Method for preparing enteromorpha oligosaccharide through composite enzymatic hydrolysis and gradient alcohol precipitation

Abstract

The invention relates to the technical field of biological catalysis engineering, and discloses a method for preparing enteromorpha oligosaccharide through composite enzymatic hydrolysis gradient alcohol precipitation, which comprises the following steps: in the enzymatic hydrolysis reaction process, establishing negative feedback coupling between stirring power and ethanol concentration increase operation, characterizing the consumption degree of a non-liquid-phase substrate in real time by using the stirring power, and calculating the concentration of the enteromorpha oligosaccharide. According to the present invention, the strong coupling relationship between the chemical reaction and the physical phase change is established, such that the enzymolysis is converted from the random collision process to the physical phase equilibrium-dominated deterministic ordered evolution process so as to achieve the selective precipitation and degradation of the product; and the capability of the system for automatically adapting to the raw material difference is improved, and the technical problems of disordered reaction and poor product selectivity in the existing enzymatic degradation are solved.

Description

Technical Field

[0001] This invention relates to a method for preparing Ulva oligosaccharides from seaweed through a complex enzymatic hydrolysis gradient alcohol precipitation, belonging to the field of biocatalysis engineering technology. Background Technology

[0002] Currently, enzymatic degradation of natural polysaccharide substrates is a common technique, which is usually carried out in a homogeneous or near-homogeneous liquid environment. Mechanical stirring promotes full contact between the enzyme and substrate molecules to achieve a high reaction conversion rate. This method is effective when the goal is to achieve a high total polysaccharide conversion rate. However, when industrial applications shift towards producing specific oligosaccharides with a narrow molecular weight distribution range, an inherent limitation of this technique becomes apparent.

[0003] The reason is that a well-mixed homogeneous reaction system provides enzyme molecules with indiscriminate contact opportunities with all sugar molecules dissolved in it. Whether it is the original long-chain substrate or the generated target intermediate oligosaccharide, they will all become the target of enzyme action according to their respective kinetic probabilities. This means that a target oligosaccharide molecule becomes a new substrate for further degradation at the same time it is generated. In a system where the reaction process and the separation process are separated, it is impossible to protect the target product in situ in time. The continuous reaction will lead to the continuous degradation of the product to smaller molecular weights, and the final result is a complex mixture.

[0004] To improve product selectivity, common improvement methods in the art include controlling reaction conditions such as temperature or pH, or employing separation techniques such as multi-stage organic solvent precipitation and chromatographic chromatography after the reaction. However, the former can only adjust the average molecular weight of the product to a limited extent and cannot change the random nature of the reaction process, while the latter is a purification step after the reaction is completed, which cannot compensate for the material loss caused by the excessive degradation of the target product that has already occurred during the reaction stage. Specifically, the existing technology mainly has the following technical problems: 1. In a homogeneous reaction environment, the enzyme's action lacks selectivity, and the formation and re-degradation of the target intermediate product occur simultaneously, making excessive degradation difficult to avoid; 2. The product formation process and its separation and protection process are independent of each other in time and space, leading to process... The process is difficult to control, and the batch stability of the final product quality is insufficient. For example, Chinese invention patent CN108003199B discloses a method for preparing and applying a hypoglycemic oligosaccharide from *Ulva prolifera*, which is prepared by a one-time precipitation with a fixed concentration of ethanol. Essentially, this method treats enzymatic hydrolysis and product separation as two completely independent sequential steps. During its relatively long reaction time, the target oligosaccharide product generated in the earlier stages cannot be separated and protected in time, and will continue to be further degraded as a substrate. This inevitably leads to a wide molecular weight distribution and poor selectivity in the final product. This technical solution fails to effectively solve the core problems of excessive degradation of the target product and disordered reaction process during enzymatic hydrolysis. Therefore, the technical problem to be solved by this invention is how to establish a new reaction mechanism under which the generation process of a specific oligosaccharide product can directly trigger its own immediate separation, and terminate its excessive degradation reaction path through separation, transforming the entire enzymatic hydrolysis process from a random reaction to an ordered, molecularly controllable, directional reaction. Summary of the Invention

[0005] This invention provides a method for preparing Ulva oligosaccharides by gradient enzymatic hydrolysis and alcohol precipitation. Its main purpose is to solve the problem of disordered reaction and poor product selectivity caused by the independent reaction process and product separation process in the existing enzymatic degradation process.

[0006] To achieve the above objectives, the present invention provides a method for preparing *Ulva prolifera* oligosaccharides through complex enzymatic hydrolysis and gradient alcohol precipitation, comprising the following steps: Step a, in a reactor equipped with a stirring device, an aqueous solution containing Ulva prolifera polysaccharide substrate is provided, and ethanol is added, so that a heterogeneous system containing a liquid phase and a non-liquid phase composed of Ulva prolifera polysaccharide substrate is formed in the reactor; Step b: Add a complex enzyme to the heterogeneous system and carry out an enzymatic hydrolysis reaction at the interface between the liquid phase and the non-liquid phase, so that the Ulva polysaccharide substrate in the non-liquid phase is degraded into oligosaccharide products that are soluble in the liquid phase. Step c: During the enzymatic hydrolysis reaction, a negative feedback coupling relationship is established between the stirring power in the reactor and the ethanol increase operation. This negative feedback coupling relationship is specifically achieved through the following operation: real-time monitoring of stirring power. When the stirring power decreases to a stable threshold due to the consumption of non-liquid phase in the enzymatic hydrolysis reaction, the operation of increasing the concentration of ethanol in the system is triggered. This increase operation selectively causes intermediate products with a molecular weight range determined by the current ethanol concentration to precipitate from the liquid phase and become new substrates for the enzymatic hydrolysis reaction until the ethanol concentration reaches the preset endpoint. In step d, after the concentration of ethanol reaches the preset endpoint, the liquid phase is separated and collected to obtain the oligosaccharide product.

[0007] Preferably, the determination of the stability threshold in step c and the triggering of the operation to increase the ethanol concentration constitute an adaptive loop. The operating mechanism of this adaptive loop is as follows: after an ethanol increase operation causes the intermediate product to precipitate and form a new non-liquid phase, the stirring power rises to a peak value; as the enzyme is consumed by the reaction at the new non-liquid phase interface, the stirring power decreases accordingly. When the stirring power is detected to decrease to a preset proportion of the peak value, it is determined that the stability threshold has been reached, and the next operation to increase the ethanol concentration is immediately triggered, so that the rate of increase of ethanol concentration directly depends on the reaction characteristics of the Ulva prolifera polysaccharide substrate itself.

[0008] Preferably, the complex enzyme added in step b is the first enzyme composition; and the method further includes: during step c, adding a second enzyme composition to the reactor with a different composition or ratio than the first enzyme composition, wherein the enzyme in the second enzyme composition has higher ethanol tolerance than the enzyme in the first enzyme composition, so as to superimpose a biocatalytic gradient composed of enzyme composition changes on top of the physical environmental gradient composed of ethanol concentration changes.

[0009] Preferably, the determination rule for the stability threshold in step c is limited by the following inequality: P_current≤(1-α)·P_peak, where P_current is the current stirring power monitored in real time; P_peak is the peak stirring power that appears immediately after an ethanol increase operation in this adaptive cycle; α is a preset power decrease ratio factor in the range of 0.05 to 0.2. When the inequality is true, it is determined that the stability threshold has been reached.

[0010] Preferably, the method further includes: monitoring an acoustic characteristic signal of the reactor after each increase in ethanol concentration in step c; and, based on the acoustic characteristic signal, determining a morphological characteristic of the precipitated non-liquid phase. When the non-liquid phase is determined to be a viscous gel-like mass with a low specific surface area, a brief interface remodeling stirring program is executed before resuming normal stirring to break up the viscous gel-like mass into loose flocs with a high specific surface area.

[0011] Preferably, the method further includes: continuously monitoring the temperature difference between a reaction temperature inside the reactor and a reference temperature outside the reactor throughout the entire process; and issuing a termination command when the temperature difference stably falls below a preset termination threshold close to zero due to the disappearance of the heat of enzymatic hydrolysis, forcibly stopping the operation of increasing the ethanol concentration and proceeding to step d, thus providing an endpoint arbitration based on the nature of the reaction for the entire adaptive process.

[0012] Preferably, in step a, the non-liquid phase is a gel phase or a solid phase; in step c, the initial concentration of ethanol in the system is 25% to 35% by volume, and the preset endpoint concentration is 55% to 75% by volume; the temperature of the enzymatic hydrolysis reaction is controlled at 40 degrees Celsius to 60 degrees Celsius, and the pH value is controlled at 4.0 to 6.0.

[0013] Preferably, the first enzyme composition mainly comprises an endonuclease responsible for rapidly reducing the molecular weight of the substrate; the second enzyme composition mainly comprises an exonuclease or debranching enzyme responsible for modifying the ends of oligosaccharides or removing branches; the addition of the second enzyme composition is carried out by continuous dropwise addition after the ethanol concentration of the system reaches an intermediate concentration point.

[0014] Preferably, the acoustic characteristic signal is the vibration signal collected by a vibration sensor installed on the outer wall of the reactor vessel; the operation to determine the morphological characteristics is as follows: after increasing the ethanol concentration, the rotation speed of the stirring paddle is briefly changed to apply an active acoustic detection excitation, and a fast Fourier transform is performed on the transient response of the vibration signal induced by the excitation. Whether a viscous gel-like mass is formed is determined based on whether a characteristic low-frequency high-amplitude resonance peak appears in the response signal.

[0015] Preferably, after step d, the collected liquid phase is further subjected to nanofiltration concentration and freeze-drying to recover the solid oligosaccharide product.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. During the enzymatic hydrolysis reaction, the concentration of organic solvent in the system is continuously controlled, establishing a strong coupling relationship between chemical reaction and physical phase transition. The degradation of the solid substrate by the enzyme directly causes the product to dissolve into the liquid phase, while the continuously increasing solvent concentration in the liquid phase selectively precipitates intermediate products of specific molecular weights as new solid-phase reaction interfaces. This process transforms the entire degradation reaction from disordered random collisions in a traditional homogeneous system into a deterministic evolutionary process continuously guided by physical phase equilibrium, proceeding sequentially from high molecular weight to low molecular weight.

[0017] 2. This invention further uses the stirring power in the reactor as a real-time physical quantity characterizing the total amount of non-liquid phase in the system, and uses the change of this physical quantity to trigger the increase of organic solvent; when the interfacial reaction causes the non-liquid phase to consume less stirring power, the system replenishes solvent to cause some intermediate products to precipitate and the stirring power to recover. This constructs a closed-loop feedback regulation based on the actual reaction process, so that the rate of increase of solvent concentration no longer depends on a fixed time program, but directly depends on the reaction characteristics of the substrate itself, thereby enabling the entire process to obtain the inherent adaptability to the differences in the physicochemical properties of different batches of raw materials.

[0018] 3. This invention further divides the complex enzyme system according to the differences in function and environmental tolerance of each component, and performs programmed supplementation at different stages of the reaction process; on top of the physical environmental gradient dominated by changes in solvent concentration, a biocatalytic gradient composed of changes in enzyme components is superimposed, so that in the early stage of the reaction, highly efficient but intolerant enzymes complete the main degradation, while in the later stage of the reaction, enzymes with better tolerance modify the product. This synergy between the physical gradient and the biological gradient allows the regulation of the final product to go beyond the single dimension of molecular weight, but extends to the control of its fine structure; and the acoustic properties of the outer wall of the reaction vessel are also considered. The resonance characteristic signal is used to determine the microscopic physical form of the newly precipitated non-liquid phase after each increase in solvent concentration, and the stirring mode is adjusted instantaneously based on the judgment result to reshape the reaction interface. This adds a mechanism to ensure the effectiveness of the microscopic reaction interface to the macroscopic process control based on stirring power. At the same time, by monitoring the differential heat flow signal inside and outside the reactor, the final disappearance of the heat of biochemical reaction is determined, providing an independent endpoint arbitration based on the nature of the reaction for the entire adaptive operation process. The introduction of these two mechanisms enables the system to get rid of dependence on indirect parameters when dealing with random disturbances in the process and determining the final reaction completion. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the process flow for the integrated multi-module feedback control of the present invention; Figure 2 This is a dynamic relationship diagram of gradient alcohol precipitation triggered by negative feedback of stirring power according to the present invention; Figure 3This is a diagram of the closed-loop control system architecture for adaptive regulation in this invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] This invention provides a method for preparing Ulva oligosaccharides by complex enzymatic hydrolysis and gradient alcohol precipitation. The technical solution establishes a correlation between the enzymatic hydrolysis chemical reaction and the precipitation of the physical phase of the product, and uses the physical quantity of stirring power to feedback and regulate the reaction process, so that the degradation process of polysaccharides proceeds sequentially from high molecular weight to low molecular weight. The method mainly includes the construction of an initial heterogeneous system, dynamic regulation of the enzymatic hydrolysis and alcohol precipitation process based on stirring power feedback, and determination of the reaction endpoint based on temperature difference signals.

[0022] In a specific implementation process, the first step is to construct a heterogeneous reaction system. This involves adding ethanol to an aqueous solution containing *Ulva prolifera* polysaccharide substrate in a reactor equipped with a stirrer, controlling the initial concentration of ethanol in the system within the range of 25% to 35% by volume. If the ethanol concentration is below 25%, the polysaccharide substrate with the largest molecular weight will not precipitate sufficiently, resulting in insufficient non-liquid interface area and affecting the initial reaction rate. If the concentration is above 35%, some intermediate products that should be degraded in subsequent stages may precipitate prematurely, interfering with the sequential degradation. Through this step, a liquid phase and a non-liquid phase composed of a gel or solid phase of high molecular weight *Ulva prolifera* polysaccharides are formed within the reactor. Simultaneously, the temperature of the reaction system is controlled between 40°C and 60°C, and the pH value is controlled between 4.0 and 6.0, thus constituting a reaction interface where the enzyme is distributed in the liquid phase and the substrate exists in the non-liquid phase. The process begins with the addition of a complex enzyme to the heterogeneous system to initiate the interfacial enzymatic hydrolysis reaction. Initially, a first enzyme composition is added, primarily consisting of endonucleases responsible for cleaving the polysaccharide backbone. These enzymes act at the interface between the liquid and non-liquid phases, degrading long-chain polysaccharides in the non-liquid phase into oligosaccharide products soluble in the liquid phase at the current ethanol concentration. As the reaction proceeds, after the ethanol concentration rises to an intermediate point according to a gradient program, for example, when the volume percentage reaches 45%, a second enzyme composition is added to the reactor. This second enzyme composition primarily consists of exonucleases or debranching enzymes responsible for modifying oligosaccharide ends or removing branches, and these enzymes exhibit higher ethanol tolerance. This arrangement of supplementing different functional enzyme systems on top of a physical environmental gradient created by changes in ethanol concentration allows endonucleases to perform the main degradation in the early stages of the reaction, while ethanol-resistant special enzymes modify the product structure in the later stages.

[0023] To adapt the process to the varying reaction characteristics of different batches of raw materials during the entire enzymatic hydrolysis reaction, this method establishes a negative feedback coupling relationship between stirring power and ethanol increase operations. The operation is as follows: stirring power is used as a physical quantity to characterize the total amount of non-liquid phase in the system in real time. When the interfacial reaction consumes the non-liquid phase, the system viscosity decreases, and the stirring power required to maintain a constant rotation speed decreases accordingly. The system monitors this stirring power P_current in real time and makes a judgment based on the inequality P_current≤(1-α)·P_peak. Here, P_peak is the peak stirring power that appears immediately after an ethanol increase operation in this adaptive cycle, and α is a power decrease proportionality factor ranging from 0.05 to 0.2. The value of α is determined based on calibration experiments, for example, in a standard system. The center value of alpha when the non-liquid substrate consumes 80 alpha is 0.1. If the value of alpha is lower than 0.05, the system is too sensitive to power changes and is easily affected by signal noise. If the value of alpha is higher than 0.2, the response is sluggish and may cause staged over-degradation. In a numerical deductive example, assuming that the power rises to the peak P_peak = 150W after one ethanol increase operation and alpha is set to 0.1, the stability threshold is calculated as (1-0.1)150W = 135W. When the enzymatic reaction causes P_current to drop to 135W, the inequality holds, and the controller immediately triggers the next operation to increase the ethanol concentration, so that the intermediate product selectively precipitates to form a new non-liquid phase, the stirring power rises again, and the next cycle begins. In this way, the rate of increase of ethanol concentration directly depends on the reaction rate of the substrate itself.

[0024] After the initial heterogeneous system is constructed and stabilized in step a, and after each ethanol addition operation in step c, the acquisition of the peak stirring power P_peak follows a standardized numerical determination procedure. Specifically, after adding the initial ethanol, stirring is continued at a preset process stirring rate until the monitored stirring power value fluctuates within ±2% of the current average value for 3 consecutive minutes. This stable average value is then used as the initial peak power P_peak of the first enzymatic hydrolysis cycle. For each subsequent new cycle triggered by the ethanol addition operation, the control system collects power data at a frequency of not less than 5Hz within a 60-second time window after the ethanol feed command is issued. The maximum value of the 5-second moving average of the power readings within this window is determined as the peak power P_peak of the new cycle. This procedure uses a moving average algorithm to filter out the hydrodynamic noise generated at the moment of ethanol injection. The value of P_peak directly corresponds to the system state after the newly precipitated non-liquid phase reaches dispersion equilibrium. Furthermore, to address the potential formation of low-surface-area viscous gel-like clumps during alcohol precipitation, this method includes an interface morphology control step based on acoustic characteristic signals. Vibration signals of the reactor vessel are collected using a vibration sensor installed on the outer wall of the vessel. After each increase in ethanol concentration, the stirring speed is briefly changed to apply an active acoustic detection excitation, and the transient response of the vibration signal induced by this excitation is analyzed using Fast Fourier Transform. If viscous gel-like clumps are formed, a characteristic low-frequency, high-amplitude resonance peak will appear in the response signal. Once the system identifies this characteristic peak, a brief interface reshaping stirring program involving rapid alternation of high and low speeds or reverse rotation is executed before resuming normal stirring. This program uses shear force to break the clumps into loose flocs, maintaining the effective area of ​​the reaction interface.

[0025] Interface morphology control based on acoustic characteristic signals is executed in a timing coordinated with conventional process stirring to avoid interference from the detection behavior on the natural precipitation morphology of the material. This coordination mechanism includes: after a single ethanol addition operation, the conventional stirring program is paused, and the system enters a settling precipitation window lasting 5 to 10 seconds; after the window ends, active acoustic detection excitation is executed, i.e., the stirring paddle is driven at a low speed of 50 rpm for 2 seconds, while simultaneously acquiring the vessel vibration signal; the control system performs a fast Fourier transform on the acquired signal. If its characteristic frequency and amplitude are within the range of the normal loose flocculent morphology obtained from calibration, conventional process stirring is directly resumed; if the signal characteristics match the calibration model of viscous gel-like clumps, an interface remodeling stirring program is executed first, and conventional process stirring is resumed after this program ends. This constitutes a closed-loop operation sequence of detection, judgment, and execution. Furthermore, to provide an endpoint determination criterion based on whether the biochemical reaction itself is complete, this method also includes an independent termination mechanism. This mechanism works by continuously monitoring the temperature difference between the reaction temperature inside the reactor and the external reference temperature. Since the enzymatic hydrolysis reaction is exothermic, as long as the reaction is in progress, the temperature inside the reactor will be slightly higher than the external reference temperature, forming a small positive temperature difference. When the substrate is exhausted and the heat of reaction disappears, this temperature difference will stably fall below a preset termination threshold, such as 0.02°C. When this phenomenon is detected, the system issues a termination command, stops adding ethanol, and proceeds to step d, i.e., separating and collecting the liquid phase. Subsequently, the collected liquid phase can be concentrated by nanofiltration and freeze-dried to recover the solid oligosaccharide product. This endpoint determination method based on the disappearance of the heat of reaction makes the process termination decision directly related to the completion of the biochemical reaction.

[0026] In a specific embodiment, the complex enzyme is further defined as follows: the complex enzyme comprises a first enzyme composition and a second enzyme composition. The first enzyme composition is mainly composed of an endonuclease responsible for random cleavage within the polysaccharide backbone, which rapidly depolymerizes the high molecular weight polysaccharide substrate at the initial non-liquid phase interface into a series of water-soluble intermediates. The second enzyme composition is mainly composed of an exonuclease responsible for cleaving monosaccharide or oligosaccharide residues from the ends of the sugar chain, or a debranching enzyme responsible for removing branched structures, which performs structural modification and fine degradation on the intermediates dissolved in the liquid phase to obtain oligosaccharides within the target molecular weight range. In a specific embodiment, the ratio of the endonuclease in the first enzyme composition to the exonuclease or debranching enzyme in the second enzyme composition, expressed in enzyme activity units (U), is in the range of (2:1) to (5:1). When this ratio is lower than 2:1, the relative content of the endonuclease in the system is insufficient, resulting in a low degradation rate at the initial non-liquid phase and a slow decrease in stirring power. The slow response of the power feedback-based ethanol concentration increase leads to a sluggish response and prolongs the total process time. When the ratio is higher than 5:1, the endonuclease is relatively excessive, which will rapidly generate a large number of intermediate products with a wide molecular weight distribution in the early stage of the reaction, exceeding the fine modification capability of the subsequent supplemented second enzyme composition, and ultimately leading to an increase in the polydispersity index (PDI) of the product. Therefore, controlling the enzyme activity ratio within the range of (2:1) to (5:1) can synergistically ensure the efficiency of the initial interface reaction and the uniformity of the molecular weight of the subsequent product. To ensure the stability of the process in long-term operation, the control system can introduce a periodic recalibration procedure. At a preset production batch interval or when the polydispersity index (PDI) of the final product deviates from the target value by more than 5% for three consecutive batches, the calibration procedure is automatically triggered to recalibrate the stirring power reduction ratio factor α, the acoustic feature judgment threshold, and the reaction endpoint temperature difference threshold, and the updated parameters are used as the control benchmark for subsequent production.

[0027] Example 1: In an industrial production process for preparing *Ulva prolifera* oligosaccharides with a specific molecular weight range, a batch of *Ulva prolifera* polysaccharide raw materials, due to differences in their source, exhibited characteristics of high average molecular weight, complex branched structure, and easy formation of viscous gels. If an enzymatic hydrolysis process based on a fixed-time program of increasing ethanol concentration was used, products with the required molecular weight distribution could not be obtained. To process this batch of raw materials, the aqueous solution of the *Ulva prolifera* polysaccharide substrate was placed in a reactor equipped with a power sensor, an external wall vibration sensor, and an internal and external temperature difference monitoring probe. Ethanol was added to make the initial concentration 30% by volume, forming a heterogeneous system. Then, the first enzyme composition was added to start the reaction. Since the degradation rate of this batch of raw materials was lower than that of the standard raw materials, the decrease in the stirring power P_current monitored by the system was correspondingly slowed down. Based on the negative feedback coupling relationship between the stirring power and the ethanol increase operation, the trigger frequency of the ethanol metering pump was automatically reduced. As a result, the rate of increase of ethanol concentration was directly determined by the degradation rate of the substrate itself. This adaptive rate adjustment provided a longer degradation time window for substrates with lower reactivity, avoiding the problem of embedding and solidifying insufficiently degraded substrates due to excessively rapid increases in ethanol concentration.

[0028] When the reaction proceeded to approximately 48% ethanol concentration, the vibration sensor on the outer wall of the reactor, after an ethanol addition operation, identified a characteristic low-frequency, high-amplitude resonance peak in the vibration signal collected by active acoustic detection excitation. Based on this, the system determined that viscous, gel-like clumps with low specific surface area had precipitated inside the reactor. At this point, the control system paused conventional stirring and executed an interface remodeling stirring program. Through several rapid alternations of high and low speeds, shear force was used to break the clumps into loose flocs until the acoustic characteristic signal returned to normal. Under this condition, the feedback adjustment of the stirring power determined the overall process rate, while the monitoring of the acoustic characteristic signal provided real-time correction for the physical morphology of the reaction interface after each phase precipitation, ensuring the reaction... The process and interface effectiveness were both considered. As the reaction continued, when the ethanol concentration exceeded 45%, the system was supplemented with a second ethanol-resistant enzyme composition according to the program to modify the newly generated oligosaccharides by branching. As the reaction entered its final stage, the stirring power changed gradually, and the system's endpoint determination was based on the internal and external temperature difference signal. When the temperature difference between the inside and outside of the reactor stabilized below the termination threshold of 0.02℃ due to the disappearance of the heat of enzymatic hydrolysis, the system issued a termination command, stopped adding ethanol, and entered the product separation step. The final collected liquid phase product was analyzed and found to have a concentrated oligosaccharide molecular weight distribution, a polydispersity coefficient of less than 1.2, and a branched structure content lower than the preset requirements. This result was obtained under the condition that the raw materials used were highly variable batches.

[0029] Example 2: To objectively verify the technical effect of the method of the present invention in improving the uniformity of product molecular weight, the following comparative experiment was designed and executed; the experiment used a 5L standard reaction vessel with a jacket, equipped with a temperature control unit with a temperature control accuracy of ±0.1℃, a stirring device connected to a power meter, and a high-performance gel electrophoresis instrument for determining the molecular weight distribution of the product; the experimental raw material was the same batch of crude *Ulva prolifera* polysaccharide, with an initial weight-average molecular weight of 2.5 x 10^5 Da and a polydispersity index (PDI) of 3.8; the enzyme preparation used was a complex cellulase, and the total enzyme activity added was kept the same in each experimental group; the experiment set up one sample group of the present invention and two controls. The sample group of this invention was carried out according to the specific implementation method, wherein the initial ethanol concentration was set to 30% by volume, the stirring power reduction ratio factor was set to 0.1, the preset endpoint ethanol concentration was 60% by volume, the reaction temperature was 50℃, and the pH was 5.0; the control group A adopted a homogeneous enzymatic hydrolysis method, reacted in a buffer solution without adding ethanol for 4 hours, and after the reaction was completed, ethanol was added at one time to achieve a final concentration of 60%, and the supernatant was collected; the control group B adopted a gradient alcohol precipitation enzymatic hydrolysis with a fixed time program, the initial conditions were the same as those of the sample group of this invention, but after the reaction started, the ethanol concentration of the system was linearly increased to 60% at a constant rate over 3 hours using a metering pump.

[0030] After the experiment, the liquid products obtained from each group were analyzed. The results showed that different preparation methods affected the yield and molecular weight distribution of the products. The control group A, which used homogeneous enzymatic hydrolysis, had a target oligosaccharide yield of only 18.2%, i.e., the component with a degree of polymerization (DP) of 5 to 15, and a final product PDI value of 2.95, indicating that the product composition was complex and the homogeneity was poor. In contrast, the control group B, which used a fixed-time gradient alcohol precipitation, had a target oligosaccharide yield of 35.7% and a PDI value of 1.81, showing that the gradient alcohol precipitation operation itself has a certain effect on improving selectivity. The sample group of this invention achieved a target oligosaccharide yield of 62.5% and a final product PDI value of 1.18 within a total process time of 3.2 hours driven by stirring power feedback. This data shows that, compared with the fixed-time program used in control group B, the adaptive feedback adjustment mechanism based on stirring power in the sample group of this invention couples the rate of increase in ethanol concentration with the actual reaction process of the substrate, which is a key factor in further improving the target product yield and molecular weight uniformity.

[0031] Example 3: This example combines Figures 1 to 3 A method for preparing *Ulva prolifera* oligosaccharides by gradient enzymatic hydrolysis and alcohol precipitation is described, as follows: Figure 1As shown, the process begins with the *Ulva prolifera* polysaccharide substrate as input. First, initial ethanol is added to construct an initial heterogeneous system containing a solid-liquid interface. Then, a complex enzyme is added to initiate interfacial enzymatic hydrolysis, degrading the polysaccharide. The core of the process is an adaptive enzymatic hydrolysis and gradient ethanol precipitation core cycle. In this cycle, the selective precipitation and orderly degradation of the product are achieved by continuously controlling the ethanol concentration. The operation of this core cycle is monitored and regulated in real time by three parallel control modules: one is a stirring power negative feedback control module, which judges the degree of consumption of the non-liquid phase by real-time monitoring of the stirring power, and accordingly... The system consists of three main components: a trigger for ethanol addition; an acoustic characteristic morphology control module, which monitors the acoustic signals of the reactor to determine whether the newly precipitated non-liquid phase is gel-like and optimizes the stirring mode accordingly to execute the interface remodeling stirring program; a temperature difference signal endpoint arbitration module, which continuously monitors the temperature difference inside and outside the reactor to determine whether the heat of reaction has disappeared and issues a termination command accordingly. Upon receiving the termination command, the core cycle ends and enters the product separation and collection step to collect the liquid phase containing the target oligosaccharide. Finally, through post-processing steps such as nanofiltration concentration and freeze drying, high-purity final product Ulva oligosaccharide is obtained.

[0032] like Figure 2 As shown in the figure, the horizontal axis represents time in minutes (min), the left vertical axis represents stirring power in watts (W), and the right vertical axis represents ethanol concentration in volume percentage (%). The solid line represents the real-time monitored stirring power, which decreases periodically due to the consumption of non-liquid phases. The dashed line represents the ethanol concentration in the system, which increases in a stepwise manner under the trigger of control commands. The dotted line represents the stability threshold calculated based on the peak stirring power. As can be seen from the figure, whenever the stirring power shown by the solid line drops to the same level as the stability threshold shown by the dotted line, the system immediately triggers an ethanol increase operation, causing the ethanol concentration shown by the dashed line to jump to the next plateau. This operation then leads to the precipitation of intermediate products and causes the stirring power to rise to a new peak, thus starting the next adaptive cycle of consumption monitoring triggering precipitation.

[0033] like Figure 3As shown, the system centers on a main reactor, whose status is collected by a series of sensors, including a power sensor for obtaining stirring power feedback, a reactor vibration sensor for analyzing non-liquid phases, an internal and external temperature difference probe for endpoint arbitration, and a temperature / pH probe for monitoring conventional process parameters. All sensor data is transmitted to a central control server, where process control software analyzes the data in real time. This software contains core modules corresponding to each control function, namely a stirring power feedback module, an acoustic morphology analysis module, and an endpoint arbitration module. Based on the analysis results, the software issues control commands to a series of actuators, including a stirring device for adjusting the reaction interface and the mixing state of the system, an ethanol feed pump for performing gradient alcohol precipitation, and a composite enzyme feeding system for supplementing special enzymes at specific stages of the process. The status monitoring and manual operation of the entire system are accomplished through a human-machine interface, namely an operator terminal, thus forming a complete automated control loop.

[0034] Example 4: To determine the key control parameters of the method of the present invention in a specific reaction device, a standardized calibration procedure needs to be performed before the process runs. This procedure is carried out in a reactor of the same specifications as that used in actual production, containing an ethanol-water solution with the same initial concentration as the process. An inert polymer microsphere with stable physicochemical properties and similar characteristics to the target non-liquid phase is used as a non-liquid phase simulant to establish the relationship between process physical quantities and control parameters under conditions without biochemical reaction interference. The calibration process first runs the reactor with only ethanol-water solution in it at a preset process stirring rate, and records the stirring power under steady-state conditions as the reference power P_liquid. Subsequently, inert polymer microspheres with a total mass of M_total are added to the reactor, which corresponds to the initial mass of the non-liquid phase precipitated in the actual process. After it is fully dispersed and stabilized, the stirring power at this time is recorded. The peak stirring power, P_peak, was used to calibrate the proportional factor of stirring power reduction. The solid-liquid mixture was removed from the reactor in batches, gradually decreasing the mass of the remaining polymer microspheres. The corresponding stirring power, P_current, was recorded after each mass point stabilized, thus plotting the power-solid content relationship curve for this system. Based on this curve, the stirring power value corresponding to a remaining microsphere mass of 0.2M_total was found. This value was defined as the stability threshold, P_threshold, when 80% of the non-liquid phase was consumed. Finally, the value of α was calculated using the formula α=(P_peak-P_threshold)P_peak. In this calibration, if P_peak was measured to be 150W and P_threshold to be 132W, the calculated α value was 0.12, which was set as a control parameter for subsequent process operation.

[0035] To calibrate the threshold for judging acoustic characteristic signals, a small amount of high-viscosity sodium alginate solution was added to a system containing M_total polymer microspheres in the reactor to simulate the formation of viscous gel-like clumps. The vibration signal of the outer wall of the reactor in this gel-like system was recorded and subjected to a fast Fourier transform to obtain the condensed acoustic resonance spectrum. This spectrum was compared with the previously recorded normal spectrum of loose flocculent material containing only polymer microspheres. It was found that the signal amplitude of the gel-like state increased sharply in the frequency band from 50Hz to 150Hz. Based on this, the judgment threshold was determined as follows: if the integrated power of the vibration signal induced by active acoustic detection excitation in the frequency band from 50Hz to 150Hz exceeds three times the integrated power of the same frequency band in the loose flocculent state during the calibration stage, it is determined that viscous gel-like clumps have formed, and the interface remodeling stirring program is triggered. This program was set to rotate forward at 300 rpm for 5 seconds, then reverse at 100 rpm for 3 seconds, and repeat three times. Through the above procedure, control parameters suitable for the system were obtained.

[0036] Example 5: During a process operation, an external equipment failure caused the ethanol feed pump valve to jam, resulting in the injection of a large amount of ethanol into the reactor in a short period of time, which was not predetermined. This sudden increase in solvent concentration caused a large amount of dissolved intermediate products in the liquid phase to precipitate out instantly and quickly accumulate near the agitator, forming a high-density semi-solid gel mass that partially encapsulates the agitator. This condition will cause conventional process control methods to fail.

[0037] Under this sudden operating condition, the coordination response mechanism of the present invention is triggered. The stirring power monitoring system detects that the power value rises in a short period of time and exceeds the normal stirring power peak P_peak in this process cycle. Based on this, the system judges that mechanical overload has occurred. At the same time, the acoustic characteristic signal collected by the vibration sensor on the outer wall of the reactor undergoes a sudden change. Its low-frequency vibration amplitude exceeds the gel state judgment threshold set in the calibration stage, indicating that a large volume of solid obstruction has formed in the reactor. Based on the comprehensive judgment of these two physical signals, the control system confirms that a large volume rapid gelation event has occurred, and then automatically enters the fault recovery program. This program first forcibly shuts down the ethanol feed pump and suspends the regular adaptive cycle. Then, it starts a high-torque interface reshaping stirring program, which mechanically breaks up the semi-solid gel clumps by alternating forward and reverse pulses until the stirring power falls back to within the equipment safety threshold and the acoustic characteristic signal recovers to the characteristic spectrum range of loose flocculents. Only then does the system exit the fault recovery program and use the stable state of the system as the new benchmark to continue to execute the adaptive enzymatic hydrolysis and alcohol precipitation process.

[0038] Example 6: To establish a baseline reference model for the method of this invention, capable of reliably determining the endpoint of the enzymatic hydrolysis reaction under specific reactor thermodynamic conditions, a standardized pre-process thermodynamic characteristic calibration procedure needs to be performed after the process is applied to new equipment or key heat exchange components are replaced. This procedure aims to quantify the background thermal noise level of the specific system and, based on this, set a statistically significant reaction termination judgment threshold, while establishing a fault-tolerant mechanism to cope with the stagnation condition at the end of the reaction. This procedure first loads the reactor with materials similar to the actual process endpoint state, i.e., materials containing inactivated enzymes and final product concentrations at the target endpoint ethanol concentration. The solution was prepared, but in the absence of active substrate, stirring and constant temperature control were initiated, and the temperature difference signal ΔT between the inside and outside of the reactor was continuously recorded for at least 1 hour. The fluctuation of the ΔT signal during this period was regarded as the background thermal noise signal of the system. Subsequently, the noise signal data was statistically analyzed to calculate its mean μ_noise and standard deviation σ_noise, and the final reaction termination threshold T_threshold was set as T_threshold=μ_noise+3σ_noise. This threshold was stored in the control system as the benchmark for judging the disappearance of the heat of biochemical reaction in all subsequent production batches.

[0039] Furthermore, to address the issue of the temperature difference signal remaining in a stagnant state slightly above T_threshold but no longer significantly decreasing due to trace substrates or slow side reactions, this procedure also sets up a set of supplementary logical judgment conditions. In the control system, a stagnation judgment timer is set up to measure the time t_elapsed that has elapsed since the last ethanol addition operation triggered by a decrease in stirring power. If t_elapsed exceeds a preset maximum reaction period, such as 60 minutes, and during this period, the temperature difference signal ΔT is always less than a low-activity state threshold set to 2T_threshold, the system will determine that the reaction has entered an ineffective stagnation state and forcibly issue a termination command. Through the construction of the above baseline model and the setting of supplementary logic, a basis for determining the reaction endpoint is provided, and the risk of abnormal process delays is avoided.

[0040] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0041] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for preparing *Ulva prolifera* oligosaccharides via complex enzymatic hydrolysis and gradient alcohol precipitation, characterized in that, Includes the following steps: Step a, in a reactor equipped with a stirring device, an aqueous solution containing Ulva prolifera polysaccharide substrate is provided, and ethanol is added, so that a heterogeneous system containing a liquid phase and a non-liquid phase composed of Ulva prolifera polysaccharide substrate is formed in the reactor; Step b: Add a complex enzyme to the heterogeneous system and carry out an enzymatic hydrolysis reaction at the interface between the liquid phase and the non-liquid phase, so that the Ulva polysaccharide substrate in the non-liquid phase is degraded into oligosaccharide products that are soluble in the liquid phase. Step c: During the enzymatic hydrolysis reaction, a negative feedback coupling relationship is established between the stirring power in the reactor and the ethanol increase operation. This negative feedback coupling relationship is specifically achieved through the following operation: real-time monitoring of stirring power. When the stirring power decreases to a stable threshold due to the consumption of non-liquid phase in the enzymatic hydrolysis reaction, the operation of increasing the concentration of ethanol in the system is triggered. This increase operation selectively causes intermediate products with a molecular weight range determined by the current ethanol concentration to precipitate from the liquid phase and become new substrates for the enzymatic hydrolysis reaction until the ethanol concentration reaches the preset endpoint. In step d, after the concentration of ethanol reaches the preset endpoint, the liquid phase is separated and collected to obtain the oligosaccharide product.

2. The method for preparing *Ulva prolifera* oligosaccharides by complex enzymatic hydrolysis and gradient alcohol precipitation according to claim 1, characterized in that, The determination of the stability threshold in step c, and the operation of triggering the increase of ethanol concentration, constitute an adaptive loop. The operating mechanism of this adaptive loop is as follows: after an ethanol increase operation causes the intermediate product to precipitate and form a new non-liquid phase, the stirring power rises to a peak value. As the enzyme is consumed in the reaction at the new non-liquid interface, the stirring power decreases accordingly. When the stirring power drops to a preset percentage of the peak value, it is determined that a stable threshold has been reached, and the next operation to increase the ethanol concentration is immediately triggered.

3. The method for preparing *Ulva prolifera* oligosaccharides by complex enzymatic hydrolysis and gradient alcohol precipitation according to claim 1, characterized in that, The complex enzyme added in step b is the first enzyme composition; Furthermore, the method also includes: during step c, adding a second enzyme composition to the reactor that has a different composition or ratio than the first enzyme composition, wherein the enzyme in the second enzyme composition has higher ethanol tolerance than the enzyme in the first enzyme composition, so as to superimpose a biocatalytic gradient composed of changes in enzyme composition on top of the physical environmental gradient composed of changes in ethanol concentration.

4. The method for preparing *Ulva prolifera* oligosaccharides by complex enzymatic hydrolysis and gradient alcohol precipitation according to claim 2, characterized in that, The determination rule for the stability threshold in step c is limited by the following inequality: P_current≤(1-α)·P_peak, where P_current is the current stirring power monitored in real time; P_peak is the peak stirring power that appears immediately after an ethanol increase operation in this adaptive cycle; α is a preset power decrease ratio factor in the range of 0.05 to 0.

2. When the inequality is true, the stability threshold is determined to have been reached.

5. The method for preparing *Ulva prolifera* oligosaccharides by complex enzymatic hydrolysis and gradient alcohol precipitation according to claim 1, characterized in that, Also includes: In step c, after each increase in ethanol concentration, an acoustic characteristic signal of the reactor is monitored. Furthermore, based on acoustic characteristic signals, a morphological characteristic of the precipitated non-liquid phase is determined. When the non-liquid phase is determined to be a viscous gel-like mass with a low specific surface area, a brief interface remodeling stirring program is executed before resuming normal stirring, which involves rapid alternation of high and low speeds or reverse rotation, in order to break up the viscous gel-like mass into loose flocs with a high specific surface area.

6. The method for preparing *Ulva prolifera* oligosaccharides by complex enzymatic hydrolysis and gradient alcohol precipitation according to claim 1, characterized in that, Also includes: Throughout the entire process, the temperature difference between a reaction temperature inside the reactor and a reference temperature outside the reactor is continuously monitored. Furthermore, when the temperature difference stably falls below a preset termination threshold close to zero due to the disappearance of the heat of enzymatic hydrolysis, a termination command is issued to forcibly stop the operation of increasing the ethanol concentration and proceed to step d.

7. The method for preparing *Ulva prolifera* oligosaccharides by complex enzymatic hydrolysis and gradient alcohol precipitation according to claim 1, characterized in that, In step a, the non-liquid phase is either a gel phase or a solid phase; in step c, the initial concentration of ethanol in the system is 25% to 35% by volume, and the preset endpoint concentration is 55% to 75% by volume; the temperature of the enzymatic hydrolysis reaction is controlled at 40°C to 60°C, and the pH value is controlled at 4.0 to 6.

0.

8. The method for preparing *Ulva prolifera* oligosaccharides by complex enzymatic hydrolysis and gradient alcohol precipitation according to claim 3, characterized in that, The first enzyme composition mainly contains an endonuclease responsible for rapidly reducing the molecular weight of the substrate; the second enzyme composition mainly contains an exonuclease or debranching enzyme responsible for modifying the ends of oligosaccharides or removing branches; the addition of the second enzyme composition is carried out by continuous dropwise addition after the ethanol concentration of the system reaches an intermediate concentration point.

9. The method for preparing *Ulva prolifera* oligosaccharides by complex enzymatic hydrolysis and gradient alcohol precipitation according to claim 5, characterized in that, The acoustic characteristic signal is the vibration signal collected by a vibration sensor installed on the outer wall of the reactor vessel; the operation to determine the morphological characteristics is as follows: after increasing the ethanol concentration, the rotation speed of the stirring paddle is briefly changed to apply an active acoustic detection excitation, and a fast Fourier transform is performed on the transient response of the vibration signal induced by the excitation. Whether a viscous gel-like mass is formed is determined based on whether a characteristic low-frequency high-amplitude resonance peak appears in the response signal.

10. The method for preparing *Ulva prolifera* oligosaccharides by complex enzymatic hydrolysis and gradient alcohol precipitation according to claim 1, characterized in that, Following step d, the collected liquid phase is further subjected to nanofiltration concentration and freeze-drying.

Images