PROCESS FOR PREPARING PRIMARY AMINES USING AN UNSUPPORTED COBALT CATALYST

Abstract

The invention relates to a process for preparing primary amines, which comprises hydrogenating at least one nitrile in an apparatus (V1) in the presence of an unsupported cobalt catalyst to obtain at least one primary amine, with recurrent or continuous addition of at least one compound (I) to the apparatus (V1), said compound (I) comprising at least one component selected from alkali metal, alkaline earth metal and rare earth metal.

Claims

1: A process for preparing a primary amine, the process comprising: hydrogenating at least one nitrile in an apparatus (V1) in the presence of an unsupported cobalt catalyst to obtain at least one primary amine, with recurrent or continuous addition of at least one compound (I) to the apparatus (V1), wherein said compound (I) comprises at least one component selected from the group consisting of an alkali metal, an alkaline earth metal, and a rare earth metal.

2: The process according to claim 1, wherein said compound (I) is an oxide, a hydroxide, a salt, or an aqueous solution.

3: The process according to claim 2, wherein aid compound (I) is an aqueous solution of an oxide, a hydroxide or a salt of lithium, potassium, cesium, magnesium, calcium, lanthanum, or cerium.

4: The process according to claim 1, wherein i) the nitrile comprises 0.01% to 10% by weight of water, or ii) the compound (I) is added to the apparatus (V1) as soon as a proportion of a secondary amine formed as a by-product is greater than 0.5 GC area %, or iii) added to the apparatus (V1) via the addition of the compound (I) is 0.01 to 500 ppm of an alkali metal, an alkaline earth metal, or a rare earth metal, based on g-atoms of nitrile in the apparatus (V1), as soon as the proportion of the secondary amine formed as the by-product in the apparatus (V1) is greater than 0.5 GC area %, at a catalyst space velocity of 0.01 to 10 kg of reactant per L of catalyst and hour.

5: The process according to claim 1, wherein the hydrogenating is conducted in the absence of ammonia.

6: The process according to claim 1, wherein the apparatus (V1) is a reactor, and a superficial velocity being 5 to 50 kg of mass flow per m.sup.2 of cross-sectional reactor area and second.

7: The process according to claim 1, wherein the hydrogenating is conducted at a pressure of 1 to 300 bar.

8: The process according to claim 1, wherein the unsupported cobalt catalyst is used in the apparatus (V1) as a fixed bed catalyst.

9: The process according to claim 1, wherein the nitrile is an aliphatic mono-, di- or trinitrile, a cycloaliphatic mono- or dinitrile, an alpha-, beta- or omega-aminonitrile, or an alkoxynitrile.

10: The process according to claim 1, wherein the at least one primary amine is N,N-dimethylaminopropylamine (DMAPA) and the nitrile is N,N-dimethylaminopropionitrile (DMAPN), or the at least one primary amine is isophoronediamine and the nitrile is isophoronenitrileimine, or the at least one primary amine is hexamethylenediamine (HMD) or 6-aminocapronitrile (6-ACN) and HMD, and the nitrile is adiponitrile (ADN).

11: The process according to claim 1, wherein the unsupported cobalt catalyst consists of catalytically active composition to an extent of at least 70% by weight.

12: The process according to claim 1, wherein the unsupported cobalt catalyst comprises, as catalytically active composition: i) cobalt (Co), ii) optionally at least one element selected from the group consisting of iron (Fe), nickel (Ni), ruthenium (Ru), rhodium (Rh), palladium (Pd), osmium (Os), iridium (Ir), and platinum (Pt), or iii) optionally at least one promoter selected from the group consisting of chromium (Cr), manganese (Mn), molybdenum (Mo), titanium (Ti), zinc (Zn), tin (Sn), an alkali metal, an alkaline earth metal, a rare earth metal, and phosphorus (P).

13: The process according to claim 12, wherein the catalytically active composition of the unsupported cobalt catalyst consists of 55% to 98% by weight of cobalt, 0.2% to 15% by weight of phosphorus, 0.2% to 15% by weight of manganese, and 0.2% to 15% by weight of an alkali metal, calculated as CoO, H.sub.3PO.sub.4, MnO.sub.2, and an alkali metal oxide prior to reduction with hydrogen, respectively.

14: The process according to claim 1, wherein the unsupported cobalt catalyst i) has a surface area of not more than 50 m.sup.2/g, or ii) a proportion of support material is not more than 10% by weight, based on a total mass of the catalyst.

15: The process according to claim 1, wherein a stream (S1) which is discharged from the apparatus (V1) comprises the primary amine and is returned at least partly to the apparatus (V1).

16: The process according to claim 2, wherein said compound (I) is an aqueous solution of lithium hydroxide.

17: The process according to claim 1, wherein the unsupported cobalt catalyst consists of catalytically active composition to an extent of at least 80% by weight.

18: The process according to claim 13, wherein the unsupported cobalt catalyst is prepared by calcining the catalytically active composition first at a final temperature of 550 to 750° C. and then at a final temperature of 800 to 1000° C.

19: The process according to claim 14, wherein the unsupported cobalt catalyst i) has a surface area of not more than 40 m.sup.2/g, or ii) a proportion of the support material is not more than 5% by weight, based on the total mass of the catalyst.

20: The process according to claim 15, wherein the compound (1) is added at first to the returned portion of the stream (S1) and then fed into the apparatus (V1) as a constituent of the stream (S1).

21: The process according to claim 15, wherein the compound (I) is added as an aqueous solution to the returned portion of the stream (S1).

Description

EXAMPLE 1

[0202] Hydrogenation of Dimethylaminopropionitrile (DMAPN) to Dimethylaminopropylamine (DMAPA) in the Presence of an Unsupported Cobalt Catalyst with Recurrent Supply of Aqueous Lithium Hydroxide Solution

a) Catalyst Preparation and Activation

[0203] The catalyst used is an unsupported cobalt catalyst composed of 90.4% by weight of cobalt, 5.1% by weight of manganese, 0.3% by weight of sodium and 3.1% by weight of phosphorus (strand diameter 2 mm), which is produced according to example 1 of EP-A 636 409. 40.2 g of this catalyst are introduced into a hydrogenation reactor and heated up to 280 C with supply of 25 L (STP) of hydrogen per hour within 12 hours, kept at this temperature while feeding in 25 L (STP) of hydrogen per hour for 12 hours and cooled under hydrogen.

b) Hydrogenation of DMAPN

[0204] The hydrogenation is conducted in trickle mode with the above-described catalyst as fixed bed catalyst in a stainless steel tubular reactor (1.4571 inner tube) having a height of one meter and a diameter of 0.6 cm. The reaction output from the hydrogenation is cooled, decompressed, partly discharged and partly recycled into the reactor.

[0205] The feedstock used for the hydrogenation is technical grade DMAPN comprising dimethylamine and water. According to GC analysis, the water content is 0.3 to 3.6 area %, the dimethylamine content 1.4 to 2.5 area % (GC column: 60 m CP Volamine/WCOT fused silica 0.32; temperature program: 50° C.-10 min-15° C./min.-240° C.-30 min).

[0206] The hydrogenation is conducted for a total of 4000 hours, for the first 149 hours without addition of aqueous LiOH solution, then for 3800 hours with intermittent addition of LiOH solution. Table 1 compiles the addition times and amounts of LiOH; table 2 shows the lithium concentration in the reactor output.

TABLE-US-00001 TABLE 1 Metered addition of lithium hydroxide Space ppm ppm LiOH Start End Duration Amount DMAPN velocity.sup.4 [LiOH/ [Li/ absolute [h] [h] [h].sup.1 Concentration.sup.2 [g/h].sup.3 [g/h] [kg/Lh] DMAPN].sup.5 DMAPN].sup.6 [g].sup.7 149 320 171 1.00% 0.0665 19.4 1.00 34.29 9.94 0.1138 794 915 121 1.00% 0.0665 19.4 1.00 34.29 9.94 0.0803 1155 1270 115 1.00% 0.0665 19.4 1.00 34.29 9.94 0.0762 1511 1845 334 0.10% 0.3326 19.4 1.00 17.14 4.97 0.1110 1845 1968 123 0.20% 0.3326 19.4 1.00 34.29 9.94 0.0820 1973 2159 186 0.50% 0.3326 19.4 1.00 85.71 24.85 0.3097 2159 2181 22 1.00% 0.3326 19.4 1.00 171.43 49.70 0.0726 2304 2379 75 1.00% 0.3326 19.4 1.00 171.43 49.70 0.2494 2449 2541 93 1.00% 0.3326 19.4 1.00 171.43 49.70 0.3079 2541 2668 127 1.00% 0.3326 19.4 0.95 171.43 49.70 0.4208 2668 3116 448 1.00% 0.3497 20.4 1.00 171.43 49.70 1.5678 3311 3621 311 1.00% 0.3497 20.4 1.00 171.43 49.70 1.0863 .sup.1Duration of addition of the aqueous lithium hydroxide solution .sup.2Lithium hydroxide concentration of the aqueous lithium hydroxide solution .sup.3Amount of the aqueous lithium hydroxide solution .sup.4Kilograms of DMAPN/liter of catalyst and hour .sup.5Ratio of lithium hydroxide to DMAPN (g/g) .sup.6Ratio of lithium to DMAPN (g/g) .sup.7Absolute amount of lithium hydroxide which has been run into the reactors within the particular addition interval.

[0207] The hydrogenation is started by pumping 200 mL of crude DMAPA into the reactor.

[0208] The hydrogenation temperature is 90 to 110° C.; the pressure is 85 bar for the first 700 hours, then 50 bar for the rest of the run time.

[0209] The catalyst space velocity over the entire run time is 1 to 1.1 kg DMAPN per liter of catalyst and hour. The superficial velocity is 41 to 42 kg per m.sup.2 and hour. In order to attain these values, a portion of the hydrogenation output is returned to the reactor after decompression.

[0210] Between 2544 and 3500 hours of run time, DMAPN comprising 5% by weight of water is employed.

[0211] Over the first 149 hours of hydrogenation time without addition of aqueous LiOH solution, the hydrogenation output contains only 87% to 90% DMAPA, 11% to 13% bis-DMAPA and 0.2% to 0.3% DMAPN (GC area %); see also comparative example 2, operation without ammonium.

[0212] After 149 hours, the first metered addition of aqueous LiOH solution is commenced (metering periods specified in table 1). After addition of aqueous LiOH solution, the DMAPN conversion from 200 hours after the start of the experiment onward is at least 99.4% over the remaining 3500 hours. The DMAPA yields are 99.1% to 99.5%. The only by-product that occurs is bis-DMAPA in amounts of 0.2% to 0.9%.

[0213] As example 1 shows, rising or high bis-DMAPA values during operation without metered addition of LiOH can be lowered again to a lower level by addition of LiOH. In this regard, FIGS. 1 to 3 show the concentrations of reactant (DMAPN), lithium hydroxide, product (DMAPA) and by-product (bis-DMAPA) measured in the reaction output over the entire experimental duration of example 1.

[0214] Example 1 additionally shows that the positive effects with small amounts of LiOH are achievable, that it is possible to work with high catalyst space velocities combined with high catalyst service lives, and that water in the hydrogenation mixture is not disruptive. Moreover, high DMAPN conversions and DMAPA yields are achieved, and it is possible to work in the absence of ammonia.

TABLE-US-00002 TABLE 2 Determination of lithium concentration in the output Run time Lithium content of [h] output [ppm] 172 14.0 246 7.0 270 17.0 294 10.0 316 12.0 342 2.8 414 1.4 438 1.6 486 1.1 510 0.9 580 1.1 654 0.8 772 0.7 172 14.0 825 22.0 846 17.0 963 2.4 988 2.2 1007 3.0 1031 2.4 1127 1.6 1201 20.0 1609 16.0 1775 2.0 1825 1.8 1871 1.8 1967 8.9 2039 1.8 2117 2.8 2188 74.0 2303 3.0 2351 5.0 2447 2.7 2543 10.0 2615 9.0 2669 11.0 2711 11.0 2813 9.0 3005 10.0 3052 9.0 3143 14.0 3215 8.0 3293 4.0 3337 6.0 3509 11.0 3629 11.0

COMPARATIVE EXAMPLE 2

[0215] Hydrogenation of DMAPN to DMAPA in the Presence of a Cobalt Catalyst with Initial Single Impregnation of the Catalyst with Aqueous LiOH Solution

[0216] 40.2 g of the catalyst charge used in example 1 are activated as described and the catalyst in the plant is flushed through with about 400 g of a 10% aqueous LiOH solution in straight pass from the bottom upward (1 mL/minute). About 7 hours after commencement of the feed of LiOH solution, in addition, the liquid circuit is switched on with a recycle rate of 1000 g/L for 2 hours, then the recycling and the metered addition of LiOH are switched off again. Subsequently, the desired reaction conditions are established (90° C., hydrogen) and the plant is started up with DMAPA. A liquid recycle rate of 814 g/h and a DMAPN feed rate of 19.4 g/h are established. The catalyst thus pretreated, as described in example 1, is used for the hydrogenation of DMAPN at 90° C. and 85 bar. The lithium concentration in the reaction output is determined regularly. After 500 hours of reaction time, the hydrogenation output consists to an extent of about 92% of DMAPA, about 8% of bis-DMAPA and less than about 0.5% of DMAPN (GC area %). Analogously to FIGS. 1 to 3 for example 1, the corresponding values for comparative example 2 are illustrated in FIG. 4.

[0217] Comparative example 2 shows that, in the case of unsupported cobalt catalysts in a fixed bed, by contrast with Raney cobalt (catalyst in suspension mode; see, for example, example 50 from EP-A 913 388), a pretreatment of the catalyst with LiOH does not bring about any increase in selectivity. The regular determination of lithium in the reaction output shows that the lithium is washed continuously out of the plant and off the catalyst. This effect is visible in comparative example 2, especially in about the first 200 hours after commencement of the hydrogenation. The reduction in selectivity with regard to DMAPA and the increase in bis-DMAPA correlate with the amount of lithium washed out of the reactor. Up to the end of the experiment after about 650 hours, the selectivity for DMAPA continues to decrease continuously and lithium is still detectable in the reaction output.

COMPARATIVE EXAMPLE 3

[0218] Hydrogenation of DMAPN to DMAPA in the Presence of an Unsupported Cobalt Catalyst and of Ammonia and in the Absence of Ammonia (without Lithium Hydroxide Addition)

[0219] 40.2 g of the catalyst charge used in example 1 are activated as described. Analogously to example 1, the plant is first started up with 200 mL of crude DMAPA. A temperature of 90° C. and a pressure of 85 bar are established and maintained over the duration of the experiment. 16 g/h of ammonia and 25 L (STP)/h of hydrogen are metered in continuously; the hydrogenation is started by feeding in 19.4 g/h of DMAPN.

[0220] The catalyst space velocity over the entire run time is 1 to 1.1 kg of DMAPN per liter of catalyst and hour. The superficial velocity is 34 to 42 kg per m.sup.2 and hour. In order to achieve these values, a portion of the hydrogenation output is returned to the reactor after decompression. After a reaction time of 301 hours, the hydrogenation output consists to an extent of 97.8% of DMAPA and less than about 0.3% of DMAPN; bis-DMAPA is not present (GC area %).

[0221] After 312 hours, under otherwise unchanged experimental conditions, the ammonia feed is stopped. The selectivity for DMAPA achieved is much lower in the absence of ammonia. After a reaction time of 442 hours, the hydrogenation output consists to an extent of 85.9% of DMAPA, less than about 0.3% of DMAPN, and 11.1% of bis-DMAPA (GC area %).

[0222] Comparative example 3 shows that large amounts of bis-DMAPA are formed over the unsupported cobalt catalyst studied in the absence of ammonia at 85 bar, and the selectivity for DMAPA is low. In the presence of ammonia, because of the vapor pressure of ammonia, a distinct drop in the reaction pressure is not possible. The corresponding concentration values according to comparative example 3 are illustrated analogously to the previous examples in FIG. 5.