JP2016017004A - Co selective methanization reactor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a CO selective methanization reactor that can maintain the outlet CO concentration at a lower value over a long period of time even for the case where the inlet CO concentration is high.SOLUTION: According to the present invention, provided is a CO selective methanization reactor for methanizing CO in a hydrogen rich gas containing CO and COand in which the reactor includes a front stage catalyst and a rear stage catalyst in order from the upstream side of the hydrogen rich gas flow within the CO selective methanization reactor and said front stage catalyst has an optimal working temperature at inlet CO concentration of 0.5% higher than the rear stage catalyst.SELECTED DRAWING: Figure 4

Description

  The present invention relates to a CO selective methanation reactor applicable to a CO removal process such as a fuel reformer for a household polymer electrolyte fuel cell.

  Since the polymer electrolyte fuel cell is operated at a low temperature of about 80 ° C., if the hydrogen rich gas as the fuel contains carbon monoxide at a certain level or more, the power generation performance is reduced due to CO poisoning of the anode platinum catalyst. There will be a problem that it will drop or eventually no power can be generated at all.

In order to avoid this CO poisoning, in a domestic polymer electrolyte fuel cell power generation system that uses city gas, LP gas, kerosene or the like after being converted to hydrogen-rich gas by a fuel reformer, CO in the fuel cell anode inlet gas It is desirable to always keep the concentration below 10 ppm. Many actual systems employ a CO selective oxidation catalyst that mixes air with the product gas at the final stage of the fuel reforming process and oxidizes CO contained in the gas to CO ₂ .
CO + 1/2 O ₂ = CO ₂ (Scheme 1)

  In this catalyst, as shown in Reaction Formula 1, since it is necessary to constantly take in air from the outside, an air blower, its control system, and a complicated gas mixing structure for uniformly mixing the supplied air with the reaction gas are provided. It is necessary to install it in the fuel reformer.

  Recently, a CO selective methanation catalyst has attracted attention as a new method to replace this CO selective oxidation catalyst (for example, Patent Documents 1 and 2).

In Patent Document 1, a non-stoichiometric Ni-Al composite oxide precursor produced by a spray plasma method is impregnated with a ruthenium salt and subjected to a reduction treatment. _A CO selective methanation catalyst capable of selectively causing a CO methanation reaction even in a high temperature region where a _2- methanation reaction and a reverse aqueous shift reaction proceed predominantly is disclosed.

  In Patent Document 2, the active component of a catalyst for methanating carbon monoxide is obtained by adsorbing or binding at least one selected from halogen, an inorganic acid, and a metal oxyacid, which are carbon dioxide reaction inhibitors. A CO selective methanation catalyst excellent in selectivity of a CO methanation reaction is disclosed.

  The catalysts disclosed in Patent Documents 1 and 2 are excellent in the selectivity of the CO methanation reaction. However, when further research was carried out for practical use of such a catalyst, the catalysts of these documents were used. When used, it was found that the catalyst is likely to deteriorate (catalytic activity decreases) particularly when the space velocity of the reaction gas is large. In addition, when the CO selective oxidation catalyst was used, the CO concentration of the outlet gas (hereinafter referred to as “outlet CO concentration”) was easy to be 10 ppm or less, but when the CO selective methanation catalyst was used, Only in a very narrow temperature range, the outlet CO concentration could be reduced to 10 ppm or less.

  In order to solve such a problem, in Patent Document 3, a coating layer having a large number of pores is provided so as to cover the surface of the supported metal catalyst, thereby reducing the CO concentration on the surface of the supported metal catalyst. As a result, the decrease in the catalytic activity is greatly suppressed, and the outlet CO concentration can be reduced to 10 ppm or less over a wide temperature range.

WO2010 / 122855 WO2011 / 142481 WO2014 / 038426

  By the way, the CO selective oxidation catalyst used in the fuel reformer of the current household fuel cell system has a durability to purify a CO concentration of 0.3 to 0.5% to 10 ppm or less over a cumulative period of 60,000 hours or more. have. Development of materials for CO selective methanation catalysts is also under the premise of this, but no practical catalyst that can withstand 60,000 hours of use has been developed, and even the catalyst disclosed in Patent Document 3 can be used for 60,000 hours. Can't withstand

  On the other hand, if the CO concentration that can be processed is increased from 0.5% to 1.0%, the amount of CO conversion catalyst in the previous stage can be greatly reduced, so that the reformer can be downsized and further cost reduction can be expected. It is expected as a catalyst for reformers. A CO selective methanation catalyst capable of maintaining performance under such high concentration CO for a long time has not yet been found.

  The present invention has been made in view of such circumstances, and even when the inlet CO concentration is high, the CO selective methanation that can maintain the outlet CO concentration at a low value over a long period of time. A reactor is provided.

According to the present invention, a CO selective methanation reactor for methanating CO in a hydrogen rich gas containing CO and CO ₂ , upstream of the flow of the hydrogen rich gas in the CO selective methanation reactor Are provided with a pre-stage catalyst and a post-stage catalyst in order, and the pre-stage catalyst is provided with a CO selective methanation reactor having an optimum operating temperature higher than that of the post-stage catalyst at an inlet CO concentration of 0.5%.

Production Example 2 of Patent Document 3 discloses a catalyst in which a coating layer made of mesoporous silica is formed on the surface of a Ni / Al-VOx catalyst (hereinafter referred to as “MS-coated catalyst”). When went, as shown in FIG. 1 (a), about 5000 hours under the conditions of a space velocity 2400h ^-1, for about 500 hours under the conditions of 4800H ^-1, outlet CO concentration was maintained at 10ppm or less.
On the other hand, paragraphs 70 to 71 of Patent Document 3 disclose a catalyst in which Ni—Fe is supported on a support made of mesoporous silica (hereinafter referred to as “Ni—Fe / MS catalyst”). As a result of the test, as shown in FIG. 1B, a result was obtained that there was almost no deterioration even under the condition of a space velocity of 10,000 h ⁻¹ . On the other hand, the outlet CO concentration was a relatively high value of about 600 ppm.
Therefore, in order to take advantage of both the advantage of the MS-coated catalyst having a high CO removal rate and the advantage of the Ni-Fe / MS catalyst having a long lifetime, a catalyst having Ni-Fe supported on the coating layer of the MS-coated catalyst (hereinafter referred to as the catalyst) "Ni-Fe / MS coated catalyst")) was prepared and evaluated. The lifetime of the Ni—Fe / MS coated catalyst was expected to be significantly improved over that of the MS coated catalyst, but actually, as shown in FIG. 1 (c), the space velocity was 4800 h ⁻¹ . The time required for the outlet CO concentration to reach 10 ppm only increased to about 1000 hours, and the expected improvement in life could not be achieved.

In order to identify the cause of the catalyst life not extending as expected, the relationship between the operating temperature and the outlet CO concentration was examined for the MS-coated catalyst and the Ni-Fe / MS catalyst. The results are shown in FIGS. As shown in FIGS. 2A to 2B, the MS-coated catalyst and the Ni—Fe / MS catalyst have a temperature at which the outlet CO concentration is minimized (that is, a temperature at which the CO removal rate is maximized) (hereinafter, “ This is referred to as “optimum operating temperature.”) It was found that T _opt was significantly different. And since the optimum operating temperature T _opt of these catalysts is greatly different, even in the Ni-Fe / MS coated catalyst that combines them, the advantages of both catalysts cannot be fully utilized, and the expected lifetime It was found that no improvement was achieved.

  In view of this, the present inventors have intensively studied a method for effectively combining a plurality of types of catalysts having different optimum operating temperatures. As a result, the CO selective methanation reactor has a downstream flow from the upstream side to the downstream side of the flow of hydrogen rich gas flowing through the reactor. There is a temperature gradient that decreases towards the side, placing the catalyst with the highest optimum operating temperature on the upstream side and the catalyst with the lower optimum operating temperature on the downstream side. It is possible to operate at or near the operating temperature, and as a result, even when the inlet CO concentration is high, it has been found that the outlet CO concentration can be maintained at a low value over a long period of time. It was completed.

Hereinafter, various embodiments of the present invention will be exemplified. The embodiments described below can be combined with each other.
Preferably, the post-stage catalyst has a CO selectivity of 0.5 or more under the condition that the outlet CO concentration is 100 ppm.
Preferably, the upstream catalyst is configured such that the outlet CO concentration at the optimum operating temperature is 0.2% or less.
Preferably, the active metal of the pre-stage catalyst includes Ni and Fe.
Preferably, the active metal of the latter catalyst includes Ni, and the support of the latter catalyst includes V oxide and Al oxide.
According to another aspect of the present invention, a CO selective methanation reactor for methanating CO in a hydrogen rich gas containing CO and CO 2, wherein the flow of the hydrogen rich gas in the CO selective methanation reactor In order from the upstream side of the catalyst, the active catalyst of the former catalyst contains Ni and Fe, the active metal of the latter catalyst contains Ni, and the support of the latter catalyst is V or Ni. A CO selective methanation reactor, which is an oxide containing Al, is provided.
Preferably, the atomic ratio of Fe / Ni is 0.01-1.
Preferably, the support of the pre-stage catalyst is an oxide containing Si.
Preferably, the latter catalyst includes a coating layer configured to reduce the CO concentration on the active metal surface.
Preferably, the pre-stage catalyst is coated on a honeycomb substrate.
Preferably, the honeycomb substrate is a metal honeycomb substrate.
Preferably, an α-alumina layer is provided on the surface of the metal honeycomb substrate.

The structure of a conventional CO methanation catalyst and the result of a life test are shown, (a) shows an MS-coated catalyst, (b) shows a Ni-Fe / MS catalyst, and (c) shows a Ni-Fe / MS-coated catalyst. It is a graph for demonstrating optimal operation temperature _Topt , (a) shows the result about a MS covering catalyst, (b) shows the result about a Ni-Fe / MS catalyst. It is a block diagram which shows schematic structure of the whole hydrogen production system. It is a block diagram which shows the detail of the CO selective methanation reactor 11 in FIG. (A)-(B) is a perspective view which shows the example of a honeycomb base material. The initial performance of the NiRu / Al—NiOx honeycomb catalyst is shown. (A) to (d) are the results when the inlet CO concentration was changed to 0.2, 0.5, and 0.8% in SV2400h- ¹ , respectively. The outlet CO concentration, outlet CH ₄ concentration, CO removal rate, and CO methanation reaction selectivity are shown. The long-term test result of a NiRu / Al-NiOx honeycomb catalyst is shown. FIG. 8 is a graph in which the rate of increase of the outlet CO concentration in FIG. 7; Δ (CO) / Δt (ppm / 100 h) is plotted against the inlet CO concentration. SV10000h ⁻¹ , initial performance of Ni / Al—VOx honeycomb catalyst and granular catalyst at inlet CO concentration of 0.8%. The dependence of the continuous operation result of the Ni / Al-VOx catalyst honeycomb catalyst on the inlet CO concentration is shown. The dependence of the outlet CO concentration increase rate of the Ni / Al-VOx catalyst honeycomb catalyst on the inlet CO concentration is shown. The initial performance of the mesoporous silica / Ni / Al-VOx honeycomb catalyst at the inlet CO concentration of 0.5% and SV2400h- ¹ is shown. The initial performance of the mesoporous silica / Ni / Al-VOx particulate catalyst was measured at SV2500 h ⁻¹ and various inlet CO concentrations. The long-term test result in 190 degreeC of a mesoporous silica / Ni / Al-VOx granular catalyst is shown. The long-term test result of the mesoporous silica / Ni / Al-VOx granular catalyst in the inlet CO concentration of 0.2% and 0.5% is shown. The result of having predicted the lifetime of a catalyst using the time change of the temperature dependence curve of FIG. 15 is shown. The result of having compared the initial performance of the catalyst which changed the mesoporous silica coating of mesoporous silica / Ni / Al-VOx to mesoporous zirconia and the performance after continuous operation is shown. The result of comparing the initial performance of the catalyst in which the mesoporous silica coating of mesoporous silica / Ni / Al-VOx is changed to mesoporous alumina and the performance after continuous operation is shown. The initial performance of Ni-Fe / MS granular catalysts obtained with an inlet CO concentration of 0.5%, SV4800 and 10000h- ¹ is shown. The initial performance of the Ni—Fe / SiO ₂ (dry high heat method ultrafine silica) granular catalyst at the inlet CO concentration of 0.5% and SV10000h ⁻¹ is shown. The initial performance of the Ni—Fe / SiO ₂ (mesoporous spherical silica) granular catalyst at an inlet CO concentration of 0.5% and SV10000h ⁻¹ is shown. The initial performance of the Ni—Fe / SiO ₂ (mesoporous spherical silica) metal honeycomb catalyst at the inlet CO concentration of 0.5% and SV10000h ⁻¹ is shown. The result of the long-term test of the Ni-Fe / MS granular catalyst in CO concentration 0.5% is shown. The change in the activity of the Ni-Fe / MS granular catalyst before and after the endurance evaluation of 0.5% CO concentration is shown. The result of having evaluated durability with the Ni-Fe / MS granular catalyst under the severer conditions of the inlet CO concentration of 1.0% and SV10000h- ¹ is shown. Showing the Ni-Fe / SiO ₂ (dry pyrogenic ultrafine silica) results of durability tests of granular catalysts in CO concentration of 0.5%. FIG. 27 (a) is a temperature dependence curve acquired at the initial time, 208 hours, 385 hours, and 535 hours. FIG. 27B is an enlarged view of a part thereof. When the Ni—Fe / SiO ₂ (dry high heat method ultrafine silica) granular catalyst is continuously operated at an inlet CO concentration of 0.5% and a reaction temperature of 235 ° C., the outlet CO concentration is 0.20, 0.21, 0.22. % Shows the time to reach each. The results of measuring initial performance of Ni—Fe / SiO ₂ (dry high heat method ultrafine silica) granular catalyst with SV of 4800 h ⁻¹ (corresponding to twice the amount of catalyst) are shown. The long-term test result of the Ni—Fe / SiO ₂ (mesoporous spherical silica) granular catalyst at an inlet CO concentration of 0.5%, SV10000 h ⁻¹ and a reaction temperature of 232 ° C. is shown. For Ni-Fe / SiO ₂ (mesoporous spherical silica) particulate catalyst at an initial, 400 hours, the temperature dependence curve obtained every 800 hours. The initial performance of Ni—Fe / SiO ₂ (mesoporous spherical silica) metal honeycomb catalyst at the inlet CO concentration of 0.5%, SV8000h ⁻¹ or 10000h ⁻¹ is shown.

  Hereinafter, embodiments of the present invention will be described.

1. FIG. 3 shows a flow for producing and purifying high-concentration hydrogen gas supplied from a raw fuel (city gas, etc.) to a fuel cell (for example, a polymer electrolyte fuel cell (PEFC stack)) and an outline of the entire system. The configuration is shown. A portion surrounded by a broken line corresponds to the fuel reforming device (fuel processing device) 14, in which the raw fuel supplied from the raw fuel supply system 4 flows and passes through each catalyst layer. It performs the removal of CO (10 ppm or less) high concentration of hydrogen gas (reformed gas: _{H 2} to about 75%, CO ₂ 20%) obtained.

The raw fuel is first removed from the sulfur component by the desulfurizer 5, and then hydrogen (H ₂ ) and carbon monoxide (CO) are generated by the reforming reaction in the reformer 7 including the reforming catalyst layer (steam generator 6). CO is converted to CO ₂ by a CO converter 8 including a CO conversion catalyst layer.

As shown in FIG. 4, the hydrogen rich gas 19 flowing into the CO selective methanation reactor 11 including the CO methanation catalyst contains about 0.3 to 1.0% CO and about 20% CO ₂ . A pre-stage catalyst 15 and a post-stage catalyst 17 are disposed in the CO selective methanation reactor 11 in order from the upstream side. The front catalyst 15 reduces the CO concentration in the hydrogen rich gas 19 to, for example, 0.2% or less, and the rear catalyst 17 reduces the CO concentration in the hydrogen rich gas 19 to, for example, 10 ppm or less. A high concentration H ₂ gas (reformed gas) with a sufficiently reduced CO concentration is supplied to the PEFC stack 13.

2. CO Methanation Catalysts CO methanation catalysts generally promote CO ₂ methanation reactions in addition to CO methanation reactions. The CO selectivity indicates the proportion of CO methanation among these methanation reactions, and is defined as CO selectivity = (reaction molar flow rate of CO) / (generated molar flow rate of CH ₄ ). If no CO ₂ methanation reaction occurs, CO reaction molar flow rate = CH ₄ production molar flow rate, so that the CO selectivity is 1. On the other hand, for example, when CO ₂ methanation occurs twice as much as CO methanation, the CO selectivity becomes 1/3. If the change in gas flow rate before and after the reaction is so small that it can be ignored, each flow rate can be replaced with a concentration.

In general, CO is more easily adsorbed and reacted by the catalyst than CO ₂ , and therefore, when sufficient CO is present in the hydrogen-rich gas, the adsorption sites on the catalyst surface are occupied by CO, and CO ₂ methanation. The reaction is difficult to occur. For this reason, the CO selectivity of the CO methanation catalyst usually shows a relatively high value when the outlet CO concentration is sufficiently high, and decreases as the outlet CO concentration decreases. Some CO methanation catalysts have a CO selectivity that drops significantly when the outlet CO concentration is reduced to about 1%, and the CO selectivity is 0.5 or more even when the outlet CO concentration is 100 ppm. Some exhibit high CO selectivity even under conditions where the outlet CO concentration is low. In the present specification, a catalyst that promotes the methanation reaction of CO is referred to as a “CO methanation catalyst”. Among them, a catalyst that has a CO selectivity of 0.5 or more even under conditions where the outlet CO concentration is 100 ppm is referred to as “CO methanation catalyst”. This is referred to as “selective methanation catalyst”. Since the CO selectivity decreases with a decrease in the outlet CO concentration, if the CO selectivity is 0.5 or more under the condition that the outlet CO concentration is 100 ppm or less, it can be said to be a CO selective methanation catalyst. The “outlet CO concentration” for each catalyst means the CO concentration in the hydrogen-rich gas discharged from each catalyst.

  If there is a catalyst having high resistance to CO poisoning and high CO selectivity at a low CO concentration (eg, CO concentration of 100 ppm or less), the use of such a catalyst makes it easy to solve the above-described problems. However, in reality, no such catalyst exists. Therefore, in the present embodiment, the pre-catalyst 15 gives priority to the CO poisoning resistance, and the CO poisoning resistance is high, but the CO removal rate is relatively low, and the CO selectivity decreases as the CO concentration decreases. By using a catalyst that is easy to use and giving priority to the CO selectivity at a low CO concentration, a catalyst having a relatively low CO poisoning resistance but a very high CO selectivity at a low CO concentration is used. Even when the inlet CO concentration is high, the CO selective methanation reactor 11 capable of maintaining the outlet CO concentration at a low value over a long period of time is realized. Further, the lower the inlet CO concentration of the rear catalyst 17 is, the greater the resistance to CO poisoning of the rear catalyst 17 is. Therefore, the CO concentration in the hydrogen-rich gas is reduced in advance by the front catalyst 15. The CO removal rate is defined as (inlet CO molar flow rate−outlet CO molar flow rate) / (inlet CO molar flow rate). In this case as well, the molar flow rate can be replaced with the concentration when the change in the volume of other gases before and after the reaction is so small that it can be ignored.

  By the way, the CO selective methanation reactor 11 has a temperature gradient in which the temperature decreases from the upstream side to the downstream side of the flow of the hydrogen-rich gas 19 flowing therethrough. The CO methanation ability of the CO methanation catalyst is temperature dependent, and a high CO removal rate can be achieved only within a narrow temperature range. Therefore, in the present embodiment, the optimum operating temperature of the front catalyst 15 is set higher than that of the rear catalyst 17. With this configuration, both catalysts can be operated at or near their optimum operating temperatures, and CO can be efficiently methanated. The optimum operating temperature can be measured under conditions where the inlet CO concentration is 0.5%. As shown in Tables 1 and 2, the optimum operating temperature of the front stage catalyst 15 and the rear stage catalyst 17 varies depending on the configuration of the catalyst, and the optimum operating temperature of the front stage catalyst 15 is, for example, a temperature within a range of 200 to 280 ° C. The optimum operating temperature of the rear catalyst 17 is, for example, a temperature within the range of 160 to 240 ° C. The difference ΔT in the optimum operating temperature (optimum operating temperature of the front stage catalyst−optimal operating temperature of the rear stage catalyst) is, for example, 5 to 70 ° C., preferably 10 to 50 ° C., and more preferably 20 to 40 ° C.

  In the present embodiment, two types of CO methanation catalysts having different optimum operating temperatures are arranged in order from the upstream in order of highest optimum operating temperature, but three or more types of CO methanation catalysts having different optimum operating temperatures are arranged. You may arrange in order from an upstream with the highest optimal operating temperature.

2-1. Pre-stage catalyst The pre-stage catalyst 15 preferably has the following requirements.
-It does not deteriorate at a CO concentration of 0.3 to 1.0% or has a durability of at least 60,000 hours.
Reduce the outlet CO concentration to a concentration that allows the rear catalyst 17 to maintain the outlet CO concentration at 10 ppm or less over 60,000 hours (eg, 0.2%).
-Maintain the proper outlet CO concentration range and avoid thermal runaway against changes in space velocity SV, inlet CO concentration, and gas temperature due to load fluctuations.

Here, the thermal runaway in the pre-stage catalyst 15 will be described. As the pre-catalyst 15, a catalyst having a relatively low CO removal rate and a CO selectivity that tends to decrease with a decrease in CO concentration is used. Therefore, when the CO concentration in the hydrogen-rich gas is less than 0.1%, for example, The methanation reaction of CO ₂ tends to occur. Methanation reaction of CO ₂ is an exothermic reaction, since the hydrogen rich gas contains a large amount of CO _2, without methanation reaction of CO ₂ is to converge, easily lead to thermal runaway. In order to prevent the occurrence of such thermal runaway, the outlet CO concentration of the pre-stage catalyst 15 is preferably set to 0.1% or more.

  The pre-stage catalyst 15 may be used as a granular catalyst, or may be used after being coated on a honeycomb substrate. An example of the honeycomb substrate is shown in FIGS. 5 (A) and 5 (B). FIG. 5A shows an example of a cordierite honeycomb substrate, and FIG. 5B shows an example of a metal honeycomb substrate. In any case, a large number of vertical, horizontal, diagonal, corrugated partition plates (partitions) arranged along the longitudinal direction of the cylinder (cylindrical, rectangular tube, etc.) are provided in an intersecting manner. A gas passage is provided between the partition plates. The pre-stage catalyst 15 is coated on the entire surface of these partition plates. A honeycomb structure having not only a hexagonal cross section but also gas passages (flow paths) (cells) having a quadrangular shape, a sinusoidal waveform, or other shapes is simply referred to as a honeycomb or a honeycomb substrate in this specification.

In order to suppress thermal runaway in the pre-stage catalyst 15, the pre-stage catalyst 15 is preferably used by coating on a metal honeycomb base material having excellent thermal conductivity. According to such a configuration, the reaction heat of the CO ₂ methanation reaction is quickly removed, and thermal runaway is suppressed. In order to improve the adhesion of the pre-stage catalyst 15, an α-alumina layer may be formed on the surface of the metal honeycomb substrate.

  The type of the support and active metal of the pre-stage catalyst 15 is not particularly limited as long as it has CO methanation ability. Specifically, for example, Ni, Ru, Fe, Co, V, Nb, Ta, Cr, Mo, W, Mn, Tc, Re, and a composite thereof can be used as the active metal. Oxides, nitrides, and carbides composed of at least one of Al, V, Ti, Zr, Si, Mg, and Ce can be used. In addition, it has been clarified through experiments that when the active metal of the pre-stage catalyst 15 contains Ni and Fe, the deterioration of the catalytic activity becomes extremely slow. Therefore, the active metal of the pre-stage catalyst 15 may contain Ni and Fe. preferable. The atomic ratio of Fe / Ni is, for example, 0.01 to 1, and preferably 0.05 to 0.5. Examples of the carrier include silica, alumina, zirconia, titania and the like, and an oxide containing Si (eg, silica) is preferable. Examples of the oxide containing Si include mesoporous silica, dry high-temperature ultrafine silica (eg, AEROSIL (registered trademark) manufactured by Nippon Aerosil Co., Ltd.), and mesoporous spherical silica (eg, Lightstar (registered trademark) manufactured by Nissan Chemical Industries). Etc. are available.

2-2. Post-stage catalyst The hydrogen-rich gas 19 supplied to the CO selective methanation reactor 11 is supplied to the post-stage catalyst 17 after the CO concentration is reduced to some extent by the pre-stage catalyst 15. For this reason, the CO poisoning resistance required for the rear catalyst 17 is lower than that of the front catalyst 15. On the other hand, the post-stage catalyst 17 is required for the post-stage catalyst 17 because the CO concentration in the hydrogen rich gas 19 discharged from the CO selective methanation reactor 11 needs to be very low (eg, 10 ppm or less). The CO selectivity at a low CO concentration is higher than that of the front catalyst 15. For this reason, the above-mentioned CO selective methanation catalyst is used as the post-stage catalyst 17.

  Further, since the outlet CO concentration of the front-stage catalyst 15 = the inlet CO concentration of the rear-stage catalyst 17, the outlet CO concentration of the front-stage catalyst 15 is equal to or lower than the upper limit of the inlet CO concentration at which the rear-stage catalyst 17 can maintain the activity for a long period. Is set to be For example, when the rear catalyst 17 can maintain the activity for a long time when the inlet CO concentration is 0.2% or less, the outlet CO concentration of the front catalyst 15 is set to 0.2% or less.

  The latter stage catalyst 17 may be used as a granular catalyst, or may be used after being coated on a honeycomb substrate. Regarding the honeycomb substrate, the description of the front catalyst 15 also applies to the rear catalyst 17.

The type of the support metal catalyst and the active metal constituting the rear catalyst 17 is not particularly limited as long as it has a CO methanation ability as described in Patent Documents 1 to 3. Specifically, for example, Ni, Ru, Fe, Co, V, Nb, Ta, Cr, Mo, W, Mn, Tc, Re, and a composite thereof can be used as the active metal. Oxides, nitrides, and carbides made of at least one of Al, Ni, V, Ti, Zr, Si, Mg, and Ce can be used. Specific examples include zeolite and silica alumina. Such a supported metal catalyst preferably contains a methanation reaction inhibitor that selectively suppresses the methanation reaction of CO ₂ . As the methanation reaction inhibitor, a material that brings the surface charge of the active metal to the δ + side or various materials that have an effect of suppressing CO ₂ methanation activity can be applied, and in particular, F, Cl, Br, I Or any one or more of inorganic acids such as halogen, HCl, HNO ₃ , H ₂ SO ₄ , H ₃ PO ₄ , and metal oxygen acids such as boric acid, vanadate acid, tungstic acid, chromic acid, etc. desirable. Since the existence form on the catalyst depends on the production process, it is not limited to the above compound, but may be a precursor, a reaction product, or a decomposition product thereof. The active metal is preferably Ni or an alloy or mixture of Ni and V. The support is preferably an oxide containing V or Ni and Al, and specifically, a composite oxide of Ni and Al or a composite oxide of V and Al is preferable. This is because it has become clear from the experiment described in Patent Document 1 that the supported metal catalyst having such a configuration can selectively methanate CO over a long period of time.

  The rear catalyst 17 may include a coating layer having a function of reducing the CO concentration on the active metal surface. A coating layer is formed on the supported metal catalyst, whereby the CO concentration on the supported metal catalyst is reduced as compared with the CO concentration in the reaction gas, so that deterioration of the supported metal catalyst is suppressed. In another aspect, by providing the coating layer, the upper limit value of the inlet CO concentration at which the rear catalyst 17 can maintain the activity over a long period of time can be increased. The advantage that is reduced.

  The principle of reducing the CO concentration on the surface of the supported metal catalyst by providing the coating layer usually involves at least one of a concentration gradient due to diffusion resistance and a concentration gradient due to methanation reaction. Since the concentration gradient due to diffusion resistance is limited in the diffusion of CO in the pores in the coating layer, the CO supply rate to the surface of the supported metal catalyst is higher than the CO consumption rate due to methanation on the surface of the supported metal catalyst. Is a concentration gradient formed when. The concentration gradient due to the methanation reaction is a concentration gradient formed when CO is methanated by the coating layer metal supported on the wall surfaces of the pores. For example, Ni-Fe can be used as the coating layer metal.

  In the present embodiment, the coating layer has a mesoporous structure. In the present specification, the “mesoporous structure” is a structure having a large number of pores (mesopores) having a diameter of 1 to 50 nm. The structure of the pores is not limited, and may be a regular structure or a random structure. Whether or not it has a mesoporous structure can be determined by whether or not the diameter of the pores observed in the electron microscope image is in the range of 1 to 50 nm.

  Examples of the material for the coating layer include mesoporous silica and mesoporous zirconia. The thickness of the coating layer is, for example, 1 to 200 nm, and preferably 5 to 50 nm.

  In order to demonstrate the effect of the present invention, various experiments shown below were conducted.

1. Production and Evaluation of Post-stage Catalyst 1-1 Preparation of NiRu / Al—NiO _x Catalyst As a synthesis example of the post-stage catalyst of the present invention, a method using a reduced pressure spray plasma method will be described first.
First, a composite oxide powder of Ni and Al was synthesized. Nickel nitrate hexahydrate against distilled water 100mL as a raw material solution _{(Ni (NO 3) 2 ·} 6H 2 O) 4.67g of aluminum nitrate nonahydrate _{(Al (NO 3) 3 ·} 9H 2 O) 17 .66 g was dissolved to prepare a mixed aqueous solution having a Ni / Al molar ratio of 0.34. This mixed aqueous solution was sprayed and conveyed in an argon plasma torch ignited at an output of 100 kW and 4 MHz in a reduced pressure space of a reduced pressure high frequency thermal plasma apparatus using an argon mixed gas of 5% oxygen. The powder produced through the plasma torch was collected by a filter and carried out until a total of 500 g of powder was obtained.

  Next, 1 wt% of Ru was supported on the Ni and Al composite oxide powder. 100 g of deionized water was added to 8.0 g of the previously synthesized powder and stirred for 10 minutes. Similarly, 28 g of deionized water was added to ruthenium (III) nitrosyl nitrate in such an amount that the metal ruthenium content after loading was 1 wt%, and the mixture was stirred for 10 minutes. The total amount of the nitrosyl ruthenium (III) nitrate solution was added to the powder suspension in about 20 minutes using a burette, and then stirred for another 10 minutes. After the suspension was introduced into the eggplant-shaped flask, the mixture was stirred for 30 minutes in a hot water bath at 35 to 40 ° C., then cooled to room temperature, and then subjected to an evaporator at 35 to 40 ° C. to evaporate water. The obtained powder was dried at 120 ° C. overnight and then calcined in air at 500 ° C. for 5 hours.

The powder obtained by this method was processed into a metal honeycomb catalyst by the following method. A metal honeycomb (manufactured by Nippon Steel & Sumikin Materials Co., Ltd., material 15Cr-4Al stainless steel NSSCHOM) having a cell number of 400 cpsi, a cell wall thickness of 30 μm, an outer diameter of 25.4 mm (1 inch φ), and a length of 15 mm was used. A dense α-Al ₂ O ₃ layer is deposited on the surface by the high-temperature oxidation treatment. A coating slurry was prepared by adding 6 g of alumina sol (manufactured by Nissan Chemical Industries, alumina sol 520) and 25 g of pure water to 3 g of 1 wt% Ru-supported NiAl composite oxide powder, and stirring and mixing. The metal honeycomb was dipped in this slurry for coating, pulled up, and then excess slurry on the inside and outside walls of the cell was removed by an air pump. After firing in air at 500 ° C. for 1 hour in an electric furnace, the coated honeycomb is weighed. This operation was repeated until the net coating amount reached 300 g per liter of honeycomb to obtain a honeycomb cam in which a catalyst layer was uniformly formed on the inner wall of each cell.

  The 1 wt% Ru-supported NiAl composite oxide coated on the metal honeycomb by this method exhibits the performance as a post-stage catalyst through reduction treatment with hydrogen gas. The structure is such that Ru fine particles supported on the surface of a composite oxide support of Ni and Al and fine Ni particles precipitated from the support are dispersed as a CO methanation active component.

1-2 Initial Characteristic Evaluation of Catalyst Conditions and procedures for evaluating the initial activity of the catalyst will be described below.
The metal honeycomb catalyst (1 inch φ-15 mmL) was installed in a quartz reaction tube having an inner diameter of 27 mm. The periphery was filled with quartz wool so that the reaction gas did not flow through the gap between the reaction tube wall and the honeycomb wall. In the case of a granular catalyst (size 1.2 to 2.0 mm), which will be described later, a catalyst having a filling capacity of 2 ml was separately measured and packed as it was on a scale plate provided in the center of a quartz reaction tube having an inner diameter of 12 mm. The temperature of the catalyst layer was measured by inserting the tip of the sheath thermocouple at a position about 5 mm from the upper end of the catalyst layer.

The catalyst installed in the reaction tube was subjected to reduction treatment in hydrogen by the following procedure before evaluating the catalyst performance. After sufficiently purging the air inside the reaction tube with N ₂ , H ₂ was flowed at 500 mL / min, and the temperature was raised to 500 ° C. at 20 ° C./min. By holding the temperature at 500 ° C. for 1 hour, Ru and Ni existing as oxides are reduced to metals and act as active components. After the reduction, the gas was switched from H ₂ to N ₂ for 5 minutes to purge H ₂ . Thereafter, the temperature was lowered to a temperature at which the activity of the catalyst was evaluated.

When the catalyst temperature reached the target reaction temperature, water vapor was introduced into the reaction tube, and the reaction gas was introduced after 5 minutes. The water vapor supply rate was set to a value corresponding to water vapor / CO = 34 (molar ratio), ion exchange water was sent to a vaporizer maintained at 200 ° C. with a micropump, and the generated water vapor was introduced into the reaction tube with N ₂ carrier. Each reaction gas was introduced into the reaction tube by a mass flow controller, and the composition was CO 0.5 vol%, CO ₂ 19 vol%, and H ₂ balance on a dry basis. In addition to 0.5%, the CO concentration was variously changed as required. The space velocity SV was selected from the range of 2000 to 10000h- ¹ . For the analysis of the reaction tube outlet gas, a non-dispersive infrared analyzer was used for CO, CO ₂ , and CH ₄ , and a thermal conductivity analyzer (all manufactured by Horiba, Ltd.) was used for H ₂ . After the gas analysis at the initial set temperature was completed, the catalyst temperature was sequentially increased, and the gas analysis was performed at each temperature.

FIG. 6 shows the initial characteristic evaluation results of the NiRu / Al—NiOx honeycomb catalyst of Example 1-1. 6A to 6D, in SV2400h- ¹ , the outlet CO concentration, outlet CH ₄ concentration, CO removal rate when the inlet CO concentration is changed to 0.2, 0.5, and 0.8%, Each CO selectivity was compared.
In the latter stage catalyst, (1) the low temperature activity of the CO methanation reaction is high, (2) the CO removal rate is high, and (3) the CO ₂ methanation reaction is well suppressed particularly at high temperatures. desired. In the present invention, the temperature “optimum operating temperature; T _opt (° C.)” at which the outlet CO concentration exhibits the minimum value has already been defined as a numerical value representing (1). A smaller value of T _opt is better. For the index (2), the outlet CO concentration CO _min at the optimum operating temperature was used. Of course, the smaller the value, the better. On the other hand, the index of (3) is as follows. First, the temperature at which the outlet CH ₄ molar flow rate becomes twice the inlet CO molar flow rate, that is, the temperature at which the CO selectivity is 50% is defined as T _{η = 50%} (° C.). The difference between this temperature and the optimum operating temperature ΔT = T _{η = 50%} −T _opt was used as an index for (3). ΔT indicates how much the CO ₂ methanation reaction is suppressed from the temperature at which the outlet CO concentration is reduced to the highest temperature in each catalyst, and the larger the value, the better.
The respective numerical values of the NiRu / Al—NiOx honeycomb catalyst are shown together in Table 1 together with other subsequent stage catalysts shown in the examples of the present invention. In the NiRu / Al—NiOx honeycomb catalyst, T _opt and CO _min decrease and ΔT increases as the inlet CO concentration decreases, indicating that both indicators show improved catalyst performance. This is consistent with the general knowledge of the CO concentration dependence on the performance of the CO selective methanation catalyst, confirming the validity of the index.

1-3 Long-term test of the catalyst After the initial characteristic evaluation of the catalyst of Example 1-2 was completed, a long-term test of the catalyst was subsequently performed. The catalyst was held at a constant temperature and the reaction was continued for thousands of hours. Meanwhile, the long-term stability of the catalyst performance was confirmed from the amount of change in the CO and CH ₄ concentrations in the catalyst outlet gas.

  If the catalyst has excellent durability and the reaction conditions are mild, there may be little change in the outlet gas composition even if the operation takes thousands of hours. In such a case, the temperature is sequentially changed within a range from about 150 ° C. to the long-term test operation temperature by the same procedure as the initial characteristic evaluation of Example 1-2 every 200 hours or 400 hours of continuous operation. Low temperature performance was measured. This method evaluates the performance at low temperatures with a slow reaction rate in the course of continuous operation, so that even if there is no significant change in the outlet gas composition, the catalyst performance can be reduced quickly and with high sensitivity. You can check it.

  FIG. 7 shows the long-term test results of the NiRu / Al—NiOx honeycomb catalyst. When the inlet CO concentrations were 0.2, 0.3, 0.5, and 0.8%, the changes in the outlet CO concentrations when operating at the respective optimum operating temperatures were compared. A relatively large increase in CO concentration was observed until about 100 hours after the start of the reaction, but thereafter, the CO concentration increased almost linearly, that is, the catalyst performance decreased. The slope of the straight line tended to increase as the inlet CO concentration increased.

1-4 Relationship between Catalyst Degradation Rate and Inlet CO Concentration FIG. 8 is a plot of the rate of increase of the outlet CO concentration; Δ (CO) / Δt (ppm / 100 h) in FIG. 7 versus the inlet CO concentration. Degradation proceeds at a rapid rate of 20-30 ppm / 100 h when the inlet CO concentration is 0.5 and 0.8%, but rapidly decreases to 3 ppm / 100 h when it reaches 0.3%, and the optimum operating temperature is 0.2%. It was found that the deterioration rate was suppressed to an extremely low value of 1 ppm / 100 h or less not only at 170 ° C. but also at 190 ° C. which was 20 ° C. higher. It has been shown that even if the same catalyst can be used at an inlet CO concentration of 0.3% or less, preferably 0.2% or less, a low outlet CO concentration can be maintained for a long time.

1-5 Preparation of Ni / Al-VOx Catalyst and Performance Evaluation Next, as another synthesis example of the latter catalyst, a method by a coprecipitation method will be described.
First, an Al-VOx catalyst powder as a catalyst carrier was prepared by a coprecipitation method. 0.60 g of ammonium vanadate (NH ₄ ) ₂ VO ₃ was placed in 61 mL of pure water and heated to dissolve. In addition, 44.1 g of aluminum nitrate was dissolved in 235 mL of pure water. After mixing these two solutions, the solution was transferred to a 2 L beaker, and an aqueous ammonium carbonate solution was added dropwise so that pH = 8 in about 15 minutes while stirring at 2500 rpm. Thereafter, stirring was continued for 30 minutes. The deposited precipitate was filtered through a 0.2 μm membrane filter and washed with 1 L of pure water. The obtained precipitate was dried under reduced pressure at room temperature for half a day and then dried in a drying furnace at 110 ° C. for 12 hours. The obtained gel was ground and then fired in air at 500 ° C. for 3 hours. As a result, an oxide carrier having a molar ratio of Al: V = 0.96: 0.04 was obtained.

6.26 g of the Al-VOx catalyst powder was added to 50 mL of pure water to prepare a suspension. The dissolved nickel nitrate _{Ni (NO 3) 2 · 6H} 2 O ( Kanto Chemical) 12.8 g of pure water 50 mL. While stirring the suspension of the oxide carrier, a total amount of nickel nitrate aqueous solution was added using a burette in about 20 minutes. The mixture was stirred at room temperature for 30 minutes and in a 45 ° C. water bath for 30 minutes, and then cooled to room temperature once. Then, it was applied to an evaporator in a hot water bath at 35 to 50 ° C. to remove all moisture. The obtained powder was dried at 110 ° C. for 12 hours and then calcined at 500 ° C. for 3 hours to obtain a 30 wt% Ni / Al-VOx catalyst powder supporting Ni 30 wt% in terms of metal.
The powder obtained by this method was coated on a metal honeycomb substrate according to the procedure described in 1-1. Part of the catalyst powder is formed into a tablet with an outer diameter of 20 mm and a thickness of about 1.5 mm using a hydraulic molding machine, and then a granular catalyst with a size of 1.2 to 2.0 mm is produced through grinding and sieving. did. These were subjected to hydrogen reduction treatment under a predetermined condition, and then evaluated as a post-stage catalyst.

FIG. 9 shows the initial performance of the honeycomb catalyst and the granular catalyst at the inlet CO concentration of 0.8%. The SV of the honeycomb catalyst is 2400 h ⁻¹ , and the SV of the granular catalyst is 10000 h ⁻¹ . Compared to the honeycomb catalyst of Example 1-1 in which the three indicators in Table 1 were evaluated under the same conditions (inlet CO concentration 0.8%, SV2400 h ⁻¹ ), T _opt increased by 20 ° C., but CO _min Decreased from 86 ppm to 65 ppm, and ΔT increased from 3 ° C. to 12 ° C., and improvements were observed for each. This is because vanadium which is a CO ₂ methanation inhibitor is added to the catalyst. 10 and 11 show the result of continuous operation of the honeycomb catalyst and the dependency of the CO increase rate on the inlet CO concentration. Similar to the catalyst of Example 1-1, the outlet CO concentration of the catalyst of Example 1-5 also increases in proportion to time, and the rate of increase strongly depends on the inlet CO concentration. Also in the case of this catalyst, when the inlet CO concentration is 0.2%, the value is suppressed to 1 ppm / 100 h or less.

1-6 Preparation and Performance Evaluation of Mesoporous Silica / Ni / Al-VOx Catalyst A mesoporous silica layer containing a trace amount of Ti was formed on the 30 wt% Ni / Al-VOx catalyst powder prepared in 1-2 by the following method. It was constructed.

  5.00 g of 30 wt% Ni / Al-VOx powder and 2.00 g of 28% ammonia water (manufactured by Kanto Chemical Co., Inc.) were added to 150 mL of ultrapure water to prepare a suspension. Further, 0.6 g of hexadecyltrimethylammonium bromide (manufactured by Acros) was dissolved in a mixed solution of 40 mL of ethanol (manufactured by Kanto Chemical Co., Ltd.) and 6 mL of ultrapure water. Next, 1.20 g of tetraethylorthosilicate (manufactured by Kanto Chemical Co., Inc.), 0.05 g of titanium isopropoxide (manufactured by Kanto Chemical Co., Ltd.) and 0.25 g of acetylacetone (manufactured by Kanto Chemical Co., Ltd.) were added to 8 mL of ethanol to obtain an alkoxide solution. While stirring the 30 wt% Ni / Al-VOx suspension, the entire amount of hexadecyltrimethylammonium bromide solution was added using a pipette in 1 minute, and the suspension was stirred at room temperature for 30 minutes. Next, while stirring the suspension, the entire amount of the alkoxide solution was added using a pipette in 1 minute, followed by stirring at room temperature for 16 hours. The suspension was then filtered and the residue was washed with 400 mL ethanol. The obtained residue was dried at room temperature under reduced pressure, further dried at 250 ° C. for 1.5 hours, and then calcined at 550 ° C. for 4 hours to form a 15 nm thick mesoporous silica layer on the surface of MS / 30 wt% Ni / Al-VOx catalyst powder was obtained.

  The catalyst powder was coated on a metal honeycomb substrate by the procedure described in 1-1. On the other hand, a part of the catalyst powder is formed into a tablet with an outer diameter of 20 mm and a thickness of about 1.5 mm by a hydraulic molding machine, and then a granular catalyst of 1.2 to 2.0 mm size is produced through grinding and sieving. did.

FIG. 12 shows the initial performance of the mesoporous silica / Ni / Al—VOx honeycomb catalyst at an inlet CO concentration of 0.5% and SV2400h- ¹ . The NiRu / Al—NiOx honeycomb catalyst (No. 1) in Table ¹ was evaluated with the same CO of 0.5% and SV2400h ⁻¹ . Therefore, when compared, T _opt increased by 24 ° C., but CO _min decreased and ΔT Are increasing, both of which have greatly improved performance. The Ni / Al-VOx honeycomb catalyst (No. 3) has a high inlet CO concentration of 0.8%, and therefore, comparison is not possible. However, the three indicators are greatly improved.

FIG. 13 shows the results of measuring the initial performance of the granular catalyst at SV2500 h ⁻¹ and various inlet CO concentrations. Although there is no granular catalyst measured under the same conditions, a strict comparison cannot be made. As is apparent from FIG. 13, this catalyst has the same tendency to improve the initial performance as the inlet CO concentration decreases. In particular, in the case of this catalyst, when the inlet CO concentration is 0.2%, the target value <10 ppm of the outlet CO concentration obtained by an actual machine is realized in a wide temperature range of 30 ° C. This result, together with the result of FIG. 16, is one of the motives leading to the present invention.

FIG. 14 shows the long-term test results of this catalyst at 190 ° C. Unlike the long-term test results of the other catalysts shown in FIGS. 7 and 10, the present catalyst showed no tendency to increase the outlet CO concentration from the initial stage at any inlet CO concentration, and showed a tendency to be constant or slightly decreased. During this period, the higher the inlet CO concentration, the shorter the CO concentration. After that period, the CO concentration begins to rise as before, but the slope increases exponentially instead of a straight line. In FIG. 14A, when the inlet CO concentration was 0.2%, no increase in CO concentration was observed even if the continuous operation time exceeded 2600 hours. For this reason, the presence or absence of performance degradation at low temperatures was confirmed using the temperature dependence data acquired during continuous operation described in Example 1-3 “Long-term test of catalyst”. The results are shown in FIG. Note that, unlike the change in CO concentration, the outlet CH ₄ concentration decreased almost monotonously at any inlet CO concentration. Both the increase in CO concentration and the decrease in CH ₄ concentration are due to the deterioration of the catalyst including Ni as an active component, including this catalyst. When the reaction continues under high concentration CO conditions, C and H are added to the Ni active site. This is because precipitates (carbon species) that are contained are deposited, and both the CO and CO ₂ methanation reaction rates are reduced.

FIG. 15A shows a temperature dependence curve of the CO concentration measured at each time during the continuous operation at the inlet CO concentration of 0.5%. As the continuous time elapses, the temperature dependence curve shifts to the high temperature side. For example, the optimum operating temperature T _opt was 172 ° C. at the start of the reaction, but increased to 188 ° C. in 1880 hours. In continuous operation of FIG. 14 (a), but was not observed an increase in the outlet CO concentration at all to around 1800 hours, as indicated by a marked increase in T _opt, that deterioration of the catalyst is clearly progressed I understood. FIG. 15B shows the time change of the temperature dependence curve when the inlet CO concentration is 0.2%. Also in this case, the temperature dependence curve tends to shift slightly toward the high temperature side. However, it can be seen that the degree is extremely small compared to 0.5%. FIG. 16 shows the result of predicting the life of the catalyst using the time change of these temperature dependence curves. The vertical axis in FIG. 16 is not T _opt of each temperature dependence curve but a temperature T _{(CO = 20 ppm)} at which the CO concentration indicates _{20 ppm} . When this T _{(CO = 20 ppm)} is plotted against the logarithm of the continuous operation time t (h) in which the temperature dependency is measured, it can be seen that it is almost in a straight line. Extrapolating this straight line, the time until T _{(CO = 20 ppm)} reaches the reaction temperature of 190 ° C. can be said to be the life of the catalyst. If T _{(CO = 10 ppm)} , the time during which the catalyst can maintain <10 ppm, which is a practically necessary concentration, can be calculated. However, at the inlet CO concentration of 0.5%, the outlet CO concentration does not fall below 10 ppm. For convenience, the life was set to 20 ppm. When the inlet CO concentration was 0.2%, T _{(CO = 20 ppm)} did not reach 190 ° C. even if the straight line was extrapolated to 100,000 hours.
This means that if the catalyst of Example 1-6 is used at an inlet CO concentration of 0.2%, a wide temperature range with an outlet CO concentration of <10 ppm can be secured as described above, and 6 required for a practical catalyst. It shows that a life of 10,000 hours can be achieved at the same time.

1-7 Preparation and Performance Evaluation of Other Mesoporous Material-Catalyzed Catalyst In 1-3, an example of a catalyst in which an MS layer was coated on a 30 wt% Ni / Al-VOx catalyst was shown. Here, a mesoporous material other than silica is used. An example of the latter catalyst coated is shown.

  The first is an example in which mesoporous zirconia is coated. 5.00 g of the core catalyst powder prepared in 1-2 was added to 150 mL of dehydrated ethanol (manufactured by Kanto Chemical) to prepare a suspension. Further, 0.6 g of hexadecyltrimethylammonium bromide (manufactured by Acros) was dissolved in 48 mL of ethanol (manufactured by Kanto Chemical). Next, 2.12 g of zirconium tetra-n-butoxide (manufactured by Kanto Chemical Co., Inc.) was added to 10 mL of ethanol to obtain an alkoxide solution. While stirring the suspension of the core catalyst, the whole amount of hexadecyltrimethylammonium bromide solution was added using a pipette in 1 minute, and the suspension was stirred at room temperature for 30 minutes. Next, while stirring the suspension, the entire amount of the alkoxide solution was added using a pipette in 1 minute, followed by stirring at room temperature for 16 hours. The suspension was then filtered and the residue was washed with 400 mL ethanol. The obtained residue was dried under reduced pressure at 50 ° C. for 3 hours, further dried at 250 ° C. for 1.5 hours, and then calcined at 550 ° C. for 4 hours to form a mesoporous zirconia layer on the surface of the core catalyst. A mesoporous zirconia-coated catalyst was obtained. In the mesoporous zirconia layer, pores of about 2 to 50 nm were randomly formed. Further, it was confirmed from scanning transmission electron microscope energy dispersive X-ray spectroscopy (STEM-EDS) that Zr was distributed almost uniformly on the surface of the core catalyst particles.

  The following is an example in which a mesoporous titania layer is coated. 5.00 g of the core catalyst powder was added to 150 mL of dehydrated ethanol (manufactured by Kanto Chemical) to prepare a suspension. Further, 0.6 g of hexadecyltrimethylammonium bromide (manufactured by Acros) was dissolved in 48 mL of ethanol (manufactured by Kanto Chemical Co., Ltd.) and 5.3 mL of ultrapure water. Next, 1.57 g of titanium tetraisopropoxide (manufactured by Kanto Chemical Co., Inc.) was added to 10 mL of ethanol to obtain an alkoxide solution. While stirring the suspension of the core catalyst, the whole amount of hexadecyltrimethylammonium bromide solution was added using a pipette in 1 minute, and the suspension was stirred at room temperature for 30 minutes. Next, while stirring the suspension, the entire amount of the alkoxide solution was added using a pipette in 1 minute, followed by stirring at room temperature for 16 hours. The suspension was then filtered and the residue was washed with 400 mL ethanol. The obtained residue was dried at 50 ° C. under reduced pressure for 3 hours, further dried at 250 ° C. for 1.5 hours, and then calcined at 550 ° C. for 4 hours to form a mesoporous titania layer on the surface of the core catalyst. A mesoporous titania-coated catalyst was obtained.

  The following is an example in which a mesoporous alumina layer is coated. 5.00 g of the core catalyst powder was added to 150 mL of dehydrated ethanol (manufactured by Kanto Chemical) to prepare a suspension. Further, 0.6 g of hexadecyltrimethylammonium bromide (manufactured by Acros) was dissolved in 48 mL of ethanol (manufactured by Kanto Chemical Co., Ltd.) and 5.3 mL of ultrapure water. Next, 1.12 g of aluminum triisopropoxide (manufactured by Kanto Chemical Co., Inc.) was added to 15 mL of toluene to obtain an alkoxide solution. While stirring the suspension of the core catalyst, the whole amount of hexadecyltrimethylammonium bromide solution was added using a pipette in 1 minute, and the suspension was stirred at room temperature for 30 minutes. Next, while stirring the suspension, the entire amount of the alkoxide solution was added using a pipette in 1 minute, followed by stirring at room temperature for 16 hours. The suspension was then filtered and the residue was washed with 400 mL ethanol. The obtained residue was dried at 50 ° C. under reduced pressure for 3 hours, further dried at 250 ° C. for 1.5 hours, and then calcined at 550 ° C. for 4 hours to form a mesoporous alumina layer on the surface of the core catalyst. A mesoporous alumina-coated catalyst was obtained.

  These catalyst powders were molded into a tablet with an outer diameter of 20 mm and a thickness of about 1.5 mm by a pressure molding machine, and then sized and granulated into 1.2 to 2.0 mm granular catalysts through pulverization and sieving. .

  FIG. 17 and FIG. 18 compare the initial performance of the catalyst in which the mesoporous silica coating of mesoporous silica / Ni / Al-VOx is changed to mesoporous zirconia and mesoporous alumina, and the performance after continuous operation, respectively. 840 hours and 1400 hours of continuous operation of each catalyst correspond to times when the outlet CO concentration starts to increase. As is clear from Table 1, it was found that equivalent performance could be obtained even if the material of the coating layer was changed.

2. Production and Evaluation of Pre-stage Catalyst 2-1 Preparation and Performance Evaluation of Pre-stage Catalyst Using Mesoporous Silica as a Support First, a pre-stage catalyst using mesoporous silica containing a small amount of Ti as a support was prepared. The method is shown below.

  First, mesoporous silica powder (Ti / Si atomic ratio was 0.03) was prepared by a sol-gel method.

  Specifically, 2.9 g of hexadecyltrimethylammonium bromide (manufactured by Across) was dissolved in a mixed solution of 298 mL of ethanol (manufactured by Kanto Chemical Co., Ltd.) and 24 mL of ultrapure water. Next, 8.1 g of 28% ammonia water (manufactured by Kanto Chemical Co., Inc.) and 537 mL of ultrapure water were mixed and added to the above solution. Next, 5.92 g of tetraethyl orthosilicate (manufactured by Kanto Chemical Co., Inc.), 0.18 g of titanium isopropoxide (manufactured by Kanto Chemical Co., Ltd.) and 1.26 g of acetylacetone (manufactured by Kanto Chemical Co., Ltd.) were added to obtain an alkoxide solution. Next, while stirring the hexadecyltrimethylammonium bromide solution, the entire amount of the alkoxide solution was added using a pipette in 3 minutes, followed by stirring at room temperature for 16 hours. Thereafter, the suspension was filtered, and the residue was washed with 400 mL of ethanol. The obtained filtrate was dried at room temperature under reduced pressure, further dried at 250 ° C. for 1.5 hours, and then calcined at 550 ° C. for 4 hours to obtain mesoporous silica powder. When the prepared mesoporous silica powder was observed with TEM, spherical particles having a diameter of approximately 300 nm were observed. Since a regular structure having a period of about 3 nm from the spherical particle center toward the particle surface can be confirmed, it is considered that almost uniform nanopores are formed in a dendritic shape from the central portion toward the outer surface.

  Next, Ni—Fe was supported on the powder to obtain a Ni—Fe / MS catalyst. This loading was performed so that the loading amount of Ni was 10 wt%. The atomic ratio of Fe to Ni was 0.1. These metals were supported by an impregnation method (incipient wetness method). Specifically, Ni acetic acid Ni hydrate (manufactured by Kanto Chemical Co., Ltd.) and iron acetate hydrate (manufactured by Kanto Chemical Co., Ltd.) prepared by adjusting the atomic ratio of Ni: Fe to 1: 0.1 with respect to 5 g of mesoporous silica powder. 4 mL of the mixed aqueous solution was dropped with a pipette and baked at 110 ° C. for 5 hours and at 500 ° C. for 3 hours. This process was repeated three times to obtain Ni-Fe / MS powder. The obtained catalyst powder was reduced in hydrogen and then observed with TEM. However, the Ni—Fe particles were not clearly observed, and are considered to be highly dispersed and supported in the MS pores.

The produced Ni-Fe / MS catalyst powder was formed into a 1.2 to 2.0 mm granular catalyst by the same method as before, and the initial performance and durability performance were evaluated. FIG. 19 shows the initial performance of the Ni—Fe / MS granular catalyst obtained at an inlet CO concentration of 0.5%, SV4800 and 10,000 h ⁻¹ . Table 2 summarizes four performance indicators (1) T _opt , (2) CO _min , (3) T _{η = 50%} , and (4) ΔT together with other pre-stage catalysts shown in the examples of the present invention. .

A feature common to the pre-stage catalyst disclosed in this example is that T _opt is high and CO _min is also large. In the case of the Ni—Fe / MS granular catalyst of this example, the inlet CO concentration is 0.5%, the top _opt at SV4800h ⁻¹ is 219 ° C., and the outlet CO concentration CO _{min at} that temperature is 289 ppm. When these values are compared with the latter catalyst in Table 1, it can be seen that the difference is remarkable.

FIG. 23 shows the results of the long-term test. SV10000h- ¹ was carried out at a reaction temperature of 250 ° C, and SV4800h- ¹ was carried out at 220 ° C. FIG. 23 (a) shows the time variation of the outlet CO concentration, but the outlet CO concentration is maintained almost constant except for the initial 100 hours in any SV. Similarly, the outlet CH ₄ concentration in FIG. 23B also maintains a constant value. FIG. 24 compares the temperature dependence curves before and after the long-term test. The temperature dependence curve after 300 hours of operation at SV10000h- ¹ is almost unchanged from the initial state. In SV4800h- ¹ , the performance is slightly improved rather than the initial performance. The cause of this is unknown at present, but in any case, the results shown in FIGS. 23 and 24 indicate that the pre-stage catalyst of this example has high deterioration resistance.

As described above, the outlet CO concentration of the catalyst under these conditions is significantly higher than that of the subsequent catalyst, and it cannot be said that the catalyst is excellent in CO removal performance. However, for example, the mesoporous / Ni / Al-VOx granular catalyst disclosed as the latter stage catalyst deteriorates at the rate shown in FIG. 15 (a) under the conditions of 0.5% CO and SV2400h- ¹ , whereas this catalyst SV is doubled to 4800h ^-1 (equivalent to halving the catalyst for the same amount of processing gas) and 4 times 10000h ^-1 (equivalent to ¼ catalyst). Nevertheless, it is a remarkable feature that it hardly deteriorates.

FIG. 25 shows the results of evaluating the durability of the present catalyst under more severe conditions of an inlet CO concentration of 1.0% and SV10000h- ¹ . The outlet CO concentration was initially 0.21% (2100 ppm) but increased to less than 0.3% (3000 ppm) in about 250 hours. However, after that, the increase in the CO concentration stopped and reached a constant value. The outlet CH ₄ concentration also decreased in the initial stage, but showed a constant value after 200 hours.

  This result is designed so that 0.3 to 0.5% of CO flows from the upstream to the inlet of the CO removal catalyst (CO selective oxidation catalyst) used in the current reformer. If the catalyst is used, it means that even if the CO concentration flowing into the CO removal catalyst is increased to 1.0%, the CO concentration can be reduced to 0.2 to 0.3% without severe deterioration of the preceding catalyst. Since the increase in the inflow CO concentration is directly connected to the reduction of the amount of the CO shift catalyst installed upstream, the economic effect is great. The results of this figure reveal that the CO methanation reactor disclosed in the present invention can be applied not only at 0.5% inflow CO concentration but also at 1.0%.

2-2 dry pyrogenic ultrafine silica as a carrier having no nonporous ultrafine particles SiO ₂ mesopores as meso porous spherical silica mesoporous silica or later Preparation and performance evaluation in the individual pre-catalyst used in the carrier used. This SiO ₂ is industrially produced by a dry high heat method. A method for supporting Ni—Fe on the carrier will be described below.

Nickel acetate (II) tetrahydrate (Ni (CH ₃ COO) ₂ .4H ₂ O, manufactured by Kanto Chemical Co., Inc.) was dissolved in 29.86 g of pure water and then made up to 200 mL, and a 0.6 M nickel acetate aqueous solution. Was prepared. Further, 0.22 g of iron (II) acetate (Fe (CH ₃ COO) ₂ , manufactured by Wako Chemical Co., Ltd.) was dissolved in pure water and made up to 25 mL to prepare a 0.05 mol / L Fe acetate aqueous solution. 6. Nitric acid Ni aqueous solution 6.50mL and Fe nitric acid aqueous solution 7.80mL were measured and mixed, respectively, and Ni-Fe mixed aqueous solution was obtained. Next, 2.06 g of SiO ₂ powder (AEROSIL (registered trademark) 380, manufactured by Nippon Aerosil Co., Ltd.) was weighed, added to 50 ml of pure water in a beaker, and stirred with a stirring blade. The Ni-Fe mixed aqueous solution prepared previously was dropped little by little to the SiO ₂ suspension stirred for about 10 minutes, and further stirred for 30 minutes. The SiO2 suspension with the Ni-Fe solution added was transferred to an eggplant-shaped flask and stirred at 45 ° C for 30 minutes, and then the temperature was lowered to 35 ° C and water was removed while sucking with an evaporator. The obtained powder was dried at 110 ° C. overnight and then calcined in air at 500 ° C. for 3 hours.

  The obtained catalyst powder was molded into a granular catalyst according to the procedure so far, and then the initial performance and durability performance were evaluated.

FIG. 20 shows the initial performance of the catalyst at an inlet CO of 0.5% and SV10000h- ¹ . Since the data for this catalyst could not be obtained up to the high temperature range, each index in Table 2 could not be calculated accurately, but the characteristics of high T _opt and large CO _min were the same. FIG. 26 shows the results of continuous operation at an inlet CO of 0.5%, an SV of 10000 h ⁻¹ , and a reaction temperature of 235 ° C. In the continuous operation for 500 hours, the low temperature activity was measured in the temperature range lower than the reaction temperature at the beginning, 208 hours, and 385 hours (arrows). Immediately after measuring the low-temperature activity, the outlet CO concentration once decreased and the CH ₄ concentration increased. However, after that, all returned to a constant value immediately. The reason why the CO concentration once decreases after the low-temperature activity measurement is unknown, but the concentration of the carbon species precursor or intermediate produced on the catalyst surface decreases because the reaction is slowed by being kept at a low temperature, and the deterioration temporarily occurs. Is expected to have recovered. In any case, it is shown that 0.5% CO can be stably maintained at 0.2% or less without significant deterioration of the catalyst in spite of severe conditions of SV of 10,000 h ⁻¹ in this catalyst. It was.

  FIG. 27 (a) is a temperature dependence curve acquired at the initial time, 208 hours, 385 hours, and 535 hours. FIG. 27B is an enlarged view of a part thereof. The catalyst life prediction performed in FIGS. 15 and 16 of Example 1-6 was performed using this figure. The results are shown in FIG. FIG. 28 shows the time required for the outlet CO concentration to reach 0.20, 0.21, and 0.22% when the catalyst is continuously operated at an inlet CO concentration of 0.5% and a reaction temperature of 235 ° C. Yes. It was found that the time required for the catalyst outlet CO concentration to reach 0.20% was about 18,000 hours, 0.21% over 65,000 hours, and 0.22% over 100,000 hours. This means that the durability of the catalyst as a pre-stage catalyst is almost at a practical level.

Figure 29 shows the results of measuring the initial performance in the the SV of the catalyst (catalyst amount equivalent to approximately 2 times the ^{SV10000h -1)} 4800h ^-1. Compared to 2400 h ⁻¹ , T _opt decreased by 218 ° C. and 20 ° C. or more, and CO _min also decreased to 1/3, 500 ppm. However, thermal runaway occurred when the temperature was raised to 241 ° C., and when the catalyst temperature rapidly increased by 40 ° C., CH ₄ generated by the CO ₂ methanation reaction increased to 10%. The temperature of the catalyst rose to around 340 ° C. until the reaction gas was stopped although the electric furnace for heating the reaction tube was stopped. This time, the thermal runaway occurred because the SV was lowered, but the thermal runaway may occur similarly even if the inlet CO concentration is lowered. This is due to the fact that the CO concentration in the catalyst layer is greatly reduced due to a decrease in the inlet CO concentration at a high temperature at which the CO methanation reaction rate is high, or because the SV decreases and the contact time between the reaction gas and the catalyst increases. This is because when the CO concentration is significantly lower than 0.1%, empty sites are generated on the Ni active sites, and the CO ₂ methanation reaction as a side reaction proceeds rapidly there.
In applying the pre-stage catalyst, it is necessary to pay particular attention to the suppression of this thermal runaway. The method will be described later.

2-3 Preparation and Performance Evaluation of Pre-Catalyst Using Mesoporous Spherical Silica as Support The previously synthesized mesoporous silica had regular mesopores of about 3 nm. SiO ₂ having mesopores of about 16 nm was used as the support.

In order to obtain a carrier to be used as a catalyst, silica sol (Nissan Chemical Light Star) corresponding to 25 g of solid SiO ₂ was put in a magnetic dish and water was blown off on a hot plate heated to 200 ° C. At this time, the mixture was mixed with a spatula so that the film was not stretched on the sol surface. After the water disappeared, it was further dried at 130 ° C. for 12 hours. The dried SiO ₂ powder was pulverized in a mortar, further heated to 500 ° C. for 3 hours in an air atmosphere, held for 3 hours and fired.

  Next, Ni—Fe was supported by the following procedure. A predetermined amount of 78.8 mL of 0.6 M Ni acetate aqueous solution diluted with ion-exchanged water and 94.7 mL of 0.05 M Fe acetate aqueous solution diluted with a mixed solution of ion-exchanged water: acetic acid = 3: 2 were mixed. And stirred for 5 minutes. The amount of Ni aqueous solution at this time was metal Ni weight: silica weight = 10: 90, and the amount of Fe aqueous solution was 1/10 of the amount of Ni substance. The Ni-Fe mixed solution was supported on 25 g of the calcined silica powder by the incipient wetness impregnation method. As a guide, 15 to 20 mg of the mixed solution was dropped with a pasteur in one loading step, and mixed while being crushed with a spatula so as not to become lumps. Thereafter, it was put in a furnace at 300 ° C. in air and dried for 30 minutes. After taking out and allowing to cool, the supporting-drying process was repeated until the mixed solution disappeared (8-9 times). Thereafter, the mixture was heated to 500 ° C. in air for 3 hours and held for 3 hours for firing. After calcination, the obtained catalyst was pulverized in a mortar. The obtained catalyst powder was molded into a granular catalyst according to the procedure so far, and then the initial performance and durability performance were evaluated.

FIG. 21 shows the initial performance of the catalyst at an inlet CO of 0.5% and SV10000h- ¹ . The performance index in Table 2 shows an intermediate value between Examples 2-1 and 2-2. FIG. 30 shows the long-term test results of the present catalyst at an inlet CO concentration of 0.5%, SV10000 h ⁻¹ , and a reaction temperature of 232 ° C. At that time, the temperature dependence curves acquired at the initial stage, every 400 hours, and every 800 hours are shown in FIG. As shown in FIG. 30 (a), it was found that the outlet CO concentration of the catalyst continued to increase and did not converge to a constant value by 800 hours, but the rate of increase gradually decreased. The CH ₄ concentration in FIG. 30 (b) showed a similar tendency.

2-4 Effect of making metal honeycomb catalyst The process of wash-coating the Ni-Fe / SiO ₂ catalyst powder produced in Example 2-3 on a metal honeycomb substrate is shown below.

Ni-Fe / SiO ₂ catalyst powder 15.0 g, silica binder (Nissan Chemical STO40) 4.2 g, and water 40 g were stirred and mixed to prepare a slurry for coating. The ratio of catalyst and binder solid content at this time was 90:10. The metal honeycomb is made of Nippon Steel & Sumikin Materials made of stainless steel (YUS205M1) having an outer diameter of 25.4 mm (1 inch φ) and a length of 15 mm, and the surface thereof is subjected to high temperature oxidation treatment. The number of cells was 400 cpsi, and the cell wall thickness was 30 μm. The metal honeycomb was dipped in the coating slurry, and the excess slurry adhering to the honeycomb after being pulled up was removed by air blow. At this time, the slurry adhering to the wall surface was wiped off. Thereafter, it was fired in an air at 500 ° C. for 15 minutes. After taking out and allowing to cool, the dry weight was measured. Thereafter, this operation was repeated until the target catalyst adhesion amount was 150 g / L (5 times). Finally, the main calcination was carried out by holding in air at 500 ° C. for 1 h.

The obtained metal honeycomb catalyst was evaluated for catalytic activity by a fixed bed normal pressure flow reaction evaluation apparatus as before. The reaction tube was made of quartz having an outer diameter of 32 mm (inner diameter: 28 mm). A honeycomb catalyst was set at the center of the reaction tube, and quartz wool was tightly packed between the inner wall of the reaction tube and the honeycomb to prevent gas from flowing around the outer periphery of the honeycomb while being fixed. Prior to the reaction, the sample was subjected to hydrogen reduction. In the reduction, 500 mL / min of H ₂ gas is allowed to flow through the reaction tube, and the temperature of the catalyst layer is raised to 500 ° C. at 20 ° C./min. After completion of the reduction, the temperature was lowered to the reaction temperature range and the reaction gas was introduced. The composition of the reaction gas was 0.5% CO, 20% CO ₂ , 6% steam carrier N ₂ and 73.5% H _{2 on a} dry basis. Water vapor corresponding to water vapor / CO = 34 (molar ratio) is added. The space velocity of the reaction gas was set to 10,000 h- ¹ .

22 and 32 show the initial performance of the metal honeycomb catalyst. This honeycomb catalyst has 150 g of catalyst powder in 1 L of space. Compared to a granular catalyst in which about 1 kg of catalyst is present in a 1 L space, it is extremely small. In SV10000h- ¹ , CO _min did not fall below 0.2%, but when SV was lowered to 8000h- ^1, a region of 0.2% or less appeared. Thus relatively freely adjustable Introduction T _opt the CO _min by adjusting the catalyst coating amount is in the honeycomb catalyst. In addition, the metal honeycomb catalyst does not cause thermal runaway even when the temperature is raised to 300 ° C., and is a suitable form for the pre-stage catalyst from the viewpoint that it can be used safely.

  In the first half of the example, it was shown that if the post-stage catalyst can be used at an inlet CO concentration of 0.2% or less, the deterioration can be remarkably suppressed and the durability of 60,000 hours or more required for a practical catalyst can be expressed. On the other hand, the latter example showed that the upstream catalyst can reduce the inlet CO concentration of 0.5 or 1.0% to an outlet CO concentration of 0.2% or less with almost no deterioration. As a result, it was shown that by installing the front stage catalyst and the rear stage catalyst, the inlet CO concentration of 0.5 to 1.0% can be stably reduced to 10 ppm or less without reducing the catalyst performance as a whole.

Claims (12)

A CO selective methanation reactor methanation of CO in the hydrogen-rich gas containing CO and CO _2,
A pre-stage catalyst and a post-stage catalyst in order from the upstream side of the flow of the hydrogen-rich gas in the CO selective methanation reactor,
The pre-stage catalyst is a CO selective methanation reactor in which the optimum operating temperature at an inlet CO concentration of 0.5% is higher than that of the post-stage catalyst. 2. The CO selective methanation reactor according to claim 1, wherein the second stage catalyst has a CO selectivity of 0.5 or more under a condition that an outlet CO concentration is 100 ppm. The CO selective methanation reactor according to claim 1 or 2, wherein the upstream catalyst is configured such that an outlet CO concentration at the optimum operating temperature is 0.2% or less. The CO selective methanation reactor according to any one of claims 1 to 3, wherein the active metal of the pre-stage catalyst contains Ni and Fe. 5. The CO selective methanation reactor according to claim 1, wherein the active metal of the latter catalyst includes Ni, and the support of the latter catalyst includes a V oxide and an Al oxide. . A CO selective methanation reactor methanation of CO in the hydrogen-rich gas containing CO and CO _2,
A pre-stage catalyst and a post-stage catalyst in order from the upstream side of the flow of the hydrogen-rich gas in the CO selective methanation reactor,
The active metal of the former catalyst contains Ni and Fe, the active metal of the latter catalyst contains Ni, and the carrier of the latter catalyst is an oxide containing V or Ni and Al. Reactor. The CO selective methanation reactor according to any one of claims 4 to 6, wherein the atomic ratio of Fe / Ni is 0.01-1. The CO selective methanation reactor according to any one of claims 4 to 7, wherein the support of the pre-stage catalyst is an oxide containing Si. 9. The CO selective methanation reactor according to claim 1, wherein the latter catalyst comprises a coating layer configured to reduce the CO concentration on the active metal surface. The CO selective methanation reactor according to any one of claims 1 to 9, wherein the upstream catalyst is coated on a honeycomb substrate. The CO selective methanation reactor according to claim 10, wherein the honeycomb substrate is a metal honeycomb substrate. The CO selective methanation reactor according to claim 11, comprising an α-alumina layer on a surface of the metal honeycomb substrate.