PARTICLE FILTER DEVICES

A device for reducing pollution of an internal combustion engine, including a collection of monolithic elements of honeycomb type connected by a jointing compound, each element incorporating a set of adjacent cells of mutually parallel axes separated by porous walls, which cells are plugged by plugs at one or other of their ends to delimit inlet chambers opening onto a gas intake face and outlet chambers opening onto a gas discharge face such that gas that is to be filtered passes through the porous walls, the collection being inserted in a metal casing by a compacted fibrous mat. A jointing compound has a three-point flexural modulus of rupture of between 0.5 and 6 MPa, the jointing compound has a dynamic Young's modulus less than or equal to 17 GPa, the mat has a mean density in the compacted state of between 0.30 and 0.54, and the mean thickness of the mat in the compacted state is between 2 and 8 mm.

Skip to: Description  ·  Claims  · Patent History  ·  Patent History
Description

The present invention relates to the field of devices for filtering the particles from an internal combustion engine, possibly comprising a catalytic component, particularly installed in an exhaust line of a diesel engine to remove the soot produced by the burning of the fuel.

Diesel engines are known to produce a large amount of soot. This is the result of phenomena of pyrolysis of hydrocarbon in the absence of oxygen actually within the combustion flame, and from insufficient temperature within the combustion chamber for all of the soot particles thus produced to be burned off. This soot, when emitted from the vehicle, acts as seeds on which unburnt hydrocarbons condense, thus forming solid particles that can be inhaled and the small size of which allows them to progress as far as the pulmonary alveoli.

In order to limit the emission of soot out of the vehicle and to meet ever tighter environmental emissions standards, it is known practice for filtration devices, possibly associated with catalytic devices, to be positioned in the exhaust line, these catalytic devices having the purpose of converting the pollutant gaseous emissions into inert gases. Featuring strongly among the pollutant gaseous emissions are the unburnt hydrocarbons together with oxides of nitrogen (NOx) or carbon monoxide (CO).

Soot filtration devices comprise “particulate filters” which generally consist of a filtering support made of porous ceramic. This support generally has a honeycomb structure, one of the faces of said structure admitting the exhaust gases that are to be filtered and the other face discharging the filtered exhaust gases. Between these faces, the filtering structure has a set of longitudinal and mutually parallel cells separated by porous walls, said cells being plugged at one of their ends to force the exhaust gases to pass through said porous walls. To ensure that the entity is correctly sealed, the peripheral part of the structure is surrounded with a cement known as a coating cement. The filter is also placed in a can, this often being known as “canning” consisting of a fibrous mat and of a metal casing. In order to afford better resistance to thermal shock, the filters are sometimes made up of a collection of monolithic and parallelepipedal elements of honeycomb structure, said elements being assembled using a material known as a “jointing compound”.

The ceramics most often used are cordierite (Mg2Al4Si2O18) or silicon carbide (SiC), the latter being preferred for its thermal conductivity and corrosion-resistance properties. Silicon carbide filters are preferably obtained by sintering, for example SiC filters connected by sintered silicon or those obtained by recrystallization (R—SiC). Examples of filters are described for example in patent applications EP 816 065, EP 1 142 619, EP 1 455 923 or alternatively WO 2004/065088 to which reference may be made for greater details regarding their structure or method of synthesis.

During engine operation, the particulate filter becomes laden with soot particles which are deposited on the porous walls. The problem of the minimum temperature needed to burn off the soot arises here just as it does in the combustion chamber. Because the soot is held in the filter, the combustion dynamics may be slower than in the combustion chamber, making it possible to lower the temperature at which the soot burns off to about 600° C. However, this reduction is not enough to ensure that the soot is burnt off within the filter throughout the entire engine operating range. It is therefore necessary, after a filtration cycle, to provide a regeneration cycle during which the soot is burnt.

The particulate filter therefore operates in the following modes:

    • filtration and quasi-simultaneous combustion of soot when the temperature of the exhaust gases so permits,
    • retention and accumulation of soot particles in the filter when the temperature of the exhaust gases is too low,
    • regeneration of the filter before the pressure drops caused by the build-up of soot become unacceptable.

The progressive plugging of the filter during the soot retention phase does in fact lead to an increase in pressure drop resulting in an increase in engine fuel consumption, or even a raised back pressure which may damage the combustion system.

The regeneration step is performed by raising the temperature of the exhaust gases using post injection, which consists in a late injection in the engine cycle of fuel which will burn in the exhaust line.

During regeneration, and because of the exothermic combustion of soot, the filter experiences high temperatures, and what is more, it experiences temperatures that are not uniform within the material, because the particles of soot prefer to deposit themselves in the central part of the filter and in the downstream part thereof. The filter is therefore subjected to intense radial and tangential thermomechanical stresses liable to give rise to microcracks within the material leading to a partial or even complete loss of its filtration capability.

In general, filter improvements involve obtaining the best possible compromise between the following properties for equivalent engine speeds. In particular, it is an object of the invention to provide a filtration device formed of an assembly of monolithic elements which simultaneously exhibits:

    • a low pressure drop caused by a structure that is filtering in operation, that is to say typically when this structure is in the exhaust line of an internal combustion engine both when said structure is free of soot and when it laden with particles,
    • a high soot storage volume so as to reduce the frequency of regeneration periods,
    • a filter mass best suited to ensure sufficient thermal mass so that the maximum regeneration temperature and the gradients experienced by the filter can be minimized,
    • a high thermomechanical strength, that is to say one that gives the filtration device an extended life.

The performance of the filtering devices comprising a filter inserted in a metal casing by means of a fibrous mat is, for its part, characterized by the following properties:

    • the mechanical integrity of the device: the individual monolithic elements of the filter, the fibrous mat and the metal casing have to remain joined together after the device has been subjected to vibrations, particularly to vibrations representative of those experienced by such a device in an exhaust line of an engine. Insufficient mechanical integrity may manifest itself in disconnection of the fibrous mat and of the filter or of the mat with respect to the metal casing, or alternatively in disconnection of one or more monolithic elements of an assembled filter.
    • sealing against the hot gases that are to be filtered: the passage of soot through the mat, between the mat and the filter or between the mat and the metal casing need to be avoided.

It would seem important to be able to obtain a device that is able to solve all of the aforementioned problems, particularly a device that has improved thermomechanical strength and improved mechanical integrity.

The inventors have discovered the key parameters that are necessary and sufficient for obtaining such a device.

In its most general form, the subject of the present invention is a device for reducing the pollution of an internal combustion engine, comprising a collection of monolithic elements of the honeycomb type connected by a jointing compound, each element incorporating a set of adjacent cells of mutually parallel axes separated by porous walls, which cells are plugged by plugs at one or other of their ends to delimit inlet chambers opening onto a gas intake face and outlet chambers opening onto a gas discharge face such that the gas that is to be filtered passes through the porous walls, said collection being inserted in a metal casing by means of a compacted fibrous mat. The device according to the invention is characterized in that:

    • the jointing compound has a three-point flexural modulus of rupture of between 0.5 and 6 MPa, preferably between 1 and 5 MPa, particularly between 2 and 4 MPa,
    • the jointing compound has a dynamic Young's modulus less than or equal to 17 GPa, preferably less than or equal to 10 GPa,
    • the mat has a mean density in the compacted state of between 0.30 and 0.54, preferably less than or equal to 0.50,
    • the mean thickness of the mat in the compacted state is between 2 and 8 mm.

It is in fact thanks to a careful combination of these various parameters that the filtration device according to the invention is able to solve the various abovementioned problems.

The porous walls are preferably made of a ceramic material, typically made of cordierite (Mg2Al4Si2O18), of aluminum titanate, or based on silicon carbide (SiC), the latter being preferred for its thermal conductivity and corrosion resistance properties. What is meant within the meaning of the present description by “material based on SiC” is that said material contains at least 30 wt % of SiC, preferably at least 70 wt % of SiC and as an extreme preference at least 98 wt % of SiC.

The material of which the walls are made preferably has an open porosity of between 35 and 65%, and more preferably still of between 40% and 60%. Particularly in an application to a particulate filter, too low a porosity leads to too high a pressure drop. Too high a porosity by contrast leads to too low a mechanical strength. The median diameter d50, by volume, of the pores constituting the porosity of the material preferably ranges between 5 and 25 microns, particularly between 10 and 30 microns. In general, in the intended applications, it is generally accepted that too low a pore diameter leads to too high a pressure drop, whereas too high a median pore diameter leads to poor filtration efficiency.

In general, the cross section of a monolithic element that makes up the assembled structure is square, the width of the element ranging between 30 mm and 50 mm. Advantageously, the thickness of the walls ranges between 200 and 500 μm. The number of cells in the filtering elements preferably ranges between 7.75 and 62 per cm2, said cells having a cross section of about 0.5 to 9 mm2. The cells may have various shapes. They may have identical or different shapes and sizes, particularly being of square, hexagonal, octagonal or triangular shape. The cells may, for example, all be square and of the same size. They may also, for example, alternate between square and hexagonal and square and octagonal shapes. The cells may also have more complex shapes associated with a corrugation of the walls, as described for example in application WO 05/016491.

The filters are preferably such that the total volume of the inlet chambers opening onto the gas inlet face is greater than the total volume of the outlet chambers opening onto the gas discharge face. For example, the inlet cells may be more numerous than the outlet cells (particularly if the inlet cells and outlet cells all have the same cross-sectional area) and/or the inlet cells may have a larger cross-sectional area than the outlet cells (particularly if the number of inlet cells is equal to the number of outlet cells). What is meant by inlet cells and outlet cells respectively, are the cells open onto the inlet face and onto the discharge face for the gases, respectively. Such filters, known as asymmetric filters, have the advantage that they are able to store a greater quantity of soot, making it possible to lengthen the time between two successive regenerations and to reduce the increase in pressure drop as the filter becomes laden with soot. Implementation of the invention has proven to be particularly advantageous in the case of such filters because the inventors have been able to demonstrate that such filters were more liable to be affected by higher thermomechanical stresses than standard filters.

The mean thickness of the jointing compound is preferably between 0.5 and 4 mm, particularly at least 1 mm. For small thicknesses, the mechanical integrity of the filter is poor and the spread on flatness of the monolithic elements may then generate local thermomechanical stresses and reduce the relaxation of stresses by the jointing compound. If the thickness is too high, the pressure drop of the filter becomes too great, especially when there are a great many monolithic elements, that is to say when the number of joints across the cross section of the filter perpendicular to the axis of the filter is high.

The jointing compound is understood here to mean a moldable composition formed of a wet or dry particulate and/or fibrous mix, able to set solid and to have sufficient mechanical strength at ambient temperature or after drying and/or heat treatment the temperature of which will not exceed the softening or collapse temperature which defines the refractoriness of the material or materials of which the monolithic elements are made.

What is meant by “moldable” is a composition capable of plastic deformation needed to spread over the face of the joint of the monolithic elements and which exhibits sufficient adhesion with respect to these elements so that it can hold them together or allow the filter to be handled in its assembled state immediately after the jointing operation or, where necessary, after a heat treatment or chemical treatment or some other treatment such as ultraviolet irradiation.

The jointing compound preferably contains particles and/or fibers of ceramic or of refractory material, chosen from non-oxides, such as SiC, aluminum and/or silicon nitride, aluminum oxynitride, or from among oxides, particularly including Al2O3, SiO2, Cr2O3, MgO, ZrO2, or any mixture thereof.

For preference, the composition contains at least 20% SiC. To encourage it to harden, the jointing compound preferably contains a thermosetting resin, in a quantity of at least 0.05 wt % and at most 5 wt % with respect to the mineral filler. A catalytic hardener intended to accelerate the setting of the resin, preferably also in the form of a powder, may be added to the mixture. The jointing compound may contain clay to encourage plasticity and its moldable nature. The jointing compound may also contain inorganic fibers and organic and/or inorganic binders. What is meant by an organic binder is, in particular, temporary binders such as derivatives of cellulose or of lignin, such as carboxymethylcelluloses, dextrin or alternatively polyvinyl alcohols. What is meant by inorganic binders is, in particular, chemical setting agents such as phosphoric acid, aluminum monophosphate or sols based on silica and/or on alumina and/or on zirconia or possibly sinter promoters such as titanium dioxide or magnesium hydroxide, or even shaping agents such as calcium stearate or magnesium stearate. The jointing compound is preferably a ceramic and/or refractory cement.

For preference, the filtering monolithic elements are based on SiC and are assembled by a jointing compound the thermal conductivity of which is greater than or equal to 0.1 W/m·K for all temperatures between 20 and 800° C. A high thermal conductivity of the jointing compound advantageously makes it possible to even out the heat transfers within the filter whereas a low thermal conductivity, particularly one of below 0.1 W/m·K (typically measured at a temperature of 600° C.) contributes to increasing the temperature gradients and the thermomechanical stresses in the joint and within the filter.

The monolithic elements are preferably assembled by partial bonding, inasmuch as the space between the monolithic elements may be not completely filled by the jointing compound, so as to relax the thermomechanical stresses in the filter, as described for example in applications EP 1 726 800 or FR 2 833 857. Jointing compound configurations like those described in applications WO 2005/084782 or WO 2004/090294, which involve regions of low or zero adhesion between the jointing compound and the filtering element and regions of strong adhesion between the jointing compound and the filtering element are also conceivable.

The assembled filter preferably has a coating cement secured to the assembled filter, particularly having the same mineral composition as the jointing compound, so as to reduce thermomechanical stresses.

The pollution-reducing device may further comprise a catalytic coating to treat pollutant gases of the CO or HC and/or NOx type.

The fibrous mat is preferably formed of inorganic fibers so as to afford the thermal insulation properties required for this application. The inorganic fibers are preferably ceramic fibers, such as fibers of alumina, of mullite, of zirconia, of titanium oxide, of silica, of silicon carbide or nitride, or alternatively glass fibers, for example glass R fibers. These fibers may be obtained by fiber drawing from a bath of molten oxides, or from a solution of organo-metallic precursors (the sol-gel method). The fibrous mat is preferably non-intumescent. It is advantageously in the form of a needled felt.

The density of the mat in the compacted state is dependent in particular on the mass per unit volume of the material of which this mat is made prior to compaction and on the thickness of the mat after compaction. Mats capable of exhibiting the required densities in the compacted state are, for example, marketed by the company Saffil Ltd, under the references 1600, 1250 or 2400 or alternatively by the company Ibiden Co., Ltd, under the references N4-1515 or N4-1253.

Particularly in the case of filters of non-circular cross section, the density in the compacted state and/or the thickness of the mat is advantageously non-uniform, in as far as it may vary according to the region of the space formed between the filter and the metal casing. During regeneration, this type of filter is in fact liable to exhibit a non-uniformity of temperature at its periphery. The difference between the temperature of certain regions of the periphery of the filter and the temperature at the center of the filter may thus be 20% or more greater than the mean difference between the peripheral temperature and the temperature at the center of the filter, and this non-uniformity of temperature is likely to give rise to high very localized stress concentrations in these regions. In order to achieve a more favorable distribution of thermomechanical stresses, the density of the mat is therefore preferably lower than the mean density and/or the thickness of the mat is preferably higher than the mean thickness in contact with the regions at which the thermomechanical stresses may become concentrated during the regeneration phases. In order best to optimize the overall thermomechanical strength of the filtration device according to the invention, the thickness of the mat in the compacted state at the peripheral regions of the filter which are subjected to the highest thermomechanical stresses is preferably at least 20%, particularly at least 50%, and even at least 100% higher than the thickness of the mat at the peripheral regions subjected to the lowest thermomechanical stresses. Alternatively or in combination, the density of the mat in the compacted state at the peripheral regions of the filter subjected to the lowest stresses is preferably at least 20%, particularly at least 50% and even at least 100% higher than the density at the regions subjected to the highest stresses. The method of the “shrinking” type (where the metal casing is shrunk around the mat) allows the density and/or the thickness of the mat to be modified, creating regions of lower density and/or greater thickness in these regions that are liable to be most highly affected by this stress concentration. In the case of an ellipsoidal or substantially ellipsoidal filter, it is particularly preferable for the density of the mat to be lower and/or for the thickness of the mat to be higher at the ends of the minor and of the major axes of the ellipse, these ends being the most highly thermomechanically stressed during regeneration.

When the density in the compacted state is non-uniform, the measurement taken corresponds to a mean value.

The mean thickness of the fibrous mat, in the compacted state, is determined on the filter placed in its metal casing, by calculating the mean of 4 thickness measurements taken in a plane perpendicular to the axis of the filter on 4 segments of two mutually perpendicular straight lines passing through the geometric center of the filter.

The density of the mat in the compacted state may be measured as follows: the filter surrounded by its mat is taken out of its metal casing then unrolled so that its surface area can be measured, and weighed so as to measure its relative density in g/cm2. The density in the compacted state is obtained by dividing the previously determined relative density by the mean thickness of the mat in cm.

Insertion into the metal casing can be done using various methods known to those skilled in the art. Mention may be made in particular of the methods known as the “tourniquet” method, the “shrinking” method, the “clamshell” method or the “stuffing” method.

The modulus of rupture of the jointing compound is measured at ambient temperature on a test specimen measuring 150×25×25 mm3. The setup for 3-point flexural testing in accordance with the standard NF B41-104 is performed with a distance of 120 mm between the two lower supports and the rate of descent of the loading plunger is equal to 0.5 mm/min. The value is a mean calculated from three successive measurements.

The dynamic Young's modulus is measured, in accordance with standard ASTM C1259-01, on test specimens of the same dimensions as those used previously, using test apparatus marketed under the reference Grindosonic MK5 by the company J.W. Lemmens. The dynamic Young's modulus is determined by measuring the natural frequency of flexural vibration at ambient temperature of a test specimen of the jointing compound in so-called “dynamic” mode. The test specimen is placed on two supports of the rubbery type, so as not to interact with the vibration mode of the test specimen being tested. The supports are positioned symmetrically with respect to the center mid-way along the test specimen. The distance between supports is 100 mm. The test specimen is excited by a mechanical impulse as close as possible to its center on its upper face the opposite face to the face resting on the supports, for example using a stick or a pencil or a small hammer supplied with the apparatus, because the excitation energy needed is small. This excitation leads to vibrations within the material of the test specimen. A piezoelectric detector positioned in contact with the test specimen then records these vibrations and converts them into an electrical signal from which the natural frequency of vibration is displayed.

The dynamic Young's modulus E is then calculated (in GPa) as a function of the mass m (in g) of the test specimen and of the flexural resonant frequency f (in Hz) using the following formula:


E=9.1584×10−9×m×f2

All the measurements (density, thickness, modulus of rupture and Young's modulus) are taken at ambient temperature.

To measure the moduli, the test specimen of jointing compound is prepared by molding the composition, and it then undergoes the same treatment (for example a heat treatment) as is undergone by the jointing compound when used to assemble the monolithic elements with one another, finally being dried at 110° C. before being cooled to ambient temperature.

The invention and its advantages will be better understood from reading the examples which follow. Of course, these examples must not be considered, in any of the aspects described, as limiting the present invention.

FIGS. 1 and 2 schematically depict non-circular filters 1 formed of a plurality of elements 2. The hatched regions 3 represent the peripheral regions in which the difference in temperature by comparison with the temperature at the center of the filter is likely, during regeneration, to be 20% or more greater than the mean difference between the peripheral temperature and the temperature at the center of the filter. This non-uniformity of temperature is likely to give rise to high very localized stress concentrations in these regions. It is therefore advantageous for the density of the mat to be lower near this region where the thickness of the mat is greater.

 EXEMPLARY EMBODIMENTS

In the examples which follow, a series of filtering devices according to the invention and which illustrate its advantages over another series of devices given for comparison purposes and which do not meet the criteria of the invention, were created.

All the monolithic filtering elements were created using the following method.

Using a mixer, powders of silicon carbide, a pore-generating agent of the polyethylene type and an organic binder of methylcellulose type were first of all mixed. Water was added and mixing was continued until a uniform paste was obtained with a plasticity that allowed extrusion through a square section honeycomb monolithic structure die the dimensional characteristics of which are given in table 1.

The raw elements obtained were then dried using microwaves for long enough to bring the chemically unbound water content down to under 1 wt %.

The cells on each face of the blocks were then alternately plugged using well known techniques, for example described in application WO 2004/065088.

The elements were then baked at an increase in temperature of 20° C./h until a temperature of the order of 2200° C. was obtained, this temperature then being maintained for 2 hours.

This finally yielded a series of monolithic filtering elements made of silicon carbide, the microstructural characteristics of which were substantially identical.

TABLE 1 Cell geometry Square Cell density 180 cpsi (cells per square inch, 1 inch = 2.54 cm) Wall thickness 350 μm Length 15.2 cm Width 3.6 cm Porosity About 47% Median pore diameter About 15 μm

According to the teaching of patent application EP 816 065, 16 filtering monolithic elements were then assembled with one another by bonding using a jointing compound of ceramic nature and were then machined to form filters of a suitable diameter. The thickness of the jointing compound was 1 mm.

In the case of comparative examples C1 to C3 and of the example according to the invention 1, the jointing compound was prepared by mixing the compound J1:

    • 81 wt % of a powder of SiC with a particle size of between 10 and 200 μm,
    • 4 wt % of a powder of calcined alumina the median diameter of which was about 5 microns, marketed by the company Almatis,
    • 8 wt % of a powder of reactive alumina the median diameter of which was about 3 microns, marketed by the company Almatis,
    • 6% of silica fume of the Elkem 971 type,
    • 0.8 wt % of a temporary and plasticizing binder of the cellulose type,
    • 0.2 wt % of a deflocculant of the STPP (sodium tripolyphosphate) type.

A quantity of water corresponding to about 15% of the weight of this mixture was added in order to obtain a paste of suitable viscosity.

Once the filter had been machined, a coating cement of the same mineral composition as used for the jointing compound was applied to the cylindrically shaped filters with a volume of the order of 2.48 liters. The assembled filter was then subjected to a heat treatment in air at 750° C. with the maximum temperature sustained for 2 h. In the case of comparative example C3, the heat treatment was performed at a temperature of 950° C. instead of 750° C., which had the effect of increasing the modulus of rupture and Young's modulus of the jointing compound.

In the case of comparative example C4 and of the examples according to the invention 2 and 3, the jointing compound was prepared by mixing the following compound J2:

    • 67 wt % of a powder of SiC with a particle size ranging between 10 and 200 μm,
    • 3 wt % of a powder of reactive alumina marketed by the company Almatis, the median diameter of which was about 3 microns,
    • 24% of hollow spheres marketed by Enviro-spheres under the name “e-spheres” which have a typical chemical composition containing 60% SiO2 and 40% Al2O3 and a median diameter of the order of 100 μm,
    • 6% of silica fume of the Elkem 971 type,
    • 0.8 wt % of a temporary and plasticizing binder of the cellulose type,
    • 0.2 wt % of a deflocculant of the STPP (sodium tripolyphosphate) type.

A quantity of water corresponding to about 15% of the weight of this mixture was added in order to obtain a paste of suitable viscosity.

Once the filter had been machined, a coating cement of the same mineral composition as used for the jointing compound was applied to the cylindrically shaped filters with a volume of the order of 2.48 liters. The assembled filter was then subjected to a heat treatment in air at 950° C. with the maximum temperature sustained for 2 h.

The filters were then coated with various fibrous mats then inserted in their metal casing in accordance with the teaching associated with FIG. 5 of patent application EP 1 382 374 (the so-called “tourniquet” method) in order to obtain densities in the compacted state and mat thicknesses in the compacted state as collated in table 2.

The metal casing was made up of two parts formed of 13% chromium refractory stainless steel sheets 1.5 mm thick.

The devices thus obtained were subjected to the following characterization tests.

A) Thermomechanical Strength Test

The devices were mounted on an exhaust line of a 2.0 l direct-injection diesel engine run at full power (4000 rpm) for 30 minutes then removed and weighed in order to determine their initial mass. The devices were then re-fitted on the engine test bed with an engine speed of 3000 rpm and a torque of 50 Nm for various lengths of time in order to obtain a soot loading of 8 g/liter (in terms of filter volume). The devices thus laden with soot were re-fitted on the line to undergo severe regeneration defined as follows: after stabilizing at an engine speed of 1700 rpm for a torque of 95 Nm for 2 minutes, post-injection was performed with a 70° phase angle for a post-injection delivery of 18 mm3/shot. Once combustion of the soot was initiated, more specifically once the pressure drop had decreased for at least 4 seconds, the engine speed was lowered to 1050 rpm for a torque of 40 Nm for 5 minutes in order to accelerate the combustion of the soot. The devices were then subjected to an engine speed of 4000 rpm for 30 minutes in order to eliminate the remaining soot.

The regenerated filters were inspected after slicing to reveal the presence of any cracks there might be that were visible to the naked eye. The thermomechanical strength of the filter was assessed in the light of the number of cracks, a low number of cracks meaning a thermomechanical strength acceptable for use as a particulate filter.

As collated in table 2, the following scores were assigned to each of the filters:

    • +++: very numerous cracks present,
    • ++: numerous cracks present,
    • +: a few cracks present,
    • −: no cracks or very rare cracks.

Because the severe regeneration was characterized by particularly extreme conditions, the presence of a few cracks (the score “+”) is acceptable. Scores “++” and “+++” by contrast are representative of poor thermo-mechanical strength.

B) Method of Assessing Integrity

The device comprising the filter with its metal casing and its fibrous mat was placed on an electrodynamic test rig equipped with accelerometers positioned at various points. A first accelerometer was placed in contact with the filter at the center of one of the planar faces thereof, a second accelerometer being positioned on the metal casing of the canning. These two accelerometers, which were at least two-axes accelerometers, are able to measure vibration in the direction of the axis of the filter and radial vibrations and any decoupling between the filter and its canning and to monitor the stability of the attachment of the canned filter to the electrodynamic test rig. The filter was subjected to a cycle of vibration at a frequency of 185 Hz comprising successive 15-minute levels each corresponding to a given acceleration. The first level corresponded to an acceleration of 5 G, the second to an acceleration of 10 G, the acceleration then being increased in steps of 10 G for each successive level. This vibration test can be carried out on an electrodynamic test rig marketed by the company LDS Test and Measurement LLC, with a capacity of 35 kN and equipped with a hydraulic ram with a maximum force of 10 kN operating in the frequency range 0-500 Hz and a 200 bar hydraulic system with a flow rate of 21 l/min.

The device was then subjected to a filtration efficiency test. The filtration efficiency of the filtering device after vibration test was determined by measuring the amount of smoke emitted at the outlet of the filter by comparison with the quantity at the inlet. To do this, a smoke meter was positioned upstream and downstream of the filtering device, the latter being positioned on an exhaust line of a diesel engine. The smoke meter made it possible to determine the amount of particles of soot emitted by measuring the blackening due to the smoke. During the measurement, the engine was preferably set at its operating point corresponding to its maximum power. If the filtering device has sufficient integrity characteristics, the filtration efficiency index needs to remain higher than 85%.

Table 2 below sets out, for comparative examples C1 to C4 and examples according to the invention 1 to 3, the following properties:

    • the nature of the jointing compound (J1 or J2, using the coding given hereinabove),
    • the temperature (in ° C.) and the length (in hours) of the heat treatment after assembly,
    • the modulus of rupture, termed “MOR”, measured using the method described hereinabove, and expressed in MPa,
    • the dynamic Young's modulus, termed “MOE”, measured according to the method described hereinabove, and expressed in GPa,
    • the density of the mat in the compacted state, measured according to the method described hereinabove,
    • the mean thickness of the mat in the compacted state, measured according to the method described hereinabove, and expressed in mm,
    • the results of the thermomechanical strength test,
    • the results of the test of integrity after vibration: the sign “X” means that the integrity of the filter was not affected by the test, and the symbol “O” means the opposite,
    • the efficiency of the filter after the integrity test, expressed in %.

TABLE 2 C1 C2 C3 C4 1 2 3 Jointing J1 J1 J1 J2 J1 J2 J2 compound Heat treatment 750° C. 750° C. 950° C. 950° C. 750° C. 950° C. 950° C. after assembly 2 h 2 h 2 h 2 h 2 h 2 h 2 h MOR (MPa) 3 3 8 3 3 3 3 MOE (GPa) 15 15 20 6 15 6 6 Mat density 0.56 0.35 0.35 0.2 0.35 0.52 0.35 Mat thickness 6 1.5 6 3 6 3 6 (mm) Thermomechanical +++ ++ +++ + + + − strength Vibration test X X X ◯ X X X Post-vibration >85% >85% >85% <65% >85% >85% >85% filtration efficiency

The various examples and comparative examples illustrate the fact that the four essential characteristics of the invention have to be met simultaneously, and therefore in combination, in order to solve all of the aforementioned technical problems. Choosing too high a density for the mat in the compacted state (example C1) leads to too low a thermomechanical strength as illustrated by the presence of very numerous cracks after severe regeneration, in spite of the choice of a jointing compound with a low modulus of rupture and Young's modulus. The same is true when the thickness of the mat in the compacted state is too small (example C2) in spite of the choice of a suitable density. Conversely, too low a density (example C4) is detrimental to obtaining good integrity: the filter obtained is unable to withstand high vibrations, causing the various elements of the device to become detached and leading to a significant drop in filtration efficiency.

Finally, choosing a suitable mat thickness and a suitable mat density does not make the filter acceptable in terms of thermomechanical strength if on the other hand, the Young's modulus and modulus of rupture of the jointing compound are too high (example C3). As illustrated by the examples according to the invention 1 to 3, it is indeed the combination, firstly of a mat of suitable density and thickness and secondly of a jointing compound that has suitable Young's modulus and suitable modulus of rupture that makes it possible to obtain a truly high-performance filtration device.

The foregoing description illustrates a few possible embodiments of the invention. Of course, this description is not however limiting and the person skilled in the art will be able to devise other variants of the invention without thereby departing from the scope thereof.

Claims

1-11. (canceled)

12. A device for reducing pollution of an internal combustion engine, comprising:

a collection of monolithic elements of honeycomb type connected by a jointing compound, each element incorporating a set of adjacent cells of mutually parallel axes separated by porous walls, which cells are plugged by plugs at one or other of their ends to delimit inlet chambers opening onto a gas intake face and outlet chambers opening onto a gas discharge face such that gas that is to be filtered passes through the porous walls, the collection being inserted in a metal casing by a compacted fibrous mat,
wherein the jointing compound has a three-point flexural modulus of rupture of between 0.5 and 6 MPa,
wherein the jointing compound has a dynamic Young's modulus less than or equal to 17 GPa,
wherein the mat has a mean density in a compacted state of between 0.30 and 0.54,
wherein the mean thickness of the mat in the compacted state is between 2 and 8 mm.

13. The device as claimed in claim 12, wherein the modulus of rupture is between 1 and 5 MPa, or is between 2 and 4 MPa.

14. The device as claimed in claim 12, wherein the dynamic Young's modulus is less than or equal to 10 GPa.

15. The device as claimed in claim 12, wherein the mean density in the compacted state is less than or equal to 0.50.

16. The device as claimed in claim 12, wherein the porous walls are made of a ceramic material based on silicon carbide (SiC).

17. The device as claimed in claim 16, wherein thermal conductivity of the jointing compound is greater than or equal to 0.1 W/m·K between 20 and 800° C.

18. The device as claimed in claim 12, wherein the mean thickness of the jointing compound is between 0.5 and 4 mm.

19. The device as claimed in claim 12, wherein the monolithic elements are assembled by partial bonding.

20. The device as claimed in claim 12, wherein the density in the compacted state and/or the thickness of the mat is non-uniform, such that the density of the mat is lower than the mean density and/or the thickness of the mat is higher than the mean thickness in contact with regions where thermomechanical stresses may become concentrated during regeneration phases.

21. The device as claimed in claim 20, wherein the thickness of the mat in the compacted state at peripheral regions of the filter subjected to a highest thermomechanical stresses is at least 20%, or is at least 50%, or at least 100% higher than the thickness of the mat at peripheral regions subjected to lowest thermomechanical stresses, and/or the density of the mat in the compacted state at the peripheral regions of the filter subjected to the lowest stresses is at least 20%, or is at least 50%, or at least 100% higher than the density in the regions subjected to the highest stresses.

22. The device as claimed in claim 12, further comprising a catalytic coating to treat polluting gases of CO or HC and/or NOx type.

Patent History
Publication number: 20110167806
Type: Application
Filed: Oct 5, 2009
Publication Date: Jul 14, 2011
Applicant: Saint-Gobain Centre De Rech Et D'Etudes Europeen (Courbevoie)
Inventors: Philippe Auroy (Gif Sur Yvette), Anthony Briot (Saint Saturnin les Avignon), David Pinturaud (Isle sur la Sorgue)
Application Number: 13/120,242
Classifications
Current U.S. Class: Reactor Plus A Washer, Sorber Or Mechanical Separator (60/297); By Sorber Or Mechanical Separator (60/311)
International Classification: F01N 3/035 (20060101); F01N 3/02 (20060101);