Patent Application
20240279076
The inventions relate to controlling the particle size, and pressurized gas flows in moving beds of particles containing metallic iron and/or nickel, wherein the metallic iron or nickel are reactants in carbonylation reactions with carbon monoxide (a component of the flowing gas).
BACKGROUND TO THE INVENTIONS The inventions are related to the production of iron carbonyl (principally Fe(CO)5 but also, possibly, small amounts of Fe(CO)4 and Fe2(CO)9), and nickel carbonyl (Ni(CO)4) using the following exothermic carbonylation reactions:
Fe(s)+5CO(g)ΔFe(CO)5(l or g) (1) ;
Ni(s)+4CO(g)ΔNi(CO)4(l or g) (2) .
These reactions are usually carried out when the input metallic iron and nickel is mixed with other solid material (impurities). And the principal practical point of performing this carbonylation is to separate the iron and nickel, as vapors (sometimes as liquids) of iron carbonyl and nickel carbonyl, from the other solids by removing the vapors and solids from the reactor through separate outlets. Carbonylation is usually followed by carbonyl decomposition (the reverse of carbonylation) to form high purity iron and nickel products. It is also possible to make other metal carbonyls with other transition metals like cobalt, manganese, molybdenum, and copper. However, the rates of formation of these other metal carbonyls are slow in comparison to those for iron and nickel carbonyl, and these other carbonyls do not vaporize so readily as iron and nickel carbonyl.
Carbonyl chemistry started when Langer first noticed the deposition of nickel in pipes in the laboratory of Ludwig Mond. They quickly connected it to carbon monoxide and nickel carbonyl (Mond, Langer, and Quincke, 1890; McNeil, 1990). In 1892 Ludwig Mond set up a pilot plant in Birmingham, England, to produce high purity nickel, using carbonylation and carbonyl decomposition, with ore from Sudbury, Ontario (McNeil, 1990). In 1900 the Mond Nickel Company was founded and started industrial nickel carbonyl processing in 1902 at Clydach, Wales (McNeil, 1990). A three-step process of nickel ore reduction, nickel carbonylation, and nickel carbonyl decomposition to (nearly) pure nickel and carbon monoxide was used by the Mond Nickel Company from its beginning (McNeil, 1990). In the 1920s, BASF/I. G. Farben developed a three-step process of iron ore reduction, iron carbonylation, and iron carbonyl decomposition to (nearly) pure iron and carbon monoxide (U.S. Pat. Nos. 1,828,376 and 1,759,659). BASF is still the largest iron carbonyl producer. It uses distillation and density separation techniques to separate iron, nickel, and other metal carbonyls.
Carbonylation suffers from several reaction slowing and self-stopping mechanisms. Stoffel (1914) discussed a self-stopping property of carbonylation reactions. I. G's Farben's 1924 U.S. Pat. No. 1,759,268 addressed this self-stopping. This patent claimed a method for running an iron carbonylation reactor such that the carbon monoxide flowed “with so great a speed that the iron carbonyl formed is substantially carried away with the gas current” Explaining this the patent said: “Such a speed of the gases is employed, according to this invention, as will prevent either completely or to a substantial amount the deposition of iron carbonyl on the iron which deposition of iron carbonyl would be obnoxious to the progress of the reaction.” In other words, unless product liquid iron carbonyl is pulled off the surface of the metal, the liquid deposited on the metal surfaces prevents access of carbon monoxide to these surfaces and stops the carbonylation reaction.
In the 1960s, ESSO research worked on a moving bed reactor for iron carbonylation (U.S. Pat. No. 3,342,588). (Here, a “moving bed” is a downward moving packed bed.) This ESSO reactor was not a success as it suffered from several carbonylation stopping and slowing mechanisms. However, there are at least two advantages of using a moving bed reactor over rotary reactors (which are the longtime standard reactors in industrial nickel carbonylation): (i) a moving bed reactor can be operated continuously, without the need for long and hazardous flushing operations; and, (ii) without the need for rotating seals, it is easier to raise the operating pressures (this raises reaction rates). With ESSO's moving bed new solid particles are dropped onto the top of the moving bed. The bed moves downward through two mechanisms. One mechanism is a combination of gravity and particle size shrinkage—the metallic bed particles are continuously size reduced by carbonylation removing surface metal atoms. The other mechanism is the outlet of material from the bottom of the reactor. Top loading and continuous particle size shrinkage create a particle size gradient within the moving bed, with the largest particles at the top and the smallest at the bottom. ESSO's moving bed reactor injected the CO-rich carbonylating gas into the bottom of their moving bed. This bottom injection had major drawbacks. The pressure gradients needed to push carbonylation gas through beds of particles with diameters of 200, 100, 50, 25, and 10 m are substantial. With the ESSO design, these high-pressure gradients were encountered right at the gas inlets and would have produced slow gas flows throughout most of the moving bed. The ESSO reactors tended to produce iron carbonyl in liquid form, which (A) slowed or stopped the carbonylation reaction in most of the reactor volume, and (B) flowed downward through the bed and probably blocked the bed's interstitial paths making upward gas flow even harder—hence creating another reaction slowing/stopping mechanism. In contrast, successful, long-lived rotary reactors such as the nickel carbonylators at Clydach, Wales and Sudbury, Ontario (both now operated by Vale) do not suffer from highly constricted/restricted gas flows. Those reactors inject carbonylation gas into a large volume above a continuously tumbling bed of carbonylation particles, so that all particle surfaces are repeatedly exposed to a reservoir of carbonylation gas.
Unstable carbonylation bed temperatures were a problem in I. G. Farden high-pressure iron carbonylation reactors (U.S. Pat. No. 1,614,625). Temperature runaway events can lead to complete halts of the carbonylation reaction. Such 1920s I. G. Farben reactors almost certainly used batch-loading and fixed-packed beds. Temperature control in large, fixed-packed beds of carbonylation particles in high-pressure vessels is difficult.
Turning to some chemical reactor art that uses gas cross-flow, in 1994, Jing Lee filed U.S. Pat. No. 5,520,891 for a reactor using horizontal gas flows (“cross-flow”) through a tall fixed packed bed of catalyst particles. Horizontal cross-flow would have substantially improved the ESSO-type carbonylation reactor. They are part of the inventions claimed here. Given this overlap, Jing Lee's reactor and reactor process and the reactions it is used for are described with enough depth to highlight differences between the inventions in U.S. Pat. No. 5,520,891 and the present inventions.
Jing Lee's reactor and process were good for exothermic reactions between gases that use a solid catalyst to form gaseous products. In particular, the reactor design and described processes were good for carrying out ammonia synthesis (the Haber-Bosch process), hydrogen or syngas production (the water-gas shift reaction), and methanol synthesis from syngas. These solid-catalyzed gas/gas reactions are large-scale industrial chemistry processes. There is a substantial patent literature presenting alternatives to Jing Lee inventions for carrying out these same reactions, including U.S. Pat. Nos. 3,918,918, 4,181,701, 4,321,234, 4,372,920, 4,423,022, 4,976,928, 5,869,011, and 7,846,417. Most of these alternatives (to Lee's) feature radial or axial/radial gas flows through beds of catalyst particles. Lee's reactor and process features (as inventive elements): (I) a reactant gas distributor, (II) a product gas collector with (III) an intervening fixed packed bed of catalyst material such that (IV) reactant gas passes through the distributor and into the fixed bed of catalyst, (V) this gas then flows across the bed of catalyst and reacts to form product gas, (VI) the product gas continues to flow through the bed of catalyst and out of the bed of catalyst through the product gas collector, further (VII) the catalyst bed is formed of layers generally axially oriented (and perpendicular to the gas cross-flows), (VIII) “banks” (i.e. many) U-bend tubes for heat extraction extend axially up and down between the distributor and collector and are embedded in the bed of catalyst, (IX) heat transfer fluid runs through the banks of U-bend tubes and collects (by heat exchange) the heat of the exothermic reactions going on in the reactor and transports this heat of reaction out of the reactor.
In preferred embodiments of Lee's process invention, the lengths of the paths of gas flows between the gas distributor and the gas collector are roughly equal. Still, the geometry shown in the figures allows for some path length variation. Further, in preferred embodiments illustrated in figures, the various up/down U-bend tubes are not evenly spaced in the horizontal cross-section. Instead, the density of tubes seen in the horizontal section is varied and varied for a purpose. This purpose was to vary the rates at which heat was extracted in different sub-volumes (of the reactor volume between the distributor and collector), and by varying the spatial heat extraction, to control the local temperature in the local sub-volumes, and, by this, to change and optimize the temperature in the catalytic bed relative to the progress of the gas through the bed. Since the optimal reaction temperature changes by locality (for the various gas/gas reactions) due to changes in gas composition as the mole fraction of product increases as the gas flows nearer and nearer to the gas collector.
In Lee's invention, the solids are not reactants but catalysts, the solid catalysts are in a fixed (non-moving) bed of particles (i.e., there is no continuous flow of catalyst particles into or out of the fixed packed bed of catalyst particles), the catalyst particles are not contained in a funnel, and there is no catalyst particle flow regulator. Further, in Lee's invention, the layers of catalyst (1) have many U-bend tubes embedded in them and (ii) are not distinguished by differing catalyst particle size or by differing catalyst material. Instead, Lee's layers of catalyst particles are distinguished by temperature layering enforced by the banks of U-bend tubes: “Successive adjacent tube banks 55 establish successive adjacent catalyst zones 57 having localized heat removal (or heat addition) to effectively partition the catalyst bed reaction zone 41 into multiple catalyst layers (or stages) in a single reactor vessel.”
Now a paragraph to outline the scope of the term “carbonylation gas.” Transition metal carbonylation involves the reaction of metals, such as iron and nickel, with carbon monoxide (see Reactions (1) and (2)). Therefore, the essential gas species in a carbonylation gas is carbon monoxide. However, carbonylation can be carried out with gas mixtures that contain gas species beyond carbon monoxide. It is known that small amounts of hydrogen sulfide, as well as vapors or suspended particles of sulfur, selenium, and tellurium, have positive catalytic effects on carbonylation (Hieber & Geisenberger, 1950); while U.S. Pat. No. 1,783,744 mentions the positive reaction rate effects of ammonia, methanol, and formaldehyde. Pichler and Walenda (1940) ran experiments with carbonylation gases with a hydrogen mole fraction as high as 0.6 and a carbon monoxide mole fraction of only 0.3, and they noted the positive reaction rate effects of hydrogen. In addition, carbonyl sulfide (COS) is added to the carbonylation gas used at Clydach in Wales to suppress the formation of thin layers of copper on the surfaces of reduced nickel particles because the copper films can significantly slow the formation of nickel carbonyl (Crundwell et al., 2011). Weak oxidizing gases such as steam and carbon dioxide can be included in a carbonylation gas mixture to selectively suppress the carbonylation of iron while having little effect on the carbonylation of nickel (see U.S. Pat. No. 2,254,158). Oxygen is such a strong oxidizer that it should be kept out of a carbonylation gas as well as possible (U.S. Pat. No. 1,783,744). Inert gas diluents such as nitrogen, argon, and other noble gases are quite acceptable in carbonylation gas and at large mole fractions, so long as the mole fraction of carbon monoxide does not become small. Trace amounts (mole fraction under 0.0003) of many other gas species are unlikely to negatively effect carbonylation. Carbonylation gas can also contain iron carbonyl vapor and nickel carbonyl vapor at well above trace levels, as well as some amount of harder-to-vaporize carbonyls of other transition metals such as cobalt, manganese, and molybdenum. A typical industrial carbonylation gas includes carbon monoxide at large mole fraction (above 0.8), with some fraction of gases with catalytic or positive reaction rate effects (see above), and also carbon dioxide. The carbonylation gas used in the industrial carbonylation reactors at Clydach has a carbon monoxide content of 80-95 vol %, while the reactors at Sudbury use carbonylation gas with 99.5 vol % carbon monoxide (Crundwell et al., 2011). Carbon dioxide tends to creep in with (a) the re-cycling of carbon monoxide liberated from carbonyl decomposition, since this carbonyl decomposition usually also generates some carbon dioxide via the Reverse Boudouard Reaction, and (b) (if used) cracking of natural gas to produce carbon monoxide. As the above discussion documents, a carbonylation gas can have a wide variety of compositions and component gas species.
Another term used in the description is “product-mixture gas.” Here a product-mixture gas is a mixture of carbonylation gas and also iron carbonyl vapor and/or nickel carbonyl vapor, wherein (A) product-mixture gas is formed inside a carbonylation reactor by some of the carbon monoxide in carbonylation gas reacting with some iron or nickel inside the reactor to form more metal carbonyl vapor (by Reactions (1) or (2)), and these carbonyl vapors then mix and flow with the remainder of the carbonylation gas forming product-mixture gas, and wherein (B) the product-mixture gas has not yet passed into a carbonyl vapor condenser, and wherein (C) condensation of some (or all) of the iron/nickel carbonyl vapors in the product-mixture gas operationally converts the uncondensed part of the gas from a product-mixture gas back into a carbonylation gas.
As used here, the phrase “carbonylation particles” can refer to any of (A) solid particles that contain metallic iron or nickel, possibly also other materials, and that are input to a carbonylation reactor, (B) the solid residue particles output from a carbonylation reactor, and (C) all the intermediates between these two. A “bed of carbonylation particles” is a collection or bed of solid particles (a granular media) that includes carbonylation particles and may also contain particles made of materials other than metallic iron and nickel.
Three times, the phrase “one or more funnels each aimed generally at the lower end of the vessel shell” appears in the claims, whereas in the specifications, many times, the phrase “downward-pointing, cross-flow funnel” appears. In the specifications and the claims, the phrase “aimed generally at the lower end of the vessel shell” should be synonymous with “downward-pointing.” Further, since in the claims a funnel is always a cross-flow funnel, the entire phrase “funnel aimed generally at the lower end of the vessel shell” should be taken to be synonymous with the phrase “downward-pointing, cross-flow funnel.” The words and phrases “funnel,” “cross-flow,” and “downward-pointing, cross-flow funnel” are all defined later in these specifications.
The inventions cover a method and an apparatus for performing carbonylation, i.e., a gas/solid surface reaction, wherein one or more downward moving beds of carbonylation particles are each held in a separate downward-pointing, cross-flow funnel, and carbonylation gas flows across each moving bed in a substantially horizontal cross-flow causing carbonylation to occur and causing iron and/or nickel carbonyl vapors to form. This turns the carbonylation gas into product-mixture gas that then is output from the carbonylation reactor.
In all embodiments, the rate of outflow of solid (carbonylation particle) residue from each downward-pointing, cross-flow funnel is regulated by a flow regulator. This regulator runs while the reactor is operating and at a regular cadence, with the regulator repeatedly alternating between (a) forcing the particles to be stopped in the funnel (most of the time) and (b) allowing the particles to flow out the funnel's bottom end surface (in brief intervals).
In preferred embodiments, the rate of outflow of solid (carbonylation particle) residue from each downward-pointing, cross-flow funnel is regulated so that this outflow rate obeys the “funnel ratios condition.” The funnel ratios condition is described fully in the detailed description section. However, briefly, when this condition is obeyed it constrains the size of the particles outlet from the funnels. These output particle size constraints are enumerated by (a) the size of the particles input to the downward-pointing, cross-flow funnels and (b) size ratio features of these funnels.
Conventional rotary kiln reactors require rotary seals for axial gas pipes to the toxic interior of the reactor—such rotary seals are difficult to make safe for reactors operating at high pressures. In contrast, the above process and apparatus is implemented in a moving bed reactor architecture which does not require the input and output gas pipes to go on a rotation axis with rotating high pressure seals—this adds simplicity, reliability and lower cost as well as safety, and enables the use of higher operating pressures (than, for example, the 70 bar used at the Sudbury reactors (Crundwell et al., 2011)). Operating pressures over 200 bar can be achieved inside the invention's reactor with its static pressurized shell. Raising the operating pressure provides a substantial extra multiple to the reactor's yield, since reaction rates scale strongly up with operating pressure. According to Pichler and Walenda (1940) carbonylation reaction rates increase with the partial pressure of carbon monoxide raised to the 1.76 power and Ludwig Mond's son found a similar pressure dependency (Mond and Wallis, 1922). So, according to the Pichler and Walenda (1940) relationship, the reaction rates for the current inventive reactor architecture operating at 200 bar will be 6.3 (=(200/70)1.76) times faster than the reaction rates inside the Sudbury reactors.
Enforcing the funnel ratios condition drives the moving beds (in the funnels) to reach desirable physical conditions. In particular, it drives the vertical variation in bed particle sizes to a dynamic steady state size distribution (measured against vertical height); secondly, this steady state size distribution is one in which, at all heights in the bed, there is a (nearly) constant number of particles from the gas inlet side of the bed to the gas outlet side of the bed; thirdly, this constancy allows the moving bed to have near-uniform rates of horizontal gas flows and nearly uniform temperatures across the bed; fourthly, given the third benefit, it is straightforward to (a) bring the whole bed to an optimal operating temperature at the given operating pressure.
While the physical conditions just described are optimized when the regulation of particle outflows from the bottoms of funnels is done according to the funnel ratios condition, just running the regulator continuously at a regular cadence will, with the funnel form, drive the carbonylation particles to adopt a size distribution (measured in the vertical direction) that provide beneficial gas flow conditions that improve the rates of the carbonylation reaction in the moving beds inside the funnels.
Speeding carbonylation can provide significant environmental and economic advantages. This is because fast carbonylation may replace highly destructive nickel ore processing methods, including heap and high-pressure acid leaching, sulfiding, and energy-intensive molten matte and slagging processes. Further, fast carbonylation applied to laterite ores should produce solid residues that include alumina and many other valuable minerals, wherein this residue can be post-processed to produce many separated, high-value concentrates. In contrast, leaching and matte/slagging methods lose the alumina to hard-to-extract aluminum compounds generally left in tailings and slag. And although the leaching and matte/slagging methods are usually good at making valuable concentrates of platinum group metals, post-processing of the solid residues of carbonylation (of laterite ores) can also recover high value concentrates of platinum group metals, rare-earth metals (Terekhov and Emmanuel, 2013) and other valuable minerals and metals. Present nickel laterite ore processing also treats the dominant iron oxy-hydroxides and hydrated iron oxide minerals in laterite ores as nuisance material. Although they spend energy on converting these iron-containing minerals into dehydrated iron oxides and then spend even more energy reducing some iron oxide to metallic iron; after this energy expenditure, the iron ends up largely unused. In contrast, carbonylation produces iron carbonyl and nickel carbonyl that are readily separable by distillation; pure iron carbonyl can then be decomposed into one of the most valuable iron products, that is, high purity iron powder, suitable for many iron/steel powder metallurgy and printing processes. There are very large global deposits of laterite ores containing hydrated iron oxides at large mass percentage (well above 50%). Laterite drying followed by hydrogen reduction, fast carbonylation, and carbonyl decomposition may produce a large amount of iron suitable for high-value iron and steel products as well as residues from carbonylation that concentrate valuable minerals. Such residues can contain platinum group metals, rare earth elements, chromium and more than five other valuable metals.
Conventional rotary kiln reactors use batch unloading and loading of the solids, extensive gas flushing operations (to remove toxic carbon monoxide and very toxic metal carbonyl vapors), and carbon monoxide re-charging. This batch operation loses a considerable amount of operating time. According to Crundwell et al. (2011), only two out of the three rotating reactors at Sudbury operate at any one time. Further, the gas flushing operations either lose large amounts of carbon monoxide or require energy-intensive gas separation methods to reclaim the carbon monoxide. The present inventions provide a reactor that can operate continuously. This continuous operation avoids the environmental safety and energy problems associated with toxic gas flushing; it is also simpler and easier to carry out.
The previous summary and the following detailed descriptions are to be read in view of the drawings which illustrate particular exemplary embodiments and features as briefly described below.
The summary and detailed descriptions, however, are not limited to only those embodiments and features explicitly illustrated.
Geometric properties of physical funnel structures are elements of the inventions presented. This detailed description starts by reviewing/defining some geometry of funnels.
Referring to the quarter-view, orthographic projection
For the present invention, a physical funnel is a structure with walls substantially shaped to follow a funnel shape such that the structure is open at the base and end surfaces (i.e., the walls do not block the base and end surfaces). Hereafter, the single word “funnel” refers to a physical funnel.
Further, the area of the end surface, denoted Ae, is less than the area of the base surface, denoted Ab (in mathematical notation: Ae<Ab). The value of √{square root over ((Ae/Ab))} for the funnels used in the present inventions is important for process control.
The inventions use downward-pointing, cross-flow funnels wherein each of these funnels is placed inside a pressure vessel. A single pressure vessel can enclose more than one downward-pointing, cross-flow funnel.
Vertical section
The example embodiment in
There are many possible ways to arrange downward-pointing, cross-flow funnels, distributor volumes, collector volumes, hopper structures, plenums/pipes, etc., inside a single pressure vessel shell. The inventive claims apply to all these possible arrangements.
In the example of
Downward-pointing, cross-flow funnels are used in both a method invention and an apparatus invention for carrying out carbonylation of moving beds of carbonylation particles using cross-flows of carbonylation gas and regulation of particle size at the bottom of the moving beds.
The narrow, black, horizontal arrows shown in
In the embodiment of
Another inventive element involves the regulation of the size of the carbonylation particles inside the moving beds of carbonylation particles (209a & 209b). This regulation sets the carbonylation particle diameters along the vertical height in the cross-flow funnels (200a & 200b). While operating a carbonylation reactor of the present inventions, the regulation of carbonylation particle diameters across the vertical height of the moving beds of carbonylation particles (209a & 209b) in the cross-flow funnels (200a & 200b) is achieved by running the regulators (217a & 217b) associated with each moving bed (217a associated with 209a; 217a associated with 209b) at a regular cadence, with the regulators (217a & 217b) repeatedly alternating between condition (A) where the regulators (217a & 217b) force the moving beds of particles (209a & 209b) to be stopped in their respective funnels (200a & 200b) (most of the time) and condition (B) where the regulators (217a & 217b) allow the particles at the bottoms of the moving beds (209a &209b) to flow out of each funnel's bottom end surfaces (202a & 202b) (in brief intervals), where conditions (A) and (B) repeat over and over while the reactor is operating. The time-averaged flow rate of particles out of the bottom end surfaces (202a & 202b) of the cross-flow funnels (200a &200b) can be regulated (varied) by the regulators (217a & 217b) by varying the details of the stopped (A) and flow (B) conditions in the repeated regulation cadence. For example, the number of times the regulators (217a & 217b) turn to the flow condition (B) in one hour can be greatly varied; for example, from once in one hour to 600 times in one hour. Another way for the flow regulators (217a & 217b) to change the time-averaged flow rate of particles out of the bottom of the cross-flow funnels (200a & 200b) is to increase or decrease the number of particles flowing down through the bottom end surfaces (202a & 202b) in each instance of the flow condition (B).
Such variation (regulation) of the time-averaged flow rate out of the bottom of the cross-flow funnels (200a & 200b) will change the distribution of particle sizes in the moving beds (209a & 209b); since (a) the moving beds are moving beds of carbonylation particles, (b) such variations will change how long each individual carbonylation particle is contained in the cross-flow funnels (200a & 200b), hence, (c) such variations will change how long each individual particle is exposed to carbonylating conditions, hence (d) such variations will change how much each individual carbonylation particle (that is dominated in composition by iron and/or nickel) shrinks in size due to carbonylation of iron and nickel on each particle's exposed surfaces. So, in particular, such variation (regulation) of the time-averaged flow rate out of the bottom of the cross-flow funnels (200a & 200b) will change the average size of the carbonylation particles, leaving these funnels through their bottom end surfaces (202a & 202b). Operating the regulators (217a & 217b) with slower time-averaged flow rates and long particle residence times in the cross-flow funnels (200a & 200b) will cause the average particle sizes of particles leaving these funnels (through their bottom end surfaces, 202a & 202b) to be smaller than when the regulators (217a & 217b) operate to produce a faster time-averaged flow rate. Note, also, that such variation of the operation of the regulators (217a & 217b) also changes the average carbonylation particle size at any height in the moving beds (209a & 209b) since there is always a gradient in average carbonylation particle sizes with their height position inside the funnels (200a & 200b) and moving beds (209a & 209b). These height-varied particle size gradients occur because the lengths of time each carbonylation particle is exposed to shrinking carbonylating conditions varies with height in the moving beds (209a & 209b), these beds are moving and downwards. In particular, the carbonylating particles at the tops of the moving beds are exposed to shrinking carbonylating conditions for the shortest times (among the particles in the beds). Hence, these top particles have the largest average size. Further, as the individual particles move downwards in the moving beds (209a & 209b), they are exposed to shrinking carbonylating conditions for longer and longer and, hence, become smaller and smaller.
The details of the particle size gradients of carbonylation particles inside the cross-flow funnels (200a & 200b) and moving beds (209a & 209b) change the details of the cross-flows of carbonylation gases through these moving beds. Changes to the particle size gradients change both the rates of cross-flows of carbonylation gases and the temperatures of the carbonylation particles at various sub-volumes of the moving beds (209a & 209b). The carbonylation bed temperatures should be stable, controllable, and uniform (U.S. Pat. No. 1,614,625). It is an inventive claim that stable, uniform temperatures for moving beds of carbonylation particles will be obtained in downward moving beds (209a & 209b) in cross-flow funnels (200a & 200b) if the regulators (217a & 217b) regulate the particle sizes to obey the funnel ratios condition. This claim is based on a particle bed feature promoted by enforcing the funnel ratios condition. This feature is that the number of particles found in any horizontal slice of the moving bed is constant, whatever the height in the bed that the slice is made. Further, if the regulators enforce the funnel ratios condition and appropriate gas pressure differences are applied between the funnel's (200a & 200b) carbonylation gas distributor volumes (204a & 204b) and their product-mixture gas collection volumes (205a & 205b) then the temperature in the moving beds (209a & 209b) can be stable, uniform, and optimal for fast carbonylation rates.
The details of the funnel ratios condition are determined by the details of the geometry of each cross-flow funnel (to which the condition is applied) and the sizes of the particles input to the top of each funnel (through its open top base surface, 201a or 201b). Before specifying the funnel ratios condition, a definition is given for a ratio for lengths of features of the bottom end (202a or 202b) surfaces and top base surfaces (201a or 201b) of each particular downward-pointing, cross-flow funnel. This ratio is called the “bottom-to-top length ratio” of a cross-flow funnel. It is defined as the square root of the fraction formed by dividing a funnel's bottom end surface area (Ae, see example in
An example of the application of the funnel ratio condition is now given. In this example, the cross-flow funnel has a base surface area (Ab) of 128 cm2 and an end surface area (Ae) of 8 cm2, and the average input particle diameter is 1 mm. This example case has a bottom-to-top length ratio of 0.25 (=√{square root over ((8 cm2/128 cm2))}) and a target diameter for output particles of 0.25 mm (=0.25×1 mm).
Table 1 provides more examples of average output particle diameters obeying the funnel ratios condition for downward-pointing, cross-flow funnels with bottom-to-top length ratios of 0.75 and 0.25 and three different, typical average diameters for input carbonylation particles to these funnels.
A practical extreme high value for the bottom-to-top length ratio is 0.99, and a practical extreme low value for the bottom-to-top length ratio is 0.01.
Some mechanical device is needed to regulate the flow of carbonylation particles out of the end surface of a downward-pointing, cross-flow funnel. A rotary valve is a suitable mechanical device for such particle flow regulation, although systems of moveable flaps may be used instead. In
This patent application claims priority to U.S. Provisional Patent Application No. 63/444,483, titled METHOD AND APPARATUS FOR FAST IRON AND NICKEL CARBONYLATION, filed on Feb. 9, 2023, the entire contents of which are incorporated by reference herein.
| Number | Date | Country | |
|---|---|---|---|
| 63444483 | Feb 2023 | US |
