CATALYST MODIFICATION WITH ALKALI METAL, ALKALINE EARTH METAL OR RARE EARTH METAL IONS IN THE CONTINUOUS LIQUID-PHASE HYDROGENATION OF NITRO COMPOUNDS

Abstract

The present invention relates to a process for continuous hydrogenation of a nitro compound to the corresponding amine in a liquid reaction mixture comprising the nitro compound in the presence of a supported catalyst which comprises as the active component at least one element from groups 7 to 12 of the periodic table of the elements, wherein the hydrogenation is performed in the presence of at least one salt selected from the group consisting of the salts of the alkali metals, alkaline earth metals and of the rare earth metals and to a supported catalyst for continuous hydrogenation of a nitro compound to the corresponding amine in a liquid reaction mixture comprising the nitro compound which comprises as the active component at least one element from groups 7 to 12 of the periodic table of the elements and one salt of the alkali metals, alkaline earth metals or of the rare earth metals.

Claims

1-20. (canceled)

21. A process for continuous hydrogenation of dinitrotoluene to toluenediamine in a liquid reaction mixture, the process comprising hydrogenating dinitrotoluene in the presence of a supported catalyst that comprises an active component comprising a mixture of nickel and platinum in an atomic ratio of from 30:70 to 70:30 and optionally one or more additional metals, wherein the hydrogenating is performed in the presence of at least one salt selected from the group consisting of a salt of an alkali metal, a salt of an alkaline earth metal and a salt of a rare earth metal.

22. The process of claim 21, wherein the active component of the supported catalyst comprises chromium.

23. The process of claim 21, wherein the active component of the supported catalyst comprises nickel in the form of nickel crystallites having a bimodal nickel crystallite size distribution and has a nickel content of 60 to 80 wt % based on a total mass of the supported catalyst and a degree of reduction of at least 70%.

24. The process of claim 21, wherein the supported catalyst comprises 1 to 5 wt % of platinum, 0.3 to 1.5 wt % of nickel, 0.05 to 1.5 wt % of the at least one additional metal and 94.65 to 97.45 wt % of a support material, based on a total weight of the supported catalyst, wherein the sum amounts to 100%.

25. The process of claim 21, wherein the at least one salt is comprised by the supported catalyst.

26. The process of claim 24, wherein the at least one additional metal is at least one metal selected from the group consisting of copper, cobalt, iron, zinc, manganese and chromium.

27. The process of claim 25, wherein the at least one salt comprised by the supported catalyst is present in a total concentration of 0.05 to 20 wt %, based on a dry weight of the supported catalyst.

28. The process of claim 21, wherein the at least one salt is comprised by the liquid reaction mixture.

29. The process of claim 28, wherein the at least one salt comprised by the liquid reaction mixture is present in a total concentration of 0.01 to 1 mol %, based on a supplied amount of dinitrotoluene for hydrogenation.

30. The process of claim 21, wherein the at least one salt comprises potassium, strontium, sodium or a mixture thereof.

31. The process of claim 21, wherein the at least one salt is a carbonate, a hydrogencarbonate, a hydroxide, an oxide, a nitrate, a carboxylate or a mixture thereof.

32. The process of claim 21, wherein the liquid reaction mixture further comprises at least one high boiler selected from the group consisting of a dinitrocresol, a trinitrocresol and a nitrophenol.

33. The process of claim 21, wherein the liquid reaction mixture comprises no high boilers selected from the group consisting of a dinitrocresol, a trinitrocresol and a nitrophenol.

34. The process of claim 21, wherein the liquid reaction mixture further comprises: at least one compound from the group consisting of nitric acid, sulfuric acid, a nitrogen oxide, dinitrogen monoxide, hydrocyanic acid, carbon monoxide and nitrobenzoic acid; and/or a degradation product of the least one compound.

35. The process of claim 21, wherein the hydrogenating is performed at a temperature of 80 C. to 250 C.

36. The process of claim 21, wherein the hydrogenating is performed in the absence of a solvent.

Description

DESCRIPTION OF THE FIGURES

[0132] In FIG. 1 to FIG. 4 and FIG. 7 the x-axis of the diagram shows the running time of the reaction in hours, during which DNT is metered in. In FIG. 5 to FIG. 6 the x-axis of the diagram shows the running time of the experiment in hours since the first DNT metered addition (including interrup-tions to the reaction). The left-hand y-axis shows the TDA yield in area % based on the total area of all peaks in the gas chromatography (GC). In FIGS. 1, 2, 5, 6 and 7 the right-hand y-axis shows the yield of low- or high-boiling components in area % and the cometering flow in centili-ters/hour. In FIGS. 3 and 4 the right-hand y-axis shows the calculated cation concentration (ideal backmixing and solubility/continuous proportional discharge) in mmol/kg. The cation concentrations are calculated for complete backmixing and continuous discharge.

[0133] FIG. 1

[0134] shows the TDA, low boiler and high boiler yields with metered addition of an aqueous solution comprising 4000 ppm of potassium bicarbonate (resulting in a concentration of 552 ppm of potassium bicarbonate).

[0135] In the curves, the data points are as follows: .diamond-solid.: yield of TDA, : yield of low-boiling components, : yield of high-boiling components and : cometering flow of KHCO.sub.3.

[0136] The vertical dashed line marks an interruption in the DNT supply for a period of almost 70 h during which the circulation stream and all other conditions are maintained.

[0137] FIG. 2

[0138] shows the TDA, low boiler and high boiler yields with metered addition of aqueous solutions of potassium bicarbonate, ammonium bicarbonate and potassium hydroxide.

[0139] In the curves, the data points are as follows: .diamond-solid.: yield of TDA, : yield of low-boiling components, : yield of high-boiling components and : cometering stream. The cometering stream consists successively of 4000 ppm of potassium bicarbonate (1), 3150 ppm of ammonium bicarbonate (2) and 2250 ppm of potassium hydroxide (3) (all aqueous solutions).

[0140] FIG. 3

[0141] shows the TDA yield with metered addition of aqueous solutions of various salts.

[0142] The data points .diamond-solid. denote the yield of TDA. The dashed curve (left-hand side) shows the calculated potassium concentration based on potassium bicarbonate (end concentration 552 weight ppm (wppm)), the solid curve (center) shows the calculated ammonium concentration based on ammonium bicarbonate (end concentration 434 wppm) and the dotted line (right-hand side) shows the calculated potassium concentration based on potassium hydroxide (end concentration 310 wppm) in the reactor stream.

[0143] FIG. 4

[0144] shows the TDA yield with pulsed metered addition of aqueous solutions of various salts.

[0145] Data sets marked with a .diamond-solid. show the yield of TDA in area % based on the total area of all peaks in the GC. Data values marked with a show the yield of TDA in area % based on the total area of all peaks as determined in an integrated online GC.

[0146] The cation concentration, i.e. potassium concentration or sodium concentration, curves result from (from left to right) a start concentration of 121 wppm of potassium bicarbonate (dashed, short dashes), a start concentration of 363 wppm of potassium bicarbonate (dashed, long dashes), a start concentration of 1089 wppm of potassium bicarbonate (solid, black) and a start concentration of 642 wppm of sodium bicarbonate (solid, gray) in the reactor stream.

[0147] The cation concentrations are calculated for complete backmixing and continuous discharge.

[0148] FIG. 5

[0149] shows the TDA, low boiler and high boiler yields with interruption of DNT metering in the absence of potassium bicarbonate

[0150] Data points characterized as follows: .diamond-solid. represent the yield of TDA, show the yield of low-boiling components and represent the yield of high-boiling components.

[0151] The vertical dashed lines mark an interruption in the DNT supply for a period of 90 h during which the circulation stream and all other conditions are maintained.

[0152] FIG. 6

[0153] shows the TDA, low boiler and high boilier yields with interruption of DNT metering in the presence of potassium bicarbonate (metered addition before interruption).

[0154] In the curves, the data points are as follows: .diamond-solid.: yield of TDA, : yield of low-boiling components, : yield of high-boiling components and : cometering stream. The cometering stream consists successively of H.sub.2O (4) and 4000 ppm of aqueous potassium bicarbonate solution (1). The vertical dashed lines mark an interruption in the DNT supply for a period of almost 70 h during which the circulation stream and all other conditions are maintained.

[0155] FIG. 7

[0156] shows the TDA yields, low boiler yields and high boiler yields with metered addition of aqueous solutions of potassium bicarbonate and strontium nitrate.

[0157] In the curves, the data points are as follows: .diamond-solid.: yield of TDA, : yield of low-boiling components, : yield of high-boiling components and : cometering stream. The cometering stream consists successively of 4000 ppm of potassium bicarbonate (1) (end concentration 5.5 mmol(K)/kg), 1600 ppm of strontium nitrate (5) (end concentration 1.0 mmol(Sr)/kg) and 10 000 ppm of strontium nitrate (6) (end concentration 6.5 mmol(Sr)/kg) (all aqueous solutions).

[0158] FIG. 8

[0159] shows the TDA yield with metered addition of increasing potassium concentrations.

[0160] The x-axis of the diagram shows the potassium concentration in the reactor in mmol/kg. The y-axis shows the TDA yield in area % based on the total area of all peaks in the GC.

[0161] The data points .diamond-solid. denote the yield of TDA. Said points form the basis for the trend curve in the form of a 3rd order polynomial which was determined by the least-squares method (dotted line).

[0162] FIG. 9

[0163] shows the TDA yield with metered addition of increasing potassium concentrations.

[0164] The x-axis of the diagram shows the potassium concentration in the reactor in mmol/kg. The y-axis shows the TDA yield in area % based on the total area of all peaks in the GC.

[0165] The data points .diamond-solid. denote the yield of TDA.

EXAMPLES

[0166] The following experiments are performed in a laboratory reactor setup. This consists of a loop reactor setup which is predominantly configured as a tubular reactor but also has a stirred tank integrated in it. In the tubular reactor part of the loop reactor dinitrotoluene (DNT) and hydrogen are injected/metered into the circulation stream consisting of product and suspended catalyst (3% Pt-1% Ni/C catalyst) and mixed therewith. Installed a short distance downstream of this feed point is a second feed point through which cometerings may be supplied to the circulation stream. The hydrogen metering is effected under pressure control so that sufficient hydrogen is always available at constant pressure. Excess product leaves the loop reactor setup through a catalyst-retaining stainless steel frit installed in the stirred tank. The filtered product is then sep-arated into a gas phase and a liquid phase in a phase separator and the gas flow set to 10 NI/h to ensure a continuous gas discharge and prevent accumulation of inert gases. In daytime operation the liquid phase is regularly withdrawn and analyzed by gas chromatography (GC with an RTX-5 amine column; injection temperature 260 C.).

Example 1

Cometering Test

[0167] The laboratory reactor setup is charged with 3 g (dry weight) of catalyst suspended in water and operated at 185 C., 27 bar, with a circulation stream of 36 kg/h and a DNT metering rate of 100 g/h. This results in a WHSV of 33.3 kg(DNT)/kg(cat)/h. As is apparent from FIG. 1, the TDA yield gradually improves by approximately 1% during the metered addition of KHCO.sub.3 solution to the reaction.

[0168] This effect gradually decreases immediately after termination of the metered addition because the salt is flushed from the reactor system with the product. The dashed line marks an interruption in the DNT supply for a period of almost 70 h during which the circulation stream and all other conditions are maintained.

Example 2

Cometering Test

[0169] The laboratory reactor setup is operated as previously except that it is charged with 5 g (dry weight) of catalyst so that the WHSV is 20 kg(DNT)/kg(cat)/h.

[0170] After a running in period of 250 h and a pronounced catalyst aging potassium bicarbonate is added and the TDA yield gradually increases by 2% (see FIG. 2). After shutdown of the metering the TDA yield gradually decreases again. Subsequent addition of ammonium bicarbonate has no effect on the TDA yield but the addition of potassium hydroxide that follows results in a grad-ual TDA yield increase. The increase in the TDA yield can therefore be attributed to the presence of potassium cations. Since the reaction is performed in the product and this is already strongly alkaline in nature (>60% TDA in water) it is possible to exclude an effect of the potassium on the pH as the cause.

[0171] FIG. 3 shows the calculated concentrations of the cations together with the TDA yield. As a result of the continuous addition and the backmixing of the system the salt concentrations in the circulation stream slowly and gradually increase and equally gradually decrease after termination of the metered addition. The reflection of this increase and decrease in the TDA yield is readily apparent.

[0172] Saturation concentrations of the salts in mmol/kg are calculated based on the concentration of pure DNT (5.49 mol/kg), i.e. 5.5 mmol/kg correspond to 0.1 wt % based on the reactant and also approximately describe the weight fraction in the product (neglecting H.sub.2 uptake).

Example 3

Cometering Test

[0173] The following experiment is performed in a Miniplant test plant. This consists of a loop reactor set up which in one part (5.6 L) has an internal circulation flow powered by a motive jet (circulation stream consisting of product and catalyst) and in another part is configured as a tubular reactor (4.4 L). The overall set up is thermostated with thermal oil to remove generated heat. DNT is mixed in close to the motive jet and hydrogen is metered into the gas space above the internal circulation flow under pressure control. Product formed is withdrawn through a catalyst-retaining membrane so that the liquid level in the reactor part with the internal circulation flow remains constant. This discharge is regularly analyzed by GC. A fixed amount of gas is discharged above the gas space to prevent unlimited accumulation of gaseous products or impurities.

[0174] The reactor is charged with 112 g (dry weight) of 3% Pt-1% Ni/C catalyst suspended in water and operated at 185 C., 25 bar overpressure, with a circulation stream of 500 kg/h and a DNT metering rate of 2 kg/h. This results in a WHSV of 17.9 kg(DNT)/kg(cat)/h.

[0175] As is apparent from FIG. 4 the TDA yield is improved by up to 0.7% by pulsed metering of potassium bicarbonate in the highest concentration. This effect gradually decreases rapidly because the salt is continuously discharged with the product. The pulsed addition is effected by addition of a volume of 0.1/0.2 L of aqueous salt solution into the reactor. To this end, after shutoff of the DNT metered addition the liquid level in the reactor is lowered and the volume introduced into the reactor circuit with overpressure by opening a valve. The DNT metered addition is subsequently resumed and the liquid level is again set to the old value.

[0176] The concentrations are calculated based on the total reactor volume. Concentrations of the salts in mmol/kg are calculated based on the concentration of pure DNT (5.49 mol/kg), i.e. 27.5 mmol/kg correspond to 0.5 wt % based on the reactant and also approximately describe the weight fraction in the product (neglecting H.sub.2 uptake). Required steady-state concentrations are in the range of examples 1 and 2.

Example 4

[0177] Potassium cometering effect during interruption of DNT addition

(=hot standby/extremely lengthy residence time of up to several days)

[0178] Interruption of the DNT addition while maintaining the other reaction conditions normally results in a severe reduction in the TDA yield through formation of high boilers. This state corresponds to an extension of the residence time and the catalyst is therefore unutilized in terms of the hydrogenation of nitro groups after reaction of the remaining DNT.

[0179] FIG. 5 shows the determined TDA yields. Over the duration of the interruption in the DNT addition the TDA content of the organic product (i.e. the TDA yield) is reduced by 13% and gradually increases after resumption of the DNT addition not to the original TDA yield value but rather to a value that is 1% lower.

[0180] FIG. 6 shows the TDA yield before and after an interruption in the DNT addition in the presence of potassium bicarbonate (metered addition before interruption). A discernible high boiler formation is not observed in this case and the TDA yield is likewise stable. The falling TDA yield after resumption of the DNT metered addition is attributable to the flushing-out of the remaining potassium bicarbonate.

Example 5

[0181] The cometering test from example 5 is performed analogously to the cometering test from example 2. The cometering stream consists successively of 4000 ppm of potassium bicarbonate (1) (end concentration 5.5 mmol(K)/kg), 1600 ppm of strontium nitrate (5) (end concentration 1.0 mmol(Sr)/kg) and 10 000 ppm of strontium nitrate (6) (end concentration 6.5 mmol(Sr)/kg) (all aqueous solutions).

Example 6

[0182] The following experiment is performed in a Miniplant test plant. This consists of a loop reactor set up which in one part (5.6 L) has an internal circulation flow powered by a motive jet (circulation stream consisting of product and catalyst) and in another part is configured as a tubular reactor (4.4 L). The overall set up is thermostated with thermal oil to remove generated heat. DNT is mixed in close to the motive jet and hydrogen is metered into the gas space above the internal circulation flow under pressure control. Product formed is withdrawn through a catalyst-retaining membrane so that the liquid level in the reactor part with the internal circulation flow remains constant. A fixed amount of gas is discharged above the gas space to prevent unlimited accumulation of gaseous products or impurities.

[0183] The reactor is charged with 112 g (dry weight) of 3% Pt-1% Ni/C catalyst suspended in water and operated at 185 C., 25 bar overpressure, with a circulation stream of 500 kg/h and a DNT metering rate of 4 kg/h. This results in a WHSV of 35.8 kg(DNT)/kg(cat)/h.

[0184] The addition of potassium carbonate is effected by supplying a continuous volume flow of a 1% aqueous salt solution into the reactor. The introduction is effected through a tube fed through at the reactor top. The solution drips into the internal circulation flow of the reactor at the edge of the push-in tube, i.e. at a point removed from the DNT feed point. Various potassium concentrations are established by variation of the volume flow of the salt solution. Once a steady-state has been established a plurality of samples are taken and analyzed by GC.

[0185] FIG. 8 plots the resulting TDA yields against the resulting steady-state potassium concentration. It is readily apparent that there is an optimum addition in terms of the improvement in the TDA yield which in this case is at about 10 mmol(K)/kg. This corresponds to 0.19 mol % of potassium based on the supplied DNT.

Example 7

[0186] The following experiment is performed analogously to example 6 except that the DNT metering rate is 2 kg/h (WHSV of 17.9 kg(DNT)/kg(cat)/h) and a 4% aqueous potassium carbonate solution is employed.

[0187] Analogously to FIG. 8, FIG. 9 plots the resulting TDA yields against the resulting steady-state potassium concentration. Once an optimum addition of the salt solution in terms of the improvement in the TDA yield has been achieved it is readily apparent that a further increase in the potassium concentration results in a deterioration of the TDA yield. The optimum addition in terms of the improvement in the TDA yield for this mode of operation is about 22 mmol(K)/kg. This corresponds to 0.36 mol % of potassium based on the supplied DNT. A further increase to about 28 mmol(K)/kg (corresponds to about 0.5 mol % of potassium based on the supplied DNT) already results in a marked deterioration in selectivity of about 0.4% in the TDA yield.