DE1468385C3 - Process for the production of methacrylonitrile from isobutylene - Google Patents

Description

It is known that unsaturated nitriles can be prepared by passing an olefin, oxygen and ammonia in the gas phase over a solid catalyst which is kept at a suitable temperature. Recently, a process for carrying out this reaction has become known in which a catalyst is used which contains bismuth salts, tin salts or antimony salts of molybdic acid, phosphomolybdic acid or phosphotungstic acid. This process is important because acrylonitrile can be obtained in relatively high yields, in particular Wismutphosphormolybdat when used as a catalyst (see FIG. Belgian Patent Document 5 ^"7 I 200, USA. Patent 2 904 580, British Patent 867,438 and Japanese Patent publication 870/1961 5). It should be noted, however, that high conversions and good yields can only be achieved if acrylonitrile is synthesized from propylene. When methacrylonitrile is to be produced from isobutylene by the above method (namely, using bismuth phosphomolybdate as a catalyst), sufficiently satisfactory results cannot be obtained. While z. B. the above Belgian patent 19 gives examples in which using bismuth phosphorus molybdate acrylonitrile is produced from propylene with a conversion (in one pass) of a maximum of 52.8% (see. Examples and 10), it brings only one example in which methacrylonitrile is produced from isobutylene with a conversion of only 10% using the same catalyst. In this case, moreover, the yield of useful products including acetonitrile and the like is only 30%. (In other words, the yield of methacrylonitrile alone can be less than this value.) U.S. Patent 2,904,580 and Japanese Patent Publication 5,870/1961, which correspond to the Belgian patent, make no reference to the conversion of isobutylene; their claims are limited to the production of acrylonitrile from propylene. It follows that the production of an unsaturated nitrile from an olefin depends on the particular olefin and that the activity of a catalyst with respect to the conversion of the olefin into the unsaturated nitrile varies depending on the structure of the olefins. The same situation exists with Australian patent 56 564 and the corresponding Belgian patent 587493, which describe a process for the preparation of acrylonitrile (or methacrylonitrile) from propylene (or isobutylene), molecular oxygen and ammonia in the presence of a cobalt molybdate catalyst. A comparison of Examples 4 and 5 of this Australian patent clearly shows the difficulties encountered in the preparation of methacrylonitrile from isobutylene. In Example 4, which relates to the production of acrylonitrile from propylene, the conversion into acrylonitrile is 54.8% (corresponding to a yield of 58.6%), based on the propylene used. In Example 5, in which the same conditions as in Example 4, but are applied to isobutylene instead of propylene, the conversion to methacrylonitrile, based on isobutylene used, is only 6.2% and the yield is 13% based on converted isobutylene.

From French patent specification 1 219 512 and the corresponding German patent specification 1 162 830 is the production of acrylonitrile by ammoxidation of propylene using Mo-Na catalysts or Mo-K catalysts. Methacrylonitrile is only mentioned there by name. Establishing this connection is not explained further there. '

The German patent specification 941 428 describes in Example 4 the production of methacrylonitrile from isobutylene, by a two step method in which using a catalyst consisting of CuO-Se from isobutylene first methacrolein and then out \ the obtained in this reaction gaseous

'Product methacrylonitrile is _{formed from MoO 3} using a catalyst, with a methacrylonitrile yield, based on the isobutylene used, of only 20%.

In the French patent 1 268 324 the oxidation or ammoxidation of propylene is using catalysts made from oxides of the Mo-As-Bi system. An embodiment for the production of methacrylonitrile is not included in this patent.

Belgian patent 598 511 and the corresponding French patent 1 278 289 also only describe the production of acrylonitrile by ammonoxidation of propylene using a catalyst with the composition Me ₀ Bi _n P ₀ O ,, (Me means Fe, Co, Ni, V, Mn or Cr) are described, but the production of methacrylonitrile is not explained using an example.
According to the French patent specification 1255 121, catalysts are used for the ammonoxidation of olefins, which are composed of oxides of at least two elements or element combinations of the group consisting of Cu, V, Mo, Te, Bi and P-Mo

stand. There, too, the production of methacrylonitrile is not explained in an exemplary embodiment.

Accordingly, these patent specifications do not provide an example of the production of methacrylonitrile by ammonoxidation of isobutylene using a one-step process. The German patent 941 428 and the above-mentioned Belgian patent 587 493 or the corresponding Australian patent 56 564 also had to be inferred due to the large differences between the yields of acrylonitrile and methacrylonitrile that persisted there that between the production of acrylonitrile and the There is no analogy to the production of methacrylonitrile. Belgian patent 596 854 (which corresponds to French patent 1268 383) must also lead to this conclusion. This patent relates to a process in which propylene and isobutylene are reacted under the same conditions and using tellurium oxide catalysts. According to Example 4 there, the conversion into acrylonitrile, based on the propylene used, is 26.0%, but the Ί conversion into methacrylonitrile, based on the isobutylene used, is only 15.8%.

The French patent specification 1098 400, similar to the German patent specification 941 428, also relates to a two-stage process using a Cu ₂ O catalyst in the first stage and a MoO ₃ catalyst in the second stage. According to Example 2 there, the yield of methacrylonitrile, based on converted isobutylene, is 64.2%, but this patent specification does not contain any information on the conversion of the starting compound achieved.

French patent 1,261,568 describes the production of acrylonitrile and methacrylonitrile by ammonoxidation of propylene or isobutylene, with oxides of the Sb-Sn system as catalysts be used. According to Example 6 there, the conversion of isobutylene into Methacrylonitrile 41%.

From French patent specification 1,269,382 a process for the ammonoxidation of olefins is disclosed 7) known using a catalyst containing a large amount of phosphoric acid and a small amount Contains amount of an activator. According to Examples 7, 8 and 9 of this patent specification does this in the process converted isobutylene 56%, 54% and 53%, the yield of methacrylonitrile (based on converted Isobutylene) 67%, 67% and 73% and the conversion of isobutylene used into methacrylonitrile 38%, 36% and 39% off. In this known method, large amounts of water are used. So are given in Examples 7.8 and 9 molar ratios of water to isobutylene of 7.0. The usage from large amounts of water is disadvantageous. But if with the known method smaller amounts If water are used, it is obtained as explained in Comparative Experiment 8 below the catalysts known from this patent have poorer results.

The invention is based on the object of a process for the production of methacrylonitrile by Implementation of a mixture of isobutylene, molecular oxygen or air and ammonia in the Gas phase at elevated temperature in the presence of water vapor and a silica containing, To provide fixed or fluidized bed catalyst containing phosphorus and molybdenum, with at the same time a high conversion and high yields of methacrylonitrile without the use of larger ones Amounts of water and can be achieved in one step.

This object is achieved according to the invention in that the reaction is carried at 400 to 500 ⁰ C, .in the starting mixture, the molar ratios of oxygen to isobutylene at least 1.6: 1, ammonium

ίο niac to isobutylene 0.8: 1 to 1.5: 1 and water to isobutylene 0.5: 1 to 3: 1 and a catalyst is used which, in addition to silica, consists of oxides of the elements phosphorus, antimony, molybdenum and bismuth , the composition with regard to the elements phosphorus, antimony, molybdenum and bismuth of the formula P _n Sb ₀ Mo ₆ Bi ₁ . where η = 0.5 to 2, a = 0.5 to 20, b = 1 to 18, c = 0.5 to 11, a + b + c = 22 and c ^ 4.4 + b .
The invention achieves that methacrylonitrile is obtained from isobutylene with an improved conversion and in yields of over 40%, without the need to use large amounts of water.

The process of the invention can convert isobutylene to methacrylonitrile up to 65% and even more can be achieved. According to the prior art, it was not possible when using a molar ratio of water to isobutylene of only 0.5: 1 to 3: 1 methacrylonitrile from isobutylene with such a high conversion in yields in excess of 40%.

As indicated above, a catalyst composed of oxides of the ternary system P-Mo-Bi can give satisfactory results in the production of acrylonitrile from propylene, while it gives only very poor yields in the production of methacrylonitrile from isobutylene. A catalyst composed of oxides of the ternary system P-Mo-Sb has a lower catalytic activity than a catalyst composed of oxides of the P-Mo-Bi system both in the production of acrylonitrile and in the production of methacrylonitrile. Taking these facts into account, it is surprising that a catalyst composed of oxides of the P-Sb-Mo-Bi system with a special atomic ratio shows a particularly excellent: catalytic activity in the conversion of isobutylene into methacrylonitrile. However, this effect is only given if the catalyst has the specified atomic ratio. A catalyst made of oxides of the P-Sb-Mo-Bi system having an unsuitable atomic ratio may be inferior to a catalyst made of oxides of the P-Mo-Bi system and even inferior to that made of oxides of the P-Mo-Sb system. So shows z. B. a catalyst made of oxides of the P-Sb-Mo-Bi system with the empirical formula PSb ₃ Mo ₅ Bi ₁₄ . almost no catalytic activity in the conversion of isobutylene to methacrylonitrile.

Of course, the above formula should

^ P "Sb _a Mo _b Bi _c

determine only the atomic ratio of four components, namely P, Sb, Mo and Bi; it does not refer to any compound that conforms to this formula. 1
⁶ S The F i g. 1 and 2 are based on the triangular coordinate system of a ternary system Sb-Mo-Bi in which the sum a + b + c = 22. F i g. 1 shows the experiments that were carried out with P = 1,

the accompanying example of the invention (O) and the comparative experiment (Δ) being plotted on the triangular coordinate system, corresponding to the elemental composition of the catalyst components used in each of these experiments. FIG. 2 shows the conversion into methacrylonitrile as it was obtained in these experiments and at the same time reveals the optimum range (in the inner zone defined by the line HIJK).

F i g. 1 concerns the case where (a + b + c) = 22.

AG stands for α ^ _20, CD for a = 0.5, BC for b = 18. GF for b = 1, AB for c = 0.5 and DE for c = 11. EF stands for c - 4.4 + b. The elemental _{composition of P n} Sb _n Mo ₆ BIc consequently covers the entire zone within the heptagon ABCDEFG. In this case this zone is independent of the value for P; but the numbers or characters in FIGS. 1 and 2 relate to those experiments that were carried out with P = 1.

A suitable catalyst can be prepared by any known method. The catalyst is particularly advantageously produced by mixing phosphoric acid (P), antimony trioxide (Sb), molybdenum oxide or ammonium molybdate (Mo) and bismuth nitrate (Bi) with a carrier and water and then drying the ^{mixture obtained at 400 to 550 ° C.} calcined. A suitable support is a siliceous material such as diatomaceous earth, silica, etc. Since some of the above catalyst components are capable of chemically combining with a silica support, precisely such silica cannot actually be referred to as a support. However, since this fact is not essential for the present invention, silica is to be referred to here as the carrier.

Any suitable source of oxygen can be used for ammonia oxidation. From the economic From this point of view, air is particularly suitable. Isobutylene as a starting material does not necessarily need high purity to be; it can be saturated hydrocarbons, e.g. B. propane, butane and the like., Contain in the reaction zone are practically inactive. Ammonia can be used in the purity in which it is used for production of fertilizers.

A particularly suitable reaction temperature is in the range from 430 to 470 ^° C. The preferred contact time is 2 to 12, in particular 4 to 10 seconds. This contact time relates to the reaction temperature given above. So z. B. a contact time of 7 seconds at a reaction ^{temperature of 450 0} C a throughput rate of about 190 h " ¹ .

To have the greatest possible degree of conversion in a single pass is preferred used a contact time of more than 4 seconds.

According to one embodiment of the invention, it has proven to be advantageous when working with a fixed arranged catalyst to supply secondary air approximately in the middle of the conversion zone.

The preferred molar ratio of oxygen to isobutylene in the starting gas mixture is between 1.6 and 2.1. If the reaction is carried out with a fluidized bed, secondary air can be used instead an increased molar ratio between oxygen and isobutylene can be used in the starting gas mixture. To avoid carbon deposition and the decomposition or combustion of ammonia, it is ensured that a small amount of water is present in the source gas. The molar ratio from water to isobutylene is as indicated 0.5: 1 to 3.0: 1, but can be in the case of a catalyst with a particularly advantageous composition between 0.5: 1 to 1: 1.

The process of the invention can be carried out in any suitable manner in fixed or fluidized beds will. Since the reaction is exothermic to a considerable extent, it must be ensured that the Heat of reaction is derived if the reaction is to be carried out in a fixed bed. To this Purpose it is very advantageous to use pipes with the smallest possible diameter or to the catalyst to give a suitable diluent (e.g. porcelain or Raschig rings).

For a better understanding of the method according to the invention, reference is made to the examples and the comparative experiments. The F i g. 1 and 2 are in addition necessary. In FIG. 1 is the number of attempts in question on a trigonal coordinate system of the ternary system Sb-Mo-Bi depending on the elemental composition of the tests used catalysts entered. So called z. B. © is the elementary composition of the

Game 2 used catalyst and Δ \ that of a catalyst used in Comparative Experiment No. 1. The F i g. 2 is used to better understand the relationship between the conversion to methacrylonitrile and the atomic ratio of Sb-Mo-Bi. In Fig. 2 the conversion into methacrylonitrile is expressed by certain symbols at the point corresponding to the elemental composition of the particular catalyst used; consequently, FIG. 2 indicates the optimum range of the elemental composition of a usable catalyst, this range being characterized by the inner zone HIJK. The symbols used here mean the following: the conversion to methacrylonitrile is 70 to 65% for, 65 to 60% for, 60 to 55% for, 55 to 50% for, 50 to 45% for, χ 45 to 40%, for χ χ 40 to 30%, for χ χ χ 30 to 20%, for χ χ χ χ 20 to 10% and for χ χ χ χ χ below 10%.

As shown in FIG. 2, the conversion to methacrylonitrile is 65% or more when a catalyst having the particularly suitable composition in terms of atomic ratio as defined by HIJK is used. This crucial range can be expressed by α = 2.3 to 8.9, b = 6.2 to 12.7 and c = 6.9 to 7.9 (sum = 22).

Depending on the deviation of α, b and c from this range, the conversion to methacrylic tril decreases. Particular attention is drawn to the gradient from JK to DE . As shown by DE , the number of Bi atoms must not be more than the total number of Mo and Sb atoms, and as can be seen from EF , the number of Bi atoms must not be more than the sum of the number of Mo Atoms plus ¹ I _{5 of} the total number of Sb, Mo and Bi atoms.

6p The point plotted on the base line Mo-Bi of the triangular coordinate system corresponds to a catalyst that appears to be most suitable for the synthesis of acrylonitrile from propylene, namely a bismuth phosphorus molybdate (oxides of PMo ₁₂ Bi ₉ ). If, however, this catalyst is used for the synthesis of methacrylonitrile from isobutylene, it corresponds to the point χ χ χ according to the tests available; in this case, however, the conversion to

I 4öb3bö

Methacrylonitrile only 22% (yield 29%) (see comparative experiment no. 6). The symbol χ χ χ χ applied to the line Sb-Mo corresponds to the conversion to methacrylonitrile when one of the antimony-phosphomolybdate _catalysts (oxides of PSb 13i5 Mo ₈₅ ) is used; its value is only 16% (yield 22%) (see comparative experiment no. 5). The symbol χ χ χ χ χ on the line Bi-Sb corresponds to the conversion to methacrylonitrile when oxides of PSb ₁₆ Bi ₆ are used; their value is equal to zero (see comparative experiment no. 4). In the particular area within the triangular coordinates, there are catalysts with excellent activity for the ammonoxidation of isobutylene.

The above explanations have been given with respect to the change in a ternary system Sb-Mo-Bi when P is taken to be 1. When η is out of the range of 0.5 to 2, the conversion into and yield of methacrylonitrile decrease to some extent depending on the Sb-Mo-Bi atomic ratio.

The following examples and comparative experiments are intended to illustrate the invention. Some experiments in which η has a value other than 1 are given in Examples 28 to 39. The conversion to methacrylonitrile from isobutylene and the yield based on isobutylene are defined as follows:

Conversion to methacrylonitrile (%)

Moles of methacrylonitrile obtained

= · IUU,

Moles of isobutylene used

Yield (%) =

Moles of methacrylonitrile obtained
Moles of isobutylene converted

Isobutylene
ammonia
air
Steam

Mole percent

7th
9

70
14th

This gas mixture is through the reaction vessel at a flow rate of 301 / hour. directed. When this gas has passed through ² J _{1 of} the catalyst layer, additional secondary air in an amount of 7 1 / h. fed.

catalyst

564 parts of 10% silica sols, 2.2 parts of 85% phosphoric acid, 4.2 parts of antimony trioxide, 46.5 parts Ammonium molybdate and 59.6 parts of bismuth nitrate are mixed together thoroughly. The received The mixture is dried by heating, calcined and then shaped into 4 × 4 mm cookies. This catalyst contains the oxides of five components namely P, Sb, Mo, Bi, Si, and has the approximate elemental composition

PSb ₁₁₅ Mo ₁₄ Bi ₆₁₅ Si ₅₀ .

This catalyst is used in Example 1.

Other catalysts in which the atomic ratio of Sb-Mo-Bi is changed are used in similarly used in Examples 2-21.

results

The results of Examples 1 to 21 are shown in Table 1; schematically they are in the F i g. 1 and 2 are shown.

100.

Examples 22-27

B e i s ρ i e 1 e 1 to 21

In these examples, P _{n is} set as P ₁ , and the number of atoms of the three components Sb, Mo and Bi is changed so as to examine the relationship between the various catalysts and their catalytic activity. The experimental conditions are the same in all of these examples.

Test conditions

130 cm ^{3 of} a catalyst as described below and 70 cm ^{3 of} a diluent (iron rings) are placed in a reaction vessel, the heating part of which is about 500 mm long and made of a drawn steel tube with an approximate inner diameter of 27.6 mm and an approximate outer diameter of 34 , 0 mm. This reaction vessel is heated to the desired temperature (about 45O ⁰ C), with the aid of a bath of an existing mixture of equal parts of potassium nitrate and sodium nitrite. The gas used as the starting mixture for the reaction has the following composition:

In these examples the same experimental conditions and the same procedures are used Preparation of the catalysts as used in Examples 1 to 21, also the atomic ratio of Sb-Mo-Bi changed. The values for Si and P are the same as in Examples 1 to 21.

The results are compiled in Table 2 and entered schematically in FIGS. 1 and 2.

Comparative experiments No. 1 to 7

These comparative tests are outside the scope of the present invention. The experimental conditions and the processes for preparing the catalysts used are the same as in FIG previous examples, only the atomic ratio of Sb - Mo - Bi is changed. The values for Si and P are the same as in the previous examples.

The results obtained in these comparative experiments are summarized in Table 3 and also schematically in FIG. 1 and 2 entered. Exceeds in all of these comparative tests the conversion to methacrylonitrile is not 35% and the yield of methacrylonitrile is not 40%. As can be seen the desired catalytic activity is considerably lower if one component of the

three components, namely Sb, Mo, Bi, precipitate; a catalyst of the Sb-Mo-Bi system that meets the requirements does not correspond to the present invention, shows only a very low catalytic activity.

309 531/501

Table 1

10

example Sb Atomic ratio Bi Methacrylonitrile Conversion to Hydrocyanic acid 1 1 Yield to ' No. 1.5 Mon 6.5 53. Methacrolein 2 2 Methacrylonitrile 1 3.0 14.0 5.0 53 4th 1 0.4 63 2 4.0 14.0 6.5 58 4th 1 0.5 55 3 5.0 11.5 6.5 63 6th 1 1 64 4th 4.0 10.5 7.5 66 6th 2 5 65 5 5.5 10.5 5.0 56 4th 1 2 68 6th 3.5 11.5 10.0 52 10 3 3 58 7th 0.5 8.5 11.0 40 3 1 2 56 8th 8.0 10.5 7.0 65 2 3 50 9 15.0 7.0 3.5 55 5 2 70 10 5.0 3.5 8.5 64 11th 4th 60 11th 10.0 8.5 3.5 49 2 67 12th 11.0 8.5 6.0 54 18th 56 13th 6.0 5.0 7.5 65 12th 64 14th 2.5 8.5 2.5 42 6th 68 15th 19.0 17.0 1.5 45 6th 49 16 15.0 1.5 5.0 44 8th 59 17th 10.0 2.0 8.0 41 4th 58 18th 9.0 4.0 1.0 42 3 52 19th 6.0 12.0 1.0 41 8th 55 20th 4.0 15.0 1.0 43 6th 54 21 17.0 7th 57

Sb Atomic ratio Table 2 Bi Methacrylonitrile Conversion to Hydrocyanic acid Yield to BeisDiel 1.0 Mon 9.0 41 Methacrolein 1 Methacrylonitrile ^ ß WA hj ^ S IV 4
No. 2.0 12.0 8.5 59 5 . 3 57 22nd 1.0 11.5 8.0 42 3 1 65 23 2.0 · 13.0 9.5 57 6th 3 54 24 2.0 10.5 7.0 59 4th 2 65 25th 3.0 13.0 7.5 67 3 2 .63 26th 11.5 4th 68 27

Sb Atomic ratio
Mon Table 3 Bi Methacrylonitrile
(%) ' Conversion to
Methacrolein
(%) Hydrocyanic acid
(%) Yield to
Methacrylonitrile
(%) Comparative experiment
No. 3.0 5.0 14.0 0 1 1 0 1 8.0 4.5 9.5 21 4th 2 28 .2 2.0 8.0 12.0 32 4th 2 37 3 16.0 - 6.0 0 0.2 2. 0 4th 13.5 8.5 --- 16 25th 2 22nd 5 - 12.5 9.5 22nd 2 2 29 6th ■ ^s i, o 18.5 2.5 31 15th 2 37 7th

B e i s ρ i e 1 e 28 to 32

The catalysts used here are prepared in the same manner as in Example 1, only the amount of phosphoric acid is changed. The same test conditions are used for the implementation

as used in example 1. The results are shown in Table 4.

Examples '33 to 36

The catalysts used are prepared in the same way as in Example 9, only the Phos-

Amount of phosphoric acid changed. The same test conditions as in Example 9 are used for implementation applied. The results are shown in Table 4.

As can be seen from Table 4, optimal results are obtained when P _{n has} the value P ₁ . If you deviate from this value, the secondary

production of methacrolein significantly. This tendency is particularly evident from Examples 33 to 36, in which catalysts with a high Sb content are used. Where P = zero or P = 4 to 7, a yield of methacrylonitrile in excess of 40% can still be obtained, but the conversion to methacrylonitrile drops.

P. Atomic ratio Mon table 4th Conversion to Hydrocyanic acid
(%) Yield to 0 Sb 14.0 Methacrolein
(%) 2. Methacrylonitrile
(%) example
No. 0.5 1.5 14.0 Bi Methacrylonitrile
(%) 8th 2 57 28 1 1.5 14.0 6.5 42 5 2 59 29 2 1.5 14.0 6.5 49 4th 1 63 1 4th 1.5 14.0 6.5 53 5 2 65 30th 7th 1.5 14.0 6.5 52 5 4th 58 31 0 1.5 7.0 6.5 43 4th 2 55 32 0.5 8.0 7.0 6.5 46 8th 1 . 42 33 1 8.0 7.0 7.0 25th 6th 1 63 34 2 8.0 7.0 7.0 56 5 1 70 9 4th 8.0 7.0 7.0 65 12th 1 65 35 8.0 7.0 50 14th 47 36 7.0 38

P. Atomic ratio Mon table 5 Conversion to Hydrocyanic acid
(%) Yield to 0 Sb 11.5 Methacrolein
(%) 1 Methacrylonitrile
(%) example
No. 0.5 2.0 11.5 Bi Methacrylonitrile
(%) 3 2 41 37. 1 2.0 11.5 8.5 34 3 3 53 38 2 2.0 11.5 8.5 50 3 2 65 23 2.0 8.5 59 2 58 39 8.5 55.

J Examples 37 to 39.

The catalysts used here are in the Prepared in the same manner as in Example 23, except that the amount of phosphoric acid is changed. To implement the same conditions as in that Example 23 applied. The results are shown in Table 5.

Examples 40 to '42

Catalysts PSb ₆ Mo _{8 5} Bi _7i5 are produced in the same way as in Example 14, only the silica concentration of the catalyst is changed. The same test conditions as in Example 14 are used for implementation. The conversion to methacrylonitrile showed the following values :

example Si / P atomic ratio conversion
. in methacrylonitrile 40
14th
41
42 29
50
65
80 63%
65%
62%
55%

B e i s ρ i e 1 43

Using the catalyst of Example 14, the procedure of Example 14 is repeated, only the following implementation conditions are applied:

1. The starting gas mixture contains 6.5% isobutylene and 9.5% ammonia. . ,

2. Added secondary air: 141 / hour.

3. Additionally added catalyst: 70 ecm.

The conversion into methacrylonitrile is 68%, that in methacrolein 1 to 2%.

Example 44

The experiment of Example 43 was repeated, but the reaction temperature is not maintained at 45O ⁰ C, but at 430 ⁰ C. There is no significant change (68 to 69%) for the conversion to methacrylonitrile.

Examples 45 and 46

The mole ratio of water to isobutylene in Example 43 is about 2.1. This molar ratio

is changed here, but otherwise proceed as in example 43. The following results will be receive:

example Molar ratio
Water to isobutylene conversion
in methacrylonitrile 45
46 1.0
0.5 66%
62%

Examples 47 and 48

The experiments of Examples 45und46 are repeated, only a reaction temperature of 430 ⁰ C instead of 450 ⁰ C is used. Practically the same results are obtained.

example Molar ratio
Water to isobutylene conversion
in methacrylonitrile 47
48 1.0
0.5 65%
61%

Examples 49 to 53

The experiments described in these examples are carried out in a fluidized bed reactor.

To silica sol with a silicon dioxide content of 20 percent by weight is 85% phosphoric acid, Antimony trichloride, ammonium molybdate, bismuth nitrate and dilute nitric acid and that Well mixed. The dispersion obtained is

in a spray dryer with rotating atomizer disc with hot air at 200 ° C (outlet temperature) dried. The spherical particles thus obtained are placed in an electric furnace for 3 hours calcined at 450 ° C. It becomes a suitable catalyst for use in a fluidized bed reactor obtained with a particle diameter between 30 and 100 μ.

The following catalysts with different compositions are produced in this way:

Catalyst A: P Sb ₃ Mo ₇ Bi ₈ O ₄₀ (SiO ₂ ) S ₀
Catalyst B: P Sb ₅ Mo ₇ Bi ₆ O ₄₀ (SiO ₂ ) S ₀
Catalyst C: P Sb ₃ 5Mo ₁₂ Bi ₇ 5O ₅₅ (SiO ₂ ) S ₀
Catalyst D: P Sb ₃ Mo ₇ Bi ₈ O ₄₀ (SiO ₂ ) S ₀ *)

* Calcined) for 3 hours at 55O ⁰ C.

Examples 49 to 52

A reactor for fluidized bed reactions with the internal diameter given in Table 6 is used. The starting materials, isobutylene, ammonia, air and water vapor, are each below the conditions specified in Table 6 implemented. The test results are in Table 7 compiled.

Example 53

The reactor used in Example 49 is with

1.2 kg of catalyst C charged. The reaction is carried out under the conditions of Example 50. The conversion to methacrylonitrile is 55 to 57% and the yield of methacrylonitrile is 60 to 62%.

catalyst crowd
(kg) temperature
ro table 6th Raw materials Molar ratio
O ₂ to isobutylene Molar ratio
H ₂ O to
Isobutylene Inside
knife of the Art 1.2 454 Reaction conditions Molar ratio
NH ₃ TO
Isobutylene 2.96 2.50 Reactor
(mm) Reisniel A. 1.2 449 1.45 2.85 2.91 53 UvJjpivl B. 31 442 mission
(1 H.) 1.41 2.74 2.45 53 49 A. 3 450 424 1.29 2.60 1.13 155 50 D. 672 1.19 81 51 8510 52 1400

Table?

Conversion ii Methacrylic
nitrile Methacrolein 1 Hydrocyanic acid Yield to
Methacrylic
nitrile (%) example 58.5 4.1 5.4 69.4 49 61.6 4.4 1.2 64.8 50 58.9 4.9 2.0 69.6 51 67.6 2.3 4.6 70.9 52

and 30 parts of Cu (NO ₃ ) ₂ · 3H ₂ O are dissolved in 400 parts of an aqueous ammonia solution while hot; the resulting solution is then added to 1000 parts of boron phosphate. The thus impregnated boron phosphate is carried out at 15O ⁰ C for 4 hours and then pellets having a diameter of 4 mm and a length of 4 mm formed. The pellets are tempered at 500 ° C. for 4 hours.

Comparative experiment No. 8

Example 1 is repeated, but instead of the catalyst there, that in Example 9 the catalyst described in French patent 1,269,382 is used. This catalyst will prepared as follows: 31 parts of tungstic acid

example Test results Action in
Methacrolein
(%) blue
acid
(%) Yield to
Methacrylic
nitrile (%) 60 450 Urm
Methacrylic
nitrile
(%) 1 4th '27 475 21 2 4th 30th 65 500 28 3 5 27 27

1 sheet of drawings

Claims (2)

Patent claims: 1 Process for the preparation of methacrylonitrile by reacting a mixture of isobutylene, molecular oxygen or air and ammonia in the gas phase at elevated temperature in the presence of steam and a silica-containing, phosphorus and molybdenum-containing fixed or fluidized bed catalyst, characterized in that the reaction is carried out at 400 to 500 ⁰ C, the molar ratios of oxygen to isobutylene in the starting mixture are at least 1.6: 1, ammonia to isobutylene 0.8: 1 to 1.5: 1 and water to isobutylene 0.5: 1 to 3: 1 and a catalyst is used which, in addition to silica, consists of oxides of the elements phosphorus, antimony, molybdenum and bismuth, the composition of the elements phosphorus, antimony, molybdenum and bismuth _{corresponding to the formula P n} Sb ₀ Mo ₆ Bi ₀ , in which η = 0.5 to 2, a = 0.5 to 20, b = 1 to 18, c = 0.5 to 11, a + b + c = 22, and c ^ 4.4 + b . 2. The method according to claim 1, characterized in that when working with a fixed Secondary air is supplied to the catalyst approximately in the middle of the reaction zone.