Sorbent from stabilized aluminium oxide including active ingredient for absorbtion of nitric and sulphur oxides and method for processing them

The present invention relates to an improved sorbent for the removal of nitrogen and sulfur oxides from the exhaust gas and to a method for removing said gas with said sorbent.

The NOXSO method described in U.S. Pat. No. 4,798,711, uses a porous alumina-containing porous sorbent for the absorption of nitrogen and sulfur oxides from flue gas. After absorbing this gas, the sorbent, loaded with nitrogen and sulfur, is regenerated by heating it in a rebuilding atmosphere.

Among other suitable restoring atmospheres, methane gas is used for sorbent regeneration. However, when using natural gas or methane, the regenerator operates at a higher temperature, about 650 ° C, which causes a further reduction in the surface area of ​​the alumina beads.

As the NOXSO sorbent for reversible-sliding movement of aluminum and sulfur oxides moves from adsorption to regeneration, it experiences fluctuations that reduce its surface area. The surface area of ​​the fresh alumina sorbent is about 200 m²/ g. After about 100 cycles, the surface area of ​​the sorbent decreases to less than 50 m²/ g. When the surface area is reduced to such a low level, the performance of the sorbent is no longer acceptable.

It is an object of the present invention to provide a sorbent having a constant surface area and a constant pore volume by regeneration in the presence of sodium or other alkali metals.

It is an object of the present invention to provide a sorbent with a constant surface area for adsorption of nitrogen and sulfur oxides from the exhaust gas.

It is also an object of the present invention to provide a sorbent with a constant surface area under hydrothermal conditions in the presence of sodium.

It is also an object of the present invention to provide a sorbent having a constant performance with respect to adsorption of nitrogen and sulfur oxides over a long period of time and after a prolonged adsorption / regeneration cycle.

It is an object of the present invention to provide a sorbent which retains its integrity when in the form of beads.

It is an object of the present invention to provide alternative methods of preparing a sorbent with a constant surface area for adsorption of nitrogen and sulfur oxides from the exhaust gas.

The object of the present invention is to provide a continuous process for the removal of nitrogen and sulfur oxides in which they are maintained by means of a constant sorbent which can be regenerated at high temperatures and hydrothermal conditions.

These and additional objects of the invention will become more apparent from the following description of the invention.

A sorbent for the removal of aluminum and sulfur oxides has been proposed, which retains a relatively large surface area after continuous use in a cyclic adsorption / regeneration process. The sorbent can also absorb other impurities such as chlorides, HCl and heavy metals such as arsenic, lead and the like. The sorbent is obtained by adding a stabilizer to the suspension forming alumina. The preferred stabilizer for alumina is silica, which can be added in the form of sodium silicate. When alumina is obtained from a suspension, the suspension usually includes, for example, aluminum sulfate and sodium aluminate, which react to form the alumina. According to a preferred embodiment of the invention, an additional alumina stabilizer, such as silica, is added to the suspension. The three components are co-precipitated. Ideally, the precipitates from the combined silica and alumina precipitates, are washed and dried to obtain a powder. The powder can be easily transported to the place where the balls are formed. The powder is then made into a suspension consisting of water / nitric acid / acetic acid. Drop the suspension drop by drop into an irrigation column containing NH phase₃ - liquid hydrocarbons over the aqueous phase of ammonia. Balls of alumina stabilized on silica are dried and annealed. Subsequently, 1-20% by weight of the active ingredient of alkali or earth alkaline metal, preferably sodium in an amount of 4-6% by weight, flashed (by initial wetting).

Short description

FIG. 1 shows a schematic diagram of a method for making a sorbent from a stabilized alumina.

FIG. Figure 2 shows the vapor resistance of the sorbents.

FIG. 3 shows an improved method of preserving the surface area of ​​a sorbent according to the present invention over a period comprising a cyclic adsorption / regeneration process.

It has been found that silica, as well as other alumina stabilizers, are the most suitable alumina stabilizer for the present sorbent. For example, lanthanum oxide, other rare earth metal oxides, titanium oxide zirconium dioxide, clay and alkaline earth metals such as calcium and barium may be used, either individually or in combination, to obtain a stabilized sorbent for a particular application.

Preferably, the amount is from effective amount to, for example, 30 mol%.

There are at least three ways to add a stabilizer to alumina by dripping beads.

Thus, if the stabilizer is mixed with the parent alumina materials without forming a highly viscous solution, then the first step of the process would be to carry out a co-precipitation to obtain a sorbent, for example by co-forming the gels. This method is described in Nozemak U.S. Pat. 4780446, which is herein incorporated by reference.

The alumina stabilized on silica is obtained in the form of a ready-to-use gel, by precipitation at specific and controlled reagent concentrations, temperature, time, pH. The preferred source of acid aluminum salt is aluminum sulphate solution. The preferred source of alkali metal aluminate is sodium aluminate solution and the most preferred source of alkali metal silicate is sodium silicate solution. The aforementioned sources are combined to produce a mixture prepared by co-production of jelly.

The overall jelly-making reaction is carried out in two steps.

(1) The initial total gel is precipitated to a pH of 7.5-8.5, most preferably 8, and (2) the precipitated total gel is stabilized at a pH of 9.6-10.3, and most preferably 10.

The required pH value in the reaction is maintained by controlling the flow rates of reagents: aluminum sulfate, sodium aluminate and sodium silicate. It also adds sufficient sodium silicate to achieve total silica (SiO₂) in the precipitated total gel obtained from an effective stabilized amount to about 30 mol%, more preferably from 5.5 to 8.5% by weight.

The alumina stabilizers that can be added by co-precipitation include silica, lanthanum oxide as well as other rare earth oxides, titanium oxide, zirconium dioxide or alkaline earth metals such as calcium or barium.

The stabilized alumina powder is then used to form substantially uniform spheroidal alumina particles. A preferred method of obtaining the shape of the spheres is described in U.S. Pat. 4279779, which is hereby incorporated by reference. According to this method, the stabilized alumina suspension is prepared in an acidic aqueous medium and the droplets of the suspension are passed through the air to a column in which the supernatant is immiscible with water, e.g. The resulting spheroidal particles ripen in aqueous ammonia to a defined hardness. Ripened particles are dried and annealed.

The improved method for obtaining the sorbent is shown in FIG. 1 and described briefly below. The suspension is prepared by mixing the stabilized alumina powder, water and acids such as acetic acid and nitric acid. Thereafter, it is desirable to reduce the particle size of the suspension, for example, to reduce the average particle size from greater than 10 microns to 2-5 microns. Although sorbent can be obtained without the desired crushing step, all of the resulting beads will have a low relative weight, high wear rate and low crushing resistance. Therefore, crushing is the preferred way to reduce particle size.

Other particle size reduction techniques can be used, such as dry milling, hammer or ball milling, and fluid energy milling. These methods can be wet or dry type.

The slurry is fed into a large industrial-type Netzsch shredder at a rate of 17.62 liters per minute (4 gallons per minute). The shredder itself holds 176.16-242.22 liters (40 to 55 gallons) of Ouackenbush beads, 1.5mm in size, which are constantly circling inside with the aid of discs mounted on the rotating axis. The suspension particles are then pulverized to the required size by the action of glass beads. The particle size is controlled by reducing or adding a specific amount of glass beads.

After grinding the suspension particles, the suspension is dropped into a humidification column, preferably ammonia-kerosene over an aqueous ammonia solution. The organic phase is necessary to form spheres while the aqueous phase additionally strengthens the beads. After the droplets have been formed, the beads are transferred to a dewatering system and then passed through a dryer, at which time they direct airflow through the beads. After drying, the sorbent is heated in a rotary kiln with indirect heating. Particles that are too large or too small are removed by a vibratory separator. It then flashes with the active ingredient, which is an alkali metal or an alkaline earth metal, most preferably in the presence of sodium,₂C0₃ form. This can be done with a horizontal rotating drum soaking device. Finally, the sorbent is dried and annealed to give the final product.

Another way to add the alumina stabilizer is to mix it with the alumina powder until the droplets form the particles. All stabilizers, including clay, can be added in this way.

A third method of adding the alumina stabilizer is to drop the alumina particles dropwise and then to immerse the particles in the alumina stabilizer. The stabilizer can be added before and after the addition of sodium, or it can be added together with sodium to make it more economical. In this way all stabilizers except clay can be added as it is insoluble.

Other ways to form beads can be used in addition to the described bead beading techniques. The ingredients may be mixed together and the mixture may be extruded until extrudates are obtained. Preferably, the ratio of length to diameter is about 1: 1. The mixture can be placed in an agglomerator for sintering the balls. Other mechanical forming methods, such as tableting, may also be used.

The sorption rate of sorbent nitrogen and sulfur oxides is measured by an adsorption / regeneration unit. The unit has a pseudosolid deposited type bed with waste gas containing 3000 particles per million SO₂, 500 particles per million NO, 4% O₂ and 10% H₂O

The cycle begins with the passage of gas through the liquid sorbent layer at a temperature of about 140 ° C. The average conditions during the first 10 minutes of periodic presence in the fluidized bed are roughly analogous to the conditions under which the sorbent will be used in industry with an estimated sulfur content of 1.5% by weight. The operation is continued for an additional 10 minutes at a sulfur content of about 2.5% by weight. The absorption characteristic of nitric oxide is continuously measured with a chemical luminescent NO analyzer (Termo Elektron Instruments). The sulfur oxide absorption characteristic is measured by the sulfur content of the sorbent samples, which are removed from the fluidized bed every 2 to 5 minutes. The adsorption phase is completed and the beads are then regenerated for 20 minutes in methane at 650 ° C followed by steam treatment at 650 ° C for a minute. The regenerated beads are cooled to an adsorption temperature of 120-160 ° C and the preferred temperature is 140 ° C.

After the given number of cycles, the beads are removed from the adsorption / regeneration unit and their surface area is measured. The desorption of nitric oxide is determined by the programmed pre-desorption temperature. Place the beads in a regenerator connected to a gas chromatograph. The desorption of sulfur oxide is measured over a period of 20 minutes, during which time methane is released at 650 ° C.

This allows the products S, SO to be dispersed₂ and H₂S. It then allows steam at 650 ° C for 20 minutes to separate any remaining H₂S.

The following sections describe the applicable test methods as defined in the invention. During abrasion and crushing tests, the balls are sieved to pass through 10 wells and retain 20 wells.

Time dependence of sorbent stability

Uses the determination of the sorbent stability over the cycle to determine the effective sorbent shelf life. This method compares the surface area after 20 cycles of adsorption / regeneration (sufficient time to reduce the initially large surface area of ​​fresh sorbent to its working level) with the surface area of ​​the sorbent after 100 cycles. This is a sufficiently long period of time to evaluate how the sorbent retains the conditions of the repetitive cycles. Sorbents made according to the present invention generally have a surface area that is reduced from more than 200 m²/ g when fresh, up to about 150 m²/ g after the twentieth cycle. Thereafter, after an additional 80 cycles, the surface area decreases to less than 60 m²/ g. Other aluminum oxide sorbent particles having a uniform sodium content will have significantly greater than 60 m²/ g surface area loss under the same test conditions and will have a lower total surface area.

Resistance to crushing

Crushing resistance is determined by placing spheroidal particles between two parallel plates of a test machine, such as an Ametek ML4434-4 model device manufactured by AMETEK Co., 8600 Somerset Drive, Largo, Florida, which is mounted on a motorized inlet platform. The lower plate is anterior and the upper plate is a flat pentium. When the release button is pressed, the panels slowly converge into one place, at which time the force measuring device slides down the drive platform at a speed of 1.24 mm / s (0.049 in. Per second). The magnitude of the effect required to chop the particles is recorded by a digital calculator, graduated in pounds of exposure. A sufficient number (for example, 25) of the grooved particles are disintegrated individually to obtain a statistical picture of the overall population. Then calculate the total size from the individual results. Measuring the balls under steady conditions, place the balls in a porcelain evaporator and heat them on a Bunzen electric burner at a temperature of at least 400 ° C. The bead is then removed from the container and, when hot, measures its resistance to crushing.

When the beads are to be used in a fluidized bed, it is desirable to have an average crush resistance of at least 1.82 kg (4 lbs). The significance of such high shredding resistance is that the beads will not collapse during physical processing under severe conditions. Under such conditions, they find themselves in a fluidized bed and other mechanical devices used in industry.

Wear detection (GM)

Sorbent particle abrasion is measured by the air current test method. This method involves placing a quantity of test substance (60 cubic centimeters) in an inverted Elenmeyer flask, fitted with a metal nozzle inlet. The flat side of the flask (bottom) is equipped with a large 25.4 mm (1 inch) outlet, covered with a 20 mesh mesh. It directs nitrogen gas through the inlet nozzle at high speed. This causes the particles to: 1) circulate on top of each other as they drop, 2) bump into each other at the top of the flask. This breaks them down depending on the resistance. The material is tested for five minutes and the remaining particles are then weighed. The weight loss after the test, expressed as a percentage of the initial weight of the material, indicates the loss of wear. The nitrogen output is about 0.088-0.1133 cubic meters per minute (3.1-4.0 cubic minutes per minute) depending on the density of the material. The flow rate must be sufficient to allow the particles to collide with each other continuously in the upper part of the flask. As a result of abrasion, the resulting fines are removed from the flask by a stream of nitrogen. Thus, it results in weight loss of the preloaded material.

The significance of low abrasion size is that the particles are not readily degraded when they are contained in a fluidized bed adsorber, in a regenerator or when moving from one installation to another. In a large installation, a cost effective sorbent is considered to have less than 1.2% daily wear.

The abrasion test air flow test method described in the Erlenmeyer flask preferably has a wear rate of less than 2% and an industry-standard rate of less than 1% when tested in this accelerated manner.

Wear during adsorption / regeneration cycle

The periodic adsorption / regeneration test is used to measure sorbent wear during the adsorption / regeneration (ARC) cycle. A sorbent with a particle size in the range of 10 to 20 meshes adsorbs at a temperature of 140 ° C in a fluidized bed exhaust gas containing 3000 particles per million SO₂, 500 particles per million NO, 4% O₂ and 10% H₂And until the sulfur reaches 1.5-2%. The sorbent is then regenerated with methane at 650 ° C for 20 minutes, followed by steam treatment at 650 ° C for 20 minutes.

This method of adsorption / regeneration is repeated a second time. The sorbent is then sifted and weighed. Any sorbent with a particle size of less than 35 meshes is considered to be worn. The daily wear percentage is based on the percentage of all material having a particle size of less than 35 mesh, which is then normalized to reflect 8 adsorption / regeneration cycles per day. It is understood that the required amount of chemical abrasion is less than 10% per day, preferably less than 5% per day, and preferably less than 1.2% per day for industrial application of a sorbent in a fluidised bed.

Surface area

The surface area referred to herein and in the definition of the invention is the surface area determined by the BET equation as described by S. Brunauer, P. Emmett, and E. Teller in J. Am. Chem. Soc., Volume 60, p. 309 (1938). This method depends on the condensation of nitrogen into the pores and is also effective in measuring pores with a diameter range of 0.001 to 0.06 micrometres (10-600 angstroms). The volume of adsorbed nitrogen depends on the surface area per unit weight of carrier.

Content density of coagulated particles

To determine the bulk density of the coagulated particles, the mass of freshly cured steroid is placed in a graduated cylinder with a graduated volume sufficient to contain the material. The cylinder is then shaken until sedimentation is complete and a constant volume is obtained. Then calculate the mass of the sample occupying the volume unit.

Total porosity

The pore distribution inside the activated spheroidal particles and the total porosity are determined by a mercury device for determining porosity. The way mercury is used is based on the principle that the smaller the pair, the higher the mercury pressure needed to get the mercury into that pair.

In this way, if the sample with aspirated air is exposed to mercury and uses increasing pressure such that the mercury volume indication disappears as the pressure increases, the pore size distribution can be determined. The relationship between the pressure and the minimum pore that mercury will pass through is given by: r = 26cos0 / p, where r is the pore radius, δ is the surface tension,

Θ - contact angle, p - pressure.

Adding a pressure of up to 166.23 kg per square centimeter (6000 pounds per square inch) and using a contact angle of 130 ° produces a pore diameter range of 0.0035 to 1 micrometer (35 to 10,000 angstroms).

The mercury porosimeter measures the total volume of pores filled with mercury in cm³/ g. Particle density is measured in g / cm³. Multiplying both of these sizes determines the porosity in cm³/ cm³.

Hydrothermal stability

The stability of the sorbent surface area according to the BET equation is measured by operating the sorbent beads in a 100 vol% vapor atmosphere at 650 ° C and regularly measuring the surface area. The sorbent must be kept at least 100 m during this test²/ g surface area after 1500 hours exposure to vapor and preferably over 140 m²/ g.

Following the disclosure of the most important aspects of the invention, the following examples will now be described to illustrate its use.

example

According to the method described in U.S. Pat. No. 4780446, prepared an alumina powder stabilized with silica. The powder contained 6.5% silica. This powder was manufactured by the Davison Chemical Division of WR Grace & Co.. Conn. as aluminosilicate. The powder has a moisture content of 28% when heated to 954 ° C (1750 ° F). The average particle size of the powder measured with a Malvern particle analyzer is 15-20 microns.

example

According to this method, which is disclosed in U.S. Pat. No. 4154812, prepared alumina powder. This powder is sold under the name SRA alumina. They are made in the Davison Chemical Division by WR Grace & Co - Conn. the powder contains 28% moisture when heated to 954 ° C (1750 ° F). Measured on a Malvern particle analyzer, an average particle size of 15 to 20 microns was obtained.

example

From 45.4 kg (100 pounds) of alumina powder (dry base) stabilized on silica prepared according to Example 1, a suspension of 132.12 liters (30 gallons) of water, 2404 grams of 70% nitric acid, and 3237 grams of glacial vinegar was prepared. acid mixture. All this was stirred for 35 minutes in a high shear mixer. The slurry was then rinsed with a Netzch mill with a capacity of 20 liters. The average particle size dropped to 5 microns and less. The suspension was then poured into a column containing kerosene-NH₃ phase above aqueous NH₃ phases. The wet spheres were dried and heated in the air. Spherical alumina particles stabilized with silica were soaked in sodium in two ways. They were soaked in one batch to 4% by weight of sodium using the moisture content of sodium carbonate, and in another batch they were soaked in up to 6% by weight of sodium. This way, the beads can adsorb a certain amount of moisture equivalent to 80-100% of their pore volume. The particles were then dried at 120 ° C in a vacuum dryer and heated in air for 6 hours. At 650 ° C.

The particles soaked in 4% sodium (named as 3A) had a surface area of ​​254m²/ g, bulk density of coagulated particles 632,3569 kg per cubic meter (39 pounds per cubic foot), average chipping size 4.6308 kg (10.2 pounds), total porosity measured with mercury, 0.712 cm³/ cm³ and wear loss (GM) of 0.2%.

Particles soaked in 6% sodium (designated as 3B) had a surface area of ​​210 m²/ g, bulk density of coagulated particles 648.5712 kg per cubic meter (40 pounds per cubic foot), average chipping size 3.632 kg (8 pounds), total porosity measured with mercury, 0.714 cm³/ cm³, wear loss (GM) 0.1% and wear loss (ARC) 0.3%.

To test the hydrothermal stability of the beads, the particles were exposed to 100 vol% steam at 650 ° C and measured surface area at defined time intervals. FIG. 2 shows that a sorbent containing 4% sodium stabilizes at 200 m²/ g and sorbent containing 6% sodium at 170 m²/ g.

a comparative example

The alumina suspension was prepared in the same manner as in Example 3 except using the alumina powder from Example 2 which did not contain any silica.

When preparing the suspension, initially used 88.08 liters (20 gallons) of water instead of 132.12 liters (30 gallons) to obtain the required viscosity. It was dripped and dipped in sodium to a consistent level. The particles were then dried and annealed as shown in Example 3A. After steam treatment at 100% by volume at 650 ° C, a sharp decrease in surface area was observed as shown in FIG. 2.

The surface area of ​​the particles was 211 m²/ g, bulk density of coagulated particles 582.09265 kg per cubic meter (35.9 lbs per cubic foot), average chipping size 3.1326 kg (6.9 lbs), total porosity measured with mercury, 0.676 cm³/ cm³, wear loss (GM) 0.7% and wear loss (ARC) 0.2%.

example

The sorbent containing silica from Example 3B and the sorbent from Comparative Example 4 which had no silica were tested by determining the sorbent stability over the cycle. These two samples had different amounts of sodium. However, the comparison of these two sorbents is believed to be correct in terms of surface area characteristics.

If one has the same base with different amounts of sodium, then the sorbent with the higher sodium content will have a smaller surface area because the extra sodium will take up (ie absorb) the extra surface area. See. Example 3, wherein sorbent 3A containing 4% sodium will have a surface area of ​​254 m²/ g then the 3B sorbent with a higher sodium content of 6% will have a smaller surface area of ​​210 m²/ g. Thus, if the sorbent according to the present invention is high in sodium, it will have a smaller surface area than the lower sodium sorbent, and thus such a comparison is advantageous since it can be estimated that the sorbent from Comparative Example 4 will have a larger initial surface area than sorbent from Example 3B with higher sodium content. However, it appears that the fresh sorbent in Example 3B (with higher sodium content) will actually have approximately the same surface area. Apparently, the reason is that the added silica, which acts as a stabilizer, also provides additional surface area, offsetting the surface area loss caused by the additional sodium.

The sorbent according to the present invention in Example 3B was 232 times closed and its surface area decreased to 70 m.²/ g. A sorbent made according to Comparative Example 4 without stabilizer already had a surface area of ​​60 m after 100 cycles²/ g. The change in surface area over the cycles is shown in FIG. 3.

The decrease in performance can be observed from the data in FIG. Table 3 and I comparing the sorbent surface area after 20 cycles with the surface area after 100 cycles, and comparing the surface area descending rate between the 20th and 100th cycles.

Table I

Surface area, m²/ g cycles 100 cycles Decrease

Example B

150 100 Comparative Example

125 60

Without stabilizer, the surface area is reduced by 65 m²/ g over 80 cycles up to 60 m²/ g, while the surface area of ​​the stabilizer sorbent according to the present invention is reduced by only 50 m²/ g over 80 cycles and is above a good 100 m²/ g level. The sorbent of the present invention, which includes a silica stabilizer, has a larger surface area. While highly desirable, an even greater advantage is that the surface area of ​​the silica-stabilized sorbent decreases more slowly than the surface area of ​​the non-stabilized sorbent.

It is to be understood that the following description of the invention is set forth in detail by way of illustration, and that many modifications are possible without departing from the scope of the invention.