HETEROGENEOUS SYNTHESIS OF METHYLENE DIANILINE

Abstract

The present invention relates to a catalytic material for the preparation of one or more of 4,4′-methylenedianiline, 2,2′-methylenedianiline, 2,4′-methylenedianiline, and oligomers of two or more thereof, the catalytic material comprising an oxidic support, wherein the oxidic support comprises an element E.sub.OS1 selected from the group consisting of Ti, Zr, Al, Si, and mixtures of two or more thereof, and further comprising a supported material supported on the oxidic support, wherein the supported material comprises an element E.sub.SM1 selected from the group consisting of Ti, Zr, V, Nb, Ta, Mo, W, Ge, Sn, Sc, Y, La, Ce, Nd, Pr, Hf, Cr, Fe, Co, Ni, Cu Zn, Pb and mixtures of two or more thereof. Further, the present invention relates in particular to a process for the preparation of a catalytic material and to a process for the preparation of one or more of 4,4′-methylenedianiline, 2,2′-methylenedianiline, 2,4′-methylenedianiline and oligomers of two or more thereof.

Claims

1. A catalytic material for the preparation of one or more of 4,4′-methylenedianiline, 2,2′-methylenedianiline, 2,4′-methylenedianiline, and oligomers of two or more thereof, the catalytic material comprising an oxidic support, wherein the oxidic support comprises an element E.sub.OS1 selected from the group consisting of Ti, Zr, Al, Si, and mixtures of two or more thereof, and further comprising a supported material supported on the oxidic support, wherein the supported material comprises an element E.sub.SM1 selected from the group consisting of Ti, Zr, V, Nb, Ta, Mo, W, Ge, Sn, Sc, Y, La, Ce, Nd, Pr, Hf, Cr, Fe, Co, Ni, Cu, Zn, Pb, and mixtures of two or more thereof.

2. The catalytic material of claim 1, wherein the oxidic support further comprises an element E.sub.OS2 selected from group 4, 13 and 14 elements of the periodic system of elements, including mixtures of two or more thereof, wherein E.sub.OS2 is different to E.sub.OS1.

3. The catalytic material of claim 2, wherein E.sub.OS2 is selected from the group consisting of Ti, Zr, Al, Si, Sn, and mixtures of two or more thereof.

4. The catalytic material of claim 1, wherein E.sub.SM1 is selected from the group consisting of Sc, Y, La, Ce, Nd, Pr, Hf, Cr, Fe, Co, Ni, Cu, Zn, Pb, Zr, and mixtures of two or more thereof.

5. The catalytic material of claim 1, wherein the catalytic material comprises E.sub.SM1, calculated as element, in an amount in the range of from 0.1 to 10 weight-%, based on the total weight of the oxidic support.

6. The catalytic material of claim 1, wherein the supported material further comprises an element E.sub.SM2 selected from groups 3-14 of the periodic system of elements, rare earth metals, and mixtures of two or more thereof, wherein E.sub.SM2 is different to E.sub.SM1.

7. The catalytic material of claim 6, wherein E.sub.SM2 is selected from groups 3, 4, 5, 6, 8, 10, 11, 12, 14, rare earth metals, and mixtures of two or more thereof.

8. The catalytic material of claim 6, wherein the catalytic material comprises E.sub.SM2, calculated as element, in an amount in the range of from 0.1 to 3 weight-%, based on the total weight of the oxidic support.

9. The catalytic material of claim 1, wherein from 95 to 100 weight-% of the catalytic material consists of the oxidic support and the supported material, based on the total weight of the catalytic material.

10. The catalytic material of claim 1, having a mesopore volume V.sub.meso in the range of from 0.50 to 1.30 cm.sup.3/g, wherein the mesopore volume is determined according to Reference Example 5.

11. The catalytic material of claim 1, exhibiting an acid site density in the range of from 0.050 to 1.000 μmol/g, determined as described in Reference Example 7.

12. A process for the preparation of a catalytic material, the process comprising (i) preparing a mixture comprising a liquid solvent system, a source of an oxidic support comprising an element E.sub.OS1 selected from the group consisting of Ti, Zr, Al, Si, and mixtures of two or more thereof, a source of an element E.sub.SM1 selected from the group consisting of Ti, Zr, V, Nb, Ta, Mo, W, Ge, Sn, Sc, Y, La, Ce, Nd, Pr, Hf, Cr, Fe, Co, Ni, Cu, Zn, Pb, and mixtures of two or more thereof, and optionally a source of an element E.sub.SM2 selected from groups 3-14 of the periodic system of elements, rare earth metals, and mixtures of two or more thereof, wherein the optional E.sub.SM2 is different to E.sub.SM1, obtaining a precursor of the catalytic material; (ii) calcining the precursor of the catalytic material in a gas atmosphere, obtaining the catalytic material.

13. A catalytic material obtainable and/or obtained by the process of claim 12.

14. A process for the preparation of one or more of 4,4′-methylenedianiline, 2,2′-methylenedianiline, 2,4′-methylenedianiline, and oligomers of two or more thereof, the process comprising (1) providing a reactor comprising a reaction zone, wherein the reaction zone comprises a catalytic material according to claim 1; (2) providing a feed into the reaction zone according to (1), wherein the feed comprises one or more of aniline, formaldehyde, and N,N′-diphenylmethylenediamine; (3) converting the feed under reaction conditions in the reaction zone; obtaining a product mixture comprising one or more of 4,4′-methylenedianiline, 2,2′-methylenedianiline, 2,4′-methylenedianiline, and oligomers of two or more thereof; (4) separating the product mixture from the reaction zone.

15. (canceled)

Description

EXAMPLES

Reference Example 1: Determination of the Lewis Acidity and of the Brønsted Acidity

[0194] The nature of the acid sites and the acid site density were determined for a sample material by pyridine adsorption using FTIR spectroscopy with pyridine as probe molecule. A self-supported wafer was placed in a cell under vacuum and activated at 250° C. for 1 h. The cell was subsequently cooled and a reference spectrum was recorded at 150° C. After this, the cell was further cooled and pyridine (25 mbar) was contacted with the wafer at 50° C. until the sample material was saturated. Weakly coordinated pyridine was removed by evacuation for 30 min, prior to IR spectrum recording at 150° C.

[0195] The difference IR spectra of adsorbed pyridine were measured for a sample and normalized to 10 mg of sample/cm.sup.2. From the difference IR spectra bands assigned to pyridine adsorbed on hydroxyl groups or physisorbed pyridine at 1595 and 1445 cm.sup.−1 were absent after evacuating the samples at 423 K, allowing for the quantification of Brønsted and Lewis acid sites, these were determined using Emeis' integrated molar extinction coefficients (see C. A. Emeis: “Determination of integrated molar extinction coefficients for infrared absorption bands of pyridine adsorbed on solid acid catalysts” in Journal of Catalysis 1993, vol. 141, p. 347-354 and M. Velthoen et al. “Probing acid sites in solid catalysts with pyridine UV-Vis spectroscopy” in Physical Chemistry Chemical Physics 2018, vol. 20, p. 21647-21659 as regards the determination of acidity properties of a material via spectroscopy and the interpretation of the resulting data) and the areas of the absorption bands at 1545 cm.sup.−1 (Brønsted acid sites) and 1450 cm.sup.−1 (Lewis acid sites).

Reference Example 2: Determination of the Thermogravimetric Profiles (TGA-MS) and Aniline-Sample Interaction

[0196] A stock solution of aniline in THF was prepared (0.0025 g.sub.Aniline/mL.sub.THF) Catalyst samples were dried at 200° C. overnight. Then, 5 weight-% aniline with respect to the dry catalyst was added to the dry catalyst using the stock solution. The mixture was sealed and stirred at room temperature until dry (so called “free-flowing”). A sample was analyzed with TGA-MS, wherein TGA-MS measurements were carried out at a heating rate of 5° C./min to a final temperature of 600° C. under a nitrogen atmosphere.

[0197] As can be taken from the results shown in FIG. 1, the TGA-MS results indicate that the aniline-catalyst interaction is strongest in zeolite Y and weakest in non-loaded silica since the maximum MS intensity is reached for SiO.sub.2, and inventive catalytic materials Hf, Zn/SiO.sub.2 and Hf/SiO.sub.2 at a temperature of about 100° C. whereas the maximum MS intensity is reached for zeolite Y at a temperature in the range of from 350 to 400° C.

Reference Example 3: Determination of N.SUB.2 .Adsorption/Desorption Isotherm

[0198] N.sub.2 physisorption isotherms were collected at 77 K on a Micromeritics 3Flex Surface Characterization Analyzer. Samples were outgassed under vacuum at 423 K for 4 h prior to data collection.

Reference Example 4: Determination of the Total Pore Volume

[0199] The total pore volume was estimated by DFT calculations (0.731 cm.sup.3/g for Hf Zn/SiO.sub.2).

Reference Example 5: Determination of the Mesopore Volume

[0200] The mesopore volume V.sub.meso was determined via BJH method.

Reference Example 6: Determination of the BET Specific Surface Area S.SUB.BET .and of the Specific External Surface Area S.SUB.EXT

[0201] The specific surface area (S.sub.BET) was determined using the BET method (0.001-0.20 p/p.sub.0 range) and the specific external surface area (S.sub.EXT) was obtained using the t-plot method.

Reference Example 7: Determination of Acidity with NH.SUB.3.-TPD

[0202] The acidity of a catalytic material was measured by temperature-programmed-desorption of NH.sub.3 (NH.sub.3-TPD), which was conducted on a Gasmet DX4000 FTIR gas analyser with Calcmet software for converting the collected spectral data into ammonia concentrations. The sample was activated at 200° C. for 2 h under nitrogen, then cooled to 100° C. and dosed with an ammonia/nitrogen gas mixture until saturation at this temperature. Non-adsorbed ammonia was then flushed away with nitrogen at 100° C. The sample was subsequently heated at a linear heating rate under nitrogen and the amount of desorbed ammonia was monitored with a downstream detector.

Reference Example 8: Determination of Metal Content

[0203] Inductively coupled plasma-optical emission spectrometry (ICP-OES) was used to determine the metal content of the samples using a Varian 720-ES.

Example 1: Preparation of a Catalytic Material Comprising Hf and Zn Supported on Silica

[0204] 0.542 g hafnium chloride (HfCl.sub.4), 0.231 g zinc nitrate (Zn(NO.sub.3).sub.2.6H.sub.2O) and 10.090 g fumed silica (Cabot CAB-O-SIL M5) were added to 100 ml ultrapure water (Ultrapure water, Millipore Milli-Q Type 1) at around 23° C. The mixture had a theoretical content of 3.0 weight-% Hf and 0.5 weight-% Zn, each with respect to the total weight of the fumed silica. The mixture was covered with filter paper (Whatman Grade 5 2.5 μm), stirred for 5 days at around 23° C., and subsequently dried at 60° C. The dried material was then heated at a rate of 2° C./min to 500° C. and held at this temperature for 2 h.

[0205] The nature of the acid sites and the acid site density of the resulting material were determined by pyridine adsorption using FTIR spectroscopy with pyridine as probe molecule. No Brønsted acid sites were observed but 62.4 μmol/g of Lewis acid sites, both determined according to Reference Example 1. Independently thereof, the Lewis acidity was determined by temperature-programmed-desorption of ammonia according to Reference Example 7 as being 0.157 mmol/g. The material had a total pore volume of 0.731 cm.sup.3/g, determined according to Reference Example 4, a BET specific surface area S.sub.BET of 176 m.sup.2/g, a specific external surface area S.sub.EXT of 169 m.sup.2/g, both determined according to Reference Example 6, and it showed a desorption peak having a maximum at about 225° C. in the temperature programmed desorption of ammonia (NH.sub.3-TPD), determined according to Reference Example 7 disclosed herein. The specific surface area (S.sub.BET) and the specific external surface area (S.sub.EXT) were similar, indicating that the material was mesoporous. Thus, the mesopore volume was determined according to Reference Example 5 as being 0.97 cm.sup.3/g. Furthermore, the metal contents were determined according to Reference Example 8 as being 2.7 weight-% Hf and 0.5 weight-% Zn, each based on 100 weight-% SiO.sub.2.

Example 2: Preparation of a Catalytic Material Comprising Hf and Zn Supported on Silica-Alumina

[0206] Example 1 was repeated wherein 112 mg hafnium chloride (HfCl.sub.4), 50 mg zinc nitrate (Zn(NO.sub.3).sub.2. 6H.sub.2O were used and 2039 mg silica-alumina (Grace, MS 13/110, 13% Al.sub.2O.sub.3, S.sub.BET=475 m.sup.2/g) were used instead of fumed silica as oxidic support.

[0207] The nature of the acid sites and the acid site density of the resulting material were determined by pyridine adsorption using FTIR spectroscopy with pyridine as probe molecule. 21.6 μmol/g of Brønsted acid sites were observed and 62.2 μmol/g of Lewis acid sites, both determined according to Reference Example 1. The Lewis acidity was also determined by temperature-programmed-desorption of ammonia according to Reference Example 7 as being 0.456 mmol/g. The material had a BET specific surface area S.sub.BET of 303 m.sup.2/g, a specific external surface area SEXT of 294 m.sup.2/g, both determined according to Reference Example 6, and it showed a desorption peak having a first maximum at about 190° C. and a second maximum at about 325° C. in the temperature programmed desorption of ammonia (NH.sub.3-TPD), determined according to Reference Example 7 disclosed herein. The specific surface area (S.sub.BET) and the specific external surface area (S.sub.EXT) were similar, indicating that the material was mesoporous. Thus, the mesopore volume was determined according to Reference Example 5 as being 0.94 cm.sup.3/g. Furthermore, the metal contents were determined according to Reference Example 8 as being 2.9 weight-% Hf and 0.5 weight-% Zn, each based on 100 weight-% SiO.sub.2.Al.sub.2O.sub.3.

Example 3: Preparation of a Catalytic Material Comprising Fe Supported on Silica-Alumina

[0208] Example 1 was repeated wherein 218 mg iron nitrate (Fe(NO.sub.3).sub.3.9H.sub.2O) were used instead of hafnium chloride and zinc nitrate, wherein further 984 mg silica-alumina (Grace, MS 13/110, 13% Al.sub.2O.sub.3, S.sub.BET=475 m.sup.2/g) has been used instead of silica as oxidic support. The mixture had a theoretical content of 3 weight-% Fe with respect to the total weight of the fumed silica-alumina.

[0209] The nature of the acid sites and the acid site density were determined by pyridine adsorption using FTIR spectroscopy with pyridine as probe molecule. 63.8 μmol/g of Brønsted acid sites were observed and 127.2 μmol/g of Lewis acid sites, both determined according to Reference Example 1. The Lewis acidity was also determined by temperature-programmed-desorption of ammonia according to Reference Example 7 as being 0.505 mmol/g. The material had a BET specific surface area S.sub.BET of 299 m.sup.2/g, a specific external surface area S.sub.EXT of 284 m.sup.2/g, both determined according to Reference Example 6, and it showed a desorption peak having a first maximum at about 180° C. and a second maximum at about 350° C. in the temperature programmed desorption of ammonia (NH.sub.3-TPD), determined according to Reference Example 7 disclosed herein. The specific surface area (S.sub.BET) and the specific external surface area (S.sub.EXT) were similar, indicating that the material was mesoporous. Thus, the mesopore volume was determined according to Reference Example 5 as being 0.80 cm.sup.3/g. Furthermore, the metal contents were determined according to Reference Example 8 as being 3.2 weight-% Fe, based on 100 weight-% SiO.sub.2.Al.sub.2O.sub.3.

Example 4: Preparation of 4,4′-methylenedianiline, 2,2′-methylenedianiline, and 2,4′-methylenedianiline by Conversion of Aniline and Formaldehyde

[0210] 155.0 mg of a catalytic material was dried overnight at 200° C. to afford 146.9 mg of a dried catalytic material. First, N,N′-diphenylmethylenediamine (also abbreviated as aminal herein) was prepared by reacting aniline with an aqueous solution of formaldehyde (36 weight-% of formaldehyde in water) in a 3:1 molar ratio of aniline to formaldehyde, for 1 h at 50° C. After phase separation, the organic layer was isolated. 2.93 g of the organic layer was added to the dried catalyst in a sealed 10 mL glass vial. The mixture was then purged with nitrogen, placed in a heated metal block at 150° C., and stirred for 24 h with a Teflon-coated stirring bar. Samples (0.2 mL each) were taken after 1 h, 5 h, and 24 h. For work-up, the catalytic material was separated from the crude solution and the latter analyzed by gas chromatography. The results are shown in table 1.

[0211] Samples were analyzed by a Shimadzu 2010 gas chromatograph equipped with a CP-Sil 5 CB column (Agilent, 100% polydimethylsiloxane, 60 m, 0.25 μm film thickness, 0.32 mm internal diameter) and a flame ionisation detector maintained at 330° C. Samples of 1 μL were injected automatically with an AOC-20s autosampler and AOC-20i auto-injector. An injection temperature of 320° C., a 1:30 split ratio, and N.sub.2 as carrier gas at a linear velocity of 24.7 cm/s were used. The initial temperature of the column was 50° C.; then it was increased at 15° C./min to 210° C. and kept for 25 min. Subsequently, the column was heated at 20° C./min to 320° C. and maintained at that temperature for 20 min. Compounds were identified with GC-MS and/or with comparing retention times of known compounds. Nitrobenzene was used as an external standard.

[0212] The isomer ratio was calculated according to the following formula:

isomer ratio=peak area of 4,4′-methylenedianiline/(peak area of 2,2′-methylenedianiline+peak area of 2,4′-methylenedianiline).

[0213] For comparative reasons, the same catalyst testing has been carried out using a conventional MCM-22 (H-MCM-22, from China Catalyst Group with Si/Al=14), a conventional zeolite Y (CBV-720, from Zeolyst International with Si/Al=15), as well as silica supporting hafnium and zinc prepared in accordance with Example 1, and silica-alumina supporting hafnium and zinc prepared in accordance with Example 2.

TABLE-US-00001 TABLE 1 Results of the catalyzed preparation of 4,4′-methylenedianiline (also abbreviated as 4,4′ or 4,4′-MDA), 2,2′-methylenedianiline (also abbreviated as 2,2′ or 2,2′-MDA) and 2,4′-methylenedianiline (also abbreviated as 2,4′ or 2,4′-MDA) using 5 weight-% of a catalytic material and a solution comprising N,N′-diphenylmethylenediamine Reaction Yield (%) Isomer ratio Catalyst time (h) Total ABA Total MDA 2,2′-MDA 2,4′-MDA 4,4′-MDA ≥3-ring 4,4′/(2,2′ + 2,4′) MCM-22 1 0.4 69.1 3.4 27.3 38.5 10.1 1.3 5 0.0 84.9 4.2 35.1 45.6 15.0 1.2 24 0.0 80.9 4.1 33.3 43.6 19.1 1.2 Hf, Zn/SiO.sub.2 1 25.4 17.3 0.0 3.3 14.0 6.2 4.2 (Example 1) 5 10.9 40.2 0.6 7.3 32.4 19.7 4.1 24 2.2 56.5 1.5 11.7 43.4 17.5 3.3 Hf, Zn/ 1 2.2 69.1 0.6 12.3 56.2 12.3 4.4 SiO.sub.2•Al.sub.2O.sub.3 5 0.8 71.5 1.4 13.8 56.2 20.6 3.7 (Example 2) 24 0.0 73.0 1.9 15.2 55.9 25.1 3.3 Zeolite Y 1 5.3 60.6 0.7 8.6 51.3 6.2 5.5 CBV-720 5 1.9 74.9 1.0 11.2 62.7 8.4 5.1 (Zeolyst) 24 0.5 77.0 1.4 13.5 62.2 14.2 4.2

[0214] As can be seen from the results shown in table 1, the conventional MCM-22 reaches the highest yield of total MDA after 24 h. However, inventive catalytic materials Hf, Zn/SiO.sub.2 (Example 1) and Hf, Zn/SiO.sub.2.Al.sub.2O.sub.3 (Example 2) both providing also good yields of total MDA reach a higher isomer ratio 4,4′/(2,2′+2,4′) than MCM-22. Zeolite Y achieved the best results in the testing with respect to the isomer ratio. However, the results shown in table 1 reflect the catalytic activity for a single cycle only.

Example 5: Recycling of a Catalytic Material

[0215] 105.4 mg of a catalytic material according to Example 1 was dried overnight at 200° C. to afford 99.2 mg of a dried catalytic material. 1.97 g of the aminal solution as prepared in Example 4 was added to the dried catalytic material in a sealed 10 mL glass vial.

[0216] For a first reaction cycle, the mixture was then purged with nitrogen, placed in a heated metal block at a temperature of 150° C., and stirred for 24 h with a Teflon-coated stirring bar. The crude product solution was separated from the catalytic material and extracted for analysis by gas chromatography as described for Example 4.

[0217] For a second reaction cycle, 1.9 mL of the aminal-aniline solution as prepared in Example 4 was added to the catalytic material retained from the first reaction cycle (wet). The mixture was then purged with nitrogen, placed in a heated metal block at 150° C., and stirred for 24 h with a Teflon-coated stirring bar. Samples (0.2 mL each) were taken after 1 h, 5 h, and 24 h and analyzed as described in Example 4. For work-up, the catalytic material was separated from the crude solution and the latter analyzed by gas chromatography as described in Example 4.

[0218] Product extraction and analysis were repeated. In total, the catalytic material obtained from the initial cycle was subjected to 4 further consecutive 24 h reaction cycles with product extraction after each cycle. The results are shown in table 2. For comparative reasons, the same procedure has been carried out similarly using a conventional zeolite Y (CBV-720, Zeolyst), a catalytic material according to Examples 1, 2 and 3 as a catalytic material, whereby in total 7 consecutive cycles were run.

TABLE-US-00002 TABLE 2 Results of the catalyzed preparation of 4,4′-methylenedianiline (also abbreviated as 4,4′ or 4,4′-MDA), 2,2′-methylenedianiline (also abbreviated as 2,2′ or 2,2′-MDA) and 2,4′-methylenedianiline (also abbreviated as 2,4′ or 2,4′-MDA) using 5 weight-% of a (recycled) catalytic material and a solution comprising N,N′-diphenylmethylenediamine Yield (%) Isomer ratio Total yield of ABA, Catalyst Cycle Total ABA Total MDA 2,2′-MDA 2,4′-MDA 4,4′-MDA ≥3-ring 4,4′/(2,2′ + 2,4′) MDA & ≥3-ring Zeolite Y 1 1.2 77.6 1.4 13.3 62.9 7.4 4.3 86.2 CBV-720 2 10.2 34.5 0.5 5.4 28.7 4.0 4.9 48.7 (Zeolyst) 3 21.9 9.6 0.0 2.0 7.6 2.2 3.8 33.7 4 24.5 7.6 0.3 1.6 5.7 1.3 3.0 33.4 5 14.4 2.8 0.0 0.8 2.0 7.6 2.4 24.8 Hf Zn/SiO.sub.2 1 2.7 61.0 0.9 11.6 48.5 17.3 3.9 81.0 (Example 1) 2 4.7 59.1 0.7 11.0 47.4 16.5 4.0 80.3 3 7.4 52.7 0.4 9.4 42.9 12.6 4.4 72.7 4 18.4 43.6 0.5 7.8 35.4 7.2 4.3 69.2 5 30.3 22.0 0.5 4.1 17.4 3.1 3.8 55.4 Hf Zn/ 1 0.0 77.8 1.9 16.2 59.6 22.2 3.3 100.0 SiO.sub.2•Al.sub.2O.sub.3 2 0.0 74.8 2.1 15.0 57.7 22.8 3.4 97.6 (Example 2) 3 0.0 76.7 2.3 15.4 59.0 23.3 3.3 100.0 4 0.0 72.8 2.2 14.5 56.1 21.4 3.4 94.2 5 0.1 75.0 2.2 14.7 58.1 21.0 3.4 96.1 6 0.2 63.2 2.1 12.7 48.4 16.8 3.3 80.2 7 0.4 64.1 2.0 12.2 49.8 17.5 3.5 82.0 Fe/ 1 0.0 72.4 1.8 15.1 55.4 18.7 3.3 91.1 SiO.sub.2•Al.sub.2O.sub.3 2 0.0 74.7 2.1 15.1 57.5 21.8 3.3 96.5 (Example 3) 3 0.1 73.5 2.2 14.8 56.5 22.0 3.3 95.7 4 0.1 72.8 2.1 14.4 56.3 19.3 3.4 92.2 5 0.2 72.6 2.0 14.2 56.4 19.0 3.5 91.8 6 0.3 62.3 1.9 12.4 48.0 15.5 3.4 78.1 7 1.1 61.1 1.8 11.3 47.9 14.2 3.6 76.4

[0219] As can be seen from the results in table 2, the yields dropped markedly after the first reaction cycle using zeolite Y whereas this was not the case for the catalytic material of the present invention. In contrast, the yield stayed almost constant for the first two cycles for the catalytic material according to Example 1, then decreased for the subsequent cycles. In its fifth recycling cycle the catalytic material achieved still more than twice as much of total MDA as zeolite Y in the third recycling cycle. A similar behavior was observed for the yield of 4,4′-MDA. Further, it can be gathered from the results shown in table 2, that Hf Zn/SiO.sub.2.Al.sub.2O.sub.3 according to Example 2 remained active for more reaction cycles compared to its all-silica counterpart according to Example 1. Though the isomer ratio was consistent over the cycles shown for Example 2, the lower isomer ratio indicates a lower selectivity towards the 4,4′-isomer. In addition to that, it can be seen from the results shown in table 2 that a catalytic material according to Example 3 wherein iron is supported on silica-alumina also shows a comparatively superior performance even after 7 consecutive cycles.

Comparative Example 6 and Examples 7 to 22: Preparation of Catalytic Materials

[0220] Commercial zeolites were used in their proton form as oxidic support. Zeolite H-Y was obtained from Zeolyst International (CBV 720, Si/Al=15), H-MCM-22 from China Catalyst Group (Si/Al=14). The silica used as a oxidic support was acquired from Cabot (CAB-O-SIL® M-5), the alumina from Condea Chemie (PURALOX® NGa-150), and the silica-alumina from Grace (MS 13/110, 13% Al.sub.2O.sub.3).

[0221] For the metal-loaded catalysts, target amounts of the metal precursor salts were first dissolved in ultrapure water at room temperature; then respective amounts of the oxidic supports were added and the mixture was left stirring until the water evaporated. The resulting powder was dried at 60° C., then calcined in air at 500° C. for 2 h at a heating rate of 2° C./min to obtain the Comparative Example and Examples 7 to 22 as presented in Table 3.

TABLE-US-00003 TABLE 3 Overview of catalytic materials according to Comparative Example 6 and Examples 7 to 18, including information on the loading and the used precursor salts. Loading [weight- % based on total weight of Oxidic Supported oxidic # support material support] Precursor salt Comp. SiO.sub.2 none none none Example 6 Example 7 SiO.sub.2 La 1.52 La(NO.sub.3)•6H.sub.2O Example 8 SiO.sub.2 Ce 1.73 Ce(NO.sub.3).sub.3•6H.sub.2O Example 9 SiO.sub.2 Sc 1.52 Sc(NO.sub.3).sub.3•xH.sub.2O Example 10 SiO.sub.2 Ta 1.32 TaCl.sub.5 (toluene used instead of water) Example 11 SiO.sub.2 Sn 1.52 Sn(CH.sub.3CO.sub.2).sub.2 Example 12 SiO.sub.2 Fe 1.32 Fe(NO.sub.3).sub.3•9H.sub.2O Example 13 SiO.sub.2 Zn 1.73 Zn(NO.sub.3).sub.2•6H.sub.2O Example 14 SiO.sub.2 Zr 1.42 ZrCl.sub.4 Example 15 SiO.sub.2 Hf 1.52 HfCl.sub.4 Example 16 SiO.sub.2 Hf 2.78 HfCl.sub.4 Example 17 SiO.sub.2 Hf, Zn 2.68, HfCl.sub.4 and 0.52 Zn(NO.sub.3).sub.2•6H.sub.2O Example 18 SiO.sub.2 Zr, Zn 1.94, ZrCl and 0.31 Zn(NO.sub.3).sub.2•6H.sub.2O Example 19 Al.sub.2O.sub.3 Hf 3.09 HfCl.sub.4 Example 20 SiO.sub.2—Al.sub.2O.sub.3 Hf 3.09 HfCl.sub.4 Example 21 Zeolite MCM-22 Hf 3.09 HfCl.sub.4 Example 22 Zeolite Y Hf 3.09 HfCl.sub.4

Example 23: Preparation of 4,4′-methylenedianiline, 2,2′-methylenedianiline, and 2,4′-methylenedianiline by Conversion of Aniline and Formaldehyde

[0222] Catalytic materials according to Comparative Example 6 and Examples 7 to 22 were dried overnight at 200° C. before use in synthesis of MDA.

[0223] An aminal solution was prepared by non-catalyzed condensation reaction between aniline and formaldehyde, and was used as starting material in the acid-catalyzed synthesis of MDA, as follows. In a 2 L two-necked round bottom flask, aniline (980 mL, Acros Organics, 99.8% purity) was stirred and heated at 50° C. Formaldehyde (274 mL, VWR Chemicals, 36% in aqueous solution, stabilized with methanol) was added dropwise until the mixture reached an aniline/formaldehyde (A/F) molar ratio of 3. The mixture was stirred for a further 1 h at 50° C. The aminal solution was collected in the organic layer and the aqueous layer was discarded by phase separation of the mixture. The NF ratio was determined by proton nuclear magnetic resonance spectroscopy (.sup.1H NMR). The NMR spectra were recorded on a Bruker AVANCE III HD 400 MHz Spectrometer and CDCl.sub.3 (Sigma-Aldrich, 99.8 atom % D) was used as the solvent.

[0224] The aminal solution was added to the dried catalytic material in a sealed vial, later purged with nitrogen. Typically, 200 mg of dried catalytic material was used in the experiments, representing 5 weight-% with respect to the aminal. The mixture was stirred and maintained at 150° C. unless stated otherwise for the required reaction time. Aliquots of ≤100 microL were taken at different time points, and the spent catalyst was separated from the crude product solution by centrifugation. The crude product was then analyzed by gas chromatography (GC). Tetrahydrofuran (Acros Organics, 99+% extra pure) was used to dilute the crude solution and nitrobenzene (Fluka, >99.5% purity) was added as an external standard. Reaction samples were analysed in a Shimadzu 2010 gas chromatograph equipped with a 60 m CP-Sil 5 CB column and an FID detector. Compounds were identified with known compounds and gas chromatography-mass spectrometry (GC-MS).

[0225] The results of the catalytic testing are presented in Table 4.

TABLE-US-00004 TABLE 4 Overview of yields and isomer ratios of the MDA synthesis products at 150° C. over catalytic materials, wherein 5 weight-% catalytic material were used and a reaction time of 24 h was applied. The yields of o-ABA and p-ABA were combined into a single ABA value (ABA), whereas the MDA yield represents the summed yields of the three MDA isomers. ABA MDA OMDA # Catalytic material [%] [%] [%] 4,4’/(2,2’ + 2,4’) Comp. Example 6 SiO.sub.2 22.8 10.0 — 3.6 Example 7 1.52 wt.-% La on SiO.sub.2 13.7 30.1 3.6 4.6 Example 8 1.73 wt.-% Ce on SiO.sub.2 16.5 25.0 — 3.8 Example 9 1.52 wt.-% Sc on SiO.sub.2 3.8 48.8 10.8 4.0 Example 10 1.32 wt.-% Ta on SiO.sub.2 17.7 21.3 — 4.2 Example 11 1.52 wt.-% Sn on SiO.sub.2 4.7 37.3 5.1 4.1 Example 12 1.32 wt.-% Fe on SiO.sub.2 6.1 55.4 10.8 4.3 Example 13 1.73 wt.-% Zn on SiO.sub.2 27.0 26.7 — 4.8 Example 14 1.42 wt.-% Zr on SiO.sub.2 2.1 53.8 12.9 3.4 Example 15 1.52 wt.-% Hf on SiO.sub.2 5.9 46.5 15.1 4.0 Example 16 2.78 wt.-% Hf on SiO.sub.2 3.1 52.7 14.6 3.4 Example 17 2.68 wt.-% Hf and 3.8 51.5 16.5 4.0 0.52 wt.-% Zn on SiO.sub.2 Example 18 1.94 wt.-% Zr and 3.6 52.3 15.4 3.0 0.31 wt.-% Zn on SiO.sub.2

[0226] As can be gathered from the results in Table 4, unloaded SiO.sub.2 only produced MDA in a 10%. Three rare-earth elements were studied, La, Ce and Sc. La.sub.1.52/SiO.sub.2 (Ex. 7) and Ce.sub.1.73/SiO.sub.2 (Ex. 8) showed a moderate activity (30% and 25% MDA yield, respectively). This activity was improved with Sc.sub.1.52/SiO.sub.2 (Ex. 9). A transition metal like Ta (Ex. 10) and a post-transition metal like Sn (Ex. 11), which are often used in other Lewis acid-catalysed transformations, were able to form MDA in modest yields (37% MDA yield with Sn.sub.1.52/SiO.sub.2, see Ex. 11). Zn.sub.1.73/SiO.sub.2 (Ex. 13) showed a comparatively better performance as catalyst (27% yield of MDA and 4,4′-MDA/(2,2′-MDA+2,4′-MDA) isomer ratio of 4.8) and further enhanced with Fe.sub.1.32/SiO.sub.2 (Ex. 12) achieving 55% yield of MDA and an isomer ratio 4,4′-MDA/(2,2′-MDA+2,4′-MDA) of 4.3. The best results were attained when other Lewis acids were employed. Both Zr.sub.1.42/SiO.sub.2 (Ex. 14) and Hf.sub.1.52;2.78/SiO.sub.2 (Ex. 15-16) were found to be outstanding catalytic materials for the synthesis of MDA, achieving high MDA and OMDA yields and a respectable 4,4′-MDA/(2,2′-MDA+2,4′-MDA) isomer ratio. Remarkably, the introduction of even a small loading of Zn (0.52 weight-% in Ex. 17, 0.31 weight-% in Ex. 18) increased both the OMDA yield and the isomer ratio with only a minor drop in MDA yield.

[0227] As can be seen from the further results in FIG. 2, an improvement was seen when supports with some Brønsted acid character such as alumina (Al.sub.2O.sub.3) and silica-alumina (SiO.sub.2—Al.sub.2O.sub.3) were employed, particularly in the case of silica-alumina, with a marked increase in the catalytic activity and higher MDA and OMDA yields obtained after just 1 h of reaction. Complete conversion was achieved as indicated by the absence of the ABA intermediates. The parent zeolites, H-MCM-22 and H-Y, were also tested as supports. For Hf.sub.3.09/MCM-22 (Ex. 21), this resulted in slightly higher ABA and OMDA contents, similar MDA yields, as well as a slightly higher 4,4′-MDA/(2,2′-MDA+2,4′-MDA) isomer ratio, compared to that obtained with the parent H-MCM-22 zeolite. As for Hf.sub.3.09/Y (Ex. 22), all the values were very similar to those obtained with its metal-free counterpart, which seems to suggest that the addition of a metal, at least in the case of Hf, did not change the catalytic behaviour of zeolite H-Y.

BRIEF DESCRIPTION OF FIGURES

[0228] FIG. 1: shows the profiles from thermogravimetric analysis (scatter plots, upper graphs) and the aniline-sample interaction strength (line plots, lower graphs) of Hf/SiO.sub.2, Hf Zn/SiO.sub.2, CBV 720 (zeolite H-Y), and SiO.sub.2 investigated by monitoring the m/z value of 93 (corresponding to aniline, C.sub.6H.sub.5NH.sub.2).

[0229] FIG. 2: shows the product yields obtained in MDA synthesis using Hf loaded on various oxidic supports as catalytic materials, where the synthesis was performed at 150° C., using 5 weight-% catalyst. The 4,4′-MDA/(2,2′-MDA+2,4′-MDA) isomer ratio is indicated by the crosses (x), the ABA yield is shown in the lower part of the respective bar, the MDA yield is shown in the middle part of the respective bar, and the OMDA yield is shown in the upper part of the respective bar, wherein the yield is given in %.

CITED LITERATURE

[0230] WO 2016/005269 A1 [0231] EP 3 263 547 A1 [0232] JP 2012-131720 A [0233] JP 2013-095724 A [0234] JP 2012-250971 A [0235] WO 2010/019844 A1 [0236] WO 2010/072504 A1 [0237] M. Haus et al. “Advanced kinetic models through mechanistic understanding: Population balances for methylenedianiline synthesis” in Chemical Engineering Science 2017, vol. 167, p. 317 [0238] D. Jin et al. “Dramatic change of methylenedianiline activity and selectivity in different pore geometry of zeolites” in Microporous and Mesoporous Materials 2016, vol. 233, p. 109-116 [0239] C. A. Emeis “Determination of integrated molar extinction coefficients for infrared absorption bands of pyridine adsorbed on solid acid catalysts” in Journal of Catalysis 1993, vol. 141, p. 347-354 [0240] M. Velthoen et al. “Probing acid sites in solid catalysts with pyridine UV-Vis spectroscopy” in Physical Chemistry Chemical Physics 2018, vol. 20, p. 21647-21659