Dark air-cured tobacco contained no pyranones or hydroxy ketones but all major terpene types, implying a more herbal, less sweet profile. Tobacco extracted from1R6F reference cigarettes showed a balanced profile across all compound classes; but had contributions from all aroma/taste classes, suggesting moderate aroma and smooth taste. Previous studies have shown that oxidative aging and storage conditions significantly impact the volatile profile of tobacco products, especially through the breakdown of carotenoids and phenolic precursors (7, 11).
The natural flavor estimation was based on the organoleptic properties of individual compounds, which represents a limitation of this approach. There is a lack of sensory panel evaluation to validate that the estimated aromas and tastes in Table 3 are actually perceived sensorially.
A total of 32 compounds were detected across all five samples within the temperature range of 200–325°C using the TSP coupled with GC-MS. Of these, fifteen compounds – Acetic acid; Furfural; 2-Furanmethanol (Furfuryl alcohol); 5-Methyl-2-furancarboxaldehyde (5-Methylfurfural); Pyridine; Guaiacol (2-Methoxyphenol); Maltol; 5-Hydroxymethylfurfural (HMF); Phenol; Nicotyrine; Megastigmatrienone; Neophytadiene; Nootkatone; Longifolene; and Farnesol – have been previously identified in the aerosol emissions of HTPs, in tobacco leaf, and in cigarette smoke. (38).
Three additional compounds – 4H-Pyran-4-one, 2,3-dihydro-3,5-dihydroxy-6-methyl-, solanone, and 2,3'-dipyridyl – have already been reported in low-temperature headspace studies of reference tobaccos and classified as natural flavors (25).
The remaining compounds identified in this study are known constituents of tobacco or products of thermal degradation under combustion or pyrrolic conditions, but their presence in aerosol from HTP has not been specifically confirmed in the literature (38).
In the current study, the authenticity of furaneol; 1,2-cyclohexanedione;
Differences in the compounds released from the five samples showed the effects of tobacco type and how those tobacco types have been processed. These differences may also arise from batch-specific variations in tobacco composition, aging effects, or oxidative degradation that selectively enhance the formation or thermal release of phenolics and ketones. Using odor descriptions from the literature, the tobacco from 1R6F cigarettes released compounds related to a more balanced woody-floral aroma with a comparatively milder sensory profile. The other tobaccos were richer in certain groups such as terpenoids, phenols, or carbonyl compounds.
These aroma descriptions came from published sources, not from direct sensory testing in this study.
Overall, this study shows that TSP–GC–MS is a useful method for studying natural flavor compounds released from tobacco at temperatures below burning. Using this approach, several additional flavor compounds were identified that may be released in heated tobacco product aerosols. These findings improve understanding of how natural tobacco chemicals influence aroma and flavor under low-temperature (200–325 °C) and provide a basis for future studies linking chemical profiles to aerosol formation and sensory perception.