Method for producing mixed oxide materials containing molybdenum

Abstract

A simple, scalable, inexpensive, and reproducible method of selectively preparing the M1 phase of a MoVNbTe mixed oxide in a hydrothermal synthesis using tellurium dioxide is disclosed which can utilize inexpensive metal oxides as starting compounds.

Claims

1. A method for preparing a mixed-oxide material containing the elements molybdenum, vanadium, niobium, and tellurium, comprising the following steps: a) preparation of a mixture of starting compounds that contains molybdenum, vanadium, niobium, and a tellurium-containing starting compound in which tellurium is in oxidation state +4, b) hydrothermal treatment of the mixture of starting compounds at a temperature of 100° C. to 300° C. to obtain a product suspension, c) separation and drying of the solid from the product suspension resulting from step b), d) activation of the solid in inert gas to obtain the mixed-oxide material, wherein the tellurium-containing starting compound is selected from tellurium dioxide and a compound of formula M.sub.x.sup.n+TeO.sub.3 in which n=1 or 2 and x=2/n and in which M is an alkali metal or alkaline earth metal, the tellurium-containing starting compound having a particle size D.sub.90 smaller than 100 μm.

2. The method as claimed in claim 1, wherein the mixture of starting compounds is an aqueous suspension.

3. The method as claimed in claim 1, wherein the tellurium-containing starting compound is tellurium dioxide.

4. The method as claimed in claim 1, wherein one of the starting compounds is ammonium heptamolybdate or molybdenum trioxide.

5. The method as claimed in claim 1, wherein one of the starting compounds is ammonium metavanadate, vanadyl sulfate or vanadium pentoxide.

6. The method as claimed in claim 1, wherein one of the starting compounds is ammonium niobium oxalate, niobium oxalate or niobium oxide.

7. The method as claimed in claim 1, wherein the particle size D.sub.50 of the tellurium-containing starting compound selected from tellurium dioxide and a compound of formula M.sub.x.sup.n+TeO.sub.3 is smaller than 35 μm.

8. The method as claimed in claim 1, wherein the particle size D.sub.50 of the niobium-containing starting compound is smaller than 100 μm.

9. The method as claimed in claim 1, wherein the particle size D.sub.50 of the starting compounds used is smaller than 50 μm.

10. The method as claimed in claim 1, wherein the tellurium-containing starting compound is a compound of formula M.sub.x.sup.n+TeO.sub.3 in which n=1 or 2 and x=2/n and in which M is an alkali metal or alkaline earth metal.

11. The method as claimed in claim 1, wherein one of the starting compounds is molybdenum trioxide.

12. The method as claimed in claim 1, wherein one of the starting compounds is vanadium pentoxide.

13. The method as claimed in claim 1, wherein one of the starting compounds is niobium oxide.

14. The method as claimed in claim 1, wherein the product suspension resulting from step b) includes the mixed oxide material, which in the XRD, when using Cu-Kα radiation, has diffraction reflections h, i, k and l whose peaks are at the diffraction angles (2θ) 26.2°±0.5° (h), 27.0°±0.5° (i), 7.8°±0.5° (k) and 28.0°±0.5° (l).

Description

(1) The MoVNbTe mixed oxide obtained is used as catalyst material in the examples and in the experimental details is therefore described as catalyst in some cases.

(2) FIG. 1: Particle size distribution of the TeO.sub.2 used in example 1 with particle size values D.sub.10=7.625 μm, D.sub.50=15.140 μm, D.sub.90=27.409 μm.

(3) FIG. 2: XRD of the MoVNbTe mixed oxide from example 1.

(4) FIG. 3: Particle size distribution of the TeO.sub.2 used in comparative example 1 with particle size values D.sub.10=16.45 μm, D.sub.50=43.46 μm, D.sub.90=236.48 μm.

(5) FIG. 4: XRD of the MoVNbTe mixed oxide from comparative example 1.

(6) FIG. 5: Particle size distribution of the TeO.sub.2 used in example 2.

(7) FIG. 6: XRD of the mixed-oxide material from example 2.

(8) FIG. 7: Comparison of the particle size distribution of the Nb.sub.2O.sub.5 used in example 3 before and after milling.

(9) FIG. 8: XRD of the MoVNbTe mixed oxide from example 3.

(10) FIG. 9: XRD of the MoVNbTe mixed oxide from comparative example 3.

METHODS OF CHARACTERIZATION

(11) The following methods were used to determine the parameters of the MoVNbTe mixed oxides obtained:

(12) 1. BET Surface Area

(13) The determination is carried out according to the BET method described in DIN 66131; the BET method is also published in J. Am. Chem. Soc. 60, 309 (1938). The sample to be determined was dried in a U-shaped quartz reactor at 200° C. under an argon atmosphere (F=50 ml (min) for 1.5 h). The reactor was then cooled to room temperature, evacuated, and immersed in a Dewar vessel of liquid nitrogen. The nitrogen adsorption was carried out at 77 K with an RXM 100 sorption system (Advanced Scientific Design, Inc.).

(14) The BET surface area of the respective MoVNbTe mixed oxide samples was determined on material dried under vacuum at 200° C. The BET surface area data for the MoVNbTe mixed oxide in the present description likewise refer to the BET surface areas of the catalyst material used in each case (dried under vacuum at 200° C.).

(15) 2. X-Ray Powder Diffractometry (XRD)

(16) The X-ray diffractogram was obtained by X-ray powder diffractometry (XRD) and analyzed according to the Scherrer formula. The XRD spectra were measured on the catalyst materials activated in nitrogen at 600° C. Measurements were obtained on a Philips PW 3710-based PW 1050 Bragg-Brentano parafocusing goniometer at 40 kV and 35 mA using Cu-Kα radiation (wavelength=0.15418 nm), a graphite monochromator, and a proportional counter. The XRD scans were digitally recorded in increments of 0.04° (2 theta, 20). SiC was added as internal standard for phase quantification. Approximately 5% SiC was added for this, but the amount was weighed out accurately. This amount is stated in the phase analyses. Phase analysis was carried out by the Rietveld method using Topas software. The result of this phase analysis is shown in the XRD figures. The exact amount of the desired M1 phase was calculated by relating the proportion of the M1 phase in the total sample (as stated) to the sample without SiC.

(17) 3. Particle Size

(18) The particle size distribution was determined by the laser scattering method. This was done using a Malvern Mastersizer 2000. Analysis was according to the Fraunhofer method.

(19) The invention is now explained in more detail on the basis of the embodiments given as examples below, which should not be understood as restrictive.

EMBODIMENTS AS EXAMPLES

Example 1

(20) An autoclave (40 L) was charged with 3.3 L of distilled H.sub.2O and heated to 80° C. with stirring. At the same time, 725.58 g of ammonium heptamolybdate tetrahydrate (from HC Starck) was added and dissolved (AHM solution). Two 5 L beakers each filled with 1.65 L of distilled H.sub.2O were likewise heated to 80° C. with stirring on a temperature-controlled magnetic stirrer. To these beakers was then respectively added and dissolved 405.10 g of vanadyl sulfate hydrate (from GfE, V content 21.2%) and 185.59 g of ammonium niobium oxalate (HC Starck, Nb content: 20.6%) (V solution and Nb solution).

(21) In successive steps, the V solution was pumped into the AHM solution, then 65.59 g of solid TeO.sub.2 powder (TeO.sub.2 of 5N+ particle size distribution, see FIG. 1) and 1.65 L of distilled H.sub.2O were added, stirring was continued for 1 h at 80° C., and finally the Nb solution was pumped into the AHM solution using a peristaltic pump. Pumping time: V solution: 4.5 min at 190 rpm (tubing diameter: 8×5 mm), Nb solution: 6 min at 130 rpm (tubing diameter: 8×5 mm).

(22) The resulting suspension was stirred for a further 10 min at 80° C. The stirrer speed during the precipitation was 90 rpm.

(23) The suspension was then blanketed with nitrogen by pressurizing with nitrogen in the autoclave to a pressure of approximately 6 bar and opening the bleed valve so far that the autoclave was flushed under pressure with N.sub.2 (5 min). Finally, the pressure was released through the vent valve to a residual pressure of 1 bar.

(24) The hydrothermal synthesis in the 40 L autoclave was carried out at 175° C. for 20 h (heating time: 3 h) using an anchor stirrer at a stirrer speed of 90 rpm.

(25) After the synthesis, the solid was filtered off with the aid of a vacuum pump using a blue ribbon filter and the filter cake was washed with 5 L of distilled H20.

(26) The solid was dried at 80° C. in a drying oven for 3 days and then milled in a hammer mill, resulting in a yield of 0.8 kg of solid.

(27) Calcination was carried out at 280° C. for 4 h in an air stream (heating rate 5° C./min, air: 1 L/min).

(28) Activation was carried out in a retort at 600° C. for 2 h in an N.sub.2 stream (heating rate 5° C./min, N.sub.2: 0.5 L/min).

(29) The particle size distribution of the TeO.sub.2 used was:

(30) D.sub.10=7.625 μm D.sub.50=15.14 μm D.sub.90=27.409 μm

(31) Analytical characterization of the product:

(32) BET=15 m.sup.2/g

(33) XRD:

(34) The XRD of the mixed-oxide material from example 1 is shown in FIG. 2 and exhibits the following phase distribution:

(35) M1=90.50%

(36) M2=2.82%

(37) (Mo.sub.0.9V.sub.1.1)O.sub.5=1.15%

(38) SiC (standard)=5.53%

Comparative Example 1

(39) An autoclave (40 L) was charged with 6.6 L of distilled H.sub.2O and heated to 80° C. with stirring. At the same time, 1451.16 g of ammonium heptamolybdate tetrahydrate (HC Starck) was added and dissolved (AHM solution). Two 5 L beakers each filled with 3.3 L of distilled H.sub.2O were likewise heated to 80° C. with stirring on a temperature-controlled magnetic stirrer. To these beakers was then respectively added and dissolved 810.21 g of vanadyl sulfate hydrate (GfE, V content 21.2%) and 370.59 g of ammonium niobium oxalate (HC Starck, Nb content: 20.6%) (V solution and Nb solution).

(40) In successive steps, the V solution was pumped into the AHM solution, then 131.18 g of solid TeO.sub.2 powder (Alfa Aesar, particle size distribution FIG. 3) and 3.3 L of distilled H.sub.2O were added, stirring was continued for 1 h at 80° C., and finally the Nb solution was pumped into the AHM solution using a peristaltic pump. Pumping time: V solution: 5 min at 290 rpm (tubing diameter: 8×5 mm), Nb solution: 5 min at 275 rpm (tubing diameter: 8×5 mm).

(41) The resulting suspension was now stirred for a further 10 min at 80° C., the stirrer speed during the precipitation was 90 rpm.

(42) The suspension was then blanketed with nitrogen by pressurizing with nitrogen in the autoclave to a pressure of approximately 6 bar and opening the bleed valve so far that the autoclave was flushed under pressure with N.sub.2 (5 min). Finally, the pressure was released through the vent valve to a residual pressure of 1 bar.

(43) The hydrothermal synthesis in the 40 L autoclave was carried out at 175° C. for 20 h (heating time: 3 h) using an anchor stirrer at a stirrer speed of 90 rpm.

(44) After the synthesis, the solid was filtered off with the aid of a vacuum pump using a blue ribbon filter and the filter cake was washed with 5 L of distilled H.sub.2O. The filtration lasted several days.

(45) The solid was dried at 80° C. in a drying oven for 3 days and then milled in a hammer mill, resulting in a yield of 0.5 kg of solid.

(46) The yield in comparative example 1 is only about half that in the example according to the present invention.

(47) Calcination was carried out at 280° C. for 4 h in an air stream (heating rate 5° C./min, air: 1 L/min).

(48) Activation was carried out in a retort at 600° C. for 2 h in an N.sub.2 stream (heating rate 5° C./min, N.sub.2: 0.5 L/min).

(49) Particle size values of the TeO.sub.2 used:

(50) D.sub.10=16.45 μm D.sub.50=43.46 μm D.sub.90=236.48 μm

(51) The XRD of the MoVNbTe mixed oxide from comparative example 1 is shown in FIG. 4 and exhibits the following phase distribution:

(52) M1=51.88%

(53) M2=8.12%

(54) (Mo.sub.0.9V.sub.1.1)O.sub.5=12.51%

(55) (V.sub.0.35Mo.sub.4.65)O.sub.14=23.19%

(56) SiC (standard)=3.59%

Example 2

(57) An autoclave (40 L) was charged with 3.3 L of distilled H.sub.2O and heated to 80° C. with stirring. At the same time, 725.58 g of ammonium heptamolybdate tetrahydrate (HC Starck) was added and dissolved (AHM solution). Two 5 L beakers each filled with 1.65 L of distilled H.sub.2O were likewise heated to 80° C. with stirring on a temperature-controlled magnetic stirrer. To these beakers was then respectively added and dissolved 405.10 g of vanadyl sulfate hydrate (GfE, V content 21.2%) and 185.59 g of ammonium niobium oxalate (HC Starck, Nb content: 20.6%) (V solution and Nb solution).

(58) On the previous day, 65.59 g of TeO.sub.2 (Alpha Aesar from comparative example 1) was milled for 3 h in 200 g of distilled H.sub.2O (Retsch PM100 ball mill) and transferred to a beaker with 1.45 L of distilled H.sub.2O (particle size after milling see FIG. 5).

(59) In successive steps, the V solution was pumped into the AHM solution, then the Te suspension milled on the previous day was added, stirring was continued for 1 h at 80° C., and finally the Nb solution was pumped into the AHM solution using a peristaltic pump. Pumping time: V solution: 5 min at 290 rpm (tubing diameter: 8×5 mm), Nb solution: 5 min at 275 rpm (tubing diameter: 8×5 mm).

(60) The resulting suspension was now stirred for a further 10 min at 80° C., the stirrer speed during the precipitation was 90 rpm.

(61) The suspension was then blanketed with nitrogen by pressurizing with nitrogen in the autoclave to a pressure of approximately 6 bar and opening the bleed valve so far that the autoclave was flushed under pressure with N.sub.2 (5 min). Finally, the pressure was released through the vent valve to a residual pressure of 1 bar.

(62) The hydrothermal synthesis in the 40 L autoclave was carried out at 175° C. for 20 h (heating time: 3 h) using an anchor stirrer at a stirrer speed of 90 rpm.

(63) After the synthesis, the solid was filtered off with the aid of a vacuum pump using a blue ribbon filter and the filter cake was washed with 5 L of distilled H20.

(64) The solid was dried at 80° C. in a drying oven for 3 days and then milled in a hammer mill, resulting in a yield of 0.8 kg of solid.

(65) Calcination was carried out at 280° C. for 4 h in an air stream (heating rate 5° C./min, air: 1 L/min).

(66) Activation was carried out in a retort at 600° C. for 2 h in an N.sub.2 stream (heating rate 5° C./min, N.sub.2: 0.5 L/min).

(67) The particle size values of the TeO.sub.2 milled for 3 h were:

(68) D.sub.10=0.569 μm D.sub.50=2.992 μm D.sub.90=6.326 μm

(69) Analytical characterization of the product:

(70) BET=12 m.sup.2/g

(71) The XRD of the MoVNbTe mixed oxide from example 2 is shown in FIG. 6 and exhibits the following phase distribution:

(72) M1=86.30%

(73) M2=2.78%

(74) (Mo.sub.0.9V.sub.1.1)O.sub.5=0.75%

(75) (V.sub.0.35Mo.sub.4.65)O.sub.14=3.75%

(76) SiC (standard)=5.01%

Example 3

(77) First, TeO.sub.2 (Alpha Aesar from comparative example 1) was slurried in 200 g of distilled H.sub.2O and milled in a ball mill (as in example 2). The portion was then transferred to a beaker with 500 ml of distilled H.sub.2O. The Nb.sub.2O.sub.5 was slurried in 200 g of distilled H.sub.2O and milled in the same ball mill. A comparison of the particle size distributions before and after milling is shown in FIG. 7.

(78) The portion was then transferred to a beaker with 500 ml of distilled H.sub.2O. The next morning it was heated to 80° C., 107.8 g of oxalic acid dihydrate was added to the Nb.sub.2O.sub.5 suspension, and the suspension was stirred for approximately 1 h. An autoclave (40 L) was charged with 6 L of distilled H20 and heated to 80° C. with stirring. Once the water had reached temperature, 61.58 g of citric acid, 19.9 g of ethylene glycol, 615.5 g of MoO.sub.3 (Sigma Aldrich), 124.5 g of V.sub.2O.sub.5, the milled TeO.sub.2, and the milled Nb.sub.2O.sub.5 in oxalic acid were successively added. 850 ml of distilled H.sub.2O was added to aid transfer and to rinse out the vessels. The total volume of water in the autoclave was 8.25 L and the stirrer speed was 90 rpm. The contents were then blanketed with nitrogen. The hydrothermal synthesis in the 40 L autoclave was carried out at 190° C. for 48 h. After the synthesis, the solid was filtered off with the aid of a vacuum pump using a blue ribbon filter and the filter cake was washed with 5 L of distilled H20.

(79) The solid was dried at 80° C. in a drying oven for 3 days and then milled in a hammer mill, resulting in a yield of 0.8 kg of solid.

(80) Calcination was carried out at 280° C. for 4 h in an air stream (heating rate 5° C./min, air: 1 L/min).

(81) Activation was carried out in a retort at 600° C. for 2 h in an N.sub.2 stream (heating rate 5° C./min, N.sub.2: 0.5 L/min).

(82) The XRD of the MoVNbTe mixed oxide from example 3 is shown in FIG. 8 and exhibits the following phase distribution:

(83) M1=85.79%

(84) M2=1.95%

(85) (Mo.sub.0.9V.sub.1.1)O.sub.5=1.43%

(86) MoO.sub.3=3.31%

(87) Nb.sub.2O.sub.5=2.86%

(88) SiC (standard)=4.66%

Comparative Example 2

(89) First, TeO.sub.2 (Alfa Aesar from comparative example 1) was slurried in 200 g of distilled H.sub.2O and milled in a ball mill (as in example 2) and then transferred to a beaker with water so that the volume of water in the beaker was 1650 ml.

(90) An autoclave (40 L) was charged with 6.6 L of distilled H.sub.2O and heated to 80° C. with stirring. As soon as the water had reached temperature, 61.58 g of citric acid, 194 g of oxalic acid dihydrate, 19.9 g of ethylene glycol, 615.5 g of MoO.sub.3 (Sigma Aldrich), 124.5 g of V.sub.2O.sub.5, the milled TeO.sub.2, and 56.8 g of Nb.sub.2O.sub.5 (unmilled with the particle size distribution from FIG. 7, which also shows particles above 100 μm) were successively added. The contents were then blanketed with nitrogen. The hydrothermal synthesis in the 40 L autoclave was carried out at 190° C. for 48 h. After the synthesis, the solid was filtered off with the aid of a vacuum pump using a blue ribbon filter and the filter cake was washed with 5 L of distilled H20.

(91) The solid was dried at 80° C. in a drying oven for 3 days and then milled in a hammer mill, resulting in a yield of 0.8 kg of solid.

(92) Calcination was carried out at 280° C. for 4 h in an air stream (heating rate 5° C./min, air: 1 L/min). Activation was carried out in a retort at 600° C. for 2 h in an N.sub.2 stream (heating rate 5° C./min, N.sub.2: 0.5 L/min).

(93) The XRD of the MoVNbTe mixed oxide from comparative example 3 is shown in FIG. 9 and exhibits the following phase distribution:

(94) M1=17.34%

(95) M2=1.75%

(96) (V.sub.0.35Mo.sub.4.65)O.sub.14=34.35%

(97) MoVO.sub.5=24.57%

(98) TeMo.sub.6O.sub.16=17.39%

(99) SiC (standard)=4.6%

(100) It can be seen clearly that only 17% of M1 phase was obtained if unmilled niobium oxide that was not reacted first with oxalic acid was used.