MXPA96004054A - Absorben - Google Patents

Abstract

The present invention relates to formed absorbent units having an average size within the range 1-10mm, and containing at least 75% by weight of an absorbent material selected from copper carbonate, basic copper carbonate, copper hydroxide and mixtures of them, the units having a mass density of at least 1.2kg / 1, and a BET surface area of at least 10m2

Description

This invention relates to absorbers and in particular to copper-containing compositions for the absorption of sulfur compounds such as hydrogen sulfide, which are frequently present as impurities in gases and liquids such as hydrocarbon streams, for example, natural gas. There have been numerous proposals for using compounds containing copper in the composition for the absorption of such sulfur compounds; in many of the previous propositions copper is present in the composition as an oxide frequently together with other components such as zinc oxide and / or alumina. Often the copper oxide is reduced to the corresponding metal before use, or it is used to treat gas streams containing a reducing gas such as carbon monoxide or hydrogen at such a high temperature that the reduction of the oxide occurs in the same place. . In our European patent EP-A-0 243 052, we demonstrate that certain compositions containing copper compounds in the form of agglomerates, for example granulates, as opposed to tablets or molded balls, have superior sulfur absorption characteristics, and can be used without a reducing step having good absorption characteristics at low temperatures. The agglomerates were formed by mixing the composition containing the copper compounds with a cement linker and a little insufficient water to form a slurry and then causing the composition to agglomerate into coarse spherical granules. Alternatively, the granules can be formed by extruding the wet composition, in that aspect, the composition containing zinc or aluminum compounds, preferably both further, of the copper compounds and the cement bond. Preferably, the zinc atoms formed from 10 to 40% of the total number of copper, zinc and aluminum atoms. Specifically, we describe the production of agglomerates using a copper compound that contains a composition produced from the basic copper and zinc carbonates precipitated together, for example, a malachite substituted with zinc with a precipitate of zinc alu-free. We describe that such agglomerates that have been dried but not calcined at a temperature high enough to decompose basic carbonates were useful for absorbing hydrogen sulfide at low temperatures and were superior to agglomerates that had been calcined. US-A-4 582 819 discloses the use of agglomerates formed by extruding a mixture of alumina and basic copper carbonate, for example malachite, followed by calcination for the absorption of sulfur compounds. Under the calcining conditions presented, for example, heating above 260 the malachite will decompose into copper oxide, the agglomerates of this reference contain substantial amounts (above 25% by weight) of alumina. The amount of sulfur that can be absorbed from a gas stream depends on the absorption capacity of the absorbent, the amount of absorbent used and the absorption profile given by a bed of the absorber. The absorption capacity of an absorbent is the theoretical amount of sulfur that can be absorbed by a given weight of an absorbent. Thus one kg of an absorber composed of copper oxide or zinc oxide and containing 10% by weight of a non-absorbent material, for example alumina, has a theoretical capacity to absorb approximately 350-360 g of sulfur, on the other hand, one kg of a malachite or malachite absorber substituted with zinc also having a content of 10% by weight of non-absorbent material, will have a theoretical sulfur absorption capacity of only about 250-260g.

It is generally not possible to achieve the total theoretical capacity of the absorbent bed, thus some sulfur compounds will be detected in the product stream that leaves the bed before the theoretical capacity has been achieved., the reason for this, is that the absorption front for the sulfur in the bed is not acute, the more acute the absorption front, the closer we will approach the theoretical capacity. We have found that despite its low theoretical capacity compared to copper oxide or copper metal, absorbents made of copper carbonate, basic copper carbonate, or copper hydroxide can actually give a higher sulfur capacity. Therefore, the present invention provides molded or formed absorbent units having an average size within the range of 1 -10mm, and containing at least 75% by weight, of an absorbent material selected from copper carbonate, basic copper carbonate, copper hydroxide and mixtures thereof; The weight of absorbent that can be used depends on the volume of the container that contains the absorbent bed and the density of mass or volume of the absorbent. It is desirable to maximize the weight of sulfur that can be absorbed by a bed of absorbent, that is, by a given volume of absorbent. Therefore, it is desirable to maximize the mass density of the absorbent as long as the other absorption characteristics are not adversely affected, the mass or mass density of the particles is the density of a bed of particles and is determined by filling a container of known volume with the particles, then stirring the container to ensure the settlement of the particles and then determining the weight of the particles in the container. The mass density is of course considerably less than the density of the particles themselves, since a considerable part of the volume of a bed of the particles is taken up by the spaces between the adjacent particles. For example, a bed of bales as described in example 3, has a density of 1.07kg / l, whereas the density of the individual balls, that is, the density of the particle was approximately 1.85kg / l]. We have found that it is possible to produce absorbers comprising copper carbonates, basic carbonates and hydroxides with a mass density of at least 0.9 kg / l, particularly at least 1.2 kg / l.

The units formed can be in the form of tablets, formed by molding a suitable powder composition into molds of suitable size, that is, as a conventional operation to form tablets. Alternatively the units formed can be in the form of extruded balls, formed by forcing a suitable composition containing the absorbent material and frequently a bond and a little water and / or a molding aid as indicated above, through a die, to the following cut the material emerging from the die in short lengths, for example, the balls or extruded pellets can be made using a pellet mill of the type used to make balls with animal feed, where the mixture will be formed into pellets it is loaded in a rotary perforated cylinder and through the perforations the mixture is forced by a bar or a roller that is inside the cylinder, the resulting extruded mixture is cut from the surface of the rotating cylinder with a scalpel placed to give the extruded pellets the desired length. Alternatively, and preferably, the units may be formed in the form of agglomerates that are formed by mixing the absorbent material with a cement bond and with a little water that is insufficient to form a slurry, and then causing the composition to agglomerate in roughly spherical granules, but generally irregular. The different shaping means have an effect on the surface area, porosity and pore structure within the molded articles and this in turn frequently has an important effect on absorption characteristics and mass density. Thus, absorbent beds in the form of molded tablets can have a relatively wider absorption front, while the agglomerates can have a more acute absorption front, this allows closer approximation to the theoretical absorption capacity. On the other hand, the agglomerates have, generally lower mass densities, than the tablet compositions; however, as described below, agglomerates with a suitably high bulk density can be produced and it is thus preferable to make the shaped units in the form of agglomerates. Preferably, the shaped units have a BET surface area of at least 10m2 / g. In the formed units of the invention, the absorbent material comprises copper carbonate, basic copper carbonate and / or copper hydroxide. Basic copper carbonate, especially malachite, is the preferred absorbent material, and the shaped units are preferably agglomerated.

As indicated above, the agglomerates are usually formed by mixing the absorbent material with a bond and a little water and then forming the composition into granules. The binder or binder is preferably a cement such as, a calcium aluminate cement. Since the binder represents a non-absorbent material, its proportion is kept to a minimum consistent with obtaining agglomerates of sufficient strength to withstand handling and loads encountered during normal use. The amount of cement is typically 5-10% by weight of the agglomerates. Similar amounts of cement can be used when the patterned articles are in the form of extruded pellets. In the aforementioned patent EP-A-0 243052, it was preferred that the agglomerates contain zinc compounds, for example, basic zinc carbonate. We have found that at low absorption temperatures that below about 100 copper compounds are superior absorbers for hydrogen sulfide compared to zinc compounds. Therefore, it is preferred that the modeled absorbent units of the present invention contain less than 10% by weight of the zinc compounds and preferably no zinc compound, although it will be appreciated that such compounds can inevitably be present as impurities in the other components. For most applications, the shaped units, that is, the agglomerates may simply comprise an absorbent material selected from copper carbonate, basic carbonate and / or hydroxide and a binder. Where the absorbent is to be used for the absorption of hydrogen sulfide, preferably at least 90% by weight of the shaped unit is such an absorbent material. However, as described below, when the absorbent is to be used for the removal of hydrolysable sulfur compounds such as carbonyl sulfide, the shaped units will also preferably contain a high surface area of gamma alumina, to catalyze the hydrolysis of the carbonyl sulfide; in this case, the shaped units preferably contain from 9 to 20% by weight of such alumina, and thus, the content of copper carbonate, basic carbonate or hydroxide, is typically in the range of 75 to 85% by weight. The absorbent material, for example, basic copper carbonate, can be naturally occurring mineral that has been ground to a powder of suitable size, or can be a synthetic product obtained by precipitation. Where it is required that alumina, in addition to that present in the cement binder, be included, that alumina may be added as a powder to the powder of the absorbent material or may be introduced by precipitation; thus, it may be coprecipitated with the absorbent material or the latter may be precipitated in a slurry of precipitated alumina or the alumina may be precipitated in a slurry of the absorbent material. Alternatively, separate slurries of precipitates can also be mixed. After such precipitation the precipitates are filtered and dried. But the drying conditions must be such that the copper carbonate or the basic carbonate does not decompose in any considerable amount. Thus the drying should be carried out at a temperature not greater than about 150 and preferably at a temperature not greater than 115. Where the molded units are agglomerated or extruded pellets, in order that those units can be made with a density of high mass, it is desirable that the powder of the absorbent material used to make the molded units have a particle size and a particle size distribution such that the particle size Dso is between about 4μm and 12μm, the size being of particle Dso of approximately 1.4 to 2.5 times the Deo value and the particle size Dio approximately 0.15 to 0.5 times the Deo value (with the previous terms Dio, Deo, DTO we understand the sizes at which 10%, 50% and 90% in volume, respectively, of the particles have a smaller size than the indicated value). With powders of particle size or particle size distributions outside those margins it can be difficult to reach agglomerates of a high mass density. The absorbent units of the invention are particularly suitable for removing hydrogen sulphide from gas or liquid streams at low temperatures. The temperature at which they are used should be less than about 150 °, at which temperature a considerable decomposition of the copper carbonate or the basic carbonate can begin to occur. Preferably, absorption takes place below 100 degrees, particularly where the absorbent material comprises copper hydroxide, the temperature is more preferably in the range of -10 to 80 °. The absorption process can be carried out at any suitable pressure, typical pressures ranging from atmospheric to approximately 200 bars absolute. When the treated gas liquid contains carbonyl sulphide and / or carbonyl disulfide or hydrogen sulphide instead, it is desirable to use absorbers that contain an alumina content. In addition, of any alumina present in the binder, the alumina must be a gamma alumina. with a high surface area, preferably having a surface area of at least 150m2. Alumina helps the reaction of carbonyl sulphide or carbon disulfide, it is not normally necessary to add water to the fluid to effect such hydrolysis, so some hydroxyl groups will normally be associated with alumina and the water develops with the absorption of hydrogen sulphide produced by hydrolysis, for example COS + HsO * CO2 + H2S Cu (0H) 2CuC03 + 2HsS + CO2 + 3H20 and thus it will be available for further hydrolysis. In fact with conventional zinc or iron sorbents, it has been found that the addition of water to the gas stream is desirable to maximize the amount of sulfur that a given volume of absorbent can absorb. The addition of water to the hydrocarbon gas streams, however, is desirably avoided due to the risk of the formation of hydrocarbon hydrates, which at high pressures, can be separated and cause blockages in the pipes. In contrast we have found that there is no need to add water when the component is a copper compound as specified above. The fluid being treated may be a hydrocarbon stream, for example natural gas, natural gas substitute, natural gas liquids, naphtha, reforming gases, for example hydrocarbon streams such as propylene separated from the naphtha cracking product. , synthesis gas produced, for example by the partial oxidation of a carbonaceous supply, organic compounds such as alcohols, chlorinated hydrocarbon esters or other gases such as carbon dioxide, hydrogen, nitrogen or helium. In order to maximize the absorption capacity of an absorbent bed, it is preferable to carry out the desulfurization process using two beds in series. By this means it is possible that the first bed can become fully saturated with sulfur compounds before an unacceptable breakdown or decomposition of the sulfur compound in the effluent occurs from the second bed. When the first bed is completely saturated, it is replaced with fresh absorbent and the order in which the treated fluid travels through the beds is inverted. Thus the partially saturated bed which was the second bed will now be used as the first bed through which the fluid passes and the fresh bed becomes the bed through which the fluid passes after passing through the partially saturated bed. The present invention is illustrated in the following examples in which samples of basic copper carbonate (malachite) powder or powders of different particle sizes and with different particle size distribution were used as follows: The mass density was measured by loading the material being tested into a cylinder until the measuring cylinder after hitting the side of it to allow settlement was filled to the 60ml mark, with the cylinder being 100 ml. The weight of the material in the cylinder was then determined. EXAMPLE 1 Units formed as agglomerates were made by mixing 93 parts by weight of the basic copper carbonate powder with seven parts by weight of calcium aluminate cement and a little water, insufficient to give a slurry and the mixture was molded into coarsely spherical agglomerates. whose mass had approximate diameters 3-5 using an agglomerated laboratory granulator of the size of that range were separated from the rest of the composition, sifting and drying at 110 ° for four hours the agglomerates made of powders A, B and C had mass densities in the range of 1.33 to 1.37 kg / 1 but the agglomerates made of powder D had a density of only 0.85 kg / l, indicating that it was desirable to employ a high density powder in order to obtain high density agglomerates. Samples of other basic copper carbonate powders of a high mass density but of a larger particle size (D? Ol6 yum Deo and 41.7 um and DTO 44.8 and 74.3μm respectively) give only low yields of agglomerates of the size wanted. The absorption characteristics of the dry agglomerates made of powders A, B and C were determined by passing natural gas containing 1% by volume of hydrogen sulfide at atmospheric pressure and 20% through a vertical cylindrical bed of the appropriate agglomerates of height 12cm and a height to diameter ratio 5 and a speed of approximately 700 / h the time taken before the hydrogen sulfide could be detected at a level of 1 ppm in the exit gas was determined and is indicated in the following table as the break-through time (B-T).

Then samples were taken from different bed heights (part 1 is the first 2cm depth of the bed, portion 2 is portion 2-4, and so on, so part 6 is the bottom of the bed, this is , the part 10-12cm) from the top and analyzed for the sulfur content the results are shown in the following table cdt > a zu f re (; - by weight) dens i d-ad t. intec r na sa agí (h s) μ > echo of 1 vo p r on BD íkg / l) time (hrs) 1 2 3 4 5 6 A 1.37 20 28 30 29 28 15 2 22 B 1.35 16 17 29 28 25 12 2 19 C 1.33 15 8 25 29 26 13 3 17 The theoretical sulfur absorption capacity of the agglomerate was approximately 360-370g / l. at the theoretical absorption capacity, each portion of the bed would have an approximate sulfur content of the beds in the range of about 210 g / l (for agglomerates made from powder C) to about 270 g / l (for agglomerates made of powder A) . It is seen from the above data that the sulfur content of at least parts 2 to 4 of the bed was close to the theoretical maximum. Therefore, if two absorbers, that is, two beds corresponding respectively to portions 1-3 and 4-6, were used in series, the first bed, this is corresponding to portions 1-3, would become completely saturated, that is, it would have a average sulfur content of approximately 29% by weight (in the case of the agglomerates made from powder A) which is close to the theoretical maximum, before an unacceptable amount of sulfur is detected in the effluent from the second bed, corresponding to the 4-6. At this stage the second bed (in the case of agglomerates made from powder A) has an average sulfur content of about 15% by weight and so if it is used as the first one in a series of beds it will have the capacity to absorb approximately Same amount of sulfur again, before it needs to be replaced. EXAMPLE 2 (Comparative) By way of comparison the agglomerates were made in accordance with the teaching of the EP-A 0 243 052, mixing 93 parts by weight of malachite powder / alumina substituted with co-precipitated zinc without calcining (where copper, zinc and aluminum were in the atomic proportions Cu 55, Zn 27, Al 18, with 7 parts by weight of calcium aluminate cement powder and granulating the mixture as described above The size of Dio, Dßo and DTO particles of the malachite / alumina powder was 34.7.7 and 17.5 um respectively The dry agglomerates had a density 0.95 kg / 1 and a sulfur absorption capacity of approximately 230g / L. The results of the sulfur absorption test of a bed of the agglomerates by the method described in example 1 were as follows: with . azu f re (in p * »so) density t .. i-nterr '----__.-_ - __ nasa agí (rs) po rci on of bed \ -C OPl 6 0.95 12 20.3 22.0 21.9 18.9 8.1 1.8 15.5 The theoretically calculated and actual absorption capacities of the sulfur were 371 and 34g / l respectively for the tablets and 175 and 33gl for the agglomerates. EXAMPLE 3 (Comparative) Following the procedure of the example of the US 4,521,387,200g of copper nitrate trihydrate 3132 of zinc nitrate hexahydrate and 414g of aluminum nonahydrate were dissolved in water or the solution was diluted to 10 1. Another solution was made by dissolving 2180 g of sodium carbonate in water and diluted to 10.28 1. The two solutions were fed separately to a continuously agitated precipitation apparatus at 80 ° controlling the feed rates so that the pH was maintained at 7 ° C. -7.5. The resulting precipitate was stirred for 45 min at 75 ° and then the precipitate was filtered and washed. The precipitate was dried in a furnace at 115"degrees for 12 hours and then calcined at 270 ° to a constant loss of ignition at 900 ° of 10% by weight The analysis of the resulting powder showed the following composition (by weight) CuO 42.7% ZnO 53.3% A portion of the calcined was then mixed with 2% by weight of graphite and the mixture was molded into cylindrical tablets of 5.4 mm diameter and 3.6 mm using a laboratory tapping machine. The resistance to crushing of the resulting tablets was obtained by applying an increasing load between the flat faces of the cylinder until the pellets were broken, the average load required to crush the tablets was 123 kg which corresponded to a resistance of 537 kg per square centimeter . The tablets had a mass density of 1.07 kg / l, while the density of the individual tablets, that is, the density of the particles was 1.85 kg / l. Another portion of the calcined precipitate was mixed with 6.5% by weight of cement, calcium aluminate and a little water and the mixture was molded into spherical agglomerates of diameters between 3.35 and 4.85 mm using a laboratory granulator. The agglomerates were dried at 115 ° for 4 hours. The load required to crush the agglomerates was 0.4 kg. The mass density of the agglomerates was 0.54 kg / 1. The test of the beds of the tablets and the agglomerates for the absorption of hydrogen sulfide by the method described in example 1 gave: EXAMPLE 4 The procedure of Example 1 was repeated using 90 parts by weight of basic copper carbonate powder E and 10 parts by weight of the calcium aluminate cement and using a larger-scale granulator that would provide greater tear on the mixture during the granulation. The resulting agglomerates had a mass density of 1.6kg / l and gave the following results when providing for the absorption of hydrogen sulfide as example 1. In this example the sulfur capacity of the bed portions is indicated g / 1 rather that the sulfur content of the portions in% by weight. a t 1 opi. sulfur content (q / l) powder BDf gl) bed portion rum, 1.6 346 346 332 258 114 12 235 The theoretical absorption capacity of the event was approximately 410/420 g / 1. EXAMPLE 5 Units formed in the form of agglomerates were made by the procedure of Example 1 using 10C parts by weight of the powder E, 10 parts by weight of alumina range of BET surface area 185m2 / g and 7 parts by weight of an aluminate cement of calcium. The resulting agglomerates had a surface area of 27.4m2 and contained approximately 85.5% by weight of malachite. A bed of the agglomerates had a mass density (BD) of 1.33kg / l. The agglomerates were tested for the absorption of hydrogen sulfide as in example 1. The breaking time (B-T) was 17.4h and the average sulfur content was 170g / l. It is seen by comparison with example 1 that the incorporation of alumina decreased the capacity of sulfur absorption by the bed. The absorption test was repeated using a fresh sample of the agglomerates and using carbonyl sulphide instead of hydrogen sulfide. The imrruption time taken before the sulfur could detect (at the lOppm level) in the exit gas was 16.4h and the average sulfur content of the discharged bed was 124g / l. EXAMPLES 6-11 The agglomerates were made as described in Example 5 with different amounts of gamma alumina and using aluminas of different types. The absorption of carbonyl sulphide was tested as described in example 5, the results together with those of example 5, are shown in the following table to luruffia aq lome idos retreat COS Fq.; art SA palak i SA BD .Pi "capacity (pAfc) (wt%) I (rrrVg) (hs) d zuf re (q / I) (kg?) 5 10 185 85.5! 27.4 1.33 16.4 124 j 6 20 185 78.7 i 45.0 1.31 20.0 163 7 5 185 89.3 18.9 1.35 3.2 25 8 10 65 85.5! 14.6 1.31 0.6 5 9 10 4 85.5 12.6 1.33 0.5 3 10 10 258 85.5 18.3 1.31 9.8 83 11 10 128 85.5 23.4 128 3.1 24 The alumina used in Examples 5-7 was gamma alumina. The aluminas used in Examples 8-11 were as follows: Example 8 range alumina calcined at 900 ° to reduce its surface area. Example 9 alumina trihydrate Example 10 Calcined alumina trihydrate at 350 ° Example 11 Calcined alumina trihydrate at 700 ° Similar results to those of Example 8 were obtained with gamma alumina calcined at 1100 ° and at 1250 ° to obtain aluminas from surface areas 32 and 8m2 . These examples indicate that it is desirable to employ a high surface area range alumina if the carbonyl sulfide is to be effectively removed. Example 7 demonstrates that the use of a small amount of alumina, for example 4.5% by weight of the agglomerates does not provide that the carbonyl sulphide is satisfactorily removed. Examples 12-14 Cylindrical tablets of 5.4mm diameter and 3.6mm height were molded using a laboratory machine to make tablets formed of 100 parts by weight malachite E powder, 2 parts graphite as a lubricant, and in examples 13 and also 7 and 11 parts by weight of calcium aluminate cement. The resulting tablets were tested in their absorption of hydrogen sulfide as in Example 1. Results were as follows. ogen ) Compared to examples 1 and 4, it is seen that despite the high bulk density the capacity to sulfur is considerably lower than that of the agglomerates, probably as a result of the low surface area of the tablets. Example 15 Articles formed as extruded with a diameter of approximately 2mm and length 5-10mm were made from a mixture of 90 parts of malachite E powder, 7 parts of extrusion aid clay and 7 parts by weight of calcium aluminate cement, mixing with a little water to form a paste that was then extruded by a suitable die using a single screw extruder. The extrudate was dried at 110ß. The extruded products were then tested for their hydrogen sulphide absorption properties as in Example 1. The results were as follows exit noise removal sulfur hydrogen Ej nalaquit. BD t.B conr,. azu capacity fre (wt%) (nrV?) (kg?) Prom (% wt) sulfur: g?) 84.1 20.3 0.95 13.3 14.8 121 It is seen that the sulfur capacity is relatively low as a result of the relatively low mass density.

Claims (8)

1. CLAIMS 1. Absorbent units formed having an average size within the range of 1-10 mm, and containing at least 75% by weight of an absorbent material selected from copper carbonate, basic copper carbonate, copper hydroxide and mixtures thereof. same, having the units a mass density of at least 1.2kg / l, and a BET surface area of at least 10m2 / g. Absorbent units formed according to claim 1, containing at least 80% by weight of copper carbonate or basic copper carbonate. Absorbent units formed according to claim 1, 2, in the form of agglomerate containing 5-10% by weight of a cement bond. 4. Absorbent units formed according to one of claims 1, 3, wherein the absorbent material is basic copper carbonate. Absorbent units formed according to one of claims 1 to 4, characterized in that they contain from 9 to 20% by weight of gamma alumina, having a surface area of at least 150 m2 / g. 6. A process for the absorption of hydrogen sulfide from a gas or liquid comprising passing the gas or liquid through a bed of absorbent units formed according to one of claims 1 to 5, at a temperature below 150 °. 7. A process for the absorption of carbonyl sulphide or carbon disulfide from a gas or liquid, characterized in that it comprises passing gas or liquid through a bed of absorbent units formed according to claim 5, at a temperature below 150. °. 8. A method according to claims 6, 7, characterized in that the gas or liquid passes through the bed at a temperature in the range of 10"below zero to 80 °.