Catalyst for the oxidative esterification of aldehydes to carboxylic esters

Abstract

The present invention relates to a novel process for oxidative esterification, generally for reaction of aldehydes with alcohols in the presence of oxygenous gases directly to give the corresponding ester in the presence of a heterogeneous catalyst, by means of which, for example, (meth)acrolein can be converted to methyl (meth)acrylate. The new catalyst has titanium dioxide as the main component of the support material. The catalysts are especially notable for high mechanical and chemical stability and for good catalytic performance even over very long periods. The process is an improvement in the catalyst service life, activity and selectivity over prior art catalysts which lose activity and/or selectivity relatively quickly in continuous operation in media having even a small water content.

Claims

1. A process for preparing a carboxylic ester from an aldehyde, the process comprising: continuously conducting an oxidative esterification of the aldehyde to the carboxylic ester via heterogeneous catalysis in a liquid phase in the presence of a catalyst particle, wherein a content of hydrogenated by-product in the carboxylic ester is from 50 ppm to 100 ppm, wherein the catalyst particle consists of 0.1% to 3% by weight of gold, 25% to 99.8% by weight of TiO.sub.2, 0% to 25% by weight of at least one oxide of an alkali metal, an alkaline earth metal, a rare earth metal, and zirconium, 0% to 20% by weight of at least one oxide selected from the group consisting of an iron oxide, a zinc oxide, and a cobalt oxide, and 0% to 5% by weight of at least one other component; gold, optionally together with the iron oxide, zinc oxide, and/or cobalt oxide, is bound to the catalyst particle in the form of particles having an average diameter between 2 and 10 nm; and the catalyst particle has an average geometric equivalent diameter between 5 μm and 10 mm, and more than 90% of the gold is present in an outer region of the catalyst particle, wherein the outer region makes up a maximum of 60% of the geometric equivalent diameter of the catalyst particle, wherein the catalyst particle does not contain silicon oxide and/or Al.sub.2O.sub.3.

2. The process according to claim 1, wherein the catalyst particle consists of 0.3% to 2.5% by weight of gold, 50% to 99.5% by weight of TiO.sub.2, 0% to 10% by weight of the at least one oxide selected from the group consisting of an iron oxide, a zinc oxide and a cobalt oxide, and 0% to 5% by weight of the at least one other component.

3. The process according to claim 1, wherein the catalyst particle consists of 0.3% to 2.5% by weight of gold, 96% to 99.5% by weight of TiO.sub.2 and not more than 3.7% by weight of other components.

4. The process according to claim 1, wherein the catalyst particles are porous and have a specific surface area between 20 and 300 m.sup.2/g and an average pore diameter of 2 to 30 rim.

5. The process according to claim 4, wherein the specific surface area of the catalyst particles is between 30 and 200 m.sup.2/g, and the average pore diameter is 5 to 25 nm.

6. The process according to claim 1, wherein the oxidative esterification is conducted in the presence of oxygen and an alcohol.

7. The process according to claim 6, wherein the aldehyde is methacrolein and the carboxylic ester is an alkyl methacrylate.

8. The process according to claim 7, wherein the alcohol is methanol; methacrolein continuously reacts with methanol in the presence of an oxygenous gas at a reaction temperature between 40-120° C., an absolute pressure between 1 and 20 bar, and with a slurry density of the catalyst, which is being stirred and swirled in a steady-state product mixture comprising MMA, methacrolein, methanol and water, between 1% and 20% by weight; and methacrolein is not fully converted in at least one reactor in each case, but after a partial conversion between 20% and 95% is separated continuously from the catalyst via filtration and fed to a workup.

9. The process according to claim 8, wherein air is supplied continuously to the reactor as the oxygenous gas and an oxygen content of gas phase that forms the reaction mixture is between 2% and 8% by volume.

10. The process according to claim 8, wherein the reaction is conducted at a reaction temperature between 50° C. and 100° C. at an absolute pressure between 2 and 15 bar.

11. The process according to claim 8, wherein the slurry density of the catalyst is between 2% and 15% by weight.

12. The process according to claim 8, wherein methacrolein is not fully converted in one reactor in each case or multiple reactors connected in series, but after a partial conversion between 40% and 80% is separated continuously from the catalyst via filtration and fed to the workup.

Description

BRIEF DESCRIPTION OF DRAWINGS

(1) FIG. 1 shows, for Table 4, the methacrolein conversion as a function of the run time in continuous operation. It is thus especially possible to demonstrate the long-term activity of the catalyst.

(2) FIG. 2 shows, for Table 4, the dependence of the formation of the hydrogenated MMA by-product depending on the experimental run time in continuous operation.

(3) FIG. 3 shows, for Table 4, the composition of the reactor output stream for components present in a concentration greater than 1% by weight in continuous operation.

(4) FIG. 4 shows, for Table 4, the composition of the reactor output stream for components present in a concentration less than 1% by weight in continuous operation.

DETAILED DESCRIPTION OF THE INVENTION

(5) A novel process for oxidative esterification of aldehydes to carboxylic esters is described. Novel catalyst particles that are surprisingly efficient and long-lived with regard to the teaching of the prior art are used in the process. The process is preferably an oxidative esterification of an aldehyde in the presence of oxygen and an alcohol to give a carboxylic ester. The process is more preferably a process for oxidative esterification of methacrolein in the presence of oxygenous gases and an alcohol to give an alkyl methacrylate, especially of methacrolein in the presence of methanol to give MMA.

(6) Alternatively, it is possible to react methacrolein with oxygen and a di-, tri- or tetrafunctional alcohol to give a hydroxyalkyl methacrylate and di-, tri- or tetramethacrylate. The di-, tri-, and tetrafunctional alcohols are known as crosslinkers. A particularly preferred example of a difunctional alcohol is ethylene glycol.

(7) The catalysts used in the process in accordance with the invention are characterized in that the catalyst particle consists of 0.1% to 3% by weight, preferably 0.3% to 2.5% by weight, of gold, 25% to 99.8% by weight, preferably 50% to 99.5% by weight, of TiO.sub.2, 0% to 50% by weight, preferably 0% to 20% by weight, of silicon oxides, 0% to 25% by weight of Al.sub.2O.sub.3, 0% to 25% by weight of oxides of alkali metals, alkaline earth metals, rare earth metals and/or zirconium, 0% to 20% by weight, preferably 0% to 10% by weight, of iron oxides, zinc oxides, nickel oxides and/or cobalt oxides and 0% to 5% by weight of further components. In general, the gold is in elemental form in the form of Au{0} and in the form of nanoparticles, while all other components in the catalyst particle are in oxidic form.

(8) In addition, it is a feature of these catalyst particles used in the process in accordance with the invention that the gold, optionally together with iron oxides, zinc oxides, nickel oxides and/or cobalt oxides, is present in the form of particles—also referred to as primary particles—having an average diameter between 2 and 10 nm and these particles are bound to the surface of the rest of the catalyst particle.

(9) In the case of mixed phases of gold with the oxides mentioned, however, this does not necessarily mean that these oxides must entirely be some of the particles having an average diameter between 2 and 10 nm in the catalyst particle. Instead, some of the oxides may also be part of the support material and, for example, be bound in another way in some other form, particularly to the surface of the support material. More preferably, however, the catalyst particle does not include any iron oxide, zinc oxide, nickel oxide and/or cobalt oxide, but has exclusively gold as the sole active component. The catalyst particle used in the process in accordance with the invention has an average geometric equivalent diameter between 5 μm and 10 mm, preferably between 10 μm and 10 mm, more preferably between 200 μm and 5 mm.

(10) According to the production method, it is possible that the gold is present either in the form of pure particles or in a mixed form, for example with one of the oxides mentioned, for example cobalt oxide. In this latter case, the gold is generally mixed only with a portion of the oxide. In addition, it is optionally also possible in both embodiments that the gold or gold-containing particles, for stabilization, are additionally provided with a thin layer, for example of TiO.sub.2, SiO.sub.2 and/or Al.sub.2O.sub.3.

(11) Furthermore, the process according to the invention is conducted continuously, by means of heterogeneous catalysis in a liquid phase. Preferably, in the continuously operated reactor, there is a triphasic system with a solid catalyst phase, a liquid reaction phase and a gas phase.

(12) In a particular embodiment of the present invention, the process is characterized in that the catalyst particle is composed predominantly or even exclusively of gold and TiO.sub.2. In this variant, the catalyst particle consists preferably of 0.3% to 2.5% by weight of gold, 96% to 99.5% by weight of TiO.sub.2 and not more than 3.7% by weight, more preferably 1.5% by weight, of further components which are neither gold nor TiO.sub.2.

(13) Preferably, the catalyst particles used in accordance with the invention are porous. In this case, the porosity generally does not relate to the gold or gold-containing phases. Such porous catalyst particles have a specific surface area between 20 and 300 m.sup.2/g, preferably between 30 and 200 m.sup.2/g, more preferably between 80 and 150 m.sup.2/g. In addition, the average pore diameter is generally 2 to 50 nm, preferably 5 to 30 nm, more preferably 10 to 25 nm.

(14) An optional component of the catalyst particle used in accordance with the invention is what are called basic elements. These basic elements are especially alkali metals (Li, Na, K, Rb, Cs, Fr), alkaline earth metals (Be, Mg, Ca, Sr, Ba), rare earth metals (Sc, Y, La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu) or mixtures of these metals. The basic element is generally also present in oxide form. It is entirely possible in the case of prolonged operation of the process with the catalyst particles used in accordance with the invention that a freshly prepared catalyst can lose a portion of the basic element during a process for preparing carboxylic esters. According to the nature of the catalyst, this may lead either to a slight improvement or to a deterioration in the activity and/or selectivity of the process, and is caused by acids formed as by-product during the process.

(15) Further suitable metal oxides, especially for the support component of the catalyst, are zirconium oxides.

(16) The further components that are optionally present in the catalyst particle at up to 5% by weight are components that first of all differ from the other compounds or elements listed. An especially suitable example of such a further component is zirconium, or a zirconium compound, such as zirconium oxide in particular.

(17) In a preferred embodiment of the present invention, the gold-containing primary particles are in the outer region of the catalyst particle. In a second, likewise preferred embodiment, these primary particles are distributed homogeneously across the catalyst particle.

(18) In the first embodiment mentioned, the catalyst particles used in the process in accordance with the invention are characterized in that the maximum gold concentration or the maximum iron, zinc or cobalt concentration of the catalyst particle is to be found in the outer region thereof. Said outer region makes up a maximum of 60%, preferably a maximum of 40% and more preferably a maximum of 30% of the geometric equivalent diameter of the catalyst particle. In this context, the gold concentration or iron, zinc and/or cobalt concentration in this outer region is at least 1.5 times, preferably at least twice and especially preferably at least 2.5 times as high as the corresponding concentration of these elements in the middle region which makes up the remaining region of the geometric equivalent diameter of the catalyst particle. More preferably, the gold in this embodiment is present to an extent of more than 90% in this outer region. Such a construction of the catalyst particle is referred to as an eggshell structure.

(19) The determination and analysis of the distribution of the concentrations of gold and/or iron, nickel, zinc and/or cobalt across the catalyst particle profile can be effected, for example, by the embedding of the particles into a polymer matrix, subsequent polishing and then SEM-EDX analysis. An analogous analysis method by means of x-ray microanalysis (EPMA) is described, for example, in EP 2 210 664 A1 on page 18.

(20) The catalyst particles used in accordance with the invention and hence the process according to the invention, particularly with regard to the prior art discussed above, are notable for a number of advantages that were not to be expected beforehand. For instance, these catalyst particles are especially notable for the retention of a high activity with simultaneously high selectivity of the catalyst particle used for the oxidative esterification over a long period. Thus, the catalyst particles used in accordance with the invention have a particularly good combination of a) low mechanical abrasion of the catalyst particle, b) low leaching of metal ions, which can lead to problems with regard to stability, for example, in the case of iron in an MMA prepared in accordance with the invention, out of the particle and c) long-term retention of catalyst performance with regard to activity and selectivity.

(21) Particularly surprisingly, it has been found that the catalysts according to the invention lead not only to a generally higher selectivity but that the formation of particularly unwanted by-products in particular, for example of methacrylic acid. In the same way, for example, in the case of the oxidative esterification of methacrolein with methanol to give MMA, a further reduction in formation of methyl methoxyisobutyrate (MMIB) was found.

(22) Furthermore, it has been found that, surprisingly, the catalysts used in accordance with the invention distinctly reduced the formation of hydrogenated by-products of the desired main products compared to the prior art. Thus, in accordance with the invention, the formation of, for example, hydrogenated MMA in the case of use of a TiO.sub.2-based catalyst according to the invention, by comparison with the best catalysts described in the prior art, is reduced by a factor of 10 from a content between 500 and 1000 ppm to a content between 50 and 100 ppm.

(23) Many of the by-products of the reaction are formed especially at high concentrations of the starting materials. Dimeric methacrolein and the esters thereof, for example, form preferentially at high concentrations of the methacrolein starting material. Dimethoxyisobutene forms especially under neutral and acidic conditions at high concentrations of methanol and methacrolein. Methyl 3-methoxyisobutyrate (3-MMIB for short) forms at high concentrations of methanol as a Michael addition product with methacrolein (and subsequent direct oxidative esterification), particularly under neutral and basic conditions.

(24) The formation of dimeric methacroleins (MAL dimer for short) can be counteracted at relatively high pH values in such a way that the conversion of the MAL is higher at high pH values and the concentration of the same is thus lower. This results in reduced formation of the MAL dimer or esters thereof. However, according to the prior art, the change in the pH, on the other hand, would increase the preferential formation of the other by-product, 3-MMIB. According to the invention, however, the formation of the MAL dimer plays a surprisingly minor role, and so it now becomes possible to improve the yield and selectivity of the process, surprisingly, with rising conversion. This is effected in such a way that an additional surprising effect in accordance with the invention is that the process can be operated at particularly high conversion without needing to extend the residence time in the reactor.

(25) The Diels-Alder reaction which leads to formation of dimeric MAL or ester thereof is especially also promoted by a higher temperature. According to the invention, this can surprisingly be additionally counteracted by more moderate reaction conditions.

(26) Equally surprisingly, the formation of a methyl isobutyrate from secondary hydrogenation reaction is also distinctly reduced, such that the catalyst present, in a continuous process, after a run-in period, retains a surprisingly high selectivity at a constant level over a very long operating period.

(27) Preferably, the catalyst particles have an average geometric equivalent diameter between 10 μm, preferably 100 μm, and 5 mm, more preferably between 500 μm and 2 mm. The diameter here is based on the average geometric equivalent diameter (also referred to synonymously hereinafter as average equivalent diameter or geometric equivalent diameter) in the form of a volume-equivalent sphere diameter. The average equivalent diameter of such an inhomogeneous body can be determined in that the average equivalent diameter corresponds to the diameter d.sub.v of a sphere of equal volume to the inhomogeneous body. For this purpose, first of all—neglecting the porosity—the volume of the catalyst particle is determined. This can be effected in the case of simple, homogeneous bodies, for example regularly shaped cylinders, by simply measuring a microscopic absorption, for example. In the case of ideally spherical particles, the determination can even be dispensed with entirely, since the result corresponds to the actual ball diameter.

(28) In the case of very inhomogeneously shaped particles, the volume is determined and—if this is done in a porosity-dependent manner—the result is corrected by the porosity to be determined additionally.

(29) The formula for calculation of the average equivalent diameter, which, in accordance with the invention, is more specifically the volume-equivalent sphere diameter d.sub.v, depending on the volume V is:

(30) $d_v=\sqrt[3]{\frac{6*V}{\pi }}$

(31) In the case that the gold-containing particles are present close to the surface of the catalyst particle, the active outer region of such a catalyst particle with eggshell structure is referred to hereinafter as outer region. The thickness of this outer region is preferably between 2 and 100 μm, preferably between 5 and 50 μm. The size of the geometric equivalent diameter is stated because the particles need not necessarily be in entirely spherical form, but may quite possibly also have more complex forms.

(32) It should also be pointed out that the boundary thus contemplated between the core and a shell in the case of such an optional eggshell structure, moreover, will not be sharp, but may especially be in the form of a gradient with varying composition. For example, the concentration of gold nanoparticles, viewed from the core, may thus increase from the inside outward. This arises merely from the fact that the particles of the invention generally have porosity. In the ideal case, an illustrative value of thickness 80 μm of the outer region, for example in the case of a particle having an equivalent diameter of 200 μm, means that, viewed over the diameter, at either outer end thereof, there is 40 μm of outer region with 120 μm of middle region in between. This size was chosen in order to describe the eggshell structure of the catalyst of the invention. The boundary between the outer and inner region can be chosen relatively freely by the person skilled in the art within the ranges specified in the examination of the particles. What is crucial in accordance with the invention is that there is a boundary within the range specified where the conditions exist with regard to the iron, zinc, nickel and/or cobalt concentration and the gold concentration. In relation to the iron, zinc and especially the cobalt concentration, this is the inventive core of the present invention.

(33) The parameters of a particularly preferred embodiment of the present invention are described hereinafter. Given compliance with the settings specified here, particularly good results of the process are obtained in relation to yield and selectivity over a long period of operation. However, this is not supposed to mean that there are not alternative settings of the present invention that lead to similar results. Instead, it should be pointed out that the catalyst used in accordance with the invention brings surprisingly good results compared to the prior art over a broad range.

(34) More particularly, it is a feature of the process in this particularly preferred embodiment that the reaction between methacrolein and an alkyl alcohol, especially methanol, is conducted continuously in the presence of an oxygenous gas observing the following parameters: I. at a reaction temperature between 40-120° C., more preferably between 50 and 100° C., II. at an absolute pressure between 1 and 20 bar, more preferably between 2 and 15 bar, III. with a slurry density of the catalyst, especially particulate catalyst, which is being stirred and swirled in the steady-state product mixture comprising MMA, methacrolein, methanol and water, between 1% and 20% by weight, more preferably between 2% and 15% by weight.

(35) Preferably, the methacrolein reactant is not completely converted in at least one reactor in each case. Instead, preferably after partial conversion between 20% and 95%, it is separated continuously from the catalyst by means of filtration and sent to the workup. In the workup, the remaining methacrolein can then, for example, be isolated and recycled into the reaction as reactant.

(36) The process can surprisingly be configured in a particularly economically viable manner when the molar excess of methanol to methacrolein fed continuously to the reactor is limited to 1:1 to 15:1. The concept of “continuous feeding” here describes a period of time appropriate for the reaction, irrespective of whether methacrolein and methanol are added simultaneously or sequentially. Both methacrolein and methanol are only partly converted using the novel catalyst systems in the process which is likewise according to the invention. This means that the resulting reaction mixture, after separation from the catalyst particle, is purified in such a way that unconverted methacrolein and methanol are separated from the product esters and fed back to the reaction.

(37) More preferably, air is fed continuously to the reactor as oxygenous gas. Likewise more preferably, the oxygen content of the gas phase which is present and forms above the reaction mixture is between 2% and 8% by volume.

(38) In a very particular variant of the reaction, the methacrolein reactant is reacted with the alkyl alcohol in multiple reactors connected in series. In this case too, it is especially preferable not to achieve full conversion, but instead, after a partial conversion between 20% and 90%, especially preferably between 40% and 80%, to continuously separate the methacrolein from the catalyst by means of filtration and feed it to the workup.

(39) In a likewise preferred, alternative variant of the reaction, the methacrolein reactant is reacted with the alkyl alcohol in exactly one reactor. In this case too, it is especially preferable not to achieve full conversion, but instead, after a partial conversion between 20% and 90%, especially preferably between 40% and 80%, to continuously separate the methacrolein from the catalyst by means of filtration and feed it to the workup.

(40) The oxidative esterification according to the invention is conducted continuously in the presence of the catalyst particle. More preferably, the catalyst is employed in suspension form (as a slurry) in a stirred reactor during the oxidative esterification.

EXAMPLES

Example 1

(41) Preparation of a Catalyst with 1.4% by Weight of Au on TiO.sub.2 Support (d50=20 μm)

(42) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 support (product name: Aeroperl P25, Evonik Industries, prepared from TiCl.sub.4 by flame pyrolysis, grain size d50=20 μm, 75% anatase, 25% rutile) are directly added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is centrifuged while hot and decanted, and the residue is converted to a slurry again and centrifuged again a total of five times with 250 ml of distilled water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h. The catalyst is heated to 70° C. under nitrogen and then reduced with a stream of 5% by volume of H.sub.2 in N.sub.2 and cooled down to room temperature, and the gas stream is ended. The performance of the catalyst was evaluated in a batch test (see Table 1).

Example 2

(43) Preparation of a Catalyst with 1.4% by Weight of Au on TiO.sub.2 Support (d50=40 μm)

(44) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 support (TP Hombikat, Huntsman, d50=about 40 μm—development product as stationary phase of HPLC columns, 95% anatase structure) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. the suspension is vacuum-filtered while still hot through a suction filter and washed 5 times with 250 ml of dist. water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h. The catalyst is heated to 70° C. under nitrogen and then reduced with a stream of 5% by volume of H.sub.2 in N.sub.2 and cooled down to room temperature, and the gas stream is ended. The performance of the catalyst was evaluated in a batch test and continuously [see Table 1 (batch test) and Table 2 (continuous)].

Example 3

(45) Preparation of a Catalyst with 1.1% by Weight of Au on TiO.sub.2 Support (d50=30 μm)

(46) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 support (product name: Aeroperl P25 H6, Evonik Industries, P25 recalcined at 600° C. under N.sub.2 atmosphere, grain size d50=30 μm) were added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is centrifuged while hot and decanted, and the residue is converted to a slurry again and centrifuged again a total of five times with 250 ml of distilled water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h. The catalyst is heated to 70° C. under nitrogen and then reduced with a stream of 5% by volume of H.sub.2 in N.sub.2 and cooled down to RT, and the gas stream is ended. The performance of the catalyst was evaluated in a batch test (see Table 1).

Example 4

(47) Preparation of a Catalyst with 1.0% by Weight of Au on TiO.sub.2 Support (2×5 mm)

(48) In a glass flask with mechanical stirring, 1.32 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.5 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 support (Aeroperl P25 with binder extruded to 2×5 mm rods and calcined, Evonik Industries) were added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The catalyst rods are filtered off with suction while still hot and washed with dist. water until the residual conductivity of the wash water obtained was below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h. The catalyst is heated to 70° C. under nitrogen with gentle circulation (6 rpm) and then reduced with a stream of 5% by volume of H.sub.2 in N.sub.2 and cooled down to room temperature, and the gas stream is ended.

Example 5

(49) Preparation of a Catalyst with 1.0% by Weight of Au on TiO.sub.2 Support (37-100 μm)

(50) The catalyst from Example 4 was comminuted in an agate mortar and the fraction <100 μm was sieved off using a wire mesh sieve. The residue was comminuted again in a mortar and the procedure was repeated until all the material had been comminuted to a size <100 μm. The residue thus obtained was freed of the fines fraction using a sieve having a mesh size of 37 μm. The catalyst with a grain size of 37-100 μm remaining on the sieve is the catalyst for Example 5 in Table 1 of the batch test and Example 2 in Table 2 of the continuous test.

Example 6

(51) Preparation of a Catalyst with 1.0% by Weight of Au on TiO.sub.2 Support (<37 μm)

(52) The catalyst from Example 4 was comminuted in an agate mortar and the fraction <100 μm was sieved off using a wire mesh sieve. The residue was comminuted again in a mortar and the procedure was repeated until all the material had been comminuted to a size <100 μm. The residue thus obtained was freed of the coarse fraction using a sieve having a mesh size of 37 μm. The catalyst that passes through the sieve with a grain size of <37 μm is the catalyst for Example 6 in Table 1.

Example 7

(53) Preparation of an Uncalcined Catalyst with 1.4% by Weight of Au on TiO.sub.2 Support

(54) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 support (product name: Aeroperl P25, Evonik Industries, prepared from TiCl.sub.4 by flame pyrolysis, grain size d50=20 μm, 75% anatase, 25% rutile) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is centrifuged while hot and decanted, and the residue is converted to a slurry again and centrifuged again a total of five times with 250 ml of distilled water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h. The performance of the catalyst thus obtained was evaluated in a batch test (see Table 1).

Example 8

(55) Preparation of a Catalyst with 1.4% by Weight of Au on TiO.sub.2 Support (Calcined for 3 h)

(56) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 support (product name: Aeroperl P25, Evonik Industries, prepared from TiCl.sub.4 by flame pyrolysis, grain size d50=20 μm, 75% anatase, 25% rutile) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is centrifuged while hot and decanted, and the residue is converted to a slurry again and centrifuged again a total of five times with 250 ml of distilled water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h. The catalyst is heated to 70° C. under nitrogen and then reduced with a stream of 5% by volume of H.sub.2 in N.sub.2 for 3 h and cooled down to room temperature, and the gas stream is ended. The performance of the catalyst was evaluated in a batch test (see Table 1).

Example 9

(57) Preparation of a Catalyst with 1.4% by Weight of Au on TiO.sub.2 Support (Calcined for 4.5 h)

(58) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 support (product name: Aeroperl P25, Evonik Industries, prepared from TiCl.sub.4 by flame pyrolysis, grain size d50=20 μm, 75% anatase, 25% rutile) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is centrifuged while hot and decanted, and the residue is converted to a slurry again and centrifuged again a total of five times with 250 ml of distilled water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h. The catalyst is heated to 70° C. under nitrogen and then reduced with a stream of 5% by volume of H.sub.2 in N.sub.2 for 4.5 h and cooled down to room temperature, and the gas stream is ended. The performance of the catalyst was evaluated in a batch test (see Table 1).

Example 10

(59) Preparation of a Catalyst with 1.4% by Weight of Au on TiO.sub.2 Support (Formate, Reduced)

(60) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 support (TP Hombikat, Huntsman, d50=about 40 μm—development product as stationary phase of HPLC columns, 95% anatase structure) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. the suspension is vacuum-filtered while still hot through a suction filter and washed 5 times with 250 ml of dist. water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h and then stirred in a 2% by weight solution of sodium formate in methanol at 70° C. for 4 h, filtered, washed to neutral pH and dried again in a thin layer at 120° C. for 16 h. The performance of the catalyst thus obtained was evaluated in a batch test (see Table 1).

Example 11

(61) Preparation of a Catalyst with 1.4% by Weight of Au on TiO.sub.2 Support (Formate, Reduced in Water)

(62) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 support (TP Hombikat, Huntsman, d50=about 40 μm—development product as stationary phase of HPLC columns, 95% anatase structure) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is vacuum-filtered while still hot through a suction filter and washed 5 times with 250 ml of dist. water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h and then stirred in a 2% by weight solution of sodium formate in water at 70° C. for 4 h, filtered, washed to neutral pH and dried again in a thin layer at 120° C. for 16 h. The performance of the catalyst thus obtained was evaluated in a batch test (see Table 1).

Example 12

(63) Preparation of a Catalyst with 1.4% by Weight of Au on TiO.sub.2 Support (Methanol, Reduced with O.sub.2)

(64) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 support (TP Hombikat, Huntsman, d50=about 40 μm—development product as stationary phase of HPLC columns, 95% anatase structure) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is vacuum-filtered while still hot through a suction filter and washed 5 times with 250 ml of dist. water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C., then stirred in methanol and sparged with oxygen for one hour, then filtered, washed to neutral pH and dried again in a thin layer at 120° C. for 16 h. The performance of the catalyst thus obtained was evaluated in a batch test (see Table 1).

Example 13

(65) Preparation of a Catalyst with 1.4% by Weight of Au on TiO.sub.2 Support (Methanol, Reduced with N.sub.2)

(66) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 support (TP Hombikat, Huntsman, d50=about 40 μm—development product as stationary phase of HPLC columns, 95% anatase structure) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is vacuum-filtered while still hot through a suction filter and washed 5 times with 250 ml of dist. water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h, then stirred in methanol and left under a nitrogen atmosphere during this time, then filtered, washed to neutral pH and dried again in a thin layer at 120° C. for 16 h. The performance of the catalyst thus obtained was evaluated in a batch test (see Table 1).

Example 14

(67) Preparation of a Catalyst with 0.7% by Weight of Au on TiO.sub.2 Support (d50=40 μm)

(68) In a glass flask with mechanical stirring, 0.92 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 support (TP Hombikat, Huntsman, d50=about 40 μm—development product as stationary phase of HPLC columns, 95% anatase structure) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is vacuum-filtered while still hot through a suction filter and washed 5 times with 250 ml of dist. water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h. The catalyst is heated to 70° C. under nitrogen and then reduced with a stream of 5% by volume of H.sub.2 in N.sub.2 and cooled down to RT, and the gas stream is ended. The performance of the catalyst was evaluated in a batch test [see Table 1 (batch test)].

Example 15

(69) Preparation of a Catalyst with 0.13% by Weight of Au on TiO.sub.2 Support (d50=40 μm)

(70) In a glass flask with mechanical stirring, 0.19 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 support (TP Hombikat, Huntsman, d50=about 40 μm—development product as stationary phase of HPLC columns, 95% anatase structure) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is vacuum-filtered while still hot through a suction filter and washed 5 times with 250 ml of dist. water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h. The catalyst is heated to 70° C. under nitrogen and then reduced with a stream of 5% by volume of H.sub.2 in N.sub.2 and cooled down to room temperature, and the gas stream is ended. The performance of the catalyst was evaluated in a batch test [see Table 1 (batch test)].

Example 16

Preparation of an Au Catalyst on TiO.SUB.2./ZrO.SUB.2 .Support

(71) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2/ZrO.sub.2 support (ZrO.sub.2 (15%) on TiO.sub.2 (anatase), Huntsman, d50=about 40 μm) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is vacuum-filtered while still hot through a suction filter and washed 5 times with 250 ml of dist. water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h. The catalyst is heated to 70° C. under nitrogen and then reduced with a stream of 5% by volume of H.sub.2 in N.sub.2 and cooled down to room temperature, and the gas stream is ended. The performance of the catalyst was evaluated in a batch test; see Table 1.

Example 17

Preparation of an Au Catalyst on TiO.SUB.2./SiO.SUB.2 .Support

(72) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2/SiO.sub.2 support (SiO.sub.2 (8.5%) on TiO.sub.2 (anatase), Huntsman, d50=about 40 μm) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is vacuum-filtered while still hot through a suction filter and washed 5 times with 250 ml of dist. water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h. The catalyst is heated to 70° C. under nitrogen and then reduced with a stream of 5% by volume of H.sub.2 in N.sub.2 and cooled down to room temperature, and the gas stream is ended. The performance of the catalyst was evaluated in a batch test; see Table 1.

Example 18

Preparation of an Au Catalyst on TiO.SUB.2./Al.SUB.2.O.SUB.3 .Support

(73) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2/Al.sub.2O.sub.3 support (Al.sub.2O.sub.3 (8.6%) on TiO.sub.2 (anatase), Huntsman, d50=about 40 μm) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is vacuum-filtered while still hot through a suction filter and washed 5 times with 250 ml of dist. water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h. The catalyst is heated to 70° C. under nitrogen and then reduced with a stream of 5% by volume of H.sub.2 in N.sub.2 and cooled down to room temperature, and the gas stream is ended. The performance of the catalyst was evaluated in a batch test; see Table 1.

Example 19

Preparation of an Au Catalyst on TiO.SUB.2 .(Rutile) Support

(74) In a glass flask with mechanical stirring, 1.82 g of auric acid (HAuCl.sub.4*3H.sub.2O) are dissolved in 600 g of water and heated to internal temperature 70° C. The pH of the solution thus obtained is about 2.2 and is adjusted to 7.5 with 0.5 M Na.sub.2CO.sub.3 solution and then maintained over a period of 10 minutes. Subsequently, 50 g of TiO.sub.2 (TiO.sub.2 (rutile), Huntsman, d50=about 40 μm) are added all at once while stirring (400 rpm). The suspension thus obtained is stirred at 70° C. for 60 min, keeping the pH constant at 7.5. The pH rises into the basic range during this period and is kept at 7.5 with 2.0 M HNO.sub.3 solution and 0.5 M Na.sub.2CO.sub.3 solution. The suspension is vacuum-filtered while still hot through a suction filter and washed 5 times with 250 ml of dist. water each time (5 ml of dist. water/g of support). The residual conductivity of the wash water obtained is to be below 100 μS/cm.sup.2. The catalyst obtained is dried in a thin layer at 120° C. for 16 h. The catalyst is heated to 70° C. under nitrogen and then reduced with a stream of 5% by volume of H.sub.2 in N.sub.2 and cooled down to room temperature, and the gas stream is ended. The performance of the catalyst was evaluated in a batch test; see Table 1.

Example 20

Comparative Example 1

Ni—Au Catalyst

(75) According to the patent literature (EP2210664, Ex. 1), a nickel-gold catalyst was prepared. The performance of the catalyst was evaluated in a batch test (see Table 1) and continuously (see Table 2).

Example 21

Comparative Example 2

Co—Au Catalyst

(76) According to the patent literature (WO2017084969, Ex. 4), a cobalt-gold catalyst was prepared. Performance of the catalyst was evaluated in a batch test (see Table 1) and continuously (see Table 2).

Example 22

(77) The batch size from Example 2 was scaled up by a factor of seven and reproduced a total of three times with 350 g of support used. The activity and selectivity of the individual reproductions—even by comparison with Example 2—did not result in any scatter beyond the limit of measurement accuracy. The catalyst batches thus obtained were combined and mixed, and 600 g that were taken from this mixture were initially charged in a 20 l tank.

(78) Methacrolein is fed continuously at a constant rate of 500 g/h into the 20 l stirred tank reactor that was sparged from below (sparging with air up to a residual oxygen content of up to 4% by volume in the offgas) under a pressure of 5 bar and at internal temperature 80° C. For every mole of methacrolein fed in, 4 mol of methanol were simultaneously fed into the reactor, and this ratio was always kept constant. At the same time, a sufficient amount of NaOH solution (in nnethanol/H.sub.2O (95:5)) was fed into this reactor that the value in the reactor remained constant at pH=7. The reaction mixture was withdrawn continuously from the reactor via a filter. After the time specified in Table 4, the product samples were taken and analysed by means of GC and HPLC. The value for methacrylic acid in particular can be determined particularly precisely in this way, even in the form of its sodium salt formed in situ. The formation of methyl isobutyrate drops very significantly within the first two days and, after about four days, reaches a steady-state value of 50 ppm which remains substantially constant over the entire duration of the experiment. The activity is constantly stable and high. Table 4 and FIGS. 1 to 4.

(79) Batch Tests for MMA Preparation (Details of Table 1)

(80) A gold-containing catalyst according to one of Examples 1 to 21 (384 mg), methacrolein (1.20 g) and methanol (9.48 g) were stirred at internal temperature 60° C. and a pressure of 30 bar in an atmosphere of 7% by volume of O.sub.2 in N.sub.2 in a 140 ml steel autoclave with a magnetic stirrer for 2 h. After 2 h, the mixture was cooled down, degassed, filtered and analysed by means of GC. Each catalyst was tested at least twice under identical conditions; the results of the respective experiments were averaged. The resulting conversion of methacrolein (C(MAL), in %) and the selectivity for MMA (S(MMA), in %) for every catalyst tested are collated in Table 1 below.

(81) Continuous Test for Preparation of MMA (General Description for Table 2)

(82) The pH of a 42.5% by weight solution of MAL in methanol is adjusted to pH=7 while stirring by the addition of a 1% by weight solution of NaOH in methanol. This solution is fed continuously at a constant rate of addition to a stirred and sparged stirred tank reactor (sparging with air) under pressure of 10 bar and at internal temperature of 80° C. At the same time, this reactor containing 20 g of powder catalyst (from a previous example) is fed with a sufficient amount of 1% by weight NaOH solution (in methanol) that the value pH=7 in the reactor remains constant. The reaction mixture was withdrawn continuously from the reactor via a filter. After the time specified in Table 2, the product samples were taken and analysed by means of GC.

(83) TABLE-US-00001 TABLE 1 Catalyst evaluation by batch test Conversion Selectivity Ex. No. Catalyst Variation [MAL] [MMA] Catalyst 1 1.4% Au on TiO2 P25, Evonik Industries 93.1% 94.9% grain 2 1.4% Au on TiO2 Hombikat, Huntsman 91.7% 90.9% 3 1.1% Au on TiO2 P25 H6, Evonik Industries 95.3% 91.0% (recalcined) pelletized & recalcined --> 4 1% Au on TiO2 without mortar crushing 75.1% 93.2% 2 × 5 mm 5 1% Au on TiO2 mortar-crushed and sieved 87.8% 93.0% to 36-100 μm 6 1% Au on TiO2 mortar-crushed and sieved 88.5% 92.7% to <36 μm Calcination 7 1.4% Au on TiO2 uncalcined 33.1% 96.6% method 8 1.4% Au on TiO2 calcined 3 h, 70° C., 92.7% 93.0% N2, H2 stream 9 1.4% Au on TiO2 calcined 4.5 h, 70° C., 93.2% 91.1% N2, H2 stream 10 1.4% Au on TiO2 2% by weight of formate 68.4% 89.1% in methanol 11 1.4% Au on TiO2 2% by weight of formate in water 12 1.4% Au on TiO2 1 h, in MeOH with O2 96.1% 93.4% 13 1.4% Au on TiO2 1 h, in MeOH with N2 Amount 14 0.7% Au on TiO2 Hombikat, Huntsman 70.2% 87.9% of gold 15 0.1% Au on TiO2 Hombikat, Huntsman 18.4% 58.7% Mixed 16 Au on TiO2/ZrO2 Huntsman, support 65.9% 81.1% support 17 Au on TiO2/SiO2 Huntsman, support 18 Au on TiO2/Al2O3 Huntsman, support 19 Au on TiO2 (rutile) Huntsman, support Comparison 20 NiAu on SiO/Al2O3/MgO EP2210664, Ex. 1 81.3% 93.2% 21 CoAu on SiO/Al2O3/MgO WO2017084969, Ex. 4 80.1% 93.2%

(84) TABLE-US-00002 TABLE 2 Catalyst evaluation by continuous testing Selectivity % Batch test Cat. from TOS Conversion Selectivity [hydrogenated No. Ex. No. Catalyst [h] % [MAL] % [MMA] MMA] (reference) 2 1.4% Au on TiO2 21 77.1 92.1 0.05 2 495 77.89 92.4 0.01 999 78.5 91.3 0.01 5 1.0% Au on TiO2, 28 85.1 92.2 0.06 5 mortar-crushed 77 83.1 92.1 0.01 240 78.2 92 0.01 500 71.3 92.1 0.01 20 NiAu catalyst 22 75.1 92 0.01 20 525 74.8 91.9 0.13 676 74.6 92.1 0.12 1000 74.1 91.9 0.13 21 CoAu catalyst 50 73.5 92.3 0.02 21 200 71.09 91.8 0.08 380 70.8 91.5 0.09 480 70.6 92.1 0.08

(85) Continuous Test for Preparation of MMA at Varied pH (Description for Table 3)

(86) The pH of a 42.5% by weight solution of MAL in methanol is adjusted to pH=7 while stirring by the addition of a 1% by weight solution of NaOH in methanol. This solution is fed continuously at a constant rate of addition to a stirred and sparged stirred tank reactor (sparging with air) under pressure of 10 bar and at internal temperature of 80° C. At the same time, this reactor containing 20 g of powder catalyst from Example 2 is fed with a sufficient amount of 1% by weight NaOH solution (in methanol) that the pH in the reactor remained at a constant setting of the value mentioned in Table 3. The reaction mixture was withdrawn continuously from the reactor via a filter. At least three product samples (with delay time 24 h) were taken in each case and analysed by means of GC before the pH was varied.

(87) TABLE-US-00003 TABLE 3 Cat. from Selectivity Selectivity Ex. Conversion Selectivity % [DiMAL & % [dimethoxy- Selectivity No. Catalyst pH % [MAL] % [MMA] DiMAL ester] isobutene] % [3-MMIB] 2 1.4% Au 6.6 78.7% 91.7% 0.248% 0.228% 0.792% on TiO2 7.0 86.3% 92.4% 0.084% 0.064% 0.813% 7.6 92.2% 93.0% 0.037% 0.017% 0.879% 8.0 96.8% 91.9% 0.011% 0.007% 0.912% 8.3 98.1% 92.1% 0.012% 0.001% 0.961%

(88) TABLE-US-00004 TABLE 4 % wt. % wt. % MAL Hydro- wt. % wt. % wt. % Di- wt. % hr mol MAL/ mol MMA/ mol/mol mol/mol mol/mol hr Conver- genated wt. % 2- Dimethoxy- wt. % wt. % 3- MAL Sum C4 Residence hr .Math. kg .Math. cat hr .Math. kg .Math. cat MeOH/MAL MeOH/MA H2O/MAL TOS sion MMA MMA MHIB isobutene MAA 3-MMIB MHIB ester Selectivities Time WHSV STY in feed L in reactor in reactor Continuous reaction regime in a 20 I pressure reactor. 14.75 82.32% 0.29% 73.64% 0.09% 0.11% 2.12% 1.14% 0.00% 0.00% 77.16% 8.74 11.51 7.11 4.05 22.31 4.81 22.75 77.75% 0.14% 87.18% 0.10% 0.15% 3.01% 1.38% 0.00% 0.00% 91.82% 8.78 11.52 8.12 4.05 15.77 3.63 30.75 75.13% 0.09% 92.06% 0.10% 0.19% 3.82% 1.62% 0.00% 0.00% 97.82% 8.77 11.50 8.31 4.06 13.50 3.15 38.75 75.00% 0.07% 92.85% 0.09% 0.21% 4.33% 1.75% 0.00% 0.02% 99.32% 8.77 11.52 8.39 4.05 13.47 3.32 46.75 74.34% 0.06% 93.32% 0.09% 0.22% 4.28% 1.82% 0.09% 0.01% 99.89% 8.78 11.51 8.34 4.06 13.22 3.27 54.75 74.36% 0.05% 93.40% 0.09% 0.21% 4.21% 1.82% 0.00% 0.04% 99.82% 8.78 11.50 8.35 4.05 12.61 2.89 62.75 74.55% 0.03% 93.72% 0.08% 0.22% 4.30% 1.87% 0.00% 0.03% 100.25%  8.78 11.48 8.39 4.06 13.15 2.96 70.75 74.60% 0.05% 93.39% 0.08% 0.21% 4.29% 1.84% 0.00% 0.05% 99.90% 8.78 11.52 8.39 4.05 12.95 3.01 78.75 74.37% 0.04% 92.89% 0.08% 0.20% 4.13% 1.83% 0.04% 0.24% 99.41% 8.78 11.55 8.33 4.04 13.07 3.43 86.75 74.25% 0.04% 93.19% 0.07% 0.19% 4.36% 1.85% 0.04% 0.25% 99.98% 8.78 11.55 8.34 4.05 13.06 3.52 94.75 74.24% 0.04% 93.40% 0.07% 0.19% 4.42% 1.84% 0.03% 0.30% 100.28%  8.78 11.51 8.34 4.06 13.14 3.50 102.75 74.26% 0.04% 92.84% 0.07% 0.19% 4.24% 1.81% 0.04% 0.26% 99.49% 8.77 11.55 8.31 4.05 13.10 3.51 110.75 74.45% 0.04% 92.76% 0.07% 0.19% 4.25% 1.83% 0.03% 0.28% 99.45% 8.78 11.55 8.33 4.05 13.19 3.58 118.75 74.15% 0.04% 93.49% 0.07% 0.19% 4.45% 1.85% 0.03% 0.27% 100.39%  8.78 11.51 8.34 4.06 13.11 3.53 126.75 74.12% 0.04% 92.53% 0.07% 0.19% 4.37% 1.85% 0.04% 0.28% 99.37% 8.77 11.55 8.26 4.05 13.02 3.41 134.75 74.21% 0.04% 92.82% 0.07% 0.19% 4.20% 1.86% 0.04% 0.27% 99.49% 8.78 11.55 8.29 4.05 13.08 3.55 142.75 74.15% 0.03% 93.20% 0.07% 0.19% 4.35% 1.85% 0.05% 0.26% 99.99% 8.78 11.53 8.31 4.05 12.88 3.57 150.75 74.19% 0.03% 92.67% 0.07% 0.19% 4.30% 1.83% 0.03% 0.26% 99.40% 8.78 11.55 8.25 4.05 12.86 3.39 158.75 74.11% 0.03% 92.99% 0.07% 0.19% 4.26% 1.82% 0.04% 0.26% 99.67% 8.78 11.54 8.27 4.05 12.92 3.36 166.75 74.15% 0.03% 93.15% 0.07% 0.20% 4.29% 1.83% 0.04% 0.28% 99.86% 8.78 11.53 8.30 4.05 13.06 3.39 174.75 74.18% 0.03% 92.55% 0.07% 0.20% 4.33% 1.82% 0.05% 0.26% 99.31% 8.78 11.55 8.22 4.05 12.87 3.36 run time 182 h-358 h 182.75 74.21% 0.03% 93.17% 0.07% 0.20% 4.52% 1.84% 0.04% 0.26% 100.11%  8.78 11.54 8.27 4.05 12.95 3.45 190.75 74.25% 0.04% 93.14% 0.07% 0.20% 4.34% 1.84% 0.05% 0.28% 99.93% 8.78 11.56 8.28 4.05 12.90 3.36 198.75 74.18% 0.04% 92.66% 0.07% 0.20% 4.53% 1.86% 0.05% 0.27% 99.66% 8.78 11.54 8.19 4.05 12.80 3.42 206.75 74.21% 0.04% 92.85% 0.07% 0.20% 4.28% 1.86% 0.04% 0.28% 99.58% 8.78 11.55 8.18 4.05 12.66 3.34 214.75 74.19% 0.04% 93.01% 0.07% 0.20% 4.31% 1.84% 0.05% 0.28% 99.79% 8.77 11.55 8.24 4.05 12.87 3.51 222.75 74.18% 0.04% 93.17% 0.07% 0.20% 4.44% 1.87% 0.05% 0.28% 100.12%  8.78 11.51 8.18 4.05 12.74 3.38 230.75 74.10% 0.04% 93.40% 0.07% 0.20% 4.47% 1.87% 0.05% 0.29% 100.38%  8.78 11.50 8.18 4.06 12.64 3.35 239.25 74.18% 0.04% 93.25% 0.07% 0.20% 4.37% 1.86% 0.05% 0.27% 100.06%  8.78 11.54 8.24 4.04 12.72 3.29 246.75 74.24% 0.04% 92.68% 0.07% 0.20% 4.37% 1.87% 0.04% 0.28% 99.50% 8.78 11.55 8.12 4.04 12.49 3.31 254.75 74.23% 0.04% 93.39% 0.07% 0.20% 4.42% 1.87% 0.04% 0.30% 100.30%  8.78 11.52 8.15 4.06 12.47 3.26 262.75 74.12% 0.04% 93.92% 0.07% 0.20% 4.42% 1.87% 0.04% 0.24% 100.77%  8.78 11.51 8.21 4.06 12.55 3.25 270.75 74.31% 0.04% 92.63% 0.07% 0.20% 4.37% 1.88% 0.04% 0.25% 99.45% 8.78 11.52 8.06 4.05 12.66 3.23 278.75 74.24% 0.04% 93.20% 0.07% 0.20% 4.43% 1.90% 0.04% 0.29% 100.14%  8.78 11.49 8.08 4.06 12.40 3.22 286.75 74.19% 0.04% 93.43% 0.07% 0.20% 4.44% 1.89% 0.04% 0.28% 100.35%  8.79 11.48 8.12 4.06 12.47 3.28 294.75 74.11% 0.04% 92.81% 0.07% 0.20% 4.46% 1.88% 0.04% 0.28% 99.75% 8.78 11.53 8.05 4.04 12.74 3.21 302.75 74.15% 0.04% 92.97% 0.07% 0.21% 4.37% 1.88% 0.04% 0.28% 99.81% 8.78 11.52 8.05 4.05 12.50 3.27 326.75 74.27% 0.04% 93.33% 0.07% 0.21% 4.45% 1.90% 0.04% 0.27% 100.29%  8.78 11.55 8.08 4.03 12.71 3.76 334.75 74.24% 0.04% 93.24% 0.07% 0.22% 4.50% 1.90% 0.04% 0.29% 100.30%  8.79 11.56 8.11 4.03 12.54 3.65 342.75 74.30% 0.04% 93.06% 0.07% 0.22% 4.76% 1.88% 0.03% 0.29% 100.41%  8.78 11.56 8.09 4.03 12.50 3.67 350.75 74.24% 0.04% 93.00% 0.07% 0.21% 4.61% 1.93% 0.04% 0.29% 100.24%  8.78 11.55 8.09 4.03 12.73 3.66 358.75 74.24% 0.04% 93.17% 0.07% 0.22% 4.59% 1.91% 0.04% 0.29% 100.45%  8.78 11.53 8.05 4.04 12.62 3.66 % wt. % wt. % MAL Hydro- wt. % wt. % wt. % Di- wt. % hr mol MAL/ mol MMA/ mol/mol mol/mol mol/mol hr Conver- genated wt. % 2- Dimethoxy- wt. % wt. % 3- MAL Total C4 Residence hr .Math. kg .Math. cat hr .Math. kg .Math. cat MeOH/MAL MeOH/MA H2O/MAL TOS sion MMA MMA MHIB isobutene MAA 3-MMIB MHIB ester selectivity time WHSV STY in feed L in reactor in reactor run time 382 h-422 h 382.75 74.15% 0.04% 91.52% 0.07% 0.22% 4.52% 1.94% 0.03% 0.27% 98.04% 8.78 11.52 8.00 4.05 12.82 3.47 390.75 74.33% 0.04% 93.16% 0.07% 0.21% 4.32% 1.90% 0.04% 0.26% 99.96% 8.78 11.50 8.04 4.07 12.53 3.53 398.75 74.25% 0.04% 93.25% 0.07% 0.22% 4.51% 1.93% 0.04% 0.29% 100.24%  8.78 11.50 8.01 4.05 12.63 3.52 406.75 74.33% 0.04% 93.88% 0.07% 0.23% 4.38% 1.93% 0.04% 0.30% 100.76%  8.78 11.52 8.04 4.04 12.67 3.46 414.75 74.15% 0.04% 93.26% 0.07% 0.24% 4.52% 1.89% 0.03% 0.31% 100.83%  8.78 11.51 7.96 4.04 12.73 3.44 422.75 74.19% 0.04% 93.35% 0.07% 0.24% 4.66% 1.94% 0.03% 0.31% 100.62%  8.79 11.51 7.97 4.04 12.65 3.49 Key: TOS = time on stream; MAL conversion = conversion of methacrolein; Hydrogenated MMA = methyl isobutyrate; MMA = methyl methacrylate; 2-MHIB = methyl 2-hydroxyisobutyrate; MAA = methacrylic acid; 3-MMIB = methyl 3-methoxyisobutyrate; 3-MHIB = methyl 3-hydroxyisobutyrate;