A rapid detection method for costus and costus root

Abstract

The application discloses a rapid detection method for costus and Sichuan costus, which comprises the following steps: 1) preparation of a test sample; 2) sealing of nC6-nC16 normal alkane mixed control samples and the test sample in a headspace sampling bottle respectively, and detection by using an electronic nose; and detection parameters are as follows: a chromatographic column is a metal chromatographic column; an initial column temperature is 80 DEG C; a temperature rising program is as follows: rising to 160 DEG C at a speed of 1 DEG C / s, rising to 250 DEG C at a speed of 2 DEG C / s, and maintaining for 21 s; and a collection time is 148 s. In the application, each odor component is effectively separated by using specific electronic nose detection parameters, an odor fingerprint is established, and it is found that costus and Sichuan costus from different sources have great differences by comparing odor fingerprint information, and the odor fingerprint can be used to distinguish costus and Sichuan costus.

Description

Technical Field

[0001] This invention specifically relates to a rapid detection method for costus root and Sichuan costus root. Background Technology

[0002] Costus root is a plant belonging to the Asteraceae family. Aucklandia lappa The dried root of *Desmodium styracifolium*, native to India, is now mainly produced in Yunnan, Shaanxi, and Sichuan provinces of my country, especially in northwestern Yunnan where it is widely cultivated and produced in large quantities. It is an important local medicinal herb of Yunnan, possessing the effects of regulating qi and relieving pain, strengthening the spleen and promoting digestion. It is a key medicine for regulating qi and relieving pain, as well as for treating damp-heat diarrhea with tenesmus. It can be used for chest and rib pain, abdominal distension and pain, tenesmus after diarrhea, indigestion, and loss of appetite. Furthermore, roasted *Desmodium styracifolium* can also solidify the intestines and stop diarrhea. *Desmodium styracifolium* is a plant of the Asteraceae family. Vladimiria souliei (Franch.) Ling or Grey-haired Sichuan Oak Vladimiria souliei The dried root of *Franch.* Ling var. *cinerea* Ling is a Tibetan medicine with distinctive ethnic characteristics, used in both Traditional Chinese Medicine and Tibetan Medicine. It is produced in Sichuan and Tibet and has the effects of regulating qi and relieving pain. It can be used to treat chest and abdominal distension and pain, borborygmus and diarrhea, tenesmus, and other ailments. Reports indicate that the traditional effects of *Franch.* Ling and *Franch.* Ling differ; *Franch.* Ling focuses on regulating qi and relieving pain, while *Franch.* Ling var. *cinerea* Ling excels at strengthening the spleen and resolving stagnation, primarily treating abdominal pain and diarrhea.

[0003] Although both *Saussurea costus* and *Saussurea involucrata* belong to the Asteraceae family, they are classified into different genera in plant taxonomy. *Saussurea costus* belongs to the *Saussurea* genus, and its original plant is a tall herb; *Saussurea involucrata* belongs to the *Saussurea* genus, and its original plant is a perennial, stemless or nearly stemless rosette-shaped herb. The original plants of both are easily identifiable, but the appearance of their roots, the medicinal parts, is quite similar (see...). Figure 1 and Figure 2 ), especially after being processed into medicinal slices, they are even more difficult to distinguish (see Figure 3 and Figure 4 ), and it requires an experienced pharmacist to detect its authenticity.

[0004] Besides morphological testing, common methods for detecting Costus root and Sichuan costus include microscopic detection, thin-layer chromatography (TLC), and high-performance liquid chromatography (HPLC). Many studies report that Costus root and Sichuan costus have extremely similar microscopic characteristics, making them difficult to distinguish using microscopic detection. Lu Luyang et al. performed TLC on 23 batches of the two medicinal materials according to the methods for content determination of Costus root and Sichuan costus as listed in the pharmacopoeia. Neither the TLC nor HPLC methods described in the pharmacopoeia could distinguish between the two medicinal materials.

[0005] Odor is an important indicator in the empirical testing of Chinese medicinal materials, and each medicinal material has its unique odor. The pharmacopoeia describes the odor of costus root as "a distinctive fragrance and a slightly bitter taste," while describing the odor of Sichuan costus root as "a slightly fragrant odor, a bitter taste, and sticky when chewed." This indicates that there is a significant difference in the odors of costus root and Sichuan costus root. Traditional sensory evaluation mainly relies on experienced pharmacists to judge by smelling and tasting. However, due to differences in the sensitivity of different people to odor perception and the fact that human senses are prone to fatigue, it is difficult to objectively quantify the characteristics of odor.

[0006] Electronic nose, also known as odor fingerprinting technology, simulates the human olfactory organ by using odor sensors to convert sample odor signals into electrical signals. Through identification, it quickly obtains the overall odor information of the sample. As a gas detection technology based on changes in volatile compounds, electronic nose has the advantages of accuracy, sensitivity, non-destructive nature, and speed. It can comprehensively and completely reflect the overall odor characteristics of Chinese medicinal materials and is widely used in fields such as the identification of Chinese medicinal material sources, quality control of processed medicinal materials, process control of Chinese medicine production, quality assessment of Chinese medicinal materials, and detection of deteriorated medicinal materials. Although electronic nose performs excellently in identifying certain specific odors, its ability to distinguish complex odor mixtures or low-concentration gases is often limited. Therefore, it is necessary to develop a method based on electronic nose for the accurate and rapid detection of costus root and Sichuan costus root. Summary of the Invention

[0007] To address the above problems, this invention provides a rapid detection method for Costus Root and Sichuan Costus Root, comprising the following steps: 1) Preparation of the test sample: Take the sample to be tested, crush it, and sieve it to obtain the test sample; 2) Take the n-alkane nC6~nC16 mixed reference standard and test sample into headspace vials and seal them. Use an electronic nose for detection. The detection parameters are: chromatographic column: metal chromatographic column; initial column temperature 80℃; temperature program: increase to 160℃ at 1℃ / s, increase to 250℃ at 2℃ / s, hold for 21s, and acquisition time 148s.

[0008] Further, the sample to be tested in step 1) is costus root and / or Sichuan costus root.

[0009] Furthermore, the sieving described in step 1) is sieving through a No. 2 sieve.

[0010] Furthermore, the metal chromatographic column is a low-polarity and / or medium-polarity metal chromatographic column; the low-polarity metal chromatographic column includes MXT-5; the medium-polarity metal chromatographic column includes MXT-1701.

[0011] Furthermore, the detection parameters in step 2) also include: a sample amount of 2.0 g of the n-alkane nC6-nC16 mixed reference standard or test sample; an incubation temperature of 40℃; an incubation time of 20 min; an injection volume of 5000 μL; and an injection rate of 125 μL·s.-1 The injection duration is 45 seconds.

[0012] Furthermore, the detection parameters also include an inlet pressure of 10 kPa; an inlet temperature of 200°C; an initial trap temperature of 50°C; and a trap split rate of 10 mL·min. -1 The final temperature of the trap is 250℃.

[0013] Furthermore, the electronic nose is the Heracles II ultrafast gas phase electronic nose.

[0014] Furthermore, using the Heracles II ultra-fast gas chromatographic electronic nose to perform Kovats retention index conversion on the odor fingerprint chromatogram of the test sample collected with a low-polarity metal chromatographic column, chromatographic peaks with retention indices of 947, 994, 1017, 1028, 986, 1035, 1069, 1195, 1438, 1415, 1470, 1490, and 1538 were observed. And / or, using Kovats retention index conversion on the odor fingerprint chromatogram of the test sample collected with a medium-polarity metal chromatographic column, chromatographic peaks with retention indices of 961, 1018, 1044, 1059, 1068, 1086, 1100, 1288, 1462, 1496, 1552, 1574, and 1703 were observed, indicating that the test sample was costus root.

[0015] Furthermore, using the Heracles II ultra-fast gas chromatographic electronic nose to perform Kovats retention index conversion on the odor fingerprint chromatogram of the test sample collected with a low-polarity metal chromatographic column, chromatographic peaks with retention indices of 1150, 1235, 1375, 1438, 1415, 1470, 1490, 1416, and 1538 were observed; and / or, using Kovats retention index conversion on the odor fingerprint chromatogram of the test sample collected with a medium-polarity metal chromatographic column, chromatographic peaks with retention indices of 1227, 1393, 1426, 1462, 1496, 1552, 1574, 1623, and 1703 were observed, thus identifying the test sample as Sichuan costus root.

[0016] Furthermore, the Heracles II ultra-fast gas chromatographic electronic nose uses chromatographic peaks with a variable contribution of 0.75 or higher in the collected chromatographic information of the test sample as sensors to establish a PCA model, and detects Costus root and Sichuan costus root through the PCA model.

[0017] Furthermore, DFA analysis or soft independent modeling analysis using PCA model were used to detect Saussurea costus and Saussurea leucantha.

[0018] This invention provides a rapid detection method for costus root and Sichuan costus root, based on the Heracles II ultra-fast gas phase electronic nose. Through specific detection parameters, it effectively separates each odor component and establishes an odor fingerprint spectrum. By comparing the odor fingerprint information, it is found that there are significant differences between costus root and Sichuan costus root from different sources. The odor fingerprint spectrum can effectively distinguish between costus root and Sichuan costus root.

[0019] This invention utilizes the Heracles II ultrafast gas phase electronic nose data analysis software to establish corresponding chemometric models based on collected odor fingerprint information. These models include principal component analysis, discriminant factor analysis, and soft independent modeling analysis to analyze the data. The results show that several models mutually validate each other, all capable of rapidly identifying *Saussurea costus* and *Saussurea involucrata*. This invention provides an effective method for quickly and accurately distinguishing *Saussurea costus* and *Saussurea involucrata*, possessing practical application value.

[0020] Obviously, based on the above description of the present invention, and according to common technical knowledge and conventional methods in the field, various other modifications, substitutions or alterations can be made without departing from the basic technical concept of the present invention.

[0021] The following detailed embodiments further illustrate the above-described content of the present invention. However, this should not be construed as limiting the scope of the present invention to the following examples. All technologies implemented based on the above-described content of the present invention fall within the scope of the present invention. Attached Figure Description

[0022] Figure 1 Image of costus root medicinal material Figure 2 Image of Sichuan costus root medicinal material Figure 3 Costus root slices Figure 4 Sichuan wood incense slices Figure 5 Heracles II aroma fingerprints of 20 batches of Costus root and 16 batches of Sichuan costus root Figure 6 The shared scent pattern of Heracles II, featuring woody and river woody notes. Figure 7 PCA plots of the Costus root and Sichuan costus samples Figure 8 DFA diagrams of the samples of Costus root and Sichuan costus root Figure 9 SIMCA diagram of samples of Costus root and Sichuan costus Detailed Implementation Example 1: Rapid detection of Costus root and Sichuan costus root according to the present invention 1) Preparation of the test sample: Take the sample to be tested, crush it, and sieve it through a No. 2 sieve to obtain the test sample; 2) Seal the n-alkanes nC6–nC16 mixed reference standard and the test sample in headspace vials, and detect them using a Heracles II ultra-fast gas chromatograph electronic nose. The detection parameters are as follows: chromatographic column: low-polarity metal column MXT-5 and medium-polarity metal column MXT-1701; sample volume of n-alkanes nC6–nC16 mixed reference standard or test sample: 2.0 g; incubation temperature: 40℃; incubation time: 20 min; injection volume: 5000 μL; injection rate: 125 μL·s. -1 The injection duration was 45 seconds; the injection port pressure was 10 kPa; the injection port temperature was 200℃; the initial temperature of the trap was 50℃; and the trap split rate was 10 mL / min. -1 The final temperature of the trap was 250℃; the initial temperature of the column was 80℃; the heating program was: 1℃ / s to 160℃, 2℃ / s to 250℃, held for 21s, and the collection time was 148s.

[0023] 3) Odor fingerprint spectra of the nC6-nC16 mixed reference standard and the test sample were collected. Qualitative analysis was performed on the odor fingerprint spectra of the test sample collected using the MXT-5 metal chromatographic column (i.e., calibration was performed using the nC6-nC16 mixed reference standard, retention time was converted into retention index, and then qualitative analysis was performed using the Kovats retention index (RI)). The results showed chromatographic peaks with RIs of 1150, 1235, 1375, 1438, 1415, 1470, 1490, 1416, and 1538. Qualitative analysis was performed on the odor fingerprint spectra of the test sample collected using the MXT-1701 medium polar metal chromatographic column. The results showed chromatographic peaks with RIs of 1227, 1393, 1426, 1462, 1496, 1552, 1574, 1623, and 1703. The test sample was identified as Aucklandia lappa. Qualitative analysis of the odor fingerprint chromatogram of the test sample collected using a metal chromatographic column MXT-5 revealed chromatographic peaks with RI values ​​of 947, 994, 1017, 1028, 986, 1035, 1069, 1195, 1438, 1415, 1470, 1490, and 1538. Qualitative analysis of the odor fingerprint chromatogram of the test sample collected using a moderately polar metal chromatographic column MXT-1701 revealed chromatographic peaks with RI values ​​of 961, 1018, 1044, 1059, 1068, 1086, 1100, 1288, 1462, 1496, 1552, 1574, and 1703, indicating that the test sample was identified as costus root.

[0024] Example 2: Rapid identification of Costus root and Sichuan costus root according to the present invention 1) Preparation of the test sample: Take the sample to be tested, crush it, and sieve it through a No. 2 sieve to obtain the test sample; 2) Seal the n-alkanes nC6–nC16 mixed reference standard and the test sample in headspace vials, and detect them using a Heracles II ultra-fast gas chromatograph electronic nose. The detection parameters are as follows: chromatographic column: low-polarity metal column MXT-5 and medium-polarity metal column MXT-1701; sample volume of n-alkanes nC6–nC16 mixed reference standard or test sample: 2.0 g; incubation temperature: 40℃; incubation time: 20 min; injection volume: 5000 μL; injection rate: 125 μL·s. -1 The injection duration was 45 seconds; the injection port pressure was 10 kPa; the injection port temperature was 200℃; the initial temperature of the trap was 50℃; and the trap split rate was 10 mL / min. -1 The final temperature of the trap was 250℃; the initial temperature of the column was 80℃; the heating program was: 1℃ / s to 160℃, 2℃ / s to 250℃, held for 21s, and the collection time was 148s.

[0025] 3) Collect the odor fingerprint chromatograms of the mixed reference standard and test sample of n-alkane nC6-nC16, screen the chromatographic peaks with a variable contribution of 0.75 or higher as sensors, establish a PCA model, and identify the costus root and Sichuan costus root through the PCA model; or: identify the costus root and Sichuan costus root through DFA analysis or soft independent modeling analysis using the PCA model.

[0026] The following experimental examples illustrate the beneficial effects of the present invention.

[0027] Experimental Example 1 1. Instruments and Materials 1.1 Instruments The Heracles II ultrafast gas chromatography electronic nose (Alpha MOS Instruments, France), equipped with an HS100 fully automated headspace sampler; a variable-speed high-speed rotary pulverizer (FRITSCH, Germany); a grinding mill (Cixi Naio Electric Co., Ltd.); an AUX220 electronic balance (Shimadzu Corporation, Japan); and pipettes (Eppendorf, Germany). 1.2 Materials The n-alkane nC6-nC16 mixed reference standard (batch number: A0168401) was purchased from RESTEK Company, USA. Twenty batches of Aucklandia lappa medicinal materials used in the experiment were identified as Aucklandia lappa by Luo Xiao, Deputy Chief Pharmacist of the Department of Traditional Chinese Medicine, Chengdu Institute for Drug Control. It is the dried root of Aucklandia lappa Decne., a plant of the Asteraceae family. Sixteen batches of Aucklandia lappa medicinal materials were identified as the dried roots of Vladimiria souliei (Franch.) Ling or Vladimiria souliei (Franch.) Ling var. cinerea Ling, both of the Asteraceae family. Specific sample information is shown in Table 1.

[0028] Table 1 Information on samples of Costus root and Sichuan costus root 2. Methods and Results 2.1 Preparation of reference solution Accurately measure 20 μL of a mixed standard solution of n-alkanes (nC6 to nC16), place it in a 20 mL headspace vial, and seal the vial with a headspace cap containing an aluminum silicone pad to obtain the mixed standard solution.

[0029] 2.2 Preparation of the test solution Randomly select appropriate amounts of each batch of Costus root and Sichuan costus root samples using the quartering method, grind them into powder, pass them through a No. 2 sieve, and seal them for later use. For analysis, accurately weigh 2.0 g of the sample powder into a 20 mL headspace vial, and seal the vial with an aluminum-silicone gasket cap to obtain the test solution.

[0030] 2.3 Establishment of odor detection methods for Costus root and Sichuan costus root To obtain good analytical results, single-factor investigations were conducted on key factors affecting the detection results, such as column temperature program, sample fineness, incubation temperature, incubation time, injection volume, and sample amount, using samples of Costus root (MX-6) and Sichuan costus root (CMX-4), to optimize the detection conditions.

[0031] 2.3.1 Column Temperature Program Investigation: The same batch of samples was taken, ground into powder and passed through a No. 2 sieve. The sample weight was fixed at 1g, the incubation temperature was 40℃, the incubation time was 20min, and the injection volume was 1000μl. Five different column temperature ramping programs were investigated. The final column temperature conditions were determined as follows: initial column temperature 80℃, ramping program: ramping to 160℃ at 1℃ / s, ramping to 250℃ at 2℃ / s, holding for 21s; acquisition time 148s.

[0032] 2.3.2 Sample Fineness Investigation: The same batch of samples was taken, with a fixed sample weight of 1g, an incubation temperature of 40℃, an incubation time of 20 min, and an injection volume of 1000 μL. The powder was then passed through No. 2, No. 3, No. 4, and No. 5 sieves for analysis. The results showed that as the sample fineness increased, the response value of the chromatographic peaks gradually decreased. The samples of Costus Root and Sichuan Costus Root had the most chromatographic peaks when passed through No. 2 sieve, with better separation and higher response values. Therefore, No. 2 sieve was selected as the optimal condition.

[0033] 2.3.3 Incubation Temperature Investigation: The same batch of samples was ground and sieved through a No. 2 sieve. Under a fixed incubation time of 20 min and an injection volume of 1000 μL, incubation temperatures of 35℃, 40℃, 50℃, 60℃, 70℃, and 80℃ were investigated. The results showed that for the *Saussurea costus* (MX-6) sample, before reaching 40℃, the chromatographic peak response value increased with increasing temperature; after 40℃, the number of chromatographic peaks decreased with increasing temperature, and the chromatogram was not intuitive. For the *Saussurea costus* (CMX-4) sample, at an incubation temperature of 35℃, the chromatographic peak response value was low and the baseline was unstable; at 40℃, the chromatographic baseline was stable; after 40℃, the chromatographic peak response value increased with increasing incubation temperature, but the number of chromatographic peaks remained consistent. Based on comprehensive comparison, 40℃ was ultimately selected as the optimal incubation temperature.

[0034] 2.3.4 Incubation Time Investigation: The same batch of samples was ground and sieved through a No. 2 sieve. Under a fixed incubation temperature of 40℃ and an injection volume of 1000 μL, incubation times of 10 min, 15 min, 20 min, 25 min, and 30 min were investigated. The results showed that the response values ​​of the chromatographic peaks gradually increased with the extension of incubation time. When the incubation time reached 20 min, the chromatographic peak response values ​​tended to saturate. Therefore, 20 min was selected as the optimal incubation time.

[0035] 2.3.5 Injection Volume Investigation: The same batch of samples was taken, ground, and sieved through a No. 2 sieve. The incubation temperature was fixed at 40℃, and the incubation time was 20 min. Injection volumes of 1000 μL, 2000 μL, 3000 μL, 40000 μL, and 5000 μL were investigated. The results showed that the response value of the chromatographic peak gradually increased with the increase of the injection volume. The peak response value was the largest at an injection volume of 5000 μL, and no peak flattening or other issues were observed. Therefore, 5000 μL was selected as the final injection volume.

[0036] 2.3.6 Sample Dosage Investigation: The same batch of samples was taken, ground, and sieved through a No. 2 sieve. The incubation temperature was fixed at 40℃, the incubation time at 20 min, and the injection volume at 5000 μL. Sample dosages of 0.5 g, 1.0 g, 1.5 g, 2.0 g, and 3.0 g were investigated. The results showed that the chromatographic peak response value gradually increased with increasing sample dosage. When the sample dosage reached 2.0 g, the chromatographic peak response value tended to saturate. Therefore, 2.0 g was selected as the optimal sample dosage.

[0037] 2.3.7 Determination of Detection Conditions: Based on the results of the single-factor investigation experiment, the specific detection parameters were determined as follows: sample vial: 20 mL; fineness of medicinal material: passing through a No. 2 sieve; sample volume: 2.0 g; incubation temperature: 40℃; incubation time: 20 min; injection volume: 5000 μL; injection rate: 125 μL·s. -1 The injection duration was 45 seconds; the injection port pressure was 10 kPa; the injection port temperature was 200℃; the initial temperature of the trap was 50℃; and the trap split rate was 10 mL / min. -1 The final temperature of the trap was 250℃; the initial temperature of the column was 80℃; the heating program was: 1℃ / s to 160℃, 2℃ / s to 250℃, held for 21s, and the collection time was 148s.

[0038] 2.4 Study on the aroma fingerprint spectrum of Costus root and Sichuan costus root 2.4.1 Establishment and Analysis of Fingerprint Spectra Following the sample preparation and detection conditions determined in section “2.3”, 20 batches of Costus Root and 16 batches of Sichuan Costus Root were tested. The collected spectral information was imported into Origin 2021 function plotting software to obtain odor fingerprint spectra for two chromatographic columns with different polarities: MXT-5 (low polarity) and MXT-1701 (medium polarity). The results are shown in Figure 5 and [Figure 6]. Figure 6 As shown in the figure, multiple odor information was detected in *Saussurea costus* from 40s to 60s, while almost none was detected in *Saussurea involucrata*. From 60s to 80s, more odor information was detected in *Saussurea involucrata* than in *Saussurea costus*. From 80s to 130s, multiple odor information was detected in both *Saussurea costus* and *Saussurea involucrata*, but they were slightly different. By comparing the odor fingerprint information, it was found that there are significant differences between *Saussurea costus* and *Saussurea involucrata* from different sources. They can be well distinguished by odor fingerprint spectrum.

[0039] 2.4.2 Precision Test: Appropriate amounts of Costus root (MX-6) and Sichuan costus root (CMX-4) were ground into powder and passed through a No. 2 sieve as test samples. Six portions of each sample, 2.0 g each, were accurately weighed and measured according to the optimized parameters in section "2.3.7". The retention times and peak areas of 13 chromatographic peaks, which accounted for more than 1% of the total peak area of ​​Costus root (MX-6) and were common to 20 batches of Costus root, were statistically analyzed. The results showed that the RSD values ​​of the peak areas of the 13 peaks were 1.19%–4.76%, all less than 5.00%; the RSD values ​​of the peak retention times of the 13 peaks were 0.02%–0.18%, all less than 1.00%. The retention time and peak area of ​​nine chromatographic peaks, which account for more than 1% of the total peak area of ​​CMX-4 and are common to 16 batches of CMX-4, were statistically analyzed. The results showed that the peak area RSD of the nine peaks was 1.30% to 3.67%, all less than 5.00%; the peak retention time RSD of the nine peaks was 0.04% to 0.08%, all less than 1.00%.

[0040] 2.4.3 Repeatability Test: Appropriate amounts of Costus root (MX-6) and Sichuan costus root (CMX-4) were randomly and evenly divided into 6 portions each. Each portion was ground into powder and passed through a No. 2 sieve as the test sample. 2.0 g of each test sample was accurately weighed and measured according to the optimized parameters in section "2.3.7". The retention times and peak areas of 13 chromatographic peaks, accounting for more than 1% of the total peak area of ​​Costus root (MX-6) and common to 20 batches of Costus root, were statistically analyzed. The results showed that the RSD values ​​of the peak areas of the 13 peaks were 1.35%–4.30%, all less than 5.00%; the RSD values ​​of the peak retention times of the 13 peaks were 0.02%–0.08%, all less than 1.00%. The retention time and peak area of ​​nine chromatographic peaks, which account for more than 1% of the total peak area of ​​CMX-4 and are common to 16 batches of CMX-4, were statistically analyzed. The results showed that the peak area RSD values ​​of the nine peaks were 1.74% to 4.67%, all less than 5.00%; the peak retention time RSD values ​​of the nine peaks were 0.04% to 0.08%, all less than 1.00%.

[0041] 2.4.4 Stability Test: Appropriate amounts of Costus root (MX-6) and Sichuan costus root (CMX-4) were ground into powder and passed through a No. 2 sieve as test samples. 2.0 g of each sample was accurately weighed at 0, 4, 8, 12, 16, and 24 h, and the results were determined according to the optimized parameters in section “2.3.7”. The retention times and peak areas of 13 chromatographic peaks, which accounted for more than 1% of the total peak area of ​​Costus root (MX-6) and were common to 20 batches of Costus root, were statistically analyzed. The results showed that the RSD values ​​of the peak areas of the 13 peaks were 1.17%–4.86%, all less than 5.00%; the RSD values ​​of the peak retention times of the 13 peaks were 0.02%–0.08%, all less than 1.00%. The retention time and peak area of ​​nine chromatographic peaks, which account for more than 1% of the total peak area of ​​CMX-4 and are common to 16 batches of CMX-4, were statistically analyzed. The results showed that the peak area RSD values ​​of the nine peaks were 2.93% to 4.79%, all less than 5.00%; the peak retention time RSD values ​​of the nine peaks were 0.02% to 0.09%, all less than 1.00%.

[0042] 2.5 Analysis and Study of the Aroma Information of Costus Root and Sichuan Costus Root The retention times of the 13 common peaks in *Saussurea costus* and the 9 common peaks in *Saussurea costus*, measured under section “2.4”, were calibrated using a mixed standard of n-alkane nC6–nC16. The retention times were converted to retention indices, and then qualitative analysis was performed using the Kovats retention index. The chromatographic peaks were matched and analyzed in the ArochemBase database. The results are shown in Table 2. Unique chromatographic peaks detected in *Saussurea costus* include peak 1 [RI(MXT-5): 947; RI(MXT-1701): 961], peak 2 [RI(MXT-5): 994; RI(MXT-1701): 1018], peak 3 [RI(MXT-5): 1017; RI(MXT-1701): 1044], and peak 4 [RI(MXT-5): 1028; RI(MXT-1701): 1059]. Peak 5 [RI(MXT-5): 986; RI(MXT-1701): 1068], Peak 6 [RI(MXT-5): 1035; RI(MXT-1701): 1086], Peak 7 [RI(MXT-5): 1069; RI(MXT-1701): 1100], Peak 9 [RI(MXT-5): 1195; RI(MXT-1701): 1288]; Unique chromatograms detected in *Saussurea costus*. Peaks 8 [RI(MXT-5): 1150; RI(MXT-1701): 1227], 10 [RI(MXT-5): 1235; RI(MXT-1701): 1393], 11 [RI(MXT-5): 1375; RI(MXT-1701): 1426], and 16 [RI(MXT-5): 1416; RI(MXT-1701): 1623] were detected; both were detected. The chromatographic peaks include peak 12 [RI (MXT-5): 1438; RI (MXT-1701): 1462], peak 13 [RI (MXT-5): 1415; RI (MXT-1701): 1496 ], peak 14 [RI (MXT-5): 1470; RI (MXT-1701): 1552], and peak 17 [RI (MXT-5): 1538; RI (MXT-1701): 1703].

[0043] Table 2. Kovats Retention Index of Costus Root and Sichuan Costus Root 2.6 Rapid Identification Study of Costus Root and Sichuan Costus Root 2.6.1 Principal Component Analysis (PCA) Multivariate statistical analysis is a method for studying datasets by simultaneously considering multiple variables. It uses statistical and mathematical methods to extract the simplest yet most telling information from complex data. The Heracles II ultra-fast gas phase electronic nose data analysis software includes principal component analysis (PCA), discriminant factor analysis (DFA), and soft independent modeling of class analysis (SIMCA), which are well-suited for analyzing odor differences. Among these, PCA is the most commonly used multivariate statistical method. It can transform and reduce the dimensionality of collected complex multidimensional data, and then perform linear classification on the reduced eigenvectors to capture the maximum differences between the entire dataset, thereby creating visually appealing 2D / 3D graphs. PCA requires no prior knowledge and can quickly browse all data to find a suitable model to maximize the representation of differences between data points.

[0044] Chromatographic information from 20 batches of *Saussurea costus* (n=3) and 16 batches of *Saussurea oleifera* (n=3) acquired using the Heracles II ultrafast gas chromatography electronic nose was divided into two groups: one for *Saussurea costus* and the other for *Saussurea oleifera*. Chromatographic peaks with a variable contribution factor greater than 0.75 were selected as sensors, and a PCA model was established. The results are as follows: Figure 7 As shown, the horizontal and vertical axes represent the contribution rates of principal component 1 (PC1) and principal component 2 (PC2) obtained through PCA transformation, respectively. The cumulative contribution rate of PC1 and PC2 is 98.2%, which can well reflect the odor information in costus root and Sichuan costus root. The higher the PCA recognition index, the greater the difference between groups. The recognition index of this model is 71, which can quickly and effectively distinguish costus root and Sichuan costus root.

[0045] 2.6.2 Discriminant factor analysis (DFA) Discrimination Analysis (DFA) is a classification technique that optimizes discriminative power by recombining sensor data. It can be used to distinguish different categories, grades, origins, intensity levels, etc. It requires optimizing the discrimination using sample data from two or more known categories. Based on principal component analysis, it shortens the distance between samples within a group and increases the difference between groups, aiming to maximize the distance between groups while minimizing the differences within groups, for qualitative discrimination.

[0046] Based on the PCA in "2.6.1", DFA was performed on the above-mentioned Costus root and Sichuan costus root samples, and the results are as follows. Figure 8As shown, the horizontal and vertical axes represent the discrimination index 1 (DF1) and discrimination index 2 (DF2) obtained by DFA transformation, respectively. The cumulative discrimination index of DF1 and DF2 is 100%, indicating that the DFA method can effectively distinguish between Costus Root and Sichuan Costus Root. Moreover, the trend and results obtained are the same as those of PCA, which further verifies the PCA results.

[0047] 2.6.3 Soft Independent Modeling of Class Analysis (SIMCA) SIMCA is a qualitative analysis method. The acceptable region in the SIMCA chart is obtained by calculating the Mahalanobis distance and standard deviation of the centroids from the reference sample to the good sample. It is used for simple pass / fail judgment, and can give the analysis results of good or bad or qualified or unqualified. The recognition percentage is calculated by verifying the cross-validity of each sample.

[0048] The above-mentioned samples of Costus root and Sichuan costus root were grouped according to "2.6.1". A SIMCA model was established for Sichuan costus root using Costus root as a reference. The results are as follows: Figure 9 As shown. By Figure 9 As can be seen, the Sichuan costus root is outside the reference area, indicating that SIMCA is an effective model that can accurately distinguish between costus root and Sichuan costus root, and can be used to quickly identify the two.

[0049] 3. Discussion Odor fingerprinting was performed on 20 batches of *Saussurea costus* and 16 batches of *Saussurea involucrata* from different sources using the Heracles II ultrafast gas chromatography electronic nose. The odor fingerprints of the 20 *Saussurea costus* batches yielded 13 common peaks, while those of the 16 *Saussurea involucrata* batches yielded 9 common peaks. Qualitative analysis of these characteristic peaks was conducted using the AroChemBase database, providing a relatively comprehensive description of the true odor and odor component composition of *Saussurea costus* and *Saussurea involucrata*. Comparison of odor fingerprint information revealed significant differences between *Saussurea costus* and *Saussurea involucrata* from different sources, allowing for effective differentiation through odor fingerprinting.

[0050] Furthermore, based on the collected odor fingerprint information, corresponding chemometric models were established using the Heracles II ultra-fast gas chromatography electronic nose data analysis software. These models included principal component analysis, discriminant factor analysis, and soft independent modeling analysis to analyze the data. The results showed that the various models mutually validated each other, all demonstrating the ability to rapidly identify *Saussurea costus* and *Saussurea involucrata*. In conclusion, the ultra-fast gas chromatography electronic nose can provide a comprehensive characteristic evaluation of the quality of *Saussurea costus* and *Saussurea involucrata*, and can also serve as an effective method for rapidly and accurately distinguishing between the two. It can also provide a reference for applying electronic nose technology to the identification of other easily confused Chinese medicinal materials, broadening the application ideas for non-destructive rapid detection.

Claims

1. A rapid detection method for *Saussurea costus* from Yunnan and Sichuan provinces, and *Saussurea costus* from Sichuan province, characterized in that... It includes the following steps: 1) Preparation of the test sample: Take the sample to be tested, crush it, and pass it through a No. 2 sieve to obtain the test sample; The sample to be tested is *Saussurea costus* from Yunnan and Sichuan and / or *Saussurea costus* from Sichuan; the *Saussurea costus* is *Saussurea costus*, a plant belonging to the Asteraceae family. Aucklandia lappa The dried root of Decne.; the aforementioned costus root is the costus root of the Asteraceae plant Costus root. Vladimiria souliei (Franch.) Ling or Grey-haired Sichuan Oak Vladimiria souliei (Franch.)Ling var. cinerea dried roots of Ling; 2) Seal the n-alkanes (nC6–nC16) mixed reference standard and the test sample in headspace vials and use a Heracles II ultra-fast gas chromatograph electronic nose for detection. The detection parameters were as follows: Column: MXT-5 and MXT-1701 columns; initial column temperature: 80℃; temperature program: increase to 160℃ at 1℃ / s, increase to 250℃ at 2℃ / s, hold for 21s, acquisition time: 148s; sample volume: 2.0g; incubation temperature: 40℃; incubation time: 20min; injection volume: 5000μL; injection rate: 125μL·s. -1 The injection duration was 45 seconds; the injection port pressure was 10 kPa; the injection port temperature was 200℃; the initial temperature of the trap was 50℃; and the trap split rate was 10 mL / min. -1 The final temperature of the trap is 250℃; Using the Heracles II ultra-fast gas chromatograph electronic nose, the odor fingerprint chromatogram of the test sample acquired with an MXT-5 column was converted to Kovats retention index, showing chromatographic peaks with retention indices of 947, 994, 1017, 1028, 986, 1035, 1069, 1195, 1438, 1415, 1470, 1490, and 1538. The odor fingerprint chromatogram of the test sample acquired with an MXT-1701 column was converted to Kovats retention index, showing chromatographic peaks with retention indices of 961, 1018, 1044, 1059, 1068, 1086, 1100, 1288, 1462, 1496, 1552, 1574, and 1703. Therefore, the test sample was identified as costus root. Using the Heracles II ultra-fast gas chromatograph electronic nose, the odor fingerprint chromatogram of the test sample acquired with an MXT-5 column was converted to Kovats retention index, showing chromatographic peaks with retention indices of 1150, 1235, 1375, 1438, 1415, 1470, 1490, 1416, and 1538. The odor fingerprint chromatogram of the test sample acquired with an MXT-1701 column was converted to Kovats retention index, showing chromatographic peaks with retention indices of 1227, 1393, 1426, 1462, 1496, 1552, 1574, 1623, and 1703. The test sample was identified as *Saussurea costus*.

Images