The sensory characteristics of cigarettes are shaped by numerous factors, with smell and taste playing pivotal roles. To mitigate the irritation, bitterness, spiciness, and offensive odors in cigarettes and enhance consumer acceptance, researchers regulate cigarette aroma by adding substances like sugar, moisturizers, flavors, and fragrances. Taste, a crucial sensory quality that garners significant attention from consumers, has always been a focal area of research.
The aroma of cigarettes is predominantly determined by analyzing the volatile compounds and their contents in the smoke and screening out those that contribute to human sensory perception [1]. Despite extensive research on identifying the chemical and aroma components of smoke, the task remains challenging due to the vast variety of these components, their trace amounts and complex interaction [2]. Researchers worldwide have employed a plethora of experimental methods to study cigarette aroma components, including chemical analysis, sensory evaluation, and biological detection. In terms of extraction methods to study cigarette aroma components, including chemical analysis, sensory evaluation, and biological detection. In terms of extraction methods, techniques such as steam distillation, thermal desorption, organic solvent extraction, supercritical extraction, gas–solid phase extraction, and headspace co-distillation, combined with various pretreatment techniques and gas chromatography–mass spectrometry (GC–MS), have been widely utilized to study cigarette fragrance components. These methods allow researchers to collect the flue gas centrally and analyze the chemical composition [3,4,5]. Among them, distillation extraction is a traditional approach. Although it is straightforward to operate, highly reproducible, and efficient, it has several drawbacks. The glassware required is costly and fragile, the extraction time is long, and heating can easily affect the essence of the extract. This method is often used to analyze the neutral aroma components in tobacco [6,7,8,9].
In recent years, significant progress has been made in tobacco aroma analysis through the application of diverse extraction and analytical techniques. Wu and Wicks identified 65 volatile compounds in enzymatically hydrolyzed tobacco buds via Maillard reaction coupled with simultaneous distillation-extraction (SDE) and GC–MS [10]. Comparative studies have emphasized the importance of methodological optimization: Li et al. demonstrated that purge-and-trap is effective for rapid screening, ultrasonic-assisted extraction is suitable for precise quantification, and liquid-liquid extraction is useful for routine flavor monitoring [11]. Deng et al. employed solvent-free microwave extraction to enhance the aroma profile of safflower tobacco [12]. Xing et al. optimized steam distillation methods and found that water distillation was superior in terms of yield and aroma richness [13]. Molecular distillation by Lian et al. classified Yunnan tobacco extracts into five volatility-based fractions [14]. Chemometric approaches by Deng et al. identified sensory markers in commercial cigarette smoke linked to cultivation practices [15], complemented by Cao et al.’s models on volatile distribution dynamics [16]. Analytical method development by Zhang et al. enabled sensitive detection of 16 polycyclic aromatic hydrocarbons in cigarettes [17], underscoring evolving capabilities in tobacco component characterization.
Solvent-assisted flavor evaporation (SAFE) is a gentle and efficient method for extracting volatile compounds. It allows for the high-fidelity separation of aroma components in complex matrices under low-temperature vacuum conditions, preserving thermolabile substances and yielding extracts that closely reflect the natural profile of the sample [18,19,24,25]. Although widely applied in food and beverage analyses – such as identifying 75 volatile compounds in seedless prickly pears via (LLE)–SAFE–GC–MS [20], differentiating green/red Zanthoxylum species using GC–MS/O [21], and tracking volatile dynamics in Gujing Gongjiu fermentation grains [22] – its use in tobacco aroma research remains underexplored. Notably, SAFE–GC–MS studies on Pu’er tea quantified trimethoxybenzene isomers, demonstrating its precision in capturing subtle compositional differences [23]. Despite its operational complexity, SAFE’s ability to retain authentic flavor profiles positions it as a promising tool for advancing cigarette aroma characterization, particularly in bridging gaps between extraction fidelity and sensory evaluation.
Previous studies on cigarette aroma compounds and sensory evaluation have often relied on traditional extraction methods like steam distillation or simple solvent extraction. These methods, however, may not fully capture the complex aroma profile due to their limited extraction efficiency or potential degradation of thermosensitive components. In contrast, this study utilizes the SAFE method. As mentioned earlier, SAFE is a gentle and comprehensive method that can effectively separate volatiles from complex matrices and extract flavors closer to the original aroma of the sample, minimizing the loss of thermosensitive volatile components.
Building on the foundation of previous research, recent studies have further advanced our understanding of cigarette aroma and sensory evaluation. Luo et al. optimized the SDE process of volatile aromatic components in flue-cured tobacco leaves using single-factor experiments and response surface methodology. By GC–MS, they qualitatively and quantitatively analyzed the aroma components. Their findings, such as the optimal extraction conditions (1:12 material to solvent ratio, 3.20 h distillation time, etc.), provided a more refined approach for extracting tobacco aroma components, enhancing the efficiency and precision of component isolation [26]. Kochhar and Warburton conducted puff-by-puff assessments of various sensory and subjective attributes for cigarettes with different tar and nicotine yields. Through principal component (PC) analysis, they identified components related to intensity-related characteristics and flavor-related aspects. This research offered insights into how different cigarette characteristics impact the smoking experience at a puff-by-puff level, deepening our understanding of the relationship between chemical composition and sensory perception [27]. The study by DiscountCiggs explored the chemistry behind cigarette scents. It detailed how compounds like terpenes and pyrazines contribute to the distinct aroma of cigarettes. Terpenes, known for their presence in natural scents such as pine and citrus, impart earthy, woody, and fruit-like notes to tobacco. Pyrazines, which are associated with the roasted and nutty fragrance in coffee, also play a role in the cigarette’s aroma profile. This work emphasized the complexity of the chemical compounds in cigarettes and their combined effect on the olfactory experience [28]. These additional studies, similar to the previously mentioned ones, have strived to either improve the extraction and analysis methods of cigarette aroma components or enhance our understanding of the connection between chemical substances and the sensory experience.
The aim of this study was to explore the aroma components in mainstream cigarette smoke. The types and contents of aroma components in different brands of cigarettes were determined by SAFE pretreatment and GC–MS analysis, and the sensory attributes of cigarettes were evaluated using sensory evaluation methods. The correlation between aroma compounds and sensory evaluation was then analyzed to reveal differences in cigarette product characteristics. The study also provides a theoretical basis for improving the comfort and quality of smoking and helps to assess the potential health risks of smoking.
In this study, the samples were from Shanghai Tobacco Group Tianjin Cigarette Factory, including Evergrande (Fully Open Type Cigarettes), Evergrande (Cigarettes 1949 Medium), Evergrande (Memory 1949 Medium), Evergrande (Hard Medium), and Evergrande (Blue Gold Medium). Normal alkanes (with 7–30 carbon atoms) and 1,2-o-dichlorobenzene (used as internal standards) were purchased from Sigma Aldrich (Shanghai, China). Dichloromethane, sodium chloride, and anhydrous sodium sulfate were purchased from Shanghai Guoyao Chemical Reagent Co., Ltd.
Before the experiment, the cigarette and Cambridge filter were placed in an environment with a temperature of 26°C and a relative humidity of 60% for 2 days to ensure that the experimental conditions were consistent.
The sample was aspirated using a PUFFMAN X200AF smoking machine (Shanghai Tobacco Group) with a smoke capture system located at the center of the machine. In this experiment, Cambridge filters were used to lock the smoke onto the filter. Before the experiment began, cigarettes and Cambridge filters were placed at a fixed temperature of 26°C and a fixed humidity of 60% for 2 days. The smoking standard of the smoking machine adopts ISO 3308, with a smoking volume of 35 mL and a smoking duration of 2 s per session. It is used to collect smoke particles and calculate the total particulate matter (TPM).
Crush a Cambridge filter with a TPM mass of approximately 0.2 g and place it in a 500 mL conical flask. Add 150 mL of dichloromethane to the conical flask. Then, set the magnetic stirrer to a rotation speed of 500 rpm and extract for 1 h. After extraction and filtration, place the upper liquid in a 500 mL round-bottom flask, retain the lower liquid, and transfer it to a separating funnel. Repeat this process for the second time, add 150 mL of dichloromethane to the separating funnel, set the magnetic stirrer to the same rotation speed of 500 rpm, extract it for 1 h, filter again, and transfer the upper liquid to a round-bottom flask while retaining the lower liquid. Transfer the lower liquid to a conical flask for the third time and add 100 mL of dichloromethane. Combine the upper liquid obtained from the three extractions into a round-bottom flask. Subsequently, extraction was performed using the SAFE device. The extract was dried over anhydrous sodium sulfate and concentrated to 5 mL on a rotary evaporator, followed by further concentration to 1 mL under mild nitrogen flow. Before conducting GC–MS, store the concentrated fraction at −20°C.
One microliter of each concentrated extract was analyzed in duplicate on an HP gas chromatograph (Hewlett Packard, Palo Alto, CA, USA) model 6890 equipped with a split 50:1. An HP 5975C mass spectrometer connected to was used with an HP-INNOWAX fused silica capillary column (60 m 0.25 mm, film thickness 0.25 mm; J & W Scientific Inc, Folsom, CA, USA). The oven temperature was held at 60°C for 2 min and then raised to 180°C at 2°C/min. and then raised to 230°C at 5 °C/min maintained 40 min. The injection temperature was 200°C, and He at a flow rate of 1.0 mL/min was used as carrier gas. GC/MS was carried out at an ionization voltage of 70 eV and an anion source temperature of 150°C. All samples were analyzed once. Peak components were identified by matching their mass spectra with those in the Wiley and NIST mass spectral library (Hewlett-Packard). The ratio of each peak area to the IS peak area was calculated, and these ratios were used in the subsequent statistical analyses.
In the sensory evaluation of cigarette smoke, the comprehensive consideration of taste and smell is essential to accurately and fairly evaluate the overall quality of cigarettes. For cigarettes, the correct evaluation of their overall quality has important reference value for new product development and old product style monitoring. Sensory testing was conducted by a panel of 12 professionals with more than 2 years of experience in sensory evaluation of smoking, selected based on performance reviews of consistency and reliability. To reduce subjectivity in the evaluation process, prior to quantitative descriptive analysis, the panel used a basic cigarette as a reference, and all members agreed on its score for each indicator and evaluated other test cigarettes on this basis.
This sensory evaluation primarily evaluated nine specific sensory attributes:
-
Basic taste: including sour (similar to citric acid), sweet (similar to sweet substances), bitter (similar to quinine), and salty (similar to sodium chloride).
-
Sensory factors: astringency (dryness or tightness in the mouth), throat roughness (discomfort on the surface of the throat), tingling (tingling or tingling), body fluid production (similar to salivation when eating sour food), and heat (similar to heat when cold).
Sensory tests are conducted in a specialized sensory evaluation laboratory. The laboratory environment is controlled at a temperature of 22 ± 2°C, a relative humidity of 60 ± 5%, and it is kept quiet and odor free. The evaluators are required to undergo an adaptation period of 30 min before the test to reduce the influence of external factors on the senses. During the test, each evaluator inhales a mouthful of cigarette smoke at a time, stays in the mouth for 3–5 s, and then exhales it. Then, according to the specified evaluation criteria, each sensory attribute is scored on a scale of 0–9 points. A score of 0 indicates that the attribute is not perceived, and a score of 9 indicates that the attribute is very strong. To avoid fatigue and sensory adaptation, each reviewer evaluates up to three cigarette samples per test, with a 10-min interval between adjacent tests, during which time the mouth is rinsed with water and tasteless biscuits are eaten to clean the mouth. Each cigarette sample is independently evaluated by 12 reviewers, and the final result is the average of all reviewers’ ratings.
Among them, PC1 and PC2 account for 25.9 and 17.1% of the total data, respectively. The aroma activity values of the five Evergrande cigarettes using the SAFE pretreatment method are mostly concentrated in the first and third quadrants, and the data distance is relatively close, and their aroma is relatively similar on the surface. Among them, Evergrande (blue gold middle branch) and Evergrande (Yankui 1949 middle branch) are the closest and concentrated in the third quadrant, which can prove that their aroma is more similar, while Evergrande (hard middle branch) is slightly farther away from the first three cigarettes, indicating that their aroma is slightly different from the first four cigarettes.
The aroma of a cigarette is determined by a variety of chemical components, which can be determined by the ratio of the relative concentration of the aroma substance to its olfactory threshold, known as the odor activity value (OAV). The higher the OAV value of the aroma substance, the greater the contribution of the aroma substance in the whole aroma system, where OAV >1, indicating a key contribution to the overall aroma, and OAV <1, indicating a modifying effect on the overall aroma. Generally speaking, the aroma contribution value of the aroma compound is positively correlated with the OAV value. After the five cigarette samples were smoked and enriched by the smoking machine, the Cambridge filter was pre-treated with SAFE and then analyzed qualitatively and quantitatively by GC–MS. The SAFE method detected 93 different aroma compounds. Each cigarette contains unique aroma compounds, the number and content of which vary. A total of 19 common aroma compounds were detected in five cigarettes under the SAFE method, including
Using threshold data, we calculated the average OAV of the aroma compounds in five cigarettes. As shown in Table S1, a total of 74 aroma compounds in these five cigarette varieties have OAV values greater than or equal to 1. Compounds such as 3-ethylphenol, p-cresol, β-methylnaphthalene, m-cresol, 4-vinylphenol, 2-methoxy-4-vinylphenol, naphthalene, and indole all have higher OAV values of 10,000 or more, which can be considered important aroma contributing compounds. Nonylaldehyde, 1-methylnaphthalene, ethyl isovalerate, 3-methylindole, 4-ethylphenol, and other compounds have high OAV values, but they are only present in specific cigarettes and can be considered the unique aroma contribution compounds of a certain cigarette, which can bring a special aroma to a specific cigarette. 3-Ethylphenol is an aroma compound contained in these seven kinds of cigarettes. Due to the low threshold, the OAV value of this compound is relatively large, which is the highest among all cigarette aroma compounds. This high OAV value of 3-ethylphenol might be attributed to its relatively high content in these cigarettes and its low olfactory threshold. The low threshold means that even a small amount of 3-ethylphenol can be easily detected by the human olfactory system, thus having a significant impact on the overall aroma. Moreover, during the smoking process, the chemical reactions and volatilization characteristics of 3-ethylphenol may also contribute to its strong influence on the aroma. It will bring an obvious muggy taste to smokers, and it has the highest OAV value among Evergrande (Yankui 1949 Middle Branch) and Evergrande (hard middle branch). It is 419823.87 and 375468.43, respectively, and it is in the remaining Evergrande (fully open tobacco), Evergrande (memory 1949 branch), and Evergrande (blue gold branch). The OAV value of the three kinds of cigarettes is between 237,820.27 and 287,811.56, which is relatively average (Tables 1 and 2).
Types of cigarettes investigated
| No. | Brand | Types of cigarettes | Abbreviations |
|---|---|---|---|
| 1 | Evergrande | Full Open Smoke Kui | FOSK |
| 2 | Evergrande | Smoke Kui 1949 Medium | SK1949M |
| 3 | Evergrande | Memory 1949 Medium | M1949M |
| 4 | Evergrande | Hard Medium | HM |
| 5 | Evergrande | Blue and Gold Medium | BGM |
Sensory attributes selected for quantitative descriptive analysis as a result of preliminary sessions
| Attributes | Characteristic | |
|---|---|---|
| Basic tastes | Sourness | The taste of citric acid alike |
| Sweetness | The taste of sweet alike | |
| Bitterness | The taste of quinine alike | |
| Saltiness | The taste of NaCl alike | |
| Feeling factors | Astringent | A feeling of dryness or tightness in the mouth |
| Throat roughing | A sensation of general discomfort on the throat surface | |
| Prickle | A prickling or tingling sensation | |
| Watering | A sensation similar to salivating at the taste of acid. | |
| Hotness | A chill-like heat sensation | |
A comprehensive quantitative descriptive sensory analysis was performed on five types of cigarettes to assess nine sensory attributes: sourness, sweetness, bitterness, saltiness, astringent, throat roughing, prickle, watering, and hotness. One-way analysis of variance tests showed significant differences between some of the sensory attributes (p < 0.05). The outcomes of these sensory evaluations are represented in the sensory radar chart depicted in Figure 1. Notably, the distinctions among the five cigarette samples were quite pronounced. Specifically, the “sourness” scores of SK1949M and FOSK were higher than those of the other samples. M1949M exhibited the highest scores in “sweetness,” whereas FOSK and SK1949M had lower scores than the other samples. In addition, SK1949M scored significantly higher in “saltiness” and “bitterness” compared to the other samples. Furthermore, the feeling factors’ (astringent, throat roughing, prickle, hotness, watering) scores of SK1949M were higher than other samples. Similarly, BGM and FOSK also scored higher than the other samples in “throat roughing,” “prickle,” and “hotness” (Figure 2).
PC analysis of cigarette aroma compounds by the SAFE pre-treatment method.
Comparison of flavor profiles composed of average scores of nine attributes identified in cigarette smoke.
General impression is a comprehensive evaluation of cigarette smoke, which includes oral comfort, harmony, softness, physiological satisfaction, and the smoker’s preference for cigarettes. PCA was performed to explore the relationship between cigarette samples and sensory attributes, including general impression.
Figure 3 shows a scatter plot of scores and factor loading from PCA. Two PCs explained 95.7% (PC1, 93.0%; PC2, 2.7%) of the total variance. This result showed that “sweetness” located closest to the general impression, which indicated that “sweetness” was associated with the general impression. Additionally, this plot suggested that the samples may be classified into two or three groups on the basis of taste characteristics. FOSK and SK1949M were grouped around “watering,” “bitterness,” “sourness,” and “throat roughing.”
Factor loadings and PC scores extracted from descriptive sensory data.
The sensory evaluation data and overall impression of five cigarette samples were processed and analyzed using SPSS. During the analysis, a variety of statistical methods were used to comprehensively explore the relationship between basic taste attributes, sensory factors, and overall impression scores. Among them, the Pearson correlation coefficient is used to measure the strength and direction of the linear relationship between two variables. The calculation principle is based on the ratio of the covariance of the two variables to their respective standard deviations, and the formula is
In this study, basic taste attributes (X1 [sour], X2 [sweet], X3 [bitter], X4 [salty]), sensory factors (X5 [astringency], X6 [laryngeal roughness], X7 [tingling], X8 [body fluid], X9 [heat]), and overall impression (X10) were included in the analysis. In addition to the Pearson correlation coefficient analysis, a significance test was used to judge the reliability of the correlation coefficient.
The correlation coefficient between X10 (general impression) and X2 (sweetness) was 0.967, showing a significant positive correlation (p < 0.01). X10 (general impression) was also highly correlated with X3 (bitterness), X4 (saltiness), X5 (astringent), X6 (throat roughing), X7 (prickle), X8 (watering), and X9 (hotness) respectively, R = −0.980, −0.968, −0.984, −0.893, −0.923, −0.953, and −0.996, which were significantly negative correlated (p < 0.05).
The correlation coefficients between X2 (sweetness) and the feeling factors (X5, X6, X7, X8, X9) were negatively correlated, while X1 (sourness), X3 (bitterness), and X4 (saltiness) were positively correlated with the feeling factors, respectively. Based on this result, we can improve the smoke taste quality of cigarettes by adjusting the basic tastes. The underlying mechanism of these correlations may involve the interaction of taste receptors in the oral cavity. Sweetness receptors may have a synergistic or inhibitory effect on other taste and somatosensory receptors. When adjusting the basic tastes, manufacturers need to consider the balance of different flavors. Over-adding sweetening components may lead to an unbalanced flavor profile if not accompanied by appropriate adjustments of other components. Moreover, the cultural background and taste preferences of consumers play a role. Different regions and consumer groups may have different sensitivities and preferences for these taste attributes, which should be taken into account in product development. Adding some degree of sweetening components might improve consumers’ general impression of cigarettes (Table 3).
Pearson correlation coefficients for smoke sensory attributes
| X1 | X2 | X3 | X4 | X5 | X6 | X7 | X8 | X9 | X10 | |
|---|---|---|---|---|---|---|---|---|---|---|
| X1 | 1 | −0.999** | 0.918* | 0.895* | 0.928* | 0.872 | 0.86 | 0.997** | 0.970** | −0.964** |
| X2 | 1 | −0.915* | −0.894* | −0.933* | −0.861 | −0.877 | −0.994** | −0.975** | 0.967** | |
| X3 | 1 | 0.985** | 0.965** | 0.926* | 0.861 | 0.914* | 0.958* | −0.980** | ||
| X4 | 1 | 0.984** | 0.849 | 0.837 | 0.895* | 0.948* | −0.968** | |||
| X5 | 1 | 0.814 | 0.896* | 0.921* | 0.980** | −0.984** | ||||
| X6 | 1 | 0.782 | 0.866 | 0.864 | −0.893* | |||||
| X7 | 1 | 0.821 | 0.941* | −0.923* | ||||||
| X8 | 1 | 0.956* | −0.953* | |||||||
| X9 | 1 | −0.996** | ||||||||
| X10 | 1 |
*On 0.05 level (two-sided test) significant correlation.
**On 0.01 level (two-sided test) significant correlation.
In the sensory evaluation of cigarettes, different compounds play a crucial role in the flavor profile of cigarettes. To gain a deeper understanding of the relationship between these compounds and the sensory attributes of cigarettes, correlation analyses were conducted. In this part, the correlations between major aromatic compounds and sensory attributes of cigarettes will be explored in detail based on the results of Pearson correlation heatmaps and correlation networks.
We constructed a heat map of the correlation between 50 major compounds and the results of sensory analysis using Pearson correlation analysis. Figure 4 illustrates the strength of correlation between these compounds and the sensory attributes of cigarettes. The results showed that there were significant correlations between most of the compounds and the sensory attributes of cigarettes.
Heat map of correlations between major aromatic compounds and sensory attributes. Note: *p ≤ 0.05, **p ≤ 0.01.
By further analyzing the heat map and correlation network (e.g., Figure 5), we found that the following major aromatic compounds were significantly correlated with the sensory attributes of cigarettes: phthalide (87-41-2) and 2,5-dimethylphenol (95-87-4): these two compounds were significantly positively correlated with the savory, bitter, and sour attributes (p ≤ 0.01) and significantly negatively correlated with the sweet taste (p ≤ 0.01). This suggests that they may contribute to the salty, bitter, and sour flavor attributes in cigarettes, while reducing the perception of sweetness. Styrene (100-42-5): positively correlated (p ≤ 0.01) with sweet flavor attributes, while significantly negatively correlated (p ≤ 0.01) with sour flavor attributes. This suggests that styrene may have increased the perception of sweetness in cigarettes while decreasing the perception of sourness. Butyric acid (107-92-6) and benzyl alcohol (100-51-6) showed significant positive correlation (p ≤ 0.01) with sourness and bitterness, while significant negative correlation (p ≤ 0.01) with sweetness. Both compounds equally contributed significantly to the sour and bitter flavors of cigarettes and reduced the sweetness perception. Benzoic acid (65-80-5) showed a significant positive correlation (p ≤ 0.01) with bitter flavor. Benzoic acid may be one of the main sources of bitter flavor in cigarettes. Masoylactone (54814-64-1), β-violetone (14901-07-6), pyridine-2-carboxaldehyde (1121-60-4), and acetylpropionic acid (123-76-2) showed significant positive correlation (p ≤ 0.01) with salty and bitter attributes. These results suggest that they equally contribute significantly to the salty and bitter flavors of cigarettes.
The correlation network between the main aroma components and sensory attributes. Note: Positive correlation is indicated by a yellow line, negative correlation by a gray line, circle size and color shade of sensory attributes indicate the number of relevant objects, and line thickness indicates the degree of correlation (|ρ| > 0.5, p < 0.01).
This study established an optimized analytical framework for cigarette aroma compounds using a smoking machine, Cambridge filter method, SAFE pretreatment, and GC–MS analysis. Five cigarette types were analyzed, identifying key aroma contributors (e.g., 3-ethylphenol, naphthalene, indole) with OAVs >10,000. Sensory evaluation of nine attributes revealed significant differences: M1949M scored highest in sweetness (positively correlated with overall preference), while SK1949M and FOSK exhibited elevated sourness. Heatmap and correlation network analyses demonstrated compound-specific sensory modulation: phthalide (CAS 87-41-2) and 2,5-dimethylphenol (CAS 95-87-4) enhanced salty/bitter/sour notes but suppressed sweetness, whereas styrene (CAS 100-42-5) inversely regulated sweetness and sourness. Statistical analysis confirmed sweetness as a critical driver of positive sensory perception, negatively correlated with throat irritation and astringency (p < 0.05). These findings provide actionable insights for optimizing cigarette sensory profiles through targeted compound modulation. Notably, identified aroma compounds pose health concerns: naphthalene (CAS 91-20-3) exhibits systemic toxicity, while benzaldehyde (CAS 100-52-7) may trigger allergic reactions at elevated concentrations. Metabolic pathways of high-OAV compounds like 3-ethylphenol (CAS 620-17-7) require further toxicological evaluation.