BRPI0810143B1 - A PROCESS FOR STARTING HETEROGENEALLY CATALYST PARTIAL OXIDATION OF ACROLEIN TO ACRYLIC ACID OR METACROLEIN FOR METACRYLIC ACID ON A FIXED CATALYST LAYER - Google Patents

Description

The Process of Starting a Heterogenously Catalyzed Partial Gas-Oxidized Phase Oxidation for Acrylic Acid or Methacrylic Acid in a Fixed Catalyst Bed Acrolein to Acrylic Acid or Metaerolein to Methacrylic Acid in a fixed catalyst bed that is arranged in a tube bundle reactor in the reaction tubes of a vertical bundle of reaction tubes surrounded by a reactor jacket, both ends of the tubes. reaction tubes being opened and the upper end of each reaction tube end sealed in a through hole of an upper tube plate sealed in the upper portion of the reactor jacket and the end of the lower end sealed in a through hole of a reactor. bottom tube plate sealed at the bottom of the reactor jacket, the outside of the reaction tubes, the top and bottom tube plate and the reactor jacket together delimiting the surrounding space of the reaction tube and each of the two tube plates being covered by a reactor hood having at least one hole in which the In order to give rise to the start, a reaction gas inlet mixture comprising> 3 vol% acrolein or metaerolein as well as molecular oxygen is fed to the tube beam reactor reaction tubes through at least one orifice. , hereinafter referred to as E, in one of the two reactor hoods and the product gas mixture which results by partial gas phase oxidation of acrolein or metaerolein to acrylic acid or methacrylic acid in the course of passage through the fixed catalyst bed disposed in the reaction tubes and comprising acrylic acid or methacrylic acid is removed through at least one orifice of the other reactor hood while in In the tube bundle reactor, at least one liquid heat exchange means is conducted around the reaction tubes such that each of the two surfaces of the two tube plates facing each other is humidified by the liquid heat exchange means and At least one liquid heat exchange medium is conducted in the surrounding space of the reaction tube at a temperature IV11 '™ ^ 1 · and is conducted out of the surrounding space of the reaction tube again at a temperature. Acrylic acid and methacrylic acid they are reactive monomers that are suitable as such or in the form of their alkyl esters, for example, for the preparation of polymers that may find uses including adhesives or water absorbing materials (e.g. for use in the hygiene sector). It is known that acrylic acid and methacrylic acid can be prepared on an industrial scale by heterogeneously catalyzed partial gas phase oxidation of their acrolein and methacrolein precursor compounds, the fixed catalyst bed being arranged in the reaction tubes of a tube beam reactor. as described at the outset, while in the tube bundle reactor side jacket at least one liquid heat exchange means is conducted around the reaction tubes such that each of the two surfaces of the two tube plates face each other is moistened by a liquid heat exchange means (as, for example, EP-A 700893, DE-A 4,431,949, WO 03/057653, EP-A 1695954, WO 03/055835, WO 03/059857, WO 03 / 076373 and DE 699 15952 T2).

In general, the tube beam reactor components are made of steel. The use for steelmaking is either stainless steel (eg DIN 1.4541 or 1.4571 material (for DIN EN 10020)) and black steel or ferritic steel (eg DIN 1.0481, 1.0315 or 1.0425 material (for DIN EN 10020)). Often all pipe beam reactor components are made from the same type of steel. The space enclosed by the exterior of the reaction tubes, the two tube plates and the reactor jacket together within which the liquid heat exchange medium is conducted shall be referred to herein as the surrounding space of the reaction tube. In the simplest way, only one liquid heat exchange medium is conducted within the surrounding space of the reaction tube (hereinafter referred to as a single zone mode in the single zone tube beam reactor). The liquid heat exchange medium is conducted into the surrounding space of the reaction tube typically at its upper end or at its lower end with an inlet temperature through the holes in the reactor jacket and driven back from the surrounding space. the reaction tube at the opposite end with a TV * 1'11 outlet temperature through the holes in the reactor jacket.

As a result of the exothermicity of the partial oxidation of the gas phase during the partial oxidation performance, THsmki> THomimla. With the aid of a heat exchanger, heat is removed from a portion or all of the liquid heat exchange medium conducted outside the surrounding space of the reaction tube before being fed back into the surrounding space of the reaction tube with the heat exchanger. temperature THen, raila. In the surrounding space of the reaction tube, the liquid heat exchange medium may in principle be conducted around the reaction tubes in simple countercurrent or countercurrent to the reaction gas mixture flowing in the reaction tube. However, this can also be conducted around the reaction tubes in a sinuous manner with the aid of deflection discs, such that a co-current or countercurrent towards the reaction gas mixture flow in the reaction tubes merely exists in the surrounding space. of the total reaction tube. The heat exchange medium should be liquid under the conditions of use, that is, suitably from an application point of view, has a melting point in the range of 50 to 250 ° C, preferably 150 to 200 ° C. .

Such useful liquid heat exchange means includes, for example, salt melts such as potassium nitrate, potassium nitrite, sodium nitrite and / or sodium nitrate and also metal melts such as sodium, mercury and alloys. Different metals. However, it is also possible to use ionic liquids or heat carrier oils. The composition of the reaction gas inlet mixture, the loading of the fixed catalyst bed disposed in the reaction tubes with acrolein or methacrolein, the entry point of the heat exchange medium into the surrounding space of the reaction tube, the flow rate. From the heat exchanger medium, the fixed catalyst bed and other reaction conditions are generally selected such that the conversion of acrolein based on a simple reaction of the reaction gas mixture across the reaction tubes is> 90 mol. %, in many cases> 95 mol%, preferably> 98 mol% and the selectivity of the attached acrylic acid formation is> 90 niol% (or methacrolein conversion based on a simple reaction of the reaction gas mixture through the tubes > 50 mol%, in many cases> 60 mol% and in some cases> 70 mol% and the selectivity of the attached methacrylic acid formation is> 70 mol%, advantageously> 80 mol%). The difference THs1, ulu - THc ", ra0a is generally maintained from> 0 to 10 ° C, often from 2 to 8 ° C, in most cases from 3 to 6 ° C.

To improve the selectivity of target product formation, heterogeneously catalyzed partial gas phase oxidation of acrolein to acrylic acid or methacrolein to methacrylic acid can also be performed as a multiple zone method (eg double zone method) in a multiple zone tube beam reactor (for example on one in a dual zone tube beam reactor).

In this case, a plurality of (e.g. two) essentially separately spatially separated liquid heat exchange means (which are normally of the same type) are conducted within the surrounding space of the reaction tube (these may be , for example, separated by separating the tube plates which are inserted into the surrounding space of the reaction tube and have suitable through-holes for the reaction tubes). The longitudinal section of reaction tube in which the particular liquid heat exchange medium extends represents a temperature or reaction zone (the corresponding single zone tube beam reactor has only one reaction zone). Within the particular temperature zone, the liquid heat exchange medium may be conducted as in the single zone method (also relative to the flow direction of the reaction gas mixture). For the difference between Tsaida and Tien, the statements made for the single-zone method with respect to the individual temperature zone apply in essentially the same way.

A graphical distinction between a single zone and the double zone method (between a single zone tube beam reactor and a double zone tube reactor) is shown schematically and by way of example by Figures 1 and 2 thereof. document. Figures 3 and 4 show the configuration of the baffle discs (these typically comprise through-holes for reaction tubes).

Multiple zone methods are, for example, described in EP-A 1734030, DE-A 10313214, DE-A 10313213, DE-A 10313211, DE-A 10313208 and prior art cited in these documents. This is advantageous, in particular, when a loading of high acrolein or acmlein or methachroline from the fixed catalyst bed is selected. Loading of the fixed catalyst bed with the reaction gas mixture or a reaction gas mixture component is understood to mean the amount of reaction gas mixture or reaction gas mixture component in standard liters (Nl); of the volume that the corresponding amount of theoretically take the gaseous form at 0 ° and 1 atm) which is driven through one liter of fixed catalyst bed (beds of pure inert material are not included) per hour. The inlet mixture of the reaction gas by itself. in different procedures, be conducted from the top down or bottom up into the reaction tubes in the tube bundle reactor (i.e. at least one hole E may be disposed in the top or bottom of the reactor). The same applies to conducting the liquid heat exchange medium. The target product is removed from the product gas mixture obtained in a heterogeneously catalyzed partial gas phase oxidation of acrolein to acrylic acid or methacrolein to methacrylic acid in general using thermal separation procedures.

To this end, the target product is usually first converted to the condensed phase by condensation and / or absorption measurements. This is subsequently removed from it as a pure product typically using extraction, rectification and / or crystallization measurements. These are realized, inter alia, in the separation columns comprising separation internals (e.g., random packages, structured packages and / or trays). It is known from the prior art that heterogeneously catalyzed partial gas phase oxidation of acrolein to acrylic acid or methacrolein to methacrylic acid in the newly substituted fixed catalyst bed can generally be performed at temperatures of at least one medium. liquid heat exchange means such that when the liquid heat exchange means moistens the tube plate (hereinafter referred to herein as reactor plate E) covered by the reactor hood having at least one hole E (also referred to later herein) document as reactor hood E), the temperature of the liquid heat exchange medium is less than 290 ° C (hereinafter referred to as humidification temperature E (Tu '); this may be a THa', mdii or a THsalda). The temperature of the reaction gas inlet mixture at the inlet of at least one orifice E (hereinafter referred to as Tge), according to the prior art claims, may in principle be below the humidification temperature E (TnH). . This is possible due to the fact that the reaction tubes in the flow direction of the reaction gas mixture are. normally charged first with a long itudinal section of formed bodies which are inert with respect to the partial oxidation of the catalytically active section of the fixed catalyst bed with formed bodies having beginnings of catalytically active composition. In the course of flow through this inert section, the mixture The reaction gas inlet may then be heated to the temperature of the heat exchange medium flowing around the corresponding catalytically active reaction tube section.

It will be estimated that the reaction gas inlet mixture at the inlet of at least one orifice E (reaction gas inlet mixture is generally fed only through an orifice E and the product gas removed only through a hole in the other reactor hood, in principle, that is, although it is also possible in each case to use 2 or 3 or more such holes in the particular reactor hood for this purpose), it may, however. already be preheated to the temperature value Τηπ which corresponds essentially to the initial reaction temperature. This is a temperature at which catalyzed sparrow oxidation in the fixed catalyst bed disposed in the reaction tube begins. Since the heterogeneously catalyzed partial gas phase oxidation of acrolein to acrylic acid or methacrolein to methacrylic acid is a markedly exothermic reaction, the temperature of the reaction gas mixture in the reactive passageway through the fixed catalyst bed is. otherwise, normally different from the temperature of the liquid heat exchange medium flowing around the catalyst bed fixed outside the catalyst tubes. This is typically above the inlet temperature of the heat exchange medium in the corresponding reaction zone (temperature zone) and generally passes through an absolute maximum (maximum hot spot) near a reaction zone or previous declines of such an absolute maximum value proceeding from such an absolute maximum value (if appropriate by means of the additional relative maximum).

It is also known from the prior art that heterogeneously catalyzed partial gas phase oxidation of acrolein to acrylic acid or methacrolein to methacrylic acid can be carried out in one and the same fixed bed and catalyst over an extended period. Typical opening times are 1 year and over.

However, a disadvantage of such a long period operation on one and the same fixed catalyst bed is that, despite intermediate measures used for its regeneration (as, for example, DE-A 10350822), an increasing irreversible reduction in the quality of the fixed catalyst bed occurs at a certain operating time with increasing operating duration (as per, for example, DE-A 102004025445).

This increasing depletion of the quality of the catalyst bed fixed in the catalyst tubes accompanying the increasing operating time can at least for a certain operating time be counteracted by gradually increasing the inlet temperature of at least one medium. net heat exchange Τη * 111π “,: ι in the surrounding space of the reaction tube. This increase is also necessarily accompanied by an increase in THH which, with increasing operating time, leads to a temperature of at least one liquid heat exchange medium, where it humidifies reactor plate E. being 290 ° C and above. . For the predominant majority of catalysts suitable for the acrogenous catalyzed partial phase gas oxidation of acrolcin to acrylic acid or methacrolein to methacrylic acid, the partial oxidation process can still be operated with fully satisfactory conversions and target product selectivity even at TVin values. up to 350 ° C (for example 300 ° C or 31 ° C or 320 ° C or 33 () ÜC or 340 ° C) and more (for example 36 ° C or 370 ° C or 380 ° C or 390 ° C or 400 ° C). Tnen values above 400CC are, however, usually exceptional. The THC increase according to the prior art procedure (as, for example, DE-A 10350822) is not normally accompanied by a simultaneous increase of To1 of the reaction gas inlet mixture. Instead, To1 is normally left at the value with which the reaction gas inlet mixture was supplied to the tube beam reactor during partial oxidation performance in the newly replaced fixed catalyst bed.

Since this Tg1 is below 29Ü ° C, the surface temperature of the reactor plate E facing the reactor hood having hole E is also below 290 ° C in normal operation of heterogeneously gas phase partial oxidation. catalyzed (even when the THentratb of that heat exchange medium using reactor plate E is> 290 ° C), since this reactor plate surface is continuously cooled in continuous operation by a gas inlet mixture (flowing, generally at a flow rate of at least 1000 m3 Nnr7 (h * nr), often at least 1500 m3 Nm-7 (h * m2) or at least 2500 m3 Nm3 / (h * nr) in the cross-sectional area of the reactor plate E), However, the above is no longer the case when heterogeneously catalyzed partial gas phase oxidation should be discontinued. A possible type of reasons for the interruption includes, for example, all possible operational interruptions detailed in EP-A 1658893 and USA 2004/00015012.

However, the reason for the discontinuation may be that, in the course of removal of the target acrylic acid or methacrylic acid products in the columns comprising the separation internals and are used for this purpose, unwanted polymer formation occurs (both acrylic acid methacrylic acid have a marked tendency to unwanted free radical polymerization). In order to be able to remove it, the partial oxidation process is suitably from an application point of view interrupted.

In many cases, the interruption of partial oxidation is also accompanied by a reduction in the inlet temperature of at least one Thc, 11 ™ Iu liquid heat exchange medium to save energy during idle time.

In many cases, however, the temperature at which the at least one liquid heat exchange medium Tni: mr; Kí; i is fed into the surrounding space of the reaction tube is essentially retained during the partial oxidation interruption in order to maintain the readiness for immediate operation. At the same time, during the interruption of partial oxidation, no gaseous stream in all or most of a significantly smaller gaseous stream (generally less than one third, often less than one fifth of the reaction gas stream input mixture). ) as compared to the reaction gas input stream mixture is conducted through the tube beam reactor through at least one port E.

Since the applicant in these cases subsequently resumed partial oxidation, immediate stoppages of partial oxidation were often suddenly triggered consequently, caused by a rapid increase in temperature in the E reactor hood gas space. Prolonged investigation and analysis of these facts showed whereas in the case of the above described variants of partial oxidation interruptions compared to the situation in the case of regular operation of partial oxidation. there is generally heating of that surface of reactor plate E facing that reactor nozzle having hole E (hereinafter referred to as reactor plate surface E) in the course of the interruption.

When, however, the reaction gas inlet mixture comprising> 3 vol% acrolein or mctaerolein, in the course of resuming heterogeneously catalyzed partial gas phase oxidation, flows towards such a comparatively reactor plate surface E hot, this can trigger (with increasing probability with increasing reactor plate surface temperature E) an exothermic (thermal) homogeneous free radical oxidation of acrolein or methacrolein that can propagate in the opposite direction to the flow direction of the gas inlet mixture reaction (and comprises a thermal decomposition of acrolein or methacrolein). At least one thermoset protruding from the reactor E shell for safety reasons (when, for example, operational interruptions occur in the removal of the reaction heat from the regular heterogeneously catalyzed partial gas phase oxidation, this may go out of control and be interrupted by immediate shutdown (termination of the reaction gas mixture input stream according to EP-A 1658893 or according to US 2004/00015012) detects the reaction heat released with exothermic free radical oxidation and consequently triggers an immediate shutdown.

In reaction tubes loaded with catalyst bodies formed and / or inert bodies formed, such thermal exothermic homogeneous free radical oxidation of acrolein or methacrolein, even at temperatures TH '> 290 ° C, essentially does not occur, since the area The larger specific surface area of the formed catalyst bodies and / or normally formed inert bodies decontaminates the free radicals that form and, as a result, normally extinguishes the free radical chain reaction. Therefore, dispersion of thermal homogeneous free radical oxidation (including decomposition) does not normally extend in reaction tubes with the catalyst bed attached.

Therefore, it was an object of the present invention to provide a process for initiating a heterogeneously catalyzed partial gas phase oxidation of acrolein to acrylic acid or methacrolein to methacrylic acid in a fixed catalyst bed disposed in the reaction tubes of a tube beam reactor. , which at least totally reduces or avoids the immediate stoppages described above.

Accordingly, a process has been observed to begin a heterogeneously catalyzed partial gas phase oxidation of acrolein to acrylic acid or methacrolein to methacrylic acid in a fixed catalyst bed that is disposed in a tube beam reactor in the vertical beam reaction. of reaction tubes surrounded by a reactor jacket, both ends of the individual reaction tubes being opened and the upper end of each reaction tube end sealed in a through hole of an upper tube plate sealed at the top of the jacket of the reactor and the end of the bottom end sealed in a through hole of a bottom tube plate sealed at the bottom in the reactor jacket, the exterior of the reaction tubes, the top and bottom reactor plate and the reactor jacket together delimiting the surrounding space of the reaction tube, and each of the two tube plates being covered by a an actor having at least one orifice in which, in order to give rise to the start, a reaction gas inlet mixture comprising> 3% by volume of acrolein or methacrolein as well as molecular oxygen is fed to the reactor reaction tubes of tube bundle through at least one orifice, hereinafter referred to as E, into one of the two reactor hoods and the product gas mixture which results through partial gas phase oxidation of acrolein or methacrolein to acrylic acid or acid Methacrylic acid in the course of passage through the fixed catalyst bed disposed in the reaction tubes and comprising acrylic acid or methacrylic acid is removed through at least one hole in the other reactor hood, while in the tube bundle reactor side jacket, at least one liquid heat exchange means is conducted around the reaction tubes such that each of the two surfaces of the two tube plates facing each other is one. and the at least one liquid heat exchange means is conducted in the surrounding space of the reaction tube with a temperature THc ", ie it is conducted out of the surrounding space of the reaction tube again. At a time when, in order to give rise to the start, the reaction gas inlet mixture comprising> 3% by volume acrolein or methacrolein enters the reactor hood through at least one orifice E, the temperature THC of at least one liquid heat exchange means that moistens the reactor hood-covered tube plate having at least one orifice E, hereinafter referred to as reactor plate E, is at least 290 ° C. the reaction gas inlet mixture entering at least one orifice E (at the entrance of at least one orifice E) has a temperature of <285 ° C and the surface temperature of the reactor plate E covers the reactor hood having at least one orifice E, hereinafter referred to as reactor plate surface E. has a value of <285 ° C. The temperature Tp "1" * 1 of at least one liquid heat exchange medium (liquid heat exchange medium useful for the process according to the invention are all those named at the beginning of this document) which humidifies reactor plate E (this temperature is also referred to herein as ΤΗϋ'ιΙ ''; κ1ίϋ '·) may, in the course of the inventive beginning, also be> 29PC or> 292 ° C or> 293 ° C or> 294 ° C or> 295 ° Or> 296 ° C or> 297 ° C or> 298 ° C or> 299cC or> 300 ° C or> 30 PC or> 302 ° C or> 303 ° C or> 304 ° C. ™ 1 * 1,1 'in the process according to the invention may also be> 305 ° C or> 310 ° C or> 315 ° C or> 320 ° C. Even the values τΗοη, ί; ιιΙ; ι · Ιί of > 330 ° C and> 340 ° C are possible in the process according to the invention, especially when this is a heterogeneously catalyzed partial oxidation of methacroline to methacrylic acid. 400 ° C, often < 375 ° C and in many cases <350 350C It will be estimated that »in an inventive beginning, each simple of the above mentioned" relations> "is also possible for The. It will be estimated that in an inventive beginning Τιι0, ηιηΜ, £ ι-κ and THH can simultaneously correspond to a single simple of the "relations>" mentioned above. The temperature of the reaction gas inlet mixture at orifice inlet E, i.e. Tob, may, at the beginning of the invention with all THiiHTOfa.re Xhl values specified individually herein, be <285 ° C or <280 ° C or < 275 ° C or <27 ° C or <26 ° C or <250 ° C or <240 ° C or <220 ° C or <210 ° C or <200 ° C. Typically, T <; f at the inventive start will however be above (preferably at least 5 ° C above) the dew point of the reaction gas inlet mixture. Which is a temperature at which condensation manifestations (liquid droplet formation) exist with the selected composition of the reaction gas inlet mixture and at the working pressure selected in the reaction gas inlet mixture within a gas inlet mixture. reaction However, a reaction gas inlet mixture comprising liquid droplets is capable of damaging the fixed catalyst bed as it flows through. Low values for Tc, 1 · are generally advantageous. This is specifically true when the reaction tubes, in the outflow direction of the reaction gas inlet mixture, first have a bed of inert bodies formed. The length of such an inert bed may absorb up to 10% or up to 20% or up to 30% of the length of the reaction tube. Overall this will not be below 5% of this length. Tt; L is. preferably from 200 to 20 ° C. especially when spontaneous oxidation of acrolefin to acrylic acid or methacrolein to methacrylic acid is not linked to a previous partial oxidation of the olefin corresponding to acrolein or methacrolein, Tgb may also be from 90 to 150 ° C. This is also favorable according to The invention was first invented when the surface temperature E facing the reactor hood has at least one hole E, ie the surface of reactor plate E, is <280 ° C, preferably <275 ° C, advantageously < 27 ° C. especially advantageous <265 ° C, very especially advantageous <26 () ° C, especially preferable <255 ° C, very especially preferable <250 ° C or <245 ° C or <240 ° C, particularly favorable <235 ° C or <230 ° C or <225 ° C or <220 ° C or <215 ° C or <210 ° C or <205 ° C or <2 () 0 ° C. This surface temperature of the reactor plate E is also referred to herein as TnH. Since the surface temperature of reactor plate E on surface E need not necessarily be homogeneous in all operating states (surface temperatures at the center and periphery of the reactor plate accessible to the reaction gas are generally slightly in this document all in particular must be the highest of the two elevated temperatures found in a temperature measurement at the aforementioned periphery (in the outermost reaction tube circle) and in the center of the tube plate.

Advantageous beginnings according to the invention are thus, for example, those for which simultaneously: - (and, if appropriate, TWH) is> 290 ° C, TcEi is <285 ° C, Tbr is <280 ° C ; or - THmUada-li (and, if appropriate, Tw! £) is> 290 ° C, T (: K is <285 ° C, Τβη is <275 ° C; or - THCMrada'E (and, if appropriate, TWE ) is> 290 ° C, TtJE is <285 ° C, Tbe is <260 ° C, or - Tnefflfada'E (and if appropriate TWR) is> 290 ° C, TGH is <285 ° C, Tbe is < 255 ° C, or - T „« “- E (and, if appropriate, TWK) is> 290 ° C, TGK is <285 ° C, Tbe is <250 ° C, or - T„ en, rada'E ( and, if appropriate, Τ% .Β) is> 290 ° C, TGE is <285 ° C, Tbe is <245 ° C; 011 - THcmrada-E (and if appropriate TWE) is> 290 ° C, T (5E is <285 ° C, Tbh is <240 ° C; or - XHcn, RiKla'E (and, if appropriate. TWE) is> 290 ° C, TGE is <285 ° C, Tbh is <235 ° C; or - (H) rada (if appropriate) TwE) is> 290 ° C, Tge is <285 ° C, Tbe is <23 ° C, or - THcn, R 1 Kla-R (and, if appropriate). , TWR) is> 290 ° C, TGR is <285 ° C, Tbe is <225 ÜC; or - THc “, wla-R <and, if appropriate, TWR) is> 290 ° C, TGE is <280 ° C »Tbe is <280Â ° C; or - TV '' (and, if appropriate, TWR) is> 290 ° C, TGt: is <275 ° C, Tbe is <280 ° C; or - THen, rada'E (and, if appropriate. TWE) is> 290 ° C, TGE is <27 ° C, Tbe is <280 ° C; 011 - THcn, r: Kla-E (and, if appropriate, TwE) is> 290 ° C, TGE is <265 ° C, TBE is Or - THc ", ryd;," h (and, if appropriate, TwE) is> 290 ° C, TGli is <260 ° C; Tbe is <280 ° C; or - T, fi, fi, of -E (and, if appropriate, Twr) is> 290 ° C, TGE is <280 ° C, Tbe is <275 ° C, or - T „eB'rada'H (and, if appropriate, TWE) is> 290 ° C, TGH is <28FC, Tbe is <270 ° C, or - T „en, rada'K (and, if appropriate, TWE) is> 29 () ° C, TGE is <28PC, Tbe is <265 ° Or - THen, radi'E (and, if appropriate, TWE) is> 290 ° C, TGE is <28 ° C, Trh is <260 ° C, or - THctrtruJ, E (c, if appropriate) TWE) is> 290 ° C, T0E is <275ÜC, TBE is <275 ° C, or - (et If appropriate, TwE) is> 290 ° C, TGb is <275 ° C, Tbh is <270 ° C; - THc ,, 'r: acid, E (and, if appropriate, TwE) is> 290 ° C, TGE is <275 ° C, Tb is <265 ° C, or THeMrada-b (e, if ap * Twb) is> 290 ° C, T (; h is <275 ° C, Tbh is <260 ° C; or THcmrada.K if appropriate, TWE) is> 290 ° C, TGb is <270 ° C, TBK is <275 ° C; or ~ Twm '™ da‘E (and, if appropriate * TwE)> 290 ° C, TGE <270 ° C, Thh <270 ° C; or - THanMl (and, if appropriate * TwE) is> 290 ° C, TGE is <270 ° C, Tbe is <235 ° C; or - THc ,, r5Kla'E (and, if appropriate * Twc) is> 290 ° C, TGE is <270 ° C * Tbe is <260T.

In addition, the above triple temperatures (or, if appropriate, quadruple temperatures) are suitable for an advantageous inventive start, except that THaitra'liLlE (and, if appropriate, Th,:) is> 291 ° C or> 292 ° Or> 293 ° C or> 294 ° C or> 295 ° C or> 297 ° C or> 299 ° C or> 3005 ° C or> 305 ° C or> 315 ° C or> 320 ° C or> 33 ° C) or> 340 ° C * Suitably from an application point of view, the Tbe in the process according to the invention will be above the dew point (preferably at least 5 ° C above) of the mixture reaction gas inlet. The process according to the invention is almost generally primarily a process for resuming heterogeneously catalyzed partial gas phase oxidation of acrolein to acrylic acid or from methacrolein to methacrylic acid. Such a restarting process is not performed in a newly prepared fixed catalyst bed recently filled in the tube beam reactor. Instead, this is performed in a fixed catalyst bed in which heterogeneously catalyzed partial gas phase oxidation of acrolein to acrylic acid or methacrolein to methacrylic acid has been performed previously.

In this document, a reaction gas inlet mixture should be understood to mean a gas mixture comprising at least 3 volume%, preferably at least 4 volume% and more preferably at least 5 volume% (based on the total volume of the reaction gas inlet mixture) of acrolein or methacrolein. Typically, the reaction gas inlet mixture will comprise from 3 to 15% by volume, in many cases from 4 to 10% by volume, in many cases from 5 to 6 to 8% by volume, of acrolein or methacrolein. When gas mixtures comprising less than 3% by volume of acrolein or methacrolein are conducted through the fixed catalyst bed (such gases may also be used as "cooling gases" which will be mentioned later), this should also be taken into account. It is understood as a partial oxidation interruption as when no gas mixture is conducted through the fixed catalyst bed of the tube beam reactor. With increasing content of acrolein or methacrolein in the reaction gas inlet mixture (and with increasing loading of the fixed catalyst bed with the reaction gas inlet mixture or with acrolein or methacrolein in the reaction gas inlet mixture), The process according to the invention gains significance. Correspondingly, the higher the acrolein or methacrolein content in the reaction gas inlet mixture, the lower the Tbl and, if appropriate, also the Tob should be selected. Even in the case of increased oxygen content in the reaction input mixture, such lower values are recommended. The molar ratio of O? : acrolein in the reaction gas inlet mixture will usually be> 0.5, often> 1. Typically, this ratio will be in values of <3. Often, the CF: aerolein molar ratio in the reaction gas inlet mixture is 0.5 or 1 to 2 or 0.6 or 1 to 1.5. The molar ratio of CF: methacrolein in the reaction gas inlet mixture will often be from 1 or 1.5 to 3 or advantageously from 1.5 to 2.5. This may also be from 0.5 to 3.

In addition, a reaction gas inlet mixture in general will comprise at least one inert diluent gas which, minus O 2 and aerolein or methacrolein, in its entirety essentially forms the remainder of the reaction gas inlet mixture. In this document, this should be understood to mean those gases which, when the reaction gas inlet mixture is conducted through the fixed catalyst bed disposed in the tube beam reactor in the course of an inventive start, remain unchanged to an extent. at least 95 mol%, preferably to a point of at least 98 mol%, more preferably to a point of at least 99 mol% or to a point of at least 99.5 mol%.

Typical inert diluent gases are molecular nitrogen, vapor, noble gases, CO2 and saturated hydrocarbons such as methane, ethane, propane, butane and pentane as well as mixtures of portions or all of the aforementioned diluent gases.

Typically, the inert diluent gas in the reaction gas inlet mixture may consist of molecular nitrogen to a point of> 20% by volume or to a point of> 30% by volume or up to a atom of> 40% by volume or up to a point of> 50% by volume or a point of> 60% by volume or a point of> 70% by volume or a point of> 80% by volume or a point of> 90% by volume or up to > 95% by volume.

Especially, when the sparse oxidation of the acrolein (or methacrolein) gas phase is the second reaction stage of a two-stage heterogeneously catalyzed partial gas phase oxidation of propylene to acrylic acid, the inert diluent gas of the inlet mixture of the The reaction gas will often consist of H2O to a point of 5 to 25% by weight or to 20% by weight (6 formed first reaction stage and added if appropriate) and N: to a point of 70 to 90% by volume. .

Often, in the processes according to the invention, the reaction gas inlet mixture will have a volume ratio of acrolein (methacrolein): oxygen: vapor: other inert gas (Nl) from 1: (1 to 3): (0 to 20): (3 to 30), preferably from I: (1 to 3): (0.5 to 10): (7 to 10), or, in the case of acrolein, from I: (0.9 to 1.3 ): (2.5 to 3.5): (10 to 12) e. in the case of methacrolein. from 1: (1.5 to 2.5): (3 to 6): (10 to 15). The inlet pressure of the reaction gas inlet mixture at orifice inlet E (absolute pressure) may be below the standard pressure (for example, up to 50 kPa (0.5 bar)) or above the standard pressure. Typically, the working pressure already mentioned will be in the range 100 to 500 kPa (1 to 5 bar), often 100 to 300 kPa (1 to 3 bar). The mentioned working pressure will normally not exceed 10,000 kPa (100 bar). Loading of the fixed catalyst bed with acrolein or methacrolein may, at the inventive start, be> 10 Nl / (l * h) or> 20 Nl / (l * h) or> 30 Nl / (l * h) or> 40 Nl / (1 * h) or> 50 Nl / (l * h) or> 60 Ml / (l * h) or> 70 Nl / (l * h) or> 80 Nl / (l * h) or> 90 Nl / (l * h) or> 110 MI / (l * h) or> 130 Nl / (Mi) or> 180 Nl / (! * H) or> 240 Nl / (l * h) or> 300 Nl / (l * h) (but normally <600 Nl / (l * h '). The loading already mentioned may also be increased in stages in the course of the inventive beginning, as described, for example, in WO 2005/016861 for the beginning of a newly replaced fixed catalyst bed and in WO 2005/047226 for the initiation of a newly regenerated fixed catalyst bed, loading of the fixed catalyst bed with reaction gas inlet mixing can, in an inventive start, simultaneously to the structure already mentioned, typically be from 500 to 10,000 Nl / (l * h), usually from 1000 to 5000 Nl / (l * h), often from 1500 to 4000 Nl / (l 'h).

Otherwise, the inventive start is advantageously from an application point of view, realized such that no point along the catalyst tubes is at the reaction gas mixture temperature in the reaction tube greater than 80 °. C greater than the temperature of the heat exchange medium at the same length of reaction tube in the surrounding space of the reaction tube. Advantageously, the temperature difference already mentioned at each point along the reaction tubes is <70 ° C, often from 20 to 70 ° C; This temperature difference is preferably small.

In addition, the process according to the invention is suitable, inter alia, in the case of tube bundle reactors meeting the requirements according to DE 20 2006 014 116 U1. The reactor jacket typically has a cylindrical geometry with a circular cross section of the tube plates. Typical cross-sectional areas of tube plates (and also ballast plate E) are 3n to 8 () nr. often from 10 nr to 50 nr. The thickness of the tube plate (and also the reactor plate E) is generally 5 to 40 cm, often 8 to 30 cm or 10 to 20 cm. The larger the cross section of the pipe plate, the greater the thickness of the pipe plate that normally c.

The reaction tubes are (just like the other elements of the tube beam reactor), as already mentioned at the beginning of this document, generally made of steel. The inventive procedure is especially advantageous when the surface of reactor plate E is made of stainless steel (austenitic steel) or black steel (ferritic steel). The preferred material (also for the other tube beam reactor elements) is ferritic steel. The wall thickness of the reaction tubes is typically 1 to 3 mm. Its internal diameter is generally (uniformly) 10 to 50 mm or 20 to 30 mm, often 21 to 26 mm. The number of catalyst tubes (reaction tubes) accommodated in the tube beam reactor is generally at least 1000 or 3,000 or 5,000, preferably at least 10,000. Often, the number of reaction tubes accommodated in the beam beam reactor. tube is from 15,000 to 30,000 or to 40,000 or to 50,000. Tube bundle reactors having several reaction tubes above 50,000 are usually exceptional. Within the reactor jacket, the reaction tubes are normally arranged in homogeneous distribution, the distribution being appropriately selected such that the separation of the central inner axes from mutually closest catalyst tubes (reaction tubes) (the so-called catalyst tube height) it is 25 to 55 mm, often 35 to 45 mm (as, for example, EP-A 468 290).

Most advantageously in accordance with the invention, the tube beam reactor is made of the type 1.0425 ferritic steel (to DIN EN 10020). For reactor plates, reaction tubes and reactor hoods, DIN material steel number 1.0481 or 1.0315 are also frequently used and for reactor liner, in many cases DIN material steel number 1.0345. The interior of the reactor hood E facing the surface of reactor plate E is, however, advantageously from an application point of view, laminated with austenitic steel (preferably of type DIN 1.4541 or type 1.4571 (to DIN EN 10020). ). Typical lamination thicknesses are about 3 mm.

Especially in the case of tube beam reactors with a relatively large cross section of their tube plates, it is appropriate from an application point of view to leave a tubeless region in the center of the tube beam reactor rather than supporting the top tube plate in this region. The length of the reaction tubes typically extends to a few meters (a typical catalyst tube length is in the range of 1 to 8 m, often 2 to 6 m, in most cases 2 to 4 m).

Within reaction tubes, a differentiation is usually made between working tubes and thermal tubes. While working tubes are those reaction tubes in which partial oxidation to be performed in the real sense is performed, thermal tubes primarily serve the purpose of monitoring and controlling the reaction temperature together with the reaction tubes in a representative manner. for all work tubes. For this purpose, the thermal tubes usually comprise, in addition to the fixed catalyst bed, a thermal well that is conducted in the center along the thermal tube and is merely charged with a temperature sensor. In general, the number of thermal tubes in a tube beam reactor is much smaller than the number of working tubes. Normally, the number of thermal tubes is <20 (cf. EP-A 873 783 and EP-A 1 270 065). The statement that the reaction tubes are sealed in the through-holes in the upper and lower tube plates expresses that there are no passageways for the heat exchange medium between the outer wall of the and the hole wall (or the wall of the through hole or outside of the through hole). Such a seal may be performed, for example, as described in DE 20 2006 014 116 U1.

Correspondingly, the circumference of the upper and lower tube plates is also incorporated into the tube bundle reactor jacket such that there is no passageway for the heat exchange medium therebetween. In the upper pipe plate, however, there is generally a connection to the heat exchange medium pump that allows to degass the surrounding space of the reaction tube and to ensure that the liquid heat exchange medium wet the upper pipe plate ( see, for example, EP-A 987 057). Otherwise, a single zone tube beam reactor is preferably configured as described in DE-A 44 31 949.

When the process according to the invention is performed on a multiple zone tube beam reactor, this advantageously occurs from an application point of view, from a double zone tube beam reactor.

When that reaction zone that the reaction gas mixture flows first is referred to as reaction zone A and that reaction zone that the reaction gas mixture flows through next as reaction zone B, the inventive start is advantageously carried out. such that a difference between the highest reaction temperature occurs in reaction zone A in the beginning course (hereinafter referred to as TimxA) and the highest reaction temperature that occurs in the course of the beginning of reaction zone B (referred to below). Advantageously, the temperature difference already mentioned is> 3 ° C and <60 ° C, especially advantageous> 5 ° C and <40 ° C. This is generally the case when the difference between the heat exchange medium inlet temperature in reaction zone A, ie THA'cmraAl and the heat exchange medium inlet temperature in reaction zone B , that is, (i.e. THAACl ,, ruda - THBjI, Irad: i) is> -20 ° C and <0 ° C.

Otherwise, the procedure for tuning TmiMA and TmaxB may be as described in DE-A 1Ü3 13 208, EP-A 1 106 598, DE-A 103 13 213, DE-A 103 13 214, DE-A 103 13 211 and US 2006/0161019 A1 and the prior art cited in these documents. In particular, the dual zone tube beam reactors written in these documents are suitable for a process according to the invention. In general, in both reaction zones, the reaction gas and the heat exchanger medium, seen in the particular reaction zone, are conducted in co-current or countercurrent.

However, it is possible without any problem to use co-current operation in one of the two reaction zones and counter-current operation in the other of the two reaction zones. , just before the reaction gas inlet mixture is fed through port E to the pipe bundle reactor, for example, for conducting an essentially inert gas or gas mixture at such a flow rate and at a sufficiently high temperature. lowers through at least one hole E that the reactor plate surface E cools to the desired surface temperature Tnh. Such useful cooling gases C include, for example, steam, molecular nitrogen, carbon dioxide, molecular oxygen, noble gases, air and mixtures of any or all of the aforementioned portions of gases. In principle, the cooling gas mixture used should comprise <3 vol%, better <2 vol% or <1 vol% and especially advantageous no acrolein and no methacrolein.

For example, useful cooling gases C include mixtures of inert gas and molecular oxygen. These may comprise at least 1 or 2 vol%, preferably at least 3 vol% and more preferably at least 4 vol% oxygen. Overall, the oxygen content of the cooling gas mixture will however be <21% by volume. For example, such a useful cooling gas mixture is thin air. This is air devoid of oxygen.

An advantageous gas according to the invention is thin air consisting of 3 to 10 vol%, preferably 4 to 6 vol%, of molecular oxygen and, as the remainder, molecular nitrogen. Often, it is advantageous when the mixture of cooling gas as well as molecular oxygen and inert gas additionally comprises steam. Suitably from an application point of view, the cooling gas mixture comprises at least 0.1 vol%, often at least 0.5 vol% and often at least 1 vol% vapor. Typically, the vapor content of the cooling gas mixture is <75%; by volume, often <50% by volume, in many cases <25% by volume. Suitable gas mixtures such as the cooling gas may thus, for example, consist of 3 to 20% by volume of molecular oxygen, 1 to 75% by volume of steam and, as the remaining inert gas, such as N2 and CO2 (eg 97% by volume of air and 3% by volume of vapor).

It will be appreciated that useful cooling gases C will also be suitable while exerting a regenerative effect on the fixed catalyst bed. In other words, useful cooling gases are, for example, all gases and gas mixtures recommended as regenerative gases in DE-A 10350822, US 2006/0161019, WO 2005/042459 and EP-A 614872.

In principle, it is advantageous when the molar specific heat of the cooling gas C is as high as possible.

Of course, it is also possible to use the cycle gas or a mixture thereof (for example with air) which comprises one or more of the other cooling gases C already mentioned as a cooling gas. Cycle gas refers to the residual gas that remains when the target product has been more or less selectively removed (e.g., by absorption in a suitable solvent) from the heterogeneously catalyzed partial gas phase oxidation product gas mixture.

In general, this consists predominantly of inert diluent gases used for partial oxidation and vapor typically formed as a byproduct of partial oxidation and as carbon and vapor oxides formed by unwanted total oxidation. If appropriate, vapor depletion of an aqueous work of the product gas mixture (an aqueous absorber for the target product) may also be present in the cycle gas. In some cases it still comprises small amounts of oxygen not consumed in partial oxidation (residual oxygen) and / or unconverted organic starting compounds (cf. WO 2004/007405).

A heterogeneously catalyzed double-stage partial oxidation gas of propylene to acrylic acid that is suitable as a cooling gas prior to the inventive start may, for example, typically have the following contents: 3 to 5% by volume of oxygen molecular weight, 1 to 5% by volume of vapor, 0 to 3% by volume of carbon monoxide, 0 to 8% by volume of carbon dioxide. 0 to 2 vol% propane, 0 to 0.5 vol% propylene, 0 to 0.1 vol% acrolein, and 85 to 95 vol% molecular nitrogen.

An example of a cycle gas composition of the structure mentioned above is as follows: component volume% oxygen 3.3, water 1.5, carbon monoxide 0.8, carbon dioxide 1.6, propane 0.3, propene 0.3, acrolein 0.05 and nitrogen 92.15.

It will be appreciated that within the aforementioned cycle gas composition structure, the propane content may also be up to 50% by volume and the nitrogen content may be correspondingly lower.

In the case of a heterogeneously catalyzed partial gas phase oxidation for the preparation of halfcrylic acid, a typical cyclic gas composition, for example, has the following contents: 5 to 12% by volume of molecular oxygen, 10 to 25% 0 to 4 vol% carbon monoxide, 0 to 6 vol% carbon dioxide, 0 to 0.5 vol% isobutene, 0 to 0.2 vol% methacrolein and 50 90% by volume of molecular nitrogen. The temperature at which cooling gas C is fed into port E to the pipe bundle reactor must, according to the invention, necessarily be <285 ° C.

Often the temperature of the cooling gas already mentioned will be <280 ° C or <270 ° C or <260 ° C or <250 ° C or <24 ° C or <230 ° C or <220 ° C or <210 ° C or <200 ° C or <190 ° C or <180 ° C or <170 ° C or <160 ° C or <150 ° C or <140 ° C or <130 ° C or <120 ° C or <110CC or < 100 ° C. In principle, the temperature of the cooling gas may, however, also be <75 ° C or <50 ° C or <25 ° C and <() ° C. What is essential, however, is that the temperature of the cooling gas is above the dew point of the cooling gas. Suitably according to the invention, the temperature of the cooling gas is at least 5 ° C above the dew point of the cooling gas. The working pressure of the cooling gas as it passes through port E can be 100 kPa (1 bar), <100 kPa (1 bar) or> 100 kPa (1 bar). Often this working pressure of the cooling gas will be> 50 kPa (0.5 bar) to 500 kPa (5 bar), in many cases> 100 to 300 kPa (I to 3 bar), often> 130 kPa. (1.3) at <200 kPa (2 bar).

Advantageously, the cooling gas C is compressed in the same manner and subjected to a mechanical separation operation for the purpose of removing solid particles e.g. if appropriate, the condensate present therein as described by WO 2005/016852 and WO 2005/100290 to the reaction gas inlet mixture. The volume flow rate of the cooling gas may correspond to that of the reaction gas input mixture.

However, this may also be greater than the volume flow rate of the reaction gas input mixture or less than the volume flow rate of the reaction gas input mixture.

Advantageously according to the invention it is from 20 to 140%, advantageously from 40 to 120% and especially advantageous from 60 to 100% of that volume flow rate that the partial gas phase oxidation reaction gas inlet mixture heterogeneously catalyzed must be stopped first.

Based on the cross-sectional area of the reactor plate E (including through-holes), the flow ratio of cooling gas C is. suitably from an application point of view, from 500 to 3000 .Nm3 / (h * m2), often from 750 to 1500 Nm3 / (h * nr). The surface temperature of the reactor plate E can be monitored with the aid of at least one thermo-dipstick bond on the surface of reactor plate E.

Advantageously according to the invention, the supply of cooling gas is terminated when the temperature 1V; considered for the surface of reactor plate E has been reached.

The higher the THCMrttda, K and the lower the flow of cooling gels, of course, the longer their cooling gas delivery duration must be.

Since the cooling gas C also flows through the reaction tubes, the cooling of the reactor plate surface E may, in an undesirable manner, also be accompanied by the cooling of the heat exchange medium conducted through the surrounding space of the reaction tube. corresponding reaction (section) (in these cases the inventive start configurations may result in that THcmrad * “bea Τηβ are different from each other; in general, the THh in these cases will be below THcnlnu, IL, i). Suitably in accordance with the invention, attempts are made to avoid this. If appropriate, a general electric heater integrated into a tube beam reactor unit will be used to counteract any excessive cooling of the heat exchange medium that fits therein.

When heterogeneously catalyzed partial gas phase oxidation of acrolefin to acrylic acid or methacrolein to methacrylic acid is in the second stage oxidation in a two-stage process or heterogeneously catalyzed partial gas phase oxidation of propylene (or propane) via acrolein to acrylic acid or, for example, isobutene (or tert-butanol or tert-butanol or isobutane methyl ether) via methacrolein to methacrylic acid, this second oxidation stage will often be carried out as a tube-tube beam reactor. single zone or independent multiple zone in a tandem reactor arrangement (series connection of two tube beam reactors). Suitably according to the invention, in this case, an intermediate (or post-cooler) cooler, which may also be integrated into the first tube bundle reactor, is disposed between the tube bundle reactor for the first oxidation stage and the tube beam reactor for the second oxidation stage. In this case, the cooling gas is normally fed to the reactor plate surface E of the second tube beam reactor via the first tube beam reactor (i.e. through their reaction tubes), which in general , has the same design as the second tube beam reactor. Since the temperature of at least one heat exchange medium in the first tube beam reactor is normally above that of the at least one heat exchange medium in the second tube beam reactor, the cooling gas cooling temperature to the reactor plate surface E is then simply adjusted to the intermediate cooler described above.

Suitably from an application point of view, the procedure will be such that in the first tube bundle reactor also, with respect to the surface of analogous reactor plate E *, cooling to a temperature below that THcntrad! L'b * Analogous (and, if appropriate, THf) of the heat exchange medium that humidifies this reactor plate is realized. However, the situation in the first oxidation stage is significantly less critical than in the second oxidation stage.

Advantageously, at the onset of the first oxidation state also, the surface temperature of reactor plate E * will be at least 5 ° C, preferably at least 1 ° C. Especially preferably at least 20 or at least 30 ° C, most especially preferably at least 40 or at least 50 ° C below that temperature as the heat exchange medium humidifies this reactor plate has a THcn, radaJ * * (and if It will be estimated that at the beginning of the first oxidation stage, the surface temperature of the E * reactor plate may also be at least 60 or 70 ° C or at least 80 or 90 ° C, in some cases even at least 1 ° C) or at least 150 ° C and below that temperature than the heat exchange medium that humidifies this reactor plate E * has a THcntrjKiiLH; (and, if appropriate, Th1 '*). Correspondingly, the temperature of the reaction gas inlet mixture for the propylene-catalyzed partially catalyzed oxidation of propylene to acrolcine or isobutene to methacrolein will also be at least 5 ° C or at least 10 ° C or at least 20 ° C or at least 30 ° C or at least 4 ° C or at least 50 ° C or at least 60 ° C or at least 70 ° C or at least 80 ° C or at least 90 ° C or at least 100 ° C below the temperature THi'mraiIiLt "(and, if appropriate, ΤηεΊ of that heat exchange medium that humidifies the reactor plate E *. The temperature (and, if appropriate, Tnbl of that heat exchange medium that humidifies the reactor plate) The E * reactor may generally be from 300 to 38 ° C, often from 310 to 360 ° C and in many cases from 320 to 340 ° C.

In the first oxidation stage also, the catalyst bed fixed in the reaction tubes in general will first have, in the flow direction, an inert bed which serves to heat the reaction gas inlet mixture.

Since, in a two-stage partial oxidation in a tandem tube beam reactor arrangement as described above, the product gas mixture of the first oxidation stage, after cooling in the post-chiller as required, forms the mixture of reaction gas inlet for the second oxidation stage directly as such (in which case the oxygen requirement for the second oxidation stage is already part of the reaction gas inlet mixture for the first oxidation stage as the so-called primary oxygen ) and / or after the addition of air (secondary air) in a mixer and an interruption in partial oxidation in the second oxidation stage is necessarily accompanied by an interruption of the first oxidation stage, the onset of the second oxidation stage in these cases normal. It is also always associated with the onset of the first oxidation stage.

Suitably from an application point of view, this is understood according to the observations in WO 2004/007405, WO 2005/016861, WO 2004/085362, WO 2004/085369, WO 2004/085363, WO 2004/085367 , WO 2004/085368, WO 2005/047224 and WO 2005/042459 such that the requirements in this application are met for an inventive start especially of the second oxidation stage.

Advantageously, in an inventive start, the procedure will be to experiment with the required cooling of the reactor plate surface E initially with the cycle gas or with the cycle gas and vapor in a mixture (this type of temperature adjustment Tb11 required according to according to the invention can be adjusted immediately upon the beginning of the partial oxidation interruption, it will be estimated that the regeneration gas may first be conducted through the tube beam reactor system according to, for example, DE-A 10350822 or DE- 10350812 for interruption of partial oxidation). The cycle rate of the gas flow and its temperature will be such that the desired temperature Tb1 is achieved without significantly impairing the circulation of the temperature control of the at least one liquid heat exchange medium. Prior to the beginning of the latter, the cycle gas flow rate or the gas / vapor mixture flow rate will be adjusted to the magnitude corresponding to its fraction of gas volume flow rate in the gas stream inlet mixture. of reaction in the inventive beginning.

Subsequently, the airflow required for the inventive start (in general, the oxygen source used is air; in principle, although it is possible to use any mixtures of oxygen and inert gas or even pure oxygen as the oxygen source) e. Finally, the reagent stream will be turned on. It will be estimated that Tbe can in principle also be adjusted using a mixture of cycle gas and air to which the reactant stream is finally changed. When the inventive initiation is carried out as the beginning of the second oxidation stage in a double-stage partial oxidation of, for example, propylene (via acrolein) to acrylic acid or, for example, isobutene (via methacrolein) to methacrylic acid in a tandem tube beam reactor arrangement, the gas streams mentioned above, as described above, are fed to the first oxidation stage tube beam reactor and supplemented, if appropriate by secondary air, via of its reaction tubes and intermediate cooler, the surface of reactor plate E. The process according to the invention is especially suitable for restarting after an interruption of the relevant partial oxidation of at least 5 minutes or at least 10 minutes or at least at least 15 minutes or at least 20 minutes or at least 30 minutes or at least 1 hour or at least 2 hours or at least 3 hours or at least 4 hours or at least 6 hours or at least 7 hours or at least 8 hours or at least 9 hours or at least 10 hours or at least 11 hours or at least 12 hours, or more.

After the inventive adjustment of the reactor plate surface temperature E and, if appropriate, E *, in the event of interruptions of the relevant partial oxidation of up to 12 hours, the inventive initiation of partial oxidation, suitably from a standpoint. application will essentially be experienced with that composition of the reaction gas inlet mixture and loading of the particular fixed catalyst bed with this mixture with which partial oxidation was operated immediately prior to its interruption, even when, within the interruption period, According to DE-A 103 50 812, DE-A 103 50 822 and WO 2005/047226, a regenerating molecular oxygen comprising gas stream is conducted through the particular fixed catalyst bed.

When the interruption period is above 12 hours and the procedure during this period occurs as in DE-Â 103 50 822, DE-A 10350812 and WO 2005/047226, the reaction gas inlet mixture composition and progress loading of the fixed catalyst bed with this mixture, suitably from an application point of view, will be practiced as described in DE-A 103 37 788. The risk of an unwanted thermally initiated homogeneous cxothermal combustion of acrolein or methacrolein present in the Reaction gas inlet mixture propagating from the surface of the reactor plate E in the opposite direction to the flow direction of the reaction gas inlet mixture may, in addition to the use of the inventive start, be reduced by application to the reaction plate. reactor E, a bed and inert material whose thermal conductivity is much lower (eg ceramic material) than that of the material whose reactor plate E is manufactured. The inert bed preferably has a high heat capacity.

Useful materials for such an inert bed are in principle all those materials which will be recommended hereinafter as support body materials for the preparation of coated catalysts. A preferred inert material from an application point of view for such an inert bed is, for example, CeramTec steatite C220. The geometric formed bodies useful for the aforementioned inert bed include, for example, spheres, cylinders and / or rings whose longest dimension (longest straight line connecting two points on its surface) may be, for example, from 2 to 40 mm, preferably from 5 to 12 mm or from 50 to 200 mm. Suitably, from an application point of view, the aforementioned inert bed, however, will not absorb more than 40% by volume of the internal volume of the reactor hood covering it. A corresponding inert bed may be applied to the corresponding E * reactor plate of a first upstream extension stage. The curvature of the reactor hoods may, for example, be torispherical in accordance with DIN 28013 or semielipsoidal in accordance with DIN 28011. In the simplest case the reactor hood may also be reduced to an open cylindrical transition leading to for example to a cooling powder or leads in addition to a cooling powder. In principle, the transition from the reactor jacket to the reactor hood (which may be fluid in the exceptional case) is configured in a gas-tight manner. In addition to the measures already described, in an inventive start, the reactor plate E can also be thermally insulated against the heat exchange medium which moistens it with the aid of a material having only low thermal conductivity. Such useful materials include, for example, liquid heat exchange means alone that essentially does not flow. This can be implemented, for example, by mounting a baffle plate on the side of the tube plate facing the heat exchange medium just before the tube plate (the reactor plate E), in addition to which the heat exchange medium liquid heat towards the reactor plate E is essentially in the rest.

Useful catalysts and inert bodies formed for loading the reaction tubes of a first upstream oxidation stage as well as a second oxidation stage to be initiated according to the invention include all those recommended prior to the art (e.g. cited in this document) for partial oxidation.

These are especially those recommended in DE A 103 50 822, WO 98/12167, DE-A 43 29 907, WO 2005/030393, DE 10 2004 025 445 EP-A 700 893 EP-A 700 714 EP-A 758 562 EP-A 1 388 533 and DE-A 103 51 269. Loading of the reaction tubes themselves can also be performed as recommended in these documents. In general, loading is effected such that the volume specific activity outside the reaction tubes increases in the direction of the reaction tube flow.

Useful inert diluent bodies formed and / or support bodies (or support bodies formed) by coated catalysts (where active composition is applied: in contrast, unsupported catalysts consist essentially of active composition only) include aluminum oxides porous and non-porous, silicon oxide, thorium dioxide, zirconium dioxide, steatite silicon carbide (eg CeramTec Type C 220) or silicates such as magnesium silicate or aluminum silicate. Once of the catalyst antibodies formed, their longest dimension may be from 1 to 20 mm, often from 2 to 15 mm and in many cases from 3 to 4 to 10 or to 8 or 6 mm.

Where the acrolein or methacrolein required for the process according to the invention is generated in a heterogeneously catalyzed partial upstream oxidation of propylene or, for example, isobuene, suitable catalysts are those whose active composition is a multi-membered oxide of formula 1 Mül2BiaFebXc, Xd2Xe-Xj4On (I) where X1 - nickel and / or cobalt, X2 = thallium, an alkali metal and / or an alkaline earth melon, X '= zinc, phosphorus, arsenic, boron, tin antimony, cerium. lead, vanadium, chromium and / or tungsten, X4 = silicon, aluminum, titanium and / or zirconium, a = 0.2 to 5 b = 0.01 to 5, c = 0 to 10, d = 0 to 2, and = 0 to 8, f = OalOe n = a number that is determined by the valence and frequency of elements in I other than oxygen.

Descriptions of the preparation of unsupported hanging cone catalysts (e.g. rings) and coated catalysts (e.g. rings or beads) can be observed, for example, in WO 02/30569, WO 2005/030393, in RD 2005 research disclosure. -497012, DE-A 10 2007 005 602 and DE-A 102007004961. In the case of ring geometry, the outside diameter may be, for example, 2 to 10 mm, or 2 to 8 mm, or 4 8 mm or 2 to 4 mm (the same applies to ball geometries). The length of this ring geometry may also be from 2 to 10 mm, or from 2 to 8 mm, or from 4 to 8 mm. The wall thickness of such ring geometries is approximately 1 to 3 mm. The aforementioned geometries are relevant especially in the case of unsupported catalysts. A particularly preferred ring geometry (especially in the case of unsupported catalysts) is, for example, the geometry of outer diameter E 5 mm x length L 3 mm x internal diameter! 2 mm.

Other unsupported catalytic ring geometries of unsupported multi-metal oxide (I) E x W x I are 5 mm x 2 mm x 2 mm, or 5 mm x 3 mm x 3 mm, or 5.5 min x 3 geometries mm x 3.5 mm, or 6 mm x 3 mm x 4 mm, or 6.5 mm x 3 mm x 4.5 mm, or 7 mm x 3 mm x 5 mm, or 7 mm x 7 mm x 3 mm , or 7mm x 7mm x 4mm.

All of these unsupported multi-metal oxide catalyst ring (I) geometries (or multi-metal oxide catalysts (1) in general) are suitable both for the catalytic spontaneous oxidation of propylene to aerolein in the gas phase and for partial oxidation. catalyst of isobutene or tert-butanol or the tert-butanol metal ester to methacrolein in the gas phase.

With respect to the active stoichiometric compositions of general formula I, the stoichiometric coefficient b is preferably from 2 to 4, the stoichiometric coefficient is preferably from 3 to 1.0, the stoichiometric coefficient is preferably from 0.02 to 2, the stoichiometric coefficient is preferably from 0 to 5 and the stoichiometric coefficient advantageously from 0.5 or 1 to 10. More preferably, the stoichiometric coefficients mentioned above are simultaneously within the preferred range mentioned.

In addition, X1 is preferably cobalt, X2 is preferably K, Cs and / or Sr, more preferably K, X3 is preferably tungsten, zinc and / or phosphors and X4 and preferably Si. Especially preferable, the variables X1 X4 simultaneously have the definitions. mentioned above.

The annular (spherical) catalyst bodies are suitably from a point of application of the aspect, diluted with inert annular (spherical) antibodies to bring about a structuring of catalyst loading activity in the catalyst tube.

For heterogeneously catalyzed partial gas phase oxidation to prepare acrolein or mettaolern, the catalyst loading in the reaction tube with the annularly-shaped antibodies described above is preferably homogeneously configured with only one type of unsupported catalyst rings in the full length of the reaction tube. or structured as follows.

At the inlet of the reaction tube (in the direction of reaction gas mixture flow), for a length of 10 to 60%. preferably from 10 to 50%, more preferably from 20 to 40% and more preferably from 25 to 35% (i.e. for a length of from 0.70 to 1.50 m, preferably from 0.90 to 1, 20 m), in each case of the total length of the catalytically active catalyst loading in the reaction tube, is placed in a homogenized mixture of only one type of unsupported annular catalysts mentioned above and only one type of inert annular antibody (both formed body types preferably have the same ring geometry), where the weight ratio of the formed diluent antibodies (the volume densities of formed catalyst antibodies and formed diluent antibodies generally differ only slightly) is usually from 5 to 40%. %; by weight, or from 10 to 40%; by weight, or from 20 to 40%; by weight, or 25 to 35% by weight. Downstream of this first section of the load is then advantageously arranged to the end of the length of the catalyst load (i.e. for a length of from 1.00 to 3.00 m or from 1.00 to 2.70 m, preferably of 1.40 to 3.00 m or dc 2.00 to 3.00 m) a bed of only one type of unsupported annular catalysts diluted only to a lesser degree (than in the first section) with only one type of inert antibody in annular form, or, more preferably, a single (undiluted) bed thereof, appetites a type of unsupported annular catalyst. Of course, it is also possible to select a homogeneous dilution over a total reaction length of all. The catalyst bed will be configured in a corresponding manner when the geometries are spherical.

Otherwise, heterogeneously catalyzed partial gas phase oxidation of propylene to acrolein or isobutene to methacrolein may be performed in a tube beam reactor having one or more temperature zones as described prior to the art (as, for example, WO 2005/03093, DE-A 10 2007 005 602 and DE-A 10 2004 025 445 and of the art cited in these documents and in the present application).

Active compositions suitable for the catalyst antibodies in geometric form for heterogeneously catalyzed partial gas phase oxidation of methacrolein to methacrylic acid include multi-element oxide of general formula II.

Mo, 2Pa VhX 'cX2t | X "cSbfRegSiA, (II) where X1 = potassium, rubidium and / or cesium, X2 = copper and / or silver, χί _ cerium, boron, zirconium, manganese and / or bismuth, a = 0 , 5 to 3, b = 0.01 to 3, c = 0.2 to 3, d = 0.01 to 2, e = 0 to 2, f = 0.01 to 2, g = 0 to 1, h = 0 or 0.001 to 0.5 and n = an integer which is determined by the valenea and frequency of elements in II other than oxygen.Preference is given to the above-mentioned formed catalyst antibodies as well as unsupported annular catalysts as obtained by For example, the procedure described in EP-A 467 144. Useful ring geometries include especially the individual geometries recommended in EP-A 467 144 and also those recommended with respect to multi-element oxide I in the present application. where E x W x I = 7 mm x 7 mm x 3 mm (see also DE-A 10 2007 ÜÜ5 602).

A structured dilution with inert antibodies in annular form may, for example, be performed as described for the case of heterogeneously catalyzed partial oxidation of propylene to acrolein. Otherwise, the conditions of the partial oxidation process described in EP-A 467 144 and DE-A 10 2007 005 602 may be used.

For heterogeneously catalyzed partial gas phase oxidation of acrolein to acrylic acid, the active multi-element oxide compositions useful for the catalyst antibodies formed for use in the process according to the invention are advantageously those of the general formula 111 Mo 12Vax1 lvX \: X1X4eX (III) wherein the variables are each defined as follows: X '= W. Nb, Ta, Cre / or Ce, X2 = Cu, Ni, Co, Fe, Mn and / or Zn, X' = Sb and / or Bi, XJ = one or more alkali metals (Li, Na, K, Rb, Cs) and / or H, X5 = one or more alkaline earth metals (Mg. Ca, Sr, Ba), X6 = Si , Al, Ti and / or Zr, a = 1 to 6 b = 0.2 to 4, c = 0 to 18, preferably 0.5 to 18, d = 0 to 40, e = 0 to 2, f = 0 at 4, g = 0 to 40 and n = a number that is determined by the valence and frequency of IR elements other than oxygen.

Vantajosumenle. These formed catalyst antibodies are spherical or annular coated catalysts, as obtained, for example, according to DE-A 10 2004 025 445, DE-A 10 350 822, DE-A 10 2007 010 422, US 2006/0205978 and EP- A7I4 700 and prior to the technique cited in these documents.

Useful ring geometries or ball geometries are especially the individual geometries recommended in the aforementioned documents. A preferred ring geometry is that where E x W x I = 7 mm x 3 mm x 4 mm for the precursor annular support bodies. The coating thickness of the active composition may be from 10 to 1000 μηι, preferably from 50 to 500 pm and more preferably from 150 to 250 μιη. Favorable coating thicknesses are those of exemplary embodiments of EP-A 714 700.

For heterogeneously catalyzed partial gas phase oxidation of acetyl to acrylic acid, the catalyst loading in the reaction tube is preferably uniformly structured with only one type of catalyst rings coated over the full length of the reaction tube, or structured as follows.

At the inlet of the reaction tube (in the direction of the reaction gas mixture flow) for a length of 10 to 60%, preferably 10 to 50% - more preferably 20 to 40% and more preferably 25 to 35%. (i.e. for a length of from 0.70 to 1.50 m, preferably from 0.90 to 1.20 m), in each case the total length of the catalytically active catalyst loading in the reaction tube is placed in a homogenized mixture of only one type of coated annular catalyst mentioned above and only one type of inert annular antibody (both types of formed bodies preferably have the same ring geometry), where the weight ratio of the diluent antibodies formed ( volume densities of catalyst antibodies formed and diluent antibodies formed generally differ only slightly) is usually 5 to 40 wt%, or 10 to 40 wt%, or 20 to 40 wt%, or from 25 to 35% by weight. Downstream of this first section of the load is then advantageously arranged to the end of the catalyst load length (i.e. for a length of from 2.00 to 3.00 m, preferably from 2.50 to 3.00 m). ) a bed of only one type of unsupported annular catalysts diluted only to a lesser extent (than in the first section) with only one type of inert annular antibody and / or, more preferably a single (undiluted) bed of the same. , only one type of annular coated catalyst. The fixed catalyst bed will be configured in a corresponding manner when the geometry of the coated catalyst is spherical.

Otherwise, heterogeneously catalyzed partial gas phase oxidation of acrolein to acrylic acid may be performed in a tube beam reactor having one or more temperature zones as described prior to the art (as, for example, DE-A 10). 2004 025 445. DE-A 103 50 822, DE-A 10 2007 010422 EP-A 700 893. US 2006/0205978 and EP-A 714 7Ü0 (prior to the technique cited in these documents).

According to the present invention above thus it relates not least to processes for the preparation of acrylic acid or methacrylic acid which comprises a process according to the invention. The process according to the invention is particularly relevant when the reaction gas inlet mixture and the heat exchange medium that moistens the reactor plate surface E are conducted in the countercurrent seen in the tube beam reactor. However, it is also of significance in the case of such a cocurrent flow thereof.

In general, the fixed catalyst bed for both the inventive partial oxidation and the preceding partial oxidation stage will be configured such that when they are initiated as described in EP-A 990636 and EP-A 1106598 both Hot spot formation regarding its thermal sensitivity are at a minimum. Otherwise, the reactant conversions and selectivity of product formation when partial oxidations are initiated (based on the simple passage through the fixed catalyst bed) correspond essentially to those in the partial oxidation regulatory operation.

The numbers in figures 1 to 4 have the following meanings: Figure 1: shows a scheme of a single zone tube beam reactor Figure 2: shows a scheme of a dual zone tube beam reactor Figures 3, 4: shows Usable baffle plate combination schemes 1 = reaction gas inlet mixture 2 = product gas mixture 3 = orifice E

4 = surface of reactor plate E 5 = lower reaction plate 6 = reactor hood E 7 = lower reactor hood 8 = motor for heat exchanger medium pump 9 = reaction tube 10: removal of heat exchanger medium heat with Ths "'Jí' 11 = supplying the heat exchange medium with THenni, di1] 2 = supplying a portion of the heat exchange medium for cooling 13 = around the cooled portion of the heat exchanger 14 = jacket of reactor 15 = space around reaction tube 16 '17> = baffle plates for heat exchanger medium 18 J

19 = E reactor plate

A = reaction zone A

B = reaction zone B 20 = tube plate separation In particular, the present invention relates to a process for resuming heterogeneously catalyzed partial gas phase oxidation, the operation of which tends to be interrupted in advance from acrolein to acid or methacrolein to methacrylic acid in a fixed catalyst bed that is disposed in a tube bundle reactor in the reaction tubes of a vertical bundle of reaction tubes surrounded by a reactor jacket, both ends of the individual reaction tubes being opened and the upper end of each reaction tube end sealed in a through hole of a upper tube plate sealed at the top in the reactor jacket and the end of the lower end sealed in a through hole in a lower tube plate sealed at the bottom of the reactor jacket, the outside of the reaction tubes, the upper and lower pipe reactor together delimiting the surrounding space of the reaction tube and each of the two tube plates being covered by a reactor hood having at least one hole in which, in order to begin the restart as well as maintain operation, an inlet mixture dc reaction gas comprising> 3 vol% acrolein or methacrolein as well as molecular oxygen is fed to the tube beam reactor reaction tubes through at least one orifice, hereinafter referred to as E in one of the two hoods of the reactor and the gas mixture of the product which results by partial gas phase oxidation of acrolein or methacrolein to acrylic acid or methacrylic acid in the course of passage through the fixed catalyst bed disposed in the reaction tubes and comprises acrylic acid or methacrylic acid is removed through at least one hole in the other reactor hood while at least one The liquid heat exchange medium is conducted around the reaction tubes such that each of the two surfaces of the two tube plate liners are moistened by a liquid heat exchange medium and at least one heat exchange medium. The liquid is conducted in the surrounding space of the reaction tube with a temperature TH «mrad« and is driven out of the surrounding space of the reaction tube again with a temperature Tifm, where A) in the period t1 when partial gas phase oxidation heterogeneously catalyzed was interrupted, the temperature Τη0, '1γ ·' "Ι; '. hereinafter referred to as Th1 '' '' '' of at least one liquid heat exchange medium which moistens the reactor plate E is at least 290 ° C, and B) in the period t> cm. In order to begin resuming heterogeneously catalyzed partial gas phase oxidation, the reaction gas inlet mixture comprising> 3 vol% acrolein or methacrolein enters the reactor hood through at least one orifice E at THL temperature, mraJ: i, hereinafter referred to as Τη0, “γ: κΙ" '1 of at least one liquid heat exchange medium which moistens the reactor plate E is at least 290 ° C, the reaction gas input mixture which enter at least one hole E has a temperature Tge of <285 ° C and the temperature Tr ['- of the reactor plate surface E has a value of <285 ° C as a result of the fact that between the period t | co period ti, a cooling of the C gas whose TrE temperature was <285 ° C was conducted at least temporarily through at least one E-hole.

It will be appreciated that in the present proceeding for the restart the following conditions may also be fulfilled at the same time: - Tn “bl]> 292 ° C, T [| mla '1> 292 ° G TGn <285 ° C, Tbe <285 ° C, Tce <285 ° C; or - T,, “b · 1J> 294CIC, T (I'in, rada-1> 294 ° C, TGK <285 TBK <285 ° C, TCE <285 ° C; or - T, fMCüda'lJ> 296 ° C, Ti 1> 296 ° C, T (, H <285 ° C, Tf <E 285 ° C, TCE <285 ° C; or - T | 1 "'1> 298 ° C, TGE <285 ° C, TBK <285 ° C, TCE <285 ° C; or - THeMr £ Kla · u> 3üü ° C, THcn, rada · 1> 30ü ° C, ToE <285 ° C, TBE <285 ° C, TCE <285 ° C.

Overall, Thuí''hI: l v and Tnc, ‘, ru <J: 1'1 in the cases mentioned above will be <400 ° C, often <375 ° C and in some cases <350 ° C. Without regard to the other temperatures in the above mentioned structure, TGh may also be <280 ° C, or <270 ° C, or <260 ° C, or <250 ° C, or <240 ° C. Similarly, Tbe within a structure mentioned above, without regard to other temperatures, may be <280 ° C, or <275 ° C, or <270 ° C, or <265 ° C, or <255 ° C, or <250 ° C, or <245 ° C, or <240 ° C. Likewise, T4, without regard to the other temperatures in the above structure, may also be <280 ° C, or <275 ° C, or <270 ° C, or <265 ° C, or <260 ° C, or <255 ° C, or <250 ° C, or <245 ° C, or <240 ° C.

Comparative examples and examples with reference to a heterogeneously catalyzed double-phase partial gas phase oxidation of propylene to acrylic acid in two series connected single zone tube reactors A) Description of the general conditions of process I. The first stage of the reaction Heat exchange medium used: Salt melt consists of 60 wt% potassium nitrate and 40 wt% sodium nitrate.

Reaction Tube Material: DIN Material Ferrous Steel

number I.Ü48L

Dimensions of reaction tubes: length 3200 mm; internal diameter 25 mm; outside diameter 30 mm (wall thickness: 2.5 mm). Number of reaction tubes in the tube bundle: 25500.

Reactor: Cylindrical container (Ferrous steel of DIN materials number 1.0345) with an external diameter of 6800 mm; shirt wall thickness = 1.8 cm at the middle, thick at 2.5 cm at the top and bottom; annular vertical tube bundle with a free central space. Diameter of free central space: 1000 mm. Distance of external reaction tubes from container wall: 150 mm. Distribution of homogeneous reaction tube in the tube bundle (6 equidistant close tubes per reaction tube).

Reaction tube height: 38 mm.

The ends of the reaction tubes were secured with the seal in the holes of the 125 mm plate thickness tube plates and their open holes in a reactor hood that covers the upper reactor plate and is connected to the container at the upper end and end. lower in cylindrical transition to cooler poses. The upper tube plate is the E * reactor plate. The covering reactor hood has an E * hole (in the form of a gas inlet tip) with a diameter of 1020 mm. The tube plates and the other elements of the tube beam reactor were made from D1N material critical steel number 1.0481. In each case, a thermolinker was admitted or introduced to the E * reactor plate surface (in the circle of the outer reaction tube) and to the top of the reactor hood (the E * reactor hood). The upper part of the reactor hood (total wall thickness = 20 mm) was placed inside type 1.4571 stainless steel (per DIN ΕΝ 10020) (rolling thickness: 3 mm).

Supply of heat exchange medium to the tube bundle: The tube bundle was divided into 4 equidistant longitudinal sections (in each case 730 mm) (zones) by three deflecting discs (thickness in each case 10 mm) mounted successively between the plates. of pipes in the longitudinal direction of this. The tiny and top of the deflector disc has ring geometry with an inner ring diameter of 1000 mm and the outer ring diameter extended up to and sealed to the container wall. The reaction tubes were checked on the deflector discs without sealing. Instead, a slot having a slot width of <0.5 mm was left such that the transverse flow rate of the salt melt within a zone was substantially constant. The median deflector disc was circular and extended to the tubing. external reactions of the tube bundle. Circulation of the salt melt was accomplished by two salt pumps each providing one half of the length of the tube bundle.

The pumps inject the salt melt into an annular channel mounted around the reactor liner, which distributes the salt melt around the circumference of the vessel. The salt melt passes through windows present in the reactor liner in the tiny longitudinal section to the beam. pipe. The then fluid salt melt, ordered by the baffle plates, subsequently from the outer interior, from the outermost part, from the outer interior, from the outermost portion.

In an essentially winding way seen on the container from the deep top. Salt melt is collected through windows mounted around the circumference of the container in a longitudinal section of the upper part (salt melt leaves the surrounding space of the reaction tube at a temperature Th l, SJ, K, a) in a channel. upper annular ring mounted around the reactor jacket and was, after cooling to the original inlet temperature injected back into the lower annular channel by the pumps. The reaction gas inlet mixture 1 was an air mixture. chemical grade propylene and gas cycle.

Reactor charge: salt melt and reaction gas mixture were conducted in countercurrent seen in the reactor. The salt melt enters the deep part, the reaction gas mixture through the hole E * in the upper part. The inlet temperature of the salt melt was Ieilratla. The outlet temperature of the salt melt was Til. Pump output was 6200 m3 salt melt / hour. Reaction gas inlet mixture 1 was fed to the reactor with a temperature of To1 '' '1 when it is passed through the EL port Fixed catalyst bed • propylene loading: This was 1Nl / (1 · h ).

Load of reaction tube with fixed catalyst bed 1 (from top to bottom): zone A: 50 cm Preliminary bed of 7 mm x geometry steatite rings. 3 mm x 4 mm (outside diameter x length x inside diameter) Zone B: 100 cm Catalytic loading with a homogeneous mixture of 30% by weight of geometry steatite rings 5 mm x 3 ne x 2 mm (outside diameter x length x diameter and 70% by weight of an unsupported annular catalyst which is similar to the unsupported catalyst EUC 3 of WO 2005/030393, was prepared using TIMREX T 44 from Timcal AG (Bodio, Switzerland) as auxiliary graphite and, without taking considering the graphite still present, has the stoichiometry Mo i. 2 B i W2 C 05 jFe 2.94S i], 59 KcmbOx with ring geometry E x W x I = 5 mm x 3 mm x 2 mm.

Zone C; 170 cm Catapult load only with unsupported annular catalyst (5 mm x 3 mm x 2 mm = outside diameter x length x inside diameter) used for zone B.

Thermal tubes (these number 10 and were evenly distributed in the central region of the tube bundle) were configured and loaded as follows in order to monitor the temperature in the reaction tubes in a representative manner.

Each of the 10 thermal tubes has a central thermal well with 40 temperature measuring points (ie each thermal tube comprises 40 thermoligors which have been integrated in different lengths into a thermal well and thus forming a multi-thermoliter with which the temperature must be determined simultaneously at different times within the heat pipe). 20 of 40 temperature measurement points in each case are present in the region of the first meter of the active section of the fixed catalyst bed (in the direction of reaction gas mixture flow). The inside diameter of a thermal tube was 29 mm. The wall thickness and pipe material were configured as for the worked pipes. The outside diameter of the thermal well was 10 mm.

The thermal tubes were filled as follows: A thermal tube was filled with zone B unsupported annular catalyst. In addition, the longest 0.5 to 5 mm piece of catalyst obtained from the annular unsupported catalyst was filled in the thermal tube. The catalyst piece was filled to the homogeneous distribution in the total active section of the fixed catalyst bed of the particular heat pipe such that the pressure drop of the reaction gas mixture in the flow stroke through the heat pipe corresponds to no gas flow stroke. of reaction through a worked tube (based on active section of fixed catalyst bed (ie excluding inert sections) from 5 to 30% by weight of catalyst piece was required for purpose in thermal tube) * In the same period the height of the particular total supply of the inert and active sections at work and thermal tubes were equalized and the ratio of the total amount of active composition present in the tube to the heat transfer area of the working tube and thermal tubes were adjusted to essentially the same value. . II. Intermediate cooling The gaseous mixture of the product comprising acrolein 1 leading to the first stage of the reaction with the temperature corresponding to an inlet temperature of the salt melt th> was, for the purpose of intermediate cooling, conducting through a heat exchanger. single-zone tube beam made of erythritic steel and cooled with the salt melt composed of 60 wt% potassium nitrate and 40 wt% sodium nitrate, which was directly lowered into the lower tube plate of the beam reactor first stage reaction tube, The distance from the lower tube plate of the tube beam reactor from an upper tube plate of the chiller was 10 cm. The salt melt and the product gas mixture were conducted countercurrently seen in the heat exchanger. The salt bath itself fluid in the same manner as in the first stage of the zone one tube fixed bed reactor in one way. winding around the cooling tubes through the gas mixture of product 1 was passed. The length of the cooling tubes was 1.65 m, their inside diameter was 2.6 cm and their wall thickness was 2.5 mm. . The cooling tubes number 8000. The outer diameter of the heat exchanger was 6.8 m; The wall thickness corresponds to that of the reactor. These were evenly distributed in the cross section with homogeneous tube height.

Stainless steel coils whose visually corresponding cross section to those cooling tubes were introduced into the cooling tube inlet (in the flow direction). Their length was 700 mm to 1000 mm (alternatively, the cooling tubes can be filled with large rings of inert material). These serve to improve heat transfer. The gaseous mixture of the product comprising acrolein 1 leaves the immediate cooler with the temperature Tcz <HU. Subsequently, compressed air (secondary air) having a temperature of 140 ° C was added in such an amount that the oxygen content of the product 2 gas mixture was 3.0 vol%, which resulted in the composition of an inlet mixture. of reaction gas 2 to the second reaction stage.

This was fed with this temperature ToL2 into the hole E of the upper portion of the second stage reaction beam tube fixed beam reactor reactor recess. IIL The Second Stage of the Reaction An identical single-zone tube beam fixed bed reactor in the design of that of the first stage except that it has an upper and lower reactor hood was used. This upper reactor plate is the reactor plate E with the reactor plate surface E covering the upper reactor hood E, The composition of a reaction gas 2 inlet mixture consists of a first stage product gas mixture. of reaction and the secondary air.

Reactor charge: salt melt and reaction gas mixture were conducted in countercurrent seen in the reactor. The salt melt entering the deep layer, the reaction mixture at the top. The inlet temperature of the salt melt was YH2, cmratb. This existing temperature was Τπ2Λ; ,, ί1; TH2.solid - T2. Tmr, "la foj> Q and <2 ° C. Pump output was 6200 nf salt melt / hour. The reaction gas inlet mixture 2 was fed to the reactor at a temperature of T g T0: when it is passed through port E. Charging of acrolein * from the fixed catalyst bed 2: This was dc L2 Nl / (> h) . The reaction tube load with Fixed catalyst bed 2 (from the descending top) was: Zone A: 20 cm preliminary bed of geometry slantite rings 7 mm x 3 mm x 4 mm (outer diameter x length x inner diameter ). Zone B: 100 cm catalyst load of a homogeneous mixture of 30% by weight of 7 mm x 3 mm x 4 mm stcatite rings (outside diameter x length x inside diameter) and 70% by weight of annular coated catalysts (approximately 7 mm x 3 mm x 4 mm) DE 10 2004 025 445 CA with the active composition M012V_1W1.2Cu2.4Ox and an active composition content of 20% by weight. Zone C: 200 cm of acceptable loading of the annular coated catalyst (approximately 7 mm x 3 mm x 4 mm) from zone B.

The thermal tubes (these number 10 and were evenly distributed in the central region of the tube bundle) were configured and loaded as follows in order to monitor the temperature in the reaction tubes in a representative manner.

Each of the 10 thermal tubes has a central thermal well with 40 temperature measuring points (ie each thermal tube comprises 40 thermoligators that have been integrated in different lengths into a thermal well and thus forming a multi-thermoliter with which the temperature must be determined. simultaneously at different heights within the thermal tube). 20 of 40 temperature measurement points in each case were present in the region of the first meter of the active section of the fixed catalyst bed (in the direction of the reaction gas mixture flow). The inside diameter of a thermal tube was 29 mm. The wall thickness and pipe material were configured as for the worked pipes. The external diameter of the thermal well was 10 mm.

The thermal tubes were filled as follows: A thermal tube was filled with the prepared annular coated catalyst. In addition, two gaps of spherical coated catalysts were filled into the thermal tube (same active composition as the annular coated catalyst; the diameter of two types of C220 backing support balls (CeramTec) was 2 to 3 mm and 4 to 5 mm; a content of the active composition in both cases was 20% by weight (preparation was performed as described by the annular coated catalyst except that the binder is a corresponding amount of water). The spherical coated catalysts were filled in homogeneous distribution in the total active section of the fixed catalyst bed of the particular heat pipe such that the pressure drop of the reaction gas mixture when it is passed through the corresponding heat tube that when the reaction gas mixture is passed through a worked tube (based on the active section of the fixed catalyst bed (ie excluding inert sections), a total of 5 to 40%; by weight of the spherical coated catalysts were required for the purpose in the thermal tube), No At the same time, the height of the particular total supply of inert and working sections and thermal tubes was equalized and the ratio of the total amount of active composition present in the tube to the heat transfer area of the working tube and thermal tubes was adjusted to the same. same value, A gas mixture of product 2 obtained in the second stage of the reaction was conducted through the bottom reactor hood of the tube beam reactor and issues for this work. Conversion was monitored and controlled at two reaction stages overall with reference to residual propylene and acrolein contents in a gas mixture of product 2, B) Results (years of constant operation refers to an L1 of 130 Nl / (l * h) and the composition of a specified reaction gas inlet mixture with respect to this charge) 1. Starting from zone one partial oxidation after the fixed catalyst bed 1 in the first reaction stage has been replaced by a catalyst bed fresh fixed 1 while the fixed catalyst bed 2 in the second reaction stage was replaced by one year of operation (including regeneration according to DE-A 10350822 c DE-A 10351269) previously by a fresh fixed catalyst bed 2 (comparative examples the life of the fixed catalyst bed 1 is generally longer than that of the fixed catalyst bed 2, which is because two fixed catalyst beds are generally exchanged at different times). However, the fixed catalyst bed 2 may, according to DE-A 10350822, also be recently regenerated.

Table 1 below indicates the conditions for starting in the operating period t [hj. Cp is a conversion of propylene based on the simple passage of the reaction gas mixture through two oxidation stages. CAC is a conversion of acrolein in the simple passage of the reaction gas mixture through the second reaction stage.

In addition, Table 1 shows a composition of a reaction gas inlet mixture 1. In Table: c (Pen) = this propylene content by volume%; cfOi) = this oxygen content in% by volume; c (HiO) = this vapor content by volume%; c (CO) = this carbon monoxide content by volume%; CICCO :) = this carbon dioxide content by volume%. c (Ni) = this nitrogen content by volume%. The selectivity of acrolein formation in the first stage of the reaction was constant in the range of 88 to 92 mol%. The selectivity of acrylic acid by product formation in the first stage of the reaction was constant in the range of 3 to 7 mol%. The selectivity of acrylic acid formation in two reaction stages was constant in the range of 88 to 92 mol%.

An acrolein content originating from the gas cycle content of a reaction gas inlet mixture (which was <0.1% by volume) was neglected.

The charges L1 and L2 are understood to mean L ± 5 Nl / (l * h). When a regular operating load (which is desired for the steady state) L1 is <190 NI / (l * h), the operating values may be kept in a steady state by obtaining this target load. Cp should, however, not be increased to 96.2 mol% before the 20th day of operation. The reaction gas inlet pressure was constant from 180 to 340 kPa (1800 to 3400 mbar) in the first reaction stage and constant from 150 to 280 kPa (1500 to 2800 mbar) in the second reaction stage.

Table 1 (Part I) The data in the last Itnlm refer to the operation nu / ona two on each of the reaction statue dots, Th1 CIII, J * 1'1, and in each case having | Vrimeira / ona of temperature in the direction of flow uncle aunts Jc reaction.

The starting conditions mentioned above can also be used when both the fixed catalyst bed 1 and the fixed catalyst bed 2 have recently been loaded. This can also be used when the fixed catalyst bed I has been recently charged and, in the fixed catalyst bed 2 according to DE-A 10232748 and WO 2004/009525, only a portion has been replaced by a fixed catalyst bed load. fresh. 2. The start-up described in B) 1. may be performed in a corresponding manner with secondary air measured when both the fixed catalyst bed 1 and the fixed catalyst bed 2 are arranged in a dual zone tube beam reactor such as WO 2004/085369, but with an intermediate cooler arranged between the two reaction stages. Under the prerequisite of a fixed catalyst bed and heating zone configuration as specified in the working examples of the above mentioned document (except using the catalysts of B) 1. of the present application), the gas mixture of the inlet compositions of the reaction and loading changes in the period of operation will be configured in a corresponding manner as described in B) 1.0 same applies to Cp and CAc. The inlet temperature of the salt melt in the first reaction zone of the first reaction stage in the flow direction c is selected such that Cp in this first reaction zone is 65 to 70 mol% (it is usually below the corresponding TH1'c "lr;“, a in zone one operation). The inlet temperature of the salt melt in the first reaction zone of the second reaction stage in the flow direction is selected such that CAc in this first reaction zone is 80 to 85 ° C. mole% (usually below the corresponding TH2x,) miiji in zone 1 operation) 3. Partial zone 1 oxidation start after the fixed catalyst bed 2 is completely replaced by a fresh catalyst bed or, as described in WO 2004/009525, partially replaced while from the waste catalyst bed I in the first stage of the reaction was replaced one year of operation previously by a fresh fixed catalyst bed 1 (comparative examples; catalyst bed fixed r 1 may, however, be recently regenerated according to WO 2005/047224).

Table 2 below indicates the conditions for starting in the operating period t) hf Similar to Table 1, it also shows a composition of a reaction gas inlet mixture 1. The selectivity of acrolein formation in the first reaction stage is constant in range 88 to 92¾ in mol. Selectivity of acrylic acid by product formation in the first stage of the reaction was constant in the range of 3 to 7 mol%. The selectivity of acrylic acid formation in two reaction stages was constant in the range of 88 to 92 mol%.

The acrolein content that originates from the gas cycle content of a reaction gas input mixture 1 (which was <0.1% by volume) was neglected.

The loads L1 and L3 are understood to mean L ± 5 Nl / (l * h). When a regular operating load (which is desired for the steady state) is <190 Nl / (l * h), the operating values may be kept in a steady state in the course of starting by obtaining this target load. . The reaction gas inlet pressure was constant from 180 to 340 kPa (1800 to 3400 mbar) in the first reaction stage and constant from 150 to 280 kPa (1500 to 2800 mbar) in the second reaction stage.

Table 2 (Part I) _________________________________ ____________________ _ The data in the last few lines of the two lines will operate at two in each of the two stages of the reaction.

Th1 '"" * 1'1 and Th' '' '' "" u refer in each case to a first / one temperature in the direction of the reaction gas. 4. The start-up described in B) 3. may be performed in a corresponding manner with secondary air measured when both the fixed catalyst bed 1 and the fixed catalyst bed 2 are arranged in a dual zone tube beam reactor as described. WO 2004/085369, but with an intermediate cooler arranged between the two reaction stages. Under the prerequisite of a fixed catalyst bed as specified in the working examples of the above-mentioned document - and temperature zone configuration (except using the catalysts of B) 1. of the present application), the inlet composition of the gas mixture reaction time and the load changes will be set over the period of operation in a manner corresponding to that in B) 3. The same applies to C1 'and AC \ The inlet temperature in the first reaction zone of the first reaction stage in the direction The flow rate is selected such that Cp in this first reaction zone is 65 to 70 mol% (it is usually below the corresponding Th1 at zone one operation). The inlet temperature of the salt melt in the first reaction zone of the second reaction stage in the flow direction is selected such that C v in this first reaction zone is 80 to 85 mol% (is usually below the corresponding τγ, 2'4 ·· "1" ·· '^ in zone one operation), 5. Partial oxidation of zone one starting after catalyst bed 1 and catalyst bed 2 (including regeneration according to WO 2005/042459) were operated together in one year and have only recently been regenerated (comparative examples). Prior to the last regeneration, TH, c, l, r was 332 ° C and TH2, c, r was 266 ° C.

Table 3 below indicates the conditions for starting at operating period t [h]. Similar to Table 1, it also shows a composition of a reaction gas inlet mixture 1. The selectivity of acrolefin formation in the first reaction stage was constant in the range of 88 to 92 mol%. The selectivity of acrylic acid by product formation in the first stage of the reaction was constant in the range of 3 to 7 mol%. The selectivity of acrylic acid formation in two reaction stages was constant in the range 88 to 92% = mol. The aerolein content that originates from the gas cycle content of a reaction gas input mixture 1 (which was <0.1% by volume) was neglected.

Loads L1 and L2 are understood to mean L ± 5 Nl / (l * h). When a regular operating load (which is desired for the steady state) is> 190 NI / (l * h), the values of operation can be kept in a steady state in the course of the start of obtaining this target load. The reaction gas inlet pressure was constant from 180 to 340 kPa (1800 to 3400 mbar) in the first reaction stage and constant from 150 to 280 kPa (1500 to 2800 mbar) in the second reaction stage. The starting procedure described below can generally be used when two fixed catalyst beds have already been operated for a prolonged period (with intermediate regeneration) and have been essentially regenerated.

Table 3 (Part 1) The data in the last few lines refer to the operation in two cm each of the two reaction stages. F · οι "· κΐ.ι tf j-ji-.i-niriiiuretea In each case a first / one temperature is in the direction of the reaction gas Πιιο. 6. The starting described in B) 5. may be performed in a corresponding manner with measured secondary air when both the fixed catalyst bed 1 and the fixed catalyst bed 2 are arranged in a dual zone tube beam reactor as described in WO 2004/085369, but with an intermediate cooler. Under the prerequisite of a fixed catalyst bed as specified in the working examples of the above-mentioned document - and temperature zone configuration (except using the catalysts of B) 1. of this application) , the inlet composition of the reaction gas mixture and the load changes will be configured over the period of operation in a manner corresponding to that in B) 3. The same applies to Cp and CAt. The inlet temperature in the first reaction zone from pr The first stage of the reaction in the flow direction is selected such that Cp in this first reaction zone is 65 to 70 mol% (it is usually below the corresponding TH1 ratio in the operation of zone one). The inlet temperature of the salt melt in the first reaction zone of the second reaction stage in the flow direction is selected such that CA1 in this first reaction zone is 80 to 85 mole% (normally below the corresponding one in operation). (Examples) Starting of partial oxidation of zone one after the fixed catalyst bed 1 and fixed catalyst bed 2 (including regeneration according to WO 2005/042459) were operated together within four years and the oxidation The partial shutdown was interrupted for 2 hours due to an operational failure. During the shutdown, a gaseous mixture composed of 4 vol% Cb and 96 vol% N2 was conducted through the arrangement of the reactor arranged with a fixed 1 N catalyst bed loading of 1 N / hr. This inlet temperature in orifice E * was 150 ° C. An inlet in port E has a temperature of 240 ° C. The THUmr: useful salt melt temperatures of 348 ° C and 299 ° C that exist prior to integration were maintained during the interruption.

Subsequently, partial oxidation will be continued again with L1 = 130 Nl / (! * H) and e (Pen) = 6.3 mol%, c (O2) = 11.0 mol% and as the essentially remaining molecular nitrogen (T (1-1 k = 348 ° C; Tir en, tllda = 299 ° C; Tl = -1 = 300 ° C; (λ = 0.1 ° C)). The temperature of the surface of reactor plate E was 298 ° C. The production plant was coated immediately, caused by a rapid rise in temperature in the R reactor hood gas space

Thereafter, for a period of 15 minutes, a vapor of 35.0ÜÜ Nm Vh of nitrogen was conducted through the first stage E of the reaction through the arrangement of the arranged reactor. An inlet in port E has a temperature of 220 ° C. The temperatures of the TH1mixed and TH2x1Urad'1 salt melt were kept unchanged. The TV 'surface temperature of the reactor plate E was reduced to 270 ° C as a result. Subsequently, the production plant must be put into operation without any problem as described above. The aerolein concentration in a reaction gas 2 inlet mixture in the second reaction stage was 4.6 mol%.

Finally, it should be established that, in case the partial oxidation consists of two partial oxidation lines operated in parallel with, if appropriate, combined work of a gas mixture of the product, one or two production lines may be initiated with the other. line in production by using the gas cycle obtained for starting the former with the last in production.

Comparative examples and examples with reference to a heterogeneously catalyzed stage one partial gas phase oxidation of methacrolein to methacrylic acid in a single zone tube beam reactor C) Description of general process conditions in the reaction stage Heat exchange medium used: The melt consists of 60 wt% potassium nitrate and 40 wt% sodium nitrate.

Reaction tube material: Ferritic steel of DIN material number 1.0315.

Dimensions of reaction tubes: length 4500 mm; internal diameter 25 mm; outside diameter 30 mm (wall thickness: 2.5 mm). Number of reaction tubes in tube bundle: 25 000. Reactor (this was the design and operating mode described in A) I, with the following essential deviations) : The cylindrical container (Ferritic steel of materials DÍN number 1.0425) with an outside diameter of 6650 mm; shirt wall thickness = 2.2 cm; annular vertical tube bundle with a free central space.

Center clearance diameter: 1367 mm. Outer tube circle diameter: 6472 mm. Distribution of homogeneous reaction tube in the tube bundle (6 equidistant close tubes per reaction tube).

Reaction tube height: 38 mm.

The ends of the reaction tubes have been checked and sealed at the holes in the plate plates of plate thickness 295 mm and their holes open in a reactor hood that covers the top tube plate and is connected to the container at the upper end and in a transition. cylindrical to the post-scrubber at the lower end. The aftercooler was of an essentially corresponding type to that of A) IL The cooling tubes number 7056. Their length was 1.4 m, their inner diameter 25 mm, their wall thickness 2 mm. The inlet temperature of the salt melt in the aftercooler was 222 ° C. The upper tube plate is the reactor plate E. The covering reactor hood has an E hole (in the form of a gas inlet tip) with an outside diameter of 1016 mm and a wall thickness of 16 mm.

The tube plates are made from DIN material 1.0481 ferritic steel and the other elements of the DIN 1.0315 ferritic material pipe beam reactor. In each case, a thermolinker was admitted or inserted into the surface of reactor plate E (in the circle of the outer reaction tube) and on top of the reactor hood (reactor hood E). The upper parle of the reactor hood was, as in A) I., placed inside with stainless steel (DIN materials number 1.4541, layer thickness 3.5 mm). The reaction gas inlet mixture was a mixture of air, fresh methacrolein (Mac), steam and gas cycle. Air, fresh methacrolein and steam were removed from the gas cycle system in the working of a gas mixture of the product. The composition of a reaction gas inlet mixture was essentially in normal operation: c (Mac) = 5.5 to 6.5% by volume, c (02) = 9.5 to 11.0% by volume, c (H2 O) - 19.0 to 22.0% by volume, c (CO) = 1.5 to 2.0% by volume, c (CO2) = 1.5 to 2.5% by volume, c ( N2) = essentially the remainder up to 100% by volume.

Reactor charge: salt melt and reaction gas mixture were conducted in countercurrent seen in the reactor. The salt melt is fed into the deep part, the reaction gas mixture through the hole E in the upper part. The inlet temperature of the salt melt was TnemtluUl. The outlet temperature of the salt melt was TiCl 3. WMil - THaura (k was> 0e <2 ° C. Pump output was 4350 m of salt melt / hr. The reaction gas inlet mixture was fed to the reactor with a temperature of ToF when it is passed through port E .

Fixed catalyst bed methacrolein loading: This was L Nl / (1 · h).

Load of the reaction tube bed Fixed catalyst from top down): Zone A: 50 cm Preliminary bed of geometry steatite rings 7 mm x 7 mm x 4 mm (outside diameter x length x inside diameter).

Zone B: 400 cm Catalytic loading of an unsupported annular catalyst which, similar to the unsupported catalyst VUC 12 of DE 102005037678, was prepared using TIMREX T 44 from Tínical AG (Bodio, Switzerland) as auxiliary graphite and, without coma graphite still present, has the stoichiometry Mo L2P1.5Vo.ftCs 1 .oCuo.sSb j So.o40s with ring geometry E x W x I = 7 mm x 7 mm x 3 mm.

Thermal tubes (these number 10 and were evenly disposed in the central region of the tube bundle) were configured and loaded in a manner analogous to that described in A) in order to monitor the temperature in the reaction tubes in a representative manner. D) Results 1. Starting of zone one partial metaerolein oxidation after the fixed catalyst bed was replaced by a fresh fixed catalyst bed (comparative examples) After the tube beam reactor was closed, first 30 t / h of air was driven through the tube bundle reactor through hole E into the reactor hood E. it was sucked from the environment by a full compressor (prior to compressor entry, any solid particles present in this and any condensate were removed as described). WO 2005/016852 and WO 2005/100290) and inlet pressure compressed from 170 to 180 kPa (1.7 to 1.8 bar), the course of which the air was heated to about 110 ° C. Finally, a proportion of air increased over the period and was conducted through a 170 kPa (17 bar) indirect steam heat exchanger prior to compression and thereby heated, which increases the temperature of the compressed air. an entry In port E gradually at 15 ° C, In order to increase this additional inlet temperature, a portion of the hot air carried to the tube bundle reactor was recycled upstream of the compressor. The portions were such that the air temperature at the compressor inlet did not exceed 9 ° C. In this way, it was possible to increase the compressed air temperature of an inlet in port E to about 180 ° C. The amount of salt melt remaining in the tube beam reactor during the fixed catalyst bed change and was converted to the solid state of the matter and was fused by the hot air from the above passage through the tube beam reactor. The total compressor was then shut down (instead of 30 t / h of compressed hot air, now only 1000 NnrVh of a mixture (it was possible to remove from the supply system with the required pressure) 5% by volume of O: and 95% by volume of N '(scarce air) with an inlet temperature of approximately 20 ° C and the appropriate autogenous pressure of approximately 180 kPa (1.8 bar) was conducted through the tube bundle reactor) and the salt was put into operation. In the route of the salt melt heat exchanger (which was not charged with chilled water in this period) an electric heater was arranged with which the salt melt was heated to Tncmr; Kb of 278 ° C within 30 hours. In the same period, the total amount of salt melt in the corresponding reservoir was supplemented with its normal operating level. The scant air provided help to bring the reactor aftercooler to operating temperature, which had also been put into operation with the hot salt melt. The total compressor was then restarted and adjusted at this steam inlet at the 45,000 Nnr1 / h tube beam reactor. This value was constantly retained in the additional process unless explicitly stated otherwise. The aqueous work that was intended for the last removal of the methacrylic acid produced from a gas mixture of the product and was charged with liquid water previously only at the beginning of the work to wash off the solid particles that hold the catalyst (powder) filler. of the gas that charged it out of the water pipe beam reactor for a short period now having the same mode being put into operation. The total compressor drives the total amount of gas already present in the system and keeps the scarce air supplied through the aqueous circulation work (top pressure at work was 1.27 bar; recompression was performed at an inlet pressure of 170 to 180 kPa (1.7 to 1.8 bar); temperature driven at top 66 ° C (this is generally the case unless explicitly stated otherwise)), in the course of which the gaseous mixture has absorbed steam (the passage to the quantities of gas exceeding the amount 45 000 Nm3 / h is present as the passage of residual gas at work) to an inlet E, at approximately a 4% by volume molecular oxygen, 21% by volume vapor and, as remaining essential nitrogen (gas temperature at orifice E inlet was about 11 ° C). The scarce air supplied has now been shut down and instead 1000 kg / h of air (previously filtered) has been supplied to the gas cycle system by means of a working air compressor (the working supply temperature was 60 ° C). The concentration of oxygen in the gas mixture supplied to orifice E originated as a result. Once it reached 5% by volume (by 21% by volume of vapor and as the essentially remaining nitrogen), 1000 kg / h of Fresh methacrolein was applied to the gas cycle system by working (the working feed temperature = 37 ° C; the gas inlet temperature of the gas mixture in port E was about 110 ° C). Freshly supplied methacrolein has a purity of> 95% by weight.

Significant impurities were 3 to 4 wt% water, approximately 1 wt% methanol and also small amounts of propionaldehyde, pentenal and formaldehyde. THcntnidí ‘has now captured at 276ÜC causing the gas to cool. As soon as the oxygen content in the waste gas began to decrease and the CO2 concentration in the essentially methacrolein-free waste gas began to increase, the pre-compressed air flow provided through the work was increased until the oxygen concentration in the waste gas. gas mixture enters orifice E reaching 5.6% by volume. While retaining the limiting conditions, partial oxidation was thus operated for an additional six minutes. At the end of this time period, the reaction gas inlet mixture has the properties listed in Table 4 below (Part 1 and Part 2). in each case in the first row of the Table.

In the Table: c (Mac) =% by volume methacrolein content, c (Oi) =% by volume oxygen content, c (H2O) = 111% vapor content by 1 ume, c (CO) = carbon monoxide content by volume%, c (Ni) = nitrogen content and volume% c, (CO :) = carbon dioxide content by volume%, in each case of an inlet gas mixture reaction. In addition, in Table 4: MMl'c = fresh methacrolein vapor supplied to the gas cycle system by working in kg / h; CMac = methacrolein conversion based on the simple passage of the reaction gas mixture through the tube bundle reactor cm% in mol; SM, IU = selectivity of methacrylic acid formation based on methacrolein converted to single passage of reaction gas mixture through tube bundle reactor in% mol. The latter at startup, the concentration of methacrolein in a reaction gas inlet mixture was gradually increased (by increase of fresh methacrolein vapor). The air vapor and THc, nratkl were each, after the increase in MMílt, readjusted corresponding to the increase of c (Mac) over time, such that the results listed in Table 4 resulted in the period of operation t [h]. were determined in each case at the end of the particular operation time interval where the quasi-operating state corresponds to the new follow-up MMat 'and was established in each case. The values for L listed in Table 4 are generally understood to mean L ± 2 Nl / (l * h).

Finally, it is possible to operate the additional partial oxidation in an essentially fixed state.

When, for example, due to the relatively low market demand, operation at lower loads L is desired, that is, it is clear that a change in the state of operation at rest is possible as soon as this is achieved. 2 (comparison) Partial oxidation of methacrolein departure after the fixed catalyst bed has been operated for a prolonged period of operation (with intermediate regeneration according to JP-A 2003/30646) (as described in D) 1.) and oxidation This part has been interrupted by an operational failure.

Prior to discontinuation, partial oxidation was operated at L = 55 Nl / (l * h). The supply of fresh methacrolein at work was 4400 kg / l. The reaction gas inlet mixture has essentially the normal operating composition and CMac = 53 mol% and SMaa = 83.4% · mol and Tc.1 During the interruption of the operation, which lasted approximately 24 hours, 4000 Nm3 / h of a gaseous mixture regenerating from the composition 5.6% by volume of oxygen, 21% by volume. Steam volume and 73.4% by volume nitrogen were conducted by meaning the total compressor through orifice E (the inlet temperature of the gas mixture was 111 ° C). TE [emimla was gradually increased to 3I4 ° C during this period with the aid of an electric heater (the salt melt vane with that aid, as in each tube beam reactor unit, the amount of salt melt that was removable). for the cooling of the salt melt was controllable and was closed during this period).

Subsequently, partial oxidation was essentially immediately continued in this normal operating state prior to discontinuation (but with normal methacrolein loading).

For this purpose, the gas vapor conducted through the pipe bundle reactor with the aid of the full compressor was the first while maintaining an oxygen content of 5.6% by volume, increased to the normal operating control value of 45,000 NnvVh. (gas cycle operation for the entire work; gas inlet temperature =] 11 ° €).

In order to obtain a surface temperature of 305 ° C reactor plate (T |, t! "MtUl = 311 ° C), 3000 kg / hr of fresh methacrolein was then returned to work as a result of a A starting methacrolein concentration of 3.6% by volume and an oxygen content of 5.4% by volume were established in a reaction gas inlet mixture, leading to an immediate suspension of a production plant as a result. of a rapid temperature rise in the reactor E.cuff. (Examples) Partial oxidation of the inventive methacrolein starting after immediate suspension in D) 2, the tube bundle reactor was subjected to 1000 NrrrVh flow of scarce air ( 5% by volume Cb, 95% by volume N 'of inlet pressure = 180 kPa (1.8 bar) (inlet temperature = 20 ° C) and TV11 ™ 1' was simultaneously increased to 32 ΓΟ with electric heater.Then the scarce air was turned off, the air compressor was put into operation. With the help of the total compressor, in the gas cycle operation it conducts through the work, a gas mixture of 5.6% by volume of oxygen, 21% by volume of steam and 73.4% by volume of nitrogen. and was conducted through the tube beam reactor at a volume flow rate of 45,000 Nrn 3 / h and an inlet temperature of 111 ° C (hole E; inlet pressure: 180 kPa (1.8 bar)). Despite opposite heating by the electric heater, THcmriul; i fell in the same period at 311 ° C and the surface temperature of the reactor plate E dropped to 257 ° C. While maintaining the O: concentration of 5.6% by volume, the 5000 kg / hr feed of fresh methacrolein through work has now begun. In parallel to the methacrolein content of a reaction gas inlet mixture which was set at these resting state values of 5.6% by volume of departure from 3.6% by volume in the period, this oxygen content was increased to 9.8% by volume. After 6 hours, the operating state of rest was achieved without incidence according to Table 5. 4. (Examples) Partial oxidation of inventive methachroline startup When methacrylic acid production was interrupted again (operating conditions before interruption ( operation as described in D) 1.): normal operating composition of a reaction gas inlet mixture; THR = = 312 ° C; CNI, a '= 52 mol%; MWa = 82.9% by volume, L = 55 Nl / (l * h)>, 45,000 NmVh of a gas mixture comprising 5.6% by volume of oxygen, 21% by volume of steam and 73.4% by volume. nitrogen volume (inlet temperature = 111 ° C inlet pressure = 180 kPa (1.8 bar)), with the help of the total compressor, by working in gas cycle mode. were first conducted through the tube beam reactor or a 2 hour period. The gas through the flow was then sparged for 48 hours (1000 Nm3 / h; 5 vol% Oi; 95 vol% Ni; inlet temperature = 20 ° C; inlet pressure = 180 kPa (1 , 8 bar)). The THcn of 312 ° C was initially maintained. Then, in the gas cycle mode, by working, with the help of the total compressor, 4000 Nm3 / h of a gas mixture comprising 5.6% by volume of oxygen, 21% by volume of steam and 73% by volume. 4%; in nitrogen volume were conducted through the tube beam reactor (inlet temperature = 111 ° C inlet pressure = 180 kPa (1.8 bar)). The TH € ltrailil was simultaneously increased to 318 ° €. After that, with the same oxygen content, the total gaseous vapor was increased to 45,000 Nm3 / h (inlet temperature = 111 ° C inlet pressure = 180 kPa (1.8 bar)).

At an E reactor plate surface temperature of 308 ° C and 315 ° C, the 1000 kg / hr feed of fresh methacrolein via work was started (this corresponds to a starting methacrolein concentration in a mixture reaction gas inlet to the beam beam reactor of only 0.7% inol; its oxygen content was adjusted to the starting value of 5.6% by volume by adjusting the supply air). The inlet temperature Tgl was 111 ° C and the inlet pressure was 180 kPa (1.8 bar). 5 minutes thereafter, fresh methacrolein vapor was increased to 2000 kg / h (this corresponds to a starting methacrolein concentration in a 1.5% by volume reaction gas inlet mixture; its oxygen content was adjusted to the starting value of 7.0% by volume by adjusting the supply air; TcH = 111 ° €; inlet pressure = 180 kPa (1.8 bar)).

After maintaining the above operating conditions for 1 hour, the surface temperature of the reactor plate E dropped to 255 ° C. Including the electric heater, it was possible to heat Tffri'r; id; i to 314 ° C. The fresh methacrolein stream value has now been increased from 500 kg / h to 5000 kg / h with equidistant steps (times) within 1 hour and the amount of air supplied through the work has been simultaneously readjusted to give fractions. volume O: in an inlet mixture of the reaction gas corresponding in each case to the particular corresponding fresh methacrolein vapor in Table 4. The THe [1, radM was left at 314 ° C. AT (; h was a constant 111 ° C and inlet pressure was a constant 180 kPa (1.8 bar). Operating conditions were then maintained. 7 hours later, the operating state indicated in Table 6 was reached .

Table (Part I) Puditld ilu PalnHi: Pnniw> ik> L 'S. N 60 / 907,954. ilqKisilailo on 04.24.2007. is ima> rp <n <u] <> nu [ursenie peJidn pdn references nu liternlura. With regard to the above meneninuiin teaching, no one: * nmercsas and cepuraijo-h from the presence of this invention are possible.

Bslc. It can therefore be assumed that the invention, within the scope of the appended claims, may be carried out dilutively in the manner described herein.

Claims (16)

1. Process for starting a heterogeneously catalyzed partial gas phase oxidation of acroleitic to acrylic acid or methacrolein to methacrylic acid in a fixed catalyst bed which is arranged in a tube bundle reactor (1) reaction tubes (9) surrounded by a reactor jacket (14), both ends of the individual reaction cubes (9) being opened and the upper end of each reaction tube end (9) sealed in a passage of an upper tube plate (19) sealed at the top of the reactor jacket (14) and the end of the lower end sealed in a through hole of a lower tube plate (5) sealed at the bottom of the reactor jacket (14), the exterior of the reaction tubes (9), the upper (19) and lower (5) tube plate and the reactor jacket (14) together delimiting the surrounding space (15) of the reaction tube (9) , and each of the two plates of you (19, 5) being covered by a reactor hood (6), (7) having at least one orifice, characterized in that a reaction gas inlet mixture (1) comprising> 3% by volume of acrolein or methacrolein as well as molecular oxygen is fed to the reaction tubes (9) of the tube beam reactor through at least one hole E (3) in one of the two reactor hoods and the product gas mixture (2). ) which results by partial gas phase oxidation of acrolein or methacrolein to acrylic acid or methacrylic acid in the passageway through the fixed catalyst bed disposed in the reaction tubes (9) and comprises acrylic acid or methacrylic acid is removed by means of the at least one orifice of the other reactor hood (7), while at the side bore (14) of the tube beam reactor at least one liquid heat exchange means is conducted around the reaction tubes (9) such that each of the two surface The two tube plates (19), (5) facing each other are wetted by the liquid heat exchange means and the at least one liquid heat exchange means is conducted in the surrounding space (15) of the reaction tube ( 9) having a temperature THentred (11) and is conducted outside the surrounding space (15) of the reaction tube (9) again with a temperature TH output (10), wherein, in the period in which the reaction gas inlet mixture (1) comprising> 3% by volume of acrolein or methacrolein enters the reactor hood (6) through at least one hole E (3), - the temperature of at least one liquid heat exchange medium which moistens the plate of tube E (19) covered by reactor hood (6) having at least one hole E is at least 290 ° C, - the reaction gas inlet mixture (1) which enters at least one hole E ( 3) has a temperature of <285 ° C and - the temperature of surface E (4) of tube plate E (19) facing the reaction hood r having at least one hole E (3) has a value of <285 ° C. Process according to claim 1, characterized in that the temperature of the at least one liquid heat exchange medium which moistens the tube plate E (19) is at least 295 ° C. Process according to Claim 1, characterized in that the temperature of the at least one liquid heat exchange medium which moistens the tube plate E (19) is at least 300 ° C. Process according to any one of Claims 1 to 3, characterized in that the reaction gas inlet mixture (1) entering the at least one port E (3) has a temperature of <280 ° C. . Process according to any one of Claims 1 to 3, characterized in that the reaction gas inlet mixture (1) entering the at least one port E (3) has a temperature of <270 ° C. . Process according to any one of Claims 1 to 5, characterized in that the surface temperature E (4) has a value of <280 ° C. Process according to any one of claims 1 to 5, characterized in that the surface temperature E (4) has a value of <275 ° C. Process according to any one of claims 1 to 7, characterized in that the reaction gas inlet mixture (1) comprises at least 4% by volume of acrolein or methacrolein. Process according to any one of claims 1 to 7, characterized in that the reaction gas inlet mixture (1) comprises at least 5% by volume of acrolein or methacrolein. Process according to any one of claims 1 to 9, characterized in that the molar ratio of O 2: (acrolein or methacrolein) in the reaction gas inlet mixture (1) is> 0.5. Process according to any one of claims 1 to 10, characterized in that the loading of the fixed catalyst bed with acrolein or methacrolein is> 10 Nl / (1 * h). Process according to any one of claims 1 to 11, characterized in that the tube plate E (19) has a cross-sectional area of 3 m2 to 80 m2. Process according to any one of claims 1 to 12, characterized in that the tube plate E (19) has a reactor plate thickness of 5 to 40 cm. Process according to any one of Claims 1 to 13, characterized in that the tube plate E (19) is made of steel. Process according to any one of claims 1 to 14, characterized in that the at least one heat exchange medium is a molten salt or a heat carrier oil. Process according to any one of Claims 1 to 15, characterized in that the reactor hood (6) covering the E-tube plate (19) on its side faces the E-tube plate (19). ) is coated with stainless steel or with a zinc silicate base.