Methods for directional regulation of flavor in anaerobic fermentation of fresh coffee cherries

Abstract

This invention relates to the field of coffee processing technology, specifically disclosing a method for directional regulation of flavor during anaerobic fermentation of fresh coffee fruit. Sodium alginate and chitosan are used to construct single-layer, double-layer, and triple-layer microcapsules to encapsulate lactic acid bacteria, yeast, and acetic acid bacteria, respectively. These three types of microcapsules are then placed into a sealed anaerobic fermentation container, where the bacterial strains are released sequentially as the pH naturally decreases, achieving metabolic relay. The entire process requires no opening of the container, thus balancing flavor enhancement and food safety.

Description

Technical Field

[0001] This invention relates to the field of coffee processing technology, specifically to a method for targeted flavor regulation during anaerobic fermentation of fresh coffee cherries. Background Technology

[0002] Coffee is the third most consumed beverage globally, after water and tea, and its flavor and quality directly determine its market value. In the post-harvest processing of coffee, fermentation is a crucial step influencing flavor formation. In recent years, anaerobic fermentation technology has received widespread attention due to its ability to significantly enhance the sensory complexity of coffee. Wang et al., in their review, systematically summarized six major factors affecting the volatile flavor compounds in coffee, pointing out the importance of inoculating specific fermenting agent strains (…). starter culture ) has become an important research direction for improving the controllability of coffee flavor (Wang X et al., J Sci Food Agric, 2022, 102:1341-1352). Research by Zhao et al. has confirmed that *Saccharomyces cerevisiae* (Saccharomyces cerevisiae) Wickerhamomyces anomalus ) and lactic acid bacteria ( Enterococcus mundtii Co-inoculation with Pichia membranifaciens significantly increased the content and types of esters, alcohols, and ketones in coffee beans, and the total cupping score of the co-inoculated group was significantly higher than that of the naturally fermented control group (Zhao N et al., J Sci FoodAgric, 2024, 104:9137-9150). Shen et al. further found that enhanced fermentation with Pichia membranifaciens can stabilize coffee flavor quality by inhibiting the growth of other microorganisms (Shen X et al., Food Sci Nutr, 2025, 13(7)). Martins et al. recently screened several non-Fermented Pichia membranifaciens strains from Arabica and Cornelian coffee. Saccharomyces Yeast is considered a potential candidate starter culture (Martins PM et al., J Food Sci, 2025, 90(7)). However, all of the above studies used direct inoculation with free cells, and there are no reports on the application of microencapsulated starter cultures to anaerobic fermentation of coffee.

[0003] In the field of microbial microencapsulation technology, a few studies have applied microencapsulated starter cultures to fermented foods. Criollo Nuñez et al. microencapsulated *Pichia kudriavzevii* yeast using alginate extrusion technology (activity 3.8 × 10⁻⁶). 8CFU / g was used in the fermentation of cocoa CCN51, and it was confirmed that a 2% microencapsulated inoculum significantly promoted the accumulation of fruit and floral aroma compounds (2,3-butanediol, phenylethanol, etc.) in the cocoa liquor (Criollo Nuñez J et al., JSci Food Agric, 2023). Phan Van et al. reported a freeze-dried microencapsulated kombucha starter (containing CFU / g) and applied it to the fermentation of cocoa CCN51. K. sugar-eating , L. brevis and S. cerevisiae This study confirmed that freeze-dried strains maintained stable fermentation capacity after three months of storage at 30°C and 4°C (Phan Van T et al., J Sci Food Agric, 2024). However, the above studies all involved encapsulating all strains or freeze-dried mixed powders with a single wall material, and did not involve selectively releasing different strains under different conditions through differentiated wall material structures.

[0004] In the field of sodium alginate-chitosan microcapsule systems, this material combination has a relatively mature research foundation in the pharmaceutical and probiotic delivery fields. Li et al. studied the inflammatory response and cell adhesion behavior after transplantation of sodium alginate-chitosan-sodium alginate (ACA) trilayer microcapsules (Li S et al., J Biomed Mater Res A, 2015, 103:2333-2343). Zheng et al. systematically studied the influence mechanism of high / low degree of deacetylation chitosan sequential assembly on microcapsule membrane formation and found that the two types of chitosan play a synergistic role in membrane performance (Zheng G et al., J Biomed Mater Res A, 2016, 104:257-263). Hua and Li used sodium alginate-chitosan microcapsules to encapsulate Lactobacillus plantarum and proved that the chitosan coating can effectively prevent the immediate release of probiotics in simulated gastric juice (Hua Q & Li D, J Food Sci, 2024, 89:8066-8076). The genipin-crosslinked sodium alginate-chitosan (GCAC) microcapsules developed by Chen et al. exhibited excellent structural stability in a simulated gastrointestinal environment (Chen H et al., Int J Polym Sci, 2010). However, the aforementioned pH-responsive microcapsule studies are all geared towards gastrointestinal delivery scenarios, utilizing the pH change from low to high (gastric acid pH 1.5–3.5 → intestinal pH 6.0–7.4) to achieve cell protection and targeted release, which is completely opposite to the pH change from high to low (initial pH 5.5–6.0 → final pH 3.5–4.0) during food fermentation.

[0005] Based on the above analysis, the existing technology faces the following three unresolved core technological bottlenecks: First, there is a fundamental contradiction between the inoculation method and the maintenance of anaerobic digestion. Both simultaneous, one-time inoculation and phased artificial inoculation have inherent drawbacks. During simultaneous inoculation, all three strains are activated at the same time. Lactic acid bacteria, due to their short generation time (approximately 30-45 minutes) and rapid acid production rate, will rapidly lower the pH to below 4.2 within the first 12 hours of fermentation, inhibiting the metabolic activity of yeast and acetic acid bacteria that have not yet fully proliferated, leading to an imbalance where lactic acid bacteria dominate the fermentation window. Zhao et al.'s research also observed significant differences in the microbial community dynamics between the inoculated and uninoculated groups, and that the metabolites of lactic acid bacteria and yeast in the co-inoculated group exhibited a complex interactive pattern (Zhao N et al., 2024, ibid.). While phased artificial inoculation can address the issue of microbial metabolic timing to some extent, each opening of the tank inevitably introduces external air, disrupting the anaerobic environment and introducing the risk of ochratoxin A (OTA) contamination. This contradiction has not yet been effectively resolved in existing literature.

[0006] Second, there is a lack of coupling mechanism between the pH change window of anaerobic fermentation of coffee cherries and the metabolic needs of microorganisms. During the fermentation of coffee cherries, the pH gradually decreases from the initial 5.5–6.0 to the final 3.5–4.0, exhibiting a phased change characteristic: in the initial stage (0 h–18 h), the pH decreases from 5.8 to 4.5–5.0; in the middle stage (18 h–48 h), the pH slowly decreases from 4.5 to 3.8–4.2; and in the final stage (48 h–96 h), the pH tends to stabilize at 3.5–4.0. This natural pH gradient precisely corresponds to the optimal metabolic window of each of the three types of microorganisms, but existing free cell inoculation methods cannot utilize this natural gradient signal to achieve automatic sequential release of the strain.

[0007] Third, current applications of microencapsulation technology in food fermentation primarily focus on the gastrointestinal targeted delivery of probiotics, utilizing pH-responsive materials to protect probiotics safely through the acidic gastric environment (pH 1.5–3.5) and release them into the intestinal environment (pH 6.0–7.4). As mentioned in the aforementioned literature, the studies by Chen et al., Hua and Li et al., and Arora et al. all focused on delivery protection scenarios from low to high pH (Arora S et al., Mediators Inflamm, 2014). The pH change direction in this application scenario is from low to high, completely opposite to the pH change direction from high to low during food fermentation. Shifting pH-responsive microcapsules from the protection-release mode of gastrointestinal delivery to the time-relay mode of food fermentation is an unexplored cross-disciplinary technology transfer direction. In particular, the use of differentiated designs with different numbers of wall material layers (single / double / triple layers) to achieve gradient release of different microorganisms during the same pH decrease process has not been reported in existing literature. Summary of the Invention

[0008] To address the aforementioned technical problems, this invention provides a method for targeted flavor regulation during anaerobic fermentation of fresh coffee cherries. The core innovation of this method lies in the creative transfer of the mature sodium alginate-chitosan pH-responsive microcapsule system from the pharmaceutical and probiotic delivery fields to the anaerobic fermentation scenario of coffee. Through differentiated design of the number of wall layers (single-layer sodium alginate / double-layer sodium alginate-chitosan / triple-layer sodium alginate-chitosan-sodium alginate), the method utilizes the natural pH decrease during fermentation to automatically achieve the timed sequential release of three functional strains—lactic acid bacteria (pH 5.5–6.0 release), yeast (pH 4.5–5.0 release), and acetic acid bacteria (pH 3.8–4.5 release)—within their respective optimal pH windows. The three types of microcapsules are mixed and added to a sealed anaerobic fermentation container at the start of fermentation, eliminating the need for opening the container throughout the process. This fundamentally solves the contradiction between inoculation timing regulation and maintaining the anaerobic environment, while simultaneously achieving a triple synergistic effect of improved flavor complexity, improved batch consistency, and elimination of OTA contamination risks. Compared to cocoa fermentation microencapsulation schemes using a single yeast and a single wall material, such as Criollo Nuñez, this invention achieves a novel inoculation mode of one-time input and three-level gradient release through differentiated wall structures, which has a fundamental technical difference and unexpected synergistic effect.

[0009] The specific technical solution of the present invention is as follows: A method for targeted flavor regulation during anaerobic fermentation of fresh coffee cherries includes the following steps: S1, Preparation of Microencapsulated Composite Bacterial Agents: Lactic acid bacteria, yeast, and acetic acid bacteria were encapsulated in microcapsules with different wall materials, respectively, to obtain three types of microencapsulated bacterial agents with differentiated pH-responsive release characteristics. Among them, the first type of microcapsule used sodium alginate as a single-layer wall material, with sodium alginate in Ca... 2+In the presence of certain microcapsules, a calcium alginate gel network is formed through ionic cross-linking. In a weakly acidic to neutral environment (pH 5.5–6.0), the carboxyl groups of this gel network are gradually protonated, causing the calcium ion bridges to break, leading to gel swelling and disintegration, and releasing lactic acid bacteria first. The second type of microcapsule is encapsulated with a sodium alginate-chitosan bilayer wall material. The inner layer of calcium alginate gel provides the initial structural framework, while the outer layer of chitosan forms a polyelectrolyte complex layer through electrostatic interactions between amino groups and the carboxyl groups of sodium alginate. Chitosan has a pKa of approximately 6.3. In an acidic environment, the degree of amino protonation increases with decreasing pH, leading to enhanced electrostatic repulsion between molecular chains and swelling of the chitosan layer. When the swelling force exceeds the interlayer bonding force, the integrity of the wall material is compromised. Due to the additional barrier effect of the chitosan layer, the second type of microcapsule requires the pH to drop to 4.5–5.0 before releasing yeast. The third type of microcapsule is encapsulated by a three-layer wall material consisting of sodium alginate, chitosan, and sodium alginate again. The outermost layer of sodium alginate provides initial protection and delays the erosion of the inner layer by the acidic environment. The middle chitosan layer expands in the acidic environment but is constrained by the inner and outer sodium alginate layers. Only when the pH further decreases to 3.8–4.5, the calcium alginate gel networks of the inner and outer layers are fully disrupted, and the expansion force of the middle chitosan layer overcomes the interlayer constraint. The synergistic disintegration of the three wall materials releases acetic acid bacteria. By controlling the concentration, thickness, and degree of cross-linking of each wall material, the pH response threshold and release kinetics of the three types of microcapsules can be precisely adjusted.

[0010] S2, Raw Material Pre-treatment: Select fresh coffee cherries with a maturity of 80% or higher, and use a combination of flotation and color grading methods for screening. Flotation involves immersing the fresh coffee cherries in clean water to remove cherries with abnormal density (including empty cherries, severely insect-infested cherries, and shriveled cherries) that float to the surface. Color grading is based on the color of the peel, retaining cherries with a deep red to purplish-red peel and removing unripe green and yellow cherries, as well as overripe brown and black cherries and diseased cherries. The screened coffee cherries are rinsed 2-3 times with clean water to remove surface dirt and impurities, and then drained until there are no visible water droplets on the surface.

[0011] S3, Establishment of Anaerobic Environment and Single-Stage Inoculation: Place the fresh coffee cherries obtained in step S2 into a sealed fermentation container, filling it to 60%–80% of its volume, leaving space at the top to accommodate the gas and liquid expansion produced during fermentation. Introduce food-grade nitrogen and / or carbon dioxide into the sealed fermentation container to replace the air at a rate of 0.5 L / min–2.0 L / min for 10–30 minutes, until the oxygen concentration in the container drops below 1% (monitored using a portable oxygen detector). Introducing nitrogen establishes an inert atmosphere to remove oxygen, while introducing carbon dioxide provides a carbonation effect to promote permeability of the cherries. Then, add the three types of microcapsule inoculants obtained in step S1 into the sealed fermentation container all at once, premixing them thoroughly before addition. After addition, gently agitate the container to distribute the microcapsules evenly among the coffee cherries, then seal it. The sealed fermentation vessel is equipped with a one-way exhaust valve with an opening pressure of 0.02 MPa to 0.05 MPa, allowing CO2 generated during fermentation to escape while preventing external air from entering and maintaining an anaerobic environment inside the vessel. The total input of the three types of microencapsulated inoculants is 0.5% to 3.0% of the weight of fresh coffee cherries, with the mass ratio of the first, second, and third types of microcapsules being (3–5):(2–3):1. This mass ratio is set based on the differences in the optimal initial inoculation density of the three strains and the encapsulation density of the microcapsules.

[0012] S4, pH-driven sequential release of strains and temperature-controlled fermentation: Sealed anaerobic fermentation was conducted at 18℃–25℃ for a total fermentation time of 48 h–96 h. After fermentation started, the organic acids naturally present in the coffee pulp, such as citric acid, malic acid, and quinic acid, maintained the initial pH of the fermentation broth within the range of 5.5–6.0, while the small amount of natural microorganisms remaining on the surface of the pericarp began initial metabolism. Under these pH conditions, the sodium alginate monolayer wall material of the first type of microcapsules began to swell and disintegrate, releasing lactic acid bacteria into the fermentation broth. The released lactic acid bacteria used glucose and fructose from the pulp as carbon sources, rapidly converting sugars into lactic acid through homolactic fermentation, causing the pH of the fermentation broth to decrease from 5.5–6.0 to 4.5–5.0 within 0 h–18 h. During this process, the viable count of lactic acid bacteria increased from approximately [missing information] at the initial release. CFU / mL proliferated to CFU / mL, lactic acid accumulation reaches 2.0 g / L~5.0 g / L.

[0013] When the pH drops to the range of 4.5–5.0, the chitosan layer of the second type of microcapsules undergoes protonation and expansion to a critical value, leading to the synergistic disintegration of the bilayer wall material and the release of yeast. Yeast initiates ethanol fermentation in a weakly acidic anaerobic environment, converting residual glucose and fructose into ethanol, glycerol, and various higher alcohols via glycolysis and pyruvate decarboxylation. Important flavor precursors produced by yeast metabolism include: phenethyl alcohol (rose aroma) generated from phenylalanine via the Ehrlich pathway through deamination, oxidation, and decarboxylation; ethyl acetate (fruity aroma) synthesized from ethanol and acetyl-CoA under the catalysis of acyltransferases; and linalool (citrus aroma) synthesized from isopentenyl pyrophosphate precursors via the mevalonate pathway. During the yeast fermentation stage (18 h–48 h), the pH further decreases from 4.5–5.0 to 3.8–4.5, the ethanol content accumulates to 0.5%–2.5% (v / v), and the phenethyl alcohol content reaches 10 μg / g–30 μg / g.

[0014] When the pH drops to the range of 3.8–4.5, the three-layer wall material of the third type of microcapsule completely disintegrates, releasing acetic acid bacteria. Acetic acid bacteria ( Acetobacter pasteurianus Acetic acid bacteria are obligate aerobic bacteria, but in the anaerobic fermentation system of this invention, the one-way exhaust valve allows trace amounts of oxygen to dissolve into the fermentation broth surface via back diffusion during CO2 discharge, forming a microaerophilic environment (dissolved oxygen concentration approximately 0.1 mg / L to 0.5 mg / L). Although the proliferation rate of acetic acid bacteria is limited under this microaerophilic environment, they can still oxidize some ethanol to acetaldehyde and acetic acid through alcohol dehydrogenase and aldehyde dehydrogenase. Acetic acid and ethanol undergo a non-enzymatic esterification reaction under acidic conditions to produce ethyl acetate. The esterases of acetic acid bacteria themselves also catalyze the synthesis of various ester compounds, including phenethyl acetate (honey floral aroma) and ethyl hexanoate (tropical fruit aroma). During the metabolic phase of acetic acid bacteria (after 48 h), the acetic acid content in the fermentation broth reaches 0.5 g / L to 2.0 g / L, and the ethyl acetate content is significantly increased.

[0015] A segmented temperature control strategy was adopted during fermentation: the temperature was controlled at 18℃~20℃ for the first 24 hours of fermentation to match the optimal growth temperature range of lactic acid bacteria. The lower temperature is conducive to the stable acid production of lactic acid bacteria without excessive consumption of reducing sugars. The temperature was controlled at 20℃~23℃ for the first 24 hours to 48 hours of fermentation to match the optimal fermentation temperature range of yeast. The moderate temperature is conducive to the balanced production of ethanol and the accumulation of higher alcohols and ester precursors by yeast. After 48 hours of fermentation, the temperature was controlled at 23℃~25℃ to match the optimal metabolic temperature range of acetic acid bacteria. The higher temperature is conducive to the oxidative metabolic activity and esterification reaction rate of acetic acid bacteria.

[0016] S5, Determination of Fermentation Endpoint: Samples are taken every 12 hours to measure the pH (using a precision pH meter, accuracy ±0.01), titratable acidity (calculated as tartaric acid, using NaOH titration), reducing sugar content (determined using the DNS method), and ethanol content (using an alcohol meter or gas chromatography). Fermentation is considered complete when the pH drops to 3.5–4.0, the titratable acidity rises to 0.8–1.2 g / 100 mL, the reducing sugar content drops to below 30%–50% of its initial value, and the ethanol content is 0.5%–3.0% (v / v). If any one of the above four indicators fails to meet the standard but the other three do, the fermentation time can be extended by 12–24 hours before retesting.

[0017] S6, Post-processing: The fermented coffee cherries are transferred to a degumming and depeeling device for mechanical degumming to remove the peel and pectin layer. After degumming, the coffee beans in their shells are soaked in clean water and washed 1-3 times, each time for 5-15 minutes. During washing, the beans are gently stirred to thoroughly remove residual microcapsule wall fragments (sodium alginate and chitosan degradation products) and organic acids and alcohols from the fermentation residue. The washing water temperature is controlled at room temperature (15℃-30℃) to avoid excessive loss of flavor compounds due to high temperatures. The beans are then dried, either by sun drying or mechanical drying. Sun drying takes 7-14 days. During sun drying, the beans in their shells are evenly spread on the drying bed, with a thickness not exceeding 3 cm, and turned every 2-4 hours to ensure even drying. They should be covered for moisture protection at night and on rainy days. Mechanical drying takes place at 35℃-45℃ for 18-36 hours, with adequate ventilation maintained during the drying process. The coffee beans are dried to a moisture content of 10%–12% (measured using a grain moisture meter) to obtain fermented green coffee beans. The resulting green coffee beans are stored in a cool, dry place at a temperature not exceeding 25°C and a relative humidity not exceeding 60%.

[0018] The beneficial effects of this invention include: (1) A novel inoculation mode of one-time input and automatic time-sequential release has been realized. Through the differentiated wall material design of pH-responsive microcapsules, the driving force of the natural decrease of pH during fermentation is utilized to automatically realize the timed release of three functional strains within their respective optimal pH windows. No tank opening operation is required throughout the process, which fundamentally eliminates the risk of anaerobic destruction and exogenous contamination caused by staged inoculation.

[0019] (2) Food safety is significantly improved. On the one hand, the microcapsule wall materials sodium alginate and chitosan are both food additives permitted by the Chinese GB2760 standard, ensuring their safety. Moreover, wall material fragments are completely removed during the post-processing cleaning step and do not remain in the final product. On the other hand, the fully sealed anaerobic operation eliminates the risk of contamination by toxin-producing fungi such as Aspergillus introduced by repeated opening of the can, effectively reducing the probability of ochratoxin A (OTA) production. In the examples, the OTA detection amount was below the detection limit of 0.2 μg / kg, which is far below the EU limit of 5 μg / kg.

[0020] (3) Flavor complexity and batch consistency are improved simultaneously. The organic acids (lactic acid), alcohols (ethanol, phenylethanol, linalool), esters (ethyl acetate, phenylethyl acetate, ethyl hexanoate), and aldehydes and ketones produced by the sequential metabolism of the three microorganisms overlap, giving coffee a multi-dimensional complex flavor of fruit, floral, and wine aromas. After verification by 5 independent and repeated experiments, the batch standard deviation of the SCA cupping score decreased from ±1.7 points for traditional natural fermentation and ±1.3 points for staged artificial inoculation to ±0.4 points.

[0021] (4) Simple to operate and suitable for promotion in production areas. Compared with the phased inoculation scheme, which requires multiple opening operations, strict time control and professional microbiology knowledge, the microencapsulated bacterial agent of the present invention can be transported to the production area as a pre-made product under cold chain conditions. When using it, it can be directly put into the fermentation container. The operation is no different from adding ordinary fermentation aids, which greatly reduces the technical threshold and operational complexity for farmers in the production area.

[0022] (5) The microcapsule wall material provides effective protection for the bacterial strains during storage and transportation. After 6 months of storage at 4°C, the survival rates of the strains in the three types of freeze-dried microcapsules remained above 92%, 88%, and 82%, respectively; after 6 months of storage at 25°C, the survival rates remained above 85%, 80%, and 75%, respectively. In contrast, the survival rate of free bacterial powder after 6 months of storage under the same conditions is usually less than 50%. The wall material of the microcapsules forms a dense protective matrix during the freeze-drying process, effectively isolating the bacterial cells from damage caused by oxygen and moisture. Attached Figure Description

[0023] Figure 1 The figures show the pH change curves during the fermentation process in the examples and comparative examples.

[0024] Figure 2 This is a comparison chart of the content of key volatile flavor compounds in the examples and comparative examples. Detailed Implementation

[0025] The present invention will be further described below with reference to specific embodiments, but the scope of protection of the present invention is not limited to the following embodiments. The strains used in the following embodiments are all standard strains preserved at the China Industrial Microbial Culture Collection Center (CICC). Food-grade sodium alginate was purchased from Qingdao Mingyue Seaweed Group, and food-grade chitosan was purchased from Zhejiang Jinke Biochemical Co., Ltd. The degree of deacetylation was provided by the supplier. The fresh coffee cherries were all harvested from the Lujiangba Coffee Plantation in Longyang District, Baoshan City, Yunnan Province.

[0026] Example 1: Please refer to Figure 1-2 A method for targeted flavor regulation during anaerobic fermentation of fresh coffee cherries, comprising the following steps: S1, Preparation of microencapsulated compound bacterial agent: (1) Lactic acid bacteria culture: Lactobacillus plantarum ( Lactiplantibacillus plantarum CICC 6002 Inoculate from cryopreserved glycerol tubes into MRS liquid medium and incubate under anaerobic conditions at 37°C for 18 h until the logarithmic growth phase. Once the concentration reaches 1.8–2.2, centrifuge (8000 rpm, 10 min, 4℃) to collect the bacterial cells, wash twice with sterile physiological saline, and resuspend to a concentration of [missing value]. CFU / mL bacterial suspension.

[0027] (2) Preparation of the first type of microcapsules: The above-mentioned lactic acid bacteria suspension and a 2% (w / w) food-grade sodium alginate solution (pre-sterilized at 121°C for 15 min) were mixed evenly at a volume ratio of 1:9 to obtain a sodium alginate mixture containing bacteria. Using a peristaltic pump with a 0.8 mm inner diameter stainless steel injection needle, the mixture was dripped into a 3% (w / w) calcium chloride solution at a constant flow rate (2 mL / min). The droplets were placed on the Ca... 2+ Under the action of [unclear], microcapsules were formed by instantaneous gelation. They were then cured in calcium chloride solution at room temperature (25℃) for 20 min to allow the gel network to fully cross-link. The microcapsules were collected using a 200-mesh stainless steel sieve, washed three times with sterile water to remove residual CaCl2, and then pre-frozen at -80℃ for 12 h before being freeze-dried for 24 h (cold trap temperature -50℃, vacuum degree 10 Pa). White spherical type I microcapsules with a particle size of 1.0 mm to 1.5 mm were obtained, and the encapsulation density was determined by plate counting method. CFU / g microcapsules with an encapsulation efficiency of 89%.

[0028] (3) Yeast culture: Brewing yeast ( Saccharomyces cerevisiae CICC 1002 Inoculate into YPD liquid medium and incubate at 30°C with shaking at 150 rpm for 16 h until the logarithmic growth phase (OD200). 600 (To reach a concentration of 1.5–1.8), collect the bacterial cells by centrifugation, wash twice with sterile physiological saline, and resuspend to a concentration of 10. 9 CFU / mL bacterial suspension.

[0029] (4) Preparation of the second type of microcapsules: After preparing sodium alginate gel microspheres using the same method as the first type of microcapsules, the solidified microspheres were immersed in a chitosan solution (90% deacetylation, dissolved in 1% acetic acid and adjusted to pH 5.5 with NaOH) with a mass concentration of 0.5% for 10 min to coat the microspheres. During this time, the microspheres were gently stirred to ensure that the chitosan was uniformly adsorbed onto the surface of the microspheres. After removal, the microspheres were washed twice with sterile water to remove unbound chitosan. The freeze-drying conditions were the same as above. Pale yellow spherical second type of microcapsules with a particle size of 1.2 mm to 1.8 mm and an encapsulation density of 3 × 10⁻⁶ were obtained. 8 CFU / g microcapsules with an encapsulation efficiency of 85%.

[0030] (5) Acetic acid bacteria culture: The culture of *Acetobacter pasteurellium* (…) Acetobacter pasteurianus CICC 20056 The cells were inoculated into GYC liquid medium containing 3% (v / v) ethanol and cultured at 30°C and 200 rpm for 24 h until the logarithmic growth phase. The cells were then collected by centrifugation, washed twice with sterile physiological saline, and resuspended to a concentration of 10. 8 CFU / mL bacterial suspension.

[0031] (6) Preparation of the third type of microcapsules: After preparing sodium alginate-chitosan bilayer microspheres using the same method as the second type of microcapsules, the bilayer microspheres were again immersed in a 1.0% sodium alginate solution for 10 min for coating, followed by a second solidification in a 2% calcium chloride solution for 15 min. The microspheres were then removed, washed twice with sterile water, and lyophilized under the same conditions. Pale yellow spherical third-type microcapsules with a particle size of 1.5 mm to 2.2 mm were obtained, with an encapsulation density of [missing information]. CFU / g microcapsules with an encapsulation efficiency of 78%.

[0032] S2, Raw Material Pretreatment: 20 kg of fresh Catimor coffee cherries, harvested in December 2024 from the Baoshan production area of ​​Yunnan Province, with a maturity of over 85%, were selected at an altitude of 1200 m–1400 m. Approximately 0.8 kg of cherries with abnormal density were removed by flotation, and approximately 1.2 kg of immature and overripe cherries were removed by color grading, ultimately yielding 18 kg of qualified fresh cherries. The cherries were rinsed twice with clean water and then drained for 30 min.

[0033] S3, Anaerobic Environment Establishment and Single-Stage Inoculation: 18 kg of fresh coffee cherries were placed into a 30 L food-grade 304 stainless steel sealed fermentation tank (filling rate approximately 72%). High-purity food-grade nitrogen (purity ≥99.9%) was introduced at a flow rate of 1.0 L / min through the bottom inlet for 20 min. The oxygen concentration inside the tank was monitored using a portable oxygen detector until it dropped below 0.5%. 200 g of type I microcapsules, 120 g of type II microcapsules, and 50 g of type III microcapsules (mass ratio approximately 4:2.4:1, total input approximately 2.1% of the fresh cherries' weight) were weighed, premixed thoroughly, and then added to the fermentation tank all at once. The tank was manually and gently turned 3–5 times to ensure even distribution of the microcapsules among the cherries, and then sealed. A food-grade silicone one-way exhaust valve with an opening pressure of 0.03 MPa was installed on the top of the fermentation tank.

[0034] S4, pH-driven sequential release of strains and temperature-controlled fermentation: The sealed fermenter was placed in a programmable temperature-controlled incubator, and the fermentation process monitoring results are as follows. At 0 h of fermentation, the initial pH was measured to be 5.8, the reducing sugar content was 12.5 g / 100 mL, and the titratable acidity was 0.35 g / 100 mL.

[0035] Fermentation period 0-6 hours (lactic acid bacteria release initiation stage): The first type of microcapsule wall material begins to swell, and lactic acid bacteria are gradually released into the fermentation broth. At 6 hours, a sample was taken and the pH was measured to be 5.4, and the viable lactic acid bacteria count was 2.3 × 10⁻⁶. 6 CFU / mL (proliferated from the initial release), the fermentation broth begins to have a slightly sour taste.

[0036] Fermentation period 6-18 hours (the stage dominated by lactic acid bacteria for acid production): Lactic acid bacteria proliferate rapidly and produce a large amount of acid. At 12 hours, the pH drops to 5.1, and the viable count of lactic acid bacteria reaches 8.5 × 10⁻⁶. 7 The concentration of CFU / mL and the lactic acid content was 3.2 g / L. After 18 h, the pH dropped to 4.7, the lactic acid content was 5.8 g / L, and the reducing sugar content dropped to 9.1 g / 100 mL. At this point, the second type of microcapsule wall material began to show obvious signs of swelling.

[0037] Fermentation period 18-48 hours (yeast-dominated fermentation stage): The second type of microcapsules began to disintegrate and release yeast around 18 hours. At 24 hours, the pH dropped to 4.3, and the viable yeast count reached 2.1 × 10⁻⁶. 6The fermentation broth contained 0.3% (v / v) CFU / mL of ethanol and 5.2 μg / g of phenylethanol. At 36 h, the pH dropped to 4.0, the ethanol content rose to 1.1% (v / v), the phenylethanol content reached 14.8 μg / g, and the linalool content was 8.5 μg / g. At this point, the third type of microcapsule wall material began to disintegrate. At 48 h, the pH dropped to 3.7, the ethanol content was 1.6% (v / v), and the yeast viable count reached a peak of 5.8 × 10⁻⁶. 6 CFU / mL.

[0038] Fermentation period 48-72 h (acetic acid bacteria modification stage): Acetic acid bacteria initiate metabolism under microaerophilic conditions. At 60 h, the viable count of acetic acid bacteria reaches 6.8 × 10⁻⁶. 4 The concentration of CFU / mL was 0.8 g / L for acetic acid, and the ethyl acetate concentration significantly increased to 22.5 μg / g. At 72 h, the pH stabilized at 3.6, the ethanol concentration was 1.8% (v / v), and the acetic acid concentration was 1.2 g / L. Segmented temperature control was used throughout the fermentation process: 19℃ from 0 h to 24 h, 22℃ from 24 h to 48 h, and 24℃ from 48 h to 72 h.

[0039] S5, Fermentation endpoint determination: After 72 h of fermentation, the following indicators were obtained: pH 3.6, titratable acidity 1.05 g / 100 mL, reducing sugar content reduced to 38% of the initial value (4.75 g / 100 mL), and ethanol content 1.8% (v / v). All four indicators met the endpoint determination criteria, and the fermentation was determined to have reached the endpoint.

[0040] S6, Post-processing: The fresh coffee cherries in the fermentation tank are transferred to a degumming machine for processing. After degumming, the husk-on beans are soaked in clean water and washed twice (10 minutes each time) to remove residual microcapsule wall material fragments and fermentation residue. The washed husk-on beans are evenly spread on an African bed drying rack (approximately 2.5 cm thick) and sun-dried for 10 days (turned over every 3 hours, covered with a rainproof cloth at night) until the moisture content reaches 11.2%, yielding approximately 3.2 kg of fermented green coffee beans.

[0041] Flavor analysis results: GC-MS (Agilent 7890B-5977B) analysis revealed 91 volatile flavor compounds in the green coffee beans obtained in Example 1. Esters accounted for 34.2% (31 compounds), mainly including ethyl acetate (29.3 μg / g), phenethyl acetate (8.7 μg / g), ethyl hexanoate (5.2 μg / g), and ethyl octanoate (3.1 μg / g); alcohols accounted for 22.5% (20 compounds), including phenylethanol (23.5 μg / g), linalool (16.1 μg / g), and isoamyl alcohol (8.3 μg / g); organic acids accounted for 15.8% (14 compounds), with lactic acid being the most abundant. The SCA cupping score was 86.2 points (blind tasting by a Q-Grader certified taster), with flavor descriptions of tropical fruit, jasmine, red wine, and dark chocolate, bright acidity, and medium-to-high body. Food safety testing: Ochratoxin A (OTA) was not detected (<0.2 μg / kg detection limit), coliform bacteria <10 CFU / g, and total mold and yeast count <100 CFU / g, all of which meet the requirements of GB 15193 food safety standard.

[0042] Example 2: The difference from Example 1 lies in the adjustment of microcapsule wall material parameters, strain replacement, and fermentation temperature scheme; the remaining steps and operating conditions are the same as in Example 1. In S1, the first type of microcapsules used sodium alginate with a mass concentration of 1.5% (the lower concentration resulted in a loose gel network and faster swelling and disintegration); the second type of microcapsules used chitosan with a mass concentration of 0.3% and an encapsulation time of 15 min (the chitosan layer was thinner, and the pH response threshold shifted upward to 4.8–5.2); the outer layer of the third type of microcapsules had a sodium alginate mass concentration of 1.5% and an encapsulation time of 15 min. Lactobacillus fermentum was used instead of Lactobacillus fermentum. Limosilactobacillus fermentum CICC 6045 The yeast strain was changed to Kluyveromycin ( Kluyveromyces marxianus CICC 1910 In S3, food-grade carbon dioxide was used instead of nitrogen for gas replacement, with an aeration rate of 1.5 L / min and an aeration time of 15 min. The mass ratio of the three types of microcapsules was adjusted to 3:2:1 (total input was 1.5% of the fresh fruit mass). In S4, a constant temperature fermentation scheme of 22℃ was adopted (without segmented temperature control), and the total fermentation time was 84 h. Due to the low chitosan concentration, the yeast release time was advanced to 14 h of fermentation. Detection results: 82 volatile flavor compounds were detected, with esters accounting for 29.5%, ethyl acetate 25.1 μg / g, phenylethanol 19.8 μg / g, and linalool 13.5 μg / g. The SCA cupping score was 84.5 points, with a flavor description of citrus, honey, and nuts, and a mild acidity. OTA was not detected.

[0043] Example 3: The difference from Example 1 lies in the use of mixed gas replacement, optimization of wall material parameters, and inoculation with mixed strains; the remaining steps and operating conditions are the same as in Example 1. In S1, the first type of microcapsules used sodium alginate with a mass concentration of 2.5%; the second type of microcapsules used chitosan with a mass concentration of 0.8% (degree of deacetylation of 85%) and an encapsulation time of 20 min (the chitosan layer was relatively thick, and the pH response threshold was precisely controlled between 4.5 and 4.8); the outer layer of the third type of microcapsules had a sodium alginate mass concentration of 2.0% and an encapsulation time of 15 min (the pH response threshold was controlled between 3.8 and 4.2). Lactic acid bacteria were a 1:1 mass ratio of *Lactobacillus plantarum* and *Lactobacillus fermentum* (balancing acid production rate and flavor diversity), and yeast was a 2:1 mass ratio of *Saccharomyces cerevisiae* and *Kluyveromyces oryzae* (balancing ethanol yield and types of higher alcohols). In S3, nitrogen gas was first introduced (0.8 L / min, 10 min) to expel most of the air, followed by carbon dioxide (0.5 L / min, 10 min) to provide carbonation and enhance the permeability of the fruit peel. The mass ratio of the three types of microcapsules was 5:3:1 (total input was 2.5% of the fresh fruit mass). In S4, temperature was controlled in stages: 18℃ from 0 h to 24 h, 21℃ from 24 h to 48 h, and 25℃ from 48 h to 72 h, with a total fermentation time of 72 h. Detection results: 96 volatile flavor compounds were found (the highest among all examples and comparative examples), with esters accounting for 37.5%, ethyl acetate 32.6 μg / g, phenylethanol 26.8 μg / g, and linalool 18.3 μg / g. The SCA cupping score was 87.8 points (the highest among all examples and comparative examples), with flavor descriptions of passion fruit, rose, red wine, and cocoa, with lively acidity and rich layers. OTA was not detected.

[0044] Comparative Example 1 (Traditional Natural Fermentation) 18 kg of fresh coffee cherries from the same batch as in Example 1 were placed in an open 60 L plastic container without any gas replacement or inoculation with any microbial agents. The container opening was covered with gauze to prevent insect infestation. Fermentation was carried out for 96 h at ambient temperature (20℃~28℃, with a diurnal temperature range of approximately 8℃). During fermentation, the natural microbial community spontaneously initiated fermentation. Due to the uncontrollable diurnal temperature range and microbial community composition, the fermentation process varied significantly between different batches. After dehulling and degumming, the cherries were sun-dried. Test results: 58 volatile flavor compounds were detected, with esters accounting for 13.1%, ethyl acetate 9.2 μg / g, phenylethanol 7.3 μg / g, and linalool 5.8 μg / g. The SCA cupping score was 79.8 points, with a flavor description of nutty and slightly fermented. The detected level of ochratoxin A (OTA) was 3.8 μg / kg (below the EU limit of 5 μg / kg but significantly higher than in the example, directly related to environmental Aspergillus contamination introduced by open fermentation).

[0045] Comparative Example 2 (SIAF unvaccinated) 18 kg of fresh coffee cherries were placed in a 30 L sealed fermentation tank. No gas replacement was performed (self-induced anaerobic digestion was achieved by gradually releasing oxygen through CO2 produced by the natural microbial community on the cherry surface). No microbial agents were inoculated, and fermentation was carried out at a constant temperature of 23℃ for 72 h. Results: 70 volatile flavor compounds were detected, with esters accounting for 19.2%, ethyl acetate 14.5 μg / g, phenylethanol 11.2 μg / g, and linalool 8.5 μg / g. The SCA cupping score was 81.5 points. Sensory evaluation after 48 h of fermentation revealed a slight musty odor, and OTA detection was 0.8 μg / kg (due to incomplete air removal during the early stages of SIAF, residual oxygen supported the growth of a small amount of Aspergillus).

[0046] Comparative Example 3 (simultaneous inoculation of three bacteria in one go, free bacterial cells) The same three bacterial strains as in Example 1 (Lactobacillus plantarum, Saccharomyces cerevisiae, and Acetobacter pasteurellium) were used, but were simultaneously inoculated into a sealed anaerobic fermenter in the form of free bacterial powder. The inoculum size of the lactic acid bacteria was... CFU / g fresh fruit, yeast inoculation amount is CFU / g fresh fruit, acetic acid bacteria inoculation amount is CFU / g fresh fruit. Nitrogen gas replacement conditions were the same as in Example 1. Fermentation was carried out at a constant temperature of 22℃ for 72 h. Detection results: 74 volatile flavor compounds were found, with esters accounting for 21.8%, ethyl acetate 16.3 μg / g, phenylethanol 13.5 μg / g, and linalool 10.2 μg / g. The SCA cupping score was 82.0. Due to the simultaneous activation of the three bacteria, lactic acid bacteria, with its much faster proliferation rate than yeast and acetic acid bacteria, rapidly lowered the pH to below 4.2 within the first 12 h. The initial activity of yeast and acetic acid bacteria was significantly inhibited by the sudden pH drop (the ethanol yield of yeast decreased by about 40% under pH < 4.0 conditions). Ultimately, the types and contents of ester flavor compounds were significantly lower than in Example 1, and the flavor was mainly a single sour taste dominated by lactic acid. OTA was not detected (the sealed anaerobic operation effectively prevented Aspergillus contamination).

[0047] Comparative Example 4 (Three strains artificially inoculated in stages, free bacterial cells) The same three bacterial strains as in Example 1 were artificially inoculated in stages using free bacterial powder. Lactic acid bacteria were inoculated at 0 h of fermentation. CFU / g fresh fruit), fermented for 18 h, then inoculated with yeast (CFU / g fresh fruit). CFU / g fresh fruit), fermented for 36 h, then inoculated with acetic acid bacteria (CFU / g fresh fruit). CFU / g fresh fruit). The initial nitrogen gas replacement conditions were the same as in Example 1, with nitrogen re-purging after each inoculation (but it could not completely restore the initial low oxygen level). The segmented temperature control scheme was the same as in Example 1, and the total fermentation time was 72 h. Detection results: 78 volatile flavor compounds, esters accounted for 25.3%, ethyl acetate 19.8 μg / g, phenylethanol 16.2 μg / g, and linalool 12.8 μg / g. The SCA cupping score was 83.5 points. Although the sequential inoculation of the strain was achieved manually, the two inoculation operations caused the oxygen concentration in the container to briefly rise from 0.5% to 8% at 18 h of fermentation (it dropped to below 2% after about 15 min of re-nitrogen purging), and rise to 5% at 36 h. A small number of Aspergillus colonies were detected in the fermentation broth. The OTA detection limit was 1.5 μg / kg (which is lower than the EU limit but significantly higher than the detection limit of <0.2 μg / kg in the example). The flavor complexity was better than Comparative Example 3 but weaker than Example 1, and the standard deviation of the cupping score between batches was ±2.1 points (5 batches), which was less consistent than ±0.4 points in Example 1.

[0048] Comparative Example 5 (microcapsule encapsulation but without chitosan layer differences) The same three bacterial strains as in Example 1 were used, but all three types of microcapsules used a 2% (w / w) monolayer sodium alginate as the wall material (excluding the chitosan layer). Specifically, the chitosan coating and outer sodium alginate coating steps were omitted in the preparation of the second and third types of microcapsules. The wall material composition and thickness of the three types of microcapsules were identical. Other conditions (gas replacement, inoculum size, segmented temperature control) were the same as in Example 1. Detection results: 69 volatile flavor compounds were detected, with esters accounting for 18.5%, ethyl acetate 13.2 μg / g, phenylethanol 10.8 μg / g, and linalool 8.0 μg / g. The SCA cupping score was 80.5. Because the wall material composition and structure of the three types of microcapsules were identical, they disintegrated almost simultaneously under the same pH conditions, releasing all bacterial strains, exhibiting a similar imbalance in bacterial community competition as Comparative Example 3 (simultaneous inoculation of free bacterial cells). This result demonstrates the necessity of the differentiated wall material hierarchical design (single-layer → double-layer → triple-layer) in this invention for achieving time-sequential release—simply encapsulating the strain in microcapsules without hierarchical differentiation cannot achieve the time-sequential release effect. OTA was not detected.

[0049] Table 1. Summary of key detection data for each of Examples 1-3 and Comparative Examples 1-5

[0050] Five independent replicate experiments were conducted on the protocol of Example 1 (each batch used fresh coffee cherries from the same region and variety but harvested on different dates: December 5, 2024, December 12, 2024, December 19, 2024, December 26, 2024, and January 2, 2025), with each batch strictly following all operating conditions of Example 1. The SCA cupping scores for the five batches were 86.2, 85.8, 86.5, 86.0, and 85.5, with an average of 86.0, a standard deviation of ±0.4, a coefficient of variation of 0.47%, and an ester compound content of 34.2% ±0.8%. As a control, the cupping scores for Comparative Example 1 (natural fermentation) were 79.8, 77.5, 81.2, 80.0, and 76.8, with an average of 79.1, a standard deviation of ±1.7, and a coefficient of variation of 2.15%. Comparative Example 4 (staged artificial inoculation) showed that the cupping scores for 5 batches were 83.5, 82.0, 84.8, 81.5, and 83.2, with an average score of 83.0, a standard deviation of ±1.3, and a coefficient of variation of 1.57%. These data indicate that the microencapsulated bacterial agent scheme of this invention is significantly superior to existing technologies in both absolute flavor quality (86.0 vs. 79.1 and 83.0) and batch consistency (±0.4 vs. ±1.7 and ±1.3).

[0051] The technical effect of this invention is based on the following triple synergistic mechanism: First, the gradient release mechanism of the pH-responsive wall material enables a relay of microbial metabolism. The natural gradient change in pH during the fermentation of fresh coffee cherries, from 5.8 to 3.5, serves as a trigger signal for the three types of microcapsules. Lactic acid bacteria are the first to release and rapidly produce acid within the pH window of 5.5–6.0, establishing a low-pH environment to create optimal fermentation conditions for subsequent yeast (pH 4.5–5.0). Yeast efficiently produces alcohols and esters in a weakly acidic anaerobic environment, generating phenethyl alcohol, linalool, and other aromatic and fruity alcohols that directly contribute to flavor, while the produced ethanol serves as a substrate for subsequent acetic acid bacteria oxidation. Acetic acid bacteria, after releasing within the pH window of 3.8–4.5, oxidize some of the ethanol to acetic acid, which then esterifies with ethanol under acidic conditions to produce key flavor compounds such as ethyl acetate. The metabolic products of the three strains are mutually substrates, forming a complete metabolic relay chain. This process is automatically achieved through the wall material design, requiring no human intervention.

[0052] Second, the fully sealed anaerobic operation synergistically enhances food safety. The microencapsulated one-time administration eliminates the need for opening the container, fundamentally blocking the invasion route of toxin-producing fungi. Comparative Example 4 provides direct evidence: even just two brief openings (each exposure lasting approximately 5 minutes) resulted in the detection of Aspergillus species (1.2 × 10⁻⁶). 2The levels of CFU / mL and OTA increased from <0.2 μg / kg to 1.5 μg / kg. OTA was not detected in any of the examples and comparative examples (Examples 1-3, Comparative Example 3, and Comparative Example 5) that employed a fully sealed approach, verifying the decisive protective effect of the fully sealed operation on food safety.

[0053] Third, the difference in wall material layers is a necessary condition for achieving sequential release. Comparative Example 5 provides crucial counter-evidence: when all three types of microcapsules used a single-layer sodium alginate wall material (without chitosan layer differences), the three strains released almost synchronously under the same pH conditions, with an SCA score of only 80.5, comparable to Comparative Example 3 (synchronous inoculation of free cells, 82.0 points), and far lower than Example 1 (86.2 points) which used a differentiated wall material design. This result clearly demonstrates that the single-layer → double-layer → triple-layer wall material layer design, by introducing the chitosan protonation expansion mechanism and multilayer constraint effect, establishes a gradient release window of pH 5.5–6.0 → pH 4.5–5.0 → pH 3.8–4.5, which is a necessary technical condition for achieving sequential release and metabolic relay of strains.

[0054] The embodiments of the present invention are not limited to the specific embodiments described above. Those skilled in the art can make various equivalent changes or substitutions based on the technical solutions of the present invention, and all such changes or substitutions should be included within the protection scope of the present invention.

Claims

1. A method for targeted flavor regulation during anaerobic fermentation of fresh coffee cherries, characterized in that, Includes the following steps: S1, Preparation of microencapsulated composite bacterial agents: Lactic acid bacteria, yeast, and acetic acid bacteria were encapsulated in microcapsules with different wall materials to obtain three types of microcapsule bacterial agents with differentiated pH-responsive release characteristics. Among them, the first type of microcapsule uses sodium alginate as the wall material and releases lactic acid bacteria under pH 5.5-6.0 conditions; the second type of microcapsule is coated with a sodium alginate-chitosan double-layer wall material and releases yeast under pH 4.5-5.0 conditions; the third type of microcapsule is coated with a sodium alginate-chitosan-sodium alginate triple-layer wall material and releases acetic acid bacteria under pH 3.8-4.5 conditions. S2, Raw material pretreatment: Select fresh coffee cherries with a maturity of over 80%, remove diseased and pest-infested cherries and unripe cherries, wash and drain until there are no obvious water droplets on the surface; S3, Establishment of anaerobic environment and one-time inoculation: Place the fresh coffee cherries obtained in step S2 into a sealed fermentation container, and fill the sealed fermentation container with food-grade inert gas to replace the air until the oxygen concentration in the sealed fermentation container drops to below 1%; then, add the three types of microcapsule inoculants obtained in step S1 into the sealed fermentation container in one go, and seal it. S4, pH-driven sequential release of bacterial strains and temperature-controlled fermentation: Sealed anaerobic fermentation is carried out at a temperature of 18℃~25℃. During the fermentation process, as the pH of the fermentation broth naturally decreases, the three types of microcapsule bacterial agents sequentially release the encapsulated bacterial strains, realizing the automatic sequential incorporation of lactic acid bacteria, yeast, and acetic acid bacteria. S5, Fermentation endpoint determination: When the pH of the fermentation broth drops to 3.5-4.0 and the titratable acidity rises to 0.8-1.2 g / 100 mL, the fermentation endpoint is determined to have been reached, and fermentation is stopped. S6, Post-processing: The fermented coffee cherries obtained in step S5 are peeled and degummed, then washed and dried to a moisture content of 10% to 12% to obtain fermented green coffee beans.

2. The method according to claim 1, characterized in that, In step S1, the preparation method of the first type of microcapsule includes: mixing a lactic acid bacteria suspension with a sodium alginate solution with a mass concentration of 1% to 3%, dripping the mixture into a calcium chloride solution with a mass concentration of 1% to 5% via a syringe for ion gelation, collecting the microcapsules after solidification for 10 min to 30 min, washing and freeze-drying; the preparation method of the second type of microcapsule includes: based on the preparation method of the first type of microcapsule, immersing the solidified sodium alginate microcapsules in a chitosan solution with a mass concentration of 0.1% to 1.0% for 5 min to 20 min to form a double-layer wall material structure; the preparation method of the third type of microcapsule includes: based on the preparation method of the second type of microcapsule, immersing the double-layer wall material microcapsules again in a sodium alginate solution with a mass concentration of 0.5% to 2.0% for 5 min to 15 min, and then solidifying them a second time with a calcium chloride solution to form a triple-layer wall material structure.

3. The method according to claim 1, characterized in that, In step S1, the particle size of the microcapsules is 0.5 mm to 3.0 mm; the encapsulation density of lactic acid bacteria in the first type of microcapsules is... CFU / g microcapsules; the encapsulation density of yeast in the second type of microcapsules is CFU / g microcapsules; the encapsulation density of acetic acid bacteria in the third type of microcapsules is CFU / g microcapsules.

4. The method according to claim 1, characterized in that, In step S1, the lactic acid bacteria are selected from at least one of Lactobacillus plantarum and Lactobacillus fermentum; the yeast is selected from at least one of Saccharomyces cerevisiae and Kluyveromyces kuhlii; and the acetic acid bacteria are Acetobacter pasteurellii.

5. The method according to claim 1, characterized in that, In step S3, the total amount of the three types of microcapsule inoculants is 0.5% to 3.0% of the weight of fresh coffee fruit; wherein the mass ratio of the first type of microcapsule, the second type of microcapsule, and the third type of microcapsule is (3 to 5): (2 to 3):

1.

6. The method according to claim 1, characterized in that, In step S3, the food-grade inert gas is nitrogen and / or carbon dioxide, the aeration rate is 0.5 L / min to 2.0 L / min, and the aeration time is 10 min to 30 min; the sealed fermentation container is equipped with a one-way exhaust valve, and the opening pressure of the one-way exhaust valve is 0.02 MPa to 0.05 MPa.

7. The method according to claim 1, characterized in that, In step S4, the temperature is controlled at 18℃~20℃ for the first 24 hours of fermentation, 20℃~23℃ for the first 24 hours to 48 hours of fermentation, and 23℃~25℃ after 48 hours of fermentation.

8. The method according to claim 1, characterized in that, In step S5, the determination of the fermentation endpoint further includes: the reducing sugar content in the fermentation broth drops to below 30% to 50% of the initial value, and the ethanol content in the fermentation broth is 0.5% to 3.0% (v / v).

9. The method according to claim 1, characterized in that, The compound microbial agent consists of three types of microcapsules: the first type of microcapsules encapsulates lactic acid bacteria with sodium alginate as the wall material and releases them at pH 5.5–6.0; the second type of microcapsules encapsulates yeast with a sodium alginate-chitosan double-layer wall material and releases them at pH 4.5–5.0; and the third type of microcapsules encapsulates acetic acid bacteria with a sodium alginate-chitosan-sodium alginate triple-layer wall material and releases them at pH 3.8–4.

5. The mass ratio of the three types of microcapsules is (3–5):(2–3):

1.

10. The method according to claim 9, characterized in that, The sodium alginate has a mass concentration of 1%–3%, the chitosan has a mass concentration of 0.1%–1.0%, and the degree of deacetylation of the chitosan is 80%–95%; the microcapsules have a particle size of 0.5 mm–3.0 mm; the encapsulation density of lactic acid bacteria in the first type of microcapsules is… CFU / g microcapsules, the encapsulation density of yeast in the second type of microcapsules is CFU / g microcapsules, wherein the encapsulation density of acetic acid bacteria in the third type of microcapsules is [missing information]. CFU / g microcapsules.

Images