Manganese-doped nickel methanization catalysts having elevated sulphur resistance

Abstract

A process for the methanation of carbon monoxide and/or carbon dioxide in a feed stream containing carbon monoxide and/or carbon dioxide is disclosed. This is achieved by a process for the methanation of carbon monoxide and/or carbon dioxide in a feed stream containing carbon monoxide and/or carbon dioxide, hydrogen and more than 1 ppb of sulfur, using a catalyst comprising aluminum oxide, an Ni active composition and Mn. It has surprisingly The Mn-containing Ni catalyst has a high sulfur resistance and also a high sulfur capacity.

Claims

1. A process for the methanation of carbon monoxide and/or carbon dioxide, comprising contacting a feed stream containing carbon monoxide and/or carbon dioxide, hydrogen and more than 4 ppb of sulfur with a catalyst comprising aluminum oxide, an Ni active composition and Mn, wherein the molar ratio of Ni/Mn in the catalyst is in the range from 1.0 to 15.0.

2. The process as claimed in claim 1, wherein the catalyst has been produced by coprecipitation.

3. The process as claimed in claim 1, wherein the molar ratio of Ni/Mn in the catalyst is in the range from 2.0 to 6.0.

4. The process as claimed in claim 1, wherein the molar ratio of Al/Ni in the catalyst is in the range from 0.1 to 0.9.

5. The process as claimed in claim 1, wherein the catalyst has been produced by impregnation of aluminum oxide with a solution comprising Ni.

6. The process as claimed in claim 5, wherein the solution comprising Ni also contains Mn.

7. The process as claimed in claim 5, wherein the molar ratio of Ni/Mn in the catalyst is in the range from 6.0 to 10.0.

8. The process as claimed in claim 5, wherein the molar ratio of Al/Ni in the catalyst is in the range from 2 to 9.

9. The process as claimed in claim 1, wherein the feed stream contains more than 10 ppb of sulfur.

10. The process as claimed in claim 1, wherein the feed stream contains from 4 ppb to 100 ppm of sulfur.

11. The process as claimed in claim 1, wherein the Ni active composition has crystallites having a diameter below 20 nm.

12. The process as claimed in claim 1, wherein the catalyst has a CO.sub.2 uptake capacity at 35° C. of greater than 200 μmol/g.

13. The process as claimed in claim 1, wherein the catalyst is brought into contact with the feed stream at a temperature above 150° C.

14. The process as claimed in claim 1, wherein the catalyst absorbs at least 90% of the sulfur present in the feed stream.

15. The process as claimed in claim 1, wherein the molar ratio of Ni/Mn in the catalyst is in the range from 6.0 to 10.0, and the molar ratio of Al/Ni in the catalyst is in the range from 2 to 9.

16. The process as claimed in claim 1, wherein the molar ratio of Ni/Mn in the catalyst is in the range from 6.0 to 10.0, and the molar ratio of Al/Ni in the catalyst is in the range from 2.3 to 5.

17. A process for the methanation of carbon monoxide and/or carbon dioxide, comprising contacting a feed stream containing carbon monoxide and/or carbon dioxide, hydrogen and more than 1 ppb of sulfur with a catalyst comprising aluminum oxide, an Ni active composition and Mn, wherein the molar ratio of Ni/Mn in the catalyst is in the range from 2.0 to 6.0, and the molar ratio of Al/Ni in the catalyst is in the range from 0.1 to 0.9.

18. The process as claimed in claim 1, wherein the molar ratio of Ni/Mn in the catalyst is in the range from 3.5 to 5.5.

19. The process as claimed in claim 2, wherein the feed stream contains more than 10 ppb of sulfur.

20. The process as claimed in claim 1, wherein the feed stream contains from 4 ppb to 100 ppm of sulfur.

Description

(1) FIG. 1: decrease in the activity of the catalysts A to C during the catalytic test reaction (Test 1).

(2) FIG. 2: increase in mass of the catalysts A to C due to uptake of sulfur during the catalytic test reaction, normalized to the weight of catalyst (Test 1).

(3) FIG. 3: decrease in the activity as a function of the uptake of sulfur during the catalytic test reaction (Test 1, with the assumption that nickel sulfide is formed).

(4) FIG. 4: decrease in the activity of the catalysts A and B during the catalytic test reaction (Test 2).

METHODS

Elemental Analysis

(5) The determination of the composition of the calcined catalysts was carried out by optical emission spectroscopy by means of inductively coupled plasma (ICP-OES). 50 mg of catalyst were dissolved in 50 ml of 1 molar phosphoric acid (VWR, A.R.) at 60° C. In order to dissolve manganese dioxide formed, 50 mg of Na.sub.2SO.sub.3 (Sigma Aldrich, A.R.) were added to the solution. After cooling, the solutions were diluted by a factor of 1/10, admixed to the same concentration with Na.sub.2SO.sub.3 and filtered by means of 0.1 μm filters (Pall). The calibration solutions were made up with concentrations of 1, 10 and 50 mg l.sup.−1 (Merck). Determination of the metal concentrations was carried out by means of an Agilent 700 ICP-OES.

Determination of the Specific Surface Area

(6) The determination of the specific surface areas of the catalysts (S.sub.BET) was carried out by means of N.sub.2-BET analysis on a NOVA 4000e (Quantachrome). For this purpose, 100 mg of catalyst were degassed at 120° C. for 3 hours and adsorption and desorption isotherms were subsequently recorded in the p/p.sub.0 range from 0.007 to 1. To determine the BET surface area, the data points in the p/p.sub.0 range from 0.007 to 0.28 were employed.

Hg Pore Volume

(7) The pore distribution and the pore volume of the catalyst particles were determined using a mercury porosimeter: Pascal 440 from Thermo Electron Corporation in accordance with DIN 66133. Here, the sample was evacuated beforehand for 30 minutes at room temperature. Samples in the range from 600 to 900 mg were measured and the pressure was increased to 2000 bar.

Chemisorption

(8) Chemisorption experiments were carried out on an Autosorb 1C (Quantachrome). Before the measurement, 100 mg of catalyst were activated at 500° C. in 10% of H.sub.2 in N.sub.2 for 6 hours. The heating ramp was 2 Kmin.sup.−1.

(9) The determination of the metal surface area (S.sub.MET) was effected in accordance with DIN 66136-2 (version 2007-01) and was carried out by means of H.sub.2 chemisorption at 35° C. For this purpose, 20 adsorption points were recorded equidistantly from 40 mmHg to 800 mmHg. The equilibration time for the adsorption was 2 minutes, and that for thermal equilibrium was 10 minutes. To determine the metal surface area, a metal atom/H stoichiometry of 1 was assumed. For CO.sub.2 chemisorption measurements, the equilibration time for the adsorption was set to 10 minutes with otherwise unchanged parameters. Before recording of the chemisorption data, a possible kinetic inhibition of the CO.sub.2 chemisorption under these conditions was experimentally ruled out. Metal surface areas and CO.sub.2 uptake capacities were extrapolated according to the extrapolation method to a pressure of 0 mmHg.

Synthesis

(10) The catalysts A and B were produced by coprecipitation at a loading with nickel of 50% by weight, which in the case of catalyst A leads to a molar ratio of aluminum to nickel of 0.75 and in the case of the manganese-containing catalyst B leads to a molar ratio of Al/Ni of 0.47. To examine the effect of manganese on the catalyst behavior, manganese(II) nitrate was added to the salt solution of nickel nitrate and aluminum nitrate during the catalyst synthesis. The purity of all chemicals used was A.R. Water was purified by means of a Millipore filter system and the purity was verified by means of conductivity measurements. The synthesis was carried out in a double-walled, 3 l capacity stirred vessel. The double wall filled with water allowed, with the aid of a thermostatic bath, maintenance of the temperature of the synthesis mixture as 30° C., and two baffles ensured improved mixing. A precision glass stirrer at 150 revolutions per minute was used for stirring. For the synthesis, 1 l of H.sub.2O was placed in the stirred vessel and set to a pH=9±0.1. The mixture of the dissolved nitrates was metered in at 2.5 ml min.sup.−1. At the same time, controlled addition of the precipitation reagent served to maintain the pH. As starting materials, use was made of one molar solutions of the respective nitrates (Ni(NO.sub.3).sub.2*6H.sub.2O, Al(NO.sub.3).sub.2*9H.sub.2O and Mn(NO.sub.3).sub.2*4H.sub.2O). For catalyst B, these were mixed in a molar ratio of Ni to Mn of 4.6:1 and of Al to Ni of 0.45 to give a total volume of 120 ml min.sup.−1, before dropwise introduction into the reactor was carried out. A mixture of equal volumes of the solutions 0.5M NaOH and 0.5M Na.sub.2CO.sub.3, which were metered using a titrator, served as precipitation reagent. The suspension was aged overnight in the mother liquor while stirring continually, the precipitate was subsequently filtered off and washed with H.sub.2O until the filtrate had a neutral pH. After drying at 80° C. overnight in a drying oven, the dried precipitate (precursor) was heated at a heating rate of 5 K min.sup.−1 to 450° C. and calcined under synthetic air for 6 hours.

(11) The catalyst C was produced by triple impregnation with subsequent calcination in each case. 3022.0 g of Ni(NO.sub.3).sub.2×6H.sub.2O (98%), 307.9 g of Mn(NO.sub.3).sub.2×4H.sub.2O (98.5%) were placed in a 5 l glass beaker. The mixture was made up to a volume of about 2900 ml with deionized water and stirred by means of a propeller stirrer until a clear solution had been obtained. The solution was then made up to the required total volume of 3441 ml with deionized water.

(12) First impregnation: 1879.3 g of gamma aluminum oxide ⅛″ extrudate (corresponding to 3.6 l; bulk density: 522 g/l; loss on ignition: 4.22%) were placed in a closable vessel. 1349 ml of impregnation solution, which corresponds to the maximum amount of solution at which a supernatant solution is not yet formed, were added slowly and in small amounts. In between, the vessel was always closed again and shaken/homogenized. After the entire solution had been added, the mixture was shaken for a further two minutes.

(13) First calcination: the impregnated extrudates were transferred to porcelain dishes and heated at 2° C./min to 120° C. and dried in air at this temperature for 6 hours. The impregnated extrudates were subsequently heated at 2° C./min to 240° C. and calcined at this temperature in air for 4 hours.

(14) Second impregnation: the extrudates which had been impregnated once were placed in a closable vessel. 1147.0 ml of impregnation solution, which corresponds to the maximum amount of solution at which a supernatant solution is not yet formed, were added slowly and in small amounts. In between, the vessel was always closed again and shaken/homogenized. When the entire solution had been added, the mixture was shaken for a further two minutes.

(15) Second calcination: the impregnated extrudates were transferred to porcelain dishes and heated at 2° C./min to 120° C. and dried at this temperature in air for 6 hours. The impregnated extrudates were then heated at 2° C./min to 240° C. and calcined at this temperature in air for 4 hours.

(16) Third impregnation: the extrudates which had been impregnated twice were placed in a closable vessel. 944.3 ml of impregnation solution, which corresponds to the maximum amount of solution at which a supernatant solution is not yet formed, were added slowly and in small amounts. In between, the vessel was always closed again and shaken/homogenized. When the entire solution had been added, the mixture was shaken for a further two minutes.

(17) Third calcination: the impregnated extrudates were transferred to porcelain dishes and heated at 2° C./min to 120° C. and dried at this temperature in air for 6 hours. The impregnated extrudates were then heated at 2° C./min to 240° C. and calcined at this temperature in air for 4 hours.

Thermogravimetric Analysis and Catalytic Test Reaction

(18) Thermogravimetric analysis (TGA) was used to examine the deactivation by hydrogen sulfide. Here, the catalyst bed to be examined is introduced into a heated reactor (1.53 ml) through which forced flow occurs and supplied with feed gases (79.5% by volume of H.sub.2, 20.5% by volume of CO.sub.2; H.sub.2S content of starting material 43 ppm at T=270° C., p=6 barg and SV (“space velocity” about 16 000 1/h). The flow behavior through the bed resembles that of tube reactors. Mass changes during the catalytic process can be detected by means of discontinuous, contactless weighing (precision=±10 μg) via a magnetic suspension coupling. At the same time, the catalytic activity was evaluated by determining the product gas composition. For this purpose, a substream of the product gas was analyzed using a mass spectrometer. The results are shown under “Catalytic test reactions”, “Test 1” and “Test 2”. The experimental conditions were kept constant during the series of experiments to ensure comparability of the various catalysts examined. Since the catalysts are obtained in oxidic form from the synthesis, they firstly have to be reduced in a stream of H.sub.2 in order to produce the catalytically active, metallic phases before commencement of the tests. Reduction is carried out until the weight remains constant. The reported weights of the catalysts used relate to this weight after reduction, which is lower than the original weight due to the conversion of NiO into Ni and MnO.sub.2 into Mn:

(19) Test 1: catalyst A: 1.27 g, catalyst B: 0.93 g, catalyst C: 0.90 g. The normalized H.sub.2 conversion (0 corresponds to 0% conversion, 1 corresponds to 100% conversion) is shown.

(20) Test 2: catalyst A: 1.02 g, catalyst B: 1.05 g of catalyst

EXAMPLES

Example 1: Synthesis of the Catalysts

(21) Three catalyst samples were prepared in accordance with the synthesis described in the method part, with catalyst A being a comparative sample without manganese, while the catalysts B and C contain manganese. The three catalysts have the properties summarized in Table 1.

(22) TABLE-US-00001 TABLE 1 Analytical data for the catalysts examined Ni Mn Surface Crystallite Ni Pore S uptake Weight [% by [% by area of Ni size dispersion volume [% by Catalyst used Catalyst weight] weight] [m.sup.2/g] [A°] [%] [mm.sup.3/g] weight] shape [g] A 50 — 35 79 10 210 1.5 1.8 × 3.6 TEST 1: 1.27 (comparison) pellet TEST 2: 1.02 B 50 10 51 55 15 255 5.1 3 × 3 TEST 1: 0.93 pellet TEST 2: 1.05 C 22 2.5 32 39 21 233 3.2 ⅛″ TEST 1: 0.9  extrudate

Example 2: Catalytic Test Reactions

Test 1

(23) The results from Test 1 are shown in FIG. 1. The H.sub.2 conversion decreases with increasing time on stream and catalyst poisoning for all catalysts examined. In the case of catalyst A and catalyst B, each having a loading of 50% by weight of Ni, the activity decrease is very similar, but the weight of Mn-promoted catalyst B at 0.93 g compared to 1.27 g for catalyst A was significantly smaller and the catalytic results are not normalized to the weight used. In the case of catalyst C, the activity decreases more quickly because of the nickel loading which is only about half as great.

(24) Test 2

(25) The results from Test 2 are shown in FIG. 4. The CO.sub.2 conversion decreases with increasing time on stream and catalyst poisoning for both catalysts examined. In contrast to the results in FIG. 1, the same catalyst weight of 1.0 g and the same size of catalyst particles were employed for the two catalysts A and B for the purpose of better comparability. The weight of catalyst A and of catalyst B was 1.02 and 1.05 g, respectively. Catalyst B displays significantly improved deactivation behavior. According to the result from Test 2, the additional amount of 10% by weight of manganese in catalyst B thus leads to an approximate doubling of the time to deactivation of catalyst B compared to catalyst A.

Example 3: Sulfur Uptake During the Catalytic Test Reaction

(26) During the catalytic test reactions in example 2, the uptake of sulfur by all three catalyst samples A, B and C was in each case measured at the same time. This can be seen from the measurement data in FIG. 2; the sulfur uptake was normalized to the catalyst weight. Surprisingly, the uptake of sulfur by catalyst B is significantly higher than in the case of catalyst A. Even in the case of catalyst C, the sulfur uptake is increased compared to the reference catalyst catalyst A without manganese, even though catalyst C has a significantly lower nickel loading. The additional content of manganese of the two catalysts B and C surprisingly leads to an increase in the sulfur uptake.

(27) This observation is also confirmed by the analytical determination of the sulfur content of the catalysts after complete deactivation. The analytical results for the S uptake from the TGA test and the chemical analysis are summarized in table 2. While the weighing of the TGA analysis gives an integrated value for the amount of S taken up, the used catalyst after Test 2 was complete was taken out in layers. By means of subsequent chemical analysis, the S uptake can then be determined in a positionally resolved manner and a corresponding loading profile can be determined. It can be clearly seen here that the sulfur uptake has the highest value in the upper part of the reactor crucible of the TGA, and the loading with sulfur decreases gradually in the direction of the reactor outlet. The chemical analysis was carried out by inductive combustion of the sample in a stream of oxygen to form SO.sub.2 and subsequent, quantitative infrared analysis of a characteristic SO.sub.2 band.

(28) TABLE-US-00002 TABLE 2 Determination of the sulfur uptake by means of TGA analysis and subsequent chemical analysis of the used catalysts after removal from the experimental reactor. Catalyst Catalyst Catalyst Unit A B C Weight of reduced catalyst Test 1 [g] 1.27 0.93 0.90 Test 2 1.0 1.0 / Nickel content [% by 50.00 50.00 22.50 weight] Molar amount of nickel [g] 0.64 0.47 0.20 Amount of nickel [mmol] 10.82 7.92 3.45 Amount of S [mmol] 1.09 1.58 0.78 Increase in mass due to uptake of S after 50 h [mg] 35 58 25 reaction time from TGA analysis Test 1.sup.1 S loading according to chem. analysis Test 1.sup.2 [%] 1.50 5.10 3.20 sampling from entire crucible [mg] 19 47 29 S loading according to chem. analysis Test 2.sup.2 [%] 8.1 12.2 — sampling from upper layer of the reactor [mg 82 128 crucible S loading according to chem. analysis Test 2.sup.2 [%] 3.8 5.2 — sampling from middle layer of the reactor [mg 39 54 crucible S loading according to chem. analysis Test 2.sup.2 [%] 2.7 2.6 — sampling from bottom layer of the reactor [mg] 27.5 27.5 crucible S loading according to chem. analysis Test 2.sup.2 [%] 4.9 6.7 — average [mg] 49 70 .sup.1Values from TGA .sup.2Values from chemical analysis

(29) In the results presented in FIG. 1, the different weights of catalyst in Test 1 were not taken into account. In order to correct this, the results presented in FIG. 3 were normalized to the weight used. The results presented in FIG. 3 show that the catalysts B and C are more suitable for use in sulfur-containing gas streams. In FIG. 3, the x axis corresponds to the uptake of sulfur relative to the mass of nickel present and the y axis corresponds to the catalytic activity. It was postulated here that the sulfur taken up is bound as NiS, but this is not confirmed by analysis but was merely employed for illustration. As FIG. 3 shows, the catalyst B according to the invention displays a higher activity than the catalyst A at the same sulfur loading, while catalyst C displays a similar activity.