Hydroformylation of naphthas with a rhodium ... - Web del Profesor - ULA

Jointly published by Akadémiai Kiadó, Budapest and Springer, Dordrecht

React.Kinet.Catal.Lett. Vol. 90, No. 2, 347–354 (2007) 10.1007/s11144-007-5023-6

RKCL5023 HYDROFORMYLATION OF NAPHTHAS WITH A RHODIUM COMPLEX IN BIPHASIC MEDIUM Marisela Reyes*, Daniel Mercades, Bernardo Fontal, Trino Suárez, Fernando Bellandi, Ricardo R. Contreras, Isolda Romero, Yuraima Fonseca and Pedro Cancines Universidad de los Andes, Facultad de Ciencias, Departamento de Química, Laboratorio de Organometálicos, Mérida 5101, Venezuela

Received November 21, 2006, accepted December 6, 2006

Abstract The chlorocarbonyl bis–[butylphenyl (meta–sulfonate-phenyl)phosphine] rhodium (I) complex shows catalytic hydroformylation activity in toluene/water biphasic medium for 1-hexene, cyclohexene, 2,3-dimethyl-2-butene and 2-methyl-2pentene, their binary mixtures and a real Venezuelan naphtha, under standardized reaction conditions (1000 psi of syngas (1:1 H2/CO), 100ºC, substrate/catalyst molar ratio (600:1) and 4 h reaction time), obtaining high percent conversion to oxygenated products. Keywords: Hydroformylation, Rh complexes, naphtha, biphasic reaction

INTRODUCTION Biphasic catalysis research has improved by ligand synthesis that leads to water-soluble transition metal complexes. Sulfonated phosphine ligands are widely used as a useful tool in modern industrial biphasic chemistry, as in the OXO process for olefin hydroformylation using hydrosoluble rhodium salts [1]. __________________________ * Corresponding author. Tel.: 0274-2401380; Fax: 0274-2401286; E-mail: [email protected] 0133-1736/2007/US$ 20.00. © Akadémiai Kiadó, Budapest. All rights reserved.

348

REYES et al.: HYDROFORMYLATION

An important problem in the petroleum industry is the high olefin levels in commercial naphthas, mostly 6 carbons atom compounds, that need to be converted to alkanes through catalytic conversion, and also the need to add oxygenated compounds to improve fuel quality. This problem has been investigated with organometallic complexes in carbonylation reactions [2]. Here we report the synthesis, characterization and catalytic hydroformylation reactions of a water-soluble rhodium complex as catalyst precursor. EXPERIMENTAL Materials and methods Reactants (Aldrich Chemical Co., Merck Co. and Riedel –de Haën), gases: H2, Ar, CO, U.A.P grade (Gases Industriales de Venezuela) were obtained and used as received. Solvents were purified before use. IR spectroscopy was done on a FT-IR Perkin Elmer Series 1600 spectrometer, in KBr pellets. Mass spectra were obtained on a Hewlett-Packard 5988A GC/MS System, with electronic impact. NMR spectra were taken in a Bruker Avance DRX 400 MHz spectrometer in d6 –DMSO. Catalytic tests were done in a 300 mL Parr reactor with internal glass liner, heating unit, temperature and stirring control and sampling valve. Products were analyzed with a Perkin-Elmer Autosystem 900 GC (software PE Nelson), FID detector and Quadrex of methyl silicone column, 50 m long, 0.52 μm diameter.

Synthesis of butylphenyl (meta–sulfonate-phenyl) PMSPP)((CH3(CH2)3)(C6H5)(m-SO3Na-C6H4)P)

phosphine

(But-

Into a 50 mL Schlenk flask with magnetic stirring, argon atmosphere and 10 mL of dry tetrahydrofuran (THF), 2.1 mL. (2.23 mmols) of n-butyllitium (5% excess of n-butyllitium) were added and kept in a dry ice-acetone bath. Into other 50 mL Schlenk flask with magnetic stirring, argon atmosphere and 15 mL of dry THF, 3.8 mL (21.1 mmols) of chlorodiphenyl phosphine were added and kept in a dry ice-acetone bath. Via cannula and argon atmosphere, the nbutyllithium solution was added dropwise to the chlorodiphenylphosphine solution for three hours. A crystalline white solid (lithium chloride), was separated (via cannula) from the solution and the solvent was evaporated at reduced pressure. Phosphine sulfonation was carried out by a modification of a method described by Ahrland et al. [3]. Into a 50 mL Schlenk cooled in an icewater bath, containing 3.5 g (14.4 mmoles) of butyldiphenylphosphine (BtDPP), 9 mL of fuming H2SO4 (95-98 % H2SO4, 35% as free SO3) were added dropwise via cannula with strong stirring for 45 minutes. When the solution


349

reached room temperature, it was placed in a boiling water bath for 75 min with occasional stirring. The solution was cooled to room temperature and added to 67 g of ice in a 1 L beaker. A milky emulsion was obtained with some gummy material around the walls of the beaker. The solution was put in an ice-water bath and neutralized with 23.3 mL of 50% NaOH until pH 6–7. A light brown precipitate was obtained, it was filtered and recrystallized from hot water and ethanol (3:1).The solid is insoluble in toluene. (Yield: 1.57g, 31.6%, m.p. 172173ºC). MS (m/z): 344, [M]+ (molecular ion); 242, [M-(SO3Na)]+; 217, [M(C3H7)(SO3Na⋅H2O)]+ (base peak); other peaks at: 199, 185, 182, 153, 152, 140. BtPMSPP shows the following FTIR bands (w = weak, s = strong, m = medium, sh = shoulder): νs(O–H) 3441cm-1 (hydration of ligand), ν(C–H) aromatic 3065 cm-1 (w), νas(C–H) 3055 cm-1, νs(C–H) 3012 cm-1 (w, CH2 butyl chain), ν(C=C) 1625 cm-1 aromatic, νas(P–C) 1432 cm-1 , νas(C–S) 1400 cm-1 (sh) and νs (C–S) 1314 cm-1 (w), νas(S=O) 1183 cm-1 (s) and νs(S=O) 1314 cm-1 (s), νs(P–C) 1059 cm-1 (s), δ (P–C) 959 cm-1 (s), δ (C–H) 725 cm-1 (m) and 692 cm-1 (s) meta di-substitution for aromatics. NMR-1H (d6-DMSO) (ppm): 7.757.67 (C6H4SO3Na); 7.52-7.42 (C6H5); 3.9 (quartet, PCH2(CH2CH2CH3)); 3.74 (quartet, PCH2CH2(CH2CH3)); 3.62 (multiplet, PCH2CH2CH2CH3); 1.0 (triplet, PCH2CH2CH2CH3). Synthesis of chlorocarbonyl bis–(butylphenyl(meta–sulfonate-phenyl) phosphine)rhodium (I) Into a three necked 50 mL Schlenk flask, under argon atmosphere 0.0851 g (0.22 mmols) of bis – (μ–chlorodicarbonyl rodium (I)) ([Rh2(CO)4Cl2]) in 10 mL of 96% ethanol, with strong stirring, 0.3020 g (0.87 mmols) of monosulfonated phosphine in 10 mL of 96% ethanol were added dropwise. The mixture was refluxed at 80ºC for 2 hours. A dark brown solid was obtained, it was filtered and recrystallized from hot water and ethanol (3:1). The solid is insoluble in toluene. (Yield: 0.09 g, 47.7%, m.p. 180 ºC, decomposition). MS (m/z): 854, [M]+ (molecular ion); 826, [M- (CO)]+; 791, [M- (CO)(Cl)]+; 510, [M- (BtPMSPP)]+; 498, [M- (CO) (BtPMSPP) ]+; 475, [M- (Cl) (BtPMSPP)]+; 447, [M- (CO)(Cl) (BtPMSPP)]+; 168, [M- (BtPMSPP)2]+. The rhodium (I) complex shows basically the same FTIR bands as the ligand, with small shifts, and a ν(C=O) band at 2065 cm-1 (w), the only terminal carbonyl group in the complex.

Catalytic test The hydroformylation reactions were carried out for the model substrates present in naphthas: 1-hexene, cyclohexene, 2-methyl-2-pentene and 2,3-

350


dimethyl-2-butene, in biphasic toluene/water medium; for the binary mixtures of these substrates; a synthetic naphtha (a mixture of approximate (% volumen) composition: 1-hexene 32.0%, cyclohexene 11.4%, 2-methyl-2-pentene 28.3% and 2,3-dimethyl-2-butene 28.3%) and also for a real naphtha from “El Palito” refinery (Venezuela). All the reactions were performed under standardized conditions established in previous studies. [4, 5, 6, 7]. The optimum parameters were: 1000 psi of syn-gas (1:1 H2/CO), 100ºC, substrate/catalyst mole ratio (600:1) and reaction time, 4 h for the real naphtha (1 mL), the reaction time was 15 h. RESULTS AND DISCUSSION Catalytics test. 1. Model compounds The results for the hydroformylation of the four-model compounds are given in Table 1. Table 1 Hydroformylation of model compounds Compound (% conversion)

Products

Selectivity*(%)

1. 1-hexene (99.3%)

2-ethyl-pentanal 2-methyl-hexanal heptanal cyclohexane cyclohexyl-carboxaldehyde cyclohexylmethanol 2-methyl-pentane 2-isopropyl-butanal 2,2-dimethyl-pentanal 2-isopropyl-butanol 2,2-dimethyl-pentanol 2,3-dimethyl-butane 2,2,3-trimethyl-butanal 2,2,3-trimethyl-butanol

7.1 40.2 51.9 12.6 1.5 85.7 3.4 8.8 47.1 17.1 23.7 4.9 48.1 47.0

2. cyclohexene (99.8%) 3. 2-methyl-2pentene (65.6%)

4. 2,3-dimethyl-2butene (10.6%)

*Selectivity = (ni / Σ ni )x100; ni = mmoles of product i; Σ ni = sum of all products; measured as areas in GC

The hydroformylation of linear 1-hexene favors slightly the linear aldehyde, and shows no hydrogenation products. The cyclic alkene and substituted olefins


351

give hydrogenation products to alcohols, a fair amount of aldehydes and a slight competitive reaction of alkene hydrogenation. The substituted alkenes give less percent conversion during the same reaction time, as expected.

Binary mixtures The results for the binary mixtures are given on Table 2.

Table 2 Hydroformylation of binary mixtures Mixture (total % conversion)

Component (partial % conversion)

Products

Selectivity* (%)

1 hexene + cyclohexene (97.0%)

Cyclohexene (97.3%)

cyclohexane cyclohexylcarboxaldehyde cyclohexyl-methanol 2-ethyl-pentanal 2-methyl-hexanal heptanal 2,2-dimethyl-pentanal 2-isopropyl-butanal 2-isopropyl-butanol 2,2-dimethyl-pentanol 2,2,3-trimethyl-butanal 2,2,3-trimethyl-butanol

4.9 0.9 27.6 4.8 25.9 35.9 15.3 3.6 6.8 69.2 2.7 2.5

1-hexene (100%) 2-methyl- 2pentene + 2,3-dimethyl-2butene (63.9%)

2-methyl-2 pentene (91.1%)

2,3-dimethyl-2 butene (72.8%)

*See Table 1; sum includes all the products for each mixture

The results observed for the binary mixtures give a product distribution similar to that obtained with the separate substrates for 1-hexene and cyclohexene. For the substituted olefins, the production of alcohols is still important.

Synthetic naphtha The results for the synthetic naphtha are shown in Table 3. The reaction products for the four olefin mixtures (synthetic naphtha) show a similar pattern of product distribution as the separate substrates and binary

352


Table 3 Hydroformylation of synthetic naphtha Naphtha (total % conversion)

Synthetic Naphtha (67.1%)

Component (partial % conversion)

Products

Selectivity* (%)

1-hexene (100%)

2-ethyl-pentanal 2-methyl-hexanal heptanal cyclohexane cyclohexyl-methanol 2-methyl-pentane 2,2-dimethyl-pentanal 2-isopropyl-butanal 2,2-dimethyl-pentanol 2,3-dimethyl-butane 2,2,3-trimethyl-butanal 2,2,3-trimethyl-butanol

3.2 18.8 20.2 1.1 18.9 2.7 8.1 1.3 9.6 7.7 6.9 1.4

cyclohexene (99.1%) 2-methyl-2- pentene (91.9%)

2,3-dimethyl-2butene (76.2%)

*See Table 1. Sum includes all products

mixtures, with some relative increase in alkene hydrogenation for substituted olefins. Aldehydes and alcohols are obtained as previously observed. “In situ” reaction. Mercury drop test A reaction was carried out using the rhodium(I) dimer precursor and the phosphine ligand to generate the active species “in situ”; the results are shown in Table 4. Table 4 In situ”, 1-hexene hydroformylation. Mercury drop test Reaction (% conversion)

Products

Selectivity* (%)

“In situ” reaction, ** 1-hexene (98.8%)

2-ethyl-pentanal 2-methyl-hexanal heptanal

6.1 39.5 52.5

Mercury drop test 1-hexene (99.8%)

2-ethyl-pentanal 2-methyl-hexanal heptanal

7.0 39.5 53.6

* See Table 1; ** bis-(μ–chlorodicarbonyl rhodium(I)): 0.91 mg (2.35x10-3 mmols) ; BtPMSPP: 3.2 mg (9.4x10-3 mmols) in water phase; substrate/catalyst mole ratio 600:1


353

The results obtained for the “in situ” reaction indicate that there is no major difference for the substrate conversion as shown in Table 1, thus it appears that the catalytic precursor can be formed in the biphasic medium and shows similar catalytic activity under similar reaction conditions. For the 1-hexene hydroformylation reaction, a mercury drop test was carried out under the same reaction conditions. The results show (Table 4) that there was no major difference in reaction products as those shown in Table 1, thus we can conclude that the catalysis is not produced by traces of rhodium metallic particles.

Hydroformylation of real naphtha One of the goals of this study is to show the possibility of applying hydroformylation conditions with a rhodium complex for real naphthas in order to reduce partially the olefin content and produce oxygenated compounds that improve gasoline formulations. The results are shown in Fig. 1, where the chromatograms for the real naphtha before and after reaction are compared.

Fig. 1. Chromatogram obtained for the hydroformylation of real naphtha

There is evidence for aldehyde production (C6 and C7 aldehydes, 20 to 25 minutes retention times) and alcohols (longer retention times) and hydrogenation products (signals below 5 min).

354


CONCLUSIONS The rhodium complex prepared presents catalytic hydroformylation activity of 1-hexene, cyclohexene, 2-methyl-2-pentene and 2,3-dimethyl-2-butene, also for the binary mixtures of substrates and a synthetic naphtha in the biphasic system toluene/water under the condition studied. The products obtained include mainly aldehydes and alcohols and small amounts of competitive alkene hydrogenation compounds. 1-hexene hydroformylation can be carried out with the rhodium complex produced “in situ”, and the Hg drop test indicates that no metallic particles are responsible for the catalysis. The reaction with the real naphtha, shows promising results that could improve gasoline quality.

Acknowledgements. Programa Iberoamericano de Ciencia y Tecnología para el Desarrollo, CYTED (Proyecto V.9). CDCHT – ULA proyects (Nº C-1252-0408-A and C–1252–04–08–F). REFERENCES 1. 2. 3. 4. 5. 6. 7.

D.E. Páez: Catálisis Homogénea en Agua: Perspectiva Industrial. Visión Tecnológica. 2. Departamento de Procesos. Intevep S.A, Vol. 4, Nº 2, 147. 1997. B. Cornils, W.A. Herrmann: Applied Homogeneous Catalysis with Organometallic Compounds. Wiley - VCH. Weinheim 1996. S. Ahrland, J. Chatt, N.R. Davies, A. Wilkinson: J. Chem. Soc., 1403 (1958). T. Suarez, G. León, B. Fontal, M. Reyes, F. Bellandi, R. Contreras, P. Cancines: React. Kinet. Catal. Lett., 76, 161 (2002). T. Suarez, B. Castillo, B. Fontal, M. Reyes, F. Bellandi, R. Contreras, A. Bahsas, G. León, P. Cancines: React. Kinet. Catal. Lett., 82, 317 (2004). T. Suarez, A. Guzmán, B. Fontal, M. Reyes, F. Bellandi, R. Contreras, P. Cancines, G. León, L. Rojas: Metal Trans. Chem., in press. 2005. G.E. León: Trabajo Especial de Maestría. (Master Thesis) Universidad de los Andes, Mérida, 2005.