Unsymmetric bisphosphite

Abstract

An unsymmetric bisphosphite of the formula (1) ##STR00001##
a process for preparation thereof, a reaction thereof with metals to give mixtures containing complexes of the unsymmetric bisphosphite and the metal, and a use thereof as a catalytically active composition in hydroformylation reactions, where the hydroformylation-active composition contains, as well as the complex of metal and unsymmetric bisphosphite, unbound bisphosphite and at least one further component.

Claims

1. A compound of formula (1): ##STR00021##

2. A process for preparing a compound of claim 1, comprising: i) oxidatively coupling 2,4-dimethylphenol to obtain 3,3,5,5-tetramethyl-2,2-dihydroxybiphenyl; ii) oxidatively coupling 3-tert-butyl-4-hydroxyanisole to obtain 5,5-dimethoxy-3,3-di-tert-butyl-2,2-dihydroxybiphenyl; iii) reacting 3,3,5,5-tetramethyl-2,2-dihydroxybiphenyl from i) with PCl.sub.3 to obtain a phosphorochloridite derivative under inert gas atmosphere; iv) reacting at least 2 equivalents of the phosphorochloridite derivative from iii) with 1 equivalent of the 5,5-dimethoxy-3,3-di-tert-butyl-2,2-dihydroxybiphenyl from ii) under inert gas atmosphere.

3. The process according to claim 2, wherein a solvent mixture is present during the reacting iv).

4. The process according to claim 3, wherein the solvent mixture comprises at least two solvents selected from the group consisting of an organic nitrogen compound, an organic ester, and an aromatic compound.

5. The process according to claim 4, wherein the organic nitrogen compound is at least one compound selected from the group consisting of a nitrile, an amine, and an amide.

6. The process according to claim 2, further comprising v), removing the compound (1) in solid form and suspending the compound (1) in an aprotic solvent mixture, recrystallizing the compound (1) in an aprotic solvent mixture, or suspending and recrystallizing the compound in an aprotic solvent mixture.

7. A compound of formula (2): ##STR00022## wherein M is selected from the group consisting of Fe, Ru, Os, Co, Rh, Ir, Ni, Pd, Pt, and M may enter into additional bonds.

8. A compound of the formula (3): ##STR00023## wherein M is selected from the group consisting of Fe, Ru, Os, Co, Rh, Ir, Ni, Pd, and Pt.

9. The compound according to claim 8, wherein M is Rh.

10. A mixture comprising a compound of formula (2) and/or (3), and a compound of formula (1) not bonded to M.

11. A composition comprising a mixture of claim 10, and a further component selected from the group consisting of bases, organic amines, buffer solutions, epoxides, and ion exchangers.

12. The composition according to claim 11, wherein the organic amine has an 2,2,6,6-tetramethylpiperidine unit.

13. The composition according to claim 11, wherein the further component is a di-4-(2,2,6,6-tetramethylpiperidinyl) sebacate.

14. A process for hydroformylating an unsaturated compound or a mixture of unsaturated compounds, comprising: a) initially charging a compound of claim 1, b) introducing a gas mixture comprising carbon monoxide and hydrogen; and c) adding an unsaturated compound or a mixture of unsaturated compounds.

15. The process according to claim 14, wherein the unsaturated compound or mixture thereof are selected from the group consisting of: hydrocarbon mixtures from steamcracking plants; hydrocarbon mixtures from catalytically operated cracking plants; hydrocarbon mixtures from oligomerization operations; hydrocarbon mixtures comprising polyunsaturated compounds; and unsaturated carboxylic acid derivatives.

16. The process according to claim 14, wherein the mixture comprises unsaturated compounds having 2 to 30 carbon atoms.

17. The process according to claim 14, wherein the mixture comprises unsaturated compounds having 2 to 8 carbon atoms.

18. The process according to claim 15, wherein the unsaturated carboxylic acid derivatives are selected from fatty acid esters.

Description

BRIEF DESCRIPTION OF SEVERAL VIEWS OF THE DRAWING

(1) FIG. 1 shows the hydridocarbonyl complex of the ligand (1), with rhodium as the metal, of the inventive compound (3).

(2) FIG. 2 shows all the coordinates, distances and angles calculated for the compound (3).

(3) FIG. 3 shows hydroformylation experiments in a continuously operated experiment system.

(4) FIG. 4 shows hydroformylation experiments in a continuously operated experiment system.

DESCRIPTION OF FIGURES

Calculation of the Complex (3)

(5) The inventive complexes of the formulae (2) and (3) are formed in situ during the hydroformylation reaction.

(6) In a particular embodiment of the invention, the complexes (2) and (3) are present alongside the unbound bisphosphite (1).

(7) The hydridocarbonyl complex of the ligand (1), with rhodium as the metal, of the inventive compound (3) was characterized by means of theoretical calculations. The result is shown in FIG. 1 in the appendix.

(8) The structure calculation was conducted with the BP86 functional and the def-SV(P) base set.

(9) The structure calculations for the model structures were effected with the Turbomole program package (R. Ahlrichs, M. Br, M. Haser, H. Horn, C. Klmel, Chem. Phys. Lett., 1989, 162, 16; TURBOMOLE V6.3 2011, a development of University of Karlsruhe and Forschungszentrum Karlsruhe GmbH, 1989-2007, TURBOMOLE GmbH, since 2007. http://www.turbomole.com) on the basis of density functional theory (DFT). The BP86 functional (S. H. Vosko, L. Wilk, M. Nusair, Can. J. Phys., 1980, 58, 1200; A. D. Becke, Phys. Rev. A, 1988, 38, 3098; J. Perdew, Phys. Rev. B, 1986, 33, 8822) and the def-SV(P) base set (A. Schfer, H. Horn and R. Ahlrichs, J. Chem. Phys., 1992, 97, 2571) were used.

(10) FIG. 2 in the appendix shows all the coordinates, distances and angles calculated for the compound (3).

(11) FIGS. 3 and 4 show hydroformylation experiments in a continuously operated experiment system. In a first experiment series (FIG. 3), the inventive compound (1) was tested. Under the selected reaction conditions, an aldehyde yield between 80% and 90% was established. It was possible to keep this state constant up to the end of the experiment. The percentage distribution between n-pentanal and 2-methylbutanal, i.e. the regioselectivity, was 92% to 8%.

(12) In the second experiment series (FIG. 4), the comparative compound biphephos was used under the same experimental conditions. Under the reaction conditions selected, the aldehyde yield fell from initially 70% to 80% after 150 h to 40% to 50%. The percentage distribution between n-pentanal and 2-methylbutanal, i.e. the regioselectivity, was 95% to 5%. In other words, this ligand features a much lower long-term stability than the inventive ligand (1). However, this is crucial for an industrial scale operation, since it strongly influences the economic viability of such an operation.

(13) Consequently, the inventive ligand features a distinct improvement in stability and activity. This result is surprising, since unsymmetric substituted bisphosphites show distinct losses of activity and selectivity compared to symmetrically substituted examples, as already stated in Rhodium-catalyzed Hydroformylation, ed. by P. W. N. M. van Leeuwen and C. Claver, Kluwer Academic Publishers 2006, A A Dordrecht, NL, pages 45-46.

(14) The inventive ligand (1) in the catalytically active composition features much better long-term stability than the ligands described in the prior art to date and thus achieves the stated object. An optimal long-term stability of the catalytically active composition is of significance especially in industrial scale use, since the ligand in the hydroformylation reaction can be replenished on the industrial scale, but any replenishment adversely affects the economic viability of an industrial scale process and may make it unfeasible.

EXAMPLES

General Reaction Equation

(15) ##STR00013##

ABBREVIATIONS

(16) DM water=demineralized water

(17) CPG=core-pulled precision glass

(18) ACN=acetonitrile

(19) EtOAc=ethyl acetate

(20) acac=acetylacetonate

(21) NEt.sub.3=triethylamine

(22) TIPB=1,2,4,5-tetraisopropylbenzene

Synthesis of 2,2-bis(3,5-dimethylphenol) (5)

(23) The biphenol (5) used as a precursor was prepared by the synthesis method which follows.

(24) ##STR00014##

(25) A 500 ml Schlenk with CPG stirrer, intermediate section and glass stirrer was initially charged with 1.42 g (0.005 mol) of iron(II) sulphate heptahydrate and 12.35 g (0.1 mol) of 2,4-dimethylphenol in 150 ml of DM water and 5 ml of cyclohexane, and the mixture was heated to 40 C.

(26) In a 100 ml beaker, 25.36 g (0.146 mol) of sodium peroxodisulphate were dissolved in 80 ml of DM water. At the start of the reaction, a small portion of Na.sub.2S.sub.2O.sub.8 solution was added to the phenol. Subsequently, a smaller portion of the solution was added every 10 min. After 30 min, the addition of Na.sub.2S.sub.2O.sub.8 solution had ended.

(27) After a reaction time of 5 h, 300 ml of cyclohexane and 200 ml of water were added to the reaction solution, which was left to stir for 20 min, then transferred while warm into a separating funnel.

(28) The organic phase was removed and concentrated to dryness. The product (5) was obtained in 69% yield (10.6 g).

(29) All the preparations which follow were conducted with standard Schlenk technology under protective gas. The solvents were dried over suitable desiccants before use (Purification of Laboratory Chemicals, W. L. F. Armarego (Author), Christina Chai (Author), Butterworth Heinemann (Elsevier), 6th edition, Oxford 2009).

(30) The products were characterized by means of NMR spectroscopy. Chemical shifts () are reported in ppm. The .sup.31P NMR signals were referenced according to: SR.sub.31P=SR.sub.1H*BF.sub.31P/BF.sub.1H)=SR.sub.1H*0.4048. (Robin K. Harris, Edwin D. Becker, Sonia M. Cabral de Menezes, Robin Goodfellow, and Pierre Granger, Pure Appl. Chem., 2001, 73, 1795-1818; Robin K. Harris, Edwin D. Becker, Sonia M. Cabral de Menezes, Pierre Granger, Roy E. Hoffman and Kurt W. Zilm, Pure Appl. Chem., 2008, 80, 59-84). By means of .sup.31P NMR, the content of the ligand (1) was determined, with characterization of this unsymmetric ligand by two phosphorus signals.

Synthesis of 2,2-bis(3,5-dimethylphenol) chlorophosphite (6)

(31) ##STR00015##

(32) A secured 2 l Schlenk with magnetic stirrer was initially charged with 440 ml of phosphorus trichloride. 120 g of 2,2-bis(3,5-dimethylphenol) were weighed into a second secured 1 l Schlenk and 500 ml of dried toluene were added while stirring. The biphenol-toluene suspension was metered into the phosphorus trichloride at 63 C. within 4 h. On completion of addition, the reaction mixture was stirred at temperature overnight. The next morning, the solution was concentrated while warm (45 C.), and the product was obtainable in 96.5% yield (153 g). .sup.31P NMR: 175.59 (94.8% 2,2-bis(3,5-dimethylphenol) chlorophosphite), 4.4% various PCl compounds, 0.8% PH compound.

(33) Inventive Synthesis Variations for Preparation of the Pure Ligand (1)

(34) ##STR00016##
Variant 1: ACN/NEt.sub.3

(35) In a 1000 ml Schlenk, under protective gas, 38.75 g (0.121 mol) of 2,2-bis(3,5-dimethylphenol) chlorophosphite were dissolved in 150 ml of degassed ACN and heated to 35 C. In a second Schlenk (500 ml), 20.1 g (0.056 mol) of 3,3-di-tert-butyl-5,5-dimethoxy-[1,1-biphenyl]-2,2-diol were dissolved in 150 ml of degassed ACN, and 40.9 ml of degassed triethylamine (0.29 mol) were added while stirring. Then the biphenol/triethylamine solution was slowly added dropwise to the chlorophosphite solution. After a further reaction time of 1 h, the reaction solution was stirred at 45 C. overnight.

(36) These solids were suspended in degassed ACN at 75 C. for 1.5 h and removed and washed with warm ACN. Subsequently, the product was suspended in dried toluene at 35 C. for 1.5 h and washed. The target product (1) was obtained as a white solid (33 g, 66%). .sup.31P NMR (202.4 MHz, toluene-d.sub.8): 142.5 and 140.9 (100%).

(37) Variant 2: EtOAc/NEt.sub.3

(38) In a 100 ml Schlenk, under protective gas, 7.3 g (21.0 mmol) of 2,2-bis(3,5-dimethylphenol) chlorophosphite were dissolved in 15 ml of degassed ethyl acetate and heated to 35 C. In a second Schlenk (100 ml), 3.9 g (9.5 mmol) of 3,3-di-tert-butyl-5,5-dimethoxy-[1,1-biphenyl]-2,2-diol were dissolved in 7.0 ml of NEt.sub.3. Subsequently, the biphenol/triethylamine solution was slowly added dropwise to the chlorophosphite solution within 20 minutes. The solution was stirred at 35 C. for a further hour and then at 45 C. overnight.

(39) These solids were suspended in degassed ACN at 75 C. for 1.5 h and removed and washed with warm ACN. Subsequently, the product was suspended in dried toluene at 35 C. for 1.5 h and removed.

(40) The target product (1) was obtained as a white solid (5.0 g, 58%). .sup.31P NMR (202.4 MHz, toluene-d.sub.8): 142.5 and 140.9 (100%).

(41) Variant 3: EtOAc/Pyridine

(42) In a 250 ml Schlenk, under protective gas, 10.07 g (31.0 mmol) of 2,2-bis(3,5-dimethylphenol) chlorophosphite were dissolved in 20 ml of degassed ethyl acetate and heated to 45 C. In a second Schlenk (50 ml), 5.54 g (15 mmol) of 3,3-di-tert-butyl-5,5-dimethoxy-[1,1-biphenyl]-2,2-diol were dissolved in 26 ml of ethyl acetate and 5.2 ml of degassed pyridine. Subsequently, the biphenol/pyridine solution was slowly added dropwise to the chlorophosphite solution within 30 minutes. The solution was stirred at 45 C. overnight.

(43) The next day, the solution was filtered and the solids were washed with ACN. The target product was obtainable as a white solid (4.2 g, 31%). .sup.31P NMR (202.4 MHz, toluene-d.sub.8): 142.2 and 141.1 (100%).

(44) Variant 4: Performance of a Low-Temperature Experiment at 20 C.

(45) In a 250 ml Schlenk, under protective gas, 8.0 g (0.025 mol) of 2,2-bis(3,5-dimethylphenol) chlorophosphite were dissolved in 30 ml of degassed ACN and cooled to 20 C. In a second Schlenk (100 ml), 4.32 g (0.012 mol) of 3,3-di-tert-butyl-5,5-dimethoxy-[1,1-biphenyl]-2,2-diol were dissolved in 30 ml of degassed ACN, and 8.5 ml of degassed triethylamine were added while stirring. Then the biphenol/triethylamine solution was slowly added dropwise at 20 C. to the chlorophosphite solution. On completion of addition, stirring was continued at 20 C. for a further 4 hours. The reaction solution was stirred overnight at 10 C. until the next day. This procedure, reaction temperature at 20 C. through the day and at 10 C. overnight, was conducted repeatedly for 3 days. Thereafter, the reaction mixture was brought to RT within 3 hours.

(46) Subsequently, the solution was filtered and the solids were washed with cold ACN. The target product was obtainable as a white solid (7.6 g, 70%). .sup.31P NMR (202.4 MHz, toluene-d.sub.8): 142.5 and 140.9 (100%).

(47) The unsymmetric bisphosphite (1) was thus obtained, completely surprisingly and contrary to the prior art, in good yields and excellent purity even at low temperatures.

(48) Purification of the Ligand (1):

(49) As well as the suspending of the ligand in various solvents (see example above), it is also possible to purify the ligand by means of recrystallization. This recrystallization was effected to WO 2012095255. Rather than o-xylene, it is also possible to use toluene for recrystallization in an analogous manner.

(50) Procedure for the Hydroformylation Experiments

(51) Experiment DescriptionGeneral

(52) The experiments were conducted in 100 ml autoclaves from Parr Instruments. The autoclaves are equipped with an electric heater. The pressure is kept constant by means of mass flow meters and pressure regulators. During the experiment duration, a syringe pump can be used to inject an exactly defined amount of reactant under reaction conditions. Capillary lines and HPLC valves can be used to take samples during the experiment duration, and these can be analysed both by means of GC analysis and by means of LC-MS analysis.

(53) Experiment DescriptionExtended Experiment

(54) The Rh precursor (Rh(acac)(CO).sub.2) (acac=acetylacetonate) and the ligand are initially charged in 40 ml of isononyl benzoate in an autoclave. The Rh concentration is 100 ppm based on the overall reaction mixture used. The ligand excess used is 4:1 in molar terms, based on rhodium.

(55) As a further component in a ratio of 2:1 to the ligand, compound (Ib) is added as the amine. As a GC standard, 0.5 g of 1,2,4,5-tetraisopropylbenzene is added.

(56) Reaction temperature is 120 C. The reaction pressure is 20 bar of synthesis gas (H.sub.2:CO=50:50% by volume).

(57) As the olefin, 4 ml of cis-2-butene each time were metered in with the syringe pump at intervals of about 1 day. GC samples were taken after 1, 2, 4 hours and before the next metered addition.

(58) The following ligands were studied with regard to their stability:

(59) ##STR00017##
ResultsExtended Experiments

(60) The relative activities are determined by the ratio of 1st order k to k0, i.e. the k value at time 0 in the reaction (start of reaction), and describe the relative decrease in activity during the experiment duration.

(61) The 1st order k values are obtained from a plot of (ln(1-conversion)) against time.

(62) TABLE-US-00002 TABLE 2 Metered Ser. addition at 1st order k k/k0 n/i No. Ligand Run time (h) (min{circumflex over ()}1) Rel. activity selectivity 1 biphephos 0 1.39E02 1 21 2 biphephos 20.5 4.45E03 0.32 21 3 biphephos 44.3 2.91E03 0.209 20 4 biphephos 66.6 1.72E03 0.124 20 5* ligand (1) 0 7.74E03 1 17 6* ligand (1) 20.8 5.10E03 0.659 16 7* ligand (1) 44.8 3.19E03 0.412 15 8* ligand (1) 117.8 2.99E03 0.386 14 9 ligand (8) 0 1.72E02 1 14 10 ligand (8) 22.4 9.00E03 0.523 13 11 ligand (8) 44.7 5.39E03 0.313 13 12 ligand (8) 68.3 3.31E03 0.192 13 *inventive

(63) Result: The decline in catalyst activity with the biphephos ligand and ligand (8) is (Table 2; entries 1-4, 9-12) much more marked than with the ligand (1). It is remarkable that the relative activity of the ligand (1) after nearly twice the reaction time (Table 2; entry 8) is still more than twice as high as for the other two ligands after half the reaction time (Table 2; entries 4 and 12), still with very good n/i selectivities. This behaviour is confirmed in the extended experiments in the continuously operated hydroformylation plant (see FIGS. 3 and 4).

(64) It was thus possible to prepare an unsymmetric ligand and use it in a hydroformylation-active composition, and this ligand, completely surprisingly and contrary to the prior art, has very good properties and achieves the technical object.

Inventive Results

Substrate Variation

Example 1

(65) In a 100 ml autoclave from Parr Instruments, 5.3 g of propene were hydroformylated at 120 C. and 30 bar. As the precursor, 0.0054 g of Rh(acac)(CO).sub.2 was initially charged in 43.89 g of toluene. As the ligand, 0.0701 g of ligand (1) was used in the catalyst mixture solution. 0.0372 g of the compound (Ib) was added as the organic amine, and 0.5016 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged. During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours. 89.6 mol % butanal, 7.9 mol % 2-methylpropanal and 2.3 mol % propane were formed. The regioselectivity for n-butanal is 92.0%.

Example 2

(66) In a 100 ml autoclave from Parr Instruments, 5.6 g of cis-2-butene were hydroformylated at 120 C. and 20 bar. As the precursor, 0.0056 g of Rh(acac)(CO).sub.2 was initially charged in 48.8 g of toluene. As the ligand, 0.0779 g of ligand (1) was used in the catalyst mixture solution. 0.0416 g of the compound (Ib) was added as the organic amine, and 0.5760 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged.

(67) During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours. 80.0 mol % pentanal, 5.2 mol % 2-methylbutanal and 3.7 mol % n-butane were formed. The regioselectivity for n-pentanal is 94.0%.

Example 3

(68) In a 100 ml autoclave from Parr Instruments, 6.3 g of isobutene were hydroformylated at 120 C. and 20 bar. As the precursor, 0.0046 g of Rh(acac)(CO).sub.2 was initially charged in 39.8 g of toluene. As the ligand, 0.0636 g of ligand (1) was used in the catalyst mixture solution. 0.0339 g of the compound (Ib) was added as the organic amine, and 0.4701 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged. During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours. 72.9 mol % 3-methylbutanal, 0.1 mol % pivalaldehyde and 4.4 mol % isobutane were formed.

Example 4

(69) In a 100 ml autoclave from Parr Instruments, 6.7 g of a C-4 mixture having the following composition: 2.9 mol % isobutane, 9.9 mol % n-butane, 28.7 mol % 1-butene, 43.5 mol % isobutene, 14.6 mol % 2-butenes and 0.2 mol % 1,3-butadiene were hydroformylated at 120 C. and 20 bar. As the precursor, 0.0049 g of Rh(acac)(CO).sub.2 was initially charged in 42.38 g of toluene. As the ligand, 0.0697 g of ligand (1) was used in the catalyst mixture solution. 0.0374 g of the compound (Ib) was added as the organic amine, and 0.5069 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged. During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours. The output comprises 32.86% 3-methylbutanal (isobutene conversion 75.6 mol %), 39.0 mol % n-pentanal and 1.8 mol % 2-methylbutanal (butenes conversion 76.5 mol %, regioselectivity for n-pentanal 95.6%). As hydrogenation products, 4.7 mol % isobutane and 11.3 mol % n-butane were found in the output.

Example 5

(70) In a 100 ml autoclave from Parr Instruments, 6.5 g of a C-4 mixture having the following composition: 5.9 mol % isobutane, 15.6 mol % n-butane, 52.9 mol % 1-butene, 0.1 mol % isobutene, 24.8 mol % 2-butenes and 0.5 mol % 1,3-butadiene were hydroformylated at 120 C. and 20 bar. As the precursor, 0.0052 g of Rh(acac)(CO).sub.2 was initially charged in 45.05 g of toluene. As the ligand, 0.0727 g of ligand (1) was used in the catalyst mixture solution. 0.0377 g of the compound (Ib) was added as the organic amine, and 0.5314 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged. During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours. The output comprises 0.14 mol % 3-methylbutanal, 69.5 mol % n-pentanal and 3.67 mol % 2-methylbutanal (butenes conversion 94.2 mol %, regioselectivity for n-pentanal 96.5%). As hydrogenation products, 5.64 mol % isobutane and 18.55 mol % n-butane were found in the output.

Example 6

(71) In a 100 ml autoclave from Parr Instruments, 7.0 g of a C-4 mixture having the following composition: the reactant comprises 5.9 mol % isobutane, 22.1 mol % n-butane, 45.5 mol % 1-butene, 2.1 mol % isobutene, 17.1 mol % 2-butenes and 0.2 mol % 1,3-butadiene were hydroformylated at 120 C. and 20 bar. As the precursor, 0.0047 g of Rh(acac)(CO).sub.2 was initially charged in 40.81 g of toluene. As the ligand, 0.0659 g of ligand (1) was used in the catalyst mixture solution. 0.0342 g of the compound (Ib) was added as the organic amine, and 0.4814 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged. During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours. The output comprises 1.5 mol % 3-methylbutanal (isobutene conversion 71.6 mol %), 61.9 mol % n-pentanal and 2.9 mol % 2-methylbutanal (butenes conversion 93.3 mol %, regioselectivity for n-pentanal 95.5%). As hydrogenation products, 5.3 mol % isobutane and 23.4 mol % n-butane were found in the output.

Example 7

(72) In a 100 ml autoclave from Parr Instruments, 7.1 g of a C-4 mixture having the following composition: 3.5 mol % isobutane, 13.0 mol % n-butane, 47.3 mol % 1-butene, 13.9 mol % isobutene, 21.6 mol % 2-butenes and 0.4 mol % 1,3-butadiene were hydroformylated at 120 C. and 20 bar. As the precursor, 0.0048 g of Rh(acac)(CO).sub.2 was initially charged in 43.88 g of toluene. As the ligand, 0.0680 g of ligand (1) was used in the catalyst mixture solution. 0.0363 g of the compound (Ib) was added as the organic amine, and 0.5092 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged. During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours. The output comprises 10.1 mol % 3-methylbutanal (isobutene conversion 72.8 mol %), 63.2 mol % n-pentanal and 3.2 mol % 2-methylbutanal (butenes conversion 96.3 mol %, regioselectivity for n-pentanal 95.2%). As hydrogenation products, 3.5 mol % isobutane and 15.1 mol % n-butane were found in the output.

Example 8

(73) In a 100 ml autoclave from Parr Instruments, 5.8 g of a C-4 mixture having the following composition: 0.1 mol % isobutane, 27.6 mol % n-butane, 27.9 mol % 1-butene, 0.1 mol % isobutene and 44.0 mol % 2-butenes were hydroformylated at 120 C. and 20 bar. As the precursor, 0.0051 g of Rh(acac)(CO).sub.2 was initially charged in 43.77 g of toluene. As the ligand, 0.0699 g of ligand (1) was used in the catalyst mixture solution. 0.0373 g of the compound (Ib) was added as the organic amine, and 0.5166 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged. During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours. The output comprises 59.9 mol % n-pentanal and 3.3 mol % 2-methylbutanal (butenes conversion 91.7 mol %, regioselectivity for n-pentanal 94.7%). As hydrogenation products, 0.1 mol % isobutane and 31.7 mol % n-butane were found in the output.

Example 9

(74) In a 100 ml autoclave from Parr Instruments, 6.0 g of a C-4 mixture having the following composition: 63.6 mol % n-butane, 1.0 mol % 1-butene and 35.8 mol % 2-butenes were hydroformylated at 120 C. and 20 bar. As the precursor, 0.0041 g of Rh(acac)(CO).sub.2 was initially charged in 35.88 g of toluene. As the ligand, 0.0573 g of ligand (1) was used in the catalyst mixture solution. 0.0306 g of the compound (Ib) was added as the organic amine, and 0.4235 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged. During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours. The output comprises 29.7 mol % n-pentanal and 1.9 mol % 2-methylbutanal (butenes conversion 85.3 mol %, regioselectivity for n-pentanal 94.0%).

Example 10

(75) In a 100 ml autoclave from Parr Instruments, 5.0 g of n-octene were hydroformylated at 120 C. and 20 bar. As the precursor, 0.0049 g of Rh(acac)(CO).sub.2 was initially charged in 41.29 g of toluene. As the ligand, 0.0669 g of ligand (1) was used in the catalyst mixture solution. 0.0378 g of the compound (Ib) was added as the organic amine, and 0.5030 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged. During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours. The output comprises 54.2 mol % aldehydes (regioselectivity for n-nonanal 90.9%). As hydrogenation products, 3.9 mol % n-octane and 3.2% nonanol were found in the output.

Example 11

(76) In a 100 ml autoclave from Parr Instruments, 7.0 g of 1,3-butadiene were hydroformylated at 120 C. and 20 bar. As the precursor, 0.0054 g of Rh(acac)(CO).sub.2 was initially charged in 46.82 g of toluene. As the ligand, 0.0770 g of ligand (1) was used in the catalyst mixture solution. 0.0413 g of the compound (Ib) was added as the organic amine, and 0.5599 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged. During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours. The output comprises 0.2 mol % n-butane, 11.3% n-butenes, 12.9% aldehydes and 11.5 mol % 4-vinylcyclohexene. The total conversion of 1,3-butadiene is 37.2%.

Example 12

(77) In a 100 ml autoclave from Parr Instruments, 5.6 g of methyl oleate were hydroformylated at 120 C. and 20 bar. As the precursor, 0.0052 g of Rh(acac)(CO).sub.2 was initially charged in 44.06 g of toluene. As the ligand, 0.0689 g of ligand (1) was used in the catalyst mixture solution. 0.0375 g of the compound (Ib) was added as the organic amine, and 0.5260 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged. During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours. From .sup.1H and .sup.13C NMR spectra, an aldehyde yield of 49.5 mol % was calculated. The regioselectivity for terminal aldehydes is 20.6 mol %. The double bond content is 35.9 mol %.

Example 13

(78) In a 100 ml autoclave from Parr Instruments, 6.9 g of a hydrocarbon mixture from catalytically operated cracking plants having the following composition: 1.5 mol % propane, 0.8 mol % propene, 28.1 mol % isobutane, 8.1 mol % n-butane, 16.4 mol % 1-butene, 16.9 mol % isobutene, 28.2 mol % 2-butenes, 0.5 mol % 1,3-butadiene and fractions of C5 olefins and hydrocarbons were hydroformylated at 120 C. and 20 bar. As the precursor, 0.0048 g of Rh(acac)(CO).sub.2 was initially charged in 43.39 g of toluene. As the ligand, 0.0672 g of ligand (1) was used in the catalyst mixture solution. 0.0359 g of the compound (Ib) was added as the organic amine, and 0.5035 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged. During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours.

(79) The output comprises 1.3 mol % propane, 0.7 mol % butanal, 27.5 mol % isobutane, 9.6 mol % n-butane, 13.1 mol % 3-methylbutanal (77.4% isobutene conversion), 39.1 mol % pentanal, 2.1 mol % 2-methylbutanal (n-butenes conversion 96.9%, regioselectivity for n-pentanal 95.0%).

Example 14

(80) In a 100 ml autoclave from Parr Instruments, 1.8 g of ethene were hydroformylated at 120 C. and 50 bar. As the precursor, 0.0050 g of Rh(acac)(CO).sub.2 was initially charged in 42.68 g of toluene. As the ligand, 0.0668 g of ligand (1) was used in the catalyst mixture solution. 0.0363 g of the compound (Ib) was added as the organic amine, and 0.5095 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged. During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 20 hours. The conversion to propanal is 98.7%.

Comparative Experiment

Unsymmetric and Symmetric Ligands

(81) As well as the testing of the inventive unsymmetric ligand (1) with various substrates, symmetric ligands and the corresponding unsymmetric isomer thereof were additionally tested under comparable conditions.

(82) First of all, the symmetric biphephos ligand already mentioned in the prior art and the unsymmetric isomer thereof (9) were tested. The compound (9) was prepared analogously to the synthesis method in WO 95/28228 on page 19.

(83) ##STR00018##

(84) The experiments were conducted by the following method:

Example 15

(85) In a 100 ml autoclave from Parr Instruments, 5.7 g of cis-2-butene were hydroformylated at 120 C. and 20 bar. As the precursor, 0.0054 g of Rh(acac)(CO).sub.2 was initially charged in 51.5 g of Diphyl (mixture of about 73.5% diphenyl oxide and 26.5% diphenyl). As the ligand, 0.0779 g of the appropriate ligand in the catalyst mixture solution was used. 0.0416 g of the compound (Ib) was added as the organic amine, and 0.5760 g of TIPB as the GC standard. The reactant was metered in after attainment of the reaction temperature envisaged.

(86) During the reaction, the pressure was kept constant via synthesis gas regulation with a mass flow meter. Samples were taken from the reaction mixture after 12 hours.

(87) The results are shown in Table 3.

(88) TABLE-US-00003 TABLE 3 Regioselectivity Entry Ligand Yield of aldehydes n-pentanal in % 1 biphephos 95.0 94.5 2 (9) 66.6 78.5

(89) The unsymmetric isomer (ligand 9; entry 2) of the symmetric biphephos thus features a much lower activity and a much poorer selectivity than the symmetric biphephos ligand. This corresponds to the prior art. The use of the two ligands, i.e. of the symmetric biphephos ligand and the unsymmetric isomer thereof, has already been described in Rhodium-catalyzed Hydroformylation, ed. by P. W. N. M. van Leeuwen and C. Claver, Kluwer Academic Publishers 2006, A A Dordrecht, NL. On pages 45-46, Table 2, the hydroformylation results for the two ligands under comparable conditions are shown. In this context, it is clearly apparent that the symmetric biphephos ligand (in the reference ligand 5a) features a much higher n/i selectivity and a higher activity than its unsymmetric isomer (in the reference ligand 7). In the hydroformylation reaction of propene, the symmetric ligand has an n/i selectivity of 53 and a reaction rate of 402, whereas the unsymmetric ligand has only an n/i selectivity of 1.2 and a reaction rate of 280. This was confirmed once again by our own results in Table 3.

(90) In addition, the inventive ligand (1) and the symmetric isomer thereof (10) were tested under comparable conditions. The experiments which follow were conducted by the method in Example 2. Only the ligands have been changed. Table 4 shows the results of hydroformylation of cis-2-butene with the inventive ligand (1) and the symmetric isomer thereof, ligand (9).

(91) ##STR00019##

(92) TABLE-US-00004 TABLE 4 mol % mol % Regioselectivity Entry Ligand pentanal 2-methylbutanal n-pentanal in % 1 (1) 80.0 5.2 94.0 2 (10) 59.9 6.35 90.4

(93) The inventive unsymmetric ligand (1) (entry 1) has very good n-pentanal regioselectivity of 94% and good aldehyde yields. The symmetric isomer thereof (entry 2), in contrast, has poorer pentanal selectivities of only 90% and much poorer activities, i.e. yields.

(94) This result is surprising since unsymmetric substituted bisphosphites show distinct losses in activity and selectivity compared to symmetrically substituted examples, as described in the prior art and confirmed by the above comparative experiments with the biphephos ligand and the unsymmetric isomer thereof (9). Thus, the inventive unsymmetric ligand (1), in complete contrast to the prior art, features very good selectivities and activities. In addition, the inventive ligand (1) is a ligand of very prolonged stability, as shown in the extended experiment in a continuously operated plant which follows.

Comparative Experiment

Extended Experiment

(95) In a first experiment series, the inventive compound (1) was tested. In the second experiment series, the comparative compound biphephos was used under the same experimental conditions.

(96) ##STR00020##

(97) A hydroformylation of butene/butane mixtures was performed in a continuously operated experiment system.

(98) This experiment system consisted essentially of a pressure reactor of capacity 20 liters with a downstream condenser and phase separation vessel (gas/liquid) for the gas phase originating from the reactor, and a cycle gas compressor which returns the gas phase from the phase separation vessel back down into the reaction zone. A portion of this cycle gas is run out of the reaction system as offgas after the phase separation.

(99) In order to achieve optimal gas distribution in the reactor system, a gas distributor ring with bores was installed here.

(100) By means of installed heating and cooling apparatuses, the temperature of the reactor could be controlled.

(101) Prior to the hydroformylation, the reactor system was purged with nitrogen to free it of oxygen. Subsequently, the reactor was charged with 12 liters of catalyst solution. This catalyst solution was composed of 12 kg of isononyl benzoate, 4.5 g of Rh(acac)(CO).sub.2, 63 g of bisphosphite ligand (1), 200 g of amine IIb, and was mixed beforehand in a vessel. The isononyl benzoate was stripped beforehand with nitrogen, in order to remove oxygen and water from the solvent.

(102) Subsequently, the reactor system was purged with synthesis gas to free it of nitrogen. Once the nitrogen content had fallen <10% by volume, the reactor system was pressurized to 1.0 MPa with synthesis gas and then heated to 120 C.

(103) On attainment of the operating temperature, the reactor system was brought to reaction pressure 1.7 MPa with synthesis gas.

(104) Thereafter, the addition of the starting materials was commenced. The crude butane was run through a vaporizer in order to run the crude butane into the cycle gas in gaseous form.

(105) The following throughputs were set:

(106) 0.3 kg/h of crude butane (a mixture of 35% 2-butenes and n-butane and 1-butene concentrations of about 1%), 75 l (STP)/h of synthesis gas (50% by vol. of H2 and 50% by vol. of CO).

(107) For the daily metered addition of the compound (1) and amine IIb, a 1.4% solution of the bisphosphite ligand I in n-pentanal, which had been freed of residual C4 hydrocarbons (<3%) beforehand by stripping with nitrogen, was made up. The amine IIb was used in a threefold molar excess relative to the compound (1). For better stabilization of this solution, the amine IIb was added to the solution before the bisphosphite ligand (1).

(108) The reaction products were removed continuously from the reactor via the cycle gas stream and partially condensed out in a condenser at 50 C. The condensed phase was run continuously out of the phase separation vessel. To determine the yield, samples were taken from the cycle gas upstream and downstream of the reactor and analysed by means of a gas chromatograph.

(109) By a daily metered addition of the above-described ligand solution, it was possible to keep the conversion and regioselectivity constant.

(110) To determine the reactor contents, samples were taken from the reactor and analysed by means of liquid chromatography (HPLC).

(111) Under the selected reaction conditions, an aldehyde yield between 80% and 90% was established. It was possible to keep this state constant up to the end of the experiment. The percentage distribution between n-pentanal and 2-methylbutanal, i.e. the regioselectivity, was 92% to 8%.

(112) In the steady-state phase of the experiment, no rhodium degradation was recorded.

(113) The results are shown in FIG. 3.

(114) FIG. 3 shows a 1500 h prolonged experiment in the hydroformylation of crude butane with the inventive unsymmetric ligand (1). Over the entire experiment duration, a constantly high activity, i.e. aldehyde yield averaging 80%, was ensured, still with very good regioselectivity.

(115) In the second experiment series, rather than the inventive compound (1), 55 g of the comparative biphephos compound were used. The results are shown in FIG. 4.

(116) Under the reaction conditions selected, the aldehyde yield fell from initially 70% to 80% after 150 h to 40% to 50%. The percentage distribution between n-pentanal and 2-methylbutanal, i.e. the regioselectivity, was 95% to 5%.

(117) In the steady-state phase of the experiment, no rhodium degradation was recorded.

(118) FIG. 4 shows a 250 h prolonged experiment in the hydroformylation of crude butane with the symmetric comparative biphephos ligand (1). In this case, compared to the inventive ligand, long-lasting activity cannot be ensured. Under the reaction conditions selected, the aldehyde yield fell from initially 70% to 80% after 150 h to 40% to 50%. The regioselectivity was still very good. In other word, this ligand features much lower long-term stability.

(119) Consequently, the inventive unsymmetric ligand (1) features much better stability than the symmetric comparative biphephos ligand.

(120) The inventive ligand (1) in the catalytically active composition features much better long-term stability than the ligands described in the prior art to date and thus achieves the stated object. An optimal long-term stability of the catalytically active composition is of significance especially in industrial scale use, since the ligand in the hydroformylation reaction can be replenished on the industrial scale, but any replenishment adversely affects the economic viability of an industrial scale process and may make it unfeasible. It is thus essential to use a ligand of maximum long-term stability, which, as well as the long-term stability, also features good activity and good n/i selectivity. This object is achieved by the inventive ligand (1).