Method for reactivating a precious metal iron catalyst and performing a chemical reaction

Abstract

Catalytic activity of a spent precious metal-iron catalyst is restored by combining the spent catalyst with an iron (III) compound. This can be performed by adding the iron (III) compound into a chemical reaction that contains the spent precious metal-iron catalyst. It is unnecessary to add more of the precious metal. The process is especially useful in a continuous process for converting a nitro compound such as nitrobenzene to the corresponding amine.

Claims

1. A process for producing an aromatic amine, comprising a) performing a continuous reduction reaction by continuously or intermittently introducing a starting nitroaromatic compound and hydrogen to a reaction vessel in the presence of a precious metal-iron catalyst supported on a carbonaceous support wherein the precious metal is palladium, platinum or a mixture of palladium and platinum and continuously or intermittently removing from the reaction vessel water and an aromatic amine produced by the reaction of the starting nitroaromatic compound and hydrogen, the reduction reaction being continued for a period of time such that the precious metal-iron catalyst becomes at least partially spent; b) thereafter adding an iron (III) compound one or more times to the reaction vessel while adding no more than 10% by weight of the precious metal, based on the weight of the iron in the added iron (III) compound wherein the iron (III) compound added is not supported and physically independent from the no more than 10% by weight of the precious metal, and thereafter continuing to perform the continuous reduction reaction in the presence of the at least partially spent precious metal-iron catalyst and the iron (III) compound.

2. The process of claim 1 wherein the iron (III) compound is an iron (III) halide, iron oxide, basic iron (III) carbonate, iron (III) carbonate, or iron (III) hydroxide.

3. The process of claim 1 wherein the carbonaceous support comprises activated carbon having a porosity of at least 3000 m.sup.2/g as measured by gas adsorption methods.

4. The process of claim 1 wherein the precious metal-iron catalyst is a carbon-supported palladium-iron or palladium-platinum-iron catalyst containing, when fresh, 3 to 7.5% by weight palladium and/or platinum and a like amount of iron, in each case based on the total dry weight of the fresh supported catalyst.

5. The process of claim 1 wherein: the precious metal-iron catalyst is a palladium iron catalyst; the iron (III) compound is ferric chloride or ferric hydroxide; and/or the iron (III) compound is added in an amount sufficient to provide 1 to 4 parts by weight of the spent palladium iron catalyst.

Description

EXAMPLES 1-4 AND COMPARATIVE SAMPLES A AND B

(1) Reaction time experiments are performed in the following general manner: A 300 mL autoclave (Autoclave Engineers Model ABA-300, steam-jacketed and equipped with an agitator) is charged with 0.027 to 0.028 gram of a palladium-iron catalyst as described below, 17.2 mL of nitrobenzene, 30 mL methanol and 50 mL water. The reactor is sealed, purged with hydrogen, and then pressurized with nitrogen to 30 bars (3040 kPa) gauge pressure. The reactor contents are heated to 100 C. by flowing steam into the jacket. The time at which the reactor contents reach 100 C. is designated time T.sub.0. The pressure in the reactor is measured continuously. The point in time at which the pressure becomes constant (T.sub.c, indicating the completion of the reaction) is determined. The reaction time is calculated as T.sub.cT.sub.0.

(2) For Comparative Sample A, the palladium-iron catalyst is a fresh catalyst sample containing about 4.6% by weight palladium, 0.4% by weight platinum and 5.15% by weight iron (all by induction-coupled plasma-mass spectroscopy (ICP-MS) on a dry weight basis). The metals are carried on carbonaceous support particles. The palladium-iron catalyst is made according to a method such as is described in U.S. Pat. No. 2,823,235.

(3) For Comparative Sample B, the palladium-iron catalyst is a spent sample of the same palladium-iron catalyst, obtained from a commercial aniline production plant.

(4) The reaction times for Comparative Samples A and B are 7.1 to 7.2 minutes and 19.3-19.7 minutes. The spent catalyst therefore has approximately one-third the activity of the fresh catalyst on this test.

(5) ICP-MS analysis of the spent catalyst reveals that the palladium and platinum levels in the spent catalyst are virtually unchanged from those of the fresh catalyst. However, the iron content of the spent catalyst is found to be reduced by about 60%, to about 2.1% by weight of the catalyst.

(6) Comparative Sample B is repeated 4 additional times, in each case adding ferric chloride (FeCl.sub.3.Math.6H.sub.2O) as a physical admixture with the spent catalyst. In Example 1 enough of the ferric chloride is added to provide 1 part iron per 100 parts by weight of the spent catalyst. In Examples 2-4, ferric chloride is added to provide, 2, 3 and 4 parts iron per 100 parts of the spent catalyst. The approximate total amount of iron (including that from the spent catalyst and ferric chloride) and reaction time are as indicated in Table 1, together with the results from Comparative Samples A and B.

(7) TABLE-US-00001 TABLE 1 Approximate Reaction Iron content Time, Designation Catalyst Description %.sup.1 minutes A* Fresh Catalyst 5.15 7.1-7.2 B* Spent Catalyst 2.1 19.3-19.7 1 Spent Catalyst + Ferric 3.1 16.0-16.4 Chloride 2 Spent Catalyst + Ferric 4.1 12.8-12.9 Chloride 3 Spent Catalyst + Ferric 5.1 7.6-7.9 Chloride 4 Spent Catalyst + Ferric 6.1 6.8-7.1 Chloride *Not an example of the invention. .sup.1By weight of the palladium-iron catalyst (spent or fresh).

(8) As the data in Table 1 shows, adding ferric chloride into the reaction reduces the reaction time very significantly compared to the spent catalyst. When the iron content in the reaction mixture is increased to approximately that of the fresh catalyst, the reaction time becomes essentially the same if not faster than that provided by the fresh catalyst. Smaller amounts of the added iron (III) compound provide a lesser but significant benefit.

EXAMPLE 5 AND COMPARATIVE SAMPLE C

(9) Although greatly improved reaction rates are obtained by adding ferric chloride into the nitrobenzene reduction reaction described in the earlier examples, the presence of chloride ion causes unwanted by-products to form in that particular reaction. Therefore, an iron (III) compound that contains few if any halide ions is preferred for reducing nitro compounds to the corresponding amine.

(10) Ferric chloride and activated carbon are combined in the presence of water at a weight ratio of iron to carbon of 1:20. Sodium bicarbonate is added, which reacts with ferric chloride to produce basic ferric carbonate, which is precipitated onto the activated carbon, and sodium chloride. The iron carbonate-activated carbon particles are separated from the liquid phase and heated to 80 C. for 30 minutes to decompose the iron carbonate to ferric hydroxide. The particles are then washed, filtered and dried.

(11) Comparative Sample C is performed according to the general reaction time procedure described above. The catalyst is a spent palladium-iron catalyst from a commercial aniline production facility, originally made according to a method as described in U.S. Pat. No. 2,823,235.

(12) In Example 5, a mixture of the same spent catalyst and the ferric hydroxide-activated carbon particles is used. The mixture is formed by adding the spent catalyst and ferric hydroxide-activated carbon particles separately to the reactor. Enough of the ferric hydroxide-activated carbon particles are added to provide about 4 wt-% iron, based on the weight of the spent catalyst particles. The total iron content in the spent catalyst and ferric hydroxide-activated carbon particles combined is 5-6.25% of the weight of the spent catalyst particles. Results are as indicated in Table 2, together with those of Comparative Sample A.

(13) TABLE-US-00002 TABLE 2 Approximate Reaction Iron content, Time, Designation Catalyst %.sup.1 minutes A* Fresh Catalyst 5.15 7.1-7.2 C* Spent Catalyst 2-2.2 19.8-21.7 5 Spent Catalyst + Iron 5-6.25 7.1-7.5 Hydroxide-Activated Carbon Particles *Not an example of the invention. .sup.1By weight of the palladium-iron catalyst (spent or fresh).

(14) As the data in Table 2 demonstrates, adding iron in the form of iron hydroxide supported on activated carbon restores the catalytic activity to approximately that of the fresh catalyst. No unwanted by-products of the reaction form.