Porous alpha-SiC-containing shaped body having a contiguous open pore structure

Abstract

The present invention relates to a porous alpha-SiC-containing shaped body with a gas-permeable, open-pored pore structure comprising platelet-shaped crystallites which are connected to form an interconnected, continuous skeletal structure, wherein the skeletal structure consists of more than 80 wt.-% alpha-SiC, relative to the total weight of SiC, a process for producing same and its use as a filter component.

Claims

1. A filter including at least one porous, open-pored, gas-permeable alpha-silicon carbide (SiC)-containing shaped body comprising: platelet-shaped crystallites connected to form an interconnected, continuous skeletal structure, wherein the skeletal structure consists of more than 80 wt.-% alpha-SiC, relative to the total weight of SiC, a porous, open-pored, gas-permeable alpha-silicon carbide (SiC) containing shaped body having an average pore diameter of 3 m to 50 m and an open porosity of 45% to 85% vol.-%, relative to a total volume of the porous, open-pored, gas-permeable alpha-silicon carbide (SiC) containing shaped body; and wherein the average pore diameter of the porous, open-pored, gas-permeable alpha-silicon carbide (SiC) containing shaped body is greater than a size of the platelet-shaped crystallites and greater than an average thickness of pore walls constructed from said platelet-shaped crystallites, wherein the shaped body is formed as a honeycomb filter element in which the inflow and outflow openings of the flow-through channels are sealed alternately.

2. The filter according to claim 1, wherein the porous, open-pored, gas-permeable alpha-silicon carbide (SiC)-containing shaped body further comprises flow-through channels which are arranged parallel to each other and have a wall thickness in a range of from 100 m to 580 m.

3. The filter according to claim 1, wherein the average pore diameter of the shaped body is 5 m to 50 m, and the open porosity is 50% to 85% vol.-%, relative to the total volume of the shaped body.

4. The filter according to claim 1, further comprising at least one ternary carbide of silicon, carbon and at least one element from main group 3 of the periodic table of the elements.

5. The filter according to claim 4, wherein the at least one ternary carbide of silicon, carbon and at least one element from main group 3 of the periodic table of the elements contains aluminum.

6. The filter according to claim 4, wherein the at least one ternary carbide is a carbide of the general empirical formula Al.sub.4C.sub.3(SiC).sub.x, wherein x is a whole number from 1 to 4, or a mixture thereof.

7. The filter according to claim 4, wherein the at least one ternary carbide is 0.1 wt.-% to 10 wt.-%, relative to the total weight of SiC.

8. The filter according to claim 1, wherein the alpha-SiC is at least partially present in at least one of 2H, 4H, 6H, 8H, 10H, 14H and 15R polytype.

9. The filter according to claim 1, further comprising at least one of nitrogen and oxygen in an amount-of-substance fraction of less than 5 at.-%, relative to the sum of all constituents of the porous, open-pored, gas-permeable alpha-silicon carbide (SiC) containing shaped body.

10. The filter according to claim 4, wherein the at least one ternary carbide is a carbide of the formula Al.sub.4SiC.sub.4, a carbide of the formula Al.sub.4Si.sub.2C.sub.5, or a mixture thereof.

11. The filter according to claim 1, wherein the porous, open-pored, gas-permeable alpha-silicon carbide (SiC) containing shaped body contains at least one catalytically active coating.

Description

FIGURES

(1) FIG. 1 shows an electron microscope 1000 magnification of a shaped body made of reaction-bonded beta-SiC.

(2) FIG. 2 schematically shows the opening of the passage openings at increasing temperature. The whole process is divided into four phases: In phase I the conversion of silicon and carbon into SiC takes place; the beta-SiC shaped body forms. In phase lithe conversion is completed. As the temperature approaches 1850 C. the conversion of beta-SiC into alpha SiC begins. Phase III: In this phase the conversion of beta-SiC into alpha-SiC takes place completely and the pore passage openings widen. In this phase the pore opening widths can be set in a targeted manner via the temperature and residence time. Phase IV: the conversion into alpha-SiC is finished. Progressive crystal growth with conventional recrystallization now begins, i.e. small crystallites dissolve and become attached to the large crystallites; very large pore passage openings form.

(3) FIGS. 3a and 3b show electron microscope magnifications of a SiC shaped body from embodiment example 1 after heating to a maximum temperature of 1450 C. for 4 hours under argon atmosphere. FIG. 3a shows the cross-section of a channel wall at 150 magnification. FIG. 3b shows a view of a channel wall at 100 magnification.

(4) FIGS. 4a and 4b show electron microscope magnifications of a SiC shaped body from embodiment example 1 after heating to a maximum temperature of 1850 C. for 20 minutes under argon atmosphere. FIG. 4a shows the cross-section of a channel wall at 150 magnification. FIG. 4b shows a view of a channel wall at 100 magnification.

(5) FIGS. 5a and 5b show electron microscope magnifications of a SiC shaped body from embodiment example 1 after heating to a maximum temperature of 1950 C. for 20 minutes under argon atmosphere. FIG. 5a shows the cross-section of a channel wall at 150 magnification. FIG. 5b shows a view of a channel wall at 100 magnification.

(6) FIGS. 6a and 6b show electron microscope magnifications of a SiC shaped body from embodiment example 1 after heating to a maximum temperature of 2000 C. for 20 minutes under argon atmosphere. FIG. 6a shows the cross-section of a channel wall at 150 magnification. FIG. 6b shows a view of a channel wall at 100 magnification.

(7) FIGS. 7a and 7b show a cut-out section of the wall cross-section represented in FIG. 3a at 400 (FIG. 7a) and 1000 (FIG. 7b) magnification.

(8) FIGS. 8a and 8b show a cut-out section of the wall cross-section represented in FIG. 5a at 400 (FIG. 8a) and 1000 (FIG. 8b) magnification.

(9) FIG. 9 shows the mass loss and the change in length, in each case given in % relative to the mass or length, respectively, of the reaction-formed beta-SiC-containing shaped body before the heat treatment, as well as the thermal mass calculated from the external dimensions (length*width*height) and the mass of the honeycombs for different maximum temperatures, in each case given as temperature of the HT step, of the porous shaped bodies produced in embodiment example 1.

(10) FIG. 10 shows the pressure loss curves without soot loading, at increasing through-flow rates, of the alternately sealed honeycomb filters (width and height of 35 mm, length 178 mm) produced in embodiment example 1 at different maximum temperatures (1450 C., 1600 C., 1850 C., 1900 C., 1950 C. or 2050 C.).

(11) FIG. 11 shows the porosities in vol.-%, determined by means of mercury porosimetry, in each case relative to the external volume of the honeycomb filter, wherein the volume of open channels was not factored in, and average pore diameters in m of the alternately sealed honeycomb filters (width and height of 35 mm, length 178 mm) produced in embodiment example 1 at different maximum temperatures, in each case given as temperature of the HT step.

(12) FIG. 12 shows the average pore diameters in m and the flow-through speed through the channel wall in m/s, at a constant drop in pressure of 60 mbar, of the alternately sealed honeycomb filters (width and height of 35 mm, length 178 mm) produced in embodiment example 1 at different maximum temperatures, in each case given as temperature of the HT step.

(13) FIG. 13 shows the true density (skeletal density) in g/I calculated from the results of the mercury porosimetry as well as the specific pore surface area in m.sup.2/g of the alternately sealed honeycomb filters (width and height of 35 mm, length 178 mm) produced in embodiment example 1 at different maximum temperatures, in each case given as temperature of the HT step.

(14) FIG. 14 shows the respective XRD spectra (X-Ray Diffraction) of the alternately sealed honeycomb filters (width and height of 35 mm, length 178 mm) produced in embodiment example 1 at different maximum temperatures.

(15) FIG. 15 shows the strength, determined on the basis of the ASTM standard designation C 1674-08, of test specimens of the dimensions 10 channels*10 channels*120 mm as an average of in each case 8 test specimens which have been produced in embodiment example 1 at different maximum temperatures, in each case given as temperature of the HT step. Also plotted are the values for the wall fracture strength and the honeycomb structure strength calculated according to ASTM C 1674-08.

(16) FIG. 16 shows the from the measured pore diameters (D) of the alternately sealed honeycomb filters (width and height of 35 mm, length 178 mm) produced in embodiment example 1 at different maximum temperatures, in each case given as temperature of the HT step together with an adaptation of these measured values to the function of the pore diameter as a function of the maximum temperature according to equation (1).

(17) FIG. 17 shows the back pressure (Dp) in mbar, the average pore size (D) in m and porosity (P) in vol.-%, in each case relative to the external volume of the honeycomb filter, wherein the volume of open channels was not factored in, of the alternately sealed honeycomb filters (width and height of 35 mm, length 178 mm) produced in embodiment example 2 at different maximum temperatures, in each case given as temperature of the HT step.

(18) FIGS. 18a, 18b, 18c, 18d, 18e, and 18f show electron microscope magnifications of a SiC shaped body from embodiment example 2b (FIG. 18a to c) and of a SiC shaped body from comparison example 3 (FIG. 18d to f). FIGS. 18a and 18d in each case shows a view of a channel wall at 100 magnification. FIGS. 18b and 18e in each case shows the cross-section of a channel wall at 150 and 160 magnification, respectively. FIGS. 18c and 18f in each case shows a cut-out section of the wall cross-section represented in FIG. 18b and FIG. 18e, respectively, at 1000 magnification.

(19) FIG. 19 shows the back pressure without soot loading and with soot loading (100 g/l), at increasing flow-through rates, of a SiC shaped body from embodiment example 2b (EE 2B and EE 2B with 100 g/l washcoat) of a SiC shaped body from comparison example 3 (comparison example).

EMBODIMENT EXAMPLE 1 (EE 1)

(20) On the basis of the processes described in documents EP1 741 685 A1, U.S. Pat. No. 7,648,932 B2 and U.S. Pat. No. 7,867,313 B2, reaction-formed beta-SiC-containing shaped bodies were produced. The siliconization was carried out at a temperature at 1450 C.

(21) The reaction-formed beta-SiC-containing shaped bodies were subsequently subjected to a heat treatment, wherein the shaped bodies were heated in each case to different maximum temperatures (1450 C., 1600 C., 1750 C., 1800 C., 1840 C., 1850 C., 1885 C., 1900 C., 1925 C., 1950 C., 1975 C., 2000 C., 2020.1 C. or 2050 C.). The effect of the temperature treatment on the pore structure of the SiC shaped bodies and their filter properties was investigated.

(22) The residence time at maximum temperature was 20 min. In some cases the influence of longer residence times was investigated.

(23) For the investigations of the influence of the temperature treatment, in each case 3 honeycomb bodies of the dimensions 35 mm*35 mm*210 mm (height*width*length) were subjected to temperature treatment simultaneously. All honeycombs used came from one extrusion run. All honeycombs used were pyrolyzed in one run.

(24) The composition of the extrusion material for producing all the shaped bodies used in this test series is given in Table 1.

(25) TABLE-US-00001 TABLE 1 Composition of the extrusion material Component Proportion, weight-% Graphite 17% Pyrolysis binder 8% Silicon 45% Aluminum 4.8% Binder 5.7% Water 16% Pore former 3% Plasticizer 0.5%

(26) A finely ground electrode graphite from Richard Anton KG (Grafelfing, Germany) with an average grain size of approx. 17 m was used as graphite. The silicon powder was obtained from Elkem AS. The powder had an average particle size of approx. 65 m.

(27) A starch solution from Zschimmer und Schwarz GmbH & Co KG (Lahnstein, Germany) was used as pyrolysis binder. A powder from Grimm Metallpulver GmbH (Roth, Germany) was used as aluminum. Methocel types from DOW were used as binder.

(28) The pore former was a polyamide powder from Evonik Industries AG (Essen, Germany). A polyoxyethylene from Zschimmer und Schwarz was used as extrusion additive. The water used was demineralized.

(29) The dry and wet components were first mixed separately in isolation for 30 minutes for homogenization. The two pre-mixtures were then kneaded in a double-Z-kneader for 4 hours. The extrusion took place with a piston extrusion press. After the extrusion, the extrudate was cut into shaped bodies with a length of 210 mm. Height and width of the honeycombs was 35 mm*35 mm. The die used for square honeycombs with square channels had a cellularity of 300 cpsi (cells per square inch). Wall thickness according to the die was 11.5 mils (milli-inch).

(30) The drying of the segments took place by means of microwaves. After complete drying (residual moisture 1-2 mass-%) in a drying oven, all honeycombs were pyrolyzed in one run. The honeycombs here were heated up slowly to 850 C. under flowing nitrogen (technical grade 2.8). The organic components were broken down and expelled or converted into non-volatile pyrolysate which remained in the shaped bodies.

(31) For the siliconization of the shaped bodies, in each case three honeycombs were first heated up to 1450 C. in a graphite oven under gentle argon flow (argon grade 5.0). The residence time was two hours. Then a residence phase at 1600 C. for one hour took place as a further intermediate stage. The heating rates above 1000 C. were in each case 5 K/min.

(32) After this residence time, the shaped bodies, in turn, were heated at a heating rate of 5 K/min to the respective maximum temperature (1450 C., 1600 C., 1750 C., 1800 C., 1840 C., 1850 C., 1885 C., 1900 C., 1925 C., 1950 C., 1975 C., 2000 C., 2020.1 C. or 2050 C.) and held for 20 minutes, or for the stated residence time, at the respective maximum temperature.

(33) The cooling then took place within 20 hours to room temperature.

(34) For the purposes of comparison, shaped bodies were also heated for the duration of four hours to a temperature of 1450 C. without a further temperature increase being carried out.

(35) A subsequent oxidation of the honeycombs to increase the strength and to burn off residual carbon is possible and customary for later use, but was not carried out for these tests.

(36) The following properties were investigated for the SiC shaped bodies obtained in each case: mass loss changes in dimension back pressure of honeycombs with alternately sealed channels mechanical stability of test specimens of the size 10*10 channels at a length of 120 mm (4-point bending test) phase analysis by means of XRD scanning electron microscopy porosity and pore-size distribution, as well as density by means of mercury porosimetry

(37) In part, the honeycomb channels were sealed alternately for this:

(38) Investigation Under Scanning Electron Microscope (SEM)

(39) The investigation under SEM clearly showed how the pore structure looks before the opening of the passage openings and after the opening of the passage openings.

(40) The pore structure of a SiC shaped body after heating to a maximum temperature of 1450 C. for 4 hours and 1850 C. for 20 minutes is represented in FIGS. 3a and 3b, and 4a and 4b, respectively. Crystalline-looking structures can already be recognized, but many amorphous-looking eggshell-shaped structures can also still be recognized.

(41) The pore structure of a SiC shaped body after heating to a maximum temperature of 1950 C. and 2000 C. for 20 minutes can be seen in FIGS. 5a and 5b, and 6a and 6b, respectively. The amorphous-looking areas have disappeared and only crystalline-looking areas can still be recognized with large pore passage openings.

(42) At temperatures above 2000 C. the size of the crystallites increases further and the microstructure becomes more unstable through further opening, as is already indicated in the 2000 C., 20 minutes' residence time sample.

(43) In the higher magnifications in FIGS. 7a, 7b, 8a, and 8b it can be recognized that the microstructure with the large cavities is preserved during the temperature treatment. The starting structure which is based on the reaction-formed SiC is thus the basis for the microstructure of the shaped body treated at higher temperatures. At the same time, it can be clearly recognized that the structure of the skeleton has passed over from amorphous-looking to fine crystalline-looking.

(44) The mass loss, the change in length, and the thermal mass calculated from the external dimensions (length*width*height) and the mass of the honeycombs, for different maximum temperatures, are represented in FIG. 9.

(45) As can be seen, the large mass loss between 1450 C. and 1600 C. is followed by only a smaller further mass loss. The cause of the mass loss is initially evaporating aluminum and silicon. Above 1800 C. SiC can also evaporate in small amounts.

(46) Above 1900 C. a slight shrinkage occurs in the segments. Between 1450 C. and 1600 C. a slight swelling is present, which is caused by rearrangement processes as well as the solidification anomaly of the residual silicon. As the thermal mass results from the volume and the weight of the segments, this remains relatively constant.

(47) The pressure loss curves of alternately sealed honeycombs (length 178 mm) without soot loading at increasing flow-through rates are represented in FIG. 10. A clear reduction as the maximum temperature of the honeycombs increases can be seen, which takes place in the range 1800 C. and 1900 C. almost as a jump.

(48) The porosities and average pore diameters, determined by means of mercury porosimetry, of the honeycombs produced at different maximum temperatures are represented in FIG. 11. The porosity increases for the temperature range between 1450 C. and 1850 C. by approx. 10% to 20%, which can be attributed to the evaporation of aluminum and silicon and to rearrangement effects. The above-described swelling effects between 1450 C. and 1600 C. also increase the porosity of the honeycombs. Above 1850 C. the porosity remains almost constant.

(49) The average pore sizes and the flow-through speed through the channel walls, at a constant drop in pressure, of the honeycombs produced at different maximum temperatures are represented in FIG. 12.

(50) The true density (skeletal density) as well as the specific pore surface area were also calculated from the measurements of the mercury porosimetry and are represented in each case in FIG. 13.

(51) Due to the evaporation of aluminum and silicon and the further reaction of remaining carbon and silicon to form SiC, the true density of the material continuously increases with increasing maximum temperature. The increase in the specific pore surface area up to 1750 C. can likewise be explained by the evaporation of material as well as the further reaction to form SiC and thus the further formation of the microstructure.

(52) After completion of the reaction, a microstructure of a framework formed from SiC with thin membranes which formed a pore system with coarse pores was present. The pores, in turn, were connected to each other via narrow passages in the membrane structure. The narrow passages explain the small pore diameter with, at the same time, high porosity and large specific pore surface area. With increasing temperature, the passage channels in the membranes slowly widened, or the membranes slowly dissolved, with the result that the average pore opening diameter increased and the pore surface area, on the other hand, became smaller, while the porosity remained almost unchanged.

(53) By means of X-ray diffraction (XRD), the crystalline phases that formed were determined for powdered samples, of the honeycombs produced at different maximum temperatures. The samples were investigated with a D500 diffractometer from Bruker AXS GmbH (Karlsruhe, Germany) using the DIFFRAC plus software in the range 5-80 2 theta at an increment of 0.02 and a measurement time of 2 seconds in each case (total measurement time: approx. 2 hours).

(54) The respective XRD spectra are represented in FIG. 14.

(55) On the basis of the measurements represented, it can be recognized that during the transition from 1450 C. to 1750 C. the proportion of free silicon and aluminum declines sharply. The cause is the further reaction to form SiC, but also the evaporation of the components as well as the formation of three-component phases such as Al.sub.4SiC.sub.4 or Al.sub.4Si.sub.2C.sub.5. As the temperature increases, the conversion of beta-SiC into alpha-SiC also takes place, wherein type 2H forms first and then later also phases 4H and others. Above 1950 C. beta-SiC can no longer be detected.

(56) In addition to these main phases, nitrogen-containing or oxygen-containing phases can also be present in small proportions. Due to the low concentrations these are not to be determined and due to the mostly isomorphic incorporation additionally have similar reflections in XRD.

(57) Due to the increasing pore opening diameters and the evaporation of material, the mechanical strength of the honeycomb bodies decreases. The strength, determined on the basis of ASTM standard designation C 1674-08, of test specimens of the dimensions 10 channels*10 channels*120 mm as an average of in each case 8 test specimens is represented in FIG. 15. It can be recognized that the mechanical strength of the honeycombs decreases as the siliconization temperature increases. Surprisingly, however, the strength is sufficient for processing to form filters.

(58) Evaluation of the Pore Diameter as a Function of the Temperature.

(59) The porosity data were adapted by means of arctangent(x) according to Formula (1):

(60) $\begin{array}{cc}{D}_{\mathrm{pore}}\left(T\right)=D_0+\frac{D_1}{}.Math.\left\{\frac{}{2}+{\tan }^{-1}\left(\frac{T-T_0}{T_1}\right)\right\}& \left(1\right)\end{array}$

(61) For the above samples, the following parameters are obtained: D.sub.0=6.35 m, D.sub.1=28.2 m, T.sub.0=1916 C., T.sub.1=85.7 C. The associated diagram is represented in FIG. 16.

(62) For the setup selected in this test series for the oven process to siliconize the honeycombs, a temperature specification of 192550 C. can thus be derived for the desired pore diameter D.sub.pore=202 m.

(63) The energy per charge, which, however, cannot be determined comprehensively enough from the oven records, is also to be seen as decisive.

EMBODIMENT EXAMPLE 2a (EE 2a)

(64) The shaped bodies used were produced as described in embodiment example 1. In the formulation, the use of pore formers was dispensed with and a graphite with an average grain size of approx. 11 m was used, which resulted in a composition according to Table 2.

(65) TABLE-US-00002 TABLE 2 Composition of the extrusion material Component Proportion, weight-% Graphite 18.5% Pyrolysis binder 8.5% Silicon 48.2% Aluminum 5.3% Binder .sup.4% Water 15% Plasticizer 0.5%

(66) In contrast to embodiment example 1, for these honeycombs a residence time of two hours at the respective final temperatures of 1925 C., 1950 C., 1965 C. and 1975 C. was selected.

(67) As in embodiment example 1, shaped bodies with a width and height of 35 mm and a length of 178 mm in which the inflow and outflow channels were sealed alternately were also produced and measured for back pressure here. The result is represented in FIG. 17.

(68) As can be seen from FIG. 17, a temperature of 1950 C. is already enough in order to optimally open the pores with a residence time of 2 hours and thus to set a minimum back pressure.

EMBODIMENT EXAMPLE 2b (EE 2b)

(69) The shaped bodies used were produced as described in embodiment example 1, wherein a composition according to Table 2 was used to produce the shaped bodies.

(70) As in embodiment example 2a, the shaped bodies produced had alternately sealed inflow and outflow channels and had a cell density of 200 cpsi (cells per square inch) and a channel wall thickness of 400 m. The width and height of the shaped bodies was 49 mm, the length was again set to 178 mm.

(71) This design is very widespread in the field of off-road uses, in which very stable and robust systems are necessary.

(72) The shaped bodies produced in this way were treated at a maximum temperature of 1965 C. for two hours and had the following properties: Porosity: 58 vol.-%, relative to the external volume of the shaped body, wherein the volume of open channels was not factored in, Average pore diameter: 23 m Specific weight: 560 g/l

COMPARISON EXAMPLE 3

(73) In accordance with document U.S. Pat. No. 4,777,152, a shaped body with 200 cpsi cell density and a wall thickness of 400 m was produced on the basis of a beta-SiC powder.

(74) The aluminum content was set to 1 wt.-%, relative to the proportion of SiC. Preliminary tests with compacts resulted in the best results for this proportion with respect to the stability of the shaped body at a sintering temperature of 2000 C.

(75) The beta-SiC powder used was obtained from Superior Graphite Europe Ltd. (Sundsvall, Sweden) and had the designation HSC-1200. The average grain size was 6.26 m. In the case of the particle size, the upper limit of the range stated in document U.S. Pat. No. 4,777,152 was selected.

(76) The formulation is specified in Table 3.

(77) TABLE-US-00003 TABLE 3 Composition of the extrusion material Component Proportion, weight-% SiC (HSC1200) 74.8% Binder 5.1% Glycerol 1.5% Aluminum 0.7% Polyethylene glycol 1.8% Water 15.9% Plasticizer 0.2%

(78) The shaped body obtained had the following properties: Porosity: 40 vol.-%, relative to the external volume of the shaped body, wherein the volume of open channels was not factored in, Average pore diameter: 8 m Specific weight: 770 g/l

(79) The investigation under SEM clearly showed that a much higher porosity in the shaped body was achieved according to the invention via the process according to the invention without addition of a porosifier. In order to be able to set a much larger pore width according to the process used in the comparison example, a beta-SiC powder with a much larger average grain size would have to be selected. The corresponding SEM photographs are represented in FIGS. 18a to 18f.

(80) Furthermore, the shaped body produced in embodiment example 2b has a different microstructure in contrast to the shaped body produced in the comparison example. Whereas the shaped body produced in embodiment example 2b has the typical structure with large primary pores and smaller passage pores analogously to a foam-like structure, the structure of the shaped body produced in the comparison example has the typical microstructure of a grain ceramic.

(81) The advantages of the higher porosity and the thereby greatly improved absorption capacity for high washcoat loadings becomes clear by comparing the drops in pressure. These are represented in FIG. 19: with the high washcoat loading of 100 g/L, the filter of EE 2b still has a much lower back pressure than that of the comparison example. This is only possible due to the much higher porosity and the larger pore diameters which arise in-situ through the production process.