NO138284B - PROCEDURE FOR THE PREPARATION OF SECONDARY ALIFATIC AMINES - Google Patents

Description

The object of the invention is a process for the production of secondary aliphatic amines, whose aliphatic chains which are saturated or unsaturated contain 7 to 24 carbon atoms by reacting the corresponding primary alcohols in the liquid phase with ammonia at normal pressure and elevated temperature

rature,! presence of hydrogenation-dehydration catalysts, the method being characterized in that the alcohol is reacted with ammonia in the presence of hydrogen at temperatures from 120 to

250°C, preferably 160 to 210°C. and pressure from 0.5 to 6 ata, preferably 0.9 to 1.5 ata, the gas velocity being 50 to approx. 1200 liters per hour per kg of alcohol, preferably

50 to 500 liters per hour per kg of alcohol and especially 150 more

300 liters per hour per kg of alcohol and the volume ratio between

ammonia and hydrogen averaged over the total reac-

sion time amounts from 80 : 20 to 20 : 80 and the water formed is removed with the gas stream.

It is known to transfer alcohols by means of aminolysis with ammonia alone or with ammonia and hydrogen in amines. According to the older known methods, one worked with dehydration catalysts such as aluminum oxide in gas phase, continuously and preferably at normal pressure. The new methods use hydrogenation-dehydration catalysts, such as nickel and cobalt catalysts, and work in the liquid phase of normal pressure up to 300 ato (see Houben-VJeyl, "Methoden der organischen Chemie'', 1957, volume 11/1, Stickstoffverbindungen II, Amine, table 15, page 118 and table 18, page 130, further DOS 1,950,604).

It appears from the known publications that the pressureless aminolysis of the higher alcohols usually generally leads to mixtures of primary, secondary and tertiary amines. The process can also be carried out in such a way that tertiary amines are preferentially formed. The aminolysis of the higher alcohols to form especially secondary amines at low pressure or on the other hand, normal or slight depression is not known.

According to US patent no. 2,078,922, example 3, octadecyl alcohol when reacted with ammonia on aluminum oxide in gas phase at atmospheric pressure and 360°C at only 45% conversion gives a mixture of mono-, di- and trioctadecylamine. From example 1 in US patent no. 2,953,601 it appears that the pressureless aminolysis in the gas phase already with isooctyl alcohol no longer proceeds in an orderly manner. The turnover is only 10%, alongside hydrocarbons and aldehyde are formed. Also in Ullmann's "Enzyklopadie der technischen Chemie", 3rd edition, 1953,

volume 3, page 454, it says with regard to the exchange of aliphatic hydroxy groups with amino groups: "At higher alcohols, the gas phase process becomes uneconomical due to side reactions (olefin formation, dehydrations, resin formations)".

Por the large-scale production of higher molecular weight secondary aliphatic amines, the gas phase method thus stands out. The pressureless aminolysis of higher alcohols with ammonia in the presence of hydrogenation-dehydration catalysts gives, according to US Patent No. 3,223,734, Example-20, with lauryl alcohol on Raney nickel at 165°C a mixture of 11% primary,. 22% secondary and 31% tertiary amines.

The turnover, referred to alcohol, thereby amounts to 70$. From this it appears-'

that with such a reaction method tertiary and not secondary amines are preferably formed.

The pressureless aminolysis of the higher alcohols could therefore only be used economically for tertiary amines. Thus, according to DOS No. 2,050,866, tertiary amines can be produced from primary alcohols with 5 to 20 C atoms in a two-stage process, the alcohol being aminated in the first stage with ammonia at normal pressure in the presence of hydration-dehydration catalysts. When using isononyl alcohol, you get, for example, a mixture of 10% residual alcohol, 19% diisononylamine and 62% triisononylamine. In another example, 14% residual alcohol, 0.5% primary, 25% secondary and 59% tertiary isononylamine appear. These mixtures are then transferred in another ammonia-free step by gas treatment with hydrogen in high-percentage triisononylamine. This method also shows that in any case tertiary and not secondary amines predominate. The same finding appears from the already mentioned US patent no. 2,953-601. According to this, alcohols with 5 to 13 C atoms are reacted with ammonia on hydrogenation-dehydration catalysts in the range of normal pressure up to 7 ato. In addition to the preferred tertiary trialkylamines, according to US patent no. 2,953-601, among other things, tertiary hydroxyamines, such as e.g. dioctyl-hydroxyoctylamines.

For the preparation of the higher secondary amines from the corresponding primary alcohols, there is no known economical implementation of the aminolysis in the range of normal pressure to 5 ato. Admittedly, some other aminolysis methods for secondary amines exist, which are partly very cumbersome, give unsatisfactory yields or an incomplete conversion. US patent no. 2,636,902 relates to the production of amines and alcohols with 2 to 12 carbon atoms at 140 to 230°C and 10 to 25 ato with ammonia and hydrogen on a - "forminate hydrogenating" catalyst from the metals in the periodic table system's Vlll group. For higher amines as referred to in this patent, in a single example, isononyl alcohol (3,5,5-trimethylhexanol) is reacted at a pressure of 17 ato and 200°C with ammonia. The turnover referred to used trimethy1-hexanol was 10%. The amount of secondary amine in the amine mixture formed was likewise at 70%, referred to isononyl alcohol, thus at 49%. US Patent No. 3-022. 349 relates to the two-stage aminolysis of primary alcohols with 2 to 10 carbon atoms in gas phase with ammonia or hydrogen. In the first reaction step, e.g. 2-ethylhexanol on a reduced copper catalyst with ammonia at 260°C to approx. "31% in the corresponding secondary amine and up to 44% in the nitrile. In another converter, hydrogenation with hydrogen takes place at 150 to 250°C on a nickel catalyst to an amine mixture, which, in addition to primary and tertiary amine, contains 68 % di-(2-ethylhexyl)-amine and 6% residual nitrile. According to US Patent No. 3-080,424, primary as well as secondary amines with 6 to 16 C atoms are obtained when aluminum alkoxides of the corresponding alcohols are reacted with at least 1 mole of ammonia per mole of alcohol at 200 to 350° C. in the presence of a solvent, such as gasoline. The simultaneous presence of aluminum chloride increases the overall yield and allows predominantly secondary amines to form.

DOS Mr. 1,950,604 relates to a process for producing amines from alcohols with ammonia or amines in the presence of hydrogen and a cobalt-containing fixed storage catalyst. In the patent claim, there is certainly no indication of the reaction conditions such as pressure and temperature, only in the description is a pressure of 10 ato mentioned. According to all eight examples, however, the reaction takes place at 300 ato. Examples 6 and 7 discuss, among other things, the hydrating aminolysis of stearyl alcohol at 300 ato and a performance of 150 g/liter/h and 215°C, which leads to a mixture of 69% primary amine, 3% intermediate and 27% secondary amine, tertiary amine not mentioned. According to example '7, this amine mixture is hydrogenated. aminating in another high-pressure step while adding the same amount of stearyl alcohol under the same conditions. The space-time yield thereby falls back to 75 g/litre/h. The resulting anhydrous amine contains 19% stearylamine, 4% intermediates and 76% distearylamine, tertiary amine is again not mentioned.

DOS no. 1,439-781 relates in detail to the amine reaction of secondary alcohols with ammonia under atmospheric pressure on nickel- .

cobalt and copper chromite catalysts. From the comparison experiments between octanol-1 and octanol-2, due to the low degree of conversion of only 16% for octanol-1 and 72% for octanol-2, the conclusion is drawn that "ammonia under the process conditions according to the invention does not react effectively with primary alcohols". The aminolysis procedure specified in the aforementioned DOS therefore only concerns 'secondary alcohols- -

In contrast to what is stated in DOS no. 1-9.50-604, according to which in order to obtain high secondary amine amounts it is necessary to operate the aminolysis in a two-stage process at 300'ato .and in opposition to what is stated in DOS no., 1.439.78l, that primary alcohols at normal pressure can only very poorly be converted to amines with ammonia on hydrogenation-dehydration catalysts, it was now found that higher secondary amines can be obtained in a very economical way from the corresponding primary alcohols at normal pressure resp. slight negative or positive pressure in high yield, when carrying out the aminolysis in the range from 120 to 250°C with ammonia in the presence of hydrogen on hydrogenation-dehydration catalysts and observing a critical minimum gas velocity.

As is known, the aminolysis of alcohols with ammonia on hydrogenation-dehydration catalysts is initiated by means of a dehydration

and it is assumed that the dehydration hydrogen is again immediately consumed by a subsequent hydration process. It was now found that under the conditions, according to the present invention, the dehydrogenation hydrogen is utilized only in the closed system for conversion to secondary amines. In the open system it is again lost.

The starting materials required for carrying out the method according to the invention are linear and/or branched, saturated and/or unsaturated primary alcohols with 7 to 24 C atoms,

such as, for example, octanol-1, 2-ethylhexanol, isooctyl alcohol, isononyl alcohol, lauryl alcohol, isotridecyl alcohol, oleyl alcohol, stearyl alcohol, further mixtures of these alcohols, especially those that arise from the hydration of natural fatty acids or their esters,

like for example. tallow fatty alcohol and palm kernel fatty alcohols. Also very suitable are the cheap ethylene-building alcohols from the Ziegler process, which give the saturated primary alcohols with up to 24 carbon atoms.. Also more or less branched alcohols, as they appear in the. various oxo processes from linear and branched olefins and from isononylaldehyde available isooctade.cyl alcohols are suitable starting products for the preparation according to the invention of secondary amines.

Preferred, technically readily available primary alcohols thus contain 7, 7-9, 9-11, 12-13,. 13, 12-15 or 16-19 C atoms, so that the compounds with different C numbers predominate, and on the other hand 8, 10, 8-10, 12-14, 16, 14-18, 16/18, 14-20, 18-22 or 20-24 C atoms^. as it constantly or predominantly contains connections with a; equal to C number. Of course, the individual "individual" alcohols with 7 to 24 C atoms can also be used individually or, if desired, mixtures. In any case, the aliphatic chains can be straight or one or more branched and saturated or one or more times, preferably once unsaturated.

Dehydration-hydrogenation catalysts include nickel catalysts in the form of the active Raney types as well as in grain or powder form with or without carrier material. Also suitable are the corresponding cobalt catalysts and catalysts of nickel and cobalt, possibly mixed with copper, as well as very complicated catalysts, e.g. of copper and chromium with additions of alkalis, barium and other metals. Nickel and cobalt catalysts of various types with or without additions of other metals, carriers and activators are suitable for the method according to the invention.

A preferred embodiment of the method according to the invention consists in passing through a flask filled with alcohol and catalyst - e.g. stearyl alcohol and Raney nickel - at 180 to 200°C with good stirring leads ammonia and hydrogen. This form is referred to as the so-called "open" mode of reaction. When the aminolysis is carried out in this way, due to the partial loss of dehydrogenation hydrogen in the interest of a higher yield of secondary amine, it is unavoidable with the ammonia to also introduce hydrogen into the system. Thereby, it is not so crucial that the hydrogen is added all at once when the ammonia is initially supplied. It can also be allowed during the course of the reaction, however hydrogen supply should begin when approx. 1/4 of the total reaction time, which amounts to approx. 3 to hours.' After approx. After 2 hours, the actual amination is finished and it is then advantageous to complete the hydration alone to feed in hydrogen. In the simplest way, the "open" reaction method of the aminolysis can be carried out when one "from the beginning to the end of the reaction works with a constant ammonia-hydrogen ratio. The most favorable... turns out to be a volume ratio between ammonia and hydrogen of 50 .: 50.' It is also possible to work with variable volume ratios NH^ : ^• Ratios of 80 : 20 to 20 : 80' integrated over the -total reaction time is

.advantageous, as the reaction can also be started with pure ammonia

and the amount of hydrogen corresponding to the conditions can be added at specific later times. '

In order to ensure the completion of the reaction, it is necessary that ammonia and hydrogen are present each time in "sufficient quantities. 50 liters of gas/h and kg of alcohol is therefore the lower limit. Upward, the amount of gas is limited mainly by the process technique and the costs. Thus, gas velocities of more than 100 litres/h and kg of alcohol, for example. 1200 litres/h.kg all the way through, possible. However, the speed should not be so great that the partial pressure of any alcohol is lowered so strongly that the temperature required for the reaction is no longer reached and/ or the alcohol is carried over into the publisher. In most cases, for lower alcohols with approx. 8 C atoms as well as for higher alcohols with, for example, 24 C atoms at a reaction time of 2 to 4 hours, it has proven particularly advantageous with gas quantities from 150 to 300 litres/kg of alcohol.

Decisive for the aminolysis of secondary amines is that the minimum gas quantity is not exceeded, as there is then the risk that the tertiary quantity increases undesirably. If the amount of gas drops too much, especially towards the end of the reaction, the end point with optimal secondary amine content is also very difficult to determine. IN

in these cases, the formed secondary amine reacts relatively quickly with itself, namely to tertiary amines and non-amine by-products such as hydrocarbons, nitriles and other cleavage products. When already at the start of the reaction there is too little ammonia or ammonia in mixture with hydrogen, the already formed secondary amines can also react with the alcohol present and -formed primary amines to undesired tertiary amines.

The case of too little ammonia supply clearly exists in the method according to the already mentioned US patent no. 2,953-601. Although here likewise higher alcohols,

namely iso-Cg-alcohol and iso-C-^-alcohol depressurized with ammonia and Raney nickel in the liquid phase at.165 to 190°C are subjected to amino-. lyse, tertiary amines are preferably obtained. Unfortunately, this method also works without additional hydrogen.' One is therefore only referred to dehydration hydrogen, which - as already stated above - is greatly lost in the ■ "open" reaction method mentioned there. According to the invention, the preferred formation of sec-

these amines are therefore excluded according to this known method.

When comparing example 3 according to US patent no. 2,953-601 with example 1 in the method according to the invention, the following amine composition appears for isooctylamine:

In addition to the small formation of secondary amines, according to US patent no. 2,953-601, the hydrogen and ammonia deficiency is also seen in the fact that, in addition to tri-isooctylamine, condensation of the aldehydes also produces tertiary amines with hydroxy groups, if Om- groups undergo the subordinate aminolysis thus

as the original alcohol used. In contrast, it could be established that with a timely and sufficient supply of ammonia and hydrogen, no hydroxyamines are formed.

The great importance of hydrogen could be demonstrated in two comparative experiments: In the first experiment, stearyl alcohol at .180 to 200°C in the presence of 2% Raney nickel was subjected to aminolysis with 200 liters of ammonia/h and kg of alcohol. Subsequently, only 12% secondary distearylamine and 35% neutral moieties, which were predominantly nitriles, occurred.

If, in another experiment, ammonia and hydrogen were used, no nitriles appeared, next to 3.5% of other neutral parts, the yield of secondary distearylamine was 88%.

These experiments show that the process according to US patent no. 2,953,601, which relates to the pressureless aminolysis of alcohols with 5 to 13 C atoms, is also unsuitable for the production of secondary amines above C-^-j of the corresponding primary alcohols.

For the method according to the invention, which in high yield leads to secondary amines, both relatively short-chain, for example C-, alcohols and long-chain, such as

- for example C2ij alcohols. However, it is appropriate to react the short-chain alcohols with 7 to 11 C atoms with e.g. other catalysts and under different reaction conditions than the long-chain ones, when you want to achieve the same high amounts of secondary amines. The. short-chain alcohols are more reactive and are therefore suitably treated with more inactive hydrogenation-dehydration catalysts than the slower, long-chain alcohols with 16 to 24 C atoms. , As a rule of thumb, one can say that the product of the reactivity of any alcohol and the activity of the catalyst should be approximately

constant. For stearyl alcohol, a highly active Raney nickel suspension is suitable, for example, and for octanol a more inactive nickel powder catalyst. The water-covered Raney-nickel catalysts can be used in the original moist state without any disadvantage for the course of the reaction.

The catalyst requirement per single mixture usually amounts to 2 to 3% by weight, referred to alcohol. At the relatively short-chain ones. C 1 to C 2 -alcohols, which are preferably treated with more inactive catalysts, it has proved advantageous to use approx. 5% catalyst. By reusing 90 to 95% of the used catalyst and adding some fresh catalyst each time with each new charge, in permanent operation a ■catalyst consumption according to

■ alcohol quality and catalyst, from 0.1 to 0.3%, referred to the weight of the alcohol.

The optimum reaction temperature for the aminolysis depends on the activity of the catalyst and the reactivity of the alcohol, which in turn is given by chain length and constitution.' It has been shown

■advantageous to coordinate both participants so that the reaction takes place at 190 to 200°C. With the C^- to' C-^ alcohols, this temperature cannot. set from the beginning, because the boiling point of these relatively low-boiling alcohols in the presence of optimally necessary ammonia and hydrogen demand drops to 140 to 150°C

It is therefore appropriate in the beginning to apply proportionately

small amounts of gas, e.g. at 120° to start with 50 liters of gas per kg Cy-alcohol per hour' and to wait with the increase until the boiling point increases due to the constant formation of the secondary amines and their precursors. You can also start with the optimal amount of gas

at temperatures around 120°C immediately and then slowly increase the temperature. According to these reactions. want. with a Cy-alcohol, the desired reaction temperature of 190 to 200°C is reached approximately half an hour to an hour later than with high-boiling alcohols. The reaction temperature, which is above the boiling point, can also be set by working at a somewhat elevated pressure to begin with, e.g. at approx. 0.5 to 1 ato, In the case of stearyl alcohol, on the other hand - the optimum temperature of 190 to 200°C can be set immediately and without special precautions.

A measure of the completion of the actual amination reaction is the separated amount of water, which is collected in a reservoir as ammonia water. Especially with the lower alcohols, easily volatile organic parts such as still unreacted alcohol and primary amines are also excreted with the reaction water. By fitting a column, the transition of the organic products can be reduced. After their separation from water, they are continuously or discontinuously returned to the reaction vessel. For the method according to the invention, the established amount of water is identical to the theoretically possible practical one, because on the one hand the alcohol is converted completely and on the other hand the high ammonia-water supply does not cause secondary condensations to hydroxyamines, as in relation to the theoretical would cause a water shortage.

Care must be taken to remove the formed water of reaction from the reaction material as best as possible. It is extracted with the help of the hot gases, ammonia and hydrogen as water vapour, directly or over a descending cooler. The gas mixture of ammonia and hydrogen may partly also contain inert gas such as Np and/or CH^ or also unsaturated water vapour.

After approx. 90% of the reaction water is excreted

the temperature is lowered to approx. 175 to 180°C. During this phase, as could be determined, predominantly hydrogenation of nitrogen-containing condensation products such as Schiff bases, into secondary amines takes place. For the final phase of the conversion, it is therefore advantageous to increase the hydrogen part or to use pure hydrogen.

The reaction time for a complete turnover lasts approximately 2 to 4 hours. The alcohols from C 1 to C 2 , where, as mentioned above, the reaction is suitably started slowly to achieve the required reaction temperature, require the longest reaction times.

The aminolysis of the higher alcohols according to the invention is normally carried out at normal pressure. This applies both to the "open" reaction method and also to the "closed" circuit gas reaction method which will be explained later. At the beginning of the reaction, however, it may be beneficial to set a certain negative pressure in the apparatus, 0.5 ata (approx. 380 mm Hg) should not, however, be exceeded, because otherwise the partial pressures will drop. ammonia and hydrogen fall too strongly. A too strong reduction in pressure could also allow the boiling point of the alcohol used to drop below the reaction temperature. The advantage of a small increase in pressure at the beginning of the aminolysis to raise the reaction temperature was already discussed above with the relatively low-boiling

-. Crj- to C-^-alcohols. For the end of the reaction, where the over-

mainly involves a catalytic hydration, brings the pressure increase. advantages with respect to time reduction. The final treatment takes place with pressure up to 6 ata. Whether H2 alone is used here or ■ H>> + NH^ depends-on- the degree of the aminolysis that has already taken place, i.e. replace-. ning of the alcoholic OH group with nitrogen. An increase in pressure towards the end of the reaction is then justified when starting from unsaturated alcohols and will obtain highly saturated secondary amines.

A "preferred embodiment" of the method according to the invention is the so-called "circular gas" or "closed" reaction method, where, in contrast to the "open" reaction method, ammonia and hydrogen are cycled and thus practically tolerated are not lost.

With equally high yields of secondary amines, one therefore comes out with significantly smaller quantities of the two gases.

In addition, it was found that with the circuit gas method, one even comes out without the addition of hydrogen. In contrast, to the open method, here dehydrogen that originates from the dehydration of the reaction process is sufficient for the production

.according to the invention of the secondary amines. The fact, not to

be referred to foreign hydrogen, is of great economic benefit.

The following table shows the advantage of "closed"

■working method, compared to "open" working method. Both experiments were carried out with stearyl alcohol as a model substance in the presence of 2% Raney nickel at normal pressure and 200°C and 200 liters of gas/hour and

kg of stearyl alcohol and a reaction time of 4 hours with pure ammonia, i.e. without addition of hydrogen.

The carbonyl number is a measure of the Schiff base still present, calculated as % CO. 1.5% CO corresponds to approx. 15%. masked primary amine and 30% secondary amine..

The results again show that the "open" reaction method with ammonia alone is unsuitable for the production of secondary amines. Since, however, in the "circuit gas" reaction mode, the partial pressure of the dehydration hydrogen constantly decreases with progressing reaction and thereby slows down the hydrogenation recognized as advantageous in the final phase of the reaction, it has proved beneficial in practice to add a certain amount of foreign hydrogen. This shortens the reaction time, increases the yield of secondary amines and reduces the formation of neutral parts. If, for example, the "closed" method of reaction is varied.e ■ according to the above experiment, then, after 2 to 3 hours of reaction time, the hydrogen partial pressure is increased by approx. 25% by adding foreign hydrogen to 80%, then instead of 84.5% you get secondary amines, 91.6%. The neutral part goes back from 6% to 2.4%. The overall turnover to amines increases from 94 to 97.6%. The carbonyl number as a measure of non-hydrated Schiff's base practically disappears completely at 0.25% - corresponding to 5% secondary distearylamine - to 0.01%.

The technical implementation of this finishing

with hydrogen can, for example, take place by more or less completely separating the ammonia from the decompressed NH^-^ mixture, for example by freezing or washing out, and replacing it with hydrogen.

The figure shows the apparatus implementation of the "circuit gas" reaction method. The apparatus consists, for example, of a 100 liter mixing vessel 1 and a circuit gas line 4 for circulating ammonia, hydrogen and possibly inert gases, a cooler 5, a separator 6 for water and

■lightly volatile organic components, as well as supplies for NH.-j 7,

H~ 8 and optionally N 2 9 (as flushing agent or possibly inert gas addition) and of an exhaust line 10. Heating and cooling of the vessel takes place over the mantle 11. After the cooler 6, a line 12 leads back to the vessel 1, over which the organic phase 14 separated from the separated aqueous phase 13 can be returned.

The method according to the invention is of course not limited to the mentioned device.

The figure also shows the form for the above-mentioned open reaction method. In this case, only the circuit gas fan is bypassed and the gas is removed via the exhaust line.

The working method of the circuit gas apparatus is described using an example for a long-chain alcohol. 50 kg of alcohol and catalyst are heated to approx. 120°C and the reaction is started with pure ammonia at a circulation of 10 to 15 m^/hour. Within 1 hour, the required temperature of 200°C is reached. The reaction water is withdrawn with the circuit gas, condensed in the cooler 5 and separated in the source 6. Any transient organic components are returned via connection line 12 to the vessel in the process. The necessary pressure in the apparatus is regulated by supplementary feeding with NH^. At the end of the 1-hour heating period, the ammonia part in the circuit gas still accounts for approx. 30%, 70% consists of hydrogen, which originates exclusively from dehydration of the alcohol. After a further 1 to 2 hours, the amount of hydrogen in the circuit gas has dropped to approx. 25°C due to the hydration that has taken place. The reaction temperature is then lowered to 180°C and in the last 30 to 60 minutes the amount of hydrogen in the circuit gas can be increased by corresponding supply of extra hydrogen. With the same result, the hydrogen can also already be added at the beginning or during the reaction.

After completion of aminolysis, it is filtered off from the catalyst.

About. 5 to .10% of this is discarded and replaced with a new catalyst.

The catalyst refreshed in this way goes back into the process.

With 25 "circuit gas" mixtures of 50 kg, the average catalyst consumption was 0.23%.

In addition to the open and closed reaction method, there are also mixtures of both methods for the aminolysis according to the invention. The mixed form of reaction exists in certain cases of circuit procedure, where one e.g. displaces the gas circulated in the apparatus towards the end of the turnover in whole or in part with hydrogen and at the same time exhausts the corresponding amount of gas, i.e. works "open". The mixed method is also available when, during the circuit gas process, a constant amount of gas is constantly withdrawn and replaced with fresh gas.

The change in the gas composition in a closed-circuit gas plant which is, for example, filled with NH-j' and H 2 can also take place by increasing the pressure or increasing the volume of gas exchange by dosing. course so that a reserve container filled with H2 of a similar size is included in the circuit gas system.

The mention of the open, closed and. mixed procedures took place under the assumption of batchwise implementation. It is of course also possible to carry out the process according to the invention completely or partially continuously, for example in a pipe system in several reactors connected one after the other or in cascade-like arranged vessels. Before they

necessary gases NH^ and by the cascade heating or in another

step reaction mode enters the next reactor, the formed resp. Entrained water is excreted. In connection with the final phase of the charge process, the last reactor can be operated under other conditions;

Instead of 200°C and a gas mixture of NH and H, for example, the last reaction chamber is kept at 180 oC and covered with pure hydrogen. The short reaction time for the final phase compared to the main reaction is thereby achieved. that the last reactor is correspondingly small.

In addition to suspendable and pumpable catalysts, solid grain or wire catalysts can also be used for the continuation. Of course, it is also possible with a corresponding "part continuum".

The statements that were made for the open, closed or mixed reaction method each time about gas composition, gas quantity, pressure-temperature and about other factors apply in the same way always for all three embodiments, furthermore also for continuous and other methods according to the invention.

With the method according to the invention, secondary amine contents from 80 to 95% can be obtained, depending on the alcohol, catalyst and process conditions used. The analytically detectable residual alcohol content is below 1%. The sum of the neutral parts is everything

depending on the alcohol used and process conditions at approx. 1 to 5%. The color quality of the products formed is very good. The iodine color number (determined according to the iodine color scale corresponding to DIN 6162) is obtained from

0.5 to 1. The secondary amines formed usually do not need to be purified by adsorption or distillation before their further processing.

The secondary amines produced by the primary alcohols from diheptylamine until di-C2^-amine is alone or in mixture, among other things desired intermediates for the production of quaternary ammonium compounds (with 2 longer-chain and 2 short-chain residues). These are again used, among other things, as softeners for textiles (laundry softener fabric softener), as components for organophilic ammonium bentonites and as moth repellants and as raw materials for cosmetic purposes, especially for hair treatment..

Example 1.

The experiment is carried out according to the so-called "open" reaction method in a 100 liter mixing vessel according to the arrangement in the figure. The circuit gas fan is bypassed in the present case and the excess gases NH^ and E^ are removed via the exhaust line.

The stirring vessel is filled with 50 kg of isooctyl alcohol (a mixture of branched Cg alcohols from the oxosynthesis, predominantly isomeric dimethylhexanols). While flushing the reaction chamber with nitrogen, 2.5 kg of Rariey nickel powder ("B 213" from Firma Degussa) is added. The vessel is then closed. Under good stirring with simultaneous nitrogen flushing, the heating then took place. At 120°C, it was then started with the supply of 50 liters of a mixture of ammonia and hydrogen in a volume ratio of 1:l/hour and kg of alcohol. Until the maximum reaction temperature of 2.00°C is reached, the gas supply is gradually increased to 250 litres/hour and kg. This condition is achieved after 3 hours.

At 200°C, the ammonia/hydrogen mixture is then carried out for a further 1 hour. The water separation had practically ended at this point. Then it was also treated for 1 hour with

250 liters of hydrogen/hour and kg of alcohol at 200°C. The hot reaction gases are cooled together with the water vapor formed after exiting the vessel in the cooler to 20 to 25°C, the water and the accompanying organic parts being condensed out in a downstream separator. The organic parts separated in the separator were continuously returned to the vessel. By the end of the reaction, 9.1 liters of ammonia water had been secreted. After the vessel contents had cooled to 100°C, it was filtered from the catalyst. The filtrate contained 42 kg of amine. The analysis yielded the following values:

Analogous results are obtained with a technical mixture of alcohols with 7-9 carbon atoms. Example 2.

In the same apparatus as in example 1, a technical oleyl alcohol was added. iodide salt 71 subjected to aminolysis. After weighing in 50 kg of oleyl alcohol and flushing the vessel with nitrogen, 1.5 kg of water-moist Raney nickel catalyst ("B 113" from Firma Degussa) was introduced. with approx. 70$ nickel content. Then it was. at 100°C started with the supply of 240 liters of a mixture of ammonia and hydrogen in volume. keep 50:'50/hour and kg alcohol. After 1 hour the reaction temperature had reached 200°C, in the same time 1.5 liters of ammonia water had formed in the separator. At 200°C, the aminolysis was continued with 240 liters of the same gas composition as before for another 2 hours. Thereby, a total of 4.8 liters of ammonia water was formed. It was then cooled and the catalyst filtered off at 80 to 100°C. The amino change amounted to 46 kg. The analysis gave:

Example 3.

The procedure was the same as in example 2, but to save hydrogen, in contrast to example 2, in this experiment

. instead of an ammonia/hydrogen ratio of 50:50, work with a gas mixture of ammonia, hydrogen and nitrogen in a volume ratio of 40:40:20. Thereby, nothing changed in the amine exchange.' Under otherwise equal conditions, the following result was obtained:

Example 4.

200 g of decanol were subjected to aminolysis in an "open" reaction mode in a glass apparatus with the addition of 10 g of stabilized nickel catalyst ("B 213" from Firma Degussa). The apparatus consisted of a 500 ml four-necked flask with stirrer, thermometer, introduction tube for ammonia and hydrogen as well as an outlet connection which was directly connected to a graduated water separator. This in turn carried a return-flow cooler with a diverter tube.

After flushing with nitrogen, the flask was heated and at 130°C, with stirring, dosing of each 20 liters of ammonia and hydrogen per hour. The heating up to approx. "170°C lasted 100 minutes. Until the reaction temperature of 200°C, another 1 hour was required. At this temperature, ammonia and hydrogen supply continued for a further 90 minutes. After this, the water separation was finished and the ammonia-hydrogen supply was stopped in favor -for a 1 hour treatment with 20 liters of hydrogen. It was then cooled to 80°C and filtered off from the catalyst. The yield was 158 g.

During the reaction with the water of reaction, it partly with- ■. expelled decanol collected in the upper part of the separator and continuously flowed back into the reaction flask. The analysis showed:

Analogous results are obtained with a technical mixture of alcohols with 10-13 carbon atoms.

Example 5.

In an analogous manner to example 4, 200 g of iso-tridecyl alcohol was converted to secondary di-(iso-C-4)-amine. As catalyst, 5.7 g of water-moist Raney nickel ("B 113" from Firma Degussa) was used, corresponding to 4 g of nickel. The heating period to a reaction temperature of 200°C during each time 20 liters of ammonia and hydrogen per hour took place under the conditions in example 4. After 1 hour the reaction temperature of 200°C had been reached, where they were then further worked on for 2 hours. The water separation was then finished. It was then further rehydrated for 1 hour at 190°C

with 20 liters of hydrogen. After the catalyst was filtered off, 171 g of iso-tridecylamine were formed. The analysis gave:

Example 6.

The aminolysis was carried out in a small circuit gas apparatus made of glass apparatus analogous to the diagram in the figure. Instead of the circuit gas fan in the technical facility, a hose pump was used. After filling the reaction flask with 400 g of stearyl alcohol and 10.8 g of water-moist Raney nickel (8.0 g of Ren nickel), the apparatus was flushed with nitrogen and heated. At 100°C, the nitrogen present was displaced with ammonia and stirred and the circulation pump was switched on. During pushing of ammonia with 25 mm Hg/pre-pressure and some gas release from the apparatus, the reaction temperature of 200°C was reached within a short hour. The exhaust gas was shut off and the apparatus remained connected to the ammonia supply line. Ammonia uptake was measured continuously. The amount of circuit gas was 100 litres/hour. After 3 hours, water separation and ammonia uptake had ended. The work was then continued for another 1 hour at 200°C while increasing the amount of circuit gas to 120 litres. The volume ratio between ammonia and hydrogen in the gas cycle amounted to 46:54 after 1 1/4 hours of reaction time and 74:26 after 3 hours. After 4 hours it was cooled and filtered off from the catalyst. The yield was 372 g. The analysis showed a 94% conversion to amine.

The same experiment was repeated in a somewhat modified form, ie

in the last phase of the 4-hour reaction, extraneous hydrogen was added. Therefore, after 3 hours of reaction time, the gas content in the apparatus was exchanged with a gas mixture of 80 volume% hydrogen and 20 volume% ammonia. With this gas mixture, the work was carried out in

1 hour under otherwise identical conditions as in the previous experiment, namely at 200°C and 120 liters of circuit gas. With this precaution, the amine conversion could be increased to 97.6% and the yield of secondary amine increased. The amine distribution was as follows:

Example 7-

50 kg of melted Myristyl alcohol (tetradecanol) was filled into a 100 liter mixing vessel of circuit gas equipment according to the figure. The flushing with nitrogen took place while the circuit gas fan was switched on. Weighing in of the catalyst and the heating took place as in example 1. A supported nickel powder catalyst ("RCH nickel catalyst 55/5" from Ruhrchemie) was used as catalyst, of which 1.2 kg was used. After the displacement of the nitrogen with ammonia, the circuit gas fan at 120 o C was set at 12 m 3/hour. After 1 hour, the required reaction temperature of 200°C had been reached, the pressure in the apparatus was kept around normal pressure by adding NH 3 via a pressure holding valve. At 200°C and 12 circuit gas/hour was. turnover continued for as long as approx. 90% of the reaction water (4.2 litres) had appeared in the form of concentrated ammonia water in the separator. This was achieved after 2 hours. At this point, the circuit gas was replaced with a mixture of H2 and NH^ in a volume ratio of 5:2. In addition, the reaction temperature was lowered to 180 to 190°C and further worked for 1 hour with 12 m2 of circuit gas. Then the water separation and the reaction were finished. It was then cooled and filtered off from the catalyst. Yield 44.5 g, conversion to amine 95% .

Analysis result:

Example 8.

The reaction was carried out analogously to example? likewise with myristyl alcohol, but the pressure conditions varied: instead of normal pressure, the one-step heating and start-up period took place at 0.8 ata and the two-hour main reaction at 0.9 ata. The water separation has already ended after this time in contrast to example 7 and amounts to 4.3 liters of NH 2 water. ' The post-reaction with Hp/NH-j volume ratio 5:2 is carried out at 1.5' ata. The amine conversion amounts to 97% and the amount of secondary amine 88%. Example 9-

This experiment was carried out analogously to example 7 with 50 kg of lauryl alcohol and 2.5 kg of Raney cobalt. The conversion to amine was 98%. Of secondary dilaurylamine it became achieved "89%". Example 10. As in example 7, 50 kg of tallow fatty alcohol was subjected to aminolysis in the circuit gas apparatus. Tallow fatty alcohol is a mixture of saturated alcohols, which is obtained by perhydration of tallow fatty acid methyl ester. The chain distribution averaged 5% Cl4, - 30% Cl6 and 65% Cl8.

The tallow fatty alcohol used for the experiment had an OH number of 215 and a molecular weight of 260. The catalyst used was Raney nickel ("B 113" from Firma Degussa), of which 1 kg was used in reference to nickel. - Filling with product, catalyst and NH 3 and heating to 200°C took place as in example 7. The packed body column and transition to the cooler were tempered at 90°C. The amount of circuit gas amounted to 13 m^/ hour. After a reaction time of 2 hours at 200°C, 4.6 liters of ammonia water had been formed. At this point, the hydrogen portion in the circuit gas was increased by feeding in H2 and NH^ in a volume ratio of 5:2 from 25 to 70%

and at the same time the reaction temperature was lowered to 180°C. After

After 30 minutes, the amount of hydrogen had increased to 4.95 liters and the aminolysis ended. It was then cooled under further gas circulation in an H 2 /NH 2 atmosphere to 160°C and then purged with N 2 . At 100°C it was filtered from the catalyst. Yield 46.8 kg. It knows

oil passed through the aminolysis amounted to 200 g and was not returned to the process. The conversion to amine was practically complete at 99%. The residual alcohol content was less than 0.5 below the detection limit.

Analysis result:

By washing out the catalyst residue with isopropanol and distilling off, the yield of 46.8 kg of amine could be increased by around 800 g. This precaution is only of importance for individual batches. In normal cases, the catalyst will be completed by adding 10 more.

20% fresh catalyst and again used. After achieving 5 to 6% total catalyst amount, referred to alcohol used, this level is kept constant by draining corresponding amounts of used catalyst and adding fresh catalyst for the further experiments. Example 11.

Following the procedure in example 10, 50 kg of C22 alcohol (behenyl alcohol) with an OH number of 175 was subjected to aminolysis in the circuit gas apparatus. As catalyst, 1 kg of nickel catalyst ("RCH catalyst 55/10" from Ruhrchemie) was used. In contrast to example 10, the amount of circuit gas during the heating period and 2 hours of main reaction amounted to 20 m^/hour. Instead of flushing and pressureless post-treatment with foreign hydrogen and NH 2 , in this case it was flushed with pure hydrogen and then post-hydrogenated with 6 ata H 2 at 180 to 200°C. It was then decanted, cooled to 120°C and filtered.

Yield: 47 kg, amine conversion 95%. ' The secondary amine content was 83%. A <p>roveta<g>nirtg before.after hydration gave 89% by weight

total amine and 73% secondary di-(behenyl)-amine.

Analogous to example 11, one obtains with technical mixtures of alcohols with 19-24 carbon atoms corresponding amounts of secondary <C>19-24~am"'"ners'

Example 12.

The reaction was carried out analogously to example 10.

As raw material, 50 kg of chain distribution alcohol from

C^g to C20 according to the Ziegler method of ethylene ("Alfol -.16/20" from Firma Condea) with an OH number of 218. Instead of the aqueous Raney nickel suspension ("B 213" from Firma Degussa) it was used 875 g of a nickel powder catalyst ("NP" from Firma Unichema). The amine turnover was 97.8%. The secondary amine content was 83.2%. Example 13.

In an apparatus consisting of a four-necked flask with stirrer, thermometer, introduction tube for NH-j and H2, an outlet connection which was equipped with a reflux cooler heated to 80°C, to which was connected a descending cooler coated with tap water. a source for the reaction water was 200 g of tallow fatty alcohol and 2% by weight Raney nickel, referred to alcohol <p>g heated under nitrogen purging. From 100°C it was flushed with NH 3 . From 150°C, a mixture of 125 liters of NH 2 and 100 liters of H 2 per hour and the reaction mixture heated to 200°C. The indicated amount of gas corresponded to 1125 litres/hour and kg of alcohol. After 2.5 hours of NHj/H2 supply at 200°C, the water separation was finished, the ammonia supply was stopped and treated for 30 minutes with 100 liters of hydrogen per hour. After filtering • the catalyst, the analysis gave:

Claims (9)

1. Process for the production of secondary aliphatic amines, whose aliphatic chains, which are saturated or unsaturated, contain 7 to 24 carbon atoms, by reacting the corresponding primary alcohols in the liquid phase with ammonia at normal pressure and elevated temperature in the presence of hydration-dehydration catalysts, characterized in that the alcohol is reacted with ammonia in the presence of hydrogen at temperatures from 120 to 250°C and pressure from 0.5 to 6 ata, the gas velocity being 50 to 1200 litres/hour per kg of alcohol, and the volume ratio between ammonia and hydrogen averaged over the total reaction time is from 80 : 20 to 20 : 80 and the water formed is removed with the gas flow.' 2. Method according to claim 1, characterized, t, in that the turnover is carried out at a gas rate of 50 to 500 litres/hour per kg of alcohol. 3. 'Procedure according to claims 1 to 2, characterized in that the turnover is carried out at a gas rate of 150 to 300 litres/hour per kg of alcohol. 4. Method according to claims 1 to 3, characterized in that the reaction is carried out at a temperature of 160 to 210°C. 5. Method according to claims 1 to 4, characterized in that the turnover is carried out at a pressure of 0.9 to 1.5 ata. 6. Method according to claims 1 to 5, characterized in that the reaction is carried out in the presence of an inert gas. 7- Method according to claims 1 to 6, characterized in that at the end of the reaction the hydrogen portion in the reaction gas is increased, preferably practically to pure hydrogen as gas phase. 8. Method according to claims 1 to 7, characterized in that the gas phase is circulated. 9. Method according to claim 8, characterized in that the reaction is carried out without the addition of extraneous hydrogen.