Multistage-pore silica filter aid for refining animal fat, preparation method and application thereof

Abstract

The application belongs to the technical field of food processing and functional inorganic porous material, and particularly relates to a multi-stage pore silica filter aid for animal oil refining, a preparation method and application thereof. The filter aid comprises a silica skeleton with three-stage pore channel structures of macro-pores (0.5-2 mu m), mesopores (6-12 nm) and micropores (<2 nm), and an aminopropyl functional group bonded by gas phase grafting. The aminopropyl forms a three-stage ladder type grafting density gradient on the macro-pore wall, the mesopore wall and the micropore wall. The mesopore wall density is 2-4 times that of the macro-pore wall, and the micropore wall density is 4-8 times that of the macro-pore wall. The three-stage pore channels and the gradient gradient cooperatively form a filter-adsorption space decoupling structure. The filter rate is increased by more than 60%, the phospholipid removal rate is greater than 96%, the free fatty acid is reduced by 0.3%-0.5%, and the oil recovery rate is greater than 97%.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary technical field of food processing and functional inorganic porous materials, specifically relating to a multi-level porous silica filter aid for refining animal fats and oils, its preparation method, and the application of the filter aid in the refining and filtration of lard, tallow, and poultry fats. Background Technology

[0002] Animal fats, including lard, tallow, and poultry fats such as chicken and duck fat, have significant applications in the food, feed, and biodiesel feedstock industries. Unrefined crude animal fats typically contain a complex series of impurities, which can be classified into four levels according to their scale. The first level consists of macroscopic impurities such as keratin residue and bone meal (5–50 μm), originating from meat tissue residues left during slaughtering and processing. The second level consists of phospholipid micelles (50–500 nm), mainly formed by the aggregation of polar phospholipids such as lysophosphatidylcholine (lysoPC) in animal fat. The third level consists of phospholipid monomers (1–3 nm) and heme-iron complexes, where heme-iron originates from animal blood residues mixed into the fat during slaughtering and processing, strongly influencing the oxidative stability and color depth of the fat. The fourth level consists of free fatty acid molecules at the 0.7~1 nm scale. Since the saturated fatty acid content of animal fats is as high as 60%~70%, which is significantly higher than that of vegetable oils (15%~30%), the removal of free fatty acids is particularly crucial for controlling the acid value and shelf life of the product.

[0003] Traditional animal fat refining and filtration processes primarily rely on diatomaceous earth filter aids. Diatomaceous earth, derived from the siliceous fossil skeletons of diatoms (single-celled algae), possesses a naturally porous structure and stable chemical properties, enabling it to form a loose, permeable filter cake layer to intercept insoluble suspended impurities. However, diatomaceous earth filter aids have three inherent drawbacks. First, diatomaceous earth has a simple pore structure, consisting only of macropores ranging from hundreds of nanometers to micrometers in size. It lacks a mesoporous and microporous system capable of chemically adsorbing small molecule impurities (such as free fatty acids and phospholipid monomers), relying solely on physical interception for primary filtration, resulting in low efficiency in removing small polar molecule impurities. Second, during diatomaceous earth filtration, as the filter cake thickness increases, the macropores inside the cake are gradually blocked by trapped phospholipid micelles. The filtration resistance increases sharply with filtration time; typically, after 30 minutes of filtration, the permeability coefficient decreases to below 40% of its initial value, leading to short filtration cycles and low equipment utilization. Third, the surface of diatomaceous earth consists of inert silanol and aluminol hydroxyl groups, lacking specific chemical adsorption activity for polar impurities, thus failing to achieve deep impurity removal.

[0004] To overcome the performance bottlenecks of diatomaceous earth, existing technologies have explored various silicon-based functional adsorbents and filter aids. US Patent 3557008A (title: Animal fat cleaning composition and method, granted in 1971) discloses a method of treating animal fat with alkali metal silicates (such as sodium silicate and potassium silicate) to adsorb protein particles by generating silica gel in situ. Although this method can remove keratin residue, its removal effect on phospholipids and free fatty acids is limited, and the silica gel has a slow settling speed and poor filterability. European patent EP0295418A2 (title: Process for the removal of chlorophyll and color bodies from glyceride oils using acid-treated silica adsorbents, published in 1988, covering animal fats and glyceride oils including animal fats and tallows) discloses the use of acid-treated amorphous silica adsorbents for the removal of phospholipids and chlorophyll from glyceride oils. Its Canadian counterpart is CA1305120C, but the silica described therein has only a single-pore structure, and the problem of increased filtration resistance remains unresolved. US patent US8066889B2 (title: Method for filtering cooking oil used in frying process) discloses a method for filtering edible oils using silicate filter aids such as calcium silicate, aluminum silicate, and magnesium silicate. However, the silicates described therein also have a single-pore structure and do not involve chemical functionalization modification for polar impurities.

[0005] In recent years, the amino functionalization modification of mesoporous silica has become a research hotspot. Scientific literature, such as the journal Microporous and Mesoporous Materials, 2020, Vol. 300, p. 110136, discloses a method for synthesizing mesoporous silica using hexadecyltrimethylammonium bromide (CTAB) and triblock copolymer P123 as a dual surfactant template, followed by liquid-phase grafting of (3-aminopropyl)trimethoxysilane. The amino functionalized mesoporous silica was used for the selective adsorption of free fatty acids and chlorophyll in olive oil. This literature verified the adsorption selectivity of amino functional groups for free fatty acids, but its silica only has a single-level mesoporous structure, lacking macroporous permeation channels and microporous high-capacity adsorption channels. The adsorption capacity is limited by the mesoporous surface area, and the filtration resistance cannot be improved. A 2016 paper published in Nanotechnology (PMC database number PMC5090446) disclosed a process for one-step synthesis of macroporous-mesoporous-microporous tertiary porous silica using polystyrene microspheres and triblock copolymers as dual templates. This process forms a tertiary porous structure but does not involve amine functionalization modification. The resulting material is only a general-purpose hierarchical porous carrier and not a functional filter aid for animal fat refining. A 2022 paper in Chemical Engineering Journal, Volume 427, page 131561, disclosed work on amine-grafted bimodal hierarchical porous silica for carbon dioxide capture and natural gas desulfurization, but it only shows a bimodal pore structure, not a trimodal pore structure, and is applied to gas-phase adsorption rather than liquid-phase filtration.

[0006] In summary, existing silicon-based filter aids and adsorbents face a fundamental technical contradiction when applied to animal fat refining: achieving high adsorption capacity necessitates a fine pore structure with a high specific surface area, but this inevitably leads to a sharp increase in filtration resistance; conversely, achieving low filtration resistance requires a macroporous structure, but the adsorption capacity of macroporous structures is insufficient for deep impurity removal. Even with materials employing hierarchical pore structures in existing technologies, the modification process (such as post-liquid-phase grafting) uniformly modifies all pore segments, making it impossible to achieve spatial decoupling between low-resistance macroporous permeation and high-density microporous chemisorption within the same material particle. Therefore, this fundamental contradiction remains unresolved. Furthermore, considering the significant differences in impurity composition between animal fats and vegetable oils (the proportions of heme iron, lysophosphatidylcholine, and saturated fatty acids in animal fats differ significantly from those in vegetable oils), existing filter aids and adsorbents developed for vegetable oils also fail to achieve optimal adsorption matching. Summary of the Invention

[0007] To address the technical contradiction of existing animal fat refining filter aids in achieving both adsorption capacity and filtration flux, as well as the technical deficiency of lacking precise matching to the specific impurity spectrum of animal fats, this invention provides a multi-level porous silica filter aid for animal fat refining and its preparation method, which significantly improves the filtration rate while deeply removing phospholipids and free fatty acids, extending the filtration cycle and increasing the fat recovery rate.

[0008] Technical Solution Summary: This invention provides a multi-level porous silica filter aid for refining animal fats, comprising an amorphous silica framework with a three-level pore structure consisting of macropores, mesopores, and micropores, and aminopropyl functional groups bonded to the surface of the silica framework pores via a vapor-phase grafting process. The macropores have a diameter of 0.5–2 μm and a volume accounting for 15%–30% of the total pore volume; the mesopores have a diameter of 6–12 nm and a volume accounting for 45%–65% of the total pore volume; and the micropores have a diameter less than 2 nm and a volume accounting for 15%–30% of the total pore volume. The macropores, mesopores, and micropores are interconnected in a three-level hierarchical topology. The specific surface area of ​​the silica framework is 400–700 m². 2 / g. The total grafting amount of the aminopropyl functional groups is 0.8~1.5 mmol / g, and a three-level stepped grafting density gradient is formed on the surfaces of macropore walls, mesopore walls, and micropore walls: the aminopropyl grafting density on the mesopore wall surface is 2~4 times that on the macropore wall surface, and the aminopropyl grafting density on the micropore wall surface is 4~8 times that on the macropore wall surface. The three-level pore structure and the three-level stepped grafting density gradient synergistically constitute a spatially decoupled structure for filtration and chemisorption. The macropores serve as low-resistance permeation channels for animal fat fluids and phospholipid micelles, the mesopores serve as chemisorption active regions for phospholipid molecule head groups, and the micropores serve as high-capacity capture regions for free fatty acid molecules.

[0009] The filter aid is prepared by a dual-template agent synergistic directed sol-gel method combined with a stepwise template agent removal process and vapor-phase grafting modification. The specific process includes: using tetraethyl orthosilicate (TEOS) as the silicon source, hexadecyltrimethylammonium bromide (CTAB) as the mesoporous template agent, and polystyrene microspheres with a particle size of 0.5–2 μm as the macroporous template agent; adjusting the pH of the reaction system to 9–11 with ammonia; carrying out a hydrolysis-condensation reaction at 50–70 °C; and aging for 18–36 h to form a gel; then sequentially removing CTAB by extraction with an ethanol-hydrochloric acid mixed solution and removing the polystyrene microspheres by calcination at 550–650 °C for 2–4 h in a stepwise template agent removal process; finally, using dry nitrogen as the carrier gas, and conducting a reaction at an aminopropyltrimethoxysilane (APTMS) vapor partial pressure of 200–800 Pa, a temperature of 100–140 °C, and a time of 4–8 hours. Under h conditions, gas-phase grafting was performed, and the differences in diffusion mechanisms of APTMS molecules in different pore sizes (bulk diffusion in macropores, transition zone diffusion in mesopores, and Knudsen diffusion in micropores) were utilized to form a step-like grafting density gradient within the same particle.

[0010] Beneficial Effects: Compared with existing technologies, this invention achieves five significant performance improvements. First, the filtration rate is increased by more than 60% compared to single-stage porous silica filter aids and by more than 120% compared to traditional diatomaceous earth filter aids. Second, the phospholipid removal rate reaches over 96%, a significant improvement compared to the 50%~70% level of traditional diatomaceous earth filter aids. Third, the free fatty acid reduction reaches 0.3%~0.5%, achieving a deep chemical deacidification effect that is difficult to achieve through physical adsorption. Fourth, the oil recovery rate is not less than 97%, an improvement of more than 5 percentage points compared to the 90%~93% of traditional diatomaceous earth filter aids. Fifth, after 30 minutes of filtration, the Darcy permeability coefficient remains above 80% of its initial value, and after 2 hours of filtration, it remains above 60%, significantly extending the filtration cycle. The aforementioned effect is not simply a linear improvement brought about by parameter optimization, but rather stems from the spatial decoupling and synergy between the three-level pore structure and the three-level stepped graft density gradient. It reflects the precise division of labor among different functional spaces within the same particle, a synergistic effect that existing technologies based on the idea of ​​uniform modification cannot predict or achieve. Attached Figure Description

[0011] Figure 1 This is a scanning electron microscope (SEM) image of the multi-porous silica filter aid obtained in Example 1 of the present invention.

[0012] Figure 2 This is a transmission electron microscope (TEM) image of the filter aid obtained in Example 1 of the present invention.

[0013] Figure 3 The nitrogen adsorption-desorption isotherm and BJH pore size distribution curve of the filter aid obtained in Example 1 of this invention are shown.

[0014] Figure 4 This is a comparison of the Fourier transform infrared (FT-IR) spectra of the filter aid obtained in Example 1 of the present invention before and after APTMS vapor-phase grafting.

[0015] Figure 5 This is an XPS N 1s signal depth distribution map of the filter aid obtained in Example 1 of the present invention after Ar ion beam depth profiling.

[0016] Figure 6 This is a comparison chart of the Darcy permeability coefficient decay curves of Example 1 and Comparative Example 1 of the present invention as a function of filtration time.

[0017] Figure 7 This is a bar chart comparing the phospholipid removal rate, free fatty acid reduction, and filtration rate improvement rate in the refining and filtration of animal fats in Examples 1 to 3 and Comparative Examples 1 to 5 of the present invention. Detailed Implementation

[0018] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to illustrate specific implementations of the present invention and do not constitute a limitation on the scope of protection of the present invention. Unless otherwise specified, the features in the following embodiments can be combined with each other.

[0019] Example 1

[0020] A method for preparing a multi-level porous silica filter aid for refining animal fats includes the following steps:

[0021] (1) Preparation of macroporous template agent for polystyrene microspheres: Polystyrene microspheres were prepared by emulsion polymerization without emulsifiers. 100 mL of deionized water was placed in a 500 mL three-necked flask, and high-purity nitrogen was introduced for deoxygenation for 30 min, followed by heating to 70 °C. 10 g of styrene, 0.15 g of potassium persulfate (as initiator), and 0.02 g of sodium dodecyl sulfate (as stabilizer) were added to the reaction system in the following ratio. The reaction was carried out at a constant temperature of 70 °C for 12 h under mechanical stirring at 300 r / min. After the reaction was completed, the mixture was cooled to room temperature, centrifuged, washed three times with deionized water, and washed twice with anhydrous ethanol. The microspheres were then vacuum dried at 60 °C for 24 h to obtain polystyrene microspheres with a particle size of 1.0 μm and a particle size distribution (CV value) of no more than 5%.

[0022] (2) Synergistic synthesis of a hierarchical porous silica framework using dual templates: 3.6 g of cetyltrimethylammonium bromide (CTAB, purity not less than 99%) was dissolved in 160 mL of deionized water, and 40 mL of anhydrous ethanol was added and stirred for 30 min to form a homogeneous mesoporous template solution. 2.0 g of the polystyrene microspheres prepared above were added to the solution and ultrasonically dispersed for 15 min to form a dual template mixture. Under stirring conditions in a 25 °C water bath, 16 mL of 25 wt% ammonia was slowly added dropwise to adjust the pH to 10.0 ± 0.2. After stirring for 5 min, 20.8 g of tetraethyl orthosilicate (TEOS) was added dropwise over a period of 30 min. The system was heated to 60 °C for hydrolysis and polycondensation reaction for 12 h, then stirring was stopped, and the system was allowed to stand at 60 °C for 24 h for aging. After aging, the gel was washed with deionized water until neutral and then filtered and collected.

[0023] (3) Stepwise template removal: The gel was transferred to an enamel beaker, and a hydrochloric acid-ethanol mixture (hydrochloric acid to ethanol volume ratio 1:7) was added. The mixture was refluxed at 70 °C for 18 h, filtered, and the extraction was repeated once to completely remove the CTAB template. After extraction, the gel was washed twice with anhydrous ethanol and once with deionized water, and then vacuum dried at 80 °C for 12 h. The dried gel was placed in a muffle furnace and calcined at a heating rate of 2 °C / min to 600 °C and held for 3 h to completely remove the polystyrene microsphere template. After natural cooling, an amorphous silica framework with a tertiary porous structure was obtained, denoted as SiO2-HP.

[0024] (4) APTMS vapor-phase grafting modification: 5.0 g of SiO2-HP powder was placed in a sample boat of a horizontal chemical vapor deposition reactor. The reactor was pretreated at 110 °C under vacuum (10 Pa) for 2 h to remove adsorbed water from the pores. After the vacuum pump was turned off, dry nitrogen was introduced as the carrier gas (flow rate 100 mL / min). At the same time, the temperature of the bubbler containing aminopropyltrimethoxysilane (APTMS) was controlled at 65 °C to stabilize the APTMS vapor pressure at about 500 Pa. The carrier gas carried the APTMS vapor into the reactor, and the reaction temperature was maintained at 120 °C for 6 h of vapor-phase grafting. After grafting, the APTMS vapor source was turned off, and nitrogen was continued to purge for 30 min to remove unreacted APTMS. After natural cooling to room temperature, APTMS functionalized hierarchical porous silica filter aid was obtained, denoted as APTMS@SiO2-HP.

[0025] (5) Particle size control: APTMS@SiO2-HP powder was classified by air classifier, and the fraction with particle size of D50=18.5 μm and D90 / D10=4.2 was collected as filter aid product.

[0026] (6) Structural characterization results: Nitrogen adsorption-desorption isotherm analysis (e.g. Figure 3 As shown in the figure, the BET specific surface area of ​​SiO2-HP is 586 m². 2 / g, total pore volume is 1.12 cm³ 3 / g. The BJH pore size distribution exhibits a typical bimodal curve (mesopore peak at 8.5 nm, micropore peak at 1.7 nm), which, combined with the macropore distribution measured by mercury intrusion spectroscopy (peak at 1.2 μm), confirms the existence of a tertiary pore structure. The calculated volume percentages of each pore level are: macropore volume 22%, mesopore volume 55%, and micropore volume 23%. Scanning electron microscopy images (e.g.) Figure 1 As shown in the image, the macropores are arranged in a quasi-spherical packing pattern, with a uniform distribution of pore diameters ranging from 0.8 to 1.5 μm. Transmission electron microscopy (TEM) images (e.g.) Figure 2 As shown in the image, an ordered worm-like mesoporous structure exists within the macropore walls, with numerous micropores smaller than 2 nm scattered throughout the structure. FT-IR spectra (as shown in the image) Figure 4 As shown in the figure, APTMS@SiO2-HP shows a difference in growth rate at 2930 cm⁻¹ compared to ungrafted SiO2-HP. -1 With 2860 cm -1 A distinct -CH2- symmetric and antisymmetric stretching vibration peak appears at 1560 cm⁻¹. -1 The presence of a -NH2 angular vibration peak nearby confirms successful APTMS grafting. XPS analysis revealed a nitrogen content of 1.42 at%, based on which the total aminopropyl grafting amount was calculated to be 1.15 mmol / g.

[0027] (7) Characterization of stepped graft density gradient: XPS (Ar) depth profiling using Ar ion beam + The N / Si atomic ratio at different depths was measured using a beam energy of 2 keV and an etching rate of approximately 0.3 nm / min. Figure 5 (As shown). From the outer layer (depth no more than 0.5 μm from the pore opening, corresponding to the macropore wall) to the middle layer (depth 0.5~2 μm, corresponding to the mesopore wall) and then to the inner layer (depth greater than 2 μm, corresponding to the micropore wall), the N / Si atomic ratios are 0.018, 0.052, and 0.128 respectively. The calculated aminopropyl grafting densities on the macropore wall, mesopore wall, and micropore wall surfaces are 0.45, 1.30, and 3.20 atoms per square nanometer, respectively. The grafting density of the mesopore wall is 2.89 times that of the macropore wall, and the grafting density of the micropore wall is 7.11 times that of the macropore wall, falling within the 2~4 times and 4~8 times ranges defined in this invention, confirming the formation of a three-level stepwise grafting density gradient. Furthermore, in solid state... 29 Si NMR analysis showed T 3 (Tridentate silicon-oxygen bond) signals account for 38% of the total grafted silicon, T 2 (Bidentate silicon-oxygen bonds) account for 52%, T 1(Single-toothed silicon-oxygen bonds) account for 10%, while double-toothed and multi-toothed bonds account for 90%, demonstrating excellent bonding stability.

[0028] (8) Animal fat refining and filtration performance test: 1000 g of pre-degummed hog fat was heated to 75 °C, and 20 g (2.0 wt%) of APTMS@SiO2-HP filter aid was added. The mixture was mechanically stirred for 20 min to ensure thorough dispersion of the filter aid. The oil-filter aid mixture was then transferred to a pressure filtration device (filter cake area 0.01 m²). 2 The filter medium was a 200-mesh stainless steel screen, and pressure filtration was performed at 0.25 MPa. The cumulative filtrate volume was recorded online at each time point, and the Darcy permeability coefficient was calculated. The test results showed that the initial Darcy permeability coefficient at the start of filtration was 4.82 × 10⁻⁶. -13 m 2 After filtration for 30 minutes, the Darcy permeability coefficient remained at 85.2% of its initial value, and after 2 hours of filtration, it remained at 65.4%. The filtration rate was 73.2% higher than that of the single-stage mesoporous silica control sample (Comparative Example 1). Systematic analysis of the filtrate after filtration showed that the phospholipid content decreased from the initial 520 mg / kg to 18 mg / kg (phospholipid removal rate 96.5%), the free fatty acid content decreased from 1.20% to 0.78% (a reduction of 0.42%), the oil recovery rate was 97.8%, and the color showed a Lovibond red value R=1.6 and a yellow value Y=12.4, meeting the national first-grade refined lard standard (GB / T 8937-2006).

[0029] The main process parameters of Example 1 (TEOS 20.8 g, CTAB 3.6 g, polystyrene microspheres 2.0 g with a particle size of 1.0 μm, pH 10.0, hydrolysis condensation temperature 60 °C, aging for 24 h, calcination at 600 °C for 3 h, APTMS vapor partial pressure 500 Pa, grafting temperature 120 °C, and grafting time 6 h) fall at the center of the parameter range of the present invention, representing the optimal implementation scheme of the present invention, and the resulting filter aid has the best overall performance.

[0030] Example 2

[0031] Some process parameters of Example 1 were adjusted to the lower limit of the range defined in this invention to verify the applicability of the technical solution at the lower boundary of the parameters. The particle size of polystyrene microspheres was 0.6 μm (reducing the macropore size to the lower limit of this invention), the amount of CTAB was reduced to 3.0 g (correspondingly making the mesopore size more inclined to the lower limit), the amount of TEOS was kept unchanged at 20.8 g, the amount of ammonia was adjusted to stabilize the pH value of the system at 9.5, the hydrolysis and polycondensation temperature was adjusted to 55 °C (more inclined to the lower limit), and the aging time was extended to 30 h to compensate for the development of gel network at the lower synthesis temperature.

[0032] The hydrochloric acid-ethanol extraction conditions remained unchanged during the stepwise template removal process. The calcination temperature was adjusted to 580 °C, the holding time was 3.5 h, and the heating rate was maintained at 2 °C / min. During the APTMS vapor-phase grafting stage, the bubbler temperature was adjusted to 55 °C to reduce the APTMS vapor partial pressure to approximately 250 Pa (close to the lower limit of this invention), the nitrogen carrier gas flow rate was adjusted to 80 mL / min, the reaction temperature was 110 °C (close to the lower limit of this invention), and the grafting time was extended to 7 h to compensate for the decrease in grafting rate caused by the lower vapor partial pressure. The resulting APTMS-functionalized hierarchical porous silica filter aid was designated as Sample 2.

[0033] Structural characterization results of sample 2: BET specific surface area is 642 m². 2 / g (close to the upper limit of this invention, due to the formation of a finer mesoporous framework at a lower synthesis temperature), BJH pore size distribution shows a mesoporous peak at 6.8 nm (close to the lower limit of this invention), and a micropore peak at 1.5 nm. Mercury piezometry shows a macropore peak at 0.7 μm (close to the lower limit of this invention). Macropore volume percentage is 17%, mesopore volume percentage is 62%, and micropore volume percentage is 21%. Total aminopropyl grafting amount is 0.92 mmol / g (close to the lower limit of this invention). Ar ion beam depth profiling XPS shows aminopropyl grafting densities of 0.38, 0.95, and 2.28 per square nanometer for macropore walls, mesopore walls, and micropore walls, respectively. The mesopore / macropore ratio is 2.50, and the micropore / macropore ratio is 6.00, falling within the range defined by this invention.

[0034] Animal fat refining and filtration performance: 1000 g of pre-degummed cow fat was mixed with 15 g (1.5 wt%) of sample 2 filter aid and pressure filtered at 70 °C and 0.2 MPa. Test results: Initial Darcy permeability coefficient 4.12 × 10⁻⁶. -13 m 2 After filtration for 30 min, the initial value was maintained at 82.1%, and after filtration for 2 h, it was maintained at 62.3%. The filtration rate was increased by 64.1% compared with the single-stage mesoporous sample, the phospholipid removal rate was 96.1% (phospholipid content decreased from 480 mg / kg to 19 mg / kg), the free fatty acid content decreased by 0.35% (from 0.90% to 0.55%), the oil recovery rate was 97.2%, and the color was Lovibond Red value R=1.8 and Yellow value Y=13.0.

[0035] Example 2 verifies that even at parameter points where both pore size and grafting density gradient are near their lower limits, the technical solution of this invention can still achieve three core performance targets: a filtration rate increase of over 60%, a phospholipid removal rate of over 96%, and an oil recovery rate of over 97%. This demonstrates that the lower boundary of the scope of protection of this invention has sufficient technical support. Comparing Example 1 and Example 2, it can be seen that changes in pore size and grafting amount within the range lead to slight variations in performance indicators, but the spatial decoupling function of the tertiary pore structure and the stepped grafting density gradient effectively functions throughout the entire range. This reflects the robustness of the innovative core of this invention.

[0036] Example 3

[0037] Some process parameters of Example 1 were adjusted to the upper limit of the range defined in this invention, and poultry fat (chicken fat) was used as the refining target to verify the applicability of the technical solution at the upper limit of the parameters and under different types of animal fats. The polystyrene microsphere particle size was 1.8 μm (close to the upper limit of this invention), the amount of CTAB was increased to 4.2 g (to make the mesopore size biased towards the upper limit), the amount of TEOS was maintained at 20.8 g, the amount of ammonia was adjusted to stabilize the pH at 10.8, the hydrolysis and polycondensation temperature was 65 °C, and the aging time was 18 h (close to the lower limit of this invention).

[0038] During the step-by-step template removal process, the calcination temperature was adjusted to 640 °C (close to the upper limit of this invention), and the holding time was 2.5 h. In the APTMS vapor-phase grafting stage, the bubbler temperature was increased to 75 °C to achieve an APTMS vapor partial pressure of approximately 750 Pa (close to the upper limit of this invention), the nitrogen carrier gas flow rate was 120 mL / min, the reaction temperature was 135 °C (close to the upper limit of this invention), and the grafting time was 5 h. The resulting filter aid was designated as Sample 3.

[0039] Structural characterization results of sample 3: BET specific surface area is 432 m². 2 / g (close to the lower limit of this invention, due to the larger pore size and higher grafting density occupying more surface area), BJH pore size distribution shows that the mesopore peak is located at 11.2 nm (close to the upper limit of this invention), and the micropore peak is located at 1.8 nm. Mercury piezoelectric method measured the macropore peak at 1.7 μm (close to the upper limit of this invention). Macropore volume accounts for 28%, mesopore volume accounts for 48%, and micropore volume accounts for 24%. Total aminopropyl grafting amount is 1.42 mmol / g (close to the upper limit of this invention). Ar ion beam depth profiling XPS shows that the aminopropyl grafting densities of macropore walls, mesopore walls, and micropore walls are 0.58, 2.20, and 4.35 per square nanometer, respectively, with a mesopore / macropore ratio of 3.79 and a micropore / macropore ratio of 7.50, falling within the upper limit range defined by this invention.

[0040] Poultry oil refining and filtration performance: 1000 g of pre-degummed chicken feather oil was added to 22 g (2.2 wt%) of sample 3 filter aid, and pressure filtration was performed at a filtration temperature of 80 °C and a filtration pressure of 0.3 MPa. Test results: Initial Darcy permeability coefficient 5.45 × 10⁻⁶. -13 m 2 After filtration for 30 min, the initial value was maintained at 86.7%, and after filtration for 2 h, it was maintained at 68.1%. The filtration rate was 81.3% higher than that of the single-stage mesoporous sample (poultry oil has lower viscosity, resulting in the largest increase in filtration rate). The phospholipid removal rate was 97.2% (phospholipids decreased from 380 mg / kg to 11 mg / kg), free fatty acids decreased by 0.48% (from 0.82% to 0.34%), the oil recovery rate was 98.1%, and the color was Lovibond Red value R=1.3 and Yellow value Y=10.8.

[0041] Example 3 verifies that at parameter points where both pore size and grafting density gradients are at their upper limits, and in the refining application of poultry fats (which have lower viscosity than lard and tallow but similar free fatty acid content), the technical solution of this invention not only achieves the three core performance targets, but also increases the filtration rate by more than 81% (significantly higher than Examples 1 and 2), demonstrating that the technical solution of this invention has good adaptability to different types of animal fats. Examples 1 to 3 collectively cover the central, lower limit, and upper limit parameter regions of this invention, as well as the three target fats: lard, tallow, and poultry fat, confirming the completeness and robustness of the technical solution.

[0042] Comparative Example 1 (Single-level mesoporous structure with homogeneous liquid phase grafting)

[0043] Comparative Example 1 was designed as a reproduction scheme corresponding to the strongest comparative document D1 (Microporous and Mesoporous Materials 2020) to verify the performance advantage of "tertiary pores plus stepped gradient" over "monotype mesoporous structures plus uniform grafting". The macroporous template agent for polystyrene microspheres was omitted; only 3.6 g of CTAB was used as a single mesoporous template agent. Other synthesis parameters (TEOS dosage, pH, temperature, aging time) were the same as in Example 1, resulting in a monotype mesoporous silica framework. APTMS modification was performed using a liquid-phase post-grafting method: 3.0 g of monotype mesoporous silica powder was dispersed in toluene solvent, 1.2 mL of APTMS was added, and the mixture was refluxed at 110 °C for 24 h. After filtration, the mixture was washed three times with anhydrous ethanol and vacuum dried at 80 °C for 12 h. The resulting sample was designated as Comparative Example 1, with a BET specific surface area of ​​798 m². 2 / g, the mesoporous peak is located at 8.2 nm, the aminopropyl grafting amount is 1.18 mmol / g and is uniformly distributed (Ar ion beam depth profiling shows that the N / Si atomic ratio fluctuation at different depths does not exceed 15%). Under the same hog bristle oil filtration conditions as in Example 1, the results for Comparative Example 1 are: initial Darcy permeability coefficient 2.78 × 10⁻⁶. -13 m 2 (57.7% of Example 1), after filtration for 30 min, the permeability coefficient remained at 51.3% of the initial value (33.9 percentage points lower than Example 1), and after filtration for 2 h, it remained at only 32.1%. The phospholipid removal rate was 88.3%, the free fatty acid reduction was 0.28%, and the oil recovery rate was 94.5%. Comparative Example 1 confirms that without macroporous permeation channels and a stepped gradient decoupling structure, both the filtration rate and the impurity removal depth are significantly lower than those of the present invention.

[0044] Comparative Example 2 (tertiary pore structure retained but liquid phase grafting employed)

[0045] Comparative Example 2 was designed to retain the tertiary porous structure but replace gas-phase grafting with liquid-phase grafting to verify the necessity of the Knudsen diffusion gradient mechanism for the formation of a step-graft density gradient. The synthesis process of the tertiary porous silica framework was exactly the same as that of Example 1, denoted as the same batch of SiO2-HP. APTMS modification was performed using liquid-phase grafting: 5.0 g of SiO2-HP powder was dispersed in 100 mL of anhydrous toluene, 2.0 mL of APTMS was added, and the mixture was refluxed at 110 °C for 24 h. After filtration, the mixture was washed three times with anhydrous ethanol and vacuum dried at 80 °C for 12 h. The resulting sample was designated as Comparative Example 2, with a BET specific surface area of ​​562 m². 2 / g (close to the level of Example 1), total aminopropyl grafting amount 1.12 mmol / g (similar to Example 1). The key difference is that Ar ion beam depth profiling XPS showed that the aminopropyl grafting densities of macropore walls, mesopore walls, and micropore walls were 1.45, 1.52, and 1.48 per square nanometer, respectively, with ratios close to 1.0, indicating a uniform distribution rather than a stepwise gradient distribution. This is due to the complete wetting of all pore surfaces by toluene solvent during liquid-phase grafting. Under the same filtration conditions as Example 1, the results for Comparative Example 2 were: initial Darcy permeability coefficient 3.12 × 10⁻⁶. -13 m 2(Due to the excessively high grafting density on the macropore walls, the flow resistance of the macropores increased.) After 30 minutes of filtration, the initial value of 62.8% was maintained, and after 2 hours of filtration, it remained at 45.6%. The filtration rate was only 12.2% higher than that of the single-stage mesoporous comparative example 1 (far lower than the 73.2% of example 1). The phospholipid removal rate was 95.1%, the free fatty acid reduction was 0.38%, and the oil recovery rate was 96.0%. Comparative example 2 confirmed that even if the tertiary pore structure is retained, if uniform liquid-phase grafting is used instead of gas-phase gradient grafting, the low-resistance permeation channel function of the macropores will be destroyed by the excessively high aminopropyl coverage, the spatial decoupling mechanism will fail, and the filtration rate improvement cannot reach the 60% or more required by this invention.

[0046] Comparative Example 3 (grafting quantity exceeds the upper limit)

[0047] Comparative Example 3 retained the tertiary porous structure and vapor-phase grafting process, but the total APTMS grafting amount exceeded the upper limit of this invention (1.5 mmol / g), serving to verify the technical necessity of the upper limit for grafting amount. The synthesis process of the tertiary porous silica framework was exactly the same as in Example 1. In the APTMS vapor-phase grafting stage, the bubbler temperature was increased to 85 °C, resulting in an APTMS vapor partial pressure of approximately 1200 Pa (exceeding the upper limit of this invention), the reaction temperature was 145 °C (exceeding the upper limit of this invention), and the grafting time was extended to 10 h. The resulting sample was designated Comparative Example 3, with the BET specific surface area reduced to 298 m². 2 / g (significantly lower than the lower limit of this invention 400 m) 2 The total aminopropyl grafting amount was 2.05 mmol / g (exceeding the upper limit of this invention). The grafting density of the micropore walls was as high as 5.35 grafts per square nanometer, close to the monolayer saturation grafting density, resulting in some micropores being filled with aminopropyl spaces and becoming blocked. Under the same filtration conditions as Example 1, the results of Comparative Example 3 were: an initial Darcy permeability coefficient of 1.85 × 10⁻⁶. -13 m 2 (Due to excessive grafting on macropore walls and micropore blockage, the filtration resistance increased significantly.) After 30 min of filtration, the initial value of 43.2% was maintained, and after 2 h of filtration, it remained at 22.5%. The filtration rate was 33.5% lower than that of the single-stage mesoporous control example 1 (i.e., this scheme was actually inferior to the single-stage mesoporous control scheme). The phospholipid removal rate was 94.2% (micropore blockage caused a decrease in adsorption capacity), free fatty acids decreased by 0.41%, and the oil recovery rate was 92.8%. Comparative example 3 confirmed that exceeding the upper limit of this invention will lead to problems such as micropore blockage, increased macropore flow resistance, and deterioration of filtration rate. The upper limit of 1.5 mmol / g in this invention has clear technical necessity.

[0048] Comparative Example 4 (without APTMS vapor-phase grafting modification)

[0049] Comparative Example 4 was designed with a tertiary porous structure, retaining but completely omitting APTMS vapor-phase grafting modification, to verify the irreplaceable role of the aminopropyl functional group in the chemisorption of specific impurities in animal fats. The synthesis process was terminated at the SiO2-HP stage without any surface functionalization modification. After particle size fractionation, a SiO2-HP filter aid with a D50 of 18.7 μm was obtained. The BET specific surface area was 912 m². 2 / g (not occupied by APTMS), with pores consisting of pure silanol surfaces. Under the same hog bristle oil filtration conditions as in Example 1, the results for Comparative Example 4 were: initial Darcy permeability coefficient 5.26 × 10⁻⁶. -13 m 2 (Due to the absence of APTMS occupying the pore surface, the permeability coefficient is higher than that of Example 1). After filtration for 30 min, the initial value of 78.4% was maintained, and after filtration for 2 h, the value of 58.1% was maintained. The filtration rate was 68.5% higher than that of the single-stage mesoporous control example 1 (slightly lower than the 73.2% of Example 1, due to the lack of pre-capture effect of amino adsorption on phospholipid micelles). The phospholipid removal rate was 78.2% (significantly lower than the 96% required by this invention, due to the lack of amino-phosphoric acid chemical adsorption, relying only on physical capture). The reduction of free fatty acids was only 0.08% (due to the lack of amino-carboxyl acid-base coordination, there is no chemical deacidification ability). The oil recovery rate was 96.8%. The color was Lovibond red value R=2.8 and yellow value Y=18.5 (far worse than R=1.6 and Y=12.4 of Example 1). Comparative Example 4 confirms that the aminopropyl functional group introduced by APTMS gas-phase grafting is a necessary component for achieving deep removal of phospholipids and chemical removal of free fatty acids. The tertiary pore structure alone can only achieve physical interception and cannot achieve the deep impurity removal target required by this invention.

[0050] Comparative Example 5 (Baseline Control of Traditional Diatomaceous Earth Filter Aid)

[0051] Comparative Example 5 used commercially available refined diatomaceous earth filter aid (brand name CelaPure 545, average particle size 25 μm, BET specific surface area 2.2 m²). 2 / g (mainly macropores of 0.5~5 μm) served as a baseline control for the traditional process. Under the same swine oil filtration conditions as in Example 1, the results for Comparative Example 5 were: an initial Darcy permeability coefficient of 2.15 × 10⁻⁶. -13 m 2After filtration for 30 minutes, the initial value of phospholipid removal was maintained at 38.7%, and after filtration for 2 hours, it remained at 21.2%, with a filtration rate only 44.6% of that in Example 1. The phospholipid removal rate was 62.8% (compared to the literature-reported 50%~70% level for diatomaceous earth), the reduction in free fatty acids was only 0.05% (almost no chemical deacidification ability), the oil recovery rate was 91.3%, and the color was Lovibond red value R=3.2 and yellow value Y=21.5. Comparative Example 5, as the baseline of the traditional process, provides a clear comparison with Example 1 of this invention, demonstrating the significant advantages of this invention in five core performance indicators: a filtration rate increase of 124.2%, a phospholipid removal rate increase of 33.7 percentage points, a free fatty acid reduction increase of 0.37 percentage points, an oil recovery rate increase of 6.5 percentage points, and a Lovibond red value reduction of 1.6 and a yellow value reduction of 9.1.

[0052] The detection method is as follows:

[0053] (1) Nitrogen adsorption-desorption isotherms and BJH pore size distribution determination: A Micromeritics ASAP 2460 fully automated specific surface area and pore size analyzer was used. Samples were degassed under vacuum at 150 °C for 8 h, followed by nitrogen adsorption-desorption testing at liquid nitrogen temperature of 77 K. The BET specific surface area was determined using multi-point linear regression results within the relative pressure range of P / P0 = 0.05–0.25, and the total pore volume was determined using a single-point value at P / P0 = 0.99. Mesopore and micropore distributions were calculated using the BJH method (desorption branch) and the t-plot method, respectively.

[0054] (2) Macropore structure characterization: A Micromeritics AutoPore V 9600 mercury pressure analyzer was used, with a pressure range of 0.01~60000 psi and a pore size range of 3 nm~400 μm, to supplement the micron-sized macropore distribution that could not be covered by nitrogen adsorption-desorption. The macropore volume ratio was calculated by deducting the contribution of mesopores from the total mercury ingress in the mercury pressure test.

[0055] (3) FT-IR and XPS surface chemical characterization: FT-IR was performed using a Thermo Nicolet iS50 spectrometer with KBr pellet method, scanning range 400~4000 cm⁻¹ -1 4 cm resolution -1 XPS was performed using a Thermo ESCALAB 250Xi spectrometer with a monochromatic Al Kα X-ray source at an energy of 1486.6 eV. Ar ion beam depth profiling was performed using Ar... + The beam energy was 2 keV, the etching area was 2 mm × 2 mm, and the etching rate was approximately 0.3 nm / min (relative to a SiO2 standard). The N / Si atomic ratio was obtained by processing the N 1s to Si 2p peak intensities at different depths using Avantage software.

[0056] (4) Solid state 29 Si NMR: A Bruker Avance III 400 MHz wide-cavity NMR spectrometer was used, with a magic angle rotation (MAS) frequency of 10 kHz. 29 Si was observed at a frequency of 79.5 MHz, using a cross-polarized pulse sequence, and T was observed. 1 / T 2 / T 3 / Q 4 Signal determination of the number distribution of grafted silicon bonding teeth.

[0057] (5) Electron microscopy characterization: SEM was performed using a Hitachi SU-8010 field emission scanning electron microscope with an accelerating voltage of 5 kV. TEM was performed using a FEI Tecnai G2 F20 field emission transmission electron microscope with an accelerating voltage of 200 kV. The sample was dispersed in anhydrous ethanol and then dropped onto a carbon film copper grid.

[0058] (6) Oil quality testing: Phospholipid content was determined using the molybdenum blue colorimetric method (GB / T 5537-2008); free fatty acid content was determined using the acid-base titration method (GB / T 5530-2005), calculated as oleic acid; peroxide value was determined using the iodometric method (GB / T 5538-2005); color was determined using the Lovibond colorimetric method (GB / T 5525-2008), with the red value (R) and yellow value (Y) read in a 133 mm cuvette at 25 °C. Filtration rate was calculated using the cumulative filtrate volume-time curve of the pressure filtration device. The Darcy permeability coefficient was calculated using the formula k=(μLQ) / (AΔP), where μ is the dynamic viscosity of the oil, L is the filter cake thickness, Q is the volumetric flow rate, A is the filter cake area, and ΔP is the filtration pressure difference.

[0059] Table 1. Comparison of key performance indicators between Examples 1-3 and Comparative Examples 1-5 in animal fat refining and filtration.

[0060] sample <![CDATA[Specific surface area (m 2 / g)]]> Grafting amount (mmol / g) Gradient ratio (medium / large, micro / large) Filtration rate increased Phospholipid removal rate FFA decreases Oil recovery rate Example 1 586 1.15 2.89 / 7.11 73.2% 96.5% 0.42% 97.8% Example 2 642 0.92 2.50 / 6.00 64.1% 96.1% 0.35% 97.2% Example 3 432 1.42 3.79 / 7.50 81.3% 97.2% 0.48% 98.1% Comparative Example 1 798 1.18 (average) ≈1.0 (no gradient) Baseline (0%) 88.3% 0.28% 94.5% Comparative Example 2 562 1.12 ≈1.0 (uniform) 12.2% 95.1% 0.38% 96.0% Comparative Example 3 298 2.05 4.10 / 8.90 -33.5% 94.2% 0.41% 92.8% Comparative Example 4 912 0 (No grafting) none 68.5% 78.2% 0.08% 96.8% Comparative Example 5 2.2 0 (diatomaceous earth) none -55.4% 62.8% 0.05% 91.3% This invention is limited 400~700 0.8~1.5 2~4 / 4~8 ≥60% ≥96% 0.3~0.5% ≥97%

[0061] The data in Table 1 reveal the nonlinear synergistic effect of this invention relative to various comparative schemes, which cannot be explained by simple parameter summation. The first nonlinear phenomenon is the leap from Comparative Example 1 (mono-mesoporous structure with uniform grafting) to Example 1 (tertiary pore structure with stepped gradient). Retaining the tertiary pore structure alone (Comparative Example 2) only increases the filtration rate from 0% to 12.2%, and increasing the grafting amount alone is not significant (the grafting amount from Comparative Example 1 to Example 1 only decreases from 1.18 to 1.15); however, the synergy between the two (Example 1) achieves a 73.2% increase in filtration rate, which is several times greater than the sum of the individual contributions. This superlinear superposition reflects the true synergistic relationship between the tertiary pore structure and the stepped gradient. The low-resistance channel function of the macropores depends on the low aminopropyl coverage on the macropore walls to function, while the high-density chemisorption function of the mesopores and micropores depends on the specific grafting space allocation to avoid affecting the flow of the macropores. The spatial decoupling of the two constitutes the underlying mechanism for the improvement in filtration performance.

[0062] The second nonlinear phenomenon is the anomalous result of Example 3 (grafting amount exceeding the upper limit to 2.05 mmol / g). Following a linear approach, increasing the grafting amount from 1.15 in Example 1 to 2.05 should result in an approximately 80% increase in adsorption capacity. However, the experimentally measured filtration rate decreased from 73.2% to -33.5% (i.e., lower than the single-stage mesoporous comparative example 1), and the phospholipid removal rate also decreased from 96.5% to 94.2%. The mechanism of this anomalous phenomenon is that when the grafting amount is too high, the micropores are completely blocked by the aminopropyl groups filling the space, and the grafting density on the macropore walls simultaneously exceeds the limit, causing a sharp increase in macropore flow resistance, and the macropores, which were originally permeation channels, lose their function. This nonlinear negative feedback confirms the technical necessity of the grafting amount upper limit of 1.5 mmol / g in this invention, and that limiting the grafting amount upper limit is not a simple empirical choice, but rather a mechanism anchored based on the critical value of micropore blockage.

[0063] The third nonlinear phenomenon is the performance spectrum of Example 4 (with tertiary pores but without APTMS modification). The filtration rate of the simple tertiary pore structure increased by 68.5% (close to 73.2% in Example 1), indicating that the improvement in filtration rate is mainly contributed by the tertiary pore structure rather than by the step gradient independently. However, the phospholipid removal rate was only 78.2% (far lower than 96.5% in Example 1), and the reduction in free fatty acids was only 0.08% (far lower than 0.42%), indicating that the deep impurity removal performance is mainly contributed by aminopropyl chemisorption. This result seems to separate the two dimensions of filtration and adsorption, but it is actually a direct verification of the spatial decoupling concept of this invention: the filtration performance and adsorption performance are spatially shared by the macroporous segment and the mesoporous / microporous segment, respectively, and the two are coupled within the same material particle through a step gradient, which avoids mutual interference and maintains their respective optimality.

[0064] The three-level pore structure and the three-level stepped grafting density gradient synergistic mechanism of this invention involve the coupling of three levels: physical filtration, chemical adsorption, and gas-phase grafting kinetics. This spatial decoupling effect is difficult to understand from the perspective of existing technologies for uniformly modified porous materials. From the perspective of filtration physics, the macropore size of 0.5~2 μm is far larger than the phospholipid micelle size (50~500 nm) in animal fats, allowing the mainstream of fats to pass through the macropore channels unimpeded. At the same time, phospholipid micelles are initially physically intercepted in the macropore section, forming pre-filtration. The Hagen-Poiseuille equation shows that the permeation resistance of the macropore section is inversely proportional to the fourth power of the pore size, and the 0.5~2 μm macropore section contributes more than 85% of the permeation flux of the entire filter aid particle.

[0065] From a chemisorption perspective, the 6–12 nm mesopore size provides a good geometric match with the head size of phospholipid molecules (1–2 nm) and the heme-iron complex size (approximately 2 nm), allowing these intermediate-scale polar impurities to fully enter the mesopores and contact the high-density aminopropyl functional groups. Strong adsorption occurs between the aminopropyl-phosphate groups through multi-point hydrogen bonding and electrostatic interactions, with association constants far exceeding reported values ​​for phospholipid adsorption in vegetable oils such as olive oil. For micropores smaller than 2 nm, the primary adsorption targets are free fatty acid molecules (molecular size 0.7–1 nm). The extremely high specific surface area within the micropores (accounting for over 60% of the total specific surface area) provides ample adsorption sites. The aminopropyl-carboxyl acid-base coordination mechanism enables saturated fatty acids (such as palmitic acid and stearic acid, major components of animal fats) to acquire strong adsorption.

[0066] From the perspective of gas-phase grafting kinetics, the mean free path of APTMS molecules at 120 °C and 500 Pa is approximately 25 nm. For macroporous segments (pore size 0.5–2 μm), the Knudsen number Kn = λ / d is approximately 0.01–0.05, belonging to the bulk diffusion region, where molecular diffusion resistance is minimal but residence time is short. For mesoporous segments (pore size 6–12 nm), Kn is approximately 2–4, belonging to the transitional diffusion region, where the frequency of molecule-pore wall collisions is moderate. For microporous segments (pore size less than 2 nm), Kn is greater than 12, completely entering the Knudsen diffusion region, where molecular motion is dominated by the pore walls. The difference in the mechanism among the three diffusion regions directly translates into differences in the contact reaction time of APTMS molecules in different pore segments, thus forming a step-like grafting density gradient. This diffusion mechanism conversion is an intrinsic physical characteristic of gas-phase grafting, which cannot be achieved in liquid-phase grafting due to the complete wetting of all pore surfaces by the solvent.

[0067] The chemical mechanism chain of this invention involves four core reaction processes, each supported by a clear thermodynamic and kinetic basis. The first process is the TEOS sol-gel hydrolysis-condensation reaction, under alkaline catalytic conditions at pH 9-11: Si(OC2H5)4 + 4H2O → Si(OH)4 + 4C2H5OH (hydrolysis), 2Si(OH)4 → (HO)3Si-O-Si(OH)3 + H2O (condensation). Ammonia water acts as an alkaline catalyst to form Si-O through deprotonation of silanol groups. - Nucleophilic centers accelerate the condensation reaction, while the spatial occupancy of macroporous and mesoporous template agents (PS microspheres and CTAB micelles) in the sol-gel process determines the final topological arrangement of the tertiary channels.

[0068] The second process involves the vapor-phase grafting reaction of APTMS on the surface of tertiary channels. APTMS molecules undergo a condensation reaction with surface silanol groups (Si-OH) via three methoxy groups, releasing methanol: SiO2-Si-OH + (CH3O)3Si-CH2CH2CH2NH2 → SiO2-Si-O-Si(OCH3)2-CH2CH2CH2NH2 + CH3OH. This reaction may further result in bidentate or tripentate bonding to form T... 2 or T 3 The structure and the number of bonding teeth increase with increasing reaction temperature and reaction time. The APTMS@SiO2-HP obtained in Example 1 at 120 °C exhibits a TT... 2 Add T 3 The proportion reached 90%, confirming that most of the aminopropyl functional groups are anchored through multidentate bonding, exhibiting excellent chemical stability.

[0069] The third process involves the acid-base coordination adsorption of aminopropyl functional groups with free fatty acids during the refining and filtration of animal fats. The pK group of aminopropyl... a The pK of free fatty acids (represented by palmitic acid) is approximately 10.3. a Approximately 4.75, ΔpK a Approximately 5.5 causes the proton transfer equilibrium to strongly favor the formation of the ammonium carboxylate product: R-CH2NH2 + R'-COOH → R-CH2NH3 + ··· - OOC-R'. Association equilibrium constant K a Approximately 10 3 ~10 4 At oil temperatures of 65–85 °C, this adsorption is nearly irreversible, thus free fatty acids are effectively captured from the oil phase. The micropore size is as high as 3.20 nm. -2 The high aminopropyl density allows the local functional group concentration to reach 6 to 7 times that of traditional mesoporous adsorbents, which is beneficial to the complete adsorption of free fatty acids from both thermodynamic and kinetic perspectives.

[0070] The fourth process involves the multimodal adsorption of aminopropyl groups to the polar head of phospholipids. Phospholipid molecules (taking lysophosphatidylcholine, abundant in animal fats, as an example) contain phosphate groups (pK). a1 ≈3.0, pK a2 ≈7.0) and quaternary ammonium groups (N + (CH3)3, permanently positively charged). Under conditions of 65–85 °C with trace amounts of residual water in the oil phase, the phosphate group is partially deprotonated to form phosphate, which then reacts with aminopropyl groups in a protonated form (R-CH2NH3). + Electrostatic adsorption occurs; simultaneously, multi-point hydrogen bonds form between the P=O group of the phosphate group and the NH group of the aminopropyl group. Mesoporous segment: 1.30 nm -2 The density of the aminopropyl groups perfectly matches the cross-sectional area of ​​the phospholipid molecule head, allowing each phospholipid head to interact simultaneously with 2-3 aminopropyl functional groups, forming multi-point synergistic anchoring. Similarly, acid-base coordination is formed between the porphyrin ring carboxyl group and the aminopropyl group in the heme-iron complex, and the weak coordination between the Fe atom and the aminopropyl nitrogen lone pair further enhances adsorption. This is the main chemical mechanism by which the filter aid of this invention reduces the color of animal fats (Lovibond Red value decreased from 3.2 in Comparative Example 5 to 1.6 in Example 1).

[0071] In summary, the technical solution of this invention achieves a dual unity of structural and chemical design at the mechanistic level: the spatial hierarchy of the three-level pore structure determines the precise distribution of impurities at different scales, and the three-level stepped grafting density gradient determines the differentiated configuration of chemical activity in each pore segment. These two mechanisms are naturally coupled during the preparation stage through a gas-phase grafting Knudsen diffusion gradient mechanism, and work synergistically during the application stage through a filtration-adsorption spatial decoupling mechanism. This mechanistic system not only explains the inevitability of the present invention's performance indicators surpassing existing technologies, but also provides a new paradigm for spatial decoupling and gradient functionalization in subsequent material design in this field.

[0072] The application of the above-mentioned filter aid in the refining of animal fats includes: adding the filter aid at a weight ratio of 0.8% to 2.5% to lard, tallow or poultry fat that has undergone degumming and initial treatment, stirring mechanically until uniform, and then performing pressure filtration at a filtration temperature of 65 to 85 °C and a filtration pressure of 0.1 to 0.4 MPa, and collecting the filtrate as refined fat.

[0073] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural modifications made using the present invention specification, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A multi-porous silica filter aid for refining animal fats, characterized in that, include: An amorphous silica framework with a three-level pore structure consisting of macropores, mesopores, and micropores. The macropores have a diameter of 0.5–2 μm and a volume accounting for 15%–30% of the total pore volume; the mesopores have a diameter of 6–12 nm and a volume accounting for 45%–65% of the total pore volume; and the micropores have a diameter of less than 2 nm and a volume accounting for 15%–30% of the total pore volume. The macropores, mesopores, and micropores are interconnected in a three-level hierarchical topology. The silica framework has a specific surface area of ​​400–700 m². 2 / g; Aminopropyl functional groups are bonded to the surface of the silica framework pores via a vapor-phase grafting process, with a total grafting amount of 0.8~1.5 mmol / g; the aminopropyl functional groups form a three-level stepped grafting density gradient on the surfaces of the macropore walls, mesopore walls, and micropore walls, with the aminopropyl grafting density on the mesopore wall surface being 2~4 times that on the macropore wall surface, and the aminopropyl grafting density on the micropore wall surface being 4~8 times that on the macropore wall surface; the three-level pore structure and the three-level stepped grafting density gradient synergistically constitute a spatially decoupled structure for filtration and chemisorption, the macropores serving as low-resistance permeation channels for animal fat fluids and phospholipid micelles, the mesopores serving as chemisorption active regions for phospholipid molecule head groups, and the micropores serving as high-capacity capture regions for free fatty acid molecules; the particle size D50 of the filter aid is 5~30. μm, the filter aid is used for refining and filtering lard, tallow or poultry fat.

2. The multi-porous silica filter aid for refining animal fats and oils according to claim 1, characterized in that, The three-level channel structure satisfies the following connectivity topology: each macropore is connected to its adjacent macropores through 3 to 8 mesopore pathways, and each mesopore is connected to its adjacent mesopores through 2 to 5 micropore pathways. The three-level hierarchical topology forms a hierarchical permeation network of macropore trunk - mesopore branch - micropore tip.

3. The multi-porous silica filter aid for refining animal fats and oils according to claim 1, characterized in that, The aminopropyl functional group is connected on the surface of the silica framework via monodentate or multidentate bonds of Si-O-Si-CH2CH2CH2NH2, wherein the monodentate bond (T 1 The aminopropyl functional group accounts for 5% to 15% of the total grafting amount, and the bidentate bond (T) 2 ) accounts for 45%~60% of the total grafting, and tridentate bonding (T) 3 The proportion of three-tooth bonding accounts for 30% to 45% of the total grafting amount. The proportion of three-tooth bonding on the surfaces of macropore walls, mesopore walls and micropore walls shows a distribution characteristic that the smaller the pore size, the higher the proportion of three-tooth bonding.

4. The multi-porous silica filter aid for refining animal fats and oils according to claim 1, characterized in that, The vapor-phase grafting process achieves the three-stage step grafting density gradient through the following parameters: the partial pressure of aminopropyltrimethoxysilane vapor is 200~800 Pa, the grafting temperature is 100~140 °C, the grafting time is 4~8 h, and the carrier gas is dry nitrogen with a flow rate of 50~200 mL / min. These process parameters enable the aminopropyltrimethoxysilane molecules to be transported in the macroporous section by bulk diffusion, in the mesoporous section by transition zone diffusion, and in the microporous section by Knudsen diffusion.

5. The multi-porous silica filter aid for refining animal fats and oils according to claim 1, characterized in that, The silica framework is prepared by a dual-template agent synergistic guided sol-gel method, specifically including: using tetraethyl orthosilicate as the silicon source, hexadecyltrimethylammonium bromide as the mesoporous template agent, and polystyrene microspheres with a particle size of 0.5~2 μm as the macroporous template agent; adjusting the pH of the reaction system to 9~11 with ammonia; carrying out a hydrolysis-condensation reaction at 50~70 °C; and aging for 18~36 h to form a gel; the gel is first extracted with an ethanol-hydrochloric acid mixed solution to remove the hexadecyltrimethylammonium bromide, and then calcined at 550~650 °C for 2~4 h to remove the polystyrene microspheres, forming the amorphous silica framework with the three-level pore structure.

6. The multi-porous silica filter aid for refining animal fats and oils according to claim 1, characterized in that, The Darcy permeability coefficient of the filter aid, within the filter cake thickness range of 1~10 mm, decreases at a rate not exceeding 0.05 units per hour with filtration time. The Darcy permeability coefficient maintains more than 80% of its initial value after 30 minutes of filtration and more than 60% of its initial value within 2 hours after filtration is completed.

7. The multi-porous silica filter aid for refining animal fats and oils according to claim 1, characterized in that, The filter aid is added at a weight ratio of 0.8% to 2.5% in the refining and filtration of lard, tallow, or poultry oil. Under filtration temperatures of 65 to 85°C, the filtration rate is increased by more than 60% compared to single-stage porous silica filter aid, the phospholipid removal rate is not less than 96%, the free fatty acid reduction is 0.3% to 0.5%, the oil recovery rate is not less than 97%, and the color of the filtered oil has a Lovibond red value R not exceeding 2.0 and a yellow value Y not exceeding 15.

8. The multi-porous silica filter aid for refining animal fats and oils according to claim 1, characterized in that, The target impurities in the animal fat refining process include a four-level spectrum: keratin residues and bone meal impurities with a size of 5-50 μm, phospholipid micelles with a size of 50-500 nm, phospholipid molecules and heme-iron complexes with a size of 1-3 nm, and free fatty acid molecules with a size of 0.7-1 nm. The filter cake layer of the filter aid intercepts the keratin residues and bone meal impurities, the macropores permeate and pre-intercept the phospholipid micelles, the mesopores chemically adsorb the phospholipid molecules and heme-iron complexes, and the micropores capture the free fatty acid molecules with high capacity.

9. A method for preparing the multi-porous silica filter aid for refining animal fats according to any one of claims 1 to 8, characterized in that, Includes the following steps: Step 1: Tetraethyl orthosilicate, hexadecyltrimethylammonium bromide, polystyrene microspheres with a particle size of 0.5-2 μm, water, and ethanol are mixed in a predetermined ratio. The pH value is adjusted to 9-11 with ammonia. The mixture is stirred and hydrolyzed at 50-70 °C for 8-16 h, and then aged for 18-36 h to obtain a gel. Step 2: The gel is extracted with a hydrochloric acid-ethanol mixture (volume ratio 1:5-1:10) at 60-80 °C for 12-24 h to remove the hexadecyltrimethylammonium bromide. Step 3: The extracted gel is calcined in air at 550-650 °C for 2-4 h to remove the polystyrene microspheres, yielding a silica framework with a tertiary pore structure. Step 4: The silica framework is placed in a gas-phase grafting reactor, first degassed under vacuum at 110 °C for 2 h, then aminopropyltrimethoxysilane vapor is introduced using dry nitrogen as the carrier gas at a vapor partial pressure of 200-800 kJ / L. Vapor grafting was carried out under the conditions of Pa, temperature 100~140 °C, and time 4~8 h to obtain the multi-level porous silica filter aid for refining animal fats.

10. The application of the filter aid according to any one of claims 1 to 8 or the filter aid obtained by the preparation method according to claim 9 in the refining of animal fats and oils, characterized in that, The application includes adding the filter aid at a weight ratio of 0.8% to 2.5% to lard, tallow or poultry fat that has undergone degumming and initial treatment, mechanically stirring until uniform, and then performing pressure filtration at a filtration temperature of 65 to 85 °C and a filtration pressure of 0.1 to 0.4 MPa, and collecting the filtrate as refined oil.

Images