The electronic effects of electron withdrawing aryl substituents on equatorial and apical diphosphines were investigated. Chelating diphosphines designed to coordinate in diequatorial or in apical−equatorial positions were synthesized, and their effects on the regioselectivity of rhodium catalyzed 1-hexene hydroformylation were observed. Only diequatorial coordination was observed for 2,2‘-bis[(diphenylphosphino)methyl]-1,1‘-biphenyl (BISBI) complexes (BISBI)Ir(CO)2H (8) and [BISBI-(3,5-CF3)]Ir(CO)2H (10), and only apical−equatorial coordination was seen for 1,2-bis(diphenylphosphino)ethane (DIPHOS) complexes (DIPHOS)Ir(CO)2H (14) and [DIPHOS-(3,5-CF3)]Ir(CO)2H (15). For the trans-1,2-bis[(diphenylphosphino)methyl]cyclopropane (T-BDCP) complexes, a mixture of diequatorial and apical−equatorial complexes was seen. For (T-BDCP)Ir(CO)2H (12), 12-ae was favored over 12-ee by 63:37, but for [T-BDCP-(3,5-CF3)]Ir(CO)2H (13) the conformational preference was reversed and a 10:90 ratio of 13-ae:13-ee was seen. The electron withdrawing groups in the equatorial positions of BISBI-(3,5-CF3) (1) and T-BDCP-(3,5-CF3) (2) led to an increase in n-aldehyde regioselectivity in rhodium catalyzed hydroformylation. However, electron withdrawing aryl substituents in the apical positions of DIPHOS-(3,5-CF3) (3) led to a decrease in n-aldehyde regioselectivity in rhodium catalyzed hydroformylation.

Electronic effects on rhodium-catalyzed hydroformylation of 1-hexene with electronically dissymmetric DIPHOS derivatives [3,5-() were investigated. Two apical-equatorial chelate isomers were observed for model (diphosphine)Ir(CO) 2 H complexes of dissymmetric diphosphines 1-4. In each case, the equatorial phosphine of the major isomer (96-60%) had electron-withdrawing aryl substituents. These dissymmetric DIPHOS derivatives were used to test the hypothesis that an electron-withdrawing substituent on an equatorial phosphine increases the hydroformylation n:i ratio while an electron-withdrawing substituent on an apical phosphine decreases the n:i ratio. In agreement with the predictions of this hypothesis, hydroformylation with the dissymmetric diphosphine ligand DIPHOS-(3,5-CF 3 ,H) (1), gave an n:i ratio of 4.2:1, higher than either of the symmetric ligands DIPHOS, 2.6:1, and DIPHOS-(3,5-CF 3 ), 1.3:1. Similar observations were made for hydroformylations with 2-4.The rhodium-phosphine-catalyzed hydroformylation reaction, 1 first reported by Wilkinson 2 in the late 1960s and implemented by Union Carbide 3 in the late 1970s, remains one of the most important homogeneously catalyzed processes. Rhodium-phosphine catalysts operate at lower temperature and pressure than earlier cobalt-based catalysts and afford the opportunity to maximize regioselectivity with ancillary phosphine ligand variations. The development of highly regioselective catalyst systems continues to be a primary research aim. Recently developed phosphorus chelates which give high regioselectivity for high linear aldehydes include TexasEastman's 2,2′-bis[(diphenylphosphino)methyl]-1,1′-biphenyl (BISBI) ligand 4 and Union Carbide's diphosphite chelates. 5 Stanley has developed a unique binuclear rhodium catalyst that gives high rates and selectivity. 6 van Leeuwen has reported a variety of diphosphite and diphosphine ligands that give high n:i regioselectivity. 7 Buchwald 8 and Wink 9 have obtained high regioselectivities in hydroformylation of functionalized alkenes using rhodium diphosphite systems. While a wide variety of phosphine and phosphite ligands have been screened, no detailed understanding of how phosphorus ligands control regioselectivity or enantioselectivity has emerged.Wilkinson's dissociative hydroformylation mechanism (Scheme 1) suggests that aldehyde regioselectivity is determined in the hydride addition step that converts a five-coordinate H(alkene)-Rh(CO)L 2 into either a primary or secondary four-coordinate (alkyl)Rh(CO)L 2 . Monodentate phosphine ligands can occupy two equatorial or one equatorial and one apical site of the key five-coordinate H(alkene)Rh(CO)L 2 intermediate. For L ) PPh 3 , Brown's NMR studies showed an 85:15 diequatorial:apicalequatorial mixture of isomers of (PPh 3 ) 2 Rh(CO) 2 H which are in rapid equilibrium at room temperature. 10 Several years ago, on the basis of Brown's findings, we set out to test the hypothesis that very different regioselectivities (1) (a) Parshall, G. W. Homogeneous For related wor...

No abstract

Reaction of (R2PC2H4PR2)Ni(C2H4) with COT gives the mononuclear complexes (R2PC2H4PR2)Ni(η2-C8H8) (R = iPr 1a, tBu 1b). The COT ligand in 1a,b is planar with alternating C−C and CC bonds, corresponding to a formal semiaromatic [C8H8]- ligand. Reactions of 1a with (iPr2PC2H4PiPr2)Ni(η2,η2-C6H10) and of 1b with stoichiometric amounts of {(tBu2PC2H4PtBu2)Ni}2(μ-C6H6) or lithium afford the dinuclear complexes {(iPr2PC2H4PiPr2)Ni}2{μ-η4(1,2,5,6):η4(3,4,7,8)-C8H8} (2a) and {(tBu2PC2H4PtBu2)Ni}2(μ-η2:η2-C8H8) (2b; two isomers). The COT ligand in 2a is tub-shaped and olefinic, whereas in 2b (as in 1a,b) it is planar and semiaromatic. The products are characterized by IR, solution and solid-state NMR spectroscopy, and by X-ray structure analysis.

No abstract