US2620356A - Production of aromatic amines - Google Patents

Description

Dec. 2, 1952 J, c, MUNDAY 2,620,356

PRODUCTION OF AROMATIC AMINES Filed July 17, 1943 REACTOR 2e P ODUCT COOLER CATALYST LEVEL R y '1' I v r. I I

40 l I /SEPARATOR |TR8 1' FEEDST CK 2 28 IO 1 l 46 32 WATER DRAW OFF CATALYST A 34 COOLER PRODUCT 5 RECYCLE FRESH HYDROGEN PRODUCT Patented Dec. 2, 1952 PRODUCTION OF AROMATIC AMINES John C. Munday, Cranford, N. J., assignor to Standard Oil Development Company, a corporation of Delaware Application July 17, 1943, Serial No. 495,177

Claims.

The present invention relates to improvements in the art of preparing gasoline blending agents, and more particularly it relates to improved methods in the preparation of amino aromatics.

The evaluation of a motor fuel and, in particular, an aviation fuel not only includes its octane number but also includes its rich mixture performance. The art has developed a test known as the A. F. D.-3C test which is indicative of the expected performance of an aviation fuel under actual operating conditions. When applied to aviation fuels under rich mixture conditions, that is with a high fuel to air ratio, the test indicates the relative value of the fuels as regards power production, which is of great importance in takeofi under high loads or at maximum speed in flight.

A great many additives have been proposed previously for increasing the rich mixture per formance of an aviation fuel. These blending agents include aromatics, such as toluene and xylene, and certain branched chain paraffins, such as 2,3 dimethyl pentane and triptane. It has also been proposed heretofore to add amino aromatics, such as aniline, toluidine and xylidine. However, the commercial preparation of amino aromatics has been accomplished so far in liquid phase, using tin and iron, also hydrochloric acid, to reduce the corresponding nitro compound to the desired amine compound as where, for example, mononitroxylene is reduced to xylidine. This operation has many disadvantages, the most outstanding of which is that it is necessarily a batch operation.

I have found that amino aromatics may be produced continuously by hydrogenation of corresponding nitro aromatics in the presence of a suitable catalyst. An important feature of my invention involves activation of the catalyst to maintain it at a high conversion level, namely, at a level wherein the product of the desired aromatic amine is from 90100% of the theoretical amount obtainable from a given nitro aromatic.

The main object of the invention, therefore, is to produce aromatic amines and/or alkylated aromatic amines continuously and in good yields.

In the accompanying drawing I have shown diagrammatically, a form and arrangement of apparatus elements in which my invention may be carried into effect.

Tests of a large number of hydrogenation catalysts have shown that certain catalysts are superior to others from, the standpoint of yield and purity of product. Of the catalysts which I have tried, I have found that metallic sulfides activated with a volatile sulfide, such as HzS or carbon disulfide, give the best results from every standpoint and, in particular, with regard to the maintenance oi the high yields and the absence of degradation products.

Catalysts which contain free metals may be highly active initially at a low temperature but they rapidly lose activity, and furthermore, they cause cracking when the temperature is raised to compensate for loss in activity. Even the initial high activity of free metals at relatively low temperature is a disadvantage commercially, since the high heat of hydrogenation of nitro aromatics (about 110,000 calories per mol) makes it desirable to operate at a relatively high temperature; this permits a large differential between the operating temperature and both the feed inlet temperature and the coolant temperature.

On the other hand, the operating temperature is limited by the thermal cracking temperature of the particular nitro aromatic. Benzene and toluene mono nitro compounds are fairly stable. In the case of nitro compounds of higher arcmatics such as the xylenes and ethyl benzene, thermal decomposition begins at about 450 F. and may be explosively violent as 600 F. is approached, although they may be heated for a short time at 600 F. without experiencing severe decomposition, particularly if hydrogen is present. It can be seen, therefore, that from a practical standpoint, the operating temperature range for hydrogenation of nitro to amino groups is from about 400 F. to about 550 F. in the case of higher aromatics, such as xylenes and ethyl benzene, and slightly higher in the case of benzene and toluene. This narrow temperature range for commercial large scale operation makes it imperative to employ catalysts which not only are highly active in the temperature range but which have a fiat temperature vs. conversion curve within the temperature range. This will be obvious from the fact that the heat of hydro genation is sufficient to raise the temperature about 3600 F., and even when diluting the nitro aromatic with a heat carrier such as recycled amine or a hydrocarbon oil and injecting cold hydrogen to the extent of 1000% or more of the theoretical amount required, the temperaturerise in the catalyst bed will be of the order of IOU- F. r

I set forth below the results of a number of runs which I performed, and it will be understood that these specific examples are to be construed EXAMPLE 1 A mixture of 2-nitro, 1,4-dimethyl benzene and 4 volumes of meta xylidine (the latter a diluent used for the purpose of absorbing heat released during the hydrogenation), was hydrogenated at 400 F. by contacting the mixture with a tungsten sulfide catalyst and added hydrogen, the amount of hydrogen fed to the catalytic zone being 300% excess over that theoretically required to reduce the nitro compound. The feed rate was 1.5 volumes of the mixture of the nitro dimethyl benzene and the meta xylidine per volume of catalyst per hour, and a pressure of 500 lbs. per square inch gauge was maintained on the reaction zone. A product was continuously withdrawn during a period of 5.5 hours, and this product was completely soluble in 10% aqueous H01, indicating that the product was 100% amino.

EXAMPLE 2 In a second run made with the same catalyst,

the same reactants, and the same diluent and car- EXAMPLE 3 Three additional runs were carried out using the catalysts indicated in the table below and under the conditions therein stated. In the first two runs the catalyst was activated by passing HzS over the catalyst for about half an hour. It -will'be noted that the use of the hydrogen sulfide increased the activity of molybdenum sulfide catalyst and thus increased the yield. In the case of the nickel sulfide-tungsten sulfide catalyst, HzS was being formed during the run by reduction of a portion of the metal sulfides, which saturated the remainder with activating H28. Over a longer period of time it would have been necessary to replace the H28 carried out with the products. 'It should be noted that this sulfide catalyst has a very flat temperature-conversion curve, and that the conversion was leveling out at a relatively high value at each temperature level.

The nickel catalyst used above is not as satisfactory as molybdenum sulfide or nickel sulfidetungsten sulfide since the nickel oxide (present prior to the reduction period) has a steep temperature-conversion curve, whereas the reduced nickel loses activity rapidly. In addition, the products obtained with nickel contained black degradation products, while the products obtained with the sulfide catalysts were light in color, indicating purity.

EXAMPLE 4 A catalyst comprising F6203, 5% CuO, 5% K20 and 78.5% MgO was employed for the hydrogenation of nitrotoluene at atmospheric pressure, 0.47 v./v./hr., 300% excess hydrogen. The percent amine in the products obtained at various temperatures were as follows: 0% at 510 F., 20% at 525 F., at 545 F., and 100% at 570 F. The products were black, even at low conversion levels, and gave a heavy black precipitate when added to HCl solution, in striking contrast to the light colored products obtained with sulfide catalysts which give almost colorless HCl solutions. A catalyst having such a steep temperature-conversion curve in the desired operating temperature range would give very unsatisfactory results in large scale operations.

EXAMPLE 5 A catalyst of the same composition as employed in Examples 1 and 2, comprising tungsten sulfide which had been prepared by decomposing ammonium thio tungstate was employed for the hydrogenation of various nitro aromatics at atmospheric pressure and at various temperatures. The feed rate was 0.24 v./v./hr. for the first 37 hours, and 0.30 v./v./hr. thereafter. The hydrogen rate was 300% excess of the theoretical amount required for hydrogenation. Feeding ortho nitrotOluene for the first 30 hours, the conversion to amine at 400 F. was 100% at the first hour and 97% at the tenth hour. The temperature was then raised to 500 F. and the conversion was 100% at the eleventh hour and 89% at the twenty-sixth hour. Carbon disulfide was then added to the feed stock in an amount equivalent to 2.5 weight percent. This activated the catalyst so that the conversion Hydrogenation of nitro-toluene300% excess hydrogen 0.47 v./v./hr. atmospheric pressure CATALYST-MOLYBDENUM DISULFIDE Hour 1 2 3 1%; 4 5 6 Temp., F- 500 550 600 600 .600 .600 .600 Percent Amine in Product 32 32 22 55 50 CATALYST NICKEL SULFIDE (1 MOL); TUNGSTEN SULFIDE (2 MOLS) H0111 1 2 3 4 5 o 7 g 1 g; 10

Temp F 400 400 400 400 500 500 510 558 610 610 15 Percent Amine in Product 46 42 38 60 57. 5 57 58 75 CATALYSTNICKEL SUPPORTED ON HF TREATED MONTMORILLONITE CLAY Hz R0- Hour 1 2 3 auction 4 5 6 7 8 9 10 11 12 Temp., F 500 610 628 850 630 618 575 600 600 600 600 600 600 Percent Amine in Product 5 40 55 82 52 55 42 35 30 26 26 l Y./v./hr.=volumes nitro-toluene per volume of catalyst per hour feed rate.

was 95% at the twenty-seventh hour and 97% at the thirtieth hour. 1

The feed stock was then changed to a mixture of 20% nitro p-xylene and 80% xylidine, which was 100% converted to amine during four hours. Pure nitro p-xylene was then fed for two hours at 600 F. and at 625 F., respectively, which gave pure p-xylidine. Previous workhad shown that nitro-p-xylene in a bomb decomposed violently below these temperatures, but in the present experiments the time on heat was merely a matter of seconds and no difficulty was encountered. Of course, in large scale commercial operation such short heating times would not be practicable. The temperature was then reduced to 500 F. and nitro p-xylene fed for eleven hours at 100% conversion. Nitro m-xylene was then fed for twelve hours, and analysis of the product showed 87% conversion. The catalyst age at this point was 60 hours.

The catalyst was then activated by passing H28 through the reactor, and the conversion of nitro o-xylene to xylidine was 100% during the sixtyfirst hour. During the next five hour period the conversion was 85%. It is evident that whereas H25 is an extremely potent activator, its life is somewhere between one and five hours when applied as a pretreatment. The continuous feeding of H28 or of an organic sulfide makes the proce'ss continuous. Two percent by weight of carbon disulfide was then added to the feed stock and in seven hours the conversion had increased to 95%. This catalystwas subsequently employed for the hydrogenation of nitro ethy1 benzene at 550 F. using continuous CS2 activation, and for the hydrogenation of crude nitro xylenes containing dinitroxylenes, using periodic or continuous H28 activation. After 110 hours of operation the catalyst was still highly active. As regards continuous activation of the catalyst, it should be noted that an excess of HZS is harmful rather than helpful to activity. While the amount required may vary with catalyst feed rate, hydrogen rate, temperature and pressure, if the H28 partial pressure is to be maintained at its optimum value, it may be pointed out that at atmospheric pressure and 450 F. when employing tungsten sulfide catalyst, and 300% excess hydrogen, the amount of HzS should be less than 100 volumes per hour per volume of catalyst, since this amount is harmful, and considerably lesser amounts are suflicient.

- The amount of hydrogen employed with the sulfur activated catalysts should be considerably in excess of the theoretical requirement. It is generally preferred to employ at least 300% of the theoretical amount required when operating at about atmospheric pressure in the vapor phase. At higher pressures such as 125 lbs. per square inch gauge, the amount should be considerably higher, for example, 1000% of the theoretical amount or higher.

There was no evidence of ring hydrogenation during the reduction of the aromatic nitro compounds in the presence of the sulfided catalysts. Thus the refractive indices of the xylidines which I obtained in liquid and vapor phase were 1.5594 and 1.5582, respectively, while that of pure 2,4-xylidine was 1.559. The closeness of the products obtained as regards refractive indices to those given in the literature indicate that the ring was not attacked during the hydrogenation of the nitro group and also that neither the nitro nor the amine group was hydrogenated off the ring.

Whereas the sulfided catalysts mentioned in the above examples are unsupported, the sulfides may be incorporated with a relatively inert base. For example, the metallic constituent, such as molybdenum, may be impregnated as a water soluble salt on a base such as magnesia or zinc oxide, or a mixture thereof, or alumina, the salt decomposed to the oxide, and the oxide sulfided. About 10% of molybdenum is sufficient to give a highly active catalyst. Other bases which may be employed are clay, synthetic gels, such as silica and alumina and mixtures thereof, pumice and activated carbon. Suitable active constituents, besides those mentioned above, which may be employed in sulfide activated form, are nickel, iron, cobalt, cadmium, lead, copper, silver, and vanadium.

The catalysts may be employed either as pellets in a fixed bed, or as a finely divided material. In the latter case the nitro aromatic is hydrogenated in vapor phase. The nitro aromatic and hydrogen are bubbled upward through the finely divided catalyst at a velocity of from 0.5 to 4 ft. per second. Under these conditions the catalyst is violently agitated and the temperature is remarkably uniform throughout. Heat can be removed from the catalyst by cooling coils situated in the reactor, or by withdrawing catalyst from the reactor as an aerated fluidized stream, passing it thru a heat exchanger and returning it to the reactor. If desired, the catalyst may be activated by treating with the sulfur-containing promoter during this circulation or the promoter may be added to the feed stock. A suitable particle size range for the catalyst is from 200 to 400 mesh. It is generally preferable to provide from 5 to 10 ft. of settling height above the catalyst in the reactor, and any catalyst which is carried out with the product gases is collected in dust separators and returned to the reactor. An important feature of the finely divided catalyst system is that the nitro aromatic feed stock can be injected as a cool liquid onto the hot catalyst, thereby becoming vaporized and simultaneously cooling the catalyst. Similarly, other liquids, for example water or oil, can be injected for cooling purposes.

Referring in detail to the drawing, the nitro feed stock, for example nitro aromatics or nitro paraffins, is introduced to the system through line it and is passed through lines 10 and [2 into the bottom portion of reactor l4, preferably below distribution grid i6 which distributes the feed evenly over the Whole cross sectional area. Hydrogen for the reduction reaction is introduced by means of line l8 and is passed through lines 20 and 12 into the bottom portion of reactor 14.

Reactor 14 contains catalyst in a finely divided state; for example, the bulk of the catalyst may be in the particle size range from 200 to 400 mesh. The catalyst is maintained as a churning, ebullient state by the hydrogen and the feed stock which pass upward through the reactor at a net velocity of from about to 3 or 4 ft, per second. The violent agitation of the catalyst by the vapors causes the temperature to be substantially uniform throughout, despite the highly exothermic nature of the reaction. Product vapors are removed overhead through line 22 after passing through dust separator 24 which may be of any convenient type such as a cyclone or a filter. Separated catalyst is returned to the reaction zone proper by means of pipe 26.

Heat of reaction is removed by cooling the catalyst in heat exchanger 28. Since the catalyst is aerated and fluidized by the reactant vapors, it flows readily through draw-off pipe 30 to heat exchanger 23 and thence through control valve 32 and line 3 into line 12. However, in some cases it may be necessary to inject additional gas, for example hydrogen, in order to maintain aeration and fluidity of the catalyst in line 30 and also in line 26 leading from the dust separator, as through taps 3|.

The catalyst in line I2 is at a lower density than that in line 30 and heat exchanger 28 because of the hydrogen introduced through line 25. For example, the density in line l2 may be from to 20 lbs. per cubic foot, while that in line 30 may be 30 to 50 lbs. per cubic foot. This difference in density causes a natural circulation of the fiuid ized catalyst from the reactor, through the heat exchanger, and back into the reactor through lines 30, 34 and I2. The rate of circulation of course depends on the amount of gas introduced on the low density side as Well as on the setting of valve 32, and may be controlled to give the optimum operating temperature in reactor it.

It is generally desirable to allow a settling space of from 5 to feet in height above the average catalyst level in reactor l4. By this means, and by controlling the upward linear gas velocity through the catalyst, the carryover of catalyst with the product vapors may be kept very low. For example, when employing a catalyst of 200- 400 mesh, which has a free settled density of about 40 lbs. per cubic foot, the catalyst carryover will be only a few thousandths of a pound per cubic foot of vapors when the upward velocity is one or two feet per second.

The sulfide catalyst activator is introduced through line 35 and passed through lines and I2 to reactor 14. It alternately may be introduced conveniently as a solution in the feed stock.

The feed stock introduced through line It may be either in the vapor or the liquid state. The latter is preferable, since the heat of vaporization absorbs some of the heat of reaction and the cooling requirement in heat exchanger 28 is somewhat less. It will be understood that the liquid feed is vaporized on coming in contact with the hot catalyst passing through line l2. Furthermore, if the feed stock contains not only monobut dinitro compounds, as is generally the case unless the nitration product has been subjected to an expensive vacuum or steam distillation, it is highly desirable to inject the feed onto the hot catalyst as liquid. The dinitro compounds are highly explosive when hot and concentrated. For example, dinitro xylene undergoes an. explosive decomposition at about 500 F. Mixtures of monoand dinitro compounds should therefore be hydrogenated either in complete liquid phase with stationary catalysts, or with fluidized finely divided catalyst at a temperature above the vaporization temperature of the mono nitro compounds and below the explo sion temperature of the dinitro compound. Sulficient catalyst should be employed to absorb the liquid dinitro compound completely. The diamino, product which is absorbed by the catalyst can be removed therefrom by extraction with a solvent, after withdrawal through line 31.

Product vapors removed overhead from the reactor through line 22 are cooled in heat exchanger 38 to condense water and amine, which are then separated from the hydrogen in separator 40. The hydrogen stream, which generally contains hydrogen. sulfide catalyst activator, is

recycled to the reactor through lines, 42, 20 and I2. Water drawn off from the lower end of separator 40 through line 42 may be heated to liberate hydrogen sulfide dissolved therein, and the latter recycled through a line (not shown) to maintain catalyst activity. The amine product is withdrawn from separator 40 through line 44, and a portion thereof may be recycled to dilute the feed stock by passing it through line 46 to line [0. The amine product withdrawn from the system may also be subjected to a flash distillation to recover hydrogen sulfide catalyst activator for recycling.

Instead of using the precise designv which I have shown and described, I may use a reactor containing a stationary bed of catalyst or the catalyst may be in the form of pills and pellets rather than powder and may move countercurrent through the reaction zone by gravity against the upflowing reactants.

In continuous operation with either stationary or moving catalyst, it is preferred to recycle hydrogen sulfide separated from the products to the inlet of the reactor. employed and after separation of the liquid product the excess is recycled to the reactor along with the hydrogen sulfide promoter, make-up hydrogen and make-up sulfide sufiicient to counter balance losses being added. It will beunderstood; that carbon disulfide or other organic sulfide promoter is largely converted to H28 in the reaction zone, and that the latter is recycled.

To recapitulate briefly, my present invention relates to improvements in the art of producing aromatic amines by hydrogenation of mino-nitro aromatics, such as nitrobenzene, nitrotoluene,

nitroxylene, and the like, to the corresponding amine, and the process may be operated continuously either in vapor or liquid phase to produce high quality amines in goods yields through the use of sulfide catalysts and of a sulfide activator such as carbon disulfide or hydrogen sulfide in order to maintain the catalyst activity at a high level.

What I claim is:

1. In a continuous method for producing a primary aromatic amine from the corresponding mono-nitro-mono-cyclic aromatic compound by hydrogenation in the presence of an active hydrogenation catalyst consisting essentially of a sulfide of a metal selected from the group consisting of molybdenum and tungsten at reaction temperatures between 400-600 F. in a reaction zone, the improvement which comprises injecting a cool liquid stream of the nitro aromatic compound onto the catalyst to cool the catalyst and vaporize the nitro aromatic compound as it enters the reaction zone, simultaneously feeding excess hydrogen gas into the reaction zone, passing a resulting mixture of the hydrogen and vaporized nitro aromatic compound through, a bed of the catalyst in the reaction zone, and withdrawing the resulting aromatic amine product from the reaction zone.

2. In the continuous method described in claim 1, admixing a stream of the amine product as liquid condensate with said stream of cool liquid nitro aromatic compound to dilute the nitro aromatic compound as it is injected onto the catalyst.

3. In the continuous method described in claim 1, recycling a stream of excess hydrogen gas withdrawn with the product from the reaction. zone,

and admixing said stream of cool liquid nitrov An excess of hydrogen isaromatic compound with the recycled stream of' hydrogen gas as it enters the reaction zone.

4. In a continuous method for producing a primary aromatic amine from the corresponding mono-nitro-mono-cyclic aromatic compound by hydrogenation in the presence of a finely divided hydrogenation catalyst consisting essentially of a sulfide of metal selected from the group consisting of molybdenum and tungsten activated by an active sulfur carrier in a reaction zone at reaction temperatures between 400-600 F., the improvement which comprises injecting a cool liquid stream of the nitro aromatic compound into a stream of hydrogen gas as it enters the reaction zone, directly contacting the liquid aromatic hydrocarbon injected into the stream of hydrogen with a portion of the catalyst to cool the catalyst and vaporize the nitro aromatic compound as it enters the reaction zone, passing the resulting mixture of the hydrogen and vaporized nitro-aromatic compound into contact with a bed of the catalyst in the reaction zone, and Withdrawing a gaseous product containing the amine from the reaction zone.

5. In the process as described in claim 4, maintaining the catayst in said reaction zone in a fluidized state, removing a portion of the catalyst, cooling the removed catalyst and returning the catalyst suspended in the stream of hydrogen gas entering the reaction zone.

JOHN C. MUNDAY.

10 REFERENCES CITED The following references are of record in the file of this patent:

UNITED STATES PATENTS Number Name Date 1,124,776 Marwedel Jan. 12, 1915 1,984,380 Odell Dec. 18, 1934 2,039,259 Pier et a1. Apr. 28, 1936 2,063,623 Pier et a1. Dec. 8, 1936 2,123,623 Brown July 12, 1938 2,127,382 Pier et a1. Aug. 16, 1938 2,198,249 Henke Apr. 23, 1940 2,432,087 Brown Dec. 9, 1947 FOREIGN PATENTS Number Country Date 462,006 France Jan. 17, 1914 101,254 Japan May 23, 1933 474,191 Great Britain Oct. 27, 1937 OTHER REFERENCES Chem. Abstrn, vol. 35 (1941), p. 1770. Groggins: Aniline and its Derivatives (1924),

Grifiitts et al.: J. Phys. Chem. 41 (1937), pp. 477-484.

Brown et al.: J. Phys. Chem. 43 (1939), pp. 383-386.

Claims (1)

1. IN A CONTINUOUS METHOD FOR PRODUCING A PRIMARY AROMATIC AMINE FROM THE CORRESPONDING MONO-NITRO-MONO-CYCLIC AROMATIC COMPOUND BY HYDROGENATION IN THE PRESENCE OF AN ACTIVE HYDROGENATION CATALYST CONSISTING ESSENTIALLY OF A SULFIDE OF A METAL SELECTED FROM THE GROUP CONSISTING OF MOLYBDENUM AND TUNGSTEN AT A REACTION TEMPERATURES BETWEEN 400*-600* F. IN A REACTION ZONE, THE IMPROVEMENT WHICH COMPRISES INJECTING A COOL LIQUID STEAM OF THE NITRO AROMATIC COMPOUND ONTO THE CATALYST TO COOL THE CATALYST

Images