Method for improving coffee flavor by combining biological fermentation with physical method

Abstract

The invention discloses a method for improving coffee flavor by combining biological fermentation with a physical method, which comprises the following steps: step (1) cleaning fresh coffee fruits, removing epicarp of coffee, and inoculating a microbial suspension for fermentation; the microorganisms are compounded by adopting lactobacillus plantarum, saccharomyces cerevisiae and leuconostoc mesenteroides; (2) carrying out ultrahigh pressure treatment or ultrasonic treatment on the fermented coffee beans and fermentation liquor; and (3) cleaning the green coffee beans subjected to ultrahigh-pressure or ultrasonic treatment, draining water, and uniformly paving the green coffee beans on a drying net for drying. Compared with natural wet fermentation, the microbial compound fermentation has the advantages that the pH drop rate is obviously increased, dominant bacterial communities such as lactic acid bacteria and saccharomycetes are increased, the degumming time is shortened by 9 hours, and the sensory quality of the obtained coffee is improved. The content of total phenols, flavones, reducing sugar and total sugar in the raw coffee beans can be increased through ultrasonic ultrahigh-pressure treatment, and the quality of the coffee after ultrahigh-pressure treatment is better improved.

Description

Technical Field

[0001] This invention relates to the field of coffee processing, and more particularly to a method for improving coffee flavor using a combination of biological fermentation and physical methods. Background Technology

[0002] Coffee cherries consist of an orange-red to red outer skin (exocarp) when ripe, a yellowish-white pulp and mucilage (mesocarp), a pure yellow parchment (endocarp), and a silvery outer skin (enum). After harvesting, coffee cherries undergo various initial processing steps to remove the outer skin and pulp, resulting in clean coffee beans. The coffee pulp is a mucilage layer composed of water, polysaccharides, and pectin, and it has strong hygroscopic properties. If not completely removed, the inner pericarp will retain pectin, which can become a breeding ground for harmful microorganisms during drying and storage, thus affecting the quality of the coffee beans. Therefore, removing the coffee cherry skin helps improve the cleanliness of the coffee beans and the quality of the coffee.

[0003] Traditional natural wet fermentation processes consume a lot of water and have a slow degumming time, resulting in coffee beans of inconsistent quality and unstable properties, with flavor and quality that need to be improved. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for improving coffee flavor using a combination of biological fermentation and physical methods. Using fresh coffee cherries as raw material, this invention selects a combination of yeast and lactic acid bacteria with high degumming efficiency for fermentation, determines the optimal fermentation process, shortens the fermentation time, and then applies ultra-high pressure and ultrasound to the fermented coffee beans for pre-drying treatment. The invention explores the impact of different pre-drying treatment methods on the flavor and quality of roasted coffee, aiming to improve coffee processing technology and enhance product flavor.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: a method for improving coffee flavor using a combination of biological fermentation and physical methods, the method comprising the following steps:

[0006] Step (1) After washing the fresh coffee cherries and removing the outer skin, a microbial suspension is introduced for fermentation; the microorganisms are a compound of Lactobacillus plantarum, Saccharomyces cerevisiae and Leuconostoc mesenteroides in a 1:1:1 ratio.

[0007] Step (2) After the fermentation is completed, the coffee beans and fermentation liquid (which can be sealed and packaged) are subjected to ultra-high pressure treatment or ultrasonic treatment; ultra-high pressure treatment is preferred.

[0008] Step (3) Clean the raw coffee beans after ultra-high pressure or ultrasonic treatment, drain the water completely, and spread them evenly on the drying net for drying.

[0009] Preferably, in step (1), the microbial fermentation time is 38-40 hours, and the inoculum size is 1-9 × 10⁻⁶.7 CFU / g, fermentation temperature 36.5-37.5℃.

[0010] Preferably, the activation process of the microorganisms is as follows: Saccharomyces cerevisiae is inoculated into YPD medium and cultured at 28-32℃ for 20-26h; Lactobacillus plantarum and Leuconostoc mesenteroides are inoculated into MRS broth medium and cultured at 36℃-37.5℃ for 10-15h; both are activated for 2-3 generations, centrifuged at 3-5℃ and 9000-11000r / min for 3-8min, washed twice with 0.8%-1% sterile physiological saline, and resuspended in sterile physiological saline for inoculation.

[0011] Preferably, the ultra-high pressure in step (2) is 250-300 MPa and the time is 20-35 min.

[0012] Preferably, in step (2), the ultrasonic temperature is 24℃-26℃, the ultrasonic power is 450-500W, and the ultrasonic time is 8-12min.

[0013] Preferably, the drying temperature in step (3) is 36.5-37.5℃, and the coffee is dried until the moisture content reaches 10-12%.

[0014] Compared with the prior art, the present invention has the following beneficial effects:

[0015] (1) This invention screened a combination of fermentation strains with high degumming efficiency by examining the pectinase activity, cellulase activity, and degumming ability of different microbial fermentation systems. The results showed that the combination of Lactobacillus plantarum, Leuconostoc mesenteroides, and Saccharomyces cerevisiae had the highest pectinase and cellulase activities and the shortest degumming time. Compared with natural wet fermentation, the pH decrease rate was significantly faster, the number of dominant bacteria such as lactic acid bacteria and yeast increased, the degumming time was shortened by 9 hours, and the sensory quality of the obtained coffee was improved.

[0016] (2) This invention analyzes the effects of ultrasonic and ultra-high pressure drying pretreatment on the physicochemical and sensory properties of coffee by examining changes in bulk density, weight loss, soluble protein, polyphenols, flavonoids, total sugar, reducing sugar, color difference, and sensory attributes of coffee beans treated with different drying pretreatment methods and direct drying. The results show that ultrasonic and ultra-high pressure treatment can increase the content of total phenols, flavonoids, reducing sugars, and total sugars in green coffee beans, while reducing the content of soluble protein. Compared with directly dried roasted beans, the treated roasted beans showed increased content of soluble protein, reducing sugars, and total sugars. The three types of coffee cupping results showed different scores in different sensory attributes. Ultra-high pressure roasted coffee showed improvements in aroma, aftertaste, and body, with the final scores being: ultra-high pressure group > ultrasonic group > direct drying group.

[0017] (4) This invention uses GC-MS technology to evaluate the differences in flavor characteristics of roasted coffee processed by ultrasound, ultra-high pressure, and direct drying. Further metabolomics techniques are used to detect metabolites to reveal the causes of these flavor differences. The results show that there are differences in volatile compounds in roasted coffee beans processed by different drying methods. A total of 92 volatile compounds were identified in the three treatments: 88 in the direct drying group, 88 in the ultrasound treatment group, and 90 in the ultra-high pressure treatment group. Pyrazines and furans were the most prevalent volatile components in roasted coffee beans. Ultrasound-treated roasted beans showed an increase in acids and phenols, resulting in richer acidity, fruitiness, and caramel aromas. Ultra-high pressure-treated beans showed an increase in the relative content of pyrazines, reaching 43.16%, resulting in richer nutty, peanut, and roasted aromas. Overall, ultra-high pressure treatment had the greatest impact on the flavor of roasted coffee. Metabolomics was used to investigate the metabolites that cause flavor differences. The results showed that 478 differential metabolites were screened in UHT green beans. The contents of 52 organic acids and their derivatives, such as flavor precursors Leu-Gly-Gly and Trp-Asp, and 66 lipids and lipid molecules, such as PC (34:5) and LPC (16:0), were significantly upregulated. 224 differential metabolites were screened in UHT roasted beans. The contents of 21 lipids and lipid molecules, such as piperidine, and 15 organic acids and their derivatives, such as N-caffeoyltryptophan and Thr-Gly-Thr, which are related to flavor, were significantly upregulated. In summary, UHT processing significantly upregulated the contents of flavor precursors in green beans and the contents of flavor-related organic acids and their derivatives, lipids, and lipid molecules in roasted beans. Attached Figure Description

[0018] Figure 1 Single-strain fermentation enzyme activity (A), fermentation system pH and degumming time (B); Note: LP is Lactobacillus plantarum, LA is Lactobacillus acidophilus, LM is Leuconostoc mesenteroides, and SC is Saccharomyces cerevisiae; letters a, b, c, d, and e indicate significant differences at the 0.05 level.

[0019] Figure 2 Compound fermentation enzyme activity (A), fermentation system pH and degumming time (B); Note: MP (Leuconostoc mesenteroides + Lactobacillus plantarum), SM (Saccharomyces cerevisiae + Leuconostoc mesenteroides), SP (Saccharomyces cerevisiae + Lactobacillus plantarum) and SMP (Saccharomyces cerevisiae + Leuconostoc mesenteroides + Lactobacillus plantarum);

[0020] Figure 3 The effect of inoculum size on pH and degumming score of coffee fermentation system;

[0021] Figure 4 The effect of fermentation time on pH and degumming score of coffee fermentation system;

[0022] Figure 5The effect of fermentation temperature on pH and degumming score of coffee fermentation system;

[0023] Figure 6 Microbial growth curves of natural fermentation (a) and compound fermentation systems (b);

[0024] Figure 7 pH change curves of natural fermentation and compound fermentation systems;

[0025] Figure 8 Sensory property curves of coffee liquids fermented using different methods;

[0026] Figure 9 The effect of ultrasonic power on the total sugar, reducing sugar, polyphenol, and flavonoid content of green coffee beans;

[0027] Figure 10 The effect of ultrasound time on the total sugar, reducing sugar, polyphenol, and flavonoid content of green coffee beans;

[0028] Figure 11 The effect of ultra-high pressure on the total sugar, reducing sugar, polyphenol, and flavonoid content of green coffee beans;

[0029] Figure 12 The effect of ultra-high pressure time on the total sugar, reducing sugar, polyphenol and flavonoid content of green coffee beans;

[0030] Figure 13 Drying curves of green coffee beans processed using different methods;

[0031] Figure 14 Bulk density (a) and weight loss rate (b) of coffee beans treated with different methods; Note: GF, GUS, and GUHP represent directly dried green beans, ultrasonically treated green beans, and ultra-high pressure treated green beans, respectively; RF, RUS, and RUHP represent directly dried roasted beans, ultrasonically treated roasted beans, and ultra-high pressure treated roasted beans, respectively; the same applies below;

[0032] Figure 15 Sensory evaluation of coffee liquids processed using different methods;

[0033] Figure 16 Relative content of volatile flavor compounds in coffee beans roasted using different processing methods;

[0034] Figure 17 Clustering heatmap of volatile flavor compounds in coffee beans processed and roasted by different methods;

[0035] Figure 18 Circular chart showing the classification and proportion of metabolites in green coffee beans (a) and roasted coffee beans (b);

[0036] Figure 19 Changes in organic acids and their derivatives in raw soybeans;

[0037] Figure 20Changes in organic acids and their derivatives in roasted beans;

[0038] Figure 21 Changes in lipids and lipid molecules in green coffee beans;

[0039] Figure 22 Changes in lipids and lipid molecules during coffee roasting. Detailed Implementation

[0040] The technical solutions of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, but the present invention is not limited to the following technical solutions.

[0041] Example 1: A method for improving coffee flavor using a combination of biological fermentation and physical methods.

[0042] The method includes the following steps:

[0043] Step (1) After washing the fresh coffee cherries, remove the outer skin and inoculate with a microbial suspension for fermentation; the microbial fermentation time is 38-40 hours, and the inoculation amount is 10. 7 CFU / g, fermentation temperature 36.5-37.5℃; each fermentation was performed in triplicate.

[0044] The microorganisms used are a compound of Lactobacillus plantarum, Saccharomyces cerevisiae and Leuconostoc mesenteroides in a 1:1:1 ratio.

[0045] The activation process of the microorganisms is as follows: Saccharomyces cerevisiae is inoculated into YPD medium and cultured at 28-32℃ for 20-26h; Lactobacillus plantarum and Leuconostoc mesenteroides are inoculated into MRS broth medium and cultured at 36-37.5℃ for 10-15h; both are activated for 2-3 generations, centrifuged at 4℃ and 9000-11000r / min for 3-8min, washed twice with 0.8%-1% sterile physiological saline, and resuspended in sterile physiological saline for inoculation.

[0046] Step (2) After fermentation, the coffee beans and fermentation liquid are sealed in a low-density polyethylene bag (or vacuum bag) and subjected to ultra-high pressure treatment or ultrasonic treatment; the ultra-high pressure is 250-300 MPa; the time is 20-35 min. The ultrasonic temperature is 24℃-26℃, the ultrasonic power is 450-500 W, and the ultrasonic time is 8-12 min.

[0047] Step (3) Clean the raw coffee beans after ultra-high pressure or ultrasonic treatment, drain the water completely, and spread them evenly on a drying net for drying at a temperature of 36.5-37.5℃. Use a halogen moisture analyzer to measure the moisture content of the coffee beans every 12 hours until the moisture content reaches 10-12%. Remove the husks from the dried coffee beans to obtain raw coffee beans, and store them at room temperature, away from light, and in an airtight container.

[0048] Example 2: Determination of Fermentation Strain

[0049] The bacterial strains were: Saccharomyces cerevisiae GIM 2.43, Lactobacillus plantarum GDMCC 1.2685, Lactobacillus acidophilus GDMCC 1.412, and Leuconostoc mesenteroides CICC 21861. All four strains were purchased.

[0050] Strain activation: Yeast and Leuconostoc mesenteroides were inoculated into YPD medium and MRS broth medium, respectively, and cultured at 30℃ for 24h; Lactobacillus plantarum and Lactobacillus acidophilus were inoculated into MRS broth medium and cultured at 37℃ for 12h. After activation for 2-3 generations, the cultures were centrifuged at 10000r / min for 5min at 4℃, washed twice with 0.85% sterile physiological saline, and resuspended in sterile physiological saline for inoculation.

[0051] Coffee bean fermentation: Wash fresh coffee cherries in clean water to remove the outer skin, and place 1 kg of peeled cherries in a glass jar; the natural fermentation microorganisms come from environmental microorganisms, and the microbial fermentation group is inoculated with a bacterial suspension at an inoculum size of 10. 7 CFU / g, each fermentation was carried out in triplicate at 30°C.

[0052] Referring to existing coffee degumming scoring standards, the pectin shedding from coffee beans in the fermentation system and the smoothness of coffee beans in the system are scored. The scores of the two indicators are added together to obtain the total score of the pectin shedding effect of coffee beans in the fermentation system. The higher the score, the better the degree of pectin shedding achieved in the system. The degumming scoring standards are shown in Table 1.

[0053] Table 1 Degumming Scoring Criteria

[0054]

[0055] Changes in enzyme activity, pH, and degumming time in the compound fermentation system

[0056] The polysaccharide composition of coffee mucilage is approximately 30% pectin, 8% cellulose, and 18% non-cellulose polysaccharides. An important aspect of selecting fermentation inoculum strains is their ability to effectively degrade the coffee mucilage layer. Pectinase refers to a collective term for various enzymes capable of breaking down pectin substances, while cellulase degrades cellulose to produce glucose. Figure 1It was found that the pectinase activity in the fermentation system inoculated with Leuconostoc mesenteriae (LM) and Saccharomyces cerevisiae (SC) was significantly increased (P < 0.05) after 24 h of fermentation, while the cellulase activity in the fermentation system inoculated with LM, SC, and Lactobacillus plantarum (LP) was significantly increased (P < 0.05). At 24 h, the pH and degumming time of the microbial fermentation groups were lower than those of the naturally fermented groups. This is because the decomposition and acidification of pectin mucus are related. In the early stages, microorganisms mainly exert their effects by producing metabolic products such as organic acids, which lowers the pH of the system. Under this environment, the water-swelling capacity of pectin tissue cells changes, and the binding capacity of cells to calcium also changes. These changes loosen the polysaccharide network of pectin, allowing the pectinase secreted by microorganisms to degrade the pectin. Therefore, one standard for evaluating the fermentation performance of microorganisms is their ability to lower the pH. It can be seen that the group inoculated with *Saccharomyces cerevisiae* had the lowest pH, and the fermentation time was shortened from 48h in the natural fermentation group to 40h. The next lowest values ​​were LP (42h) < LM (44h) < LA (46h), and the degumming time and pH changes showed the same trend. LM had higher pectinase activity than LP, but its degumming time and pH reduction ability were less than LP. This may be because LM's acid production capacity is lower than LP's, resulting in lower pectin swelling capacity, preventing pectinase and cellulase from entering the pectin layer to exert their effects. *Saccharomyces cerevisiae*, *Leuconostoc mesenteriae*, and *Lactobacillus plantarum* were selected for a combined fermentation study.

[0057] Changes in enzyme activity, pH, and degumming time in the compound fermentation system

[0058] The dominant fermentation strains selected from single-strain screening were combined to obtain four compound fermentation groups: MP (LM+LP), SM (SC+LM), SP (SC+SP), and SMP (SC+LM+LP). Figure 2It can be seen that the pectinase activity of SMP was significantly higher than that of other compound fermentation groups (P<0.05). SMP had the highest utilization rate of pectin mucilage, the fastest pH decrease rate, the strongest acid production capacity, and the lowest degumming time. Comparing the compound fermentation combination and the single-strain fermentation system, the enzyme activity of the compound fermentation group was improved compared with that of the single-strain fermentation group, and the acid production capacity was enhanced. The SMP blend fermentation enzyme activity was higher than that of the single-strain fermentation enzyme activity, and the degumming time was 40 hours. Although the degumming time was not significantly different from that of SC alone, the beans obtained from the blended fermentation were not sticky and had a rough texture. This indicates that the degumming was more thorough and complete (which is conducive to flavor substances penetrating the barrier of the coffee polysaccharide mucus layer under external physical action and entering the coffee bean interior). The viscosity of the fermentation liquid was significantly reduced, indicating that the polysaccharides in the coffee mucus layer were significantly degraded (the molecular weight distribution of coffee mucus polysaccharides was degraded from 4.3kDa-25.8kDa to 0.5kDa-11.3kDa, and the proportion of small molecule polysaccharides (<3kDa) was significantly increased to 82.1%), laying a good foundation for the subsequent improvement of coffee flavor and quality. In summary, fermentation using *Saccharomyces cerevisiae*, *Lactobacillus plantarum*, and *Leuconostoc mesenteroides* resulted in a faster pH decrease and higher pectinase activity, leading to better degumming. In particular, the polysaccharides in the coffee mucilage layer were significantly hydrolyzed, facilitating the entry of small-molecule reducing sugars into the coffee bean interior. This lays a solid foundation for subsequent enhancements in coffee flavor and quality based on Maillard and caramelization reactions. Therefore, *Saccharomyces cerevisiae*, *Lactobacillus plantarum*, and *Leuconostoc mesenteroides* were selected as the optimal combination of fermentation strains for further optimization of the fermentation process.

[0059] Single-factor experiment of fermentation process

[0060] Effects of inoculum size on pH and degumming score of coffee fermentation system: Coffee fermentation process was optimized using *Saccharomyces cerevisiae*, *Lactobacillus plantarum*, and *Leuconostoc mesenteroides* as fermentation strains. The inoculum size of the strains had a significant impact on the fermentation process. Figure 3 It can be seen that the degumming score of the fermentation system shows a trend of first increasing and then decreasing. When the inoculum size is small, there are fewer dominant microorganisms, the pH is higher, and the production of pectinase and cellulase is less, resulting in slower degumming. When the inoculum size is 10... 7 At CFU / g, the degumming score reached its maximum of 9.33 points. When the inoculum size was higher than 10... 7 At a concentration of CFU / g, the degumming score of the fermentation system began to decrease, and excessively high inoculum sizes significantly affected the degumming time. Therefore, based on single-factor experiments and after comprehensive consideration, the optimal inoculum size was selected as 10. 7 CFU / g, the inoculum size was selected as 10 in the response surface methodology design. 6 10 7 10 8 CFU / g.

[0061] The effect of fermentation time on pH and degumming score of coffee fermentation system: The effect of different fermentation times on coffee fermentation system is as follows Figure 4 As shown, the pH of the fermentation system decreased significantly with increasing fermentation time. At 24 hours, due to the short fermentation time, less acidic substances were produced by microbial degradation of nutrients such as pectin, resulting in a higher pH. After 40 hours of coffee fermentation, the pH dropped to 3.39, indicating that lactic acid bacteria and yeast were in a stable phase with relatively stable acid production. After 48 hours of fermentation, the pH decreased to 3.30, gradually leveling off after that point. The degumming score showed a trend of first increasing and then leveling off. At 40 hours of fermentation, the degumming score was 9.67. Beyond 40 hours, the degumming score slope slowed, as the prolonged fermentation time led to the consumption of nutrients such as pectin, resulting in a less significant increase in the degumming score. Insufficient fermentation time meant that the required fermentation was not achieved, and the pectin and mucilage from the coffee beans could not be completely removed. The degumming score of the system increased slightly after 48 hours of fermentation compared to 40 hours. After comprehensive consideration, 40 hours was selected as the optimal fermentation time. Therefore, in the response surface optimization design, the experimental fermentation times were 32 hours, 40 hours, and 48 hours.

[0062] The effect of fermentation temperature on the pH and degumming score of coffee fermentation systems: such as Figure 5 As shown, with increasing fermentation temperature, the pH of fermented coffee beans exhibits a trend of first decreasing and then increasing, reaching a minimum inflection point of 3.35 at 32℃. When the fermentation temperature is below 32℃, lactic acid bacteria cannot reproduce and grow well, producing less acid, thus the pH decreases more slowly. When the fermentation temperature is above 32℃, yeast reproduction and growth are inhibited, reducing acid production, thus the pH increases. Degumming in the fermentation system shows a trend of first increasing and then leveling off. At 36℃, the degumming score reaches its maximum of 10 points. This may be because appropriately increasing the temperature is beneficial for lactic acid bacteria fermentation, promoting the production of pectinase, increasing enzyme activity, and thus increasing the degumming score. When the temperature is above 36℃, the degumming score no longer changes. Therefore, in this single-factor experiment, 36℃ was selected as the optimal fermentation temperature after comprehensive consideration. Thus, in the response surface methodology, the experimental fermentation temperatures were selected as 32℃, 36℃, and 40℃.

[0063] Results of Response Surface Experiment for Fermentation Process Optimization: Based on single-factor experiments, a response surface design was conducted using the Box-Behnken experimental design principle in Design-Expert 13 software. Fermentation time (A), inoculum size (B), and fermentation temperature (C) were considered as factors, and the degumming score at the end of fermentation was used as the response value. The regression equation for the degumming score as a function of the process conditions (fermentation time, inoculum size, and fermentation temperature) was obtained as follows: Y = 9.87 - 0.1667A + 0.58B + 0.83C - 0.083AB - 0.083AC - 0.25BC - 1.06A 2-1.06B 2 -0.725C 2

[0064] The equations were solved using Design Eepert 13 software, and the predicted optimal conditions were found to be a fermentation time of 38.95 h, an inoculum size of 7.36 lg CFU / g, and a fermentation temperature of 37.27 °C. Under these conditions, the degumming score at the end of fermentation was 10.12. For the convenience of subsequent experiments, the optimal conditions were modified to a fermentation time of 39 h and an inoculum size of 10... 7 The fermentation temperature was 37℃ and the CFU / g was used. To test the reliability of the fitting results, a verification experiment was conducted using the modified experimental conditions. The test results showed that the degumming score was 10 points, which was close to the predicted value.

[0065] Quality testing of fermented coffee beans

[0066] Microbial growth curve of fermentation system: Figure 6 It can be seen that in a natural fermentation system, the initial number of natural microorganisms in coffee berries is 10. 2 -10 5 CFU / g, the microbial population showed a slight decrease in the initial fermentation stage (0-16h), a slight decrease in the number of microorganisms due to competition (16-24h), and a stable total microbial count (24-48h) at 10. 8 In the CFU / g order of magnitude, the viable counts of the dominant microorganisms, lactic acid bacteria and yeast, exceeded the viable bacterial count after 40 hours of fermentation. In the compound fermentation system, the initial microbial count reached 10. 6 -10 7 At 32 hours, the number of viable yeast and lactic acid bacteria reached their maximum values ​​of CFU / g, at 3.47 × 10⁻⁶ CFU / g. 8 CFU / g and 3.58×10 8 CFU / g, when fermentation exceeds 32 hours, the dominant bacterial populations such as yeast and lactic acid bacteria gradually decrease. The total bacterial count reaches its highest concentration of 3.30 × 10⁻⁶ CFU / g at 40 hours. 8 After 40 hours, the CFU / g level showed a significant decrease due to nutrient consumption.

[0067] pH change curve of fermentation system: from Figure 7 It can be seen that the initial pH value in both fermentation systems was about 4.5 to 4.7, which gradually decreased after the start of fermentation. At 48 hours, the pH value of the natural fermentation group reached 3.30, while the pH value of the compound fermentation group reached 3.21 at 40 hours of fermentation. Due to the growth and reproduction of yeast and lactic acid bacteria in the fermentation system, the acid production increased, and the compound fermentation group was characterized by higher acidity.

[0068] Sensory evaluation of coffee liquids from different fermentation methods

[0069] Following the Specialty Coffee Association of America (SCAA) cupping system, 8.5g of freshly ground coffee powder was accurately weighed into a cupping bowl, and 150mL of 93℃ hot water was poured in. The sensory evaluation experiment was conducted by an evaluation panel of 10 trained members (5 women and 5 men), aged between 20 and 35 years old. They scored the coffee on 10 indicators, including aroma, flavor, acidity, body, aftertaste, balance, consistency, purity, sweetness, and overall character. According to SCAA standards, full marks were given for consistency, purity, and sweetness. The sum of the scores for all 10 sensory indicators was the final score for a single sample.

[0070] Figure 8 The sensory attribute curves of roasted coffee liquids treated with blended fermentation and natural fermentation are shown. As can be seen from the graph, coffee inoculated with blended strains and naturally fermented coffee received different scores on various sensory attributes. The overall score improved from 78.5 points in the naturally fermented group to 80 points, indicating that the quality of coffee beans after blended fermentation improved from superior commercial beans to specialty coffee beans (according to the Specialty Coffee Association (SCAA) standards, coffee with a total score of 75-79 points is classified as superior commercial beans, and a total score of 80-84 points is classified as specialty coffee). The aroma (7.7) and acidity (6.9) of blended fermented coffee were higher than those of naturally fermented coffee (6.6) and acidity (6.5). These results are consistent with other research reports, which indicate that coffee fermented with *Lactobacillus plantarum* and *Leuconostoc mesenteroides* has higher acidity and more aroma than naturally fermented coffee. The flavor (8.3) and body (7.7) intensity of blended fermented coffee were higher than those of naturally fermented coffee (6.7) and body (7.5). Because the final metabolites produced by microbial activity affect the chemical composition of coffee, these microbial-derived metabolites can diffuse into the coffee beans and remain after roasting, playing a crucial role in the quality of the coffee beverage. The results show that the inoculation-based compound fermentation method used in this experiment imparts a fruity aroma and a high acidity to the coffee beans. This is closely related to the different precursor substances produced in the green beans under the influence of the inoculated compound microorganisms during fermentation. The inoculation-based compound fermentation method in this invention brings higher sensory quality to the coffee beans.

[0071] Example 3: Determination of Ultrasonic High-Pressure Drying Pretreatment Parameters

[0072] Based on the response surface methodology for fermented coffee, ultrasonic or ultra-high pressure (UHP) treatment was introduced to improve coffee flavor after fermentation. Coffee beans that had not been cleaned of the fermentation mucus after fermentation were divided into three equal groups: the first group was a control group that was dried directly without treatment; the second group underwent ultra-high pressure (UHP) treatment; and the third group underwent ultrasonic (US) treatment. The latter two groups were vacuum-packed in low-density polyethylene (LDPE) bags. One group of packaged coffee beans was placed in a UHP container for ultra-high pressure treatment using water as the medium; the other group was placed in an ultrasonic bath for ultrasonic treatment.

[0073] Ultrasonic pretreatment time: The ultrasonic treatment time was set to 0-40 min, the ultrasonic treatment power was 480 W, and the constant ultrasonic temperature was 25℃. The weighted comprehensive score of total sugar content, reducing sugar content, total phenolic content, and flavonoid content in green coffee beans was used as the evaluation value. The weighting coefficients for total sugar content, reducing sugar content, total phenolic content, and flavonoid content were 0.3, 0.3, 0.2, and 0.2, respectively. The same applies below. Samples of green coffee beans treated with different ultrasonic powers were taken and measured to investigate the effect of different ultrasonic times on the comprehensive score of green coffee beans.

[0074] Ultrasonic pretreatment power: Fermented coffee beans were subjected to ultrasonic treatment before drying to improve the flavor of the fermented coffee. The ultrasonic treatment power was set from 0 to 600W, the ultrasonic treatment time was 30 minutes, and the ultrasonic temperature was kept constant at 25℃. The weighted comprehensive score of total sugar content, reducing sugar content, total phenol content and flavonoid content in green coffee beans was used as the evaluation value. Samples of green coffee beans treated with different ultrasonic powers were taken and measured to investigate the effect of different ultrasonic powers on the comprehensive score of green coffee beans.

[0075] Ultra-high pressure pretreatment pressure: Based on the response surface methodology for fermented coffee, ultra-high pressure treatment was introduced before drying to improve the flavor of fermented coffee. The ultra-high pressure treatment pressure was set from 0 to 600 MPa, and the ultra-high pressure treatment time was 20 min. The weighted comprehensive score of total sugar content, reducing sugar content, total phenolic content, and flavonoid content in green coffee beans was used as the evaluation value. Samples of green coffee beans treated with different ultrasonic powers were taken and measured to investigate the effect of different ultra-high pressure treatment pressures on the comprehensive score of green coffee beans.

[0076] Ultra-high pressure pretreatment time: The ultra-high pressure treatment time was set to 0-40 min, and the ultra-high pressure treatment pressure was 300 MPa. The weighted comprehensive score of total sugar content, reducing sugar content, total phenol content and flavonoid content in green coffee beans was used as the evaluation value. Samples of green coffee beans treated with different ultrasonic powers were taken and measured to investigate the effect of different ultra-high pressure treatment times on the comprehensive score of green coffee beans.

[0077] Total polyphenol content determination: Spectrophotometry was used. Approximately 0.25 g of ground coffee beans (sieved through an 80-mesh sieve) was mixed with 5 mL of 85% methanol and filtered. Calibration curves were plotted using gallic acid standard solutions (0.00, 0.02, 0.04, 0.06, 0.08, and 0.1 mg / mL). 0.5 mL of the standard solution and filtrate were added, mixed with 0.5 mL of Folin-Ciocalteu reagent for 5 min, then 1.0 mL of saturated sodium carbonate was added. The mixture was reacted with the sample in the dark for 30 min, and the absorbance detected on the spectrometer was 760 nm. The standard equation is: y = 4.7767x + 0.0108, R0 2 =0.9979.

[0078] Determination of flavonoid content: Extracted coffee (with the same polyphenol content) was prepared using rutin standard solutions (0.00, 0.02, 0.04, 0.06, 0.08, 0.10 mg / mL), and a standard curve was plotted. 1.0 mL of the standard and sample were added, followed by 0.5 mL of 5% NaNO₂ and mixed for 6 min. Then, 0.5 mL of 10% AlNO₃ was added and mixed for 6 min. Finally, 4 mL of 4% NaOH was added and mixed for 10 min. The absorbance measured on the spectrometer was 510 nm. The standard equation is: y = 0.9148x + 0.0024, R₀ = 0.9148x + 0.0024. 2 =0.9977.

[0079] Determination of total sugar: Weigh approximately 1.0 g of ground coffee beans (sieved through an 80-mesh sieve) and mix with 20 mL of water, then filter. Plot a standard curve using glucose standard solutions (0, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08 mg / mL). Take 2 mL of standard, sample, and water, add 1.0 mL of 6% phenol and 5.0 mL of concentrated sulfuric acid, shake well, cool, and allow to stand at room temperature for 30 minutes. Measure the optical density at 490 nm. The standard equation is: y = 9.7055x + 0.0018, R0 2 =0.9997.

[0080] Determination of reducing sugar content: The sample preparation method is the same as for total sugar. Prepare glucose standard solutions (0, 0.2, 0.4, 0.6, 0.8, 1.0 mg / mL) and plot a standard curve. Take 1.0 mL of the standard and sample, add 2 mL of DNS reagent to each, heat in a boiling water bath for 2 min, cool under running water, and bring the volume to 15 mL. Measure the absorbance at 540 nm. Determine the glucose mg / mL value from the standard curve and calculate the reducing sugar content in the sample. The standard equation is: y = 0.606x + 0.046, R 2 =0.9996.

[0081] Determination of soluble protein: The Coomassie Brilliant Blue method was used. The sample preparation method was the same as for total soluble sugar. 0.5 mL of a 50 mg / mL sample solution was added to 5 mL of Coomassie Brilliant Blue (G-250), mixed well, and allowed to stand at room temperature for 2-5 min. The absorbance was then measured at 595 nm. A standard curve was plotted using bovine serum albumin (BSA), with the standard equation: y = 2.9453x + 0.0082, R0 2 =0.9989.

[0082] Measurement of bulk density and weight loss of soybeans

[0083]

[0084] In the formula, the mass before baking is m1, and the mass after baking is m2.

[0085] Bulk density (g / L): 250 mL of green and roasted coffee beans were measured using a 250 mL graduated cylinder, weighed, and the weight was multiplied by 4 to obtain the density value. Effect of ultrasonic power on the total sugar, reducing sugar, polyphenol, and flavonoid content of green coffee beans.

[0086] After response surface methodology optimization, coffee beans with unwashed fermentation mucus were subjected to ultrasonic treatment. Sugars, as precursors to the Maillard reaction, significantly influence the coffee roasting process. Flavonoids directly affect the color and flavor of the coffee, while polyphenols are important bioactive substances in coffee, possessing high nutritional value but also contributing to bitterness and astringency. The ultrasonic temperature was fixed at 25℃, the ultrasonic time at 30 min, and the ultrasonic power was set to 0, 240, 360, 480, and 600 W. Figure 9 As shown, the results indicated that ultrasonic treatment significantly increased the total sugar, reducing sugar, and polyphenol content of green coffee beans (P < 0.05). With increasing ultrasonic power, the total sugar content initially increased and then decreased, reaching its maximum at 360W. Beyond 360W, the total sugar content gradually decreased. The highest overall score was achieved at 480W (with weighting coefficients of 0.3 for total sugar content, 0.3 for reducing sugar content, 0.2 for flavonoid content, and 0.2 for total phenol content). Therefore, 480W was the optimal power for ultrasonic treatment of coffee beans after fermentation, and subsequent experiments selected 480W as the optimal ultrasonic treatment power.

[0087] Effects of ultrasound time on the total sugar, reducing sugar, polyphenol, and flavonoid content of green coffee beans

[0088] The ultrasonic temperature was kept constant at 25℃, and the ultrasonic power was 480W. The ultrasonic power was set to 0, 10, 20, 30, and 40 min, respectively. The results are as follows: Figure 10As shown, the ultrasonic treatment time has a certain impact on the total sugar, reducing sugar, polyphenol, and flavonoid content of green coffee beans. With the gradual increase of ultrasonic time, the contents of total sugar, reducing sugar, polyphenol, and flavonoids all show a trend of first increasing and then decreasing, reaching their maximum at 10 minutes. This suggests that a suitable ultrasonic time allows sugars and phenols in the fermentation broth to enter the coffee beans. When the treatment time exceeds 10 minutes, the contents of each substance begin to gradually decrease. This is because the cell walls of the coffee beans are broken down, and the cell contents flow out, resulting in a reduction in the contents of total sugar, reducing sugar, polyphenols, and flavonoids. The highest overall score was achieved at 10 minutes of ultrasonic treatment; therefore, 10 minutes is the optimal ultrasonic treatment time for fermented coffee beans, and subsequent experiments selected 10 minutes as the optimal ultrasonic treatment time.

[0089] The effect of ultra-high pressure on the total sugar, reducing sugar, polyphenol, and flavonoid content of green coffee beans

[0090] The ultra-high pressure time is 20 minutes, and the ultra-high pressure is set to 0, 150, 300, 450, and 600 MPa respectively. Figure 11 As shown, the results indicate that with the gradual increase of ultra-high pressure, the contents of total sugar, reducing sugar, polyphenols, and flavonoids all show a trend of first increasing and then decreasing, reaching their highest levels at an ultra-high pressure of 300 MPa. When the ultra-high pressure exceeds 300 MPa, the contents of total sugar, reducing sugar, polyphenols, and flavonoids all decrease significantly. When the ultra-high pressure is within an appropriate range, sugars, phenols, and flavonoids in the fermentation broth enter the coffee bean tissue due to the high pressure. However, as the pressure exceeds 300 MPa, the cell structure in the green coffee beans is also destroyed, and substances within the green bean cells begin to flow out, leading to a decrease in the contents of total sugar, reducing sugar, polyphenols, and flavonoids. The highest overall score was achieved at an ultra-high pressure of 300 MPa, therefore 300 MPa is the optimal pressure for ultra-high pressure treatment of fermented coffee beans, and subsequent experiments selected 300 MPa as the optimal ultra-high pressure treatment pressure.

[0091] Effects of ultra-high pressure time on the total sugar, reducing sugar, polyphenol, and flavonoid content of green coffee beans

[0092] The ultra-high pressure is 300 MPa, and the ultra-high pressure treatment time is set to 0, 10, 20, 30, and 40 minutes respectively. Figure 12 As shown, the results indicate that the contents of total sugar, reducing sugar, polyphenols, and flavonoids generally showed a trend of first increasing and then decreasing with increasing ultra-high pressure (UHPP) time. Within a suitable holding time, substances such as total sugar, reducing sugar, polyphenols, and flavonoids in the coffee mucilage enter the green coffee beans due to high pressure. However, exceeding the suitable holding time leads to the destruction of the cell structure and the outflow of cell contents. The highest overall score was achieved when the UHPP treatment time was 30 minutes; therefore, 30 minutes is the optimal time for UHPP treatment of fermented coffee beans, and subsequent experiments selected 30 minutes as the optimal UHPP treatment time.

[0093] Physicochemical composition analysis of green and roasted coffee beans under different drying pretreatments

[0094] Coffee quality can be comprehensively evaluated from physical, chemical, and sensory perspectives. Physical analysis involves assessing the properties of coffee beans, such as color, shape, size, and the presence of defects; chemical analysis involves various non-volatile compounds that act as precursors to the formation of volatile compounds during the roasting process.

[0095] Variations in drying curves of green coffee beans under different treatments

[0096] To control the moisture content of the obtained green coffee beans within a certain range and improve the flavor and taste of the beans during the roasting stage, after the wet fermentation and physical processing, the coffee beans were washed with water to remove excess residue and then dried at a constant temperature of 37°C. The moisture content of the green coffee beans was measured every 12 hours during the drying process to study the changing trends of moisture content in green coffee beans treated with different drying pretreatments. Figure 13 It can be seen that there are significant differences among the three drying curves. Initially, at 0 hours, the moisture content of the coffee beans is the highest, reaching approximately 49.88%. As fermentation and drying time continue to extend, the moisture content gradually decreases. The rate of moisture content decreases rapidly within the 0-48 hour range, while the drying rate slows down somewhat during the 48-60 hour stage, which is the later stage of drying. At 40 hours, the ultra-high pressure treatment group completes drying, with a moisture content of 10.57%. The rapid decrease in moisture content in the ultra-high pressure group may be due to the high pressure creating a loose internal space within the beans, facilitating moisture evaporation. The ultrasonic treatment group and the direct drying group complete drying at 60 hours, with moisture contents of 12.64% and 10.73%, respectively, achieving a bean moisture content of 10%-12%, which meets international standards.

[0097] Analysis of changes in weight loss and bulk density of coffee beans under different treatments

[0098] The physical changes caused by coffee roasting are mainly reflected in drastic alterations in shape, color, moisture content, density, and internal structure. Due to the different processing methods used for the three types of roasted coffee samples, their bulk density and weight loss also exhibit certain differences. For example... Figure 14 The graph shows the changes in bulk density and weight loss of green and roasted coffee beans. As can be seen from the graph, the bulk density of the green coffee beans in the ultra-high pressure treatment group was significantly higher than that of the other two groups (P < 0.05), while the bulk density of the roasted coffee beans in the ultra-high pressure treatment group was significantly lower than that of the other two groups (P < 0.05), indicating that the volume expansion of the ultra-high pressure treatment group after roasting was greater than that of the other two groups. In this experiment, the weight loss of coffee beans after roasting was significant. The weight loss rates of the ultrasonic treatment group and the ultra-high pressure treatment group were significantly lower than those of the direct drying group (P < 0.05). This roasting loss mainly includes water evaporation, conversion of organic matter into gas, and physical losses.

[0099] Chemical composition analysis of coffee beans processed in different ways

[0100] Table 2 Effects of different treatments on the chemical composition of coffee beans

[0101]

[0102] Note: Different lowercase letters in the data of different sample types indicate that the differences are significant at the 0.05 level. The data are mean ± standard deviation, and the number of replicates is n=3. The same applies below.

[0103] Table 2 shows that the protein, total phenols, flavonoids, reducing sugars, and total sugars of the three groups of treated green and roasted coffee beans all exhibited significant changes (P < 0.05). Specifically, the soluble protein content increased in the ultrasonic treatment group and decreased in the ultra-high pressure treatment group. This is because ultra-high pressure denatures the proteins in coffee; however, the ultra-high pressure treatment group experienced the least protein loss after roasting, with a loss rate of 39.74%, while the protein loss rates in the direct drying and ultrasonic groups were 73.68% and 65.19%, respectively. The reducing sugar and total sugar content of the three groups of green and roasted coffee beans was in the order of ultra-high pressure treatment group > ultrasonic treatment group > direct drying group. The flavonoid content increased in green coffee beans after ultrasonic and ultra-high pressure treatment. Compared with directly dried roasted beans, the flavonoid content increased in ultra-high pressure roasted beans, while the flavonoid content decreased in ultrasonic roasted beans. The total phenol content increased in green coffee beans after ultra-high pressure and ultrasonic treatment; however, the total phenol content decreased in the ultra-high pressure and ultrasonic groups after roasting. In conclusion, ultrasonic and ultra-high pressure processing can increase the content of flavor precursors and flavor compounds in coffee beans after roasting.

[0104] Sensory index analysis of coffee beans under different drying pretreatments

[0105] The effect of different drying pretreatments on the appearance quality of coffee beans: The color of green and roasted coffee beans was evaluated using a colorimeter, and the parameters L*, a*, and b* were measured 5 times for each group.

[0106] Table 3. Analysis of color parameters of green and roasted beans under different treatment methods.

[0107]

[0108] In this invention, the pre-drying treatment method affects the color of both green and roasted coffee beans. Table 3 shows that, overall, the color change of coffee beans treated with ultra-high pressure is the most significant, while the colors of coffee beans in the direct-drying group and the ultrasonic-treated group are relatively similar. The L* value represents brightness and is inversely proportional to the sample color; a higher L* value indicates a lighter sample color. The L* values ​​of ultra-high pressure treated green beans are significantly lower than the other two groups (P < 0.05), while the L* values ​​of ultra-high pressure treated roasted beans are significantly higher than the other two groups (P < 0.05). The a* value represents red-green hue, and the b* value represents yellow-blue hue (yellow b+, blue b-). Compared with directly dried roasted beans, the b* values ​​of ultra-high pressure and ultrasonic-treated roasted beans are significantly higher (P < 0.05). Phenolic substances are the main source of the yellow hue in the samples. The phenolic content in coffee is directly proportional to the b* value; because the phenolic content of coffee beans treated with ultrasound and ultra-high pressure is increased, the b* value of roasted coffee beans increases. In conclusion, ultrasonic and ultra-high pressure drying pre-treatments affect the color of coffee. The effect of different pre-drying treatments on the sensory quality of coffee beans

[0109] To determine the actual sensory differences between coffee samples with different pretreatments, cupping was performed on the main sensory characteristics of coffee according to SCAA standards. Figure 15 Sensory attribute curves of coffee beverages obtained from three different pre-drying treatments are shown. According to the SCAA scoring criteria, all coffees received a final score higher than 80, allowing them to be classified as specialty coffee. Furthermore, the final scores of coffee beverages treated with ultrasound (82 points) and ultra-high pressure (UHPP) (82.6 points) were higher than those of the directly dried group (80 points). The aroma (7.9) and body (7.9) scores of the UHPP-treated samples were higher than those of the directly dried samples. Aftertaste refers to the aroma perceived after inhaling the coffee, lingering in the mouth, and can be perceived through taste, touch, and even residual nasal aroma. The aftertaste scores of the ultrasound- and UHPP-treated samples were improved to 8.7 and 8.8 respectively, while the directly dried samples had the lowest aftertaste (8.5). Flavor refers to the overall flavor perception perceived upon inhalation, including aroma and taste; the UHPP-treated samples had the highest flavor score (8.3). The consistency of the ultrasound-treated (8.7) and UHPP (8.3) samples was higher than that of the directly dried samples (7.9). After ultra-high pressure treatment, the intensity of flavor, acidity, aroma, consistency, and aftertaste increased, while ultrasonic treatment increased the intensity of consistency, acidity, and balance. Ultra-high pressure beans also showed an increase in acidity and bitterness. The results indicate that the ultrasonic and ultra-high pressure treatments used in this experiment brought new flavor characteristics to the coffee beans, giving them a higher acidity and a nutty aroma. This is attributed to the characteristic aroma compounds such as pyrazines and esters formed after the final roasting reaction following ultrasonic and ultra-high pressure treatment. In conclusion, ultrasonic and ultra-high pressure drying pre-treatments improved the overall sensory qualities of the coffee.

[0110] Analysis of volatile flavor components of roasted coffee beans under different drying pretreatments: Solid phase microextraction coupled with mass spectrometry can be used for identification in food. The volatile substances of roasted coffee beans under three different drying pretreatments were identified and analyzed by HS-SPME / GC-MS. The results were compared with the NIST 2018 standard spectral library, and a total of 92 volatile substances were identified. 88, 88 and 90 volatile flavor substances were detected in directly dried roasted coffee beans, ultrasonically treated roasted coffee beans and ultra-high pressure treated roasted coffee beans, respectively. Among them, 7 aldehydes, 8 acids, 4 esters, 5 alcohols, 17 ketones, 9 furans, 25 pyrazines, 5 pyridines and 5 phenols constitute the main aroma of roasted coffee beans, as shown in Table 4.

[0111] Table 4. Volatile flavor compounds of coffee beans roasted under different treatments

[0112]

[0113]

[0114]

[0115]

[0116]

[0117] Note: The Duncan test was used, and the result is expressed as the mean ± standard deviation of three replicates. The Tukey test was used on the samples; in the row corresponding to each substance name, ND indicates that the substance was not detected.

[0118] The aroma description comes from http: / / www.perflavory.com / search.htmL; to clarify the specific categories of volatile flavor compounds in roasted coffee, the relative content of each type of volatile flavor compound is calculated using relative peak areas, such as... Figure 16 As shown in the figure, pyrazines account for more than 30% of the total volatile flavor compounds in the three types of roasted coffee beans, making them the most abundant volatile flavor compounds in roasted coffee beans. In RUHP, pyrazines account for 43.16%, with the main components being furans, acids, aldehydes, esters, alcohols, and ketones. From the percentage packing plots of the flavors of the three groups of roasted coffee beans, the flavor compositions of the ultrasonically treated group and the direct-drying group are relatively similar. Compared with the other two groups, the ultra-high pressure group shows an increase in alcohols and pyrazines, and a decrease in aldehydes, esters, ketones, furans, and phenols. By analyzing the relative content and types of volatile flavor compounds in roasted coffee beans under different treatments, the influence of pre-drying treatment on the volatile flavor compounds of roasted coffee beans and the substances with significant differences can be clearly identified.

[0119] Cluster heatmaps provide a visual comparison of differences in volatile components among different samples. To intuitively observe the differences in volatile flavor compounds of roasted coffee beans under different treatments, cluster heatmaps were generated as follows: Figure 17 Cluster heatmaps revealed significant differences in some volatile flavor compounds across groups. Lateral clustering showed that volatile compounds were clustered into five distinct regions.

[0120] Region A primarily corresponds to pyrazines, including 5H-5-methyl-6,7-dihydrocyclopentylpyrazine, isobutyraldehyde, 2,5-dimethylisopentylpyrazine, indole, 3-ethylpyridine, 2-isobutyl-3-methoxypyrazine, ethanol acetate, 2,5-dimethylpyrazine, and 2,3-ethylpyrazine. These flavor compounds exhibit high peak values ​​in ultrasonically and ultra-high pressure roasted beans. Region B involves pyrazines, mainly including 2,6-dimethylpyrazine, 2-acetyl-3-methylpyrazine, 3,5-diethyl-2-methylpyrazine, 2,3-dimethylpyrazine, 2-ethyl-3,6-dimethylpyrazine, 2-ethyl-6-methylpyrazine, furfuryl alcohol, 2-ethyl-5-methylpyrazine, 2,3,5-trimethylpyrazine, and 2-ethylpyrazine, which are present in high amounts in ultra-high pressure roasted beans. Region C exhibits an irregular pattern, with pyrrole-2-carboxaldehyde, 2,5-dimethyl-4-hydroxy-3(2H)-furanone, 2-methylbutyraldehyde, 5-methyl-2-furfural, and pyridine being flavor compounds present in higher concentrations in directly dried roasted coffee samples. Region D primarily consists of ketones, including hydroxyacetone, isovaleric acid, acetoxy-2-propanone, 2-acetylfuran, furfural acetate, and 2,3-pentanedione; these substances are present in higher concentrations in both directly dried and ultrasonically processed roasted beans. Region E mainly comprises ketones and acids, including 2,5-dimethyl-4-hydroxy-3(2H)-furanone, 2-acetylpyrrole, 2-methylpyrazine, propionic acid, 5-acetoxymethyl-2-furanaldehyde, furfural, and 2-methoxy-4-vinylphenol; these substances are present in higher concentrations in ultrasonically processed roasted coffee.

[0121] The main contributors to coffee aroma from pyrazines are 2-methylpyrazine, 2,5-dimethylpyrazine, 2,6-dimethylpyrazine, and 2,3,5-trimethylpyrazine. The relative contents of these four flavor compounds in the three groups of roasted coffee were: direct drying group (20.40%) < ultrasonic treatment group (21.22%) < ultra-high pressure treatment group (23.32%). The increase in pyrazine content mainly enhanced the nutty and roasted flavors of coffee. Roasted coffee beans treated with ultrasound and ultra-high pressure showed increased pyrazine content, with the ultra-high pressure treatment group exhibiting the highest levels. Therefore, ultra-high pressure treatment of coffee beans before drying can increase the pyrazine content in roasted beans.

[0122] Furans were the second most volatile compounds in roasted coffee. The furan content in the direct-drying group, ultrasonic treatment group, and ultra-high pressure treatment group was 29.43%, 30.26%, and 25.91%, respectively, indicating that ultra-high pressure treatment reduced the furan content in roasted coffee. Furfural and furfuryl alcohol provide a sweet caramel flavor to coffee, and their content increased in ultrasonically and ultra-high pressure roasted beans, respectively. Ketones and aldehydes decreased in roasted coffee after ultrasonic and ultra-high pressure treatment. Esters were significantly reduced in ultra-high pressure treated coffee samples because esters are converted into carbon-responsive acids and alcohols after esterase-catalyzed hydrolysis. Ultrasonically treated coffee had relatively high phenolic and acid content, with a relative acid concentration of 6.08%. Pyruvic acid, isovaleric acid, and propionic acid were relatively high, described as having caramel, acid, fruity, creamy, and sweet flavors, and significantly contributing to the acid content of the coffee samples. Guaiacin (a pungent phenolic compound) has been reported as a key aromatic volatile compound, unique to ultrasonically roasted beans, and has a positive impact on coffee flavor. Therefore, ultrasonic and ultra-high pressure processing have a positive effect on the volatile flavor of roasted coffee beans.

[0123] Metabolomics analysis of green and roasted coffee beans under different drying pretreatments

[0124] Qualitative and quantitative analysis of metabolites: To gain a deeper understanding of the changes in non-volatile compounds between green and roasted coffee beans treated by the three methods, UHPLC-QE-MS was used to perform non-target metabolomics analysis on green and roasted beans treated with different methods. A total of 6859 metabolites were extracted from six coffee bean samples, classified into 20 categories. Among them, 4110 substances were identified by primary chromatograms, and 2749 substances were identified by secondary chromatograms. In green coffee beans, the metabolites mainly consist of 20 categories of primary metabolites, including amino acids and peptides, benzene compounds, fatty acids, lipids and lipid molecules, organic acids and their derivatives, organic heterocyclic compounds, ketones, and esters. In roasted coffee beans, the main metabolites include alkaloids, benzene compounds, fatty acids, lipids and lipid molecules, organic acids and their derivatives, organic oxygen compounds, organic heterocyclic compounds, ketones, esters, and terpenes. Their classification and proportions are detailed below. Figure 18 .

[0125] Principal Component Analysis (PCA): To understand the overall metabolic differences among different processed samples and the magnitude of variability within each group, PCA was performed. PCA analysis was conducted on all coffee extracts, resulting in PCA score maps for the three coffee sample groups. The maps show that all samples in each PCA score map fall within the 95% confidence ellipse. The two major components contributed 55.7% and 46.5% to the green coffee bean group and the roasted coffee bean group, respectively. The distribution results indicate significant differences in metabolites among the different samples, allowing for effective separation.

[0126] Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA): OPLS-DA models enable supervised classification to monitor the conversion of metabolites during coffee roasting. This study established OPLS-DA models for GUS vs. GF, GUHP vs. GF, RUS vs. RF, and RUHP vs. RF treatments. Permutation tests showed that the OPLS-DA models did not exhibit overfitting, and R0 was [not specified in the original text]. 2 Y and Q 2 Y values ​​are all above 0.7, therefore, the OPLS-DA model is effective and has high predictive power, and can be used to explore metabolic differences in coffee samples with different treatments.

[0127] Screening and Identification of Differential Metabolites: To gain a deeper understanding of the impact of the three treatments on the non-volatile qualities of coffee beans and the reasons for flavor changes, based on the established OPLS-DA model, differential metabolites between green and roasted coffee beans under different treatments were further screened and identified. The screening criteria were a VIP value ≥ 1, a fold change factor (FC) value less than 0.5 or greater than 2, and P < 0.05 to identify differential metabolites. The screening results showed that multiple up- and down-regulated differential metabolites were present in all coffee groups treated differently. Compared to the GF group, the GUS group had 1489 differential metabolites (115 down-regulated, 1374 up-regulated), and the GUHP group had 3206 differential metabolites (363 down-regulated, 2843 up-regulated). In roasted coffee beans, compared to RF, the RUS group had 883 differential metabolites (109 down-regulated, 73 up-regulated), and the RUHP group had 1724 differential metabolites (673 down-regulated, 1051 up-regulated).

[0128] To identify metabolites, their exact molecular weights were confirmed. Fragment information obtained using MS / MS (including the mass ratio, retention time, and peak area of ​​the identified metabolites) was matched with the Human Metabolome Database (HMDB) (Metlin, MassBank, Lipid Maps, mzclound) and a self-built standard library to obtain accurate metabolite information. To understand the changes in differential metabolites under different processing methods, the annotated differential metabolites and their up- and down-regulation were statistically analyzed, as shown in Table 5. The GUHP vs GF and RUHP vs RF comparison groups showed the most differential metabolites, indicating that ultra-high pressure processing has the most significant impact on coffee non-volatile metabolites.

[0129] Table 5. Quantities and up- and down-regulation of differentially expressed metabolites in coffee beans processed using different methods.

[0130]

[0131] Differential metabolite analysis of coffee samples from different treatments

[0132] Organic acids and their derivatives

[0133] Organic acids are important chemical components in coffee beans, accounting for approximately 11% and 6% of green and roasted coffee beans, respectively, and are among the substances that significantly contribute to the quality and flavor of coffee beans. In this study, 63 differential organic acid metabolites were identified from three types of green coffee beans, and cluster heatmaps were generated as follows: Figure 19 The main metabolites are amino acids and carboxylic acids. In green coffee beans, Leu-Gly-Gly, Trp-Asp, Val-Ile, Phe-Glu, Leu-Tyr, Arg-Phe, 1-α-acetylalanine, and Thr-Gly-Thr are the differentially expressed metabolites contributing significantly to the GUHP group. Leu-Tyr, 9,10-dihydrolysergic acid, Thr-Gly-Thr, phenylpropionic acid, and N,N-dimethylacrylamide are the differentially expressed organic acids and their derivatives contributing significantly to the GUS group. During coffee bean roasting, amino acids react with an equal amount of glucose at 180℃ to produce caramel aroma, and through the Sterk degradation reaction, they generate aldehydes and ketones, forming unique aromas and flavors. After roasting, Val produces a chocolate aroma, Phe produces a special violet aroma, Arg produces a roasted sucrose aroma, Glu produces a caramel aroma, Leu produces a roasted cheese aroma, and Ile produces a roasted cheese aroma, etc. The content of organic acids and their derivatives in green and roasted coffee beans treated with ultra-high pressure (UHPP) is generally higher than that in the other two treatment methods. Val-Ile, Val-Glu, and N-caffeoyltryptophan are substances with relatively high content of differential organic acid metabolites in green coffee beans. They are all differential metabolites unique to GUHP, and their content increased by 2.83, 4.48, and 2.78 times, respectively, compared with GF.

[0134] Organic acids and their derivatives decompose during the roasting process, resulting in a decrease in the content of organic acids and their derivatives in roasted coffee beans. Figure 20 It was found that the number of differentially expressed organic acids and their derivative metabolites decreased for all three groups of roasted coffee beans. A total of 21 differentially expressed organic acid metabolites were identified, mainly belonging to the amino acid class, primarily Trp, Glu, and Gly. Some free amino acids present in roasted coffee can directly contribute to the flavor of coffee beverages; Glu contributes to the sour taste, while Trp and Gly contribute to the sweet taste. Among them, N-caffeoyltryptophan, threonine-glycyl-threonine, and Thaxtomin A were relatively abundant substances among the differentially expressed organic acid metabolites in roasted coffee beans, all of which are RUHP-specific differentially expressed metabolites. N-caffeoyltryptophan and Thr-Gly-Thr increased by 2.14 and 2.81 times respectively compared to RF, while Thaxtomin A decreased by 0.476 times.

[0135] In general, the differential organic acid content in the coffee beans studied was GUHP > GUS > GF; while the HS-SPME-GC-MS results of the roasted coffee beans showed that the relative content of RUHP pyrazines was higher than that of the other two groups. The amino acids in the roasted beans affect the flavor quality of the coffee beans through their own flavor properties or by forming flavor proteins, giving the coffee certain sweetness and body.

[0136] Lipids and lipid molecules

[0137] Lipids are an important factor in evaluating coffee quality. During roasting, lipids undergo auto-oxidation and decomposition reactions, which contribute to aroma formation and are crucial to the flavor of coffee beverages. Lipids in roasted coffee absorb hydrophobic flavor components, thus making a significant contribution to the product's taste.

[0138] In the three groups of raw soybeans, a total of 86 differentially expressed lipids and lipid metabolites were screened, of which 81 were significantly elevated. Figure 21 GUS, accounting for 94.19% of the total differential lipids, mainly includes fatty acids, glycerophospholipids, glycerides, and steroids and their derivatives. Only cannabidiol and CPA (16:0 / 0:0) were relatively low in GUS. PC (34:5), PS (14:1 / 20:2), PA (20:3 / 15:0), linoleic acid DG (22:6 / 22:6 / 0:0), and DG (20:4 / 18:4 / 0:0) were the differential lipid metabolites that contributed significantly to GUS.

[0139] Twenty-three differential lipids and lipid metabolites were screened from roasted beans, such as Figure 22 Of these, RUHP contains 22 different lipid metabolites, while RUS contains only one. Piperidine is the only differential lipid metabolite found in RUS. Apart from ganoderic acid, RUHP generally exhibits higher levels of differential lipid metabolites such as terpene lactones and cucurbitacin E-2-O-glucoside.

[0140] In general, the lipids and lipids in green beans obtained after ultrasonic and ultra-high pressure treatment undergo oxidation and decomposition during roasting, resulting in more differentially volatile compounds. Roasted beans have the highest content of RUHP differentially volatile lipids, leading to increased oil content and high aroma retention during brewing. Therefore, ultra-high pressure treated coffee has a higher sensory score.

[0141] In summary, adding microorganisms during coffee fermentation can shorten fermentation time. Furthermore, the co-fermentation of microorganisms helps increase the secretion of pectinase and cellulase, and increases the number of microorganisms. This may be because *Leuconostoc mesenteroides* and *Lactobacillus plantarum* are lactic acid bacteria that can grow in low-oxygen environments and ferment carbohydrates under anaerobic conditions. Yeast can promote the growth of lactic acid bacteria through autolysis, providing them with essential amino acids and vitamins for development. In addition, the organic acids (such as lactic acid) produced by lactic acid bacteria lower the pH, which is beneficial to yeast growth. The effect of co-inoculation of lactic acid bacteria and yeast on sensory properties may be due to their symbiotic interaction with the entire microbial community. During fermentation, ethanol and carbon dioxide are produced, along with other byproducts including esters, organic acids, and volatile compounds. These flavor compounds all have a positive impact on the sensory properties of coffee.

[0142] In this invention, the drying curves of the three types of coffee beans after ultrasonic and ultra-high pressure drying pretreatment showed significant differences. Compared with the untreated group and the ultrasonically treated group, the ultra-high pressure group had a drying time shortened by 12 hours, which is related to the increased permeability of the coffee cell walls and the accelerated mass and heat transfer process during pretreatment. The weight loss rate and bulk density of the three types of coffee beans differed considerably. The bulk density of the green coffee beans treated with ultra-high pressure was greater than that of the untreated group and the ultrasonically treated group because ultra-high pressure compressed the internal pores of the green coffee beans; however, after roasting, the bulk density of the ultra-high pressure group was less than that of the other two groups, indicating an increased expansion rate. This is because ultra-high pressure treatment altered the cell structure and cell wall composition of the coffee beans, making them more prone to expansion. The weight loss rate of coffee beans treated with ultrasonic ultra-high pressure decreased.

[0143] The contents of polyphenols, flavonoids, reducing sugars, and total sugars in green coffee beans treated with ultrasound and ultra-high pressure (UHAP) all increased. The contents of reducing sugars and total sugars in roasted beans also increased. This is likely because ultrasound and UHAP diffuse substances and microbial metabolites from the coffee fermentation mucus into the coffee seeds, increasing the sugar and flavonoid content in the green coffee beans. Phenolic compounds are important bioactive substances in coffee, responsible for its bitterness and astringency. In this study, ultrasound and UHAP increased the polyphenol content in green coffee beans. While the soluble protein content in UHAP-treated green coffee beans decreased, the soluble protein content in roasted coffee beans increased compared to the other two groups. This is because UHAP denatures proteins in green beans, converting them into smaller amino acids, a finding confirmed in subsequent metabolomics analysis. The increase in total phenols, flavonoids, reducing sugars, and total sugars in the ultrasound- and UHAP-treated samples may be related to increased permeability, leading to the release of certain bioactive substances from cells. Ultrasonic and high-pressure treatment increases the permeability of coffee bean cell walls and cell walls in the fermentation mucus, allowing more sugars and phenols from the fermentation mucus to enter the cells, and more soluble solids to pass through the cell membrane, thus increasing the cell contents of green coffee beans. However, excessively high ultrasonic time, ultrasonic power, high-pressure treatment, and high-pressure treatment duration can reduce the content of these substances in coffee. This is because excessive ultrasonic and high-pressure treatments damage the cell structure, causing cell contents to leak out. Therefore, high-pressure and ultrasonic treatment can enhance the flavor precursors and taste compounds in green coffee beans.

[0144] Volatile flavor components of coffee constitute important sensory characteristics for high-quality coffee brewing. Different post-harvest processing methods lead to differences in the volatile components of coffee. In this study, roasted coffee beans treated with different drying pre-processing methods exhibited different volatile flavors. The content of characteristic volatile flavor compounds in roasted coffee beans treated with ultrasound and ultra-high pressure changed. Compared to the untreated group, the ultrasound and ultra-high pressure treated group showed an increase in pyrazine compounds, with the ultra-high pressure group showing the largest increase. This is because ultrasound and ultra-high pressure treatments increase the content of flavor precursors such as soluble proteins and sugars in green beans. Furthermore, since ultrasound and ultra-high pressure alter the coffee bean structure, the high heat transfer of the coffee bean particles promotes the release of more volatile components from the beans. Therefore, ultrasound and ultra-high pressure treatments can enhance the flavor precursors in green coffee beans, thereby increasing the volatile flavor compounds in roasted coffee.

[0145] Metabolomics is the science of the overall composition and changes of endogenous substances in organisms, enabling a more comprehensive, accurate, and efficient study of differences in metabolite content among samples. This study's metabolic results show that, compared to directly dried green beans, ultrasonically treated green beans exhibit unique differences in metabolites primarily consisting of lipids, heterocyclic organic compounds, and organic acids and their derivatives. Similarly, ultra-high pressure (UHP) green beans, compared to directly dried green beans, show unique differences in metabolites primarily consisting of organic acids and their derivatives, heterocyclic organic compounds, lipids, and benzenes. UHP treatment mainly alters benzene levels in roasted beans. The main reason for these differences is that ultrasonic treatment activates the linoleic acid and α-linolenic acid metabolic pathways in green beans, leading to the upregulation of a large number of lipid metabolites; UHP activates the arginine, tryptophan, and linoleic acid metabolic pathways in green beans, leading to the upregulation of a large number of amino acid and lipid metabolites. Ultrasonic and UHP treatments upregulate flavor precursors in green beans, thus altering the volatile compounds such as pyrazines, aldehydes, and acids in roasted beans.

Claims

1. A method for improving coffee flavor using a combination of biological fermentation and physical methods, characterized in that, The method includes the following steps: Step (1) After washing the fresh coffee cherries and removing the outer skin, a microbial suspension is introduced for fermentation; the microorganisms are a compound of Lactobacillus plantarum, Saccharomyces cerevisiae and Leuconostoc mesenteroides in a 1:1:1 ratio. Step (2) After the fermentation is completed, the coffee beans and fermentation liquid are subjected to ultra-high pressure treatment or ultrasonic treatment; Step (3) Clean the coffee beans after ultra-high pressure or ultrasonic treatment, drain the water completely, and spread them evenly on the drying net for drying.

2. The method for improving coffee flavor using a combination of biological fermentation and physical methods according to claim 1, characterized in that, Step (1) Microbial fermentation time: 38-40 hours; total inoculum size: 1-9 × 10⁻⁶ 7 CFU / g, fermentation temperature 36.5-37.5℃.

3. The method for improving coffee flavor using a combination of biological fermentation and physical methods according to claim 1, characterized in that, The activation process of the microorganisms is as follows: Saccharomyces cerevisiae is inoculated into YPD medium and cultured at 28-32℃ for 20-26h; Lactobacillus plantarum and Leuconostoc mesenteroides are inoculated into MRS broth medium and cultured at 36℃-37.5℃ for 10-15h; after activation for 2-3 generations, centrifuge at 3-5℃ and 9000-11000r / min for 3-8min, wash twice with 0.8%-1% sterile physiological saline, and resuspend in sterile physiological saline for inoculation.

4. The method for improving coffee flavor using a combination of biological fermentation and physical methods according to claim 1, characterized in that, The ultra-high pressure in step (2) is 250-300 MPa; the time is 20-35 min.

5. The method for improving coffee flavor using a combination of biological fermentation and physical methods according to claim 1, characterized in that, Step (2) Ultrasonic temperature 24℃-26℃, ultrasonic power 450-500W, ultrasonic time 8-12min.

6. The method for improving coffee flavor using a combination of biological fermentation and physical methods according to claim 1, characterized in that, Step (3) The drying temperature is 36.5-37.5℃, and the coffee is dried until the moisture content reaches 10-12%.

Images