HK1236152A1 - Tangential flow separation element integrated with flow obstacles, and manufacture method - Google Patents

Description

Element for separation by tangential flow with built-in flow obstacle and method of manufacture

Technical Field

The present invention relates to the technical field of elements for tangential flow separation of a fluid medium to be treated in order to produce a filtrate and a retentate, these elements being commonly referred to as filtration membranes. More precisely, the invention relates to a novel shape of a multi-channel porous support (carrier) for reducing or even eliminating clogging problems, and also to an additional manufacturing method of the support, and a tangential flow separation element comprising the support.

Background

Separation methods using membranes are used in various industries, particularly in the case of the production of drinking water (potable water) and the treatment of industrial sewage, and in the fields of chemistry, petrochemistry, pharmaceuticals, the agricultural product industry, and biotechnology.

The membrane constitutes a selective barrier and, under the action of the conveying force, it serves to permeate or block specific components of the medium to be treated. The size of these components relative to the size of the pores in the membrane causes them to be permeated or blocked, and thus the membrane functions like a filter. Depending on the pore size, such techniques are known as "microfiltration", "ultrafiltration", or "nanofiltration".

Existing films have different textures and textures. Generally, the membrane is composed of a porous support (carrier) that provides mechanical strength to the membrane and imparts shape thereto, which in turn determines the filtration area of the membrane. On this support one or more separation-performing layers are deposited, each of which has a thickness of a few microns and is called "separation layer", "filtration layer" or "active layer". During separation, the fluid being filtered is conveyed through the separation layer, and the fluid is subsequently spread out through the porous structure of the support so as to travel towards the outer surface of the porous support. The portion of the treated fluid that has passed through the separation layer and porous support is referred to as "permeate" or "filtrate" which is recovered by the collection chamber surrounding the membrane. The remaining part is called "retentate" and, according to the general rule, it is reinjected into the fluid to be treated upstream of the membrane via a recirculation loop.

In conventional manner, the support is initially manufactured by extrusion (extrusion) into the desired shape and then sintered at a temperature and for a time sufficient to ensure the desired strength, while leaving the desired structure of open and interconnected pores in the resulting ceramic. This method entails obtaining one or more rectilinear channels, after which a plurality of the above-mentioned separation layers are deposited and sintered in succession. The support body is generally tubular and has one or more linear channels arranged parallel to the central axis of the support body. In general, the inside surfaces of these channels are smooth and free of any irregularities.

However, it has been found that filtration membranes made from supports of this shape face problems of clogging, as a result of which they are limited in their performance in terms of flow rate. In particular, small particles and large molecules may be absorbed on the surface of the separation layer, or may be deposited thereon in the form of a sol or precipitate, and may even penetrate into and block some of the pores.

All tangential separations using filter elements rely on the principle of selective transfer, the efficiency of which depends on the selectivity of the membrane (active layer) and the permeability (flow) of the filter element considered as a whole (support + active layer). Selectivity and permeability are determined by (but not by) the characteristics of the active layer and the filter element, as selectivity and permeability may be reduced or limited due to concentration polarization, precipitation, and/or pore blocking.

Concentration polarization occurs when macromolecules present in the liquid to be treated become concentrated at the membrane/solution interface during the filtration operation, when the macromolecules exert an osmotic back pressure opposite to the separation force, or diffuse back to the centre of the liquid to be treated under the influence of fick's law. The cause of this concentration polarization phenomenon is the accumulation of residual components in the vicinity of the membrane due to the permeation of the solvent.

When the concentration of particles at the surface of the membrane during the filtration operation increases sufficiently to cause the appearance of a condensed phase (condensed phase) in the form of a sol or viscous precipitate, precipitation occurs, causing the fluid resistance to rise in addition to the membrane resistance.

Clogging of the pores occurs when particles having a size smaller than or equal to the size of the pores intrude, thereby causing a reduction in the filtration area.

Clogging and its reversibility or irreversibility are complex phenomena depending on the filter element, in particular on its separation layer, the liquid to be treated and the operating parameters.

Clogging is a major obstacle to the economic attractiveness of filtration, since in designing a filtration installation it is first directed to an increased installation area to meet the requirements in terms of process capacity, and then it is necessary to use specific technical means to compensate for the late clogging, such as regular cleaning cycles using detergents or back flushing.

Several solutions have been proposed in the prior art to mitigate the clogging phenomenon by creating turbulent flow conditions inside the channels of the filtering element.

It was originally proposed to introduce turbulence-generating means in the tubular filter element. See Journal of membrane Science, p.303-314 of 208 th (2002) of Journal of membrane Science, authored by d.m. krstic et al. By limiting clogging, these devices function to improve permeate flow and thereby increase filtration efficiency. Nevertheless, installing and attaching such devices in tubular elements is a difficult and complex operation. Moreover, they induce harmful vibrations, which are detrimental to the reliability of the device.

Jaffrin also proposes other more complex systems at Journal of Membrane Science, 324 (2008), pages 7-25, which use circular membranes and central modules that rotate relative to each other in order to create turbulence. Nevertheless, this work demonstrates that the large shear rates generated are utilized to reduce clogging.

Other solutions include modifying the shape of the tubular element. Patent FR 2503615 describes a cylindrical tube for filtering a gas mixture injected under pressure, the inner wall of the tube having indentations for generating turbulence which prevents one of the gas phases from accumulating on the wall of the tube and promotes separation by gaseous diffusion. These indentations are formed by passing the tube exiting the extrusion die between a plurality of rollers or tools to locally deform the tube through the thickness of the tube wall. Patent FR 2503616 describes a method based on the same principle, consisting in deforming the wall of the tube as it leaves the extrusion die, using knurling wheels placed facing each other on both sides of the tube or in staggered positions.

In both these patent documents, the final shaping step is carried out by plastic deformation immediately after the previous step of extruding the single-channel tube, so as to obtain indentations inside the single channel by pressing on the outer surface of the tube using a rotary punch or similar device. Whether such "impressions" are easier or more difficult to obtain depends on the ductility of the material, i.e. its property of being able to undergo permanent deformation without breaking. However, the compounds used for the manufacture of ceramic membranes do not have good ductility: they are easily extruded but their elongation at break is usually less than 5%. Moreover, with such a technique, only small-sized indentations can be obtained. Finally, the deformation caused over the entire thickness of the tube leads to high levels of stress in the material and to the risk of cracking, thus degrading the mechanical strength considerably. Reference may also be made to patent application FR 2736843, which proposes a perforated tube with a single channel, the wall of which comprises indentations, but the wall of the periphery of the support is smooth. To this end, the perforated tube is shaped by means of an extrusion die comprising cylindrical pins arranged along its axis, the pins or outlets of the extrusion die being mounted so as to be rotatable and having a non-circular cross-section. The manufacturing technique is still limited to a specific type of indentation, i.e. an indentation that extends continuously from one end of the separating element to the other, and which does not produce any change in the flow section of the channel. Moreover, it cannot be transferred to the manufacture of separation elements with a series of internal channels. Nevertheless, there is an increasing demand for multi-channel separation elements, since they allow to increase the filtration surface area and thus the performance.

In the same way, document EP 0813445 describes a filter element having one or more passages, each of which comprises a helical groove, which can be a single start groove, a double start groove or a triple start groove. This filter element has the same drawbacks as the filter element described in document FR 2736843.

On this background, the present invention aims to provide novel filter elements having a single-channel or multi-channel structure and a shape suitable for reducing or even eliminating clogging phenomena, as well as manufacturing techniques suitable for manufacturing these filter elements. It is an object of the present invention to provide a novel filter element, the shape of which can be adjusted to have regularity (order) to generate high surface shear stress and strong turbulence inside the channels, but without the drawbacks of the prior art solutions.

Disclosure of Invention

To achieve this object, the invention provides a monolithic tangential flow separation element for separating a fluid medium to be treated into a filtrate and a retentate, said separation element comprising a three-dimensionally structured, rectilinear, rigid porous support in which at least one channel is provided for the passage of the fluid medium to be treated in order to recover the filtrate from an outer surface of the support, the outer surface of the support having a fixed profile.

According to the invention, the monolithic rigid porous support comprises, on or in the inner wall of the channel, a plurality of obstacles acting on the flow of the fluid to be filtered, which have the same identity and continuity in terms of material and porosity as the support, which, by emerging between a first position and a second position, chosen along the longitudinal axis of the channel, force the fluid to flow around them, thus obstructing or disrupting the flow of the fluid.

Furthermore, the elements of the invention may also incorporate at least one and/or a plurality of the following additional features:

at least one obstacle of the passage creates a sharp narrowing or a tapered nozzle portion in the passage in the flow direction of the fluid to be treated;

at least one obstacle of the channel produces a locally narrower flow cross section at the location of said obstacle, perpendicular to the longitudinal axis of said channel and having a shape different from the shape of the portions of the channel upstream and downstream of the obstacle;

between the portion of the passage upstream of the narrowest flow cross-section and the narrowest flow cross-section, one of the conditions (criteria) chosen from the following remains constant while the others vary, this condition being chosen from: the shape, area, wetted perimeter (wetted perimeter), and hydraulic diameter (hydrodynamic diameter) of the flow right cross-section of the channel;

all conditions, between the portion of the passage upstream of the narrowest flow cross-section and the narrowest flow cross-section, chosen from: the shape, area, wet circumference and hydraulic diameter were all kept constant.

At least one obstacle of the channel has a flow cross section perpendicular to the longitudinal axis of the channel, which rotates (turns) around the longitudinal axis of the channel between two positions taken along the longitudinal axis of the channel;

the flow front section of at least one obstacle of a channel rotates in a discontinuous manner about the longitudinal axis of said channel between the ends of said channel;

the element comprises at least one separation layer deposited continuously on the inner wall of the channel and completely covering the obstacle;

the porous support is made of an organic or inorganic material;

the separating or intermediate layer is made of organic or inorganic material; and

the three-dimensional structure of the porous support has a plurality of different stacks (ply ) that can be observed by optical microscopy or by scanning electron microscopy.

The invention also provides a method of manufacturing a monolithic separation element according to the invention.

The method of manufacturing a tangential flow separation element of the present invention comprises: the three-dimensional structure of the support is made by forming a plurality of individual stacks, which are stacked and successively joined to each other, so as to progressively build up the desired three-dimensional shape.

Furthermore, the method of the invention may also incorporate at least one and/or a plurality of the following additional features:

producing the three-dimensional structure by repeating the following steps:

-making a continuous substrate (bed) of material for forming the porous support, having a fixed thickness over a region greater than the section of the porous support at the aforesaid lamination;

locally consolidating some of the above-mentioned materials to form a pattern determined for each laminate, so as to bond the individual laminates made in this way to the previous laminate while creating them;

a continuous substrate made of a solid material in powder form or a liquid material such as a photopolymerizable resin;

a continuous substrate made of solid material in the form of an organic or inorganic powder;

a continuous substrate made of a medium in the form of a photopolymerizable liquid precursor in which an inorganic powder is deposited;

each stack is made by continuously or discontinuously melting a strand (string) of a hot-melt solid precursor, either a hot-melt organic polymer used per se for the preparation of the organic support and of the organic layer, or a mixture of hot-melt organic polymer and ceramic inorganic powder (support is of inorganic nature); and

producing a continuous string of material (bead) by spraying the powder melted in the laser beam.

The tangential flow separation element obtained by the method defined in the context of the present invention allows to form the three-dimensional structure of the support. It will be seen that this structure is visualised by observation of the various stacks described above by means of an optical microscope or by means of a scanning electron microscope. Of course, it is desirable that the boundaries between the stacks be as thin as possible.

Drawings

Other various features will appear from the following description with reference to the accompanying drawings, which show, by way of non-limiting example, embodiments of the invention.

Fig. 1A is a longitudinal sectional view of the support body, showing one embodiment of the barrier portion.

Fig. 1B and 1C are cross-sections of the support body taken at a position level with the obstacle and at an upstream of the obstacle with respect to a flow direction of the fluid, respectively.

Fig. 2 is a longitudinal section of the support body showing the obstruction creating a sharp constriction and a tapered nozzle portion.

Fig. 3A and 3B are cross-sections of the support body taken at the upstream of the obstacle and at the obstacle, respectively, showing the variation of the flow section of the channel (its area is constant).

Fig. 4A and 4B are cross-sections of the support body taken at the upstream of the obstacle and at the obstacle, respectively, showing the variation in the flow section of the channel (its wet circumference is constant).

Fig. 5A and 5B are cross-sections of the support body taken at the upstream of the obstacle and at the obstacle, respectively, showing the variation of the flow section of the channel (the hydraulic diameter thereof is constant).

Fig. 6A and 6B are a longitudinal sectional view and a cross sectional view of the support body, respectively, showing the property that the shape of the flow section thereof is constant while the size of the channel is varied.

Fig. 7A and 7B are a longitudinal sectional view and a cross sectional view of the support body, respectively, illustrating the property that the shape of the flow section of the channel having a constant size is constant.

FIG. 8 is a partial longitudinal cross-sectional view of the support body illustrating the rotation of a partially constant flow cross-section.

Detailed Description

First, definitions are given for some terms used in the context of the present invention.

The term "average particle size" is used to indicate the d50 value of the volume distribution (volume distribution) in which 50% of the total volume of the particles corresponds to the volume of particles having a diameter smaller than this d50 value. The volume distribution is a curve (analytical function) plotted with the particle volume frequency as a function of particle diameter. This d50 value corresponds to the median value between two equal parts of the area located below the frequency curve obtained by laser diffraction of the particle size, which is the reference technique used in the context of the present invention for measuring the mean diameter of particles. For the measurement technique of d50, see in particular:

ISO standard 13320: 2009, for laser particle size measurement techniques;

ISO standard 14488: 2007, techniques for sampling the powder analyzed; and

ISO standard 14887: 2000, for reproducibly dispersing a powder sample in a liquid prior to particle size measurement by laser.

The term "average pore diameter" is used to denote the d50 value of the capacity distribution, where 50% of the total capacity of the pores corresponds to the capacity of pores having a diameter smaller than this d50 value. The capacity distribution is a curve plotted with the particle volume frequency as a function of pore diameter (analytical function). This d50 value corresponds to the median value between two equal parts of the area lying below the frequency curve obtained by mercury intrusion for an average diameter of the order of a few nanometers (nm), or by adsorption of gases, in particular of N, for pores of smaller diameter₂(to obtain a frequency curve), both techniques being used as references in the context of the present invention for measuring the average diameter of the holes.

In particular, the techniques described in the following documents may be used:

ISO Standard 15901-1: 2005, about mercury intrusion measurement techniques; and

ISO Standard 15901-2: 2006 and 15901-3: 2007, for gas and absorption measurement techniques.

The invention proposes a tangential flow separation element for separating a fluid medium to be treated into a filtrate and a retentate, such element comprising a monolithic porous support, single or multi-channel, the shape of which is chosen to define a plurality of obstacles on the inner side walls of the channels in order to impede the flow of the fluid to be filtered. As will be described below, such monolithic supports with obstacles (which form an integral part of the monolithic porous structure) are made by additive technology (additive technology).

In the context of the present invention, separation elements are used for separating fluid media by tangential filtration and they are often referred to as filtration membranes. Such a separating element comprises a porous support in which one or more flow channels for the fluid to be filtered are provided. Conventionally, the support body is tubular. Each flow channel has an inlet and an outlet. In general, the inlet of the flow channel is located at one end of the support body, which end serves as an inlet zone for the fluid medium to be treated, while the outlet is located at the other end of the support body, which end is an outlet zone for the concentrate.

In such a separating element, the body constituting the support body has a porous structure. Such a porous structure is characterized by an average diameter of the pores, as derived from the distribution of pores measured by mercury intrusion porosimetry.

The porous structure of the support is open and forms an array of interconnected pores, so that fluid filtered by the filtration separation layer can pass through the porous support and be recovered at its periphery. It is common practice to measure the water permeability of the support in order to demonstrate the hydraulic resistance of the support. Specifically, in porous media, the steady state flow of incompressible viscous fluid is determined by Darcy's law. The velocity of the fluid is proportional to the pressure gradient and inversely proportional to the dynamic viscosity of the fluid, by means of a characteristic parameter known as "permeability", which can be measured, for example, according to french standard NFX45-101, month 12 1996.

The permeate is thereby recovered from the outer peripheral surface of the porous support. The walls of the channels are continuously covered by at least one filter separation layer for filtering the fluid medium to be treated. By way of limitation, the average pore size of the filtration separation layer must be smaller than the average pore size of the support. These separation layers define the following surfaces of the tangential flow separation element: the surface is in contact with the fluid to be treated and the fluid to be treated flows over the surface.

Fig. 1 shows an example of such a tubular tangential flow separating element 1 in which channels are provided, however, many other shapes can also be constructed using the method of the invention. The tangential flow separation element 1 comprises a porous support 2, the porous support 2 being made elongated extending along a longitudinal central axis a, which is why the structure of the porous support is said to be "linear". The porous support 2 shown in fig. 1 has a circular right cross-section and thus a cylindrical outer surface 5, although the right cross-section may be of any shape or polygonal shape. The term "cross-section" refers to a shape defined by a volume being cut by a plane, wherein a right cross-section of a cylinder is a shape defined by a cylinder being cut by a plane perpendicular to the longitudinal center axis.

According to a characteristic of the invention, the outer surface or peripheral surface 5 of the support body has a fixed profile. It will be appreciated that this profile corresponds to the outer shape of the porous support 2 taken in a transverse plane containing the longitudinal central axis a. In the example shown, the support body 2 has a linear profile and is constant from the inlet to the outlet. In other words, a constant (fixed) profile means that all external generatrices parallel to the central axis of the support are straight lines parallel to each other.

In other words, the outer surface 5 is free of any surface irregularities, except those created by the porosity of the material itself, or by the surface roughness inherent to a suitable forming process. Thus, the outer surface 5 does not have any deformations or indentations.

The porous support 2 is arranged to comprise at least one channel 3, which in the example shown in fig. 1 has one channel 3 and in the example shown in fig. 2 has two channels 3. Each channel 3 is parallel to the axis a of the support extending along a longitudinal axis T, which advantageously coincides with the axis a of the support in a single-channel support. Each channel 3 has a surface covered by at least one separation layer 4, which separation layer 4 is in contact with the fluid medium to be treated flowing inside the channel 3. A portion of the fluid medium passes through the separation layer 4 and the porous support 2 such that the portion of the fluid being treated (referred to as the "permeate") flows over the outer surface 5 of the porous support. The fluid to be filtered flows between the inlet region and the outlet region in a flow direction indicated by arrow f. In the example shown, the inlet zone 6 is located at one end of the tubular support body, while the outlet zone 7 is located at the other end.

Typically, the thickness of the filter separation layer is in the range of 1 micrometer (μm) to 100 μm. Of course, in order to be able to perform its separation function and to function as an active layer (effective layer), the average pore diameter of the separation layer is smaller than that of the support. Typically, the pore size of the filtration separation layer is at least 3 times smaller than the average pore size of the support (i.e. at most one third of the average pore size of the support), preferably at least 5 times (at most one fifth).

The concept of separation layers for microfiltration, for ultrafiltration and for nanofiltration is well known to the person skilled in the art. It is generally believed that:

the mean pore size of the microfiltration separation layer is in the range of 0.1 μm to 2 μm;

the mean pore size of the ultrafiltration separation layer is in the range of 0.1 μm to 0.01 μm; and

the mean pore diameter of the nanofiltration separation layer is in the range of 0.5nm to 2 nm.

This so-called "active" microfiltration or ultrafiltration layer can be deposited directly on the porous support (for a single-layer separation layer), or it can even be deposited on an intermediate layer with a smaller average pore size, itself deposited directly on the porous support (for a single-layer separation layer).

For example, the separation layer may be based on one or more metal oxides, carbides or nitrides, or ceramics, or may consist of these materials only, wherein the term "ceramic" covers all non-metallic inorganic materials. In particular, the separating layer may be based on TiO₂，Al₂O₃And ZrO₂Or may consist of TiO only₂，Al₂O₃And ZrO₂Individually or in mixtures.

For example, the separation layer may also be based on or consist only of polymeric collodion deposited on a porous support of organic nature. For example, the separating layer may also be based on, or consist only of, a metal deposit deposited on the metallic porous support.

According to an essential feature of the invention, the support body is shaped with an inner wall 3 from the passage₁The initial (extended) at least one obstacle 9, more generally a series of obstacles 9, suitable for generating disturbances to the flow and for generating shear forces of sufficient amplitude to cause the occurrence of a backflow, so as to limit the clogging phenomena, or even eliminate them altogether. The barriers form an integral part of the monolithic porous support, i.e. they are formed by the shape given by the porous support itself, and not by any way of separate element fitted to the support. The support and the barrier together form a one-piece porous monolithic element without any kind of connection, interface or joint.

The material and the porous structure of the barrier 9 and the porous support 2 are the same, and the material and the porous structure between the barrier 9 and the porous support 2 are continuous. Therefore, the barrier 9 is mechanically integrated with the support body, and the barrier 9 has the same chemical resistance as the support body. The obstacles 9 are completely covered by the separating layer so that they do not reduce but rather increase the filtering area of the separating element.

The same nature of the material between the barrier 9 and the support 2 means that all of their parts are chemically identical, i.e. they are identical both in the porous support and in the barrier.

The same porous structure includes porosity, tortuosity and size and distribution of pores, which are the same in all parts of the element, i.e. in the obstacles and in the porous support.

By material continuity is meant that all portions of the element are chemically identical, i.e. there is no chemical discontinuity between the barrier and the porous support.

The continuity of the porous structure means that the porosity, the tortuosity and the size and distribution of the pores are the same in all parts of the element, so that there is no discontinuity of the porous structure between the obstacle and the porous support.

The obstruction serves to be disposed in a path of a fluid flowing in the channel. The obstacles 9 obstruct or interfere with the passage of the fluid to be treated, causing the fluid to have to flow around them, since they are located between a first position P1 and a second position P2, taken along the longitudinal axis T of the channel. Thus, the first position P1 is taken immediately upstream of the obstacle 9 in the flow direction of the fluid to be treated indicated by the arrow f, as defined by the cross-sectional view C-C of the channel (fig. 1C), while the second position P2 is taken at the position of the obstacle 9, downstream of the first position P1 in the flow direction f of the fluid to be treated, as defined by the cross-sectional view B-B of the channel (fig. 1B). These obstacles 9 therefore cause an increase in the flow speed with the liquid respectively flowing towards them, thus generating high levels of surface shear stress and turbulent zones in which the clogging phenomena are reduced or even eliminated. These obstacles act to promote turbulence.

In general, the obstacle 9 has a length L along the longitudinal axis a of the channel, and has an inner wall 3 along a direction perpendicular to the longitudinal axis a and starting from the channel₁Height h of (a). In the example shown in fig. 1, the passage 3 is at the barrierThe upstream and downstream of the obstacle 9 have the same diameter D.

These obstacles 9 may have regular or irregular intervals. The shape of the novel support body contemplated in the present invention presents a repetitive arrangement of one or more obstacles starting from the wall of each channel where the obstacles are disposed.

In particular, the inner wall of the channel containing the obstacles may have relief-like portions, such as recesses, protrusions, grooves, striations and/or any other shape suitable as a corresponding number of obstacles, so as to promote turbulence when the fluid flows within said channel.

In general, it will be understood that the obstacle 9 produces a flow cross section that is locally modified in its shape, its area, its wetted perimeter or its fluid diameter, or is locally offset, or even undergoes a rotation in the channel 3 in a position relative to the channel portions located upstream and downstream of said obstacle, this flow cross section of the fluid being taken perpendicularly to the longitudinal axis T of said channel.

As can be seen more clearly in fig. 2, the obstacle 9 can produce a sharp narrowing or a tapered nozzle portion in the direction of fluid flow in the passage indicated by the arrow f, as indicated by the top and bottom passages in fig. 2, respectively. The abrupt constriction has a radial wall 9 extending perpendicularly to the longitudinal axis T from the inner wall of the passage_aThe tapered nozzle portion has a wall 9 inclined at an angle α strictly greater than 0 ° and less than 90 ° with respect to the longitudinal axis T_a. Of course, the radial wall or inclined wall 9_aMay optionally be connected to the inner wall of the channel via a connecting fillet (filet).

Of course, the obstacle 9 may have a wide variety of shapes that can obstruct or interfere with fluid flow. The following examples describe various shapes of the obstacle 9 present between the portion of the channel located upstream of the smallest or narrowest flow front section of the channel and said smallest or narrowest front section portion (corresponding to the first and second positions, respectively).

Fig. 3A and 3B show a first variant embodiment, in which the shape of the flow cross-section of the passage between the first and second position is changed, while the area of the flow cross-section remains unchanged.

In the first position, the right section of the channel is square with a side length a, so that the area of the right section is equal to a²(FIG. 3A). At this point, the hydraulic diameter Dh of the channel is 4A/P, where a is the area of the flow cross section of the channel and P is the wet circumference of the flow cross section. In this example, the area A is equal to a²The wet circumference is equal to 4a, so that the hydraulic diameter Dh is a.

In the second position the passage has a square obstacle 9 with a side length of a/2₁And a complementary concave obstacle 9₂(FIG. 3B). The area of the flow normal section at the second position is A ═ a²-(a/2)²+(a/2)²＝a². The area of the right cross-section of the channel is constant. Instead, its hydraulic diameter changes because it is equal to Dh-4 a²2/3a, the wet cycle P6 a also changes.

Fig. 4A and 4B show a second modified embodiment in which the shape of the flow cross section of the channel is changed while the wet circumference P is kept constant, while the hydraulic diameter Dh and the area a of the flow cross section of the channel are changed. In this example, the right section of the channel 3 in the first position as shown in FIG. 4A is a square with a side length of 3a, giving an area A equal to 9a²The wet cycle P is 12a and the hydraulic diameter Dh is 3 a.

In the second position, the channel 3 has four portions 9₁Forming obstacles, these parts 9₁The positive sectional shape is constituted as a square with a side of a in each corner of the section of the channel (fig. 4B). In this second position, the wet circumference P is equal to 12a and constant, while the area a is 5a²The hydraulic diameter Dh is changed to 5/3 a.

Fig. 5A and 5B show a third modified embodiment in which the shape of the flow cross-section of the channel is changed, and in spite of the channel cross-sectionThe area A and the wet circumference P were varied, but the hydraulic diameter Dh was unchanged. In this example, at the first position shown in fig. 5A, the shape of the right cross-section of the channel is a square with a side length a, giving an area a ═ a²Wet cycle P4 a and hydraulic diameter Dh 4a²/4a＝a。

In the second position (fig. 5B), the right section of the channel is circular with radius r ═ a/2, so that its area a ═ r²＝πa²(ii)/4, wet cycle P ═ pi a, hydraulic diameter Dh ═ pi a²/πa＝a。

Thus, its hydraulic diameter remains unchanged, while the shape of the cross section of the channel changes. It should be noted that between the first position and the second position the dimensions change while the cross section is not rotated and not offset with respect to the central axis of the support body, however, it may of course also be provided that the cross section can be rotated and/or eccentric with respect to the central axis of the support body.

Fig. 6A and 6B show a fourth modified embodiment in which the sectional shape of the passage is constant, and the area, wetted perimeter, and hydraulic diameter of the section of the passage vary. In the first position P1, the channel is rectangular in shape, while in the second position P2, its frontal cross-sectional shape is still rectangular but smaller. It should be noted that between the first position and the second position the dimensions are varied, but the cross-section is not rotated and not offset with respect to the central axis of the support body, however, it is of course also possible to provide that the cross-section can be rotated and/or eccentric with respect to the central axis of the support body.

In the foregoing embodiment, one of the conditions taken from the following remains unchanged, while the other condition is changed, the condition taken from: shape, area, wetted perimeter, and hydraulic diameter.

In the example shown in fig. 7A, 7B, the shape of the flow cross section of the channel 3 remains unchanged, as does the wetted perimeter and hydraulic diameter of the flow cross section of the channel. In the first position P1, the channel is circular in shape, and in the second position P2, it is circular in shape of the same size, but offset with respect to the flow cross-section in the first position. The obstacle 9 is formed by offsetting the circular flow cross section. Of course, the flow cross-section may be of any shape. The obstacle 9 thus creates a sharp constriction. It should be noted that at the intersection between two portions of an offset flow section, the area of the flow cross-section varies. Thus, the area of the flow cross-section of the channel 3 remains unchanged except for the intersection between the two above-mentioned portions of the flow cross-section. In the example shown, the flow cross-section of the channel is disk-shaped.

It should be observed that the same function as creating the obstacle 9 can be obtained by rotating a non-circular flow section. This applies, for example, to the flow cross section of the channel 3 formed as an isosceles triangle whose shape remains unchanged, and also the area, wetted perimeter and hydraulic diameter of the flow cross section of the channel. In the first position, the channel bus is triangular in shape, while in the second position it is likewise triangular, but is angularly offset by a given value, for example equal to 90 °.

Fig. 8 shows an example of another embodiment involving the orientation arrangement of the obstacle 9 within the channel 3. In this example, the flow cross section of the obstacle 9 is perpendicular to the longitudinal axis T of the channel, about which it rotates between two positions P1 and P2 taken along the longitudinal axis T of the channel. The flow cross-section rotates in a discontinuous manner between the ends of the channel, i.e. the length of the obstacle is shorter than the length of the channel. For example, the obstacle 9 has the shape of at least one surface spiral, so that a plurality of spiral sections are present between the inlet and the outlet of the channel. In this embodiment, the shape and area of the right cross section of the channel 3 varies in particular between the inlet and the outlet of the channel 3. Thus, at least one such obstacle creates a sharp constriction in the flow direction of the fluid to be treated.

In the context of the present invention, the porous support (or even the tangential flow separation element as a whole) is manufactured by additive technology. The method of the invention comprises making the three-dimensional structure of the support body by forming a plurality of individual superposed layers, which are successively superposed and joined to each other, so as to progressively build up the three-dimensional structure of the support body.

The advantage of the method compared to the prior art is that the support body is made in a single production step without the need for any tools or machining, so that a larger range of support body shapes can be obtained and the shape and size of the obstacles in the channel can be changed.

When using a solid material such as a powder, the thickness of the powder substrate and hence each stack consolidated in succession is relatively small, it being possible for the guest to bond it to the underlying stack by applying energy or by spraying a liquid. In particular, the powder is deposited in a thickness ranging from 20 μm to 200 μm, depending on the additional technique chosen.

The desired three-dimensional shape can be built up layer by repetition of the binary sequence. The consolidation pattern (pattern) may vary from one ply to another. A desired three-dimensional shape is built along the selected build axis.

The size of the particle size of the deposited powder is one of the factors that determine the minimum thickness of each powder substrate and also one of the factors that determine the average diameter of the pores finally obtained. In particular, the powder used is a powder of the material constituting the support, for example a metal oxide powder, or a powder of one of its precursors. For example, the average particle size of the deposited powder may be about 35 μm to obtain an average pore size of about 10 μm within the ceramic support.

Applicants have found that by adjusting various parameters, such as the materials selected; and for a given material, adjusting the average particle size of the powder used; and adjusting the thickness of the powder substrate repeated layer by layer for a given material and a given particle size; it is also possible to obtain residual interconnected pore formations within the entirety of the consolidation in a well-controlled manner by adjusting various parameters specific to the technique selected for consolidation. This residual pore architecture is the result of controlled sintering of the powder particles to leave interconnected voids between the particles.

When using an energy beam, the main parameters that can work are its focal point, i.e. the diameter of the energy beam impinging (acting on) the powder substrate, but also the speed at which the powder substrate is swept by the photon or electron beam, and the percentage overlap ratio between the impact areas of the energy beam even while constructing the stack.

When using liquid sprays, the main parameters that can work are the weight of the droplets, their frequency, the speed at which the powder substrate is swept by the "jet" of droplets, and even the percentage overlap rate during successive passes.

The applicant has also observed that by regulating the various parameters described above, it is possible to adjust the size distribution of the pores and to control the number and tortuosity of the pores for each given population of pores.

Once the powder has agglomerated (caked) in the selected areas, non-agglomerated material can be eliminated by any suitable technique. The initial flowability of the powder used facilitates this operation. Water jet techniques or vibration may also be used to remove the last remaining powder remaining on the surface of the fabricated shape.

The final consolidation of the filter element and the final state of the porous structure are generally obtained by one or more subsequent heat treatments aimed at eliminating the binder (debonding) and/or subjecting the material to a suitable sintering. The temperature selected for such final sintering depends on the nature of the inorganic material used and the average particle size of the powder used.

In this way, the support (or even the entire tangential flow separation element) is built up layer by layer. For this purpose, computer-aided design (CAD) software is used before the subdivision of the three-dimensional structure of the support or tangential flow separation element to be manufactured in pieces is started. Thereby, the virtual three-dimensional object to be manufactured is subdivided into two-dimensional slices of very small thickness. The sheets are then made one by one in the form of individual stacks which are stacked and bonded together to build up the desired three-dimensional shape in stages.

This three-dimensional structure is made by:

or by repeating the following steps:

manufacturing a substrate of solid material (organic or inorganic powder) or liquid material (organic precursor or liquid in which the powder, which may be organic or inorganic, is dispersed) forming a porous support, the thickness of which is fixed over a larger area than the cross section of said porous support taken at the level of the stack; and

locally consolidating some of the material to form the pattern defined by each laminate, so as to produce a single laminate, while bonding this single laminate produced in this way to the previous one

Or by melting an organic or inorganic powder injected into the laser beam to produce a continuous string of material, thereby forming a predetermined pattern for each layer;

or by continuous or discontinuous (drop-wise) melting of a strand of hot-melt solid precursor. When the precursor is a hot-melt organic polymer used per se, the support is of organic nature and can be used directly for depositing layers of organic nature. When the precursor is a mixture of a hot-melt organic polymer and a ceramic or metallic inorganic powder, then the support is of inorganic nature after the polymer used as binder has been removed and after the particles of the inorganic powder have been sintered.

In general, in the first case, the material used is a solid or a liquid, and the individual stacks are consolidated by delivering energy or by spraying the liquid in the form of fine droplets. Energy can be delivered locally by: by using a directed beam of light (by a Light Emitting Diode (LED) or by a laser), or by using a directed beam of electrons, or even any energy source that can be focused and swept on the powder substrate in a pattern selected by the CAD. The subsequent interaction between the energy and the material causes sintering, or causes the material to melt and solidify, or even causes the material to undergo photopolymerization or photocrosslinking, depending on its nature and the nature of the energy source used.

The liquid can be delivered in a localized manner using micro-droplets produced by means of a piezoelectric system, which droplets are optionally charged and guided using an electrostatic field. The liquid should be a binder or an agent for activating a binder that has been previously added to the ceramic powder.

With the additional techniques illustrated in the context of the present invention, it is possible firstly to obtain an improvement in terms of production reliability and rate, and secondly to obtain a large range of shapes of the support body and of shapes of the relief portions that can be formed in the channels inside the support body, compared with the prior art.

In the context of the present invention, various additional techniques may be used to design the three-dimensional shape, as described below.

Selective Laser Sintering (SLS) or Selective Laser Melting (SLM)

Using this technique, powders of the material constituting the support or the tangential flow separation element, organic or preferably inorganic materials made of ceramics of the metal or oxide, nitride or carbide type, or even precursors thereof, are deposited to form a continuous substrate. Subsequently, a powerful laser beam is locally applied in a selected pattern to agglomerate the powders so as to form a stack corresponding to the support or tangential flow separation element and to bond it to the previous stack by sintering. Under the action of the local energy delivery, the powder particles are partially melted and welded together to bond the stack, thereby performing a pre-sintering of the produced shape. Thereafter, a new powder substrate is unrolled and the process begins again.

The laser beam is swept across the surface of the powder to consolidate the material layer by layer in a desired pattern. The sweeping is performed by moving the laser along a plurality of parallel paths. Advantageously, the impact area of the laser overlaps between two successive parallel paths. The amount of energy that the powder substrate receives at the impact position of the laser beam must be such that the powder particles remain partially melted or, in any case, such that each particle is sufficiently melted to be combined with its nearest neighbor, but at the same time not to close the porous structure.

The adjustment of the machine is therefore dependent in particular on the intrinsic properties of the powder substrate and on the material properties defining the effect of the interaction between photons and the material.

By way of illustration, conditions corresponding to the ranges listed in table 1 below may be used:

TABLE 1

By locally adjusting the focus of the laser beam and/or the travelling speed of the beam, the amount of energy received by the powder substrate can be adjusted, thereby adjusting the compactness of the obtained ceramic material and thus its porous structure. Thus, a desired porous structure corresponding to the filtration separation layer can be obtained at certain specific positions, and a desired porous structure corresponding to the support body can be obtained at other positions.

Although sintering is performed progressively while applying the laser to build the support or tangential flow separation element, it is advantageous to employ a final sintering step after the support or tangential flow separation element has been completely built in order to relieve residual mechanical stresses and make the porous structure more uniform. The temperature selected for such final sintering should depend on the nature of the inorganic material itself used and the average particle size of the powder used, for example when titanium oxide is used, temperatures in the range 1300 ℃ to 1500 ℃ should be used.

It should be observed that the selective melting of the powders described above can be obtained in a similar way by means of an electron beam, which corresponds to the Electron Beam Melting (EBM) technique.

3D printing

The principle remains unchanged, but the deposited stack can be made by printing to correspond to a mixture of powders, which may be organic or inorganic, ceramic or metallic, which may be the material constituting the support, or a precursor of this material and provided with a binder, which itself may be in the form of a powder or a coating on the inorganic powder itself. Preferably, the mixture is homogeneous and the powder particles of the material constituting the support or the precursor of the material and the particles of the binder all have similar particle sizes. Mention may be made, as examples of binders, of furans, phenolic resins and amino resins. The weight percentage of binder should be in the range of 1% to 25%, depending on its nature and the average diameter of the powder used. Thereafter, the agent for activating the binder is sprayed in the form of very fine droplets according to a selected pattern and the powder is locally agglomerated. The activator may be a solvent for the binder, which after almost instantaneous drying serves to bind the inorganic particles together or enclose them within a solid lattice.

It is also possible to deposit only the powders of organic or inorganic, ceramic or metallic powders or precursors thereof of the material constituting the support, so as to form a continuous substrate, and then to spray locally the binder, which should be a quick-drying liquid binder or a thermosetting liquid resin.

The sweeping (jetting) is carried out by moving the print head along a plurality of parallel paths, by using any suitable means to jet the binder or activator in liquid form, in particular a piezoelectric system of the type used in inkjet type printers. It may be advantageous that the impact area of the droplet overlaps between two consecutive parallel paths.

After removal of the unagglomerated powder, the binder is eliminated during the sintering heat treatment, and this debinding process is usually completed before 500 ℃ is reached.

In the case where the average size of the particles of the ceramic powder is in the range of 30 to 100 μm, the 3D printing may make the thickness of the powder substrate in the range of 80 to 300 μm, and enable the speed of obtaining a linear configuration of a desired shape to be in the range of 25 to 100 mm/h.

Lithography based ceramic fabrication (LCM)

LCM is a technique in which ceramic powder is premixed with photopolymerizable resin, and curing by polymerization is obtained using an LED or laser light source. As in the above-described techniques, the non-crosslinked powder must be removed, i.e., the photopolymerizable resin is eliminated, prior to the sintering thermal cycle for binder removal, followed by appropriate sintering.

The use of LCM is limited by the following factors: in order to achieve bulk polymerization below and around the point of impact of light, the powder particles must be transparent at the wavelength of interest.

Fused Deposition Modeling (FDM)

FDM is a technique that uses a hot-melt solid organic polymer, optionally added with an inorganic powder. This technique is intended to form strings of successively deposited material from a strand or strip of material. The rope is made in a continuous (extruded) or discontinuous (dropped) manner by softening or melting the end of the strand or strip of material. Unlike the techniques described above, there is no prior shaping (preforming) of the material substrate. The material stack or material string is consolidated by heating.

In one variation of this technique, the powder may be sprayed into the laser beam to melt prior to impact by spraying an inorganic powder to create a continuous string of material.

Stereolithography using a stereolithography apparatus (SLA)

This technique is similar in principle to the above-described technique, and uses a liquid material as a photocurable liquid precursor containing an inorganic powder. A photon beam (LED or laser) sweeps the liquid layer and causes it to polymerize locally.

The filter separation layer is deposited after the support is made, by 3D printing or LCM, after the final sintering operation. By depositing a suspension containing at least one sinterable composition, which after baking constitutes the filter layer, on the support, the separating layer is deposited, in particular on the surfaces of the channels and on the obstacles in the channels of the support. Such compositions have ingredients conventionally used in the production of inorganic filter membranes. The composition comprises at least one oxide, nitride, carbide or other ceramic material or mixtures thereof, preferably metal oxides, nitrides and carbides. The sinterable composition is placed in suspension (e.g., in water). In order to eliminate the risk of agglomeration (coagulation) present and to optimize the distribution of the particles in the liquid, the resulting suspension is ground to break up the agglomeration and obtain a composition essentially consisting of separate particles. The rheology of the suspension is then adjusted with organic additives to meet the hydrodynamic requirements required for penetration into the channels of the support. Once the layer has been deposited, it is dried and then sintered at a temperature that depends on its nature, the average size of its particles and the desired cut-off threshold.

With SLS or SLM, the separation filter layer can be created at the same time as the support is built up, or it can be subsequently deposited using conventional deposition methods used in membrane production. Likewise, the separation filtration layer may be deposited from a suspension of particles of the inorganic material to be deposited or a precursor thereof. Such suspensions are conventionally used for the production of ceramic filter elements. After drying, the layer or layers are each subjected to a sintering operation which serves to consolidate the layers and bond them to the surface on which they are deposited. The size of the particles present in the suspension depends on the desired porosity of the final separation filtration layer.

The invention is illustrated below by means of a number of examples, which are not limiting.

Tubular elements for tangential flow separation of the type shown in the figures are manufactured according to the invention. The support body has a tube shape having a length in the range of 300mm to 1200mm, a circular right cross section having a diameter in the range of 10mm to 42mm, and a plurality of linear channels formed therein parallel to the axis of the tube.

Example 1: SLS/support only

Example 2: SLS/support + layer

Example 3: SLS/support only

In this case, final sintering is not necessary.

Example 4: 3D printing

Material	Titanium oxide
		Average particle diameter of ceramic powder	30μm-35μm
Thickness of powder substrate	80μm
		Type of adhesive	Furan resin
Percentage of adhesive	20％
		Linear speed of formation of shape	30mm/h
Final sintering temperature	1500℃
		Residence time at 1500 deg.C	6h
Average diameter of pores obtained	10μm-12μm

In examples 1, 3 and 4, the fabrication of the tangential flow separation element was completed by depositing a separation layer on the surface of the channel using the following suspension:

preparation of the suspension by grinding in a ball mill

Material	Titanium oxide
		Average particle diameter of powder before grinding	3.6μm
Ratio of titanium oxide/water	0.4
		Grinding time	5h
Average particle diameter of the milled powder	1μm
		Adding water to adjust rheology	200 centipoise (cps) to 400cps

After performing the direct deposition on the support in the following manner, a microfiltration separation layer with a cut-off threshold of 1.4 μm was obtained.

The suspension is infiltrated into the channel by pumping so that it is in contact with the surface of the channel. The mechanism driving the deposition is: liquid from the suspension is drawn through the pores of the porous support.

The thickness of the titanium oxide particles deposited on the surface, and thus the weight per unit area, depends on the time that the suspension has elapsed in the channels of the support.

The time that the suspension passes in the channels of the support	30 seconds
		Deposition weight	50g/m2To 60g/m2

This operation was repeated twice to reach about 110g/m²Final deposition weight of (c).

Baking cycle for a sinter bed

Tangential flow microfiltration separation elements with a cut-off threshold of less than 1.4 μm as well as tangential flow ultrafiltration and nanofiltration separation elements can be manufactured by successive multiple depositions on such a first layer while using increasingly finer suspensions in connection with suitable thermal treatment cycles.

The invention is not limited to the examples described and shown, since various modifications can be made to these examples without departing from the scope of the invention.

Claims (18)

1. A monolithic tangential flow separation element for separating a fluid medium to be treated into a filtrate and a retentate, said separation element comprising a three-dimensionally structured, rectilinear, rigid porous support (2) in which at least one channel (3) is provided for the fluid medium to be treated to flow through in order to recover the filtrate from an outer surface (5) of said support, said outer surface (5) of the support having a fixed profile, and said element being characterized in that said monolithic, rigid, porous support (2) comprises, on or in the inner walls of said channel, a plurality of obstacles (9) arranged for the flow of the fluid to be filtered, said obstacles (9) having material and porosity identity with said support and continuity with said support, said obstacles (9) being continuous with a third obstacle by emerging at a first position chosen along the longitudinal axis (T) of said channel Between two positions (P1, P2) to create a sharp narrowing or a tapered nozzle portion in the passage in the direction of flow of the fluid to be treated, in order to obstruct or disturb the flow of the fluid.

2. The integral tangential flow separation element according to claim 1, characterized in that at least one channel (3) provided with obstacles (9) whose flow right section taken between its inlet (6) and its outlet (7) varies in shape and/or area and/or wetted perimeter and/or hydraulic diameter.

3. The monolithic tangential flow separation element according to claim 1 or 2, characterized in that at least one obstacle (9) of the channel (3) produces a locally narrower flow cross section at the location of said obstacle, which is perpendicular to the longitudinal axis (T) of said channel and has a shape different from the shape of the partial channels upstream and downstream of said obstacle.

4. The monolithic tangential flow separation element according to any of claims 1 to 3, characterized in that between the portion of the passage (3) upstream of the narrowest flow cross section and the narrowest flow cross section, one of the conditions chosen from: the shape, area, wetted perimeter and hydraulic diameter of the flow cross-section of the channel.

5. The monolithic tangential flow separation element of claim 3, wherein between the upstream portion of the passage located at the narrowest flow cross-section and the narrowest flow cross-section, all conditions chosen from: the shape, area, wet circumference and hydraulic diameter were all kept constant.

6. The monolithic tangential flow separation element according to any of claims 1-5, wherein at least one obstacle (9) of the channel (3) has a flow cross section perpendicular to the longitudinal axis (T) of the channel, which is rotated about the longitudinal axis of the channel between two positions taken along the longitudinal axis of the channel.

7. The monolithic tangential flow separation element according to claim 6, characterized in that the flow front section of at least one obstacle (9) of the channel (3) rotates in a discontinuous manner around the longitudinal axis of the channel between its ends.

8. The monolithic tangential flow separation element according to any of claims 1-7, characterized in that it comprises at least one separation layer (4), said separation layer (4) being continuously deposited on the inner wall (3) of said channel (3)₁) And completely covers the obstacle (9).

9. The monolithic tangential flow separation element according to any of claims 1 to 8, characterized in that said porous support (2) is made of an organic or inorganic material.

10. The monolithic tangential flow separation element according to claim 8 or 9, wherein the separation layer or intermediate layer is made of an organic or inorganic material.

11. The monolithic tangential flow separation element according to any of claims 1 to 10, characterized in that the three-dimensional structure of the porous support (2) has a stack of different materials observable by optical microscopy or by scanning electron microscopy.

12. A method of manufacturing an integral tangential flow separation element (1) according to any one of claims 1-11, wherein the three-dimensional structure of the support body is made by forming a plurality of individual stacks, which are stacked and successively joined to each other, so as to build up progressively the desired three-dimensional shape.

13. The method of claim 12, comprising making the three-dimensional structure by repeating the steps of:

-making a continuous substrate of material for forming a porous support, said substrate having a fixed thickness over an area greater than the cross-section at said lamination of said porous support;

locally consolidating a portion of said material to form the pattern determined for each laminate, so as to produce a single laminate, while at the same time bonding the single laminate produced in this way to the previous one.

14. A method of manufacturing as claimed in claim 13, comprising manufacturing a continuous substrate of solid material in powder form or liquid material such as a photopolymerizable resin.

15. A method of manufacturing according to any one of claims 12-14, characterized in that the method comprises preparing a continuous substrate of solid material in the form of an organic or inorganic powder.

16. A method of manufacturing according to claim 12, comprising preparing a continuous substrate of the medium in the form of a photopolymerizable liquid precursor in which the inorganic powder is disposed.

17. A manufacturing process according to claim 12, characterized in that each stack is made by continuously or discontinuously melting a strand of hot-melt solid precursor, either a hot-melt organic polymer used as such for the preparation of organic supports and organic layers, or a mixture of hot-melt organic polymer and ceramic inorganic powder used for the preparation of supports of inorganic nature.

18. A method of manufacturing according to claim 12, comprising producing the continuous string of material by spraying a powder melted in a laser beam.