Bone trabecula-imitating gradient random porous structure model construction method, white calcium stone piezoelectric bone scaffold and preparation method of white calcium stone piezoelectric bone scaffold

Abstract

The invention discloses a model construction method of a gradient random porous structure of a simulated bone trabecula, a white calendite piezoelectric bone scaffold and a preparation method of the white calendite piezoelectric bone scaffold, and the model construction method provided by the invention is a design method based on a Thiessen polygon porous structure with seed point release gradient change and volume scaling coefficient linear gradient change. In the body center scaling process, interpolation calculation is conducted on scaling coefficients from the axis to the outermost end in the cylindrical domain in the radial direction, and therefore the model of the stochastic gradient porous structure with the pore diameter decreasing from inside to outside in the radial direction and the pore edge diameter increasing from inside to outside in the radial direction is finally formed, by designing the formula and process parameters of the white calcium stone ceramic slurry, the possible problems of over-curing, cracking and the like are avoided, and the white calcium stone piezoelectric ceramic bone scaffold with excellent bonding performance is prepared by adopting a 3D printing photocuring printing technology.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of bone scaffold design and preparation, and particularly relates to a model construction method of a gradient random porous structure simulating bone trabeculae, a white phosphorite piezoelectric bone scaffold and a preparation method thereof. BACKGROUND

[0002] Traditional grafts are mainly metal and bioceramic scaffolds. They have sufficient density and strength to replace the bone in the body for a long time, but metal materials are not degradable and prone to stress shielding, and have poor biological performance; bioceramic materials are usually degradable, but cannot provide an electric physiological microenvironment for bone repair to help the bone defect area achieve good tissue regeneration. Piezoelectric ceramic materials (such as lead zirconate titanate, barium titanate, KNN, etc.) have shown outstanding advantages in electromagnetic reconstruction, but compared with bioceramic materials, the biological performance of piezoelectric ceramics is poor, and most of the metal ions released by piezoelectric ceramics during degradation are toxic, which limits the use of piezoelectric ceramics. The existing structure of ceramic bone scaffolds is mostly regular structure such as grid structure, three-period minimal surface structure, etc., and the scaffold structure lacks bionics and functionality, making the scaffold function single and the performance limited. In summary, the bone repair scaffold is limited in terms of material and structure. White phosphorite is the second most abundant inorganic mineral in bone and teeth, accounting for about 25%-35% of the weight of bone. Compared with hydroxyapatite, which degrades very slowly, white phosphorite has a more moderate degradation rate in a physiological environment. It can gradually degrade with the growth of new bone, providing space and calcium and phosphorus sources for bone cell ingrowth, directly stimulating the proliferation and differentiation of osteoblasts. At the same time, white phosphorite has a certain piezoelectricity and can mimic the electric physiological microenvironment in the body to promote bone regeneration. At the same time, the random structure based on the Voronoi polygon has irregular multi-pore diameters simulating natural bone trabeculae, superior mechanical properties compared with ordinary grid structures, and adjustable porosity. Therefore, it has a great optimization space for piezoelectric scaffolds. SUMMARY

[0003] In view of the deficiencies of the prior art, a first object of the present application is to provide a model construction method of a gradient random porous structure simulating bone trabeculae.

[0004] A second object of the present application is to provide a preparation method of a white phosphorite piezoelectric bone scaffold of a gradient random porous structure simulating bone trabeculae.

[0005] A third object of the present application is to provide a white phosphorite piezoelectric bone scaffold of a gradient random porous structure simulating bone trabeculae prepared by the preparation method.

[0006] To achieve the above objects, the present application adopts the following technical solutions:

[0007] This invention provides a method for constructing a gradient random porous structure for trabecular bone. Based on a cylindrical domain, several spherical domains of radius r are formed within it. The spherical domains are divided into i layers from the inside out according to their distance from the central axis of the cylindrical domain. A radial layering seeding method is used to plant N seed points on the surface of a single spherical domain, generating primary cells of Thiessen polygons at these seed points. These primary cells are then exploded, and their vertices, volume centers, and face centers are extracted. A radially dependent volume center scaling method is used to scale the primary cells around their volume centers to form secondary cells, with a scaling factor K1. The distance between the body center and the axis of the cylinder is calculated by interpolation; the secondary cells are exploded to extract their vertices, and then the primary faces of the primary cells are scaled with the face center as the center to extract the vertices of the secondary faces. All vertex sets are collected so that adjacent vertices form a quadrilateral mesh and the quadrilaterals are smoothly connected to generate a smooth surface STL model; the STL model is exported and Boolean operation is performed with a cylindrical model with a radius larger than the STL model to obtain a random porous structure. Finally, by fitting the regression model, the gradient random porous structure model of the bone trabeculae is obtained.

[0008] The model construction method provided by this invention first uses a radial layered seeding method to determine the number of seed points. This radial layered seeding method is a discretization method that uses the radial distance from the central axis of the sphere to the cylindrical domain to create layers, and distributes seed points on each spherical surface according to a rule specific to the number of layers. This discretization method, based on the radial distance from the sphere to the cylindrical axis and the distribution of seed points on each spherical surface according to a rule specific to the number of layers, achieves gradient changes. Then, a Thiessen polygon porous structure design method based on linear gradient changes using a volume scaling factor is employed. During the volume-center scaling process... The invention employs a radially dependent body-centered scaling method, which dynamically and continuously adjusts the scaling ratio of each cell based on its radial distance from the reference axis (such as the central axis of a cylinder). This achieves gradient spatial discretization driven by geometric features, resulting in a non-uniform secondary cell mesh. In this invention, the scaling coefficients from the axis radially to the outermost end within the cylindrical domain are interpolated, ultimately forming a stochastic gradient porous structure with a radially decreasing aperture and radially increasing pore edge diameter, exhibiting a structure distribution similar to human bone. Furthermore, precise control of the model porosity is achieved through fitting a regression model.

[0009] The preferred approach is to use a GhPython Script battery to write a Python program to generate the cylindrical fields.

[0010] The preferred solution is N=ki+1 or N=k. i , where k is the seed point coefficient, and the value of k is a natural number between 1 and 5.

[0011] It is experimentally found that by using the relationship between the above-mentioned seed point coefficients k and i to confirm the number of seeded points, the seeded points can be controlled to exhibit a certain gradient change.

[0012] In a preferred solution, the Voronoi 3D cell and the Solid Difference cell are used to generate the primary cells of the Thiessen polygon at the seed points, and the Deconstruct Brep cell is used to explode the primary cells to extract their vertices, body centers, and face centers.

[0013] In a preferred solution, the Scale cell and the GhPython Script cell are used to write a Python program to scale the primary cells centered on the body center to form secondary cells.

[0014] In a preferred solution, the calculation formula of the scaling coefficient K1 is shown in Equation (1):

[0015] Equation (1);

[0016] Where, d is the distance between the body center and the central axis of the cylindrical domain, d min is the minimum value of the distance from the body center of each primary cell of the Thiessen polygon to the central axis of the cylindrical domain, d max is the maximum value of the distance from the body center of each primary cell of the Thiessen polygon to the central axis of the cylindrical domain, the scaling coefficient f1 is the volume scaling multiple of the primary cell of the Thiessen polygon farthest from the central axis of the cylindrical domain, 0.5 ≤ f1 < 1; f2 is the volume scaling multiple of the primary cell of the Thiessen polygon closest to the central axis of the cylindrical domain, 0 < f2 ≤ 0.5.

[0017] Through the above interpolation calculation formula, the scaling coefficient of the primary cell body center can be associated with its distance to the central axis of the cylindrical domain, realizing that the volume of the secondary cells decreases from the inside to the outside along the radial direction, the distance between the vertices of the primary and secondary cells gradually increases, and the diameter of the pore edges formed by connecting the vertices of the quadrilateral mesh gradually increases. Thus, a bionic structure distribution with a gradually decreasing porosity from the inside to the outside can be achieved. Therefore, the maximum scaling coefficient f1 and the minimum scaling coefficient f2 should not only be within the scaling feasible range (0, 1), but also ensure that there is enough scaling gap to make the bionic structure distribution feasible. It is experimentally found that when 0.5 ≤ f1 < 1 and 0 < f2 ≤ 0.5 are set, the obtained model is closest to the natural trabecula.

[0018] In a preferred solution, the Scale cell and the GhPython Script cell are used to write a Python program to scale the primary faces of the primary cells centered on the face center, and the scaling coefficient K2 is 0.6 - 0.8, and the vertices of the secondary faces are extracted. It is experimentally found that directly fixing the scaling coefficient K2 to 0.6 - 0.8 can make there be a part of reliable pore edges with a uniform large diameter inside the structure, ensuring the mechanical properties of the scaffold.

[0019] The preferred approach is to use Construct Mesh cells to construct quadrilateral meshes from adjacent vertices, and to use Weaverbird's Catmull-Clark Subdivision cells to achieve smooth connections between quadrilaterals, generating an STL model with a smooth surface.

[0020] In a preferred embodiment, after exporting the STL model, the non-closed surface is repaired using 3-magic, and then the STL model is subjected to Boolean operation with a cylindrical model with a radius 0.1-0.4 mm larger to obtain a random porous structure.

[0021] In the preferred embodiment, the process of fitting the regression model is as follows: using the response surface methodology to fit the relationship between k, f1, f2 and the porosity of the gradient random porous structure model of the simulated bone trabeculae, and drawing a three-dimensional graph, thus obtaining the model.

[0022] By fitting the porosity of k, f1, f2 and the gradient random porous structure model of the bone trabeculae, precise control of porosity is achieved, and finally a model of the gradient random porous structure of the bone trabeculae with radius r, height h, and porosity of 30%-80% is obtained.

[0023] This invention also provides a method for preparing a white phosphogypsum piezoelectric bone scaffold with a gradient random porous structure that mimics bone trabeculae. The method involves mixing white phosphogypsum nanopowder, curing raw materials, and additives according to a designed ratio to obtain a ceramic slurry. The ceramic slurry is then added to a photopolymerization printing device, and photopolymerization printing is performed based on a model of the gradient random porous structure that mimics bone trabeculae to obtain a ceramic preform. The ceramic preform is then degreased and sintered to obtain the white phosphogypsum piezoelectric bone scaffold.

[0024] In a preferred embodiment, the process of obtaining the white phosphogypsum nanoparticles is as follows: zircon beads with a diameter of 10-15 mm, zircon beads with a diameter of 5-8 mm, and white phosphogypsum powder are placed in a ball mill jar at a mass ratio of 1-3:1-2:1 and wet ball milled. During the wet ball milling process, anhydrous ethanol is used as the medium, the ball milling speed is 300-400 r / min, and the time is 8-12 h. After the wet ball milling is completed, the particles are dried, ground through a 60-100 mesh sieve, and the material passing through the sieve is obtained. The white phosphogypsum powder is obtained by chemical synthesis.

[0025] In a further preferred embodiment, the drying temperature is 70-100℃ and the drying time is 20-24h.

[0026] This invention uses white phosphogypsum powder obtained by chemical synthesis, which is then ball-milled and passed through a 60-100 mesh sieve. The powder obtained from the sieve is used as a ceramic powder raw material. Experiments have shown that the WH particles obtained by chemical synthesis and drying still exhibit small particle agglomeration after direct manual grinding, affecting the slurry performance and printing accuracy. After ball milling, the WH particles can be better dispersed, and after drying, sieving and grinding will yield a finer WH powder.

[0027] In a preferred embodiment, the curing raw material comprises polyethylene glycol (200) diacrylate (PEG200DA), trimethylolpropane triacrylate resin (TMPTA), and polyethylene glycol (PEG200); by volume ratio, polyethylene glycol (200) diacrylate (PEG200DA): trimethylolpropane triacrylate resin (TMPTA): polyethylene glycol (PEG200) = 45~55: 20~40: 5~35.

[0028] Experiments revealed that PEG200DA, with its hydrophilic PEG segments, exhibits good compatibility with the hydrophilic surface of leucoxene powder, facilitating uniform dispersion of the powder in the resin and reducing agglomeration. This is crucial for preparing high-solids-content, low-viscosity slurries. Furthermore, as a bifunctional resin monomer, it forms a linear cross-linked network after curing, endowing the ceramic green body with a certain degree of flexibility and toughness. TMPTA, a trifunctional photosensitive resin, rapidly participates in the photocuring reaction, forming a highly cross-linked three-dimensional network. This significantly improves the mechanical properties of the cured ceramic green body, neutralizes the flexibility of PEG200DA, and accelerates the curing reaction rate. Adding a relatively small amount of PEG200 can reduce slurry viscosity, while its hydrophilic segments further aid in the dispersion of leucoxene powder, stabilizing the slurry. During the debinding stage, it melts and volatilizes / decomposes at lower temperatures, leaving tiny voids between ceramic particles to provide gas escape paths and reducing the possibility of debinding cracking.

[0029] In a preferred embodiment, the additive comprises a dispersant, an anti-settling agent, a leveling agent, a defoamer, a photoinitiator, and a light absorber, wherein, by mass ratio, the dispersant: anti-settling agent: leveling agent: defoamer: photoinitiator: light absorber = 1~5: 0.1~2: 0.1-2: 1~4: 0.1~3: 0.01~1;

[0030] The dispersant is KMT-3331, the anti-settling agent is Sago-8810X, the leveling agent is Rad2500, the defoamer is SRE-2022A, the photoinitiator is photoinitiator 184, and the light absorber is UV-531.

[0031] Experiments revealed that white phosphogypsum powder has low absorbance, and the cured thickness and printing accuracy obtained by commonly used formulas are not suitable for high-precision printing of porous scaffolds. Therefore, this invention innovatively introduces a light absorber into the formula used for white phosphogypsum powder. Since the absorbance of the light absorber is greater than that of the resin, which is greater than that of white phosphogypsum, the addition of the light absorber increases the overall UV absorption rate of the slurry, thereby reducing the light penetration depth, reducing the light scattering effect, significantly reducing the cured thickness, and significantly improving the printing accuracy. In addition, in this invention, the dispersant KMT-3331 is adsorbed on the powder surface to achieve primary dispersion and prevent agglomeration, laying the foundation for obtaining a uniform slurry. The anti-settling agent Sago-8810X establishes a three-dimensional thixotropic network, which works synergistically with the dispersant to ensure that the slurry has good stability and rheological properties. Defoamer SRE-2022A is responsible for eliminating internal defects and ensuring that the slurry body is dense and bubble-free; while leveling agent Rad-2500 quickly adjusts the surface tension, which can promote the smoothness of the liquid layer after scraping and ensure smooth printing. Photoinitiator 184 acts as a high-efficiency trigger, absorbing ultraviolet light of a specific wavelength to trigger the rapid polymerization and cross-linking of resin monomers and complete curing.

[0032] In a preferred embodiment, the ceramic slurry comprises, by volume ratio: 30-45 vol% ceramic powder, 50-65 vol% curing raw material, and 5-8 vol% additives.

[0033] In a preferred embodiment, the mixing speed is 1000-1800 r / min, and the mixing time is 7-15 min.

[0034] In a preferred embodiment, the parameters for the photopolymerization printing are: laser exposure energy 0.3~0.9 J / cm². 2 The slice thickness is 40-60 μm. By controlling the laser exposure energy within the above range and the slice thickness is 40-60 μm, micron-level high-precision processing of white phosphogypsum can be achieved.

[0035] In a preferred embodiment, the sintering process is as follows: the degreased billet is first placed in a high-temperature furnace and kept at 500-700℃ for 1.5-2.5h, and then placed in a microwave sintering furnace and heated to 650-1100℃ at a heating rate of 20-40℃ / min for sintering for 10-20min.

[0036] This invention first performs decarburization through sintering in a conventional high-temperature sintering furnace, followed by microwave sintering for densification. Conventional sintering furnaces heat from the outside in through heat conduction and radiation, and the heating rate, holding time, and atmosphere (such as air) can be precisely controlled to ensure that organic matter is fully oxidized into gas (such as CO2) and safely discharged, avoiding blistering and cracking of the green body due to rapid internal gas production or carbon residue. After preliminary sintering, neck connections begin to form between particles, achieving a certain mechanical strength. This process requires a uniform heating environment to ensure uniform shrinkage and minimal stress in all parts of the green body, preventing deformation. Therefore, a conventional box furnace is chosen. Microwave sintering can achieve complete densification in a short time. This process can effectively inhibit abnormal grain growth, which is beneficial for obtaining a microstructure with fine, uniform grains and better mechanical properties. At the same time, it avoids the loss of piezoelectricity of white phosphogypsum after long-term high-temperature sintering due to dehydrogenation.

[0037] After sintering, a high-precision white phosphogypsum piezoelectric ceramic support with a porosity of 50-80% is obtained. Its piezoelectric properties are measured by atomic force microscopy, showing a distinct butterfly curve and a compressive strength of 10-50 MPa.

[0038] The present invention also provides a white phosphogypsum piezoelectric bone scaffold prepared by the above preparation method.

[0039] Beneficial effects

[0040] This invention provides a white phosphogypsum piezoelectric ceramic bone scaffold with a gradient random porous structure mimicking bone trabeculae. The structural advantages are verified through a novel structural design method and finite element simulation. Combined with 3D printing photopolymerization technology, a novel and biomimetic white phosphogypsum piezoelectric ceramic scaffold is obtained. Specifically, addressing the problems of regular scaffold structures in existing technologies leading to weak correlation with human bone structure and lack of biomimicry, a porous structure design method based on Thiessen polygons is proposed, using gradient changes at seed points and linear changes in volume scaling coefficients. Addressing the issues of existing bone scaffold bioceramic materials failing to provide an electrophysiological microenvironment and lacking functionality, and piezoelectric ceramic materials having poor biocompatibility and easily releasing toxic metal ions, a porous scaffold using white phosphogypsum biopiezoelectric ceramic as the material is proposed. The mechanical and electrical properties of the scaffold are verified through finite element simulation. These methods not only solve the problems of existing bone scaffolds having a single structure and lacking inductive properties, and existing bone scaffold materials not being able to simultaneously achieve both bioactivity and electroactivity to promote osteogenesis, but also provide a new design scheme for bone repair scaffolds by accurately determining the scaffold performance through finite element simulation. To address the potential forming problems that may occur in 3D printing photopolymerization printing technology that does not involve white calcium phosphate, we can avoid potential problems such as over-curing and cracking by designing the white calcium phosphate ceramic slurry formula and process parameters, thereby obtaining a high-precision, high-performance white calcium phosphate piezoelectric ceramic bone scaffold.

[0041] This invention achieves outstanding results in the design of 3D-printed ceramic bone scaffolds by combining random porous structure design with photopolymerization technology:

[0042] Advantages of Random Porosity Controllability: Based on the linear scaling of the Thiessen polyhedron, the synergistic effect of the seed point placement coefficients, inner scaling coefficient, and outer scaling coefficient on the structural porosity is obtained using the response surface methodology. Furthermore, precise control of the model porosity is achieved through a fitted regression model. R0 2 Reaching 0.9999, adjust R 2 The accuracy of the regression model reached 0.9998, demonstrating that the prediction results are highly accurate and can achieve precise control of total porosity from 30% to 80%. The model pore size range is 100μm-2000μm, providing good conditions for bone and blood vessel ingrowth.

[0043] High-precision manufacturing process and high-performance evaluation of white phosphogypsum ceramic scaffolds: The optimal formula for the white phosphogypsum ceramic slurry was achieved by adjusting the content and type of dispersant, thus improving its high fluidity and printing accuracy. Adjusting the light absorber content in the ceramic slurry formula improved the photocuring accuracy of the white phosphogypsum ceramic by 90%, with over-curing as low as 15μm. Finite element simulation results predicted the voltage the bone scaffold could provide; the scaffold could provide an electric field strength of 1mV / cm-100mV / cm under stress, a range consistent with the field strength required for promoting bone repair. Finite element simulation and experimental verification yielded the following mechanical properties of the scaffold: compressive strength 1-20MPa, Young's modulus 200MPa-2GPa, sufficient to cover the mechanical strength required for human cancellous bone. Attached Figure Description

[0044] Figure 1 A schematic diagram of the process of constructing a gradient random porous structure for bone-like trabeculae, wherein (a) a spherical domain is generated and random points are added; (b) Thiessen polygons are generated; (c) the unit cell is scaled with the body center as the center; (d) the unit cell surface is scaled with the face center as the center; (e) all vertex sets are obtained; (f) four adjacent points are connected to form a surface; (g) smoothing is performed; and (h) the derivation process is performed.

[0045] Figure 2 It is the correspondence between the response surface methodology and the seed point coefficient k, scaling factor f1, f2.

[0046] Figure 3 It is a model diagram of a 50%-80% gradient random porous structure.

[0047] Figure 4 A photograph of the white phosphogypsum piezoelectric bone scaffold prepared in Example 1. Detailed Implementation

[0048] Example 1

[0049] 1. Model construction of gradient random porous structure for bone-like trabeculae

[0050] First, a Python program is written using the GhPython Script battery to generate a cylindrical region with a radius of 5mm and a height of 10mm. Based on this cylindrical region, a spherical region with a radius of r=1mm is generated within it. According to the distance of the sphere from the central axis of the cylinder, the entire sphere is divided into i=3 layers from the inside out, allowing seed points to be randomly distributed within the sphere. The number of seed points N projected onto the surface of a single sphere is... , where k=2, thereby controlling the seed point to exhibit a gradient change that increases radially from the inside to the outside.

[0051] Using a Voronoi 3D cell and a Solid Difference cell, primary cells of a Thiessen polygon are generated at the seed point. The primary cells are then exploded using a Deconstruct Brep cell to extract their vertices, volume center, and face center. A Python program is written using a Scale cell combined with a GhPython Script cell to scale the primary cells around their volume center to form secondary cells. The scaling factor K1 is calculated by interpolating the distance from the volume center to the cylinder's axis. The secondary cells are then exploded using a Deconstruct Brep cell to extract their vertices.

[0052] The scaling factor K1 is calculated using the formula shown in equation (1):

[0053] Equation (1);

[0054] Where d is the distance between the body center and the central axis of the cylindrical domain, d min d represents the minimum distance from the body center of the primary cell of each Thiessen polygon to the central axis of the cylindrical domain. max f1 is the maximum distance from the volume center of the primary cell of each Thiessen polygon to the central axis of the cylindrical domain. The scaling factor f1 is the volume scaling factor of the primary cell of the Thiessen polygon farthest from the central axis of the cylindrical domain, and f2 is the volume scaling factor of the primary cell of the Thiessen polygon closest to the central axis of the cylindrical domain.

[0055] The scaling factors are set to f1=0.5 and f2=0.1. By adjusting the scaling factors f1 and f2, the radial gradient change of the porosity of the support can be achieved.

[0056] A Python program was written using Scale cells and GhPython Script cells to scale the primary faces of primary cells around their center, with a fixed scaling factor of 0.8, and extract the vertices of secondary faces. All vertex sets were collected, and Construct Mesh cells were used to construct quadrilateral meshes from adjacent vertices. Weaverbird's Catmull-ClarkSubdivision cells were used to achieve smooth connections between quadrilaterals, generating a smooth surface STL model. After exporting, 3-magic was used to repair unclosed surfaces. Then, Boolean operations were performed on this STL model with a cylindrical model with a radius 0.2 mm larger, resulting in a random porous structure. Finally, the response surface methodology was used to fit the relationship between k, f1, f2, and the porosity of the support structure, achieving precise control of the porosity P.

[0057]

[0058] The final result is a radially gradient random porous scaffold with a radius of 5 mm, a height of 10 mm, and a porosity of 50%, exhibiting increasing porosity from the inside out.

[0059] 2. Fabrication of a gradient random porous white phosphogypsum piezoelectric bone scaffold with simulated bone trabeculae.

[0060] 0.34 mol of calcium hydroxide (Ca(OH)2) and 0.15 mol of magnesium hydroxide (Mg(OH)) were added to 300 ml of deionized water and heated at 100 °C for 30 min with stirring. Phosphoric acid was then slowly added dropwise until the pH reached 3.2. The mixture was stirred and refluxed at 100 °C for 10 h and aged at room temperature for 14 h. After filtration through filter paper, the mixture was dried in a drying oven for 24 h to obtain WH powder. The dried WH powder was then mixed with anhydrous ethanol as a medium and ball-milled at 350 r / min for 12 h with a mass ratio of 12 mm zirconium beads: 7 mm zirconium beads: powder of 2:1:1. The resulting mixture was then placed in a ball mill and milled for 24 h with a mass ratio of 350 r / min. The resulting mixture was then placed in a drying oven at 80 °C for 24 h and ground and sieved to obtain WH nanoparticles.

[0061] Take 100g of WH nanoparticles, 13.537g of TMPTA, 20.511g of PEG200DA, 9.025g of PEG200, 0.720g of Rad2500, 0.720g of 8810X, 1.439g of 2022, 3.167g of KMT-3331, 0.2154g of 184, and 0.0431g of UV-531. Mix the above raw materials directly at 1800 r / min for 10 minutes to obtain a ceramic slurry with a solid content of 35 vol%. Use a laser exposure energy of 0.313 J / cm². 2The resulting ceramic preform had a cured thickness of 86 μm and an overcured width of <36 μm.

[0062] The ceramic blank was then degreased by heating it to 350℃ at a rate of 1℃ / min and holding it for 120 min, then heating it to 500℃ at a rate of 0.1℃ / min and holding it for 120 min, then heating it to 600℃ at a rate of 0.2℃ / min and holding it for 120 min, and finally cooling it to 100℃ at a rate of -1℃ / min. The degreasing stage was then completed. After degreasing, the sample was first sintered at 600℃ for 2 hours in a box furnace for decarburization, and then sintered at 950℃ for 10 minutes in a microwave sintering furnace at a rate of 30℃ / min. This yielded a high-precision white phosphogypsum piezoelectric ceramic scaffold with a porosity of 50%, a compressive strength of 11.36 MPa, and an elastic modulus of 1.58 GPa, which meets the mechanical requirements of human cancellous bone. It also has piezoelectric response capability, which can simulate the electrophysiological microenvironment of human bone and promote bone repair.

[0063] Example 2

[0064] 1. Model construction of gradient random porous structure for bone-like trabeculae

[0065] First, a Python program is written using the GhPython Script battery to generate a cylindrical region with a radius of 5mm and a height of 10mm. Based on this cylindrical region, a spherical region with a radius of r=1mm is generated within it. According to the distance of the sphere from the central axis of the cylinder, the entire sphere is divided into i=3 layers from the inside out, allowing seed points to be randomly distributed within the sphere. The number of seed points N projected onto the surface of a single sphere is... , where k=3, thereby controlling the seed point to exhibit a gradient change that increases radially from the inside to the outside.

[0066] Using a Voronoi 3D cell and a Solid Difference cell, primary cells of a Thiessen polygon are generated at the seed point. The primary cells are then exploded using a Deconstruct Brep cell to extract their vertices, volume center, and face center. A Python program is written using a Scale cell combined with a GhPython Script cell to scale the primary cells around their volume center to form secondary cells. The scaling factor K1 is calculated by interpolating the distance from the volume center to the cylinder's axis. The secondary cells are then exploded using a Deconstruct Brep cell to extract their vertices.

[0067] The scaling factor K1 is calculated using the formula shown in equation (1):

[0068] Equation (1);

[0069] Where d is the distance between the body center and the central axis of the cylindrical domain, d mind represents the minimum distance from the body center of the primary cell of each Thiessen polygon to the central axis of the cylindrical domain. max f1 is the maximum distance from the volume center of the primary cell of each Thiessen polygon to the central axis of the cylindrical domain. The scaling factor f1 is the volume scaling factor of the primary cell of the Thiessen polygon farthest from the central axis of the cylindrical domain, and f2 is the volume scaling factor of the primary cell of the Thiessen polygon closest to the central axis of the cylindrical domain.

[0070] The scaling factors are set to f1=0.9 and f2=0.5. By adjusting the scaling factors f1 and f2, the radial gradient change of the porosity of the support can be achieved.

[0071] A Python program was written using Scale cells and GhPython Script cells to scale the primary faces of the primary cells around their center, with a fixed scaling factor K2 of 0.8, and extract the vertices of the secondary faces. All vertex sets were collected, and Construct Mesh cells were used to construct quadrilateral meshes from adjacent vertices. Weaverbird's Catmull-ClarkSubdivision cells were used to achieve smooth connections between the quadrilaterals, generating a smooth surface STL model. After exporting, 3-magic was used to repair the unclosed surfaces. Then, Boolean operations were performed between this STL model and a cylindrical model with a radius 0.2 mm larger, resulting in a random porous structure. Finally, the response surface methodology was used to fit the relationship between k, f1, f2, and the porosity of the support structure, achieving precise control of the porosity P.

[0072]

[0073] The final result is a radially gradient random porous scaffold with a radius of 5 mm, a height of 10 mm, and a porosity of 75%, exhibiting increasing porosity from the inside out.

[0074] 2. Fabrication of a gradient random porous white phosphogypsum piezoelectric bone scaffold with simulated bone trabeculae.

[0075] 0.34 mol of calcium hydroxide (Ca(OH)2) and 0.15 mol of magnesium hydroxide (Mg(OH)) were added to 300 ml of deionized water and heated at 100 °C for 30 min with stirring. Phosphoric acid was then slowly added dropwise until the pH reached 3.1. The mixture was stirred and refluxed at 100 °C for 10 h and aged at room temperature for 14 h. After filtration through filter paper, the mixture was dried in a drying oven for 24 h to obtain WH powder. The dried powder was then mixed with anhydrous ethanol as a medium and ball-milled at 400 r / min for 8 h with a mass ratio of 12 mm zirconium beads: 7 mm zirconium beads: powder of 3:2:1. The resulting mixture was then placed in a ball mill and milled for 8 h with a mass ratio of 400 r / min. The mixture was then placed in a drying oven at 80 °C for 24 h and ground and sieved to obtain WH nanoparticles.

[0076] Take 100g WH nanoparticles, 15.218g TMPTA, 23.058g PEG200DA, 10.145g PEG200, 0.775g Rad2500, 0.775g 8810X, 1.550g 2022, 3.410g KMT-3331, 0.2421g 184, and 0.0484g UV-531. Mix the above raw materials directly at 1000 r / min for 15 minutes to obtain a ceramic slurry with a solid content of 32.5 vol%. Use a laser exposure energy of 0.358 J / cm². 2 The resulting ceramic preform had a cured thickness of 100 μm and an overcured width of <35 μm. The preform was then degreased by heating to 350 °C at 1 °C / min and holding for 120 min, followed by heating to 500 °C at 0.1 °C / min and holding for 120 min, then heating to 600 °C at 0.2 °C / min and holding for 120 min, and finally cooling to 100 °C at -1 °C / min. The degreasing stage was then completed. After degreasing, the sample was sintered at 600 °C for 2 h in a box furnace for decarburization, and then sintered at 850 °C for 20 min in a microwave sintering furnace at a rate of 10 °C / min. This yielded a high-precision white phosphogypsum piezoelectric ceramic scaffold with a porosity of 75%, a compressive strength of 1.24 MPa, and an elastic modulus of 248 MPa, meeting the mechanical requirements of human cancellous bone. It also exhibits piezoelectric response capabilities, simulating the electrophysiological microenvironment of human bone and promoting bone repair.

[0077] Example 3

[0078] 1. Model construction of gradient random porous structure for bone-like trabeculae

[0079] First, a Python program is written using the GhPython Script battery to generate a cylindrical region with a radius of 5mm and a height of 10mm. Based on this cylindrical region, a spherical region with a radius of r=1mm is generated within it. According to the distance of the sphere from the central axis of the cylinder, the entire sphere is divided into i=3 layers from the inside out, allowing seed points to be randomly distributed within the sphere. The number of seed points N projected onto the surface of a single sphere is... , where k=2, thereby controlling the seed point to exhibit a gradient change that increases radially from the inside to the outside.

[0080] Using a Voronoi 3D cell and a Solid Difference cell, primary cells of a Thiessen polygon are generated at the seed point. The primary cells are then exploded using a Deconstruct Brep cell to extract their vertices, volume center, and face center. A Python program is written using a Scale cell combined with a GhPython Script cell to scale the primary cells around their volume center to form secondary cells. The scaling factor K1 is calculated by interpolating the distance from the volume center to the cylinder's axis. The secondary cells are then exploded using a Deconstruct Brep cell to extract their vertices.

[0081] The scaling factor K1 is calculated using the formula shown in equation (1):

[0082] Equation (1);

[0083] Where d is the distance between the body center and the central axis of the cylindrical domain, d min d represents the minimum distance from the body center of the primary cell of each Thiessen polygon to the central axis of the cylindrical domain. max f1 is the maximum distance from the volume center of the primary cell of each Thiessen polygon to the central axis of the cylindrical domain. The scaling factor f1 is the volume scaling factor of the primary cell of the Thiessen polygon farthest from the central axis of the cylindrical domain, and f2 is the volume scaling factor of the primary cell of the Thiessen polygon closest to the central axis of the cylindrical domain.

[0084] The scaling factors f1 and f2 are set to 0.788 and 0.205, respectively. By adjusting the scaling factors f1 and f2, the radial gradient change of the porosity of the support can be achieved.

[0085] A Python program was written using Scale cells and GhPython Script cells to scale the primary faces of primary cells around their center, with a fixed scaling factor of 0.8, and extract the vertices of secondary faces. All vertex sets were collected, and Construct Mesh cells were used to construct quadrilateral meshes from adjacent vertices. Weaverbird's Catmull-ClarkSubdivision cells were used to achieve smooth connections between quadrilaterals, generating a smooth surface STL model. After exporting, 3-magic was used to repair unclosed surfaces. Then, Boolean operations were performed on this STL model with a cylindrical model with a radius 0.2 mm larger, resulting in a random porous structure. Finally, the response surface methodology was used to fit the relationship between k, f1, f2, and the porosity of the support structure, achieving precise control of the porosity P.

[0086]

[0087] The final result is a radially gradient random porous scaffold with a radius of 5 mm, a height of 10 mm, and a porosity of 65%, exhibiting increasing porosity from the inside out.

[0088] 2. Fabrication of a gradient random porous white phosphogypsum piezoelectric bone scaffold with simulated bone trabeculae

[0089] 0.34 mol of calcium hydroxide (Ca(OH)2) and 0.15 mol of magnesium hydroxide (Mg(OH)) were added to 300 ml of deionized water and heated at 100 °C for 30 min with stirring. Phosphoric acid was then slowly added dropwise until the pH reached 3.2. The mixture was stirred and refluxed at 100 °C for 10 h and aged at room temperature for 14 h. After filtration through filter paper, the mixture was dried in a drying oven for 24 h to obtain WH powder. The dried powder was then mixed with anhydrous ethanol as a medium and ball-milled at 350 r / min for 12 h with a mass ratio of 12 mm zirconium beads: 7 mm zirconium beads: powder of 2:1:1. The resulting mixture was then placed in a ball mill and milled for 24 h with a mass ratio of 350 r / min. The resulting mixture was then placed in a drying oven at 80 °C for 24 h and ground and sieved to obtain WH nanoparticles.

[0090] Take 100g of WH nanoparticles, 17.179g of TMPTA, 26.029g of PEG200DA, 11.453g of PEG200, 0.840g of Rad2500, 0.840g of 8810X, 1.679g of 2022, 3.694g of KMT-3331, 0.2733g of 184, and 0.0547g of UV-531. Mix the above raw materials directly at 1400 r / min for 7 minutes to obtain a ceramic slurry with a solid content of 30 vol%. Use a laser exposure energy of 0.417 J / cm². 2 The resulting ceramic preform had a cured thickness of 130 μm and an overcured width of <52 μm. The preform was then degreased by heating to 350 °C at 1 °C / min and holding for 120 min, followed by heating to 500 °C at 0.1 °C / min and holding for 120 min, then heating to 600 °C at 0.2 °C / min and holding for 120 min, and finally cooling to 100 °C at -1 °C / min. The degreasing stage was then completed. After degreasing, the sample was sintered at 600 °C for 2 h in a box furnace for decarburization, and then sintered at 650 °C for 15 min in a microwave sintering furnace at a rate of 30 °C / min. This yielded a high-precision white phosphogypsum piezoelectric ceramic scaffold with a porosity of 60%, a compressive strength of 3.66 MPa, and an elastic modulus of 305 GPa, meeting the mechanical requirements of human cancellous bone. It also exhibits piezoelectric response capabilities, simulating the electrophysiological microenvironment of human bone and promoting bone repair.

[0091] Comparative Example 1

[0092] Other conditions are the same as in Example 1, except that the slurry formulation in Content 2 is different. 100g of white phosphogypsum powder, 24.412g of TMPTA, 15.26g of HDDA, 15.257g of HEA, 6.103g of PEG200, 0.5g of Rad2500, 0.5g of 8810X, 0.5g of Tenda foam N, 8g of 41000, and 0.2421g of 184 are used to obtain the slurry. The final printed and cured thickness is greater than 240μm, and the over-cured width is too large, making it impossible to form a precision model with a pore size of 300-500μm, and thus unable to complete the high-precision processing of random porous structures.

[0093] Comparative Example 2

[0094] Other conditions were the same as in Example 1, except for the slurry formulation in Content 2. 100g of white phosphogypsum powder, 15.218g of TMPTA, 23.058g of PEG200DA, 10.145g of PEG200, 0.775g of Rad2500, 0.775g of 8810X, 1.550g of 2022, 3.410g of 41000, 0.2421g of 184, and 0.0484g of UV-531 were used. The resulting slurry with a solid content of 35 vol% had excessively high viscosity, making it unable to flow and severely affecting the processing. If the viscosity were reduced by lowering the solid content, the mechanical and electrical properties of the scaffold would be affected.

[0095] Comparative Example 3

[0096] Other conditions are the same as in Example 2, except that the model construction part of Content 1 is different. A Python program is written using Scale battery combined with GhPython Script battery to scale the primary face of the primary cell with the face center as the center. The scaling factor K2 is calculated by interpolation based on the distance from the body center to the cylinder axis to extract the vertices of the secondary face. The formula for calculating the scaling factor K2 is shown in Equation (1):

[0097] Equation (1);

[0098] Where d is the distance between the body center and the central axis of the cylindrical domain, d min d represents the minimum distance from the body center of the primary cell of each Thiessen polygon to the central axis of the cylindrical domain. max The scaling factor f3 represents the maximum distance from the body center of each primary cell of the Thiessen polygon to the central axis of the cylindrical region. The scaling factor f4 is the scaling factor for the primary cell of the Thiessen polygon furthest from the central axis of the cylindrical region, and f5 is the scaling factor for the primary cell of the Thiessen polygon closest to the central axis of the cylindrical region. Because face-center scaling results in a 4-6 times greater change than body-center scaling, the positions of secondary face vertices are uncontrollable, leading to program errors and the inability to obtain results.

Claims

1. A method for constructing a model of a gradient random porous structure mimicking bone trabeculae, characterized in that: Based on a cylindrical domain, several spherical domains with radius r are formed within it. The spherical domains are divided into i layers from the inside out based on their distance from the central axis of the cylindrical domain. A radial layering seeding method is used to plant N seed points on the surface of a single spherical domain, generating primary cells of Thiessen polygons at these seed points. These primary cells are then exploded, and their vertices, body centers, and face centers are extracted. A radially dependent body center scaling method is used to scale the primary cells around their body centers to form secondary cells. The scaling factor K1 is calculated by interpolating the distance from the body center to the axis of the cylinder. The secondary cells are then exploded, and their vertices are extracted. The primary faces of the primary cells are then scaled around their face centers to extract the vertices of the secondary faces. All vertex sets are collected, and adjacent vertices are used to construct a quadrilateral mesh, creating smooth connections between the quadrilaterals to generate a smooth STL model. This STL model is exported and subjected to Boolean operations with a cylindrical model having a larger radius than the STL model to obtain a random porous structure. Finally, by fitting a regression model, a model of the gradient random porous structure of the bone-like trabeculae is obtained.

2. The method for constructing a model of a gradient random porous structure for trabecular bone according to claim 1, characterized in that: Use the GhPython Script battery to write a Python program to generate cylindrical fields; N=ki+1 or N=k i , where k is the seed point coefficient, and the value of k is a natural number between 1 and 5; At the seed point, the primary cells of the Thiessen polygon are generated using the Voronoi 3D cell and the Solid Difference cell. The primary cells are then exploded using the Deconstruct Brep cell to extract their vertices, volume centers, and face centers.

3. The method for constructing a model of a gradient random porous structure for trabecular bone according to claim 2, characterized in that: A Python program was written using Scale batteries and GhPython Script batteries to scale primary cells around the body center to form secondary cells. The scaling factor K1 is calculated using the formula shown in equation (1): Equation (1); Where d is the distance between the body center and the central axis of the cylindrical domain, d min Let d be the minimum distance from the body center of the primary cell of each Thiessen polygon to the central axis of the cylindrical domain. max Let f1 be the maximum distance from the volume center of each primary cell of the Thiessen polygon to the central axis of the cylindrical region. The scaling factor f1 is the volume scaling factor of the primary cell of the Thiessen polygon furthest from the central axis of the cylindrical region, where 0.5 ≤ f1 < 1; f2 is the volume scaling factor of the primary cell of the Thiessen polygon closest to the central axis of the cylindrical region, where 0.5 ≤ f1 < 1. <f2≤0.5; Write a Python program using Scale cells and GhPython Script cells to scale the primary faces of the primary cell around the face center, with a scaling factor K2 of 0.6-0.8, and extract the vertices of the secondary faces.

4. The method for constructing a gradient random porous structure for trabecular bone according to claim 3, characterized in that: Construct Mesh cells are used to build quadrilateral meshes from adjacent vertices, and Weaverbird's Catmull-Clark Subdivision cells are used to achieve smooth connections between quadrilaterals, generating an STL model with a smooth surface. After exporting the STL model, use 3-magic to repair the non-closed surface, and then perform Boolean operations on the STL model and the cylindrical model with a radius larger than 0.1-0.4 mm to obtain a random porous structure. The process of fitting the regression model is as follows: use the response surface methodology to fit the relationship between k, f1, f2 and the porosity of the gradient random porous structure model of the simulated bone trabeculae, and draw a three-dimensional graph to obtain the model.

5. A method for preparing a white phosphogypsum piezoelectric bone scaffold with a gradient random porous structure mimicking bone trabeculae, characterized in that: White phosphogypsum nanopowder, solidification raw materials, and additives are mixed according to the design ratio to obtain a ceramic slurry. The ceramic slurry is added to a photopolymerization printing device, and a model of a gradient random porous structure of a bone trabeculae constructed according to any one of the model construction methods in claims 1-4 is photopolymerized to obtain a ceramic blank. The ceramic blank is degreased and sintered to obtain a white phosphogypsum piezoelectric bone scaffold.

6. The method for preparing a gradient random porous white phosphogypsum piezoelectric bone scaffold with a simulated bone trabeculae structure as described in claim 5, characterized in that: The process for obtaining the white phosphogypsum nanopowder is as follows: zircon beads with a diameter of 10-15 mm, zircon beads with a diameter of 5-8 mm, and white phosphogypsum powder are placed in a ball mill jar at a mass ratio of 1-3:1-2:1 and wet ball milled. During the wet ball milling process, anhydrous ethanol is used as the medium, the ball milling speed is 300-400 r / min, and the time is 8-12 h. After the wet ball milling is completed, the powder is dried, ground through a 60-100 mesh sieve, and the material passing through the sieve is obtained. The white phosphogypsum powder is obtained by chemical synthesis. The drying temperature is 70-100℃, and the drying time is 20-24 hours.

7. The method for preparing a gradient random porous white phosphogypsum piezoelectric bone scaffold with a simulated bone trabeculae structure as described in claim 5, characterized in that: The curing raw material comprises polyethylene glycol (200) diacrylate, trimethylolpropane triacrylate resin, and polyethylene glycol; by volume ratio, polyethylene glycol (200) diacrylate: trimethylolpropane triacrylate resin: polyethylene glycol = 45~55: 20~40: 5~35; The additives include dispersants, antisettling agents, leveling agents, defoamers, photoinitiators, and light absorbers, in a mass ratio of dispersant: antisettling agent: leveling agent: defoamer: photoinitiator: light absorber = 1~5: 0.1~2: 0.1-2: 1~4: 0.1~3: 0.01~1; The dispersant is KMT-3331, the anti-settling agent is Sago-8810X, the leveling agent is Rad2500, the defoamer is SRE-2022A, the photoinitiator is photoinitiator 184, and the light absorber is UV-531; The ceramic slurry, by volume ratio, comprises: 30-45 vol% ceramic powder, 50-65 vol% solidifying raw material, and 5-8 vol% additives. The mixing speed is 1000-1800 r / min, and the mixing time is 7-15 min.

8. The method for preparing a gradient random porous white phosphogypsum piezoelectric bone scaffold with a bone-like trabecular structure as described in claim 5, characterized in that: The parameters for the photopolymerization printing are: laser exposure energy 0.3~0.9 J / cm². 2 The slice thickness is 40-60μm.

9. The method for preparing a gradient random porous white phosphogypsum piezoelectric bone scaffold with a simulated bone trabeculae structure as described in claim 5, characterized in that: The sintering process is as follows: the degreased billet is first placed in a high-temperature furnace and kept at 500-700℃ for 1.5-2.5h, and then placed in a microwave sintering furnace and heated to 650-1100℃ at a heating rate of 20-40℃ / min for sintering for 10-20min.

10. A white phosphogypsum piezoelectric bone scaffold prepared by any one of claims 1-9.

Images