Heterobimetallic complexes supported by phosphinoamide ligands

Ayres, Alexander James (2017) Heterobimetallic complexes supported by phosphinoamide ligands. PhD thesis, University of Nottingham.

[img] PDF (Thesis - as examined) - Repository staff only until 18 July 2019. Subsequently available to Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (7MB)


Many major advances in our fundamental chemical knowledge have emerged from novel metal-metal chemistry. Such complexes which exhibit metal-metal interactions, have provided excellent tools for developing our understanding of chemical structure and bonding, catalysis, metal surface chemistry and bioinorganic chemistry. However, whilst metal-metal bonding is now well understood in the p- and d- blocks and to some extent the s-block of the Periodic Table very little is known, in comparison, about that for the f-block and especially the actinide elements. Traditionally the chemistry of molecular f-element compounds has been dominated by the use of carbon, nitrogen, oxygen and halogen based ligands with the implementation of metal-based fragments as ligands considerably less developed.

In recent years dinuclear species containing a metal-metal bond between two different transition metal centres bridged by a supporting ligand structure have rapidly been gaining interest, as they would be expected to exhibit different reactivity to those of monometallic or homobimetallic complexes. Early/late heterobimetallic complexes, featuring metal-metal interactions supported by a phosphinoamine ligand system, have received particular attention due to their two very different reaction sites, inherent bond polarity, and synergy between the two centres resulting in unique reactivity.

A range of novel phosphinoamines, including MesN(H)PPh2, have been successfully synthesised to accompany the previously reported phosphinoamines DippN(H)PPh2 and MesN(H)PiPr2. All phosphinoamines were successfully deprotonated using nBuLi to afford the corresponding lithium salts as ligand transfer reagents. Reaction of uranium tetrachloride or thorium tetrachloride with three molar equivalents of either [Li(MesNPPh2)(Et2O)2] or [{LiNMes(PiPr2)(Et2O)}2] resulted in the respective tris(phosphinoamide) actinide(IV) chloride complexes [AnCl(MesNPPh2)3] and [AnCl(MesNPiPr2)3] (An = U, Th). Treatment of the tris(phosphinoamide) actinide(IV) halide complexes with Me3SiI afforded the corresponding tris(phosphinoamide) actinide(IV) iodide complexes.

The installation of cobalt iodide into the coordination sphere of the uranium was achieved in the presence of zinc powder to afford the two uranium-cobalt complexes [UCl(MesNPPh2)3CoI] and [(MesNPiPr2)U(μ-X)(MesNPiPr2)2CoI] (X = 41% I, 59%Cl). [UCl(MesNPPh2)3CoI] adopts a paddle-wheel structure whereas [(MesNPiPr2)U(μ-X)(MesNPiPr2)2CoI] exhibits a structure with at best C2v symmetry and two bridging phosphinoamides and one bridging halide ligand. These structural differences confer very different magnetic behaviour. At low temperature [UCl(MesNPPh2)3CoI] can be formulated as an S = 1 spin system resulting from a combination of a magnetic singlet uranium(IV) and triplet cobalt(I). However, in contrast [(MesNPiPr2)U(μ-X)(MesNPiPr2)2CoI] is an S = 0 spin system at low temperature with antiferromagnetic exchange between uranium and cobalt proposed. Density functional theory calculations and topological bond analyses support the notion of formally dative Co ->U bonds in both complexes. In an attempt to further support the claim of antiferromagnetic exchange between uranium and cobalt the thorium analogues were synthesised. To date, investigations into the respective electronic structures are still on going and conclusions about the exact nature of the cobalt centres cannot be drawn. There is however sufficient evidence which warrants further investigation that would be expected to be very detailed and that is thus unfortunately beyond the timeframe of this PhD.

Reaction of the same four tris(phosphinoamide) actinide(IV) chlorides, that resulted in the synthesis of the actinide-cobalt complexes, with [Mo(MeCN)3(CO)3] in dichloromethane afforded the heterobimetallic uranium- and thorium-molybdenum complexes [AnCl(MesNPR2)2(MesNPR2{μ-NCMe})Mo(CO)3] (An = U, R = iPr; An = Th, R = iPr; An = U, R = Ph; An = Th, R = Ph). In contrast treatment of the tris(phosphinoamide) actinide(IV) iodides [IAn(MesNPiPr2)3] (An = Th, U) with [Mo(MeCN)3(CO)3] in toluene resulted in the formation of the complexes [(η2-MesNPiPr2)AnI(MesNPiPr2)({μ-NCMe}MesNPiPr2)Mo(CO)3 (An = U, Th). These compounds show unprecedented acetonitrile insertion as a bridging ligand between the actinide metal and Mo with structural analysis suggests an activation of the nitrile group.

To synthesise the analogous actinide-molybdenum paddlewheel complexes to [UCl(MesNPPh2)3CoI], the appropriate tris(phosphinoamide) actinide(IV) halides were treated with [Mo(CO)3(NCMe)3] to afford the heterobimetallic uranium- and thorium-molybdenum complexes [M(X)(MesNPPh2)3Mo(CO)3] (M = U, X = Cl; M = U, X = I; M = Th, X = Cl; M = Th, X = I,). Orbital- and density-based quantum chemical calculations reveal dative MoM σ-interactions in all cases and so these complexes constitute unprecedented actinide-group 6 metal-metal bonds, where before heterobimetallic uranium-metal bonds were restricted to a few group 7-10 metals.

With a synthetic route towards actinide-cobalt complexes established future work must look towards the reduction of these complexes to result in highly reactive and polarised early-late heterobimetallic compounds. The subsequent reaction of these compounds with small molecules such as CO, CO2 and H2 etc has the potential to result in novel and unusual chemistry which can help develop our fundamental understanding of the actinides. Although the actinide-molybdenum complexes were relatively straightforward to synthesise, surprisingly, the analogous chromium and tungsten complexes were not accessible by the same method, although it has been anticipated that photolysis could be an alternative synthetic route. Photolysis could also lead to the removal of one or more of the carbonyl groups situated on the group 6 metal and result in the targeted highly reactive and polarised early-late heterobimetallic compounds suitable for the activation of small molecules.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Kays, D.L.
Liddle, S.T.
Subjects: Q Science > QD Chemistry > QD241 Organic chemistry
Faculties/Schools: UK Campuses > Faculty of Science > School of Chemistry
Item ID: 41568
Depositing User: Ayres, Alexander
Date Deposited: 18 Jul 2017 04:40
Last Modified: 19 Jul 2017 01:09
URI: http://eprints.nottingham.ac.uk/id/eprint/41568

Actions (Archive Staff Only)

Edit View Edit View