Exploratory Study on Distinguishing Dendrobium Stem and Five Species of Dendrobium Using Heracles Neo Ultra-Fast Gas Phase Electronic Nose

: Dendrobium stem is a valuable food with medicinal and edible properties. Due to its high medicinal value and price, closely related Dendrobium varieties are often sold as imitations on the market. Therefore, there is an urgent need to develop new methods that can quickly identify Dendrobium stem and its closely related species. The Heracles Neo ultra-fast gas phase electronic nose was used in this study to determine and analyze the composition and contents of volatile organic compounds (VOCs) in Dendrobium stem and samples of five other species closely related to it. A total of 20 VOCs were identified, and a fingerprint map of the VOCs was constructed. Principal component analysis (PCA), Euclidean distance, and other methods were used to comprehensively process and analyze the obtained VOC information. The AroChemBase database was also used for qualitative analysis of the compounds. The results showed that there are significant differences in the odor fingerprint spectra of Dendrobium stem and the five other closely related species. The main types of compounds in Dendrobium stem and its five closely related species were organic esters, aldehydes, ketones, and olefins. Among them, 3-methylbutanal and n-butanol were characteristic compounds of the Dendrobium stem sample, while the VOCs acetonitrile and trometamol were present in the five related Dendrobium species samples. The Heracles Neo ultra-fast gas phase electronic nose can quickly and accurately identify Dendrobium stem and its five closely related species. It can also be used for the quality evaluation of Dendrobium stem, providing a theoretical reference for reducing the phenomenon of medicinal confusion in the Dendrobium stem market.


Introduction
Dendrobium officinale Kimura et Migo (Fam.Orchidaceae) is a herbaceous plant that has been the subject of research for many years [1].Dendrobium stem is the dried stem of Dendrobium officinale Kimura et Migo (Fam.Orchidaceae).It has been widely used as a medicinal and edible product for thousands of years, and it is listed as a top-grade traditional Chinese medicine along with Ganoderma lucidum, Ginseng, and Cordyceps sinensis.At the same time, Dendrobium officinale is often used as an ingredient in food.Common Dendrobium stem foods include wine, yogurt, and extract [2].
Dendrobium stem benefits the stomach and fluid generation.It is commonly used for heat-related conditions consuming fluids like dry mouth, irritation, and thirst [3].
Separations 2024, 11, 211 2 of 13 Dendrobium is a collective term for various medicinal plants in the Dendrobium genus, Orchidaceae family.Besides Dendrobium stem, common medicinals include Dendrobium nobile Lindl, Dendrobium huoshanense C.Z.Tang et S.J.Cheng, Dendrobium chrysotoxum Lindl, Dendrobium fimbriatum Hook, and fresh/dried stems of similar species in the same genus.Dendrobium stem, known as the head of the "Nine Immortal Grasses of China," is expensive due to its scarce wild resources and high medicinal value.
Modern pharmacological research has shown that plants of the Dendrobium genus are rich in polysaccharides, alkaloids, volatile oils, various beneficial amino acids, and trace elements for humans [4][5][6].They provide pharmacological activities such as antioxidant, anti-tumor, hypoglycemic, liver protection, and bone enhancement properties and the improvement of endocrine and metabolic disorders and bowel movements [7][8][9][10][11].The seeds of Dendrobium stem have a low germination rate and slow growth under natural conditions.In addition, due to long-term overharvesting and other reasons, the wild resources of Dendrobium stem are scarce [12].Due to the rich nutritional value and high price of Dendrobium stem, closely related species such as Dendrobium fimbriatum Hook.and Dendrobium chrysotoxum Lindl., which have similar appearances and significant differences in medicinal value, are often sold as Dendrobium stem on the market.
At present, Dendrobium stem is typically identified using methods such as source identification, microscopic identification, physicochemical identification, and PCR-RFLP DNA technology [13][14][15][16].The first two methods can be carried out quickly but require the evaluators to have extensive experience, professional knowledge, and strong subjectivity.The latter two have high accuracy and good precision; however, the processes are cumbersome and costly, making it difficult to achieve rapid identification.
The electronic nose, also known as the artificial olfactory system, is an emerging analytical instrument that simulates human olfaction.It is specifically used to detect, analyze, and identify complex flavors and volatile organic compounds (VOCs) [17].Its principle is that when complex gases touch the embedded gas sensor inside, the sensor begins to detect and recognize the gas components, generating an odor fingerprint, thereby achieving the purpose of distinguishing and verifying the authenticity of different samples [18].
The Heracles Neo electronic nose uses two gas chromatography columns with different polarities to separate odor substances.The chromatographic peaks obtained from the gas phase are used as sensors and processed using appropriate models to qualitatively or quantitatively analyze volatile compounds in the sample [19].Compared with ordinary gas detection, its advantages lie in its extremely short analysis time, high sensitivity, real-time detection, high throughput, etc.Its retention index qualitative library, combined with n-alkane calibration, converts retention time into retention index and can qualitatively analyze odor components [20,21].
To establish an accurate and efficient identification method for Dendrobium stem and related species, this study compared differences in volatile organic compounds (VOCs) between Dendrobium stem and samples of five closely related species using the Heracles Neo ultra-fast electronic nose, AroChemBase professional database, principal component analysis (PCA), Euclidean distance analysis, and compound qualitative analysis.Sensory descriptions of flavor characteristics for Dendrobium stem and the five Dendrobium samples are additionally provided.

Materials
Dendrobium stem and the five Dendrobium samples were identified by Prof. Zhaoming Xie at the Hunan University of Chinese Medicine.A voucher specimen (HNUCM2023-SH01) was deposited in Science and Technology Innovation Center of Hunan University of Chinese Medicine.Dendrobium officinale Kimura et Migo (collected from Fenghuang, China, referred to as TPSH), Dendrobium nobile Lindl (collected from Chishui, China, referred to as JCSH), Dendrobium huoshanense C.Z.Tang et S.J. Cheng (collected from Huoshan, China, referred to as HSSH), Dendrobium chrysotoxum Lindl (collected from Jinghong, China, re-ferred to as GZSH), Dendrobium fimbriatum Hook (collected from Wuming, China, referred to as LSSH), and Dendrobium denneanum Kerr (collected from Xingyi, China, referred to as DQSH) were selected as samples in this study in order to explore and study the VOCs of different varieties of Dendrobium.

Sample Preparation
Different varieties of Dendrobium stem and five types of Dendrobium samples were crushed in sequence, passed through a No. 3 sieve, sealed, and refrigerated for later use.The preparation of the samples for Heracles Neo ultra-fast gas phase electronic nose detection was as follows: First, 1.0 g of Dendrobium stem or Dendrobium sample was weighed and placed into a 20 mL headspace bottle attached to the electronic nose and then sealed with PTFE spacers.For each sample, we set up five parallel samples in order to prevent accidental errors and ensure the accuracy of the experiment.The prepared samples were placed on an automatic sampler device for further analysis.

Heracles Neo Ultra-Fast Gas Phase Electronic Nose Analysis Conditions
The detection conditions for the Heracles Neo were determined by optimizing the detection parameters used in previous research: a 20 mL sample bottle; sample weight of 1.0 g; incubation temperature of 80  C; and FID gain of 12. Calibration was performed using a standard solution of n-alkanes (nC6 nC16), and the retention time was converted into a retention index.The chromatographic information was recorded within 275 s.There are two chromatographic columns in the Heracles Neo ultra-fast gas phase electronic nose (Alpha MOS Corporation, Toulouse, France), namely, the low-polarity MXT-5 column and the medium-polarity MXT-1701 column (Alpha MOS Corporation, Toulouse, France).After deducting the blank running reference, the odor chromatograms of Dendrobium stem and the other Dendrobium species could be obtained.Qualitative analysis of the compound was then performed using the AroChemBase database.

Statistical Analysis
Principal component analysis (PCA) and compound qualitative analysis of experimental data were performed using AlphaSoft 17.0 (Alpha MOS Corporation, Toulouse, France), and bar charts were drawn using Origin Pro 2023 software (OriginLab Corporation, Northampton, MA, USA).

Heracles Neo Analysis
The six samples of Dendrobium stem and related species were detected and analyzed using the optimized electronic nose detection parameters mentioned above.Figures 1 and 2 show the different samples represented by different colors.From the gas chromatography overlay graphs, it can be intuitively seen that the detection results of the two chromatography columns are generally similar.There are differences in retention time and peak area among the Dendrobium stem and Dendrobium samples.In the spectrum, the blue color represents a prominent chromatographic peak in the HSSH sample between 0 and 50 s, and there is a characteristic peak near 50-100 s.The yellow color represents a higher chromatographic peak in the TPSH sample than in the other five Dendrobium samples, and the red color represents a lower chromatographic peak in the GZSH sample.The peak height is relatively low between 100 s and 300 s, with the green JCSH sample showing a significant difference in peak height, which is higher than that of the other samples at different peak times.Through the analysis of the original spectra, it can be concluded that the differences in the six Dendrobium stem and Dendrobium samples are mainly reflected in the changes in peak height, which indicates the difference in VOCs.Overall, there are significant differences in odor among the Dendrobium stem and Dendrobium samples, which can be effectively distinguished through odor fingerprint spectra.To further verify the differences between sample groups, PCA statistics were first used to identify the odor differences between sample groups in this study.The differential chromatographic peaks were determined, and then the specific chromatographic peaks were qualitatively analyzed through searching the AroChemBase database.
higher chromatographic peak in the TPSH sample than in the other five Dendrobium samples, and the red color represents a lower chromatographic peak in the GZSH sample.The peak height is relatively low between 100 s and 300 s, with the green JCSH sample showing a significant difference in peak height, which is higher than that of the other samples at different peak times.Through the analysis of the original spectra, it can be concluded that the differences in the six Dendrobium stem and Dendrobium samples are mainly reflected in the changes in peak height, which indicates the difference in VOCs.Overall, there are significant differences in odor among the Dendrobium stem and Dendrobium samples, which can be effectively distinguished through odor fingerprint spectra.To further verify the differences between sample groups, PCA statistics were first used to identify the odor differences between sample groups in this study.The differential chromatographic peaks were determined, and then the specific chromatographic peaks were qualitatively analyzed through searching the AroChemBase database.higher chromatographic peak in the TPSH sample than in the other five Dendrobium samples, and the red color represents a lower chromatographic peak in the GZSH sample.The peak height is relatively low between 100 s and 300 s, with the green JCSH sample showing a significant difference in peak height, which is higher than that of the other samples at different peak times.Through the analysis of the original spectra, it can be concluded that the differences in the six Dendrobium stem and Dendrobium samples are mainly reflected in the changes in peak height, which indicates the difference in VOCs.Overall, there are significant differences in odor among the Dendrobium stem and Dendrobium samples, which can be effectively distinguished through odor fingerprint spectra.To further verify the differences between sample groups, PCA statistics were first used to identify the odor differences between sample groups in this study.The differential chromatographic peaks were determined, and then the specific chromatographic peaks were qualitatively analyzed through searching the AroChemBase database.

Principal Component Analysis (PCA)
Principal component analysis (PCA) is a multivariate statistical method that obtains the maximum difference between data groups based on the original data of the sample without knowledge of their characteristics.After dimensionality reduction and linearization, the data are presented intuitively as two-or three-dimensional graphs [22].Principal component analysis (PCA) is a multivariate statistical method that obtain the maximum difference between data groups based on the original data of the sample without knowledge of their characteristics.After dimensionality reduction and lineariza tion, the data are presented intuitively as two-or three-dimensional graphs [22].

Euclidean Distance Analysis
Table 1 and Figure 4 show the Euclidean distance of Dendrobium stem and the five Dendrobium samples.The larger the distance value, the greater the difference, and the smaller the distance value, the smaller the difference.The distance between DQSH and LSSH samples is the smallest, indicating that these samples have the least difference.The distance between JCSH and TPSH is the largest, indicating that these samples have the most difference, which is consistent with the results of the visual analysis in the PCA dia gram.The second and third largest distances are between HSSH and LSSH and DQSH and HSSH, respectively.The first three components with the smallest distance are com posed of DQSH, LSSH, and HSSH samples in pairs, indicating that these three sample have the least difference in VOCs and the most similar odor.Similarly, the components

Euclidean Distance Analysis
Table 1 and Figure 4 show the Euclidean distance of Dendrobium stem and the five Dendrobium samples.The larger the distance value, the greater the difference, and the smaller the distance value, the smaller the difference.The distance between DQSH and LSSH samples is the smallest, indicating that these samples have the least difference.The distance between JCSH and TPSH is the largest, indicating that these samples have the most difference, which is consistent with the results of the visual analysis in the PCA diagram.The second and third largest distances are between HSSH and LSSH and DQSH and HSSH, respectively.The first three components with the smallest distance are composed of DQSH, LSSH, and HSSH samples in pairs, indicating that these three samples have the least difference in VOCs and the most similar odor.Similarly, the components with the secondand third-smallest relative distance are GZSH and TPSH and HSSH and TPSH, respectively.Among the three components with the largest distance, TPSH has the largest difference in VOCs compared to the other three samples.A loading plot was added based on PCA in order to further investigate the differences in odor between different samples.It was determined that chromatographic peaks with good separation effect, high discrimination ability (discrimination ability > 0.900, peak area > 1000), and clear peak sample differentiation would be the most important.The closer the factor is to the sample, the greater the contribution rate.The experimental results are shown in Figure 5. Based on the chromatographic peak screening, the sample s overall distribution trend remained the same as that of Figure 3, indicating that the chromatographic peaks selected reflect the sample s overall odor.The chromatographic peaks were searched in the AroChemBase database based on their retention indices.These chromatographic peaks were analyzed qualitatively to determine the compounds that caused the differences between Dendrobium stem and Dendrobium samples.A loading plot was added based on PCA in order to further investigate the differences in odor between different samples.It was determined that chromatographic peaks with good separation effect, high discrimination ability (discrimination ability > 0.900, peak area > 1000), and clear peak sample differentiation would be the most important.The closer the factor is to the sample, the greater the contribution rate.The experimental results are shown in Figure 5. Based on the chromatographic peak screening, the sample's overall distribution trend remained the same as that of Figure 3, indicating that the chromatographic peaks selected reflect the sample's overall odor.The chromatographic peaks were searched in the AroChemBase database based on their retention indices.These chromatographic peaks were analyzed qualitatively to determine the compounds that caused the differences between Dendrobium stem and Dendrobium samples.

Compound Identification
Following the abovementioned screening, retention times (Rts) were converted to retention indices (RIs) for chromatographic peaks that meet retention conditions.Data from the AroChemBase database were used to qualitatively analyze the chromatographic peaks that show significant differences.The compounds included in the qualitative analysis are shown in Table 2.The threshold values in Table 1 represent the strength of the substance odor.The threshold values in the air medium are in mg/m 3 , and those in the oil medium are in mg/kg.For two compounds with the same content, the lower the threshold in the same medium, the stronger the odor.The odor activity value (OAV), which is the ratio of the concentrations of the individual substances in the sample to their threshold concentrations, is used for two substances with different concentrations and thresholds.A high ratio indicates that the odor activity value is high, contributing significantly to the overall odor.A low ratio indicates that the odor activity value is small and contributes little to the overall odor.The odor activity value of each compound (odor contribution value) of the sample can be calculated by determining its content and the odor threshold.This value can be used to adjust the sample s odor and trace the source of the odor.
As shown in Tables 2 and 3, a total of 20 odor components were identified.In order to more intuitively compare the content differences of compounds in Dendrobium stem and Dendrobium samples, a bar chart of differential compound contents was drawn based on the data in Table 3, with VOCs as the x-axis and average peak area as the y-axis, as shown in Figure 6.In the bar chart, we can observe that the most obvious feature is GZSH, represented by red, with hexanal compounds, which are the characteristic compounds of GZSH samples, having the highest content.TPSH, represented by orange, has two characteristic compounds with higher content, namely, 3-methylbutanal and n-butanol.Some volatile organic compounds are generally lower in content than they are in other samples.The purple color represents JCSH, which has higher contents of compounds such as propenal, but-(E)-2-enal, and 2,3-pentanedione compared to the other Dendrobium samples.The content changes in DQSH and LSSH samples are relatively similar, and no LSSH components were detected to be higher than in the other Dendrobium samples.The compound content is generally lower than that of other samples.This also reveals that the abovementioned components may be biomarkers for distinguishing odor differences in Dendrobium stem and Dendrobium samples.

Compound Identification
Following the abovementioned screening, retention times (Rts) were converted to retention indices (RIs) for chromatographic peaks that meet retention conditions.Data from the AroChemBase database were used to qualitatively analyze the chromatographic peaks that show significant differences.The compounds included in the qualitative analysis are shown in Table 2.The threshold values in Table 1 represent the strength of the substance odor.The threshold values in the air medium are in mg/m 3 , and those in the oil medium are in mg/kg.For two compounds with the same content, the lower the threshold in the same medium, the stronger the odor.The odor activity value (OAV), which is the ratio of the concentrations of the individual substances in the sample to their threshold concentrations, is used for two substances with different concentrations and thresholds.A high ratio indicates that the odor activity value is high, contributing significantly to the overall odor.A low ratio indicates that the odor activity value is small and contributes little to the overall odor.The odor activity value of each compound (odor contribution value) of the sample can be calculated by determining its content and the odor threshold.This value can be used to adjust the sample's odor and trace the source of the odor.   2 and 3, a total of 20 odor components were identified.In order to more intuitively compare the content differences of compounds in Dendrobium stem and Dendrobium samples, a bar chart of differential compound contents was drawn based on the data in Table 3, with VOCs as the x-axis and average peak area as the y-axis, as shown in Figure 6.In the bar chart, we can observe that the most obvious feature is GZSH, represented by red, with hexanal compounds, which are the characteristic compounds of GZSH samples, having the highest content.TPSH, represented by orange, has two characteristic compounds with higher content, namely, 3-methylbutanal and n-butanol.Some volatile organic compounds are generally lower in content than they are in other samples.The purple color represents JCSH, which has higher contents of compounds such as propenal, but-(E)-2-enal, and 2,3-pentanedione compared to the other Dendrobium samples.The content changes in DQSH and LSSH samples are relatively similar, and no LSSH components were detected to be higher than in the other Dendrobium samples.The compound content is generally lower than that of other samples.This also reveals that the abovementioned components may be biomarkers for distinguishing odor differences in Dendrobium stem and Dendrobium samples.

Discussion
Heracles Neo ultra-fast gas phase electronic nose technology was used in this study to conduct in-depth chemical composition analysis of Dendrobium stem and its related species, and 20 compounds were successfully identified.It is worth noting that in the TPSH samples, we found two unique compounds: 3-methylbutanal and n-butanol.The presence of these two compounds may be the key factor leading to significant differences between Dendrobium stem and its related species.In order to further analyze and distinguish the odor characteristics of different Dendrobium samples, we established an odor fingerprint map containing Dendrobium samples and Dendrobium stem based on the chromatographic peaks of key chemical components that affect their odor.Using these data, a stoichiometric PCA model was also established with a cumulative contribution rate of 79.158% (the sum of the contribution rates of the first principal component (PC1) and the second principal component (PC2)).According to the standards of chemometrics, if the overall contribution rate of a model exceeds 70% to 85%, the model has good discrimination ability.This indicates that the model can quickly distinguish, identify, and analyze the odor of Dendrobium stem and its related species, providing a scientific basis for quality control and source tracing.
In addition, the AroChemBase database was used to qualitatively identify the chromatographic peaks with significant differences, and fingerprint and qualitative substance analyses were performed on Dendrobium stem and the five types of Dendrobium samples.Different varieties of Dendrobium samples showed differences in component types.By combining the bar charts of differential compound contents, it can be concluded that the composition of Dendrobium stem and the five types of Dendrobium samples is complex, which is mainly reflected in the relative contents of ethanol, propylene, 2,3-pentanedione, and hexanal.In the evaluation of flavor characteristics, the main sensory properties of Dendrobium stem and the five related species are fruity, sweet, green, and fresh.3-Methylbutanal and n-butanol are characteristic compounds of the TPSH sample which endow TPSH with unique malt, nut, and chocolate flavors [23].Therefore, these two compounds can be used to distinguish Dendrobium stem from other sampled varieties.In addition to these two characteristic compounds, the contents of individual compounds in Dendrobium stem samples are generally lower than those of the other Dendrobium samples.But-(E)-2-enal, acetonitrile, and trometamol are differential VOCs with a light aroma found in all five Dendrobium samples but not the TPSH sample.However, their contents are similar in the five Dendrobium samples, so they can only be used to distinguish Dendrobium stem from the other five Dendrobium samples.Hexanal is a characteristic compound of GZSH with a special herbal fragrance.Its content is much higher in GZSH samples than in the other five samples.Using a gas chromatography overlay, it can be seen that the peak of this compound appears at around 95 s.This makes hexanal the main characteristic compound distinguishing GZSH from the other Dendrobium samples.In addition, 2,4-octadience was detected in all samples.It has a strong herbal odor and is a key substance contributing to the flavor of the Dendrobium samples.Based on all the analysis results, it can be concluded that the compound differences in TPSH samples are the greatest, which is consistent with existing studies on the quality evaluation of Dendrobium stem [24].
It is important to recognize that odor, as a criterion evaluating medicinal materials and food, is closely related to the intrinsic composition of substances.In the past few decades, new biomimetic sensing technologies have increasingly been applied to quality detection, safety control, and other aspects of life with the development of science and technology [25][26][27][28].In this study, the Heracles Neo ultra-fast gas phase electronic nose was used instead of the human olfactory system.This study confirms the importance of electronic nose technology in the analysis of traditional Chinese medicine odors.The method effectively avoids human factors and increases evaluation accuracy by converting complex odor information into quantifiable data.This approach is in contrast to traditional methods that rely on appearance, color, and the sense of smell to distinguish small differences between samples [29,30].

Conclusions
A rapid detection method, Heracles Neo ultra-fast gas phase electronic nose technology, for Dendrobium stem and its closely related species was preliminarily explored in this study.The following conclusions can be drawn: 1.
The VOCs of Dendrobium stem and five types of Dendrobium samples were analyzed, and a total of 20 odor chemical components were identified.The gas chromatogram results obtained from the MXT-1701 and MXT-5 chromatographic column analysis were similar.It was observed that there are certain differences in the retention time and peak area of the six Dendrobium stem and Dendrobium samples.2.
PCA also verified that there are obvious odor differences between Dendrobium stem and its related Dendrobium species.

3.
According to the Euclidean distance, the difference between the Dendrobium stem sample and the JCSH sample was the largest.4.
The AroChemBase database was used to characterize the screened chromatographic peaks with large differences, and two unique compounds in the Dendrobium stem sample were found: 3-methylbutanal and n-butanol.
This work provides a reference for the classification of the odor patterns, characteristic description, and origin classification of Dendrobium.The results of this study will provide technical support for the identification and quality control of Dendrobium stem and five species of Dendrobium and are of great significance for ensuring the quality and value of Dendrobium stem on the market.This study also lays a solid theoretical foundation for the application of Heracles Neo ultra-fast gas phase electronic nose technology in other fields.

Figure 3
Figure 3 shows the PCA of Dendrobium stem and the five Dendrobium samples.The horizontal and vertical coordinates represent the contribution rates of the first principal component (PC1) and the second principal component (PC2) obtained via PCA, respectively.From the figure, it can be seen that the sum of the contribution rates of the first principal component (PC1) and the second principal component (PC2) reaches 79.158%, which can better reflect the actual situation of the sample.A recognition index between 80 and 100 indicates effective differentiation.The recognition index of the sample on the electronic nose principal component analysis chart reaches 93, indicating that the difference in odor can effectively distinguish Dendrobium stem and the five Dendrobium samples.The position distribution and distance of Dendrobium stem and the five Dendrobium samples in the PCA chart reflect the degree of odor difference between samples.It can be seen that the degree of odor difference between the samples is different.In the PCA chart, the closer the distance, the smaller the sample difference, and the farther the distance, the greater the sample difference.The TPSH sample is located in the left area of the figure, while the other Dendrobium samples are located in the right area, indicating a significant difference in overall odor between Dendrobium stem and the other Dendrobium samples.The close distribution of LSSH and DQSH indicates that the overall odor difference between the two groups of samples is relatively small.The farthest distance between GZSH and JCSH indicates the greatest difference in VOCs between the two.
shows the PCA of Dendrobium stem and the five Dendrobium samples.The horizontal and vertical coordinates represent the contribution rates of the first principa component (PC1) and the second principal component (PC2) obtained via PCA, respec tively.From the figure, it can be seen that the sum of the contribution rates of the firs principal component (PC1) and the second principal component (PC2) reaches 79.158% which can better reflect the actual situation of the sample.A recognition index between 80 and 100 indicates effective differentiation.The recognition index of the sample on the elec tronic nose principal component analysis chart reaches 93, indicating that the difference in odor can effectively distinguish Dendrobium stem and the five Dendrobium samples The position distribution and distance of Dendrobium stem and the five Dendrobium sam ples in the PCA chart reflect the degree of odor difference between samples.It can be seen that the degree of odor difference between the samples is different.In the PCA chart, the closer the distance, the smaller the sample difference, and the farther the distance, the greater the sample difference.The TPSH sample is located in the left area of the figure while the other Dendrobium samples are located in the right area, indicating a significan difference in overall odor between Dendrobium stem and the other Dendrobium samples The close distribution of LSSH and DQSH indicates that the overall odor difference be tween the two groups of samples is relatively small.The farthest distance between GZSH and JCSH indicates the greatest difference in VOCs between the two.

Figure 3 .
Figure 3. Principal component analysis of Dendrobium stem and five Dendrobium samples.

Figure 3 .
Figure 3. Principal component analysis of Dendrobium stem and five Dendrobium samples.

Figure 5 .
Figure 5. Principal component analysis and loading diagram of Dendrobium stem and five Dendrobium samples.

Figure 5 .
Figure 5. Principal component analysis and loading diagram of Dendrobium stem and five Dendrobium samples.

Figure 6 .
Figure 6.Histogram of differential compound contents.Figure 6. Histogram of differential compound contents.

Figure 6 .
Figure 6.Histogram of differential compound contents.Figure 6. Histogram of differential compound contents.
• C; incubation time of 20 min; initial temperature of the trap of 40 • C; final temperature of the trap of 240 • C; trap diversion rate of 0 mL•min −1 ; capture duration of 55 s; injection port temperature of 200 • C; injection volume of 5000 µL; injection speed of 250 µL/s; injection duration of 25 s; initial temperature of the column of 40 • C; column temperature program heating method of 0.5 • C/s −150 • C and 3 • C/s to 250 • C; collection time of 290 s; detector temperature of 260

Table 1 .
Euclidean distance between samples of Dendrobium stem and the five Dendrobium samples.

Table 2 .
Differential chromatographic peak qualitative results and odor descriptions.

Table 3 .
Average peak area of differential chromatographic peaks.

Table 3 .
Average peak area of differential chromatographic peaks.