A method for separating EPA and DHA from fish oil

Abstract

This invention relates to a method for separating EPA and DHA from fish oil, belonging to the field of separation technology. First, utilizing the difference in esterification rates of EPA and DHA by lipase RM, a selective esterification reaction of fish oil fatty acids with lauryl alcohol is catalyzed, thereby constructing a polar binary system of low-polarity EPA-lauryl ester and high-polarity DHA fatty acid. Then, a PDA-mSiO2 / PDMS composite membrane, prepared from polydimethylsiloxane, mesoporous silica, and polydopamine in a specific mass ratio, is used as the filter medium. The system is filtered using a dual-action mechanism of "size sieving-hydrogen bond affinity," achieving efficient permeation of EPA-lauryl ester and selective retention of DHA fatty acid. Finally, the crude EPA-lauryl ester solution on the permeation side and the concentrated DHA fatty acid solution on the retention side are purified by low-temperature distillation, achieving efficient and green separation of EPA-lauryl ester and DHA fatty acid.

Description

Technical Field

[0001] This invention belongs to the field of separation technology, and in particular relates to a method for separating EPA and DHA from fish oil. Background Technology

[0002] Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), as core polyunsaturated fatty acids in fish oil, have irreplaceable application value in health foods, pharmaceuticals, and infant formula. Among them, EPA is in high demand in cardiovascular health products due to its strong antioxidant stability and high bioavailability, while DHA, as a key nutrient for infant brain development, has extremely high purity requirements.

[0003] Existing EPA and DHA separation technologies still face numerous bottlenecks: Molecular distillation requires operation at high temperatures of 150℃-180℃ and high vacuum conditions of ≤1Pa, resulting in extremely high energy consumption and a high risk of fatty acid oxidation and degradation, with the final product purity typically only reaching 65%-70%. Urea inclusion methods require 3-5 times the amount of urea and ethanol as solvents, followed by multiple water washes to remove residual urea, generating large amounts of organic wastewater and posing a high risk of solvent residue (≥0.5wt%). Furthermore, urea is not listed in the current catalog of usable food additives, presenting a compliance deficiency. In conventional enzymatic methods combined with traditional membrane separation, existing enzymatic methods primarily focus on glycerol ester synthesis, failing to utilize the difference in esterification rates between EPA and DHA to establish a separation basis. Additionally, conventional PDMS membranes rely solely on hydrophobic interactions or size sieving mechanisms, without introducing hydrogen bonding affinity, resulting in a separation factor ≤2.5, which cannot meet the requirements for high-purity separation. Moreover, these membranes suffer from rapid flux decay (≥15% / 8h), making continuous industrial operation difficult.

[0004] In summary, existing technologies cannot simultaneously achieve the separation goals of "low energy consumption, high purity, and no solvent residue," so there is an urgent need to develop a new separation process to meet these requirements. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method for separating EPA and DHA from fish oil. First, utilizing the difference in esterification rates between EPA and DHA by lipase RM, a selective esterification reaction of fish oil fatty acids with lauryl alcohol is catalyzed. The EPA esterification rate is controlled at ≥90%, and the DHA esterification rate is limited to ≤20%, thereby constructing a polar binary system of low-polarity EPA-lauryl ester and high-polarity DHA fatty acid. Subsequently, a PDA-mSiO2 / P... is prepared using polydimethylsiloxane, mesoporous silica, and polydopamine in a specific mass ratio. Using the DMS composite membrane as the filtration medium, the system is filtered under certain conditions through a dual-action mechanism of "size sieving-hydrogen bond affinity" to achieve efficient permeation of EPA-laureol esters and selective retention of DHA fatty acids. Finally, the crude EPA-laureol ester solution on the permeation side and the concentrated DHA fatty acid solution on the retention side are purified by low-temperature distillation to obtain EPA-laureol esters with a purity of 75%-84% and DHA fatty acids with a purity of 68%-72%, thus achieving efficient and green separation of EPA-laureol esters and DHA fatty acids.

[0006] The purpose of this invention is to provide a method for separating EPA and DHA from fish oil, comprising the following steps: S1. Under the action of lipase RM, fish oil fatty acids and lauryl alcohol undergo a catalytic esterification reaction to obtain a mixture; the fish oil fatty acids contain 30%-37% EPA and 28%-33% DHA by mass. S2. Polydimethylsiloxane, mesoporous silica and polydopamine are ultrasonically dispersed in a solvent. After uniform dispersion, a membrane preform is prepared by a scraping method. After cross-linking and curing treatment, a PDA-mSiO2 / PDMS composite membrane is obtained. S3. Preheat the mixture described in S1 to 32℃-38℃, and then pass it through a filter with an effective filtration area of ​​0.1m². 2 -0.2m 2 The filter is filtered by a flat sheet membrane cross-flow filtration device. After filtration, crude EPA-laurate liquid is collected on the permeate side and concentrated DHA fatty acid liquid is collected on the retrieval side. The flat sheet membrane cross-flow filtration device uses the PDA-mSiO2 / PDMS composite membrane described in S2 as the filter medium. S4. The crude EPA-laurate solution and the concentrated DHA fatty acid solution described in S3 are distilled separately to obtain EPA-laurate and DHA fatty acids.

[0007] In one embodiment of the present invention, before S1, the fish oil fatty acids are distilled at a distillation temperature of 120°C-150°C, a vacuum of 0.4Pa-0.6Pa, a feed rate of 1.5L / h-2.0L / h, and a scraper rotation speed of 200r / min-300r / min.

[0008] In one embodiment of the present invention, in S1, the molar ratio of the fish oil fatty acid to lauryl alcohol is 1:(1.2-1.5). And / or, the amount of lipase RM used is 0.8%-1.2% of the total mass of fish oil fatty acids and lauryl alcohol; And / or, the enzyme activity of the lipase RM is 8000U / g-10000U / g.

[0009] In one embodiment of the present invention, in S1, the temperature of the catalytic esterification reaction is 30℃-40℃, the stirring rate is 200r / min-300r / min, and the time is 4h-6h. When the reaction temperature is below 30℃, the activity of lipase RM is insufficient, making it difficult for the EPA esterification rate to reach the target value of ≥90%, and failing to fully construct the polar difference system between EPA-lauryl ester and DHA fatty acid. When the temperature is above 40℃, the spatial conformation of lipase RM will change slightly, and its catalytic selectivity for EPA and DHA will decrease accordingly, thereby destroying the polar binary system of "EPA completely esterified and DHA remaining free".

[0010] In one embodiment of the present invention, in S1, the esterification rate of EPA is ≥90% and the esterification rate of DHA is ≤20%.

[0011] In one embodiment of the present invention, in S2, the mesoporous silica has a pore size of 5nm-10nm and a specific surface area of ​​500m². 2 / g-700m 2 / g; The pore size of mesoporous silica forms a precise size difference that matches the molecular size of EPA-lauryl ester (approximately 6nm) and DHA fatty acid (approximately 8nm). Its high specific surface area can construct continuous mesoporous channels, thereby enhancing the size sieving effect of the composite membrane on the two substances.

[0012] In one embodiment of the present invention, in S2, the mass ratio of polydimethylsiloxane, mesoporous silica, and polydopamine is 100:(5-8):(1.5-2.0). When the amount of mesoporous silica is less than 5, the density of the mesoporous channels constructed is insufficient, which will lead to a weakening of the size sieving effect. When the amount ratio is greater than 8, the mesoporous silica is prone to agglomeration, which will destroy the dense structure of the membrane and thus cause a decrease in membrane flux. At the same time, the surface of polydopamine (PDA) is rich in -NH2 and -OH groups, which can form intermolecular hydrogen bonds with -COOH in DHA fatty acid molecules. PDA, as a hydrogen bond acceptor, can significantly enhance the retention effect of DHA. When the amount ratio of PDA is less than 1.5, it provides insufficient hydrogen bond sites, resulting in a low membrane separation factor, which cannot meet the requirements of high selective separation. When the amount ratio is greater than 2.0, PDA is prone to agglomeration, which will block the mesoporous channels and significantly reduce the initial membrane flux.

[0013] In one embodiment of the present invention, in S2, the ultrasonic dispersion power is 300W-500W, the time is 30min-40min, and the temperature is 25℃-30℃. If the power is too low, the mesoporous silica and polydopamine will not be uniformly dispersed in the solvent, while if the power is too high or the dispersion temperature exceeds 30℃, it will cause premature cross-linking of polydimethylsiloxane, thereby destroying the preset structure of the composite membrane. At the same time, the ultrasonic time can ensure that the components of polydimethylsiloxane, mesoporous silica and polydopamine are fully and uniformly dispersed, avoiding excessively high local concentrations that may affect the membrane forming effect and separation performance.

[0014] In one embodiment of the present invention, in S2, the thickness of the preform is 80μm-100μm.

[0015] In one embodiment of the present invention, in S2, the crosslinking curing treatment is carried out at a temperature of 60°C-70°C for 8-10 hours. When the temperature is below 60°C, the crosslinking of polydimethylsiloxane will be incomplete, resulting in insufficient mechanical strength of the composite membrane, which is difficult to meet the requirements of industrial operation. When the temperature is above 70°C, the amino groups of polydopamine will be oxidized, resulting in a reduction of hydrogen bond sites on the membrane surface and weakening the selective retention effect on DHA. At the same time, the curing time allows the composite membrane to form a dense and stable structure, ensuring that the membrane can operate continuously for ≥300 hours while maintaining excellent separation performance.

[0016] In one embodiment of the present invention, in S3, the mixture is preheated to 32°C-38°C before filtration. This temperature range can prevent the viscosity of the mixture from increasing and causing a decrease in membrane flux when the temperature is below 32°C, and can also prevent the hydrogen bonding from being destroyed when the temperature is above 38°C, thereby ensuring the stability of the separation factor of the composite membrane and ensuring a high-selectivity separation effect.

[0017] In one embodiment of the present invention, in S3, the filtration pressure is 0.25MPa-0.35MPa, and the crossflow velocity is 1.5m / s-2.0m / s. When the pressure is below 0.25MPa, the contact efficiency between the mixture and the membrane surface is insufficient, which leads to insufficient dynamic process of hydrogen bond adsorption-retention and weakens the synergistic effect of the "size sieving-hydrogen bond affinity" dual mechanism. When the pressure is above 0.35MPa, the molecular thermal motion intensifies, which reduces the accuracy of size sieving and thus leads to a decrease in the permeability of EPA-laurate, affecting the separation efficiency. And / or, during the filtration process, the filter media surface is backwashed with ethanol solution for 10-15 minutes every 2 hours; ethanol can compete with DHA to bind to hydrogen bond sites on the surface of polydopamine (PDA), effectively desorbing DHA adsorbed on the membrane surface, while also dissolving a small amount of oil; the polarity of ethanol solution can reduce the hydrophobic adsorption on the membrane surface, reduce the adhesion of impurities on the membrane surface and mesoporous channels, thereby maintaining the unobstructed mesoporous channels and ensuring stable membrane filtration performance.

[0018] In one embodiment of the present invention, in S3, the permeability of the crude EPA-lauryl ester solution is ≥90%, and the retention rate of the DHA fatty acid concentrate is ≥88%.

[0019] In one embodiment of the present invention, in S4, the distillation process of the crude EPA-laurate liquor is as follows: vacuum degree is -0.095MPa, temperature is 80℃-85℃; And / or, the distillation process of the DHA fatty acid concentrate is as follows: vacuum degree of -0.098MPa, temperature of 120℃-125℃.

[0020] In one embodiment of the present invention, in S4, the purity of EPA-laureyl ester is 75%-84%, and the purity of DHA fatty acid is 68%-72%.

[0021] The technical solution of the present invention has the following advantages compared with the prior art: (1) The method described in this invention couples “selective catalysis of hydrogen bond differential system by lipase RM” with “mSiO2-PDA dual-mechanism modified membrane”. The specificity constant of lipase RM for EPA is 5.4 times that of DHA. The structure of the active center is more compatible with the molecular configuration of EPA. It can preferentially recognize the carboxyl group of EPA and catalyze its esterification with lauryl alcohol. DHA has a longer carbon chain and greater steric hindrance, making it difficult to enter the active center, and the esterification rate is significantly lower. By controlling the catalytic esterification reaction conditions, EPA is preferentially esterified (esterification rate ≥90%) to generate EPA-lauryl ester (containing only ester group -COO-) without hydrogen bond ability. DHA is retained in the form of free fatty acid (esterification rate ≤20%) and retains the carboxyl group (-COOH) containing hydrogen bond donor. It actively constructs a “nonpolar / weakly polar” polar binary system, laying the foundation for hydrogen bond separation. Meanwhile, mesoporous silica and polydopamine are introduced into the PDMS membrane to form a dual-mechanism synergistic structure: the mesoporous channels of the 5nm-10nm mesoporous silica have low resistance to the permeation of EPA-lauryl ester with a molecular dynamic diameter of about 6nm (permeability ≥90%), but high resistance to the permeation of DHA fatty acids with a diameter of about 8nm, thus achieving preliminary size sieving; the hydrophilic layer formed by PDA on the membrane surface forms intermolecular hydrogen bonds with the -COOH of DHA through -NH2 and -OH, which enables DHA to be adsorbed and retained, significantly improving the retention rate.

[0022] (2) The method described in this invention has low energy consumption throughout the process, does not require the use of urea and a large amount of organic solvents, reduces wastewater discharge by more than 90%, has environmental advantages, meets the needs of continuous industrial production, and successfully solves the pain points of existing technologies such as low separation factor, fast flux decay, high energy consumption and poor environmental performance. Detailed Implementation

[0023] The present invention will be further described below with reference to specific embodiments, so that those skilled in the art can better understand and implement the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. It should be understood that the specific embodiments are only used to explain the present invention, but the embodiments are not intended to limit the present invention.

[0024] In this invention, unless otherwise stated, the technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0025] In this invention, unless otherwise stated, the term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0026] In this invention, unless otherwise specified, the experimental methods used in the embodiments of this invention are conventional methods, and the materials and reagents used are commercially available unless otherwise specified.

[0027] In this invention, unless otherwise stated, the lipase RM used in the embodiments of this invention was purchased from Novozymes and has an enzyme activity of 9200 U / g.

[0028] In this invention, unless otherwise stated, the lipase Novozym 435 used in the comparative examples of this invention was purchased from Novozymes and has an enzyme activity of 9200 U / g.

[0029] In this invention, unless otherwise stated, the apparatus used for short-path distillation in the embodiments of this invention is a short-path molecular distillation apparatus, the scraper material is polytetrafluoroethylene, and the condensation area is 0.2 m². 2 .

[0030] In this invention, unless otherwise stated, the membrane components of the flat-sheet membrane cross-flow filtration device used in the embodiments of this invention are made of 316L stainless steel, the seals are made of fluororubber, and the inlet temperature fluctuation of the mixed liquid during the filtration process is ≤±1℃ and the pressure fluctuation is ≤±0.02MPa.

[0031] Example 1

[0032] The method for separating EPA and DHA from fish oil in this embodiment specifically includes the following steps: S1. Pretreatment of fish oil fatty acids: 150 kg of crude fish oil fatty acids (EPA mass fraction 33.0%, DHA mass fraction 29.0%) were weighed and subjected to short-path distillation at a distillation temperature of 135℃, a vacuum degree of 0.5 Pa, a feed rate of 1.8 L / h, and a scraper rotation speed of 250 r / min. After removing 15% of the light component non-target fatty acids, the distillate was collected to obtain 140 kg of pretreated fish oil fatty acids (EPA mass fraction 34.6%, DHA mass fraction 30.8%). S2. Enzyme-catalyzed esterification: Pretreated fish oil fatty acids (140 kg) and lauryl alcohol were added to a 500 L reactor at a molar ratio of 1:1.3. Lipase RM (1.1% of the total mass) was added and stirred at 32 °C and 250 r / min for 5 h to obtain 242.6 kg of mixed solution (EPA esterification rate of 90% and DHA esterification rate of 18%). S3. Preparation of PDA-mSiO2 / PDMS composite film: Take 20 kg of polydimethylsiloxane (PDMS) and 1.2 kg of [a specific material] with a pore size of approximately 7.5 nm and a specific surface area of ​​approximately 620 m². 2 / g of mesoporous silica (mSiO2), 0.35kg of polydopamine (PDA), and 80L of tetrahydrofuran were ultrasonically dispersed at 28℃ for 35min using a power of 350W. After uniform dispersion, a film preform with a thickness of 90μm was prepared on polyester nonwoven fabric using a scraping method. The film preform was then crosslinked and cured at 65℃ for 9h to obtain a film preform with an area of ​​0.15m². 2 PDA-mSiO2 / PDMS composite film; S4. Filtration: After preheating the mixture to 35°C, pass it through a filter with an effective filtration area of ​​0.15m². 2 The flat sheet membrane cross-flow filtration device uses a PDA-mSiO2 / PDMS composite membrane as the filter medium. Cross-flow filtration is carried out under the conditions of 0.3MPa operating pressure and 1.8m / s cross-flow rate. At the same time, the membrane surface is backwashed for 12 minutes every 2 hours with 5vol% ethanol aqueous solution. After 4 hours of filtration, 135kg of crude EPA-laurate ester solution is collected on the permeate side, and 54kg of DHA fatty acid concentrate is collected on the retentate side. The lauryl alcohol is separated and recycled. S5. Purification: The crude EPA-laurate solution was subjected to vacuum distillation at -0.095 MPa and 82℃, while the condenser temperature was controlled at 42℃. After distillation, 75 kg of EPA-laurate was obtained with a purity of 83.2%. The DHA fatty acid concentrate was subjected to short-path distillation at -0.098 MPa and 123 °C, while the condenser temperature was controlled at 52 °C. After distillation, 42 kg of DHA fatty acid was obtained with a purity of 71.4%.

[0033] Example 2

[0034] The process is basically the same as in Example 1, except that in S1, the crude fish oil fatty acids contain 35.1% EPA and 28.3% DHA by mass; and the fish oil fatty acids obtained after pretreatment contain 36.8% EPA and 29.9% DHA by mass.

[0035] Example 3

[0036] The process is basically the same as in Example 1, except that in S4, the effective filtration area is 0.2m². 2 Filtration time: 3 hours.

[0037] Comparative Example 1

[0038] The process is basically the same as in Example 1, except that the 242.6 kg mixture is directly distilled, specifically including the following steps: A molecular distillation apparatus was used to separate 242.6 kg of the mixture, with the vacuum degree controlled at 0.5 Pa and the evaporation temperature at 165 °C, to obtain EPA-lauryl ester and DHA fatty acid.

[0039] Comparative Example 2

[0040] The process is basically the same as in Example 1, except that a composite film, i.e., an mSiO2 / PDMS composite film, is prepared without the use of PDA.

[0041] Comparative Example 3

[0042] The basic structure is the same as in Example 1, except that the PDA-mSiO2 / PDMS composite film is modified into a PDMS film.

[0043] Comparative Example 4

[0044] The results are basically the same as in Example 1, except that the lipase RM is adjusted to lipase Novozym 435 (EPA esterification rate reaches 68.5%, DHA esterification rate reaches 42.3%).

[0045] Comparative Example 5

[0046] Basically the same as Example 1, except that: the pore size is approximately 7.5 nm and the specific surface area is approximately 620 m². 2 / g of mesoporous silica (mSiO2) was adjusted to have a pore size of approximately 15nm and a specific surface area of ​​approximately 480m². 2 / g of mesoporous silica (mSiO2).

[0047] Comparative Example 6

[0048] The reaction was basically the same as in Example 1, except that the stirring reaction at 32°C and 250 r / min for 5 hours was changed to stirring reaction at 60°C and 250 r / min for 5 hours (the esterification rate of EPA reached 82.1% and the esterification rate of DHA reached 31.7%).

[0049] Comparative Example 7

[0050] The process is basically the same as in Example 1, except that the filtration is adjusted from operating pressure of 0.3 MPa and crossflow velocity of 1.8 m / s to operating pressure of 0.2 MPa and crossflow velocity of 1.8 m / s.

[0051] Comparative Example 8

[0052] The process is basically the same as in Example 1, except that molecular distillation pretreatment is not performed.

[0053] Test Example 1

[0054] Based on the examples and comparative examples, relevant parameters in the filtration, purification and other processes were tested; (1) Permeation rate: Record the mass of crude EPA-lauryl ester collected on the permeate side during membrane filtration, denoted as m1; record the total mass of EPA-lauryl ester in the mixture before membrane filtration as m0, and calculate it using the formula m0 = initial mass of EPA in fish oil fatty acids × EPA esterification rate × 470 / 302, where 470 is the molecular weight of EPA-lauryl ester and 302 is the molecular weight of EPA; calculate the membrane filtration permeation rate using the formula permeation rate (%) = (m1 / m0) × 100%; (2) Retention rate: Record the mass of DHA in the DHA fatty acid concentrate collected on the retention side during membrane filtration, and record it as m2; calculate the total mass of unesterified DHA in the mixture before membrane filtration according to the formula, and record it as m3, m3 = initial mass of DHA in fish oil fatty acids × (1 − DHA esterification rate); then calculate the retention rate according to the above two masses according to the formula: retention rate (%) = (m2 / m3) × 100%; (3) Flux decay rate: After the membrane filtration is started, it runs continuously for 1 hour (without flushing). The volume of liquid on the permeate side is monitored in real time by the flow meter. The stable flux is calculated according to the formula J=V / (A×T), where V is the permeate volume, A is the effective filtration area of ​​the membrane, and T is the filtration time. This flux is taken as the initial flux J0. Then, filtration continues according to the patented process. Backwashing is performed every 2 hours. After filtration for 8 hours, the flux J8 is tested before the next backwash. The flux decay rate is calculated according to the following formula: Flux decay rate (%) = (J0-J8) / J0×100%; (4) Purity: Gas chromatography (GC) was used for testing. The chromatographic column used was an HP-5 capillary column (30m×0.32mm×0.25μm). The test conditions were as follows: the column temperature was programmed, with an initial temperature of 80℃ and held for 2 min, then increased to 250℃ at a rate of 10℃ / min and held for 15 min. The injection port temperature was 260℃, the flame ionization detector (FID) temperature was 280℃, the carrier gas was nitrogen, the flow rate was 1.0mL / min, the injection volume was 1μL, and the split ratio was 10:1. The sample pretreatment was to take 0.1g of the purified product, dilute it with n-hexane to 10mL, filter it through a 0.22μm organic phase membrane, and then perform gas chromatography analysis. The purity was calculated by the area normalization method, and the percentage of the peak area of ​​the target component (EPA-lauryl ester / DHA fatty acid) to the total peak area was used as the sample purity. (5) Separation factor: The mass fractions of EPA-laureyl ester and DHA fatty acid in the permeate side and the retrieval side after membrane filtration were tested respectively. The mass fraction of EPA-laureyl ester in the permeate side was recorded as x1 and the mass fraction of DHA fatty acid was recorded as y1. The mass fraction of EPA-laureyl ester in the retrieval side was recorded as x2 and the mass fraction of DHA fatty acid was recorded as y2. The separation factor α was calculated according to the formula α=(x1 / y1) / (x2 / y2). (6) Contribution of hydrogen bonding to the separation factor: The experimental group was the PDA-mSiO2 / PDMS composite membrane prepared in Example 1, the control group 1 was the mSiO2 / PDMS composite membrane of Comparative Example 2, and the control group 2 was the PDMS membrane of Comparative Example 3. The separation factors of the three groups of membranes were tested respectively and recorded as α total, α sieve fraction, and α blank. The separation factor α hydrogen bond contributed by hydrogen bonding was α total - α sieve fraction. The proportion of hydrogen bonding was calculated according to the formula: proportion of hydrogen bonding (%) = (α hydrogen bond / (α total - α blank)) × 100%. (7) Energy consumption calculation: Record the energy consumption of each piece of equipment in the entire process, including the molecular distillation device, enzyme catalytic reactor, membrane filtration device and distillation device; among them, record the heating energy consumption of the enzyme catalytic reactor at 30℃-40℃ and the stirring energy consumption at 200r / min-300r / min, record the pressure pump energy consumption of the membrane filtration device at 0.25MPa-0.35MPa and the cross-flow pump energy consumption at 1.5m / s-2.0m / s, and record the vacuum distillation at 80-85℃ and short-path distillation at 120℃-125℃. The energy consumption of heating, vacuum pump to maintain the corresponding vacuum level, and cooling medium of condenser were recorded in the molecular distillation apparatus. The energy consumption of heating to maintain 120℃-150℃, vacuum pump to maintain 0.1Pa-1Pa, and feed pump and scraper stirring were recorded. The energy consumption was measured by recording the total power consumption (kWh) through the equipment’s built-in energy consumption monitoring instrument. The energy consumption per unit product (kWh / t) was calculated by combining the total mass of the final product (t). The ambient temperature was controlled at 25±2℃ during the test to eliminate the influence of environmental factors on the energy consumption data. Table 1 shows the final measured performance: Table 1

[0055] As can be seen from Table 1, the performance indicators of Examples 1-3 exhibit high stability and superiority. The transmittance is stable at 91.5%-92.5%, the rejection rate is 84.6%-85.5%, the flux decay rate is only 4.3%-4.5%, the separation factor is stable at 4.4-4.6, the hydrogen bonding ratio is 64.8%-65.5%, and the unit energy consumption is controlled at 242kWh / t-248kWh / t. There are no significant fluctuations in the parameters, which proves that the process parameters of the present invention have good repeatability and industrial adaptability. The synergistic design of parameters throughout the process ensures the stability of the separation effect.

[0056] Comparing Example 1 and Comparative Example 1, it can be seen that when only high-temperature molecular distillation is used without membrane filtration and enzyme-catalyzed esterification, there is no data related to permeability, rejection rate, and flux decay rate. The product purity is only 58.5% and 52.3%, respectively, the separation factor is as low as 1.8, there is no contribution from hydrogen bonding, and the unit energy consumption is as high as 420 kWh / t. All performance aspects are far inferior to Example 1. This is because traditional molecular distillation relies solely on differences in molecular motion rates for separation, lacking the polarity difference basis for selective catalysis. Furthermore, high temperatures easily cause fatty acid oxidation and degradation, and high-vacuum operation significantly increases energy consumption. This demonstrates that the coupled process of this invention, "enzyme-catalyzed construction of a polar system + dual-mechanism membrane filtration + low-temperature distillation," has significant performance and energy consumption advantages compared to traditional single distillation processes.

[0057] Comparing Example 1 and Comparative Example 2, it can be seen that when using the mSiO2 / PDMS composite membrane containing only a size sieving mechanism, the permeability drops to 88.5%, the rejection rate is 72.0%, the flux decay rate rises sharply to 18.3%, the product purity is only 68.5% and 58.2%, the separation factor is 3.1 and there is no contribution from hydrogen bonding, and the unit energy consumption of 302 kWh / t is higher than that of Example 1. This is because the composite membrane relies only on the size sieving effect of mSiO2 and lacks the hydrogen bonding affinity of PDA, resulting in a significant decrease in the selective rejection capacity for DHA. Moreover, the membrane surface is prone to rapid flux decay due to oil accumulation. This proves that hydrogen bonding affinity is the core of improving membrane separation selectivity and maintaining stable membrane filtration performance, and the synergy of the "size sieving-hydrogen bonding" dual mechanism is the key to achieving efficient separation.

[0058] Comparing Example 1 and Comparative Example 3, it can be seen that when using a pure PDMS membrane, the permeability further decreased to 85.2%, the rejection rate was only 62.5%, and the flux decay rate of 19.5% was the highest among all samples. The product purity of 60.3% and 54.8% were close to that of Comparative Example 1, the separation factor of 2.1 had no contribution from hydrogen bonding, and the unit energy consumption of 310 kWh / t continued to increase. This is because the pure PDMS membrane has neither a mesoporous structure for size sieving nor hydrogen bond affinity sites. Relying solely on hydrophobic interactions, it is impossible to effectively distinguish between EPA-lauryl esters and DHA fatty acids. The most severe fouling on the membrane surface led to the fastest flux decay, further confirming the necessity of dual-mechanism modification for improving the separation performance of PDMS membranes, and also proving that a single mechanism cannot meet the separation requirements of high purity and low decay.

[0059] Comparing Example 1 and Comparative Example 4, it can be seen that after replacing the enzyme with the conventional lipase Novozym435, the permeability was 82.0%, the rejection rate was 68.0%, the flux decay rate was 12.8%, the product purity was 70.5% and 62.1%, the separation factor decreased to 3.5, the hydrogen bonding effect accounted for only 45.2%, and the unit energy consumption was 295 kWh / t. This is because Novozym435 has low catalytic specificity for EPA and DHA, and cannot achieve the goal of high esterification of EPA and low esterification of DHA. The polar binary system in the feed solution is not fully constructed, the content of free DHA that can form hydrogen bonds is reduced, resulting in the inability to fully exert the hydrogen bonding effect of the composite membrane, and the synergistic efficiency of the two mechanisms is greatly reduced. This proves that the selective catalysis of lipase RM is a prerequisite for the effective implementation of subsequent membrane separation processes, and the polar difference system constructed by enzymatic methods is the basis for hydrogen bond separation.

[0060] Comparing Example 1 and Comparative Example 5, it can be seen that when using large-pore size and low specific surface area mSiO2, the permeability is 87.0%, the rejection rate is 75.0%, the flux attenuation rate is 8.7%, the product purity is 75.8% and 65.5%, the separation factor is 4.2, and the hydrogen bonding ratio is 62.5%, which is close to Example 1, with a unit energy consumption of 288 kWh / t. This is because the pore size of mSiO2 increases to 15 nm, and the decrease in specific surface area leads to a reduction in size sieving accuracy, making it impossible to accurately distinguish the molecular size of EPA-lauryl ester and DHA fatty acids. Although the hydrogen bonding effect of PDA is basically retained, the synergistic effect of the dual mechanism of size sieving and hydrogen bonding affinity is destroyed, and the separation factor decreases significantly. This proves that precise control of mSiO2 pore size and specific surface area is a necessary condition to ensure the size sieving effect and achieve the synergistic effect of the dual mechanism.

[0061] Comparing Example 1 and Comparative Example 6, it can be seen that when the enzyme-catalyzed reaction temperature is increased to 60℃, the permeability is 85.5%, the rejection rate is 73.0%, the flux decay rate is 7.2%, the product purity is 76.5% and 66.8%, the separation factor is 4.8, the hydrogen bonding ratio is 60.8%, and the unit energy consumption is 295kWh / t. This is because high temperature causes changes in the spatial conformation of lipase RM, resulting in decreased catalytic selectivity, the DHA esterification rate exceeding the design range of ≤20%, insufficient construction of the polar binary system, reduced free DHA content in the feed solution, and inability to fully utilize the hydrogen bonding rejection effect of the membrane. At the same time, the increase in non-target esterification products leads to increased adsorption of impurities on the membrane surface. This proves that precise control of the enzyme catalytic temperature (30℃-40℃) is the key to ensuring the effective construction of the polar binary system and maintaining membrane separation performance.

[0062] Comparing Example 1 and Comparative Example 7, it can be seen that when the membrane filtration operating pressure is reduced to 0.2 MPa, the permeability is 88.0%, the rejection rate is 76.0%, the flux decay rate is 6.5%, the product purity is 78.2% and 68.5%, the separation factor is 5.1, the hydrogen bonding ratio is 63.5%, and the unit energy consumption is 265 kWh / t, which is slightly higher than that of Example 1. This is because insufficient pressure leads to a decrease in the contact efficiency between the mixture and the membrane surface, resulting in insufficient dynamic processes of hydrogen bonding adsorption-retention. At the same time, the driving force for size sieving is insufficient, weakening the synergistic effect of the two mechanisms. However, the hydrogen bonding sites of the membrane material itself are not affected, so the hydrogen bonding ratio does not decrease significantly. This proves that precise matching of membrane filtration pressure (0.25 MPa-0.35 MPa) is a necessary condition for fully utilizing the synergistic effect of the two mechanisms and improving separation efficiency.

[0063] Comparing Example 1 and Comparative Example 8, it can be seen that when only the molecular distillation pretreatment step of fish oil fatty acids is omitted, the permeability is 87.5%, the rejection rate is 74.5%, the flux decay rate is 7.8%, the product purity is 76.8% and 67.2%, the separation factor is 5.3, the hydrogen bonding ratio is 62.8%, and the unit energy consumption is 272 kWh / t. This is because non-target impurities in crude fish oil fatty acids are not removed, which interfere with the catalytic efficiency of lipase RM during esterification, and accumulate and block the mesoporous channels of the membrane and occupy the hydrogen bonding sites on the membrane surface during filtration, resulting in a decrease in the synergistic efficiency of the two mechanisms. Both membrane filtration performance and product purity are reduced. This proves that the "impurity reduction and concentration" effect of molecular distillation pretreatment is an important support for ensuring the efficient implementation of subsequent enzyme catalysis and membrane filtration processes, and the synergistic cooperation of all processes is the basis for achieving excellent separation results.

[0064] Test Example 2

[0065] The PDA-mSiO2 / PDMS composite membrane of Example 1, the mSiO2 / PDMS composite membrane of Comparative Example 2, and the PDMS membrane of Comparative Example 3 were all cut into 2cm×2cm square samples. After removing residual moisture and impurities by vacuum drying at 60℃ for 12h, 0.1g of each sample was accurately weighed and placed in a 50mL stoppered centrifuge tube. 20mL of a 100mg / L DHA fatty acid ethanol solution was added to each tube, and the tubes were sealed and placed in a 25℃ constant temperature water bath shaker. The tubes were shaken at 150r / min for 24h until adsorption equilibrium was reached. After adsorption equilibrium, the PDA-mSiO2 / PDMS composite membrane, mSiO2 / PDMS composite membrane, and PDMS membrane were... The corresponding residual DHA concentrations in the solutions were 2.8 mg / L, 82.5 mg / L, and 88.6 mg / L, respectively. The residual DHA concentration in the supernatant of each centrifuge tube was then determined by high-performance liquid chromatography (HPLC). The static adsorption capacity of the three membrane materials for DHA was calculated using the formula: "Static adsorption capacity (mg / g) = (Initial concentration - Equilibrium concentration) × Solution volume (L) / Membrane sample mass (g)" (20 mL solution volume is converted to 0.02 L). The average value was taken from three parallel measurements. The final static adsorption capacities of the PDA-mSiO2 / PDMS composite membrane, mSiO2 / PDMS composite membrane, and PDMS membrane for DHA were 19 mg / L, 19 mg / L, and 88.6 mg / L, respectively. At concentrations of 0.4 mg / g, 3.5 mg / g, and 2.3 mg / g, the adsorption capacity of the PDA-mSiO2 / PDMS composite membrane for DHA was 5.5 times that of the mSiO2 / PDMS composite membrane and 8.4 times that of the PDMS membrane, respectively. Its adsorption capacity for DHA was significantly superior to that of the PDA-free mSiO2 / PDMS composite membrane and the pure PDMS membrane. This is because the surface of PDA molecules is rich in -NH2 and -OH groups, which can form intermolecular hydrogen bonds with the -COOH groups in DHA molecules, achieving highly efficient adsorption of DHA. In contrast, the mSiO2 / PDMS composite membrane relies solely on the porous structure of mSiO2 and the hydrophobic effect of PDMS for adsorption. Relying solely on hydrophobic interactions, neither of them possesses hydrogen bond affinity sites, resulting in weak adsorption and retention capacity for DHA. This test result corroborates the membrane separation performance data in Test Example 1, further demonstrating that the introduction of PDA endows the PDA-mSiO2 / PDMS composite membrane with crucial hydrogen bonding. This bonding, in synergy with the size sieving mechanism of mSiO2, significantly enhances the selective adsorption and retention capacity of the composite membrane for DHA. This is also a key reason why the PDA-mSiO2 / PDMS composite membrane exhibits a high separation factor and low flux decay rate. Simultaneously, it directly confirms the core role of hydrogen bonding in enhancing the adsorption and retention performance of the composite membrane and optimizing the separation effect of EPA and DHA.

[0066] Test Example 3

[0067] Using the process parameters of Example 1, during the continuous filtration of the mixed liquid into the flat sheet membrane cross-flow filtration device, the membrane running time, the collection volume of crude EPA-laurate solution on the permeate side and concentrated DHA fatty acid solution on the retrieval side were recorded simultaneously. The permeate rate, retrieval rate and flux decay rate were measured every 24 hours. At the beginning of the experiment (0h) and after 300h of continuous operation, the products on the permeate side and retrieval side were collected respectively. After the S5 purification step, the product loss rate was calculated (product loss rate = (theoretical product mass - actual obtained product mass) / theoretical product mass × 100%), and the purity of the final product was tested. Test results show that during 300 hours of continuous operation, the permeability of the PDA-mSiO2 / PDMS composite membrane remained stable at 91.0%-92.5%, the rejection rate remained at 84.5%-85.5%, and the flux decay rate only slowly increased from the initial 1.0% to 3.5%, without any sharp decay. After 300 hours of operation, the purity of the purified EPA-laureol ester was 82.5%, and the purity of the DHA fatty acid was 70.5%, which was very similar to the purity of the product in the initial stage (0h) (EPA-laureol ester 83.2%, DHA fatty acid 71.4%), and the product loss rate was controlled at 2.5%, meeting the industrial requirement of ≤3.5%. This is because, firstly, the selective catalysis of lipase RM achieves "active differentiation," preferentially esterifying EPA to form the less polar EPA-lauryl ester, while DHA retains its fatty acid form (higher polarity), constructing a clearly defined polar binary system that lays the foundation for subsequent separation. Secondly, the "dual-mechanism differential capture" of the PDA-mSiO2 / PDMS composite membrane is utilized. The size sieving effect of mSiO2 precisely distinguishes the size differences between the two molecules. Simultaneously, the hydrogen bond affinity between the -NH2 and -OH groups on the PDA surface and the carboxyl groups of DHA fatty acids enhances the retention of DHA and the permeation selectivity of EPA-lauryl ester. Furthermore, the dynamic adsorption-desorption characteristics of hydrogen bonding reduce membrane surface fouling and decrease flux decline. Finally, the synergistic optimization of parameters such as enzyme catalysis temperature, membrane filtration pressure, and cleaning frequency throughout the entire process avoids the decrease in separation efficiency or membrane performance degradation caused by mismatch in parameters at a single stage, forming a stable separation system.

[0068] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for separating EPA and DHA from fish oil, characterized in that, Includes the following steps: S1. Under the action of lipase RM, fish oil fatty acids and lauryl alcohol undergo a catalytic esterification reaction to obtain a mixture; the fish oil fatty acids contain 30%-37% EPA and 28%-33% DHA by mass. S2. Polydimethylsiloxane, mesoporous silica and polydopamine are ultrasonically dispersed in a solvent. After uniform dispersion, a membrane preform is prepared by a scraping method. After cross-linking and curing treatment, a PDA-mSiO2 / PDMS composite membrane is obtained. S3. Preheat the mixture described in S1 to 32℃-38℃, and then pass it through a filter with an effective filtration area of ​​0.1m². 2 -0.2m 2 The filter is filtered by a flat sheet membrane cross-flow filtration device. After filtration, crude EPA-laurate liquid is collected on the permeate side and concentrated DHA fatty acid liquid is collected on the retrieval side. The flat sheet membrane cross-flow filtration device uses the PDA-mSiO2 / PDMS composite membrane described in S2 as the filter medium. S4. The crude EPA-laurate solution and the concentrated DHA fatty acid solution described in S3 are distilled separately to obtain EPA-laurate and DHA fatty acids.

2. The method for separating EPA and DHA from fish oil according to claim 1, characterized in that, Before S1, the fish oil fatty acids are distilled at a temperature of 120℃-150℃, a vacuum of 0.4Pa-0.6Pa, a feed rate of 1.5L / h-2.0L / h, and a scraper rotation speed of 200r / min-300r / min.

3. The method for separating EPA and DHA from fish oil according to claim 1, characterized in that, In S1, the molar ratio of the fish oil fatty acids to lauryl alcohol is 1:(1.2-1.5). And / or, the amount of lipase RM used is 0.8%-1.2% of the total mass of fish oil fatty acids and lauryl alcohol; And / or, the enzyme activity of the lipase RM is 8000U / g-10000U / g.

4. The method for separating EPA and DHA from fish oil according to claim 1, characterized in that, In S1, the temperature of the catalytic esterification reaction is 30℃-40℃, the stirring rate is 200r / min-300r / min, and the time is 4h-6h.

5. The method for separating EPA and DHA from fish oil according to claim 1, characterized in that, In S2, the mesoporous silica has a pore size of 5nm-10nm and a specific surface area of ​​500m². 2 / g-700m 2 / g.

6. The method for separating EPA and DHA from fish oil according to claim 1, characterized in that, In S2, the mass ratio of polydimethylsiloxane, mesoporous silica and polydopamine is 100:(5-8):(1.5-2.0).

7. The method for separating EPA and DHA from fish oil according to claim 1, characterized in that, In S2, the thickness of the preform is 80μm-100μm.

8. The method for separating EPA and DHA from fish oil according to claim 1, characterized in that, In S2, the cross-linking curing treatment is performed at a temperature of 60℃-70℃ for 8h-10h.

9. The method for separating EPA and DHA from fish oil according to claim 1, characterized in that, In S3, the filtration pressure is 0.25MPa-0.35MPa, and the crossflow velocity is 1.5m / s-2.0m / s; And / or, during the filtration process, the surface of the filter medium is backwashed with ethanol solution for 10-15 minutes every 2 hours.

10. The method for separating EPA and DHA from fish oil according to claim 1, characterized in that, In S4, the distillation process of the crude EPA-laurate liquor is as follows: vacuum degree is -0.095MPa, temperature is 80℃-85℃; And / or, the distillation process of the DHA fatty acid concentrate is as follows: vacuum degree of -0.098MPa, temperature of 120℃-125℃.

Images