TERNARY INTERMETALLIC COMPOUND CATALYST

Abstract

The present invention relates to a catalyst comprising particles of a ternary intermetallic compound of the following formula (I): X.sub.2YZ wherein X, Y, and Z are different from one another; X being selected from the group consisting of Mn, Fe, Co, Ni, Cu, and Pd; Y being selected from the group consisting of V, Mn, Cu, Ti, and Fe; and Z being selected from the group consisting of Al, Si, Ga, Ge, In, Sn, and Sb; wherein the particles of the ternary intermetallic compound are supported on a support material, as well as to a method for its production and to its use as a catalyst, and more specifically as a catalyst in a process for the condensation of a carbonyl compound with a methylene group containing compound or for the selective catalytic reduction of nitrogen oxides in exhaust gas.

Claims

1: A catalyst comprising particles of at least one ternary intermetallic compound selected from the group consisting of Co.sub.2FeAl, Co.sub.2FeSi, Co.sub.2FeIn, Cu.sub.2FeAl, Cu.sub.2FeSi, Fe.sub.2MnGa, Fe.sub.2MnSi, Co.sub.2CuAl, and Fe.sub.2TiGa, wherein the particles of the ternary intermetallic compound are supported on a support material, wherein an average particle size D50 of the ternary intermetallic compound particles is from 3 nm to 2 ?m.

2: The catalyst of claim 1, wherein the intermetallic compound is a Heusler phase intermetallic compound.

3: The catalyst of claim 1, wherein the support material comprises at least one metal oxide and/or metalloid oxide selected from the group consisting of silica, alumina, silica-alumina, titania, and zirconia.

4: The catalyst of claim 3, wherein the BET surface area of the at least one metal oxide and/or metalloid oxide comprised in the support material ranges from 150 to 500 m.sup.2/g, wherein the BET surface area is determined according to ISO 9277 or DIN 66131.

5: The catalyst of claim 3, wherein a weight ratio of the ternary intermetallic compound to the at least one metal oxide and/or metalloid oxide comprised in the support material ranges from 0.5:99.5 to 50:50.

6: A method for the preparation of a catalyst comprising at least one ternary intermetallic compound selected from the group consisting of Co.sub.2FeAl, Co.sub.2FeSi, Co.sub.2FeIn, Cu.sub.2FeAl, Cu.sub.2FeSi, Fe.sub.2MnGa, Fe.sub.2MnSi, Co.sub.2CuAl, and Fe.sub.2TiGa, the method comprising: adding a support material to a solution comprising a precursor compound for Fe, Co, and Cu, a precursor compound for Mn, Cu, Ti, and Fe, a precursor compound for Al, Si, Ga, and In, and a solvent to obtain a mixture; evaporating the mixture to dryness to obtained a dried mixture; and heating the dried mixture in a hydrogen-containing atmosphere.

7: The method of claim 6, wherein the support material comprises at least one metal oxide and/or metalloid oxide selected from the group consisting of silica, alumina, silica-alumina, titania, and zirconia.

8: The method of claim 7, wherein the BET surface area of the at least one metal oxide and/or metalloid oxide ranges from 150 to 500 m.sup.2/g, wherein the BET surface area is determined according to ISO 9277 or DIN 66131.

9: A catalyst obtained according to the method of claim 6.

10: A process for the condensation of a carbonyl compound with a methylene group-containing compound, the process comprising simultaneously contacting a carbonyl compound and a methylene group-containing compound with a catalyst according to claim 1.

11: The process of claim 10, wherein the carbonyl compound is selected from the group consisting of aldehydes and ketones.

12: The process of claim 10, wherein the contacting of the carbonyl compound and the methylene group-containing compound with the catalyst is performed at a temperature in the range of from 30 to 150? C.

13: A process for the selective catalytic reduction of nitrogen oxides in exhaust gas, the process comprising performing selective catalytic reduction of nitrogen oxides in exhaust gas with the catalyst according to claim 1.

Description

DESCRIPTION OF THE FIGURES

[0093] FIGS. 1a to 14a, and 15 to 17 show the X-Ray Diffraction (XRD) pattern of the catalyst sample obtained from Examples 1-17, respectively. In the figures, the diffraction angle 2 theta in ? is shown along the abscissa and the intensities are plotted along the ordinate.

[0094] FIG. 14b displays the XRD pattern of gamma-alumina, wherein the diffraction angle 2 theta in ? is shown along the abscissa and the intensities are plotted along the ordinate.

[0095] FIGS. 1b to 13b show the scanning electron micrograph (SEM) of particles of the ternary intermetallic compound contained in the catalyst samples obtained from Examples 1-13, respectively.

[0096] FIG. 18 shows the results from catalyst testing performed on the catalyst samples from Examples 1-3 in the Knoevenagel condensation reaction of benzaldehyde with malononitrile to benzylidenemalononitrile (BMDN). In the Figure, the yield of BMDN in % is shown along the ordinate and the reaction time in hours is plotted along the abscissa. The results for Example 1 are indicated with the symbol ?, those for Example 2 with the symbol ?, and those for Example 3 with the symbol . The results from testing using the support material (SiO.sub.2) by itself are indicated with the symbol ?, and those from the control experiment conducted in the absence of a catalyst are indicated by the symbol ?.

[0097] FIGS. 19 and 20 respectively show the results from catalyst testing performed on the catalyst samples from Examples 4-7 in the Knoevenagel condensation reaction of benzaldehyde with malononitrile to benzylidenemalononitrile (BMDN). In the figures, the yield of BMDN in % is shown along the ordinate and the reaction time in hours is plotted along the abscissa. The results for Example 4 are indicated with the symbol ?, those for Example 5 with the symbol ?, those for Example 6 with the symbol ?, and those for Example 7 with the symbol . The results from testing using the support material (SiO.sub.2) by itself are indicated with the symbol ?, and those from the control experiment conducted in the absence of a catalyst are indicated by the symbol ?.

[0098] FIGS. 21 and 22 respectively show the results from catalyst testing performed in Example 18 as performed on the catalyst samples from Examples 8-10 in the Knoevenagel condensation reaction of benzaldehyde with malononitrile to benzylidenemalononitrile (BMDN). In the Figure, the yield of BMDN in % is shown along the ordinate and the reaction time in hours is plotted along the abscissa. The results for Example 8 are indicated with the symbol ?, those for Example 9 with the symbol ?, and those for Example 10 with the symbol . The results from testing using the support material (SiO.sub.2) by itself are indicated with the symbol 0, and those from the control experiment conducted in the absence of a catalyst are indicated by the symbol ?.

[0099] FIGS. 23 to 28 respectively show the results from selective catalytic reduction (SCR) testing performed in Example 19 as performed on the catalyst samples from Examples 12-17 wherein the values for the conversion of NO.sub.x is displayed by the symbol ? and those for the yield of N.sub.2O is displayed by the symbol ?, wherein the conversion rate/yield in % are shown along the ordinate and the reaction temperature in ? C. is plotted along the abscissa. In the respective figure, the results from SCR testing performed with the fresh catalyst samples are displayed on the left, those from testing performed on the catalyst samples aged at 750? C. for 5 hours are displayed in the middle, and those from testing performed on the catalyst samples aged at 850? C. for 6 hours are displayed on the right, respectively.

[0100] FIGS. 29 to 35 display High Angle Annular Dark Field-Scanning Transmission Electron Microscopy (HAADF-STEM) images obtained for the sample from Example 8. In the images, selected examples of individual ternary intermetallic compound particles of the Heusler phase Co.sub.2FeGa are indicated by arrows.

[0101] FIGS. 36 to 38 display Scanning Electron Microscopy images obtained with detection of backscattered electrons (SEM-BSE). In FIG. 36, selected examples of individual ternary intermetallic compound particles of the Heusler phase Co.sub.2FeGa are indicated by arrows.

[0102] FIG. 39 displays the particle size distribution for the particles mainly having a particle diameter of less than 400 nm as obtained from the HAADF-STEM images in FIGS. 29 to 35. The minimum diameter of the particles in nm is shown along the abscissa and the relative number of the particles having a given minimum diameter is plotted along the ordinate.

[0103] FIG. 40 displays the particle size distribution for the particles mainly having a particle diameter of 400 nm or greater as obtained from the SEM-BSE images in FIGS. 36 to 38. The minimum diameter of the particles in ?m is shown along the abscissa and the relative number of the particles having a given minimum diameter is plotted along the ordinate.

EXPERIMENTAL SECTION

[0104] The structure of the samples was characterized by powder x-ray diffraction (XRD) using Cu K-alpha radiation at 40 kV and 30 mA (Siemens D5005) at room temperature. The measurement of the powder patterns of the catalysts was carried out in the range of 3?2??100? with a step size of 0.05?.

[0105] The BET surface areas of the Heusler compounds were analyzed by nitrogen physisorption at 77 K with a Quantachrome AUTOSORB-1. The samples were pre-activated for 12 hours at 200? C. (Examples 1-10) or 100? C. (Examples 11 and 12). The BET surface area of pure ?-Al.sub.2O.sub.3 (Fa. Sasol Puralox SCFa-230) is 230 m.sup.2.Math.g.sup.?1. The BET surface area of the metal-loaded materials decreases to 170-180 m.sup.2.Math.g.sup.?1.

[0106] Scanning electron microscopy (SEM, SU 8000 Hitachi) was used to study the size and surface morphology of nanoparticles. The materials were coated with 5 nm chromium layer and measured at a voltage of 5 kV (Examples 1-10) or 20 kV (Examples 11 and 12).

[0107] Particle Size Analysis

[0108] The particle size D50 of the ternary intermetallic compound particles was determined by a combination of High Angle Annular Dark Field-Scanning Transmission Electron Microscopy (HAADF-STEM) and Scanning Electron Microscopy with detection of backscattered electrons (SEM-BSE) at 20 kV.

[0109] For conducting the HAADF-STEM analysis, samples were dispersed in ethanol. In view of the bimodal distribution of particle sizes for the ternary intermetallic compound particles in the inventive samples which may be divided into particles with a particle diameter of less than 400 nm and particles with a particle diameter of 400 nm or greater, the particle diameters of particles having a particle diameter of less than 400 nm was analyzed by HAADF-STEM, whereas the particle diameters of the particles having a particle diameter of 400 nm or greater was analyzed by SEM-BSE.

[0110] For determining the average particle diameter D50, multiple HAADF-STEM and SEM-BSE images were prepared and the particles in the images manually analyzed by a technical expert. For statistical analysis, a total of 10-20 HAADF-STEM and SEM-BSE images were prepared and evaluated. The respective images of the samples were enlarged such that the smallest particle dimensions were represented by at least 10 pixels. Individual particles identified in the images were then measured and their minimum diameter respectively recorded in accordance with Recommendation 2011/696/EU of the European Commission. Agglomerates of particles were treated as particles, i.e. the minimum diameter of the agglomerate was recorded. In the case of irregularly shaped particles or agglomerates, the minimum Feret diameter was determined.

[0111] The results from the analysis of the respective HAADF-STEM and SEM-BSE images was then respectively compiled and the D50 value for the average diameter calculated for the range of particle diameters from 0 nm to <400 nm and from 400 nm to 7 ?m.

Example 1: Co.SUB.2.FeGa on SiO.SUB.2 .(Co.SUB.2.FeGa@SiO.SUB.2.)

[0112] Methanol (500 ml) was supplied to CoCl.sub.2.6H.sub.2O (2.57 g, 10.8 mmol), Fe(NO.sub.3).sub.3.9H.sub.2O (1.62 g, 4.0 mmol) and Ga(NO.sub.3).sub.3.xH.sub.2O (1.21 g, 3.2 mmol). The round bottom flask containing the solution was placed in an ultrasonic bath and treated for 5 minutes. Fumed silica (10.00 g, primary particle average particle size=14 nm) was added to the precursor solution and the suspension was sonicated for 2 h at the room temperature. Then, the methanol from the orange suspension was removed on a rotary evaporator. Water bath temperature was adjusted to 40? C. The orange residue was transferred to a crystallizing dish and dried at 100? C. for 12 hours. The sand-colored solid was cooled to room temperature and grounded to a powder. A part of this powder was distributed in three ceramic shells and placed in a horizontally arranged quartz glass tube reactor mounted in a heating furnace. First, the reactor was rinsed thoroughly with nitrogen (36 ml.Math.min.sup.?1) for 10 minutes at room temperature. The annealing was carried out in a hydrogen atmosphere with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded silica was heated within 75 min to 850? C. and this temperature was maintained constant for 5 h. Finally, the gray samples were cooled to room temperature and characterized.

[0113] The crystal structure of the Heusler-compounds was determined by X-ray powder diffraction. The X-ray diffraction pattern of Co.sub.2FeGa on SiO.sub.2 for the angle range 2?=3-100? is shown in FIG. 1a. The sharp reflections between 2?=40-100? are caused by crystalline nanoparticles, and display the crystalline structure of the Heusler compound. Based on the results of simulation calculations, an assignment of the experimentally observed reflections could be made. The reflexes indicate an ordered superstructure. However, because of the strong noise and the small intensity in the range 2?=10-40? the characteristic signals for the L2.sub.1 phase may not observed.

[0114] FIG. 1b displays a particle of Co.sub.2FeGa on SiO.sub.2 as obtained from scanning electron microscopy of the sample from Example 1.

Example 2: Co.SUB.2.FeAl on SiO.SUB.2 .(Co.SUB.2.FeAl@SiO.SUB.2.)

[0115] Methanol (250 ml) was supplied to CoCl.sub.2.6H.sub.2O (1.28 g, 5.4 mmol), Fe(NO.sub.3).sub.3.9H.sub.2O (0.81 g, 2.0 mmol) and AlCl.sub.3.6H.sub.2O (0.39 g, 1.6 mmol). The round bottom flask containing the solution was placed in an ultrasonic bath and treated for 5 minutes. Fumed silica (5.03 g, primary particle average particle size=14 nm) was added to the precursor solution and the suspension was sonicated for 2 h at the room temperature. Then, the methanol from the orange suspension was removed on a rotary evaporator. Water bath temperature was adjusted to 40? C. The orange residue was transferred to a crystallizing dish and dried at 100? C. for 12 hours. The sand-colored solid was cooled to room temperature and grounded to a powder. A part of this powder was distributed in three ceramic shells and placed in a horizontally arranged quartz glass tube reactor mounted in a heating furnace. First, the reactor was rinsed thoroughly with nitrogen (36 ml.Math.min.sup.?1) for 10 minutes at room temperature. The annealing was carried out in a hydrogen atmosphere with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded silica was heated within 75 min to 850? C. and this temperature was maintained constant for 5 h. Finally, the gray samples were cooled to room temperature and characterized.

[0116] The X-ray diffraction pattern of Co.sub.2FeAl on SiO.sub.2 for the angle range 2?=3-100? is shown in FIG. 2a. The sharp reflections between 2?=40-100? are caused by crystalline nanoparticles, and display the crystalline structure of the Heusler compound.

[0117] FIG. 2b displays a particle of Co.sub.2FeAl on SiO.sub.2 as obtained from scanning electron microscopy of the sample from Example 2.

Example 3: Co.SUB.2.FeSi on SiO.SUB.2 .(Co.SUB.2.FeSi@SiO.SUB.2.)

[0118] Methanol (250 ml) was supplied to CoCl.sub.2.6H.sub.2O (1.29 g, 5.4 mmol), Fe(NO.sub.3).sub.3.9H.sub.2O (0.81 g, 2.0 mmol) and TEOS (tetraethyl orthosilicate) (0.33 g, 1.6 mmol). The round bottom flask containing the solution was placed in an ultrasonic bath and treated for 5 minutes. Fumed silica (5.02 g, primary particle average particle size=14 nm) was added to the precursor solution and the suspension was sonicated for 2 h at the room temperature. Then, the methanol from the orange suspension was removed on a rotary evaporator. Water bath temperature was adjusted to 40? C. The orange residue was transferred to a crystallizing dish and dried at 100? C. for 12 hours. The sand-colored solid was cooled to room temperature and grounded to a powder. A part of this powder was distributed in three ceramic shells and placed in a horizontally arranged quartz glass tube reactor mounted in a heating furnace. First, the reactor was rinsed thoroughly with nitrogen (36 ml.Math.min.sup.?1) for 10 minutes at room temperature. The annealing was carried out in a hydrogen atmosphere with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded silica was heated within 75 min to 850? C. and this temperature was maintained constant for 5 h. Finally, the gray samples were cooled to room temperature and characterized.

[0119] The X-ray diffraction pattern of Co.sub.2FeSi on SiO.sub.2 for the angle range 2?=3-100? is shown in FIG. 3a. The sharp reflections between 2?=40-100? are caused by crystalline nanoparticles, and display the crystalline structure of the Heusler compound.

[0120] FIG. 3b displays a particle of Co.sub.2FeSi on SiO.sub.2 as obtained from scanning electron microscopy of the sample from Example 3.

Example 4: Co.SUB.2.FeIn on SiO.SUB.2 .(Co.SUB.2.FeIn@SiO.SUB.2.)

[0121] Methanol (250 ml) was supplied to CoCl.sub.2.6H.sub.2O (1.29 g, 5.4 mmol), Fe(NO.sub.3).sub.3.9H.sub.2O (0.81 g, 2.0 mmol) and InCl.sub.3.xH.sub.2O (0.38 g, 1.6 mmol). The round bottom flask containing the solution was placed in an ultrasonic bath and treated for 5 minutes. Fumed silica (5.04 g, primary particle average particle size=7 nm) was added to the precursor solution and the suspension was sonicated for 2 h at the room temperature. Then, the methanol from the orange suspension was removed on a rotary evaporator. Water bath temperature was adjusted to 40? C. The orange residue was transferred to a crystallizing dish and dried at 100? C. for 12 hours. The sand-colored solid was cooled to room temperature and grounded to a powder. A part of this powder was distributed in three ceramic shells and placed in a horizontally arranged quartz glass tube reactor mounted in a heating furnace. First, the reactor was rinsed thoroughly with nitrogen (36 ml.Math.min.sup.?1) for 10 minutes at room temperature. The annealing was carried out in a hydrogen atmosphere with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded silica was heated within 75 min to 850? C. and this temperature was maintained constant for 5 h. Finally, the gray samples were cooled to room temperature and characterized.

[0122] The X-ray diffraction pattern of Co.sub.2FeIn on SiO.sub.2 for the angle range 2?=3-100? is shown in FIG. 4a. The sharp reflections between 2?=40-100? are caused by crystalline nanoparticles, and display the crystalline structure of the Heusler compound.

[0123] FIG. 4b displays a particle of Co.sub.2FeIn on SiO.sub.2 as obtained from scanning electron microscopy of the sample from Example 4.

Example 5: Co.SUB.2.FeGa on SiO.SUB.2 .(Co.SUB.2.FeGa@SiO.SUB.2.)

[0124] In a typical example, distilled water (500 ml) was supplied to CoCl.sub.2.6H.sub.2O (2.57 g, 10.8 mmol), Fe(NO.sub.3).sub.3.9H.sub.2O (1.62 g, 4.0 mmol) and Ga(NO.sub.3).sub.3.xH.sub.2O (1.21 g, 3.2 mmol). The round bottom flask containing the solution was placed in an ultrasonic bath and treated for 5 minutes. Fumed silica (10.02 g, primary particle average particle size=7 nm) was added to the precursor solution and the suspension was sonicated for 2 h at the room temperature. Then, the water from the orange suspension was removed on a rotary evaporator. Water bath temperature was adjusted to 60? C. The orange residue was transferred to a crystallizing dish and dried at 100? C. for 12 hours. The sand-colored solid was cooled to room temperature and grounded to a powder. A part of this powder was distributed in three ceramic shells and placed in a horizontally arranged quartz glass tube reactor mounted in a heating furnace. First, the reactor was rinsed thoroughly with nitrogen (36 ml.Math.min.sup.?1) for 10 minutes at room temperature. The annealing was carried out in a hydrogen atmosphere with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded silica was heated within 75 min to 850? C. and this temperature was maintained constant for 5 h. Finally, the gray samples were cooled to room temperature and characterized.

[0125] The X-ray diffraction pattern of Co.sub.2FeGa on SiO.sub.2 for the angle range 2?=3-100? is shown in FIG. 5a. The sharp reflections between 2?=40-100? are caused by crystalline nanoparticles, and display the crystalline structure of the Heusler compound.

[0126] FIG. 5b displays a particle of Co.sub.2FeGa on SiO.sub.2 as obtained from scanning electron microscopy of the sample from Example 5.

Example 6: Co.SUB.2.FeAl on SiO.SUB.2 .(Co.SUB.2.FeAl@SiO.SUB.2.)

[0127] Distilled water (500 ml) was supplied to CoCl.sub.2.6H.sub.2O (2.57 g, 10.8 mmol), Fe(NO.sub.3).sub.3.9H.sub.2O (1.62 g, 4.0 mmol) and AlCl.sub.3.6H.sub.2O (0.77 g, 3.2 mmol). The round bottom flask containing the solution was placed in an ultrasonic bath and treated for 5 minutes. Fumed silica (10.07 g, primary particle average particle size=7 nm) was added to the precursor solution and the suspension was sonicated for 2 h at the room temperature. Then, the water from the pink suspension was removed on a rotary evaporator. Meanwhile, the color of the suspension has changed from pink to orange. Water bath temperature was adjusted to 60? C. The orange residue was transferred to a crystallizing dish and dried at 100? C. for 12 hours. The sand-colored solid was cooled to room temperature and grounded to a powder. A part of this powder was distributed in three ceramic shells and placed in a horizontally arranged quartz glass tube reactor mounted in a heating furnace. First, the reactor was rinsed thoroughly with nitrogen (36 ml.Math.min.sup.?1) for 10 minutes at room temperature. The annealing was carried out in a hydrogen atmosphere with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded silica was heated within 75 min to 850? C. and this temperature was maintained constant for 5 h. Finally, the gray samples were cooled to room temperature and characterized.

[0128] The X-ray diffraction pattern of Co.sub.2FeAl on SiO.sub.2 for the angle range 2?=3-100? is shown in FIG. 6a. The sharp reflections between 2?=40-100? are caused by crystalline nanoparticles, and display the crystalline structure of the Heusler compound.

[0129] FIG. 6b displays a particle of Co.sub.2FeAl on SiO.sub.2 as obtained from scanning electron microscopy of the sample from Example 6.

Example 7: Co.SUB.2.FeSi on SiO.SUB.2 .(Co.SUB.2.FeSi@SiO.SUB.2.)

[0130] Distilled water (500 ml) was supplied to CoCl.sub.2.6H.sub.2O (2.57 g, 10.8 mmol), Fe(NO.sub.3).sub.3.9H.sub.2O (1.61 g, 4.0 mmol) and TEOS (tetraethyl orthosilicate) (0.67 g, 3.2 mmol). The round bottom flask containing the solution was placed in an ultrasonic bath and treated for 5 minutes. Fumed silica (10.07 g, primary particle average particle size=7 nm) was added to the precursor solution and the suspension was sonicated for 2 h at the room temperature. Then, the water from the pink suspension was removed on a rotary evaporator. Meanwhile, the color of the suspension has changed from pink to orange. Water bath temperature was adjusted to 60? C. The orange residue was transferred to a crystallizing dish and dried at 100? C. for 12 hours. The sand-colored solid was cooled to room temperature and grounded to a powder. A part of this powder was distributed in three ceramic shells and placed in a horizontally arranged quartz glass tube reactor mounted in a heating furnace. First, the reactor was rinsed thoroughly with nitrogen (36 ml.Math.min.sup.?1) for 10 minutes at room temperature. The annealing was carried out in a hydrogen atmosphere with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded silica was heated within 75 min to 850? C. and this temperature was maintained constant for 5 h. Finally, the gray samples were cooled to room temperature and characterized.

[0131] The X-ray diffraction pattern of Co.sub.2FeSi on SiO.sub.2 for the angle range 2?=3-100? is shown in FIG. 7a. The sharp reflections between 2?=40-100? are caused by crystalline nanoparticles, and display the crystalline structure of the Heusler compound.

[0132] FIG. 7b displays a particle of Co.sub.2FeSi on SiO.sub.2 as obtained from scanning electron microscopy of the sample from Example 7.

Example 8: Co.SUB.2.FeGa on SiO.SUB.2 .(Co.SUB.2.FeGa@SiO.SUB.2.)

[0133] Supported Co.sub.2FeGa nanoparticles on SiO.sub.2 were prepared by synthesis as described in Example 5. The sample was placed in the quartz glass tube reactor, rinsed thoroughly with nitrogen (36 ml.Math.min.sup.?1) for 10 minutes and then annealed in a hydrogen/nitrogen (5/95) atmosphere with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded silica was heated within 75 min to 850? C. and this temperature was maintained constant for 5 h.

[0134] The X-ray diffraction pattern of Co.sub.2FeGa on SiO.sub.2 for the angle range 2?=3-100? is shown in FIG. 8a. The sharp reflections between 2?=40-100? are caused by crystalline nanoparticles, and display the crystalline structure of the Heusler compound.

[0135] FIG. 8b displays a particle of Co.sub.2FeGa on SiO.sub.2 as obtained from scanning electron microscopy of the sample from Example 8.

[0136] FIGS. 29 to 35 display High Angle Annular Dark Field-Scanning Transmission Electron Microscopy (HAADF-STEM) images obtained for the sample from Example 8.

[0137] FIGS. 36 to 38 display Scanning Electron Microscopy images obtained with detection of backscattered electrons (SEM-BSE) for the sample from Example 8.

[0138] FIG. 39 displays the particle size distribution for the particles mainly having a particle diameter of less than 400 nm as obtained from the HAADF-STEM images. Analysis of the results affords an average particle size D50 of 86.6 nm for the ternary intermetallic compound particles in the sample of Example 8.

[0139] FIG. 40 displays the particle size distribution for the particles mainly having a particle diameter of 400 nm or greater as obtained from the SEM-BSE images.

Example 9: Co.SUB.2.FeAl on SiO.SUB.2 .(Co.SUB.2.FeAl@SiO.SUB.2.)

[0140] Supported Co.sub.2FeAl nanoparticles on SiO.sub.2 were prepared by synthesis as described in Example 6. The sample was placed in the quartz glass tube reactor, rinsed thoroughly with nitrogen (36 ml.Math.min.sup.?1) for 10 minutes and then annealed in a hydrogen/nitrogen (5/95) atmosphere with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded silica was heated within 75 min to 850? C. and this temperature was maintained constant for 5 h.

[0141] The X-ray diffraction pattern of Co.sub.2FeAl on SiO.sub.2 for the angle range 2?=3-100? is shown in FIG. 9a. The sharp reflections between 2?=40-100? are caused by crystalline nanoparticles, and display the crystalline structure of the Heusler compound.

[0142] FIG. 9b displays a particle of Co.sub.2FeAl on SiO.sub.2 as obtained from scanning electron microscopy of the sample from Example 9.

Example 10: Co.SUB.2.FeSi on SiO.SUB.2 .(Co.SUB.2.FeSi@SiO.SUB.2.)

[0143] Supported Co.sub.2FeSi nanoparticles on SiO.sub.2 were prepared by synthesis as described in Example 7. The sample was placed in the quartz glass tube reactor, rinsed thoroughly with nitrogen (36 ml.Math.min.sup.?1) for 10 minutes and then annealed in a hydrogen/nitrogen (5/95) atmosphere with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded silica was heated within 75 min to 850? C. and this temperature was maintained constant for 5 h.

[0144] The X-ray diffraction pattern of Co.sub.2FeSi on SiO.sub.2 for the angle range 2?=3-100? is shown in FIG. 10a. The sharp reflections between 2?=40-100? are caused by crystalline nanoparticles, and display the crystalline structure of the Heusler compound.

[0145] FIG. 10b displays a particle of Co.sub.2FeSi on SiO.sub.2 as obtained from scanning electron microscopy of the sample from Example 10.

Example 11: Co.SUB.2.FeIn on SiO.SUB.2 .(Co.SUB.2.FeIn@SiO.SUB.2.)

[0146] Supported Co.sub.2FeIn nanoparticles on SiO.sub.2 were prepared by synthesis as described in Example 4. The sample was placed in the quartz glass tube reactor, rinsed thoroughly with nitrogen (36 ml.Math.min.sup.?1) for 10 minutes and then annealed in a hydrogen/nitrogen (5/95) atmosphere with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded silica was heated within 75 min to 850? C. and this temperature was maintained constant for 5 h.

[0147] The X-ray diffraction pattern of Co.sub.2FeIn on SiO.sub.2 for the angle range 2?=3-100? is shown in FIG. 11a. The sharp reflections between 2?=40-100? are caused by crystalline nanoparticles, and display the crystalline structure of the Heusler compound.

[0148] FIG. 11b displays a particle of Co.sub.2FeIn on SiO.sub.2 as obtained from scanning electron microscopy of the sample from Example 11.

Example 12: Cu.SUB.2.FeAl on SiO.SUB.2 .(Cu.SUB.2.FeAl@SiO.SUB.2.)

[0149] Distilled water (500 ml) was supplied to Cu(NO.sub.3).2?H.sub.2O (2.51 g, 10.8 mmol), Fe(NO.sub.3).sub.3.9H.sub.2O (1.62 g, 4.0 mmol) and AlCl.sub.3.6H.sub.2O (0.77 g, 3.2 mmol). The round bottom flask containing the solution was placed in an ultrasonic bath and treated for 5 minutes. Fumed silica (10.03 g, primary particle average particle size=7 nm) was added to the precursor solution and the suspension was sonicated for 2 h at the room temperature. Then, the water from the light green suspension was removed on a rotary evaporator. Water bath temperature was adjusted to 60? C. The green residue was transferred to a crystallizing dish and dried at 100? C. for 12 hours. The yellow brown red colored solid was cooled to room temperature and grounded to a powder. A part of this powder was distributed in three ceramic shells and placed in a horizontally arranged quartz glass tube reactor mounted in a heating furnace. First, the reactor was rinsed thoroughly with nitrogen (43 ml.Math.min.sup.?1) for 10 minutes at room temperature. The annealing was carried out in a hydrogen/nitrogen (5/95) atmosphere with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded silica was heated within 75 min to 850? C. and this temperature was maintained constant for 5 h. Finally, the red samples were cooled to room temperature and characterized.

[0150] The X-ray diffraction pattern of Cu.sub.2FeAl on SiO.sub.2 for the angle range 2?=3-100? is shown in FIG. 12a. The sharp reflections between 2?=40-100? are caused by crystalline nanoparticles, and display the crystalline structure of the Heusler compound. Based on the results of simulation calculations, an assignment of the experimentally observed reflections could be made. The reflexes indicate an ordered superstructure. However, because of the strong noise and the small intensity in the range 2?=10-40? the characteristic signals for the L2.sub.1 phase may not observed.

[0151] FIG. 12b displays a particle of Cu.sub.2FeAl on SiO.sub.2 as obtained from scanning electron microscopy of the sample from Example 12.

Example 13: Cu.SUB.2.FeSi on SiO.SUB.2 .(Cu.SUB.2.FeSi@SiO.SUB.2.)

[0152] In a typical example, distilled water (500 ml) was supplied to Cu(NO.sub.3).2?H.sub.2O (2.51 g, 10.8 mmol), Fe(NO.sub.3).sub.3.9H.sub.2O (1.62 g, 4.0 mmol) and TEOS (tetraethyl orthosilicate) (0.67 g, 3.2 mmol). The round bottom flask containing the solution was placed in an ultrasonic bath and treated for 5 minutes. Fumed silica (10.02 g, primary particle average particle size=7 nm) was added to the precursor solution and the suspension was sonicated for 2 h at the room temperature. Then, the water from the light green suspension was removed on a rotary evaporator. Water bath temperature was adjusted to 60? C. The green residue was transferred to a crystallizing dish and dried at 100? C. for 12 hours. The brown red colored solid was cooled to room temperature and grounded to a powder. A part of this powder was distributed in three ceramic shells and placed in a horizontally arranged quartz glass tube reactor mounted in a heating furnace. First, the reactor was rinsed thoroughly with nitrogen (45 ml.Math.min.sup.?1) for 10 minutes at room temperature. The annealing was carried out in a hydrogen/nitrogen (5/95) atmosphere with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded silica was heated within 75 min to 850? C. and this temperature was maintained constant for 5 h. Finally, the red samples were cooled to room temperature and characterized.

[0153] The X-ray diffraction pattern of Cu.sub.2FeSi on SiO.sub.2 for the angle range 2?=3-100? is shown in FIG. 13a. The sharp reflections between 2?=40-100? are caused by crystalline nanoparticles, and display the crystalline structure of the Heusler compound. Based on the results of simulation calculations, an assignment of the experimentally observed reflections could be made. The reflexes indicate an ordered superstructure. However, because of the strong noise and the small intensity in the range 2?=10-40? the characteristic signals for the L2.sub.1 phase may not observed.

[0154] FIG. 13b displays a particle of Cu.sub.2FeSi on SiO.sub.2 as obtained from scanning electron microscopy of the sample from Example 13.

Example 14: Fe.SUB.2.MnGa on ?-Al.SUB.2.O.SUB.3 .(Fe.SUB.2.MnGa@Al.SUB.2.O.SUB.3.)

[0155] In a typical example, water (1.5 mL) was supplied to Fe(NO.sub.3).sub.3.9H.sub.2O (0.36 g, 0.89 mmol), Mn(NO.sub.3).sub.2.4H.sub.2O (0.11 g, 0.45 mmol) and Ga(NO.sub.3).sub.3.xH.sub.2O (0.19 g, 0.45 mmol). The mixture was placed in an ultrasonic bath and treated for 5 minutes to form a solution. Aluminium oxide (?-Al.sub.2O.sub.3, 2.00 g, particle size D50=25 ?m; Fa. Sasol, Puralox SCFa-230) was supplied to a crystallizing dish and the precursor solution was added drop wise under constant steering (incipient wetness impregnation). The wet solid was dried at 100? C. for 18 hours. The solid was cooled to room temperature and grounded to a powder. The powder was distributed in three ceramic shells and placed in a horizontally arranged quartz glass tube reactor mounted in a heating furnace. First, the reactor was rinsed thoroughly with nitrogen (45 mL.Math.min.sup.?1) for 10 minutes at room temperature. The annealing was carried out with 10 vol % hydrogen in nitrogen with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded aluminium oxide was heated with a rate of 11.5 K.Math.min.sup.?1 to 850? C. and this temperature was maintained constant for 5 h. Finally, the sand-colored samples were passive cooled to room temperature and characterized.

[0156] The X-ray diffraction pattern of Fe.sub.2MnGa on ?-Al.sub.2O.sub.3 for the angle range 2?=3-100? is shown in FIG. 14a. As may be taken from a comparison of the diffraction pattern in FIG. 14a with the XRD pattern of pure ?-Al.sub.2O.sub.3 shown in FIG. 14b, the pattern of the latter overlays the reflections of the ternary intermetallic compound Fe.sub.2MnGa.

Example 15: Fe.SUB.2.MnSi on ?-Al.SUB.2.O.SUB.3 .(Fe.SUB.2.MnSi@Al.SUB.2.O.SUB.3.)

[0157] In a typical example, water (1.4 mL) was supplied to Fe(NO.sub.3).sub.3.9H.sub.2O (0.44 g, 1.08 mmol), Mn(NO.sub.3).sub.2.4H.sub.2O (0.14 g, 0.54 mmol) and Si(OC.sub.2H.sub.5).sub.4 (0.11 g, 0.54 mmol). The mixture was placed in an ultrasonic bath and treated for 5 minutes to form a solution. Aluminium oxide (?-Al.sub.2O.sub.3, 2.00 g, particle size D50=25 ?m; Fa. Sasol, Puralox SCFa-230) was supplied to a crystallizing dish and the precursor solution was added drop wise under constant steering (incipient wetness impregnation). The wet solid was dried at 100? C. for 18 hours. The solid was cooled to room temperature and grounded to a powder. The powder was distributed in three ceramic shells and placed in a horizontally arranged quartz glass tube reactor mounted in a heating furnace. First, the reactor was rinsed thoroughly with nitrogen (45 mL.Math.min.sup.?1) for 10 minutes at room temperature. The annealing was carried out with 10 vol % hydrogen in nitrogen with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded aluminium oxide was heated with a rate of 11.5 K.Math.min.sup.?1 to 850? C. and this temperature was maintained constant for 5 h. Finally, the light gray samples were passive cooled to room temperature.

[0158] The X-ray diffraction pattern of Fe.sub.2MnSi on ?-Al.sub.2O.sub.3 for the angle range 2?=3-100? is shown in FIG. 15. As may be taken from a comparison of the diffraction pattern in FIG. 15 with the XRD pattern of pure ?-Al.sub.2O.sub.3 shown in FIG. 14b, the pattern of the latter overlays the reflections of the ternary intermetallic compound Fe.sub.2MnSi.

Example 16: Co.SUB.2.CuAl on ?-Al.SUB.2.O.SUB.3 .(Co.SUB.2.CuAl@Al.SUB.2.O.SUB.3.)

[0159] In a typical example, water (1.5 mL) was supplied to CoCl.sub.2.6H.sub.2O (0.24 g, 1.01 mmol), Cu(NO.sub.3).sub.2.2.5H.sub.2O (0.12 g, 0.51 mmol) and AlCl.sub.3.6H.sub.2O (0.18 g, 0.51 mmol). The mixture was placed in an ultrasonic bath and treated for 5 minutes to form a solution. Aluminium oxide (?-Al.sub.2O.sub.3, 2.00 g, particle size D50=25 ?m; Fa. Sasol, Puralox SCFa-230) was supplied to a crystallizing dish and the precursor solution was added drop wise under constant steering (incipient wetness impregnation). The wet solid was dried at 100? C. for 18 hours. The solid was cooled to room temperature and grounded to a powder. The powder was distributed in three ceramic shells and placed in a horizontally arranged quartz glass tube reactor mounted in a heating furnace. First, the reactor was rinsed thoroughly with nitrogen (45 mL.Math.min.sup.?1) for 10 minutes at room temperature. The annealing was carried out with 10 vol % hydrogen in nitrogen with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded aluminium oxide was heated with a rate of 11.5 K.Math.min.sup.?1 to 850? C. and this temperature was maintained constant for 5 h. Finally, the light blue samples were passive cooled to room temperature and characterized.

[0160] The X-ray diffraction pattern of Co.sub.2CuAl on ?-Al.sub.2O.sub.3 for the angle range 2?=3-100? is shown in FIG. 16. As may be taken from a comparison of the diffraction pattern in FIG. 16 with the XRD pattern of pure ?-Al.sub.2O.sub.3 shown in FIG. 14b, the pattern of the latter overlays the reflections of the ternary intermetallic compound Co.sub.2CuAl.

Example 17: Fe.SUB.2.TiGa on ?-Al.SUB.2.O.SUB.3 .(Fe.SUB.2.TiGa@Al.SUB.2.O.SUB.3.)

[0161] In a typical example, water (1.5 mL) was supplied to Fe(NO.sub.3).sub.3.9H.sub.2O (0.37 g, 0.92 mmol), TiCl.sub.4 (0.07 g, 0.46 mmol) and Ga(NO.sub.3).sub.3.xH.sub.2O (0.18 g, 0.46 mmol). The mixture was placed in an ultrasonic bath and treated for 5 minutes to form a solution. Aluminium oxide (?-Al.sub.2O.sub.3, 2.00 g, particle size D50=25 ?m; Fa. Sasol, Puralox SCFa-230) was supplied to a crystallizing dish and the precursor solution was added drop wise under constant steering (incipient wetness impregnation). The wet solid was dried at 100? C. for 18 hours. The solid was cooled to room temperature and grounded to a powder. The powder was distributed in three ceramic shells and placed in a horizontally arranged quartz glass tube reactor mounted in a heating furnace. First, the reactor was rinsed thoroughly with nitrogen (45 mL.Math.min.sup.?1) for 10 minutes at room temperature. The annealing was carried out with 10 vol % hydrogen in nitrogen with a flow rate of 50 ml.Math.min.sup.?1. The metal-loaded aluminium oxide was heated with a rate of 11.5 K.Math.min.sup.?1 to 850? C. and this temperature was maintained constant for 5 h. Finally, the sand-colored samples were passive cooled to room temperature and characterized.

[0162] The X-ray diffraction pattern of Fe.sub.2TiGa on ?-Al.sub.2O.sub.3 for the angle range 2?=3-100? is shown in FIG. 17. As may be taken from a comparison of the diffraction pattern in FIG. 17 with the XRD pattern of pure ?-Al.sub.2O.sub.3 shown in FIG. 14b, the pattern of the latter overlays the reflections of the ternary intermetallic compound Fe.sub.2TiGa.

Example 18: Catalytic Testing Experiments Based on the Knoevenagel Condensation Reaction

[0163] The synthesized nanoparticles supported on SiO.sub.2 as obtained from Examples 1-10 were used in a Knoevenagel condensation for the reaction of benzaldehyde with malononitrile to benzylidenemalononitrile (BMDN) and the composition of the product mixture are analyzed by gas chromatography. In a typical catalytic experiment 0.26 g (4 mmol) malononitrile, 0.42 g (4 mmol) of freshly distilled benzaldehyde, 10 ml of toluene as a solvent and 0.2 g of 1,4-dichlorobenzene as internal standard were mixed in a 50 ml two-necked flask equipped with a reflux condenser. The mixture was heated in an oil bath at 80? C. In general, 0.2 g of dried (12 h at 100? C.) catalyst was added. At regular time intervals the reaction mixture was analyzed by gas chromatography. The samples (0.2 ?l) were injected into the heated GC injector block of a HP 6890 Series gas chromatograph (Hewlett-Packard). The assignment of the peaks of the analyzed mixture was compared with that of the calibration. A solution from each of the components of the reaction mixture with toluene and 1,4-dichlorobenzene was injected in the GC and analyzed. The gas chromatographic conditions are listed in Table 1 below.

TABLE-US-00001 TABLE 1 Gas chromatographic conditions employed in Example 18 Sample volume 0.2 ?l Injector temperature 250? C. Heating rate Start at 70? C., 2 min isothermal Heating rate of 10 K .Math. min.sup.?1 to 250? C. Carrier gas Helium Flow 2.3 ml .Math. min.sup.?1 Column head pressure 0.8 bar Split ratio 50:1 Column HP-5 Trace Analysis 5% Phenyl Methyl Capillary (Length: 30 m, Inner diameter: 320 ?m, Film thickness: 0.25 ?m) Detector FID

[0164] The activity of synthesized Heusler compounds from the respective examples were tested in the base-catalyzed reaction. Before the start of the test series benzaldehyde was distilled under reduced pressure to remove benzoic acid. The freshly distilled benzaldehyde was then stored under an inert gas atmosphere. In addition, for comparison, the reaction of benzaldehyde with malononitrile was carried out only over SiO.sub.2. For the graphical analysis the yield of product was applied against the reaction time. The results obtained for Examples 1-3 are shown in FIG. 18, those obtained for Examples 4-6 are shown in FIGS. 19 and 20, respectively, and those obtained for Examples 9-10 are shown in FIGS. 21 and 22, respectively.

[0165] Thus, as may be taken from FIG. 18, in the reference reaction only using SiO.sub.2, a low yield of product was detected. For Co.sub.2FeGa@SiO.sub.2 (Example 1) and Co.sub.2FeSi@SiO.sub.2 (Example 3) a low catalytic activity was also detected. The significantly higher activity of Co.sub.2FeAl@SiO.sub.2 (Example 2) is tentatively attributed to the high catalytic activity of aluminum.

[0166] As may be taken form FIG. 19, in the catalytic reaction of benzaldehyde with malononitrile with the catalyst samples which were prepared in water as the solvent (see synthetic procedures of Examples 4-6, respectively) a general increase of the product yield for all samples is observed compared to those prepared in methanol. Most active is Co.sub.2FeAl@SiO.sub.2 (Example 6) with approximately 95% yield, followed by Co.sub.2FeSi@SiO.sub.2 (Example 7) with 88% yield, Co.sub.2FeGa@SiO.sub.2 (Example 5) with 62% yield and Co.sub.2FeIn@SiO.sub.2 (Example 4) with 60% yield of BMDN. Upon repeating the reactions, it is observed that the order of activity of the prepared compounds is almost the same (see results displayed in FIG. 20). In this respect it is however noted that in FIG. 20, the Co.sub.2FeIn@SiO.sub.2 catalyst sample used was obtained according to Example 4 yet using water instead of methanol.

[0167] The compounds prepared in water and annealed in H.sub.2/N.sub.2 atmosphere were also investigated in Knoevenagel reaction. In FIG. 21 it can be seen that aluminum-containing compound Co.sub.2FeAl@SiO.sub.2 (Example 9) is most active. Then Co.sub.2FeGa@SiO.sub.2 (Example 8) follows with 82% yield and Co.sub.2FeSi@SiO.sub.2 (Example 10) with 48% yield of product. The results of the repeated reactions are shown in FIG. 22. One difference from the other samples (Example 1-7 in FIGS. 18-20) is that in reactions with Co.sub.2FeGa@SiO.sub.2 (Example 8) more product is formed than in those with Co.sub.2FeSi@SiO.sub.2 (Example 10).

Example 19: SCR (Selective Catalytic Reduction) Testing

[0168] For the SCR test, the catalyst samples from Examples 12-17 were first mixed with a slurry of premilled gamma alumina (30 wt % Al.sub.2O.sub.3, 70 wt % catalyst). The slurry was dried under stirring on a magnetic stirring plate at 100? C., calcined (1 h, 600? C., air), and the resulting cake crushed and sieved to a target fraction of 250-500 ?m for testing. Fractions of the respective shaped powders were aged in a muffle oven for 5 h at 750? C. in 10% steam/air and for 6 h at 850? C. in 10% steam/air.

[0169] SCR tests were then performed using a 48 fold parallel testing unit equipped with ABB LIMAS NOx/NH3 and ABB URAS N.sub.2O analyzers. For each fresh and aged catalyst, 170 mg of the shaped powder diluted with corundum to a total volume of 1 mL were placed in each reactor. Under isothermal conditions (T=150, 200, 250, 300, 450, 500, 575? C.) a feed gas consisting of 500 ppm NO, 500 ppm NH3, 5% O2, 10% H.sub.2O balance N.sub.2 was passed at a GHSV of 80,000 h.sup.?1 through the catalyst bed. In addition to 30 min equilibration time for thermal equilibration of the parallel reactor at each temperature, every position was equilibrated for 3.5 min followed by 30 sec sampling time. Data recorded by the analyzers at a frequency of 1 Hz was averaged for the sampling interval and used to calculate NO conversions and N.sub.2O yield.

[0170] The results obtained for the samples prepared from Examples 12-17 are displayed in FIGS. 23-28, respectively. Thus, as may be taken from the results, the samples from Examples 12 and 13 on silica (Cu.sub.2FeAl@SiO.sub.2 and Cu.sub.2FeSi@SiO.sub.2) only display a moderate activity when employed in SCR, which nevertheless is not diminished after ageing. Furthermore, the aforementioned samples display a certain activity relative to the conversion of N.sub.2O which is not observed by the samples prepared from Examples 14-17 on gamma-alumina.

[0171] As regards the results obtained for the samples from Examples 14-17 on gamma-alumina, on the other hand, these display a surprisingly high acitivity with respect to the conversion of NO.sub.x, wherein it is observed that the samples containing Fe display a progressive increase in their ability to reduce NO.sub.x emission, whereas the sample containing Co displays a rapid increase in acitivity at lower temperatures which decreases at higher temperatures. In particular, as for the samples from Examples 12 and 13, it has quite unexpectedly been found that the activity of the inventive catalysts does not decrease upon aging. In fact, as concerns the Co containing sample of Example 16, it is even observed that the maximum activity level in the reduction of NO.sub.x acutally increases upon aging compared to the fresh sample when employed in selective catalytic reduction.