Abstract
This study aimed to establish a comprehensive evaluation system for assessing the quality and classification of licorice from six different origins in China using multiple indicators, including combustion heat, combustion stability, fat, crude fiber, ash, trace element, glycyrrhizic acid (GLA), liquiritin (LI), and amino acid contents. Six types of licorice from Dongsheng District in Ordos, Inner Mongolia, Huimin District in Hohhot, Inner Mongolia, Guyang County in Baotou, Inner Mongolia, Linhe District in Bayanzhuoer City, Xinjiang, Kashgar Prefecture in Xinjiang, and Yining County in Yili, Xinjiang, were selected as research objects to determine the combustion heat, combustion stability, fat, crude fiber, ash, trace element, glycyrrhizic acid (GLA), liquiritin (LI), and amino acid contents. The heat of combustion is a crucial metric for assessing the energy content of substances, thereby reflecting the quality of medicinal materials. The thermal stability of traditional Chinese herbal medicines is associated primarily with their ability to preserve their chemical structure and biological activity when subjected to heating processes, such as boiling, brewing, heating, or storage. Using gray pattern recognition (GPR), entropy factor analysis (EFA), and systematic clustering analysis (SCA), comprehensive evaluation systems were constructed for six licorice samples based on nutritional indicators, including multiple indicators of the heat of combustion, combustibility (licorice combustion stability), fat content, amino acid content, ash content, crude fiber content, glycyrrhizic acid (GLA), liquiritin (LI), and trace element content. Using entropy factor analysis based on 39 variables, the quality ranking of the six licorice samples was as follows: licorice from Dongsheng District, Ordos, Inner Mongolia > licorice from Yining County, Yili, Xinjiang> licorice from Guyang County, Baotou, Inner Mongolia> licorice from Hohhot Huimin District, Inner Mongolia> licorice from Kashgar Prefecture, Xinjiang > licorice from Linhe District, Bayanzhuoer City, Inner Mongolia. The results of the systematic clustering analysis (SCA) indicated that the licorice samples could be classified into three distinct categories and that the 39 variables could be grouped into four groups. This study has significant theoretical and practical implications for the determination and comprehensive evaluation of multiple quality indicators of licorice.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12870-026-08467-0.
Keywords: Licorice, thermogravimetric analysis, entropy factor analysis (EFA), amino acids, trace elements
Introduction
Licorice (Glycyrrhiza uralensis Fisch), a traditional Chinese medicinal herb, has been used in Chinese medicine [1] for thousands of years. It has a wide range of pharmacological effects, including anti-inflammatory, antiviral, antioxidant, and hepatoprotective properties [2, 3]. In recent years, licorice has been increasingly studied, and its applications in food and cosmetics have attracted attention. However, the quality and efficacy of licorice are closely related to its chemical composition [4], and there are significant differences in the components of licorice samples from different sources, which directly affects their effectiveness. Therefore, establishing a scientific and objective comprehensive evaluation system to assess the quality of licorice samples is essential.
Nutritional indicators are core components [5, 6]. The combustion heat is an important index for measuring the energy content of a substance [7] and reflects the total amount of combustible materials in licorice. The combustibility, particularly the combustion stability, of licorice is directly related to its safety and effectiveness during processing and use [8].Fat content is an important parameter for evaluating the nutritional value of licorice, as it affects its pharmacological activity and stability [9, 10]. Trace elements and amino acids present in Chinese herbal medicines constitute two fundamental chemical components that establish a connection between their material basis and pharmacological effects, indicating the quality of the medicinal materials [11]. The crude fiber content represents the quantity of indigestible plant material [12], which affects the digestibility and potential therapeutic effectiveness of the herb. The ash content denotes the total mineral content [13, 14], offering insights into the presence of essential inorganic compounds and the potential for contamination.
Owing to differences in their planting environments, cultivation methods, and harvesting processes, licorice from different regions exhibit significant quality variations. These differences impact not only the medicinal efficacy of licorice but also its market value. Therefore, studying the quality differences of licorice from various regions and systematically evaluating them through multi-index quantitative analysis methods is essential for the standardized cultivation and quality control of licorice [15, 16].Traditionally, research on licorice quality has focused on the analysis of single indicators, such as active components (e.g., glycyrrhizic acid and liquiritin) [17].
In summary, no detailed reports have focused on constructing a quality evaluation system for licorice based on the heat of combustion, combustion stability, fat, crude fiber, ash, trace element, glycyrrhizic acid (GLA), liquiritin (LI), and amino acid contents of licorice. To address this gap, we can refer to methods used for evaluating other plants and adapt them to the unique characteristics of licorice for a comprehensive assessment [18–20].This study builds on indicators such as combustion heat, combustion stability, fat content, crude fiber content, ash content, trace element content, glycyrrhizic acid (GLA), liquiritin (LI), and amino acid content to explore the quality differences of licorice from different production areas and establish an evaluation system. This approach precisely addresses the shortcomings of previous studies on licorice quality. Currently, there is extensive research on licorice, both domestically and internationally.
The study selected six types of licorice from various regions as research objects, including Dongsheng District of Ordos in Inner Mongolia, Huimin District of Hohhot in Inner Mongolia, Guyang County in Baotou in Inner Mongolia, Linhe District of Bayanzhuoer City in Xinjiang, Kashgar District in Xinjiang, and Yining County, Yili in Xinjiang. This study assessed parameters such as the heat of combustion, combustion stability, fat content, crude fiber, ash, trace elements, glycyrrhizic acid (GLA), liquiritin (LI), and amino acids in licorice from six different origins [21]. This study established a multi-index analysis and evaluation system for licorice from different sources, providing a reference for quality control and the development and utilization of licorice byproducts.
Materials and methods
Materials and instruments
Test material
Six types of licorice (cultivated), the dried roots of the leguminous plant Glycyrrhiza uralensis Fisch, Licorice samples from Dongsheng District of Ordos Inner Mongolia (Figure S4), Huimin District of Hohhot of Inner Mongolia (Figure S5), Guyang County of Baotou of Inner Mongolia (Figure S6), Linhe District of Bayanzhuoer City (Figure S7), Xinjiang Kashgar Prefecture (Figure S8), and Yili Yining County of Xinjiang (Figure S9) were used as analysis samples and were purchased from Laibin Chinese herbal medicine market in Guangxi, China. The corresponding Figures S4–S9 are available in the Supplementary Materials. Licorice samples of six different origins were dried, ground using a mortar, and screened through 40 mesh pharmacopoeias. Identification was conducted by Prof. Caiyun Jiang of Guangxi Science & Technology Normal University, and the samples were stored at Guangxi Science & Technology Normal University.
Licorice originated from Dongsheng District, Ordos, Inner Mongolia, production date: January 2021; licorice originated from Hohhot Huimin District, Inner Mongolia, production date: January 2021; licorice originated from Guyang County, Baotou, Inner Mongolia, production date: November 2020; licorice originated from Linhe District, Bayanzhuoer City, Xinjiang, production date: December 2020; licorice originated from Kashgar Prefecture, Xinjiang, production date: November 2020; licorice originated from Yining County, Yili, Xinjiang, production date: October 2020.
Instrument
The combustion heat determination included an HR-15 oxygen bomb calorimeter, ignition wire (nickel-chromium wire), tablet press (Hunan Changsha Changxing Higher Education Instrument and Equipment Development Co., Ltd.), pulverizer (FW135 type, Tianjin Tester Instrument Co., Ltd.), crucible, STA2500 synchronous thermal analyzer (NETZSCH, Germany), F1600 automatic fiber tester (Jinan Alva Instrument Co., Ltd.), GZX-GF101-3-S drying oven (Shanghai Huyueming Scientific Instrument Co., Ltd.), SE206 fat analyzer (Jinan Alva Instrument Co., Ltd.), SX-4-10P muffle furnace (Tianjin Teste Instrument Co., Ltd.), FA2004 electronic analytical balance (Shanghai Sunny Hengping Scientific Instrument Co., Ltd.), ICP‒OES spectrometer (iCAP 7000 SERIES, Thermo Scientific, USA), 300 amino acid analyzer (German Mann Bohr), LC-2050 high-performance liquid chromatograph (Shimadzu Corporation, Japan), circulating water vacuum pump (Shanghai Lingde Instrument Co., Ltd.), 2 ml sample vial, and disposable needle filter (mixed fiber‒aqueous system) (Sinopharm Chemical Reagent Co., Ltd.).
Reagent
Benzoic acid (analytical pure, Tianjin Chemical Reagent Co., Ltd.); petroleum ether (boiling range of 30 ~ 60 ℃, Chengdu Jinshan Chemical Reagent Co., Ltd.); 2,006,003 pharmaceutical capsules (Guangdong Biotech Co., Ltd.); potassium hydroxide (Tianjin Damao Technology Co., Ltd.), hydrochloric acid (Xilong Science Co., Ltd.), sulfuric acid (Xilong Science Co., Ltd.), multielement mixed liquid standard samples (GNM-M218741-2013), 30% hydrogen peroxide (Xilong Science Co., Ltd.), nitric acid (Chengdu Jinshan Chemical Reagent Co., Ltd.), amino acid mixed standards 650 − 0037 Feedstuff (German Mannmabol Co., Ltd.), ninhydrin hydrate (Shanghai Aladdin Biochemical Technology Co., Ltd.), methanol (Guangzhou Tongyuan Chemical Technology Co., Ltd.), phosphoric acid (Tianjin Damao Chemical Reagent Factory), liquiritin (Shanghai Yuanye Biotechnology Co., Ltd.), glycyrrhizic acid (Shanghai Yuanye Biotechnology Co., Ltd.), and acetonitrile (Guangzhou Tongyuan Chemical Technology Co., Ltd.) were used.
Related experimental methods
The ash content, trace element content, heat of combustion, fat content, thermogravimetric parameters [22], and crude fiber content of the six types of licorice were determined according to previously described methods [23].
Thermogravimetric analysis conditions: The experimental procedure involved increasing the test temperature from 25 ℃ to 600 ℃ at a rate of 20 K/min under a high-purity nitrogen gas atmosphere. The specific conditions for the analysis were as follows: a heating rate of 2.5 ℃/min and the use of high-purity nitrogen gas as the atmosphere. An empty crucible was used as the reference material.
Methods for determining amino acids [24, 25] involve the use of wavelengths of 570 nm and 440 nm. The procedure requires the addition of ascorbic acid, which serves as a coreagent activator, to the ninhydrin solution, which acts as a reagent for the derivator. This mixture was thoroughly shaken and stored in the dark for a minimum of four hours. The mixture was subsequently transferred to a reagent bottle, resulting in a color change from yellow to purple. The recommended ratio is 40 mg of activator to 100 ml of ninhydrin. A 0.4 g sample (accurate to four decimal places) was accurately weighed in a dry 50 mL volumetric flask, and a diluted hydrochloric acid solution (1 mL of hydrochloric acid was measured, and ultrapure water was added to a 1000 mL volumetric flask) was used for 0.5 h, after which the diluted hydrochloric acid was used to dissolve the sample in a 50 mL volumetric flask for backup.
At wavelengths of 570 and 440 nm, the 20 amino acids were optimally separated under the given conditions [26, 27]. Amino acid profiles were automatically generated, and the amino acid content was calculated by processing peak areas. The peak spectrum of the amino acid content of licorice in Dongsheng District, Ordos, Inner Mongolia is shown in Fig. 1. A comprehensive evaluation system was developed to analyze multiple indicators of licorice from six different origins, including combustion heat, combustion stability, fat content, ash content, crude fiber, trace elements, and amino acids.
Fig. 1.
Peak spectrum of the amino acid content of licorice in Dongsheng District, Ordos, Inner Mongolia
Determination of glycyrrhizic acid (GLA) and liquiritin (LI) contents [28, 29]: A 2 g sample was accurately weighed onto filter paper and dried at 60 ℃ for 12 h. The dried sample (0.1 g) was accurately weighed into a 25 mL volumetric flask, and a 67% methanol aqueous solution was added. The mixture was soaked at room temperature for 24 h, sonicated for 30 min, allowed to return to room temperature, diluted to the mark with a 67% methanol aqueous solution, filtered through a 0.22 μm membrane, and analyzed according to the HPLC operating procedures. The GLA and LI HPLC chromatograms of licorice in Dongsheng District, Ordos, Inner Mongolia, are shown in Figures S10–S11 (Supplementary Materials).
The test was repeated three times for each sample (n = 3, RSD%<2%).
Results and discussion
Licorice thermogravimetric analysis
Licorice in Dongsheng District, Ordos, Inner Mongolia.
Figure 2a and Table S1(a) show that licorice lost weight at 36.3 ℃, with a loss of 3.34%. When the temperature reached 146.5 ℃, licorice entered the second weight-loss stage, and when the temperature rose to 394.3 ℃, the loss rate of licorice reached 42.29%. The licorice decomposed further, resulting in weight loss and a residual mass of 42.17%. The DTG curve [30] of Dongsheng District, Ordos, Inner Mongolia, licorice exhibited two peak morphologies with inflection points at 119.5 °C and 310.2 °C. The DTA curve showed two exothermic peaks in the temperature range of 41.0–151.2 ℃, with peak areas of 154.6 J/g and 14.56 J/g and peak temperatures of 114.7 ℃ and 375.4 ℃, respectively.
Fig. 2.
a Thermal analysis of licorice in Dongsheng District, Ordos, Inner Mongolia b Thermal analysis of licorice in the Hohhot Huimin area, Inner Mongolia c Thermal analysis of licorice in Guyang County, Baotou, Inner Mongolia d Thermal analysis of licorice in Linhe District, Bayanzhuoer city, Xinjiang e Thermal analysis of licorice in Kashgar Prefecture, Xinjiang f Thermal analysis of licorice in Yining County, Yili, Xinjiang
-
(2)
Licorice in Hohhot Huimin District, Inner Mongolia.
Figure 2b and Table S1(b) show that licorice began losing weight at 40.7 ℃, and the loss rate was 4.10% at this temperature. At 182.3 ℃, licorice entered the second weight loss stage, and the loss rate of licorice was 52.23%. When the temperature reached 429.7 ℃, it entered the third weight loss stage. At 599.1 ℃, the loss rate of licorice was 7.28%. Ultimately, licorice continued to decompose, resulting in a remaining mass as low as 34.91%. The DTG curve of licorice from the Hohhot Huimin area in Inner Mongolia showed a single peak with an inflection point at 312.7 °C. The DTA curve exhibited a significant exothermic peak at 117.1 ℃, with a peak area of 129.5 J/g and a temperature range of 44.4–155.0 ℃.
-
(3)
Licorice in Guyang County, Baotou, Inner Mongolia.
As shown in Fig. 2c and Table S1 (c), licorice began to lose weight at 41.3 ℃, and the loss rate of licorice was 3.02% at this temperature. When the temperature reached 147.6 ℃, licorice entered the second weightlessness stage, and when the temperature rose to 407.4 ℃, the loss rate of licorice reached 50.49%. Licorice was further decomposed weightlessly, and its residual mass was 35.01%.
The DTG curve of licorice in Guyang County, Baotou, Inner Mongolia, showed two peaks, with inflection points at 108.5 and 308.7 °C. The DTA curve showed a significant exothermic peak at 110.6 ℃, with a peak area of 91.57 J/g and a temperature range of 40.8–146.7 ℃.
-
(4)
Licorice in Linhe District, Bayanzhuoer city, Xinjiang.
As illustrated in Fig. 2d and Table S1(d), the weight loss in licorice commenced at 49.3 ℃, with a loss rate of 6.10%. Upon reaching 153.8 ℃, the second stage of weight loss commenced, and upon reaching 397.6 ℃, the loss rate of licorice reached 50.31%. As the temperature continued to increase, the licorice became increasingly weightless and decomposed, resulting in a final mass of 27.28%.
The DTG curve exhibited two distinct peak shapes, with inflection points at 85.6 °C and 305.0 °C. The DTA curve of licorice in Linhe District, Bayanzhuoer City, Xinjiang, exhibited a pronounced exothermic peak, with a peak value of 113.5 ℃, a temperature range of 34.7–166.5 ℃, and a peak area of 260.7 J/g.
-
(5)
Licorice in Kashgar Prefecture, Xinjiang.
As shown in Fig. 2e and Table S1(e), licorice in the Kashgar region of Xinjiang reached the weight-loss stage at 45.4 ℃, corresponding to a loss rate of 3.22%. At 150.7 ℃, the licorice entered the second weight-loss stage. Upon further increasing the temperature to 397.6 ℃, the rate of licorice loss was 49.57%. The licorice was then subjected to further weight loss, resulting in a residual mass of 34.18%.
The DTG curve of licorice in Kashgar Prefecture, Xinjiang, exhibited two peaks, with inflection points at 108.9 and 309.6 °C. The DTA curve demonstrated a pronounced exothermic peak at 117.9 ℃, with a peak area of 167.5 J/g and a temperature range of 36.4–158.2 ℃.
-
(6)
Licorice in Yining County, Yili, Xinjiang.
As shown in Fig. 2f and Table S1(f), weight loss in licorice commenced at 44.8 ℃ in Yili Yining County, Xinjiang, and the loss rate of licorice was 8.61% at this temperature. Upon reaching 162.6 ℃, licorice commenced the second weight-loss stage, with a loss rate of 8.61% at this temperature. At 390.6 ℃, the loss rate reached 50.16%. As the temperature continued to increase, licorice entered the third weight-loss stage, with a loss rate of 19.25% at 598.7 °C. The final quality of licorice was 21.50%.
The DTG curve of licorice in Yining County, Yili, Xinjiang, exhibited two peaks, with inflection points at 95.5 and 312.7 °C. The DTA curve exhibited a pronounced exothermic peak at 108.9 ℃, spanning a temperature range of 34.4–177.7 ℃, and peaked area of 287.70 J/g. Additionally, an endothermic peak was observed at 295.9 °C, with a peak area of 96.00 J/g and a temperature range of 251.5–367.4 °C.
When the temperature reached a maximum of 182.3 ℃, the first stage of water loss ended, with the water loss rate for licorice ranging from 3.2% to 8.7%. Within the temperature range of 305.0 to 312.7 ℃, licorice began to decompose proteins, crude fats, carbohydrates, and other substances. In the third stage, the crude fiber components of licorice decompose. The higher the heating rate is, the more pronounced the temperature lag phenomenon, yielding higher measured values for both the initial and final temperatures of weight loss [31, 32] and a broader decomposition temperature range [33, 34].
The combustion stability of licorice from six different origins was analyzed using combustion parameter data. The licorice samples were sourced from the following locations: Dongsheng District of Ordos, Inner Mongolia; Huhhot Huimin District, Inner Mongolia; Guyang County, Baotou, Inner Mongolia; Linhe District, Bayanzhuoer City, Inner Mongolia; Kashgar Prefecture, Xinjiang; and Yining County, Yili, Xinjiang. Thermogravimetry was employed to study the combustion characteristics of licorice at various heating rates using a thermogravimetric analyzer, which helped in determining the combustion stability. This study utilized the gray pattern recognition [35] method, and EXCEL was used to calculate the F values of the six licorice samples, which were 0.7127, 0.6933, 0.6621, 0.7225, 0.6777, and 0.8770. The combustion stability ranking of the six licorice samples was as follows: licorice from Yining County, Xinjiang > licorice from Linhe District, Bayanzhuoer City, Inner Mongolia > licorice from Dongsheng District, Ordos, Inner Mongolia > licorice from Hohhot Huimin District, Inner Mongolia > licorice from Kashgar Prefecture, Xinjiang > licorice from Guyang County, Baotou, Inner Mongolia. The ranking of combustion stability serves as a guide for processing and storage decisions, identifying licorice from Yining County, Xinjiang as possessing superior thermal stability. Variations in combustion stability can be attributed to diverse chemical compositions influenced by the geographic and climatic conditions of each region [36]. These findings offer valuable insights for selecting licorice sources based on their thermal stability for industrial use.
Determination of combustion heat, fat, ash, trace element, amino acid, glycyrrhizic acid (GLA), liquiritin (LI), and crude fiber contents
The optimal wavelengths, correlation coefficients (R), and instrument detection limits for the elements are listed in Table S2 (supplementary materials). The results of the combustion heat, fat, ash, trace element, amino acid, glycyrrhizic acid (GLA), liquiritin (LI), and crude fiber contents of the six types of licorice are shown in Table S3 (supplementary materials). Table S3 shows that As, Cd, Hg, and Sc were not detected in the samples from the six different licorice sources, indicating that these samples do not contain the toxic or harmful elements As, Cd, and Hg. Only licorice in Dongsheng District, Ordos, Inner Mongolia, contained 0.0883 µg/g Co. The Pb content ranged from 1.84 to 6.01 µg/g, with an average value of 4.0567 µg/g. The order of the Pb content from highest to lowest was as follows: liquor in Dongsheng District, Ordos, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Guyang County, Baotou Guyang County, Inner Mongolia> licorice in Yili Yining County, Xinjiang> licorice in Linhe District, Bayanzhuoer City, Xinjiang.
All 6 types of licorice contained seven essential amino acids (Trp was not detected), namely, Thr, Val, Met, Ile, Leu, Phe and Lys. The ratio of the content of essential amino acids to the total content of amino acids in the human body ranged from 6.34% to 26.58%, and the ratio of the content of essential amino acids to the content of non-essential amino acids in the human body ranged from 6.77% to 36.21%.
There are differences in the amino acid content of licorice in different production areas, which may be related to the climate, planting environment, and processing methods of each licorice production area.
Development of a multi-indicator system for evaluating licorice quality
Entropy factor analysis
Entropy factor analysis (EFA) integrates information entropy and factor analysis to reduce multivariate data and extract relevant features. The entropy coefficients of the samples were used to replace the sample correlation coefficient matrix in factor analysis, the entropy correlation coefficients were used to participate in the computation and obtain the eigenvalues and eigenvectors, and the variance contribution ratio and total variance contribution ratio of the entropy factor analysis were computed [37, 38]. Given the extensive number of indicators (39) incorporated in this study, entropy factor analysis was employed to reduce the dimensionality and identify the most informative factors.
After performing entropy factor analysis, the cumulative contribution (Table 1) of the first 5 entropy factors reached 100.00%>85%, so the first 5 entropy factors were selected, which represented 100.00% of the 39 variables, including the stability of combustion, QV, ash, crude fiber, fat, Al, Ba, Cr, Cu, Fe, K, Li, Mg, Mn, Na, Ni, Sr, Zn, CySO3H, Asp, MetSON, Thr, Ser, Glu, Gly, Ala, (Cys)2, Val, Met, Ile, Leu, Tyr, Phe, His, Lys, glycyrrhizic acid (GLA), liquiritin (LI), Arg and Pro, in six types of licorice from Dongsheng District in Ordos, Inner Mongolia, Huimin District in Hohhot, Inner Mongolia, Guyang County in Baotou, Inner Mongolia, Linhe District, Bayanzhuoer City, Xinjiang, Kashgar Prefecture in Xinjiang, and Yining County in Yili, Xinjiang.
Table 3.
Entropy factor scores and composite entropy factor scores for the 39 variables in the 6 licorice samples
| Place of production | F1 | F2 | F3 | F4 | F5 | F | Ranking |
|---|---|---|---|---|---|---|---|
| Dongsheng District, Ordos, Inner Mongolia | 0.0496 | 1.9919 | 0.3065 | 0.2572 | 0.1913 | 0.5827 | 1 |
| Hohhot Huimin District, Inner Mongolia | 0.4952 | -0.3227 | 0.6137 | -0.1956 | -1.8445 | -0.0303 | 4 |
| Guyang County, Baotou, Inner Mongolia | 0.4480 | -0.4521 | 0.7224 | -1.4642 | 1.0468 | 0.0281 | 3 |
| Linhe District, Bayanzhuoer City, Xinjiang | -2.0073 | -0.3668 | -0.0191 | 0.0352 | -0.0331 | -0.7000 | 6 |
| Kashgar Prefecture, Xinjiang | 0.4458 | -0.0875 | -1.9713 | -0.2691 | -0.0406 | -0.3084 | 5 |
| Yining County, Yili, Xinjiang | 0.5687 | -0.7627 | 0.3478 | 1.6365 | 0.6801 | 0.4280 | 2 |
Table 1.
Characteristic roots of the entropy correlation coefficients for the 39 variables in the 6 licorice samples
| Entropy factors | Characteristic root | Contribution rate% | Cumulative contribution rate % |
|---|---|---|---|
| 1 | 11.993 | 30.751 | 30.751 |
| 2 | 8.698 | 22.301 | 53.052 |
| 3 | 7.415 | 19.012 | 72.064 |
| 4 | 6.812 | 17.467 | 89.531 |
| 5 | 4.083 | 10.469 | 100.000 |
| … | … | … | … |
The first entropy factor loading matrix (Table 2) contains information on fat, Fe, MetSON, Gly, Met, Ile, His, and LI. The fat content (%) of six licorices, including licorice in Dongsheng District of Ordos in Inner Mongolia, licorice in Hohhot Huimin District in Inner Mongolia, licorice in Guyang County in Baotou, Linhe District, Bayanzhuoer City, licorice in Kashgar District, Xinjiang, and licorice in Yining County, Xinjiang, was between 0.34 ~ 1.31%, and the average content was 0.7419%. The size order was licorice in Hohhot Huimin District in Inner Mongolia> licorice in Guyang County in Baotou, Inner Mongolia> licorice in Dongsheng District of Ordos in Inner Mongolia> licorice > licorice in Kashgar Prefecture, Xinjiang licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Yining County, Yili, Xinjiang.The Fe content ranged from 95.86 to 228.93 µg/g, with an average of 168.7352 µg/g. The order of Fe content was as follows: licorice in Hohhot Huimin District, Inner Mongolia> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Yining County, Yili, Xinjiang> licorice in Kashgar Prefecture, Xinjiang> licorice in Linhe District, Bayanzhuoer City, Xinjiang.The MetSON content ranged from 174.49 to 2208.14 µg/g, with an average value of 596.3787 µg/g. The MetSON content order was as follows: licorice in Kashgar Prefecture, Xinjiang> licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Hohhot Huimin District in Inner Mongolia> licorice in Yining County, Yili, Xinjiang> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Guyang County, Baotou, Inner Mongolia.The Gly content ranged from 133.39 to 3741.55 µg/g, with an average value of 1767.3455 µg/g. The order of Gly content was as follows: licorice in Kashgar Prefecture, Xinjiang> Licorice in Yining County, Yili, Xinjiang> Licorice in Guyang County, Baotou, Inner Mongolia> Licorice in Hohhot Huimin District, Inner Mongolia> Licorice in Dongsheng District, Ordos, Inner Mongolia> Licorice in Linhe District, Bayanzhuer City, Xinjiang.The Met content ranged from 69.73 to 824.97 µg/g, with an average value of 315.4176 µg/g. The content order was as follows: licorice in Hohhot Huimin District, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang> licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Yining County, Yili, Xinjiang.The Ile content ranged from 102.90 to 339.56 µg/g, with an average value of 189.8277 µg/g. The content order was as follows: Guyang County, Baotou, Inner Mongolia> Yining County, Yili, Xinjiang> Hohhot Huimin District, Inner Mongolia> Kashgar Prefecture, Xinjiang> Dongsheng District, Ordos District, Inner Mongolia> Linhe District, Bayanzhuoer City, Xinjiang.His content was between 101.70 ~ 365.12 µg/g, the average value was 197.7819 µg/g, and the content order was: licorice in Kashgar Prefecture, Xinjiang> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Yining County, Xinjiang> licorice in Linhe District, Bayanzhuoer City, Xinjiang.
Table 2.
Entropy factor loading matrix for the 39 variables in the 6 licorice samples
| Variables | Entropy factor loading matrix 1 | Entropy factor loading matrix 2 | Entropy factor loading matrix 3 | Entropy factor loading matrix 4 | Entropy factor loading matrix 5 |
|---|---|---|---|---|---|
| Crude fiber | 0.066 | -0.248 | 0.199 | 0.914 | 0.245 |
| Ash | 0.105 | -0.298 | 0.517 | 0.725 | 0.326 |
| Combustion stability | 0.148 | 0.395 | 0.006 | 0.906 | 0.044 |
| QV | -0.647 | 0.221 | 0.531 | -0.412 | -0.286 |
| Fat | 0.407 | 0.269 | 0.379 | -0.522 | -0.588 |
| Al | 0.214 | 0.906 | 0.313 | 0.186 | -0.011 |
| Ba | 0.394 | 0.044 | 0.785 | -0.451 | -0.151 |
| Cr | 0.216 | 0.909 | -0.016 | 0.039 | -0.354 |
| Cu | 0.063 | 0.969 | -0.237 | 0.035 | 0.023 |
| Fe | 0.756 | 0.127 | 0.547 | -0.012 | -0.336 |
| K | 0.006 | 0.024 | 0.977 | 0.129 | 0.166 |
| Li | 0.469 | 0.180 | 0.414 | 0.729 | -0.212 |
| Mg | 0.533 | 0.539 | -0.110 | -0.548 | -0.336 |
| Mn | 0.459 | 0.789 | 0.146 | -0.373 | -0.080 |
| Na | -0.944 | 0.049 | -0.021 | -0.257 | -0.201 |
| Ni | 0.040 | 0.957 | -0.284 | 0.011 | 0.044 |
| Sr | -0.290 | -0.556 | -0.402 | 0.665 | 0.048 |
| Zn | 0.071 | 0.591 | -0.514 | -0.570 | 0.238 |
| CySO3H | 0.418 | -0.143 | -0.161 | 0.308 | 0.827 |
| Asp | -0.947 | -0.224 | 0.079 | 0.007 | -0.217 |
| MetSON | 0.031 | -0.092 | -0.987 | -0.121 | -0.042 |
| Thr | -0.886 | -0.331 | 0.059 | 0.285 | 0.142 |
| Ser | -0.297 | -0.446 | 0.376 | -0.156 | 0.739 |
| Glu | -0.973 | -0.205 | -0.057 | 0.084 | 0.018 |
| Gly | 0.618 | -0.417 | -0.512 | -0.058 | 0.423 |
| Ala | 0.566 | -0.217 | -0.268 | 0.010 | 0.749 |
| (Cys)2 | 0.401 | 0.730 | 0.341 | -0.252 | 0.355 |
| Val | -0.059 | 0.960 | 0.272 | -0.009 | 0.018 |
| Met | 0.009 | -0.212 | -0.201 | -0.175 | -0.940 |
| Ile | 0.548 | -0.467 | 0.499 | -0.308 | 0.371 |
| Leu | 0.415 | 0.286 | 0.467 | 0.138 | 0.713 |
| Tyr | 0.386 | 0.113 | 0.473 | 0.313 | 0.718 |
| Phe | -0.964 | -0.246 | 0.097 | 0.030 | -0.024 |
| His | 0.349 | 0.112 | -0.921 | -0.127 | -0.030 |
| Lys | 0.368 | -0.036 | 0.447 | -0.050 | -0.813 |
| Arg | 0.021 | 0.761 | 0.008 | 0.602 | 0.241 |
| Pro | 0.303 | -0.326 | 0.374 | -0.813 | -0.028 |
| GLA | -0.575 | -0.380 | -0.138 | 0.645 | 0.300 |
| LI | 0.423 | -0.252 | -0.527 | -0.691 | -0.040 |
The second entropy factor loading matrix contains information on Al, Cr, Cu, Mg, Mn, Na, Ni, Zn, (Cys)2, Val, and Arg. The Al content ranged from 12.46 to 309.19 µg/g, with an average value of 86.5052 µg/g. The factor loadings represent the extent to which each variable contributes to the extracted underlying entropy factors. This matrix is crucial for interpreting the multivariate relationships among the measured elements and amino acids. The order of Al content was as follows: Dongsheng District, Ordos, Inner Mongolia> Hohhot Huimin District, Inner Mongolia> Yining County, Yili, Xinjiang> Guyang County, Baotou, Inner Mongolia> Kashgar Prefecture, Xinjiang> Linhe District, Bayanzhuoer City, Xinjiang. The Cr content ranged between 0.39 and 1.24 µg/g, with an average value of 0.3709 µg/g. The order of Cr content was as follows: Dongsheng District, Ordos, Inner Mongolia> Hohhot Huimin District, Inner Mongolia> Kashgar Prefecture, Xinjiang > Yili Yining County, Xinjiang > Guyang County, Baotou, Inner Mongolia > Linhe District, Bayanzhuoer City, Xinjiang. The Cu content ranged from 17.56 to 155.48 µg/g, with an average of 55.5278 µg/g. The order of Cu content was as follows: licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Yining County, Xinjiang. The Mg content ranged from 1352.10 to 1759.72 µg/g, with an average value of 1049.9198 µg/g. The order of Mg content was as follows: licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang> licorice in Guyang County, Baotou Guyang County, Inner Mongolia, Licorice in Linhe District, Bayanzhuoer City, Xinjiang, and licorice in Yining County, Xinjiang, which was not detected. The Mn content ranged from 5.98 to 17.14 µg/g, with an average value of 10.9459 µg/g. The order of Mn content from highest to lowest was as follows: licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang> licorice in Yining County, Xinjiang> licorice in Linhe District, Bayanzhuoer City, Xinjiang. The content of Na was between 552.42 ~ 2000.98 µg/g, the average value was 756.2076 µg/g, and the content order was as follows: licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Hohhot Huimin District, Inner Mongolia > > licorice in Guyang County, Baotou, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang, and no Na was detected in licorice in Yining County, Xinjiang. The Asp content ranged from 34.90 to 1860.99 µg/g, with an average value of 430.7022 µg/g. The Ni content ranged from 31.79 to 344.88 µg/g, with an average of 123.4596 µg/g. The order of Ni content was as follows: licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang> licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Yining County, Yili, Xinjiang. The Zn content ranged from 254.78 to 943.18 µg/g, with an average of 629.5785 µg/g. The order of Zn content was as follows: licorice in Kashgar Prefecture, Xinjiang> licorice in Dongsheng District, Ordos District, Inner Mongolia> licorice in Hohhot Huimin District in Inner Mongolia> licorice in Hohhot Huimin District in Inner Mongolia> licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Yining County, Xinjiang. The (Cys)2 content ranged from 730.78 to 7059.65 µg/g, with an average value of 3324.7865 µg. The content order was as follows: Dongsheng District, Ordos, Inner Mongolia> Guyang County, Baotou, Inner Mongolia> Yining County, Yili, Xinjiang> Hohhot Huimin District, Inner Mongolia> Kashgar Prefecture in Xinjiang> Linhe District, Bayanzhuoer City, Xinjiang. The Val content ranged from 151.45 to 1612.14 µg/g, with an average value of 520.4444 µg/g. The content order was as follows: licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Kashgar Prefecture, Xinjiang> licorice in Yining County, Yili, Xinjiang. The Arg content ranged from 453.62 to 15749.05 µg/g, with an average value of 5927.4617 µg/g. The order of Arg content was as follows: Dongsheng District, Ordos, Inner Mongolia> Yining County, Yili, Xinjiang> Kashgar Prefecture, Xinjiang> Linhe District, Bayanzhuoer City, Xinjiang> Hohhot Huimin District, Inner Mongolia> Guyang County, Baotou, Inner Mongolia. These variables represent the key biochemical and elemental components relevant to the study.
The third entropy factor loading matrix contained information on ash, Ba, K, Asp, Phe, Lys, and Pro. The ash content (%) was between 0.07% and 6.21%, the average content was 3.1805%, and the order of size was as follows: licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Yining County, Xinjiang> licorice in Kashgar Prefecture, Xinjiang.The Ba content ranged from 3.17 to 7.18 µg/g, with an average value of 5.1929 µg/g. The Ba content order was as follows: Guyang County, Baotou, Inner Mongolia> Hohhot Huimin District, Inner Mongolia> Dongsheng District, Ordos, Inner Mongolia> Yining County, Yili, Xinjiang> Linhe District, Bayanzhuoer City, Xinjiang> Kashgar Prefecture, Xinjiang.The K content ranged from 3638.32 to 5262.26 µg/g, with an average value of 4843.8996 µg/g. The order of K content was as follows: Guyang County, Baotou, Inner Mongolia> Yili Yining County, Xinjiang> Dongsheng District, Ordos, Inner Mongolia> Hohhot Huimin District, Inner Mongolia> Linhe District, Bayanzhuoer City, Xinjiang> Kashgar Prefecture, Xinjiang.The order of Asp content was as follows: licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Yining County, Yili, Xinjiang> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Dongsheng District, Ordos District, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang.The Phe content ranged from 61.98 to 613.28 µg/g, with an average value of 188.8930 µg/g. The order of Phe content was as follows: licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Yining County, Yili, Xinjiang> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang.The Lys content ranged from 60.25 to 1476.40 µg/g, with an average value of 434.2088 µg/g. The content order was as follows: licorice in Hohhot Huimin District, Inner Mongolia> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Yining County, Yili, Xinjiang> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang> Licorice in Linhe District, Bayanzhuoer City, Xinjiang.The Pro content ranged from 243.27 to 9577.09 µg/g, with an average value of 3042.7781 µg/g. The Pro content order was as follows: licorice in Guyang County, Baotou, Inner Mongolia> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Kashgar Prefecture in Xinjiang> licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Dongsheng District, Ordos, Inner Mongolia. Pro was not detected in licorice from Yining County, Yili, Xinjiang.
The fourth entropy factor loading matrix contains information on the stability of combustion, QV, crude fiber, Li, Sr, Thr, Glu, and GLA. These variables contribute to characterizing the underlying factors that influence the combustion process. The loadings indicate the strength and direction of each variable’s association with the entropy factor. This matrix serves as the foundation for interpreting the stability and chemical composition relationships within the system. The order of the combustion heat of licorice in the six production areas was as follows: licorice in Yining County, Yili, Xinjiang> licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang. The combustion heat of the six licorice samples ranged from 850.69 to 38072.94 J/g, of which the combustion heat of licorice in Yining County, Xinjiang, was 38072.9327 J/g, with the highest energy, and the combustion heat of licorice in Kashgar Prefecture, Xinjiang, was 850.6997 J/g, with the lowest energy. The crude fiber content (%) ranged between 17.04% and 42.04%, with an average of 29.96%. The order of size was as follows: licorice in Yining County, Yili, Xinjiang> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Guyang County, Baotou, Inner Mongolia. The Li content ranged from 0.24 to 0.53 µg/g, with an average value of 0.3623 µg/g. The order of Li content was as follows: licorice in Yili Yining County, Xinjiang> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Dongsheng District, Ordos Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang> licorice in Guyang County, Baotou, Inner Mongolia> Licorice in Linhe District, Bayanzhuoer City, Xinjiang. The Sr content ranged from 126.61 to 271.91 µg/g, with an average value of 194.6393 µg/g. The order of Sr content was as follows: licorice in Yining County, Xinjiang> licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Kashgar Prefecture, Xinjiang> licorice in Hohhot Huimin District, Inner Mongolia> licorice in Dongsheng District, Ordos, Inner Mongolia> Licorice > in Guyang County, Baotou, Inner Mongolia. The Thr content ranged from 70.22 to 3875.44 µg/g, with an average value of 997.8519 µg/g. The order of Thr content was as follows: licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Yining County, Yili, Xinjiang> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Hohhot Huimin District, Inner Mongolia.The Glu content ranged from 107.77 to 5652.71 µg/g, with an average value of 1203.9734 µg/g. The content order was as follows: licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Yining County, Yili, Xinjiang> licorice in Kashgar Prefecture, Xinjiang> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Hohhot Huimin District, Inner Mongolia.
The fifth entropy factor loading matrix contains information on CySO3H, Ser, Ala, Leu, and Tyr. The content of CySO3H was between 335.12 ~ 389.06 µg/g, and the average value was 362.8142 µg/g, and the content order was: licorice in Yining County, Xinjiang> licorice in Guyang County, Baotou, Inner Mongolia> licorice in Kashgar Prefecture, Xinjiang> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Hohhot Huimin District, Inner Mongolia.The Ser content ranged from 62.13 to 704.80 µg/g, with an average value of 352.3097 µg/g. The order of licorice content was as follows: Guyang County, Baotou, Inner Mongolia> Linhe District, Bayanzhuoer City, Xinjiang> Yining County, Yili, Xinjiang> Dongsheng District, Ordos, Inner Mongolia> Kashgar Prefecture, Xinjiang> Huimin District, Hohhot, Inner Mongolia.The Ala content was between 35.22 ~ 719.23 µg/g, the average value was 420.5496 µg/g, and the content order was: licorice in Guyang County, Baotou, Inner Mongolia> licorice in Yining County, Yili, Xinjiang> licorice in Kashgar Prefecture, Xinjiang> licorice in Dongsheng District, Ordos, Inner Mongolia> licorice in Linhe District, Bayanzhuoer City, Xinjiang> licorice in Hohhot Huimin District, Inner Mongolia.The Leu content ranged from 58.11 to 592.29 µg/g, with an average value of 328.9080 µg/g. The order of Leu content from highest to lowest was as follows: licorice from Guyang County, Baotou, Inner Mongolia> licorice from Dongsheng District, Ordos, Inner Mongolia> licorice from Yining County, Yili, Xinjiang> licorice from Kashgar Prefecture, Xinjiang> licorice from Hohhot Huimin District in Inner Mongolia> licorice from Linhe District, Bayanzhuoer City, Xinjiang.The Tyr content ranged from 113.77 to 421.90 µg/g, with an average value of 243.1832 µg/g. The order of Tyr content from highest to lowest was as follows: licorice from Yining County, Xinjiang> licorice from Guyang County, Baotou, Inner Mongolia> licorice from Dongsheng District, Ordos, Inner Mongolia> licorice from Linhe District, Bayanzhuoer City, Xinjiang> licorice from Kashgar Prefecture, Xinjiang> licorice from Hohhot Huimin District, Inner Mongolia.
This ranking was determined from composite entropy factor scores, which were calculated via the weighted least squares approach [39, 40], with weights representing the eigenroots of the entropy factors (F1-F5) and composite entropy factor scores (F), with weights representing the eigenroots of the entropy factors [41, 42]. The formula employed was F = 0.3075 F1 + 0.2230 F2+ 0.1901 F3+ 0.1747 F4+ 0.1047 F5.
The entropy factor scores and composite entropy factor scores [43] (Table 3) of the 39 variables, including the stability of combustion, QV, ash, crude fiber, fat, Al, Ba, Cr, Cu, Fe, K, Li, Mg, Mn, Na, Ni, Sr, Zn, CySO3H, Asp, MetSON, Thr, Ser, Glu, Gly, Ala, (Cys)2, Val, Met, Ile, Leu, Tyr, Phe, His, Lys, glycyrrhizic acid (GLA), liquiritin (LI), Arg, and Pro, the entropy factor score order of the six licorice samples was as follows: licorice of Dongsheng District, Ordos, Inner Mongolia > licorice of Yining County, Yili, Xinjiang> licorice of Guyang County, Baotou, Inner Mongolia> licorice of Hohhot Huimin District, Inner Mongolia> licorice of Kashgar Prefecture, Xinjiang > licorice of Linhe District, Bayanzhuoer City, Inner Mongolia. The results of the multi-indicator evaluation of 39 variables revealed that the licorice of Dongsheng District, Ordos, Inner Mongolia, had the highest quality, followed by the licorice of Yining County, Yili, Xinjiang. The findings indicate substantial variations in the quality of licorice from different regions, with Dongsheng District, Ordos, Inner Mongolia being identified as the source of the highest quality. These insights can inform sourcing decisions for medicinal, nutritional, and industrial applications, thereby enhancing product efficacy and safety. All the samples were purchased from the same herbal medicine market rather than being collected directly from their places of origin. These findings suggest that the samples may be subject to postharvest influences, such as storage and transportation, which may affect their ability to accurately represent differences attributable solely to their origin [44].
Systematic clustering analysis (SCA)
Systematic clustering analysis (SCA) is a statistical method for grouping data based on similarities and correlations [45, 46]. The 39 variables, including the stability of combustion, QV, ash, crude fiber, fat, Al, Ba, Cr, Cu, Fe, K, Li, Mg, Mn, Na, Ni, Sr, Zn, CySO3H, Asp, MetSON, Thr, Ser, Glu, Gly, Ala, (Cys)2, Val, Met, Ile, Leu, Tyr, Phe, His, Lys, glycyrrhizic acid (GLA), liquiritin (LI), Arg and Pro, and six types of licorice from Dongsheng District in Ordos, Inner Mongolia, Huimin District in Hohhot, Inner Mongolia, Guyang County in Baotou, Inner Mongolia, Linhe District, Bayanzhuoer City, Xinjiang, Kashgar Prefecture in Xinjiang, and Yining County in Yili, Xinjiang, can be classified into three categories (Fig. 3a). Licorice from Huimin District in Hohhot, Inner Mongolia, Guyang County in Baotou, Inner Mongolia, and Kashgar Prefecture in Xinjiang were classified into one category. Licorice from Dongsheng District in Ordos, Inner Mongolia, was grouped into one category, and licorice from Yining County in Yili and Linhe District, Bayanzhuoer City, Xinjiang, was grouped into another category. This is consistent with the results of the entropy factor analysis.
Fig. 3.
a Systematic clustering analysis (SCA) of 6 licorice samples b Systematic clustering analysis (SCA) for 39 variables of 6 licorice samples
The 39 variables were classified into four groups (Fig. 3b). The variables Cu, Ni, Cr, Al, Val, Mg, Mn, (Cys)2, fat, Lys, Ba, Fe, Ile, Pro, and Met were clustered into one category. The analysis primarily conveys information regarding energy and trace elements, which are essential components of nutrition and play crucial roles in maintaining overall health and well-being. Crude fiber, Arg, combustion stability, QV, Li, Leu, Tyr, K, and Ser were clustered into one category, which reflected the quality of licorice from the perspective of combustion. CySO3H, Ala, Gly, MetSON, His, LI, and Zn were clustered into a category that broadly encompasses biochemistry and chemistry, with a particular focus on amino acids, chemical compounds, and elements involved in biological processes. These components, including ash, Na, Sr, Asp, Thr, Glu, GLA, and Phe, are clustered into one category, which is essential for various metabolic processes and overall physiological health. Minerals (ash, Na, and Sr) are inorganic elements that play crucial roles in maintaining physiological functions, such as fluid balance (Na), bone health (Sr), and overall mineral balance. Amino acids (Asp, Thr, Glu, and Phe) are protein building blocks involved in a wide range of biological processes [47, 48].
Conclusion
Through entropy factor analysis, on the basis of 39 variables, including the stability of combustion, QV, ash, crude fiber, fat, Al, Ba, Cr, Cu, Fe, K, Li, Mg, Mn, Na, Ni, Sr, Zn, CySO3H, Asp, MetSON, Thr, Ser, Glu, Gly, Ala, (Cys)2, Val, Met, Ile, Leu, Tyr, Phe, His, Lys, glycyrrhizic acid (GLA), liquiritin (LI), Arg and Pro, the quality rankings of the six licorice samples were as follows: licorice of Dongsheng District, Ordos, Inner Mongolia > licorice of Yining County, Yili, Xinjiang> licorice of Guyang County, Baotou, Inner Mongolia> licorice of Hohhot Huimin District, Inner Mongolia> licorice of Kashgar Prefecture, Xinjiang > licorice of Linhe District, Bayanzhuoer City, Inner Mongolia. The multi-indicator evaluation of 39 variables revealed that Dongsheng District, Ordos, Inner Mongolia, produced the highest quality licorice.
The gray pattern recognition (GPR) method was used to assess the stability of licorice combustion. The results demonstrated that the combustion stability rankings of the six licorice samples were as follows: licorice of Yining County, Yili, Xinjiang > licorice of Linhe District, Bayanzhuoer city, Inner Mongolia > licorice of Dongsheng District, Ordos, Inner Mongolia > licorice of Hohhot Huimin District, Inner Mongolia > licorice of Kashgar Prefecture, Xinjiang > licorice of Guyang County, Baotou, Inner Mongolia.
Systematic clustering analysis (SCA) revealed that the licorice samples could be classified into three categories and that the 39 variables could be divided into four groups. In this study, we constructed a comprehensive evaluation system for six licorice samples based on nutritional indicators, including multiple indicators of heat of combustion, combustibility (licorice combustion stability), fat content, amino acid content, ash content, crude fiber content, glycyrrhizic acid (GLA) content, liquiritin (LI) content, and trace element content, using gray pattern recognition (GPR), entropy factor analysis (EFA), and systematic clustering analysis (SCA).This study has substantial theoretical and practical implications for the determination and comprehensive evaluation of multiple quality indicators of licorice, particularly in assessing its quality and classification.
Supplementary Information
Acknowledgements
We would like to thank AJE (https://China.aje.com) for English language editing.
Authors’ contributions
L.B. and L.L. were responsible for the experiments, and L.B. was responsible for data processing and writing the manuscript. C.Y. was responsible for sample identification, and Z.D. and Z.P. were responsible for additional testing of glycyrrhizic acid and liquiritin. All the authors reviewed the manuscript.
Funding
This research was supported by the High-Level Talents Project of Guangxi Science & Technology Normal University (GXKS2020GKY006), Laibin Scientific Research and Technological Development Program Project (Laikeneng 240203), and Guangxi Science & Technology Normal University Key Laboratory of Speciality Food Evaluation and Application (GXKSKYPT2024011).
Data availability
Studies complied with the relevant institutional, national, and international guidelines and legislation, as well as with local and national regulations. All relevant data are included in this paper.
Declarations
Ethics approval and consent to participate
Not applicable.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Zhao X, Li X, Bao A, Zhang X, Xu Y, Li Y. Metabolome and transcriptome integration explored the mechanism of browning in Glycyrrhiza uralensis Fisch cells. Front Plant Sci. 2024;15:1305871. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hui S, Kan W, Qin S, He P, Zhao J, Li H, Bai J, Wen J, Mou W, Hou M, et al. Glycyrrhiza uralensis polysaccharides ameliorates cecal ligation and puncture-induced sepsis by inhibiting the cGAS-STING signaling pathway. Front Pharmacol. 2024;15:1374179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Tao X, Hu Y, Mao N, Shen M, Fang M, Zhang M, Lou J, Fang Y, Guo X, Lin Z. Echinatin alleviates inflammation and pyroptosis in hypoxic-ischemic brain damage by inhibiting TLR4/ NF-kappaB pathway. Int Immunopharmacol. 2024;136:112372. [DOI] [PubMed] [Google Scholar]
- 4.Yang FL, Zhang YJ, Yu Y, Gao QN, Sun GX. Quality assessment of licorice extract powder through geometric linear quantified fingerprint method combined with multicomponent quantification and chemometric analysis. Microchem J. 2019;146:239–49. [Google Scholar]
- 5.Roosta HR, Estaji A, Khadivi A, Shams M. Balanced ammonium-nitrate nutrition enhances photosynthetic efficiency, micronutrient homeostasis, and antioxidant networks via ROS signaling in Glycyrrhiza glabra across soil and soilless systems. Sci Rep. 2025;15(1):25404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Yang S, Zhang K, Ji R, Chen X, Wang J, Raja IH, AnShanShan, Zhang S. Assessment of nutritional value, aerobic stability and measurement of in vitro fermentation parameters of silage prepared from several leguminous plants. BMC Plant Biol. 2025;25(1):641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Sun G, Yin R, Li X. Assessing Baizi Yangxin Pills Quality by Merging Chromatographic Fingerprints,Spectrum Fingerprints and Combustion Heat. Chin J Pharmaceuticals. 2013;44(08):789–95. [Google Scholar]
- 8.Zhou L, Huang C. Stoichiometric analysis of 20 amino acids, thermogravimetric parameters and trace elements in five types of tea from Guizhou, China, based on entropy analysis (EA) and factor cluster analysis (FCA). Food Chemistry-X. 2025;27:102457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Cattaneo F, Rodriguez IF, Zampini IC, Burgos-Edwards A, Schmeda-Hirschmann G, Isla MI. Neltuma nigra cotyledon and seed flour: nutritional, phytochemical, and techno-functional characterization and nutraceutic potential of polyphenolic enriched extract. Food Funct. 2024;15(18):9446–56. [DOI] [PubMed] [Google Scholar]
- 10.Robert P, Zamorano M, Gonzalez E, Silva-Weiss A, Cofrades S, Gimenez B. Double emulsions with olive leaves extract as fat replacers in meat systems with high oxidative stability. Food Res Int. 2019;120:904–12. [DOI] [PubMed] [Google Scholar]
- 11.Zhou L, Jiang C, Lin Q. Entropy analysis and grey cluster analysis of multiple indexes of 5 kinds of genuine medicinal materials. Sci Rep. 2022;12(1):6618. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Alamri E, Rozan M, Bayomy H. A study of chemical Composition, Antioxidants, and volatile compounds in roasted Arabic coffee. Saudi J Biol Sci. 2022;29(5):3133–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bakhsh A, Mustapha NM, Mohamed S. Catechin-rich oil palm leaf extract enhances bone calcium content of estrogen-deficient rats. Nutrition. 2013;29(4):667–72. [DOI] [PubMed] [Google Scholar]
- 14.Duan P, Han S, Wang W, Tang Y. Distribution and Migration of Trace Elements during Flotation in Ge-Rich Low-Rank Coal from Wulantuga Coal Mine, Inner Mongolia, China. ACS Omega. 2022;7(2):2023–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Li Y, Du X, Pian H, Fan X, Zhang Y, Wang T, Zhai F, Abro SM, Yu D. Effects of dietary supplement with licorice and rutin mixture on production performance, egg quality, antioxidant capacity, and gut microbiota in quails (Turnix tanki). Poult Sci. 2023;102(11):103038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Moussa SG, El Hoffy NM, Mouselhy YY, Mubarak R, Attia RT, Khalil N, Amer SA. Sustainable Treatment of Oral Traumatic Ulcers with Licorice Containing Hydrogels: Integrating Computational Modeling, Quality by Design, Green Synthesis, and Molecular Biological Evaluation. Pharmaceutics. 2023;15(12):2734. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sharifi-Rad J, Quispe C, Herrera-Bravo J, Belen LH, Kaur R, Kregiel D, Uprety Y, Beyatli A, Yeskaliyeva B, Kirkin C et al. Glycyrrhiza Genus: Enlightening Phytochemical Components for Pharmacological and Health-Promoting Abilities. Oxid Med Cell Longev. 2021; 2021:7571132. [DOI] [PMC free article] [PubMed]
- 18.Schlessinger J, Saxena S, Mohr S. Assessment of a Novel Anti-Aging Hand Cream. J Drugs Dermatol. 2016;15(4):496–503. [PubMed] [Google Scholar]
- 19.Tsai HH, Lin HW, Lu YH, Chen YL, Mahady GB. A review of potential harmful interactions between anticoagulant/antiplatelet agents and Chinese herbal medicines. PLoS ONE. 2013;8(5):e64255. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Yang F, Chu T, Zhang Y, Liu X, Sun G, Chen Z. Quality assessment of licorice (Glycyrrhiza glabra L.) from different sources by multiple fingerprint profiles combined with quantitative analysis, antioxidant activity and chemometric methods. Food Chem. 2020;324:126854. [DOI] [PubMed] [Google Scholar]
- 21.Zhou LB, Jiang CY, Lin QX, ENTROPY ANALYSIS AND GREY CLUSTER ANALYSIS, OF 8 CHINESE MEDICINAL HERBS. Fresenius Environ Bull. 2022;31(8B):9045–62. [Google Scholar]
- 22.Cui HY, Zhang CH, Li CZ, Lin L. Preparation and antibacterial activity of Litsea cubeba essential oil/dandelion polysaccharide nanofiber. Ind Crops Prod. 2019;140:111739. [Google Scholar]
- 23.Zhou LB, Zhang Q. Multiple indicators metrological analysis for 5 kinds of tea produced in Yunnan, China. Food Sci Technol. 2022;42:e70922. [Google Scholar]
- 24.Li X, Chen Z, Wang J, Mu J, Ma Q, Lu X. Symbiosis of acetic acid bacteria and yeast isolated from black tea fungus mimicking the kombucha environment in bacterial cellulose synthesis. Int Food Res J. 2023;30(6):1504–18. [Google Scholar]
- 25.Dai L, Yang L, Wang YW, Li Y, Zhao JN, Pan SX, Li YX, Yang D, He D. An Optimized Microwave-Assisted Digestion Method to Analyze the Amino Acids Profile of Quisqualis Fructus from Different Planted Origins. Foods. 2024;13(11):1645. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Kaysheva AL, Kopylov AT, Stepanov AA, Malsagova KA, Izotov AA, Shurubor YI, Krasnikov BF. Chromatomass-Spectrometric Method for the Quantitative Determination of Amino- and Carboxylic Acids in Biological Samples. Metabolites. 2023;13(1):16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Yang G, Zhang J, Dai R, Ma X, Huang C, Ren W, Ma X, Lu J, Zhao X, Ji R, et al. Comparative Study on Nutritional Characteristics and Volatile Flavor Substances of Yak Milk in Different Regions of Gannan. Foods. 2023;12(11):2172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Dang H, Zhang T, Wang Z, Li G, Zhao W, Lv X, Zhuang L. Succession of endophytic fungi and arbuscular mycorrhizal fungi associated with the growth of plant and their correlation with secondary metabolites in the roots of plants. BMC Plant Biol. 2021;21(1):165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Kuang Y, He B, Du Y, Li J, Wang D, Bi K, Li Q. Simultaneous Determinations of Eleven Bioactive Components in Suanzaoren Decoction Granules by High-Performance Liquid Chromatography and Its Application to the Quality Control in Productive Processes. Anal Sci. 2016;32(9):931–6. [DOI] [PubMed] [Google Scholar]
- 30.Cui HY, Surendhiran D, Li CZ, Lin L. Biodegradable zein active film containing chitosan nanoparticle encapsulated with pomegranate peel extract for food packaging. Food Packaging Shelf Life. 2020;24:100511. [Google Scholar]
- 31.Munhoz DR, Meng K, Wang L, Lwanga EH, Geissen V, Harkes P. Exploring the potential of earthworm gut bacteria for plastic degradation. Sci Total Environ. 2024;927:172175. [DOI] [PubMed] [Google Scholar]
- 32.Wang Y, Jia L, Guo B, Li J, Bai T, Jin Z, Jin Y. Investigation on the interaction mechanism during co-combustion of sewage sludge and coal slime: The effect of coal slime type and pretreatment method. Sci Total Environ. 2024;927:172419. [DOI] [PubMed] [Google Scholar]
- 33.Bikyashev EA, Kubrin SP, Popov AV, Nazarenko AV, Ter-Oganessian NV, Raevski IP. Evolution of the structure of MSnO(3) (M = Ba, Sr) perovskites during hydrothermal synthesis and their photocatalytic activity. Dalton Trans. 2023;52(47):17881–93. [DOI] [PubMed] [Google Scholar]
- 34.Giri NA, Gaikwad P, Gaikwad NN, Manjunatha N, Krishnakumar T, Kad V, Raigond P, Suryavanshi S, Marathe RA. Development of fiber-enriched muffins using pomegranate peel powder and its effect on physico-chemical properties and shelf life of the muffins. J Sci Food Agric. 2024;104(4):2346–58. [DOI] [PubMed] [Google Scholar]
- 35.Zhou L. Study on the Amino Acids in different Varieties of Osmanthus Fragrans Using Grey Pattern Recognition and Grey Hierarchical Clustering Analysis. IOP Conf Series: Earth Environ Sci. 2020;526(1):012079. [Google Scholar]
- 36.Szymanski J, Chlosta M, Dudek P, Rajwa P, Krajewski W, Bryniarski P, Krzanowska K, Chlosta P, Przydacz M. Prevalence, correlates, and treatment behaviors for urolithiasis and renal colic-like pain symptoms at the population level in Poland. Sci Rep. 2025;15(1):10827. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Kim K, Lee H, Lee M, Bae YH, Kim HS, Kim S. Analysis of Weather Factors on Aircraft Cancellation using a Multilayer Complex Network. Entropy (Basel). 2023;25(8):1209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Robnik J, Seljak U. Statistical Significance Testing for Mixed Priors: A Combined Bayesian and Frequentist Analysis. Entropy (Basel). 2022;24(10):1328. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Diao Y, Wang L, Chen S, Barnett LM, Mazzoli E, Essiet IA, Wang X, Wang L, Zhao Y, Li X, et al. The validity of the Physical Literacy in Children Questionnaire in children aged 4 to 12. BMC Public Health. 2024;24(1):869. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Graffelman J, de Leeuw J. Improved approximation and visualization of the correlation matrix. Am Stat. 2023;77(4):432–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Koval B, Fruhwirth-Schnatter S, Sogner L. Bayesian Reconciliation of Return Predictability. Stud Nonlinear Dyn Econom. 2024;28(2):337–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Lins-Kusterer L, Aguiar I, Santos-Lins LS, Lins-Rocha M, Aguiar CVN, Menezes MS, Sampaio AS, Quarantini LC, Cotrim HP, Brites C. Validation of the RAND 36-Item Health Survey questionnaire in Brazil. Arq Gastroenterol. 2022;59(2):193–7. [DOI] [PubMed] [Google Scholar]
- 43.Noorbaloochi S, Clothier BA, Murdoch M. Using joint probability density to create most informative unidimensional indices: a new method using pain and psychiatric severity as examples. BMC Med Res Methodol. 2024;24(1):171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Gao S, Bao J, Liu X, Stenmarck A. Life cycle assessment on food waste and its application in China IOP Conference Series: Earth and Environmental Science. 2018; 108(4):042037.
- 45.Tan C, Huang S, Xu L, Zhang T, Yuan X, Li Z, Chen M, Chen C, Yan Q. Cross-talk between oxidative stress and lipid metabolism regulators reveals molecular clusters and immunological characterization in polycystic ovarian syndrome. Lipids Health Dis. 2024;23(1):248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Xiong H. Regional coordinated development of low-carbon sports industry in China. Sci Rep. 2024;14(1):18689. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Chen Y, Yan L, Chi Y, Liu Y, Zhao Y. Protocell Self-Assembly Driven by Sodium Trimetaphosphate. Chemistry. 2023;29(39):e202300512. [DOI] [PubMed] [Google Scholar]
- 48.Yemelyanov VV, Puzanskiy RK, Shishova MF. Plant Life with and without Oxygen: A Metabolomics Approach. Int J Mol Sci. 2023;24(22):16222. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Studies complied with the relevant institutional, national, and international guidelines and legislation, as well as with local and national regulations. All relevant data are included in this paper.