METHOD OF PRODUCING DIAMINES AND POLYAMINES OF THE DIPHENYLMETHANE SERIES

20200087243 · 2020-03-19

    Inventors

    Cpc classification

    International classification

    Abstract

    The invention relates to a method for producing diamines and polyamines of the diphenylmethane series, by condensing aniline and formaldehyde followed by an acid-catalysed rearrangement at different production capacities with alteration of the isomer composition in the resulting diamines of the diphenylmethane series (altering the 2,4-MDA content). Adapting the molar ratios of the total used aniline to the total used formaldehyde and of the total used acid catalyst to the total used aniline, and adapting the reaction temperature, allows the rearrangement reaction to be fully completed despite the change in dwell time inevitably associated with a change in production capacity, and allows the formation of undesired by-products to be avoided as far as possible; the intended modification to binuclear content is likewise achieved.

    Claims

    1. A process for preparing di- and polyamines of the diphenylmethane series from aniline (1) and formaldehyde (2) in a production plant (10 000), where the molar ratio of total aniline used (1) to total formaldehyde used (2), n(1)/n(2), is always not less than 1.6, comprising: (A-I) reacting aniline (1) and formaldehyde (2) in the absence of an acidic catalyst to obtain a reaction mixture (4) comprising an aminal (3), and then at least partly separating an aqueous phase (6) from the reaction mixture (4) to obtain an organic phase (5) comprising the aminal (3); (A-II) contacting the organic phase (5) which comprises the aminal obtained in step (A-I) with an acidic catalyst (7) in a reactor cascade (3000) composed of i reactors connected in series (3000-1, 3000-2, . . . , 3000-i), where i is a natural number from 2 to 10, wherein: the first reactor (3000-1) in flow direction is operated at a temperature T.sub.3000-1 in the range from 25.0 C. to 65.0 C. and is charged with stream (5) and acidic catalyst (7) and optionally with further aniline (1) and/or further formaldehyde (2), every reactor downstream in flow direction (3000-2, . . . , 3000-i) is operated at a temperature of more than 2.0 C. above T.sub.3000-1 and is charged with the reaction mixture obtained in the reactor immediately upstream; (B) isolating the di- and polyamines of the diphenylmethane series from the reaction mixture (8-i) obtained from step (A-II) in the last reactor (3000-i) by a process comprising: (B-I) adding a stoichiometric excess of base (9), based on the total amount of acidic catalyst used (7), to the reaction mixture (8-i) obtained in the last reactor (3000-i) in step (A-II) to obtain a reaction mixture (10); and (B-II) separating the reaction mixture (10) obtained in step (B-I) into an organic phase (11) comprising di- and polyamines of the diphenylmethane series and an aqueous phase (12); wherein in the event of a change in the production capacity from a starting state A with a mass flow rate m.sub.1 of total aniline used in the starting state of m.sub.1(A)0, a mass flow rate m.sub.2 of total formaldehyde used in the starting state of m.sub.2(A)=X(A).Math.m.sub.2(N), where X(A) is a dimensionless number >0 and 1 and m.sub.2(N) denotes the nameplate load of the production plant (10 000), a molar ratio n(1)/n(2) of total aniline used (1) to total formaldehyde used (2) in the starting state of n(1)/n(2)(A), a molar ratio n(7)/n(1) of total acidic catalyst used to total aniline used in the starting state of n(7)/n(1)(A), a proportion by mass .sub.MMDA, based on the total mass of di- and polyamines of the diphenylmethane series, of diamines of the diphenylmethane series of .sub.MMDA(A), and a proportion by mass .sub.2,4-MDA, based on the total mass of diamines of the diphenylmethane series, of 2,4-methylenediphenyldiamine in the starting state of .sub.2,4-MDA(A) to an end state E with a mass flow rate m.sub.1 of total aniline used of in the end state m.sub.1(E)0, a mass flow rate m.sub.2 of total formaldehyde used in the end state of m.sub.2(E)=X(E).Math.m.sub.2(N), where X(E) is a dimensionless number >0 and 1, a molar ratio n(1)/n(2) of total aniline used (1) to total formaldehyde used (2) in the end state of n(1)/n(2)(E), a molar ratio n(7)/n(1) of total acidic catalyst used to total aniline used in the end state of n(7)/n(1)(E), a proportion by mass .sub.MMDA, based on the total mass of di- and of the diphenylmethane series, of diamines of the diphenylmethane series the end state of .sub.MMDA(E), and a target proportion by mass .sub.2,4-MDA for the end state, based on the total mass of diamines of the diphenylmethane series, of 2,4-methylenediphenylenediamine, of .sub.2,4-MDA(E); by a quantity X=|X(E)X(A)|, with X0.10, wherein the process comprises at least one change in production capacity that commences at a time t.sub.1 and concludes at a time t.sub.2, wherein .sub.2,4-MDA is also altered in such a way that, for the target value of .sub.2,4-MDA(E) for the end state, either 1.15.Math..sub.2,4-MDA(A) or 0.85.Math..sub.2,4-MDA(A), and wherein:
    0.95.Math..sub.MMDA(A).sub.MMDA(E)1.05.Math..sub.MMDA(A); characterized in that, in the period from t.sub.1 to t.sub.2, the transition state T, with a molar ratio of total aniline used (1) to total formaldehyde used (2) of n(1)/n(2)(T) and a molar ratio of total acidic catalyst used to total aniline used of n(7)/n(1)(T), (i) the temperature in the first reactor (3000-1) in flow direction from step (A-II) is adjusted to a value that differs from the temperature in that reactor during the starting state A by not more than 10.0 C.; (ii-1) in the case that m.sub.2(E)>m.sub.2(A), the temperature in at least one of the reactors downstream in flow direction (3000-2, . . . , 3000-i), by comparison with the starting state A, is increased by more than 2.0 C. in such a way that the target end temperature is reached at the latest at time t.sub.2, and in all reactors (3000-2, . . . , 3000-i) in which the temperature is not increased it is kept the same within a range of variation of 2.0 C.; (ii-2) in the case that m.sub.2(E)<m.sub.2(A), the temperature in at least one of the reactors downstream in flow direction (3000-2, . . . , 3000-i), by comparison with the starting state A, is lowered by more than 2.0 C. in such a way that the target end temperature is reached at the latest at time t.sub.2, and in all reactors (3000-2, . . . , 3000-i) in which the temperature is not lowered it is kept the same within a range of variation of 2.0 C.; (iii-1) in the case that .sub.2,4-MDA(E)1.15.Math..sub.2,4-MDA(A), n(1)/n(2)(T) and n(7)/n(1)(T) are adjusted in such a way that, at time t.sub.2:
    1.01.Math.n(1)/n(2)(A)n(1)/n(2)(T)1.50.Math.n(1)/n(2)(A), and
    0.50.Math.n(7)/n(1)(A)n(7)/n(1)(T)0.95.Math.n(7)/n(1)(A); wherein, over the entire transition state before reaching time t.sub.2, it is always the case that:
    0.80.Math.n(1)/n(2)(A)n(1)/n(2)(T)2.50.Math.n(1)/n(2)(A), and
    0.40.Math.n(7)/n(1)(A)n(7)/n(1)(T)1.15.Math.n(7)/n(1)(A); (iii-2) in the case that .sub.2,4-MDA(E)0.85.Math..sub.2,4-MDA(A), n(1)/n(2)(T) and n(7)/n(1)(T) are adjusted in such a way that, at time t.sub.2:
    0.75.Math.n(1)/n(2)(A)n(1)/n(2)(T)0.99.Math.n(1)/n(2)(A), and
    1.05.Math.n(7)/n(1)(A)n(7)/n(1)(T)3.50.Math.n(7)/n(1)(A), wherein, over the entire transition state before reaching time t.sub.2, it is always the case that:
    0.40.Math.n(1)/n(2)(A)n(1)/n(2)(T)2.00.Math.n(1)/n(2)(A), and
    0.80.Math.n(7)/n(1)(A)n(7)/n(1)(T)4.50.Math.n(7)/n(1)(A).

    2. The process of claim 1, in which the temperature in the reactors of the reactor cascade 3000 increases from reactor 3000-1 to reactor 3000-i in all states of operation (A, T, E).

    3. The process of claim 1, in which it is always the case that T.sub.3000-1 is set to a value in the range from 25.0 C. to 65.0 C. and the temperature in each of the reactors downstream in flow direction (3000-2, . . . , 3000-i) is set to a value in the range from 35.0 C. to 200.0 C.

    4. The process of claim 3, in which it is always the case that T.sub.3000-1 is set to a value in the range from 30.0 C. to 60.0 C. and the temperature in each of the reactors downstream in flow direction (3000-2, . . . , 3000-i) is set to a value in the range from 50.0 C. to 180.0 C.

    5. The process of claim 1, in which the acidic catalyst (7) is a mineral acid.

    6. The process of claim 1, in which step (B) further comprises: (B-III) washing the organic phase (11) with washing liquid (13); (B-IV) separating the mixture (14) obtained in step (B-III) into an organic phase (16) comprising di- and polyamines of the diphenylmethane series and an aqueous phase (15); and (B-V) distilling the organic phase (16) from step (B-IV) to obtain the di- and polyamines of the diphenylmethane series (18), with removal of a stream (17) comprising water and aniline.

    7. The process of claim 6, further comprising: (C) recycling stream (17), optionally after workup, into step (A-I) and/or, if the optional addition of further aniline (1) in step (A-II) is conducted, into step (A-II).

    8. The process of claim 1, in which the molar ratio of total aniline used (1) to total formaldehyde used (2), n(1)/n(2), in all states of operation (A, T, E) is adjusted to a value of 1.6 to 20.

    9. The process as of claim 1, in which the values of n(1)/n(2), n(7)/n(1) and of the temperature of each reactor j of the reactor cascade (3000), T.sub.3000-j, that exist in each case at time t.sub.2 are retained for the duration of the production with the formaldehyde mass flow rate m.sub.2(E).

    10. The process of claim 1, in which the target end value for the molar ratio of total aniline used to total formaldehyde used, n(1)/n(2)(E), and the target end value for the molar ratio of total acidic catalyst used to total aniline used, n(7)/n(1)(E), are established by continuously adjusting the value of m.sub.2 and at least one of the values of m.sub.1 and/or m.sub.7 until time t.sub.2.

    11. The process as of claim 1, in which the period from t.sub.1 to t.sub.2 lasts from 1.00 minute to 120 minutes.

    Description

    EXAMPLES

    [0131] The results outlined in the examples for the bicyclic content, the isomer composition and the content of N-methyl-4,4-MDA are based on calculations. The calculations are based partly on theoretical models and partly on process data collected in real operational experiments, the statistical evaluation of which created a mathematical correlation of running parameters and result (e.g. bicyclic content). The content of N-formyl-4,4-MDA is reported on the basis of operational experience values. All percentages and ppm values reported are proportions by mass based on the total mass of the respective stream of matter. The proportions by mass in the real operational experiments that gave the basis for the theoretical model were ascertained by HPLC.

    [0132] Reactor temperatures are based on the temperature of the respective process product at the exit from the reactor.

    [0133] The MDA prepared, in all examples, has a residual aniline content in the range from 50 ppm to 100 ppm and a water content in the range from 200 ppm to 300 ppm.

    A. Reduction in Load Proceeding from Production with Nameplate Load

    I. Starting State: Description of the Conditions Chosen for the Preparation of MDA at Nameplate Load

    [0134] In a continuous reaction process, 23.20 t/h of feed aniline (containing 90.0% by mass of aniline, 1) and 9.60 t/h of 32% aqueous formaldehyde solution (corresponding to a molar ratio of aniline (1):formaldehyde (2) of 2.25:1) are mixed and converted to the aminal (3) at a temperature of 90.0 C. and a pressure of 1.40 bar (absolute) in a stirred reaction tank (1000). The reaction tank is provided with a cooler having a cooling circuit pump. The reaction mixture leaving the reaction tank is guided into a phase separation apparatus (aminal separator, 2000) (step (A-I)).

    [0135] After the phase separation to remove the aqueous phase (6), the organic phase (5) is admixed in a mixing nozzle with 30% aqueous hydrochloric acid (7) (protonation level 10%, i.e. 0.10 mol of HCl is added per mole of amino groups) and run into the first rearrangement reactor (3000-1). The first rearrangement reactor (called vacuum tank) is operated at 50.0 C., which is ensured by means of evaporative cooling in a reflux condenser at a pressure of 104 mbar (absolute). The reflux condenser is charged with 0.50 t/h of fresh aniline. The rearrangement reaction is conducted to completion in a reactor cascade composed of a total of seven reactors at 50.0 C. to 156.0 C. (i.e. 50.0 C. in reactor 3000-1/60.0 C. in reactor 3000-2/83.0 C. in reactor 3000-3/104.0 C. in reactor 3000-4/119.0 C. in reactor 3000-5/148.0 C. in reactor 3000-6/156.0 C. in reactor 3000-7) (step (A-II)).

    [0136] On completion of reaction, the reaction mixture (8-i) obtained is admixed with 32% sodium hydroxide solution in a molar ratio of 1.10:1 sodium hydroxide to HCl and reacted in a stirred neutralization vessel (4000) (step (B-I)). The temperature here is 115.0 C. The absolute pressure is 1.40 bar. The neutralized reaction mixture (10) is then separated in a neutralization separator (5000) into an aqueous lower phase (12), which is guided to a wastewater collection vessel, and into an organic phase (11) (step (B-II)).

    [0137] The organic upper phase (11) is guided to the washing and washed with condensate (13) in a stirred washing vessel (6000) (step (B-III)). After the washing water (15) has been separated from the biphasic mixture (14) obtained in the washing vessel (6000) in a washing water separator (7000, step (B-IV)), the crude MDA (16) thus obtained is freed of water and aniline (removed together as stream 17) by distillation, and 17.00 t/h of MDA (18) were obtained as bottom product (step (B-V)).

    [0138] MDA prepared in this way has an average composition of 45.2% 4,4-MDA, 5.5% 2,4-MDA, 0.3% 2,2-MDA, i.e. a total bicyclic content of 51.0% and a proportion by mass of 2,4-MDA of 10.8%, and also 0.3% N-methyl-4,4-MDA and 0.3% N-formyl-4,4-MDA, the remainder to 100% consisting essentially of higher homologs (PMDA) and isomers thereof.

    II. Target End State: Production at Half-Load

    Example 1 (Comparative Example): Reduction in Load of the MDA Plant from Nameplate Load to Half-Load (=50% of Nameplate Load), where the n(1)/n(2) Ratio and the n(7)/n(1) Ratio are Kept the Same, the Aniline, Formalin, Hydrochloric Acid and Sodium Hydroxide Solution Feedstocks are Reduced Simultaneously and the Temperatures in the Vacuum Tank (3000-1), the Downstream Reactors and the Exit Temperature of the Crude MDA Solution in the Last Rearrangement Reactor (3000-7) Remain the Same

    [0139] The MDA plant, as described above under A.I, is operated at a production capacity of 17.00 t/h of MDA. Now the proportion by mass of 2,4-MDA is to be distinctly reduced, combined with halving of the load owing to lower product demand. For this purpose, at the same time, the feed rates of aniline and formalin to the aminal reactor (1000) are adjusted to the new production load within 120 minutes. The formalin rate is reduced to 4.80 t/h. The aniline feed rate is reduced to 11.35 t/h. At the same time, the flow rate of hydrochloric acid into the mixing nozzle in the feed to the first rearrangement reactor (3000-1) is halved. The first rearrangement reactor is still operated at 50.0 C., which is ensured by means of the evaporative cooling in the reflux condenser at 104 mbar (absolute). The reflux condenser is still charged with 0.50 t/h of fresh aniline. The rearrangement reaction is conducted to completion in a reactor cascade at 50.0 C. to 156.0 C. (50.0 C. in reactor 3000-1/60.0 C. in reactor 3000-2/83.0 C. in reactor 3000-3/104.0 C. in reactor 3000-4/119.0 C. in reactor 3000-5/148.0 C. in reactor 3000-6/156.0 C. in reactor 3000-7). On completion of reaction, the reaction mixture obtained, as described in the general conditions for preparation of MDA, is neutralized with sodium hydroxide solution, with reduction of the amount of sodium hydroxide solution within the same time window as formalin and aniline, and then worked up to give MDA (18).

    [0140] 40 hours after commencement of the change in load, the MDA in the feed to the MDA tank has a composition of 44.7% 4,4-MDA, 5.2% 2,4-MDA, 0.2% 2,2-MDA, i.e. a total bicyclic content of 50.1% and a proportion by mass of 2,4-MDA of 10.3%, and also 0.3% N-methyl-4,4-MDA and 0.3% N-formyl-4,4-MDA, the remainder to 100% consisting essentially of higher homologs (PMDA) and isomers thereof. The MDA has an elevated proportion of unwanted by-products with acridine and acridane structure. In this regard, see also section C further down. The bicyclic content is essentially the same as that in the starting state.

    Example 2 (Inventive): Reduction in Load of the MDA Plant from Nameplate Load to Half-Load, where the n(1)/n(2) Ratio is Lowered, the Aniline and Formalin Feedstocks are Reduced Simultaneously and where the n(7)/n(1) Ratio is Increased with Retention of the Hydrochloric Acid Volume, Sodium Hydroxide Solution is Reduced Simultaneously with the Aniline and Formalin Feedstocks, the Temperature in the Vacuum Tank (3000-1) Remains the Same and the Exit Temperature of the Crude MDA Solution in the Last Rearrangement Reactor (3000-7) is Lowered

    [0141] The MDA plant, as described above under A.I, is operated at a production capacity of 17.00 t/h of MDA. Now the proportion by mass of 2,4-MDA is to be distinctly reduced, combined with halving of the load owing to lower product demand. For this purpose, at the same time, the feed rates of aniline and formalin to the aminal reactor (1000) are adjusted to the new production load within 120 minutes. The formalin rate is reduced to 4.80 t/h. The aniline feed rate is reduced to 10.85 t/h. The flow rate of hydrochloric acid into the mixing nozzle in the feed to the first rearrangement reactor is kept the same, i.e. the protonation level is increased to 21%. The first rearrangement reactor (3000-1) is still operated at 50.0 C., which is ensured by means of the evaporative cooling in the reflux condenser at 104 mbar (absolute). The reflux condenser is still charged with 0.50 t/h of fresh aniline. The rearrangement reaction is conducted to completion in a reactor cascade at 50.0 C. to 146.0 C. (50.0 C. in reactor 3000-1/60.0 C. in reactor 3000-2/81.0 C. in reactor 3000-3/95.0 C. in reactor 3000-4/116.0 C. in reactor 3000-5/144.0 C. in reactor 3000-6/146.0 C. in reactor 3000-7) (step (A-II)). On completion of reaction, the reaction mixture obtained, as described in the general conditions for preparation of MDA, is neutralized with sodium hydroxide solution, with reduction of the amount of sodium hydroxide solution within the same time window as formalin and aniline and HCl with retention of the molar ratio of 1.10:1 sodium hydroxide solution to HCl, and then worked up to give the desired MDA type, obtaining 8.5 t/h of MDA (18) as the bottom product from the distillation.

    [0142] 40 hours after commencement of the change in load, the MDA in the feed to the MDA tank has a composition of 46.9% 4,4-MDA, 4.3% 2,4-MDA, 0.2% 2,2-MDA, i.e. a total bicyclic content of 51.4% and a proportion by mass of 2,4-MDA of 8.3%, and also 0.2% N-methyl-4,4-MDA and 0.2% N-formyl-4,4-MDA, the remainder to 100% consisting essentially of higher homologs (PMDA) and isomers thereof. The product differs corresponds to the desired target product. An elevated proportion of unwanted by-products with acridine and acridane structure is not formed.

    [0143] Table 1 below compares the results from section A.

    TABLE-US-00001 TABLE 1 Comparison of the examples from section A Starting state End state Half-load End state Half-load Nameplate load Example 1 (comp.) Example 2 (inv.) Aniline (90%) in reactor 1000 [t/h] 23.20 11.35 10.85 Aniline in reactor 3000-1 [t/h] 0.50 0.50 0.50 Formalin (32%) in reactor 1000 [t/h] 9.60 4.80 4.80 n(1)/n(2) 2.25 2.25 2.16 [n(1)/n(2)(T)(t = t.sub.2)]/[n(1)/n(2)(A)] 1.00 0.96 Protonation level [%] 10 10 21 n(7)/n(1) 0.10 0.10 0.21 Temp. gradient in reactor cascade 50.0 .fwdarw. 156.0 50.0 .fwdarw. 156.0 50.0 .fwdarw. 146.0 3000 [ C.] T(3000-1) [ C.] 50.0 50.0 50.0 T(3000-2) [ C.] 60.0 60.0 60.0 T(3000-3) [ C.] 83.0 83.0 81.0 T(3000-4) [ C.] 104.0 104.0 95.0 T(3000-5) [ C.] 119.0 119.0 116.0 T(3000-6) [ C.] 148.0 148.0 144.0 T(3000-7) [ C.] 156.0 156.0 146.0 Production capacity [t/h] 17.00 8.50 8.50 4,4-MDA [%] 45.2 44.7 46.9 2,4-MDA [%] 5.5 5.2 4.3 2,2-MDA [%] 0.3 0.2 0.2 N-methyl [%] 0.3 0.3 0.3 N-formyl [%] 0.3 0.3 0.3 Bicyclic [%] 51.0 50.1 51.4 (2,4-MDA) [%] 10.8 10.3 8.3 (2,4-MDA)(E)/(2,4-MDA)(A) 0.96 0.77 Comment Elevated As desired, product proportion of has a distinctly unwanted by- lowered 2,4-MDA products with content with very acridine and similar bicyclic acridane structure; content and a 2,4-MDA content similar by-product virtually spectrum to the unchanged product in the starting state

    B. Increase in Load Proceeding from Production at Half-Load

    I. Starting State: Description of the Conditions Chosen for the Preparation of MDA at Half-Load

    [0144] (N.B. There are of course various options for operating a production plant 10 000 at half-load.) The conditions set in example 2 are one option; for examples 3 and 4 which follow, another option was chosen for the half-load starting state.)

    [0145] The reaction is operated as described above under A.I for nameplate load with the following differences:

    [0146] 11.35 t/h of feed aniline (containing 90.0% by mass of aniline);

    [0147] 4.80 t/h of 32% aqueous formaldehyde solution (i.e. the molar ratio of aniline:formaldehyde is 2.25:1);

    [0148] The first rearrangement reactor (3000-1) is still operated at 50.0 C., which is ensured by means of the evaporative cooling in the reflux condenser at 104 mbar (absolute). The reflux condenser is still charged with 0.50 t/h of fresh aniline.

    [0149] 50.0 C. in reactor 3000-1/60.0 C. in reactor 3000-2/81.0 C. in reactor 3000-3/95.0 C. in reactor 3000-4/116.0 C. in reactor 3000-5/144.0 C. in reactor 3000-6/146.0 C. in reactor 3000-7;

    [0150] Bottom product of 8.50 t/h of MDA (18).

    [0151] MDA prepared in this way has an average composition of 46.3% 4,4-MDA, 5.0% 2,4-MDA, 0.2% 2,2-MDA, i.e. a total bicyclic content of 51.5% and a proportion by mass of 2,4-MDA of 9.7%, and also 0.3% N-methyl-4,4-MDA and 0.3% N-formyl-4,4-MDA, the remainder to 100% consisting essentially of higher homologs (PMDA) and isomers thereof.

    II. Target End State: Production at Nameplate Load

    Example 3 (Comparative Example): Increase in Load of the MDA Plant from Half-Load to Nameplate Load, where the n(1)/n(2) Ratio and the n(7)/n(1) Ratio are Kept the Same, the Aniline, Formalin, Hydrochloric Acid and Sodium Hydroxide Solution Feedstocks are Increased Simultaneously and the Temperature in the Vacuum Tank (3000-1) Remains the Same and the Exit Temperature of the Crude MDA Solution in the Last Rearrangement Reactor (3000-7) is Increased

    [0152] The MDA plant, as described above under B.I, is operated at a production capacity of 8.50 t/h of MDA. Now the proportion by mass of 2,4-MDA is to be distinctly increased, combined with a rise in load to nameplate load owing to higher product demand. For this purpose, at the same time, the feed rates of aniline and formalin to the aminal reactor are adjusted to the new production load within 120 minutes. The formalin rate is increased to 9.60 t/h. The aniline feed rate is increased to 23.20 t/h. The flow rate of hydrochloric acid into the mixing nozzle in the feed to the first rearrangement reactor is increased within the same period as aniline and formalin with retention of the protonation level of 10% (i.e. with retention of the n(7)/n(1) ratio). The first rearrangement reactor (3000-1) is still operated at 50.0 C., which is ensured by means of the evaporative cooling in the reflux condenser at 104 mbar (absolute). The reflux condenser is charged with 0.50 t/h of fresh aniline. The rearrangement reaction is conducted to completion in the reactor cascade at 50.0 C. to 156 C. (50.0 C. in reactor 3000-1/60.0 C. in reactor 3000-2/83.0 C. in reactor 3000-3/104.0 C. in reactor 3000-4/119.0 C. in reactor 3000-5/148.0 C. in reactor 3000-6/156.0 C. in reactor 3000-7).

    [0153] After the reaction, the reaction mixture obtained is admixed with 32% sodium hydroxide solution in a molar ratio of 1.10:1 sodium hydroxide solution to HCl and reacted in a stirred neutralization vessel, increasing the amount of sodium hydroxide solution within the same time window as formalin, aniline and HCl with retention of the molar ratios. The further workup is effected as described above under A.I. At the end of the transition state, 17.0 t/h of a bottom product are obtained.

    [0154] 20 hours after commencement of the change in load, the MDA in the feed to the MDA tank has a composition of 45.2% 4,4-MDA, 5.5% 2,4-MDA, 0.3% 2,2-MDA, i.e. a total bicyclic content of 50.0% and a proportion by mass of 2,4-MDA of 10.8%, and also 0.3% N-methyl-4,4-MDA and 0.3% N-formyl-4,4-MDA, the remainder to 100% consisting essentially of higher homologs (PMDA) and isomers thereof. The desired distinct increase in the 2,4-MDA content was not achieved.

    Example 4 (Inventive)

    [0155] Increase in Load of the MDA Plant from Half-Load to Near-Nameplate Load, where the n(1)/n(2) Ratio is Increased, the Aniline and Formalin Feedstocks are Increased Simultaneously, and where the n(7)/n(1) Ratio is Lowered, Hydrochloric Acid and Sodium Hydroxide Solution are Increased Simultaneously, the Temperature in the Vacuum Tank (3000-1) is Kept the Same and the Exit Temperature of the Crude MDA Solution in the Last Rearrangement Reactor (3000-7) is Increased

    [0156] The MDA plant, as described above under B.I, is operated at a production capacity of 8.50 t/h of MDA. Now the proportion by mass of 2,4-MDA is to be distinctly increased, combined with a distinct rise in load owing to higher product demand. For this purpose, at the same time, the feed rates of aniline and formalin to the aminal reactor are adjusted to the new production load within 120 minutes. The formalin rate is increased to 9.00 t/h. The aniline feed rate is increased to 22.30 t/h. The flow rate of hydrochloric acid into the mixing nozzle in the feed to the first rearrangement reactor is adjusted within the same period as aniline and formalin with lowering of the protonation level to 6%. The first rearrangement reactor (3000-1) is still operated at 50.0 C., which is ensured by means of the evaporative cooling in the reflux condenser at 104 mbar (absolute). The reflux condenser is charged with 0.50 t/h of fresh aniline. The rearrangement reaction is conducted to completion in the reactor cascade at 50.0 C. to 156 C. (50.0 C. in reactor 3000-1/60.0 C. in reactor 3000-2/83.0 C. in reactor 3000-3/104.0 C. in reactor 3000-4/119.0 C. in reactor 3000-5/148.0 C. in reactor 3000-6/156.0 C. in reactor 3000-7) (step A-II)).

    [0157] After the reaction, the reaction mixture obtained is admixed with 32% sodium hydroxide solution in a molar ratio of 1.10:1 sodium hydroxide solution to HCl and reacted in a stirred neutralization vessel, increasing the amount of sodium hydroxide solution within the same time window as formalin, aniline and hydrochloric acid. The temperature is 115.0 C. The absolute pressure is 1.40 bar. The further workup is effected as described further up under A.I. At the end of the transition state, the bottom product obtained is 16.00 t/h of MDA (18).

    [0158] 20 hours after commencement of the change in load, the MDA in the feed to the MDA tank has a composition of 45.1% 4,4-MDA, 5.9% 2,4-MDA, 0.3% 2,2-MDA, i.e. a total bicyclic content of 51.3% and a proportion by mass of 2,4-MDA of 11.5%, and also 0.3% N-methyl-4,4-MDA and 0.3% N-formyl-4,4-MDA, the remainder to 100% consisting essentially of higher homologs (PMDA) and isomers thereof. The product differs only insignificantly from the MDA stream (18) which is obtained on average in the starting state as described in B.I. An elevated proportion of unwanted by-products with acridine and acridane structure is not formed.

    [0159] Table 2 below compares the results from section B.

    TABLE-US-00002 TABLE 2 Comparison of the examples from section B Starting state End state Nameplate End state Nameplate Half-load load Example 3 (comp.) load Example 4 (inv.) Aniline (90%) in reactor 1000 [t/h] 11.35 23.20 22.30 Aniline in reactor 3000-1 [t/h] 0.50 0.50 0.50 Formalin (32%) in reactor 1000 [t/h] 4.80 9.60 9.00 n(l)/n(2) 2.25 2.25 2.30 [n(1)/n(2)T)(t = t.sub.2)]/[n(1)/n(2)(A)] 1.00 1.02 Protonation level [%] 10 10 6 n(7)/n(1) 0.10 0.10 0.06 Temp. gradient in reactor cascade 50.0 .fwdarw. 146.0 50.0 .fwdarw. 156.0 50.0 .fwdarw. 156.0 3000 [ C.] T(3000-1) [ C.] 50.0 50.0 50.0 T(3000-2) [ C.] 60.0 60.0 60.0 T(3000-3) [ C.] 81.0 83.0 83.0 T(3000-4) [ C.] 95.0 104.0 104.0 T(3000-5) [ C.] 116.0 119.0 119.0 T(3000-6) [ C.] 144.0 148.0 148.0 T(3000-7) [ C.] 146.0 156.0 156.0 Production capacity [t/h] 8.50 17.00 16.00 4,4-MDA [%] 46.3 45.24 45.1 2,4-MDA [%] 5.0 5.5 5.9 2,2-MDA [%] 0.2 0.3 0.3 N-methyl [%] 0.3 0.3 0.3 N-formyl [%] 0.3 0.3 0.3 Bicyclic [%] 51.5 51.0 51.3 2,4-MDA content [%] 9.7 10.8 11.5 (2,4-MDA)(E)/(2,4-MDA)(A) 1.12 1.19 Comment Insufficient As desired, product increase in the has a distinctly 2,4-MDA content elevated 2,4-MDA content with very similar bicyclic content and a similar by-product spectrum to the product in the starting state

    C. Fundamental Experiments for Formation of by-Products with Acridine and Acridane Structure

    [0160] In a series of experiments, the amount of aqueous 30% hydrochloric acid required in each case to achieve the desired protonation level (see table 3 below) was added to a 2,2-MDA solution in aniline preheated to 100.0 C. The 2,2-MDA concentration in each of the individual experiments was 1.0% by mass; in addition, the solutions contained octadecane as an internal standard for gas chromatography (GC) analysis. The resulting mixture was transferred as quickly as possible by means of a peristaltic pump to a Bchi glass autoclave preheated to 120.0 C. and heated to the reaction temperature envisaged (see table 3). On attainment of the desired reaction temperature, the first sample was taken (time=zero). In order to monitor the progress of the reaction, further samples were taken after 30, 60, 120 and 240 minutes and analyzed by means of GC analysis. Reaction conditions and experimental results are collated in table 3 below.

    ##STR00002##

    TABLE-US-00003 TABLE 3 Laboratory experiments for formation of the acridane and acridine secondary components Sum total Reaction Protonation (acridine + temperature level Time acridane) Experiment [ C.] [%] [min] [ppm] 1 160 C. 10% 15 79 30 208 60 421 120 925 240 1775 2 170 C. 25% 0 977 30 2554 60 3950 120 5647 240 8894 3 160 C. 25% 0 364 30 930 60 1594 120 2681 240 4354 4 180 10% 0 606 30 1630 60 2613 120 3932 240 5294 5 170 10% 0 292 30 794 60 1533 120 2406 240 3932 6 160 5% 0 0 30 0 60 0 120 0 240 0 7 170 5% 0 0 30 0 60 0 120 0 240 218 8 180 5% 0 0 30 402 60 834 120 1355 240 2380

    [0161] It is found that, with rising temperature and rising hydrochloric acid concentration, the formation of the acridine and acridane secondary components from 2,2-MDA occurs to an increased degree.