Production process of organic red oil broad bean with improved flavor

Abstract

The application relates to the food processing technical field and discloses an organic red-oil broad bean production process capable of improving flavor, which comprises the following steps: N2 purging is performed on fermented broad bean matrix loaded in a material basket to remove free water and establish a near-infrared monitoring benchmark; plant oil is pumped in under a negative pressure state and mixed with N2 pulse back pressure to complete pre-wetting of internal pores of the material; two-stage gradient extraction is adopted, low-temperature collection of aroma oil in the first stage is performed first, after indexes such as absorbance and pressure difference are stable, temperature collection of pigment oil in the second stage is performed; negative pressure degassing and fractional condensation are performed on the aroma oil, hydrophobic esters are intercepted and returned to the oil phase; the aroma oil and the pigment oil are blended to form red oil, the rich-aroma water phase obtained through condensation is atomized and sprayed back to the extracted oil residue, and finally, mixing and forming are performed. The process retains heat-sensitive flavor substances and reduces the oil emulsification rate.

Description

Technical Field

[0001] This invention relates to the field of food processing technology, specifically to a process for producing organic red chili bean paste with enhanced flavor. Background Technology

[0002] In the production and processing of red bean paste, the extraction of red oil typically employs high-temperature boiling or hot oil extraction processes. High-temperature operations cause a significant amount of heat-sensitive flavor compounds, such as esters and alcohols, within the bean paste system to volatilize or undergo thermal degradation, resulting in the loss of the product's characteristic ester aroma. Simultaneously, the fermented bean paste matrix retains free water, which easily enters the oil phase during extraction with oil, triggering an emulsification reaction. This leads to turbidity in the extracted red oil, reducing its clarity.

[0003] In fixed-bed extraction, the presence of gases in the pores of the matrix and the removal of moisture due to heat can cause solvent deviation within the bed, leading to instability in the extraction flow path and affecting the uniformity of solid-liquid extraction. Furthermore, existing extraction and separation methods result in the loss of some volatile flavor compounds and hydrophilic components during processing. The solid residue and separated liquid oils are difficult to effectively retain and precisely recombine in subsequent processing, resulting in low utilization of materials and flavor components throughout the system and negatively impacting the overall sensory quality of the finished product.

[0004] Therefore, this invention proposes a production process for organic red chili bean paste that enhances flavor, in order to address the shortcomings of existing technologies. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a production process for organic red chili bean paste that enhances flavor. This process solves the problems in existing red chili bean paste production processes, such as the volatilization and thermal degradation of heat-sensitive aroma substances caused by high-temperature oil extraction, and the low utilization rate of components in the entire system due to free water inside the matrix causing oil emulsification and turbidity and unstable extraction channels.

[0006] To address the above problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a production process for organic red chili bean paste that enhances flavor, employing the following technical solution:

[0008] An enhanced flavor production process for organic chili bean paste includes the following steps:

[0009] S1. Spread the fermented soybean substrate in a split porous media basket and place it in the extraction vessel; purge the extraction vessel with N2 and obtain the integrated absorbance of the discharged free brine phase in the near-infrared band as the background reference value of the polar aqueous phase.

[0010] S2. The extraction vessel is placed under negative pressure. Preheated vegetable oil is pumped into the extraction vessel from the bottom and mixed into the N2 back pressure extraction vessel. The pressure recovery curve is collected and the time constant is calculated by fitting. Pulse cycle is executed. When the change rate of the time constant of the adjacent pulse cycle drops to 3.0% to 5.0%, the pre-wetting is terminated.

[0011] S3. Pump vegetable oil into the bottom of the extraction vessel and collect the first-stage aroma oil; monitor the normalized absorbance, transmittance and turbidity of the first-stage aroma oil, and the fluctuation rate of the differential pressure between the inlet and outlet of the extraction vessel bed for five consecutive minutes. When the normalized absorbance of the first-stage aroma oil drops to 2.0% to 5.0%, the transmittance and turbidity of the first-stage aroma oil drops to 10.0 to 15.0 NTU, and the fluctuation rate of the differential pressure between the inlet and outlet of the extraction vessel bed for five consecutive minutes drops to 0.002 to 0.005 MPa / min, raise the temperature of the vegetable oil and collect the second-stage pigment oil.

[0012] S4. The front-end aroma oil is subjected to negative pressure degassing treatment; the extracted gas phase is condensed and collected in the first-stage condensation zone to collect the fragrant phase; the gas phase that is not condensed in the first-stage condensation zone is condensed and collected in the second-stage cryogenic zone after passing through the demister to collect the liquefied hydrophobic esters, and the liquefied hydrophobic esters are returned to the front-end aroma oil after negative pressure degassing treatment.

[0013] S5. Mix the front-stage aroma oil after reflux liquefaction of hydrophobic esters with the rear-stage pigment oil after cooling treatment to make red oil; take out the fermented soybean matrix in the extraction vessel as the oil extraction residue, spray the aroma-rich phase into the oil extraction residue to obtain the sprayed residue; mix the red oil with the sprayed residue to obtain the finished product.

[0014] By adopting the above technical solution, the process combines multi-stage thermodynamic state control and a material phase-segmentation and circulation mechanism to complete the replacement of pore fluids in the porous matrix and the directional separation of characteristic components, thereby improving the retention rate of flavor substances and the purity of color in the final product. The specific reactions and mechanisms of action are manifested in each process stage as follows:

[0015] In step S1, fluidized, isothermal N2 penetrates the porous media basket from top to bottom. The airflow removes free oxygen from the pores of the fermented soybean substrate and free water from the outer layer, eliminating interference from water and oxygen and significantly reducing the risk of lipid oxidation and phase emulsification during subsequent oil extraction. Simultaneously, the system collects the overtone absorption peak signal specific to water molecules in the near-infrared band and converts it into optical integrated absorbance. This serves as the initial optical benchmark for the current batch's moisture content, used to accurately determine whether the fluid has reached a steady state in subsequent processes.

[0016] As the process enters step S2, a negative pressure is established inside the extraction vessel, and preheated vegetable oil is injected. Due to the pressure difference, the fluid is forced into the micron-sized pores of the soybean matrix, followed by the introduction of nitrogen (N2) to induce system backpressure. The pressure change during this process strictly follows a first-order exponential decay model: P(t) = P ∞ +(P0-P ∞ )e -t / τ In the initial stage, because the fluid is not yet fully saturated inside the pores, the existing cavitation is violently compressed or dissolved, resulting in a high value of the fitted time constant τ. However, after multiple pressure pulse cycles, the gas in the micropores is gradually discharged and filled by the oil phase, the fluid network resistance of the entire bed begins to stabilize, and the rate of change of the time constant drops to the range set for termination. This negative pressure pulse pre-wetting method effectively eliminates solvent deviation and internal dead zones, maximizing the mass transfer specific surface area of ​​liquid-solid extraction.

[0017] After thorough pre-wetting, step S3 officially initiates the two-stage gradient extraction. Initially, relatively low-temperature vegetable oil is slowly injected from the bottom. The rising liquid flow from the low-frequency chromatography gently dissolves and extracts highly volatile and heat-sensitive ester and alcohol flavor compounds from the matrix, forming the initial aroma oil and fundamentally avoiding thermal degradation caused by high temperatures. During extraction, the system continuously monitors the normalized absorbance, turbidity, and inlet / outlet pressure difference fluctuation rate of the top overflow liquid. These three parameters reflect the concentration of free water and hydrophilic components carried out in the fluid, the amount of mechanically entrained suspended fines, and the structural stability of the internal flow channels of the fixed bed, respectively. Once these three indicators all drop to the set low range, it indicates that the hydrated emulsion phase in the system has been drained and the bed flow pattern has been completely stabilized. Based on this, the equipment switches to the high-temperature extraction stage. The increased inlet oil temperature intensifies the thermal motion of molecules, effectively breaking the physical binding force between large fat-soluble pigments such as carotenoids and the soybean matrix, thus carrying them out with the liquid flow and collecting them as the subsequent pigment oil.

[0018] For the collected initial aroma oils, step S4 is specifically designed to treat the entrained volatile phases. Under the combined effects of a specific negative pressure and temperature field, residual moisture and some low-boiling-point volatiles within the aroma oils are converted into a gaseous phase and extracted. The extracted mixed gas first enters the first-stage condensation zone, where the temperature is maintained above the freezing point of water. A large amount of water vapor, along with the accompanying hydrophilic flavor compounds, undergoes a phase change under the influence of the temperature gradient, liquefying and accumulating into a rich aroma phase. The uncondensed residual gaseous phase flows through a demister to remove micron-sized droplets and then directly enters the extremely low-temperature second-stage cryogenic zone, where trace amounts of hydrophobic ester compounds are forcibly captured and liquefied. Finally, this high-concentration hydrophobic ester liquid phase is directly refluxed into the degassed aroma oils to compensate for the losses caused by the degassed process and to reshape the complete characteristic aroma profile of the system.

[0019] In the final S5 blending stage, the aroma oils retained from the initial stage are combined with the cooled pigment oils from the later stage in a specific ratio to create a complex red oil with a bright red color and rich ester aroma. Simultaneously, the separated, rich-fragrant phase is evenly sprayed back onto the surface of the oil extraction residue, which has lost its oil content, allowing the residue to reabsorb the hydrophilic fermentation flavor substances lost earlier. Finally, the complex red oil is mechanically stirred and blended with the treated residue, ensuring that while the oil phase is purified, the core flavor components within the material system are reconstituted without damage, guaranteeing the sensory consistency of the final product.

[0020] Preferably, the fermented soybean substrate has a moisture content of 55.0wt% to 65.0wt%, a total salinity of 12.0wt% to 16.0wt%, and an internal solid particle size distribution range of 0.5mm to 5.0mm; the vegetable oil is refined organic pressed rapeseed oil, with a free fatty acid content of no more than 0.2wt% and a moisture content of no more than 0.05wt%.

[0021] By employing the above technical solutions, the moisture content and total salinity of the fermented soybean substrate are controlled within a certain range, which effectively maintains the rheological rigidity of the porous structure framework of the biomass material, making the substrate less prone to clumping or collapse under hydraulic pressure. Simultaneously, the solid particle size distribution within this range not only ensures moderate resistance during liquid-phase permeation but also provides a sufficient mass transfer interface for the extraction process. Strict limits are imposed on the free fatty acids and moisture content of the vegetable oil used as a solvent, reducing the absorption interference of its own impurities on the spectral monitoring signal and cutting off the potential hydrolytic rancidity pathway during the high-temperature extraction stage.

[0022] Preferably, in S1, the thickness of a single layer of fermented soybean substrate in the split-type porous media basket is controlled to be 5.0cm to 15.0cm; the split-type porous media baskets filled with fermented soybean substrate are stacked in multiple layers in the extraction vessel; the conditions for purging the extraction vessel with N2 are as follows: the jacket temperature of the extraction vessel is controlled to be 35℃ to 45℃, the gauge pressure of N2 introduced from the top of the extraction vessel is controlled to be 0.01MPa to 0.05MPa, the flow rate of N2 is controlled to be 2.0L / min to 5.0L / min, and the continuous N2 introduction time is 10min to 20min; the free brine phase is discharged from the separation valve at the bottom of the extraction vessel; the near-infrared band is 1400nm to 1450nm.

[0023] By adopting the above technical solution, the process controls the thickness of a thin single-layer material layer and combines it with multi-layer material basket stacking, mainly to shorten the pressure drop loss during axial penetration of the fluid and make the solvent distribution more uniform in the radial direction. Specific combinations of nitrogen pressure and flow parameters can establish stable and effective forced convection within the reactor, accelerating the vaporization and removal of moisture adhering to the pore surface. In terms of monitoring technology, the near-infrared band of 1400nm to 1450nm precisely covers the frequency doubling absorption region of the OH bonds within polar water molecules, resulting in a very high signal-to-noise ratio for the acquired aqueous phase optical signals.

[0024] Preferably, in S2, the negative pressure state is achieved by introducing recycled N2 to reduce the O2 volume fraction in the headspace region of the extraction vessel to 1.0%–2.0%, turning on the dry vacuum pump to reduce the gauge pressure in the extraction vessel to -0.08 MPa–-0.06 MPa, and maintaining the pressure reduction for 5–10 minutes; the flow rate of preheated vegetable oil pumped from the bottom of the extraction vessel is 0.5 BV / h–1.0 BV / h, the temperature of the preheated vegetable oil is 40°C–45°C, and the N2 is mixed into the extraction vessel to return the pressure to atmospheric pressure at a pressurization rate of 0.01 MPa / min–0.03 MPa / min.

[0025] By employing the above technical solution and strictly controlling trace amounts of oxygen in the headspace region, the peroxidation of polyunsaturated fatty acid chains during processing can be suppressed. Maintaining an initial vacuum of -0.08 MPa to -0.06 MPa within the extraction vessel provides sufficient pressure differential, allowing preheated vegetable oil pumped in at a low flow rate to quickly overcome capillary resistance and penetrate the microporous structure. Subsequently, nitrogen gas is slowly introduced to provide backpressure. This gradual increase in pressure exerts additional mechanical compression on the gas-liquid-solid three-phase interface within the pores, causing the old fluid boundary layer to break down and new mass transfer channels to form.

[0026] Preferably, in S2, the pressure recovery curve is fitted using a first-order exponential decay model to calculate the time constant. The first-order exponential decay model is P(t) = P ∞ +(P0-P ∞ )e -t / τ P(t) is the pressure at time t, P0 is the pressure at the start of the back pressure, P ∞ The equilibrium pressure after the pressure recovers and stabilizes is given by τ, which is a time constant; at least three pulse cycles are executed.

[0027] By employing the above technical solution, the pressure curve is fitted using a first-order exponential decay model, and the time constant τ is extracted as the core judgment index. Unlike simply monitoring the pressure difference, the time constant, as a kinetic parameter, can effectively shield against system fluctuations caused by differences in the hardware volume of the extraction vessel or variations in the amount of material loaded each time. Furthermore, the mandatory execution of at least three cycles of pulsed backpressure ensures that even micropores in deep and blind-end regions can complete the capillary degassing and solvent replacement process, thereby constructing a complete and continuous solid-liquid contact surface.

[0028] Preferably, in step S3, the flow rate of the vegetable oil pumped from the bottom of the extraction vessel is 1.0 BV / h to 2.0 BV / h, the temperature of the vegetable oil is 40℃ to 50℃, and both the aroma oil in the first stage and the pigment oil in the second stage are collected by overflow from the top of the extraction vessel.

[0029] By employing the above technical solution, vegetable oil is injected from bottom to top at a low rate of 1.0 BV / h to 2.0 BV / h, coupled with a suitable temperature of 40℃ to 50℃, forming a typical laminar flow and flat-push flow pattern within the fixed bed. This flow state avoids localized tumbling and fine powder entrainment phenomena that may be caused by high-velocity turbulence. Simultaneously, the top overflow collection method utilizes the density difference of the fluid and the upward buoyancy to cleanly and efficiently push the initially turbid leading liquid section, containing some trace bubbles and particulate impurities, away from the solid phase bed region.

[0030] Preferably, in step S3, the temperature of the vegetable oil is increased to 80℃ to 85℃ at a heating rate of 2.0℃ / min to 5.0℃ / min; the absorbance of the pigment oil at 460nm is monitored, and the oil pump is turned off when the rate of change of absorbance of the pigment oil at 460nm drops to 0.001AU / min.

[0031] By employing the above technical solution, the rate of increase in oil inlet temperature is limited primarily to prevent the solid biomass skeleton from shrinking rapidly upon encountering instantaneous thermal shock, thus avoiding the re-closure of previously opened channels or the triggering of violent boiling of deep bound water. The system utilizes a 460nm wavelength to specifically track the dissolution dynamics of pigments in the downstream stage. Once the rate of change in absorbance at this wavelength approaches a small limit, the oil inlet pump is immediately shut off. This ensures complete pigment removal while preventing subsequent meaningless solvent idling and heat energy waste.

[0032] Preferably, in S4, the negative pressure degassing treatment conditions are as follows: the front-end aroma oil is pumped into the degassing tank, and N2 is introduced from the bottom of the degassing tank for aeration at a temperature of 35℃~40℃ and the system pressure in the degassing tank is reduced to -0.095MPa~-0.090MPa. The aeration rate of N2 is controlled at 0.05vvm~0.10vvm; the temperature of the first-stage condensation zone is 2℃~5℃; the demister is a 316L knitted mesh demister; and the temperature of the second-stage cryogenic zone is -20℃~-15℃.

[0033] By employing the above technical solution, in addition to maintaining a vacuum negative pressure, a small amount of nitrogen aeration is superimposed at the bottom of the degassing tank. The rising nitrogen bubbles act as a carrier gas, greatly expanding the gas-liquid contact area and lowering the partial pressure of volatile components in the gas phase, forcing water vapor and low-boiling-point substances to escape more quickly. The discharged mixed gas undergoes a two-stage temperature reduction process: the first-stage condensation zone is kept above the freezing point, liquefying water vapor while avoiding the risk of icing and pipe blockage; subsequently, the demister uses mechanical inertial impaction to intercept suspended micro-droplets; finally, only truly difficult-to-liquefy hydrophobic small-molecule esters enter the second-stage cryogenic zone, where they lose kinetic energy and undergo phase transition at extremely low temperatures. This staged interception ensures the purity of the substances in each reflux stream.

[0034] Preferably, in S4, the liquefied hydrophobic esters are refluxed back to the pre-fragmented aroma oil after negative pressure degassing at a volume ratio of 0.1 vol% to 0.5 vol%. The non-condensable gas discharged from the second-stage cryogenic zone passes through the 3A-type zeolite molecular sieve drying tower and is then compressed and transported back to the extraction vessel by a dry vacuum pump.

[0035] By employing the above technical solution, the captured high-purity hydrophobic esters are refluxed into the aroma oil at an extremely low volume ratio. Strictly limiting the addition amount is to prevent excessive local accumulation of similar components, which could lead to microscopic stratification or precipitation within the oil phase. The pore size of the 3A-type molecular sieve installed on the exhaust pipe is perfectly suited to intercept escaping water molecules. The purified, dried, non-condensable gas is recompressed and returned to the extraction vessel, establishing a closed-loop gas circulation system and completely eliminating the environmental pressure of emitting volatile organic compounds (VOCs) to the outside.

[0036] Preferably, in step S5, the temperature of the pigment oil after cooling treatment is 40°C, and the aroma oil from the first stage and the pigment oil after cooling treatment are mixed and blended in a mass ratio of 1:2 to 1:4 to obtain red oil; the aroma-rich phase is sprayed into the oil extraction residue with an atomized particle size of 20μm to 50μm using an ultrasonic atomizing nozzle; the mixing ratio of red oil to sprayed residue is 20.0wt% to 30.0wt% of red oil and 70.0wt% to 80.0wt% of sprayed residue, and the conditions for mixing to obtain the finished product are stirring for 15 to 20 minutes under a gauge pressure of -0.05MPa.

[0037] By adopting the above technical solution, the pigment oil in the later stage is required to be cooled to 40°C before being mixed with the aroma oil in the earlier stage. This directly eliminates the problem of secondary flash evaporation and loss of aroma esters caused by temperature differences between hot and cold liquid flows. The subsequent mass ratio combination found a sensory balance point in color, aroma, and texture. It is worth noting that during compounding, the ultrasonic atomizing nozzle disperses the liquid, rich-fragrant phase into micron-sized droplets, allowing it to spread evenly and densely on the pore surface of the oil extraction residue. Finally, mechanical stirring is performed under a slightly negative pressure environment, which in turn removes any air bubbles that may have been entrained during the mixing process, promoting a tight physical fusion of the free oil, water phase, and solid fibrous skeleton.

[0038] This invention provides a production process for organic red chili bean paste that enhances flavor. It has the following beneficial effects:

[0039] 1. This invention effectively preserves heat-sensitive flavor compounds. The process employs a two-stage gradient extraction. The initial stage extracts aroma oils at a lower temperature, avoiding the volatilization and thermal degradation of flavor compounds caused by traditional high-temperature continuous boiling. Combined with subsequent negative pressure degassing, the escaped gas is liquefied in a cryogenic zone, trapping hydrophobic esters and then refluxed, compensating for the unavoidable aroma loss during degassing and thus maintaining the integrity of the product's characteristic flavor.

[0040] 2. This invention reduces the probability of oil emulsification and turbidity, and improves the clarity of the oil phase product. In the initial extraction stage, nitrogen purging combined with near-infrared absorbance monitoring removes free water from the surface and shallow pores of the soybean matrix. Subsequently, in the negative pressure degassing process, water vapor entrained in the oil phase is liquefied and separated through the first-stage condensation zone. This physical dehydration mechanism before and after extraction reduces the mixing and contact between liquid water and oil, preventing oil-water emulsification.

[0041] 3. This invention improves the uniformity of the solid-liquid extraction process. The process introduces a negative pressure pulse pre-wetting operation. After establishing negative pressure in the extraction vessel, vegetable oil is injected, followed by nitrogen backpressure. Relying on the periodic alternating pressure changes, the vegetable oil is forced to penetrate and fill the internal pores of the soybean matrix, expelling blind-end gas. Based on this, a first-order exponential decay model is used to monitor the time constant to determine the fluid displacement progress, reducing fluid deviation within the fixed bed and expanding the actual solid-liquid mass transfer area.

[0042] 4. This invention achieves complete utilization of the material components within the system. The process combines the aroma oil from the initial extraction stage with the pigment oil from the subsequent extraction stage to create a compound red oil. Simultaneously, the rich aroma phase recovered through degassing and condensation is re-attached to the residual mash after oil extraction via atomized spraying. Finally, the red oil and residual mash are mechanically mixed under a slight negative pressure environment. This recombination process restores and mixes the separated oil phase, aqueous flavor components, and solid matrix in a proportional manner, avoiding the waste of processing by-products and ensuring the sensory consistency of the final product. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the production process of organic red oil broad bean paste according to the present invention;

[0044] Figure 2 This is a schematic diagram illustrating the dynamic changes in the pressure difference between the inlet and outlet of the bed during the initial extraction process in Examples 1 to 3 of the present invention.

[0045] Figure 3 This is a schematic diagram showing the distribution of moisture and flavor substance retention rates in Examples 1 to 3 of the present invention;

[0046] Figure 4 This is a schematic diagram comparing the dynamic evolution of fluid pressure difference and extraction yield during the extraction process of Example 1 and Comparative Example 1 of the present invention;

[0047] Figure 5 This is a schematic diagram of the comparative test of the effects of purity and thermal degradation on Example 1 of the present invention and related comparative examples. (a) shows the distribution of light transmittance turbidity of the front aroma oil of different batches of each test object, and (b) shows the comparative results of the 5-hydroxymethylfurfural content in the final red oil product.

[0048] Figure 6 The following is a schematic diagram of the comparison test between cryogenic anti-blocking and flavor retention in Example 1 and the comparative example of the present invention. (a) shows the distribution of the total retention rate of polar characteristic flavors after the fifth batch of operation. (b) shows the trend of the change of the gas phase pressure drop in the cryogenic zone with the operation batch. Detailed Implementation

[0049] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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.

[0050] The main raw materials and reagents used in the following examples and comparative examples have the following sources and specifications. Reagents not specifically mentioned are all commercially available analytical grade or higher grade products.

[0051] Mature organic fermented broad bean paste is a composite biomass matrix made from a mixture of broad beans and chili peppers after fermentation and maturation. It has a moisture content of 55.0wt% to 65.0wt%, a total salinity of 12.0wt% to 16.0wt%, and an internal solid particle size distribution range of 0.5mm to 5.0mm.

[0052] The refined organic pressed rapeseed oil has the CAS number 8002-13-9. As an extraction solvent phase, the free fatty acid content does not exceed 0.2wt% and the moisture content does not exceed 0.05wt%.

[0053] High-purity nitrogen gas, CAS number 7727-37-9, with a purity of not less than 99.99%, is used as an inert protective medium and a gas scavenging medium.

[0054] The 3A type zeolite molecular sieve has the CAS number 308080-99-1 and an effective pore size of 0.3 nm. It is used as a gas drying and protective medium.

[0055] In this manual, BV / h refers to the hourly oil inflow volume multiple, measured by the apparent bulk volume of the fermented soybean substrate packed in the split porous media basket; vvm refers to the minutely aeration volume multiple, measured by the volume of the aroma oil to be treated in the deaeration tank. Both the online near-infrared process analyzer and the online turbidity meter are located in a switchable bypass sampling loop, switching between the bottom separation valve drain branch and the top overflow branch via a valve assembly. Signals in the 1400nm to 1450nm band are analyzed using normalized absorbance A. * Characterization, A * =A t / A ref ×100%, where A t A represents the integrated absorbance measured at any time t during the extraction process in the wavelength range of 1400 nm to 1450 nm. ref This represents the integrated absorbance of the free brine phase discharged from the bottom separation valve at the end of the gas sweep in the 1400 nm to 1450 nm wavelength range. The rate of change of the time constant between adjacent pulse cycles is expressed as |τ n -τ n-1 | / τ n-1 Calculate by multiplying by 100%, where τ n τ is the time constant obtained by fitting the nth pulse cycle. n-1 The time constant is obtained by fitting the pulse cycle for the (n-1)th time. The pressure difference fluctuation rate over 5 consecutive minutes is calculated by dividing the difference between the maximum and minimum pressure difference between the inlet and outlet of the bed within the corresponding time window by 5 minutes. The volume fraction of O2 in the headspace of the extraction vessel is determined by an online oxygen content analyzer located at the headspace sampling port. An online UV-Vis detector is located in the top overflow branch or its bypass sampling loop to monitor the absorbance change of the downstream pigment oil at 460 nm.

[0056] Preparation Example 1: This preparation example provides a method for assembling and calibrating a split-type porous shallow bed extraction assembly, including the following steps:

[0057] Immerse a 316L stainless steel split-type porous media basket in a 95% ethanol aqueous solution for 15 minutes. Rinse the surface thoroughly with deionized water until the conductivity of the washing solution deviates from the background value of the deionized water by no more than 5%. Transfer it to a 60℃ hot air drying oven and dry it for 30 minutes. Lay a commercial 0.5mm PTFE mesh liner flat at the bottom of the dried basket. Connect the online near-infrared process analyzer and the online turbidimeter to the switchable bypass sampling loop. Pass deionized water into the sampling loop. After the liquid phase fills the flow cell of the online near-infrared process analyzer, perform instrument zero-point calibration in the 1400nm to 1450nm wavelength range. Then, inject 0.1 NTU of formazan polymer standard suspension into the flow cell of the online turbidimeter and perform baseline calibration of the transmitted light signal.

[0058] Preparation Example 2: This preparation example provides a standardized pretreatment method for fermented broad bean substrate, including the following steps:

[0059] 100 kg of mature organic fermented soybean mash was weighed and placed in a double ribbon mixer. The mixture was mechanically stirred continuously at 30 r / min for 20 minutes at an ambient temperature of 25°C. 100 g of material was weighed from five different spatial locations inside the mixer and mixed thoroughly. 50 g of the mixed sample was placed in a 105°C atmospheric pressure drying oven and dried continuously for 4 hours to determine the moisture content of the calibration batch. Another 10 g of the mixed sample was used to determine the total salinity of the calibration batch using silver nitrate volumetric precipitation titration. The remaining homogenized material was sealed and stored at 25°C for later use.

[0060] Example 1: This example provides a production process for organic red chili bean paste that enhances flavor, combined with... Figure 1 As shown, it includes the following steps:

[0061] The homogeneous fermented soybean substrate that underwent standardized pretreatment in Preparation Example 2 was spread evenly and filled into the assembled and calibrated split-type porous media basket in Preparation Example 1, with a single layer thickness controlled at 10.0 cm. The four filled baskets were stacked together and hoisted into a jacketed extraction vessel. The extraction vessel was sealed, and water was circulated through the jacket to maintain the internal temperature at 40°C. N2 was introduced from the top of the vessel for purging, with the top gauge pressure controlled at 0.03 MPa and the gas flow rate at 3.5 L / min, for a continuous purging time of 15 minutes. The free brine phase discharged from the bottom separation valve was then introduced into a switchable bypass sampling loop, and its integrated absorbance A in the 1400 nm to 1450 nm wavelength range was recorded. ref It is defined as the 100% polar aqueous phase background reference value.

[0062] Recycled N2 was introduced through a closed-loop circulation system to reduce the O2 volume fraction in the headspace of the extraction vessel to 1.5%. A dry vacuum pump was then activated to reduce the gauge pressure inside the extraction vessel to -0.07 MPa and maintain this pressure for 8 minutes. Preheated refined organic pressed rapeseed oil (42°C) was pumped in from the bottom at a flow rate of 0.75 BV / h, while N2 was simultaneously introduced at a pressurization rate of 0.02 MPa / min to restore the pressure to atmospheric pressure. Pressure recovery curves were collected for each pressure recovery process. The first-order exponential decay model of the pressure recovery curve can be expressed as P(t) = P ∞ +(P0-P ∞ )e -t / τ Where P(t) is the pressure at time t, P0 is the pressure at the start of the back pressure, and P ∞ The equilibrium pressure after the pressure recovers and stabilizes is given, and τ is the fitted time constant. Pre-wetting is performed for at least 3 pulse cycles; starting from the 3rd cycle, pre-wetting is terminated when the rate of change of the time constant of the adjacent cycle drops to 4.0%.

[0063] The bottom injection oil inlet temperature was set at 45℃, and the inlet flow rate was 1.5 BV / h. The aroma oil was collected from the top overflow of the extraction vessel. The top overflow oil phase was continuously monitored; when the online near-infrared normalized absorbance A... * When the turbidity of the overflow phase measured by the online turbidity meter is not higher than 12.0 NTU, and the fluctuation rate of the differential pressure between the inlet and outlet of the bed is not higher than 0.003 MPa / min for 5 consecutive minutes, the bottom-injection vegetable oil heater is heated to 82℃ at a heating rate of 3.5℃ / min to perform the second stage of extraction. The absorbance of the pigment oil in the downstream section of the overflow is monitored at 460 nm by an online UV-Vis detector. When the absorbance change rate drops to 0.001 AU / min, the oil pump is turned off.

[0064] The initial aroma oil is pumped into the degassing tank, where the temperature is maintained at 38°C. The system pressure is reduced to -0.092 MPa, and N2 is introduced through the commercial sintered metal microporous gas distributor at the bottom of the tank, with the aeration rate controlled at 0.08 vvm. The extracted gas phase enters the first-stage condensation zone at 3°C, where the rich aroma phase is condensed and collected. The uncondensed gas phase passes through a 316L knitted mesh demister and enters the second-stage cryogenic zone at -18°C, where the liquefied hydrophobic esters are collected and refluxed back to the degassed initial aroma oil at a volume ratio of 0.3 vol%. The non-condensable tail gas passes through a 3A-type zeolite molecular sieve drying tower and is compressed and returned to the extraction vessel by a dry vacuum pump.

[0065] The aroma oil from the refluxing hydrophobic esters was mixed with the pigment oil from the heat exchanger (cooled to 40°C) at a mass ratio of 1:3 to prepare red oil. The extraction residue was removed from the feed basket and poured out. The retained aroma-rich phase was then sprayed into the extraction residue using an ultrasonic atomizing nozzle with a particle size of 35 μm. The two were then added to a mixer at a ratio of 25.0 wt% red oil and 75.0 wt% treated extraction residue, and stirred for 18 minutes under a gauge pressure of -0.05 MPa to obtain the final product.

[0066] Example 2: This example provides a process for producing organic red chili bean paste with enhanced flavor, including the following steps:

[0067] The pretreated fermented soybean substrate from Preparation Example 2 was spread evenly in the calibrated material basket from Preparation Example 1, with a single layer thickness controlled at 5.0 cm. Five layers were stacked and then hoisted into the extraction vessel. The jacket temperature was set to 35℃, the N2 gas pressure was 0.01 MPa, and the flow rate was 2.0 L / min for 10 minutes. The free brine phase discharged from the bottom separation valve was introduced into the switchable bypass sampling loop, and its integrated absorbance A in the 1400 nm to 1450 nm wavelength range was recorded. ref It is defined as the 100% polar aqueous phase background reference value.

[0068] The headspace O2 volume fraction was reduced to 2.0%, and the gauge pressure inside the extraction vessel was reduced to -0.06 MPa and maintained for 5 minutes. Refined organic pressed rapeseed oil preheated to 40°C was pumped in from the bottom at a flow rate of 0.5 BV / h, while N2 was simultaneously introduced at a pressurization rate of 0.01 MPa / min to return the pressure to atmospheric pressure. Pre-impregnation was performed for at least 3 pulse cycles; from the 3rd cycle onwards, pre-impregnation was terminated when the rate of change of the time constant between adjacent pulse cycles decreased to 5.0%.

[0069] The bottom injection oil inlet temperature was set at 40℃, and the inlet flow rate at 1.0 BV / h. The overflow oil from the front section was collected. The top overflow oil phase was continuously monitored; when the online near-infrared normalized absorbance A... * When the turbidity of the overflow phase measured by the online turbidity meter is not higher than 5.0%, and the fluctuation rate of the differential pressure between the inlet and outlet of the bed is not higher than 0.005 MPa / min for 5 consecutive minutes, the bottom-injection vegetable oil heater is heated to 80℃ at a heating rate of 2.0℃ / min to perform the second stage of extraction. When the absorbance change rate of the pigment oil in the overflow section at 460nm drops to 0.001 AU / min as monitored by the online UV-Vis detector, the oil pump is turned off.

[0070] The initial aroma oil enters the degassing tank at 35℃ and -0.090MPa. The commercial sintered metal microporous gas distributor at the bottom of the tank is turned on to control the N2 aeration rate at 0.05vvm. The gas phase enters the first-stage condensation zone at 5℃ to collect the water-rich phase. The uncondensed gas phase passes through a 316L knitted mesh demister and enters the second-stage cryogenic zone at -15℃. The liquefied hydrophobic esters are collected and refluxed to the degassed initial aroma oil at a volume ratio of 0.1 vol%. The tail gas is returned to the extraction vessel via a drying tower.

[0071] The aroma oil from the first stage and the pigment oil from the second stage, cooled to 40°C by a heat exchanger, are mixed and blended at a mass ratio of 1:2. The residual mash is removed, and the aroma-rich phase is sprayed into the residual mash with a particle size of 50μm using an ultrasonic atomizing nozzle. The finished product is obtained by stirring for 15 minutes under a gauge pressure of -0.05MPa at a ratio of 30.0wt% red oil and 70.0wt% sprayed residual mash.

[0072] Example 3: This example provides a process for producing organic red chili bean paste with enhanced flavor, including the following steps:

[0073] The pretreated fermented soybean substrate from Preparation Example 2 was spread evenly in the calibrated material basket from Preparation Example 1, with a single layer thickness controlled at 15.0 cm. Three layers were stacked and then hoisted into the extraction vessel. The jacket temperature was set to 45℃, the N2 gas pressure was 0.05 MPa, and the flow rate was 5.0 L / min for 20 minutes. The free brine phase discharged from the bottom separation valve was introduced into the switchable bypass sampling loop, and its integrated absorbance A in the 1400 nm to 1450 nm wavelength range was recorded. ref It is defined as the 100% polar aqueous phase background reference value.

[0074] The headspace O2 volume fraction was reduced to 1.0%, and the gauge pressure inside the extraction vessel was reduced to -0.08 MPa and maintained for 10 minutes. Refined organic pressed rapeseed oil preheated to 45°C was pumped in from the bottom at a flow rate of 1.0 BV / h, while N2 was simultaneously introduced at a pressurization rate of 0.03 MPa / min to return the pressure to atmospheric pressure. Pre-impregnation was performed for at least 3 pulse cycles; from the 3rd cycle onwards, pre-impregnation was terminated when the rate of change of the time constant between adjacent pulse cycles decreased to 3.0%.

[0075] The bottom injection oil inlet temperature was set at 50℃, and the inlet flow rate at 2.0 BV / h. The overflow oil from the front section was collected. The top overflow oil phase was continuously monitored; when the online near-infrared normalized absorbance A... *When the turbidity of the overflow phase measured by the online turbidity meter is not higher than 2.0%, and the fluctuation rate of the differential pressure between the inlet and outlet of the bed is not higher than 0.002 MPa / min for 5 consecutive minutes, the bottom-injection vegetable oil heater is heated to 85℃ at a heating rate of 5.0℃ / min to perform the second stage of extraction. When the absorbance change rate of the pigment oil in the downstream section at 460nm drops to 0.001 AU / min as monitored by the online UV-Vis detector, the oil pump is turned off.

[0076] The initial aroma oil enters the degassing tank at 40℃ and -0.095MPa. The commercial sintered metal microporous gas distributor at the bottom of the tank is turned on to control the N2 aeration rate at 0.10vvm. The gas phase enters the first-stage condensation zone at 2℃ to collect the water-rich phase. The uncondensed gas phase passes through a 316L knitted mesh demister and enters the second-stage cryogenic zone at -20℃. The liquefied hydrophobic esters are collected and refluxed to the degassed initial aroma oil at a volume ratio of 0.5 vol%. The tail gas is returned to the extraction vessel via a drying tower.

[0077] The aroma oil from the first stage and the pigment oil from the second stage, cooled to 40°C by a heat exchanger, are mixed and blended at a mass ratio of 1:4. The residual mash is removed, and the aroma-rich phase is sprayed into the residual mash with a particle size of 20μm using an ultrasonic atomizing nozzle. The finished product is obtained by stirring for 20 minutes under a gauge pressure of -0.05MPa at a ratio of 20.0wt% red oil and 80.0wt% sprayed residual mash.

[0078] Comparative Example 1:

[0079] Compared with Example 1, the difference is that: instead of using a split porous media basket, the same total weight of homogeneous fermented soybean substrate is directly filled into the porous plate at the bottom of the extraction vessel to form a single continuous bed layer with a thickness of 40.0 cm. All other aspects are the same.

[0080] Comparative Example 2:

[0081] Compared with Example 1, the difference is that in the pre-soaking stage, the dry vacuum pump is turned on to reduce the pressure inside the extraction vessel to -0.098 MPa and maintain it for 8 minutes, while the rest are the same.

[0082] Comparative Example 3:

[0083] Compared with Example 1, the difference is that: the online near-infrared process analyzer, online turbidity meter and bed inlet and outlet differential pressure are not monitored together; after the bottom injection oil is continuously extracted at 45°C for 60 minutes, the set value of the bottom injection vegetable oil heater is directly raised to 82°C and the heating rate is not controlled; all other aspects are the same.

[0084] Comparative Example 4:

[0085] Compared with Example 1, the difference is that: no first-stage condensation zone and second-stage cryogenic zone are set up, the gas phase extracted from the degassing tank directly enters the single-stage condensation zone at a temperature of 20°C for condensation, and the collected liquid phase is directly discharged; after taking out the material basket and pouring out the oil extraction residue, no liquid phase spraying operation is performed, and the oil extraction residue and red oil are directly put into the mixer according to the ratio, and the rest are the same.

[0086] Comparative Example 5:

[0087] Compared with Example 1, the difference is that: in the gas phase pipeline between the first-stage condensation zone and the second-stage cryogenic zone, a 316L knitted mesh demister is not installed, and the uncondensed gas phase directly enters the second-stage cryogenic zone with a temperature of -18°C, while the rest are the same.

[0088] Test Example 1:

[0089] Test objective: To verify the engineering stability of the fluid seepage process under different parameter settings, and the accuracy of the multi-sensor joint matrix in determining the fluid phase reversal node.

[0090] The experimental steps are as follows:

[0091] The extraction vessel and piping system in Examples 1, 2, and 3 were used as test objects. In the initial aroma oil bottom injection extraction stage, differential pressure data of the inlet and outlet of the bed were collected at a sampling period of 1 second by a differential pressure transmitter connected in parallel at both ends of the bottom liquid inlet main pipe and the top overflow pipe of the extraction vessel.

[0092] Based on the collected pressure difference time series, the maximum inlet and outlet pressure difference peak values ​​of each embodiment during the fluid immersion and elution cycle are extracted, and the highest pressure difference fluctuation rate within any consecutive 5-minute interval is calculated.

[0093] At the moment when the control system triggers the bottom-injection vegetable oil heater to heat to the set value, the absorbance value of the characteristic band of the online near-infrared process analyzer and the transmittance turbidity value of the online turbidimeter are recorded simultaneously, and the data are summarized and compared.

[0094] The experimental results are shown in Table 1:

[0095] Table 1: Test results of fluid dynamic stability and phase identification accuracy of the whole system in Examples 1 to 3

[0096]

[0097] Combination Figure 2Based on the data in Table 1, Examples 1 to 3 demonstrated hydrodynamic stability within the set parameter boundaries. In actual scale-up production, when processing high-viscosity, wet fermentation biomass, solid particles are prone to compaction under fluid scouring and gravity. This physical compaction phenomenon causes the system pressure drop to increase exponentially. Test results show that the maximum inlet and outlet pressure difference throughout the entire seepage cycle is below 0.0521 MPa, not reaching the set threshold for triggering process shutdown or pump stalling, and the highest pressure difference fluctuation rate is also controlled within the range of several thousand Pascals per minute. The spatial structure design of the porous shallow bed provides physical support, blocking the cumulative transmission channel of longitudinal compressive stress in the deep space. Whether in a 5cm thin bed or a 15cm thick bed condition, the bottom-injected fluid wavefront can maintain a laminar flow state within its independent compartment, alleviating the problem of reduced mass transfer efficiency caused by local channeling short-circuiting.

[0098] The stability of the fluid permeation boundary provides a foundation for the subsequent intervention of a multi-sensor optical judgment system. During the final stage of fixed-bed extraction with polar aqueous phase replacement, water and hot oil easily form an emulsion drag layer. Traditional processes, relying solely on time estimation or manual observation, struggle to accurately determine the timing of segmentation, easily leading to the boiling entrainment of residual aqueous phase or thermal degradation of the matrix. This process employs a combined judgment logic of near-infrared absorbance and transmittance turbidity to quantify the characteristics of free water discharge and changes in oil phase transparency. When a water-in-oil to oil-in-water transition occurs within the bed, the optical attenuation and scattering response data collected by the sensors in the three embodiments reach the preset trigger threshold range. This data-driven judgment mechanism controls the free water content at the matrix interface during subsequent heated oil bottom injection jumps, limiting flavor compound decomposition and browning side reactions caused by thermal shock at the source.

[0099] Unless otherwise specified, in Test Examples 2 and 5, solid-phase microextraction gas chromatography-mass spectrometry (SPME-GC-MS) analysis was performed using 50 / 30 μm DVB / CAR / PDMS extraction fibers; the samples were equilibrated at 40 °C for 10 min, extracted for 30 min, and desorbed at 250 °C for 5 min at the gas chromatograph injection port. The chromatographic column was a DB-WAX capillary column (30 m × 0.25 mm × 0.25 μm), the carrier gas was high-purity helium, and the flow rate was 1.0 mL / min; the temperature program was as follows: initial temperature of 40 °C held for 3 min, increased to 220 °C at 5 °C / min and held for 5 min; quantification was performed using the external standard method. In Test Example 4, 5-hydroxymethylfurfural was detected using a C18 reversed-phase column (250 mm × 4.6 mm, 5 μm), with a mobile phase of methanol-water (volume ratio 15:85), a flow rate of 1.0 mL / min, a column temperature of 30 °C, a detection wavelength of 284 nm, and an injection volume of 10 μL. An external standard method was used to establish a standard curve for quantification.

[0100] Test Example 2:

[0101] Test objective: To verify the effectiveness of the multi-stage fractionation mechanism in retaining volatile substances during the dehydration and degassing stages of the process, and the impact of end-of-pipe atomization reprocessing on the overall flavor retention rate.

[0102] The experimental steps are as follows:

[0103] The initial organically fermented soybean mash and the final product after compound processing in Examples 1, 2, and 3 were used as test objects. Karl Fischer titration combined with gravimetric method was used to determine the total moisture content of the initial material at the start of the process and the total moisture content of the finished product at the end of the process in each example, and the moisture recovery rate was calculated.

[0104] The moisture recovery rate is calculated by multiplying the mass ratio of the total moisture content of the final product to the total moisture content of the initial material by 100%; the retention rates of polar aroma components and ethyl linoleate are calculated by multiplying the mass ratio of the total mass of the corresponding components in the final product to the total mass of the corresponding components in the initial material by 100%, respectively.

[0105] Solid-phase microextraction gas chromatography-mass spectrometry (SPME-GC-MS) was used to quantitatively analyze the initial organic fermented soybean mash and the final product of each embodiment. The total mass content of water-soluble polar aroma components isoamyl alcohol and phenethyl alcohol was separated and quantified in the chromatograms, and the retention rate of polar aroma components was calculated based on the mass ratio before and after processing.

[0106] Using the above-mentioned chromatography-mass spectrometry testing conditions, the content of hydrophobic ester component ethyl linoleate in the test object was quantitatively detected, the data were recorded, and the retention rate of ethyl linoleate was calculated.

[0107] The experimental results are shown in Table 2:

[0108] Table 2: Results of Mass Conservation and Flavor Closed-Loop Tracking Tests in Examples 1 to 3

[0109]

[0110] Combination Figure 3Based on the test data in Table 2, Examples 1 to 3 exhibited clear material retention patterns after the entire process chain was run. High-moisture fermentation materials typically experience the release of volatile substances during the heating and vacuum extraction stages. The measured data show that the moisture recovery rate in each example remained above 95%, reflecting the system's ability to limit the loss of gaseous materials. The first-stage condensation zone in the process induced a cooling phase change in the gaseous moisture, and polar components entrained in the gas stream were retained as water vapor liquefied. The total retention rate of polar aroma molecules such as isoamyl alcohol and phenylethanol exceeded 90%, which would easily lead to the extraction and loss of these low-boiling-point substances during conventional negative pressure exhaust processes. The cascade fractionation temperature zone caused a thermodynamic phase change in polar volatiles on the surface of the condenser tube, and the resulting droplets were collected as a fragrant aroma phase, mitigating aroma loss during venting.

[0111] When determining hydrophobic flavor compounds, the retention rate of ethyl linoleate ranged from 92% to 95%. Hydrophobic molecules in the gas phase that failed to be liquefied in the primary temperature zone passed through the demister and were carried into the cryogenic zone. The low temperature environment caused the lipid-soluble gaseous molecules to condense and liquefy. The rich aroma phase intercepted at the front end was re-sprayed into the oil extraction residue using an ultrasonic atomizing device, and the hydrophobic esters recovered in the cryogenic zone were mixed into the aroma oil from the front end. This physical recombining mechanism did not introduce external chemical components; it relied on gas-liquid thermodynamic phase change regulation to recombine the flavor components separated in the intermediate stages. Test data verified the practical engineering role of the separation condensation and material re-injection engineering path in maintaining the overall flavor quality of the fermented soybean matrix.

[0112] Test Example 3:

[0113] Test objective: To analyze the influence of the split porous shallow bed structure on the fluid state of the extraction system and the mass transfer efficiency of oil extraction.

[0114] The experimental steps are as follows:

[0115] The fixed-bed systems of Example 1 and Comparative Example 1 were used as test objects. The extraction program was started while keeping the oil inlet temperature and bottom injection pump speed consistent in both systems, and the inlet and outlet pressure difference threshold for pump shutdown was set to 0.15 MPa in the control system.

[0116] Record the operating time from the start of bottom injection to the triggering of the 0.15MPa pump stop threshold as the limit operating time. If the system differential pressure is lower than the set threshold during the complete process cycle, record the actual total operating time of that batch.

[0117] After the pigment extraction stage is completed, the overflowing pigment oil is collected and weighed. The total yield per batch is calculated by combining this yield with the initial input of vegetable oil. This operation is performed in three parallel experiments, and the average value is taken.

[0118] The experimental results are shown in Table 3:

[0119] Table 3: Runtime and Quality Yield Test Results of Example 1 and Comparative Example 1

[0120]

[0121] Combination Figure 4 Based on the test data in Table 3, the packing method altered the fluid boundary conditions within the extraction system, affecting solvent permeation and mass transfer efficiency. When viscous, wet materials with yield stress are deeply packed, the bottom layer structure is easily compressed by the weight of the material above. Comparative Example 1 used a single continuous packing with a thickness of 40 cm, resulting in compressed bottom layer porosity. At the initial stage of the oil injection process, the fluid resistance in Comparative Example 1 increased rapidly, and the system reached the 0.15 MPa stop-pump warning line between 21.8 and 27.5 minutes. The increased pressure drop altered the propagation state of the fluid wavefront within the bed, causing the extraction solvent to flow towards areas of lower resistance, forming localized short-circuit channeling.

[0122] After the channel is formed, the bottom layer of vegetable oil flows out along the channel, and the material in the surrounding compacted section forms a contact dead zone, making it difficult for the pigment components inside the solid particles to enter the oil phase for mass transfer. In Comparative Example 1, the average extraction yield of pigment oil in the later stage was 34.23%, with some extractable components remaining inside the matrix. Example 1 used a split-type material basket to divide the continuous deep bed into physically isolated 10cm shallow beds. The single-layer thickness limitation reduced the accumulation of stress in the vertical direction, and the material in each layer maintained its original loose porosity distribution. This structural adjustment enabled the system to maintain stable operation within a 125-minute extraction cycle, with an average pigment oil yield of 91.45% in the later stage. The split shallow bed structure reduced the fluid short-circuiting phenomenon caused by the accumulation of sticky and wet materials, increased the contact area between the oil and solid phases, and improved the extraction effect of the extraction medium on the matrix components.

[0123] Test Example 4:

[0124] Test objective: To analyze the effects of vacuum parameters and temperature control conditions on the amount of particulate entrainment in the extracted oil phase and the degree of thermal degradation side reactions.

[0125] The experimental steps are as follows:

[0126] Fixed-bed oil extraction was carried out in three consecutive batches according to the process conditions of Example 1, Comparative Example 2, and Comparative Example 3. At the end of the aroma oil collection stage in the early stage of each batch, 500 mL of oil phase sample was collected from the overflow pipeline and allowed to stand and cool to 25°C for later use.

[0127] The transmittance turbidity of the collected aroma oil samples was measured using a benchtop light scattering turbidimeter. Before measurement, the instrument was calibrated using a silicone oil standard solution. The oil sample was then poured into a cuvette, and the turbidity value was read and recorded.

[0128] Each batch of downstream red oil products was collected and centrifuged and filtered. The 5-hydroxymethylfurfural (5-HMF) content in the red oil was quantitatively determined using high-performance liquid chromatography (HPLC) with a UV detector. After ultrasonic extraction with methanol and subsequent volume adjustment, the samples were injected into the chromatographic column, and the peak area was recorded at 284 nm. The mass concentration of 5-HMF was calculated by referring to a standard curve.

[0129] The experimental results are shown in Table 4:

[0130] Table 4: Test results of transmittance and 5-hydroxymethylfurfural content in Example 1 and Comparative Examples 2 to 3

[0131]

[0132] Combination Figure 5 Based on the data in Table 4, the vacuum parameters and heating timing altered the amount of entrained physical particles and the content of thermal degradation products during the oil extraction process. The hydration boundary layer rupture state on the surface of the high-water-content particle matrix is ​​directly related to the system pressure conditions. In Comparative Example 2, the dry vacuum pump was turned on, reducing the system gauge pressure to -0.098 MPa. This pressure environment is below the boiling point critical value of free water within the system. The pressure reduction caused the water inside the pores to boil violently, and the vaporization and expansion of liquid water generated physical impact, leading to the shedding of the cell matrix and colloidal substances. Insoluble particles entered the first-stage aroma oil with the boiling fluid, and their transmittance turbidity measurements ranged from 75.9 to 82.1 NTU. In Example 1, the pulsed vacuum was set near the critical point of aqueous phase boiling, relying on gas expansion stress to displace water, reducing the physical damage to the matrix skeleton, and the turbidity of the first-stage aroma oil remained in the range of 10.8 to 13.5 NTU.

[0133] Different heating sequences resulted in variations in the degree of thermodynamic side reactions. 5-Hydroxymethylfurfural (5-HMF) is a characteristic product of the thermal degradation and Maillard reaction of sugars and is often used to measure the thermal shock of materials. In Comparative Example 3, the bottom heating temperature was raised to 82°C for a fixed time, at which point the free water inside the fixed bed was not completely drained. Localized heating caused thermal stress concentration at the water-oil interface, accelerating the dehydration and cracking of sugars. The 5-HMF content in the red oil product was measured to be between 85.3 and 92.1 mg / kg. In Example 1, the heating program was initiated after the fluid phase reversal based on real-time monitoring data from turbidity and near-infrared sensors. The data-feedforward controlled heating method avoided the thermodynamic environment when a large amount of free water was present, reducing the content of thermal degradation products to 12.8 to 15.1 mg / kg and mitigating adverse thermodynamic side reactions during oil extraction.

[0134] Test Example 5:

[0135] Test objective: To test the effects of multi-stage fractionation and condensation mechanisms and interstage demisting structures on the retention rate of polar flavor components, and to analyze the changes in gas phase pressure drop in the cryogenic section under continuous operation.

[0136] The experimental steps are as follows:

[0137] The process systems of Example 1, Comparative Example 4, and Comparative Example 5 were used as test objects. In Comparative Example 4, the rich water phase generated by the first-stage condensation was directly discarded during the dehydration and degassing stage without end-of-pipe reprocessing; in Comparative Example 5, no demister structure was installed between the first-stage condensation zone and the cryogenic zone.

[0138] Each test object ran the production process for 5 consecutive batches. At the end of the depressurization, dehydration and degassing process of each batch, a differential pressure transmitter was used to record the peak value of the gas phase pressure drop at the inlet and outlet of the cryogenic zone. The differential pressure transmitter was set between the inlet pipe section and the outlet pipe section of the second-stage cryogenic zone.

[0139] The fifth batch of the final product was extracted, and the total mass of isoamyl alcohol and phenylethanol was determined using solid-phase microextraction gas chromatography-mass spectrometry. The polar flavor retention rate was calculated based on the baseline flavor content of the initial materials.

[0140] The experimental results are shown in Table 5:

[0141] Table 5: Test results of pressure drop and flavor retention rate of Example 1 and related comparative examples

[0142]

[0143] Combination Figure 6 Based on the results in Table 5, in Comparative Example 4, the rich-water phase was directly discarded after the first-stage condensation, resulting in a polar flavor retention rate of 46.82%. In Example 1, the rich-water phase obtained from the first-stage condensation was recycled back into the residual mash, resulting in a polar flavor retention rate of 92.47%. This indicates that staged condensation combined with aqueous phase recycling is beneficial for improving the retention level of polar flavor components. In Comparative Example 5, after removing the 316L knitted mesh demister between the first-stage condensation zone and the second-stage cryogenic zone, the pressure drop in the cryogenic section increased from 1.21 kPa in the first batch to 26.84 kPa in the fifth batch. In Example 1, after installing the 316L knitted mesh demister, the pressure drop in the cryogenic section for all five batches remained within the range of 1.15 to 1.29 kPa, indicating that interstage demisting is beneficial for reducing the risk of frost formation and blockage caused by droplets entering the cryogenic zone.

[0144] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A process for producing organic red chili bean paste with enhanced flavor, characterized in that, include: S1. Spread the fermented soybean substrate in a split porous media basket and place it in the extraction vessel; purge the extraction vessel with N2 and obtain the integrated absorbance of the discharged free brine phase in the near-infrared band as the background reference value of the polar aqueous phase. S2. The extraction vessel is placed under negative pressure. Preheated vegetable oil is pumped into the extraction vessel from the bottom and mixed into the N2 back pressure extraction vessel. The pressure recovery curve is collected and the time constant is calculated by fitting. Pulse cycle is executed. When the change rate of the time constant of the adjacent pulse cycle drops to 3.0% to 5.0%, the pre-wetting is terminated. S3. Pump vegetable oil into the bottom of the extraction vessel and collect the first-stage aroma oil; monitor the normalized absorbance, transmittance and turbidity of the first-stage aroma oil, and the fluctuation rate of the differential pressure between the inlet and outlet of the extraction vessel bed for five consecutive minutes. When the normalized absorbance of the first-stage aroma oil drops to 2.0% to 5.0%, the transmittance and turbidity of the first-stage aroma oil drops to 10.0 to 15.0 NTU, and the fluctuation rate of the differential pressure between the inlet and outlet of the extraction vessel bed for five consecutive minutes drops to 0.002 to 0.005 MPa / min, raise the temperature of the vegetable oil and collect the second-stage pigment oil. S4. The front-end aroma oil is subjected to negative pressure degassing treatment; the extracted gas phase is condensed and collected in the first-stage condensation zone to collect the fragrant phase; the gas phase that is not condensed in the first-stage condensation zone is condensed and collected in the second-stage cryogenic zone after passing through the demister to collect the liquefied hydrophobic esters, and the liquefied hydrophobic esters are returned to the front-end aroma oil after negative pressure degassing treatment. S5. Mix the front-stage aroma oil after reflux liquefaction of hydrophobic esters with the rear-stage pigment oil after cooling treatment to make red oil; take out the fermented soybean matrix in the extraction vessel as the oil extraction residue, spray the aroma-rich phase into the oil extraction residue to obtain the sprayed residue; mix the red oil with the sprayed residue to obtain the finished product.

2. The production process for enhancing the flavor of organic red chili bean paste according to claim 1, characterized in that, The fermented soybean substrate has a moisture content of 55.0wt% to 65.0wt%, a total salinity of 12.0wt% to 16.0wt%, and an internal solid particle size distribution range of 0.5mm to 5.0mm; the vegetable oil is refined organic pressed rapeseed oil, and the free fatty acid content of the refined organic pressed rapeseed oil does not exceed 0.2wt%, and the moisture content does not exceed 0.05wt%.

3. The production process for enhancing the flavor of organic red chili bean paste according to claim 1, characterized in that, In S1, the thickness of a single layer of the fermented soybean substrate in the split-type porous media basket is controlled to be 5.0cm to 15.0cm; the split-type porous media baskets filled with the fermented soybean substrate are stacked in multiple layers in the extraction vessel; the conditions for purging the extraction vessel with N2 are as follows: the jacket temperature of the extraction vessel is controlled to be 35℃ to 45℃, the gauge pressure of N2 introduced from the top of the extraction vessel is controlled to be 0.01MPa to 0.05MPa, the flow rate of N2 is controlled to be 2.0L / min to 5.0L / min, and the continuous N2 introduction time is 10min to 20min; the free brine phase is discharged from the separation valve at the bottom of the extraction vessel; the near-infrared band is 1400nm to 1450nm.

4. The production process for enhancing the flavor of organic red chili bean paste according to claim 1, characterized in that, In S2, the negative pressure state is achieved by introducing recycled N2 to reduce the O2 volume fraction in the headspace region of the extraction vessel to 1.0%–2.0%, turning on the dry vacuum pump to reduce the gauge pressure in the extraction vessel to -0.08 MPa–-0.06 MPa, and maintaining the pressure reduction for 5–10 minutes. The flow rate of preheated vegetable oil pumped from the bottom of the extraction vessel is 0.5 BV / h–1.0 BV / h, and the temperature of the preheated vegetable oil is 40°C–45°C. The N2 is mixed into the extraction vessel to reduce the pressure to atmospheric pressure by introducing N2 into the extraction vessel at a pressurization rate of 0.01 MPa / min–0.03 MPa / min.

5. The process for producing flavor-enhancing organic red chili bean paste according to claim 1, characterized in that, In S2, the pressure recovery curve is fitted using a first-order exponential decay model to calculate the time constant. The first-order exponential decay model is P(t) = P ∞ +(P0-P ∞ )e -t / τ P(t) is the pressure at time t, P0 is the pressure at the start of the back pressure, P ∞ The equilibrium pressure after the pressure recovers and stabilizes is given by τ, which is a time constant; at least three pulse cycles are executed.

6. The production process for enhancing the flavor of organic red chili bean paste according to claim 1, characterized in that, In S3, the flow rate of the vegetable oil pumped into the bottom of the extraction vessel is 1.0 BV / h to 2.0 BV / h, the temperature of the vegetable oil is 40℃ to 50℃, and the aroma oil in the first stage and the pigment oil in the second stage are both collected by overflow from the top of the extraction vessel.

7. The process for producing flavor-enhancing organic red chili bean paste according to claim 6, characterized in that, In S3, the temperature of the vegetable oil is increased to 80℃~85℃ at a heating rate of 2.0℃ / min~5.0℃ / min; the absorbance of the pigment oil in the later stage is monitored at 460nm, and the oil pump is turned off when the rate of change of absorbance of the pigment oil in the later stage at 460nm drops to 0.001AU / min.

8. The production process for enhancing the flavor of organic red chili bean paste according to claim 1, characterized in that, In S4, the negative pressure degassing treatment conditions are as follows: the front-end aroma oil is pumped into the degassing tank, and under the conditions of a temperature of 35℃~40℃ and a system pressure reduction of -0.095MPa~-0.090MPa in the degassing tank, N2 is introduced from the bottom of the degassing tank for aeration, and the aeration rate of N2 is controlled at 0.05vvm~0.10vvm; the temperature of the first-stage condensation zone is 2℃~5℃; the demister is a 316L knitted mesh demister, and the temperature of the second-stage cryogenic zone is -20℃~-15℃.

9. The process for producing flavor-enhancing organic red chili bean paste according to claim 8, characterized in that, In S4, the liquefied hydrophobic esters are refluxed back to the front aroma oil after negative pressure degassing at a volume ratio of 0.1 vol% to 0.5 vol% of the front aroma oil volume; the non-condensable gas discharged from the second-stage cryogenic zone passes through the 3A-type zeolite molecular sieve drying tower and is then compressed and transported back to the extraction vessel by a dry vacuum pump.

10. The production process for enhancing the flavor of organic red chili bean paste according to claim 1, characterized in that, In S5, the temperature of the pigment oil after cooling treatment is 40℃. The aroma oil from the first stage and the pigment oil after cooling treatment are mixed and blended at a mass ratio of 1:2 to 1:4 to form red oil. The aroma-rich phase is sprayed into the oil extraction residue with an atomized particle size of 20μm to 50μm using an ultrasonic atomizing nozzle. The mixing ratio of the red oil and the sprayed residue is 20.0wt% to 30.0wt% of red oil and 70.0wt% to 80.0wt% of sprayed residue. The conditions for mixing to obtain the finished product are stirring for 15 to 20 minutes under a gauge pressure of -0.05MPa.

Images