Process for converting oxygenates to olefins using molecular sieve catalysts comprising desirable carbonaceous deposits

ES 2 247 702 T3

DESCRIPTION

Process for converting oxygenates to olefins using molecular sieve catalysts comprising desirable carbonaceous deposits.

Field of the invention

The present invention relates to methods for selectively converting oxygenates to light olefins, preferably ethylene and propylene, in which desired carbonaceous deposits are maintained in a total reaction volume of the catalyst by fully regenerating only a portion of the total reaction volume of the catalyst. catalyst, and mixing the regenerated part with the total unregenerated reaction volume of the catalyst.

Background of the invention

Light olefins (defined as "ethylene, propylene, and butylene") serve as feeds for the production of numerous chemicals. Traditionally, light olefins are produced by the catalytic pyrolysis of petroleum. Due to the limited supply and / or the high cost of petroleum sources, the cost of olefin production from petroleum sources has continually increased.

Alternative raw materials for the production of light olefins are oxygenates, such as alcohols, particularly methanol, dimethyl ether, and ethanol. Alcohols can be produced by fermentation, or from synthesis gas derived from natural gas, petroleum liquids, carbonaceous materials, including coal, recycled plastics, municipal waste, or any organic material. Due to the wide variety of sources, alcohols, alcohol derivatives, and other oxygenates are presented as an inexpensive non-petroleum source for olefin production.

The catalysts used to promote the conversion of oxygenates to olefins are molecular sieve catalysts. Since ethylene and propylene are the most sought after products of such a reaction, research has focused on which catalysts are the most selective to ethylene and / or propylene, and on methods to increase the selectivity of molecular sieve catalysts to ethylene. and / or propylene. The selectivity of certain molecular sieve catalysts to ethylene and propylene is known to increase if the level of coke in the total reaction volume of the molecular sieve catalyst is kept in the range of about 2% by weight to about 30% by weight. . Some have suggested maintaining this desired level of coke by removing all or part of the total reaction volume of the catalyst, partially regenerating the catalyst thus removed, and returning the partially regenerated catalyst to the reactor. However, partial regeneration may not result in maximum selectivity of the catalyst to light olefins.

Methods are needed that maintain a desired level of coking in molecular sieve catalysts during the conversion of oxygenates to olefins, while maintaining maximum catalyst activity.

Compendium of the invention

The present invention provides a method of treating a molecular sieve catalyst, comprising: contacting a feed comprising oxygenates with a total reaction volume of a molecular sieve catalyst under conditions effective to produce a stream comprising C olefins.₂-C₃, wherein said total reaction volume comprises desirable carbonaceous deposits that make said catalyst more selective to C olefins.₂-C₃ that in the absence of such desirable carbonaceous deposits; and wherein, after the accumulation of undesirable carbonaceous deposits, effective to interfere with the activity of the catalyst, said desirable carbonaceous deposits are maintained in said molecular sieve catalyst by a process comprising: separating said total reaction volume of sieve catalyst molecular in a part and a remnant; treating said part with a regeneration medium under conditions effective to remove said undesirable carbonaceous deposits, forming a regenerated part comprising the range of about 0% by weight to a regenerated amount of carbonaceous deposits; and, mixing said regenerated portion with said carryover, wherein said regenerated quantity of carbonaceous deposits comprises a sufficient quantity, after said mixing, to produce a total regenerated reaction volume comprising said desirable carbonaceous deposits.

Brief description of the drawings

FIG. 1 is a diagram of a preferred embodiment of a high speed fluid bed reactor with catalyst recirculation for use in the present invention.

Detailed description of the invention

The conversion of oxygenates to light olefins is catalyzed by various molecular sieve catalysts. Due to the high temperatures required during the conversion process, carbonaceous deposits known as "coke" inevitably form on the surface of the molecular sieve catalyst. In order to avoid

ES 2 247 702 T3 a significant reduction in catalyst activity, the catalyst must be regenerated by burning coke deposits.

One objective during the conversion of oxygenates to olefins is to maximize the production of light olefins, preferably ethylene and propylene, and to minimize the production of methane, ethane, propane, and C materials.₅+. The present invention uses the coke that inevitably deposits on the catalyst to accomplish this goal by allowing "desirable carbonaceous deposits" to build up on the molecular sieve catalyst while removing undesirable carbonaceous deposits.

One method that has been suggested to maintain desirable carbonaceous deposits on the catalyst is to only partially regenerate some or all of the total reaction volume of the molecular sieve catalyst. Without limiting the present invention to any particular theory, it is believed that only partially regenerating a portion or only partially regenerating an entire reaction volume of coked molecular sieve catalyst has a serious drawback. The coke that is produced during the conversion of oxygenates to olefins is known to deposit both on the surface and in the "micropores" of molecular sieve catalysts. Reactions that selectively convert oxygenates to ethylene and propylene occur at the micropores of the molecular sieve catalyst. It is relatively difficult, for a regeneration medium (usually oxygen), to access the micropores. Because of this, the coke that forms in the micropores is more difficult to remove during the regeneration process. Most likely, partial regeneration will not remove coke from the micropores of the catalyst, resulting in an adverse impact on the selectivity of the catalyst to ethylene and propylene.

The present invention maintains "desirable carbonaceous deposits" on the catalyst by removing only a portion of the total reaction volume of the coked molecular sieve catalyst and fully regenerating only that portion of the catalyst. Full regeneration is believed to remove coke from both the micropores and the less selective surface areas of the regenerated part of the catalyst. When the regenerated part of the catalyst is mixed with the unregenerate remainder of the catalyst, the result is the maintenance of desirable carbonaceous deposits that block the less selective surface areas on the unregenerate part of the catalyst, and an increase in the sites available to convert selectively the oxygenates in light olefins (surface area of ​​micropores) in the regenerated part of the catalyst.

As used herein, the term "desirable carbonaceous deposits" is defined to comprise an amount of at least about 2% by weight of carbonaceous deposits, preferably in the range of from about 2% by weight to about 30% by weight. carbonaceous deposits, based on the weight of the total reaction volume of the coked catalyst. The "desirable carbonaceous deposits" - even if they comprise more than 30% by weight of the total reaction volume of the molecular sieve catalyst - are carbonaceous deposits that mainly block parts of the catalyst surface that are not selective for the production of C2- olefins. C3.

Substantially, any small or medium pore molecular sieve catalyst, and equivalents thereof, can be used in the present invention. "Small pore" molecular sieve catalysts are defined as catalysts with pores having a diameter less than about 5.0 Angstroms. "Medium pore" molecular sieve catalysts are defined as catalysts with pores having a diameter in the range of 5 to 10 Angstroms. "Their equivalents" are defined to refer to catalysts having a pore size that performs substantially the same function in substantially the same way to achieve substantially the same result as catalysts having the foregoing pore diameter or size.

One group of suitable molecular sieve catalysts is the group of zeolites. There are several types of zeolites, each of which exhibits different properties and different utilities. Structural types of small pore zeolites that are suitable for use in the present invention with various levels of efficacy include, but are not necessarily limited to, AEI, AFT, APC, ATN, ATT, ATV, AWW, BIK, CAS, CHA, CHI, DAC, DDR, EDI, ERI, GOO, KFI, LEV, LOV, LTA, MON, PAU, PHI, RHO, ROG, and THO and substituted examples of these structural types, described in WM Meier and DH Olsen, Atlas of Zeolite Structural Types (Butterworth Heineman-3^to ed. 1997). Preferred zeolite catalysts include, but are not necessarily limited to, ZSM-5, ZSM-34, erionite, and chabazite.

Silicoaluminophosphates ("SAFOs") are another group of molecular sieve catalysts that are useful in the invention. SAFOs have a three-dimensional microporous crystal structure of PO tetrahedral units₂+, AlO₂^-, and SiO₂. SAFOs suitable for use in the invention include, but are not necessarily limited to, SAfO-34, SAFO-17, and SAFO-18. A preferred SAFO is SAFO-34, which can be synthesized according to US-A-4,440,871 and Zeolites, Vol. 17, pp. 512-522 (1996).

SAFOs with added substituents may also be useful in the present invention. These substituted SAFOs form a class of molecular sieves known as "MeAFSOs." Substituents can include, but are not necessarily limited to, nickel, cobalt, strontium, barium, and calcium.

Structural types of medium pore molecular sieves useful in the present invention include, but are not necessarily limited to, MFI, MEL, MTW, EUO, MTT, HEU, FER, AFO, AEL, TON, and substituted examples of these structural types , described in the Atlas of Zeolite Types.

ES 2 247 702 T3

The process for converting oxygenates to olefins employs an organic starting material (raw material) which preferably comprises "oxygenates". As used herein, the term "oxygenates" is defined to include, but is not necessarily limited to, aliphatic alcohols, ethers, and carbonyl compounds (aldehydes, ketones, carboxylic acids, carbonates, and the like). The aliphatic moiety should preferably contain the range of about 1-10 carbon atoms, and more preferably the range of about 1-4 carbon atoms. Representative oxygenates include, but are not necessarily limited to, straight chain or branched chain lower aliphatic alcohols, and their unsaturated counterparts. Examples of suitable compounds include, but are not necessarily limited to: methanol, ethanol, n-propanol, isopropanol, C alcohols₄-C₁₀, methyl ethyl ether, dimethyl ether, diethyl ether, diisopropyl ether, formaldehyde, dimethyl carbonate, dimethyl ketone, acetic acid, and mixtures thereof. As used herein, the term "oxygenate" designates only the organic material used as feed. The total feed charge to the reaction zone may contain additional compounds, such as diluents.

Preferably, the oxygenate feedstock should be contacted in the vapor phase, in a reaction zone, with the defined molecular sieve catalyst, under effective process conditions, to produce the desired olefins, i.e., a temperature, effective pressure, VEPH (Weight Space Velocity per Hour), and optionally an effective amount of diluent. Alternatively, the process can be carried out in a liquid phase or in a mixed vapor / liquid phase. When the process is carried out in the liquid phase or in a mixed vapor / liquid phase, different conversion rates and food-product selectivities can result, depending on the catalyst and the reaction conditions.

The temperature employed in the conversion process can vary over a wide range, depending, at least in part, on the selected catalyst. Although not limited to a particular temperature, the best results will be obtained if the process is performed at temperatures in the range of about 200 ° C to about 700 ° C, preferably in the range of about 250 ° C to about 600 ° C, and most preferably in the range of from about 300 ° C to about 500 ° C. Lower temperatures generally result in lower reaction rates, and the formation of the desired light olefin products can become remarkably slow. However, at higher temperatures the process may not form the optimal amount of light olefin products, and the coking rate may become too high.

Light olefins will be formed - although not necessarily in optimal amounts - over a wide range of pressures, including autogenous pressures and pressures in the range of about 0.1 kPa to about 100 MPa. A preferred pressure is in the range of about 6.9 kPa to about 34 MPa, most preferably in the range of about 48 kPa to about 0.34 MPa. The preceding pressures do not include that of the diluent, if any is present, and refer to the partial pressure of the raw material in relation to the oxygenates and / or their mixtures. Pressures outside of the ranges set forth may be used and are not excluded from the scope of the invention. Lower and higher extremes of pressure can adversely affect selectivity, conversion, rate of coking, and / or rate of reaction; however, light olefins, such as ethylene, can still be formed.

The process must continue for a period of time sufficient to produce the desired olefin products. The reaction time can vary from tens of seconds to several hours. Reaction time is largely determined by reaction temperature, pressure, selected catalyst, hourly weight space velocity, phase (liquid or vapor), and selected process design characteristics.

A wide range of weight space velocities per hour (VEPH) for the feedstock will work in the present invention. VEPH is defined as the weight of feed (excluding diluent) per hour per weight of a total reaction volume of molecular sieve catalyst (excluding inerts and / or fillers). VEPH should generally be in the range of approximately 0.01 h¹ at about 500 h^-1, preferably in the range of about 0.5 h^-1 at about 300 h^-1, and most preferably in the range of about 0.1 h^-1 at about 200 h^-1.

One or more diluents with the oxygenates can be fed to the reaction zone, such that the total feed mixture comprises diluent in a range from about 1% by mole to about 99% by mole. Diluents that can be employed in the process include, but are not necessarily limited to, helium, argon, nitrogen, carbon monoxide, carbon dioxide, hydrogen, water, paraffins, other hydrocarbons (such as methane), aromatics, and mixtures of them. The preferred diluents are water and nitrogen.

A preferred embodiment of a reactor system for the present invention is a circulating fluid bed reactor with continuous regeneration, similar to a modern fluid catalytic pyrolyzer. Fixed beds are not viable for the process because the conversion of oxygenate to olefin is a highly exothermic process that requires several stages with heat exchangers or other cooling devices. The reaction also results in a high pressure drop due to the production of low pressure, low density gas.

As the catalyst must be regenerated frequently, the reactor must allow easy removal of part of the

ES 2 247 702 T3 catalyst to a regenerator, where the catalyst is subjected to a regeneration medium, preferably a gas comprising oxygen, most preferably air, to burn the coke off the catalyst, which restores the activity of the catalyst. The conditions of temperature, oxygen partial pressure, and residence time in the regenerator should be selected to achieve a coke content in the regenerated catalyst of less than about 0.5% by weight. At least a part of the regenerated catalyst must be returned to the reactor.

It is important that the reactor be designed such that a relatively high average coke level is maintained in the reactor - an amount greater than about 1.5% by weight, preferably in the range of from about 2% by weight to about 30%. % by weight, most preferably in the range of about 2% by weight to about 20% by weight. If the reactor is a high speed fluidized bed reactor (sometimes called a riser reactor), then a portion of the catalyst that exits the top of the reactor must be returned to the inlet of the reactor. This is different from a typical Fluid Catalytic Pyrolyzer (PCF) riser reactor, where all or most of the catalyst exiting the top of the reactor is sent to the regenerator. The return of coked catalyst directly to the reactor, without regenerating the coked catalyst, allows the average level of coke in the reactor to increase to a preferred level. By adjusting the flow ratio of the coked catalyst between the regenerator and the reactor, a preferred level of coking, or "desirable carbonaceous deposits", can be maintained.

If the fluidized bed reactor is designed with low gas velocities, below about 2 m / s, then cyclones can be used to return the catalyst fines to the fluidized bed reaction zone. Such reactors generally have high rates of recirculation of solids within the fluidized bed, which allows the level of coke in the catalyst to increase to a preferred level. Desirable carbonaceous deposits are maintained by removing the catalyst from the bed and regenerating the catalyst in the manner described above, and then returning at least a portion of this regenerated catalyst to the reactor.

A preferred embodiment of a riser reactor configuration for use in the present invention is depicted in Figure 1. A methanol feed 12 is at least partially vaporized in a preheater (not shown). The methanol feed is mixed with regenerated catalyst 28 and coked catalyst 22 at the bottom of riser reactor 14. An inert gas and / or steam can be used to dilute the methanol, elevate catalyst streams 22 and 28, and keep pressure gauge lines clear of catalyst. This inert gas and / or steam is mixed with the methanol and catalyst in reactor 14. The reaction is exothermic, and the preferred reaction temperature, in the range of about 300 ° C to about 500 ° C, is maintained by removing the hot. The heat can be removed by any suitable means, including, but not necessarily limited to, cooling the reactor with a catalyst cooler (not shown), feeding some of the methanol in liquid form, cooling the catalyst feed to the reactor, or any combination of these methods.

The reactor effluent 16, containing the products, coked catalyst, diluents, and unconverted feed, must flow into a separation zone 18. In the separation zone 18, the coked catalyst is separated from the gaseous materials by means of gravity and / or cyclone separators. A part of the coked catalyst 22 is returned to the inlet of the reactor. The part of catalyst 22 coked to be regenerated is first sent to a purification zone 29, where steam or inert gas is used to recover the hydrocarbons adsorbed from the catalyst. The purified coked catalyst 23 should flow into regenerator 24. The portion of the catalyst sent to regenerator 24 must be contacted with a regeneration medium, preferably a gas 30 comprising oxygen, at temperatures, pressures, and residence times that are capable of burning the catalyst coke and to a lesser level. than about 0.5% by weight. The preferred temperature in the regenerator is in the range of about 550 ° C to about 700 ° C, the preferred concentration of oxygen in the gas leaving the regenerator is in the range of about 0.1% by volume to about 5% by volume. volume, and the preferred residence time is in the range of about 1 to about 100 minutes.

Coke burn is exothermic. The temperature can be maintained at a suitable level by any acceptable method, including, but not limited to, feeding refrigerant gas, cooling the catalyst in the regenerator with a catalyst re-generator 26, or a combination of these methods.

Regenerated catalyst 28 is sent to reactor 14, where it is mixed with recirculated coked catalyst 22 and methanol feed 12. Regenerated catalyst 28 can be raised to reactor 14 via inert gas, steam, or methanol vapor (not shown). The process must be repeated in a continuous or semi-continuous manner. The hot product gases 20 from the reactor must be cooled, the by-product water condensed and collected, and the desired olefin product gases recovered for further processing.

In order to determine the level of coke in the reactor and in the regenerator, small samples of catalyst can be periodically withdrawn from various points of the recirculation system, for the measurement of the carbon content. The reaction parameters can be adjusted accordingly.

The following examples illustrate, but do not limit, the present invention.

ES 2 247 702 T3

Example 1

A continuous circulation fluid bed reactor was charged with 3200 g of catalyst, which was spray dried from a SAFO-34 powder mixture (obtained from UOP, Des Plaines, Illinois) with alumina and clay binders, which had a mean particle size 90-100 microns. In three different tests, pure methanol was charged at a rate of 900 grams / hour and vaporized in a preheater. The vaporized feed was mixed with 20,000 to 25,000 grams / hour of catalyst, and fed into a reactor with an inner diameter of 1.02 cm and a length of 6.71 meters. Approximately 268.21 liters / hr of nitrogen was used to lift the catalyst and keep the pressure gauges clear of catalytic fines. Nitrogen was mixed with methanol and catalyst in the reactor. The temperature in the reactor was kept at 450 ° C by means of electric heaters. The effluent from the reactor flowed into a scrubber, where the catalyst was removed from the product gas. The catalyst was contacted with nitrogen at the bottom of the scrubber to recover volatile hydrocarbons from the catalyst. The purified catalyst was sent to a regenerator, where the catalyst was contacted with a mixture of nitrogen and air. The temperature in the regenerator was maintained at 620 ° C with electric heaters, and the air speed could be varied to adjust the level of coke in the regenerated catalyst. The catalyst was returned from the regenerator to the reactor, where it was mixed with the methanol feed. The process was repeated continuously. The hot product gases from the reactor were cooled, and the by-product water condensed and collected. The hydrocarbon gases were separated from the water and analyzed by gas chromatography. The regenerator exhaust gas was analyzed for oxygen, carbon monoxide, and carbon dioxide, and the velocity was measured on a dry test meter. Small samples of catalyst were taken periodically, both from the purifier and the regenerator, for the measurement of the carbon content. Based on these measurements, the yield on the products, including coke, was calculated.

In Test 1, the air velocity was set so that almost all of the carbon in the catalyst was removed during each pass through the regenerator. The carbon content in the catalyst leaving the reactor was 0.5% by weight, and the regenerator removed all but 0.2% by weight of this carbon. The ethylene selectivity was 10.8% by weight, and the selectivity to heavy compounds and coke was 34.9% by weight and 14.3% by weight, respectively.

In test 2, the air velocity to the regenerator was reduced so that all the catalyst was only partially regenerated at each pass. The carbon in the recirculating catalyst increased, eventually reaching a stable state, such that the carbon was removed at the same rate as it was deposited. At this point, the carbon content in the catalyst going to the regenerator was 5.5%, and the carbon content of the catalyst leaving the regenerator was 4.9%. Ethylene selectivity improved to 26.7%, and undesirable heavy compound selectivity decreased to 17.9%. The coke yield remained relatively unchanged at 13.6%. The methanol conversion was 91.3%, showing a decrease in catalyst activity due to coke in the catalyst.

In Run 3, the methanol feed was stopped and the recirculating catalyst was allowed to fully regenerate (to 0.15 wt% carbon). The air to the regenerator was then stopped, the methanol feed was reintroduced, and the coke was allowed to form on the catalyst without regeneration for about 5 hours. The carbon content in the circulating catalyst after 5 hours was 5.8%. At this point, the ethylene selectivity had improved to 35.0%, the heavy compound selectivity was further reduced to 13.4%, and the coke selectivity was reduced to 4.1%. Coke yield was calculated from measurements of carbon build-up on the catalyst during the previous hour in oil. The conversion at this point was 89.9%, showing that the catalyst had approximately the same activity as the catalyst in run 2.

After test 3, the reactor was returned to the same operating conditions as in test 1, and then test 2. Methanol conversions and yields to product were essentially the same as the original yields in tests 1 and 2 after a total of 150 hours in oil, showing that the results were not simply the effect of catalyst aging.

Column 4 represents a calculated product selectivity for a commercial reactor using the present invention, based on data from Tests 1 and 3. Column 4 assumes that 10% of the methanol is converted over freshly regenerated catalyst (selectivities according to assay 1), and the remaining 90% of the methanol is converted over coked catalyst (selectivities according to test 3). The calculated selectivities are slightly worse than the results of test 3, but are nevertheless significantly better than those that would be obtained if the catalyst were only partially regenerated, as in test 2:

ES 2 247 702 T3

<td>Test No.</td><td> 1</td><td> 2</td><td> 3</td><td>4 (calculated)</td>

<td>Regeneration mode</td><td>Complete regeneration of all catalyst</td><td>Partial regeneration of the entire catalyst</td><td>None</td><td>Complete regeneration of a part of the catalyst</td>

<td>Carbon in the catalyst coming out of the reactor</td><td> 0,5%</td><td> 5,5%</td><td> 5,8%</td><td> 5,3%</td>

<td>Carbon in the catalyst coming out of the generator</td><td> 0,2%</td><td> 4,9%</td><td> 5,8%</td><td> 0,2%</td>

<td>Methanol conversion</td><td> 97,8%</td><td> 91,3%</td><td> 89,9%</td><td> 90,7%</td>

<td>Selectivities *:</td><td></td><td></td><td></td><td></td>

<td>Hydrogen</td><td> 0,1</td><td> 0,2</td><td> 0,1</td><td> 0,1</td>

<td>Methane</td><td> 3,1</td><td> 2,1</td><td> 4,0</td><td> 3,9</td>

<td>Ethane</td><td> 0,4</td><td> 0,5</td><td> 0,9</td><td> 0,9</td>

<td>Ethylene</td><td> 10,8</td><td> 26,7</td><td> 35,0</td><td> 32,6</td>

<td>Propane</td><td> 5,1</td><td> 0,7</td><td> 0,2</td><td> 0,7</td>

<td>Propylene</td><td> 16,0</td><td> 24,4</td><td> 27,5</td><td> 26,4</td>

<td>Butenos</td><td> 14,9</td><td> 14,0</td><td> 14,1</td><td> 14,2</td>

<td>Comp. heavy</td><td> 34,9</td><td> 17,9</td><td> 13,4</td><td> 15,6</td>

<td>CO₂</td><td> 0,3</td><td> 0,1</td><td> 0,6</td><td> 0,6</td>

<td>Coke</td><td> 14,3</td><td> 13,6</td><td> 4,1</td><td> 5,1</td>

* "Selectivities" are on a water-free basis

The preceding results demonstrated that coked catalysts (~ 5% on catalyst) achieved higher selectivities to ethylene and propylene than catalysts having coke levels less than 1%. The results also showed that the fully regenerated catalyst that was allowed to accumulate coke achieved higher selectivities than the catalyst that was partially regenerated to reduce the same coke to the same level. These results are consistent with the theory that partial coke regeneration selectively removes coke that is blocking undesirable surface reactions that form propane and C materials.₅+. The reactions that selectively prepare ethylene and propylene occur in the small pores, and the coke that accumulates in these pores is more difficult to remove than the coke on the outer surface (macropore). When a "clean" catalyst is allowed to coke, the coke deposits faster on the surfaces of the macropores than on the micropores, slowing down non-selective surface reactions, thus justifying the improved selectivities for the "coked" versus the "clean" catalyst. . As more and more coke builds up, however, the catalyst eventually becomes inactive. Activity is restored by regeneration with air, but it is important that the catalyst is completely burned to remove as much carbon as possible, and then allowed to coke again in the reactor. Partial catalyst regeneration, taught in US-A-4,873,390 to Lewis, was not nearly as effective in maintaining selectivity to ethylene and propylene in the reactor.

ES 2 247 702 T3

Based on the foregoing, it was concluded that better selectivity to light olefins can be achieved in an oxygenate to olefin conversion if a desirable coke is maintained in a total molecular sieve catalyst reaction volume by fully regenerating only a part of the catalyst and returning at least a part of the regenerated part to the total reaction volume.

The embodiment described herein is only intended to be illustrative, and should not be construed as limiting the invention, which is defined in the following claims.