A METHOD FOR MANUFACTURING AT LEAST ONE POROUS MONOLITHIC INORGANIC SUPPORT, A METHOD FOR PREPARING A TANGENTIAL FILTRATION MEMBRANE COMPRISING THE MANUFACTURE OF SAID SUPPORT, A POROUS MONOLITHIC INORGANIC SUPPORT MANUFACTURED BY SAID METHOD, AND A TANGENTIAL FILTRATION MEMBRANE COMPRISING SAID SUPPORT

Abstract

A method for producing a porous monolithic inorganic support (1) using a 3D printing machine comprising at least one extrusion head (6) movably mounted in space.The method comprises: Controlling the extrusion head according to a digital path such that: by means of an overlapping path, the material being deposited partially overlaps with at least one edge of a previously deposited material deposit, by means of a portion of overlapping material presented in its thickness and having a thickness strictly less than the nominal height (e) so as to avoid gaps between the rounded edges of the current material deposit and those of the previously deposited material deposit; by means of a crossing path, the material being deposited crosses at least one previously deposited material deposit with a complete overlap of said previously deposited material deposit, by means of a portion of overlapping material presented in its thickness and having a thickness strictly less than the nominal height (e).

Description

METHOD FOR THE PREPARATION OF AN INORGANIC FILTRATION SUPPORT BY INTERLACE AND MEMBRANE OBTAINED THROUGH THE SAME Technical mastery [1] The present invention relates to a method for preparing a porous monolithic inorganic support, which can be used in particular to obtain a filtration membrane, and in particular a tangential filtration membrane. More specifically, the porous support is prepared by a technique carried out by adding material. Prior art [2] The filtration membrane constitutes a selective barrier and allows, under the action of a transfer force, the passage or detention of certain components of the liquid medium to be treated. The passage or detention of the components may be the result of their size compared to the pore size of the membrane, which then functions as a filter. Depending on the pore size, these techniques are called microfiltration, ultrafiltration, or nanofiltration. [3] A membrane consists of a porous support on which one or more separation layers are deposited. Typically, the support is first molded by extrusion. The support is then subjected to sintering in order to obtain the required strength, while maintaining an open and interconnected porous texture. This method makes it necessary to obtain rectilinear channels within which the separation layer(s) are then deposited and sintered. The membrane produced in this way is therefore subjected to a minimum of two sintering operations. 2102481 of 47 sintering. The organic binders added during paste preparation, before extrusion, are completely burned during the sintering of the support. [4] In application FR 3 006 606, the Applicant has described the production of a filtration membrane whose porous support is produced by an additive technique, by the repetitive deposition of a continuous bed of powder followed by localized consolidation according to a predetermined pattern. This technique makes it possible to produce mechanically resistant filtration membranes suitable for use in tangential filtration. However, this technique has the disadvantage that it is necessary to adjust the fluidity of the powder to allow it to flow perfectly when the powder bed is deposition. Furthermore, this technique makes it necessary to remove the unconsolidated powder, eventually also for its recycling, which can be difficult, time-consuming and expensive, especially when said unconsolidated powder is present within the non-rectilinear channels of the porous support. [5] The Applicant has also proposed, through patent applications W02020 / 109715 and W02020 / 109716, new methods for preparing a porous support which do not have the drawbacks of the prior art, and in particular which are fast, easy to carry out, which allow obtaining a mechanically resistant porous support and whose shape, and in particular that of the non-rectilinear channels, can be easily varied. The porous support obtained is homogeneous, mechanically resistant and has a porosity suitable for use in filtration, i.e. a porosity between 10 and 60% and which is 2102481 of 47 open and interconnected with an average pore diameter in the range between 0.5 pm and 50 µm. [6] For this purpose, the methods use a 3D printing machine comprising an extrusion head mounted movably in space relative to and above a fixed horizontal plate. An inorganic composition exits the extrusion head in the form of a ribbon or bead of material that allows, from a digital 3D model, a manipulable raw three-dimensional structure intended to form the porous monolithic inorganic support(s). The manipulable raw three-dimensional structure is then subjected to a sintering step. [7] As explained in these patent applications, the raw manipulable three-dimensional structure is obtained from the superposition of strata, each corresponding to a set of continuous or discontinuous, juxtaposed or non-juxtaposed strands, which are extruded at the same altitude according to the digital 3D model. The different strata can be stacked along the vertical axis in different ways. [8] In the state of the art, a method for producing a 3D object by means of modelling is also known from patent application WO 2020 / 109716, by depositing a molten wire in which a superposition path is followed. Similarly, patent application US 2017 / 165917 describes a method for producing a 3D object by means of modelling, implementing two nozzles for a controlled deposit of material to create in particular a crossing between the two strands of material. 2102481 of 47 [9] The Applicant has therefore considered producing the strata by vertically stacking contiguous deposits of strands of material c as illustrated in Figure 1A. The Applicant observed that after the sintering operation (Figure 1B), said structure contained voids of material v which could have a size larger than the pores, and which could reduce the mechanical strength of the porous support.

[10] The Applicant has also considered producing the strata by depositing strands of material c arranged in a staggered manner as illustrated in Figure 1C. This solution makes no contribution to the elimination of the residual hollow spaces v between the strands of material. After the sintering operation (Figure 1D), the structure still has internal voids of material v. Furthermore, this solution deteriorates the quality of the lateral surfaces because every second stratum has a gap which leads to the deposit of material of the upper stratum sinking. Disclosure of the invention

[11] The object of the invention is precisely to remedy the drawbacks of the state of the art, providing a new method for the production of a porous monolithic inorganic support, designed to avoid the creation of hollow material spaces that can produce a reduction in the mechanical resistance of the porous support.

[12] Another object of the invention is to provide a method for the production of a three-dimensional structure, designed to be able to control the dimensions of this three-dimensional structure and in particular the profile of the walls of this three-dimensional structure. 2102481 of 47

[13] Another object of the invention is to provide a method for the production of a three-dimensional structure provided with at least one circulation channel for a fluid medium to be treated, which has a wall adapted for the deposit of separation layers.

[14] Another object of the invention is to provide a method for the production of a three-dimensional structure provided with at least one circulation channel for a fluid medium to be treated, which has a wall that presents an uninterrupted succession of rounded reliefs that generate variations in the passage section of the channel, thus avoiding insurmountable pressure drops.

[15] To achieve these objectives, the invention relates to a method for the production of at least one porous monolithic inorganic support having at least one channel for the circulation of fluid to be treated and having a porosity comprised between 10% and 60% and an average pore diameter that is included within the range between 0.5 pm and 50 pm, by using a 3D printing machine comprising at least one extrusion head mounted in a movable manner in the space above a fixed horizontal plate, which moves successively with a nominal height according to a predefined digital trajectory to carry out the deposits of material in an overlapping manner, where each one has a nominal width and a thickness defined between a lower surface and an upper surface, wherein said 3D printing machine allows the deposit of material in the form 2102481 of 47 - a cord with rounded edges so as to create in the wall of at least one circulation channel, rounded perimeter reliefs that contribute to the generation of turbulence, - of a composition to build, on said horizontal plate, from the predefined digital path, walls of a manipulable raw three-dimensional structure intended to form the porous monolithic inorganic support(s), where the method consists of: - Using a wall whose width is greater than the nominal width of the material tank, break down the digital path exclusively into overlapping paths and crossing paths; - Feed the extrusion head of the 3D printing machine with a composition, - Control the extrusion head according to the digital path so that: * by means of an overlapping path, the material being deposited partially overlaps at least one edge of a previously deposited material deposit, by a part of the overlapping material presented in its thickness and having a thickness strictly less than the nominal height so as to avoid, in the internal part of the manipulable raw three-dimensional structure, the spaces between the rounded edges of the current material deposit and those of the previously deposited material deposit, * by means of a crossing path, the material being deposited crosses at least one previously deposited material deposit with a complete overlap of said previously deposited material 2102481 of 47 deposited, by means of a portion of superimposed material presented in its thickness and having a thickness strictly less than the nominal height so as to avoid in the internal part of the manipulable raw three-dimensional structure, the spaces between the rounded edges of the deposit of current material and those of the deposit of previously deposited material.

[16] According to a characteristic advantageous embodiment, the extrusion head is mounted in a movable manner according to the predefined digital trajectory to carry out a succession of turns where each one corresponds to the path made to find the same position in the horizontal plane with an elevation corresponding to a nominal height.

[17] According to another advantageous embodiment, the extrusion head is controlled such that, in at least one rotation, the extrusion head is raised at least once by an intermediate height which is a fraction of the nominal height.

[18] According to another characteristic of the invention, the extrusion head is controlled such that, in at least one rotation, the extrusion head is positioned at different points of the path, rising at each point by an increment in height whose sum corresponds to the nominal height.

[19] Advantageously, the extrusion head is controlled such that, by means of a succession of rotations, the extrusion head is raised to build the manipulable raw three-dimensional structure according to the digital 3D model, following a path with a deposit of material. 2102481 of 47 uninterrupted from the beginning to the end of the construction of the manipulable raw three-dimensional structure.

[20] According to an advantageous feature, the extrusion head is controlled such that, during each rotation, the extrusion head moves in the horizontal plane such that the material being deposited partially overlaps an edge of a previously deposited deposit of material.

[21] For a crossover path, the extrusion head is controlled to reduce the amount of deposited material to produce the overlapped material portion having a thickness strictly less than the nominal height.

[22] For a path with the same overlap, the extrusion head is controlled to keep the flow rate of deposited material constant.

[23] For an overlap path, the extrusion head is controlled to adjust its position in the plane to define the degree of overlap of the overlapped material portion over the previously deposited material deposit.

[24] For an overlapping stroke, the extrusion head flow rate is adjusted according to the degree of overlap of the overlapping portion of material over the previously deposited material deposit.

[25] Typically, the extrusion head is mounted movably according to the predefined digital path to carry out the material deposits in a superimposed manner by means of strata each rising to a nominal height. 2102481 of 47

[26] According to another characteristic of the invention, the extrusion head is controlled to provide in the manipulable raw three-dimensional structure, at least one circulation channel for a fluid medium to be treated, which has a wall presenting a succession of rounded reliefs that generate variations in the passage section of the channel, where the rounded reliefs are formed by the part of material located on the opposite side of the part of superimposed material.

[27] According to another feature, the extrusion parameters are adjusted and the extrusion head is configured such that the material portion located on the opposite side of the overlapping material portion has a rounded edge.

[28] Usually, the manipulable raw three-dimensional structure is placed in a heat treatment furnace in order to perform a sintering operation on it.

[29] Another object of the invention is to provide a method for preparing a tangential filtration membrane comprising the preparation of a porous monolithic inorganic support in which at least one circulation channel of the fluid medium to be treated is formed, followed, after sintering of said support, by a step of creating at least one separation layer on the walls of one or more of the channels.

[30] Another object of the invention is to provide a porous monolithic inorganic support made according to the method according to the invention and having an outer surface presenting a succession of rounded perimeter reliefs and a circulation channel whose wall presents 2102481 of 47 rounded perimeter reliefs that contribute to the generation of turbulence.

[31] Another object of the invention is to provide a tangential filtration membrane comprising a porous monolithic inorganic support provided with at least one circulation channel for the fluid medium to be treated, the wall of which with rounded perimeter reliefs is covered by at least one separation layer. Brief description of the figures

[32] [Figure 1A] Figure 1A is a cross-sectional view of a wall produced by vertically stacking deposits of strands of material showing void spaces.

[33] [Figure 1B] Figure 1B is a view of the wall shown in Figure 1A showing the persistence of void spaces.

[34] [Figure 10] Figure 10 is a cross-sectional view of a wall produced by vertically stacking in a staggered manner deposits of strands of material showing void spaces.

[35] [Figure 1D] Figure 1D is a view of the wall shown in Figure 10 after the sintering operation and showing the persistence of voids.

[36] [Figure 2A] Figure 2A is a schematic view of a 3D printing machine that allows the implementation of the invention.

[37] [Figure 2B] Figure 2B is a perspective view of an exemplary embodiment of a manipulable three-dimensional raw structure. 2102481 of 47

[38] [Figure 3] Figure 3 is a schematic top view showing the material reservoir in conventional mode to produce a tubular wall of width equal to the nominal width of a material reservoir.

[39] [Figure 3A] Figure 3A is a cross-sectional view taken substantially along lines AA of Figure 3 showing successive deposits of material in conventional mode to produce a tubular wall.

[40] [Figure 4] Figure 4 is a schematic top view showing the material reservoir in vessel mode for producing a tubular wall of width equal to the nominal width of a material reservoir.

[41] [Figure 4A] Figure 4A is a cross-sectional view taken substantially along lines AA of Figure 4 showing successive deposits of material in a cup-like manner to produce a tubular wall.

[42] [Figure 5] Figure 5 is a schematic top view showing the method according to the invention of depositing material by means of an overlapping path of the extrusion head.

[43] [Figure 5A] Figure 5A is a cross-sectional view taken substantially along lines AA of Figure 5 showing the overlapping of material deposits by an overlapping path of the extrusion head.

[44] [Figure 6] Figure 6 is a schematic top view showing material deposits through a path with crossing of the extrusion head. 2102481 of 47

[45] [Figure 6A] Figure 6A is a cross-sectional view taken substantially along lines AA of Figure 6 showing the superposition, in accordance with the invention, of material deposits by a crossing path of the extrusion head.

[46] [Figure 6B] Figure 6B is a cross-sectional view taken substantially along lines BB of Figure 6 showing the superposition in accordance with the invention of material deposits by a crossing path of the extrusion head.

[47] [Figure 7] Figure 7 is a schematic top view showing a step of the sequential vessel mode material deposition method for producing a tubular wall with a width greater than the nominal width of a material reservoir.

[48] ​​[Figure 7A] Figure 7A is a cross-sectional view taken substantially along lines AA of Figure 7 showing a step of the sequential vessel mode material deposition method for producing a tubular wall having a width greater than the nominal width of a material reservoir.

[49] [Figure 7B] Figure 7B is a schematic top view showing the path of the material reservoir in sequential vessel mode to produce a tubular wall with a width greater than the nominal width of a material reservoir.

[50] [Figure 8] Figure 8 is a schematic top view showing another step of the sequential vessel mode material deposition method for producing a tubular wall with a width greater than the nominal width of a material reservoir. 2102481 of 47

[51] [Figure 8A] Figure 8A is a cross-sectional view taken substantially along lines AA of Figure 8 showing another step of the sequential vessel mode material deposition method for producing a tubular wall having a width greater than the nominal width of a material reservoir.

[52] [Figure 8B] Figure 8B is a schematic top view showing the path of the material reservoir in sequential vessel mode to produce a tubular wall with a width greater than the nominal width of a material reservoir.

[53] [Figure 8C] Figure 8C is a view showing another step of the sequential vessel mode material deposition method for producing a tubular wall with a width greater than the nominal width of a material reservoir.

[54] [Figure 9] Figure 9 is a schematic top view showing a step of the continuous vessel mode material deposition method for producing a tubular wall with a width greater than the nominal width of a material reservoir.

[55] [Figure 9A] Figure 9A is a cross-sectional view taken substantially along lines AA of Figure 9 showing a step of the continuous vessel mode material deposition method for producing a tubular wall having a width greater than the nominal width of a material reservoir.

[56] [Figure 9B] Figure 9B is a schematic top view showing the material reservoir path in continuous vessel mode for 2102481 of 47 produce a tubular wall with a width greater than the nominal width of a material tank.

[57] [Figure 10] Figure 10 is a schematic top view showing another step of the continuous vessel mode material deposition method for producing a tubular wall with a width greater than the nominal width of a material reservoir.

[58] [Figure 10A] Figure 10A is a cross-sectional view taken substantially along lines AA of Figure 10 showing another step of the continuous vessel mode material deposition method for producing a tubular wall having a width greater than the nominal width of a material reservoir.

[59] [Figure 10B] Figure 10B is a schematic top view showing the path of the material reservoir in continuous vessel mode to produce a tubular wall with a width greater than the nominal width of a material reservoir.

[60] [Figure 10C] Figure 10C is a view showing another step of the continuous vessel mode material deposition method for producing a tubular wall with a width greater than the nominal width of a material reservoir.

[61] [Figure 11] Figure 11 is a diagram that allows to illustrate an example of an extrusion head path for the method called “sequential cup mode” and for the method called “continuous cup mode”.

[62] [Figure 12A] Figure 12A is a top view showing another example of an embodiment of a wall delimiting two channels 2102481 of 47 contiguous buildings and which has a width greater than twice the nominal width of a material deposit.

[63] [Figure 12B] Figure 12B is a cross-sectional view taken substantially along lines BB of Figure 12A.

[64] [Figure 13A] Figure 13A is a cross-sectional view showing the deposition of material carried out in accordance with a prior art method, by means of a path with a crossing, before crossing the crossing.

[65] [Figure 13B] Figure 13B is a cross-sectional view showing the deposition of material carried out in accordance with a prior art method, by means of a path with a crossing, at the time of crossing the crossing.

[66] [Figure 13C] Figure 130 is a cross-sectional view showing the deposition of material made according to a prior art method, by means of a path with a crossing, after crossing the crossing.

[67] [Figure 13D] Figure 13D is a cross-sectional view showing the deposit of material made according to a prior art method, by means of a path with a crossing, and superimposed on the deposit of material previously produced.

[68] [Figure 13E] Figure 13E is a cross-sectional view showing the deposit of material made according to a method of the prior art, by means of a path with a crossing, superimposed on the deposit of material previously produced and showing the creation of hollow spaces at the level of the crossing of material deposits.

[69] [Figure 14] Figure 14 is a cross-sectional view showing an example of an embodiment according to the invention with 2102481 of 47 material deposits produced in superimposed form by strata according to the principle of the invention.

[70] [Figure 15] Figure 15 is an oblique sectional view of a porous monolithic inorganic support with eight filtration channels showing the absence of void spaces.

[71] [Figure 16] Figure 16 is a cross-sectional view of a tangential filtration membrane comprising a porous monolithic inorganic support with eight channels whose channel wall with rounded perimeter reliefs is covered by a separation layer. Description of the embodiments

[72] The object of the invention relates to the development of a porous monolithic inorganic support 1 (Figure 2B) in order to constitute an element for the separation by tangential flow of a liquid medium to be treated in a filtrate (or permeate) and a retentate, commonly called a tangential filtration membrane.

[73] In numerous applications, said porous supports have a tubular geometry and comprise at least one channel or circulation path for the fluid to be filtered, provided with at least one separation layer. These circulation channels have an inlet and an outlet. In general, the inlet of the circulation channels is located at one end of the porous support, this end acting as an inlet zone for the fluid medium to be treated and its outlet is located at the other end of the porous support fulfilling the role of outlet zone for the retentate. The inlet zone and the outlet zone are connected by a continuous peripheral zone at the level of which the permeate is recovered. 2102481 of 47

[74] In other applications, the porous supports may be presented in the form of a block, for example, of parallelepiped shape, in which at least one channel or circulation path is provided for the fluid to be filtered, provided with at least one separation layer. The permeate is recovered at the periphery of the block or with the aid of a collection circuit provided in the block. In the illustrated examples, the porous support has a tubular geometry, but it is clear that the object of the invention can be applied to porous supports of any shape.

[75] When the porosity (the average diameter of the pores) of the sintered support is adapted to the fluid medium to be treated (filtration threshold) said sintered support can be used directly in filtration and is called an automembrane or homogeneous membrane.

[76] When the porosity of the sintered support is not suitable for the fluid medium to be treated (pores with a dimension too large with respect to the necessary filtration threshold), the walls of the circulation channel(s) are then continuously coated by at least one separation layer which ensures the filtration of the fluid medium to be treated. The separation layer(s) are porous and have an average pore diameter smaller than that of the support. The separation layer can be deposited directly on the porous support (in the case of a single-layer separation layer), or even on an intermediate layer with a smaller average pore diameter, itself deposited directly on the porous support (in the case of a multi-layer separation layer). In this way, a part of the fluid medium to be filtered passes through the separation layer(s) and the porous support, such that said treated part of the fluid, called permeate, flows through the 2102481 of 47 outer periphery of the porous support. The separation layers delimit the surface of the filtration membrane that is intended to be in contact with the fluid to be treated and in contact with the surface through which the fluid to be treated circulates.

[77] The porosity of the monolithic inorganic support 1 is open, i.e. it forms a network of interconnected pores in three dimensions, which allows the fluid filtered by the separation layer(s) to pass through the porous support and be recovered at the periphery. Therefore, the permeate is recovered at the peripheral surface of the porous support.

[78] The porous monolithic inorganic support 1 has an average pore diameter which is included within the range between 0.5 µm and 50 µm. The porosity of the porous monolithic inorganic support 1 is comprised between 10 and 60%, preferably between 20 and 50%.

[79] The term mean pore diameter designates the d50 value of a volume distribution for which 50% of the total pore volume corresponds to the volume of pores with a diameter less than this d50. The volume distribution is the curve (analytical function) representing the frequencies of the pore volumes as a function of their diameter. The d50 corresponds to the median that separates the area under the frequency curve obtained by mercury penetration into two equal parts. In particular, the technique described under ISO 15901-1:2005 as it relates to the mercury penetration measurement technique can be used.

[80] The porosity of the support, which corresponds to the total volume of the interconnected voids (pores) present in the material considered, is a physical quantity between 0 and 1 or between 0% and 100%. This 2102481 of 47 determines the flow and retention capacities of said porous body. For the material to be used in filtration, the total interconnected open porosity must be at least 10% for adequate filtrate flow through the support, and at most 60% to ensure adequate mechanical strength of the porous support.

[81] The porosity of a porous body can be measured by determining the volume of a liquid contained in said porous body by weighing said material before and after a prolonged stay in said liquid (water or other solvent). Knowing the respective volumetric masses of the material considered and the liquid used, the difference in mass, converted to volume, is directly representative of the pore volume and therefore of the total open porosity of the porous body.

[82] There are other techniques that allow the specific measurement of the total open porosity of a porous body, among which we can mention: - mercury intrusion porosimetry (previously cited ISO 15901-1 standard): when injected under pressure, mercury fills the pores accessible for the pressures used, and the volume of mercury injected then corresponds to the pore volume, - Small-angle scattering: This technique, which uses neutron or X-ray radiation, provides physical quantities averaged over the entire sample. The measurement consists of analyzing the angular distribution of the intensity scattered by the sample, - the analysis of 2D images obtained by microscopy, - the analysis of 3D images obtained by X-ray tomography. 2102481 of 47

[83] Furthermore, the porous monolithic inorganic support 1 has a mechanical strength suitable for use in tangential filtration. More specifically, the porous monolithic inorganic support 1 withstands an internal pressure of at least 10 bars without bursting, and preferably at least 30 bars without bursting and advantageously at least 50 bars without bursting. The bursting pressure corresponds, according to the invention, to the pressure at which a support, the porosity of which has been previously plugged (with a hot-melt material such as, for example, paraffin) bursts under the effect of an internal overpressure compared to the external pressure on the support, where this overpressure is applied in the channels with water and the pressure external to the support is atmospheric pressure.

[84] As can be seen more specifically in Figure 2A, the porous monolithic inorganic support 1 according to the invention is produced by sintering a manipulable raw three-dimensional structure 2, which is built up according to a digital 3D model M by superimposing deposits of material 3 falling within a composition 4 in a general sense. The deposits of material are produced by using a three-dimensional printing machine comprising in particular a horizontal plate 5, optionally stationary, above which is arranged at least one extrusion head 6.

[85] The term “raw three-dimensional structure” 2 designates a three-dimensional structure obtained from the superposition of deposits of a composition 4 and which has not yet been subjected to sintering. The shape and dimensions of this raw structure are determined level by level by the digital 3D model M, as will be explained later. 2102481 of 47 detailed in the following description. The digital 3D model M is determined by a computer design program to construct the raw three-dimensional structure 2.

[86] This raw three-dimensional structure 2 is described as “manipulable” because it does not deform under its own weight, and can also present slopes, thanks to an accelerated consolidation that gives it a stable mechanical rigidity over time, as will be explained later. This raw three-dimensional structure 2 can thus be separated from the horizontal plate 5 to move without deformation or breakage, in particular to subsequently undergo a heat treatment operation necessary to obtain a monolithic porous support according to the invention.

[87] The extrusion head 6 of the three-dimensional printing machine is supported by a displacement mechanism (not shown in the figures), such as a robot, which allows its displacement according to at least three axes (x, y, yz). In this way, the extrusion head 6 can be displaced according to a horizontal plane (x and y axes) and a vertical plane (z axis), thanks to the displacement mechanism which is controlled by a computer R of any known type. This computer R controls the movements of the displacement system and consequently of the extrusion head 6, according to a predetermined trajectory based on the digital 3D model M from which the raw three-dimensional structure 2 was produced that allows obtaining the porous monolithic inorganic support 1 after a heat treatment operation. 2102481 of 47

[88] The extrusion head 6 comprises an inlet for the composition 4 (not shown in the figures). As shown in the figures, the extrusion head 6 also comprises an extrusion nozzle comprising a calibrated flow orifice 8 from which the composition 4 exits. The extrusion head 6 and consequently also the extrusion nozzle with its flow orifice 8 which is integral with the extrusion head 6, can be moved according to said digital 3D model M. According to the method of the invention, the composition 4 is introduced into the extrusion head 6 of the machine through an inlet for feeding the flow orifice 8. A mechanical action can be applied to introduce the composition 4 into the head 6 through this inlet.

[89] In the context of the invention, the term “mechanical action” refers to the application of pressure by any known technique, such as a piston, a pump or an extrusion screw. This step can be carried out in a conventional manner by a person skilled in the art and will not be detailed here. The material exits the flow orifice 8 thanks to a pressure somewhat higher than atmospheric pressure. This is a force exerted vertically on the material from top to bottom at the time of extrusion and which contributes to the crushing of the material between the extrusion nozzle and the underlying hardened material.

[90] The flow orifice 8 is placed in front of and close to the horizontal plate 5. The flow orifice 8 can be moved, vertically (i.e. along the z-axis) and horizontally (i.e. along the x and y axes), with respect to the horizontal plate 5 which is fixed. The vertical and / or horizontal displacement of the flow orifice 8 with respect to the horizontal plate 5 can be achieved by means of a horizontal orifice 8 which is fixed to the horizontal plate 5. 2102481 of 47 fixed horizontal plate 5 allows the construction according to the digital 3D model M of the manipulable raw three-dimensional structure 2 resting on the horizontal plate 5 after extrusion of a material bead through the flow orifice 8.

[91] According to the embodiment illustrated in the figures, the extrusion head 6 is provided with a flow orifice 8 of circular section. When the flow orifice 8 has a circular section, its diameter is advantageously between 0.1 mm and 10 mm, preferably between 0.1 mm and 1 mm and preferably between 0.2 and 0.8 mm.

[92] It is recalled that for a predefined extrusion flow rate E (determined for example by the rotation speed of the extrusion screw), a predefined travel speed F of the extrusion head 6 and a nominal travel height e of the extrusion head 6, there corresponds to a material reservoir 3 with a nominal width L such that L=f(E, F, e). The material reservoir 3 has a cross section with a nominal height e defined between a substantially flat upper surface 3s and a substantially flat lower surface 3¡ connected to each other on both sides by two rounded edges 3b (FIGS. 3 and 3A). The nominal width L of the material reservoir 3 can therefore be adjusted or modified as a function of the travel speed F and / or the extrusion flow rate of the extrusion head 6.

[93] The extrusion head 6 is fed by a composition 4 which is presented for example in the form of a paste as described for example in document PCT / FR2019 / 052807 or in the form of a filament or granules as described in document PCT / FR2019 / 052808. 2102481 of 47 advantageously, composition 4 is an inorganic composition typically of a ceramic and / or metallic nature.

[94] The ceramic composition is composed of a solid inorganic powder phase and a matrix.

[95] The solid inorganic powder phase of the ceramic composition comprises one or more solid inorganic materials, each in the form of particles with an average diameter between 0.1 pm and 150 pm.

[96] The concept of mean diameter is associated with that of particle distribution. In fact, powder particles rarely have the same size or are monodisperse, and a powder is therefore often characterized by a distribution of particle sizes. The mean diameter therefore corresponds to the average of the particle size distribution. The distribution can be represented in different ways, such as a frequency or cumulative distribution. Certain measurement techniques directly provide a distribution based on number (microscopy) or mass (sieving). The mean diameter is a measure of the central tendency.

[97] Among the most commonly used central tendencies are the mode, the median, and the mean. The mode is the most frequent diameter in a distribution: it corresponds to the maximum of the frequency curve. The median represents the value for which the total frequency of the values ​​above and below it is identical (in other words, we find the same number or total volume of particles above and below the median). The mean must be calculated and this determines the point where the moments of the distribution are equal. For a distribution 2102481 of 47 normal, the mode, mean and median coincide, while they are different for the case of a non-normal distribution.

[98] The average diameter of the particles constituting an inorganic powder can be measured in particular by: - laser light diffraction for particles in the range between 3 mm and approximately 0.1 pm; - sedimentation / centrifugation; - dynamic light scattering (DLS) for particles in the range between 0.5 pm and 2 nm; - analysis of images obtained by microscopy; - X-ray diffraction at small angles.

[99] The term granularity of the solid inorganic powder phase designates the dimensions of the particles that make up the solid inorganic powder phase. Granularity is characterized by the concept of average diameter described above.

[100] Often, the ceramic composition comprises, as powdered ceramic material(s), alone or as a mixture, an oxide and / or a nitride and / or a carbide. As examples of oxides that may be suitable in the context of the invention, mention may be made in particular of metal oxides, and in particular of titanium oxide, zirconium oxide, aluminum oxide and magnesium oxide, with titanium oxide being preferred. As examples of carbides, mention may be made in particular of metal carbides, and in particular of silicon carbide. As examples of nitrides that may be used, mention may be made in particular of titanium nitride, aluminum nitride, and boron nitride. According to a preferred embodiment, the ceramic composition 2102481 of 47 comprises at least one metal oxide as an inorganic powder material, and preferably titanium oxide.

[101] In the context of the invention, the ceramic composition has a suitable rheology in terms of plasticity for its extrusion through the extrusion head 6.

[102] According to a first embodiment, the matrix of the ceramic composition comprises one or more solvents. The solvent(s) may be aqueous or organic. Examples that may be mentioned are water, ethanol or acetone.

[103] Furthermore, the matrix of the ceramic composition comprises one or more organic additives. Advantageously, these organic additives are soluble in the solvent(s) of the matrix. The organic additive(s) that are suitable in the context of the invention may be selected, as non-limiting examples, from: - binders, and for example among cellulose ethers such as hydroxyethylcellulose which is a polymer, gum arabic which is a polysaccharide, or polyethylene glycol (PEG); - lubricants and plasticizers, and for example between glycerol or stearic acid; - thickeners and gelling agents, and for example between xanthan gum or agar, which is a polymer of galactose.

[104] The mass content of the inorganic powder material(s) in the ceramic composition may be between 50 and 90%, preferably between 80 and 85% by weight, with respect to the total weight of the ceramic composition. 2102481 of 47

[105] The mass content of the matrix in the ceramic composition may be between 10% and 50% by weight, preferably between 15 and 20% by weight, with respect to the total weight of the ceramic composition.

[106] This ceramic composition is not a powder but a paste. The rheology of this ceramic composition can be regulated thanks to the granularity of the solid inorganic powder phase, and / or thanks to the nature of the organic additives when these are present and / or thanks to their respective proportions. In fact, for example, the use of a matrix comprising one or more organic additives soluble in one or more solvents included in the matrix makes it possible to modify the rheology of the ceramic composition.

[107] According to a second embodiment, the ceramic composition comprises a matrix consisting of one or more thermo-melt polymers. The matrix is ​​organic in nature and solid at room temperature.

[108] The term “hot-melt polymer” refers to a polymer that softens under the effect of heat.

[109] As examples of hot-melt polymers that may be suitable in the context of the invention, the following polymers or families of polymers, optionally functionalized, used alone or in combination in the matrix, may be mentioned: polylactic acid (PLA), polyvinyl alcohol (PVA), acetonitrile butadiene styrene (ABS), polypropylene (PP), polyethylene, polyethylene terephthalate (PET), thermoplastic polyurethane (TPU), polyolefins, thermoplastic elastomers (TPE), polyolefin-based elastomers (TPE-O) and polycarbonate. 2102481 of 47

[110] The mass content of the inorganic powder material(s) in the ceramic composition may be between 40 and 95%, preferably between 70 and 90% by weight, relative to the total weight of the ceramic composition.

[111] In the context of the invention, the ceramic composition, preferably in the form of granules, is preheated upstream so that the hot-melt polymer(s) are softened so that the ceramic composition can be placed under pressure upstream of the flow orifice 8. In a conventional manner, the extrusion head 6 is heated to soften the hot-melt polymer(s) which then allows extrusion of the ceramic composition. The temperature of the extrusion head 6 and of the flow orifice 8 can be regulated according to the hot-melt polymer(s) present in the ceramic composition.

[112] Furthermore, in the context of the invention, the rheology of the ceramic composition can be regulated thanks to its temperature within the extrusion head and / or the granularity of the solid inorganic powder phase, and / or thanks to the nature of the hot-melt polymer(s) and / or thanks to their proportions.

[113] According to the invention, the method according to the invention aims at breaking down the digital trajectory into paths so as to avoid, in the internal part of the manipulable raw three-dimensional structure, the creation of spaces between the rounded edges of the current deposit of material and those of the deposit of material previously deposited. As explained for Figures 1A and 1C, the deposit of strands of material produced contiguously according to the state of the art leads to the creation of 2102481 of 47 material voids that persist after sintering and can lead to a reduction in the mechanical strength of the porous monolithic inorganic support.

[114] The method according to the invention aims at determining whether the manipulable raw three-dimensional structure comprises a wall whose width is equal to or greater than the nominal width L of the material reservoir. Indeed, for a wall whose width is equal to the nominal width L of the material reservoir, this wall is built up with the aid of single superimposed material reservoirs without the creation of material voids. For this type of wall, it should be noted that the superimposed deposition of material strands does not lead to the creation of material voids except in the case where the path comprises a route with a crossing as given in relation to Figures 6 and 13A to 13E.

[115] As shown in Figure 6, a path with crossing is shown for a figure-eight structure. A crossing occurs when the path, looping around itself, returns to point I of the coordinates X¡, Y¡ where a material deposit (order n1) has already been made. The crossing of a material deposit (order n-1) is carried out by the current material deposit (order n) coming into contact with the previously produced material deposit (Figure 13A). The extrusion of the material is then stopped (Figure 13B) by sliding at the same altitude and resuming extrusion after the downstream edge of the material deposit of order n-1 (Figure 13C). The material deposit of order n+1 is superimposed on the material deposit of order n-1 and the crossing of this new material deposit is produced as described above (Figures 2102481 of 47 13D and 13E). As is clear from the figures, the juxtaposition or joining of two material deposits at the time of complete crossing of a previously deposited material deposit leads to the creation of gaps.

[116] Figures 3 and 3A illustrate a first variant of an embodiment of said conventional mode for producing a wall 10 whose width is equal to the nominal width L of the material reservoir. By way of example, this wall 10 forms part of a manipulable raw three-dimensional structure which takes the form of a support with a circular section having a single central channel 11. The path of the extrusion head 6 is broken down into circular paths of mean diameter D, in each case of length πϋ and with an elevation of the extrusion head according to the vertical axis in increments of e for each rotation. The extrusion head 6 carries out a first rotation t being positioned at an elevation e with respect to the level of the plate 5, such that it extrudes a constant quantity of material to produce a material reservoir of constant thickness e.For example, arrow F indicates the counterclockwise direction of travel of the extrusion head 6 in this case. Once the first rotation t has been completed, the extrusion head 6 is raised by an amount equal to the thickness e of the material deposit to produce a second rotation, extruding a constant quantity of material to deposit the material continuously. The material deposits 3 are produced in this manner until the desired height is obtained.

[117] Figures 4 and 4A illustrate a second variant of embodiment called vessel mode for producing a wall 10 in the case for example of a manipulable raw three-dimensional structure which is 2102481 of 47 is presented in the form of a support with a circular section having a single central channel 11. The path of the extrusion head 6 is broken down into circular paths of average diameter D, with a length in each case equal to πϋ, with an elevation of the extrusion head according to the vertical axis by means of a succession of increments such that at the end of each rotation, the sum of these increments makes it possible to obtain, for each rotation, a deposit of material with a thickness e.

[118] During the first rotation t, the extrusion head extrudes an increasing amount of material proportional to its position on the average slope ramp e / mD. Once the first rotation t has been completed, the extrusion head 6 makes its second rotation following the same average slope e / πϋ and extruding a constant amount of material to continuously deposit the material according to an elevation corresponding to the thickness e. The material deposits 3 are produced in this way until the desired height is obtained.

[119] In the case where the manipulable raw three-dimensional structure comprises a wall whose width is greater than the nominal width L of the material reservoir, then the material reservoirs must be produced according to the invention because it is no longer possible to construct this wall using only superimposed material reservoirs. According to the invention, the method provides for interlacing of material reservoirs 3 as a result of interlacing by one or more superpositions and / or crossings of said material reservoirs 3 in order to avoid the creation of material spaces or voids between the material reservoirs. In the context of the invention, interlacing is understood to mean, 2102481 of 47 regardless of the definition given by quantum mechanics, the state of entangled material deposits due to superpositions and / or crossings.

[120] Furthermore, for a wall 10 whose width is greater than the nominal width L of the material tank 3, the method according to the invention aims to break down the digital trajectory of the extrusion head 6 exclusively into paths with overlapping and / or paths with crossing. It is to be understood that the extrusion head 6 (or more specifically its carrying carriage) follows, point by point, a completely defined path from the beginning to the end of the process of producing the manipulable raw three-dimensional structure 2, through a succession of points. Each of these points is spatially predefined by its Cartesian coordinates Xi, Yi, Z¡. X¡ and Yi define exactly the position of the center of the circular orifice 8 of the extrusion nozzle.The extrusion head 6 thus moves from a point A (Xa, Ya, Za) to a point B (Xb, Yb, Zb) along a vector AB, then from point B to a point C according to the vector AC, and thus according to a succession of lines of instructions called G-code instructions of the computer program that links the 3D digital model and the printing machine. It should be noted that in addition to the determination of the trajectory at each of its points determined by the Cartesian coordinates Xi, Yi, Zi, all the parameters, namely in particular the extrusion flow rate E and the travel speed F of the extrusion head 6, are defined for each of these points. 2102481 of 47

[121] It is to be understood that the digital path of the extrusion head 6 is broken down into overlapping paths and / or crossing paths. An overlapping path corresponds, as illustrated in Figures 5 and 5A, to a portion of the path for which a material deposit 3 (of order n) partially overlaps at least one edge 3b of a previously deposited material deposit 3 (of order n-1) so as to avoid, on the inside of the manipulable raw three-dimensional structure, gaps between the rounded edges of the current material deposit and those of the previously deposited material deposit.As can be seen more specifically from Figure 5A, the material being deposited (of order n) is partially superimposed on at least one edge 3b of a deposit of previously deposited material (of order n-1) by a portion of superimposed material 3c present in its thickness, i.e., between the upper face 3s and the lower face 3¡ of said deposit in progress. In addition, this portion of superimposed material 3c has a thickness strictly less than the nominal height e.

[122] According to a feature of the method according to the invention, for an overlapping stroke, the extrusion head 6 is controlled to adjust the extrusion flow rate as a function of the degree of overlapping of the portion of overlapping material on the previously deposited deposit of material.

[123] A crossing path corresponds, as illustrated in Figures 6A and 6B, to a portion of the trajectory for which a deposit of material being deposited (of order n) crosses at least one deposit of previously deposited material (of order n-1), with an overlap 2102481 of 47 completes said deposit of previously deposited material so as to avoid, on the internal part of the manipulable raw three-dimensional structure, the spaces between the rounded edges of the current deposit of material and those of the previously deposited deposit of material. In a crossing path, the material being deposited (order n) completely overlaps, by means of a portion of overlapping material 3c, the previously deposited deposit of material (order n-1), i.e. extending at least from one edge 3b to the other of the previously deposited deposit of material. This crossing between these two deposits of material can occur according to all possible angles between the directions of these two deposits of material. The portion of overlapping material 3c is considered within the thickness of the current deposit of material, i.e. between the upper face and the lower face of said deposit in progress.Furthermore, this part of the overlying material 3c has a thickness that is strictly less than the nominal height e.

[124] The above description results in an interlacing of the material deposits 3 with each other. This interlacing exists from the moment when there is, in a turn, at least one path with overlap and / or a path with crossing. Each of the following turns that necessarily comes to “cover” the previous overlap(s) and / or the previous crossing(s) produces an interlacing of the material deposits 3. The interlacing of the material deposits is therefore the result of the multiple turns for each of which there is at least one path with overlap and / or a path with crossing. 2102481 of 47

[125] According to a feature of the method according to the invention, by means of a crossing path, the extrusion head 6 is controlled to reduce the quantity of deposited material to produce the part of superimposed material having a thickness strictly lower than the nominal height e. In that way, as clearly emerges from Figures 6A and 6B, each deposit of material has a nominal height e except at or when passing over a deposit of material already produced.

[126] Thus, in the example illustrated in Figures 6A and 6B, a first material deposit n is produced with a thickness less than the nominal height e, for example equal to e / 2. For a material deposit n crossing the material deposit n-1 previously produced, this material deposit n is produced with a nominal thickness e except when it overlaps the preceding material deposit n-1. When the extrusion head protrudes from the material deposit, the extrusion flow rate is reduced so that the overlapping part 3d of the deposit has a thickness less than the nominal thickness e, equal to e / 2 in the illustrated example. In this way, the crossing paths produced according to the method of the invention do not have material voids, contrary to the prior art, as illustrated in Figures 13C to 13E.

[127] According to a preferred implementation variant of the invention, the extrusion head 6 is mounted in a movable manner according to the predefined digital path T to carry out a succession of turns t each corresponding to the path taken to find the same position in the horizontal plane with an elevation corresponding to a nominal height e. In other words, the digital path of the extrusion head 6 is 2102481 of 47 is broken down into turns, considering that for each turn, the extrusion head rises from the nominal height e. Of course, the number of turns t is selected to obtain the desired height for the raw three-dimensional structure being manipulated.

[128] In each of these rotations t, the course of the extrusion head 6 is made up of one or more overlapping paths and / or one or more crossing paths. For each of these rotations, the extrusion head 6 follows a path formed by a succession of points with coordinates Xi, Yi, Z¡ as explained above, with the possibility of elevation along the vertical axis, at one or more of these points.

[129] According to a first implementation example which is described more specifically in relation to Figures 7 to 7B and 8 to 80, the method provides for controlling the extrusion head 6 such that, in at least one rotation, the extrusion head is raised at least once by an intermediate height which is a fraction of the nominal height e.

[130] According to a second implementation example which is described more specifically in relation to Figures 9 to 9B and 10 to 10C, the method provides for controlling the extrusion head 6 such that, in at least one rotation, the extrusion head is positioned at different points of the path, rising at each point by an increment in height whose sum corresponds to the nominal height e.

[131] The following description in relation to figures 7 to 7B and 8 to 80, describes the first example of implementation of the method called “sequential vessel mode” which consists of building a part of the route at the same altitude a (for example a=e / 2) then increasing in elevation by a 2102481 of 47 value a with the aim of completing the turn and overlapping the preceding deposit with a portion of overlapping material of value a. According to this method, the nominal height e is an integer multiple of a.

[132] Figure 7B shows the first half of the path of the extrusion head corresponding to a circle of diameter D1. As illustrated in Figures 7 and 7A, a first bead of material 3 is deposited on the horizontal plate 5 with an elevation equal to e / 2 continuously according to the complete course of the circle of diameter D1. In the context of the invention, this course covering a circumference of 360° corresponds to a half turn. Once this half turn has been made to the same altitude e / 2, the extrusion head 6, and therefore the center of the flow orifice 8, moves radially towards the center of the circle by a distance equal to ad = (D2-D1) / 2 and is raised by an increment equal to e / 2 following a circle of diameter D2 (Figures 8A, 8B).This is followed by a material deposit which partially overlaps the inner edge 3b of the previously deposited material deposit by a portion of overlapping material 3c presented in its thickness. It should be noted that the portion of overlapping material 3c has a thickness strictly less than the nominal height e. The extrusion head 6 has therefore made a rotation t in the context of the invention which corresponds to the path with the circle of diameter D1 followed by the circle of diameter D2 (Figure 8B).

[133] In the position shown in Figure 8A, the extrusion head 6 (and therefore the center of the flow orifice 8) is displaced radially outward from the circle D2 by a distance equal to a and is raised again by an increment e / 2 so as to partially overlap the circle D2. 2102481 of 47 deposit of preceding material according to circle D1. This is followed by a deposit of material that partially overlaps the outer edge 3b of the previously deposited material deposit, by a portion of overlapping material 3d presented in its thickness. The portion of overlapping material 3d has a thickness strictly less than the nominal height e. The extrusion head 6 moves radially towards the interior of circle D1 by a distance equal to ad and is raised again by an increment e / 2 so as to partially overlap the preceding material deposit according to circle D2. This is followed by a deposit of material that partially overlaps the inner edge 3b of the previously deposited material deposit, by a portion of overlapping material 3c presented in its thickness.The extrusion head 6 has therefore made a second rotation t in the context of the invention which corresponds to the path with the circle of diameter D1 followed by the circle of diameter D2 (Figure 80).

[134] The extrusion head 6 is thus controlled according to a number of turns which enables construction according to the desired height for the manipulable raw three-dimensional structure 2. It should be noted that in the illustrated example, the courses according to the first turn and the second turn are identical. Of course, it can be foreseen that the courses of the different turns are different from each other. Likewise, the nominal height e for each of the turns which is identical in the illustrated example, can be different for the different turns.

[135] The following description in relation to Figures 9 to 9B and 10 to 10C, describes the second example of implementation of the method called 2102481 of 47 “continuous cup mode” for which the extrusion head 6 extrudes an increasing amount of material proportional to its position.

[136] Figure 9B shows the first half of the path to be performed by the extrusion head corresponding to a circle of diameter D1. As illustrated in Figures 9 and 9A, a first bead of material is deposited on the horizontal plate 5 with a progressive elevation in a continuous manner according to the complete course of the circle of diameter D1 to reach at the end of the course of the circle D1, a height equal to e / 2. In the context of the invention, this course corresponds to a half turn and it should be noted that for the quarter of a turn, the thickness of the material deposit is equal to e / 4. Once this half turn has been performed to reach the altitude e / 2, the extrusion head 6 moves radially towards the center of the circle by a distance equal to ad=(D2-D1) / 2 and is progressively elevated in a continuous manner, following a circle of diameter D1 to reach the altitude e at the end of the path performed according to this diameter D1 (Figures 10, 10A, 10B).This is followed by a material deposit that partially overlaps the inner edge 3b of the previously deposited material deposit by a portion of overlapping material 3d presented in its thickness. It should be noted that the portion of overlapping material has a thickness strictly less than the nominal height e. The extrusion head 6 has therefore made a rotation t in the context of the invention, which corresponds to the path with the circle of diameter D1 followed by the circle of diameter D2.

[137] In the position illustrated in Figures 10A, the extrusion head 6 moves radially towards the outside of the circle D2 by a distance equal to and is progressively raised by an increment e / 2 of 2102481 of 47 such that it partially overlaps the preceding material deposit according to circle D1. This is followed by a material deposit which partially overlaps the outer edge 3b of the previously deposited material deposit by a portion of overlapping material 3d presented in its thickness. The portion of overlapping material has a thickness strictly less than the nominal height e. The extrusion head 6 moves radially towards the interior of circle D1 by a distance equal to a and is progressively raised again in an increment e / 2 such that it partially overlaps the preceding material deposit according to circle D2. This is followed by a material deposit which partially overlaps the inner edge of the previously deposited material deposit by a portion of overlapping material presented in its thickness.The extrusion head has therefore made a second rotation t in the context of the invention which corresponds to the path with the circle of diameter D1 followed by the circle of diameter D2 (Figure 10C).

[138] The extrusion head is thus controlled according to a number of revolutions which allows construction according to the desired height for the manipulable raw three-dimensional structure. It should be noted that in the illustrated example, the courses according to the first revolution and the second revolution are identical. Of course, it can be foreseen that the courses of the different revolutions are different from each other. Likewise, the nominal height for each of the revolutions which is identical in the illustrated example, can be different for the different revolutions.

[139] Figure 11 allows to illustrate by way of example, using curve A, the course of the extrusion head 6 for the method called “mode of 2102481 of 47 “sequential vessel” described in connection with Figures 7 to 7B and 8 to 8C. The path is carried out, for example, in a point-to-point manner by means of two increment stages e / 2. According to this example, the extrusion head 6 is controlled such that, in at least one rotation t, the extrusion head is raised at least once by an intermediate height which is a fraction of the nominal height.

[140] Curve B describes the course of the extrusion head 6 for the method called “continuous cup mode” illustrated in Figures 9 to 9B and 10 to 10C. In the example illustrated, each path (a rotation t) is carried out according to fifteen points (of coordinates in X, Y) that is to say with fifteen increments or steps of height dZ proportional to the distance separating two successive points, that is to say fifteen increments of equal height dZ=e / 15. The method has as its objective to control the extrusion head 6 so that, in at least one rotation, the extrusion head is positioned at different points of the path, rising at each point by an increment of height whose sum corresponds to the nominal height e. It should be noted that Figure 11 shows the continuous progressive elevation curve C which corresponds to the progressive elevation according to the Z axis of the extrusion head.

[141] It is evident from the above description that the extrusion head 6 is controlled such that, by succession of rotations, the extrusion head is raised to build the manipulable raw three-dimensional structure 2 according to the digital 3D model M, following a path with an uninterrupted material deposition 3 from the start to the end of the construction of the manipulable raw three-dimensional structure 2. 2102481 of 47

[142] On the other hand, the extrusion head 6 is controlled such that, during each revolution t, the extrusion head moves in the horizontal plane X,Y such that the material being deposited partially overlaps an edge of a previously deposited deposit of material. Advantageously, for an overlapping stroke, the extrusion head 6 is controlled to adjust its position in the horizontal plane X,Y to define the degree of overlapping of the overlapping material portion 3d on the previously deposited deposit of material. As described above, the displacement d in the horizontal plane X,Y between consecutive strokes makes it possible to select the degree of overlapping between the material deposits.

[143] In the exemplary embodiments illustrating the method of the invention, it should be noted that the wall to be constructed has a width that can be produced from two partially superimposed material reservoirs. Of course, the method of the invention can be implemented to construct a wall whose width requires a greater number of material reservoirs, as illustrated by way of example in Figures 12A and 12B with three reservoirs. According to this exemplary embodiment, the wall to be constructed delimits two adjacent channels and has a width with the superposition of three reservoirs. The construction of this wall is obtained in vessel mode according to the path a1+b1+c1+d1+e1+f1+g1 then the path a2+b2+c2+d2+e2+f2+g2 then a3+b3+... The flow of the extrusion head 6 has dropped to zero when the extrusion head 6, ending at fi, meets e¡ up to the bifurcation (following gi) to meet ai+1.The thicknesses of the material deposits of the first turn are. 2102481 of 47 proportional to the distance traveled while the thickness of the material deposits from the second turn onwards are equal to the nominal thickness.

[144] According to an implementation feature of the invention, it should be noted that by means of an overlapping path, the extrusion head 6 is controlled to keep the flow rate of deposited material constant while the flow rate of the extrusion head 6 decreases by means of a crossing path.

[145] The method according to the invention thus makes it possible to construct a manipulable raw three-dimensional structure 2 without creating material voids, regardless of the shapes or dimensions of the walls. Figure 15 illustrates an example of a manipulable raw three-dimensional structure 2 which, after sintering, does not have any material voids in the walls produced. The examples described above explain a first advantageous embodiment variant called conventional mode or cup mode, for which the extrusion die follows a path with an uninterrupted material deposition 3 from the start to the end of the construction of the manipulable raw three-dimensional structure 2.

[146] Figure 14 illustrates a second embodiment variant for which the extrusion die 6 is mounted movably according to the predefined digital path T in order to carry out the material deposition in superimposed manner by means of layers each rising by a nominal height e. The layered material deposition is described in particular in patent applications WO 2020 / 109715 and WO 2020 / 109716. This method consists in forming a first layer 3i and 32, according to the 2102481 of 47 digital 3D model M predetermined by the computer design program, thanks to the horizontal displacement of the flow hole 8 above the horizontal plate 5.

[147] The extrusion head 6 moves horizontally, and therefore parallel to the horizontal plate 5, according to a path predetermined based on the digital 3D model M, to form the first layer. After depositing the first layer, the extrusion head 6 moves so that the deposited bead forms the second layer 3s according to the digital 3D model M. To do so, the extrusion head 6 moves vertically (i.e. along the Z axis) and horizontally (i.e. along the X and / or Y axes) to the desired position. The extrusion of the inorganic composition 4 through the extrusion head 6 can be continued or discontinued. In this way, the second layer is deposited on the first layer by superimposing the deposit of material on the previously deposited layer, according to the digital 3D model M.

[148] Of course, the deposition of material occurs in such a way as to ensure an overlap (partial or complete) as described above in order to avoid the creation of gaps between the material deposits. Thus, in the example illustrated in Figure 14, the material deposit 3s of the second stratum partially overlaps, from its two opposite edges, the neighbouring edges of the two previously deposited material deposits 3i and 32, thus avoiding the creation of empty spaces.

[149] The method according to the invention allows to build manipulable crude three-dimensional structures 2 intended to form 2102481 of 47 porous monolithic inorganic supports 1, which have the mechanical strength characteristics suitable for use in tangential filtration. Advantageously, the extrusion head 6 is controlled to provide, in the manipulable raw three-dimensional structure 2, at least one channel 11 for the circulation of a fluid medium to be treated. The wall of this channel 11 is produced by means of the rounded edges 3b of the material deposits located opposite the superimposed material portion. As illustrated in the figures, the extrusion head 6 is configured such that the material deposits have rounded edges 3b. In this way, a channel 11 has a wall having rounded reliefs, each extending over the entire periphery of the channel, being defined by the rounded edge 3b of the material deposits.These rounded perimeter reliefs extend in an overlapping manner along the channel, as illustrated in Figures 15 and 16. On the scale of the thickness of the material deposits e, these rounded perimeter reliefs participate at their level, in this channel, during tangential filtration, in the generation of turbulence along the entire channel. Figure 16 shows that these rounded perimeter reliefs are present even when the internal wall of the channels 11 is covered by separation layers whose profile appears in the form of a white line in Figure 16. The separation layer or layers thus conform to the rounded perimeter reliefs whose rounded profile persists to participate in the creation of turbulence within the channel 11. It should be noted that the porous monolithic inorganic support 1 produced according to the method according to the invention has an external surface which also has. 2102481 of 47 a succession of rounded perimeter reliefs St that extend in an overlapping manner along the support, as illustrated in figures 2B, 15 and 16. 2102481 of 47 G. BREUER - 30525624826 Digitally signed by PORTALTRAMITES - INPI Date: 2022.12.28 10:43:21 -03:00 Reason: Digitally signed by the INPI Location: Buenos Aires, Argentina 2102481

Claims

1. A method for producing at least one porous monolithic inorganic support (1) having at least one channel for the circulation of the fluid to be treated and having a porosity between 10% and 60% and a mean pore diameter between 0.5 µm and 50 µm, by using a 3D printing machine (I) comprising at least one extrusion head (6) movably mounted in the space above a fixed horizontal plate (5), which moves successively with a nominal height (e) according to a predefined numerical path (T) to make overlapping material deposits where each has a nominal width and a defined thickness between a lower and an upper surface, said 3D printing machine allowing the deposition of material in the form of: - a bead with rounded edges so as to create,rounded perimeter reliefs that contribute to the generation of turbulence on the wall of at least one circulation channel. - a composition (4) for constructing, on said horizontal plate (5), from the predefined numerical path (T), walls of a manipulable raw three-dimensional structure (2) to form the porous monolithic inorganic supports (1), characterized in that the method comprises: - for a wall whose width is greater than the nominal width (L) of the material deposit, decomposing the numerical path exclusively into overlapping paths and crossing paths; - feeding the extrusion head (6) of the 3D printing machine (I) with a composition (4), - controlling the extrusion head according to the numerical path such that: * for an overlapping path,The material being deposited partially overlaps at least one edge of a previously deposited material deposit by means of an overlapping portion of material presented in its thickness and having a thickness strictly less than the nominal height (e), so as to avoid, on the inside of the three-dimensional raw, workable structure, gaps between the rounded edges of the deposit of material being deposited and those of the previously deposited material deposit. * For a crossing path, the material being deposited intersects at least one previously deposited material deposit with a complete overlap of said deposit of material, by means of an overlapping portion of material presented in its thickness and having a thickness strictly less than the nominal height (e), so as to avoid, on the inside of the three-dimensional raw, workable structure,spaces between the rounded edges of the deposit of material being deposited and those of the deposit of material previously deposited. 17 claims follow,