JP5260300B2 - A process for producing acrolein from propane, acrylic acid, or a mixture thereof - Google Patents

Description

The present invention relates to a process for producing acrolein from propane, or acrylic acid, or a mixture thereof,
A) Feeding first reaction zone A with at least two gas feed streams containing propane (of which at least one stream contains fresh propane) to produce reaction gas A;
In reaction zone A, reaction gas A is passed through at least one catalyst bed where molecular hydrogen and propylene are produced by partially heterogeneously catalyzed dehydrogenation of propane,
The reaction zone A is supplied with molecular oxygen that oxidizes the molecular hydrogen contained in the reaction gas A in the reaction zone A to water vapor, and the reaction gas A contains molecular hydrogen, water vapor, propylene, and propane. Take out A,
B) At least one oxidation reaction of the product gas A taken out from the reaction zone A in the reaction zone B together with the reaction gas B containing molecular hydrogen, water vapor, propane, propylene and molecular oxygen under the supply of molecular oxygen Acrolein as a target product, or acrylic acid, or a mixture thereof, unreacted propane as a by-product by heterogeneously catalyzed partial gas phase oxidation of propylene contained therein Producing a product gas B containing molecular hydrogen, water vapor, carbon dioxide, and other subcomponents boiling below or above water,
C) The product gas B is discharged from the reaction zone B, the target product contained therein, water and secondary components boiling higher than water are separated therefrom in the first separation zone I, and unreacted propane, Carbon dioxide, molecular hydrogen, secondary components boiling below water and residual gas I optionally containing unreacted propylene and possibly unreacted molecular oxygen in reaction zone B,
D) As post-treatment means 1, in the second separation zone II, the carbon dioxide contained in the residual gas I is washed out, and the water still optionally contained in the residual gas I is optionally condensed,
As post-processing means 2, a partial amount of residual gas I is discharged,
Optionally, as post-treatment means 3, molecular hydrogen contained in the residual gas I is separated through a separation membrane in the third separation zone III, and optionally as post-treatment means 4, optionally contained in the residual gas I Chemical reduction of molecular oxygen
In this case, the order of application of the post-treatment means 1 to 4 is arbitrary, and E) a residual gas I containing unreacted propane remaining after the application of the post-treatment means 1 and 2 and optionally 3 and / or 4 (Also referred to herein as circulating gas I) is refluxed into reaction zone A as at least one stream of at least two feed streams containing propane.

  Acrylic acid is an important basic chemical used in particular as a monomer for the production of polymers which are used, for example, as binders present in dispersible distribution in aqueous media. Other applications of acrylic acid polymers are superabsorbents in hygiene and other applications.

  Acrolein is an important intermediate for the production of eg glutardialdehyde, methionine, 1,3-propanediol, 3-picoline, folic acid and acrylic acid.

  The preparation of acrolein from propane or acrylic acid, or mixtures thereof, described in the introduction of the present specification is known in an analogous manner (see, for example, DE-A3313573 and EP-A117146).

  In this method, the product gas A discharged from the reaction zone A is not subjected to substance separation before being used for charging the reaction zone B, so that DE-A102004032129, EP-A731077, DE-A102005049699, DE-A102005052923 , WO01 / 96271, WO03 / 011804, WO03 / 076370, WO01 / 96270, DE-A102005009891, DE-A102005013039, DE-A102005022798, DE-A102005009885, DE-A1020050101111, DE-A102455585, DE-A10316039, WO03 / 011804 To the similar methods described in. This usually means that valuable products associated with such material separations (especially in the case of thermal separation processes) and energy loss and necessary equipment consumption are avoided in this respect. Is advantageous. Of course, the method according to the introduction of the present description cannot completely avoid such material separation. Rather, as with all circulating gas processes, not only at least one target product discharge, but also at least one secondary component discharge is required. However, the method described at the beginning of this specification brings the subcomponent separation closer to the target product separation, and therefore, in order to reach the target, anyway, inevitably under the energy and equipment consumption. -Interesting as long as a thermal gradient is forced. Linking subcomponent separation to such consumption reduces all required consumption.

  The secondary component that protrudes individually in the relevant circulating gas process is molecular hydrogen produced in reaction zone A.

Forced by the oxygen present and no intermediate free hydrogen is produced (hydrogen drawn from the hydrocarbon to be dehydrogenated is drawn directly as water (H ₂ O)) or undetectable Unlike some exothermic, heterogeneously catalyzed oxydehydrogenation, the heterogeneously catalyzed dehydrogenation to be carried out in reaction zone A should be understood as ("conventional") dehydrogenation. Yes, the change in calorie is endothermic, unlike oxydehydrogenation (which may include exothermic hydrogen combustion in reaction zone A as a continuation step), and at this time liberated at least intermediately Molecular hydrogen is produced. To that end, it usually requires different reaction conditions and a different catalyst than oxydehydrogenation.

That is, the relevant process inevitably progresses in reaction zone A under H ₂ − generation. Since molecular hydrogen so generated in reaction zone A is not strictly a reactant of the target reaction occurring in reaction zone B, it does not naturally deplete in the relevant circulating gas process. Thus, in the described circulating gas process, where and in what form the molecular hydrogen produced in reaction zone A (and optionally supplied to reaction zone A) should be discharged again. , Should be determined by the operator's decision (in the ideal case, one hydrogen molecule is produced per molecule of propylene produced in reaction zone A).

A proposal made in a recent state-of-the-art specification (see, for example, US Pat. It is introduced as an inert gas until after the separation zone, and then it is completely burned into water by unreacted molecular oxygen and heterogeneous catalysis in reaction zone B, and the total amount of the generated water is returned to reaction zone A. Thus, the molecular hydrogen produced in the reaction zone A is separated for the first time in the second pass through the circulation runway and exclusively in its oxidized form in the separation zone, ie as water. Such a method is disadvantageous in various respects. For one thing, the non-separated molecular hydrogen has a special dispersibility in the reaction gas B (some structural materials are permeable to molecular hydrogen), as well as a prominent reduction potential (special inspection is in the reaction zone B). The presence of an increased amount of molecular hydrogen has been shown to negatively affect the duration of the catalyst to be used for heterogeneously catalyzed partial gas phase oxidation) and On the other hand, the target product must be separated from a significant amount of water in separation zone I, which requires significant energy because the target product has a high affinity for water (typically, The increased amount of water in reaction gas B also reduces the duration of the catalyst used for partial oxidation, which typically contains Mo in the oxide form, because H ₂ O is Mo-oxidized. Because it promotes the sublimation of objects). In addition, with the hydrogen combustion described, very high combustion heat is released which is difficult to use locally and advantageously locally after the separation zone I. The use of an excess of molecular oxygen relative to the target reaction stoichiometry, which favors an increased catalyst duration in the partial oxidation of reaction zone B, is inseparable in connection with the above problems, Naturally follows the increased explosion risk in reaction zone B (reactions when acrylic acid is the desired product and the reaction zone B is pre-fed with all the necessary molecular oxygen. A molar ratio of molecular oxygen contained in zone B to propylene contained therein of at least> 1.5 is necessary (in this case, little propylene complete combustion has already been taken into account)).

Furthermore, as an alternative, the known state of the art provides a variant in which, in that range, the total amount of molecular hydrogen produced in reaction zone A is as such after separation zone I. Will be introduced. Thereafter, all propane and propylene contained in the residual gas I are separated from all other components of the residual gas I, including molecular hydrogen, and only the C ₃ -hydrocarbon stream thus separated is simply separated. It is recommended to separate and reflux to reaction zone A. However, the disadvantage of this variant is that the entire amount of C ₃ -hydrocarbons must first be transferred over the entire range to the condensed phase and then again returned to the gas phase.

  As another proposal, DE-A 3331573 oxidizes molecular hydrogen generated in reaction zone A completely to water by molecular oxygen supplied in reaction zone A and still reacts at least a portion of the generated water. It includes the possibility of condensing before zone B and separating in this way, or introducing the entire amount produced into reaction zone B. Both methods have drawbacks. For one, the condensation of water requires the cooling of product gas A to a temperature well below that required in reaction zone B.

  On the other hand, this temperature drop must start from a very high temperature level, because the combustion of molecular hydrogen is about twice that amount of heat consumed in the range of dehydrogenation for its production. It is because it releases.

  Furthermore, when leaving the total amount of water produced in the reaction gas in the separation zone I, the increased separation consumption as described above and, as a rule, the reduced catalyst duration in the reaction zone B results. Cause it to occur.

  In contrast to the known state of the art described, the object of the present invention was to obtain a method according to the preamble of the present description which has the disadvantages described above, optionally further reduced.

  A relatively simple elucidation of the established problem is that both the discharge of molecular hydrogen generated in reaction zone A and the generation of discharge forms extend over a larger range of methods. In this case, a partial discharge is advantageously realized (only) in the separation zone I in the form of water that is exclusively discharged after the reaction zone B and water previously produced by hydrogen combustion in the reaction zone A. The partially oxidized water that is absolutely generated in the separation zone I in the reaction zone B is separated together in any case by a known target product separation method. Naturally, a corresponding device effective for the separation is already included.

Therefore, as an elucidation of the problem of the present invention, a process for producing acrolein from propane, or acrylic acid, or a mixture thereof is obtained.
A) Feeding first reaction zone A with at least two gas feed streams containing propane (of which at least one stream contains fresh propane) to produce reaction gas A;
In reaction zone A, reaction gas A is passed through at least one catalyst bed where molecular hydrogen and propylene are produced by partially heterogeneously catalyzed dehydrogenation of propane,
The reaction zone A is supplied with molecular oxygen that oxidizes the molecular hydrogen contained in the reaction gas A in the reaction zone A to water vapor, and the reaction gas A contains molecular hydrogen, water vapor, propylene, and propane. Take out A,
B) At least one oxidation reaction of the product gas A taken out from the reaction zone A in the reaction zone B together with the reaction gas B containing molecular hydrogen, water vapor, propane, propylene and molecular oxygen under the supply of molecular oxygen Acrolein as a target product, or acrylic acid, or a mixture thereof, unreacted propane as a by-product by heterogeneously catalyzed partial gas phase oxidation of propylene contained therein Producing a product gas B containing molecular hydrogen, water vapor, carbon dioxide, and other subcomponents boiling below or above water,
C) The product gas B is discharged from the reaction zone B, the target product contained therein, water and secondary components boiling higher than water are separated therefrom in the first separation zone I, and unreacted propane, Carbon dioxide, molecular hydrogen, secondary components boiling below water and residual gas I optionally containing unreacted propylene and possibly unreacted molecular oxygen in reaction zone B,
D) As post-treatment means 1, in the second separation zone II, the carbon dioxide contained in the residual gas I is washed out, and the water still optionally contained in the residual gas I is optionally condensed,
As post-processing means 2, a partial amount of residual gas I is discharged,
Optionally, as post-treatment means 3, molecular hydrogen contained in the residual gas I is separated through a separation membrane in the third separation zone III, and optionally as post-treatment means 4, optionally contained in the residual gas I Chemical reduction of molecular oxygen
In this case, the order of application of the post-treatment means 1 to 4 is arbitrary, and E) the post-treatment means containing unreacted propane remaining after the application of the post-treatment means 1 and 2 and optionally 3 and / or 4. Residual gas I (also referred to herein as circulating gas I) is refluxed into reaction zone A as at least one stream of at least two feed streams containing propane,
This process comprises at least 5 mol% of the total amount of molecular hydrogen produced in reaction zone A and optionally fed to reaction zone A, but not more than 95 mol%, in an amount M of molecular hydrogen in reaction zone A. And is characterized by being oxidized to water vapor.

  According to the invention, the amount M is preferably at least 10 mol% but not more than 90 mol% (always related in a corresponding manner). The amount M is particularly preferably at least 15 mol%, but not more than 85 mol%. The amount M is even better still at least 20 mol% but not more than 80 mol%. The amount M is very particularly preferably at least 25 mol% but not more than 75 mol%. The amount M is still better at least 30 mol% and at most 70 mol%. More advantageous is an amount M of at least 35 mol% and at most 65 mol%. Even better still, the amount M is at least 40 mol%, but not more than 60 mol%. The amount M is best at least 45 mol% and not more than 55 mol%. M = 50 mol% is very particularly advantageous according to the invention.

  The essential advantage of the process according to the invention is that the propane remains in its entire circulation, mainly in a gaseous state of aggregation, ie it should not be transferred to a condensed liquid phase. This is of course advantageous because propane is a relatively non-polar molecule and its condensation is relatively expensive.

  Another advantageous feature of the process according to the invention is that the reaction gas B necessarily contains molecular hydrogen in an advantageous amount according to the invention. This is because, in addition to the above-mentioned disadvantageous properties according to the theory of DE-A 3313573 and EP-A 117146, it also has the advantage that it remains chemically inert in the reaction zone B. That is, at least 95 mole percent, usually rather at least 97 mole percent or at least 99 mole percent of the molecular hydrogen fed to reaction zone B remains chemically unchanged during reaction zone B circulation.

  However, this also has the best thermal conductivity advantage in the gas at the same time. According to Walter J. Moore, Physikalische Chemie, WDEG publication, Berlin (1973), page 171, at standard conditions, for example, the thermal conductivity of carbon dioxide. It is more than 10 times larger and about 8 times larger than that of molecular nitrogen or molecular oxygen.

According to Table 1 of DE-A 3313573, in reaction zone B this increase in CO _x -production selectivity is significantly reduced in the presence of molecular hydrogen than in the absence of molecular hydrogen. The cause is the thermal conductivity. It conditions a relatively low catalyst surface temperature and thereby a lower proportion of complete propylene combustion by faster heat exhaustion from the reaction sites in reaction zone B. This is especially true when, as in the process according to the invention, the highest thermal conductivity molecular hydrogen and the highest heat-absorbing propane are synergistically shared in the heat exhaustion.

  As used herein, fresh propane refers to propane that has not yet participated in dehydrogenation in reaction zone A. This is usually supplied as a component of crude-propane (which advantageously meets the specifications according to DE-A 10256119 and DE-A 10245585) which contains a small amount of components different from propane. Such crude-propane can be obtained, for example, by the method described in DE-A 102005022798. As a rule, the reaction zone A is fed as a feed stream containing other propane in addition to fresh propane and also with the residual gas I which has been post-treated only according to the invention.

  In the process according to the invention, the supply of fresh propane exclusively to reaction zone A is advantageously carried out as a component of the charge gas mixture for reaction zone A. In principle, however, a portion of fresh propane can be fed to the charge gas mixture of the first and / or second stage of reaction zone B for explosion safety reasons.

  The reaction gas B loaded in the reaction zone B advantageously meets the specifications likewise recommended in DE-A 10246119 and DE-A 10245585. Furthermore, between the reaction zone A and the reaction zone B, a mechanical separation operation according to DE-A 10316039 is advantageously connected according to the invention.

  As used herein, the loading of the catalyst bed that catalyzes the reaction stage with the reaction gas is the standard liter (= Nl; corresponding reaction gas) that is passed through 1 liter of catalyst bed (eg, catalyst solid bed) per hour. The amount of reaction gas at the standard conditions (capacity liters taken in at 0 ° C., 1 bar) is solved.

  The load may be related only to the components of the reaction gas. In that case, this amount of component Nl / l · h passed through 1 liter of catalyst bed per hour (a pure inert material bed is not counted in the catalyst solid bed). If the catalyst bed comprises a mixture from the catalyst and the inert dilution molding, the loading may only be related to the volume unit of the contained catalyst, as long as it is described accordingly.

  As an inert gas herein, it is generally advantageous that it is actually chemically inert under the corresponding reaction conditions and up to more than 95 mol% (in view of each inert reaction gas component itself). Reactive gas components that remain chemically unchanged up to 97 mol% or more or 99 mol% or more should be understood.

A typical reaction gas B that can be charged to the reaction zone B in the process according to the invention has the following contents:
Propylene 4-25% by volume
Propane 6-70% by volume
H ₂ O 5~60% by volume
O ₂ 5 to 65 volume% and H ₂ 0.3 to 20 volume%.

The reaction gas B advantageous according to the invention has the following contents:
Propylene 6-15% by volume
Propane 6-60% by volume
H ₂ O 5~30% by volume
O ₂ 8 to 35 volume% and H ₂ 2 to 18 volume%.

The reaction gas B which is very particularly advantageous for said loading according to the invention has the following contents:
Propylene 8-14% by volume
Propane 20-55% by volume, preferably 20-45% by volume
H ₂ O 10 to 25 volume%
O ₂ 10 to 25 volume%, preferably 15 to 20 volume% and H ₂ 6 to 15% by volume.

  The advantage of the average propane content in the reaction gas B is apparent, for example, from DE-A 10245585.

Within the above composition raster, it is advantageous if the molar ratio V _{1 of} propane contained in the reaction gas B to propylene contained in the reaction gas B is ₁ to 9 (ie advantageously according to the invention). Propane / propene separation according to WO 04/094041 can be omitted). Furthermore, it is advantageous when the ratio V _{2 of} molecular oxygen contained in the reaction gas B to propylene contained in the reaction gas B is 1 to 2.5 in the composition raster. Furthermore, in the meaning of the present invention, it is advantageous that the ratio V _{3 of} propylene contained in the reaction gas B to molecular hydrogen contained in the reaction gas B is 0.5 to 20 in the composition raster. It is. Similarly, the molar ratio V ₄ of the total molar amount of propane and propylene contained in the reaction gas B to the water vapor contained in the reaction gas B may be 0.005 to 10 in the composition raster. It is advantageous.

V ₁ in the reaction gas B used in the charging of the reaction zone B (reaction gas B—in the starting mixture) is particularly preferably 1 to 7 or to 4, or 2 to 6 and particularly preferably 2 to 5 or 3. .5 to 4.5. Furthermore, for the reaction gas B- starting mixture, if _{V 2} is a 1,2～2.0 or 1.4 to 1.8 are preferred. Furthermore, for the reaction gas B- starting mixture, if _{V 3} is 0.5 to 15, or 0.5 to 10 or 0.5 to 1.5 are preferred. Correspondingly, V ₄ in the reaction gas B-starting mixture is preferably 0.01-5, better 0.05-3, preferably 0.1-1 and particularly preferably 0.1-0. 5 or similarly 0.3.

  A non-explosive reaction gas B-starting mixture is advantageous according to the invention.

  Reaction gas B-For the answer to whether the starting mixture is explosive or not, under a mixture under certain conditions (pressure, temperature), by a local ignition source (eg a burning platinum wire) It is decisive whether the introduced combustion (ignition, explosion) spreads (see inspection record in DIN 51649 and WO 04/007405). If spreading is done, the mixture should be characterized as explosive. If there is no spreading, the mixture is positioned herein as non-explosive. If the partial oxidation reaction gas starting mixture according to the invention is non-explosive, this also applies to the reaction gas mixture produced therefrom during the course of partial oxidation (see WO 04/007405).

  As a source of molecular oxygen that is required in reaction zone B and not contained in product gas A, molecular oxygen itself or a gas that is chemically inert in molecular oxygen and reaction zone B (or Mixtures containing such inert gas mixtures (e.g. noble gases such as argon, molecular nitrogen, water vapor, carbon dioxide, etc.) can be used (e.g. air). According to the invention, other gas (different from molecular oxygen) 30% by volume or less, preferably 25% by volume or less, preferably 20% by volume or less, particularly preferably 15% by volume or less, better still 10% by volume or less and particularly advantageous The molecular oxygen containing 5% by volume or less is supplied as a gas. It is particularly advantageous if the molecular oxygen to be supplied to reaction zone B is supplied in its pure form.

  The same applies in principle to the molecules to be fed in excess of the molecular oxygen optionally contained in the residual gas I which is post-treated and refluxed to the reaction zone A in the process according to the invention in the reaction zone A. The same applies to oxygen. However, since the oxygen demand in reaction zone A is relatively low, the use of air as an oxygen source for the oxygen demand in reaction zone A is also advantageous in the process according to the invention, especially for economic reasons. .

Indeed, in the process according to the invention, in general, the post-treated residual gas I refluxed in the reaction zone A still contains residual molecular oxygen in the reaction zone B and, in principle, the reaction This oxygen content of the post-treated residual gas I refluxed in zone A can then be distributed such that no further supply of molecular oxygen to reaction zone A is necessary in the process according to the invention. However, in principle, the post-treated residual gas I refluxed in the reaction zone A does not necessarily contain oxygen according to the invention, and in many examples according to the invention, in the reaction zone A, Requires additional oxygen supply. In that case, this may be pure oxygen or molecular oxygen and a gas that is chemically inert in one or several reaction zones A, B (eg N ₂ , H ₂ O, noble gases and / or Or as a mixture containing CO ₂ ) (eg as air). That is, according to the invention, other gases different from molecular oxygen are 30% by volume or less, preferably 25% by volume or less, preferably 20% by volume or less, particularly preferably 15% by volume or less, better still 10% by volume or less and particularly advantageous Is advantageously performed as a gas containing 5% by volume or less or 2% by volume or less. For this, a supply of pure oxygen is particularly advantageous.

  The supply of pure oxygen to reaction zone A, as well as to reaction zone B, which is advantageous according to the present invention as described above, should be discharged again in the further course of the circulation process, according to the invention. This is particularly advantageous since it is not unavoidably loaded with an inert gas.

Very generally, the total content of components different from propylene, molecular hydrogen, water vapor, propane and molecular oxygen in the reaction gas B-starting mixture is usually ≦ 40% by volume, or ≦ 35% by volume, or ≦ 30% by volume, Alternatively, it is typical for the process according to the invention that ≦ 25% by volume, or ≦ 20% by volume, furthermore ≦ 15% by volume, often ≦ 10% by volume. Thus, a total content of ≦ 5% by volume is difficult to realize within the scope of the method according to the invention. Among this other component of the reaction gas B-starting mixture, ethane and / or methane may be present up to 80% by volume. Among such inclusions, in particular, carbon oxides (CO ₂ , CO) and noble gases, but oxygen supply subcomponents such as formaldehyde, benzaldehyde, methacrolein, acetic acid, propionic acid, methacrylic acid, etc. are important. is there. Of course, ethylene, iso-butene, n-butane, n-butene and molecular nitrogen also belong to the possible other components of the reaction gas B-starting mixture. However, the advantage of the process according to the invention (see EP-A293224) is basically that the reaction gas B-starting mixture is 0.1-30% by volume of CO ₂ , or 1-25% or 20% by volume, often 5 to 15% by volume can be contained. The CO content is typically ≦ 5% by volume, or ≦ 4% by volume, or ≦ 3% by volume, or ≦ 2% by volume, usually ≦ 1% by volume. However, as a rule, N ₂ ≦ 20% by volume, preferably ≦ 15% by volume, particularly preferably ≦ 10% by volume, and very particularly preferably ≦ 5% by volume.

  In principle, for example, WO03 / 076370, WO01 / 96271, EP-A117146, WO03 / 011804, EP-A731077, US-A31316670, WO01 / 96270, DE-A3315573, DE-A10245585, DE-A10316039, DE-A102005009891, All known heterogeneous catalyst parts of propane in reaction zone A, as already known from DE-A102005013039, DE-A102005022798, DE-A102005009885, DE-A102005010111, DE-A102005049699 and German application DE-A102004032129. This is the case for dehydrogenation.

  That is, for one pass of propane fed to reaction zone A through reaction zone A, reaction zone A is fluid (ie, liquid or gaseous) introduced outside reaction zone A. It can be made isothermal by suitable heat exchange with the heat carrier. However, this can be carried out under comparable conditions, adiabatically, ie without such a purposeful heat exchange with a heat carrier that is actually introduced outside the reaction zone A. In the latter case, the total heat change for one pass through reaction zone A of propane fed to reaction zone A is recommended in the above specification and is then endothermic by means to be further described. It can be (negative) or autothermal (actually zero) or exothermic (positive). Similarly, the catalysts recommended in the above specification can be applied in the process according to the invention. In principle, heterogeneously catalyzed propane dehydrogenation is independent of whether it is an adiabatic or isothermal operation and is carried out in a fixed bed reactor, in a moving bed-, fluidized bed- or vortex bed reactor. Is possible (the latter is based on the backmixing, particularly when the circulating gas I contains molecular oxygen, the reaction gas A-starting to the reaction temperature in the reaction zone A by hydrogen combustion in the reaction gas A) Suitable for heating the mixture).

Typically, heterogeneously catalyzed partial dehydrogenation of propane to propylene requires relatively high reaction temperatures. The conversion rates achievable here are usually limited by thermodynamic equilibrium. Typical reaction temperatures are 300-800 ° C or 400-700 ° C. At this time, one molecule of hydrogen is generated per one molecule of propane dehydrogenated to propylene. The operating pressure in reaction zone A is typically 0.3-5 or -3 bar. According to the invention, the operating pressure in reaction zone A is preferably 2-5 or -4 bar. However, it may be up to 20 bar. The amount of hydrogen corresponding to at least 50 mol% of the total amount of molecular hydrogen produced in reaction zone A, where reaction zone A is operated at a very high pressure (for example> 5 or 10-20 bar) When the gas is combusted, the product gas A is released by expansion in the expansion turbine, and together with the operation performed, the CO ₂ -heat is released together from the compressor for the residual gas I in advance. Is advantageous. At the same time, product gas A is allowed to cool to the temperature required for further use in reaction zone B. High temperature and removal of the reaction product H ₂ promotes an equilibrium state in the reaction zone A in the sense of the desired product.

  As the heterogeneously catalyzed dehydrogenation proceeds with increasing capacity, the conversion can be increased by lowering the partial pressure of the dehydrogenated product. This can be done in a simple manner, for example by dehydrogenation at reduced pressure (naturally increased pressure is usually advantageous for catalyst duration) and / or inactive dilution. It can be achieved by mixing a gas, for example water vapor, which is an inert gas in the standard case for dehydrogenation reactions. Since steam reacts with the product coke on the principle of coal gasification, dilution with steam is usually further contingent on reduced coking of the catalyst used. The heat capacity of water vapor can also balance a portion of the dehydrogenation endotherm.

  Due to the activity of the partial oxidation catalyst, it has been found that a limited amount of water vapor in the next reaction zone B is usually found to be effective, so for that reason a further amount in reaction zone B is It is disadvantageous. In accordance with the present invention, further in the reaction zone A, the hydrogen content should be oxidized to water vapor. Therefore, according to the present invention, the amount of water vapor supplied to the reaction zone A relative to the reaction gas A supplied to the catalyst loading in the reaction zone A is ≦ 20% by volume, preferably ≦ 15% by volume and particularly preferably ≦ The case of 10% by volume is advantageous. However, as a rule, the amount of water vapor fed to the reaction zone A is correspondingly usually ≧ 1% by volume, often ≧ 2% by volume, or ≧ 3% by volume and often ≧ 5% by volume.

Suitable diluents for heterogeneously catalyzed propane dehydrogenation are, for example, nitrogen, noble gases such as He, Ne and Ar, but also compounds such as CO, CO ₂ , methane and ethane. All of the above diluents may be used in combination in themselves or in different mixtures. As long as the dilution gas is produced as a by-product in the circulating gas process according to the invention or supplied as fresh gas (or fresh gas component), a corresponding fresh gas supply is advantageous in small quantities according to the invention. Thus, in the discharge method according to the invention, it is required in a corresponding amount. However, in principle, in the method according to the invention, a certain amount of dilution gas can actually be circulated, with the loss lost being newly replenished.

  For the heterogeneously catalyzed propane dehydrogenation, in principle, all known dehydrogenation catalysts are known in the state of the art. They are broadly divided into two groups: those that are oxidative (eg chromium oxide and / or aluminum oxide) and at least one, usually deposited on an oxidic support, typically relatively rare metals ( For example, platinum is included. Thus, in particular, WO01 / 96270, EP-A731077, DE-A10211275, DE-A10131297, WO99 / 46039, US-A4788371, EP-A070705136, WO99 / 29420, US-A4220091, US-A543020, US-A5877369, EP- All dehydrogenation catalysts recommended in A0117146, DE-A 19937196, DE-A 19937105 and DE-A 19937107 can be used. In particular, the catalysts according to Examples 1, 2, 3, and 4 of DE-A 19937107 can also be used.

  In this case, 10 to 99.9% by mass of zircon dioxide, 0 to 60% by mass of aluminum oxide, silicon dioxide and / or titanium dioxide, and at least one element of the first or second main group of the periodic table of elements, Dehydrogenation containing a third subgroup element, an eighth subgroup element, lanthanum and / or tin in an amount of 0.1 to 10% by mass (wherein the sum of the mass% is 100% by mass) The catalyst is important.

  The dehydrogenation catalyst used in the examples herein is also particularly suitable.

  In general, the dehydrogenation catalyst is a catalyst cord (diameter, typically 1-10 mm, preferably 1.5-5 mm; length, typically 1-20 mm, preferably 3-10 mm), tablet (advantageously The same shape as the cords) and / or catalyst annulus (outer diameter and length, typically 2-30 mm or 10 mm, respectively, wall thickness, preferably 1-10 mm, or -5 mm, or -3 mm) It may be. For the performance of heterogeneously catalyzed dehydrogenation in a vortex bed (or fluidized or moving bed), a correspondingly finely divided catalyst is used. According to the invention, a catalytic solid bed is advantageous in the reaction zone A.

  Typically, the dehydrogenation catalyst (especially used illustratively herein and recommended in DE-A 19937107 (especially this DE-A illustrative catalyst)) is a dehydrogenation of propane. It is procured so that it can also catalyze the combustion of molecular hydrogen. At this time, hydrogen combustion proceeds much more rapidly in the case of a competitive situation on the catalyst compared to propane dehydrogenation.

  For the implementation of heterogeneously catalyzed propane dehydrogenation, all catalyst types and variants of the state of the art are in principle applicable. Descriptions of such variants are included, for example, in all specifications cited with respect to the prior art and the prior art dehydrogenation catalysts cited at the beginning of this specification.

  A relatively detailed description of dehydrogenation processes suitable according to the present invention includes: Catalytica® Study Division, Oxidative Dehydrogenation and Alternative Dehydrogenation Processes, Study Number 4192OD, 1993, 430 Ferguson Drive, Mountain View, California, 94043-5272U. S. Also included in A.

  For the partially heterogeneously catalyzed dehydrogenation of propane, it is characterized by an endothermic process as described above. That is, the heat (energy) required for adjusting the required reaction temperature and for the reaction should be supplied in particular to the reaction gas in advance and / or during the course of heterogeneously catalyzed dehydrogenation. is there. In some cases, the reaction gas A should recover the necessary heat of reaction from itself.

  Furthermore, for heterogeneously catalyzed dehydrogenation of propane, based on the high reaction temperature required, a small amount of high-boiling polymeric organic compound is produced up to carbon, which precipitates on the catalyst surface and thereby Typically, the catalyst is deactivated. In order to minimize this disadvantageous accompanying phenomenon, the propane-containing reaction gas A to be introduced onto the catalyst surface at a temperature elevated to the heterogeneous catalyzed dehydrogenation may be diluted with water vapor as described above. it can. The precipitated carbon is thus partially or completely eliminated under the given conditions, according to the principle of coal gasification.

  Another possibility for removing the precipitated carbon compounds is to pass an oxygen-containing gas (preferably in the absence of hydrocarbons) through the dehydrogenation catalyst at an occasionally elevated temperature, so that the precipitated carbon is apparently burned. It is in. However, a certain suppression of the formation of carbon precipitates is possible by supplying molecular hydrogen to the propane to be dehydrogenated by heterogeneous catalysis before introducing it onto the dehydrogenation catalyst at an elevated temperature. is there.

  Of course, it is also possible to supply a mixture containing steam and molecular hydrogen to propane to be dehydrogenated by heterogeneous catalysis. The provision of molecular hydrogen for the heterogeneously catalyzed dehydrogenation of propane also reduces the unwanted production of allene (propadiene), propyne and acetylene as by-products.

  Thus, in the simplest case for the production of reaction gas A (also referred to herein as reaction gas A charge gas mixture or reaction gas A-starting mixture) to be introduced through at least one catalyst bed. Only fresh propane and circulating gas I can be fed to reaction zone A. In this case, the latter can already contain the required amount of molecular oxygen in the reaction zone A in order to condition the hydrogen combustion necessary according to the invention. This is because molecular oxygen is used in reaction zone B, preferably in excess, relative to the reaction stoichiometry, and this is usually extensively retained in the circulation gas I in the process according to the invention. Derived from.

  In said case, the circulating gas I is regularly positive, such that the steam can develop its advantageous properties for the whole process together with the steam produced in the reaction zone A in the range of hydrogen combustion. Contains water vapor in an amount. The supply of a further gas stream to reaction zone A is not necessary in this case. The desired conversion in reaction zone A takes place in a single pass of reaction gas A through it.

  But of course, in order to develop this advantageous effect described here, in addition to fresh propane and circulating gas I for the production of reaction gas A to be passed through at least one catalyst bed. Further, water vapor and / or molecular hydrogen can be supplied. The molar ratio of molecular hydrogen to propane in the charge gas mixture of reaction zone A is typically ≦ 5. The molar ratio of water vapor to propane in the charge gas mixture of reaction zone A can be, for example, ≧ 0-30, preferably 0.1-2 and conveniently 0.5-1. Further, the charged gas mixture in reaction zone A can be supplied with extra molecular oxygen (in pure form and / or as a mixture with an inert gas) and / or extra inert gas as required. In that case, the desired return in the reaction zone A can be obtained again without the supply of further gas flow along the reaction path, again with the simple flow of the reaction gas A (of the charge gas in the reaction zone A). Can be done in one pass. In the present specification, the reaction path in the reaction zone A means the dehydrogenation conversion rate (heterogeneous catalysis) of propane supplied to the reaction zone A before the first passage through at least one catalyst bed in the reaction zone A. Depending on the conversion rate in dehydrogenation, the propane flow path through reaction zone A is solved.

  Suitable reactor configurations for such heterogeneously catalyzed propane dehydrogenation in a single pass through reaction zone A of the charge gas mixture without an intermediate gas supply are for example fixed bed tubes or bundle tubes Reactor. At this time, the dehydrogenation catalyst is present in one reaction tube or bundle tube as a solid bed. When the hydrogen combustion in the reaction zone A required by the present invention is distributed so that all reactions occurring in the reaction zone A endothermically, the reaction tube is advantageously heated from the outside according to the present invention ( But obviously, it can also be cooled if necessary). This is done, for example, by burning a gas, for example a hydrocarbon, for example methane, in the space surrounding the reaction tube. This direct form of contact tube heating is applied only over the first 20-30% of the fixed bed and the remaining bed is heated to the required reaction temperature by radiant heat released in the range of combustion. It is advantageous. In this way, a nearly isothermal reaction course can be achieved. A suitable reaction tube inner diameter is about 10-15 cm. A typical dehydrogenation bundle reactor includes 300-1000 or more reaction tubes. The temperature inside the reaction tube varies in the range of 300 to 700 ° C, preferably in the range of 400 to 700 ° C. It is advantageous to feed the reaction gas A-starting mixture to a tubular reactor preheated to the reaction temperature. The product gas (mixture) A can exit the reaction tube at a temperature lower by 50-100 ° C. However, this discharge temperature may be higher or the same as the level. Within the scope of the process, the use of oxide-based dehydrogenation catalysts based on chromium oxide and / or aluminum oxide is advantageous. Dehydrogenation catalysts are usually used undiluted. In large industry, several such bundle tube reactors can be operated in parallel and the product gas A can be mixed and used to charge reaction zone B. In some cases, only two of the reactors can be involved in the dehydrogenation operation, while the catalyst charge is regenerated in the third reactor.

  However, a single pass of charge gas through reaction zone A can be used in moving bed or fluidized bed reactors, as described, for example, in DE-A 10245585 and the literature cited in this regard. It can also be done.

  Reaction zone A of the process according to the invention can in principle also comprise two sections. Such a configuration of reaction zone A is especially recommended when the charge gas of reaction zone A does not contain molecular oxygen (which may be the case, for example, when the circulating gas I does not contain molecular oxygen). There is.

  In this case, the original heterogeneously catalyzed dehydrogenation takes place in the first section, and after the intermediate feed of molecular oxygen and / or a mixture comprising molecular oxygen and inert gas in the second section, The necessary heterogeneously catalyzed hydrogen combustion can take place.

  Very generally, in the process according to the invention, for one pass through reaction zone A, ≧ 5 mol% to ≦ 60 mol%, preferably ≧ 10 mol, of propane fed in total into reaction zone A % To ≦ 50 mol%, particularly preferably ≧ 15 mol% to ≦ 40 mol% and very particularly preferably ≧ 20 mol% to ≦ 35 mol% in the reaction zone A To operate advantageously. Such a limited conversion rate in reaction zone A is usually sufficient according to the invention, since the residual amount of unreacted propane actually acts as a diluent gas in the subsequent reaction zone B, And in the further course of the process according to the invention, it can be refluxed into the reaction zone A without loss. The advantage of a process with a low propane conversion is that a single pass of reaction gas A through reaction zone A requires a relatively low amount of heat for endothermic dehydrogenation and a relatively low reaction to achieve conversion. The temperature is sufficient.

  Advantageously according to the invention, as mentioned above, it may be advantageous to carry out propane dehydrogenation in reaction zone A (e.g. relatively low propane conversion) (in appearance) adiabatically. That is, the charge gas mixture in reaction zone A is typically first heated to a temperature of 500-700 ° C (or 550-650 ° C) (eg, by direct heating of the surrounding walls). In that case, it is usually sufficient to pass adiabatically once through the catalyst bed in order to achieve the desired dehydrogenation conversion rate and also the hydrogen combustion required by the present invention, the reaction gas being Depending on the quantitative ratio of endothermic dehydrogenation and exothermic hydrogen combustion in the mechanical evaluation, it is heated, cooled or neutralized. An adiabatic operating method in which the reaction gas is cooled at about 30-200 ° C. in a single pass is advantageous according to the invention. If necessary, in the second section of reaction zone A, the hydrogen produced in the range of dehydrogenation can be combusted after heterogeneous catalysis with intermediately supplied molecular oxygen. This combustion can likewise be carried out adiabatically.

  It should be noted that, in particular in adiabatic operation, a single shaft furnace reactor that is passed axially and / or radially by the reaction gas A is sufficient as a fixed bed reactor.

  In this case, in the simplest case, a single closed reaction space, for example having an inner diameter of 0.1 to 10 m, optionally 0.5 to 5 m, in which the catalyst solid bed is a support device (for example, The container on the grid is important. At this time, the thermal reaction gas A containing propane is passed in the axial direction through a reaction space loaded with a catalyst that is actually adiabatic in the adiabatic operation. At this time, the catalyst shape may be spherical, tubular or cord-like. In this case, all catalyst configurations with a particularly low pressure drop are advantageous, since the reaction space can be realized by a very cost-effective apparatus. This is in particular a catalyst shape in which a large gap is obtained or structurally constructed, for example a monolithic structure or a honeycomb. In order to achieve a radiant flow of the propane-containing reaction gas A, the reactor comprises, for example, two cylindrical grids that are concentrically arranged with each other present in the jacket covering, and the catalyst bed is It may be arranged in the annular slit. In the case of insulation, the metal coating is optionally re-insulated.

  As catalyst loading for the heterogeneously catalyzed propane dehydrogenation according to the invention, in particular, the all-catalyst disclosed in DE-A 19937107, in particular geometrically inert with respect to the heterogeneously catalyzed dehydrogenation. A mixture thereof with the body is also suitable.

After a long period of operation, the catalyst is, for example, in a simple manner, at an entry temperature of 300-600 ° C., often 400-550 ° C., first in the first regeneration stage, with nitrogen and / or steam (advantageously). It can be regenerated by introducing diluted air onto the catalyst bed. At this time, the catalyst (bed) load in the regeneration gas (for example, air) is, for example, 50 to 10,000 h ⁻¹ , and the oxygen content of the regeneration gas is 0.1 or 0.5 to 20% by volume. Good.

In the next further regeneration stage, air can be used as regeneration gas under the same regeneration conditions. From an application technical point of view, it is recommended that the catalyst be washed with an inert gas (eg N ₂ ) before its regeneration.

  Subsequently, typically with pure molecular hydrogen or molecular hydrogen diluted with inert gas (preferably water vapor and / or nitrogen) (hydrogen content should be ≧ 1% by volume), otherwise under the same conditions It is recommended to play.

The heterogeneously catalyzed propane dehydrogenation in reaction zone A of the process according to the invention is in all cases relatively low propane conversion (≦ 30 mol%) and high propane conversion (> 30 mol%). It can be operated with the same catalyst (bed) loading as in the variant (all reactive gases as well as the propane contained therein). The loading of this reaction gas A can be, for example, 100 to 40000 or 10000 h ⁻¹ , often 300 to 7000 h ⁻¹ , i.e. often about 500 to 4000 h ⁻¹ .

  In a particularly refined manner, the heterogeneously catalyzed propane dehydrogenation in reaction zone A (especially with a propane conversion of 15 to 35 mol% per pass) is carried out in a bedplate reactor. Can be realized.

  This includes one or more catalyst beds that catalyze dehydrogenation in spatial succession. The number of catalyst beds can be 1-20, preferably 2-8, but 3-6. As the number of floorboards increases, gradually increased propane conversion is easily achieved. The catalyst beds are preferably arranged continuously in a radial or axial direction. From a technical point of view, a catalyst solid bed type is applied in such a bed reactor.

  In the simplest case, the catalyst fixed bed is arranged in a shaft furnace reactor in the axial direction of concentrically arranged cylindrical grids or in an annular slit. However, it is also possible to arrange the fan-shaped annular slits together and send the gas to the next higher or lower fan after passing through the fan in a radial pattern.

  In an advantageous manner, the reaction gas A is introduced into the next catalyst bed on its way, for example by introduction onto the heat exchanger surface heated with hot gas, or in a floor plate reactor, Intermediate heating is carried out by introduction into a tube heated in (if necessary, intermediate cooling can also be carried out in a corresponding manner).

  When the floor-plate reactor is otherwise operated adiabatically, in particular for propane conversion ≦ 30 mol%, the catalyst of the exemplary embodiment described in DE-A 19937107 in particular. It is sufficient to preheat the reaction mixture to 450-550 ° C. (preferably 450-500 ° C.), send it to the dehydrogenation reactor and keep it in this temperature range in the bed reactor. That is, total propane dehydrogenation can be achieved between the two regenerations, at significantly lower temperatures that have proven to be effective, especially for the duration of the catalyst bed. For higher propane conversion, the reaction gas mixture is preheated at an advantageously higher temperature and sent to the dehydrogenation reactor (which can be up to 700 ° C.), and this increased temperature range in the floorboard reactor Hold on.

  It is more advisable to carry out the intermediate heating by a direct method (as much as possible by an autothermal method). For this purpose, the reaction gas A contains molecular hydrogen which has already been advantageously fed (for example as a component of the circulation gas I) before the flow of the first catalyst bed (in which case the reaction gas A-starting mixture is advantageously fed). And / or provide a limited range of molecular oxygen between subsequent catalyst beds. Thus (typically catalyzed by the dehydrogenation catalyst itself), contained in the reaction gas A, produced during the course of the heterogeneously catalyzed propane dehydrogenation and / or fed to the reaction gas A The limited combustion required by the present invention of molecular hydrogen can be caused in a particularly controlled manner (especially a catalyst loaded with a catalyst that specifically (selectively) catalyzes the combustion of hydrogen. It can also be applied technically advantageous to charge the bed into a floor-plate reactor (for example, US Pat. No. 4,788,371, US Pat. No. 4,886,928, US Pat. No. 5,430,209, US Pat. No. 5,530,171, US Pat. No. 5,527,979 and Each of US-A 5,563,314 corresponds to this; for example, such a catalyst bed is selectively placed in a bedplate reactor, selectively to a bed containing a dehydrogenation catalyst. This catalyst is also suitable for the hydrogen combustion described above in the second section of reaction zone A.) The heat of reaction released in this way is Depending on it, a totally exothermic or totally autothermal (total heat change is actually zero) or total endothermic operation of heterogeneously catalyzed propane dehydrogenation is possible, ie in the catalyst bed. With increasing selected residence time of the reaction gas, propane dehydrogenation is possible at a reduced temperature or indeed a constant temperature, which allows a particularly long duration between the two regenerations, And is advantageous according to the invention.

In general, according to the invention, the oxygen supply as described above is such that the oxygen content of the reaction gas A is 0.5-50 or -30, preferably 10-25% by volume, based on the amount of molecular hydrogen contained therein. Should be done to be. The source of oxygen can be pure molecular oxygen (which is advantageous according to the invention) or molecular oxygen diluted with an inert gas, for example CO, CO ₂ , N ₂ and / or a noble gas, but in particular also air. This is the case. Molecular oxygen is less than 30% by volume of other gas (different from molecular oxygen), preferably less than 25% by volume, preferably less than 20% by volume, particularly preferably less than 15% by volume, better still less than 10% by volume and particularly advantageous. It is advantageous according to the invention to be supplied as a gas containing up to 5% by volume. It is particularly advantageous for the oxygen supply to take place in pure form.

Combustion of 1 mole of molecular hydrogen into H ₂ O results in about twice the energy (about 240 kJ / mol) consumed by dehydrogenation of 1 mole of propane to propylene and H ₂ (about 120 kJ / mol). Thus, the autothermal situation as described above of reaction zone A in an adiabatic floor plate reactor achieves a particular purpose with respect to the intended advantages of the process according to the invention, but it is exactly in reaction zone A. Combustion of about 50 mol% of the amount of molecular hydrogen produced in the range of dehydrogenation of is required.

  However, the advantages of the process according to the invention are not only achieved when the amount of hydrogen of about 50 mol% of the amount of molecular hydrogen produced in reaction zone A is combusted in reaction zone A. Rather, this advantage is rather from 5 to 95 mol%, preferably from 10 to 90 mol%, particularly preferably from 15 to 85 mol%, very particularly preferably from 20 to 80 mol%, of the amount of molecular hydrogen produced in reaction zone A. More preferably 25 to 75 mol%, more preferably 30 to 70 mol%, still more preferably 35 to 65 mol% and most preferably 40 to 60 mol% or 45 to 55 mol% of hydrogen in reaction zone A It is also already effective when it is burned into water (preferably in the manner of operation of the insulated floor plate reactor described above).

  Typically, the oxygen supply should be carried out such that the oxygen content of the reaction gas A is 0.01 or 0.5 to 3% by volume, based on the amount of propane and propylene contained therein. .

  The isothermal nature of the heterogeneously catalyzed propane dehydrogenation is an internal device (e.g., tubular) that is advantageously pre-filled, but inevitably closed vacuum, in the space between the catalyst beds in the bedplate reactor. It can be further improved by installing Such internal devices can be installed in each catalyst bed. This internal device contains a suitable solid or liquid that evaporates or melts above a certain temperature, in which case it consumes heat and condenses again below this temperature, in which case it releases heat To do.

The possibility of heating the heterogeneously catalyzed propane dehydrogenation charge gas mixture in reaction zone A to the required reaction temperature is achieved by converting a portion of the propane and / or H ₂ contained therein to the charge gas mixture. Burning with the molecular oxygen contained therein upon entry into reaction zone A (eg with a suitable specifically acting combustion catalyst, eg by simple introduction and / or passage), and so on The combustion heat released in this way also causes heating to the reaction temperature desired for the dehydrogenation (such a method is advantageous in particular in a vortex bed reactor). .

  Correspondingly, the reaction zone A of the process according to the invention is used as described in DE-A 102004032129 and DE-A 10200501339, but as a charge gas mixture of reaction zone A, steam, fresh It can be formed with the difference that a mixture comprising propane and circulating gas I is used. In this case, the reaction zone A is realized as a (preferably adiabatic) bed reactor, in which the catalyst bed (preferably a fixed bed) is arranged radially or axially continuously. The number of catalyst bed plates in such a bed plate reactor is advantageously three. In this case, it is advantageous to carry out the heterogeneously catalyzed partial propane dehydrogenation autothermally. For this purpose, molecular oxygen is added to the charged gas mixture in reaction zone A to a limited extent after the first pass catalyst (solid) bed and between the catalyst (solid) bed following the first catalyst (solid) bed in the flow direction. Or the mixture with the inert gas containing it is supplied. Thus, typically the hydrogen produced during the course of heterogeneously catalyzed propane dehydrogenation (and possibly a very limited range of propane and / or propylene), catalyzed by the dehydrogenation catalyst itself, is limited. Combustion is caused and the exothermic heat change actually maintains the dehydrogenation temperature.

  In this case, the partially heterogeneously catalyzed dehydrogenation of propane is actually about 20 moles of propane introduced into the reactor on three catalyst bed plates per one reactor pass. % Is advantageously distributed (but obviously, in the process according to the invention, it may be 30 mol%, or 40 mol%, or 50 mol%). Here, the selectivity achieved for the production of propylene is typically 90 mol%. The highest conversion contribution of a single floorboard moves from front to back in the flow direction with increasing operating time. The catalyst loading is typically regenerated before the third floor plate provides the highest conversion contribution in the flow direction. Regeneration is advantageously performed when coking of the entire floorboard is achieved to an ideal degree.

  Due to the above heterogeneously catalyzed partial dehydrogenation of propane, the loading of the total amount of catalyst (total over the whole bed) including propane and propylene is ≧ 500 Nl / l · h and ≦ 20000 Nl / l · h. Some cases (typical values are 1500 Nl / l · h to 2500 Nl / l · h) are very generally advantageous. In this case, the maximum reaction temperature in the individual catalyst fixed bed is advantageously kept at 500 ° C. to 600 ° C. (or ˜650 ° C.). The reaction zone A charge gas mixture is dehydrated from fresh propane and partial oxidation refluxed from partial oxidation to dehydrogenation during the heterogeneously catalyzed partial propane dehydrogenation in the bed reactor. In order to condition the satisfactory duration of the elementary catalyst bed, it is advantageous to include only the circulating gas I which usually contains a sufficient amount of water vapor.

  The disadvantage of the above process is that almost all catalysts that catalyze the dehydrogenation of propane also catalyze the combustion of propane and propylene with molecular oxygen (complete oxidation of propane and propylene to carbon oxide and steam), and As described above, molecular oxygen is usually contained in the circulation gas I from the partial oxidation refluxed in the heterogeneously catalyzed partial dehydrogenation of propane.

  On the other hand, according to DE-A 10211275, the product gas discharged from the dehydrogenation zone is divided into two partial quantities of the same composition, and one of the two partial quantities is supplied as partial gas A to partial oxidation, while the other This can be countered by refluxing a partial amount as a component of reaction gas A to dehydrogenation. In this case, the molecular hydrogen contained in the circulating gas II derived from this dehydrogenation itself is a molecule which likewise contains propane and possibly propylene contained in the charge gas mixture of the reaction zone A. Should be protected from oxygen. This protection is usually based on the heterogeneously catalyzed combustion of molecular hydrogen to water with the same catalyst being kinetically advantageous compared to complete combustion of propane and / or propylene.

  According to the editorial of DE-A 10211275, dehydrogenation circulation gas induction is preferably realized by the principle of the injection pump (this is also referred to as the loop method). Furthermore, the possibility of adding additional molecular hydrogen to the charge gas mixture in reaction zone A as an additional oxidation protection is claimed here. From DE-A 102005049699, corresponding information is obtained on the requirement to distribute molecular hydrogen with a constant class supply during propulsion injection of the injection pump.

  The circulation of the circulating gas I containing molecular oxygen derived from heterogeneously catalyzed partial oxidation into the charge gas mixture of heterogeneously catalyzed partial propane dehydrogenation is carried out according to the editorial of DE-A102004032129 and DE-A102005013039 It is advantageous not to. Rather, it is advantageous to carry out this reflux into the reaction gas A in the reaction zone A for the first time after a certain dehydrogenation conversion. In this context, DE-A 102004032129 proposes to add additional molecular hydrogen from the outside, preferably additionally to the charge gas mixture in reaction zone A, prior to this reflux. Furthermore, DE-A102004032129 also discloses a dehydrogenation loop method. In this case, the propulsion injection is a circulating gas I that is diverted to the process according to the invention and is exclusively refluxed from partial oxidation to dehydrogenation.

  According to the editorial of Example II of DE-A 102005009885, the loop method is preferably applied to the reaction zone A according to the invention, wherein the charge gas mixture is derived from circulating gas I, from fresh propane, from external molecular hydrogen. It is composed of a minimum amount of external water vapor and from circulating gas II which is refluxed to the dehydrogenation itself (the external water vapor can also be discarded). As propulsion radiation, a mixture containing fresh propane, external molecular hydrogen, circulating gas I from partial oxidation and external water vapor is used. DE-A102005049699 is advantageously recommended with regard to the order of distribution that should be advantageously adhered to when generating propulsion radiation.

Advantageously according to the invention, the circulating gas I typically contains not only molecular oxygen and water vapor but also molecular hydrogen, carbon monoxide and carbon dioxide. That is, it typically also contains CO. The circulation gas I preferably contains 5 to 15% by volume of CO ₂ according to the invention.

  Advantageously according to the invention, the product gas A discharged from the reaction zone A has the following contents in the process according to the invention.

Propane 25-60% by volume
Propene 8-25% by volume
H ₂ 0.04~25% by volume, often to 15% by volume
H ₂ O 2 to 25 volume% and CO _2> 0 to 30 volume%, and often 15% by volume.

  The temperature of the product gas A is typically 400-700 ° C, preferably 450-650 ° C.

  The pressure of the product gas A leaving the reaction zone A is preferably 2 to 4 bar according to the invention. However, as mentioned above, it may be up to 20 bar.

  By the way, according to the invention, the product gas A is used together with the reaction gas B to charge at least one oxidation reactor in the reaction zone B without further separation of subcomponents.

In an advantageous manner, it is sufficient to add to the product gas A the amount of molecular oxygen necessary to achieve the set target in the reaction zone B. This addition comprises in principle one or more gases (for example N ₂ , H ₂ O, noble gases, CO ₂ ) as pure oxygen or chemically inert in molecular oxygen and reaction zone B. It can be done as a mixture (eg air). It is advantageously according to the invention that a gas different from molecular oxygen is not more than 30% by volume, preferably not more than 25% by volume, preferably not more than 20% by volume, particularly preferably not more than 15% by volume, better still not more than 10% by volume and particularly advantageous. Is performed as a gas containing 5% by volume or less or 2% by volume or less. For this, the use of pure oxygen is very particularly advantageous. Typically, the molecular oxygen feed is allocated in the reaction zone B charge gas (in the reaction gas B-starting mixture) such that the molar ratio of molecular oxygen to propylene is ≧ 1 and ≦ 3. The Prior to the supply of molecular oxygen to product gas A, product gas A is preferably cooled according to the invention to a temperature in the range of 250-350 ° C., preferably in the range of 270-320 ° C. This cooling is advantageously achieved according to the present invention by means of an indirect heat exchanger (preferably in countercurrent operation; this statement is very generally applied to the indirect heat exchangers herein, unless stated otherwise). Is true). In this case, the reaction gas A-starting mixture in reaction zone A is preferably used as coolant, which is set in this way at the same time as the reaction temperature required in reaction zone A (but in principle). In addition, cooling can also be performed by releasing pressure in the expansion turbine as long as the starting pressure is sufficiently high in this respect, as described above).

  In accordance with the present invention, the supply of molecular oxygen-containing gas to the product gas A, which has been advantageously cooled as previously described, is preferably propelled by a radiant pump comprising a propulsion nozzle, a mixing zone, a diffuser and a suction tube. It can be carried out by operating with the product gas A as radiation, in this case the transport path of the propulsion radiation and the suction action of the suction tube which are released through the propulsion nozzle and through the mixing zone and diffuser The direction of supply of oxygen-containing gas is indicated, and at this time, molecular oxygen-containing gas is sucked by the reduced pressure generated in the suction pipe and transported through the diffuser through the mixing zone by simultaneous mixing with propulsion radiation. The resulting reaction gas B-starting mixture is discharged at the inlet of the second reaction zone B or at the inlet of at least one oxidation reactor in the second reaction zone B. The variant described above is particularly applicable when the product gas A has a pressure of 2-5 bar or higher, or -4 bar.

  Of course, however, the molecular oxygen-containing gas can also be mixed with the product gas A in a conventional manner to obtain a charge gas mixture in the reaction zone B (reaction gas B-starting mixture). Advantageously, according to the invention, a mechanical separation operation according to DE-A 10316039 may be connected between reaction zone A and reaction zone B.

  In a manner known per se, the heterogeneously catalyzed gas phase-partial oxidation of propylene to acrylic acid using molecular oxygen proceeds in principle in two successive stages along the reaction coordinates, of which the first stage is Acrolein and the second stage is led from acrolein to acrylic acid.

  This reaction course in two temporally continuous stages is a method known per se, in which the process according to the invention is interrupted in reaction zone B at the acrolein stage (mainly the stage of acrolein formation) and at this stage The possibility of carrying out product separation or continuing the process according to the invention mainly until acrylic acid production, at which time the target product separation is carried out is revealed.

  If the process according to the invention is mainly led to acrylic acid production, it is advantageous according to the invention to carry out the process in two stages, i.e. in two successive oxidation stages. The catalyst solid bed to be used in each of the three oxidation stages and preferably other reaction conditions, for example the temperature of the catalyst solid bed, are optimally adapted.

  A multi-metal oxide containing elements Mo, Fe and Bi, which is suitable as an active substance for the catalyst in the first oxidation stage (propylene → acrolein), catalyzes the second oxidation stage (acrolein → acrylic acid) to a certain extent. While it is certainly possible, for the second stage, a catalyst is usually advantageous in which the active substance is a multimetallic oxide containing at least one of the elements Mo and V.

  Accordingly, the heterogeneously catalyzed portion of propylene in a catalyst solid bed, in which the catalyst is to be carried out in reaction zone B according to the invention, the catalyst having a multimetallic oxide containing at least one element Mo, Fe and Bi as active substances The oxidation process is particularly suitable as a one-step process for the production of acrolein (and optionally acrylic acid) or as the first reaction step for the two-step production of acrylic acid.

  In this case, using the reaction gas B produced according to the present invention, a one-stage heterogeneously catalyzed partial oxidation from propylene to acrolein and optionally acrylic acid, or a two-stage heterogeneous catalyst from propylene to acrylic acid. The realization of oxidised partial oxidation can be achieved individually by means of EP-A 700714 (first reaction stage; as described therein, but also in the corresponding countercurrent process of the salt bath and the reaction gas starting mixture via a bundle tube reactor. ), EP-A 700893 (second reaction stage; as described therein, but also in the corresponding countercurrent process), WO 04/085369 (in particular, this description corresponds as a component of the present description) (As a two-step method), WO 04/85363, DE-A10313212 (first reaction step), EP-A 1159248 (as a two-step method), EP-A 1159246 ( Two reaction steps), EP-A 1159247 (as a two step method), DE-A 1994248 (as a two step method), DE-A 10101695 (one step or two steps), WO04 / 085368 (as a two step method), DE 102004021764 (two steps) ), WO 04/085362 (first reaction stage), WO 04/085370 (second reaction stage), WO 04/085365 (second reaction stage), WO 04/085367 (two stages), EP-A990636, EP-A1007007 and EP- It can be carried out as described in the specification of A1106598.

  This is especially true for all the examples contained herein. These can be carried out as described herein, with the difference that the reaction gas B produced according to the invention is used as the reaction gas starting mixture in the first reaction stage (propylene to acrolein). Accompany. The relevant other parameters are treated in the same way as in the examples of the above specification (in particular the relevant catalyst solid bed and the catalyst solid bed reactant-load). If the above-mentioned prior art embodiment is treated in two stages and secondary oxygen (secondary air which is less advantageous according to the invention) is supplied between the two reaction stages, this is a corresponding method. In that amount, the molar ratio of molecular oxygen to acrolein in the charge gas mixture of the second reaction stage is adapted to correspond to that in the examples of the specification.

  Advantageously according to the invention, the amount of oxygen in the reaction zone B is distributed such that the product gas B still contains unreacted molecular oxygen (preferably ≧ 0.5 to 6% by volume, preferably 1 to 5%. Volume%, preferably 2 to 4 volume%). In the case of the two-stage process, the above applies for each of the two oxidation stages.

  Particularly suitable multi-metal oxide catalysts for each oxidation stage have been previously described variously and are well known to those skilled in the art. For example, EP-A 253409 refers to the corresponding US-patent on page 5.

  Suitable catalysts for each oxidation stage are also apparent from DE-A 4431957, DE-A 102004025445 and DE-A 4431949. This applies in particular to that of general formula I in the two above-mentioned older specifications. Particularly advantageous catalysts for each oxidation stage are evident from the specifications of DE-A 10325488, DE-A 10325487, DE-A 10353594, DE-A 10344149, DE-A 10351269, DE-A 10350812 and DE-A 10350822.

  In the reaction stage according to the invention of heterogeneously catalyzed gas phase partial oxidation of propylene to acrolein or acrylic acid or mixtures thereof, in principle, all multi-metal oxide materials containing Mo, Bi and Fe as active substances This is the case.

  These are in particular the multi-metal oxide active substances of the general formula I of DE-A 19955176, the multi-metal oxide active substances of the general formula I of DE-A 1948523, Listed in oxide active materials, multi-metal oxide active materials of general formula I, II and III of DE-A 1948248 and multi-metal oxide active materials of general formula I, II and III of DE-A 199556618 and EP-A 700714 It is a multimetal oxide active substance.

Further, this reaction step includes Research Disclosure No. 497012 with 29.08.2005, DE-A10046957, DE-A10063162, DE-C3338380, DE-A19902562, EP-A15565, DE-C2380765, EP-A807465, EP-A279374, DE-A3300044, EP-A575897, US-A4438217, DE-A19855913, WO98 / 24746, DE-A19746210 (of general formula II), JP-A91 / 294239, EP-A293224 and EP -Multi-metal oxide catalysts containing Mo, Bi and Fe, which are published in the specification of A700714, are preferred. This is particularly suitable for the illustrated embodiments herein, of which EP-A 15565, EP-A 575897, DE-A 19746210 and DE-A 1855913 are particularly advantageous. In this connection, the catalyst according to Example 1c from EP-A 15565 and the corresponding process should be prepared, but the active substance has the composition Mo ₁₂ Ni _6.5 Zn ₂ Fe ₂ Bi ₁ P _0.0065 K _0. Particular mention may be made of a catalyst having ₀₆ O _x .10SiO ₂ . Further, examples of serial number 3 from DE-A19855913 _{(stoichiometry: Mo 12 Co 7 Fe 3 Bi} 0.6 K 0.08 Si 1.6 O x) the shape 5mmx3mmx2mm (external diameter x height x internal diameter) Mention may be made, as the hollow tubular complete catalyst, and of the multimetal oxide II complete catalyst according to Example 1 of DE-A 19746210. Furthermore, mention may be made of the multimetal oxide catalysts of US-A 4,438,217. The latter is particularly true when the hollow cylinder has the shape 5.5 mm x 3 mm x 3.5 mm, or 5 mm x 2 mm x 2 mm, 5 mm x 3 mm x 2 mm, or 6 mm x 3 mm x 3 mm, or 7 mm x 3 mm x 4 mm (each outer diameter x height x inner diameter). In this regard, further possible catalyst shapes are cords (eg length 7.7 mm and diameter 7 mm; or length 6.4 mm and diameter 5.7 mm).

A number of multimetal oxide actives suitable for the propylene to acrolein and optionally acrylic acid stage can include the general formula IV:
^{_{Mo 12 Bi a Fe b X 1}} c X 2 d X 3 e X 4 f O n (IV)
[Where the variables have the following meanings:
X ¹ = nickel and / or cobalt,
X ² = thallium, alkali metal and / or alkaline earth metal,
X ³ = zinc, phosphorus, arsenic, boron, antimony, tin, cerium, lead and / or tungsten,
X ⁴ = silicon, aluminum, titanium and / or zirconium,
a = 0.5-5,
b = 0.01-5, preferably 2-4,
c = 0 to 10, preferably 3 to 10,
d = 0-2, preferably 0.02-2,
e = 0-8, preferably 0-5,
f = 0 to 10 and n = number determined by the valence and frequency of an element different from oxygen in IV].

  These are obtained in a manner known per se (see, for example, DE-A 4023239) and are usually shaped with materials, spheres, rings or cylinders, or coated with shell-type catalysts, ie active materials. Used in the form of a preformed inert carrier. Of course, however, they can also be applied as catalysts in powder form (for example in a vortex bed reactor).

The active substances of the general formula IV are dried in a simple manner, in principle, from a suitable source of their elemental constituents, as well mixed as possible, preferably in powder form, commensurate with their stoichiometry. It can be produced by forming a mixture and calcining it at a temperature of 350-650 ° C. Calcination can be under an inert gas or in an oxidizing atmosphere such as air (a mixture comprising an inert gas and oxygen) and a reducing atmosphere (eg an inert gas, NH ₃ , CO and / or H _2. Can also be carried out. The calcination time can be from a few minutes to a few hours and typically decreases with temperature. The source of the elemental component of the multi-metal oxide active substance IV corresponds to a compound in which an oxide and / or a compound that can be converted into an oxide in the presence of at least oxygen by heating is important.

In addition to oxides, such starting compounds include in particular halogenides, nitrates, formates, oxalates, citrates, acetates, carbonates, amine complexes, ammonium salts and / or hydroxides. Applicable (at the latest, upon subsequent calcination, can be decomposed and / or decomposed into gaseous leaking compounds such as NH ₄ OH, (NH ₄ ) ₂ CO ₃ , NH ₄ NO ₃ , NH ₄ CHO ₂ , CH ₃ COOH, NH ₄ CH ₃ CO ₂ and / or ammonium oxalate can additionally be admixed into the well-mixed dry mixture).

  Thorough mixing of the starting compounds for the production of the multimetal oxide active substance IV can be carried out in dry or wet form. When carried out in dry form, the starting compound is preferably used as a fine powder and calcined after mixing and optionally pressing. However, it is advantageous that sufficient mixing takes place in wet form. In this case, the starting compounds are usually mixed with one another in the form of an aqueous solution and / or suspension. Particularly well-mixed dry mixtures are obtained in the case of the above-mentioned mixing methods exclusively when they are derived from a source existing in dissolved form of the elemental components. It is advantageous to use water as the solvent. The resulting aqueous material is subsequently dried, wherein the drying process is preferably carried out by spray drying an aqueous mixture having an exhaust temperature of 100 to 150 ° C.

  The multimetal oxide active material of the general formula IV can be used in the stage “propylene → acrolein (and optionally acrylic acid)”, either in powder form or formed into a catalyst of a certain shape, Molding can take place before or after the final calcination. For example, producing a complete catalyst from a powdered form of the active substance or its uncalcined and / or partially calcined precursors by compression into the desired catalyst shape (eg by tableting, extrusion or squeezing) Optionally adding auxiliary agents such as graphite or stearic acid as lubricants and / or forming aids and reinforcing fibers such as glass, asbestos, silicon carbide or potassium titanate. be able to. However, instead of graphite, hexagonal boron nitride, as recommended by DE-A 102005037678, can be used as a molding aid. A suitable complete catalyst shape is, for example, a solid cylinder or a hollow cylinder having an outer diameter and a length of 2 to 10 mm. In the case of a hollow cylinder, a wall thickness of 1 to 3 mm is advantageous. Of course, the complete catalyst can also have the shape of a sphere, in which case the sphere diameter can be 2 to 10 mm. Particularly in the case of a complete catalyst, a particularly advantageous hollow cylinder shape is 5 mm × 3 mm × 2 mm (outer diameter × length × inner diameter).

  Of course, the shaping of the powdered active substance or its powdered, still uncalcined and / or partially calcined precursor can also be carried out by application onto a preformed inert catalyst support. The coating of the support for the production of the shell-type catalyst is usually carried out in a suitable rotatable container, as is known, for example, from DE-A 2909671, EP-A 293859 or EP-A 714700. For the coating of the carrier, the powder substance to be applied is wetted and, after application, dried again, for example with hot air. The layer thickness of the powder material applied on the support is preferably chosen to be in the range from 10 to 1000 μm, preferably in the range from 50 to 500 μm and particularly preferably in the range from 150 to 250 μm.

  In this case, customary porous or nonporous aluminum oxide, silicon dioxide, thorium dioxide, zirconium dioxide, silicon carbide or silicates such as magnesium silicate or aluminum silicate can be used. These are usually practically inactive with respect to the target reaction based on the process according to the invention. The support may be shaped regularly or irregularly, with preference given to regularly shaped supports having a clearly shaped rough surface, for example spheres or hollow cylinders. Preference is given to using actually non-porous rough surface spherical carriers made of steatite whose diameter is 1-10 mm or -8 mm, preferably 4-5 mm. However, it is also preferable to use a cylindrical member having a length of 2 to 10 mm and an outer diameter of 4 to 10 mm as the carrier. In the case of an annular material suitable according to the invention as a carrier, the wall thickness is usually 1 to 4 mm. The annular carrier to be used advantageously according to the invention has a length of 2 to 6 mm, an outer diameter of 4 to 8 mm and a wall thickness of 1 to 2 mm. According to the present invention, an annular member having a shape of 7 mm × 3 mm × 4 mm (outer diameter × length × inner diameter) is also suitable as the carrier. The fineness of the catalytically active oxide material to be applied on the surface of the support is naturally adapted to the desired shell thickness (see EP-A 714700).

Furthermore, the multi-metal oxide active material to be used in the stage from propylene to acrolein (and possibly acrylic acid) is expanded in three dimensions and its local environment based on its composition differs from its local environment. It has a range of chemical composition Y ¹ _{a ′} Y ² _{b ′} O _{x ′} to be segmented, and its maximum diameter (the longest bond line through the center of gravity of the two points on the surface (interface) of the range) 1 nm Substances of general formula V that are ˜100 μm, often 10 nm to 500 nm or 1 μm to 50 or 25 μm:
[Y ¹ _{a ′} Y ² _{b ′} O _{x ′} ] _p [Y ³ _{c ′} Y ⁴ _{d ′} Y ⁵ _{e ′} Y ⁶ _{f ′} Y ⁷ _{g ′} Y ² _{h ′} O _{y ′} ] _q (V)
[Where the variables have the following meanings:
Y ¹ = bismuth alone or bismuth and at least one element tellurium, antimony, tin and copper,
Y ² = molybdenum, or tungsten, or molybdenum and tungsten,
Y ³ = alkali metal, thallium and / or samarium,
Y ⁴ = alkaline earth metal, nickel, cobalt, copper, manganese, zinc, tin, cadmium and / or mercury,
Y ⁵ = iron or iron and at least one element chromium and cerium,
Y ⁶ = phosphorus, arsenic, boron and / or antimony,
Y ⁷ = rare earth metal, titanium, zirconium, niobium, tantalum, rhenium, ruthenium, rhodium, silver, gold, aluminum, gallium, indium, silicon, germanium, lead, thorium and / or uranium,
a ′ = 0.01-8,
b ′ = 0.1-30,
c ′ = 0 to 4,
d ′ = 0 to 20,
e ′> 0-20,
f ′ = 0 to 6,
g ′ = 0-15,
h ′ = 8-16,
x ′, Y ′ = number determined by the valence and frequency of an element different from oxygen in V, and p, q = number whose ratio p / q is 0.1-10.

Particularly advantageous multimetal oxide materials V according to the invention are those in which Y ¹ is only bismuth.

Among them, again those corresponding to the general formula VI are advantageous:
^{_{[Bi a "Z 2 b"}} O x "] p" [Z 2 12 Z 3 c "Z 4 d" Fe e "Z 5 f" Z 6 g "Z 7 h" O y "] q" (VI)
[Where the variables have the following meanings:
Z ² = molybdenum, or tungsten, or molybdenum and tungsten,
Z ³ = nickel and / or cobalt,
Z ⁴ = thallium, alkali metal and / or alkaline earth metal,
Z ⁵ = phosphorus, arsenic, boron, antimony, tin, cerium and / or lead,
Z ⁶ = silicon, aluminum, titanium and / or zirconium,
Z ⁷ = copper, silver and / or gold,
a "= 0.1-1;
b "= 0.2-2,
c "= 3-10,
d "= 0.02-2,
e "= 0.01-5, preferably 0.1-3,
f "= 0-5,
g "= 0-10,
h "= 0-1
x ″, y ″ = number determined by the valence and frequency of an element different from oxygen in VI,
p ″, q ″ = number whose ratio p ″ / q ″ is 0.1-5, preferably 0.5-2]: Z ² _b ″ = (tungsten) _b ″ and Z ² ₁₂ = The substance VI which is (molybdenum) ₁₂ is particularly advantageous.

Furthermore, all the components [Y ¹ _{a ′ of the} multimetal oxide material V (polymetal oxide material VI) preferred according to the invention in the multimetal oxide material V (polymetal oxide material VI) suitable according to the invention _{^{Y 2 b 'O x']}} p at least 25 mole% of _{^{([Bi a "Z 2 b}} " O x "] p") ( preferably at least 50 mol%, and particularly preferably 100 mol%) is, the three-dimensional enlarged, of its chemical and local environment different from its local environment based on its chemical composition is divided composition _{^{Y 1 a 'Y 2 b'}} O x '[Bi a "Z 2 b" O x "] It is advantageous if it exists in the form of a range and its maximum diameter is in the range of 1 nm to 100 μm.

  For molding with respect to the multi-metal oxide material V-catalyst, the statements in the multi-metal oxide material IV-catalyst apply.

  The preparation of multimetal oxide materials V-active materials is described, for example, in EP-A 575897 and DE-A 1985913.

  The inert support material recommended above applies in particular as an inert material for the dilution and / or limitation of the corresponding catalyst solid bed or as a pre-fill bed for its protection and / or gas mixture heating. To do.

  For the second stage (second reaction stage), heterogeneously catalyzed gas phase-partial oxidation of chlorein to acrylic acid, as described above, in principle, all multimetal oxide materials containing Mo and V, For example, DE-A corresponds to this as the required catalytic active.

A number of them, such as that of DE-A 198 15281, can be encompassed by the general formula VII:
^{_{Mo 12 V a X 1 b X}} 2 c X 3 d X 4 e X 5 f X 6 g O n (VII)
[Where the variables have the following meanings:
X ¹ = W, Nb, Ta, Cr and / or Ce,
X ² = Cu, Ni, Co, Fe, Mn and / or Zn,
X ³ = Sb and / or Bi,
X ⁴ = one or more alkali metals,
X ⁵ = one or more alkaline earth metals,
X ⁶ = Si, Al, Ti and / or Zr,
a = 1 to 6,
b = 0.2-4,
c = 0.5-18,
d = 0 to 40,
e = 0-2,
f = 0-4,
g = 0 to 40 and n = number determined by the valence and frequency of an element different from oxygen in VII].

Preferred embodiments according to the invention within the active multimetal oxide VII are those encompassed by the following meanings of the variables in the general formula VII:
X ¹ = W, Nb and / or Cr,
X ² = Cu, Ni, Co and / or Fe,
X ³ = Sb,
X ⁴ = Na and / or K,
X ⁵ = Ca, Sr and / or Ba,
X ⁶ = Si, Al and / or Ti,
a = 1.5-5,
b = 0.5-2,
c = 0.5-3,
d = 0-2,
e = 0-0.2,
f = 0-1 and n = a number determined by the valence and frequency of an element different from oxygen in VII].

However, a very particularly advantageous multimetal oxide VII according to the invention is that of the general formula VIII:
^{_{Mo 12 V a 'Y 1 b}} ' Y 2 c 'Y 5 f' Y 6 g 'O n' (VIII)
[Where:
Y ¹ = W and / or Nb,
Y ² = Cu and / or Ni,
Y ⁵ = Ca and / or Sr,
Y ⁶ = Si and / or Al,
a ′ = 2 to 4,
b ′ = 1 to 1,5,
c ′ = 1 to 3,
f ′ = 0 to 0.5,
g ′ = 0-8 and n ′ = number determined by the valence and frequency of an element different from oxygen in VIII].

  Multimetal oxide active substances (VII) suitable according to the invention are obtained in a manner known per se, for example as disclosed in DE-A 4335973 or EP-A 714700.

In principle, multi-metal oxide active substances suitable for the stage “acrolein → acrylic acid”, in particular those of the general formula VII, are advantageously mixed in a simple manner as far as possible from suitable sources of their elemental components. Can be produced by producing a finely divided dry substance commensurate with its stoichiometry and calcining it at a temperature of 350-600 ° C. Calcination can be performed in a reducing atmosphere (eg, inert gas and reducing gas, eg, H ₂ ), either under an inert gas or in an oxidizing atmosphere, eg, air (a mixture comprising an inert gas and oxygen). , NH ₃ , CO, methane and / or acrolein or the gas having the reducing action itself). The calcination time can be from a few minutes to a few hours and typically decreases with temperature. The source of the elemental component of the multimetal oxide active substance VII corresponds to a compound in which an oxide and / or at least oxygen and a compound that can be converted into an oxide by heating are important.

Thorough mixing of the starting compounds for the production of the multimetal oxide material VII can be carried out in dry or wet form. When carried out in dry form, the starting compound is preferably used as a fine powder and calcined after mixing and optionally pressing. However, it is advantageous that sufficient mixing takes place in wet form.
In this case, the starting compounds are usually mixed with one another in the form of an aqueous solution and / or suspension. A particularly well-mixed dry mixture is obtained when the above-mentioned mixing method is derived from a source that exists exclusively in dissolved form of the elemental components. It is advantageous to use water as the solvent. The resulting aqueous material is subsequently dried, wherein the drying process is preferably carried out by spray-drying the aqueous mixture with a discharge temperature of 100-150 ° C.
The resulting multimetal oxide material, in particular that of the general formula VII, is used for acrolein oxidation, either in powder form (for example in a vortex bed reactor) or formed into a catalyst of a certain shape. In this case, the shaping can take place before or after calcination. For example, a complete catalyst can be produced from a powder form of the active substance or its uncalcined precursor by compression into the desired catalyst form (eg by tableting, extrusion or squeezing), in which case According to the invention, auxiliary agents such as graphite or stearic acid can be added as lubricants and / or forming aids and reinforcing fibers such as glass, asbestos, silicon carbide or potassium titanate can be added. A suitable complete catalyst shape is, for example, a solid cylinder or a hollow cylinder having an outer diameter and a length of 2 to 10 mm. In the case of a hollow cylinder, a wall thickness of 1 to 3 mm is advantageous. Of course, the complete catalyst can also have the shape of a sphere, where the sphere diameter can be 2 to 10 mm (for example 8.2 mm or 5.1 mm).
Of course, the shaping of the powdered active substance or its powdery, still uncalcined precursor can also be carried out by application onto a preformed inert catalyst support. The coating of the support for the production of a shell-type catalyst is usually carried out in a suitable rotatable container, as is known, for example, from DE-A 2909671, EP-A 293859 or EP-A 714700.
For the coating of the carrier, it is advantageous to wet the powder material to be applied and to dry it again after application, for example with hot air. The layer thickness of the powder material applied on the support is preferably chosen to be in the range from 10 to 1000 μm, preferably in the range from 50 to 500 μm and particularly preferably in the range from 150 to 250 μm.

  In this case, customary porous or nonporous aluminum oxide, silicon dioxide, thorium dioxide, zirconium dioxide, silicon carbide or silicates such as magnesium silicate or aluminum silicate can be used. The support may be regularly or irregularly shaped, in which case a regularly shaped support with a clearly shaped rough surface, for example a sphere or hollow cylinder with a coarse layer. Is advantageous. Preference is given to using actually non-porous rough surface spherical carriers made of steatite whose diameter is 1-10 mm or -8 mm, preferably 4-5 mm. That is, a suitable sphere shape may have a diameter of 8.2 mm or 5.1 mm. However, it is also preferable to use a cylindrical member having a length of 2 to 10 mm and an outer diameter of 4 to 10 mm as the carrier. In the case of an annular material as the carrier, the wall thickness is usually 1 to 4 mm. The annular carrier to be used advantageously has a length of 2 to 6 mm, an outer diameter of 4 to 8 mm and a wall thickness of 1 to 2 mm. In particular, an annular member having a shape of 7 mm × 3 mm × 4 mm (outer diameter × length × inner diameter) is also suitable as the carrier. The fineness of the catalytically active oxide material to be applied on the surface of the support is naturally adapted to the desired shell thickness (see EP-A 714700).

Furthermore, suitable multimetal oxide actives to be used in the stage “acrolein → acrylic acid” are those of the general formula IX:
[D] _p [E] _q (IX)
[Where the variables have the following meanings:
^{_{D = Mo 12 V a "Z}} 1 b" Z 2 c "Z 3 d" Z 4 e "Z 5 f" Z 6 g "O x",
E = Z ⁷ ₁₂ Cu _h " _Hi " _Oy ",
Z ¹ = W, Nb, Ta, Cr and / or Ce,
Z ² = Cu, Ni, Co, Fe, Mn and / or Zn,
Z ³ = Sb and / or Bi,
Z ⁴ = Li, Na, K, Rb, Cs and / or H,
Z ⁵ = Mg, Ca, Sr and / or Ba,
Z ⁶ = Si, Al, Ti and / or Zr,
Z ⁷ = Mo, W, V, Nb and / or Ta, preferably Mo and / or W,
a "= 1-8,
b "= 0.2-5,
c "= 0-23,
d "= 0 to 50,
e "= 0-2,
f "= 0-5,
g "= 0 to 50,
h "= 4-30,
i "= 0 to 20 and x", y "= number determined by the valence and frequency of an element different from oxygen in IX, and p, q = the ratio p / q is 160: 1 to 1: 1. , A number different from zero]
This is a multi-metal oxide material E:
Z ⁷ ₁₂ Cu _h " _Hi " _Oy ", (E)
Is separately prepared in fine powder form (starting material 1) and subsequently the solid starting material 1 prepared previously is converted to the stoichiometric formula D:
^{_{Mo 12 V a "Z 1 b}} " Z 2 c "Z 3 d" Z 4 e "Z 5 f" Z 6 g "(D)
An aqueous solution, an aqueous suspension or a finely powdered dry mixture of the elements Mo, V, Z ¹ , Z ² , Z ³ , Z ⁴ , Z ⁵ , Z ⁶ , containing the elements (starting material 2) During which the desired amount ratio p: q is admixed, in which case the resulting aqueous mixture is dried and the resulting dry precursor is subjected to a temperature of 250-600 ° C. before or after its drying. And calcining to the desired catalyst shape.

  Preference is given to the multimetal oxide material IX, wherein the premixing of the solid starting material 1 into the aqueous starting material 2 is carried out at a temperature <70 ° C. Details of the preparation of the multi-metal oxide material VI-catalyst are contained, for example, in EP-A 668104, DE-A 197 36 105, DE-A 10046928, DE-A 19740493 and DE-A 19528646. For molding with respect to the multimetal oxide material IX-catalyst, the statement of the multimetal oxide material VII-catalyst applies.

  A multimetal oxide catalyst suitable for the stage “acrolein → acrylic acid” in an excellent manner is that of DE-A 19815281, in particular with a multimetal oxide active material of the general formula I herein.

  It is advantageous to use a fully catalytic cyclic for the propylene to acrolein stage and a shell-type catalytic cyclic for the acrolein to acrylic acid stage.

  The partial oxidation of the process according to the invention from propylene to acrolein (and optionally acrylic acid) can be carried out with the above-mentioned catalyst, for example, a single zone-multiple, such as that described in DE-A 4431957. It can be carried out in a contact tube-solid bed reactor. In this case, the reaction gas mixture and the heat medium (heat exchange agent) can be introduced in a cocurrent or countercurrent manner as viewed from above the reactor.

  The reaction pressure is typically in the range of 1 to 3 bar, and the total space load of the catalyst solid bed with reaction gas B is typically 1500 to 4000 or 6000 Nl / l · h or more. The propylene loading (the propylene loading of the catalyst fixed bed) is typically 90 to 200 Nl / l · h or ˜300 Nl / l · h or more. According to the invention, propylene loadings of 135 Nl / l · h or more or ≧ 140 Nl / l · h, or ≧ 150 Nl / l · h, or ≧ 160 Nl / l · h are particularly advantageous, because according to the invention This is because the reaction gas starting mixture in reaction zone B conditions advantageous hot spot properties due to the presence of unreacted propane and molecular hydrogen (all the above are independent of the particular choice of the fixed bed reactor). Is true).

Advantageously, the charge gas mixture flows through a single zone-multi-contact tube-solid bed reactor from above. As a heat exchanger, it preferably contains 60% by weight of potassium nitrate (KNO ₃ ) and 40% by weight of sodium nitrite (NaNO ₂ ), or 53% by weight of potassium nitrate (KNO ₃ ), 40% by weight of sodium nitrite (NaNO ₂ ) And a salt melt containing 7% by weight of sodium nitrate (NaO ₃ ) is advantageous.

  As mentioned above, the salt melt and the reaction gas mixture can be conducted in cocurrent or countercurrent with the reactor viewed from above. It is advantageous for the salt melt itself to circulate around the contact tube.

When flowing through the contact tube from top to bottom, it is advantageous to load the catalyst into the contact tube from bottom to top as follows (for flow from bottom to top, the loading sequence is an advantageous method). Is reversed):
First, the length of the contact tube is 40 to 80 or -60%, the catalyst alone or a mixture containing the catalyst and the inert substance (the latter being up to 30% by weight or up to 20% by weight, based on the mixture) Mass ratio) (Category C);
Subsequent to the length of 20-50 or -40% of the total tube length, the catalyst alone or a mixture containing the catalyst and an inert substance (where the latter is a mass proportion up to 40% by weight with respect to the mixture) (Category B); and finally, a built-in bed containing an inert substance advantageously selected so that it is conditioned to a pressure loss as low as possible to 10-20% of the total pipe length (Category A).

  Section C is advantageously undiluted.

  Said loading variants are as catalysts according to Research Disclosure No. 497012 with 29.0.2005, according to DE-A 10046957 example 1 or according to DE-A 10046957 example 3 and inert This is advantageous when a steatite ring having a shape of 7 mm × 7 mm × 4 mm (outer diameter × height × inner diameter) is used as the substance. Regarding the salt bath temperature, the statements in DE-A 4431957 apply.

  The implementation of partial oxidation in the reaction zone B from propylene to acrolein (and optionally acrylic acid) can be carried out using the catalysts described above, but for example DE-A19910506, DE-A102005009885, DE-A102004032129, DE-A102005013039. And DE-A 102005009891 and DE-A 102005010111, for example, can also be carried out in a two-zone-multi-contact tube-fixed bed reactor. In the above two examples (and very generally in the process according to the invention), the propene conversion achieved in a single pass is typically a value of ≧ 90 mol% or ≧ 95 mol% and The selectivity is a value of ≧ 90 mol%. According to the invention, the partial oxidation according to the invention from propene to acrolein or acrylic acid or mixtures thereof is preferably as described in EP-A 1159244 and very particularly preferably in WO 04/085363 and in WO 04/085362. Done.

  EP-A 1159244, WO 04/085363 and WO 04/085362 are considered components of this specification.

  That is, the partial oxidation of propylene to acrolein (and optionally acrylic acid) to be carried out in reaction zone B is particularly advantageously carried out with an elevated propylene load in a catalyst solid bed having at least two temperature zones. obtain.

  In this connection, reference is made, for example, to EP-A 1159244 and WO 04/085362.

  The implementation of the second stage in the case of a two-stage partial oxidation of propylene to acrolein, i.e. the partial oxidation of acrolein to acrylic acid, is carried out using the catalyst described above, for example as described in DE-A 443 949. It can be carried out in a single zone-multi-contact tube-fixed bed reactor. In this case, the reaction gas mixture and the heat carrier can be guided in parallel flow as viewed from above the reactor. As a rule, the product gas mixture of the partial oxidation of propylene to acrolein according to the invention is essentially as such (optionally after intercooling (this is indirectly or directly, e.g. secondary oxygen- (Or by secondary air—which is less advantageous according to the invention —) — supplied))), ie without separation of the secondary components, leading to a second reaction stage, ie acrolein partial oxidation.

The molecular oxygen required for the second stage, acrolein partial oxidation, may (in this case) already be contained in the reaction gas B-starting mixture for the partial oxidation of propylene to acrolein. However, this may be partly or completely fed directly to the product gas mixture of the first reaction stage, ie propylene partial oxidation to acrolein (this is preferably as secondary air but preferably of pure oxygen In the form or in the form of a mixture comprising an inert gas and oxygen (preferably ≦ 50% by volume, or ≦ 40% by volume, or ≦ 30% by volume, or ≦ 20% by volume, or ≦ 10% by volume, or ≦ 5 Volume%, or ≦ 2 volume%). Irrespective of that, as described above, such a partial oxidation charge gas mixture from acrolein to acrylic acid (reaction gas starting mixture) advantageously has the following contents:
Acrolein 3-25% by volume
Molecular oxygen 5-65% by volume
Propane 6-70% by volume
Molecular hydrogen 0.3-20 volume% and water vapor 8-65 volume%.

Said reaction gas starting mixture advantageously has the following contents:
Acrolein 5-14% by volume
Molecular oxygen 6-25% by volume
Propane 6-60% by volume
Molecular hydrogen 2-18 vol% and water vapor 7-35 vol%.

It is very particularly advantageous that the reaction gas starting mixture has the following contents:
Acrolein 6-12% by volume
Molecular oxygen 8-20% by volume
Propane 20-55% by volume
6-15% by volume of molecular hydrogen and 11-26% by volume of water vapor,
In this case, the priority ranges are independent of each other, but it is advantageous to realize them simultaneously. The nitrogen content in the mixture is generally ≦ 20% by volume, preferably ≦ 15% by volume, particularly preferably ≦ 10% by volume and very particularly preferably ≦ 5% by volume. The molar ratio of O ₂ and acrolein contained in the charge gas mixture of the second oxidation stage O _2: acrolein, advantageously by the present invention, usually ≧ 0.5 and ≦ 2, often ≧ 1 and ≦ 1.5.

CO ₂ in the charge gas mixture of the second oxidation stage - content is 1 to 40, or 2-30, or 4-20, or often may be 6 to 18% by volume.

  As in the first reaction stage (propylene → acrolein), even in the second reaction stage (acrolein → acrylic acid), the reaction pressure is typically 1 to 3 bar and the total catalyst solid bed in the reaction gas (starting) mixture is The space load is preferably 1500 to 4000 or 6000 Nl / l · h or more. The acrolein load (acrolein load of the catalyst fixed bed) is typically 90 to 190 Nl / l · h, or more than 290 Nl / l · h. An acrolein load of 135 Nl / l · h or more, or ≧ 140 Nl / l · h, or ≧ 150 Nl / l · h, or ≧ 160 Nl / l · h is particularly advantageous, which should also be used according to the invention This is because the reaction gas starting mixture conditions similarly advantageous hot spot properties due to the presence of propane and molecular hydrogen.

  The acrolein conversion for a single pass of the reaction gas mixture through the catalyst bed of the second oxidation stage is preferably typically ≧ 90 mol% and the resulting selectivity for acrylic acid production is ≧ 90 mol%.

It is advantageous to let the charge gas mixture flow through the single zone-multi-contact tube-solid bed reactor as well. As a heat exchanger, also in the second stage, in an advantageous manner, preferably contains 60% by weight of potassium nitrate (KNO ₃ ) and 40% by weight of sodium nitrite (NaNO ₂ ) or 53% by weight of potassium nitrate (KNO ₃ ), nitrous acid A salt melt containing 40% by weight of sodium (NaNO ₂ ) and 7% by weight of sodium nitrate (NaNO ₃ ) is used. As described above, the salt melt and the reaction gas mixture can be guided in a co-current or counter-current as viewed from above the reactor. It is advantageous for the salt melt itself to circulate around the contact tube.

When flowing through the contact tube from top to bottom, it is advantageous to load the contact tube from bottom to top as follows:
First, 50-80 or 70% of the length of the contact tube, and the catalyst alone or a mixture containing the catalyst and an inert substance (the latter being a mass up to 30 (or 20) mass% based on the mixture) (Category C);
Subsequent to 20-40% of the total tube length, the catalyst alone or a mixture containing the catalyst and an inert substance (where the latter is in a mass proportion of up to 50 or 40% by weight with respect to the mixture). (Finally) (section B); and finally a built-in bed containing an inert material (section) that is advantageously selected to condition as little pressure loss as possible to a length of 5-20% of the total pipe length A).

  Section C is advantageously undiluted. As is very common with heterogeneously catalyzed gas phase partial oxidation of acrolein to acrylic acid (especially with high acrolein loading of the catalyst bed and high water vapor content of the charge gas mixture), section B is Continuous catalyst dilution can also be included (for minimizing hot point temperature and hot point temperature sensitivity). From bottom to top, up to 30% (or 20)% by weight of inert material, followed by inert material> 20% to 50 or -40% by weight. In that case, section C is advantageously undiluted.

  Due to the flow of the contact tube from bottom to top, the contact tube loading is reversed in an advantageous manner.

  Said loading variant has in particular the shape 7 mm × 7 mm × 4 mm or 7 mm × 7 mm × 3 mm (each outer diameter x height x inner diameter) as catalyst, according to DE-A 10046928 Preparation 5 or according to DE-A 19815281 and as inert substance This is advantageous when a steatite ring is used. Regarding salt bath temperature, the statements in DE-A 44 31 949 apply. This is typically chosen such that the acrolein conversion achieved in a single pass is typically ≧ 90 mol%, or ≧ 95 mol% or ≧ 99 mol%.

  The partial oxidation of acrolein to acrylic acid can be carried out with the above-mentioned catalyst, but also in a two-zone-multi-contact tube-solid bed reactor, for example as described in DE-A 19910508. it can. Regarding the acrolein conversion rate, the above statements apply. The charge gas mixture (reaction gas starting mixture) is also used in the case of the acrolein partial oxidation described above as the second reaction stage of the two-stage propylene partial oxidation to acrylic acid in a two-zone-multi-contact tube-solid bed reactor. The production gas mixture is preferably directly intermediated in the partial oxidation directed to the first stage (propylene → acrolein), optionally indirectly or directly (eg by supplying secondary oxygen). Used after cooling) (as already mentioned above). The oxygen required for the partial oxidation of acrolein is preferably as pure molecular oxygen (but optionally contains air or molecular oxygen and an inert gas, with an inert gas component, preferably ≦ 50% by volume, particularly preferably ≦ 40% by volume. %, Or ≦ 30% by volume, or ≦ 20% by volume, even better as a mixture having ≦ 10% by volume, or ≦ 5% by volume, or ≦ 2% by volume) It is added directly to the product gas mixture of the first stage (propylene → acrolein). However, as already mentioned, it may already be contained in the reaction gas starting mixture of the first reaction stage.

  In the two-stage partial oxidation of propylene to acrylic acid with direct further use of the product gas mixture of the first stage of partial oxidation for the second stage of partial oxidation, typically two single zones -Multi-contact tube-fixed bed reactor (between the reaction gas and the salt bath (heat carrier) as seen from above the bundle tube reactor, as is very generally advantageous at high reactant loads in the catalyst bed) A parallel flow method is advantageous) or two two-zone-multi-contact tube-fixed bed reactors connected in series. Mixed series connections (single band / 2 bands or vice versa) are also possible.

  There may be intercoolers between the reactors that may optionally contain an inert bed that can perform a filter function. The salt bath temperature of the multi-contact tube reactor for the first stage of the two-stage partial oxidation of propylene to acrylic acid is typically 300-400 ° C. The salt bath temperature of the multi-contact tube reactor for the second stage of partial oxidation of propylene to acrylic acid, partial oxidation of acrolein to acrylic acid, is usually 200-350 ° C. Furthermore, the heat exchange agent (preferably salt melt) flows through the associated multi-contact tube fixed bed reactor in such an amount that the difference between its entry and exit temperatures is typically ≦ 5 ° C. Is done. However, as mentioned above, the two stages of partial oxidation of propylene to acrylic acid can also be carried out with one charge in one reactor, as described in DE-A 101 21592.

  In this case, part of the reaction gas B—starting mixture of the first stage (“propylene → acrolein”) is the residual gas I originating from the separation zone I and / or the residual gas I post-treated according to the invention. Good things are also mentioned above.

  However, the residual gas I is exclusively used as the circulating gas I for the heterogeneously catalyzed propane dehydrogenation in the reaction zone A, which is advantageously exclusively used as a component of the reaction gas A-starting mixture. As reflux. For completeness, it is now possible to supply fresh propane additionally to the first oxidation stage charge gas mixture, to the second oxidation stage charge gas mixture, or to only one of the two. Is emphasized. However, although not advantageous according to the present invention, this is sometimes advantageous in order to eliminate the flammability of the charge gas mixture.

  A two-stage propylene partial oxidation in a bundle tube reactor (such as a so-called “Single-Reactor”) in which the catalyst loading changes correspondingly along the individual contact tubes as the first reaction stage ends, EP-A 9111313, EP-A 979813, EP-A 990636 and DE-A 2830765 for example) form the simplest implementation of a two-oxidation stage for the two stages of partial oxidation of propylene to acrylic acid. At this time, loading of the catalyst into the contact tube is interrupted in some cases by the inert material bed.

  However, the two oxidation stages are realized in the form of two continuously connected bundle pipe systems. These are present in one reactor, in which the movement from one bundle tube to the other bundle tube containing an inert substance that is not accommodated (advantageously) in the contact tube. This is done with a laid floor. The contact tube is swirled around, typically with a heating medium, but this does not reach the inert material bed as previously described. Thus, the two contact bundle tubes are advantageously accommodated in spatially separate reactors. In this case, an intercooler is usually present between the two bundle reactors in order to reduce the post-acrolein combustion that may occur in the product gas mixture leaving the first oxidation zone. The reaction temperature in the first reaction stage (propylene → acrolein) is usually from 300 to 450 ° C., preferably from 320 to 390 ° C. The reaction temperature in the second reaction stage (acrolein → acrylic acid) is typically 200-370 ° C., often 220-330 ° C. The reaction pressure in the two oxidation zones is preferably 0.5 to 5, preferably 1 to 3 bar. In two reaction stages, the oxidation catalyst loading (Nl / l · h) in the reaction gas is often 1500-2500 Nl / l · h or ˜4000 Nl / l · h. At this time, the load with propylene or acrolein is 100 to 200 or 300 Nl / l · h or more.

  The two oxidation steps in the process according to the invention can in principle be carried out as described, for example, in DE-A 19837517, DE-A 19910506, DE-A 19910508 and DE-A 19837519.

  In this case, in two reaction stages, an excess of molecular oxygen compared to the reaction stoichiometrically required amount is effective for the rate of each gas phase partial oxidation and the duration of the catalyst.

  However, the heterogeneously catalyzed gas phase-partial oxidation of propylene to acrylic acid to be carried out according to the present invention can in principle be realized in one single zone bundle tube reactor as follows. The two reaction stages are carried out in one oxidation reactor charged with one or more catalysts, the active substance being able to catalyze the conversion of the two reaction stages, elements Mo, Fe and Bi Is a multi-metal oxide containing Of course, this catalyst loading may change continuously or intermittently along the reaction coordinates. Of course, in the embodiment of the two-stage partial oxidation of propylene to acrylic acid according to the invention, in the structure of two consecutively connected oxidation stages, contained in the product gas mixture leaving the first oxidation stage, The carbon oxide and water vapor produced as by-products in the first oxidation stage can be partially or completely separated from the carbon dioxide, if necessary, before further introduction into the second oxidation stage. According to the present invention, it is advantageous to select a method that does not intend such separation.

As mentioned above, in addition to air (which is less advantageous according to the invention), pure molecular oxygen can also be used as an inert gas, for example CO 2 as a source of intermediate oxygen supply carried out between the two oxidation stages. This also applies to molecular oxygen diluted with ₂ , CO, noble gases, N ₂ and / or saturated hydrocarbons. According to the invention, the oxygen source is preferably a gas different from molecular oxygen ≦ 50% by volume, preferably ≦ 40% by volume, particularly preferably ≦ 30% by volume, very particularly preferably ≦ 20% by volume, better ≦ 10% by volume. % Or ≦ 5% by volume, or ≦ 2% by volume.

  In the scope of the process according to the invention, the cooling of the first partial oxidation stage directly before its further use as a reaction gas starting mixture component of two partial oxidation stages, for example by supplying a cold oxygen source. Can be enabled.

  According to the invention, the partial oxidation of acrolein to acrylic acid is preferably carried out as described in EP-A 1159246, and very particularly preferably in WO 04/085365 and WO 04/085370. However, in accordance with the invention, the reaction gas starting mixture which is the product gas mixture of the first stage partial oxidation of propylene to acrolein according to the invention is preferably used as the reaction gas starting mixture containing acrolein, The ratio of molecular oxygen to acrolein in the reaction gas starting mixture is optionally supplemented with an amount of secondary air, in each case 0.5-1.5. EP-A 1159246, WO 04/08536 and WO 04/085370 are considered as components of this description.

  That is, the partial oxidation of acrolein to acrylic acid according to the present invention can be carried out in a catalyst solid bed with advantageously increased acrolein loading having at least two temperature zones.

  Overall, the two-stage partial oxidation of propylene to acrylic acid is preferably carried out as described in EP-A 1159248 or WO 04/085367 or WO 04/085369.

The product gas B leaving the partial oxidation (according to the first and / or second oxidation stage) to be carried out according to the invention is actually acrolein as the target product, or acrylic acid, or a mixture thereof, unreacted propane, molecules Hydrogen, water vapor (generated as a by-product and / or used as a diluent gas), optionally unreacted molecular oxygen (this is the product gas of the two partial oxidation stages, taking into account the duration of the catalyst used) Boils below and above water (usually advantageous if the oxygen content in the mixture is still at least 1.5-4% by volume, for example), and other (normal pressure, i.e. at 1 bar) by-product or - components (e.g., CO, CO _2, lower aldehydes, lower alkanecarboxylic acids (e.g., acetic acid, formic acid and propionic acid), maleic anhydride, benzaldehyde, aromatic Carboxylic acids and aromatic carboxylic anhydrides (e.g., phthalic acid, and benzoic acid anhydride), optionally other hydrocarbons, e.g., C 4 _- other by a hydrocarbon (e.g., other butene by butene-1 and optionally) and optionally An inert diluent gas, such as N ₂ ).

  As a method for separating the target product contained in the product gas B, water, and a secondary component boiling higher than water, the method according to the present invention in the first separation zone I is basically known in the prior art. All methods apply to this. These methods actually transfer the target product, for example by absorption and / or condensation methods, from the gaseous to the condensed phase, and subsequently extract the condensate or absorption for further target product separation. Characterized by treatment by distillation, crystallization and / or desorption. Together with the target product and / or following the transition of the target product to the condensed phase, the process typically also includes subcomponents boiling above the water vapor and water contained in the reaction gas B in the condensed phase. Transferred and thus separated (preferably according to the invention, preferably at least 70 mol% of the amount of water vapor produced in reaction zone B (or particularly preferably all contained in product gas B), preferably at least 80 The amount of water vapor which is mol%, better at least 90 mol% and particularly preferably 95 mol% is transferred to the condensed phase).

In this case, examples of the absorbent include water, an aqueous solution, and / or an organic solvent (for example, a mixture containing diphenyl and diphenyl ether or a mixture containing diphenyl, diphenyl ether and o-dimethylphthalate). In the scope of this “condensation” (separation) of the target product, water and subcomponents boiling higher than water, typically including components of product gas B that are relatively hard to condense (boiling lower than water), Residual gas that does not enter the condensed phase remains. These are usually components whose boiling point is typically ≦ −30 ° C. at standard pressure (1 bar) (its total proportion in the residual gas is usually ≧ 60% by volume, often ≧ 70% by volume, often ≧ 80 Volume%, but usually ≦ 90 volume%). First of all, unreacted propane, molecular hydrogen, carbon dioxide, possibly unreacted propylene, possibly unreacted molecular oxygen and other secondary components boiling below water, such as CO, ethane, methane N _{2 and} , in some cases, rare gases (eg, He, Ne, Ar, etc.) belong. The residual gas may still contain acrylic acid, acrolein and / or H ₂ O to a narrow extent. The residual gas preferably contains according to the invention water vapor ≦ 10% by volume, preferably ≦ 5% by volume and particularly preferably ≦ 3% by volume. This said residual gas is usually at least the main amount (usually at least 80% or at least 90%, or at least 95%) of the residual gas produced in the first separation zone I (relative to the amount of propane contained therein) Above) and is characterized herein as (main) residual gas.

  In particular, if the target product is condensed by absorption with an organic solvent, the separation zone I typically contains at least one second residual gas containing unreacted propane and optionally unreacted propylene. The resulting (its amount relative to the propane contained therein is usually less in comparison to the amount of (main) residual gas I (typically ≦ 20%, usually ≦ 10%, or ≦ 5%, or ≦ 1 %)). This should be due to the fact that the condensed phase produced contains unreacted propane and possibly unreacted propylene to a certain extent.

  In the further course of extraction of the desired product from the condensed phase, distillation, crystallization and / or desorption separation, this unreacted propane and optionally propylene is usually recovered as at least one other gas phase component, Characterized herein as (secondary) residual gas I.

  In this case, the total including the (main) residual gas I and the (sub) residual gas I is the total amount of the residual gas I. When the (sub) residual gas I is not generated in the separation zone I, the (main) residual gas I is automatically the total amount of the residual gas I (also referred to as (total) residual gas I).

  Advantageously according to the invention, the transition of the target product from product gas B to the condensed phase takes place by fractional condensation. This is especially the case when the target product is acrylic acid. However, in principle, for the purpose of product separation, for example, DE-A 102 13998, DE-A 2263696, US 3343840 (the method described in the above specification is particularly recommended for acrolein separation), EP-A 1388533, EP-A1388532, DE-A10235847, EP-A792867, WO98 / 01415, EP-A1015411, EP-A1015410, WO99 / 50219, WO00 / 53560, WO02 / 09839, DE-A10235847, WO03 / 041833, DE-A10223058, DE- A102443625, DE-A103336386, EP-A854129, US-A4317926, DE-A19837520, DE-A196606877, DE-A1905013 5, DE-A10247240, DE-A19740253, EP-A695736, EP-A982287, EP-A1041062, EP-A117146, DE-A43008087, DE-A4335172, DE-A4436243, DE-A19924532, DE-A10332758 and DE-A19924533 Preference is given to all absorption methods described in the literature and / or “condensation of target products” and condensation methods for the further processing of “condensates”. Acrylic acid separation can be performed using EP-A 982287, EP-A 982289, DE-A 10336386, DE-A 10115277, DE-A 19606877, DE-A 19740252, DE-A 19627847, EP-A 920408, EP-A 1068174, EP-A 10626239, EP-A 1066240, WO00. / 53560, WO 00/53561, DE-A 10053086 and EP-A 982288. Advantageously, they are separated as described in FIG. 7 of WO / 0196271 or DE-A102004032129 and its equivalent patents. An advantageous separation method is also the method described in WO 04/063138, WO 04/35514, DE-A 10243625 and DE-A 10235847. Further treatment of the crude acrylic acid obtained at this time can be carried out, for example, as described in WO 01/77056, WO 03/041832, WO 02/055469, WO 03/078378 and WO 03/041833.

  A common feature of the separation method is that, as described above, for example, the product gas B in the lower part is typically supplied after, for example, direct and / or indirect cooling before it is separated. At the head of the separation column containing the equipment having, the residual gas I stream typically remains, which has a boiling point lower than that of water at standard pressure (1 bar), usually ≦ −30 It mainly contains that component of product gas B that is at 0 ° C. (ie, a component that is difficult to condense or is also easily volatile). However, the residual gas I can also contain further components, for example water vapor and acrylic acid.

  In the lower part of the separation column, typically, the hardly volatile components of the product gas B containing the respective target products are produced in the condensed phase. The condensed water phase is usually discharged via the side port and / or via the bottom of the column.

Typically, the (main) residual gas I has the following contents:
H ₂ O 1 to 20 volume%
Propane 40-90% by volume
O ₂ 0-10% by volume
CO ₂ 5 to 30% by volume and CO> 0 to 2% by volume.

  The contents of acrolein and acrylic acid are typically <1% by volume each.

  By the way, according to the present invention, it is necessary to treat the residual gas I generated in the separation zone I by a certain post-treatment method. In this case, according to the present invention, it is essential that each post-treatment method should not necessarily be carried out with the total amount of residual gas I. That is, it is advantageous according to the invention to carry out the aftertreatment process only with a fraction of the total amount of residual gas I (for example only with (main) residual gas I). In this case, the total from the untreated partial amount of the total amount of residual gas I and from the residual gas containing propane remaining after the implementation of the post-treatment process with other partial amounts of total residual gas I is It constitutes the total amount of post-treatment residual gas I ("new" residual gas I) that can be processed in a further post-treatment according to the invention in the process (the said partial amounts can also have different chemical compositions). Of the total amount of unreacted propane-containing (post-treatment) residual gas I remaining after the implementation of all post-treatment methods deemed necessary in each case, at least a partial amount according to the invention is at least partially Reflux into reaction zone A as at least one of two propane-containing feed streams. The other partial quantities (and possibly also the untreated partial quantities of the residual gas I) may optionally be adjusted in order to advantageously adjust the explosive properties of the reaction gas as a component of the respective charge gas. And / or refluxed to the second oxidation stage.

  However, the reflux amount of the residual gas I post-treated according to the present invention as at least one stream of at least two propane-containing feed streams to reaction zone A according to the present invention is advantageously increased according to the present invention. Propane exiting reaction zone B is at least 80 mol%, preferably at least 85 mol%, better at least 90 mol%, or at least 92 mol%, or at least 94 mol%, or at least 96 mol%, or at least 98 mol% % Is allocated. At this time, the reflux does not necessarily have to be performed at one place and the same place in the reaction zone A. Rather, this can be done over a number of differently distributed feed points on reaction zone A.

Again in the above-mentioned range, it is essential according to the invention that the order of the aftertreatment processes carried out on the residual gas I is arbitrary according to the invention. That is, instead of first discharging a portion of the residual gas I and then continuing to CO ₂ -washing out the remaining amount, for example, first CO ₂ -washing is performed, and the residual gas I thus washed out A part of the amount can be discharged.

  However, with the advantageous method according to the invention, the following sequence of aftertreatment methods is observed (in particular for the aftertreatment of the (main) residual gas I).

  First, a portion of the residual gas I is discharged (discharging is necessary in particular for discharging inert components introduced with the supply of molecular oxygen to the whole process). At this time, this emission amount (usually, the emission amount of the residual gas I is supplied to the exhaust gas combustion) has the same composition as the residual gas I itself. This is usually because the amount of propane contained in the exhaust residual gas I per unit time is ≦ 30 mol%, more preferably ≦ 25 mol, relative to the amount of fresh propane supplied per unit time during the process. Examples thereof are% or ≦ 20 mol%, or ≦ 15 mol%, preferably ≦ 10 mol%, particularly preferably ≦ 5 mol% and very particularly preferably ≦ 3 mol% or ≦ 1 mol%.

  If the amount of propane to be discharged is higher than the above value or is an example of the above value, the propane contained in the amount of residual gas I to be discharged and possibly propylene are discharged before the discharge. It is also possible to carry out the discharge in such a way that the propane thus separated and optionally the propylene remain as components of the residual gas I post-treated according to the invention.

A simple possibility for said separation is, for example, that a corresponding amount of residual gas I is absorbed (preferably for preference over other components of residual gas I) propane and propylene (advantageous). (For example, by simple passage introduction) in contact with an organic solvent (highly boiling) which is advantageously hydrophobic (for example, tetradecane or a mixture of C ₈ -C ₂₀ -alkanes). Subsequent desorption (under reduced pressure), distillation and / or with non-hindered gases (eg, water vapor, molecular hydrogen, molecular oxygen and / or other inert gases (eg, air as well)) in reaction zone A The propane and (optionally) propylene are recovered and added to the post-treatment residual gas I by the following stripping. Individually, for example, a similar absorption separation of propane from product gas A can be carried out as described in DE-A 102004032129. When said separation, which is not very advantageous according to the invention, is carried out in the process according to the invention, it always relates only to a partial amount of residual gas I and thus only to a partial amount of propane contained in product gas B. To do. The amount of the propane part is preferably ≦ 50 mol%, particularly preferably ≦ 40 mol%, particularly preferably ≦ 30 mol%, particularly preferably ≦ 20 mol%, very particularly preferably ≦ 10 mol%, according to the invention. And at best ≦ 5 mol%.

After the partial amount of the residual gas I is discharged, the CO ₂ contained therein is washed out from at least a partial amount of the residual gas I (in the second separation zone II). If the residual gas I contains also CO and O ₂ are selected by the introduction of the CO contained in (simultaneous chemical under reduction of molecular oxygen) First residual gas I, for example, on the catalyst to the corresponding Can be oxidized to CO ₂ (if necessary, the residual gas I can be pre-fed with the molecular oxygen required for it). This is done using the catalyst recommended in WO 01/60738 (complex oxide of stoichiometric formula Cu _x Ce _1-x O _{2 -y} x = 0.01-0.3 and y ≧ x 1). It is possible even if exists. However, such CO-conversion is not advantageous according to the present invention. This is because the CO remaining in the circulation gas I is anyway at least partially oxidized to CO ₂ in the subsequent annular circuit in the reaction zone A and / or B, so that said CO ₂ -wash Conditioned by having natural emissions on the outing. However, simultaneously with the CO-conversion, reduction of O ₂ contained in the residual gas I with H ₂ contained in the residual gas I can be accompanied at the same time. This reduction can be performed alone.

Typically, according to the present invention, it is sufficient to wash out only the CO ₂ contained therein from a portion of the residual gas I. CO ₂ -washing is preferably carried out at elevated pressure (typically 3-5 bar, better 5-30, preferably 8-20, particularly preferably 10-20 or -15 bar) and the book The invention is preferably carried out with a basic (in the sense of Bronsted) liquid.

Such basic liquids include, for example, organic amines such as monoethanolamine, organic amines such as aqueous solutions of monoethanolamine, aqueous alkali metal hydroxides or aqueous alkali metal carbonates or alkali metal carbonates and alkalis. This is the case with aqueous solutions of metal hydrogen carbonate (see, for example, WO 05/05347). Advantageously according to the invention, an aqueous solution containing K ₂ CO ₃ is used for CO ₂ -washing. Examples of such an aqueous solution include an aqueous solution of K ₂ CO _3, an aqueous solution of K ₂ CO ₃ and KHCO ₃ , or an aqueous solution of K ₂ CO ₃ , KHCO ₃ and KOH. Such an aqueous potassium carbonate solution preferably has a solids content of 10 to 30 or 40% by weight, particularly preferably 20 to 25% by weight. In order to neutralize the carboxylic acid (eg acetic acid, formic acid, acrylic acid) optionally contained in the residual gas I in a small amount, an alkali metal carbonate can be obtained by supplying a small amount of alkali metal hydroxide (eg KOH). The aqueous solution can additionally be adjusted to alkaline before washing. Suitable aqueous washing solutions also contain KHCO ₃ and K ₂ CO ₃ dissolved in a mass ratio of about 1: 2, with a solids content of preferably 20 to 25% by weight. CO ₂ and under accommodation of water in the range of _washing, K 2 CO ₃ 1 mol unit converts into KHCO ₃ 2 molar units _{(K 2 CO 3 + H 2} O + CO 2 → 2KHCO 3).

CO ₂ - the washout instead be carried out in a basic liquid, CO ₂ - the washout, for example, by editorial EP-A900121 CO ₂ - may also be carried out under clathrate formation.

The CO ₂ -washing is carried out countercurrently in the washing column, advantageously in terms of the application technique. The residual gas I to be cleaned typically flows from bottom to top and the cleaning liquid from top to bottom in the cleaning column. In a manner known per se, the washing column has an internal structure with an enlarged exchange surface. This may be a packing, a mass exchange bed (eg a sieve bed) and / or a package.

The washed residual gas I is discharged at the top of the washing column, and, for example, an aqueous solution containing mainly potassium hydrogen carbonate (or CO ₂ -clathrate aqueous solution) is advantageously extracted from the bottom of the washing column. For example, the tower bottom solution may contain KHCO ₃ and K ₂ CO ₃ dissolved in a mass ratio of 2.5: 1.

By introducing hot water vapor into the aqueous potassium hydrogen carbonate solution, the hydrogen carbonate is thermally dissociated (to carbonate and CO ₂ , H ₂ O), with the liberated CO ₂ being expelled. The aqueous potassium carbonate solution recovered in this way can usually be refluxed for CO ₂ -washing to be carried out as described above after evaporation of the water vapor condensed in the range of dissociation. The evaporation of the condensed water vapor can be operated continuously through an evaporator built in the column. Said dissociation is preferably carried out in a column (for example a (sieving) bed column, a package column and / or a packed column) which likewise contains an internal structure with an enlarged exchange surface. Here, a countercurrent operation is also advantageous. Hot water vapor is preferably fed to the lower part of the column and an aqueous potassium carbonate solution is preferably charged to the top of the column column.

CO ₂ -washing is preferably carried out at low temperatures, whereas hydrogen carbonate dissociation is advantageously carried out at elevated temperatures. In the context of the CO ₂ -washing, dissociation of the cleaning solution and reuse of the cleaning solution, the dissociation is preferably carried out with water vapor having a temperature of 130-160 ° C., preferably 140-150 ° C. The solution to be dissociated is preferably charged at a temperature of 80 to 120 ° C., particularly preferably 90 to 110 ° C. Conversely, the residual gas I is preferably introduced according to the invention into the washing column at a temperature of 60-90 ° C., particularly preferably at 70-80 ° C., and the washing liquid is introduced at a temperature of 70-90 ° C., preferably 75-85 ° C. Charge.

In particular, only a portion of residual gas I CO 2 _- when cleaning, CO 2 _- is advantageously carried out in several steps the compressed residual gas I to favorable pressure washout (e.g., By means of a multistage radial compressor; the compressed gas can be discharged after each compression stage).

Typically, the residual gas I is discharged from the separation zone I in the process according to the invention at a pressure of ≧ 1 bar and ≦ 2,5 bar, preferably ≦ 2.0 bar. The heterogeneously catalyzed dehydrogenation in reaction zone A is preferably carried out at a pressure of ≧ 2 to ≦ 4 bar, preferably ≧ 2.5 to ≦ 3.5 bar. Reaction zone B is preferably operated according to the invention at pressures of ≧ 1 bar and ≦ 3 bar, preferably ≧ 1.5 and ≦ 2.5 bar. CO ₂ - advantageous operating pressure of the washout is 3-50 bar, or from 5 to 30 bar, or from 8 to 20 bar, preferably from 10 to 20 or 15 bar.

By the way, it is advantageous in terms of application technology, in the first compression stage, a multistage type, for example, a turbo compressor (radial compressor; for example, Typ MH4B of Manesmann DEMAG, DE) (in this case) , Also referred to as circulating gas I compressor), the residual gas I is compressed to the operating pressure in reaction zone A (for example from 1.2 to 3.2 bar). If the total amount of residual gas I is not subjected to CO ₂ -washing, the residual gas I compressed as described above is advantageously divided into two partial amounts of the same composition. At this time, the amount ratio is, for example, 70 to 30% of the total amount. CO ₂ -A partial amount that should not be washed out is prepared for reflux directly into reaction zone A. Only the other partial quantity is further compressed to the operating pressure watched for CO ₂ -washing. This can be done in just another compression stage. To that end, however, at least two further compression stages are advantageously applied according to the invention. This can be attributed to an increase in temperature as the gas is compressed. Conversely, CO 2 to a suitable operating pressure in reflux into the reaction zone A _- with relief of the washout residual gas I (expansion), CO 2 _- cooling the washout residual gas I occurs. If the pressure relief is likewise carried out in multiple stages, an indirect heat exchange is carried out before each subsequent stage, with the heated compressed unwashed residual gas I and the cooled expanded CO ₂ -washed residual gas I. And can be done between. Such a heat exchange can also be applied already before the first expansion. As a result of the indirect heat exchange, condensation of the water vapor optionally still contained in the residual gas I can occur. Said expansion is preferably carried out in an expansion turbine (preferably likewise in multiple stages), which is used for the recovery of compression energy. At each compression-or expansion stage, the difference between the entry- and exit pressures can be, for example, 2-10 bar.

Typically, at least 50 mol%, usually at least 60 mol%, or at least 80 mol%, and often at least 90 mol% or 90 mol% or more of the CO ₂ contained in the residual gas I in the separation zone II is washed out.

The amount of CO ₂ -washed out advantageously corresponds to the amount of CO ₂ washed out in the reaction zones A and B.

Typically, CO ₂ - washout residual gas is still _CO 2 1 to 20, often still contains 5-10% by volume.

In order to separate at least a portion of the molecular hydrogen contained therein before refluxing the residual gas I into the reaction zone A before the CO ₂ -washed residual gas I is released, a third Membrane separation can be effected in separation zone III (especially when only a relatively small amount of hydrogen is burned in reaction zone A). Suitable separation membranes in this regard are, for example, aromatic polyimide membranes such as those of UBE Industries Ltd. Among the latter, in particular, membrane types A, BH, C and D fall under this category. For such separation, the use of Ube Industries BH type polyimide membrane is particularly advantageous. For this membrane, the hydrogen permeability of H ₂ is 0.7 · 10 ⁻³ [STP · cc / cm ² · sec · cmHg] at 60 ° C. For this purpose, the CO ₂ -washing residual gas I is usually shaped into a tube (but also a plate-like or scroll-like module), through a membrane that is permeable only for molecular hydrogen. be introduced. The molecular hydrogen thus separated can be used further in the scope of other chemical syntheses or, for example, supplied to the combustion together with the residual gas I discharged in the process according to the invention. This combustion also includes free CO ₂ (which can easily be released to the atmosphere) and aqueous condensate produced in separation zones I and II.

  Furthermore, the high boiling body separated in the separation zone I can be supplied to the combustion. Combustion is typically performed with an air supply. Combustion can be performed, for example, as described in EP-A 925272.

Membrane separation of molecular hydrogen is likewise advantageously carried out under high pressure (for example 5-50 bar, typically 10-15 bar).
Therefore, in order to take advantage of the once increased pressure level advantageously twice, it is possible to carry it out immediately before (or advantageously) the CO ₂ -washing of the residual gas I or immediately after the CO ₂ -washing. It is advantageous according to the invention.

In addition to plate-like membranes, scroll-like membranes or capillary membranes, this is especially the case for hose-like membranes (hollow fiber membranes) for hydrogen separation. The inner diameter may be, for example, a small number of μm to a small number of mm. To that end, for example, a bundle of such hose-like membranes is poured into one plate at each of the two hose ends, similar to the tubes of a bundle reactor. The exterior of the hose membrane is preferably dominated by reduced pressure (<1 bar). Residual gas I under elevated pressure (> 1 bar) is introduced opposite one of the two plate ends and pumped to the hose outlet at the opposite plate end through the hose interior . In this way, molecular hydrogen is released to the outside through a H ₂ -permeable membrane along a predetermined flow path inside the hose.

  In other respects, the residual gas I post-treated as described above (total including post-treated partial amount and possibly unpost-treated partial amount), at least 2 containing gaseous propane, as required. It can be refluxed into reaction zone A as one of the two feed streams. Advantageously according to the invention, the total amount of residual gas I worked up according to the invention is fed to the reaction gas A-starting mixture (reaction zone A charge gas mixture).

Preference is given according to the invention to the process according to the invention in which the highest set operating pressure level is present in the CO ₂ -washing process stage of the residual gas I. This is the combination of only one compressor (preferably a multistage radial compressor) that can be situated between the production of the residual gas I and the CO ₂ -washing of the residual gas I, as described above. Allows the implementation of the method according to the invention below. In particular, between the reaction zone A and the reaction zone B, in the process according to the invention, as mentioned above, it is not necessary to use another compressor in combination. Of course, however, another such compressor can be incorporated in the method according to the invention.

  Another advantage of the method according to the invention is that it enables the realization of the smallest possible circulation flow. Furthermore, the reaction gas A—starting mixture fed to reaction zone A naturally protects the hydrocarbons contained in this starting mixture, simultaneously protecting the molecular hydrogen before combustion with the molecular oxygen contained in this mixture. Clarify the possibility of inclusion. In connection with the fresh propane to be converted, this results in a relatively high target product selectivity. Finally, the basis of the process according to the invention is based on the fact that non-inert subcomponents produced or introduced during the course of the process according to the invention are connected to the outlet of the process according to the invention. It is emphasized that it is naturally present in the condensate produced in zones I and II.

The circulating gas I refluxed to the reaction zone A typically has the following:
50-90% by volume of propane,
CO> 0-5, usually ~ ≦ 2 or ≦ 1% by volume,
H ₂ 1 to 20% by volume,
CO ₂ 1 to 20% by volume,
H ₂ O 1 to 20% by volume,
O ₂ 0-10, often ˜5% by volume and propylene 0 (often ≧ 0.1) -5% by volume.

  Finally, it is further emphasized that a polymerization inhibitor is supplied whenever a condensed phase appears in either separation zone I or separation zone II. As such, in principle all known process inhibitors fall under this category. For example, phenothiazine and methyl ether of hydroquinone are particularly suitable according to the invention. The presence of molecular oxygen increases the effectiveness of the polymerization inhibitor. The acrolein produced by the process according to the invention can be converted into the acrolein reaction products mentioned in US-A 6,166,263 and US-A 6,187,963. Illustratively, the acrolein reaction products are 1,3-propanediol, methionine, glutaraldehyde and 3-picoline.

Examples and comparative examples Long-term operation of heterogeneously catalyzed two-stage partial oxidation of propylene to acrylic acid in the absence and presence of molecular hydrogen A) General test structure of the reactor First oxidation stage reactor The reactor is made of Seiko (A cylindrical guide tube surrounded by a cylindrical outer container). The wall thickness was 2-5 mm overall.

  The inner diameter of the outer cylinder was 91 mm. The inner diameter of the guide tube was about 60 mm.

  The double-walled cylinder was sealed with a lid or floor above and below.

  The contact tube (total length 400cm, inner diameter 26mm, outer diameter 30mm, wall thickness 2mm, refined steel) protrudes straight through the lid or floor (sealed) at the top or bottom in the guide tube of the cylindrical container. It was installed as it was. The heat exchange agent (a salt melt containing 53% by mass of potassium nitrate, 40% by mass of sodium nitrite and 7% by mass of sodium nitrate) was sealed in a cylindrical container. The heat exchange agent was pumped with a propeller pump in order to ensure as uniform thermal limit conditions as possible on the outer wall of the contact tube over the entire length of the contact tube (400 cm) present in the cylindrical vessel.

  An electric heater mounted on the outer jacket allowed the temperature of the heat exchange agent to be adjusted to the desired level. In other respects it was air cooling.

  Reactor loading: Viewed from the top of the first stage reactor, the salt melt and charge gas mixture of the first stage reactor were introduced in cocurrent. The charge gas mixture entered the first stage reactor at the bottom. Each was introduced into the reaction tube at a temperature of 165 ° C.

The salt melt ^entered the cylindrical guide tube at a temperature T at the bottom, and was discharged from the cylindrical guide tube at a temperature T ^discharge that was 2 ° C. higher than the T ^entry at the top.

The T ^entry was adjusted at the outlet of the first oxidation stage to always result in a propylene conversion of 97.8 ± 0.1 mol% in a single pass.

Contact tube loading: (from bottom to top)
Section A: pre-filled floor containing steatite-spheres with a length of 90 cm and a diameter of 4-5 mm,
Section B: catalyst loading with a homogeneous mixture containing 30% by weight steatite annulus of length 100 cm, shape 5 mm × 3 mm × 2 mm (outer diameter × length × inner diameter) and 70% by weight of complete catalyst from section C,
Section C: length 200 cm, annular (5 mm × 3 mm × 2 mm = outer diameter × length × inner diameter) complete catalyst according to DE-A10046957 example 1 (stoichiometric formula: [Bi ₂ W ₂ O ₉ x2WO ₃ ] _0.5 [Mo ₁₂ Catalyst loading with Co _5.5 Fe _2.94 Si _1.59 K _0.08 O _x ] ₁ ),
Section D: Post-embedded floor containing steatite-annular material having a length of 10 cm and a shape of 7 mm × 3 mm × 4 mm (outer diameter × length × inner diameter).

Intermediate cooling and oxygen intermediate supply (pure O ₂ as secondary gas)
The product gas mixture leaving the first fixed bed reactor is connected to a connecting tube (length 40 cm, inner diameter 26 mm, outer diameter 30 mm, wall thickness 2 mm, refined steel, 1 cm of insulating material for intermediate cooling (indirectly by air). This connecting tube is centered up to a length of 20 cm and is an inert built-up bed containing steatite-annular material of shape 7 mm x 3 mm x 4 mm (outer diameter x length x inner diameter) It was loaded and flanged directly to the first stage reactor.

The product gas mixture always ^{entered the} connecting tube with a temperature> T ^approach (first stage) and was discharged at temperatures set at 200 ° C. and 270 ° C.

At the end of the connecting tube, the cooled product gas mixture was fed with molecular oxygen at the pressure level of the product gas mixture. The resulting gas mixture (first dioxide stage charge gas mixture) was fed directly into a second stage contact tube with the connecting tube flanged to it at the other end as well. The amount of molecular oxygen fed was distributed such that the molar ratio of O ₂ present in the resulting gas mixture to acrolein present in the resulting gas mixture was 1.3.

The reactor in the second oxidation stage A contact tube-solid bed reactor that is structurally the same as that in the first oxidation stage was used. The salt melt and charge gas mixture were introduced in cocurrent flow as viewed from the top of the reactor. The salt melt entered at the bottom and the charge gas mixture entered as well. The salt melt entry temperature T ^entry was always adjusted to produce an acrolein conversion of 99.3 ± 0.1 mol% in a single pass at the outlet of the second oxidation stage. The T ^discharge of the salt melt was high from T ^entry to 2 ° C.

Contact tube loading (from bottom to top) was as follows:
Section A: Pre-filled floor containing steatite-annular material of length 70 cm, shape 7 mm x 3 mm x 4 mm (outer diameter x length x inner diameter),
Section B: catalyst loading with a homogeneous mixture containing 30% by weight steatite annulus of length 100cm, shape 7mm x 3mm x 4mm (outer diameter x length x inner diameter) and 70% by weight shell-type catalyst from section C,
Section C: Length 200 cm, cyclic shell-type catalyst (7 mm × 3 mm × 4 mm = outer diameter × length × inner diameter) according to Production Example 5 of DE-A10046928 (stoichiometric formula: Mo ₁₂ V ₃ W _1.2 Cu _2.4 O loading of catalyst at _x ),
Section D: Post-filled floor containing steatite-annular material having a length of 30 cm and a diameter of 4 to 5 mm.

  B) Results achieved depending on the composition of the charge gas mixture of the first oxidation stage (propene loading is adjusted to 150 Nl / l · h, selectivity for acrylic acid production (averaged over the two reaction stages, and (Based on converted propylene) was always ≧ 94 mol%).

a) The composition of the charge gas mixture of the first oxidation stage was actually as follows:
7% by volume of propylene,
O ₂ 12% by volume,
H ₂ 20% by volume,
H ₂ O 5% by volume and N ₂ 56% by volume.

At the beginning of the newly loaded reactor, the entry temperature was as follows:
T ^approach (first oxidation stage): 320 ° C
T ^approach (second oxidation stage): 268 ° C
After three months of operation time, the entry temperature was as follows:
T ^approach (first oxidation stage): 330 ° C
T ^approach (stage 2): 285 ° C
b) The composition of the charge gas mixture of the first oxidation stage was actually as follows:
7% by volume of propylene,
O ₂ 12% by volume,
H ₂ O 5 vol% and N ₂ 76% by volume.

At the beginning of the newly loaded reactor, the entry temperature was as follows:
T ^approach (first oxidation stage): 320 ° C
T ^approach (second oxidation stage): 268 ° C
After three months of operation time, the entry temperature was as follows:
T ^approach (first oxidation stage): 324 ° C
T ^approach (stage 2): 276 ° C
II. Acrylic acid production method from propane (state the steady operation state)
Reaction zone A consists of a furnace furnace, which is implemented as an adiabatic reactor with three catalytic solid beds arranged in succession in the flow direction and is adiabatically formed. Each catalyst solid bed was placed on a fine steel wire mesh (arranged in the above-mentioned order in the flow direction), an inert substance (filled bed height: 26 mm; steatite spheres having a diameter of 1.5 to 2.5 mm) ) And a mixture of fresh dehydrogenation catalyst and steatite spheres (diameter 1.5-2.5 mm) (bed bed volume ratio, dehydrogenation catalyst: steatite spheres = 1: 3 (selection) Alternatively, the same amount of catalyst can be used instead, but it can also be used undiluted)).

There is a static gas mixer in front of each fixed bed. The dehydrogenation catalyst is a Pt / Sn-alloy, which is promoted by the oxide forms of the elements Cs, K and La, the element stoichiometry (mass ratio including support) Pt _0.3 Sn _0.6 La _3.0 Cs _0.5 K _0.2 (ZrO ₂ ) _88.3 (SiO ₂ ) _7.1 ZrO ₂ / SiO ₂ mixed oxide support extrudate (average length (up to about 6 mm, 3-12 mm) On the outer and inner surfaces (catalyst precursor production and activation to an active catalyst is similar to example 4 of DE-A 102 19879).

  Heterogeneously catalyzed partial propane dehydrogenation is carried out in the above-mentioned floor reactor in a straight line. The total catalyst loading (calculated without inert material) in propane is 1500 Nl / l · h.

The reaction gas A-starting mixture (T = 438 ° C., P = 2.92 bar) 62473 Nm ³ / h is fed to the first catalyst bed in the flow direction:
Circulating gas I50870 Nm ³ / h with the following contents:
Acrolein 0.01% by volume,
Acrylic acid 0.01% by volume,
63.49% by volume of propane,
Propylene 0.86% by volume,
H ₂ 14.50% by volume,
4.72% by volume of O ₂ ,
H ₂ O 2.05% by volume,
CO 1.17% by volume,
CO ₂ 11.17% by volume,
Crude-propane 8512 Nm ³ / h containing water vapor 3091 Nm ³ / h; and propane> 98% by volume.

  The reaction gas A—starting mixture is heated from 80 ° C. to 438 ° C. by indirect heat exchange with product gas A (T = 581 ° C .; P = 2.88 bar). The evaporation of the crude-propane is effected by the aqueous condensate generated in separation zone I (acidic water (T = 33 ° C.), indirect heat exchange). The condensate cooled in this case can be further used for acid water condensation in the separation zone I (for example as a coolant in the range of direct cooling). The water vapor is obtained at a temperature of 152 ° C. and 5 bar.

Reactant gas A-the bed height of the first catalyst bed passed by the starting mixture is discharged with the following contents at a temperature of 549 ° C and a pressure of 2.91 bar. Is the state to:
59.0% by volume of propane,
5.94% by volume of propylene,
H ₂ 10.25% by volume,
H ₂ O 12.98% by volume
CO ₂ 9.97% by volume and O ₂ 0% by volume.

  The maximum temperature in the first catalyst bed is 592 ° C.

The discharge amount is 63392 Nm ³ / h. After the first catalyst bed, molecular oxygen (purity> 99% by volume) 986 Nm ³ / h is fed to the reaction gas A. The oxygen is preheated to 176 ° C. It is compressed to a pressure of 3.20 bar, so that the pressure produced by the reaction gas A produced is still 2.91 bar.

The bed height of the second catalyst bed is such that the reaction gas A is discharged from the second catalyst bed at a temperature of 566 ° C. and a pressure of 2.90 bar with the following contents:
Propane 51.91% by volume,
9.55% by volume of propylene,
H ₂ 10.88% by volume,
H ₂ O 15.38% by volume
CO ₂ 9.57% by volume and O ₂ 0% by volume.

  The maximum temperature in the second catalyst bed is 595 ° C.

The discharge amount is 66058 Nm ³ / h. This reaction gas A is fed with molecular oxygen (purity> 99% by volume) 918 Nm ³ / h in front of a static mixer in front of the third catalyst bed (T = 176 ° C., pressure of 3.20 bar) Compress to state). The production pressure of the reaction gas I produced is further 2.90 bar.

The bed height of the third catalyst bed is a state in which the reaction gas A is discharged with the following content at the temperature of 581 ° C. and the pressure of 2.88 bar with the third catalyst bed as the product gas A. :
Propane 47.19% by volume,
12.75% by volume of propylene,
H ₂ 11.38% by volume,
H ₂ O 17.46% by volume
CO ₂ 9.21% by volume and O ₂ 0% by volume.

  The maximum temperature in the third catalyst bed is 612 ° C.

The discharge amount is 68522 Nm ³ / h. By indirect heat exchange with the reaction gas A-starting mixture, the product gas A is cooled to a temperature of 290.degree.

Molecular oxygen (purity> 99% by volume) 13306 Nm ³ / h is supplied to the product gas A (T = 176 ° C., compressed to a pressure of 3.20). For the first partial oxidation stage of the two-stage partial oxidation of propylene contained in product gas B to acrylic acid, the production pressure and production temperature of the charge gas mixture thus obtained (reaction gas B-starting mixture) are 283 ° C. and 1.75 bar.

The first oxidation stage is a multi-tube reactor having two temperature zones. The reaction tube is formed as follows: V2A steel; outer diameter 30 mm, wall thickness 2 mm, inner diameter 26 mm, length: 350 cm. The reaction tubes are loaded from top to bottom as follows:
Category 1: Steatite ring having a length of 50 cm and a shape of 7 mm × 7 mm × 4 mm (outer diameter × length × inner diameter) as a pre-stacked floor,
Category 2: Steatite annulus with a length of 140 cm, shape 5 mm x 3 mm x 2 mm (outer diameter x length x inner diameter) 20% by mass (optionally 30% by mass) and 80% by mass of complete catalyst from category 3 (optionally 70% by mass) Catalyst loading in a homogeneous mixture containing
Category 3: 160 cm long, cyclic complete catalyst according to DE-A10046957 Example 1 (5 mm × 3 mm × 2 mm = outer diameter × length × inner diameter) (stoichiometric formula: [Bi ₂ W ₂ O ₉ x2WO ₃ ] _0.5 [Mo _{12 Co 5.5 Fe 2.94 Si 1.59 K} 0.08 O x] 1). Optionally, one of the catalysts BVK1 to BVK11 from Research Disclosure No. 497012 with 29.08.2005 can also be used (otherwise it is numerically accurate). , the unit of specific surface area shown in the examples, although numerically accurate, the cm 2 / ^g without a m 2 ^{/ g).}

  From top to bottom, the first 175 cm is held isothermally by salt bath A pumped countercurrently to reaction gas B. The second 175 cm is held at a constant temperature by the salt bath B pumped countercurrently to the reaction gas B.

The second oxidation stage is a multi-tube reactor which likewise has two temperature zones. The reaction tubes are loaded from top to bottom as follows:
Category 1: Steatite ring having a length of 20 cm and a shape of 7 mm x 7 mm x 4 mm (outer diameter x length x inner diameter)
Category 2: Steatite annulus with a length of 90 cm, shape 7 mm x 3 mm x 4 mm (outer diameter x length x inner diameter) 25% by mass (optionally 30% by mass) and 75% by mass of shell type catalyst from category 4 (optionally 70% Catalyst loading with a homogeneous mixture containing
Category 3: Steatite annulus with a length of 50 cm, shape 7 mm x 3 mm x 4 mm (outer diameter x length x inner diameter) 15% by mass (optionally 20% by mass) and 85% by mass of shell-type catalyst from category 4 (optionally 80%) Catalyst loading with a homogeneous mixture containing
Section 4: Length 190 cm, annular shell-type catalyst (7 mm × 3 mm × 4 mm = outer diameter × length × inner diameter) according to Production Example 5 of DE-A10046928 (stoichiometry: Mo ₁₂ V ₃ W _1.2 Cu _2.4 O catalyst loading at _x ).

  From top to bottom, the first 175 cm is held isothermally by a salt bath C pumped countercurrently to the reaction gas. The second 175 cm is held at a constant temperature by a salt bath D pumped countercurrently to the reaction gas.

  Between the two oxidation stages, there is a bundle heat exchanger cooled by a salt bath, so that the product gas of the first oxidation stage can be cooled. There is a valve for the supply of molecular oxygen (purity> 99% by volume) prior to entry into the second dioxide stage.

The propylene load for catalyst loading in the first oxidation stage is selected to be 145 Nm ³ / h. The salt melt (53% by weight of KNO ₃ , 40% by weight of NaNO ₂ , 7% by weight of NaNO ₃ ) has the following entry temperature:
T _A = 324 ° C T _B = 328 ° C
T _C = 265 ° C T _D = 269 ° C
Molecular oxygen (176 ° C., 3.20 bar) is added to the product gas mixture of the first oxidation stage such that the molar ratio of O ₂ : acrolein in the product charge gas mixture for the second dioxide stage is 1.25. Supply by quantity. The acrolein load for the catalyst loading of the second stage is 140 Nl / l · h. The pressure at the inlet of the second oxidation stage is 1.61 bar. The reaction gas exits the intercooler at a temperature of 260 ° C. and the entry temperature of the charged gas mixture into the second oxidation stage is 258 ° C. The product gas mixture of the first oxidation stage has the following contents:
Acrolein 9.02% by volume,
Acrylic acid 0.61% by volume,
39.5% by volume of propane,
0.53% by volume of propylene,
H ₂ 9.49% by volume,
O ₂ 3.90% by volume,
H ₂ O 25.75% by volume and CO ₂ 8.77% by volume.

  The temperature of the product gas in the first oxidation stage is 335 ° C. before entering the aftercooler.

The product gas mixture of the second stage of oxidation (product gas B) has a temperature of 270 ° C. and a pressure of 1.55 bar and the following contents:
Acrolein 0.049% by volume,
9.12% by volume acrylic acid,
Acetic acid 0.26% by volume,
Propane 39.43% by volume,
0.53% by volume of propylene,
H ₂ 9.52% by volume,
O ₂ 2.94% by volume,
H ₂ O 26.10% by volume,
CO 0.72% by volume and CO ₂ 9.3% by volume.

  The product gas B is fractionally condensed in a bed column (separation zone I) as described in WO 2004/035514.

  For the residue combustion, 118 kg / h of a high boiling body (polyacrylic acid (Michael-adduct) or the like) is supplied as the first combustion substance.

  From the second collection bed on the feed of product gas B into the bed column, 235.7 kg / h of condensed crude acrylic acid having a temperature of 15 ° C. and 96.99% by weight of acrylic acid are withdrawn. As described in WO 2004/035514, this was suspended and crystallized after feeding a small amount of water, the suspended crystals were separated from the mother liquor in a hydraulic wash column, and the mother liquor was separated as described in WO 2004/035514. Reflux into the condensation column. The purity of the washed suspended crystals is> 99.87% by weight and is readily suitable for the production of water “superabsorbent” polymers for hygiene-range use.

The amount of acidic water condensate removed from the third collection bed on the feed port of product gas B into the condensation column and not refluxed into the condensation column is 18145 kg / h, having a temperature of 33 ° C. and the following contents:
Acrolein 0.11% by volume,
Acrylic acid 1.30% by volume,
Acetic acid 0.95% by volume and water 95.6% by volume.

At the top of the condensation column, the residual gas I53379Nm ³ / h leaves separation zone I with a temperature of 33 ° C. and a pressure of 1.20 bar and the following contents:
Acrolein 0.03% by volume,
Acrylic acid 0.02% by volume,
61.07% by volume of propane,
Propylene 0.82% by volume,
H ₂ 14.02% by volume,
4.5% by volume of O ₂
H ₂ O 2.00% by volume,
CO 1.13% by volume and CO ₂ 14.36% by volume.

  Residual gas I 0.084% by volume is discharged with the composition of residual gas I (as third combustion product).

The remaining amount of residual gas I (from now on linguistically simply referred to as “residual gas I”) is compressed from 1.20 bar to 3.20 bar in the first compression stage of the multistage radial compressor, At this time, the temperature of the residual gas I rises to 92 ° C. Next, the remaining gas I is divided into two halves with the same composition. One half is directly a part of the circulating gas I refluxed into the reaction zone A. The other half of the residual gas I is compressed from 3.20 bar to 5.80 bar in the second compression stage. At this time, heat is generated at 127 ° C. At this time, the condensate is not formed, but it is cooled to 78 ° C. by an indirect heat exchanger (the coolant is CO ₂ -washed out and molecular hydrogen is depleted by subsequent membrane separation, and The residual gas I, which was released in the expansion turbine at a temperature of 42 ° C. and a pressure of 4.25 bar). By indirect air cooling, the residual gas I is cooled from 78 ° C. to 54 ° C. without producing condensate. In another compression stage, the residual gas I is compressed from 5.80 bar to 12.0 bar, with an exotherm from 54 ° C. to 75 ° C.

The residual gas I thus compressed is introduced into the lower part of the packed column (separation zone II). At the top of this washing column, 18000 kg / h of a K ₂ CO ₃ -water solution having a temperature of 82 ° C. and containing phenothiazine as a polymerization inhibitor is charged. At the top of the washing column, CO ₂ -washing residue gas I leaks out with the following contents at a temperature of 12.00 bar and 85 ° C .:
Propane 65.94% by volume,
Propylene 0.89% by volume,
H ₂ 15.13% by volume,
4.91% by volume of O ₂
H ₂ O 2.16% by volume,
CO 1.22% by volume and CO ₂ 7.75% by volume.

CO ₂ -washing residual gas I (24528 Nm ³ / h) on a bundle of cast hose membranes (external pressure 0.1 bar; Typ BH polyimide membrane from Fa. UBE Industries Ltd.) It is introduced and discharged with the following contents (p = 12.00 bar; T = 85 ° C.):
66.1% by volume of propane,
Propylene 0.89% by volume,
H ₂ 15.03% by volume,
4.92% O _{2 by} volume,
H ₂ O 2.10% by volume,
CO 1.22% by volume and CO ₂ 7.7% by volume.

By the way, in the multistage expansion turbine, the residual gas I (24461 Nm ³ / h) is released from 12.00 bar to 4.25 bar and cooled from 85 ° C. to 42 ° C. CO ₂ -cleaned and H ₂ depleted residual gas I is generated by heat generation from 42 ° C. to 97 ° C. by indirect heat exchange with residual gas I compressed in the first stage originating from separation zone I Together with the other fraction of residual gas I remaining from the fractional condensation of gas B (CO ₂ -non-washed out), it is refluxed as circulating gas I into the reaction gas A-starting mixture in reaction zone A.

The permeate stream (67.2 Nm ³ / h) has the following contents:
Propane 0.12% by volume,
0.01% by volume propylene,
H ₂ 56.2% by volume,
H ₂ O 27.2% by volume,
CO 0.15% by volume and CO ₂ 12.39% by volume.

  This is supplied to a common combustion apparatus as the fourth combustion product.

At the bottom of the CO ₂ -washing column, an aqueous potassium hydrogen carbonate solution of 1846 kg / h is taken out at a temperature of 75 ° C. and heated from 75 ° C. to 100 ° C. by indirect heat exchange (as a heat exchanger, dissociated hydrogen carbonate A potassium solution is used), loaded on top of a second packed column (dissociation column). Steam 26 Nm ³ / h (144 ° C., 4 bar) is introduced in countercurrent at the bottom of the dissociation column. The gas stream leaking at the top of the column is cooled directly (coolant: condensate previously formed) and the resulting aqueous condensate is charged to the column as reflux. The leaked CO ₂ is supplied to the common combustion apparatus as the fifth combustion substance. From the dissociation column stream (32968 kg / h), 14968 kg / h of water is evaporated. The remaining amount (optionally post-neutralized with KOH) is refluxed as a wash solution onto the top of the CO ₂ -wash column.

Combustion materials 1-5 are combusted together in a combustor under a supply of air (17674 Nm ³ / h).

Claims (24)

To produce acrolein, or acrylic acid, or mixtures thereof from propane,
A) Feeding first reaction zone A with at least two gas feed streams containing propane (of which at least one stream contains fresh propane) to produce reaction gas A;
In reaction zone A, reaction gas A is passed through at least one catalyst bed where molecular hydrogen and propylene are produced by partially heterogeneously catalyzed dehydrogenation of propane,
The reaction zone A is supplied with molecular oxygen that oxidizes the molecular hydrogen contained in the reaction gas A in the reaction zone A to water vapor, and the reaction gas A contains molecular hydrogen, water vapor, propylene, and propane. Take out A,
B) In the reaction zone B, the product gas A taken out from the reaction zone A is at least one oxidation reaction together with the reaction gas B containing molecular hydrogen, water vapor, propane, propylene and molecular oxygen under the supply of molecular oxygen. Acrolein as a target product, or acrylic acid, or a mixture thereof, unreacted propane as a by-product by heterogeneously catalyzed partial gas phase oxidation of propylene contained therein Producing a product gas B containing molecular hydrogen, water vapor, carbon dioxide, and other subcomponents boiling below or above water,
C) The product gas B is discharged from the reaction zone B, the target product contained therein, water and secondary components boiling higher than water are separated therefrom in the first separation zone I, and unreacted propane, Carbon dioxide, molecular hydrogen, secondary components boiling below water and residual gas I optionally containing unreacted propylene and possibly unreacted molecular oxygen in reaction zone B,
D) As post-treatment means 1, in the second separation zone II, the carbon dioxide contained in the residual gas I is washed out, and the water still optionally contained in the residual gas I is optionally condensed,
As post-processing means 2, a partial amount of residual gas I is discharged,
Optionally, as post-treatment means 3, molecular hydrogen contained in the residual gas I is separated through a separation membrane in the third separation zone III, and optionally as post-treatment means 4, optionally contained in the residual gas I The molecular oxygen to be reduced is chemically reduced, the post-treatment means 1 to 4 being applied in any order, and E) remaining after the application of the post-treatment means 1 and 2 and optionally 3 and / or 4 In a process for producing a post-treated residual gas I containing unreacted propane by refluxing it into reaction zone A as at least one stream of at least two feed streams containing propane,
A molecular hydrogen amount M, which is produced in reaction zone A and is optionally fed to reaction zone A from at least 35 mol% but not more than 65 mol%, is oxidized to water vapor in reaction zone A. In the separation zone I, the amount of water that is at least 80 mol% of the total amount of water contained in the product gas B is separated from the product gas B, and is contained in the residual gas I released per unit of time. A method characterized in that the amount of propane present is not more than 20 mol% with respect to fresh propane fed to the process per unit of time . In reaction zone A, an amount of molecular hydrogen M, which is produced in reaction zone A and optionally supplied to reaction zone A, is at least 40 mol%, but not more than 60 mol%, is oxidized to water vapor. The method according to claim 1, wherein: In reaction zone A, an amount of molecular hydrogen M, which is produced in reaction zone A and optionally supplied to reaction zone A, is at least 45 mol% but not more than 55 mol%, is oxidized to water vapor. The method according to claim 1, wherein: In the reaction zone A, the molecular hydrogen amount M, which is 50 mol % of the total amount from the molecular hydrogen produced in the reaction zone A and optionally supplied to the reaction zone A , is oxidized to steam. Item 2. The method according to Item 1. The reaction gas B charged to the at least one oxidation reactor contains the following:
4-25% by volume propylene,
6-70% by volume of propane,
H ₂ O 5~60% by volume,
O ₂ 8 to 65% by volume and H ₂ 0.3 to 20% by volume
5. The method according to claim 1, comprising: The reaction gas B charged to the at least one oxidation reactor contains the following:
6-15% by volume propylene,
6-60% by volume of propane,
H ₂ O 5~30% by volume,
O2 _{8 to} 35% by volume and H2 _{2 to} 18% by volume
5. The method according to claim 1, comprising: Wherein the molar ratio V ₁ for propylene contained in the reaction gas B pulp propane is 1-9, A method according to claim 5 or 6. Wherein the molar ratio V ₁ for propylene contained in the reaction gas B pulp propane is 1-7, A method according to claim 5 or 6.   9. The source of molecular oxygen required in the reaction zone B is pure molecular oxygen or a mixture containing molecular oxygen and an inert gas, and containing 10% by volume or less of an inert gas. The method according to any one of the above.   10. The source of molecular oxygen required in the reaction zone A is characterized by using pure molecular oxygen or a mixture containing molecular oxygen and an inert gas and containing 10% by volume or less of an inert gas. The method according to any one of the above.   Air as a source of molecular oxygen required in reaction zone A, and pure molecular oxygen or a mixture containing molecular oxygen and an inert gas as sources of molecular oxygen required in reaction zone B, and containing 10% by volume or less of inert gas The method according to any one of claims 1 to 8, characterized in that is used. The reactive gas B to be loaded in at least one oxidation reactor, characterized in that it contains a CO ₂ 0.1 to 30 volume%, The method according to any one of claims 1 to 11. The reactive gas B to be loaded in at least one oxidation reactor, characterized in that it contains a CO ₂ 1 to 20 volume%, The method according to any one of claims 1 to 11.   14. The process according to claim 1, wherein the reaction gas B charged into at least one oxidation reactor contains molecular nitrogen ≦ 5% by volume.   The process according to claim 1, wherein the reaction zone A is formed adiabatically.   The process according to any one of claims 1 to 15, characterized in that the reaction zone A is implemented as a floorboard reactor. The post-treatment residual gas I refluxed in the reaction zone A contains molecular oxygen, water vapor, molecular hydrogen, CO and CO ₂ , according to any one of claims 1 to 16. Method. 18. The process according to claim 1, wherein the post-treatment residual gas I refluxed in the reaction zone A contains 5 to 15% by volume of CO ₂ . The method according to any one of claims 1 to 18, characterized in that the washing of CO ₂ from the residual gas I is performed with an aqueous solution containing potassium carbonate. The process according to any one of claims 1 to 19, characterized in that the scavenging of CO ₂ from the residual gas I is carried out at a pressure of 3 to 50 bar. In separation zone I, from the product gas B, and characterized in that the separation of water is at least 90 mole% of the total water volume that is contained in the product gas B, any one of claims 1 to 20 The method described in 1. In separation zone I, from the product gas B, and characterized in that the separation of water is at least 95 mole% of the total water volume that is contained in the product gas B, any one of claims 1 to 20 The method described in 1. The amount of propane contained in the residual gas I discharged per unit time is 10 mol% or less with respect to the amount of fresh propane supplied per unit time during the process. Item 23. The method according to any one of Items 1 to 22.   23. The amount of propane refluxed in the reaction zone A in the aftertreatment residual gas I is at least 90 mol% of the amount of propane contained in the product gas B. The method according to any one of the above.