SYNTHESIS AND CATALYSES OF TRANSITION-METAL COMPLEXES WITH HEMILABILE LIGANDS TEO SHIHUI (B Sc (Merit), NUS) A THESIS SUBMITTED FOR THE DEGREE OF DOCTOR OF PHILOSOPHY DEPARTMENT OF CHEMISTRY NATIONAL UNIVERSITY OF SINGAPORE 2010 i Table of Contents Abbreviations and Symbols vii Summary viii List of Publications x List of Figures xi List of Tables xv Acknowledgements xvii Chapter General Introduction 1.1 Transition metal catalysts 1.2 Advantages of incorporating ferrocene in catalysts’ design 1.3 Types of mixed donor hybrid ligands 1.3.1 Bidentate [P, O] ligands 1.3.1.1 Phosphine-ether ligands 1.3.1.2 Phosphine-ester and phosphine-ketone ligands 11 1.3.1.3 Phosphine-alcohol and phosphine-phenol ligands 12 1.3.1.4 Phosphine-phosphonate ligands 13 1.3.2 Bidentate [P, N] ligands 13 1.3.2.1 Phosphine-oxazoline ligands 14 1.3.2.2 Phosphine-pyridine ligands 15 1.3.2.3 Phosphine-amine ligands 15 1.3.2.4 Phosphine-imine ligands 17 1.3.3 Bis(phosphine)amine ligands 18 1.4 Complexes with mixed donor hybrid ligands of interest 19 1.4.1 Reported complexes with [P, O] ligands 19 ii 1.4.1.1 Ni complexes 19 1.4.1.2 Pd complexes 20 1.4.1.3 Cr complexes 21 1.4.1.4 Rh complexes 22 1.4.2 Reported complexes with [P, N] ligands 23 1.4.2.1 Ni complexes 23 1.4.2.2 Pd complexes 24 1.4.2.3 Cr complexes 25 1.4.2.4 Rh complexes 25 1.4.3 Complexes with bis(phosphine)amine ligands 26 1.5 Future Perspective 27 1.6 Objectives 28 Chapter Synthesis of Bidentate Mixed donor Ligands 29 2.1 Synthesis of 1,1’-disubstituted phosphine-imine and phosphine-ether ferrocenyl ligands (L1 and L2) 29 2.1.1 Synthesis of phosphine-imine ferrocenyl ligands (L1) 31 2.1.2 Synthesis of phosphine-ether ferrocenyl ligands (L2) 34 2.2 Synthesis of phenyl iminophosphines 35 2.3 Synthesis of bis(phosphino)amine ligands 36 2.4 Conclusion 38 Chapter Low Valent Nickel Complexes Supported by [P, N] ferrocenyl Ligands And Their Catalytic Applications in Ethylene Oligomerization 3.1 Introduction 39 3.2 Earlier work on nickel catalyzed ethylene oligomerization 41 iii 3.2.1 SHOP catalysts 42 3.2.2 Ni(II) α-diimine and related [N, N] catalysts 43 3.2.3 [N, O] chelating neutral nickel catalysts 44 3.2.4 [P, N] chelating nickel catalysts 45 3.2.5 Development of low valent nickel complexes and catalytic contributions towards ethylene oligomerization 47 3.3 Objectives of this work 49 3.4 Results and discussion 50 3.4.1 Synthesis and characterization of new low valent nickel catalysts 3.4.1.1 Reaction of Ni(COD)2 with ferrocene iminophosphane ligands and diphenylacetylene 50 3.4.1.2 Reaction of Ni(COD)2 with ferrocene iminophosphane ligand under CO 53 3.4.1.3 Reaction of Ni(COD)2 with ferrocene iminophosphane ligand and AlEtCl2 or AlCl3 57 3.4.2 Catalytic Studies 60 3.5 Conclusion 64 Chapter Chromium Precursors And [PNP] Ligands: Complexation and Potential Catalytic Applications in Ethylene Selective Tetra-, Oligo- and Polymerization 4.1 Introduction 4.1.1 Global demand for 1-octene 65 4.1.2 Ethylene tetramerization 66 4.2 Objectives 70 4.3 Results and discussion 71 4.3.1 Chromium(III) [PNP] complexes syntheses and structures 4.3.1.1 Reaction of [PNP] ligand L4d with CrCl3(THF)3 and RCN 71 iv 4.3.1.2 Reaction of reported binuclear Cr(III)-PNP complexes with acetonitrile 76 4.3.1.3 Hydrolytic disintegration and ligand transformation of [PNP] ligand 79 4.3.2 Catalytic studies 85 4.3.2.1 Potential applications of compounds C5 – C10 and L5 to ethylene oligomerization and polymerization 85 4.3.2.2 Potential applications of [PNP] ligand L4 towards catalyzing Selective Ethylene Tetramerization in the presence of Cr(acac)3 89 4.4 Conclusion 92 Chapter Palladium Catalyzed C-C bond formations 5.1 Introduction 93 5.1.1 Palladium catalyzed Suzuki-Miyaura cross coupling reactions 93 5.1.2 Palladium catalyzed 1,2-addition of aryl boronic acids to aldehydes 95 5.1.3 Palladium catalyzed 1,4-conjugate addition of organoboronic acids to α,βunsaturated ketones 97 5.2 Results and Discussion 99 5.2.1 Catalytic performance and studies 5.2.1.1 Suzuki cross coupling reactions of aryl halides and organoboronic acids 99 5.2.1.2 1,2- and 1,4-addition reactions of organoboronic acid to carbonyl compounds 105 5.2.2 Coordination studies of potential catalytic active intermediates formed between the [P, N] ferrocenyl ligands and Pd(0) precursors 112 5.2.2.1 [P, N] palladium complexes formed from oxidative addition in Suzuki-Miyaura cross coupling reactions 112 5.2.2.2 Contribution of chloroform in directing palladium catalyzed 1,2 addition of phenylboronic acid to aldehydes 118 v 5.3 Conclusion 120 Chapter Potential Applications of Chiral 1,1’-Iminophosphine Ferrocenyl Ligands in Asymmetric Rhodium-catalyzed Hydrosilylation of ketones to alcohols 6.1 Introduction 121 6.1.1 Importance of optically active secondary alcohols 121 6.1.2 Asymmetric rhodium-catalyzed hydrosilylation of ketones to alcohols 122 6.1.3 Reported ligands in rhodium-catalyzed asymmetric hydrosilylation of ketones to alcohols 124 6.1.3.1 Nitrogen-based ligands 124 6.1.3.2 Diphosphine ligands 125 6.1.3.3 Heterobidentate [P, N] ligands 126 6.2 Objectives 129 6.3 Results and Discussion 129 6.3.1 Catalytic evaluation of [P,N] ferrocenyl ligands with Rh and Ir precursors towards asymmetric hydrosilylation of ketones to alcohols 129 6.3.2 Synthesis of Rh/ligand complexes 133 6.3.3 Catalytic performance of Rh/complex towards asymmetric hydrosilylation of ketones 136 6.4 Conclusion 138 Experimental 139 References 169 Appendix 187 vi ABBREVIATIONS AND SYMBOLS Å acac Ar BBN BINAP Bn nBu tBu °C cat δ COD COE Cp Cy dba dea DME DMF dppb DPPF dppp ee eqn ESI FAB h HPLC IR M MAO mg mL mmol MS m/z Nap NMR r.t THF TMEDA Tol Angstrom Acetonylacetate Aryl 9-Borabicyclo[3.3.1]nonane 2,2’-Bis(diphenylphosphino)-1,1’-binaphthyl Benzyl Primary butyl Tertiary Butyl Degree Celsius Catalyst Chemical shift 1,5-Cyclooctadiene Cyclooctene η5-Cyclopentadienyl Cyclohexyl Dibenzylideneacetone Diethanolamine 1,2-Dimethoxyethane N,N-Dimethylformamide Bis(diphenylphosphino)butane 1,1’-Bis(diphenylphosphino)ferrocene Bis(diphenylphosphino)propane Enantiomeric excess Equation Electrospray ionization Fast atom bombardment Hours High pressure liquid chromatography Infra red Molar (moles per litre) Methylaluminoxane Milligram Millilitre Millimole Mass-to-charge Mass-to-charge ratio Naphthyl Nuclear magnetic resonance Room temperature Tetrahydrofuran N,N,N’,N’-Tetramethyl-1,3-ethylenediamine Tolyl vii SUMMARY The aim of the project is the development of ligands with mixed hybrid donors and their corresponding metal (namely nickel, chromium, palladium and rhodium) complexes Structural and catalytic studies on these transition metal complexes towards ethylene oligomerization, C-C bond formations and asymmetric reduction are also performed This thesis contains six chapters Chapter is a general introduction on the importance of mixed hybrid multi-dentate ligands and transition metal catalysts in industries and laboratories The ability of these multidentate mixed hybrid ligands to bind in more than one fashion to the metal centre, which is known as the hemilabile behaviour, has greatly improved both catalytic and organometallic model reactions Chapter focuses on the synthesis of new mixed donor hybrid ligands namely: 1,1’disubstituted iminophosphino ferrocenes, 1,1’-disubstituted phosphino-ether ferrocenes, phenyl iminophosphines and bis(phosphino)amines Chapter describes the development of Ni(0), Ni(I) and Ni(II) catalysts towards ethylene oligomerization to produce linear alpha-olefins which are versatile intermediates for production of co-polymers, plasticizers, alcohols, detergents, synthetic lubricants and surfactants Through the use of ferrocenyl iminophosphine ligands, appropriately stabilized Ni(0) and binuclear Ni(I)-Ni(0) complexes are synthesized, structurally identified and activated for catalyzing ethylene oligomerization The use of an alkynestabilized Ni(0) catalyst could be viewed as a mimic for olefin entry to nickel-promoted olefin oligomerization and thus, enhance the catalytic activity and oligomeric selectivity The isolation of binuclear Ni(I)-Ni(0) complex highlights a balance between coordinative ability, complex stability and catalytic activity can be achieved through the use of mixeddonor hybrid ligands Chapter is about developing suitable catalysts for producing 1-octene which is a challenge for industrial linear alpha-olefin production Promising bis(phosphino)amine ligands with functional side arm at the N-site are developed These ligands display good viii activities (TON: 9000 – 61000) in the chromium catalyzed selective ethylene tetramerization in the presence of Cr(acac)3, toluene and MAO, under 30 bar of ethylene pressure at 80°C Attempts to understand the coordination nature of these bis(phosphino)amine ligands with chromium precursors have led to an unusual ligand destruction followed by reconstruction process which the ligand is reconfigured from a bidentate [PNP] to a tridentate [ONO] donor set This discovery has revealed a direct pathway to prepare new type of hybrid ligand Chapter concerns the application of palladium precursors and the [P, N] mixed hybrid ligands we have developed as catalysts for Suzuki-Miyaura cross coupling reactions, and 1,2- & 1,4- addition reactions of organoboronic acids to carbonyl compounds Further attempts to understand the contribution of the [P, N] ligands towards the Palladium intermediates in the Suzuki-Miyaura cross coupling reaction and 1,2-addition reaction of arylboronic acids to aldehydes have been made Potential useful spectroscopic data of the palladium intermediates has also been obtained Chapter describes our attempts to apply chiral 1,1’-iminophosphine ferrocenes in asymmetric rhodium-catalyzed hydrosilylation of ketones Among them, the ferrocenyl ligand which contains (R)-ethylnaphthylimine and the diphenylphosphine group and its related rhodium complex catalyzed the asymmetric hydrosilylation of ketones efficiently under mild conditions The related square-planar rhodium complexes have been synthesized and structurally characterized ix LIST OF PUBLICATIONS S Teo, Z Weng, T S A Hor, Iminophosphine Ligands in Palladium Catalyzed Addition Reactions of Arylboronic Acids to Carbonyl Compounds, Manuscript in preparation S Teo, Z Weng, T S A Hor, Substituent effect of ferrocenyl Iminophosphine in Pd(II) catalyzed Suzuki coupling, J Organomet Chem 2011, submitted S Teo, C.-Y Tan, Z Weng, T S A Hor, Complexation of chiral 1,1’Iminophosphine Ferrocene to [RhI(COD)] and their catalytic effects on asymmetric hydrosilylation to ketones to alcohols, Manuscript in preparation S Teo, Z Weng, T S A Hor, Unusual Ligand Transformation Mediated by Chromium(III): Hydrolytic Disintegration of a [PNP] Hybrid Ligand with CH3CN Insertion, Organometallics, 2008, 27, 4188 Z Weng, S Teo, T S A Hor, Metal Unsaturation and Ligand Hemilability in Suzuki Coupling, Acc Chem Res 2007, 40, 676 Z Weng, S Teo, T S A Hor, Chromium(III) catalyzed ethylene tetramerization promoted by bis(phosphino)amines with an N-functionalized pendant, Dalton Trans 2007, 3493 Z Weng, S Teo, T S A Hor, Stabilization of Nickel (0) by hemilabile P, NFerrocene Ligands and their ethylene oligomerization Activities Organometallics, 2006, 25, 4878 Z Weng, S Teo, L L Koh, T S A Hor, A structurally characterized Ni-Al methyl-bridged complex with catalytic ethylene oligomerization activity, Chem Comm 2006, 1319 x Chapter Figure 4.9 ORTEP diagram of complex C8 (R = Cy) with thermal ellipsoids at the 40% probability level Hydrogen atoms have been omitted for clarity Table 4.2 Selected bond lengths (Å) and angles (º) of C8 Complex C8 Bond lengths (Å) Cr(1)-P(1) Cr(1)-P(2) Cr(1)-N(1) Cr(1)-Cl(1) Bond angles (º) 2.489(1) 2.485(1) 2.051(4) 2.278(1) Cr(1)-Cl(2) Cr(1)-Cl(3) N(1)-C(1) 2.300(1) 2.295(1) 1.501(5) P(1)-N(1)-P(2) Cl(1)-Cr(1)-Cl(3) Cl(3)-Cr(1)-P(2) N(2)-Cr(1)-Cl(1) 104.7(2) 91.62(5) 96.14(5) 177.2(1) P(1)-Cr(1)-P(2) Cl(2)-Cr(1)-Cl(3) Cl(2)-Cr(1)-P(1) N(2)-Cr(1)-Cl(3) 65.95(4) 102.46(6) 94.85(5) 89.7(1) 78 Chapter 4.3.1.3 Hydrolytic Disintegration and ligand transformation of [PNP] ligand Complex C8 is stable in pure and dry CH3CN, but in the presence of trace of H2O, the [PNP] ligand breaks down and transforms to a new class of tridentate [ONO] hybrid ligand in [Ph2P(O)C(CH3)(NH2)P(O)Ph2]CrCl3 (C9, see Equation 4.1) IR analysis on C9 indicates the loss of the nitrile function Presence of νN-H at 3082 and 3058 cm-1 is in agreement with the formation of a new –NH2 function CH3 R N Ph2P PPh2 trace of Cr Cl Cl Ph 2P Cl N C CH3 C8 O C H 2N PPh O + RNH2 Cr H2O Cl CH H2O Cl Ph2P O C PPh NH2 O Cl L5 C9 Equation 4.1 Hydrolytic Disintegration and ligand transformation of [PNP] ligand mediated by Cr(III) Structural confirmation of complex C9 comes from single-crystal X-ray diffraction studies (Figure 4.10 and Table 4.3) It has a distorted octahedral geometry with a fac-[CrCl3] core completed by [ONO] tridentate coordination The Cr–O lengths (2.003(6) and 2.039(6) Å) are significantly shorter than that of the related PNP complex CrCl3{[η3-(P,P,O)-(PNPOMe)]} with an “agostic” Cr-O interaction (2.1562(15) Å).118 The Cr-N bond (2.141(7) Å) is comparable to those in the trimerization catalysts, viz CrCl3[HN(CH2CH2PPh2)2] (2.137(3) Å)51a and CrCl3[HN(CH2CH2SEt)2] (2.1059(18) Å).113 This hybrid tridentate [ONO] ligand 79 Chapter coordinates to a fac-Cr(III) core with two fused 5-membered rings such that two phenyl groups across the two rings are oriented face-to-face with some degree of π···π stacking (interplanar distance = 3.52 Å) Figure 4.10 ORTEP diagram of complex C9 with thermal ellipsoids at the 40% probability level Table 4.3 Selected Bond Lengths (Å) and Angles (deg) of Complex C9 Complex C9 Cr(1)-N(1) Cr(1)-Cl(1) Cr(1)-Cl(2) Cr(1)-Cl(3) Cr(1)-O(1) 2.141(7) 2.307(3) 2.278(3) 2.313(3) 2.039(6) Cr(1)-O(2) C(1)-C(2) N(1)-C(1) P(1)-O(1) P(2)-O(2) 2.003(6) 1.51(1) 1.52(1) 1.521(6) 1.505(6) P(1)-C(1)-P(2) N(2)-Cr(1)-Cl(1) N(2)-Cr(1)-Cl(3) Cl(2)-Cr(1)-Cl(3) 109.3(4) 177.2(1) 89.7(1) 102.46(6) O(1)-Cr(1)-O(2) N(2)-Cr(1)-Cl(2) Cl(1)-Cr(1)-Cl(2) Cl(1)-Cr(1)-Cl(3) 87.7(2) 88.3(1) 93.78(5) 91.62(5) 80 Chapter In a single reaction, a bidentate [PNP] aminodiphosphine remarkably transformed to a tridentate [ONO] amino-α-diphosphoryl ligand This was driven by the favorable conversion of the reactive [CrCl3L2(solvate)] to the stable [CrCl3(L2L’)] core with extrusion of free amine RNH2 To our knowledge, there is no precedent of such ligand transformation A possible mechanism is the hydro-heterolytic cleavage of the reactive PIII-N bonds in C8, followed by CH3CN insertion to the void between the two phosphoryl functions (A), resulting in proton and electron transfer (B) to create the macrocyclic ligand (C9) (Scheme 4.4) Effectively, the first step is ligand dissection whereas the second is a ligand reconstruction step Scheme 4.4 Proposed mechanism for the formation of C9 from C8 81 Chapter This ligand transformation has several favorable driving forces A strained 4membered ring is replaced by the more stable fused 5-membered rings with additional macrocyclic effect The reactive (P(1)-Cr(1)-P(2) 65.95(4)°) P−N bonds are replaced by the more robust C−P and C−N with additional P−O bonds Removing the labile CH3CN and ejection of stable amine RNH2 also contribute to the thermodynamic drive The potential use of this ligand transformation as a synthetic method for new [ONO] ligand Ph2P(O)C(CH3)(NH2)P(O)Ph2 (L5) was demonstrated in the hydrolytic reaction of C9 In the presence of excess H2O, complex C9 broke down to release the [NON] ligand L5 (65% yield) (Scheme 4.4) L5 was characterized via 1H, 13C-{1H}, 31P-{1H} NMR and mass analyses The chemical shift of amine protons on L5 occurred at 2.13 ppm in the 1H-NMR spectrum The characteristic signals for amine carbon resonated at 58.1 ppm and all the signals for the carbons on L5 could be allocated The characteristic 31P-{1H} NMR signal for phosphorus in ligand L5 was a singlet resonance at 32.7 ppm and the chemical shift corresponded accordingly to the Ph2P=O group.27 Mass signals in positive mode at m/z = 446 corresponded to [L5+H]+ species Structure of L5 was confirmed crystallographically (Figure 4.11, Table 4.4) The dihedral angle between the O(1)-P(1)-C(1) and P(1)-C(1)-P(1A) planes expands beyond perpendicular, viz 113.7(3)o The anti-orientation of the two P=O bonds move the phenyl rings away from each other and remove the π stacking as observed in the complexed state (L5) 82 Chapter In the absence of Cr, the free [PNP] ligand Ph2PN(Ph)PPh2 in CH3CN could react with excess water (>1,000 fold) to give L5 but only very slowly (~30 days), with much lower yield (ca 20%) and with other unidentified by-products There was no evidence of the formation of L5 when the PNP ligand was added to a stoichiometric mix of CH3CN and water under ambient conditions, even after prolonged stirring of > 30 days Figure 4.11 ORTEP diagram of complex L5 with thermal ellipsoids at the 40% probability level Table 4.4 Selected Bond Lengths (Å) and Angles (°) of Complexes L5 Bond Lengths (Å) Angles (°) P(1)-O(1) 1.490(3) O(1)-P(1)-C(1) 112.47(12) P(1)-C(1) 1.879(3) N(1)-C(1)-C(2) 107.4(5) C(1)-N(1) 1.476(11) N(1)-C(1)-P(1) 112.6(4) C(2)-C(1)-P(1) 107.6(4) 83 Chapter Compounds with the P-C(N)-P functions have been shown to be biologically active.119 For example, α-aminophosphates can mimic the action of natural amino acids and act as enzyme inhibitors.119a However, only a few methods of their preparation are known.120 This metal-mediated pathway could provide a new alternative one-step methodology to amino-diphosphine oxides and diphosphonates such as R2P(O)-C(NH2)-R’-P(O)R2 under ambient conditions Attempts to develop the use of Cr catalysts to promote the hydrolysis of PNP compounds to form new amino-diphosphonates are currently in progress Chloride abstraction reaction of C9 with AgSbF6 in CH3CN was attempted However, it did not give the expected product such as [{Ph2P(O)C(CH3)(NH2)P(O)Ph2}CrCl2(CH3CN)]+ presumably the solvate on a [CrIIICl2] core is too reactive to be isolated Its formation triggers a ligand cleavage to release Ph2POH which subsequently attacks the intermediate to yield {[Ph2P(O)C(CH3)(NH2)P(O)Ph2]CrCl2[P(O)(H)Ph2]}+SbF6- (C10) (Scheme 4.5) Other decomposition by-products are not isolated or identified The structure of C10 had been confirmed by X-ray single-crystal crystallographic analysis but the quality of data was too poor to justify any meaningful structural discussions 84 Chapter Scheme 4.5 Chloride abstraction reaction of C9 with AgSbF6 in CH3CN 4.3.2 Catalytic Studies 4.3.2.1 Potential applications of compounds C5 – C10 and L5 towards catalyzing ethylene oligomerization and polymerization Current knowledge points to a metallacyclic pathway involving coordinatively unsaturated intermediates and perhaps a reversible Cr(I)/Cr(III) process.121 If the hybrid character of the multidentate ligand is expressed in terms of a dynamic formation and cleavage of the M-L bond, the resultant hemilability could offer additional support to the unsaturated intermediate The detection of C5 is an illustration of this model It was formed directly from ligand addition to CrCl3(THF)3 This 13-e mononuclear complex would be 5-coordinated and unsaturated without the solvate or ligation from the internal donor (i.e N) of the hybrid ligand Coordination of solvate therefore provided a simple model for the access of a range of potentially unsaturated and catalytically-active Cr(III) with the needed hybrid ligand This type of catalyst in olefin oligomerization and polymerization has many advantages 85 Chapter Preliminary data collected at room temperature under 430 psi pressure of ethylene with [Al]/[Cr] ratio at 420 indicated low – moderate activites (1300 – 29100 g prod./g Cr) of C5 – C10 towards oligomerization and polymerization as shown in Table 4.5 Table 4.5 Activities and selectivities of different catalytic systems towards polymerization and oligomerization of ethylenea Activityb Catalyst Total PE Oligomers Oligomer distribution (wt%) product Entry (%)c (%) C4 C6 C8 C10 (α-C4) (α-C6) (α-C8) (α-C10) (mg) C12+ C5 500 11 89 11 1(75) 25(43) 39(22) 35 C6 5400 112 93 6(90) 14(49) 80 3d C6 5400 113 89 11 18(87) 6(23) 76 C7a 900 19 78 22 1(57) 17(19) 82 C7b 1200 24 95 21(99) 28(22) 51 C8 (R = Cy) 13700 286 100 - - - - - C9 1300 28 86 14 76 (83) (43) (42) 19 Cr(THF)3Cl3 600 13 100 - - - - - 9e L5/ 2500 52 100 - - - - - 10 Cr(acac)3 1500 32 94 (60) (40) (45) 11 (42) 76 11e L5/ Cr(acac)3 7400 154 10 90 (43) 15 (73) 13 (72) 21 (60) 43 4500 94 84 16 70 (14) 18 (26) 12 12300 255 96 0 0 100 Cr(THF)3Cl3 C10 12 13f C10 a Conditions: 0.4 µmol of Cr loading, toluene, 430 psi of ethylene, 420 equiv of MAO, room temperature, h Average at least runs bActivity = g prod./g Cr c% = weight % d The run was performed at 80°C instead of room temperature e0.4 µmol of L5 is used f Chlorobenzene was used as solvent instead of toluene 86 Chapter In our preliminary studies, it was observed the catalytic activities are significantly influenced by the ligand environment The combination of Cr(III) with the hybrid ligand was essential to achieve ethylene oligomerization and polymerization and influence their distribution as well as the α-alkenes selectivity Without these ligands, Cr precursors like Cr(acac)3 and CrCl3(THF)3 gave no or little oligomers They gave polymers but with mediocre activities (Table 4.6, Entry and 7) The mono-solvated Cr-PNP complexes C5, C7a, C7b, and C8 as catalysts in the presence of MAO gave low oligomeric products and the dimeric complex C6 gave high oligomeric products at room temperature (Table 4.6, entries – and - 6) At elevated temperature of around 80°C, although the dimeric complex C6 produced an increase amount of C6-C8 products (overall mass% = 24%, 1-C6 = 87%, 1-C8 = 23%, Table 4.5, entry 3), the mono-solvated complexes C5, C7a, C7b, and C8 gave no oligomers but polymers up to TOF of 18700 h-1, 24300 h-1, 29000 h-1 and 13700 h-1 respectively It was therefore noted that the introduction of a coordinating solvent such as THF, MeCN, EtCN would drastically alter the selectivity of the catalyst to exclusive production of polyethylene (PE) when dry toluene was used as solvent (Table 4.5, entry 1, – 6) The role of coordinating solvent remained unclear but it could stabilize the metallacyclic intermediates118,122 by suppressing the β-hydrogen elimination and therefore encourage growth of the metallacyclic ring Typical ONO ligands, represented by amine bis(phenolate), N-alkoxy-βketoiminate and Schiff base, have shown good catalytic activity in polymerization 87 Chapter of alkenes and enantioselective catalysis.123 However, we are not aware of any report on metal complexes with the [ONO] donor-set on a P-C-P backbone The catalytic activity and production of PE decreased significantly when the bidentate [PNP]-Cr precatalyst such as 132 was replaced by the tridentate [ONO]Cr complex C8 and C10 (1300 and 4500 g prod./g Cr respectively, Table 4.6, Entry and 12) Isolation of C8 and C10 enabled a direct comparison between a neutral (C9) and ionic complex of similar makeup of hybrid ligands Both showed similar activities towards oligomerizations, but the ionic complex C10 showed at least 7-fold increase compared to the neutral complex C8 in terms of PE activity In general both complexes were not very active, probably because of their low solubility in the reaction solvent toluene When it was replaced by a more polar solvent such as chlorobenzene in which C10 was significantly more soluble, the PE activity of improves by nearly 2-fold to 12300 g/g (Table 4.5, Entry 13) Attempts to characterize the polymer product by GPC analysis were not successful as it was insoluble under standard analytical procedure which involves dissolving the PE in 1,3,5-trichlorobenzene at 145°C 124 The PE product could be high-density polyethylene (HDPE) The nature of the metal complex also influenced the oligomeric selectivity Although the [ONO]-ligand L5 with CrCl3(THF)3 catalyzed ethylene polymerization (Table 4.5, Entry 9), it worked with Cr(acac)3 to promote good activity (7400 g prod./g Cr) towards ethylene oligomerization in which the 88 Chapter distribution of the liquid products (C4-C20) (Entry 7) followed a typical SchulzFlory distribution with a K value of 0.82.125 Activation of the L5/Cr(acac)3 system with other Lewis acid co-catalysts such as EtAlCl2, AlMe3 and BEt3 or variation of the ethylene pressure would affect the overall catalytic activity but not the oligomers distribution At higher temperatures (up to 80°C), the activity increased by 2-fold, but the selectivity favoured PE over oligomers, with the latter production greatly reduced to 15% 4.3.2.2 Potential application of [PNP] ligand L4 towards catalyzing selective ethylene tetramerization in the presence of Cr(acac)3 Although the related Cr complexes synthesized displayed poor – moderate activities for ethylene oligomerization and polymerization, the bis(phosphino)amines ligands L4a – L4h were found to display good activities in the Chromium catalyzed selective trimerization or tetramerization of ethylene in the presence of Cr(acac)3, toluene, MAO, under 30 bar of ethylene pressure at 80ºC as summarized in Table 4.6 The selectivity of 1-octene over other octenes was excellent (96-99%) when a equivalent amount of ligand L4a – L4g were utilized (Table 4.6, Entries –7) In the preliminary catalytic findings, the isomeric distributions however depended sensitively on the nature of the functionalized side-chain at N The thioether ligands showed the best selectivity with respect to octenes formation (33-55%, Table 4.6, Entries 1-4), while the pyridyl derivative exhibited the lowest selectivity (19%, Table 4.6, Entry 7) However, the ether-functionalized ligands (L4e-f and L4g) displayed varying 89 Chapter catalytic performance While ligand L4h together with Cr(acac)3 appeared to be more selective towards formation of 1-C6 (mass% = 83, Table 4.6 entry 8), the ligands with alkyl ether-substituents L4e-f favoured the formation of 1-C8 (mass% ~ 98%, Table 4.6, entries and 6) and higher oligomer formation in the presence of Cr(acac)3 The length of the side chain also influenced the selectivity Use of a methylpropyl thioether residue in conjunction with an electron-donating and less hindered ethyl substituent at the phosphine gave 1-octene in excess of 55 mass% (Table 4.6, entry 3) This is almost comparable to the high performance catalysts in the Sasol process with selectivities up to 70%,111,113 both of which exceed the conventional one-step ethylene oligomerization or tetramerization technology that generally give 1-octene in yields