Organometallic Chemistry, Volume 36 (Specialist Periodical Reports)

Organometallic Chemistry Volume 36 A Specialist Periodical Report Organometallic Chemistry Volume 36 A Review of the ...

Author: J. M. Lynam | I. J. S. Fairlamb

199 downloads 1292 Views 10MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Organometallic Chemistry Volume 36

A Specialist Periodical Report

Organometallic Chemistry Volume 36 A Review of the Literature Published between January 2007 and December 2008 Editors I. Fairlamb and J. Lynam, University of York, UK Authors J.G. Brennan, State University of New Jersey, USA T.H. Bullock, University of Cambridge, UK M.P. Cifuentes, Australian National University, Canberra, Australia V. Engels, University of Cambridge, UK M.G. Humphrey, Australian National University, Canberra, Australia D.L. Kays, University of Nottingham, UK S.T. Liddle, University of Nottingham, UK R.L. Melen, University of Cambridge, UK D.P. Mills, University of Nottingham, UK B.E. Moulton, Manchester, UK N.J. Patmore, University of Sheffield, UK A. Sella, University College London, Uk A.E.H. Wheatley, University of Cambridge, UK A.J. Wooles, University of Nottingham, UK C.E. Willans, University of Leeds, UK D.S. Wright, University of Cambridge, UK

If you buy this title on standing order, you will be given FREE access to the chapters online. Please contact [email protected] with proof of purchase to arrange access to be set up. Thank you.

ISBN 978-1-84755-950-0 ISSN 0301-0074 DOI 10.1039/9781847559616 A catalogue record for this book is available from the British Library & The Royal Society of Chemistry 2010 All rights reserved Apart from fair dealing for the purposes of research or private study for non-commercial purposes, or for private study, criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988 and the Copyright and Related Rights Regulations 2003, this publication may not be reproduced, stored or transmitted, in any form or by any means, without the prior permission in writing of The Royal Society of Chemistry, or in the case of reproduction in accordance with the terms of the licences issued by the Copyright Licensing Agency in the UK, or in accordance with the terms of the licences issued by the appropriate Reproduction Rights Organization outside the UK. Enquiries concerning reproduction outside the terms stated here should be sent to The Royal Society of Chemistry at the address printed on this page. Published by The Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge CB4 0WF, UK Registered Charity Number 207890 For further information see our web site at www.rsc.org

Preface Ian J. S. Fairlamba and Jason M. Lynama DOI: 10.1039/9781847559616-FP005

The format for this Volume follows on from the modiﬁcations made to the structure of this journal series, which were introduced in Volumes 34 and 35. The Volume is split into two sections: critical reviews and comprehensive reviews. A series of critical reviews and perspectives which focus on speciﬁc aspects of organometallic chemistry, which interface with other ﬁelds of study, are provided. For this Volume, the critical reviews examine the role played by new ligands motifs for example Charlotte Willans (University of Leeds, UK) highlights the cutting-edge applications of N-heterocyclic carbenes in non-transition metal chemistry. This is coupled with an exciting and comprehensive review of bis(phosphorus-stabilised)methanide and methandiide derivatives ligands in early transition metals and f-elements by Stephen Liddle, David Mills and Ashley Wooles (University of Nottingham, UK). The use of bulky terphenyl ligand systems in stabilising low valent metal complexes has been elegantly described by Deborah Kays (University of Nottingham, UK), and the chemistry multiply bonded paddlewheel compounds with potential applications in molecular electronic devices has been reviewed by Nathan Patmore (University of Sheﬃeld, UK). For the ﬁrst time in this series, structural, synthetic and mechanistic aspects pertaining to the versatile and widely used Pauson-Khand reaction of alkynes, alkenes and dicobalt(0) octacarbonyl is examined by Benjamin Moulton (Reaxa Ltd., Manchester, UK). Comprehensive reviews of the organometallic chemistry of a wide range of elements from across the periodic table are included in this Volume. These articles cover the literature from 2006 and 2007. Signiﬁcant contributions are made by John Brennan and Andrea Sella (covering the f-elements). Volker Engels and Andrew Wheatley have reviewed recent developments in alkali and coinage metals with a focus on organolithium and organocuprate chemistry. Thomas Bullock, Rebecca Melen and Dominic Wright discuss the key aspects in Group 2 (Be-Ba) and Group 12 (Zn-Hg) compounds. Mark Humphrey and Marie Cifuentes discuss a selection of highlights on metal cluster chemistry. In summary, this volume covers a wide range of organometallic chemistry aligned with important applications in other key areas. We have tried to ensure that a broad spectrum of topics which illustrate the diverse nature of modern organometallic chemistry are included in this Specialist Periodical Report.

a

Department of Chemistry, University of York, York YO51 5DD, UK

Organomet. Chem., 2010, 36, v–v | v

c

The Royal Society of Chemistry 2010

CONTENTS Cover

Ball and stick representation of Grubbs generation II catalyst.

Preface Ian J. S. Fairlamb and Jason M. Lynam

v

Non-transition metal N-heterocyclic carbene complexes Charlotte E. Willans 1. Introduction 2. s-Block-carbenes 3. p-Block-carbenes 4. f-Element-carbenes Conclusion References

1

Bis(phosphorus-stabilised)methanide and methandiide derivatives of group 1–5 and f-element metals Stephen T. Liddle, David P. Mills and Ashley J. Wooles 1. Introduction 2. Methane syntheses 3. Group 1 methanides and methandiides

1 1 9 21 25 26

29

29 30 31

Organomet. Chem., 2010, 36, vii–x | vii

c

The Royal Society of Chemistry 2010

4. Group 2 methanides and methandiides 5. Group 3 and f-element methanides and methandiides 6. Group 4 methanides and methandiides 7. Group 5 methanides and methandiides 8. Conclusions References

35 38 47 51 53 53

The stabilisation of organometallic complexes using m-terphenyl ligands

56

Deborah L. Kays 1. Introduction 2. Ligands 3. Complexes of 4. Complexes of 5. Complexes of 6. Complexes of 7. Complexes of 8. Complexes of 9. Conclusions References

group 3 groups 4 and 5 groups 6-9 group 10 group 11 group 12

Recent developments in the chemistry of metal-metal multiply bonded paddlewheel compounds Nathan J. Patmore 1. Introduction 2. Chromium 3. Molybdenum and tungsten 4. Rhenium complexes 5. Diruthenium complexes 6. Dirhodium complexes 7. Conclusions References

The Pauson-Khand reaction Benjamin E. Moulton 1. Introduction References

viii | Organomet. Chem., 2010, 36, vii–x

56 56 57 59 59 66 67 70 72 72

77

77 78 78 81 82 85 90 90

93 93 118

121

Scandium, Yttrium and the Lanthanides John G. Brennan and Andrea Sella 1. Introduction 2. Hydrocarbyl chemical reactivity 3. Redox chemistry 4. Catalysis 5. Organolanthanides in organic synthesis 6. Organolanthanides in materials synthesis 7. Polymer chemistry 8. Synthesis and characterization of new compounds 9. Theory 10. Spectroscopy/electronic structure Acknowledgements Abbreviation References

121 121 129 130 132 133 133 136 141 142 143 143 144

Alkali/coinage metals – organolithium, organocuprate chemistry

148

Volker Engels and Andrew E. H. Wheatley 1. The alkali metals 2. Group II Abbreviations References

148 153 161 162

168

Group 2 (Be-Ba) and group 12 (Zn-Hg) Thomas H. Bullock, Rebecca L. Melen and Dominic S. Wright 1. Scope and organisation of the review 2. Group 2 3. Group 12 References

168 168 173 177

Organo-transition metal cluster complexes

182

Mark G. 1. 2. 3. 4. 5. 6.

182 182 182 182 183 184

Humphrey and Marie P. Cifuentes Introduction Reviews Theory Spectroscopy High-nuclearity clusters Group 7

Organomet. Chem., 2010, 36, vii–x | ix

7. Group 8 8. Group 9 9. Group 10 10. Group 11 11. Mixed-metal clusters Abbreviations References

x | Organomet. Chem., 2010, 36, vii–x

184 194 195 196 196 202 203

Abbreviations Ac acac acacen Ad AIBN ampy Ar Ar* Ar0 f arphos ATP Azb 9-BBN BHT Biim BINAP bipy Bis bma BNCT Bp bpcd bpk Bpz4 But2bpy t-bupy Bz Bzac cbd 1,5,9-cdt chd chpt CIDNP [Co] (Co) cod coe cot CP/MAS Cp CpR

acetate acetylacetonate N,N0 -ethylenebis(acetylacetone iminate) adamantyl azoisobutyronitrile 2-amino-6-methylpyridine aryl 2,4,6-tri(tert-butyl)phenyl 3,5-bis(trifluoromethyl)phenyl 1-(diphenylphosphino)-2-(diphenylarsino)ethane adenosine triphosphate azobenzene 9-borabicyclo[3.3.1]nonane 2,6-dibutyl-4-methylphenyl biimidazole 2,20 -bis(diphenylphosphino)-1,10 -binaphthyl 2,20 -bipyridyl bis(trimethylsilyl)methyl 2,3-bis(diphenylphosphino)maleic anhydride boron neutron capture therapy biphenyl 4,5-bis(diphenylphosphino)cyclopent-4-ene-1,3-dione benzophenone ketyl (diphenylketyl) tetra(1-pyrazolyl)borate 4,40 -di-tert-butyl-2,20 -bipyridine tert-butylpyridine benzyl benzoylacetonate cyclobutadiene cyclododeca-1,5,9-triene cyclohexadiene cycloheptatriene chemically induced dynamic nuclear polarisation cobalamin cobaloxime [Co(dmg)2 derivative] cycloocta-1,5-diene cyclooctene cyclooctatriene cross polarisation/magnetic angle spinning Z5-cyclopentadienyl Z5-alkylcyclopentadienyl

Organomet. Chem., 2010, 36, xi–xv | xi c

The Royal Society of Chemistry 2010

Cp* Cp0 Cp00 CV CVD Cy Cyclam Cym Cyttp dab dabco dba dbpe DBU DCA depe depm DFT diars diarsop dien diop DIPAMP diphos dipp dipyam DMAD DMAP dmbpy DME DMF dmg dmgH dmgH2 DMP dmpe dmpm dmpz DMSO dpae dpam dppa dppb

Z5-pentamethylcyclopentadienyl trimethylsilylcyclopentadienyl tetramethylethylcyclopentadienyl cyclic voltammetry(ogram) chemical vapour deposition cyclohexyl 1,4,8,11-tetraazacyclotetradecane p-cymene PhP(CH2CH2CH2PCy2)2 1,4-diazabutadiene 1,4-diazabicyclo[2.2.2]octane dibenzylideneacetone 1,2-bis(dibutylphosphino)ethane 1,8-diazabicyclo[5.4.0]undec-7-ene 9,10-dicyanoanthracene 1,2-bis(diethylphosphino)ethane 1,2-bis(diethylphosphino)methane density functional theory o-phenylenebis(dimethyl)arsine {[(2,2-dimethyl-1,3-dioxolan-4,5-diyl)bis(methylene)]bis-[diphenylarsine]} diethylenetriamine {[(2,2-dimethyl-1,3-dioxolan-4,5-diyl)bis(methylene)]bis-1-[diphenylphosphine]} 1,2-bis(phenyl-o-anisoylphosphino)ethane 1,2-bis(diphenylphosphino)ethane 2,6-diisopropylphenyl di-(2-pyridyl)amine dimethyl acetylenedicarboxylate 2-dimethylaminopyridine dimethylbipyridine 1,2-dimethoxyethane N,N-dimethylformamide dimethylglyoximate monoanion of dimethylglyoxime dimethylglyoxime dimethylpiperazine 1,2-bis(dimethylphosphino)ethane bis(dimethylphosphino)methane 1,3-dimethylpyrazolyl dimethyl sulfoxide 1,2-bis(diphenylarsino)ethane bis(diphenylarsino)methane 1,2-bis(diphenylphosphino)ethyne 1,4-bis(diphenylphosphino)butane

xii | Organomet. Chem., 2010, 36, xi–xv

dppbz dppe dppf dppm dppp DSD edt EDTA ee EELS EH MO ELF en ES EXAFS F6acac Fc Fe* Fp Fp0 FTIR FVP glyme GVB HBpz3 HBpz*3 H4cyclen HEDTA hfa hfacac hfb HMPA HNCC HOMO IGLO im Is* ISEELS KTp LDA LiDBB LMCT LNCC MAO Me2bpy

1,2-bis(diphenylphosphino)benzene 1,2-bis(diphenylphosphino)ethane 1,10 -bis(diphenylphosphino)ferrocene bis(diphenylphosphino)methane 1,3-bis(diphenylphosphino)propane diamond–square–diamond ethane-1,2-dithiolate ethylenediaminetetraacetate enantiomeric excess electron energy loss spectroscopy extended Hu¨ckel molecular orbital electron localisation function ethylene-1,2-diamine MS electrospray mass spectrometry extended X-ray absorption fine structure hexafluoroacetylacetonate ferrocenyl Fe(CO)2Cp* Fe(CO)2Cp Fe(CO)2Z5-(C5H4Me) Fourier transform infrared flash vacuum pyrolysis ethyleneglycol dimethyl ether generalised valence bond tris(pyrazolyl)borate tris(3,5-dimethylpyrazolyl)borate tetraaza-1,4,7,10-cyclododecane N-hydroxyethylethylenediaminetetraacetate hexafluoroacetone hexafluoroacetylacetonato hexafluorobutyne hexamethyl phosphoric triamide high nuclearity carbonyl cluster highest occupied molecular orbital individual gauge for localised orbitals imidazole 2,4,6-triisopropylphenyl inner shell electron energy loss spectroscopy potassium hydrotris(1-pyrazolyl)borate lithium diisopropylamide lithium di-tert-butylbiphenyl ligand to metal charge transfer low nuclearity carbonyl cluster methyl alumoxane 4,40 -dimethyl-2,20 -bypyridyl Organomet. Chem., 2010, 36, xi–xv | xiii

Me6[14]dieneN4

5,7,7,12,14,14-hexamethyl-1,4,8,11-tetraazacyclotetradeca-4,11-diene Me6[14]N4 5,5,7,12,12,14-hexamethyl-1,4,8,11-tetraazacyclotetradecane 4,7-dimethyl-1,10-phenanthroline 4,7-Me2phen 3,4,7,8-Me4phen 3,4,7,8,-tetramethyl-1,10-phenanthroline Mes mesityl Mes* 2,4,6-tributylphenyl MeTHF methyltetrahydrofuran mcpba metachloroperbenzoic acid MLCT metal–ligand charge transfer MTO methylrhenium trioxide nap 1-naphthyl nb norbornene nbd norbornadiene NBS N-bromosuccinimide NCS N-chlorosuccinimide NCT neutron capture theory Neo neopentyl Np 1-naphthyl np3 N(CH2CH2PPh2)3 nta nitrilotriacetate OEP octaethylporphyrin OTf trifluoromethanesulfonate (triflate) OTs p-toluenesulfonate (tosylate) Pc phthalocyanin PES photoelectron spectroscopy PMDT pentamethylenediethylenetetramine pd pentane-2,4-dionate phen 1,10-phenanthroline pic pyridine-2-carboxylic acid Pin ( þ )-pinanyl Pmedta pentamethyldiethylenetriamine pp3 P(CH2CH2PPh2)3 [PPN] þ [(Ph3P)2N] þ py pyridine pydz pyridazine pz pyrazolyl R-PROPHOS (R)-( þ )-1,2-bis(diphenylphosphino)propane R,R-SKEWPHOS (2R,4R)-bis(diphenylphosphino)pentane RDF radial distribution function ROMP ring opening metathesis polymerisation sal salicylaldehyde salen N,N0 -bis(salicylaldehydo)ethylenediamine saloph N,N-bisalicylidene-o-phenylenediamine xiv | Organomet. Chem., 2010, 36, xi–xv

SCF TCNE TCNQ terpy tetraphos TFA tfbb tfacac THF thsa tht TMBD TMEDA tmp TMS tol TP TP* TPP Trip Triph triphos TRIR Tsi TTF vi WGSR XPS Xyl

self consistent field tetracyanoethylene 7,7,8,8-tetracyanoquinodimethane 2,20 ,200 -terpyridyl 1,1,4,7,10,10-hexaphenyl-1,4,7,10-tetraphosphadecane trifluoroacetic acid tetrafluorobenzobarrelene trifluoroacetylacetonato tetrahydrofuran thiosalicylate (2-thiobenzoate) tetrahydrothiophen NNN0 N00 -tetramethyl-2-butene-1,4-diamine (tmena) tetramethylethylenediamine 2,2,6-6-tetramethylpiperidino tetramethylsilane tolyl hydrotris(1-pyrazolyl)borate hydrotris(2,5-dimethylpyrazolyl)borate meso-tetraphenylporphyrin 2,4,6-triisopropylphenyl 2,4,6-(triphenyl)phenyl 1,1,1-tris(diphenylphosphinomethyl)ethane time resolved infrared (spectroscopy) tris(trimethylsilyl)methyl (Me3Si)3C tetrathiafulvalene vinyl water gas shift reaction X-ray photoelectron spectroscopy xylyl

Organomet. Chem., 2010, 36, xi–xv | xv

Non-transition metal N-heterocyclic carbene complexes Charlotte E. Willansa DOI: 10.1039/9781847559616-00001 Since their isolation in 1991,1 N-heterocyclic carbenes (NHCs) have become ubiquitous in organometallic chemistry. In more recent years investigations into the coordination of NHCs to other elements have expanded, and there are examples of their coordination to elements across the whole periodic table. This report gives an overview of NHC complexes of non-transition metal elements, ranging from the s-block elements, through the p-block and on to the lanthanides.

1.

Introduction

Due to their steric and electronic properties N-heterocyclic carbenes (NHCs) are a rapidly expanding area of research, particularly in transition metal chemistry.2–7 Beside their role as excellent ligands in metal-based catalytic reactions, organocatalytic carbene catalysis has emerged as an exceptionally fruitful research area in synthetic organic chemistry, and this area has recently been reviewed.8 Coordination of NHC ligands is not only limited to transition metals; there is an expanding range of examples in which NHCs have been used in combination with groups 1, 2, 13, 14, 15, 16, 17 and also the lanthanides. NHCs are Lewis base 2-electron donors and don’t necessarily require backbonding in their complexes, making them suitable for coordinating to a range of diﬀerent centres. This report describes many of the interesting NHC complexes formed with elements of the s-block and p-block and also the lanthanide ions, most of which have emerged during the past decade, particularly over the past few years. The carbene interactions range from being covalent to more ionic in nature, and can be evaluated by comparison of bond lengths and angles in the solid state, and in solution by chemical shift changes in the NMR spectra. This report focuses on singlet 5-membered NHCs, and the diagrams of the complexes are represented in the same way as the paper they correspond to. 2.

s-Block-carbenes

In 1998, Siebert and co-workers reported a 3-borane-1,4,5-trimethylimidazol-2-ylidene (1) which was formed through deprotonation of 3-borane-1,4,5trimethylimidazole using n-butyl lithium.9 The 13C NMR spectrum shows a shift in the C2 carbon from 213.7 ppm in the non-coordinating carbene to 191.3 ppm, which is consistent with the carbene interacting with a lithium metal centre. The authors described this as a Li(thp) þ 1 (thp=tetrahydropyran) salt, and the solid state structure reveals dimeric units of two carbene centres which are connected by two lithium ions (Fig. 1). The N1C2N3 angle of 104.0(2)1 is enlarged by approximately 2.51 when compared to the noncoordinating carbene, which can be explained by the interaction of the a

School of Chemistry, University of Leeds, Woodhouse Lane, Leeds, LS2 9JT, UK

Organomet. Chem., 2010, 36, 1–28 | 1 c

The Royal Society of Chemistry 2010

H3B N thp

Li N

C C

N Li

thp

N BH3 1

þ

Fig. 1 Diagram of (Li(thp) 1 )2 as observed in the solid state.

carbene centres with the lithium ions. For each of the lithium cations a short (Li1C2 0 2.169(5) A˚) and a longer (Li1C2 2.339(5) A˚) contact with the carbene C2 atoms are observed. The coordination around the lithium cation is completed by the thp molecule and a weak interaction with the anionic BH3 group. This was the ﬁrst reported example in which an NHC interacts with a group 1 metal. The ﬁrst example of lithium-NHC complexes, in which the lithium is coordinated only to carbon centres, was reported by Arduengo and coworkers.10 Stable NHCs were reacted with lithium 1,2,4-tris(trimethylsilyl) cyclopentadienide to give 2 (Fig. 2). A single crystal X-ray structure reveals a complex in which the lithium centre is coordinated in a Z5-fashion to the cyclopentadienyl ring, with a single s-interaction between the lithium and carbene centre. The lithium centre lies 2.155(4) A˚ from the carbene centre hence has a closer contact than in the previous example, possibly as a result of the carbene interacting with only one lithium centre. Polydentate ligands that combine NHC ligands with an anionic functional group, to stabilise higher oxidation states and Lewis acidic metal centres, have become a popular area of research over the past few years. One of the common routes used to generate NHCs is deprotonation at the C2 position of the imidazolium precursor using a base, of which potassium t-butoxide is the most commonly reported. As more N1- and N3-functional groups are incorporated into NHC ligands to improve catalysis and increase diversity, more complexes in which the group 1 metal cation is incorporated have been reported. In fact, group 1 NHC salts are becoming competitors to silver(I) adducts as eﬀective and less costly transmetallation reagents in systems where the non-coordinating NHC ligand is unavailable. One of the ﬁrst groups to demonstrate this was Arnold and co-workers. An

t-Bu N

N t-Bu Li SiMe3

Me3Si

SiMe3 2

Fig. 2

Diagram of 2 as observed in the solid state.

2 | Organomet. Chem., 2010, 36, 1–28

t-Bu NH2 2

t-Bu HN

2Br H

2 n-BuLi N t-Bu

N

THF

Li

Br

2

N t-Bu

N 3

Scheme 1 Synthesis of 3.

alkylammonium imidazolium bromide was deprotonated in sequential steps using n-butyl lithium, to aﬀord a lithium bromide adduct of an amine carbene (3) (Scheme 1).11 The solid state structure comprises a dimeric unit of two amine carbene groups which are connected by two lithium bromide bridging groups. The lithium-NHC distance of 2.197(4) A˚ is relatively long when compared to the previous structures, possibly as a result of the ligand being relatively bulky and too large for closer contact with the lithium ion. The ligand was successfully transferred on to lanthanide ions, an area which is discussed later in this review. The ﬁrst group 1 carbene complex with an N-bound anionic functional group was reported in 2004.12 An alkylamino carbene is readily deprotonated using n-butyl lithium to aﬀord 4 (Fig. 3). The solid state structure comprises a discrete dimer via bridging amido groups. Although there is severe distortion of the lithium-NCN bond (147.91 compared to the closer to linear 161.81 in 3), the lithium-NHC bond distance of 2.124(4) A˚ is still short, suggesting that the interaction is predominantly ionic.

N N t-Bu

t-Bu t-Bu N N Li Li N N t-Bu 4

180°

R N

Li

R 148°

Li N R

N N R

Fig. 3 Diagram of 4 as observed in the solid state. (Li-NCN bond of 1801 and distortion of a Li-NCN bond).

A few years later Arnold and co-workers also reported the synthesis of lithium complexes of the neutral and anionic salts of a tridentate amino biscarbene ligand (Scheme 2).13 Treatment of the cationic amino bis-imidazolium salt with three equivalents of n-butyl lithium aﬀords the lithium amino bis-carbene chloride complex (5). Deprotonation with four equivalents of n-butyl lithium aﬀords the lithium amide salt (6). Although the complexes were not characterised in the solid state, characteristic shifts in the multinuclear NMR spectra and elemental analysis are consistent with the lithium complexes being formed. NMR spectra of 5 suggest formation of a cluster of lithium chloride ions with lithium-NHC bonds (13C NMR: NCN 203.9 ppm) and NH-chloride bonding interactions. Following further deprotonation to form 6 the complex also retains lithium chloride and exhibits a similar C2 chemical shift (13C NMR: NCN 203.4 ppm). Organomet. Chem., 2010, 36, 1–28 | 3

N H

N

N 3 LiCl

N H2

N N Mes

3 Cl

H

3 n-BuLi

N

5

N Mes

4 n-BuLi

N Mes

H

N Mes

N Li

N

N 3 LiCl

N Mes

6

N Mes

Scheme 2 Synthesis of 5 and 6.

Arnold and co-workers also reported the deprotonation of alkoxy imidazolium iodides with n-butyl lithium to yield lithium alkoxide carbenes (Scheme 3).14 Single crystals of one of the complexes were grown from a diethyl ether solution, and revealed a dimer of LiL with lithium iodide incorporated to form a tetramer of lithium cations (7). The lithium-NHC bond distance of 2.131(6) A˚ is similar to that of the lithium amide carbene 4. Also as in 4 there is distortion of the lithium-NCN bond which has an angle of 152.31. The C2 carbon resonates at 200 ppm in the 13C NMR spectrum which is a relatively high-frequency, possibly as a result of the incorporated lithium iodide. The lithium salts were able to act as ligand transfer reagents and react with copper (II) chloride or triﬂate to aﬀord mono- or bis-substituted copper(II) alkoxy carbene complexes. Et2O t-Bu

H OH

2 N

I H

4 n-BuLi N Me

t-Bu

Li H O

I Li

THF N

Br Me N 2

7 Scheme 3 Synthesis of 7.

Similar alkoxide ligands enabled the ﬁrst crystallographic characterisation of a potassium-NHC complex.15 Previous attempts to isolate potassium-NHC complexes had led to migration of the N-hydrocarbyl group to the C2-position to form cyclic imines.16 By using O-functionalised N-alkyl arms, thermally stable potassium salts of NHCs were isolated. Following reaction of an alcohol-functionalised imidazolium iodide with excess potassium hydride in THF, single crystal X-ray crystallography revealed a polymeric structure, based on a network of potassium-NHC tetramers with cube-shaped K4O4 cores (8) (Scheme 4). Each potassium centre is four-coordinate, with three O atoms and a carbene from the ligands. The average potassium-NHC bond-length is 3.045 A˚ which is, as expected, 4 | Organomet. Chem., 2010, 36, 1–28

I

Me OH H Me

THF

N i-Pr

N

Me O Me

2 KH

K N i-Pr

N 8

Scheme 4 Synthesis of 8.

longer than when interacting with the smaller lithium centre. The backbone C4 and C5 atoms also display close intermolecular contact with potassium centres, which is noteworthy due to the number of ‘abnormal’ carbenes arising due to a [1,4] H shift.17 It is unlikely, however, that an ‘abnormal’ carbene is generated in this case, as upon dissolution in NMR solvent the original ‘normal’ carbene is generated. The 13C NMR spectrum shows a high-frequency C2 chemical shift of 208.4 ppm. When a bulky bis(adamantylethoxy) imidazolium salt was treated with potassium hydride the reaction did not aﬀord the expected potassiumcarbene.18 Instead, elimination of one alcohol arm produced a mono (adamantylethoxy) imidazole (9) (Scheme 5). Treatment of this with isopropyl iodide resulted in the alcohol imidazolium iodide salt, which undergoes deprotonation with lithium hexamethyldisilazide to aﬀord the lithium alkoxy carbene (10) which was characterised by mass spectrometry and multinuclear NMR spectroscopy. The C2 carbon in 10 resonates at 186.3 ppm in the 13C NMR spectrum, which is a signiﬁcantly lower frequency than the similar ligand in 7 which has lithium iodide incorporated into the structure.

Ad

OH

H

N

O

Ad

2 KH THF

N

i-PrI

Ad

Ad

OH N

I

H N i-Pr

OH N

MeCN

N

9

2 LiN(SiMe3)2 THF

Ad

O

Li(THF)n N i-Pr

N 10

Scheme 5 Synthesis of 9 and 10.

Danopoulos and co-workers reported on the preparation of NHC ligands with pendant indenyl and ﬂuorenyl groups.19 Deprotonation of the alkylindene or -ﬂuorene imidazolium salts with one equivalent of potassium hexamethyldisilazide leads to NHCs functionalised with neutral indene or ﬂuorene moieties (IndH-NHC and FlH-NHC). Further deprotonation with Organomet. Chem., 2010, 36, 1–28 | 5

a second equivalent of potassium hexamethyldisilazide aﬀords anionic indenyl and ﬂuorenyl NHC species [(Ind-NHC) K þ and (Fl-NHC) K þ ] (11) (Scheme 6). The 13C NMR spectra for (Ind-NHC)–K þ and (FlNHC)–K þ exhibit C2 resonances at 211.0 ppm and 206.0 ppm respectively, which is consistent with the previously described potassium-NHC complex 8. The structure of (Fl-NHC) K þ (11) was determined by single crystal X-ray diﬀraction, and comprises polymeric zigzag chains with potassium atoms and bridging ﬂuorenyl units. The potassium coordination sphere comprises two phenyl rings which sandwich the metal and a tethered NHC group, with a potassium-NHC bond distance of 2.896(5) A˚. Br

N DiPP

N

K

2 KN(SiMe3)2

H

benzene

N

N DiPP

11 Scheme 6 Synthesis of 11.

The ﬁrst magnesium-NHC adducts were reported by Arduengo and coworkers in 1993.20 Stable NHCs were reacted with diethyl magnesium to aﬀord the corresponding magnesium-NHC complexes in good yields (Scheme 7). Relative to the non-coordinating carbenes the C2 carbons are shifted substantially upﬁeld by 25–30 ppm in the 13C NMR spectra (R=adamantyl 180.1ppm, R=mesityl 194.8 ppm). Single crystal X-ray structures revealed that the N-adamantyl-substituted carbene complex has a monomeric solid state structure, while the less bulky N-mesityl complex is dimeric in the solid state. The dimeric structure is formed through bridging of one ethyl unit from each magnesium centre, and is likely as a result of the mesityl units providing insuﬃcient steric protection in the imidazole plane. The dimer comprises a magnesium-carbene bond distance of 2.279(3) A˚. R

N

N

Mg(C2H5)2

N R

N R R = 1-adamantyl, mesityl

Scheme 7

R Mg

C2H5

C 2H 5

Synthesis of magnesium-NHC complexes.

In 1998, the same group reported the synthesis of magnesium metallocene NHC complexes, in addition to metallocene NHC complexes of other group 2 elements.21 The adducts show an interesting trend in the nature of the metalcarbene bonds which increase as metal radii increase. The trends are reﬂected in both the solid state structures (metal-carbene bond length increases) and the NMR spectra (downﬁeld shift of C2) of the adducts (Table 1). In addition to mono-NHC complexes, the heavier alkaline earth elements (Sr and Ba) are capable of forming stable bis-NHC adducts. 6 | Organomet. Chem., 2010, 36, 1–28

Table 1 Comparison of the metal-NHC bond in group 2-NHC complexes

N

N

M(Cp*)2L

N

N

M

M = Mg, L = x M = Ca, L = OEt2 M = Sr, L = OEt2 M = Ba, L = THF2

Mg Ca Sr Ba Ligand

Cp*

Cp*

Metal-carbene bond length (A˚)

13

2.194(2) 2.562(2) 2.861(5) 2.951(3)

185.7 196.2 198.2 203.5 213.7

C NMR: C2 (ppm)

In 2001, Schumann and co-workers reported a similar set of metallocene complexes with 1,3-di-iso-propyl-4,5-dimethylimidazoly-2-ylidene.22 X-ray crystallography and NMR studies conﬁrmed a similar trend between metalcarbene bond strength and alkaline earth metal as that found by Arduengo and co-workers. Additionally they showed that as the steric bulk of the cyclopentadienyl ligand increases the metal-carbene bond distance is elongated. The ﬁrst group 2 amido NHC complex was reported in 2004.12 An amino carbene is readily deprotonated by half an equivalent of dimethylmagnesium to aﬀord MgL2 (12) (Scheme 8). Despite the diagonal relationship between Li þ and Mg2 þ there is virtually no distortion about the M-NCN bond as seen in complex 4. The magnesium bis-(amido NHC) chelate has short magnesium-N bond lengths and average magnesium-NHC bond distances (2.263(2) A˚ and 2.2697(16) A˚). An anionic aryloxy-bound NHC ligand was reported by Zhang and coworkers (Scheme 9).23 Initially, attempts were made to isolate the sodium aryloxy NHC through deprotonation of the imidazolium-substituted phenol using sodium hexamethyldisilazide. A monoanionic carbene ligand is generated at 78 1C, though on warming to room temperature a 1,2-migration rearrangement results in an aryloxy substituted imidazole. The sodium salt was therefore generated in situ at 78 1C and transferred on to magnesium forming an [ML]2 dimer with bridging aryloxy groups (13). The

t-Bu

t-Bu 2

NH N

MgMe2 N t-Bu

N

N

t-Bu N

THF

Mg N t-Bu N

N

t-Bu

12 Scheme 8 Synthesis of 12.

Organomet. Chem., 2010, 36, 1–28 | 7

N N ONa

t-Bu

t-Bu

N

(2,4,6-Me3C6H2)MgBr

N

t-Bu

Mg O

O

t-Bu

Mg t-Bu

t-Bu

N N 13

Scheme 9 Synthesis of 13.

magnesium centre is tetrahedral with two bridged oxygen atoms, an NHC carbon and a mesitylene carbon. The closest magnesium-carbon interaction of 2.224(4) A˚ is with that of the NHC. Magnesium complexes of a tridentate monoanionic bis-carbene ligand have recently been reported (Scheme 10).24 Treatment of the cationic amino bis-imidazolium salt with methylmagnesium chloride leads to the formation of a magnesium chloride adduct (14), with a C2 chemical shift of 194.0 ppm in the 13C NMR spectrum. The remaining amino proton can be removed through heating the magnesium chloride adduct in THF to give Mg2(L)Cl3 NH

N N Mes

3 Cl

N H2 H

N

N 3 MeMgCl

N Mes

H

N Cl Mg Mg

N Mes

Cl Cl

THF 80 °C

Cl

Mg

Cl

14

N Mes

Cl

N N N Mes

N Mg

Mg Cl

Cl

2 LiN(SiMe3)2 N Mes N

3 LiN(SiMe3)2

N N N Mes

Mg

N Mes

N Mg

Mg

Cl N(SiMe3)2 Cl

KN(SiMe3)2

N Mg

N

Cl

15

N Mes

Cl N(SiMe3)2 N(SiMe3)2

Scheme 10 Magnesium complexes of a tridentate bis-NHC ligand.

8 | Organomet. Chem., 2010, 36, 1–28

N Mes

(15). Deprotonation of 14 can also be achieved by reaction with lithium hexamethyldisilazide. Following deprotonation of the amino group the 13C NMR spectrum exhibits a resonance at 182.3 ppm for the C2 carbon, suggesting a stronger interaction of the carbene with the magnesium centre. A number of heavier amido group 2-NHC complexes have been reported in which the NHC ligand does not posses an anionic tether.25 These are prepared by addition of the group 2-amide to the corresponding imidazolium salt, or addition of the stable NHC to a solvent-free group 2-amide. Solid state studies are consistent with the formation of monomeric threecoordinate group 2 species in which the NHC binds through donation of the lone pair to the electrophillic metal centre. Multinuclear NMR studies suggest that this coordination is retained in solution, though the labile NHC ligand is readily displaced upon reaction with protic substrates or other Lewis bases. The metal-carbene bond lengths and C2 chemical shifts in the 13 C NMR spectra are similar both in values and trends to those reported in the previous group 2 metal-carbene complexes (Table 2).

Table 2 Comparision of the metal-NHC bond in group 2-NHC complexes

N N Ar

Ar

Cl N

H

Ar M = Ca, Sr, Ba

Ca Sr Ba

N

2 M[N(SiMe3)2]2

Ar M

N(SiMe3)2

N(SiMe3)2

Metal-carbene bond length (A˚)

13

2.6285(16) 2.731(3) 2.915(4)

194.1 199.0 Not observed

C NMR: C2 (ppm)

NHCs have been shown to form stable complexes with many of the s-block metals in both groups 1 and 2. Complexes both with and devoid of an anionic tether have been reported and demonstrate that a tether does not necessarily bring the carbene into closer proximity with the metal centre. Other factors such as steric bulk and other substituents should be considered. In general, moving from group 1 to group 2 decreases the metalcarbene bond distance as the Lewis acidity of the metal centre increases. Moving down the groups increases the metal-carbene bond distance as the metal radius increases. 3.

p-Block-carbenes

In 2002, Jones and co-workers reported bidentate NHC complexes of group 13 trihydrides and trihalides.26 It was found that the MH3 fragments form mono-NHC four-coordinate complexes, with the ligand bridging two metal centres, whereas with InBr3 and TlCl3 the bis-NHC ligand coordinates in a chelating fashion (Scheme 11). This suggests that the halides are stronger Lewis acids than the hydrides, and highlights the ability of the larger metals Organomet. Chem., 2010, 36, 1–28 | 9

t-Bu N

N

N t-Bu

N

H 3M

MH3(NMe3)

t-Bu N

N

N t-Bu MH3

MX3

M = In, X = Br M = Tl, X = Cl

N

M = Al, Ga, In N

X

N

N M X N t-Bu X t-Bu

Scheme 11 Reaction of group 13 metals with a bis-NHC ligand.

to achieve higher coordination numbers. Solid state structures of aluminium-mono-NHC, indium-mono-NHC and indium-bis-NHC exhibit metal-carbene bond distances of 2.067(2) A˚, 2.3069(16) A˚ and 2.233(6) A˚ respectively. The same group also reported on the reaction of GaI and InCl with mono-NHCs.27 A product was isolated using GaI only when it was reacted with a bulky NHC, and resulted in an anionic complex and imidazolium cation, with the imidazolium proton likely being abstracted from the solvent (16) (Fig. 4). The gallium-NHC bond length of 2.070(7) A˚ is very similar to the GaH3(NHC) discussed previously. Reaction of a less sterically hindered NHC with InCl aﬀords the unusual oxo-bridged dimer 17, likely as a result of adventitious oxygen in the reaction mixture, as a product could not be isolated when the reaction was done under strict anaerobic conditions. The average indium-NHC bond length of 2.234 A˚ is similar to the InBr3(NHC)2 discussed previously, and shorter than the InH3(NHC).

I I Ga Ga I I

N

N

I

N

Ar N

Ar N Ar = 2,6-(iPr)2C6H3 16

N Cl In O In Cl Cl N N N N 17

Fig. 4 Anionic components of 16 and 17.

In 2004 an indium-NHC complex was reported which was prepared from an air stable imidazolium salt precursor.29 Reaction of one equivalent of imidazolium salt with InMe3 aﬀords (NHC)InMe2Cl 18 (Scheme 12). The indium-NHC bond distance of 2.267(2) A˚ is shorter than that of InH3(NHC), though longer than the bis-NHC InBr3(NHC)2. Mono-triﬂate (19) and bis-triﬂate (20) complexes can also be prepared by treatment of 18 with trimethyl silyl triﬂate and treatment of 19 with triﬂic acid respectively. 10 | Organomet. Chem., 2010, 36, 1–28

Cl N

InMe3 CH2Cl2

N

N

N In

Me TMS-OTf N

Me

Cl

18

N In

Me

Me

OTf

19

HOTf

N

N In

Me

OTf OTf 20

Scheme 12 Indium-NHC complexes.

The indium-NHC bond distance of 2.264(2) A˚ in 19 barely changes from the trichloride adduct 18, though the indium-NHC bond distance of 2.183(2) A˚ in 20 is the shortest of the three complexes, reﬂecting the dicationic nature of the metal centre. The ﬁrst example of a thallium-NHC complex was reported by Meyer and co-workers in 2003.28 A tris-NHC ligand, of which there were previously no metal complexes reported, was reacted with Tl(OTf) in THF at –35 1C, resulting in a thallium-tris-NHC complex (21) (Scheme 13). The highly temperature sensitive complex was characterised by single crystal Xray diﬀraction which conﬁrmed the tridentate conformation. The three carbene ligands are not symmetrically bound to the thallium as they exhibit slightly diﬀerent thallium-NHC bond distances, with an average bond length of 2.952 A˚. t-Bu N

t-Bu N

N Tl(OTf)

N

t-Bu N N

THF, -35 °C

N

t-Bu N

Tl N

N

t-Bu t-Bu N

N 21 Scheme 13 Synthesis of 21.

The 13C NMR spectra of the group 13-NHC adducts exhibit C2 resonances that are upﬁeld from the uncoordinated carbene and downﬁeld from Organomet. Chem., 2010, 36, 1–28 | 11

that of the imidazolium salt. This, and the relatively long M-NHC bond lengths, suggest that the electronic structures are intermediate between those of the stable free NHC and the imidazolium ion. The ﬁrst group 14 adduct of an NHC was reported by Arduengo and coworkers in 1993.30 Germanium diiodide was reacted with a stable carbene to aﬀord the germanium-NHC adduct 22 (Fig. 5). The C2 carbon is shifted upﬁeld in the 13C NMR spectrum by 60.88 ppm relative to the uncoordinated carbene, from 219.69 ppm to 158.81 ppm. The most interesting feature of this compound is that the geometry around the germanium centre is pyramidal, with a germanium-NHC bond distance of 2.102(12) A˚. This is quite diﬀerent to the geometry of a germaethene where the germanium and carbon atoms exhibit trigonal planar coordination and a shorter germanium-carbon bond length of 1.803 A˚.31 The length and orientation of the germanium-NHC bond in 22 and the NMR spectroscopy data indicate a highly polarised structure rather than a double bond.

N

N Ge

I

I 22

Fig. 5

Diagram of 22 as observed in the solid state.

A similar pyramidal structure was reported two years later when a stable carbene was reacted with bis(2,4,6-triisopropylpheny1)stannylene (Scheme 14).32 The C2 carbon of the complex 23 is shifted upﬁeld in the 13C NMR spectrum compared to the uncoordinated carbene, though only by 28 ppm which is signiﬁcantly less than in the germanium adduct 22. The solid state structure has a tin-NHC bond length of 2.379(5) A˚.

N

N

Sn(Ar)2

N

N Sn

Ar = 2,4,6-(iPr)3C6H2

Ar

Ar 23

Scheme 14 Synthesis of 23.

In the same year Kuhn and co-workers reported a series of silicon- and tin-NHC complexes (Scheme 15).33 The pentacordinated silicon and tin structures 24 and 28 were determined by single crystal X-ray crystallography, and possess metal-NHC bond lengths of 1.911(7) A˚ and 2.179(3) A˚ respectively. The monomeric stannylene complex 29 was also 12 | Organomet. Chem., 2010, 36, 1–28

R N

N R

R N

SiCl4 24

Ph2SiCl2 R N

SiPh2Cl2 Ph2SnCl2 26

R N

25

Me3SiI R N

N R

SiMe2Cl

Me2SiCl2

SiCl4

N R

N R

R = Me, Et, iPr

R N

SiMe3 I 27

SnCl2

R N

N R

N R

N R

SnPh2Cl2

SnCl2

28

29

Scheme 15 Preparation of silicon- and tin-NHC complexes.

characterised by single crystal X-ray crystallography and has a tin-NHC bond distance of 2.290(5) A˚. As in previous cases the geometry around the metal centre is pyramidal. The single crystal X-ray structure of the silylene-NHC adduct 30 also comprises a pyramidal geometry around the silicon, with a long siliconNHC bond of 2.162(5) A˚ (Fig. 6).34 The NMR spectral data indicate signiﬁcant C þ –Si bond polarity and DFT calculations are also consistent with this. The single crystal X-ray structure of 31 is also consistent with a zwitterionic species made up from a partially cationic carbene and a partially anionic stannylene.35 Compound 32 was the ﬁrst example of an NHCstabilised transient diorganogermylene and exhibits the expected pyramidal geometry around the germanium centre.36 The germanium-NHC bond distance is 2.078(3) A˚.

t-Bu

N

t-Bu

N

N Si N N

N t-Bu t-Bu 30

N

N

Sn N N

N

31

32

Ge

Mes Mes

N

Fig. 6 Pyramidal geometries of group 14-NHC complexes.

The same group also described the synthesis and structural characterisation of a number of NHC-stabilised germanium(II) compounds derived from the dichloro derivative 33 via substitution chemistry (Fig. 7).37 They demonstrated that the length of the germanium-NHC bond is signiﬁcantly Organomet. Chem., 2010, 36, 1–28 | 13

N

N Ge

Cl

Cl 33

Fig. 7 Diagram of 33.

inﬂuenced by the p-donating ability of the substituents on germanium. For example, when one of the chloride substituents in 33 is replaced by triﬂate the germanium-NHC bond is reduced in length from 2.106(3) A˚ to 2.068(2) A˚. The observations are consistent with the germanium having a d þ charge due to the electron-withdrawing triﬂate group. In 2006, Jones and co-workers reported the reaction of anionic gallium(I)-NHC analogues with the heavier group 14 (E) alkene analogues (Scheme 16).38 The complexes formed exhibit long gallium-E bonds with that of E=Sn being 2.7186(6) A˚ in the mono-NHC and an average of 2.6485 A˚ in the dianionic bis-NHC. The nature of the gallium-E bond was probed by DFT calculations and was shown to be closely related to the neutral NHC adducts of group 14 dialkyls. Various other group 14-NHC analogues have also been reported and the area has recently been reviewed.39–41

2

Ar N GaN Ar [K(tmeda)]+

{E[CH(SiMe3)2]2}2 E = Ge, Sn R = CH(SiMe3)2 Ar = 2,6-iPrC6H3

E = Sn Ar N Ga N Ar E K(tmeda) R R

Ar N Ga N Ar R Sn R Ar N Ga N Ar

-KR Ar N Ga N Ar Ar

R

Sn

[K(tmeda)]2+

K(tmeda)

N Ga N Ar

Scheme 16 Group 14 adducts of gallium(I) NHC analogues.

The ﬁrst group 15 adduct of an NHC was reported by Arduengo and coworkers in 1997.42 A stable carbene was reacted with pentaphenylcyclopentaphosphine to form the carbene-phosphonidene 34 (Scheme 17). The C2 chemical shift in the 13C NMR spectrum appears at 169 ppm, 44.7 ppm upﬁeld from the uncoordinated carbene, and the single crystal X-ray structure shows a phosphorus-NHC bond length of 1.794(3) A˚. This is relatively long for typical phosphaalkenes indicating that the bond is highly polarised, which is also indicated in the NMR data.43 14 | Organomet. Chem., 2010, 36, 1–28

Ph

N

N

Ph P P + 0.2 Ph P P P Ph Ph

THF N

N

N P

N P

Ph

Ph

34 Scheme 17 Synthesis of 34.

Similar phosphorus- and arsenic-NHC adducts have also been reported by Arduengo and co-workers with varying N-substituents (Me, Mes), C3 and C4 substituents (H, Me) and group 15 substituents (Ph, CF3, C6F5). The high ﬁeld 31P NMR chemical shift, the upﬁeld shift of the C2 carbon in the 13C NMR spectra and the long phosphorus- or arsenic-NHC bond length are consistent with all the adducts being highly polarised.42,44 The same group reported the ﬁrst carbene complex of a phosphorus(V) centre.45 A stable carbene was reacted with phenyltetraﬂuorophosphorane aﬀording 35 (Scheme 18). The 13C NMR signal for the C2 carbon appears at 164.7 ppm, 55 ppm upﬁeld from the uncoordinated carbene. The single crystal X-ray structure comprises an octahedral geometry around the phosphorus, with a phosphorus-NHC bond length of 1.91(4) A˚. The longer phosphorus-NHC bond length and larger upﬁeld shift of the C2 resonance compared to compound 34 is consistent with 35 being even more polarised.

Mes N

F

+ N Mes

F F P F

Mes FF N P N F F

THF

Mes 35 Scheme 18 Synthesis of 35.

Kuhn and co-workers reported the reaction of a stable carbene with POCl3 to yield the [(NHC)POCl2][Cl] salt which, following partial hydrolysis, aﬀords 36 (Scheme 19).46 The solid state single crystal X-ray structure comprises a phosphorus-NHC bond length of 1.843(2) A˚, which is consistent with the P–C bond being polarised.

Cl N

N

N

N

O P Cl Cl

N N

O P Cl O

36 Scheme 19 Synthesis of 36.

Organomet. Chem., 2010, 36, 1–28 | 15

Clyburne and co-workers investigated the reaction of a stable carbene with diazoalkanes to aﬀord azines.47 Reaction with diazaﬂuorene aﬀords 37, and reaction with diphenyldiazamethane aﬀords 38 (Fig. 8). Both the CN bond lengths in both compounds (1.325(3) A˚ and 1.304(3) A˚ in 37 and 1.312(3) A˚ and 1.294(3) A˚ in 38) are longer than typical CQN double bonds (ca. 1.29 A˚), with the asymmetry suggesting that the compounds are polarised. The longer C-N bond length in both compounds is between the nitrogen and the NHC fragment, and compound 37 appears to be the most ionic of the two, likely as a result of increased delocalisation.

Mes N N N N Mes

Mes N N N N Mes

Mes N N N N Mes

37

38 Fig. 8 NHC-stabilised azines.

In 2005, Bielawski and co-workers reported the reaction of a stable carbene with an aryl azide to give the triazine 39 (Scheme 20).48 Both the Eand the Z- isomers were identiﬁed in the solid state and were found to have nitrogen-NHC bond lengths of 1.339(3) A˚ and 1.330(3) A˚ respectively, hence is even more polarised than compound 37.

N3 t-Bu N

N t-Bu

t-Bu N THF

N t-Bu N 39

N2Bn

Scheme 20 Synthesis of 39.

While many stable carbenes tend to be unreactive towards oxygen in the absence of a catalyst, Denk and co-workers found that a stable carbene could be oxidised to the urea 40 in the presence of a catalyst or by reaction with NO (Scheme 21).49 The solid state structure exhibits an oxygen-NHC bond length of 1.237(3) A˚, and the C2 carbon resonates at 152.7 ppm in the 13C NMR spectrum, 60.2 ppm upﬁeld from the uncoordinated carbene. The bond length is elongated compared to a typical CQO double bond NO t-Bu N

N t-Bu

t-Bu N

N t-Bu O 40

Scheme 21 Synthesis of 40.

16 | Organomet. Chem., 2010, 36, 1–28

(ca. 1.20 A˚) and the large shift of the C2 carbon in the 13C NMR spectrum is consistent with a polarised compound. Reaction of a stable carbene with SCl2 results in the hypervalent sulfur compound 41 (Scheme 22).50 The sulfur-NHC bond length of 1.732(3) A˚ is signiﬁcantly longer than that of a typical CQS double bond (ca. 1.60 A˚) indicating a highly polarised compound.

SCl2 N

N

N Cl

N S 41

Cl

Scheme 22 Synthesis of 41.

Bildstein and co-workers prepared an N-iminoisopropyl NHC ligand that forms adducts with sulfur and selenium through reaction of them in their elemental form (42) (Fig. 9).49 The solid state structures exhibit a sulfurNHC bond length of 1.681(2) A˚, and a selenium-NHC bond length of 1.840(2) A˚. The C2 resonance in the 13C NMR spectrum where E=S is 161.0 ppm, an upﬁeld shift of 55.3 ppm compared to the uncoordinated carbene, and the C2 carbon when E=Se resonates at 153.8 ppm, an upﬁeld shift of 62.5 ppm. This indicates that on going down Group 16 the E–NHC bond becomes more polarised. The sulfur-NHC bond does not appear to be as polarised as in the hypervalent sulfur compound 41, in which the sulfur atom also possesses electron withdrawing groups. The seleniumNHC bond in 43 also appears to be highly polarised, with a bond length of 1.884(9) A˚.51 This is signiﬁcantly longer than a typical SeQC double bond (SeQCQSe=1.698 A˚)52 and nearer to that of an Se–C single bond (ca. 1.94 A˚).

N N

N E

N E = S, Se

42 Fig. 9

N Se 43

Sulfur- and selenium-NHC adducts.

The tellerium-NHC adduct 44 was isolated and described as having a mesomeric structure (Fig. 10).53 This is conﬁrmed by the dramatic upﬁeld shifts in the 13C NMR spectrum and also the 125Te NMR spectrum. The solid state structure exhibits a long tellerium-NHC bond length of 2.087(4) A˚. One of the ﬁrst Group 17-NHC adducts was isolated by Arduengo and co-workers in 1991.54 A stable carbene and iodopentaﬂuorobenzene exist in Organomet. Chem., 2010, 36, 1–28 | 17

N

N

N

Te

N Te

44 Fig. 10 A stable tellurium-NHC adduct.

equilibrium with the adduct in solution as evidenced by averaged NMR chemical shifts in the presence of excess of either reagent (Scheme 23). After several hours in solution at room temperature it appears that the I–C bond is cleaved resulting in pentaﬂuorobenzene and the iodo-imidazolium ion. Compound 45 was isolated and characterised by single crystal X-ray diffraction and comprises an iodine-NHC bond distance of 2.754(3) A˚ and iodine-C(phenyl) bond distance of 2.159(3) A˚. Ad N

Ad N

THF

+ IC6F5

F

F

I N Ad

N Ad

F F

F

45 Scheme 23 Synthesis of 45.

From the above reaction it can be considered that the lone pair of an NHC interacts with the s*-orbital of a halogen to generate a reverse ylide (Scheme 24). The resulting product is ionic, similar to hydroimidazolium salts. Crystalline adducts of NHCs with iodine, bromine and chlorine have all been reported.

R

R N

R

N R

R

R N

R

N R

X2

R

R N

R

N R

X

X X

X

Scheme 24 Interaction of an NHC with a halogen.

Reaction of a stable carbene with iodine results in compound 46, which can be considered an isolated transition state which models the nucleophilic attack of the carbene on the iodine molecule (Scheme 25).55 The iodineiodine bond is signiﬁcantly lengthened and the carbon iodine bond distance of 2.104(3) A˚ is slightly elongated when compared to that of iodoarenes. The authors report that protic solvents promote ionic dissociation to the 2-iodoimidazolium ion which is isoelectronic with the tellerium adduct 44. A stable carbene reacts with 2-iodo-1,3-dimesityl imidazolium salt to form the bis(NHC) iodine complex 47 (Scheme 26).56 The central C-I-C unit 18 | Organomet. Chem., 2010, 36, 1–28

N

N

I2

I

I

N

N

46 Scheme 25 Synthesis of 46.

is almost linear with a small diﬀerence between the two iodine-NHC bond distances (2.286 A˚ and 2.363 A˚). In solution the iodine anion does not appear to compete with the carbene for complexation to the 2-iodoimidazolium salt. This structure is in contrast to the previous compound 46 in which the iodine does not appear to exchange between cations. Mes N

Mes N +

N Mes

I

Mes N

THF

Mes N

I N N I Mes Mes 47

N I Mes Scheme 26 Synthesis of 47.

Reaction of a stable carbene with sulfuric chloride results in abstraction of the chloride cation to give the adduct 48 (Scheme 27).57 The C2 carbon resonates at 133.05 ppm in the 13C NMR spectrum and the solid state structure exhibits a chlorine-NHC bond distance of 1.696(9) A˚. The ﬂuorine-NHC analogue was prepared by Kuhn and co-workers by reaction of the stable carbene with SO2F2.50 The solid state structure exhibits a ﬂuorineNHC bond distance of 1.291(14) A˚.

N N

SO2Cl2

N Cl

SO2Cl

N 48 Scheme 27 Synthesis of 48.

Reports from Kuhn and co-workers identiﬁed the reaction of stable carbenes with 1,2-dichloroethane to yield 2-chloro-1,3-disubstituted imidazolium chloride salts.58 The versatility of these salts has been demonstrated by Ishikawa and co-workers.59 Due to its strong electrophilicity, 2-chloro-1, 3-dimethylimidazolium chloride can be used in chlorination, oxidation, reduction and rearrangement reactions, in addition to being used as a dehydrating agent. Jones and co-workers investigated reactions of the stable carbene 1,3bis(2,4,6-trimethylphenyl)imidazol-2-ylidene with a series of halide sources Organomet. Chem., 2010, 36, 1–28 | 19

Mes Br N H + N Mes

Mes Br N Br N Mes

51

50

(CH2Br)2

Mes N

Mes Cl N Cl N Mes

C2Cl6

N Mes

49

Br2 Mes Br N Br N Mes 50

Scheme 28 Reactions of 1,3-bis(2,4,6-trimethylphenyl)imidazol-2-ylidene with a series of halide sources.

(Scheme 28).60 Compound 49 displays a chlorine-NHC bond length of 1.677(5) A˚ which is comparable to that of 48. The C2 carbon resonates at 135.7 ppm in the 13C NMR spectrum. Reaction with 1,2-dibromoethane yields the bromo-analogue (50) of the product reported by Kuhn and coworkers through reaction with 1,2-dichloroethane. Following the reaction by NMR spectroscopy using 1 molar equivalents of 1,2-dibromoethane reveals that the 2-hydroimidazolium salt (51) is also formed in the reaction. Reaction with dibromine yields compound 50 only. The bromine-NHC bond distance is 1.861(4) A˚ and the C2 carbon resonates at 126.4 ppm in the 13 C NMR spectrum. Kuhn and co-workers reported on the syntheses and structures of some 2-bromo-1,3-diisopropyl-4,5-dimethylimidazolium derivatives.61 The stable carbene reacts with bromine to give the bromine adduct 52 (Fig. 11). The bromine-NHC bond length of 1.881(5) A˚ is as expected for imidazolium ions. By use of excess tetrabromomethane instead of bromine the CBr4 adduct 53 was isolated, and the bromotellurate salt 54 is obtained by reaction with TeBr4. Incorporation of the tetrabromomethane molecule into the unit cell does not signiﬁcantly inﬂuence the structure of the Br–Br–NHC unit, though coordination of the bromide ion at TeBr4 lowers the nucleophilic character of the anion and the closest Br–Br bond length increases signiﬁcantly.

N

N Br Br

N

N

52

53

N Br Br CBr4

Fig. 11 2-bromo imidazolium derivatives.

20 | Organomet. Chem., 2010, 36, 1–28

Br N

54

TeBr5

In general, on moving down the groups of the p-block, the M-NHC bond length increases as the M centre becomes softer. On moving across the rows the M-NHC bond length decreases and the bonding becomes more ionic, with the bonding of X–NHC (X=halogen) being as expected for imidazolium ions. Other factors that inﬂuence the bonding are the sterics and electronics of substituents both on the NHC fragment and the M centre. 4.

f-Element-carbenes

The ﬁrst lanthanide-NHC complexes were isolated by Arduengo and coworkers in 1994.62 A stable carbene displaces THF in bis(pentamethylcyclopentadieny1)-samarium-THF to form the samarium(II)-NHC complex 55 (Scheme 29). The addition of a second equivalent of NHC resulted in the isolation of the bis(NHC) adduct 56. Compound 56 was characterised in the solid state by single crystal X-ray diﬀraction and exhibits samarium–NHC bond distances of 2.837(7) A˚ and 2.845(7) A˚, which are longer than the M–C bond in s-bonded alkyl lanthanide complexes.

O

N +

N Sm

Sm

N

N 55

N

N

N

N

Sm N N 56 Scheme 29 Synthesis of 55 and 56.

Addition of the same NHC to Eu(thd)3 (thd=tetramethylheptanedioate) aﬀords the europium(III) adduct Eu(thd)3(NHC). The europium-NHC bond distance of 2.663(4) A˚ is shorter than that of the samarium(II) complex and is consistent with the higher oxidation state of the lanthanide centre. The yttrium(III) analogue was also prepared and characterised by NMR spectroscopy. The C2 carbon resonates at 199 ppm in the 13C NMR spectrum, with a 1JYC coupling constant of 33 Hz. This indicates that the NHC remains bound to the metal centre in solution and does not dissociate on the NMR timescale. The synthesis of Y[N(SiHMe2)2]3(NHC)x (NHC=1,3-dimethylimidazolin-2-ylidene, x=1, 2) was achieved by displacement of THF ligands in Y[N(SiHMe2)2]3(THF)2 by the stable carbene.63 The structural data reveal Organomet. Chem., 2010, 36, 1–28 | 21

that the carbene ligands aﬀect the coordination mode of the bis(dimethylsilyl)amide ligands by forcing them to form b-H-yttrium agostic interactions. Organometallic uranyl complexes of monodentate NHC adducts have also been reported (Fig. 12).64–66

Mes Mes O Cl N N U N Cl O N Mes Mes X = H, Cl

X X

X

N

X

(Me3Si)N

X

N U

U

N N

N(SiMe3) N(SiMe3)

X = I, O Fig. 12 NHC adducts of uranium.

NHCs are Lewis base 2-electron donors, but have no necessary requirements for back-bonding, making them perfectly suited for the f-elements. NHCs are relatively soft ligands, thus a tethered anionic moiety represents a viable method for covalent attachment to the hard electropositive metal centre. The anionic component forms a strong covalent interaction with the metal centre and brings the NHC into close proximity. As previously discussed, NHC ligands with an OH or NH tether may be deprotonated to form s-block alkoxide or amido salts, in which the NHC group in the chelate binds to the metal centre. These adducts can be used as transmetallation agents for early metal and f-element complexes. Treatment of a potassium alkoxide with UI3(THF)4 in THF aﬀords tetravalent uranium(IV) complexes 57 or 58, with the outcome being dependent upon stoichiometry (Scheme 30).67 Variable temperature NMR studies on 57 display a ﬂuxional process in solution which is assumed to be the exchange of free for uranium-coordinated NHC. The same potassium alkoxide adduct reacts with uranyl dichloride [UO2Cl2(THF)2]2 to aﬀord [UO2(L)2],68 and the lithium amide adduct aﬀords the analogous [UO2(L2)].12

i-Pr

i-Pr

N N N i-Pr

N O U O

N N N i-Pr

O

O N

UI3(THF)4

2.5 KL THF

i-Pr

N N i-Pr

N

O

N 3 KL THF

O U O N

57

I

N i-Pr 58

Scheme 30 Synthesis of 57 and 58.

The imidazolium protons and the alcohol and amino protons are sufﬁciently acidic that monoprotonated proligands can be used in transamination reactions to aﬀord f-element functionalised-NHC adducts. 22 | Organomet. Chem., 2010, 36, 1–28

Transamination of the lithium bromide NHC amine 3 (Scheme 1) with Sm[N(SiMe3)2]3 proceeds cleanly to aﬀord the dark yellow air-sensitive Sm(L)[N(SiMe3)2]2 (59) (Scheme 31).11 The lithium bromide adduct gives better product yields than the free base, and no lithium or bromide ions remain in the coordination sphere of the lanthanide metal. The samarium– NHC bond length of 2.588(2) A˚ is shorter than those of the monodentate lanthanide-NHC adducts. The yttrium(III),11 europium(III)68 and neodymium(III)69 analogues have also been isolated. The yttrium-NHC bond distance of 2.501(5) A˚ is even shorter, reﬂecting the smaller size and increased Lewis acidity of the yttrium(III) centre, and the adduct exhibits a large 1JYC coupling constant of 54.7 Hz in solution. The larger neodymium (III) centre renders complex 62 signiﬁcantly more air-sensitive than the others. The free base could not be used to prepare NHC adducts of the larger cerium(III) metal, hence the cerium(III) analogue of 59 could only be prepared via the lithium bromide adduct.70 Ligand exchange between the product and lithium bromide resulted in the bridged complex {Ce(L)[N (SiMe3)2](m-Br)}2. Presumably this is due to the lower Lewis acidity of cerium(III) enabling lithium to compete for the amide ligand.

t-Bu HN

Li

Br

t-Bu N

2

N t-Bu

N

+

PhMe

Ln[N(SiMe3)2]3

Reflux

Ln

N

3

N(SiMe3)2 N(SiMe3)2

N t-Bu

Ln = Sm (59), Y (60), Eu (61), Nd (62) Scheme 31 Synthesis of lanthanide amide-NHC complexes.

The transamination of the anionic amido ytterbium complex LiYb (NiPr2)4 with aryloxo-functionalised NHC imidazolium salt precursors affords bis-aryloxo-NHC monoamido ytterbium(III) complexes (Scheme 32)71 The complexes are isostructural with one another in the solid state and exhibit average ytterbium–NHC bond distances of 2.487 A˚ (R=Me) and 2.535 A˚ (R=iPr). These are comparable to the monodentate lanthanide(III)NHC bond lengths where the ligand possesses and amido tether, though a direct comparison with like-for-like metal centre is not possible. The longer ytterbium-NHC bond length where R=iPr is likely as a result of the increased bulk of the ligand.

OH t-Bu

Cl N

t-Bu

N R

NiPr2

t-Bu

i

LiYb(N Pr2)4

Yb

O

2

THF R = iPr, Me

t-Bu

N

N R

Scheme 32 Synthesis of bis-aryloxo-NHC monoamido ytterbium complexes.

Organomet. Chem., 2010, 36, 1–28 | 23

Treatment of the yttrium(III) adduct 60 with potassium naphthalenide in dme-diethyl ether mixture results in deprotonation of the C4 carbon and migration to aﬀord the ‘abnormal’ carbene complex 63 (Fig. 13).72 The C2 binding carbon migrates from the yttrium(III) centre to the incorporated potassium(I) cation. The C4 carbanion forms a short bond with the yttrium(III) centre in the solid state (2.447(2) A˚) and exhibits a large 1JYC coupling constant of 62 Hz in solution. Complex 63 may be quenched with a variety of electrophiles. For example, reaction with Me3SiCl silylates the NHC backbone to aﬀord 64. O

N (Me3Si)2N (Me3Si)2N

Y

O

K

N

t-Bu N

t-Bu N

N t-Bu O

N

K

N(SiMe3)2 Y

N

O

Y

N(SiMe3)2 N t-Bu

N

N(SiMe3)2

N(SiMe3)2

Me3Si

63

64

Fig. 13 ‘Abnormal’ lanthanide-NHC complexes.

Transamination between the lithium salt of the tridentate amino biscarbene (6) and Y[N(SiMe3)2]3 aﬀords an yttrium bis-NHC complex Y(L)[N(SiMe3)2]Cl.13 The 13C NMR spectrum exhibits a C2 resonance at 194.3 ppm with a 1JYC coupling constant of 48.0 Hz, which is slightly lower than that of the mono-NHC yttrium complex bearing two amido groups. The yttrium-NHC bond length of 2.574(3) A˚ is also slightly longer than in the mono-NHC complex. Lanthanide complexes of NHC ligands bearing indenyl groups have also been reported. Transamination of the imidazolium-bromide salt in Scheme 6 with Y(CH2SiMe3)3(THF)2 aﬀords the bromide bridged complex {Y(L) (CH2SiMe3)(m-Br)}2.73 Reaction with anionic LiLn(CH2SiMe3)4(THF)4 (Ln=Y, Lu, Sc), however, yields the monomeric halide-free Y(L) (CH2SiMe3)2.74 This product can also be achieved by deprotonation of the imidazolium salt using lithium hexamethyldisilazide to give the stable carbene, followed by reaction with Ln(CH2SiMe3)3(THF)2. In all the complexes the monoanionic Ind-NHC ligand bonds to the metal centre in an Z5- fashion through the 5-membered ring of the indenyl unit, and the strong electron-donating carbene coordinates to the metal centre preventing THF coordination. The yttrium-NHC bond distance of 2.501(3) A˚ in Y(L) (CH2SiMe3)2 is the same as that of the amide tethered ligand in 60. Shen and co-workers reported a method for preparing lanthanide-NHC halides through protonolysis.75 Reaction of the imidazolium bromide salt with anionic LiLn(NiPr)4 aﬀords salicylaldiminato-functionalized NHClanthanide bromides (Scheme 33). The complexes were all characterised by single crystal X-ray diﬀraction and exhibit capped octahedral geometries. The NHC–Ln–NHC bond angles decrease NdWSmWErW which is consistent with decreasing ionic radii, and the Ln-NHC bond lengths also 24 | Organomet. Chem., 2010, 36, 1–28

t-Bu 2

t-Bu

t-Bu Br

OH

i-Pr N

N

LiLn(NiPr2)4

t-Bu

O

N

Ln

Br

2

N Ln = Nd, Sm, Er

N

N

i-Pr

Scheme 33 Synthesis of salicylaldiminato-functionalized N-heterocyclic carbene lanthanide bromides.

decrease 2.717(3) A˚ (Nd), 2.685(6) A˚ (Sm), 2.568(7) A˚ (Er) with decreasing ionic radii. Cui and co-workers have reported the ﬁrst xylene-bridged bis-NHC-ligated CCC-pincer lanthanide metal dibromides.76 They were prepared by reaction of the imidazolium salts with LnCl3 in the presence of n-butyl lithium (Scheme 34). Single crystal X-ray diﬀraction analysis revealed that the overall molecular structure of these complexes is an isostructural monomer of a THF solvate. The monoanionic xylene-bridged bis-NHCs bond to the central metal as a tridentate CCC-pincer moiety in a kC:kC:kC 0 mode which, in combination with the two trans-located bromo units, generates a twisted tetragonal bipyramidal geometry. The average lanthanideNHC bond lengths of 2.585(7) A˚ (Sm), 2.484(7) A˚ (Lu) and 2.309(8) A˚ (Sc) are on the short side when compared to other lanthanide-NHC bonds.

2Br N Mes

N

Br

+ LnCl3

N N

Mes

3nBuLi THF

Ln = Sm, Lu, Sc Mes

N N

Ln Br

Br

THF

N N

Mes

Scheme 34 Synthesis of bis-NHC-ligated CCC-pincer lanthanide metal dibromides.

The metal-carbene bond in lanthanide-NHC complexes is clearly weaker and more reactive than in late metal systems. The use of bidentate and tridentate ligands bearing anionic tethers can be eﬀective in stabilising lanthanide-NHC complexes, generating systems with shorter metal-carbene bonds. The area is still relatively underdeveloped, though the lanthanideNHC complexes are already displaying a range of chemistry in homogeneous catalysis and small molecule activation, some of which is unseen in late transition metal-NHC chemistry. Conclusion The coordination chemistry of NHCs with non-transition metal centres has expanded rapidly in recent years. It is clear that NHCs are able to form a range of interactions, ranging from being covalent to more ionic in nature. Their interaction with elements across the whole periodic table has improved our understanding of these highly tuneable and versatile ligands. Organomet. Chem., 2010, 36, 1–28 | 25

References 1 A. J. Arduengo, R. L. Harlow and M. Kline, J. Am. Chem. Soc., 1991, 113, 361–363. 2 L. Cavallo, A. Correa, C. Costabile and H. Jacobsen, J. Organomet. Chem., 2005, 690, 5407–5413. 3 C. M. Crudden and D. P. Allen, Coord. Chem. Rev., 2004, 248, 2247–2273. 4 W. A. Herrmann, J. Schutz, G. D. Frey and E. Herdtweck, Organometallics, 2006, 25, 2437–2448. 5 D. Pugh and A. A. Danopoulos, Coord. Chem. Rev., 2007, 251, 610–641. 6 N. M. Scott and S. P. Nolan, Eur. J. Inorg. Chem., 2005, 1815–1828. 7 C. E. Willans, K. M. Anderson, P. C. Junk, L. J. Barbour and J. W. Steed, Chem. Commun., 2007, 3634–3636. 8 D. Enders, O. Niemeier and A. Henseler, Chem. Rev., 2007, 107, 5606–5655. 9 A. Wacker, H. Pritzkow and W. Siebert, Eur. J. Inorg. Chem., 1998, 843–849. 10 A. J. Arduengo, M. Tamm, J. C. Calabrese, F. Davidson and W. J. Marshall, Chem. Lett., 1999, 1021–1022. 11 P. L. Arnold, S. A. Mungur, A. J. Blake and C. Wilson, Angew. Chem.-Int. Edit., 2003, 42, 5981–5984. 12 S. A. Mungur, S. T. Liddle, C. Wilson, M. J. Sarsﬁeld and P. L. Arnold, Chem. Commun., 2004, 2738–2739. 13 I. S. Edworthy, A. J. Blake, C. Wilson and P. L. Arnold, Organometallics, 2007, 26, 3684–3689. 14 P. L. Arnold, M. Rodden, K. M. Davis, A. C. Scarisbrick, A. J. Blake and C. Wilson, Chem. Commun., 2004, 1612–1613. 15 P. L. Arnold, M. Rodden and C. Wilson, Chem. Commun., 2005, 1743–1745. 16 G. Steiner, A. Krajete, H. Kopacka, K. H. Ongania, K. Wurst, P. PreishuberPﬂugl and B. Bildstein, Eur. J. Inorg. Chem., 2004, 2827–2836. 17 M. R. Crittall, C. E. Ellul, M. F. Mahon, O. Saker and M. K. Whittlesey, Dalton Trans., 2008, 4209–4211. 18 P. L. Arnold and C. Wilson, Inorg. Chim. Acta, 2007, 360, 190–196. 19 S. P. Downing and A. A. Danopoulos, Organometallics, 2006, 25, 1337–1340. 20 A. J. Arduengo, H. V. R. Dias, F. Davidson and R. L. Harlow, J. Organomet. Chem., 1993, 462, 13–18. 21 A. J. Arduengo, F. Davidson, R. Krafczyk, W. J. Marshall and M. Tamm, Organometallics, 1998, 17, 3375–3382. 22 H. Schumann, J. Gottfriedsen, M. Glanz, S. Dechert and J. Demtschuk, J. Organomet. Chem., 2001, 617, 588–600. 23 D. Zhang and H. Kawaguchi, Organometallics, 2006, 25, 5506–5509. 24 P. L. Arnold, I. S. Edworthy, C. D. Carmichael, A. J. Blake and C. Wilson, Dalton Trans., 2008, 1, 3739–3746. 25 A. G. M. Barrett, M. R. Crimmin, M. S. Hill, G. Kociok-Kohn, D. J. MacDougall, M. F. Mahon and P. A. Procopiou, Organometallics, 2008, 27, 3939– 3946. 26 R. J. Baker, M. L. Cole, C. Jones and M. F. Mahon, J. Chem. Soc.-Dalton Trans., 2002, 1992–1996. 27 R. J. Baker, H. Bettentrup and C. Jones, Eur. J. Inorg. Chem., 2003, 2446–2451. 28 H. Nakai, Y. J. Tang, P. Gantzel and K. Meyer, Chem. Commun., 2003, 24–25. 29 J. H. Cotgreave, D. Colclough, G. Kociok-Kohn, G. Ruggiero, C. G. Frost and A. S. Weller, Dalton Trans., 2004, 1519–1520. 26 | Organomet. Chem., 2010, 36, 1–28

30 A. J. Arduengo, H. V. R. Dias, J. C. Calabrese and F. Davidson, Inorg. Chem., 1993, 32, 1541–1542. 31 M. Lazraq, J. Escudie, C. Couret, J. Satge, M. Drager and R. Dammel, Angew. Chem.-Int. Edit. Engl., 1988, 27, 828–829. 32 A. Schafer, M. Weidenbruch, W. Saak and S. Pohl, J. Chem. Soc.-Chem. Commun., 1995, 1157–1158. 33 N. Kuhn, T. Kratz, D. Blaser and R. Boese, Chem. Berichte, 1995, 128, 245–250. 34 W. M. Boesveld, B. Gehrhus, P. B. Hitchcock, M. F. Lappert and P. V. Schleyer, Chem. Commun., 1999, 755–756. 35 F. E. Hahn, L. Wittenbecher, M. Kuhn, T. Lugger and R. Frohlich, J. Organomet. Chem., 2001, 617, 629–634. 36 P. A. Rupar, M. C. Jennings, P. J. Ragogna and K. M. Baines, Organometallics, 2007, 26, 4109–4111. 37 P. A. Rupar, M. C. Jennings and K. A. Baines, Organometallics, 2008, 27, 5043–5051. 38 S. P. Green, C. Jones, K. A. Lippert, D. P. Mills and A. Stasch, Inorg. Chem., 2006, 45, 7242–7251. 39 A. V. Zabula and F. E. Hahn, Eur. J. Inorg. Chem., 2008, 1, 5165–5179. 40 F. E. Hahn, D. Heitmann and T. Pape, Eur. J. Inorg. Chem., 2008, 1039–1041. 41 S. M. Mansell, C. A. Russell and D. F. Wass, Inorg. Chem., 2008, 47, 11367– 11375. 42 A. J. Arduengo, H. V. R. Dias and J. C. Calabrese, Chem. Lett., 1997, 143–144. 43 A. Jouaiti, M. Geoﬀroy, G. Terron and G. Bernardinelli, J. Am. Chem. Soc., 1995, 117, 2251–2258. 44 A. J. Arduengo, J. C. Calabrese, A. H. Cowley, H. V. R. Dias, J. R. Goerlich, W. J. Marshall and B. Riegel, Inorg. Chem., 1997, 36, 2151–2158. 45 A. J. Arduengo, R. Krafczyk, W. J. Marshall and R. Schmutzler, J. Am. Chem. Soc., 1997, 119, 3381–3382. 46 N. Kuhn, M. Strobele and M. Walker, Zeitschrift Fur Anorganische Und Allgemeine Chemie, 2003, 629, 180–181. 47 J. M. Hopkins, M. Bowdridge, K. N. Robertson, T. S. Cameron, H. A. Jenkins and J. A. C. Clyburne, Journal of Organic Chemistry, 2001, 66, 5713–5716. 48 D. M. Khramov and C. W. Bielawski, Chem. Commun., 2005, 4958–4960. 49 M. K. Denk, J. M. Rodezno, S. Gupta and A. J. Lough, Conference of the Canadian-Society-for-Chemistry, Calgary, Canada, 2000. 50 N. Kuhn, H. Bohnen, J. Fahl, D. Blaser and R. Boese, Chem. Ber.-Recl., 1996, 129, 1579–1586. 51 D. J. Williams, M. R. Fawcettbrown, R. R. Raye, D. Vanderveer, Y. T. Pang, R. L. Jones and K. L. Bergbauer, Heteroatom Chemistry, 1993, 4, 409–414. 52 B. M. Powell and B. H. Torrie, Acta Crystallographica Section C-Crystal Structure Communications, 1983, 39, 963–965. 53 N. Kuhn, G. Henkel and T. Kratz, Chem. Ber.-Recl., 1993, 126, 2047–2049. 54 A. J. Arduengo, M. Kline, J. C. Calabrese and F. Davidson, J. Am. Chem. Soc., 1991, 113, 9704–9705. 55 N. Kuhn, T. Kratz and G. Henkel, J. Chem. Soc.-Chem. Commun., 1993, 1778–1779. 56 A. J. Arduengo, M. Tamm and J. C. Calabrese, J. Am. Chem. Soc., 1994, 116, 3625–3626. 57 N. Kuhn, H. Bohnen, D. Blaser, R. Boese and A. H. Maulitz, J. Chem. Soc.-Chem. Commun., 1994, 2283–2284. 58 N. Kuhn, J. Fahl, R. Fawzi, C. Maichle-Mossmer and M. Steimann, Z. Naturforsch.(B)., 1998, 53, 720–726. Organomet. Chem., 2010, 36, 1–28 | 27

59 T. Isobe and T. Ishikawa, Journal of Organic Chemistry, 1999, 64, 5832–5835. 60 M. L. Cole, C. Jones and P. C. Junk, New Journal of Chemistry, 2002, 26, 1296–1303. 61 N. Kuhn, A. Abu-Rayyan, K. Eichele, S. Schwarz and M. Steimann, Inorg. Chim. Acta, 2004, 357, 1799–1804. 62 A. J. Arduengo, M. Tamm, S. J. McLain, J. C. Calabrese, F. Davidson and W. J. Marshall, J. Am. Chem. Soc., 1994, 116, 7927–7928. 63 W. A. Herrmann, F. C. Munck, G. R. J. Artus, O. Runte and R. Anwander, Organometallics, 1997, 16, 682–688. 64 W. J. Oldham, S. M. Oldham, B. L. Scott, K. D. Abney, W. H. Smith and D. A. Costa, Chem. Commun., 2001, 1348–1349. 65 H. Nakai, X. L. Hu, L. N. Zakharov, A. L. Rheingold and K. Meyer, Inorg. Chem., 2004, 43, 855–857. 66 T. Mehdoui, J. C. Berthet, P. Thuery and M. Ephritikhine, Chem. Commun., 2005, 2860–2862. 67 P. L. Arnold, A. L. Blake and C. Wilson, Chemistry-a European Journal, 2005, 11, 6095–6099. 68 P. L. Arnold and S. T. Liddle, Chem. Commun., 2006, 3959–3971. 69 P. L. Arnold and S. T. Liddle, Chem. Commun., 2005, 5638–5640. 70 S. T. Liddle and P. L. Arnold, Organometallics, 2005, 24, 2597–2605. 71 Z. G. Wang, H. M. Sun, H. S. Yao, Y. M. Yao, Q. Shen and Y. Zhang, J. Organomet. Chem., 2006, 691, 3383–3390. 72 P. L. Arnold and S. T. Liddle, Organometallics, 2006, 25, 1485–1491. 73 S. P. Downing, S. C. Guadano, D. Pugh, A. A. Danopoulos, R. M. Bellabarba, M. Hanton, D. Smith and R. P. Tooze, Organometallics, 2007, 26, 3762–3770. 74 B. L. Wang, D. Wang, D. M. Cui, W. Gao, T. Tang, X. S. Chen and X. B. Jing, Organometallics, 2007, 26, 3167–3172. 75 J. G. Zhang, H. S. Yao, Y. Zhang, H. M. Sun and Q. Shen, Organometallics, 2008, 27, 2672–2675. 76 K. Lv and D. M. Cui, Organometallics, 2008, 27, 5438–5440.

28 | Organomet. Chem., 2010, 36, 1–28

Bis(phosphorus-stabilised)methanide and methandiide derivatives of group 1–5 and f-element metals Stephen T. Liddle,a David P. Millsa and Ashley J. Woolesa DOI: 10.1039/9781847559616-00029

In the past 11 years bis(phosphorus-stabilised)methanides and methandiides have emerged as valuable ligands for metals across the periodic table. This Review, focussing on structurally characterised examples, covers the synthesis, bonding, and reaction chemistry of {CH(PR2NR 0 )2} methanide and {C(PR2NR 0 )2}2 and {C(PPh2S)2}2 methandiide complexes of group 1–5 and f-element metals reported between 1999 and January 2010. 1.

Introduction

Early metals, which for the purpose of this Review are deﬁned as belonging to either groups 1–5 or the f-elements, are characterised by their hard, polarising nature and large separation of frontier orbital energies which renders their chemical bonding highly polarised and predominantly ionic.1 Consequently, early metal-ligand bonds are expected to be labile and very reactive. This is exempliﬁed by the behaviour of early metal-alkyl bonds which are easily exchanged by protonolysis or s-bond metathesis reactions and they are vulnerable to a- and b-hydride elimination and a-abstraction reactions.2 Strategies to overcome the inherent reactivity of early metal-alkyl bonds have classically relied on saturating the coordination sphere of the metal with multi-dentate or sterically demanding ligands to block vacant coordination sites or by employing ligands with no b-hydrogens.3 Given that PH2 has been calculated to stabilise a carbanion centre by –89.1 kJ mol 1,4–6 attention innevitably turned to phosphorus-stabilised methanides.7 Bis (phosphorus-stabilised) methanides represented a logical extension which naturally developed to include bis(iminophosphorano)methanides (I) as P(V) is superior to P(III) for stabilising carbanion charges.8 R R R′

P

H C

N

R P N

I

R R

R

R′ R′

P

2 C

N

R P N

II

R R

R

R′

R′

P

2 C

N

R P N

III

R R

R

R′ R′

P

C

N

R P N

IV

R R

R

R′

R′

P

C

N

R P N

V

R R′

Placing two P(V) centres a to a methylene group renders the two hydrogens acidic, and, in addition to the expected facile mono-deprotonation by mild bases such as lithium amides, the second hydrogen can routinely be removed using a strong base, such as an organolithium, to aﬀord the corresponding geminal dianionic methandiide. The resulting dianion is of considerable interest when formal resonance forms are considered: (i) a geminal carbon dianion and two imino groups (II); (ii) a dipolar form with a a

School of Chemistry, University of Nottingham, University Park, Nottingham NG7 2RD, UK

Organomet. Chem., 2010, 36, 29–55 | 29 c

The Royal Society of Chemistry 2010

geminal carbon dianion and two anionic amides (III); (iii) a phosphaalkeneamide-iminophosporano (IV); and (iv) a carbodiphosphorane with two anionic amides (V). The later resonance form is particularly interesting because it is predicated on the assignment of a P(V) oxidation state, but an alternative assignment is that of two amido-functionalised P(III) centres datively coordinated to a captodative, zero-valent carbon centre.9 Germane to this discussion is the closely related bis(thiophosphinoyl)methandiide {C(PPh2S)2}2 , which is analogous to II and has only recently been used with early metals. Together, these phosphorus-stabilised methandiides can be regarded as an intriguing class of early metal carbene that is distinct to traditional early metal carbenes such as Tebbe’s reagent,10 Schrock’s alkylidenes,11 and Mindiola’s group 4 and 5 alkylidenes.12 In this Review we cover the syntheses of methane precursors and the synthesis, bonding, and reactions of bis(phosphorus-stabilised)-methanide and -methandiide complexes of group 1–5 and f-element metals (early metals) organised by group number with the f-elements incorportated into the group 3 section. 2.

Methane syntheses

Two routes for the preparation of bis(iminophosphorano)methane precursors are commonly employed, namely the Phospha-Staudinger and Kirsanov methods. The bis(diphenylthiophosphinoyl)methane is prepared by a straightforward oxidation method. 2.1

Phospha-Staudinger method R2 PCH2 PR2 þ 2 N3 R0 ! CH2 ðPR2 NR0 Þ2 þ 2 N2

ð1Þ

The Phospha-Staudinger reaction between a bis(phosphino)methane and an organoazide to give the corresponding bis(iminophosphorano)methane with concomitant elimination of dinitrogen, Eq. 1, is applicable to a wide range of silyl, alkyl, and aryl azides, e.g. R 0 =SiMe3; 2,4,6-Me3C6H5; 2,6Pri2C6H3; adamantyl, and a range of bis(phosphino)methanes, e.g. R=Me, Ph, cyclohexyl.13 For N-SiMe3 derivatives the reactions can be done in neat azide as this is a stable liquid. However, for other N-substituents it is usual to use an arene co-solvent to prevent explosion since more covalent azides are prone to detonation when heated neat. 2.2

Kirsanov method

Alternatively, quaternisation of bisphosphinomethanes with bromine, followed by treatment with a primary amine and subsequent reaction with a base aﬀords the corresponding methanes as a mixture of bis-imino and imino-amine tautomers; the latter is favoured by N-alkyl substituents, Scheme 1.14 Irrespective of which valence tautomer is isolated, or whether a mixture is obtained, both aﬀord the same methanide and methandiide following deprotonation so the Kirsanov method provides a valuable alternative to the preferred Phospha-Staudinger method when the required azide is unavailable or too unstable to be used. 30 | Organomet. Chem., 2010, 36, 29–55

H2 C

R R

P

2 Br2 R

R

R P

R

P

H2 C

Br

Br

H2 C

R

R P

2 R′NH2 R Base

R 2 Br

P N

R′

R P N

R R

R

+

R′

P N

R′

H C

R P N

H

R R′

Scheme 1

2.3

Oxidation method R2 PCH2 PR2 þ 2 S ! CH2 ðPR2 SÞ2

ð2Þ

The bis(diphenylthiophosphinoyl)methane CH2(PR2S)2 is prepared by simple oxidation of bis(diphenylphosphino)methane with excess elemental sulfur, Eq. 2.

3.

Group 1 methanides and methandiides

3.1

Group 1 methanides Ph PhPh Ph

SiMe3 SiMe3 Ph Ph Cy SiMe3 Ph SiMe3 SiMe3 Ph P N P H P Ph P N THF N P Ph Cy P N N C N HC Li OEt2 Li Li HC Na HC Na Na CH N N C Ph P N THF N P Ph Cy P N Me3Si P H P SiMe3 Ph P N

Me3Si

Ph PhPh 1

Ph

Ph

Cy

SiMe3 SiMe3 Ph 2

SiMe3

3

Ph

4

SiMe3

Although the ﬁrst bis(iminophosphorano)methanides were reported by Elsevier,15 the ﬁrst structurally characterised examples were reported by Cavell.16 Solvent-free 1 and 2 were prepared from the reaction between the parent methane and lithium or sodium bis(trimethylsilyl)amides in aromatic solvents. By avoiding Lewis base solvents such as ethers, dimeric complexes were isolated. Treatment with excess quantities of alkali metal amide did not eﬀect a second deprotonation, even under reﬂux conditions over days, which was attributed to the less basic nature of amides compared to alkyls (see section 3.2 below). In addition to the expected methanide-alkali metal bonds, methine C–H?Li interactions were observed in 1 in the solid state but the analogous C–H?Na interactions appeared to be weak in 2. Ph Ph

SiMe3 P

N

HC Ph Me3Si

THF Ph

K

N

SiMe3 P

Ph

P

5

Ph

N SiMe3 6

P HC

Li THF

Ph

SiMe3 SiMe3 Ph

Ph Ph

N

HC THF

P

Ph

P

Ph

N

N

K

K

N

N

P

Ph Ph

SiMe3 SiMe3 Ph 7

P

N

HC

CH P

SiMe3

Ph

Ph Ph

P

K N

Ph

O O O

SiMe3 8

Cavell subsequently reported Lewis base adducts of lithium (3), sodium (4), and potassium (5) bis(iminophosphorano)methanides.17 Complex 3, with P-cyclohexyl substituents, was noted to resist a second deprotonation even Organomet. Chem., 2010, 36, 29–55 | 31

when treated with organolithium reagents which demonstrates the signiﬁcant eﬀect on acidity of the C–H bonds that the P-substituents can have. Complexes 4 and 5 were prepared from the respective hydrides. Although the sodium complex did not undergo a second deprotonation with excess sodium hydride and reﬂux conditions, the potassium complex reacted further with excess potassium hydride and extended reﬂux to give a complex mixture of products. Structural investigations revealed that whereas a lithium-methanide contact was present in 3, no sodium-methanide contact was present in 4, and in 5 no potassium-methanide contact was present but the potassium centre was found to be bound in an Z2 manner to one of the P-phenyl rings. Roesky extended the number of bis(iminophosphorano)methanides with the report of the lithium complex 6,18 which was also described by Leung two years later,19 and the two potassium complexes 7 and 8.18 The lithium complex was prepared by treatment of the parent methane with n-butyl lithium in THF resulting in a THF-coordinated complex similar to 3. Structural investigation of 6 showed that a weak lithium-methanide interaction was present. Interestingly, complex 7 was prepared in an analogous manner to 5 but a diﬀerent solid state structure was obtained which was found to be solvent-free with a weak potassium-methanide contact and the monomers arranged as ‘‘loose’’ dimers through weak bridging K?N contacts (cf the Na?N contacts in 2). Recrystallisation of 7 from heptane/diglyme resulted in the isolation of 8 in which no potassium-methanide interaction was observed in the solid state presumably because of the presence of the k3-coordinated tridentate diglyme. Ph Ph

P

N

HC Ph

Ph

Mes

P

Ph

Li OEt2 N

Ph

Mes 9

Ph

P

H C

N Mes

Ph

Mes

P K

N N

K

P Ph

N C H Ph

10

P

Pr i

Ph Ph

Ph

P HC

Ph Ph

P

N

Ph Ph

Li OEt2 N

Ph

Pr 11

HC Ph

i

SiMe3 P

P

N

THF

Li S

THF

Ph 12

Hill expanded the range of bis(iminophosphorano)methanides with the lithium20 and potassium21 complexes 9 and 10. Complex 9 was prepared from the reaction of n-butyl lithium with the parent methane in ether and was crystallised as a mono-etherate which was found to not exhibit a lithiummethanide contact. The potassium complex 10 was prepared from the parent methane and potassium bis(trimethylsilyl)amide in toluene. A structural investigation revealed 10 to be a dimer formed by Z6-mesityl?potassium interactions, but a potassium-methanide contact was not observed. Le Floch reported complex 11 which was prepared from the parent methane and three equivalents of methyl lithium which is noteable as at this stoichiometry no methandiide was apparently formed.22 A structual investigation showed 11 to exhibit a structure very similar to 3, 6, and 9. The structure of the hybrid lithium methanide complex 12 was reported by So.23 Complex 12 is similar to 3, 6, 9, and 11, except an extra molecule of solvent is coordinated to the lithium centre (in this case THF), presumably due to the diminished steric demands of a thio group compared to a substituted imino group. 32 | Organomet. Chem., 2010, 36, 29–55

3.2

Group 1 methandiides Ph PhPh Ph

Me3Si

N

P

P

C

Ph PhPh Ph SiMe3 Me3Si

N

P

Ph PhPh Ph Et2O S Li S

P Li P

P

C

Li C

Ph PhPh 21

P

S Li S OEt2 Ph

C

P

N Li Li Li Li M M M M N N N C N C P P P P Me3Si SiMe3 Me3Si SiMe3 Ph PhPh Ph Ph PhPh Ph 13

N

Ph PhPh Ph R

SiMe3

M = Na4 14 Li2Na2 15 LiNa3 16 Li2K2 17 Na2K2 18 Na3K 19 R = SiMe3

P N

C

P

R N

Li Li Li Li N C P P R R Ph PhPh Ph N

20 (R = MeCHPri)

Ph Et2O Ph Li S

Ph Ph P P S C S

Li Li OEt Li 2 C S OEt 2 P P

Ph PhPh

Ph

22

The ﬁrst rational synthesis of a dilithium salt of a bis-phosphorus-stabilised methandiide was simultaneously reported by Cavell24 and Stephan.25 Complex 13 was prepared from the reaction of two molar equivalents of organolithium reagents with the parent methane in arene solvent. A structural study showed complex 13 to be dimeric in the solid state with the four lithium atoms arranged in a square plane. With the two methandide ligands above and below this plane a near perfect Li4C2 octahedron is formed. The two ligand planes deﬁned by the NPCPN units, which were themselves essentially coplanar, were found to be orthogonal to each other. Henderson extended the range of di-alkali metal methandiides with the synthesis of tetra-sodium (14), di-lithium-di-sodium (15), lithium-tri-sodium (16), di-lithium-di-potassium (17), di-sodium-di-potassium (18), and trisodium-potassium (19) complexes.26,27 The homometallic species were prepared by treatment of the parent methane with two equivalents of the corresponding organo alkali metal reagent. Additionally, as Cavell found, the use of amides gave only the methanide complexes which could be selectively used as precursors to heterometallic methandiide compounds. The heterometallic complexes were prepared by stepwise metallation with organo alkali metal reagents or by careful addition of alkali metal alkoxides in the relevant stoichiometries and could also be prepared by cation exchange between mixtures of the homometallic derivatives. Complex 15 was found to be in dynamic equilibria in solution with the ﬁve possible tetrametallic derivatives Li4, Li3Na, Li2Na2, LiNa3, and Na4 coexisting. Although the gross structures appear the same for 14–19, the core of the structures gradually expands as the alkali metal size increases and in the heterometallic systems the harder metal(s) occupies the central zone of the core with bonds to the methandiides but the softer larger alkali metal is pushed out to the edge of the core and in most instances no longer has contacts to the methandiide centres. A computational analysis of 13 found that the dipolar resonance form III most accurately described the electronic structure of the methandiide ligand. Organomet. Chem., 2010, 36, 29–55 | 33

Le Floch has reported three di-lithium methandiide complexes with iminophosphorano and thiophosphinolyl ligands. Complex 20 was prepared by treatment of the parent bis(iminophosphorano)methane with four equivalents of methyl lithium in diethyl ether.22 A structural study of 20 showed it to possess a structure very similar to 13 constructed around a distorted Li4C2 octahedral core. Complexes 21 and 22 were prepared from the corresponding bis(thiophosphinoyl)methane and two equivalents of methyl lithium in diethyl ether.28 In gross terms, complexes 21 and 22 diﬀer only in the number of coordinated diethyl ether molecules which result from the crystallisation conditions. In detail, the structures of the cages clearly distort in response to the number of coordinated diethyl ether molecules which presumably coordinate due to the less sterically demanding nature of the thio groups compared to 13–20 which contain bulky imino groups. Ph PhPh Ph

Ph PhPh Ph S THF Li S

P

C

Li P

P Li

C

Ph PhPh 23

P

Me3Si H2 THF Ph P P C N H2C N L C N Cs THF Cs Cs L M M M M L Ph P N C SiMe3 P P Ph N C L N Me3Si Ph C N P P P Ph Cs Ph Ph Cs THF Cs CH2 N Ph PhPh Ph C THF H2 SiMe

SiMe3 Ph N Li N SiMe3 Ph

24 (K4, L = THF), 25 (Rb4, L = η2-Ph)

26

3

So reported the di-lithium methandiide complex 23 which employs the hybrid ligand used in 12.23 Interestingly, complex 23, which is dimeric and constructed around a Li4C2 core, was found to crystallise with the two imino nitrogens ‘up’ and the two thio sulfurs ‘down’ which enables a molecule of THF to coordinate to the bis-thio-coordinated lithium centre, as in 21 and 22. Harder extended the di-alkali metal methandiides reported by Cavell, Stephan, and Henderson to include the tetra-potassium and tetra-rubidium methandiides by treatment of the parent methane with two equivalents of benzyl potassium or rubidium, respectively, but crystalline products could not be isolated.29 However, replacing the N-SiMe3 substituent with N-phenyl resulted in the isolation of complexes 24 and 25.29 Complexes 24 and 25 are dimeric in the solid state exhibiting K4C2 and Rb4C2 cores which are signiﬁcantly distorted from octahedral such that the clusters display approximate S4 symmetry. Each potassium centre in 24 is coordinated by a molecule of THF whereas in 25 the rubidium centres display Z2-contacts to neighbouring phenyl rings. Although NMR spectroscopy indicated the di-caesium analogue of 24 and 25 was formed using benzyl caesium, all attempts to isolate it were unsuccessful. The only crystalline product that could be isolated of any caesium derivative was the three-dimensional network structure 26 which derived from the treatment of 13 with a caesium alkoxide. Complex 26 was proposed to arise from the attack of the methandiide centre on a P-phenyl ring to aﬀord a caesium methanide and caesiated phenyl ring which attacked a phosphorus centre to give ring formation and elimination of phenyl caesium which subsequently deprotonated the methanide centre to regenerate the methandiide centre. The presence of benzyl caesium in 26 was suggested to originate from the fact the starting material, 13, co-crystallises with a moleule of toluene and that this was metallated by phenyl caesium formed during the reaction. 34 | Organomet. Chem., 2010, 36, 29–55

4.

Group 2 methanides and methandiides

4.1

Group 2 methanides Ph Ph P

N

P Ph Ph

Ph

I

CH2 Mg N

SiMe3 SiMe3 Ph

Ph

SiMe3

P

N

HC Mg

I

Ph

SiMe3

N

P

Cl Cl

N

P

Ph

Mg CH N

SiMe3 N

THF

HC Mg

P

Ph

SiMe3 SiMe3 Ph

Ph

27

Ph Ph P P Ph Ph

28

N

I

SiMe3 29

The ﬁrst structurally characterised group 2 complexes were reported by Stephan.30 The neutral coordination complex [MgI2{H2C(PPh2NSiMe3)2}] 27 was isolated in low yield from the reaction of the parent methane with MeMgI and was rationalised on the basis of facile Schlenk equilibria of the Grignard precursor. In contrast, complex 28 was isolated from the reaction of the parent methane with MeMgCl. Complex 29, the original target of the reaction which produced 27, was isolated from the reaction of the lithium complex 1, prepared in situ, and MeMgI, which infers the elimination of MeLi. Both 28 and 29 exhibit magnesium-methanide contacts in the solid state but the chloro complex 28 was found to be dimeric whereas the iodo complex 29 was found to be a solvated monomer. Ph Ph P HC P

Ph Ph

Mes Mg N

Cl

Mes 30

Mes

Ph

Mes

Ph

Ph Ph P

Mes

Ph

Mes

Mes Ph N P Ph Mg CH HC Mg HC Ba CH HC M Ph Cl N(SiMe3)2 P N N P Ph Ph P N Ph P N N P Ph Ph Mes Mes Mes Mes Ph Ph Mes Ph Ph 31 M = Ca 32 34 Sr 33

THF Ph

N

P

N

Cl

N

P

Ph

N

THF

Ph

P

N

Hill subsequently reported a series of alkaline earth complexes covering magnesium (30, 31), calcium (32), strontium (33), and barium (34).21,31 Complexes 30–34 were prepared by reaction of MeMgCl with the parent methane or by combining the parent methane, the relevant metal diiodide and two equivalents of potassium bis(trimethylsilyl)amide; the formation of 34 is likely the result of ligand scrambling due to the large size of barium. Methanide-metal contacts were observed in 31–33 but not in 30 and 34. Whereas 32 and 33 were found to polymerise up to two hundred equivalents of rac-lactide in less than one minute, 34 displayed no activity. Reactivity studies of 32 with potentially bidentate, protic hydroxy and amino reagents to form prospective lactide catalysts resulted in indiscrimniate protonation of the methanide centre and this approach was subsequently abandoned. SiMe3 SiMe3 Ph THF THF N P Ph Ph P N THF N P Ph Ph P N N THF I I Sr CH Ba CH HC Sr HC Ba HC Ca I I I P N THF Ph P N THF N P Ph Ph P N THF N P Ph Ph THF SiMe3 Ph Ph Ph SiMe3 SiMe3 Ph SiMe3 SiMe3 Ph 37 35 36 Ph Ph P

SiMe3

Ph

SiMe3

SiMe3 Ph

Ph

Organomet. Chem., 2010, 36, 29–55 | 35

Roesky used 7 to prepare heteroleptic calcium (35), strontium (36), and barium (37) complexes from the corresponding metal diiodides.32 Complex 35 was found to be monomeric, exhibit a calcium-methanide contact, and to be isostructural to the analogous ytterbium(II) complex 93 (vide infra). In contrast, complexes 36 and 37 were found to be dimeric, the former is isostructural to the samarium and europium analogues 91 and 92 (vide infra), and constructed around four-membered M2I2 rings but the larger size of barium compared to strontium is reﬂected by the inclusion of one coordinated THF molecule per strontium in 36 but two coordinated molecules of THF per barium in 37. Both 36 and 37 exhibit metal-methanide contacts which for 37 contrasts with the situation in 34. Complex 38 was prepared from two equivalents of 7 and strontium diiodide and, in contrast to 34, was found to exhibit metal-methanide contacts.33 SiMe3 SiMe3 Ph P N N P HC M CH Ph P N N P SiMe3 Ph SiMe3 M = Sr 38 Ca 39 Ba 40 Ph

Ph Ph Ph Ph

Harder prepared the two bis(methanide) complexes 39 and 40 from the respective reactions between the parent methane and the relevant bis-amide [M{N(SiMe3)}2].34,35 Under no circumstances could a second deprotonation be observed. In common with 38, 39 and 40 were found to contain calcium- and barium-methanide contacts in the solid state. 4.2

Group 2 methandiides Ph

Ph Ph P

41

Ph

Ph Ph

Ph

SiMe3 Me3Si

P

Ph

Ph Ph

Ph P P SiMe3 Me3Si N Me3Si N N N SiMe3 N N C C C Ph Ca Ca AdC N Ca Ca N CAd Ca Ca O C C C Ph N N N SiMe N N Me3Si Me3Si N SiMe3 Me3Si 3 SiMe3 P P P P P P Ph Ph Ph Ph Ph Ph Ph Ph Ph Ph Ph Ph P

Ph

P

42

43

Ph Ph P Ph N P CyN C O C Ca Ph Ph Ca C O C NCy P N Ph P N SiMe3 Ph SiMe3 Ph Me3Si

Me3Si

N

44

36 | Organomet. Chem., 2010, 36, 29–55

Harder reported the ﬁrst calcium methandiide 41 from reaction between the parent methane and dibenzyl calcium.34 The methandiide centres in dimeric 41 were found to bridge the two calcium centres and the structure overall is reminiscent of 13–20 except only two metals are present for charge neutrality. The reactivity of 41 towards benzophenone was examined but all that could be isolated was the mono-benzophenone adduct 42. No alkene metathesis product was found, and even after reﬂux only a small amount of the alcohol HOC(Ph)2C(PPh2NSiMe2)2 could be identiﬁed amongst other products. The reaction of 41 with adamantyl cyanide gave the adduct 43 as the sole isolable product and no [2 þ 2] cycloaddition products were observed. The sluggish reactivity of 41 prompted an investigation of stronger electrophiles and consequently the reaction of cyclohexylisocyanate with 41 was examined which resulted in the isolation of the [2 þ 2] cycloaddition product 44 which was found to be a dimer in the solid state. A DFT study of 41 concluded that, like 13, resonance form III most accurately depicts the electronic structure of the methandiide ligand in 41. Ph Ph

Dipp P

N

C Ca Ph

P

N

Ph

Ph

THF THF

Dipp P

Ph

N

C Ca Ph

Dipp

P

N

Ph

45

Cy Dipp

THF O

C

Ph Dipp Ph

N N Ph2P

Ph2P

Dipp

THF

46

C

Cy N

Ca O

C O

N

O Ca C C N N Cy Cy 47

C O Ph2P N

C

PPh2 N

Dipp

Dipp

Harder subsequently expanded the range of calcium methandiides to include complex 45.36 Complex 45 was prepared from the parent methane and a dibenzyl calcium reagent and crystallises as a bis(THF) solvated monomer which contrasts to 41. However, sluggish behaviour was again observed towards benzophenone with the isolation and characterisation of the benzophenone adduct 46. Bond lengths within the methandiide and benzophenone ligands are hardly aﬀected in 46 compared to 45 which indicates that the benzophenone coordination is weak. The behaviour of 45 as an initiator for the anionic trimerisation of isocyanates was examined and although trimerisation was observed it was extremely slow and consequently the double insertion intermediate 47 could be reproducibly crystallised as a dinuclear species. Ph Me3Si

Ph Ph P

N

P C

THF Ba P Ph

Ph Ph

Ba THF C

N Me3Si

Ph SiMe3 N N

P

Ph Ph 48

SiMe3 Ph

Ph

Dipp P

N THF C Ba THF P N THF

Ph

Dipp 49

Dipp N

Dipp Ph N Ph P P Ph Ba C C Ph C O O Ph P C NCy P Ph Ba CyN N Ph N Dipp THF Dipp 50 Ph

Harder also reported the barium methandiide complexes 48 and 49.35 Complex 48 was prepared from the parent methane and crystallises as a dimeric species with bridging methandiide centres and is similar to the calcium analogue 41 except the larger size of barium compared to calcium Organomet. Chem., 2010, 36, 29–55 | 37

allows a molecule of THF to coordinate to each barium centre. Replacement of the N-SiMe3 groups in 48 with N-Dipp groups results in complex 49 crystallising as a THF solvated monomer similar to 45. Sluggish reactivity of 49 with benzophenone and adamantyl cyanide was observed and heating reaction mixtures led to complex and intractable mixtures of unidentiﬁable products. The reactivity of 49 with cyclohexyl isocyanate was assessed and resulted in the production of the dimeric complex 50 which contained the methanide derivative of a [2 þ 2] cycloaddition of the isocyanate. 5.

Group 3 and f-element methanides and methandiides

5.1

Group 3 and f-element methanides SiMe3

Ph Ph

Ph Ph

SiMe3 Ph

Cl N P Cl M CH HC M Cl P N Cl N P P

N

Ph

Ph Ph P HC

Ph

SiMe3 SiMe3 Ph M = Sm 51 Dy 52 Er 53 Yb 54 Lu 55 Y 56

P

NPh2

Ph Ph P

NPh2

P

SiMe3 N

HC

M N

SiMe3 N

Ph SiMe3 Ph M = Sm 57 Y 58

M N

Ph SiMe3 Ph M = Sm 59 Er 60 Y 61

Ph Ph P

SiMe3 N

HC P

M N

Cl Ph SiMe3 Ph M = Sm 62 Yb 63 Y 64

Roesky introduced bis(iminophosphorano)methanides to rare earth chemistry with a comprehensive study of trivalent rare earth bis(iminophosphorano)methanide dichlorides by the synthesis of samarium (51), dysprosium (52), erbium (53), ytterbium (54), lutetium (55), and yttrium (56) derivatives.37 Complexes 51–56 were prepared from the corresponding anhydrous rare earth trichlorides and 7 in THF and 51 and 56 were further derivatised with two equivalents of potassium diphenylamide to produce 57 and 58, respectively.37 Additionally, treatment of 51, 53, and 56 with two equivalents of sodium cyclopentadienyl resulted in the formation of the bis(cyclopentadienly) derivatives 59–61.38 In 51–61 a metal-methanide bond was observed in the solid state, and for 56 this was shown to persist in solution by 13C NMR spectroscopy (dCH 17.6 ppm, JYC=3.6; 2 JPY=89.1 Hz). However, for 61 the NMR data suggested the yttriumcarbon bond was lost in solution. DFT calculations supported the presence of an yttrium-methanide contact in 56 with a calculated shared electron number (SEN) of 0.40 for the yttrium-carbon bond in a monomeric gas phase model of 56; for comparison, the yttrium-nitrogen bond SEN was calculated to be 0.41. Extending the theme of cyclopentadienyl-substituted methanides, complexes 51, 54, and 56 were used to prepare the mono-pentamethyl-cyclopentadienyl complexes 62, 63, and 64 in which the metal-methanide bond was maintained.39 Complexes 62–64 were tested for their activity in the polymerisation of e-caprolactone and they exhibited low activities. However, treatment of 62–64 with one equivalent of iso-propyl alcohol resulted 38 | Organomet. Chem., 2010, 36, 29–55

in alcoholyses to give active initiators which gave high molecular weights and low polydispersities. The polymerisations were found to be pseudoliving in nature, and MALDI-TOF analyses showed a caproyl hydroxy group at one end and an isopropyl ester group at the other. Mechanistic investigations showed partial protonation of the methanide and cyclopentadienyl groups following addition of the iso-propyl alcohol indicating a complex mixture of products was formed rather than simple chloride-alkoxide exchange, but it was suggested that the low polydispersities were the result of continued coordination of the methanide ligand to the metal centre. Ph Ph P HC

SiMe3 N M

PPh2

Ph Ph P HC

N P P N Cl PPh2 Ph Ph SiMe3 Ph Ph M = La 65 Nd 66 Yb 67 Y 68

N

SiMe3 PPh2

La N PPh 2 N NPh2 SiMe3 69

Ph Ph P HC

SiMe3 N M

PPh2

N * P N Cl Ph SiMe3 Ph Ph M = Er 70a/b (R/S) Yb 71a/b (R/S) Lu 72a/b (R/S)

Ph Ph P HC P Ph Ph

SiMe3 N

Cl

THF Li

Yb Cl

N

THF

SiMe3 73

Closely related to 57 and 58, complexes 65–68 were prepared from the dihalide precursors or by one pot reactions of 7, a rare earth trihalide and potassium bis(diphenylphosphanyl)amide.40 Although the bis(iminophosporano)methanide ligand is sterically demanding, the bis(diphenylphosphino)amide ligand was found to coordinate in a k2-P,N manner (with the remaining phosphane centre not coordinated) in the solid state and the metal-methanide contact was described as weak. Complex 65 was further derivatised with potassium diphenyl amide to give the sterically crowded, heteroleptic complex 69,40 and it was noted that the order of substitution was important to access the desired complexes. Complexes 65–69 were tested in the polymerisation of e-caprolactone and methyl methacrylate (MMA). For e-caprolactone, poly-caprolactone was obtained in good yields in minutes. The larger metals were unsurprisingly found to be more active, but the solutions rapidly became viscous and this was suggested to be impeding the reaction. Additionally, polydispersities in the range 1.8–4.4 and molecular masses up to three times the theoretical indicated that one initiator molecule was starting several polymer chains. For MMA, variable yields of polymethylmethacrylate (PMMA) were obtained and the addition of tri-ethyl aluminium was found to increase yields and the larger metals were more active. For 65, syndiotactic PMMA was obtained but when complex 69 was used a larger fraction of atactic polymer was formed. In some instances the molecular masses vastly exceeded the calculated masses and cooling of solutions was required but this resulted in masses being lower than expected with modest polydisperisites. The use of phosphanylamides was extended by Roesky to include chiral analogues. Treatment of 53–55 with R- and S-[N(CHMePh)-(PPh2)Li] aﬀorded the amide derivatives 70a/b, 71a/b, and 72a/b in which the phosphane and methanide centres were found to coordinate to the metal centres in the solid state.41 Interestingly, an attempt to prepare 71a/b by a one pot reaction between 7, ytterbium(III) Organomet. Chem., 2010, 36, 29–55 | 39

trichloride and the above lithium salts resulted in isolation of the lithium chloride occluded complex 73.41 Ph Ph P HC

SiMe3 N

N(SiHMe2)2 M

Ph Ph P

SiMe3 N

HC

M

N(SiHMe2)2 P N P N Ph Ph SiMe3 SiMe3 Ph Ph M = La 74 M = Sm79 Er 80 Sm75 Yb 81 Ho 76 Lu 82 Lu 77 Y 83 Y 78

Ph Ph P HC

SiMe3 SiMe3

N M

P N SiMe3 Ph SiMe3 Ph M = Er 84 Y 85

Underscoring the high synthetic utility of the rare earths in stoichiometric and catalytic bond forming reactions, Roesky prepared complex 74 and demonstrated its eﬃcacy in catalytic hydroamination/cyclisation, hydrosilylation, and sequential hydroamination/-hydrosilylation reactions.42 Complex 74 was prepared from the parent methane and the corresponding homoleptic tris(dimethylsilyl)amido lanthanum complex by amine elimination in toluene. Complex 74 catalysed a range of hydroamination/cyclisation reactions in good yields at moderate temperatures to generate 5- and 6-membered rings using amino-tethered alkenes and alkynes. A variety of terminal alkenes and dienes were catalytically hydrosilylated by 74 in high yields giving the antiMarkovnikov products for aliphatic substrates but a mixture of products for styrene. Lastly, the two types of catalysis were combined for amino-alkynes to give the corresponding products (5- and 6-membered ring derivatives) in quantitative yields. This range was subsequently extended to include samarium (75), holmium (76), lutetium (77), and yttrium (78) by the amine elimination route described above or by salt elimination reactions between the corresponding metal trichloride, 7, and two equivalents of the potassium amide.43 From this extended study several conclusions could be drawn. Firstly, reaction rates were unsurprisingly observed to increase with metal size. Secondly, the reaction kinetics were found to be ﬁrst order with respect to catalyst and zero order with respect to substrate as is the case with rare earth metallocene systems. Thirdly, 31P NMR spectroscopy indicated that the methanide ligand was not protonated and remained coordinated to the metal centres during reactions. Fourthly, Thorpe-Ingold eﬀects were found to favour heterocycle formation and increased rates were observed for substrates with bulky geminal substituents. Fifthly, the rate of cyclisation for 5-membered ring products is faster than for the 6-membered products. As described above, 75–78 were also all found to be eﬀective catalysts for hydrosilylation and sequential hydroamination/-hydrosilylation in excellent yields. The reaction between 51 and one equivalent of di-potassium cyclooctatetraenide aﬀorded 79.44 Complex 79 was also reported to be preparable from [Sm(Z8-C8H8)(I)(THF)] and 7, perhaps indicating the similar steric demands of these two organometallic ligands, giving straightforward synthesis where less sterically demanding ligands render the sequence of addition of ligands important (cf. 69). Roesky demonstrated that, surprisingly, although 79 contains no obvious leaving group, it is an active catalyst for the hydroamination/ 40 | Organomet. Chem., 2010, 36, 29–55

cyclisation of one amino-tethered alkene and several amino-tethered alkynes at elevated temperatures. NMR studies showed that the catalyst remained intact during the reactions with no loss of the methanide or cyclooctatetraenide ligands. In terms of mechanism, it was suggested that complex 79 weakly polarises the substrates, thus activating them towards hydroamination/ cyclisation. Roesky subsequently extended the range to include erbium (80), ytterbium (81), lutetium (82), and yttrium (83).45 As described above, the kinetics of the reactions with amino-tethered alkenes and alkynes was determined to be zero order with respect to the substrate. Although these systems were found to be not as active as metallocene pre-catalysts, they do provide examples of catalysts for hydroamination/cyclisation which appear to operate by a distinct pathway compared to the commonly accepted protonolysis/ s-bond metathesis/protonolysis mechanism for metallocene-based hydroamination/-cyclisation pre-catalysts. Roesky also demonstrated that bulky cyclooctatetraenide ligands could be substituted onto bis(iminophosporano)methanide rare earth complexes. The one pot reaction between 7, erbium or yttrium trichloride, and di-lithio 1,4-bis(trimethylsilyl)cyclooctatetraenide aﬀorded complexes 84 and 85.46 In the solid state metal-methanide bonds were observed, and for yttrium this was inferred to persist in solution as adjudged by 31P NMR spectroscopy (dP 20.3, 2JPY=7.4 Hz). Ph Ph P

SiMe3 N

HC P Ph Ph

I Y

THF I

N

SiMe3 86

Ph Ph P

Pr i

Ph

Pr i Ph Ph P N I N P Ph HC Nd CH Ph Ph P N N P Ph i Pr i Pr Ph 88

N THF I

HC Nd P Ph Ph

Pr i

I N THF i Pr 87

Ph Ph P

Pr i N

N(SiMe3)2

HC Nd P Ph Ph

N(SiMe3)2

N Pr

i

89

Liddle reported the synthesis of complex 86 from the reaction between 5 and yttrium triiodide.47 The reaction was surprisingly sluggish but 86 was obtained in good yield. In the solid state 86 was found to be monomeric, which contrasts with the dimeric formulations of 51–56 and the yttriummethanide bond observed in the solid state was found to be maintined in solution by 13C NMR spectroscopy (dCH 15.8, JPC=133; 2JYC=7.4 Hz). Le Floch reported that treatment of [K{CH(PPh2NPri)2}] with one molar equivalent of neodymium triiodide aﬀorded 87, but when reacted with half a molar equivalent complex 88 was instead obtained.48 Complex 87 was converted to the bis-amide derivative 89 without any evidence of ligand scrambling. Ph Mes

Mes Ph N P Ph HC Sm CH Ph Ph P N N P Ph Ph Mes Mes Ph

P

N

90

Hill reported the ﬁrst divalent, homoleptic rare earth bis(iminophosphorano)-methanide complex which exhibited two samarium-methanide bonds in the solid state.49 A one pot reaction between samarium diiodide, Organomet. Chem., 2010, 36, 29–55 | 41

the parent methane, and two equivalents of potassium bis(trimethylsilyl)amide aﬀorded complex 90. Attempts to prepare the heteroleptic amido-methanide complex were unsuccessful, which is surprising given the straighforward isolation of the strontium analogue 33, and this was rationalised on the basis that samarium(II) is slightly larger than strontium(II) and thus the behaviour is more reminiscint of barium for which only a homoleptic methanide complex could be isolated (cf. 34). Ph

SiMe3 Ph

SiMe3 P

Ph

N THF N P I HC M M CH I P N THF N P

Ph

Ph

Ph Ph P

Ph

P

SiMe3 SiMe3 Ph M = Sm 91 Eu 92

Ph

Ph Ph P HC P

M

Yb

I

N THF

Ph Ph

SiMe3 93

PPh2

N N THF PPh2 SiMe3

Ph Ph M = Eu 95 Yb 96

P

SiMe3 Ph N

I

HC Sm

N

P

Ph

Sm CH N P Ph Ph SiMe3 Ph

94

SiMe3 Ph N P Ph HC M CH Ph Ph P N N P Ph Ph Me3Si SiMe3

Ph

P

Ph P N Ph Ph Me3Si

Ph Me3Si

SiMe3 N

Ph

N THF

HC

SiMe3 N

Ph

SiMe3

N

Ph Ph P

SiMe3 N THF

HC

THF BPh4

Eu

P

N THF

Ph Ph

SiMe3 100

M = Sm 97 Eu 98 Yb 99

Roesky also reported a range of divalent rare earth methanide complexes. Dimeric complexes 9150 and 9232 were prepared from 7 and one equivalent of samarium or europium diiodide, respectively and they are isostructural to the corresponding strontium analogue (36). In contrast, the ytterbium(II) complex 9332 was found to be monomeric and isostructural to the calcium analogue (35). Complexes 91–93 are easily derivatised. Complex 9450 was prepared from the reaction between dimeric 91 and one equivalent of potassium diphenylamide. Similar reactions between dimeric 92 and monomeric 93 with two and one equivalents of potassium bis(diphenylphosphanyl) amide, respectively, aﬀorded complexes 95 and 96 which were found to exhibit k2-P,N amides in the solid state (with the remaining phosphane centre not coordinated).32 The homoleptic complexes 97–99 were also prepared from the reactions between 7 and the corresponding rare earth diiodide.32,50,51 Like 90, 97–99 all exhibited metal-methanide bonds in the solid state. Lastly, the cationic europium(II) complex 100 was prepared from the reaction of 92 with one equivalent of sodium tetraphenylborate in dichloromethane.51 Surprisingly, in the solid state the europium-methanide contact was essentially unchanged compared to 92 which is surprising when considering the neutral versus cationic natures of 92 and 100, respectively. Ph Ph

NO

HC U Ph Ph

SiMe3 Ph

SiMe3 P

P

NO

Cl Cl

ON

P

Ph

U CH ON

SiMe3

P

Ph Ph P HC

Ph

SiMe3 Ph

P

Ph Ph

101

42 | Organomet. Chem., 2010, 36, 29–55

SiMe3 N O U N O

THF Cl

SiMe3 102

Ph Ph P HC P

Ph Ph

SiMe3 N O U

N(SiMe3)2

N O SiMe3 103

Sarsﬁeld reported the only three known examples of uranyl bis(iminophosphorano)-methanides reported to date.52,53 Treatment of the tris-THF adduct of anhydrous uranyl dichloride with one molar equivalent of 2 resulted in the isolation of dimeric 101 when the recrystallisation was carried out in dichloromethane, but monomeric 102 when recrystallised from THF. Both 101 and 102, when treated with two and one equivalents, respectively, of potassium bis(trimethylsilyl)amide aﬀord 103. Complexes 101–103 all exhibit uranyl-methanide contacts in the solid state and these represent noteable examples of out-of-equatorial-plane uranyl coordination. A DFT study concluded that the methanide centre coordinates to the uranyl centre with an orbital of p-type character by a highly polarised s-type interaction. 5.2

Group 3 and f-element methandiides Ph Ph

SiMe3 P

N

C Sm Ph Ph

P

N

THF NCy2

SiMe3 104

Ph Ph

SiMe3 P C

Ph Ph

P

N Y N

THF CH2SiMe3

SiMe3 105

Cavell reported the ﬁrst rare earth bis(iminophosphorano)methandiide complex 104 in 2000.54 Complex 104 was prepared from the reaction between the parent methane and the homoleptic complex samarium(III) tris(dicyclohexylamide) under forcing conditions. The samarium-methandiide bond was ascribed double bond character although shorter single samarium-carbon bonds are known. No reactivity studies of 104 have ever been reported. Liddle reported complex 105 in 2008.55 Complex 105 was prepared from the reaction between the parent methane and the bis(THF) adduct of ytrrium tris(trimethylsilylmethanide). The straightforward preparation of 105 contrasts to the synthesis of 104 since the latter required forcing conditions to eﬀect double deprotonation but this is achieved in the former at room temperature and no methanide intermediates are observed. In common with 104, the yttrium-methandiide bond distance in 105 is short and can be formally ascribed to possess double bond character, but shorter yttrium-carbon bond distances are known. This bond is maintained in solution as evidenced by 13C NMR spectroscopy (dC 60.08, JPC=131.86; JYC=4.88 Hz). A DFT study showed that the bonding in the methandiide ligand is best represented by resonance form III, as Henderson and Harder concluded (see above), and that the geminal dianion charge is primarily localised on the methandiide centre. Since 105 contains methandiide and methanide groups it was possible to directly compare the two yttrium-carbon bond orders which were found to be 0.61 and 0.51, respectively. Thus whilst 105 can be formally described as exhibiting a yttrium-carbon double bond it is more accurate to regard it as a highly polarised, latent double bond, i.e a geminal methandiide. This is underscored by inspection of the compositions of the HOMO and HOMO–2 of 105, which represent the ‘non-bonding’ and ‘pseudo-s-bonding’ molecular orbital ‘lone pairs’ on the methandiide centre which are each predominantly carbocentric (B50%) with minimal yttrium contributions (B4%). Organomet. Chem., 2010, 36, 29–55 | 43

Ph Ph

SiMe3 P

N Y

C Ph

P

THF

SiMe3

SiMe3 P C

CH2Ph

N

Ph

Ph Ph Ph

P

Ph

106

Ph

N

Ph

SiMe3 P

N

THF THF CH2Ph CH2Ph C Y N Ph N N Ph P N O Ph Ph SiMe3 Ph SiMe3 Ph Y

107

108

Ph Ph Ph P P Ph SiMe3 Me3Si N N PhH2C C CH2Ph Y O Ph O Y Ph C Ph Ph N N Me3Si SiMe3 Ph P P Ph Ph Ph 109

Since the preparation of rare earth tris(trimethylsilylmethanides) can be capricious and occasionally fail for no obvious reason, and their synthesis is limited to the smaller, heavier rare earths, Liddle reported the benzyl analogue of 105, complex 106.56 Complex 106 was prepared in a straightforward manner from the corresponding tris(THF) yttrium tribenzyl and the parent methane. In the solid state the structure of 106 is very similar to 105 except the benzyl is bound with a weak yttrium-ipso-carbon interaction. The yttrium-methandiide bond is maintained in solution as evidenced by 13C NMR spectroscopy (dC 61.81, JPC=207.3; JYC=5.03 Hz). A preliminary study of the reactivity of 106 has been reported.56 The reaction of 106 with one equivalent of diphenyldiazene gave the 1,2-migratory insertion product 107 in which the PhNNPh unit inserts into the yttrium-alkyl bond. In the solid state the newly formed PhNN(Ph)(CH2C6H5) ligand coordinates in a k2-N,N manner. The yttriummethandiide bond observed in the solid state is maintained in solution as evidenced by 13C NMR spectroscopy (dC 60.0, JPC=207.3; JYC=5.03 Hz). Complex 106 does not react with further equivalents of diphenyldiazene. The reaction of 106 with benzophenone does not furnish the expected alkene Ph2CQC(PPh2NSiMe3)2; instead, 1,2-migratory insertion at the yttrium-alkyl bond occurs to give the corresponding alkoxide 108 (13C NMR of methandiide centre: dC 51.04, JPC=134.85; JYC=7.04 Hz). Addition of an additional equivalent of benzophenone still does not give the Wittig-product, rather a dimerisation of 108 to 109 occurs on heating which is, remarkably, directly proportional to the quantity of benzophenone added. Ph Ph

SiMe3 P C

Ph

P

N M N

THF CH2Ph

Ph SiMe3 M = Dy 110 Er 111

Ph Me3Si Ph P N

SiMe3 Ph N P Ph C M CH Ph Ph P N N P Ph Ph Me3Si SiMe3 M = La 112 Ce 113 Pr 114 Nd 115 Sm 116 Gd 117

44 | Organomet. Chem., 2010, 36, 29–55

Ph Ph P C

SiMe3 N THF M

I

P N THF Ph SiMe3 Ph M = Er 118 Y 119

Ph Ph P C P Ph Ph

SiMe3 Ar

N THF

N Ga N N THF Ar SiMe3 Y

120

Liddle extended the range of reactivity between rare earth tribenzyls and the parent methane but encountered selectivity problems associated with the size of the metal.57 The dysprosium (110) and erbium (111) analogues of 106 were prepared in a straighforward manner.57 However, attempts to prepare the analogous lanthanum, cerium, praseodymium, neodymium, samarium, and gadolinium complexes resulted in the isolation of 112–117 as the sole isolable rare earth containing compounds.57 Apparently, the large size of those metals prevents selective formation of analogues of 106 and deprotonation of a second ligand proceeds to give 112–117. Solid state structures of 112–117 showed that in all cases the metal is bound to the methanide and methandiide centres, and for the former a boat conformation of the CP2N2M ring is exhibited but for the latter the CP2N2M ring is almost planar. Since the benzyl methodology proved to have limitations, Liddle investigated the synthesis of 118 and 119 as the presence of an iodide ligand aﬀords great synthetic utility. Complexes 118 and 119 were prepared in a straightforward manner from the parent methane and the tris(THF) adducts of iododibenzyl yttrium and erbium which are themselves noteable for their kinetic stability with respect to ligand scrambling by Schlenk-type equilibria and their thermal stability.58 Complexes 118 and 119 exhibit metal-methandiide bonds in the solid state, which for yttrium was shown to persist in solution (13C NMR of methandiide centre: dC 60.28, JPC=207.3; JYC=5.03 Hz), and the respective CP2N2M rings are planar and exhibit approximately T-shaped methandiide centres where the possibility of p-bonding is maximised. However, a DFT study of 119 again showed that resonance form III is most representative of the electronic structure of the methandiide ligand and although the two methandiide ‘lone pairs’ are polarised towards yttrium no p bond is in fact manifested and the bonding is highly polarised in nature. However, the same study reported a DFT study of the archetypal carbodiphosphorane Ph3PQCQPPh3 and the similarity of the compositions of the frontier orbitals of this carbodiphosphorane to the methandiide ligand renders the assignment of these methandiide ligands as ‘captodative’ carbon(0) complexes of erbium and yttrium an intriguing possibility which cannot be ruled out. Liddle demonstrated the synthetic utility of 119 by its conversion to the gallyl-yttrium-methandiide complex 120,47 which contains the ﬁrst example of a molecular gallium-yttrium bond. Complex 120 can be prepared from the reaction between 119 and [K(tmeda)][Ga{(NArCH)2}] (tmeda= Me2NCH2CH2NMe2; Ar=2,6-diisopropylphenyl) or alternatively from the in situ treatment of 86 with one equivalent of benzyl potassium to generate 119 in situ followed by addition of the gallyl reagent. A DFT study of 120 showed little change in the nature of the bonding of the CP2N2Y ring on conversion from 119 to 120. Pri

Ph

Ph Ph THF THF Pri Ph P Ph S P Ph S N P Ph I C Nd CH C M M C Ph I Ph P N N P Ph P S S P Ph Ph i THF THF Ph Ph Pri Pr Ph 121 M = Sm 122 Tm 123 Ph

P

N

Ph Ph

P C

Ph

P

S

Ph S

M S

P

Ph

P

Ph

C S

Ph Ph [Li(THF)4)] M = Sm 124 Tm 125

Organomet. Chem., 2010, 36, 29–55 | 45

THF Ph Ph Li Ph C O O C Ph Ph Ph M C C P S S P Ph P S S P Ph Ph Ph Ph Ph M = Sm 126 Tm 127

Simultaneously to the disclosure of 112–117 by Liddle, Le Floch reported that complex 88 could be converted to complex 121, which is analogous to 112–117, by reaction with potassium bis(trimethylsilyl)amide.48 The second deprotonation of one of the methanide ligands to give a methandiidemethanide complex is noteable because amides, especially silyl-amides, are not usually basic enough to eﬀect a second deprotonation at a carbon centre (cf. synthesis of 1–8) and it was suggested that coordination to the electropositive neodymium polarised the C–H bond suﬃciently to activate it. Le Floch, Nief, and Me´zailles reported the ﬁrst applications of the bis (diphenylthiophosphinoyl)methandiide ligand C(PR2S)22 in rare earth chemistry with the synthesis of 122–125. Complexes 122 and 123 were prepared from the reactions between dimeric 21 and two molar equivalents of samarium(III) or thulium(III) triiodide.59,60 In the solid state 122 and 123 are dimeric, constructed around M2I2 four-membered rings and the CP2S2M rings are planar. By using only one molar equivalent of samarium(III) or thulium(III) triiodide the separated ion pair ‘ate’ complexes 124 and 125 were isolated. In common with 122 and 123, complexes 124 and 125 possess planar CP2S2M cores, but for 125 a solid state dimorphism was observed where boat and planar conformations of one of the CP2S2M rings was observed at 150 and 230 K, respectively. The reactivity of 122 and 123 towards benzophenone was tested and, in contrast to 106, the anticipated Wittig-product Ph2CQC(PPh2S)2 was produced within one hour. Complexes 124 and 125 also reacted with benzophenone to produce Ph2CQC(PPh2S)2 but the ‘open’ metallo-oxetane intermediates 126 and 127 could be isolated in which the methandiide centres have nucleophilically attacked the electrophillic ketyl carbon of benzophenone and the metalmethandiide bonds have been cleaved due to steric constraints. These ‘open’ metallo-oxetanes are noteworthy because they are often proposed as intermediates in the reactions of early metal carbene complexes with carbonyl compounds but they are seldom isolated due to their instability with respect to the metal oxide and alkene reaction products. BH4 P P BH4 U Li Et O 2 P S S P S S C C P BH4 C C P P P S S BH4 S S S U S U P C P C BH4 U Et2O Li S S P P H4B P-Ph groups omitted for clarity P-Ph groups omitted for clarity 128

46 | Organomet. Chem., 2010, 36, 29–55

129

Ph Ph

Ph

P

S BH4 THF C U THF P S BH4

Ph 130

Me´zailles and Ephritikhine reported the only three known examples of the bis(diphenylthiophosphinoyl)methandiide ligand C(PR2S)22 in actinide chemistry to date.61 Noting that reaction of complex 21 with uranium(IV) tetrachloride gave mixtures of products in toluene or diethyl ether and THF is incompatible with 21, the borohydride [U(BH4)4] was employed. When the reaction was conducted in diethyl ether complex 128 was isolated in which three methandiide ligands are associated with one uranium centre and two lithium counter cations are occluded. Changing the reaction solvent to toluene aﬀorded complex 129 in which three methandiide ligands are primarily associated with one uranium centre with two ‘capping’ uranium centres which are each ligated by three tetrahydroborate anions. However, additon of THF to complex 129 aﬀorded rearrangement to give complex 130 as a monomeric compound. The solid state structure of 130 showed a short uraniummethandiide bond and the two tetrahydroborate ligands bind in a k3-mode. The reactivity of 130 towards benzophenone was assessed and the expected alkene Wittig-product Ph2CQC(PPh2S)2 was isolated. The nucleophillic nature of the methandiide centre in 130 was underscored by its reactivity towards 9-anthracene carboxaldehyde which reacted more rapidly than benzophenone. DFT calculations on the nature of the uranium-methandiide bond were carried out and showed the involvement of 5f and 6d orbitals from uranium in the s and p components of the UQC double bond. Covalent overlap in the molecular orbital manifold could be visualised in the NBO plots, the contributions to the p and s components were found to be a carbon 2p lone pair (82.9%) and uranium metal hybrid orbital (17.1% composed of 59.0% 5f and 40.9% 6d) and a carbon lone pair (80.7%) and uranium metal hybrid (19.3% composed of 52.6% 5f and 37.0% 6d), respectively. 6. 6.1

Group 4 methanides and methandiides Group 4 methanides

Perhaps remarkably, there are few examples of bis(iminophosphorano)methanide complexes of group 4 metals and those that are known are derived from the corresponding methandiide complexes as a result of 1,2-addition of protic reagents across the metal-carbon double bond.62 Additionally, only one report of a structurally authenticated bis(iminophosphorano)methanide complex of hafnium has been published.63 These additions conﬁrm the expected polarity of the metal-carbon double bond in that the proton is always captured by the methandiide centre and the anion component always coordinates to the metal centre. Since these compounds are generated from the corresponding methandiide complexes they are presented in Section 6.2 below. 6.2

Group 4 Methandiides Ph Ph

SiMe3 P C

Ph Ph

P

N Ti N

Cl Cl

SiMe3 131

Ph Ph

SiMe3 P C

Ph Ph

P

N Zr N

Cl Cl

SiMe3 132

Ph Ph

SiMe3 P

N

C Ph

P

Hf N

Ph

Cl Cl

SiMe3 133

Cy Cy

SiMe3 P

N

C Cy

P

Hf N

Cy

Cl Cl

SiMe3 134

Organomet. Chem., 2010, 36, 29–55 | 47

Cavell initiated a comprehensive study of bis(iminophosphorano)methandiide group 4 complexes by the synthesis of the triad of complexes 131– 133.64,65 Complexes 131–133 were prepared from the reactions between 13 and the corresponding group 4 tetrachloride in diethyl ether or toluene. In all three complexes solid state studies showed the CP2N2M rings to be essentially planar and the methandiide centres approach T-shaped geometries. Complexes 131–133 exhibited excellent stabilities and were found to be stable in toluene solutions heated to 140 1C for seven days. In contrast, complex 104 was found to decompose to unidentiﬁable products under the same conditions. To underscore the utility of this synthetic approach, Cavell also prepared the P-cyclohexyl analogue of 133, namely complex 134.65 Interestingly, the 13C NMR chemical shifts of the methandiide centres were found to range from vastly diﬀerent to being very similar to the yttrium analogues (105–109, 119, and 120), appearing at 191.0, 101.7, 84.6, and 66.6 ppm for 131–134, respectively. Ph

Ad

Ph

P

Ph

P

C Ph

N Zr N

CH2Ph CH2Ph

Ad 135

Ph P

Ph

P

C Ph

Ph

SiMe3

Ph

N Zr N

CH2Ph CH2Ph

SiMe3

Ph

P

Ph

P

N

C

SiMe3 136

Hf N

Ph

CH2But CH2But

SiMe3 137

Me Me3Si Me P N

SiMe3 Me N P Me C Zr C Me Me P N N P Me Me Me Si SiMe3 3 138

Cavell also reported alkyl derivatives. Tetrabenzyl zirconium was found to react separately with one molar equivalent of two parent methanes to aﬀord complexes 135 and 136 in excellent yields.66 The 13C NMR chemical shifts of the methandiide centres in 135 (Ad=adamantyl) and 136 of 82.8 and 84.4 ppm are clearly diﬀerent from the dichloride congeners and reﬂect the substitution of the chlorides by alkyls. The corresponding hafnium dialkyl 137 was also prepared from 133 and neopentyl lithium and the methanide resonance for this complex appeared at 71.6 ppm in its 13C NMR spectrum.63 Noteably, these alkyls displayed far greater stability than their homoleptic tetraalkyl cousins. By adjusting the reaction stoichiometry, Cavell was able to access the bismethandiide complex 138.67 Complex 138 was prepared from tetrabenzyl zirconium and two molar equivalents of the parent methane. In the solid state the two methandiide ligands are disposed orthogonal with respect to each other, and each CP2N2M ring is essentially planar with close to T-shaped methandiide centres. The 13C NMR chemical shift of the methandiide centre was found to be 77.8 ppm and is consistent with the above examples. Cavell investigated the reactivity of 132 and 133 towards a wide range of substrates including THF, nitriles, iso-nitriles, amines, alcohols, alkyl halides, carbon dioxide, isocyanates, and carbodiimides. Three classes of reactivity emerged: (i) base adduct formation; (ii) 1,2-addition across the MQC bond; (iii) [2 þ 2] cycloadditions across the MQC double bond.62,63

48 | Organomet. Chem., 2010, 36, 29–55

Ph Ph

SiMe3 P C

Ph

P

Ph

Ph Ph

Ph Ph

SiMe3 P

N

Cl HC Hf Cl P N HN SiMe3 142

N Hf N

Ph Ph

Cl THF Cl

C Ph

SiMe3 139

P

Ph

Ph Ph

N Hf N

N

Ph

SiMe3 143

SiMe3 P C

Ph

P

Ph

SiMe3 140

Cl HC Zr Cl P N O Ad

Ph

Ph

Cl Cl NCAd

SiMe3 P

Ph

SiMe3 P

Ph Ph

N Zr N

SiMe3 141

SiMe3 P

N

Cl MeC Hf Cl Ph P N I Ph

Cl CN Cl

SiMe3 144

O

C O Cl Ph Hf Cl C Ph P P N N SiMe3 Ph SiMe3 Ph 145

Ad Ad Cy p-Tolyl O N N N Cl C N Cl C N Cl C C N Cl Cy p-Tolyl Ph Ph Ph Ph Zr Cl Hf Cl Zr Cl Hf Cl C C C C P N P N Ph P P N N Ph P P N N Ph Ph SiMe3 P N SiMe3 SiMe3 P N SiMe3 Ph Ph Ph Ph SiMe3 Ph SiMe3 SiMe3 Ph Ph SiMe3 Ph O

146

147

148

149

The reaction between 133 and THF or adamantyl cyanide aﬀorded the base adducts 139 and 140, respectively. Complex 140 was characterised by X-ray crystallography and in the solid state one chloride was found to be trans to the methandiide centre, mutually cis to the other chloride which was in turn trans to the coordinated nitrile. The base adduct 141 resulted from the addition of 2,6-dimethylphenyl isocyanide to 132. Neither 139 or 141 were characterised in the solid state so it is not known which geometric isomers were formed. The addition of para-tolylamine to 133 aﬀorded the 1,2-addition complex 142 which was characterised in the solid state conﬁrming the cis orientation of the amide to the newly formed methanide centre. Likewise, reaction of 132 with adamantyl alcohol and 133 with methyl iodide, respectively, afforded the 1,2 addition products 143 and 144. The regioselective nature of these 1,2-additions conﬁrmed the nucleophilic nature of the methandiide centres and this contrasts to electrophilic Fischer carbene centres which are usually attacked by incoming nucleophiles. In contrast, complexes 142–144 are formed by attack of electrophilic centres at the methandiide centres. Complexes 132 and 133 reacted with a range of unsaturated substrates to aﬀord [2 þ 2] cycloaddition products. Addition of carbon dioxide to 133 gave complex 145 as adjudged by FTIR spectroscopy, mass spectrometry, and combustion analysis. Complexes 133 and 132 aﬀorded complexes 146 and 147, respectively, after addition of adamantyl isocyanate. The formation of 146 and 147 is noteworthy from a regioselective point of view in that 1,2-addition gives the carbonyl-amide product. However, the alkoxideimine product might have been expected given that the Lewis acidity of

Organomet. Chem., 2010, 36, 29–55 | 49

group 4 metals should make M–O bond formation more favourable than M-N bond formation. Addition of carbodiimides to 133 and 132 aﬀorded the [2 þ 2] cycloaddition products 148 and 149, respectively, as expected. Lastly, three compounds [Zr(Cl)3{CH(PMe2NSiMe3)2}], [Zr(Cl)3{CH (PPh2N-SiMe3)2}], and [Hf(Cl)2{(X)CH(PPh2NSiMe3)2}] are listed in a review by Cavell62 but their preparation and the identity of X are not described. These compounds were apparently characterised in the solid state but are not reported in the open literature. Ph Ph P

SiMe3 N

C P

Ph Ph

Zr N

Cl SiMe3

150

The only other zirconium bis(iminophosphorano)methandiide was reported by Roesky. Complex 150 was prepared from zirconocene dichloride and 7 with concomitant elimination of KCl and C5H6.68 Complex 150 is similar to 132 except, for the obvious diﬀerence that one chloride is replaced by a cyclopentadienyl group, the CP2N2Zr ring is not planar. Ph Ph

Ph P

S

C Ph

P

Ph

Zr S

Ph

Ph

Ph S Cl THF S P Cl C Zr Zr C Cl P S THF Cl S P P

Ph 151

Ph 152

Ph

Ph Ph

P C

Ph

Ph

P

S Cl Zr S Cl

py py

Ph 153

Le Floch reported the only three known applications of the {C(PPh2S)2}2 methandiide ligand in group 4 chemistry.69 The reaction beween zirconocene dichloride and 21 aﬀorded complex 151. Complex 151 was characterised in the solid state and was found to have a planar CP2S2Zr ring. For comparison, the dimeric complex 152 was prepared from 21 and the bis(THF) adduct of zirconium tetrachloride. However, following the reaction by NMR spectroscopy showed that 152 was ﬁrst formed as a monomeric species solvated by two THF molecules but over two hours one THF molecule is lost and dimerisation occurs to give the much less soluble 152. Complex 152 was consequently diﬃcult to characterise. However, addition of pyridine to 152 aﬀorded the monomeric complex 153. Interestingly, complex 151 was found to be unreactive towards aldehydes and ketones, presumably because of the sterically crowded zirconium centre and the fact that it is, formally, a twenty valence electron complex. However, complexes 152 and 153 were found to be highly reactive towards a range of aldehydes and ketones giving the corresponding alkene Wittig-reaction products. A DFT study of a simpliﬁed model of 151 showed the ZrQC bond to have a Wiberg bond index of only 0.69, and whilst the two ‘lone pairs’ of the methandiide centre show overlap with the zirconium-based dxz 50 | Organomet. Chem., 2010, 36, 29–55

orbital, this interaction is depleted by interaction of the sulfur lone pairs with the zirconium centre. Additionally the summed contribution of the dxz orbital is only 11%. A study of the isomer where the sulfurs are not coordinated to the zirconium found an increase in the ZrQC Wiberg bond index to 1.16 compensating the loss of donation of two sulfur lone pairs but that this isomer is 30.0 kcal/mol 1 higher in energy. Although the coordination of the two sulfur lone pairs reduces the ZrQC bonding interaction this is more than compensated by the formation of Zr–S s-bonds. An analysis of the bonding manifold showed that complex 151 can in fact be considered an eighteen valence electron complex. This arises because the methandiide ligand donates four electrons from two sulfur lone pairs and two electrons from a methandiide lone pair. The other lone pair remains essentially non-bonding and/or undergoes polarisation towards the phosphorus centres by polarisation eﬀects and/or negative hyperconjugation. A DFT analysis of a simpliﬁed model of 153 showed a similar bonding picture although the presence of chloride ligands rendered the dxz orbital contribution slightly higher at 13% and the Wiberg bond index of the ZrQC bond was found to be higher at 0.80. Although formally containing ZrQC double bonds, these systems in fact contain highly polarised metal carbon bonds of low bond order. However, their reactivity with aldehydes and ketones shows they may be regarded as ‘latent’ MQC bonds. 7.

Group 5 methanides and methandiides

7.1

Group 5 methanides

Ph

SiMe3 P

Ph

N V

HC P

Ph Ph

Ph Ph

Ph

Cl

P

P

N V N

H Cl H

SiMe3

P

P

Ph Ph

N V N

N

N

Me Me

Ph

P

Ph

Ph

CH P

Ph

Ph Ph

SiMe3

SiMe3 Ph

N

N

V N

H

V

H

P CH

N

SiMe3

Ph

P

Ph Ph

Ph

SiMe3 Ph

Me3Si P

Ph

Ph

HC P

N Me3Si

P

P

N V N

N

Cl H

V

Cl

N

SiMe3

158

SiMe3 Ph P CH Ph

N V

N P Ph O SiMe3 Ph 157

SiMe3 Ph

SiMe3

O

N V

156

HC

SiMe3 Ph

P

Ph

Ph

Ph

P HC

SiMe3 155

SiMe3 Ph V

Ph

SiMe3 HC

SiMe3 154

SiMe3 HC

Ph

N

Cl

Ph Ph

P

Ph Ph

CH P

SiMe3

Ph P HC

Ph Ph

SiMe3 Ph

159

Ph

P

N V N

SiMe3 Ph

H H H

N V N

P

Ph

CH P

Ph

SiMe3 Ph

SiMe3 160

Ph CH SiMe3 Ph CH P Ph N Ph N H P V V CH H Ph HC N P Ph H N P Ph SiMe3 Ph Me3Si Ph Me3Si

161

Organomet. Chem., 2010, 36, 29–55 | 51

Group 5 bis(phosphorus-stabilised)methanides are surprisingly sparse, and the only known examples were reported by Gambarotta.70,71 Reaction of 1 with the tris(THF) adduct of vanadium trichloride gave 154 in good yield. Complex 154 was easily converted to complex 155 by treatment with two equivalents of methyl lithium and this complex was characterised in the solid state, which showed a typical boat conformation of the CP2N2V ring, and it was found to be isostructural to 154. Since the aim of the study was to investigate vanadium hydride complexes as models for intermediates in the catalytic production of elastomers, complex 155 was converted to complex 156 by hydrogenolysis. During the hydrogenation the vanadium centre was reduced from V(III) to V(II) and the complex was found to be a dimer in the solid state. The magnetic moment for 156 was found to be low and this, together with the close approach of the two vanadium centres to each other, suggested a direct vanadium-vanadium interaction. Complex 156 was described as highly reactive and reaction with water converted divalent 156 into trivalent 157 with concomitant elimination of three equivalents of hydrogen and cleavage of the N–Si bond. Thus, in addition to acid-base chemistry a one-electron oxidation reaction occurs. In common with 156, complex 157 displayed a low mangetic moment which was attributed to a vanadiumvanadium interaction. In the vanadium catalysed production of elastomers the divalent state is regarded as catalytically deactivated and oxidation to the trivalent state restores reactivity. This is usually achieved by addition of halogenated substances which implies transfer of a chloride to the vanadium centre during oxidation. Thus the synthesis of mixed halide/hydride complexes of vanadium was attempted to evaluate their stability. Attempts to stepwise partially alkylate 154 followed by hydrogenolysis were not successful so the steps were combined without work up and depending on the ﬁnal work up conditions the mixed valence complexes 158 and 159 were isolated which infers the presence of chloride ligands can prevent complete reduction. The reaction of alkenes with 156 was investigated. Ethylene was found to react with 156 to give the mixed valence dimer 160. An examination of the reaction mother liquor revealed the presence of 2-butene and 3-methyl-1,4-pentadiene, indicating that 156 promotes ethylene di- and tri-merisation. Addition of styrene to 156 aﬀorded the oxidative addition product 161. The two bridging hydrides were located in the diﬀerence map of the crystal structure. However, the terminal hydride was not located, but its presence was supported by the otherwise vacant coordination site at one of the vanadium centres and NMR and FTIR spectroscopy. Since divalent 156 is produced from hydrogenolysis of trivalent 155 it was reasoned that hydrogenation of 161 would give catalytic hydrogenation of styrene. Accordingly, catalytic production of ethylbenzene was achieved at low turn overs and although this activity is modest compared to late metal hydrogenation systems it was the ﬁrst example of catalytic hydrogenation of styrene by vanadium. 7.2

Group 5 methandiides

Surprisingly, considering the prevalence of Schrock alkylidenes, there are no structurally characterised group 5 bis(phosphorus-stabilised)methandiide complexes. 52 | Organomet. Chem., 2010, 36, 29–55

8.

Conclusions

In this Review, which has focussed mainly on structurally authenticated examples, we have covered a burgeoning area of early metal chemistry covering synthesis, bonding, and reactions supported by, or involving, {CH(PR2NR 0 )2} methanide and {C(PR2NR 0 )2}2 and {C(PPh2S)2}2 methandiide ligands which has emerged in just over ten years. These ligand systems are interesting as supporting ancillary ligands but also in their own right in terms of their bonding and reactivity which is distinct and complimentary to other carbenes which do not exist in the absence of a metal centre. Novel reactivities have been observed and the utility of these ligand classes is demonstrated by the fact they have found applications from the lightest metal lithium all the way to the heaviest naturally occuring metal uranium. Furthermore, these ligands have utility in late metal chemistry which is covered elsewhere.72–74 The combination of novel structures, bonding, and reactivity gives great promise to future endeavours in the area.

References 1 C. Elschenbroich, Organometallics, 3rd edition, 2005, Wiley-VCH, Weinham, Germany. 2 P. J. Davidson, M. F. Lappert and R. Pearce, Chem. Rev., 1976, 76, 219. 3 M. Bochmann, Organometallics 1 Complexes with Transition Metal-Carbon s-Bonds, 1994, Oxford University Press, Oxford, UK. 4 B. Romer, G. G. Gatev, M. Zhong and J. I. Brauman, J. Am. Chem. Soc., 1998, 120, 2919. 5 A. M. El-Nahas and P. von R. Schleyer, J. Comput. Chem., 1994, 15, 596. 6 P. von R. Schleyer, T. Clark, A. J. Kos, G. W. Spitznagel, C. Rohde, D. Arad, K. N. Houk and N. G. Rondan, J. Am. Chem. Soc., 1984, 106, 6467. 7 K. Izod, Adv. Inorg. Chem., 2000, 50, 33. 8 K. Izod, Coord. Chem. Rev., 2002, 227, 153. 9 M. Alcarazo, C. W. Lehmann, A. Anoop, W. Thiel and A. Fu¨rstner, Nature Chem., 2009, 1, 295. 10 J. Scott and D. J. Mindiola, Dalton Trans., 2009, 8463. 11 R. R. Schrock, Acc. Chem. Res., 1979, 12, 98. 12 D. J. Mindiola, B. C. Bailey and F. Basuli, Eur. J. Inorg. Chem., 2006, 3135. 13 R. Appel and I. Ruppert, Z. Anorg. Allg. Chem., 1974, 406, 131. 14 A. V. Kirsanov, Izv. Akad. Nauk. SSSR, 1950, 426. 15 P. Imhoﬀ, R. Vanasselt, C. J. Elsevier, K. Vrieze, K. Goubitz, K. F. Vanmalssen and C. H. Stam, Phosphorus Sulfur and Silicon and the Related Elements, 1990, 47, 401. 16 R. P. Kamalesh Babu, K. Aparna, R. McDonald and R. G. Cavell, Inorg. Chem., 2000, 39, 4981. 17 R. P. Kamalesh Babu, K. Aparna, R. McDonald and R. G. Cavell, Organometallics, 2001, 20, 1451. 18 M. T. Gamer and P. W. Roesky, Z. Anorg. Allg. Chem., 2001, 627, 877. 19 W.-P. Leung, C.-W. So, Z.-X. Wang, J.-Z. Wang and T. C. W. Mak, Organometallics, 2003, 22, 4305. 20 M. S. Hill, P. B. Hitchcock and S. M. A. Karagouni, J. Organomet. Chem., 2004, 689, 722. 21 S. A. Ahmed, M. S. Hill and P. B. Hitchcock, Organometallics, 2006, 25, 394. Organomet. Chem., 2010, 36, 29–55 | 53

22 M. Demange, L. Boubekeur, A. Auﬀrant, N. Me´zailles, L. Ricard, X. Le Goﬀ and P. Le Floch, New. J. Chem., 2006, 30, 1745. 23 J.-H. Chen, J. Guo, Y. Li and C. W. So, Organometallics, 2009, 28, 4617. 24 A. Kasani, R. P. Kamalesh Babu, R. McDonald and R. G. Cavell, Angew. Chem. Int. Ed., 1999, 38, 1483. 25 C. M. Ong and D. W. Stephan, J. Am. Chem. Soc., 1999, 121, 2939. 26 K. L. Hull, B. C. Noll and K. W. Henderson, Organometallics, 2006, 25, 4072. 27 K. L. Hull, I. Carmichael, B. C. Noll and K. W. Henderson, Chem. Eur. J., 2008, 14, 3939. 28 T. Cantat, L. Ricard, P. Le Floch and N. Me´zailles, Organometallics, 2006, 25, 4965. 29 L. Orzechowski, G. Jansen and S. Harder, Angew. Chem. Int. Ed., 2009, 48, 3825. 30 P. Wei and D. W. Stephan, Organometallics, 2003, 22, 601. 31 M. S. Hill and P. B. Hitchcock, Chem. Commun., 2003, 1758. 32 T. K. Panda, A. Zulys, M. T. Gamer and P. W. Roesky, J. Organomet. Chem., 2005, 690, 5078. 33 M. Wiecko, S. Marks, T. K. Panda and P. W. Roesky, Z. Anorg. Allg. Chem., 2009, 635, 931. 34 L. Orzechowski, G. Jansen and S. Harder, J. Am. Chem. Soc., 2006, 128, 14676. 35 L. Orzechowski and S. Harder, Organometallics, 2007, 26, 5501. 36 L. Orzechowski and S. Harder, Organometallics, 2007, 26, 2144. 37 M. T. Gamer, S. Dehnen and P. W. Roesky, Organometallics, 2001, 20, 4230. 38 M. T. Gamer and P. W. Roesky, J. Organomet. Chem., 2002, 647, 123. 39 M. T. Gamer, P. W. Roesky, I. Palard, M. Le Hellaye and S. M. Guillaume, Organometallics, 2007, 26, 651. 40 M. T. Gamer, M. Rasta¨tter, P. W. Roesky, A. Steﬀens and M. Glanz, Chem. Eur. J., 2005, 11, 3165. 41 T. K. Panda, M. T. Gamer and P. W. Roesky, Inorg. Chem., 2006, 45, 910. 42 M. Rasta¨tter, A. Zulys and P. W. Roesky, Chem. Commun., 2006, 874. 43 M. Rasta¨tter, A. Zulys and P. W. Roesky, Chem. Eur. J., 2007, 13, 3606. 44 A. Zulys, T. K. Panda, M. T. Gamer and P. W. Roesky, Chem. Commun., 2004, 2584. 45 T. K. Panda, A. Zulys, M. T. Gamer and P. W. Roesky, Organometallics, 2005, 24, 2197. 46 T. K. Panda, P. Benndorf and P. W. Roesky, Z. Anorg. Allg. Chem., 2005, 631, 81. 47 S. T. Liddle, D. P. Mills, B. M. Gardner, J. McMaster, C. Jones and W. D. Woodul, Inorg. Chem., 2009, 48, 3520. 48 A. Buchard, A. Auﬀrant, L. Ricard, X. F. Le Goﬀ, R. H. Platel, C. K. Williams and P. Le Floch, Dalton Trans., 2009, 10219. 49 M. S. Hill and P. B. Hitchcock, Dalton Trans., 2003, 4570. 50 M. Wiecko, P. W. Roesky, V. V. Burlakov and A. Spannenberg, Eur. J. Inorg. Chem, 2007, 876. 51 M. Wiecko and P. W. Roesky, Organometallics, 2009, 28, 1266. 52 M. J. Sarsﬁeld, M. Helliwell and D. Collison, Chem. Commun., 2002, 2264. 53 M. J. Sarsﬁeld, H. Steele, M. Helliwell and S. J. Teat, Dalton Trans., 2003, 3443. 54 K. Aparna, M. Ferguson and R. G. Cavell, J. Am. Chem. Soc., 2000, 122, 726. 55 S. T. Liddle, J. McMaster, J. C. Green and P. L. Arnold, Chem. Commun., 2008, 1747. 56 D. P. Mills, O. J. Cooper, J. McMaster, W. Lewis and S. T. Liddle, Dalton Trans., 2009, 4547. 54 | Organomet. Chem., 2010, 36, 29–55

57 A. J. Wooles, D. P. Mills, W. Lewis, A. J. Blake and S. T. Liddle, Dalton Trans., 2010, 39, 500. 58 D. P. Mills, A. J. Wooles, J. McMaster, W. Lewis, A. J. Blake and S. T. Liddle, Organometallics, 2009, 28, 6771. 59 T. Cantat, F. Jaroschik, F. Nief, L. Ricard, N. Me´zailles and P. Le Floch, Chem. Commun., 2005, 5178. 60 T. Cantat, F. Jaroschik, L. Ricard, P. Le Floch, F. Nief and N. Me´zailles, Organometallics, 2006, 25, 1329. 61 T. Cantat, T. Arliguie, A. Noe¨l, P. Thue´ry, M. Ephritikhine, P. Le Floch and N. Me´zailles, J. Am. Chem. Soc., 2009, 131, 963. 62 R. G. Cavell, R. P. Kamalesh Babu and K. Aparna, J. Organomet. Chem., 2001, 617, 158. 63 R. P. Kamalesh Babu, R. McDonald and R. G. Cavell, Organometallics, 2000, 19, 3462. 64 R. G. Cavell, R. P. Kamalesh Babu, A. Kasani and R. McDonald, J. Am. Chem. Soc., 1999, 121, 5805. 65 R. P. Kamalesh Babu, R. McDonald and R. G. Cavell, Chem. Commun., 2000, 481. 66 R. P. Kamalesh Babu, R. McDonald, S. A. Decker, M. Klobukowski and R. G. Cavell, Organometallics, 1999, 18, 4226. 67 K. Aparna, R. P Kamalesh Babu, R. McDonald and R. G. Cavell, Angew. Chem. Int. Ed., 2001, 40, 4400. 68 M. T. Gamer, M. Rasta¨tter and P. W. Roesky, Z. Anorg. Allg. Chem., 2002, 628, 2269. 69 T. Cantat, L. Ricard, N. Me´zailles and P. Le Floch, Organometallics, 2006, 25, 6030. 70 G. Aharonian, K. Feghali, S. Gambarotta and G. P. A. Yap, Organometallics, 2001, 20, 2616. 71 G. Aharonian, S. Gambarotta and G. P. A. Yap, Organometallics, 2001, 20, 5008. 72 N. D. Jones and R. G. Cavell, J. Organomet. Chem., 2005, 690, 5485. 73 T. K. Panda and P. W. Roesky, Chem. Soc. Rev., 2009 J. Organomet. Chem., 2001, 38, 2782. 74 G. Ma, R. McDonald and R. G. Cavell, Organometallics, 2010, 29, 52.

Organomet. Chem., 2010, 36, 29–55 | 55

The stabilisation of organometallic complexes using m-terphenyl ligands Deborah L. Kaysa DOI: 10.1039/9781847559616-00056

The use of bulky monodentate m-terphenyl ligands in the stabilisation of d-block organometallic compounds is surveyed. Importantly, these ligands have facilitated the isolation of hitherto unknown species containing low-coordinate centres and metal-metal multiple bonds. This review reports on these advances with emphasis on the synthesis, structural characterisation and, where possible, reactivity studies of complexes featuring metal-carbon bonds between m-terphenyl ligands and the transition metals.

1.

Introduction

The use of bulky monodentate ligand systems in the stabilisation of highly reactive metal centres and unusual bonding modes is a key research theme in organometallic and coordination chemistry. As such, synthetic strategies involving the utilisation of sterically demanding chalcogenolato (ER, E=O, S, Se),1 pnictide (ER2, E=N, P, As),2 and to a lesser extent, alkyl,3 aryl4 and silyl5 ligands in the stabilisation of highly reactive transition metal complexes is well established.6–8 Over the last decade or so, the synthetic utility of m-terphenyl ligands has been investigated, leading to a number of compounds featuring unusual and highly reactive bonding modes for the main group elements. Examples of such compounds include the ﬁrst stable species with a triple bond to a heavier group 14 element, one-coordinate In and Tl centres in the solid state, a homologous series of dipnictines, heavier group 14 alkyne analogues, the ﬁrst stable, well characterised group 14 divalent hydrides, stable molecular hydrides of elements of the sixth period, multiple bonding for the group 13 elements, an unsaturated main group compound which can undergo facile activation of dihydrogen and unusual bonding motifs for main group clusters.9–17 Thus the synthesis and investigation of m-terphenyl compounds of the main group elements have been of signiﬁcant recent interest and as such are the subject of a number of reviews.18 Complexes featuring M–C s bonds between a transition metal and a monoanionic m-terphenyl ligand are somewhat rarer; however, a surge of reports of complexes for these elements is now gathering momentum, with a number of spectacular results now being seen for low-coordinate and multiply bonded transition metal centres. In this review the m-terphenyl complexes of the d-block metals will be discussed, with emphasis on their synthesis, structural investigations and, where possible, reactivity studies. 2.

Ligands

m-Terphenyl ligands are generally of the type 2,6-Ar2C6H3 in which two aryl groups are situated meta to each other on a central aryl ring, giving rise to a a

School of Chemistry, University of Nottingham, University Park, Nottingham, U.K.

56 | Organomet. Chem., 2010, 36, 56–76 c

The Royal Society of Chemistry 2010

steric pocket with shielding provided by the ﬂanking aryl groups. Related ligands such as 2,4,6-Ph3C6H2 have also been investigated for the stabilisation of highly reactive d-block complexes, and will also be discussed in this review. The synthetic route towards m-terphenyls was described by Hart and co-workers; it involves the reaction of a 1,3-dihalo-2-iodobenzene with three equivalents of an aryl Grignard, which proceeds through the sequential creation and capture of two aryne intermediates with Grignard reagents, followed by quenching with a halogen.19 This route is being superseded by a methodology involving the generation of the lithium complex of 1,3-dichlorobenzene at low temperatures followed by reaction with two equivalents of an aryl Grignard and subsequent quenching with a halogen.20 Both synthetic routes are shown in Scheme 1. Power has suggested that there appears to be a limit to the size of the ﬂanking aryl group that can be added to the central aryl ring.21 Thus, although it is possible to obtain the 2,4,6-triisopropylphenyl (2,4,6-iPr3C6H2, Trip) substituted m-terphenyl iodide, attempts to synthesise the analogous 2,4,6-tri-tert-butylphenyl (2,4,6-tBu3C6H2, Mes*) substituted m-terphenyl iodides have met without success. The most commonly used m-terphenyl ligands are of the type shown in Scheme 1, but the synthetic routes towards these ligands are conducive to substituent changes on the central aryl allowing the formation of a range of ligands with varying steric and electronic properties. Additionally, the range of ﬂanking aryl groups on m-terphenyls is increasing in order to both confer diﬀerent steric demands on the resulting complex and incorporate intramolecular donors. Such variations on this theme are leading to new and interesting discoveries (vide infra). Cl

Ar

Ar I

3 ArMgBr

Cl

MgBr

E

E Ar

Ar 2 ArMgBr Cl

Cl n

BuLi

Cl

Li Cl

Scheme 1 General synthetic routes towards m-terphenyl halides starting with 1,3-dichloro-2iodobenzene (top) or 1,3-dichlorobenzene (bottom)

Lithium salts of m-terphenyl ligands are easily generated in good yields;21,22 these reagents along with the corresponding Grignard complexes are used as starting materials for the generation of the majority of the compounds within this review. 3.

Complexes of group 3

A number of m-terphenyl complexes of the trivalent group 3 metals have been reported. Use of naphthyl moieties as ﬂanking groups in m-terphenyls Organomet. Chem., 2010, 36, 56–76 | 57

has been investigated in order to stabilise mono-arylated species. Reaction of one equivalent of 2,6-Nap2C6H3Li (Nap=1-naphthyl) with YCl3 in THF (THF=tetrahydrofuran) yields the trigonal bipyramidal complex 2,6Nap2C6H3YCl2(THF)2 (1) (Fig. 1); these systems are chiral with both enantiomers being present in the solid state.23 In addition to the ﬁve-fold coordination provided by the m-terphenyl ligand, two chlorides and two THF molecules, the distance between the yttrium and the naphthyl ipsocarbon and the ortho-CH [Y–C=3.283(6) and 3.496(6) A˚, respectively] indicates the possibility of weak interactions with the metal. Utilising this ligand, it was not possible to synthesise the analogous scandium complex, the product from the reaction of 2,6-Nap2C6H3Li with ScCl3 was a heterobimetallic complex with no m-terphenyl ligand present. Use of the 2,6-Mes2C6H3 (Mes=2,4,6-Me3C6H2) ligand facilitates the stabilisation of the corresponding scandium m-terphenyl complex as well as the yttrium compound 2,6-Mes2C6H3MCl2(THF)x (M=Sc, x=2 2; M=Y, x=3 3) according to Equation 1.24 2;6-Mes2 C6 H3 Li þ MCl3 þ THF —! 2;6-Mes2 C6 H3 MCl2 ðTHFÞx þ LiCl

ð1Þ

The development of the 2,6-Anis2C6H3 (Anis=2-MeOC6H4) ligand has conferred further stability on the resulting group 3 and lanthanide complexes facilitated by the incorporation of an intramolecular donor into the m-terphenyl ligand system. The heterobimetallic complex [2,6-Anis2 C6H3Y(m2-Cl)2(m2-Cl)Li(THF)2]2 (4, Fig. 1) has been synthesised via reaction of 2,6-Anis2C6H3Li with YCl3 in THF.25 4 dimerises via two chlorides and two Cl–Li(THF)2–Cl units bridging the seven-coordinate yttrium centres. Interestingly, 4 exhibits a diﬀerent molecular structure to the Yb analogue {which forms the complex [2,6-Anis2C6H3Yb(m2-Cl)2 (m3-Cl)Li(THF)]2}, despite the fact that the ionic radius of Y(III) is only slightly larger than Yb(III) for a given coordination number.26 The analogous reaction with ScCl3 did not proceed due to the polymerisation of the THF solvent. The one pot reaction of YCl3, 2,6-Anis2C6H3Li and the aryloxide DippOK (Dipp=2,6-iPr2C6H3) or amides KN(SiMe2R)2 gives rise to the complexes 2,6-Anis2C6H3Y(ODipp)2THF (5) or 2,6-Anis2C6H3Y[N(SiMe2R)2]2 (R=H, 6; R=Me, 7).27 Amide complexes 6 and 7

THF THF Nap THF Y THF Nap

Cl Cl Cl

O

Li Cl

Y O Cl

O

DippO O THF Y

O

DippO

Y

Cl

O

Cl Li

THF 1

Cl

THF 4

Fig. 1 m-Terphenyl complexes of yttrium stabilised by the ligands

58 | Organomet. Chem., 2010, 36, 56–76

5

2,6-Nap2C6H3

and 2,6-Anis2C6H3

were not crystalline; however, it was possible to obtain the crystal structure of aryloxide 5. The central Y atom in 5 displays distorted trigonal prismatic geometry with the 2,6-Anis2C6H3 ligand occupying three coordination sites, with two aryloxide moieties and a THF molecule taking up the other three sites (Fig. 1). The single-pot reaction of K2COT (COT=cyclooctatetraene), YCl3 and 2,6-Ph2C6H3Li has led to the formation of the complex 2,6-Ph2C6H3Y(Z8C8H8)ClLi(THF)3 (8) where the yttrium centre is bonded in an Z8-fashion to the COT, and the remaining coordination sites are occupied by the m-terphenyl ligand and a ClLi(THF)3 moiety.28 4.

Complexes of groups 4 and 5

The only complex featuring a metal-carbon s bond between a group 4 metal and an m-terphenyl was reported in 2008 by Robinson and co-workers; the highly sterically encumbered complex 2,6-(4-MeC6H4)2C6H3Ti(Z5-C5H5)2 (9) was synthesised by the reaction of 2,6-(4-MeC6H4)2C6H3Li with [(Z5C5H5)2Ti(m2-Cl)]2.29 Paramagnetic 9 may be used as a one electron reductant. Examples of group 5 complexes featuring trianionic ligands such as 2,6-(2-O-3,5-tBu2-C6H2)2C6H33 have been reported,30 but as this review concentrates on the use of m-terphenyls as monoanionic ligand frameworks these complexes will not be discussed here. 5.

Complexes of groups 6-9

Although the aryl chemistry of the metals from groups 6-9 is well known, the utilisation of m-terphenyl ligands in the stabilisation of these species has only been explored recently. New developments within this area of chemistry have led to reports of complexes featuring novel coordination modes for the transition metals. The ﬁrst report of an m-terphenyl complex featuring a metal from groups 6-9 was that by Power and Ellison, who described a rare example of a Co(II) aryl species; dimeric [2,6-Mes2C6H3Co(m2-Br)THF]2 (10) was synthesised via the reaction of the Grignard complex [2,6-Mes2C6H3Mg(m2-Br)THF]2 with CoCl2.31 10 is found to be isostructural with the Grignard complex [2,6Mes2C6H3Mg(m2-Br)THF]2, consistent with the close similarity of the Shannon-Prewitt radii of four-coordinate Mg2 þ and high spin four-coordinate Co2 þ ions (0.71 A˚ and 0.72 A˚, respectively).32 The Co2Br2 unit is essentially planar and despite the steric demands of the 2,6-Mes2C6H3 ligand the Co(II) centre in 10 was found to be four coordinate, with the other three sites being ﬁlled by two bridging bromides and a coordinating THF molecule. Indeed, the reaction of two equivalents of [2,6-Mes2C6H3Mg(m2Br)THF]2 with CoBr2 was attempted in order to synthesise the homoleptic, two-coordinate Co(II) aryl species, but only the blue bromide-bridged dimer 10 could be isolated in a pure state via this reaction.31 Recently, Power and co-workers have extended this chemistry to form heteroleptic complexes featuring more sterically demanding m-terphenyl ligands. Reaction of [2,6-Dipp2C6H3Li]2 with two equivalents of CrCl2(THF)2 in diethyl ether led to the formation of the solvent-free chloride-bridged dimer [2,6-Dipp2C6H3Cr(m2-Cl)]2 11 (Scheme 2), which Organomet. Chem., 2010, 36, 56–76 | 59

possesses a central Cr2Cl2 moiety.33 Despite the lack of donor solvent in the coordination sphere of 11, the complex features a quasi-four-coordinate Cr(II) centre due to an interaction with the ipso-carbon of one of the ﬂanking Dipp substituents on the m-terphenyl ligand, leading to a square planar coordination environment around the metal dication. Concomitant with this interaction is the distortion of the bond angles around the ipso-carbon on one Dipp substituent of each m-terphenyl ligand. Interesting results have come from the use of metal diiodide precursors in metathesis chemistry. Reaction of MI2 (M=Mn, Fe, Co) with [2,6-Dipp2C6H3Li]2 yields [2,6Dipp2C6H3MI2Li(OEt2)]2 (M=Mn, 12; M=Fe, 13; M=Co, 14). 12-14 are isostructural, exhibiting distorted cubane structures with M2Li2I4 cores where the transition metal atoms are bonded to an m-terphenyl ligand and the lithium centres are coordinated by ether moieties (Scheme 2). Although the incorporation of lithium halides into transition metal complexes has literature precedent,34 cubane structures incorporating transition metals, lithium cations and halides in this manner remain relatively rare. Magnetic susceptibility measurements on 11 are consistent with some weak antiferromagentic coupling between the Cr(II) centres via the bridging chloride ions. In the case of 12 and 14 antiferromagnetic coupling was observed between the bridging iodides, however this interaction was found to be very weak due to the near 901 I–M–I bond angles. Conversely, stronger antiferromagnetic exchange was observed in 12, which has an I–M–I bond angle closer to 901 [for 12 +I–Mn–I=87.04(5)1 and for 14 +I–Co–I=85.58(2)1]. [2,6-Dipp2C6H3Li]2 2 MI2

2 CrCl2(THF)2

Ar

Dipp

i

Pr

I i

Pr Cl Cr Cr Cl iPr

Et2O i

Pr

M

I Li M Ar I

I

Li

Dipp

M = Mn (12) M = Fe (13) M = Co (14)

OEt2

11 Scheme 2

Synthesis of complexes 11-14

Work within our research group has led to the discovery that the reaction of CoBr2(DME) (DME=1,2-dimethoxyethane) with the lithium salt [2,6Mes2C6H3Li]2 in a mixture of toluene and THF (Equation 2) gives rise to a deep claret red coloured solution and the formation of the pure homoleptic diaryl species (2,6-Mes2C6H3)2Co (15) in good yields.35 Structural studies have revealed that 15 is a rare example of a two-coordinate Co centre, and is the only structurally authenticated example of such a complex to feature Co–C s bonds, illustrating the stabilising inﬂuence of sterically demanding m-terphenyl ligands. The applicability of this metathetical route is demonstrated by the synthesis of rare examples of homoleptic, two-coordinate Mn and Fe diaryls (2,6-Mes2C6H3)2M (M=Mn, 16; M=Fe, 17), which could 60 | Organomet. Chem., 2010, 36, 56–76

also be formed in good yields by reaction of MnCl2 and FeCl2(THF)1.5 with [2,6-Mes2C6H3Li]2. Two-coordinate Mn and Fe diaryls have been isolated previously by Power and Seidel, who reported MMes*2 (M=Mn, Fe) in 1995.36,37 ½2;6-Mes2 C6 H3 Li2 þ CoBr2 ðDMEÞ! ð2;6-Mes2 C6 H3 Þ2 Co þ 2LiBr

ð2Þ

Like MnMes2* and FeMes2*, 15–17 are signiﬁcantly deviated from D2d symmetry, [+C–M–C in the range 162.8(1)-173.0(1)1] the bending of these angles has been attributed to crystal packing eﬀects or intramolecular interactions. However, given the long distances between the central metal atoms and the ipso-carbon atoms of the ﬂanking mesityl substituents in these complexes [falling in the range 2.679(2)–3.007(3) A˚], any such intramolecular interactions are likely to be weak. Recently, heteroleptic complexes featuring C–M–N moieties have been reported. Formed from the reaction of 2,6-Mes2C6H3N(H)Li with transition metal complexes of the 2,6-Dipp2C6H3 ligand, 2,6-Dipp2C6H3MN(H) C6H3Mes2-2,6 [M=Mn, 18; M=Fe, 19; M=Co, 20] are monomeric due to the steric demands of the ligands.38 In the crystal structures of 18 and 19 there is an interaction between the metal centres and the ipso-carbon of one of the ﬂanking mesityl rings of the anilido ligand. However, in the case of the Co(II) system (20) two diﬀerent structures were obtained depending on the temperature used for data collection. At 240 K the majority (95%) of the structure is similar to that for 18 and 19, i.e. an Z1-interaction of the central Co with the ipso-carbon of one of the ﬂanking mesityl rings of the anilido ligand [d(Co–Cipso)=2.393(2) A˚, 20a], whereas cooling to 90 K leads to a shift to a strong cobalt Z6-interaction with one of the ﬂanking mesityl rings [d(Co–Cipso)=2.077(3) A˚, d(Co–centroid)=1.636(3) A˚, 20b] (Fig. 2). Magnetic measurements on 20 are in agreement with the observed structural changes at 240 K; the complex undergoes an electronic spin-state transition from the low spin S=1/2 state (20b) to the high spin S=3/2 state above this temperature (20a). Mes

Dipp Co

H N

Mes

Dipp Co

H N

Dipp

Dipp 20a

20b

Fig. 2 The two diﬀerent coordination modes for 2,6-Dipp2C6H3CoN(H)C6H3Mes2-2,6; 20a (left) and 20b (right)

Power and co-workers have utilised the steric demands of bulky monodentate m-terphenyls to protect highly reactive transition metal centres, whilst minimising the number of orbitals involved in metal-ligand bonding in order to form complexes with very high M–M bond orders. The ﬁrst complex to feature a quintuple bonding interaction between two Organomet. Chem., 2010, 36, 56–76 | 61

2,6-iPr2

2,6-iPr2 Dipp

Cr

Cr

Dipp

Dipp 2,6-iPr2

21

M

M

Dipp 2,6-iPr2 M = Fe, 22 M = Co, 23

Fig. 3 Diﬀerences in the solid state structures for the complexes [2,6-Dipp2C6H3M]2 (M=Cr, 21; M=Fe, 22; M=Co, 23)

metals, [2,6-Dipp2C6H3Cr]2 (21, Fig. 3), was synthesised via the reduction of 11 with KC8 (Equation 3).39 In 21 the highly electropositive chromium centres are each s bonded to one m-terphenyl ligand [Cr–C distance=2.131(1) A˚], but also experience a secondary interaction with an ipso-carbon atom of one of the ﬂanking Dipp substituents [Cr?C distance=2.2943(9) A˚]. The metal-metal bonding in binuclear 21 involves the interaction of the ﬁve d electrons on each Cr(I) centre via s, 2p and 2d orbital overlaps and 21 features an extremely short Cr–Cr bond [1.8351(4) A˚].40 The eﬀective bond order (EBO) of 21 is, however, much lower than ﬁve due to the weak coupling between the metal 3d orbitals giving rise to an EBO of 3.52.41 ½2;6-Dipp2 C6 H3 Crðm2 -ClÞ2 þ x=sKC8 ! ½2;6-Dipp2 C6 H3 Cr2 þ 2KCl

ð3Þ

A theoretical investigation into the nature of the chromium-chromium bond in 21 has been performed. The structure possesses a planar, trans-bent rather than a linear Cipso–Cr–Cr–Cipso core; this is predicted for the dimer HCrCrH on the basis of DFT calculations and a natural bond analysis.42,43 Theoretical investigations performed on the model complex PhCrCrPh have found that the trans-bent conﬁguration is only 1 kcal mol 1 higher in energy than the linear structure; it seems that the preference for the trans-bent conﬁguration in 21 is presumably due to the weak, but not negligible secondary interaction which occurs between the chromium centres and the ﬂanking aryl rings of the m-terphenyl ligand (calculated as ca. 1-2 kcal mol 1 for the Cr–Cipso interaction).41 Indeed, inspection of the potential energy surface (PES) of PhCrCrPh indicates that the trans-bent conﬁguration is not a minimum but a transition point. It is therefore possible that the sterically demanding m-terphenyl ligands in 21 can stabilise a trans-bent geometry, turning a transition point into an energy minimum in an analogous manner to that observed in [2,6-Ph2C6H3Pb]2.44 Interestingly, the metal-metal distances in the analogous Fe and Co dimers [2,6-Dipp2C6H3M]2 (M=Fe, 22; M=Co, 23) (Fig. 3) are signiﬁcantly longer than in 21 [2.5151(9) A˚ for 22 and 2.8033(5) A˚ for 23].45 Although the Fe–Fe distance in 22 may be indicative of a single bond between the two iron centres (S covalent radii for Fe–Fe=2.48 A˚), the Co–Co distance in 23 is signiﬁcantly longer than the sum of the covalent radii (2.46 A˚) indicating that the cobalt(I) centres are either very weakly bonding or non-bonding.46 The structural diﬀerences between the Cr complex 21 and 62 | Organomet. Chem., 2010, 36, 56–76

its Fe and Co analogues 22 and 23 are also evidenced by the metal-carbon distances within. Unlike the weak additional Z1-coordination exhibited between the Cr centres and the ﬂanking Dipp rings in 21, the metal centres in 22 and 23 are Z6-bonded to one of the ﬂanking Dipp rings of the mterphenyl ligand bonded to the other metal centre. Additionally, the metalcentroid distances in 22 and 23 [1.7333(18) and 1.7625(19) A˚ for 22 and 1.7638(16) A˚ for 23] are signiﬁcantly shorter than that for 21 [2.203(6) A˚]. The structural diﬀerences between 21 and the iron and cobalt analogues suggest that the very short bond lengths in 21 are due to the strong interaction between the d5 chromium centres. This is in agreement with theoretical calculations which have shed light on diﬀerences in the bonding interactions within 21 compared to Fe and Co systems 22 and 23.47 The calculated Wiberg bond order (WBO) values for the M–M bonds are 4.11, 0.44 and 0.10 for 21, 22 and 23, respectively; in the case of the 23 this leads to essentially the absence of a Co–Co bond. Calculations on the model complexes suggest that the strength of the Cr–Cr bonds in complexes such as 21 are related to the d5 nature of the Cr(I) species, which precludes strong interactions with nearby arenes resulting from the exact match of valence electrons to bonding orbitals. In order to probe how electronic changes within m-terphenyl ligands inﬂuence the bonding in dimeric Cr–Cr complexes, [4-X-2,6-Dipp2C6H2Cr]2 (X=SiMe3, 24; X=OMe, 25; X=F, 26) which feature electron donating and withdrawing substituents in the para position of the central aryl ring have been synthesised.48 The Cr–Cr distances for 21 and 24-26 are remarkably similar [occurring in the narrow range 1.8077(7) A˚ (for 24) to 1.8351(4) A˚ (for 21)]; these small variations do not appear to follow the electron donating/withdrawing abilities of the para substituents indicating that the strong metal-metal bonding within these systems is insensitive to changes in the electronics of the m-terphenyl ligands. It appears that the small diﬀerences in the Cr–Cr distances in 21 and 24-26 are due to crystal packing eﬀects; this suggestion is supported by the results of DFT calculations on the model complexes [4-X-C6H4Cr]2. Interestingly, the reduction of [4-CF3-2,6-Dipp2C6H2Cr(m2-Cl)]2 with KC8 does not give rise to the Cr–Cr bonded complex, but instead yields the ﬂuoride bridged dimer [4-CF3-2,6-Dipp2C6H2Cr(m2-F)THF]2 (27, Equation 4) in low yields. As the by-products for this reaction could not be identiﬁed the mechanism for the formation of 27 has yet to be ascertained. ½4-CF3 -2;6-Dipp2 C6 H2 Crðm2 -ClÞ2 þ x=sKC8 ! ½4-CF3 -2;6-Dipp2 C6 H2 Crðm2 -FÞ2

ð4Þ

Further increase of the steric congestion around the metal centres in the chromium halide precursors leads not to the formation of metal-metal bonds upon reduction with KC8, but instead yields the monomeric Cr(I) complexes.48,49 The 3,5-iPr2-2,6-Trip2C6H ligand includes iPr substituents in the meta positions of the central aryl ring in addition to ﬂanking Trip groups to increase the crowding around the metal atom. Reduction of [3,5-iPr2-2,6-Trip2C6HCr(m2-Cl)]2 in the presence of THF and PMe3 leads to the formation of 3,5-iPr2-2,6-Trip2C6HCr(L) (L=THF, 28; L=PMe3, 29, Organomet. Chem., 2010, 36, 56–76 | 63

Fig. 4), the ﬁrst examples of two-coordinate, open shell transition metal complexes in a þ 1 oxidation state.48,49 These complexes are unstable in aromatic solvents; it appears that the arene molecules displace the coordinated THF or PMe3, the resulting complex then decomposes. Indeed, reduction of [3,5-iPr2-2,6-Trip2C6HCr(m-Cl)]2 followed by treatment with toluene leads to the formation of (3,5-iPr2-2,6-Trip2C6H2)Cr(Z3:Z6-CH2Ph) Cr(2,6-Trip2C6H2-3,5-iPr2) (30) (Fig. 4), which features a þ 2 oxidation state for one Cr (bound to the m-terphenyl and benzyl anions) and a zero oxidation state for the other Cr atom which exhibits a structure which is similar to that of (bis)benzenechromium sandwich complexes.48

Trip Cr

2,4,6- i Pr3 L

Trip

Cr Trip

Trip L = THF, 28 L = PMe3, 29 Fig. 4

Cr 2,4,6- i Pr3 30

Cr complexes featuring the sterically demanding 3,5-iPr2-2,6-Trip2C6H ligand

Similar reaction conditions for manganese, iron and cobalt complexes lead to more stable arene systems, presumably due in part to a higher valence electron count for these metal centres. Reduction of [3,5-iPr2-2,6Trip2C6HMX]n (M=Mn, X=I, n=1; M=Fe, X=Cl, n=1; M=Co, X=Cl, n=2) gives rise to the Mn(I) inverted sandwich complex [3,5-iPr22,6-Trip2C6HMn]2(m-Z6:Z6-C6H5Me) (31) and the M(I) half sandwich complexes 3,5-iPr2-2,6-Trip2C6HFe(Z6-C6H6) (32) and 3,5-iPr2-2,6Trip2C6HCo(Z6-C6H5Me) (33) after extraction into toluene or benzene (e.g. Eqution 5).50 In 31 the two Mn centres are bridged via an Z6-bound toluene molecule; the results of magnetic measurements indicate that there is no magnetic exchange between the two Mn centres via the bridging toluene. 3;5-i Pr2 -2;6-Trip2 C6 HMnI þ x=sKC8 þ C6 H5 Me ! 0:5½3;5-i Pr2 -2;6-Trip2 C6 HMn2 ðm-Z6 : Z6 -C6 H5 MeÞ þ KI

ð5Þ

Recent investigations into the reactivity of 32 with organoazides have led to the formation of amido and imido complexes. Reaction of 32 with 2,6-Mes2C6H3N3 leads not to the formation of the expected Fe(III) imido complex, but instead yields an Fe(II) amido species 34 (Scheme 3), resulting from dimerisation of 2,6-Mes2C6H3N(H)FeC6H-2,6-Trip2-3,5-iPr2 through the formation of a C–C bond between two ortho-methyl groups on the ﬂanking mesityl groups of the m-terphenyl ligands.51 The steric demands of the aryl and amido ligands lead to a rare example of a two-coordinate Fe(II) centre within 34 and signiﬁcant N–Fe–C bending is observed [+N–Fe– C=138.91(8)1]. The long Fe?C interaction observed for 34 [2.866(2) A˚] is 64 | Organomet. Chem., 2010, 36, 56–76

Trip Fe

Mes C6H3-2,6-Mes2N3

ArFe

Mes NH

HN

FeAr

Trip 32

Ar = 3,5-iPr2-2,6-Trip2C6H, 34

(1-Ad)N3 Trip N(1-Ad) Fe N(1-Ad) Trip 35 Scheme 3 Reaction of 3,5-iPr2-2,6-Trip2C6HFe(Z6-C6H6) (32) with azides 2,6-Mes2C6H3N3 and (1-Ad)N3

similar to that reported for 17.35 Reaction of 32 with the less sterically demanding (1-Ad)N3 (1-Ad=1-adamantyl) leads to the imido complex 3,5-iPr2-2,6-Trip2C6HFe[=N(1-Ad)]2 (35), notable as a rare example of a well-characterised Fe(V) system. A 17 valence electron compound, the steric demands of the m-terphenyl ligand allow the coordination of two (1-adamantyl)imido moieties to the metal centre without exceeding 18 valence electrons. 33 has been shown to react with CO and NO, with insertion of these molecules into the Co–C s bond.52 Reaction of 33 with CO leads to the formation of the acyl cobalt(I) carbonyl complex (3,5-iPr2-2,6-Trip2C6H)C(O)Co(CO) (36), which features an Z6-interaction between the Co and one of the ﬂanking Trip rings (Scheme 4). Such insertion chemistry into the M–C bond is in contrast with that exhibited by related Co(I) and Fe(I) b-diketiminate complexes.53 The analogous reaction of 33 with NO leads to the formation of (3,5-iPr2-2,6-Trip2C6H)N(NO)OCo(NO)2 (37) which features a Co(I) centre bound in an Z2-O,O mode by the bidentate anion [(3,5-iPr2-2,6-Trip2C6H)N(NO)O]– (Scheme 4). 37 is formed via a rare example of double NO insertion into a late transition metal-carbon

2,4,6-iPr3

NO

ON Trip Co CO

CO

O

NO

Co

Co O

N N

O

Trip

Trip

Trip

33

Trip 36

37 i

6

Scheme 4 Reaction of 3,5- Pr2-2,6-Trip2C6HCo(Z -C6H5Me) (33) with CO and NO

Organomet. Chem., 2010, 36, 56–76 | 65

s bond,54 the insertion of NO into the Co–C bond taking place with concomitant NO coupling and N–N bond formation. Monoanionic m-terphenyl complexes of groups 6-9 are currently restricted to the ﬁrst row metals; the trianionic ligand 2,6-(2-O-3-t Bu-C6H3)2C6H33 has been used to stabilise Mo(IV) species of the type 2,6(2-O-3-tBu-C6H3)2C6H3Mo[N(iPr2)H]2X (X=NiPr2, Cl).55 As this review concentrates on the use of m-terphenyls as monoanionic ligand frameworks these complexes will not be discussed here. 6.

Complexes of group 10

m-Terphenyl complexes of the group 10 metals featuring M–C s bonds are currently restricted to the pincer-type species developed by Protasiewicz as potential reaction catalysts. Utilisation of the m-terphenyl framework as a platform for substitution with nitrogen and phosphorus donors has allowed the formation of twisted metal complexes. 2,6-(2-R2PCH2C6H4)2C6H3MBr [M=Ni, R=Ph 38; M=Pd, R=Ph 39 (Fig. 5), R=Cy 40 (Cy=cyclohexyl), R=tBu 41] are synthesised via the insertion of Ni(0) and Pd(0) into the carbon-bromine bond of the ligand (e.g. Equation 6).56 Structurally authenticated 39 reveals a distorted square-planar Pd(II) centre with a twist angle57 (F=76.01) which is signiﬁcantly larger than the related complex 2,6-(Ph2PCH2)2C6H3PdBr (F=18.01)58 and that of N-heterocyclic carbene pincer complexes such as [PdCl(CNC)]BF4 [where CNC=C,N,C’-2,6-bis[(3-methylimidazolin-2-yliden-1-yl)methyl]pyridine] (which has Fr41.91).59 NMR experiments reveal that the pincer framework in 39 remains structurally rigid at temperatures up to 130 1C. The analogous diphosphinite complex 2,6-(2-Ph2POC6H4)2C6H3PdBr (42, Fig. 5) has a similar twist angle (F=73.81). 2;6-ð2-R2 PCH2 C6 H4 Þ2 C6 H3 Br þ Pd2 ðdbaÞ3

ð6Þ

! 2;6-ð2-R2 PCH2 C6 H4 Þ2 C6 H3 PdBr

O PPh2

PPh2 Pd

Br

Pd

Br

NMe2 Pd

PPh2

PPh2

Br

N Pd

NMe2

N

Mes Br Mes

O

39

42

43

51

Fig. 5 Pincer complexes of Pd based on the m-terphenyl framework

The amine pincer complex 2,6-(2-Me2NCH2C6H4)2C6H3PdBr (43, Fig. 5), which features a twist angle of 73.81 has been synthesised, along with a small amount of the trimetallic complex {[2,6-(2-Me2NCH2C6H4)2C6H3Pd]3 (m3-CO3)} þ I3 (44).60 Investigation of 42 and the related complexes 66 | Organomet. Chem., 2010, 36, 56–76

2,6-(2-R2POC6H4)2C6H3PdX (R=Ph, X=I 45; R=iPr, X=I 46; R=iPr, X=Br, 47) as catalysts for the Suzuki-Miyaura C–C cross coupling reaction have showed that although 42 was able to promote the C–C coupling between p-tolylboronic acid with aryl bromides, it was unsuccessful in the analogous coupling reaction with aryl chlorides.61 However, 47 was eﬀective in the promotion of the C–C coupling reaction between p-tolylboronic acid with both aryl bromides and chlorides. Mechanistic investigation of this catalysis using elemental mercury suggest that these complexes may serve as a source of highly active colloidal Pd(0) as the catalytic species for these reactions. Diimine pincer complexes 2,6-(2-RNQCHC6H4)2C6H3PdI [R=Ph, 48; R=Cy, 49; R=m-Xyl 50 (m-Xyl=meta-xylyl); R=Mes, 51 (Fig. 5)] feature twist angles in the range 61.5–65.11, smaller than those for 39-41 (F=73.8–76.01), presumably due to the diﬀering steric and rigidity demands of these donor groups.62 NMR experiments reveal hindered rotation around the N–R bond in 50 and 51 due to the proximity of the methyl groups on the xylyl and mesityl substituents to the iodide (rotation barrier=15.2 and 15.7 kcal/mol for 50 and 51, respectively). Introduction of chiral groups onto the imine moiety allows the isolation of both diastereomers of the complex 2,6-[2-Ph(H)CH3CNQCHC6H4]2C6H3PdI (52) evidence that these pincer systems are conﬁgurationally stable; NMR experiments suggest that these conﬁgurations are also maintained in solution. Interestingly, the twist angles for the two diastereomers diﬀer by 21 despite having the same substituents. 7.

Complexes of group 11

Early investigations into the terphenyl chemistry of group 11 involved the use of the 2,4,6-Ph3C6H2 ligand;63 the complex 2,4,6-Ph3C6H2Cu(m2-2,4,6Ph3C6H2)Cu(SMe2)2 (53, Fig. 6) exists as a dimer where the two copper centres are symmetrically bridged by a 2,4,6-Ph3C6H2 ligand.64 A short contact exists between the metal atoms [Cu?Cu=2.443(1) A˚], with one of the copper centres bound in an Z1 fashion to a 2,4,6-Ph3C6H2 ligand and the other solvated by two SMe2 molecules. NMR studies have revealed that 53 is monomeric in SMe2 solution, presumably as SMe2-solvated 2,4,6Ph3C6H2Cu molecules. This is exhibited in the solid state structure of Mes*CuSMe2, which features the more sterically demanding supermesityl ligand.64 The use of sterically demanding phosphine and N-heterocyclic carbene donors gives rise to the monomeric species 2,6-Ar2C6H2CuPPh3 Ph Ph Ph

Trip

Ph Cu

Trip

Li Ph Cu

Ph Me2S 53

THF THF

SMe2

Cu C N

N C Cu Li

Trip

THF THF 57

Trip

Fig. 6 m-Terphenyl complexes 2,4,6-Ph3C6H2Cu(m2-2,4,6-Ph3C6H2)Cu(SMe2)2 (53) and [2,6Trip2C6H3Cu(CN)Li(THF)2]2 (57)

Organomet. Chem., 2010, 36, 56–76 | 67

(Ar=Ph, 54; Ar=Mes, 55) and 2,6-Mes2C6H2CuC[N(iPr)CMe]2 (56).65 Presumably due to the lower steric demands of ﬂanking phenyls compared to mesityl substituents, 54 displays an intermolecular interaction between the copper centre and part of the arene ring of the PPh3 ligand on a neighbouring molecule. The use of 2,6-Trip2C6H3 allowed the isolation of [2,6-Trip2C6H3Cu(CN)Li(THF)2]2 (57, Fig. 6), helping to facilitate the deduction of the interaction between the Li þ and the [RCu(CN)] anion in lithium cyanoorganocuprate salts.66 57 features a near-linear two-coordinate geometry for the Cu atoms and exists as a dimer through two bridging Li(THF)2 moieties as had been previously shown for Li[PhCu(CN)] via cryoscopy.67 The structure of 57 was also found to be broadly in agreement with the results of EXAFS spectroscopic measurements on Li[MeCu(CN)] in THF solution.68 Intriguingly, 2,6-Mes2C6H3Cu(CN)Li(THF)3 (58) is a monomer,69 while 57, which features the more sterically demanding 2,6-Trip2C6H3 ligand exists as a lithium-bridged dimer. Although the reasons for this diﬀerence remain unclear, it has been proposed that the type and concentration of solvent has a major eﬀect on the degree of aggregation of some organocopper complexes.70 The reaction of one equivalent of 2,6-Ar2C6H3Li (Ar=Ph, Mes) with CuOtBu yielded the copper aryl complexes [2,6-Ar2C6H3Cu]n [Ar=Ph, n=3 59; Ar=Mes, n=2 60] (Equation 7). The existence of 60 as a dimer compared to trimeric 59 is a reﬂection of the greater steric demands of the ﬂanking mesityl groups compared to phenyl.71 Trimer 59 features two distinct copper environments (Fig. 7). There are two m-terphenyl ligands forming bridging interactions between two of the Cu(I) centres; the third m-terphenyl ligand forms a s-bond to one Cu and an Z2-p arene interaction via one of its ﬂanking phenyl substituents. Due to the alternating short and long C–Cipso distances, together with larger and smaller C–Cu–C angles (Fig. 7), leads to a bonding situation within 59 best described as an alternating 2c-2e Cu–Cipso and p-type Cu–Cipso interaction. With values of 2.3758(7), 2.4224(10) and 2.9136(11) A˚, the Cu?Cu distances in 59 are similar to that displayed by 53. The Cu(I) centres in 60 are s-bonded to the ipso-carbon of the 2,6-Mes2C6H3 ligand, dimerising through an Z2-interaction to an ipso- and ortho-carbon of a ﬂanking mesityl substituent. The Cu?Cu distance of 2.5953(12) A˚ in 60 is similar to those in 53 and 59. 1 2;6-Ar2 C6 H3 Li þ CuOt Bu ! ½2;6-Ar2 C6 H3 Cun þ LiOt Bu n

Cu

Mes Li

Cu

59

Mes Cu Cu

Cu Mes

Cu 2,6-Ph2

Mes

Mes

Ph

2,6-Ph2

62

ð7Þ

Mes 63

Fig. 7 Solid state structures of 59, 62 and 63 (methyl groups on the mesityl substituents in 62 and 63 omitted for clarity)

68 | Organomet. Chem., 2010, 36, 56–76

A minor product of the reaction between 2,6-Mes2C6H3Li and CuOtBu is the tetranuclear complex (m2-2,6-Mes2C6H3)2Cu4(m2-OtBu)(m2-I) (61) which features a near-planar C2Cu4OI ring.72 As is seen in 60, the alternating long and short Cu–Cipso distances with accompanying large and small C–Cu–C angles can also be described as an alternating 2c-2e Cipso–Cu and p-type C–Cipso interaction; alternatively 61 can also be described as a [Cu2I] þ [(2,6Mes2C6H3)2Cu2(OtBu)]– contact ion pair. In contrast to the asymmetry seen in the copper-carbon bonding, symmetric bridging Cu–I [2.4457(6) and 2.4499(6) A˚] and Cu–OtBu [1.856(3) and 1.861(3) A˚] distances are seen. Reaction of CuOtBu with two equivalents of 2,6-Mes2C6H3Li yields the lithium cuprate (2,6-Mes2C6H3)2CuLi (62, Fig. 7), the ﬁrst example of a neutral monomeric cuprate with simple R2CuLi stoichiometry. The central copper atom is Z1-bonded to two terphenyl ligands, the Li þ is bonded in an Z6 fashion to the ﬂanking mesityl substituent on one terphenyl and in an Z1 fashion to the Cipso on the other terphenyl ligand. Lithium cuprate 62 cocrystallised with the copper cuprate (2,6-Mes2C6H3)2Cu2 (63, Fig. 7), which features one copper Z1-bonded to two terphenyl ligands, while the other copper is bonded in an Z2 fashion to the ﬂanking mesityl substituent on one terphenyl and in an Z1 fashion to the Cipso on the other terphenyl ligand. Reaction of the lithium salts of 2,6-Ar2C6H3 (Ar=Mes, Trip) with CuI produces organoiodocuprate complexes [Li(THF)4][(m2-2,6-Mes2C6H3)(CuI)2] (64) and [2,6-Trip2C6H3CuI]Li(OEt2)2 (65).73 Solvent separated ion pair 64 features an m-terphenyl ligand unsymmetrically bridging two CuI moieties. The Cu?Cu separation of 2.391(3) A˚ is relatively short, however, DFT calculations on the close d10–d10 interactions in complexes involving Cu(I) centres indicate that short distances such as those seen in 65 are not necessarily indicative of strong Cu–Cu bonding.74 65 features a rather small Cu–I–Li angle [95.5(2)1]; it appears that such metal-halide-alkali metal interactions do not have a strong angular dependence and as such are easily inﬂuenced by weak interactions such as crystal packing eﬀects. Lithium organoargentates [Li(THF)4][(2,4,6-Ph3C6H2)2Ag] (66) and [Li(THF)4][(2,6-Mes2C6H3)2Ag] (67) exist as solvent separated ion pairs with two-coordinate Ag centres and near linear Cipso–Ag–Cipso angles [176.7(4)1 and 176.8(8)1 for 66 and 67, respectively].75 The only example of a structurally authenticated gold m-terphenyl complex, two-coordinate 2,6-Mes2 C6H3AuPPh3 (68, Equation 8) features a M–C distance [d(Au–C)= 2.045(6) A˚] which is somewhat shorter than those for 66 and 67 [ave. d(Ag– C)=2.097(9) and 2.09(3) A˚].76 The likely explanation for these diﬀerences is the increased steric crowding in bis(m-terphenyl) complexes 66 and 67 compared to 68; this result may also be in agreement with a previous study that for a given coordination number the ionic radius of Au(I) is smaller than that for Ag(I) due to relativistic eﬀects.77 The structural data for the as yet unreported Ag(I) analogue of 68 would shed more light on the reasons for this diﬀerence. 2; 6-Mes2 C6 H3 Li þ Ph3 PAuCl ! 2; 6-Mes2 C6 H3 AuPPh3 þ LiCl

ð8Þ

The structurally authenticated, monomeric alkyne-stabilised complexes (Z5-C5H4SiMe3)2Ti(CCSiMe3)M(C6H2Ph3-2,4,6) were synthesised via the Organomet. Chem., 2010, 36, 56–76 | 69

reaction of (Z5-C5H4SiMe3)2Ti(CCSiMe3)MX with 2,4,6-Ph3C6H2MgBr (M=Cu, X=SC6H4CH2NMe2-2, 69; M=Ag, X=OC(O)Me, 70).78 The group 11 metal centres in these complexes are stabilised by Z2-coordination to the two alkynes and Z1-coordination to the 2,4,6-Ph3C6H2 ligand, and also feature remarkably short Ti?Cu [2.994(3) A˚] and Ti?Ag [3.220(1) A˚] distances. An interesting structure in the m-terphenyl chemistry of the group 11 elements is that of [2,6-Mes2C6H3Cu(OSn)(OtBu)Li(OEt2)]2 (71), where the oxide bridge of a heterocubane structure acts as a donor atom to a 2,6Mes2C6H3Cu moiety.79 8.

Complexes of group 12

The ﬁrst example of a group 12 m-terphenyl complex, (2,6-Mes2C6H3)2Hg (72) was synthesised as a potential transmetallation reagent to form group 1 aryl complexes.80 Unfortunately, 72 was shown not to undergo any signiﬁcant reaction with sodium or potassium in benzene at 75 1C. Interestingly, the C–Hg–C moiety in 72 is near linear [+C–Hg–C=178.6(3)1], which diﬀers signiﬁcantly from that exhibited by the group 7-9 complexes 15-17 [which display +C–M–C in the range 162.8(1)–173.0(1)1].35 Following from the reports of the ﬁrst Zn–Zn bond in [(Z5-C5Me5)Zn]2,81 the use of the 2,6-Dipp2C6H3 ligand has also facilitated the isolation of group 12 complexes containing metal-metal interactions. Starting materials for the generation of these complexes are the iodides [2,6-Dipp2C6H3MI]n (M=Zn, n=2 73; M=Cd, n=2 74; M=Hg, n=1 75); these halide complexes were reduced yielding the dimeric M(I)–M(I) complexes [2,6-Dipp2C6H3M]2 (M=Zn 76; M=Cd 77; M=Hg 78) (Scheme 5).82–84 The reducing agents which were utilised were Na, NaH and KC8 which gave rise to 76, 77 and 78, respectively. In the case of 77, the reaction proceeds through the weakly associated hydride dimer [2,6-Dipp2C6H3Cd(m2-H)]2 (79, Fig. 8) which Dipp

Dipp

Dipp

(i) M I

M

Dipp n M = Zn, n = 2 73 M = Cd, n = 2 74 M = Hg, n = 1 75

Dipp

M Dipp

M = Zn 76 M = Cd 77 M = Hg 78

Scheme 5 Synthesis of homologous series of group 12 M–M bonded complexes; (i)=Na, NaH and KC8 to yield 76, 77 and 78, respectively

Dipp

Dipp

Dipp

Cd

Cd

Dipp

Zn

Dipp

Zn

79 Fig. 8

Dipp

Zn

Zn H

Dipp

Dipp

80 m-Terphenyl hydride complexes of cadmium and zinc

70 | Organomet. Chem., 2010, 36, 56–76

Dipp Na

H

H Dipp

Dipp

H

H

Dipp 81

forms 77 with concomitant loss of H2. Indeed, the use of Na or Na/naphthalene reducing agents with 74 led to over-reduction giving rise to 2,6Dipp2C6H4 and Cd metal.83,84 The ﬁrst reported homologous series of group 12 M(I)–M(I) complexes, 76–78 have provided a unique opportunity to study the structure and bonding in such systems. These molecules feature near-linear Cipso–M–M–Cipso moieties and there is signiﬁcant interaction between the metal centres in these complexes; the M–M distances in 76-78 [2.3591(9) A˚, 2.6257(5) A˚ and 2.5738(3) A˚, respectively] are ca. 0.14 A˚, 0.20 A˚ and 0.30 A˚ shorter than the sum of Pauling’s single bond metallic radii for these metals.82–85 The Zn–Zn distance in 76 is within the range of that found in related complexes [2.295(3)– 2.399(11) A˚].81,86 The shortness of the Hg–Hg distance in 78 compared to the silylated analogue [(Me3SiMe2Si)3SiHg]2 [which has d(Hg–Hg)=2.6569(8) A˚] is suggestive of a large inﬂuence on the Hg–Hg bonding interaction in these systems by the ligand environment.87 The Cd–Cd distance in 77 is signiﬁcantly longer than the Hg–Hg distance in 78, which is consistent with a stronger M–M bonding interaction in 78 than 77 and a larger HOMO-LUMO gap. Relativistic calculations on model systems indicate that the Hg–Hg distance is contracted ca. 0.1 A˚ compared to the Cd–Cd distance in such compounds due to relativistic eﬀects contributing to a higher bond energy for Hg–Hg.84 The dimeric hydride complex, [2,6-Dipp2C6H3Zn(m2-H)]2 (80, Fig. 8), formed by the reaction of 73 with NaH (Equation 9) displays a very different coordination to the weakly associated cadmium hydride 79.82,84 The bridging hydrides in 80 are very symmetrical [d(Zn–H)=1.67(2) A˚, 1.79(3) A˚] and the Zn?Zn distance is short [2.4084(3) A˚], whereas 79 has an unsymmetrical bridging structure [d(Cd–H)=1.78(6) A˚, 2.27(6) A˚] and the Cd?Cd distance is long [2.9196(5) A˚] compared to 77. Further reduction of 80 with NaH leads to the formation of the unusual complex 2,6Dipp2C6H3Zn(m2-H)(m2-Na)ZnC6H3Dipp2-2,6 (Equation 10) which features a NaH bridge (81, Fig. 8).82,84 The Zn(m-H)Zn unit in 81 can be viewed as a s-antiaromatic ring. The arylmercury hydride 2,6-Dipp2C6H3HgH (82) can also be synthesised by the reaction of 75 with KH, its monomeric formulation in solution conﬁrmed by 1H and 199Hg NMR spectroscopy.84 ½2; 6-Dipp2 C6 H3 Znðm2 -IÞ2 þ 2NaH —! ½2; 6-Dipp2 C6 H3 Znðm2 -HÞ2

ð9Þ

½2; 6-Dipp2 C6 H3 Znðm2 -HÞ2 þ NaH —! 2; 6-Dipp2 C6 H3 Znðm2 -HÞðm2 -NaÞZnC6 H3 Dipp2 -2; 6

ð10Þ

Sterically demanding m-terphenyl ligands have also been used to stabilise novel Zn–Zr bonds.88 The precursor for the synthesis of a trimetallic Zn/Zr complex is the dimeric iodide [2,6-Trip2C6H3Zn(m2-I)]2 (83), formed by the vigorous stirring of 2,6-Trip2C6H3Zn(m2-I)2Li(OEt2)2 (84) in toluene. The sodium reduction of (Z5-C5H5)2ZrCl2 with 83 yields (Z5-C5H5)2Zr (ZnC6H3Trip2-2,6)2 (85), which features a rare example of a symmetric Zn–M–Zn (M=transition metal) unit.89 85 has a pseudo-tetrahedral Zr centre bonded to two two-coordinate Zn atoms [+Zr–Zn–C=172.84(13)1]. The Zn–Zr bond distance [2.7721(7) A˚] for 85 is very close to the sum of the covalent radii for these elements (2.70 A˚).90 Organomet. Chem., 2010, 36, 56–76 | 71

9.

Conclusions

m-Terphenyl ligands have been shown to be very eﬀective in the stabilisation of a range of unusual and highly reactive bonding modes for the dblock elements due to the variety and versatility of these bulky ligand frameworks. The ease of synthesis of m-terphenyls has allowed the manipulation of steric bulk, electronic properties, donor functionality and structural rigidity of the ligands and resultant complexes. The expansion of the repertoire of these ligands available to the organometallic chemist will certainly lead to further interesting developments in this area. In particular it is envisaged that low-coordination modes and metal-metal bonds for the transition metals will feature heavily in these new developments, especially for the metals of the ﬁfth and sixth periods, as m-terphenyl complexes of these elements have not been researched extensively.

References 1 See, for example: (a) D. C. Bradley, R. C. Mehrotra, I. P. Rothwell and A. Singh, Alkoxo and Aryloxo Derivatives of Metals, Academic, San Diego, 2001; and (b) G. Parkin, Prog. Inorg. Chem., 1998, 47, 1. 2 See, for example: (a) D. H. Harris and M. F. Lappert, J. Organomet. Chem. Libr., 1976, 2, 13; (b) D. C. Bradley and M. H. Chisholm, Acc. Chem. Res., 1976, 9, 273; (c) P. G. Eller, D. C. Bradley, M. B. Hursthouse and D. W. Meek, Coord. Chem. Rev., 1977, 24, 1; (d) M. F. Lappert, P. P. Power, K. A. R. Sanger and R. C. C. Srivastava, Metal and Metalloid Amides: Syntheses, Structures and Physical and Chemical Properties, Ellis Horwood, Chichester, 1979; and (e) R. Waterman, Dalton Trans., 2009, 1. 3 See, for example: (a) P. J. Davidson, M. F. Lappert and R. Pearce, Acc. Chem. Res., 1974, 7, 209; (b) P. J. Davidson, M. F. Lappert and R. Pearce, Chem. Rev., 1976, 76, 219; and (c) A. Yamamoto, J. Organomet. Chem., 1986, 300, 347. 4 See, for example: (a) R. S. Nyholm, Q. Rev. Chem. Soc., 1970, 24, 1; (b) S. U. Koschmieder and G. Wilkinson, Polyhedron, 1991, 10, 135; and (c) M. A. Garcı´ a-Monforte, P. J. Alonso, J. Fornie´s and B. Menio´n, Dalton Trans., 2007, 3347. 5 See, for example: (a) C. S. Cundy, B. M. Kingston and M. F. Lappert, Adv. Organomet. Chem., 1973, 11, 253; (b) C. Eaborn, J. D. Smith and K. Izod, J. Organomet. Chem., 1995, 500, 89; (c) C. Eaborn and J. D. Smith, Coord. Chem. Rev., 1996, 154, 125; and (d) X. Ziling, Comments Inorg. Chem., 1996, 18, 223. 6 P. P. Power, Comments Inorg. Chem., 1989, 8, 177. 7 P. P. Power, Chemtracts: Inorg. Chem., 1994, 6, 181. 8 P. P. Power, J. Organomet. Chem., 2004, 689, 3904. 9 R. S. Simons and P. P. Power, J. Am. Chem. Soc., 1996, 118, 11966. 10 S. T. Haubrich and P. P. Power, J. Am. Chem. Soc., 1998, 120, 2202. 11 B. Twamley, C. D. Soﬁeld, M. M. Olmstead and P. P. Power, J. Am. Chem. Soc., 1999, 121, 3357. 12 (a) L. Phu, B. Twamley and P. P. Power, J. Am. Chem. Soc., 2000, 122, 3524; (b) A. D. Phillips, R. J. Wright, M. M. Olmstead and P. P. Power, J. Am. Chem. Soc., 2002, 124, 5390; and (c) M. Stender, A. D. Phillips, R. J. Wright and P. P. Power, Angew. Chem. Int. Ed., 2002, 41, 1785. 72 | Organomet. Chem., 2010, 36, 56–76

13 (a) B. E. Eichler and P. P. Power, J. Am. Chem. Soc., 2000, 122, 8785; and (b) A. F. Richards, A. D. Phillips, M. M. Olmstead and P. P. Power, J. Am. Chem. Soc., 2003, 125, 3204. 14 N. J. Hardman, B. Twamley and P. P. Power, Angew. Chem. Int. Ed., 2000, 39, 2771. 15 (a) N. J. Hardman, R. J. Wright, A. D. Phillips and P. P. Power, Angew. Chem. Int. Ed., 2002, 41, 2842; (b) R. J. Wright, A. D. Phillips, N. J. Hardman and P. P. Power, J. Am. Chem. Soc., 2002, 124, 8538; and (c) R. J. Wright, A. D. Phillips, S. Hino and P. P. Power, J. Am. Chem. Soc., 2005, 127, 4794. 16 G. H. Spikes, J. C. Fettinger and P. P. Power, J. Am. Chem. Soc., 2005, 127, 12232. 17 E. Rivard, Y. Steiner, J. C. Fettinger, J. R. Giuliani, M. P. Augustine and P. P. Power, Chem. Commun., 2007, 4919. 18 (a) P. P. Power, J. Chem. Soc., Dalton Trans., 1998, 2939; (b) B. Twamley, S. T. Haubrich and P. P. Power, Adv. Organomet. Chem., 1999, 44, 1; (c) J. A. C. Clyburne and N. McMullen, Coord. Chem. Rev., 2000, 210, 73; (d) P. P. Power, Struct. Bonding (Berlin), 2002, 103, 57; (e) P. P. Power, Chem. Rev., 2003, 103, 789; (f) P. P. Power, Appl. Organomet. Chem., 2005, 19, 488; (g) P. P. Power, Organometallics, 2007, 26, 4362; (h) E. Rivard and P. P. Power, Inorg. Chem., 2007, 46, 10047; and (i) E. Rivard and P. P. Power, Dalton Trans., 2008, 33, 4336. 19 (a) K. Hanoda, H. Hart and C.-J. F. Du, J. Org. Chem., 1985, 50, 5521; and (b) C.-J. F. Du, H. Hart and K.-K. D. Ng, J. Org. Chem., 1986, 51, 3162. 20 A. Saednya and H. Hart, Synthesis, 1996, 1455. 21 B. Schiemenz and P. P. Power, Organometallics, 1996, 15, 958. 22 See, for example: (a) K. Ruhlandt-Senge, J. J. Ellison, R. J. Wehmschulte, F. Pauer and P. P. Power, J. Am. Chem. Soc., 1993, 115, 11353; and (b) B. Schiemenz and P. P. Power, Angew. Chem., Int. Ed. Engl., 1996, 35, 2150. 23 G. W. Rabe, C. D. Be´rube and G. P. A. Yap, Inorg. Chem., 2001, 40, 2682. 24 G. W. Rabe, C. D. Be´rube, G. P. A. Yap, K.-C. Lam, T. E. Concolino and A. L. Rheingold, Inorg. Chem., 2002, 41, 1446. 25 G. W. Rabe, C. D. Be´rube and G. P. A. Yap, Inorg. Chem., 2001, 40, 4780. 26 R. D. Shannon, Acta Cryst., 1976, A32, 752. 27 G. W. Rabe, M. Zhang-Preße, F. A. Riederer and G. P. A. Yap, Inorg. Chem., 2003, 42, 3527. 28 G. W. Rabe, M. Zhang-Preße, F. A. Riederer, C. D. Incarvito, J. A. Golen and A. L. Rheingold, Acta Cryst., 2004, E60, m1442. 29 B. Quillian, Y. Wang, P. Wei and G. H. Robinson, J. Coord. Chem., 2008, 61, 137. 30 T. Agapie and J. E. Bercaw, Organometallics, 2007, 26, 2957. 31 J. J. Ellison and P. P. Power, J. Organomet. Chem., 1996, 526, 263. 32 (a) R. D. Shannon and C. T. Prewitt, Acta. Cryst., 1969, B25, 925; and (b) R. D. Shannon and C. T. Prewitt, Acta. Cryst., 1970, B26, 1046. 33 A. D. Sutton, T. Nguyen, J. C. Fettinger, M. M. Olmstead, G. J. Long and P. P. Power, Inorg. Chem., 2007, 46, 4809. 34 See, for example: (a) C. Eaborn, P. B. Hitchcock, J. D. Smith, S. Zhang, W. Clegg, K. Izod and P. O’Shaughnessy, Organometallics, 2000, 19, 1190; (b) S. S. Al-Juaid, C. Eaborn, P. B. Hitchcock, M. S. Hill and J. D. Smith, Organometallics, 2000, 19, 3224; (c) J. M. Smith, R. Lachiotte and P. L. Holland, Chem. Commun., 2001, 1542; and (d) J. Prust, K. Most, I. Mu¨ller, A. Stasch, H. W. Roesky and I. Uson, Eur. J Inorg. Chem., 2001, 1613. 35 D. L. Kays and A. R. Cowley, Chem. Commun., 2007, 1053. 36 H. Mu¨ller, W. Seidel and H. Go¨rls, Angew. Chem., Int. Ed. Engl., 1995, 34, 325. Organomet. Chem., 2010, 36, 56–76 | 73

37 R. J. Wehmschulte and P. P. Power, Organometallics, 1995, 14, 3264. 38 C. Ni, J. C. Fettinger, G. J. Long and P. P. Power, Inorg. Chem., 2009, 48, 2443. 39 T. Nguyen, A. D. Sutton, M. Brynda, J. C. Fettinger, G. J. Long and P. P. Power, Science, 2005, 310, 844. 40 Other very short Cr–Cr bonds stabilised by bridging N-ligands have subsequently been reported; (a) K. A. Kreisel, G. P. A. Yap, O. Dmitrenko, C. R. Landis and K. H. Theopold, J. Am. Chem. Soc., 2007, 129, 14162; (b) A. Noor, F. R. Wagner and R. Kempe, Angew. Chem. Int. Ed., 2008, 47, 7246; and (c) Y.-C. Tsai, C.-W. Hsu, J.-S. K. Yu, G.-H. Lee, Y. Wang and T.-S. Kuo, Angew. Chem. Int. Ed., 2008, 47, 7250. 41 M. Brynda, L. Gagliardi, P.-O. Widmark, P. P. Power and B. O. Roos, Angew. Chem. Int. Ed., 2006, 45, 3804. 42 (a) F. Weinhold and C. R. Landis, Chem. Ed. Res. Pract. Eur., 2001, 2, 91; and (b) F. Weinhold and C. R. Landis, J. Am. Chem. Soc., 2006, 128, 7335. 43 F. Weinhold and C. R. Landis, in Valency and Bonding: A Natural Bond Order Donor-Acceptor Perspective, Cambridge University Press, Cambridge, 2005. 44 Y. Chen, M. Hartmann, M. Diedenhofen and G. Frenking, Angew. Chem. Int. Ed., 2001, 40, 2051. 45 T. Nguyen, W. A. Merrill, C. Ni, H. Lei, J. C. Fettinger, B. D. Ellis, G. J. Long, M. Brynda and P. P. Power, Angew. Chem. Int. Ed., 2008, 47, 9115. 46 L. Pauling, Proc. Natl. Acad. Sci. USA, 1976, 73, 4290. 47 G. La Macchia, L. Gagliardi, P. P. Power and M. Brynda, J. Am. Chem. Soc., 2008, 130, 5104. 48 R. Wolf, C. Ni, T. Nguyen, M. Brynda, G. J. Long, A. D. Sutton, R. C. Fischer, J. C. Fettinger, M. Hellman, L. Pu and P. P. Power, Inorg. Chem., 2007, 46, 11277. 49 R. Wolf, M. Brynda, C. Ni, G. J. Long and P. P. Power, J. Am. Chem. Soc., 2007, 129, 6076. 50 C. Ni, B. D. Ellis, J. C. Fettinger, G. J. Long and P. P. Power, Chem. Commun., 2008, 1014. 51 C. Ni, J. C. Fettinger, G. J. Long, M. Brynda and P. P. Power, Chem. Commun., 2008, 6045. 52 H. Lei, B. D. Ellis, C. Ni, F. Grandjean, G. J. Long and P. P. Power, Inorg. Chem., 2008, 47, 10205. 53 (a) X. Dai, P. Kapoor and T. H. Warren, J. Am. Chem. Soc., 2004, 126, 4798; and (b) J. M. Smith, A. R. Sadique, T. R. Cundari, K. R. Rogers, G. LukatRogers, R. J. Lachiotte, C. J. Flaschenriem, J. Vela and P. L. Holland, J. Am. Chem. Soc., 2006, 128, 756. 54 S. C. Puiu and T. H. Warren, Organometallics, 2003, 22, 3974. 55 S. Sarkar, A. R. Carlson, M. K. Veige, J. M. Falkowski, K. A. Abboud and A. S. Veige, J. Am. Chem. Soc., 2008, 130, 1116. 56 L. Ma, R. A. Woloszynek, W. Chen, T. Ren and J. D. Protasiewicz, Organometallics, 2006, 25, 3301. 57 The ‘‘twist’’ angle (F) is deﬁned as the angle between the plane of the anchoring ring and the square plane containing the metal and its four directly attached atoms. 58 F. Bachechi, Struc. Chem., 2003, 14, 263. 59 D. J. Nielsen, K. J. Cavell, B. W. Skelton and A. H. White, Inorg. Chim. Acta, 2002, 327, 116. 60 L. Ma, S. D. Wobser and J. D. Protasiewicz, J. Organomet. Chem., 2007, 692, 5331. 61 M. C. Lipke, R. A. Woloszynek, L. Ma and J. D. Protasiewicz, Organometallics, 2009, 28, 188. 74 | Organomet. Chem., 2010, 36, 56–76

62 L. Ma, P. M. Imbesi, J. B. Updegraﬀ III, A. D. Hunter and J. D. Protasiewicz, Inorg. Chem., 2007, 46, 5220. 63 Lingnau and Stra¨hle performed the reaction of 2,4,6-Ph3C6H2MgBr with MCl to form the air stable, isostructural complexes 2,4,6-Ph3C6H2M (M=Cu, Ag). Subsequent investigation of these complexes theoretically and crystallographically have concluded that the crystals used by Lingnau and Stra¨hle to determine the structures of these species consisted partly or entirely of the starting material, 2,4,6-Ph3C6H2Br. R. Lignau and J. Stra¨hle, Angew. Chem. Int. Ed. Engl., 1988, 27, 436; A. Haaland, K. Rydpal, H. P. Verne, W. Scherer and W. R. Thiel, Angew. Chem. Int. Ed. Engl., 1994, 33, 2443; C. W. Bauschlicher, Jr., S. R. Langhoﬀ, H. Partridge and L. A. Barnes, J. Chem. Phys., 1989, 91, 2399. 64 X. He, M. M. Olmstead and P. P. Power, J. Am. Chem. Soc., 1992, 114, 9668. 65 M. Niemeyer, Z. Anorg. Allg. Chem., 2003, 629, 1535. 66 C.-S. Hwang and P. P. Power, J. Am. Chem. Soc., 1998, 120, 6409. 67 A. Gerold, J. T. B. H. Jastrzebski, C. M. P. Kronenburg, N. Krause and G. van Koten, Angew. Chem., Int. Ed. Engl., 1997, 36, 755. 68 (a) T. J. Stemmler, J. E. Penner-Hahn and P. Knochel, J. Am. Chem. Soc., 1993, 115, 348; and (b) T. M. Barnhart, H. Huang and J. E. Penner-Hahn, J. Org. Chem., 1995, 60, 4310. 69 C.-S. Hwang and P. P. Power, Bull. Korean Chem. Soc., 2003, 24, 605. 70 N. Krause, Angew. Chem., Int. Ed. Engl., 1999, 38, 79. 71 M. Niemeyer, Organometallics, 1998, 17, 4649. 72 M. Niemeyer, Acta Cryst., 2001, E57, m416. 73 C.-S. Hwang and P. P. Power, Organometallics, 1999, 18, 697. 74 F. A. Cotton, X. Feng and D. J. Timmons, Inorg. Chem., 1998, 37, 4066. 75 C.-S. Hwang and P. P. Power, J. Organomet. Chem., 1999, 589, 234. 76 G. W. Rabe and N. W. Mitzel, Inorg. Chim. Acta, 2001, 316, 132. 77 A. Bayler, A. Schier, G. A. Bowmaker and H. Schmidbauer, J. Am. Chem. Soc., 1996, 118, 7006. 78 H. Lang, K. Ko¨hler, G. Rheinwald, L. Zsolnai, M. Bu¨chner, A. Driess, G. Huttner and J. Stra¨hle, Organometallics, 1999, 18, 598. 79 M. Niemeyer, Z. Anorg. Allg. Chem., 2004, 630, 252. 80 M. Niemeyer and P. P. Power, Organometallics, 1997, 16, 3258. 81 I. Resa, E. Carmona, E. Gutierrez-Puebla and A. Monge, Science, 2004, 305, 1136. 82 Z. Zhu, R. J. Wright, M. M. Olmstead, E. Rivard, M. Brynda and P. P. Power, Angew. Chem. Int. Ed., 2006, 45, 5807. 83 Z. Zhu, R. C. Fischer, J. C. Fettinger, E. Rivard, M. Brynda and P. P. Power, J. Am. Chem. Soc., 2006, 128, 15068. 84 Z. Zhu, M. Brynda, R. J. Wright, R. C. Fischer, W. A. Merrill, E. Rivard, R. Wolf, J. C. Fettinger, M. M. Olmstead and P. P. Power, J. Am. Chem. Soc., 2007, 129, 10847. 85 L. Pauling, in The Nature of the Chemical Bond, 3rd Edn., Cornell University Press, Ithica, NY, 1960, p. 394. 86 (a) Y. Wang, B. Quillian, P. Wei, H. Wang, X.-J. Yang, Y. Xie, R. B. King, P. v. R. Schleyer, H. F. Schaefer III and G. H. Robinson, J. Am. Chem. Soc., 2005, 127, 11944; (b) A. Grirrane, I. Resa, A. Rodriguez, E. Carmona, E. Alvarez, E. Gutierrez-Puebla, A. Monge, A. Galindo, D. del Rı´ o and R. A. Andersen, J. Am. Chem. Soc., 2007, 129, 693; (c) Y.-C. Tsai, D.-Y. Lu, Y.-M. Lin, J.-K. Hwang and J.-S. K. Yu, Chem. Commun., 2007, 4125; and (d) X.-J. Yang, J. Yu, Y. Xie, H. F. Schaefer, Y. Liang and B. Wu, Chem. Commun., 2007, 2365. Organomet. Chem., 2010, 36, 56–76 | 75

87 D. Bravo-Zhivotovskii, M. Yuzefovich, M. Bendikov, K. Klinkhammer and Y. Apeloig, Angew. Chem. Int. Ed., 1999, 38, 1100. 88 Y. Wang, B. Quillian, C. S. Wannere, P. Wei, P. v. R. Schleyer and G. H. Robinson, Organometallics, 2007, 26, 3054. 89 See, for example (Z5-C5H5Zn)2Co(Z5-C5H5)PPh3; P. H. M. Budzelaar, J. Boersma, G. J. M. Van der Kerk, A. L. Spek and A. J. M. Duisenberg, Inorg. Chem., 1982, 21, 3777 90 J. Emsley, in The Elements, Oxford University Press, Oxford, 1989.

76 | Organomet. Chem., 2010, 36, 56–76

Recent developments in the chemistry of metal-metal multiply bonded paddlewheel compounds Nathan J. Patmorea DOI: 10.1039/9781847559616-00077

This review summarises the main developments in the chemistry of metal-metal multiply bonded paddlewheel compounds reported in the literature between 2007 and 2008.

1.

Introduction

Metal-metal multiply bonded paddlewheel compounds typically consist of two metal atoms bridged by up to four bidentate three-atom ligands, such as a carboxylate or amidinate, as shown in Scheme 1 for a [M2(O2CR)4] compound. Complexes of this type incorporate transition metals from groups 5–9, and have been isolated with formal metal-metal bond orders (n) ranging from 0.5 to 4. In some exciting recent developments, the isolation of quintuply bonded paddlewheel compounds with two- or three-bidentate ligands has also been achieved.1,2 The current popularity and continued expansion of this ﬁeld is due to the intriguing electronic structure and variety of useful physical properties that are often inherent in paddlewheel compounds. These features, such as a well deﬁned coordination environment about the M2 core and interesting spectroscopic properties, results in their application in a wide variety of areas such as catalysis, antitumour metallopharmaceuticals, molecular wires and magnetic materials. A book published in 2005 comprehensively summarised the major developments in the area, and also provides an excellent introduction for those unfamiliar to the ﬁeld.3 R O M O

O

O O n O

R O

M O

R R Scheme 1

This review will summarise some of the main developments in the chemistry of M2n þ paddlewheel compounds of Cr, Mo, W, Re, Ru, Os and Rh, that have appeared more recently in the literature, between 2007 and 2008. a

Department of Chemistry, University of Sheﬃeld, Sheﬃeld, UK, S3 7HF

Organomet. Chem., 2010, 36, 77–92 | 77 c

The Royal Society of Chemistry 2010

2.

Chromium

Signiﬁcant attention has been paid recently to the isolation and characterisation of dichromium compounds with quadruple and quintuple bonds and ‘‘supershort’’ metal-metal distances. The crystal structure of the diazadiene complex [Cr2(L1)2] revealed a short Cr-Cr bond distance of 1.8028(9) A˚. DFT calculations at the BLYP/6-311g level indicate some degree of quintuple bonding, with natural resonance theory analysis giving an eﬀective bond order of 4.28.4 However, a more recent theoretical study on this compound employing multiconﬁgurational quantum chemical calculations (CASSCF/CASPT2) gave an eﬀective Cr-Cr bond order of 3.43, indicating a quadruple bond and a formal Cr oxidation state of þ 1.5 instead.5 The Cr–Cr bond length of 1.7397(9) A˚ in the D3-symmetric trigonal paddlewheel compound [Cr2(ArXylN{H}NArXyl)3] [ArXyl=2,6-(CH3)2-C6H3] is the shortest metal-metal bond length found to date in an isolable compound. DFT calculations (BP86) indicate that two d-bonds are formed from overlap of (dx2-y2 þ dx2-y2) and (dxy þ dxy) orbitals, to give a quintuply bonded electronic conﬁguration of s2p4d4.1 Bulky amidinate ligands were also used to stabilise a series of related dichromium(I) quintuply bonded complexes [Cr2(ArN{R}NAr)2] (Ar=ArXyl, 2,6-Et2-C6H3, 2,6-iPr2-C6H3; R=H or CH3), with Cr-Cr distances of B1.74 A˚.6 A formal quintuple bond between Cr(I) atoms in [Cr2(L2)2] results in a short Cr-Cr bond length [1.749(2) A˚]. The smaller calculated (DFT) Cr-Cr bond order of 4.2 is attributed to the weak contribution to the bond from the d orbitals.2

N

N

L1

N

N H

HL2

Four quadruply bonded Cr(II) complexes, [Cr2(formamidinate)n (acetate)4-n] (n=2–4), have been structurally characterised and exhibit Cr–Cr bond lengths in the range 1.8897(5) to 2.012(1) A˚, with distances dependent on the presence (long) or absence (short) of axial interactions.7 The heterotrinuclear complex [Cr2Fe(dpa)4Cl2] (dpa=2,2 0 -dipyridylamide) incorporates a quadruply bonded Cr2 unit into a linear Cr–Cr?Fe framework, with electrochemical studies showing a dramatic stabilisation of the higher iron oxidation states.8 3. 3.1

Molybdenum and tungsten Synthetic studies

Quadruply bonded dimolybdenum tetracarboxylates are typically synthesised by extended reﬂux (20–24 hours) of Mo(CO)6 and the carboxylic acid in high-boiling solvents (e.g. diglyme, 1,2-dichlorobenzene) under an 78 | Organomet. Chem., 2010, 36, 77–92

inert atmosphere. It was shown that closed-vessel microwave-assisted synthesis can be used to prepare [Mo2(O2CCH3)4], [Mo2(O2CCH2CH3)4] and [Mo2(O2CC6H4)] in less than one hour, with superior yields to traditional methods. In addition, this new method does not require the use of ‘‘dry’’ solvents, or an inert gas.9 A number of mixed amidate-carboxylate complexes of formula trans[M2(O2CMe)2(iPrNC{R}NiPr)2] (M=Mo, W; R=CCtBu, CCPh, CCCpFeCp) and cis-[Mo2(O2C-9-anthracene)2(iPrNC{R}NiPr)2] were prepared from reaction of [M2(O2CMe)4] and [Li(iPrNC{R}NiPr)]. Studies showed that both the amidinate and carboxylate ligands are kinetically labile to ligand exchange reactions in solution.10 The reaction of the amidate complexes [M2(mhp)4] (Hmhp=2-OH-6-Me-pyridine) with carboxylic acids gives and equilibrium mixture of complexes [M2(mhp)n(O2CR)4 n] (R= 2-thieny, Ph). However, when R=9-anthracene, the [W2(O2CR)4] product can be isolated cleanly, as it precipitates from solution driving the reaction. In addition, when the bulky ligand 2,4,6-triisopropylbenzoic acid (HTiPB) is employed, the thermodynamic product, [M2(mhp)2(TiPB)2], can be cleanly isolated.11 The mixed carboxylate-amidinate complexes [Mo2(DAniF)3 (O2CC6H4N)] and cis-[Mo2(DAniF)2(O2CC6H4N)2] (DAniF=N,N 0 -di-panisylformamidinate) that contain the bifunctional isonicotinate ligand [O2CC6H4N] have been used as building blocks in the formation of heterometallic arrays. Molecular rods have been synthesised by self-assembly with [Ni(acac)2] or [Rh2(O2CCH3)4], whereas self-assembly of ZnCl2 and the cis-Mo2 complex generated a cyclic oligomer.12 The reaction between [Mo2(TiPB)4] and 2 equivalents of [HO2CC6H4N] also leads to the formation of a ‘bis,bis’ complex, although substitution occurs in the trans position in this instance to generate trans-[Mo2(TiPB)2(O2CC6H4N)2]. This molecule is a redox-active analogue of 4,4-bipyridine, and was used to generate the rodlike complex [(1,5-COD)MePt]2[m-Mo2(TiPB)2(O2CC6H4N)2] and a 1-D polymeric chain [{Rh2(O2CMe)4}{Mo2(TiPB)2(O2CC6H4N)2]N.13 The role of axial donors in ligand isomerisation processes was investigated for the [Mo2(L3)2(O2CCH3)2][(BF4)2] series of complexes. The naphthyridinebased ligands with the stronger pyridyl (X=N, Y=CH2) and thiazolyl (X=S, Y=N) donors result in the formation of the cis-isomer, whereas the weaker thienyl (X=S, Y=CH2) and furyl (X=O, Y=CH2) groups result in a trans orientation of the ligands.14 The gas-phase reaction of [Mo2(O2CCF3)4] and [Rh2(O2CCF3)4] results in the formation of a heterometallic 1-D polymeric chain in the solid-state, comprising of a statistical distribution of homobimetallic [Mo2] and [Rh2] rather than heterobimetallic [MoRh] species.15

N

Y

N X L3

Oxidation of [Mo2(DAniF)3]2(m2-MoO4) (DAniF=N,N 0 -di-p-anisyl-formamidinate) leads to the formation of the unusual trigonal D3 assembly Organomet. Chem., 2010, 36, 77–92 | 79

{[Mo2(DAniF)3]3(m3-MoO4)2}2 þ , consisting of three oxidised Mo25 þ units connected by molybdate dianions. Electrochemical studies show three reversible one-electron processes corresponding to reduction of the Mo25 þ units, with DE1/2 of 0.36 and 0.41 V.16 The reaction of {Mo2(DAniF)2}2 þ units with hydride ions results in the formation of the tetranuclear complex [Mo2(cisDAniF)2]2(m-H)4, that has a Mo4H4 core with elongated tetrahedral symmetry and the shortest [Mo2]?[Mo2] separation (2.718 A˚) found to date.17 3.2

Photophysical properties

The photophysical properties of a series of fused and nonfused thienylbridged complexes [(tBuCO2)3M2]2[m-L] (M=Mo, W; L=L4, L5, L6) were investigated using steady-state absorption, transient absorption and emission spectroscopies. Dual emission was observed for both Mo and W complexes, with relaxation dynamics of the S1 states (tB10 ps) dominated by intersystem crossing to long-lived T1 states (Mo, tB70 ms; W, tB3 ms) that emit in the NIR region.18 The oligothiophene complexes [Mo2(TiPB)2(O2C-Th)2] (Th=2-thienyl, 2,2 0 -bithiophene, 2,2 0 :5 0 ,200 -terthiophene) show dual emission at room temperature in THF solution mainly originating from 1MLCT and 3MM(dd*) states, with 1MLCT lifetimes that increase from 4 to 12 ps upon increasing the number of thienyl rings, and phosphorescence lifetimes of B80 ms.19

O

OH

HO

OH S

O

S

O

S

S

3.3

O

O HO

L4

S

S

O

S

L5

HO

L6

HO

Electronically coupled MM quadruple bonds

Quadruply bonded ‘‘dimers of dimers’’ of the type [M2]-B-[M2], where B is a conjugated bridge, have been attracting signiﬁcant interest as models for the study of electron transfer and mixed valence. Electronic coupling in dicarboxylate bridged mixed-valence complexes of the form [(tBuCO2)3M2]2 (m-O2C-X-CO2) þ (M=Mo, W; X=p conjugated organic group) has been reviewed. The origin of the electronic coupling in these systems is strong interaction between the M2-d and bridge-p* orbitals, the extent of which was determined by EPR and absorption spectroscopies. Mo4 compounds were shown to typically exhibit valence trapped (class II) behaviour, whereas for W4 compounds full electron delocalisation over both M2 units (class III) was observed despite separations of up to 14 A˚.20 An example of this was shown for 2,5-dianilinoterephthalate (L7) bridged complexes [(tBuCO2)3M2]2 (m-L7) þ . The mixed-valence cations were shown to be valence localised for M=Mo and fully delocalised for M=W, with cross coupling matrix element (Hab) values of 383 cm 1 and 1500 cm 1 respectively.21 80 | Organomet. Chem., 2010, 36, 77–92

O−

O

H N N H −O

O L7

Exceptionally strong electronic communication between {Mo2(DAniF)3} þ fragments (DAniF=N,N 0 -di-p-anisyl-formamidinate) bridged by N-CH3 substituted benzoquinonemonoimines (L8 and L9) was observed by EPR spectroscopy and electrochemistry (KC up to 2.34 1014).22 Changing bridges from oxamidate (E=O) to dithiooxamidate (E=S) in the isomorphous [(DAniF)3Mo2]2(m-HN{E}CC{E}NH) complexes results in a signiﬁcant enhancement of the electronic coupling, due to a lowering of the bridge p* orbital energy that in turn decreases the energy barrier in an electron hopping exchange mechanism.23 Only weak electronic interactions are observed in complexes linked by squarate dianions of form [(DAniF)3Mo2]2(m-C4O4), with magnetic studies on the doubly oxidised complex suggesting only weak antiferromagnetic coupling (J= 121 cm 1) between the two unpaired electrons. This is in sharp contrast to the strong coupling observed for the analogous dioxolene bridged species, attributed to localisation of the squarate p-electrons to minimise antiaromaticity in the central C4 ring and energy mismatch in the frontier orbitals of the linker and [Mo2] units.24 It was also demonstrated that MM quadruple bonds could facilitate electron transfer between redox-active centres. Single-electron reduction of [W2(TiPB)2(L10)2] results in the population of an azulene p* orbital, with NIR electronic absorption spectroscopy indicating that the radical is fully delocalised over both azulene ligands by ligand p-M2d-ligand p conjugation.25 CH3 O

O

O

N

−O

N

N

N

O

O

CH3

CH3

CH3

L8

4.

L9

O OEt L10

Rhenium complexes

The Re-Re bond distances in the series of tetraguanidinate quadruply bonded paddlewheel complexes [Re2(L11)4X2] (X=CF3SO3 , CF3CO2 , F ) are signiﬁcantly eﬀected by the s- and p-donating electron abilities of the axial (X) ligands, with the complex containing the weakly coordinating [CF3SO3] ligand having the shortest Re-Re distance found to date [2.1562(7) A˚]. In addition, detailed EPR spectroscopy studies on related one-electron oxidised [Re2(L11)4]3 þ species show that the unpaired electron Organomet. Chem., 2010, 36, 77–92 | 81

lies in an MO of predominantly metal character with a little mixing from the guanidinate ligand orbitals.26 N N

N H HL11

The quadruply bonded dirhenium(III) phosphate complex [Re2(HPO4)2 (H2PO4)2(H2O)2] was synthesised by reaction of [Re2(O2CCH2CH3)4Cl2] with phosphoric acid.27 A mass spectrometric study was used to identify adducts of [Re2(O2CCH3)2Cl4] and purine dinucleotides, and the results suggest a preference for guanine coordination.28 Dirhenium carboxylates of general formula [Re2(O2CnH2n þ 1)4Cl2] (n=5–8) have been deposited as thin ﬁlms on silicon surfaces. Photolysis of these ﬁlms at 254 nm led to the formation of ReO3 thin ﬁlms, with quantum yields of up to 0.004.29 5.

Diruthenium complexes

5.1

Carboxylate ligands

Theoretical calculations using density functional theory employing the B3LYP functional have been used to investigate the electronic structure of Ru2(II,II) tetracarboxylates axially coordinated with nitric oxide. The calculations gave the expected triplet ground-state electronic conﬁguration of s2p4d2d*2p*2 for [Ru2(O2CR)4(L)2] (R=H, CH3, CF3; L=nothing, H2O, THF) species, with two unpaired electrons.30 DFT calculations have also been used to interpret the structural, spectroscopic and electrochemical data in Ru2(II,III) complexes of form [Ru2(O2CCH3)4X2] (X=Cl , Br , I ). This study also included calculations on [Ru2(O2CCH3)4(L)2] þ (L=nothing, H2O) species, which were found to have a quartet ground-state electronic conﬁguration of s2p4d2(d*p*)3. The three unpaired electrons are a result of the near-degeneracy of the p*-d* manifold, in agreement with previous studies.31 Carboxylate exchange reactions were used to prepare three 3-thiophenecarboxylate and one 2,2 0 -bithiophene-5-carboxylate complexes, [Ru2(L)4 (MeOH)2][PF6] (L=L12, L13, L14, L15). Electrochemical studies were used to assess their ability to undergo oxidative polymerisations, with the bithiophene ring system showing the greatest promise as it has the lowest thiophene oxidation potential.32 Pyrrole-2-carboxylate complexes [Ru2(O2CC4H4N)4Cl] have been crystallised as water and acetone solvates that display diﬀerent hydrogen bonding and p-p stacking interactions in the solid-state.33 Porous, crystalline solids of dicarboxylate bridged Ru2(II,III) complexes [Ru2(terephthalate)2Cl] 3.5H2O and [Ru2(adipate)2Cl] 1.5H2O have high BET surface areas (235 m2g 1 and 281 m2g 1, respectively) and occlude similar numbers of mol of N2 per mol of metal (B0.8).34 Rate constants for the axial-ligand-substitution reactions of [Ru2(O2CCH3)4(MeCN)2] þ with the phosphanes PCy3 and PCy2Ph were investigated using UV/vis spectroscopy, and found to be two orders of magnitude less than similar axial-ligand substitutions for [Rh2(O2CR)4(L)2] 82 | Organomet. Chem., 2010, 36, 77–92

complexes. In addition, variable temperature magnetic and X-ray crystallography studies on the phosphane diadduct [Ru2(O2CCH3)4(PCy3)2] þ are consistent with a s2p4d2p*3 conﬁguration at low temperature and a s2p4d2p*2d*1 conﬁguration at room temperature.35 Ru2(II,III) compounds containing one o-alkene-a-carboxylate ligand of form [Ru2(RNC{H} NR)3(L16)Cl] (R=3,5-Cl2-Ph; n=1 or 2) undergo oleﬁn cross metathesis reactions catalysed by [(Cy3P)2Cl2Ru(=CHPh)] to give dimeric products. By contrast, Ru2(II,III) compounds containing two cis-oriented o-alkene-acarboxylates (L16) undergo an intramolecular ring closing reaction to yield cis[Ru2(RNC{H}NR)3(O2C{CH2}nCH=CH{CH2}nCO2)Cl].36 S

S

S

S

O

O OH

HO

HL12

n

S

O

O

OH HL13

HO

O

HO

HL14

HL15

HL16

A slow diﬀusion reaction of the Ru2(II,II) complex [Ru2(O2CCF3)4 (THF)2] with 7,7,8,8-tetracyanoquinodimethane (TCNQ) results in the formation of a ladder chain network, having {Ru2} rails and TCNQ rungs in a 2:1 ratio. Magnetic measurments indicated relatively strong antiferromagnetic interactions between the [Ru2] units.37 Self-assembly of [Ru2(O2CPh-m-F)4] (O2CPh-m-F=m-ﬂuorobenzoate) with BTDA-TCNQ (L17) results in the formation of an inﬁnite 3D network of form [{Ru2(O2CPh-m-F)4}2(L17)]n. Structural studies suggest full electron transfer from one [Ru2] unit to BTDA-TCNQ, and the compound exhibits longrange ferromagnetic order with Tc=107 K.38 The compound [Ru2(L18)2Cl] is an active catalyst for the oxygenation of organic sulﬁdes by tbutyl hydroperoxide, with good selectivity for formation of the sulfoxide over the sulfone.39 The eﬀect on tumor-cell proliferation of Ru2(II,III) complexes incorporating the deprotonated carboxylate from the non-steroidal anti-inﬂammatory drugs (NSAIDs) ibuprofen, aspirin, naproxen and indomethacin of form [Ru2(NSAID)4Cl] and [Ru2(NSAID)4(H2O)2][PF6] was investigated. No signiﬁcant eﬀects were found in the Hep2 human larynx or T24/83 human bladder tumor, but the naproxen and ibuprofen complexes increased signiﬁcantly C6 rat glioma cell antiproliferation activity.40 S N

N

NC

CN

NC

CN N

N S L17

HO

OH O

O H2L18

Organomet. Chem., 2010, 36, 77–92 | 83

5.2

N,N 0 -bidentate ligands

Complexes of [Ru2(CO)4]2 þ containing 2-aryl substituted napthyridine ligands undergo aryl C-H activation at room temperature to form [Ru2(CO)4(L19)(HL19)], [Ru2(CO)4(L20)(HL20)] and [Ru2(CO)4(L21)(HL21)], in which the aryl group from one of the naphthyridine ligands is orthometalated while the second is bound in an agostic fashion.

N

N

N

N

H HL19

O

N

H

H HL20

S

N

HL 21

The remarkable Ru-RuN nitrido complex [Ru2(dPhf)4N] (dPhf=N,N 0 diphenylformamidinate) was synthesised from thermal or photolytic decompostion of a Ru-Ru-N3 precursor. The complex has one unpaired electron, and is stabilised by the population of new nonbonding MO combinations by electrons that would otherwise occupy antibonding orbitals to give a s2p4s(n.b.)2p(n.b.)4d*1 electronic conﬁguration.41 The ﬁrst open-paddlewheel structures in diruthenium chemisty, [Ru2(dPhf)3X2] (X=Cl, Br), have been reported, and can be used to generate the unsaturated unit {Ru2(dPhf)3}2 þ stabilised by MeCN, BF4 , or NO3 . The magnetic behaviour of the complexes is intermediate between low- and high-spin conﬁgurations.42 In another ‘‘ﬁrst’’, a carbonyl Ru2(II,III) complex [Ru2(dPhf)3(O2CCH3)(CO)] was isolated along with its isomorphous nitrosyl analogue, with magnetic measurements indicating one and three unpaired electrons respectively.43 The isothiocyanate Ru2(II,III) complex (3,1) [Ru2(F3ap)4(NCS)] (F3ap=2,4,6-triﬂuoroanilinopyridinate) has three unpaired electrons, whereas the related Ru2(III,III) complex (3,1) [Ru2(F3ap)3(F2Oap)(NCS)] was found to be diamagnetic.44 Interconversion between the (4,0) and (3,1) isomers of diruthenium anilinopyridinate (ap) complexes was demonstrated by reaction of (4,0) [Ru2(ap)4Cl] with LiCCC5H4N to give (3,1) [Ru2(ap)4(CCC5H4N)], and reaction of (3,1) [Ru2(F3ap)4Cl] with TBACl H2O to give (4,0) [Ru2(F3ap)4Cl].45 Diruthenium s-alkynyl compounds are showing promise as building blocks for organometallic molecular wires. Electrochemical measurements on unsymmetric donor-Ru2-acceptor compounds trans-[(ArCC)Ru2(dmba)4 (CCAr 0 )] (Ar=NMe2-substituted phenyl; Ar 0 =NO2-substituted phenyl; dmba=N,N 0 -dimethylbenzamidinate) were used to illustrate the potential of these types of species to act as organometallic Aviram-Ratner diodes.46 Electron transport through trans-[(-S-C6H4-CC)Ru2(ap)4(CC-C6H4-S-)] was measured by self-assembly into nanogap molecular junctions.47 Synthesis of the diruthenium heterocycle-acetylide complexes [Ru2(L22)4(CC-Ar)2] (Ar=2-pyrimidine, 4-N-methylpyridinium) has been reported, with structural and electrochemical data that are consistent with the heterocycles being moderate electron acceptors.48 A family of Ru2(II,III) alkynyl compounds containing aniliopyridinato ligands modiﬁed with methoxy groups (L23) of 84 | Organomet. Chem., 2010, 36, 77–92

form [Ru2(L23)4({CC}n-R)] (n=1 or 2; R=Fc, SiMe3; X, Y=H or OMe) have been synthesised and structurally characterised.49 CH3

X

Y

N HN

Y

CH3

N H

N

HL23

HL22: Y = H, OMe

Peripheral covalent modiﬁcation of diruthenium compounds has been achieved using a variety of cross-coupling reactions. The compounds [Ru2(D{3,5-Cl2Ph}F)(L24)Cl] and [Ru2(D{3,5-Cl2Ph}F)(L25)Cl] [D{3,5Cl2Ph}F=N,N 0 -bis(3,5-dichlorophenyl)formamidinate] undergo Heck cross-coupling reactions with terminal oleﬁns; under diﬀerent conditions, the compound trans-[(PhCC)2Ru2(D{3,5-Cl2Ph}F)(L25)] also underwent the Heck reaction. Structural and electrochemical studies indicate that the peripheral functionalisation with oleﬁns does not signiﬁcantly perturb the geometry or electronic properties of the diruthenium core.50 Peripheral modiﬁcation of [Ru2(D{3,5-Cl2Ph}F)(L24)Cl] and [Ru2(D{3,5-Cl2Ph}F) (L25)Cl] could also be achieved under more mild conditions by employing Negishi coupling reactions with [BrZnAr] (Ar=C6H5, 4-OCH3-C6H4, 4-CF3-C6H4) to give the biphenyl products.51 Diruthenium dendritic compounds have been prepared via a facile Cu(I)-catalysed cycloaddition between [Ru2(D{3,5-Cl2Ph}F)4 n(L26)n] (n=1 or 2) and azidopoly(benzyl ether) dendrons.52 A short review has been published on the use postligation modiﬁcation in diruthenium compounds using Sonogashira, Suzuki, Negishi and Heck cross-coupling reactions to generate molecular architectures.53

I I

I N

N H HL24

6.

H3C

N

N H HL25

CH3

H3C

N

N H

CH3

HL26

Dirhodium complexes

The synthesis, isolation and structural characterisation of a Rh2(III,III) paddlewheel complex with no metal-metal bond has been reported for the ﬁrst time. Aerobic oxidation of [Rh2(cap)4(MeCN)2] (cap=caprolactamate, L27) with [Cu(I)(OTf)] (10 mol%) in the presence of [NaBPh4] results in the formation of the Rh2(III,III) complex [Rh2(cap4)(Ph)2]. Structural (Rh?Rh= 2.519 A˚) and spectroscopic studies indicate cleavage of the Rh-Rh bond, proposed to result from a change in metal hybridisation to d2sp3 and a Organomet. Chem., 2010, 36, 77–92 | 85

p4d2p*4d*2 electronic conﬁguration.54 Equilibrium constants for the acid-base reaction of [Rh2(O2CR)2(PC)2(H2O)2] (R=alkyl or aryl group; PC=cyclometalated aryl phosphine) with diﬀerent Lewis bases that can displace the water ligands and axially coordinate to the Rh2(II,II) core were determined by UV/vis spectroscopy, and agree with trends from the calculated (DFT) interaction energies. The inductive eﬀect of base coordination at the axial site weakens the Lewis acidity of the second metal, reducing the equilibrium constant for the substitution of the second Lewis base.55 DFT calculations were used to investigate the inﬂuence of carboxylate, carboxamidate and orthometalated arylphospines on the electronic structure of Rh2(II,II) carbene complexes. The amidate and ortho-metalated phosphine ligands are better electron donors, resulting in greater Rh - Ccarbene p back-bonding which can account for the greater selectivity of these complexes in catalytic carbene transfer reactions.56

N H

O

HL27

The pyridyl groups of the Rh2(II,II) tetraamidinate complex [Rh2(L28)4] were coordinated to four Re(I) chromophores, {Re(CO)3(bpy)} þ , to generate a hexametallic light-harvesting assembly, with electrochemical studies indicating no communication between the Re(I) centers.57 Self-assembly of the Ru(II) polypyridyl chromophores I, II, and III about Rh2(II,II) cores was used to generate light-harvesting systems of form [Rh2(O2CCH3)4-n (RuNN)n(MeCN)2]2n þ (n=1, 2, 3, 4; RuNN=I, II, III). Photophysical studies show eﬃcient energy transfer from the Ru-based MLCT triplet state of the I-based complexes to the lowest energy excited state of the Rh2(II,II) core takes places in acetonitrile at 298 K, whereas such a process is ineﬃcient for the II- and III-based complexes. This is rationalised in terms of the driving force and reorganisation energies of the complexes, and places the energy of the non-emissive state of the dirhodium tetracarboxylate core near 1.77 eV.58

N

N

N

X N

R N H

N

O−

Y

O

N

Ru

X N

Y

N

HL28 I: R = H; X, Y = CH II: R = Ph; X = N; Y = CH III: R = H; X = CH; Y = N

86 | Organomet. Chem., 2010, 36, 77–92

6.1

Catalytic applications

One of the attractions of dirhodium paddelwheel complexes is their ability to catalyse a wide variety of organic transformations such as C–H insertions, cyclopropanations and ylide formation. A review on the application of high symmetry chiral Rh2(II,II) paddlewheel compounds highlights their application as catalysts for asymmetric metal carbenoid and nitrenoid reactions, and as Lewis acids.59 Their impressive performance as catalysts in C–H functionalisation reactions has been exploited in the synthesis of complex natural products and pharmaceutical agents. A recent review on catalytic C–H functionalisation by metal carbenoid and nitrenoid insertion demonstrates the important role of dirhodium species in this ﬁeld.60 Dirhodium(II,II) nitrenoids of [Rh2(O2CC3F7)4] can be formed under mild conditions from azides, which was exploited for catalytic intramolecular C–H amination of vinyl azides61 and aryl azides62 into indoles and other N-heterocycles. The chiral dirhodium carboxamidate complex [Rh2(S-nap)4], IV, is an eﬀective catalyst for enantioselective intramolecular amination of benzylic and allylic C–H bonds.63 A family of chiral dirhodium complexes containing amino acids with diﬀerent nitrogen protecting groups were evaluated as catalysts in stereoselective intermolecular C–H amination reactions. A highly reactive chiral metallanitrene derived from sulfonimidamides is formed in situ and leads to highly eﬃcient (up to 92% yield) functionalisation of benzylic and allylic substrates with stereoselectivities of up to 99%.64 The performance of the Rh2(II,II) catalyst [Rh2(O2CCH3)4] in intramolecular C–H insertion reactions of diazo-acetamides is highly inﬂuenced by axial coordination of an N-heterocyclic carbene (NHC). The rates and selectivity of the monocoordinated NHCRh2 catalyst are diﬀerent to those of the parent [Rh2(O2CCH3)4] species and, interestingly, a new reaction mode was discovered that leads to a decarbonylation pathway.65 Dirhodium N-heterocyclic carbene complexes were also shown to be highly eﬃcient catalysts (up to 99% yield) for the arylation of aldehydes under mild conditions.66 Ts H

N O

N

Rh

Rh IV

Dirhodium tetracetate has been employed as an eﬃcient catalyst for the synthesis of highly functionalised pyridines and 1,4-dihydropyridines via a rhodium carbenoid induced ring expansion of isoxazoles,67 and synthesis of cyclobutanes by highly selective ring expansion of cyclopropanes.68 Dirhodium caprolactomate, [Rh2(L28)4] has also proven to be a versatile catalyst for a number of transformations including, oxidation of secondary amines to imines by tert-butyl hydroperoxide with high chemo- and regioselectivity,69 and allylic oxidation of D5-steroids to 7-keto-D5-steroids by 70% tert-butyl Organomet. Chem., 2010, 36, 77–92 | 87

hydroperoxide in water.70 Cooperative catalysis between a chiral phosphoric acid and [Rh2(O2CCH3)4] was used for three-component reaction of diazo compounds with alcohols and imines to generate syn-b-amino-a-hydroxyl acid derivatives with excellent enantioselectivity.71 The mechanism of Rh(II)-catalysed cycloproponation of alkenes has been investigated computationally using the model [Rh2(O2CH)4] complex, with CH2N2 and C2H4 as model organic substrates, and MeCl as a model for coordinating solvent. Three potential carriers of catalysis were identiﬁed, with complexes resulting from CH2 insertion into the Rh-O bond reducing the activation energy for the rate-determining CH2-N2 bond cleavage step the most.72 Chiral dirhodium compounds, generated using a chiral metal-metal bridging ligand, are being successfully employed highly selective catalysts. For example, chiral cationic dirhodium(II,III) carboxamidates were shown to enhance selectivity in Lewis acid catalysed transformations,73 and diastereoselective imination of sulﬁdes was achieved using a chiral dirhodium tetracarboxylate catalyst.74 Dirhodium ortho-metalated phosphine complexes were immobilised on a carboxyethylpolystyrene polymer and evaluated as catalysts in the asymmetric cycloproponation of styrene with ethyl diazoacetate. The reaction yields were improved by comparison to the homogenous chiral triﬂuoroacetate derivatives, although their diastereoand enantioselectivities were lower.75 6.2

Biological applications

Dirhodium(II,II) complexes are receiving signiﬁcant interest for biological applications, because of their signiﬁcant antitumour activity and notable interactions with DNA. The eﬀect of axial coordination on the electronic structure and biological activity was investigated using the cis-[Rh2(O2CCH3)2 (np)2]2 þ , cis-[Rh2(O2CCH3)2(np)(pynp)]2 þ and cis-[Rh2(O2CCH3)2(pynp)2]2 þ [np=1,8-napthyridine; pynp=2-(2-pyridyl)1,8-naphthyridine] series of complexes, in which the pnyp bridging ligand occupies one or two of the axial coordination sites. The ability of the complexes to stabilise duplex DNA and inhibit transcription in vitro was greatly aﬀected by the availability of the axial site, with the binding constant for the complex with two axial positions blocked approximately two-orders of magnitude smaller than the complexes which have open axial positions that may bind covalently to DNA.76 One- and twodimensional NMR spectroscopy, pH titrations and molecular modelling studies were used to characterise the adduct formed from the reaction of the anticancer active compound [Rh2(DTolF)2(CH3CN)6](BF4)2 (DTolF=N,N 0 di-p-tolylformamidinate) and the DNA fragment d(ApA), H2L.29 The studies show that the adenine bases are deprotonated, and bind to the Rh2-core in equatorial positions via N7/N6, which stabilises the rare imino form of the ligand.77 High level computations employing combined DFT and continuum dielectric model approach, were used to study the reaction of three adenine tautomers with dirhodium tetraformate. In agreement with the previously described experimental study, the thermodynamic product was found to be the N7/N6 bound tautomer in the rare imino form, although transition state predictions suggested that the multiple step mechanisms leading to formation of this species are kinetically challenging.78 88 | Organomet. Chem., 2010, 36, 77–92

NH2 7

6

N 5

8 9N 4

N1 2

N 3

HO

NH2 7

O

6

N 5 8 9N

O

4

N

2

3

O

O P

N1

O

O−

OH H2L29

Dirhodium dinucleotide adducts [Rh2(DTolF)2{d(GpA)}2] [d(GpA)=L30] and [Rh2(DTolF)2{d(ApG)}2] [d(ApG)=L31] exhibit head-to-head arrangements of the bases that bridge the dirhodium core. The adenine bases coordinate via N6 and N7 which again stabilises the rare imino tautomer, and the guanine bases coordinate at positions N7 and O6.79 The mechanism of the reaction between guanine and dirhodium tetracarboxylate leading to the guanine-N7,O6 equatorially coordinated product was investigated using quantum chemical calculations. The calculations suggest a multi-step reaction mechanism via an axial guanine-N7 adduct and an axial-equatorial carboxylate chelate as unexpected key intermediates.80 A computational study was performed to determine the thermodynamic products arising from the replacement of one or two acetyl-ligands from [Rh2(O2CCH3)4(H2O)2] by adenine and guanine DNA bases. Higher stabilisation energies as well as bonding energies were found for adenine coordination, and the smaller HOMO-LUMO gap and higher-lying HOMO of the adenine complexes indicates they are likely to be more reactive.81

O

NH2

7

N 5

1

6

8

7 8

2

9N 4

N 7

O

N 5

6

8 9N

O O

P O−

9N 4

NH2 NH2

3

HO

4

N 3

O O OH H2L30

6

N 5

NH

N1 2

N 3

HO O

N1

O 7

N 5 8

2

9N

O O

P O−

4

6

1

NH

N 2 NH2 3

O O OH H2L31

Organomet. Chem., 2010, 36, 77–92 | 89

7.

Conclusions

This review has demonstrated that metal-metal multiply bonded paddlewheel compounds continue to attract attention from a number of research groups due to their potential application in a variety of ﬁelds, such as the use of metal-metal quadruply bonded complexes for the study of mixed-valency and the incorporation of diruthenium complexes into materials with interesting magnetic properties. The ability of dirhodium and diruthenium compounds to catalyse a variety of transformations is likely to continue to be of interest, along with further study of the biological applications of dirhodium complexes. Other future research endevours are likely to include the application of quadruply bonded Mo2 and W2 paddlewheel complexes in optoelectronic devices, based on the recent investigations into their remarkable photophysical properties and discovery of their long-lived excited state lifetimes.

References 1 Y.-C. Tsai, C.-W. Hsu, J.-S. K. Yu, G.-H. Lee, Y. Wang and T.-S. Kuo, Angew. Chem. Int. Ed., 2008, 47, 7250. 2 A. Noor, F. R. Wagner and R. Kempe, Angew. Chem. Int. Ed., 2008, 47, 7246. 3 F. A. Cotton, C. A. Murillo and R. A. Walton, ‘Multiple Bonds Between Metal Atoms’, Springer Science and Business Media, Inc, 2005. 4 K. A. Kreisel, G. P. A. Yap, O. Dmitrenko, C. R. Landis and K. H. Theopold, J. Am. Chem. Soc., 2007, 129, 14162. 5 G. La Macchia, F. Aquilante, V. Veryazov, B. O. Roos and L. Gagliardi, Inorg. Chem., 2008, 47, 11455. 6 C.-W. Hsu, J.-S. K. Yu, C.-H. Yen, G.-H. Lee, Y. Wang and Y.-C. Tsai, Angew. Chem. Int. Ed., 2008, 47, 9933. 7 F. A. Cotton, Z. Li and C. A. Murillo, Eur. J. Inorg. Chem., 2007, 3509. 8 M. Nippe and J. F. Berry, J. Am. Chem. Soc., 2008, 129, 12684. 9 K. D. Johnson and G. L. Powell, J. Organomet. Chem., 2008, 693, 1712. 10 D. J. Brown, M. H. Chisholm and J. C. Gallucci, Dalton Trans., 2008, 1615. 11 D. J. Brown, M. H. Chisholm and C. W. Gribble, Dalton Trans., 2008, 1793. 12 F. A. Cotton, J.-Y. Jin, Z. Li, C. Y. Liu and C. A. Murillo, Dalton Trans., 2008, 2328. 13 M. H. Chisholm, A. S. Dann, F. Dielmann, J. C. Gallucci, N. J. Patmore, R. Ramnauth and M. Scheer, Inorg. Chem., 2008, 47, 9248. 14 M. Majumdar, S. K. Patra, M. Kannan, K. R. Dunbar and J. K. Brera, Inorg. Chem., 2008, 47, 2212. 15 B. Li, H. AZhang, L. Huynh, M. Shatruk and E. V. Dikarev, Inorg. Chem., 2007, 46, 9155. 16 J. P. Donahue and C. A. Murillo, Dalton Trans., 2008, 1547. 17 F. A. Cotton, C. A. Murillo and Q. Zhao, Inorg. Chem., 2007, 46, 6858. 18 M. H. Chisholm, P.-T. Chou, Y.-H. Chou, Y. Ghosh, T. L. Gustafson and M.-L. Ho, Inorg. Chem., 2008, 47, 3415. 19 G. T. Burdzinskim, M. H. Chisholm, P.-T. Chou, Y.-H. Chou, F. Feil, J. C. Gallucci, Y. Ghosh, T. L. Gustafson, M.-L. Ho, Y. Liu, R. Ramnauth and C. Turro, Proc. Natl. Acad. Sci. USA, 2008, 105, 15247. 20 M. H. Chisholm and N. J. Patmore, Acc. Chem. Res., 2007, 40, 19. 21 M. H. Chisholm and N. J. Patmore, Dalton Trans., 2007, 91. 90 | Organomet. Chem., 2010, 36, 77–92

22 F. A. Cotton, J.-Y. Jin, Z. Li, C. A. Murillo and J. H. Reibenspies, Chem. Commun., 2008, 211. 23 F. A. Cotton, Z. Li, C. Y. Liu and C. A. Murillo, Inorg. Chem., 2007, 46, 7840. 24 F. A. Cotton, C. A. Murillo, M. D. Young, R. Yu and Q. Zhao, Inorg. Chem., 2008, 47, 219. 25 M. V. Barybin, M. H. Chisholm, N. J. Patmore, R. E. Robinson and N. Singh, Chem. Commun., 2007, 3652. 26 F. A. Cotton, N. S. Dalal, P. Huang, S. A. Ibragimov, C. A. Murillo, P. M. B. Piccoli, C. M. Ramsey, A. J. Schultz, X. Wang and Q. Zhao, Inorg. Chem., 2007, 46, 1718. 27 A. V. Shtemenko, V. G. Stolyarenko and K. V. Domasevich, Russ. J. Coord. Chem., 2007, 33, 79. 28 E. F. Day, T. A. Payne and C. A. Holt, Rapid Commun. Mass Spectrom., 2007, 21, 903. 29 J. P. Bravo-Vasquez and R. H. Hill, J. Photochem. Photobio. A, 2008, 196, 1. 30 O. V. Sizova, L. V. Skripnikov, A. Y. Sokolov and O. O. Lyubimova, J. Struct. Chem., 2007, 48, 28. 31 M. A. Castro, A. E. Roitberg and F. D. Cukiernik, Inorg. Chem., 2008, 47, 4682. 32 A. H. Murray, Z. Yue, A. I. Wallbank, T. S. Cameron, R. Vadavi, B. J. MacLean and M. A. S. Aquino, Polyhedron, 2008, 27, 1270. 33 M. d. C. Barral, R. Gonza´lez-Prieto, S. Herrero, R. Jime´nez-Aparicio, J. L. Priego, E. d. C. Royer, M. R. Torres, F. A. Urbanos and F. Zamora, J. Clust. Sci., 2008, 19, 219. 34 G. Ribeiro, F. M. Vichi and D. d. O. Silva, J. Mol. Struct., 2008, 890, 209. 35 T. J. Burchell, T. S. Cameron, D. H. Macartney, L. K. Thompson and M. A. S. Aquino, Eur. J. Inorg. Chem., 2007, 4021. 36 W.-Z. Chen, J. D. Protasiewicz, S. A. Davis, J. B. Updegraﬀ, L.-Q. Ma, P. E. Fanwick and T. Ren, Inorg. Chem., 2007, 46, 3775. 37 N. Motokawa, T. Oyama, S. Matsunaga, H. Miyasaka, K. Sugimoto, M. Yamashita, N. Lopez and K. R. Dunbar, Dalton Trans., 2008, 4099. 38 N. Motokawa, H. Miyasaka, M. Yamashita and K. R. Dunbar, Angew. Chem. Int. Ed., 2008, 47, 7760. 39 J. E. Barker and T. Ren, Inorg. Chem., 2008, 47, 2264. 40 G. Ribeiro, M. Benadiba, A. Colquhoun and D. d. O. Silva, Polyhedron, 2008, 27, 1131. 41 J. S. Pap, S. D. George and J. F. Berry, Angew. Chem. Int. Ed., 2008, 47, 10102. 42 M. C. Barral, T. Gallo, S. Herrero, R. Jime´nez-Aparicio, M. R. Torres and F. A. Urbanos, Chem. Eur. J., 2007, 13, 10088. 43 M. C. Barral, S. Herrero, R. Jime´nez-Aparicio, M. R. Torres and F. A. Urbanos, J. Organomet. Chem., 2008, 693, 1597. 44 M. Nguyen, T. Phan, E. V. Caemelbecke, X. Wei, J. L. Bear and K. M. Kadish, Inorg. Chem., 2008, 47, 4392. 45 M. Nguyen, T. Phan, E. V. Caemelbecke, W. Kajonkijya, J. L. Bear and K. M. Kadish, Inorg. Chem., 2008, 47, 7775. 46 J.-W. Ying, A. Cordova, T. Y. Ren, G.-L. Xu and T. Ren, Chem. Eur. J., 2007, 13, 6874. 47 A. K. Mahapatro, J.-W. Ying, T. Ren and D. B. Janes, Nano Lett., 2008, 8, 2131. 48 J.-W. Ying and T. Ren, J. Organomet. Chem., 2008, 693, 1449. 49 B. Xi, G.-L. Xu, J.-W. Ying, H.-L. Han, A. Cordova and T. Ren, J. Organomet. Chem., 2008, 693, 1656. 50 W.-Z. Chen, P. E. Fanwick and T. Ren, Organometallics, 2007, 26, 4115. Organomet. Chem., 2010, 36, 77–92 | 91

51 L. Zhang, Z. Huang, W.-Z. Chen, E.-i. Negishi, P. E. Fanwick, J. B. Updegraﬀ, J. D. Protasiewicz and T. Ren, Organometallics, 2007, 26, 6526. 52 W.-Z. Chen, P. E. Fanwick and T. Ren, Inorg. Chem., 2007, 46, 3429. 53 T. Ren, C. R. Chemie, 2008, 11, 684. 54 J. M. Nichols, J. Wolf, P. Zavalij, B. Varughese and M. P. Doyle, J. Am. Chem. Soc., 2007, 129, 3504. 55 P. Hirva, J. Esteban, J. Lloret, P. Lahuerta and J. Pe´rez-Prieto, Inorg. Chem., 2007, 46, 2619. 56 J. Lloret, J. J. Carbo´, C. Bo, A. Lledo´s and J. Pe´rez-Prieto, Organometallics, 2008, 27, 2873. 57 D. Chartrand and G. S. Hanan, Chem. Commun., 2008, 727. 58 M. W. Cooke, G. S. Hanan, F. Loiseau, S. Campagna, M. Watanabe and Y. Tanaka, J. Am. Chem. Soc., 2007, 129, 10479. 59 J. Hansen and H. M. L. Davies, Coord. Chem. Rev., 2008, 252, 545. 60 H. M. L. Davies and J. R. Manning, Nature, 2008, 451, 417. 61 B. J. Stokes, H. Dong, B. E. Leslie, A. L. Pumphrey and T. G. Driver, J. Am. Chem. Soc., 2008, 129, 7500. 62 M. Shen, B. E. Leslie and T. G. Driver, Angew. Chem. Int. Ed., 2008, 47, 5056. 63 D. N. Zalatan and J. Du Bois, J. Am. Chem. Soc., 2008, 130, 9220. 64 C. Liang, F. Collet, F. Robert-Peillard, P. Mu¨ller, R. H. Dodd and P. Dauban, J. Am. Chem. Soc., 2008, 130, 343. 65 L. F. R. Gomes, A. F. Trindade, N. R. Candeias, P. M. P. Gois and C. A. M. Afonso, Tet. Lett., 2008, 49, 7372. 66 P. M. P. Gois, A. F. Trindade, L. F. Veiros, V. Andre´, T. Duarte, C. A. M. Afonso, S. Caddick and F. G. N. Cloke, Angew. Chem. Int. Ed., 2007, 46, 5750. 67 J. R. Manning and H. M. L. Davies, J. Am. Chem. Soc., 2008, 130, 8602. 68 H. Xu, W. Zhang, D. Shu, J. B. Werness and W. Tang, Angew. Chem. Int. Ed., 2008, 47, 8933. 69 H. Choi and M. P. Doyle, Chem. Commun., 2007, 745. 70 H. Choi and M. P. Doyle, Org. Lett., 2007, 9, 5349. 71 W. Hu, X. Xu, J. Zhou, W.-J. Liu, H. Huang, J. Hu, L. Yang and L.-Z. Gong, J. Am. Chem. Soc., 2008, 130, 7782. 72 J. A. S. Howell, Dalton Trans., 2007, 1104. 73 Y. Wang, J. Wolf, P. Zavalij and M. P. Doyle, Angew. Chem. Int. Ed., 2008, 47, 1439. 74 F. Collett, R. H. Dodd and P. Dauban, Org. Lett., 2008, 10, 5473. 75 J. Lloret, F. Estevan, K. Biegar, C. Villanueva and M. A. U´beda, Organometallics, 2007, 26, 4145. 76 J. D. Aguirre, D. A. Lutterman, A. M. Angeles-Boza, K. R. Dunbar and C. Turro, Inorg. Chem., 2007, 46, 7494. 77 H. T. Chifotides and K. R. Dunbar, J. Am. Chem. Soc., 2007, 129, 12480. 78 D. V. Deubel, J. Am. Chem. Soc., 2008, 130, 665. 79 H. T. Chifotides and K. R. Dunbar, Chem. Eur. J., 2008, 14, 9902. 80 D. V. Deubel and H. T. Chifotides, Chem. Commun., 2007, 3438. 81 J. V. Burda and J. Gu, J. Inorg. Biochem., 2008, 102, 53.

92 | Organomet. Chem., 2010, 36, 77–92

The Pauson-Khand reaction Benjamin E. Moultona DOI: 10.1039/9781847559616-00093

The metal mediated synthesis of cyclopentenones via a [2 þ 2 þ 1] cycloaddition between an alkyne, an alkene and carbon monoxide has become commonly known as the Pauson-Khand (PK) reaction. This report will brieﬂy summarise some of the major developments since its initial discovery including an intramolecular variant of the reaction, the progress made towards making the process catalytic and examples of how the reaction has been utilised. The proposed mechanism for the reaction and the factors that inﬂuence the product distribution will also be introduced.

1.

Introduction

The formation of ketonic products when acetylene hexacarbonyl dicobalt is treated with norbornadiene was a serendipitous discovery ﬁrst reported in 1971 by Pauson and co-workers.1,2 The group were investigating the reaction between acetylene or phenylacetylene hexacarbonyl dicobalt and norbornadiene and found that varying the nature of the solvent changed the major product of the reaction (Scheme 1).

Scheme 1 Original products identiﬁed from reaction between acetylene hexacarbonyl dicobalt (1) and norbornadiene.

They noted that in dimethoxyethane or in iso-octane (path a), the major product was dicarbonylcyclopentadienylcobalt (2) which must arise as a result of a retro Diels-Alder reaction of the norbornadiene (which would lead to the formation of acetylene and cyclopentadiene). When the solvent was changed to an aromatic hydrocarbon such as benzene or toluene (path b), the major cobalt-containing product was shown to be a complex derived from Co4(CO)12 , with three CO ligands on an apical cobalt being replaced by a molecule of the aromatic solvent (3). The group noted that they were also obtaining ‘hydrocarbon and ketonic products derived from norbornadiene, acetylene and carbon monoxide’.1,2 In 1973, the group reported the preparation and characterisation of a number of ketonic products obtained from the reaction between variously substituted hexacarbonyl dicobalt complexes and norbornadiene.3 They were able to deduce that the products contained the original acetylene, one a

Reaxa Ltd., Hexagon Tower, Blackley, Manchester, UK

Organomet. Chem., 2010, 36, 93–120 | 93 c

The Royal Society of Chemistry 2010

Fig. 1 General structure of the ketonic products.

carbonyl group and the alkene and, from the NMR spectra of the products, that the general structure was a substituted cyclopentenone ring (Fig. 1). It was also discovered that the exo-isomer was formed exclusively in the reaction (except in cases involving norbornadiene where the formed exoisomer underwent irreversible isomerisation to the endo-isomer, catalysed by dicobalt octacarbonyl) and noted ‘when the products are derived from unsymmetrical acetylenes (MeCCH or PhCCH) only a single position isomer is isolated. . .it is readily shown that the single substituent always occurs on the carbon atom adjacent to the carbonyl group’.3

2.1

Development of the Pauson-Khand reaction

The formal [2 þ 2 þ 1] cycloaddition between an alkyne, an alkene and carbon monoxide has become commonly known as the Pauson-Khand (PK) reaction and has undergone extensive investigation since its initial discovery.4–7 Recent improvements in the reaction conditions and an increase in substrate scope has led to the reaction becoming an important method for the preparation of cyclopentenones. This section will brieﬂy summarise some of the major developments from the last 30 þ years, including the discovery of the intramolecular variant of the reaction, the progress made towards making the process catalytic and examples of how the reaction has been utilised. 2.1.1 The intramolecular Pauson-Khand reaction. It was not until 10 years after the ﬁrst reports of the intermolecular PK reaction that Schore and co-workers published their ﬁndings on an intramolecular variant of the reaction.8 It was found that by applying the same reaction conditions as for the intermolecular reaction, they were able to obtain moderate yields of bicyclic products (Scheme 2). Starting from hept-1-en-6-yne, they were able to generate the 5,5-ringed bicyclo[3.3.0]oct-1-en-3-one (4). Increasing the carbon tether by one carbon unit and using oct-1-en-7-yne led to the generation

Scheme 2 Intramolecular PK reaction.

94 | Organomet. Chem., 2010, 36, 93–120

of the 6,5-ringed bicycle[4.3.0]non-1(9)-en-8-one (5). They note that all attempts to make the 4,5-ring system proved unsuccessful. Unlike the intermolecular reaction, there is only one possible regioisomeric product that results from the intramolecular PK reaction. Although discovered later than the intermolecular variant, greater progress has been made in the intramolecular reaction towards improving reactivity and making the reaction catalytic and stereospeciﬁc.7 2.1.2 Increasing reactivity in the Pauson-Khand reaction. The PK reaction originally suﬀered from a lack of substrate scope and low reaction yields which prevented it from being widely employed. The discovery of new reaction conditions (additives and modiﬁed methods) led to an improvement in yields and reaction times, allowing the scope of the reaction to be expanded. Smit and co-workers reported one of the ﬁrst breakthroughs in improving the reaction by employing dry state adsorption conditions (DSAC) in an intramolecular variant of the reaction.9 It was demonstrated that by adsorbing the cobalt-alkyne complex onto silica-gel, the reaction could be performed in the absence of solvent and at lower reaction temperatures (and times) to aﬀord cyclopentenone products (6-7 - Scheme 3).

Scheme 3 Employment of DSAC in the intramolecular Pauson-Khand reaction.

One of the most commonly employed methods to promote the reaction is to add an amine N-oxide, such as triethylamine N-oxide or N-methylmorpholine N-oxide.10–12 These N-oxides act by oxidising a CO ligand to CO2 and creating a vacant site into which the alkene is able to bind. The loss of CO from the hexacarbonyl dicobalt fragment is a key step in the mechanism of the reaction (see section 2.2); the majority of additives are employed to aﬀect this stage of the pathway. Other chemical additives include Lewis bases such as n-butyl methyl sulﬁde and cyclohexylamine, while molecular sieves have also been employed in order to trap CO out of the reaction mixture.13–16 Evans and co-workers were able to observe a signiﬁcant rate enhancement in the formation of 8 when the reactions were heated using microwave irradiation rather than conventional heating methods (Scheme 4).17,18 Groth and co-workers independently reported rate enhancement when using microwaves in the reaction.19 They also discovered that they were able to use a sub-stoichiometric quantity of Co2(CO)8 when cyclohexylamine was employed as an additive. Organomet. Chem., 2010, 36, 93–120 | 95

Thermal Microwave

Toluene, 110 °C, 16 h CH2Cl2, 5 eq. NMO, 25 °C, 16 h Toluene, 90 °C, 20 mins DCE, 90 °C, 20 mins

70 % 73 % 72 % 89 %

Scheme 4 The employment of microwave heating in PK reactions.

The use of solid supported reagants in the PK reaction has also been investigated, though results have shown that the method is not always beneﬁcial in comparison to traditional homogeneous conditions.20–22 2.1.3 Substrate scope. The PK reaction typically involves an alkyne and an alkene, and most simple alkynes and alkenes are able to undergo the reaction. For alkenes, strained molecules possess the highest reactivity (see section 2.3). One method for increasing the range of alkenes suitable for the PK reaction is to use so called ‘traceless tethers’. By connecting the alkene to the alkyne via a group that can easily be removed, the reaction becomes an intramolecular reaction, meaning there is no regioselectivity issue and the range of suitable alkenes is increased. If the conditions are correct, it is possible that the PK reaction and tether removal can occur in a one-pot domino-like process. An example of this was reported by Pagenkopf and coworkers in 2002 who obtained a monocyclic cyclopentenone in one step utilising a silicon-containing tether (9-11 via 10 - Scheme 5).23 The use of silicon based tethers has been widely investigated.24

Scheme 5 Use of a silicon tether in the PK reaction.

Itami, Yoshida and co-workers demonstrated that this methodology can also be employed in the intermolecular variant of the PK reaction.25 They showed that using a dimethyl(2-pyridyl)silyl group as a directing group allows the reaction to proceed catalytically with respect to the metal complex and gives some control over the regioselective outcome of the reaction. 96 | Organomet. Chem., 2010, 36, 93–120

Alkenes that contain electron-withdrawing groups are often poor substrates for the PK reaction as a hydrogen migration pathway, leading to 1,3-diene formation, competes with the CO insertion step (that leads to the formation of the cyclopentenone). Pauson and co-workers showed how the use of styrene (13) as alkene led to the formation of both the cyclopentenone (14) and the 1,3-diene (15) in a 3:1 ratio in preference of the diene (Scheme 6).26

Scheme 6

Competing 1,3-diene formation when electron-poor alkenes are employed.

It is possible to use other carbon-carbon double bond containing functional groups. In 1985, Pauson reported the ﬁrst example of the use of an allene in the reaction and noted that it reacted ‘readily’, although the structure of the resulting product could not be conﬁrmed.27 The ﬁrst intramolecular example involving an allene was reported in 1994 by Shibata and co-worker using an iron carbonyl catalyst.28 Around the same time, Brummond and co-workers reported the same process on similar substrates using Mo(CO)6 as catalyst.29 The examples given show how the outcome of the reaction is dependant on the conditions employed. Two diﬀerent products (16 and 17) can be formed as coordination to either double bond of the allene is possible (Scheme 7).

Scheme 7 Two possible products of intramolecular allenyne PK-type reaction – (i) product of reaction of the internal p-bond (16); (ii) product of reaction of external p-bond (17).

The use of allene substrates in the intermolecular reaction has also been reported.30–34 1,3-Dienes are suitable substrates for both the inter- and intramolecular reactions, although care is needed in the choice of reaction conditions so as to avoid unwanted side reactions.35 Wender and co-workers reported the reaction between 1,4-dimethoxybut-2-yne (18) and 2,3-dimethylbuta-1,3diene (19) and observed a large temperature dependence on the outcome of the reaction (Scheme 8).36 Organomet. Chem., 2010, 36, 93–120 | 97

Scheme 8 Eﬀect of temperature on outcome of reaction with a 1,3-diene.

By lowering the reaction temperature, the [2 þ 2 þ 1] reaction became favoured (20) over the unwanted [4 þ 2] reactions (21 and 22). In the intramolecular variant of the PK reaction, similar side reactions have been avoided by reducing reaction temperatures. In 2006, the Wender group reported an example of an intramolecular PK-type reaction between an allene and a diene (23).37 The unwanted [4 þ 2] cycloadduct (25) could again be avoided if the reaction was performed at room temperature (Scheme 9).

Scheme 9 Intramolecular PK-type reaction between an allene and a tethered 1,3-diene.

2.1.4 Stereoselectivity in the Pauson-Khand reaction. For the PK reaction to be useful in synthetic chemistry, it is important that the stereochemical outcome of the reaction can be controlled. Four diﬀerent approaches have been employed in attempts to control the stereoselectivity of the PK reaction, namely the use of chiral precursors, chiral auxiliaries, chiral promoters and chiral metal complexes. The use of chiral precursors involves starting with chiral substrates, usually prepared from amino acids or carbohydrates, with the chirality transferred to the ﬁnal product. In 1998, Marco-Contelles and co-workers showed that the carbohydrate derived enol ether 26 could be used to give enyne 27, which was then subsequently reacted in an intramolecular PK reaction to aﬀord the tricyclic unit 28 which was central to their target molecule (Scheme 10).38 98 | Organomet. Chem., 2010, 36, 93–120

Scheme 10 A stereoselective PK reaction using a chiral precursor.

The employment of chiral precursors is a useful method for introducing stereochemistry provided the synthesis of the substrates is straightforward from easily accessible chiral starting materials. A more general method is required though. Chiral auxiliaries can be attached to either the alkene or the alkyne in order to inﬂuence the reaction. They should be easily removed following the reaction to aﬀord the stereochemically pure product. In 1997, Moyano, Pericas and co-workers used a phenylpropynoyl derivative of Oppolzer’s bornane-10,2-sultam to bring about complete diastereoselectivity in the intermolecular PK reaction (29-30 - Scheme 11).39

Scheme 11 Use of Oppolzer’s bornane-10,2-sultam as a chiral auxiliary.

Quantitative removal of the auxiliary with LiAlH4 aﬀorded an enantiomerically pure cyclopentenone product. Auxiliaries of this type are thought to act by chelating to a cobalt atom, forcing a diastereoselective coordination of the alkene to control the conﬁguration of the ﬁnal cyclopentenone product. Another approach involves the use of a chiral additive in the reaction. This method is appealing as it allows the use of two achiral starting materials in the formation of one optically pure product. In 1995, Kerr and coworkers reported the use of a chiral N-oxide, brucine N-oxide (31), leading to enantiomerically enriched products (32-33 - Scheme 12).40 Two possible explanations for the observed enantioselectivity were suggested. It was proposed that the N-oxide is able to selectively decarbonylate the hexacarbonyl dicobalt complex, or alternatively (or additionally), that the amine (brucine) resulting from the attack of the N-oxide is able to stabilise the coordinatively unsaturated cobalt atom that results from the loss of a CO ligand, presumably through coordination to the vacant site. In Organomet. Chem., 2010, 36, 93–120 | 99

Scheme 12 The use of brucine N-oxide as a chiral additive.

2000, the same group reported results showing that if 31 is stirred with a hexacarbonyl dicobalt complex in the presence of triphenylphosphine or trimethylphosphite, a desymmetrised complex is formed where one of the CO ligands is exchanged for a phosphorous-containing ligand.41 If this complex is then utilised in the PK reaction with an achiral N-oxide, enantiomeric excesses are still observed, but the opposite enantiomer is favoured to when 31 is present in the reaction mixture. In 1998, Christie and co-workers reported the ﬁrst example where stereoselectivity was induced by a chiral metal complex.42 By making diastereomerically pure heterobimetallic ‘‘Mo-Co alkyne’’ complexes possessing a menthyl group (35 and 36), they were able to obtain diastereomeric excesses of W99% (determined by 13C NMR spectroscopy) of the resulting cyclopentenone (37 and 38 - Scheme 13). In the intermolecular PK reaction, if an alkene is used that can lead to the formation of two facial-stereoisomers, the exo-isomeric product is usually always favoured. In 2002, Moyano and co-workers reported that they were able to obtain reversal in the facial-stereoselectivity of the intermolecular PK reaction if they used Mo-Co complexes derived from N-(2-alkynoyl) oxazolidinones or sultams (39-40 - Scheme 14).43 For the ﬁrst time, the endo-stereoisomer was obtained as the major product of the reaction (40). The reaction was also regioselective, placing the oxazolidinone group in the a-position of the cyclopentenone. They were able to explain these observations in terms of steric eﬀects. The presence of the oxazolidinone, or sultam, forces the Z5-Cp ligand to sit in a pseudo-equatorial position. This forces the norbornadiene to coordinate from the endo-face as coordination from the exo-face leads to steric repulsion between its methylene bridge and the Z5-Cp ligand. The [Mo(Z5-Cp)(CO)2]2 complex was also tested, using the same reaction conditions, but no cyclopentenone products were obtained, suggesting that alkene coordination occurs at cobalt. An alternative method for using chiral metal complexes is to substitute one or more of the ligands present on the metal with a chiral ligand. The use of chiral phosphine ligands in the reaction, such as BINAP, has been well investigated.44–46 In 2000, Pericas and co-workers showed that by using bidentate phoshine/nitrogen or phosphine/sulfur ligands, they were able to obtain enantiomeric excesses of up to 99% (41-8 - Scheme 15).47,48 100 | Organomet. Chem., 2010, 36, 93–120

Scheme 13 Use of a chiral metal complex to give diastereomerically pure cyclopentenones.

Scheme 14 Preferential formation of the endo-isomer from an N-(2-alkynoyl)oxazolidinone substituted Mo-Co complex.

Scheme 15 Use of PuPHOS in a chiral metal complex.

Organomet. Chem., 2010, 36, 93–120 | 101

It is proposed that the observed stereoselectivity seen in the reactions involving this bidentate ligand is a result of the diﬀerence in bond strength between Co–P and Co–S. The more labile Co–S bond is able to dissociate, leaving a free coordination site on one of the cobalt atoms onto which the alkene is able to coordinate, leading to a stereoselective reaction. 2.1.5 Development of the catalytic Pauson-Khand reaction. One disadvantage of the PK reaction, as ﬁrst reported, was its need for stoichiometric amounts of dicobalt octacarbonyl. In order to make the reaction more attractive, both synthetically and environmentally, eﬀorts were made to reduce the quantities of the transition metal species, although this often led to the requirement for high pressures of CO in order to obtain respectable yields. This section will brieﬂy summarise advances made towards catalytic variants of the PK reaction, including the development of cobalt as a catalyst, the use of other transition metals and the generation of CO in situ.49,50 The initial catalytic version of the PK reaction actually appeared in 1973, in the ﬁrst paper detailing the preparation of cyclopentenones by Pauson and co-workers.3 By stirring a solution of dicobalt octacarbonyl (10 mol % with respect to the alkene) and norbornene (or norbornadiene), ﬁrst under a pressure of acetylene and then under a pressure of mixed acetylene/CO (1:1), it was possible to obtain a good yield of the desired cyclopentenone products (42 and 8 - Scheme 16). It was possible to lower the loading of dicobalt octacarbonyl to 2.5 mol % and still obtain a 62% yield of the cyclopentenone. When norbornadiene was employed as the alkene, a lower yield of the desired cyclopentenone was obtained but this was due to the product reacting further under the conditions employed. In 1990, Rautenstrauch and co-workers reported the catalytic reaction of 1-heptyne (43) with ethene (Scheme 17).51 They were able to use just 0.22 mol % of dicobalt octacarbonyl (with respect to the alkyne) in the reaction and obtained a moderate yield of the desired cyclopentenone (44).

Scheme 16 The ﬁrst examples of a catalytic PK reaction

102 | Organomet. Chem., 2010, 36, 93–120

Unfortunately, a reaction temperature of 150 1C and high pressures of both CO and ethene were required leading to overall pressures of between 310 and 360 atmospheres. Livinghouse and co-workers were successfully able to employ 5 mol% dicobalt octacarbonyl and CO at atmospheric pressure by irradiating the reaction mixture with high-intensity visible light.52 Later, they conﬁrmed that it was possible to perform the reaction in the absence of the irradiation,

Scheme 17 Catalytic use of Co2(CO)8 in the formation of 2-pentylcyclopent-2-en-1-one

but that the reaction was very sensitive to temperature.53 In both cases, the purity of the dicobalt octacarbonyl was shown to be critical in order to observe eﬃcient catalysis. The high toxicity and ﬂammability of pure dicobalt octacarbonyl, required for optimum catalytic ability, make it a less then ideal catalyst for the PK reaction. Following the ﬁndings by Jeong and co-workers that triphenylphosphite could be successfully used as an additive in the catalytic PK reaction, Gibson and co-workers have shown that substituting a carbonyl ligand in dicobalt octacarbonyl with a phosphine or phosphite ligand forms a stable metal complex capable of catalysis under atmospheric pressure of CO.54–56 It was thought that the formation of inactive cobalt clusters such as Co4(CO)12, formed by dimerisation of the remaining cobalt carbonyl species after release of the cyclopentenone product, were responsible for the shutdown of the catalytic cycle when dicobalt octacarbonyl was employed.51 Kraﬀt and co-workers were able to show that Co4(CO)12 can actually be exploited as a catalytic species in the PK reaction and were able to obtain excellent yields if cyclohexylamine was introduced as an additive alongside the metal cluster.57,58 The use of metal clusters as catalysts for the reaction has been extended to involve mixed metal clusters.59 There has been signiﬁcant interest in substituting cobalt for other transition metals giving rise to what are known as Pauson-Khand type reactions.60 In 1996, Buchwald and co-workers showed that a titaniumcyclopentadienyl complex (45) could be used sub-substoichiometrically in the reaction (46-47 - Scheme 18).61,62 Since this discovery, a whole range of transition metals have been exploited, with ruthenium (Ru3(CO)12), rhodium ([RhCl(CO)2]2), iridium ([Ir(COD)Cl]2), molybdenum (Mo(CO)6) and iron (Fe(CO)4(NMe3)) all having been successfully employed as catalysts for the reaction.28,29,36,63–65 One of the major practical issues associated with the PK reaction is the use of carbon monoxide. In 2002, two groups independently published details of intramolecular PK-type reactions of enynes and alkenes where CO Organomet. Chem., 2010, 36, 93–120 | 103

Scheme 18 Titanium catalysed PK-type reaction.

was generated in situ. If the correct transition metal species was employed, it was possible to decarbonylate an additive, such as an aldehyde or ketone, and carry out the PK-type reaction with the same metal complex in one pot. Rhodium was found to be the best metal for the process, with low yields obtained when using iridium and ruthenium based catalysts. Morimoto and co-workers used pentaﬂuorobenzaldehyde as their source of CO while Shibata and co-workers were able to perform the reaction in solvent free conditions and found that cinnamaldehyde gave the best results.66,67 The two additives gave very similar yields when used on comparable substrates. In Morimotos and Shibatas examples, pentaﬂuorobenzene and styrene are generated, respectively, as unwanted by-products. Chung was able to make the reaction more atom economic by using a,b-unsaturated aldehydes as a source of both CO and alkene (48 þ 49-50 - Scheme 19).

Scheme 19 PK-type reaction with CO generated in situ.

As well as the atom economy of using a,b-unsaturated aldehydes, the cobalt/rhodium heterobimetallic nanoparticles used as catalyst had the advantage of being recoverable from the reaction mixture. On re-employment in the reaction, no loss in their catalytic activity could be observed.68 It should be noted that the development of catalytic systems has been most successful in the intramolecular reaction, with limited progress being made with the intermolecular variant. 2.1.6 Applications of the Pauson-Khand reaction. The cyclopentenone moiety (and derivatives thereof) are abundant in nature and the PK reaction has led to improved synthetic pathways to a number of natural products and compounds exerting interesting biological eﬀects. Synthetic chemists have been attracted to the reaction because it has the potential to create highly substituted cyclopentenone rings in one step, potentially reducing the number of steps required to access complex cyclopentenone frameworks. 104 | Organomet. Chem., 2010, 36, 93–120

Prostaglandins are unsaturated carboxylic acids made up of a 20-carbon skeleton and contain three key structural elements – a seven membered carbon chain, an eight membered carbon chain and a cyclopentane ring. Cyclopentenone prostaglandins have been shown to exhibit greater biological activity then those that are a,b-saturated and, because of this, there is much interest in their synthesis. An example of a prostaglandin synthesised utilising the PK reaction was reported by Brummond and coworkers in 2004.63 They employed an allenic PK-type reaction (51-52) to install the cyclopentenone as the key step in the total synthesis of 15-deoxyD12,14-prostaglandin J2 (53 - Scheme 20).

Scheme 20 Key step in synthesis of 15-deoxy-D12,14-prostaglandin J2.

Steroids are naturally occurring terpenoids that are based on a system consisting of four fused rings. In 2006, Chung and co-workers published the synthesis of the basic steroidal skeleton (54-56) and employed an intramolecular PK reaction as the key step (55-56 - Scheme 21).69

Scheme 21 Synthesis of a steroidal framework.

The PK reaction allows access to complex molecular frameworks. In 2007, Mukai and co-worker reported the synthesis of three related alkaloids by using the PK reaction twice in the same molecule – initially to construct a bicyclo[4.3.0]nonane carbon framework (57-58) and then a second time to fuse a bicyclo[3.3.0]octane skeleton onto the initial framework (59-60 Scheme 22).70 Organomet. Chem., 2010, 36, 93–120 | 105

Scheme 22 Total synthesis of the alkaloids magellaninone (61), paniculatine (62) and magellanine (not shown).

( )-Asteriscanolide (65) is a cyclooctane sesquiterpene lactone that has been synthesised by a number of groups who have employed a variety of diﬀerent strategies. In 2001, Kraﬀt and co-workers developed a strategy that utilised the intermolecular PK reaction as a key step in its synthesis (63-64 - Scheme 23).71

Scheme 23 Cyclopentenone formation in the total synthesis of ( )-asteriscanolide.

The reaction is frequently employed in the synthesis of complex molecules and these are just a few examples demonstrating how both the intermolecular and intramolecular PK reactions can be utilised.72–78 2.2

Proposed mechanism of the Pauson-Khand reaction

In 1985, 14 years after the ﬁrst reports of the PK reaction, the currently accepted mechanism for the reaction was proposed by Magnus (Scheme 24).79,80 106 | Organomet. Chem., 2010, 36, 93–120

Scheme 24 Magnus’s postulated mechanism for the Pauson-Khand reaction.

The ﬁrst step in the mechanism is the coordination of the alkyne, across the triple bond, with dicobalt octacarbonyl (I). A CO ligand dissociates from one of the cobalt atoms in the resulting m2-hexacarbonyl dicobalt complex (II) creating a vacant coordination site into which the alkene is able to coordinate (III). The alkene then inserts into a Co–C bond to form a cobaltacycle (IV). This is the key step in terms of determining the regiochemical and stereochemical outcome of the reaction. A CO ligand then coordinates and inserts into the cobaltacycle (V) and reductive elimination and decomplexation (through steps VI and VII) from this intermediate gives the cyclopentenone. The ﬁrst DFT studies performed on the PK reaction pathway were published in 2001 by Nakamura and co-workers.81 They used ethyne and ethene as substrates and conﬁrmed that the alkene insertion step is the critical stereo- and regiochemical determining step in the mechanism and that it is reversible. They also showed that although the coordination and insertion steps take place on one cobalt atom, the other cobalt atom exerts an electronic inﬂuence through the Co–Co bond by donating and excepting electron density. Greene and co-workers have proposed that the formation of the cobaltacycle may not be reversible. In 2005 they showed, using DFT calculations, that the accelerating eﬀect seen on the reaction by the addition of Lewis bases is in fact due to the lowering of the activation barrier to alkene insertion, making cobaltacycle formation irreversible rather then facilitating the loss of a CO ligand from the hexacarbonyl complex.82 Although the Magnus mechanism is widely accepted, there is a lack of evidence for many of the proposed intermediates. Until recently, the only Organomet. Chem., 2010, 36, 93–120 | 107

proposed intermediates that had been isolated as stable complexes and fully characterised were the alkyne-hexacarbonyl dicobalt species.83 In 2007, a second intermediate was identiﬁed and characterised when Evans, McGlinchey and co-workers were able to isolate a stable alkene-pentacarbonyldicobalt-alkyne species (68 - Scheme 25).84

Scheme 25 Formation of a stable alkene-pentacarbonyldicobalt-alkyne complex.

The novel cobalt complex came about as a result of the intramolecular coordination of a double bond, present in one of the R groups on the acetylene, to one of the cobalt atoms – taking the place of a CO ligand (6768). They found that the new pentacarbonyl complex could be readily formed in CDCl3 at room temperature from the hexacarbonyl dicobalt precursor. Attempts to use the pentacarbonyl complex as a substrate in the PK reaction led to no formation of cyclopentenone product. It was proposed that this is due to the alkene occupying a pseudo-equatorial site – alkene insertion is thought to occur from the axial position (see section 2.4). In 2008, Fox and co-workers were able to isolate and fully characterise 2 cobalt-complexes of the proposed alkene-insertion intermediates, one of which (72) is shown below (Scheme 26).85

Scheme 26 Isolation of an alkene-insertion Pauson Khand intermediate.

The complexes are binuclear, contain a bridging carbonyl and are coordinated by m-bonded, ﬁve-carbon ‘ﬂyover’ carbene ligands. Their formation has been explained as being the result of the fragmentation of the cyclopropane ring after alkene insertion, providing indirect evidence for the proposed Magnus mechanism. Whilst studying the eﬀects of additives in the PK reaction, Kraﬀt and coworkers were able to isolate an internally stabilised pentacarbonyl intermediate (75 - Scheme 27).86 108 | Organomet. Chem., 2010, 36, 93–120

Scheme 27 Intramolecular stabilisation of a pentacarbonyl intermediate by a sulfur atom.

The expected hexacarbonyl dicobalt complex was formed when the alkyne and dicobalt octacarbonyl were stirred in CH2Cl2 (73-74). If the same reaction was performed in the presence of ten equivalents of NMO, the novel pentacarbonyl complex could be isolated (73-75). On dissociation of the CO ligand, accelerated by the addition of NMO, the sulfur atom is able to coordinate to the vacant site creating a stable complex. Treatment of the complex with CO resulted in regeneration of the hexacarbonyl complex and, unlike the isolated alkene-pentacarbonyldicobaltalkyne species 68, heating of the pentacarbonyl complex 75 led to the formation of the desired cyclopentenone. A similar sulfur stabilised intermediate was isolated by Perica´s and co-workers while investigating tethered chiral auxiliaries in the PK reaction.87,88 Other pentacarbonyl intermediates have been identiﬁed spectroscopically. Gordon and co-workers studied phenylacetylene hexacarbonyl dicobalt photochemically.89 By irradiating the complexes at 254 nm at 12 K in various matrices, they were able to observe a pentacarbonyl intermediate by IR spectroscopy. Irradiation in an argon matrix led to the observation of the unsaturated pentacarbonyl complex (76-77). Irradiation in mixed argon/nitrogen or in pure nitrogen matrices allowed dinitrogen complexes to be observed (76-78 - Scheme 28).

Scheme 28 The photolysis of phenylacetylene hexacarbonyl dicobalt in various matrices

They tentatively assigned the CO as being lost from the axial position after comparison with other pentacarbonyl analogues. This stands in Organomet. Chem., 2010, 36, 93–120 | 109

contrast to the calculations reported by Gimbert, Greene and co-workers who have shown, using DFT calculations, that CO is lost from the pseudoequatorial position under thermal conditions (see section 2.4).90,91 Electrospray ionisation mass spectrometry studies by Gimbert and coworkers have shown that loss of CO from the hexacarbonyl dicobalt complex does occur prior to alkene coordination and insertion.92 They used a dppm substituted complex (79), previously shown to exhibit some activity in the PK reaction, as the substrate in the investigation (Fig. 2).93

Fig. 2

The dppm complex used by Gimbert to investigate alkene coordination.

The energy pathway for the PK reaction with acetylene as the alkyne and ethylene as the alkene has been calculated and is shown below (Fig. 3).12

Fig. 3 Calculated energy pathway for ligand exchange and cobaltacycle formation in Magnus’s proposed mechanism. Values in kcal mol 1.12

The calculations by Perica`s and co-workers show that the most energetically demanding step, and hence the rate determining step, is the loss of CO from the hexacarbonyl dicobalt complex. Once this loss has occurred, the remaining steps of the pathway are much less energy demanding, though it should be noted that the insertion of the alkene has a moderate activation energy and is often problematic, meaning that dissociation of CO from the complex does not necessarily lead to cyclopentenone formation. The mechanism of PK-type reactions involving metal sources other then Co2(CO)8 has not been the subject of as much theoretical investigation, though DFT studies have been carried out on Ru3(CO)12 and [Rh(CO)2Cl]2 110 | Organomet. Chem., 2010, 36, 93–120

catalysed PK-type reactions.94,95 Instead of the traditionally accepted alkene-alkyne coupling mechanism, Wu and co-workers suggest that the Ru3(CO)12 catalysed intramolcular PK-type reaction proceeds through an alkyne-CO insertion pathway which involves coupling between a CO molecule and the acetylene to form a ruthenacyclobutenone, with binding of the alkene and its insertion, followed by reductive elimination, leading to the cyclopentenone product. 2.3

Importance of the alkene in the Pauson-Khand reaction

The range of alkenes which can be used in the intermolecular PK reaction is limited and there is a large variation in reaction rate between alkenes. It is found that strained alkenes (cyclohexeneocyclopenteneonorbornene) are the best substrates for the reaction and there has been extensive investigation into the reasons behind this. Perica`s and co-workers calculated the exothermicity of the cobaltacycle formation step and were able to correlate this with alkene strain (Table 1).12

Table 1

Calculated energy for cobaltacycle formation with varying alkenes

Alkene

DH1r (Kcal/mol)

Cyclohexene Cyclopentene Norbornene Cyclobutene Cyclopropene

21.6 21.7 27.8 31.3 47.3

The explanation given for these results was that between coordination and insertion into the cobaltacycle, most of the energy associated with the strain is released and it is this liberation of energy that drives the reaction. The calculations showed that cyclopropene would be an ideal alkene and they proceeded to obtain good to excellent yields of bicyclo[3.1.0]hex-3-en2-one adducts from a range of alkyne substrates at low temperatures ( 35 1C). Milet, Gimbert and co-workers used DFT calculations to study the formation of the cobaltacycle and calculated activation energies and energies for the orbitals involved.96 From their calculations, they were able to deduce that the p*-orbital (LUMO) of the alkene is very important in the PK reaction. On coordination of the alkene to the (m2-alkyne)Co2(CO)5 fragment, possessing a vacant coordination site, the p*-orbital of the alkene is able to accept electron density from the dp-orbital (HOMO) of the Co2(CO)5 fragment. This orbital is also involved in the formation of the new C–C bond (Fig. 4). Their calculations show that the LUMO energy level of the free alkene and the HOMO-LUMO energy gaps mirror the observed reactivity in the PK reaction. In a comparison between cyclohexene, cyclopentene and norbornene, cyclohexene has the largest HOMO-LUMO gap, the highest Organomet. Chem., 2010, 36, 93–120 | 111

Fig. 4 Bonding between the Co2(CO)5-acetylene fragment and the alkene.

LUMO energy and is the least reactive substrate while norbornene has the smallest HOMO-LUMO gap, the lowest LUMO energy and is the most reactive. They note that the LUMO energy level of an alkene is a useful approximation of its reactivity. They also note a correlation between a smaller CQC–C bond angle and a lowering of the LUMO energy level – a link with Perica`s’ theory of strain driving the reaction.12 It was also possible to calculate the number of electrons involved in back donation from the metal to the alkene and it was found that this correlates with the barrier to cobaltacycle formation - the greater the back donation, the lower the barrier to cobaltacycle formation and therefore the higher the reactivity of the alkene. Not all alkenes able to act as p-acceptors are good substrates though, suggesting that a subtle combination of donor and acceptor properties, alongside a release of ring strain, must combine for an alkene to function eﬀectively in the PK reaction. The electrons in the p-orbital (HOMO) of the alkene are involved in the formation of the new Co-Calkyl bond, but the importance of this orbital on the outcome of the reaction was not discussed. 2.4

Explaining the regioselectivity in the Pauson-Khand reaction

While the intramolecular PK reaction can lead to the formation of only one regioisomer (see section 2.1.1), the intermolecular reaction (involving a symmetrical alkene) can lead to two possible regioisomeric products (Scheme 29).

Scheme 29 Possible regioisomeric products of intermolecular PK reaction.

In the majority of examples of intermolecular PK reactions, the bulkiest of the acetylenic substituents favours being situated a to the carbonyl in the ﬁnal cyclopentenone product. This was typically explained as being the 112 | Organomet. Chem., 2010, 36, 93–120

result of steric interactions in the alkene insertion step of the reaction cycle, with the alkene positioning itself adjacent to the least bulky of the substituents to avoid any steric clash (Scheme 30).

Scheme 30 Steric interactions when the alkene coordinates to the cobalt-acetylene complex directs insertion into least hindered Co–C bond.

There are examples in the literature, however, that cannot be explained by steric eﬀects alone. Kraﬀt and co-workers presented contrasting results from two alkyne substrates that can be regarded as being sterically similar (Scheme 31).97,98 When ethyl propiolate (80) was employed in the PK reaction, the major regioisomeric product contained the carbethoxy group a to the carbonyl (82). When ethyl butynoate was employed (81), the major regioisomeric product contained the carbethoxy group b to the carbonyl (83). This result was unexpected as in the other substrates investigated, it was found that the major product always contained the sterically bulkier substituent in the a-position.

Scheme 31 Contrasting results observed with ethyl propiolate (80-82) and ethyl butynoate (81-83) as substrates in the PK reaction (norbornene used as alkene).

To explain this observation, it was suggested that electron density differences at the acetylenic carbons can combine with steric eﬀects to determine the regiochemical outcome of the reaction. Kraﬀt concludes ‘electronic eﬀects seem to play a contributing role in determining the regiochemical outcome of the Pauson-Khand reaction, but, steric inﬂuences over the transition state apparently exert a more powerful directing eﬀect.’ In 2001, Gimbert, Greene and co-workers investigated, using DFT calculations, whether it is possible that electronic eﬀects resulting from the substitution of the acetylene could aﬀect the position from which CO is lost and, in turn, determine the regiochemistry of the reaction.90,99 It was possible to calculate the atomic charges of the acetylenic carbons CR and CR 0 in Organomet. Chem., 2010, 36, 93–120 | 113

Fig. 5 The complexes analysed by DFT calculations and the labelling of the CO ligands in the complexes (cis- and trans- are labelled relative to R’)

three dicobalt complexes (Fig. 5). They were able to demonstrate that different substituents aﬀected the electron density present on each of the carbon atoms which, in turn, aﬀected the polarisation of the acetylenic C–C bond. From these calculations they were able to rationalise the observed regiochemistry of the reaction in terms of the bonding within the metalacetylene complex. The bonding between the bent acetylene and Co2(CO)6 fragments is made up primarily of back bonding to the acetylene – the interaction between the empty p*-orbitals of the acetylene and the bonding and antibonding combination of the dx2-y2 orbitals of cobalt.100–102 These d-orbitals are also involved in bonding to the equatorial CO ligands, meaning back-donation to the acetylene will increase Co-COpseq bond distances and the lability of the Co-COpseq bonds. This could be seen clearly if the bond dissociation energies of the pseudoaxial (psax) and cis- and trans-pseudoequatorial (pseqcis and pseqtrans) CO bonds were analysed. In each of the complexes 84–86, the Co-COpseq bond dissociation energies were calculated to be 10 kcal/mol lower then the dissociation energies for Co-COpsax bonds, showing that the dissociation of an equatorial CO will always be favoured. They proposed that the p-accepting CO ligand trans-equatorial to the acetylene carbon containing the greatest electron density will be expected to accept the most electron density and become more tightly bound, leaving the CO ligand cis-equatorial to this as the most labile CO (as observed in the calculated bond dissociation energies). If the reacting alkene now displaces the CO from this position, the substituent on the more electron deﬁcient acetylene carbon will ﬁnish a to the carbonyl in the ﬁnal cyclopentenone ring. This proposal mirrored the results obtained from the PK reactions of 84 and 86, while the lack of polarisation observed in the C–C bond of 85 meant that the outcome of the reaction was led by the steric nature of the two substituents. In order to test their hypothesis, they performed a PK reaction with a substrate bearing groups that were sterically similar but electronically diﬀerent (87 - Scheme 32). They observed, as calculated, the exclusive formation of the cyclopentenone containing the tolyl group in the a-position (88). Gimbert and Greene were able to successfully explain the experimentally observed regioselectivity as a combination of both steric and electronic factors. They conclude that electronic diﬀerences in the acetylenic substituents can lead to the discriminant loss of a CO ligand, which in turn controls the regiochemistry of the reaction. Following these ﬁndings, Gleiter and co-workers reported that in a series of S-alkyl-substituted alkynes, the regiochemical outcome of the reaction 114 | Organomet. Chem., 2010, 36, 93–120

Scheme 32 Example of electronic eﬀect on PK reaction when groups are sterically similar.

could not be predicted on the basis of this ‘trans-eﬀect’ alone, suggesting that any electronic eﬀect may be more complex then previously proposed.103 The calculations by Gimbert, Greene and co-workers showed that the pseudoequatorial CO ligands are the most labile and hence the most likely to dissociate from the Co2(CO)6 fragment. Also in 2001, Gimbert, Milet and co-workers showed that the barrier for rotation for a propyneCo2(ethene)(CO)5 complex is relatively low and can be easily overcome at room temperature.91 This in turn means that although the initial CO dissociation and alkene coordination may take place from an equatorial position, the complex is free to rotate and the alkene free to insert from either equatorial position or from the axial position to form the cobaltacycle. The energy levels for cobaltacycle formation were also calculated, starting with the alkene coordinated in either the cis-equatorial, trans-equatorial or axial positions, through four determined transition states to give one of two cobaltacycles (each leading to a diﬀerent regioisomeric product). It was shown that the reaction pathway which passes through the lowest lying transition state for alkene insertion to the most stable cobaltacycle (which is the cobaltacycle that would aﬀord the experimentally observed regioisomer) originates from the complex with the alkene coordinated in the axial position. They conclude that the initial position of alkene coordination does not determine the regiochemistry of the reaction if the complex is able to freely rotate. In 2000, Laschat and co-workers reported their experimental results into the investigation of the regiochemical outcome of the intermolecular PK reaction involving unsymmetrically substituted bridged bicyclic alkenes.104 They observed that when using these sterically bulky alkenes, they were able to alter the regiochemical outcome of the reaction by varying the temperature. In order to explain their results, they proposed a modiﬁcation to the reaction mechanism, suggesting coordination of the alkene occurs at the axial position (Scheme 33). At higher temperatures, and when the alkyne substituent is bulky, the formation of the exo-Re complex (90) is favoured as it removes any steric interaction between the methyl group of the alkene and the R group of the alkyne. In general, the alkene will preferably insert into the Co–CH bond over insertion into the Co–CR bond for steric reasons, but in the case of the exo-Re complex, insertion into the Co–CH bond leads to the methyl group pointing into the complex. In this case, the alkene insertion step is disfavoured. In the exo-Si complex (89), alkene insertion leads to the methyl Organomet. Chem., 2010, 36, 93–120 | 115

Scheme 33 Proposed mechanism involving insertion from the axial position.

group pointing away from the complex and is favoured. At lower temperatures, or when the R-group is small, the alkene insertion step is more important then the initial coordination and so formation of the exo-Si complex is favoured. Laschat points to the work of Greene and co-workers on the synthesis and utilisation of diphosphinoamine complexes as evidence for their proposed mechanism.93 Greene reported the synthesis of a set of dicobalt tetracarbonyl complexes containing P–N–P bidentate ligands and showed, by X-ray crystallographic studies, that the trans-equatorial CO ligands had been replaced by the bidentate ligand. It was also shown that the complexes are active in the PK reaction. The activity of the complexes, together with the location of the bidentate ligand in the complex, led Laschat and co-workers to conclude that the alkene must coordinate at the axial position. In 2006, Gimbert and Milet returned to the examples of propyne, methyl 2-butynoate and methyl propiolate (84-86) and investigated, using DFT calculations, whether the insertion of the alkene was aﬀected by the polarisation of the acetylenic bonds rather then just a preferential loss of a CO ligand.99 They found that the acetylenic carbon carrying the most electron density was that involved in the carbon insertion and subsequent C–C bond formation. They concluded ‘the acetylenic carbon that carries the larger electron density will, in general, be that involved in forming the crucial C–C bond, even when geometrical factors are less favourable.’ Recent ﬁndings from Overgaard, Platts and co-workers have shown that the traditional view of the cobalt-cobalt bond present in the dicobalt complex intermediates may be incorrect and that it is better thought of as a singlet diradical species.105,106 The involvement of radical species would 116 | Organomet. Chem., 2010, 36, 93–120

have signiﬁcant implications for the mechanism of the PK reaction as is currently understood, and the authors note that these ﬁndings should be borne in mind when using theoretical methods to investigate the mechanism of the reaction. 2.5

Summary

The Pauson-Khand reaction has developed considerably since its discovery in 1971. Advances have been made which mean the reaction has become a useful tool for the synthetic chemist, with the range of suitable substrates ever increasing and the improvement in reaction conditions meaning that the reaction can be performed under standard laboratory conditions. Developments in the catalytic variant and improvements in the ability to control the regio- and stereochemical outcomes of the reaction mean it will become more attractive for large-scale synthesis. Considering these extensive developments, the mechanism of the reaction remains debated with only a handful of the proposed intermediates having been observed/characterised. While it was originally thought that the intermolecular reaction was controlled solely by the steric nature of the substituents, it has now been proposed that their electronic nature can also aﬀect the regiochemical outcome of the reaction. The extent to which each of the factors inﬂuences the reaction, and which is dominant, is still open to debate. References 1 I. U. Khand, G. R. Knox, P. L. Pauson and W. E. Watts, Chem. Comm., 1971, 36. 2 P. L. Pauson and I. U. Khand, Ann. N.Y. Acad. Sci., 1977, 295, 2. 3 I. U. Khand, G. R. Knox, P. L. Pauson, W. E. Watts and M. I. Foreman, J. Chem. Soc. Perkin Trans. 1, 1973, 977. 4 A. J. Fletcher and S. D. R. Christie, J. Chem. Soc. Perkin Trans. 1, 2000, 1657. 5 J. Blanco-Urgoiti, L. An˜orbe, L. Pe´rez-Serrano, G. Domı´ nguez and J. Pe´rezCastells, Chem. Soc. Rev, 2004, 33, 32. 6 S. Laschat, A. Becheanu, T. Bell and A. Baro, Synlett, 2005, 17, 2547. 7 S. E. Gibson and N. Mainolﬁ, Angew. Chem. Int. Ed., 2005, 44, 3022. 8 N. E. Schore and M. C. Croudace, J. Org. Chem., 1981, 46, 5436. 9 W. A. Smit, A. S. Gybin, A. S. Shashkov, L. G. Kyzmina, G. S. Mikaelian, R. Caple and E. D. Swanson, Tetrahedron Lett., 1986, 27, 1241. 10 S. Shambayati, W. E. Crowe and S. L. Schreiber, Tetrahedron Lett., 1990, 31, 5289. 11 N. Jeong, Y. K. Chung, B. Y. Lee, S. H. Lee and S.-E. Yoo, Synlett, 1991. 12 M. A. Perica`s, J. Balsells, J. Castro, I. Marchueta, A. Moyano, A. Riera, J. Vazquez and X. Verdaguer, Pure Appl. Chem., 2002, 74, 167. 13 T. Sugihara, M. Yamada, M. Yamaguchi and M. Nishizawa, Synlett, 1999, 771. 14 J. Blanco-Urgoiti, D. Abdi, G. Dominguez and J. Pe´rez-Castells, Tetrahedron, 2008, 64, 67. 15 T. Sugihara, M. Yamada, H. Ban, M. Yamaguchi and C. Kaneko, Angew. Chem. Int. Ed., 1997, 36, 2801. 16 L. Pe´rez-Serrano, L. Casarrubios, G. Domı´ nguez and J. Pe´rez-Castells, Org. Lett., 1999, 1, 1187. Organomet. Chem., 2010, 36, 93–120 | 117

17 M. Iqbal, N. Vyse, J. Dauvergne and P. Evans, Tetrahedron Lett., 2002, 43, 7859. 18 M. Iqbal, Y. Li and P. Evans, Tetrahedron, 2004, 60, 2531. 19 S. Fischer, U. Groth, M. Jung and A. Schneider, Synlett, 2002, 2023. 20 G. L. Bolton, J. C. Hodges and J. R. Rubin, Tetrahedron, 1997, 53, 6611. 21 J. L. Spitzer, M. J. Kurth and N. E. Schore, Tetrahedron, 1997, 53, 6791. 22 A. Stobrawe, P. Makarczyk, C. Maillet, J.-L. Muller and W. Leitner, Angew. Chem. Int. Ed., 2008, 47, 6674. 23 J. F. Reichwein, S. T. Iacono and B. L. Pagenkopf, Tetrahedron, 2002, 58, 3813. 24 A. P. Dobbs, I. J. Miller and S. Martinovic´, Beilstein Journal of Organic Chemistry, 2007, 3, No. 21. 25 K. Itami, K. Mitsudo and J. Yoshida, Angew. Chem. Int. Ed., 2002, 41, 3481. 26 I. U. Khand, P. L. Pauson and M. Habib, J. Chem. Res. Miniprint, 1978, 4434. 27 P. L. Pauson, Tetrahedron, 1985, 41, 5855. 28 K. Narasaka and T. Shibata, Chemistry Letters, 1994, 315. 29 J. L. Kent, H. Wan and K. M. Brummond, Tetrahedron Lett., 1995, 36, 2407. 30 A´. Gonza´lez-Go´mez, L. An˜orbe, A. Poblador, G. Dominguez and J. Pe´rezCastells, Eur. J. Org. Chem., 2008, 1, 1370. 31 F. Antras, M. Ahmar and B. Cazes, Tetrahedron Lett., 2001, 42, 8153. 32 F. Antras, M. Ahmar and B. Cazes, Tetrahedron Lett., 2001, 42, 8157. 33 M. Ahmar, O. Chabanis, J. Gauthier and B. Cazes, Tetrahedron Lett., 1997, 38, 5277. 34 M. Ahmar, F. Antras and B. Cazes, Tetrahedron Lett., 1995, 36, 1995. 35 P. A. Wender, N. M. Deschamps and G. G. Gamber, Angew. Chem. Int. Ed., 2003, 42, 1853. 36 P. A. Wender, N. M. Deschamps and T. J. Williams, Angew. Chem. Int. Ed., 2004, 43, 3076. 37 P. A. Wender, M. P. Croatt and N. M. Deschamps, Angew. Chem. Int. Ed., 2006, 45, 2459. 38 J. Marco-Contelles and J. Ruiz, Tetrahedron Lett., 1998, 39, 6393. 39 S. Fonquerna, A. Moyano, M. A. Perica`s and A. Riera, J. Am. Chem. Soc., 1997, 119, 10225. 40 W. J. Kerr, G. G. Kirk and D. Middlemiss, Synlett, 1995, 1085. 41 D. R. Carbery, W. J. Kerr, D. M. Lindsay, J. S. Scott and S. P. Watson, Tetrahedron Lett., 2000, 41, 3235. 42 D. T. Rutherford and S. D. R. Christie, Tetrahedron Lett., 1998, 39, 9805. 43 R. Rios, M. A. Perica`s, A. Moyano, M. A. Maestro and J. Mahı´ a, Org. Lett., 2002, 4, 1205. 44 K. Hiroi, T. Watanbe, R. Kawagishi and I. Abe, Tetrahedron Lett., 2000, 41, 891. 45 K. Hiroi, T. Watanbe, R. Kawagishi and I. Abe, Tetrahedron: Asymmetry, 2000, 11, 797. 46 S. E. Gibson, K. A. C. Kaufmann, J. A. Loch, J. W. Steed and W. A. J. P., Chem. Eur. J., 2005, 11, 2566. 47 X. Verdaguer, A. Moyano, M. A. Perica`s, A. Riera, M. A. Maestro and J. Mahı´ a, J. Am. Chem. Soc., 2000, 122, 10242. 48 A. Lledo´, J. Sola`, X. Verdaguer, A. Riera and M. A. Maestro, Adv. Synth. Catal., 2007, 349, 2121. 49 S. E. Gibson and A. Stevenazzi, Angew. Chem. Int. Ed., 2003, 42, 1800. 50 T. Shibata, Adv. Synth. Catal., 2006, 348, 2328. 51 V. Rautenstrauch, P. Me´gard, J. Conesa and W. Ku¨ster, Angew. Chem. Int. Ed., 1990, 29, 1413. 52 B. L. Pagenkopf and T. Livinghouse, J. Am. Chem. Soc., 1996, 118, 2285. 118 | Organomet. Chem., 2010, 36, 93–120

53 D. B. Belanger, D. J. R. O’ Mahony and T. Livinghouse, Tetrahedron Lett., 1998, 39, 7637. 54 N. Jeong, S. H. Hwang and Y. Lee, J. Am. Chem. Soc., 1994, 116, 3159. 55 A. C. Comely, S. E. Gibson, A. Stevenazzi and H. N. J, Tetrahedron Lett., 2001, 42, 1183. 56 S. E. Gibson, C. Johnstone and A. Stevenazzi, Tetrahedron, 2002, 58, 4937. 57 M. E. Kraﬀt, L. V. R. Bonaga and C. Hirosawa, J. Org. Chem., 2001, 66, 3004. 58 M. E. Kraﬀt and L. V. R. Bonaga, Angew. Chem. Int. Ed., 2000, 39, 3676. 59 R. Paolillo, V. Gallo, P. Mastrorilli, C. F. Nobile, J. Rose´ and P. Braunstein, Organometallics, 2008, 27, 741. 60 I. Omae, Appl. Organometal. Chem., 2008, 22, 149. 61 F. A. Hicks, N. M. Kablaoui and S. L. Buchwald, J. Am. Chem. Soc., 1996, 118, 9450. 62 F. A. Hicks, N. M. Kablaoui and S. L. Buchwald, J. Am. Chem. Soc., 1999, 121, 5881. 63 K. M. Brummond, P. C. Sill and H. Chen, Org. Lett., 2004, 6, 149. 64 T. Morimoto, N. Chatani, Y. Fukumoto and S. Murai, J. Org. Chem., 1997, 62, 3762. 65 T. Shibata and K. Takagi, J. Am. Chem. Soc., 2000, 122, 9852. 66 T. Morimoto, K. Fuji, K. Tsutsumi and K. Kakiuchi, J. Am. Chem. Soc., 2002, 124, 3806. 67 T. Shibata, N. Toshida and K. Takagi, Org. Lett., 2002, 4, 1619. 68 K. H. Park, I. G. Jung and Y. K. Chung, Org. Lett., 2004, 6, 1183. 69 D. H. Kim, K. Kim and Y. K. Chung, J. Org. Chem., 2006, 71, 8264. 70 T. Kozaka, N. Miyakoshi and C. Mukai, J. Org. Chem., 2007, 72, 10147. 71 M. E. Kraﬀt, Y. Y. Cheung and K. A. Abboud, J. Org. Chem., 2001, 66, 7443. 72 S.-J. Min and S. J. Danishefsky, Angew. Chem. Int. Ed., 2007, 46, 2199. 73 M. K. Palleria and J. M. Fox, Org. Lett., 2007, 9, 5625. 74 T. Honda and K. Kaneda, J. Org. Chem., 2007, 72, 6541. 75 T. Honda and K. Kaneda, Tetrahedron, 2008, 64, 11589. 76 C. E. Madu and C. J. Lovely, Org. Lett., 2007, 9, 4697. 77 K. A. Miller and S. F. Martin, Org. Lett., 2007, 9, 1113. 78 K. A. Miller, C. S. Shanahan and S. F. Martin, Tetrahedron, 2008, 64, 6884. 79 P. Magnus, C. Exon and P. Albaugh-Robertson, Tetrahedron, 1985, 41, 5861. 80 P. Magnus and L. M. Principe, Tetrahedron Lett., 1985, 26, 4851. 81 M. Yamanaka and E. Nakamura, J. Am. Chem. Soc., 2001, 123, 1703. 82 C. Perez del Valle, A. Milet, Y. Gimbert and A. E. Greene, Angew. Chem. Int. Ed., 2005, 44, 5717. 83 H. Greenﬁeld, H. W. Sternberg, R. A. Friedel, J. H. Wotiz, R. Markby and I. Wender, J. Am. Chem. Soc., 1956, 78, 120. 84 E. V. Banide, H. Mu¨ller-Bunz, A. R. Manning, P. Evans and M. J. McGlinchey, Angew. Chem. Int. Ed., 2007, 46, 2907. 85 M. K. Pallerla, G. P. A. Yap and J. M. Fox, J. Org. Chem., 2008, 73, 6137. 86 M. E. Kraﬀt, I. L. Scott, R. H. Romero, S. Feibelmann and C. E. Van Pelt, J. Am. Chem. Soc., 1993, 115, 7199. 87 X. Verdaguer, A. Moyano, M. A. Perica´s, A. Riera, V. Bernardes, A. E. Greene, A. Alvarez-Larena and J. F. Piniella, J. Am. Chem. Soc., 1994, 116, 2153. 88 I. Marchueta, E. Montenegro, D. Panov, M. Poch, X. Verdaguer, A. Moyano, M. A. Perica´s and A. Riera, J. Org. Chem., 2001, 66, 6400. 89 C. M. Gordon, M. Kiszka, I. R. Dunkin, W. J. Kerr, J. S. Scott and J. Gebicki, J. Organomet. Chem., 1998, 554, 147. 90 F. Robert, A. Milet, Y. Gimbert, D. Konya and A. E. Greene, J. Am. Chem. Soc., 2001, 123, 5396. Organomet. Chem., 2010, 36, 93–120 | 119

91 T. J. M. de Bruin, A. Milet, F. Robert, Y. Gimbert and A. E. Greene, J. Am. Chem. Soc., 2001, 123, 7184. 92 Y. Gimbert, D. Lesage, A. Milet, F. Fournier, A. E. Greene and J.-C. Tabet, Org. Lett., 2003, 5, 4073. 93 Y. Gimbert, F. Robert, A. Durif, M.-T. Averbuch, N. Kann and A. E. Greene, J. Org. Chem., 1999, 64, 3492. 94 C. Wang and Y.-D. Wu, Organometallics, 2008, 27, 6152. 95 W. H. Pitcock Jr, R. L. Lord and M.-H. Baik, J. Am. Chem. Soc., 2008, 130, 5821. 96 T. J. M. de Bruin, A. Milet, A. E. Greene and Y. Gimbert, J. Org. Chem., 2004, 69, 1075. 97 M. E. Kraﬀt, R. H. Romero and I. L. Scott, J. Org. Chem., 1992, 57, 5277. 98 M. E. Kraﬀt, R. H. Romero and I. L. Scott, Synlett, 1995, 577. 99 T. J. M. de Bruin, C. Michel, K. Vekey, A. E. Greene, Y. Gimbert and A. Milet, J. Organomet. Chem., 2006, 691, 4281. 100 M. Elian and R. Hoﬀmann, Inorg. Chem., 1975, 14, 1058. 101 D. M. Hoﬀmann, R. Hoﬀmann and C. R. Fisel, J. Am. Chem. Soc., 1982, 104, 3858. 102 R. Hoﬀmann, T. A. Albright and D. L. Thorn, Pure Appl. Chem., 1978, 50, 1. 103 J. H. Schulte, R. Gleiter and F. Rominger, Org. Lett., 2002, 4, 3301. 104 V. Derdau, S. Laschat and P. G. Jones, Eur. J. Org. Chem., 2000, 4, 681. 105 J. A. Platts, G. J. S. Evans, M. P. Coogan and J. Overgaard, Inorg. Chem., 2007, 46, 6291. 106 J. Overgaard, H. F. Clausen, J. A. Platts and B. B. Iversen, J. Am. Chem. Soc., 2008, 130, 3834.

120 | Organomet. Chem., 2010, 36, 93–120

Scandium, Yttrium and the Lanthanides John G. Brennana and Andrea Sellab DOI: 10.1039/9781847559616-00121

1.

Introduction

This review covers the synthesis, characterization, and reaction chemistry of organometallic complexes of Sc, Y and the lanthanides reported in the year 2008. We have decided this year to focus on what we see as the signiﬁcant developments. We have been less rigid in discussing reactions by ancillary, but rather have tried to pull out a number of themes, organizing by chemical reactivity whenever possible. A striking development this year is the increasingly widespread use of the ortho-aminobenzyl ligands as hydrocarbon starting materials for protonolysis chemistry. The starting materials are cheap and the complexes are easy to isolate in salt-free form. This has also been the year for the use of AlMe4 as a means of introducing Ln-C interactions. A high point is the extraordinary molecular carbide species prepared by Anwander and the related mixed lanthanide aluminium clusters. For this astonishing work we would like to award the ‘SellaBren’ prize for 2009. This year has also seen a variety of new clusters and aggregates bridged by oxo, hydroxo and nitrogen ligands. Of particular interest are a pair of apparent imido clusters. But unsupported terminal imido ligands remain the ﬁeld’s araba fenice.w 2. 2.1

Hydrocarbyl chemical reactivity C–H bond activation

C-H bonds were activated in a variety of systems. Treatment of anilidophosphinimine (L) ligated Y mono(alkyl) complex LYR 0 (THF) [1] with 2 equivalents of phenylsilane in DME aﬀorded the methoxy-bridged complex [LY(m-OCH3)]2, via the corresponding hydrido intermediate, while in the presence of excess isoprene, an Z3-isopentene product, LY(CH2CMeQ CHCH3)(THF), was isolated. A Lu chloride, LLuCl(DME), was generated through the reaction of Lu mono(alkyl) complex LLuR 0 (THF), with [Ph3C][B(C6F5)4]. A metathesis reaction of the Y mono(alkyl) compound with excess AlMe3 at room temperature gave a Me-terminated counterpart, LLuMe(THF)2. In all these reactions, the Ln-C phenyl bonds remained untouched. However, protonolysis with 2 equivalents of phenylacetylene in DME provided a Lu bis(acetylide), LHLu(CCPh)2(DME), and the linkage of the Ln-C(phenyl) bond was cleaved, indicating that the activated C-H bond was recovered.1 The same ligand bound to scandium gave LScCl2 compounds that were alkylated with LiMe to cleanly give a

Department of Chemistry and Chemical Biology, Rutgers, the State University of New Jersey, 610 Taylor Road, Piscataway, NJ 08854-8087, USA Christopher Ingold Laboratories, University College London, 20 Gordon Street, London WC1H 0AJ, UK w ‘‘Che ci sia ciascun lo dice. Dove sia nessun lo sa.’’ (All say that it exists, but where it might be no one knows). Metastasio. b

Organomet. Chem., 2010, 36, 121–147 | 121 c

The Royal Society of Chemistry 2010

dialkyl complexes LScMe2. Thermolysis of these materials under Ar and H leads to decomposition products as a result of C-H activation of the ligand.2 Y dialkyl complexes coordinated by 6-aryl-substituted amidopyridinate ligands undergo selective intramolecular sp2 or sp3 C-H bond activation and, upon further treatment with PhSiH3, underwent a s-bond metathesis reaction to give extremely uncommon dimeric aryl-hydride [2] or benzylhydride Y complexes.3 i

N

i

Pr

SiMe3 P

N

THF

Y

H

Y

Y

THF

H

N

i

N

Pr

Pr

N

THF i

Pr

N

i

Pr

i

Pr

[1]

[2]

The imido ligand of a transient imidoscandium compound activated the C–H bond in pyridine and benzene, and it was complexed with Al(CH3)3 to yield an imide zwitterion [3]. A combination of isotopic labeling, reactivity, and DFT analysis implied the formation of a terminal Sc-NR functional group.4 The reaction of ScR30 (THF)2 with 1,4,7-trithiacyclononane ([9]aneS3) gave Sc([9]aneS3)R30 [4]. The corresponding reaction for yttrium did not go to completion. Reaction of MR30 (THF)2 with [NHMe2Ph] [BArF4] (ArF=C6F5) in the presence of a face-capping ligands L (LQHC(Me2pz)3, Me3[9]aneN3, or [9]aneS3) gave cationic complexes that had b-Si-C agostic alkyl groups in most instances. The isolated cation [Sc(fac-N3)(CH2SiMe3)2(THF)] þ underwent THF substitution reactions with OPPh3 or pyridine, Sc-alkyl migratory insertion with carbodiimides, and C–H bond metathesis with PhCCH. The scandium complexes polymerized ethylene when activated with one equivalent [CPh3][BArF4], and 1hexene when activated with two equivalents. DFT calculations were used to support speculation about structural chemistry and reactivity.5

Pri

Pri i P Pr N

Sc

N

C H3

P Pri

S S

i

Pr

S Sc

Al Me3Si

SiMe3 SiMe3

Pri

[3]

122 | Organomet. Chem., 2010, 36, 121–147

[4]

2.1.1 CH bond activation: carbides. An important series of compounds with multiply charged carbanions were reported. Reactions of [La(AlMe4)3] or [Y(AlMe4)3] with PMe3 show that the phosphine can cleave relatively weak 3c,2e Ln-CH3-Al linkages, and the La compound reacted with one equivalent of PMe3 by Me group C-H bond cleavage to give brilliant methylene, methine, or carbide moieties. [La4Al8(CH)4(CH2)2Me20(PMe3)] [5] has a {La4(CH)4} cuboid core supported by AlMe3, Me2AlCH2AlMe2, and PMe3 ligands. [La4Al8(C)(CH)2(CH2)2Me22(toluene)] [6] also contains a cuboid core, and [La3Al(C)(CH)2(CH2)], which includes one exo cubic La atom, is supported by AlMe3, Me3AlCH2AlMe2, (AlMe4) , and toluene ligands. The La atoms in [La5Al9(CH)6Me30] [7] are arranged in a trigonalbipyramidal fashion with (CH) functionalities capping each face. The {La5(CH)6}3 core is formally balanced by three AlMe2þ moieties.6 Me Al

Me

CH2 Me

Me Al

Me

Me CH

La

Al

C

Al

Me2 Al Me

Me

Me Al

Me

Me

PMe3

Me

Al C

Me Me Me

AlMe C

Al

Me

C H2

Me

CH

La Me

Al Al

La

La

Me

La

C

Me

La

Al

Me CH

CH2

Me

Me

Me

Al Me

Me

C

La

H2 C

Al

Me

La Me

Me

Me

Me

Al

Me

La

Me

Me

[5]

[6] Me Al Me La

Me

Me Al Me Me Me Me Al CH CH Me CH

Me Al Me Me

CH

La La AlMe2

Al

Me

Al

La Me

CH Al Me C Al

La Me

Me

Me Me

Me

Me Me

[7]

Another methylidene complex, (PNP)Sc(m3-CH2)(m2-CH3)2[AlMe2]2 [8] (PNPQN[2-P(CHMe2)2-4-MeC6H3]2 ) was prepared from the reaction of (PNP)ScMe2 and 2 equivalents of AlMe3. (A closely related La species is reported later [25]). The CH2 ligand reacts with H2NAr and OCPh2, the latter reagent yielding the oleﬁn H2CQCPh2 along with the novel oxoscandium complex (PNP)Sc(m3-O)(m2-CH3)2[AlMe2]2.7 Finally, while on the topic of non-classical carbon ligands, the d0 yttrium N-heterocyclic carbene compound mer-Y(OCMe2CH2[C{N(CHCH)NPri}]3 [9] was prepared and structurally characterized.8 Organomet. Chem., 2010, 36, 121–147 | 123

Pri

Pri

Pri

H 3C Sc

N

H3C

N

Al

Y

CH2

O

O

Al Pri

N

N

Pri

[8]

2.2

N O

N

P Pri

N Pr

i

P

[9]

Reactions of Ln-R with protic sources

2.2.1 Protonolysis to give alkoxides. The Et substituted aluminum hydroxide LAlEt(OH) (L=HC[C(Me)N(2,6-iPr2C6H3)]2) was prepared by the hydrolysis of LAlEt(Cl) in the presence of a N-heterocyclic carbene. This hydroxide reacts with Cp2ZrMe2 in toluene to give LAlEt(m-O)ZrMeCp2 and CH4, or with Cp3M in THF to give LAlEt(m-O)M(THF)Cp2 and HCp (M=Yb, Er, Dy, Y). In both cases, the displacement of the Al and the g-C atom out of the NCCN plane is observed in a boat conformation, but with converse direction.9 The protonolysis reaction of heterobimetallic peralkylated complexes [Ln(AlR4)2]n (Ln=Sm, Yb; R=Me, Et) with 2 equivalents of a sterically demanding alcohol (HOR) aﬀords the bis(trialkylaluminum) adducts Ln[(m-OR)(m-R)AlR2]2. Analogous reactions with the less sterically demanding iPr-substituted HOR 0 result in ligand redistributions and formation of Ln[(m-OR 0 )2AlR2]2 (Ln=Yb, Sm), Yb[(m-OR 0 ) (m-Et)AlEt2]2(THF), and [Et2Al(m-OR 0 )2Yb(m-Et)2AlEt2]2. The solid-state structures of a serendipitous aluminoxane complex Sm[(m-OR)AlEt2OAlEt2(m-OR)] and a dimeric AlMe3-adduct complex [(AlMe3)(m-OR)Sm (m-OR)2Sm(m-OR)(AlMe3)] were also determined. While the former can be discussed as a typical hydrolysis product of Sm[(m-OR)(m-Et)AlEt2]2, the latter was isolated from the 1:1 reaction of [Sm(AlEt4)2]n with HOR.10 A series of hydroxyamide compounds were prepared by a number of approaches. The amido derivative Y[N(SiHMe2)2]2[ONBn2][THF] was prepared by the silylamide elimination pathway from the reaction of N,N-dibenzylhydroxylamine, Bn2NOH (Bn=CH2C6H5), with Y[N(SiHMe2)2]3[THF]2. Organometallic derivatives [Cp2Y(ONBn2)]2 and [Cp2Sm(ONBn2)]2 were obtained by the reactions of MCp3 (M=Y, Sm) with Bn2NOH. The trinuclear compound Cp5Y3[ON(Me)CH2CH2(Me)NO]2 [10] was obtained by the reaction of YCp3 and the bis-hydroxylamine HON(Me)CH2CH2(Me)NOH. In all these complexes side-on-coordination of the hydroxylamide units was observed. The compounds are dynamic in solution.11 CH3

CH3 N

N

Y O N

O

O

Y

O N

Y CH3

CH3

[10]

124 | Organomet. Chem., 2010, 36, 121–147

The synthesis, characterization and reactivity of heteroleptic rare earth metal complexes supported by the carbon-bridged bis(phenolate) ligand 2,2 0 -methylenebis(6-tert-butyl-4-methylphenolate) (MBMP)2 was described. Reaction of Cp3Ln(THF) with a carbon bridged bisphenol (MBMPH2) in a 2:3 molar ratio in THF at 501 produced the heteroleptic rare earth metal bis(phenolate) complexes CpLn(MBMP)(THF)n (Ln=La, Yb, Y) in nearly quantitative yields. The remaining Cp groups can be replace at elevated temperature to give the neutral rare earth metal bis(phenolate) complexes that vary with metal radius. Lanthanum gave dinuclear (MBMP)La(THF)(m-MBMP)2La(THF) while Y and Yb formed (MBMP) Ln(MBMPH)(THF)2 as the ﬁnal products, in which one hydroxyl group of the phenol is coordinated to the rare earth metal as a neutral donor. These compounds react with AlEt3 and ZnMe2 in toluene at room temperature to give ligand redistribution products, and with LiR to give ionic products.12 2.2.2 Protonolysis to give amides. Aminolysis of a tribenzylscandium with ferrocenediamine 1,1 0 -(R3SiNH)2Fc gave the diamido scandium complex [[1,1 0 -Fc(NSiR3-kN)2]Sc(CH2C6H3Me2-3,5)(THF)] (Fc=ferrocenediyl, SiR3=SiMe2tBu). Attempts to remove the coordinated THF with AlMe3 led to the isolation of a scandium Me complex [[m-1,1 0 -Fc(NSiR3kN:kN)2]Sc[AlMe2(m-Me)]2] [11], which then led to a scandium Me complex [[1,1 0 -Fc(NSiR3-mN)2]ScMe(THF)2] that polymerizes lactide.13 Lanthanum dibenzyl complexes with triazanonane-silylamide, amidine and 1,3-diketiminate monoanionic ancillary ligands were also prepared by in situ peralkylation in the presence of the protonated ligand or by complexation (for 1,3-diketiminate) with lanthanum tribromide and subsequent alkylation. The coordination polymer {[m-Z2:Z1-ArNC(Me)CHC(CH2)NAr]2La[K(THF)4]}n formed by H-abstraction from one of the diketiminate Me groups and a ligand redistribution.14 Reaction of Cp2LnCl with KTpMe2 in toluene gave the mixed TpMe2/Cp lanthanide complexes Cp2Ln(TpMe2) (Ln=Yb, Er, Dy). Reactions with protic compounds were explored: benzotriazole (C6H4NHN2) gave the lanthanide metallomacrocyclic complex [(TpMe2)CpEr(m-N3C6H4)]3, while 2-aminopyridine in THF gave an unexpected oxide complex [(TpMe2)Yb (2-HNC5H4N)]2(m-O). The oxide ligand was thought to originate with adventitious water. Two equivalents of 3,5-dimethylpyrazole yields a completely Cpabstracted product (TpMe2)Dy(PzMe2)2(THF), that can also be directly obtained from a three-component reaction of Cp2DyCl, KTpMe2, and 3,5-dimethylpyrazole in THF.15 Reaction of the lanthanide metallocene allyl complexes, Cp*2Ln(Z3-CH2CHCH2)(THF) (Ln=Ce, Sm, Y) with 1,3,4,6,7,8-hexahydro2H-pyrimido[1,2-a]pyrimidine (Hhpp) forms a series of metallocene complexes, (Cp*)2Ln(k2-hpp) (Ln=Ce, Sm, Y). The coordination mode of the hpp anion was compared with the acetonitrile adduct, Cp*Sm(hpp)(MeCN), the capro2 lactamate Cp*2Y(ONC6H10) and the dithiocarbamate Cp*2Sm(S2CNEt2).16 Two reports included multidentate ligands as supports for amination chemistry. First, 4,40 -Di-tert-butyl-2,20 -bipyridyl (tBu2bpy) stabilizes the thermally sensitive [LuR 0 3] unit, giving the [(tBu2bpy)Lu(R 0 )3]. This tris(alkyl) complex readily reacts with Ph3COH, 2,6-iPr2C6H3NH2, 2,4,6-tBu3C6H2NH2, and N,N 0 -dicyclohexylcarbodiimide to aﬀord a variety of Lu(III) tris(alkoxide), Organomet. Chem., 2010, 36, 121–147 | 125

tris(amide), mono(amide) bis(alkyl), and amidinate bis(alkyl) compounds. Reaction of an amide bis(alkyl) complex with OPPh3 gave (Ph3PO)2Lu(NHC6H2tBu3-2,4,6)(R0 )2, showing that the bidentate tBu2bpy ligand can be displaced.17 Lu alkyl complexes supported by a monoanionic, tridentate ligand system formed by the dearomatization and functionalization of a 2,2 0 :6 0 ,200 -terpyridine have been prepared. These were reacted with H2NC6H2Ph3-2,4,6 or the ﬂuorinated anilines H2NC6H4F-4 and C6F5NH2 to give both terminal alkaly/mono(amide) [12] and bis(amide) Lu(III) complexes, which were fully characterized. Once again, elimination does not occur and no evidence was seen for a terminal (or bridging) imido species.18

SitBuMe2

N

N Ph

SiMe3

N Sc

Fe

Lu

N H

N

N SiMe3

N Ph SitBuMe2

N

[11]

ArF

H

[12]

2.2.3 Protonolysis to give other bonds. The reaction of the half-sandwich Lu dialkyl complex Cp*Lu(R0 )2(THF) with Ru-trihydrido-phosphine complexes Cp*Ru(PR3)H3 (R3=Ph3, Ph2Me, PhMe2, and Me3) aﬀorded the corresponding Lu/Ru heterobimetallic dihydride complexes [Cp*M(m-H)2 (RuCp*)2(m-PR2C6H4)] [13] which were accompanied by selective C–H bond activation of the phosphine ligand. The reaction of these phosphinomethylbridged Lu/Ru complexes with PhSiH3 led to selective dehydrogenative silylation at the CH2 unit.19 Similarly, the reaction of monohydride complexes of the late transition metals with lanthanoid alkyls aﬀords heterobimetallic compounds with direct metal-metal bonds. These covalent bonds are strongly polar and can be considered as donor-acceptor bonds. Thus, reaction of [Cp*RuH2]2 with [Cp2Y(R0 )(THF)] gave [H(Cp*Ru)2H2YCp2], [14].20 The heterobimetallic peralkylated complexes [Ln(AlR4)2]n (Ln=Sm, Yb; R=Me, Et) were synthesized and the two compounds are isomorphous. Polymeric [Yb(AlMe4)2]n was examined by 1H and 13C MAS NMR spectroscopy revealing the presence of distinct bridging Me groups. The Yb compound reacts wtih 1,10-phenanthroline (Phen) to give the donor adduct Yb(AlMe4)2(Phen), while the protonolysis reaction with 2 equivalents of HCp* yielded ionic [Cp*Yb(THF)4][AlMe4].10

P THF

Y

Ph Ph

H

Ru

Y

H

Ru H

[13]

126 | Organomet. Chem., 2010, 36, 121–147

H

Ru

H

[14]

2.3

Insertion reactions

The polar nature of lanthanide bonds often leads to chemistry where unsaturated molecules with relatively electronegative components react by insertion into the Ln-X bonds of relatively electropositive anions. This year witnessed insertions with an unusually diverse range of unsaturated molecules. For example, the reaction of [(Cp*)2Ln][(m-Ph)2BPh2] complexes with the Li salt of (trimethylsilyl)diazomethane, Li[Me3SiCN2], gave products formulated as the dimeric isocyanotrimethylsilyl amide complexes {Cp*2Ln[mN(SiMe3)NC]}2 (Ln=Sm, La). Reactions of Cp*2Sm and [Cp*2Sm(m-H)]2 with Me3SiCHN2 also form the same product. These compounds react with Me3CCN to form the 1,2,3-triazolato complexes Cp*2Ln(NCCMe3)[NNC(SiMe3)C(CMe3)N] and the La derivative reacts further with Me3SiN3 to make the isocyanide ligated azide complex {Cp*2La[CNN(SiMe3)2](m-N3)}3 [15].21 Reaction of Cp2LnNH(nBu) with Ph2CQCQO in toluene aﬀords dimeric complexes [Cp2Ln(OC(CHPh2)NnBu)]2 [Ln=Yb, Dy], derived from a formal insertion of the CQC bond of the ketene into the N-H bond. Treatment of CpErCl2 with 2 equivalents of LiNHnBu followed by reacting with Ph2CCO aﬀords a rearrangement product [Cp2Er(OC(CHPh2)NnBu)]2. Treatment of [Cp2Ln(m-Im)]3 (Im=imidazolate) with PhRCCO gives [Cp2Ln(mOC(Im)QCPhR)]2 [R=Et, Ln=Yb; R=Ph, Ln=Yb, Er]. In contrast to the previous observations that [Cp2ErNiPr2]2 and [Cp2ErNHEt]2 react with ketenes to give di-insertion products, in the present cases the presence of excess of ketene has no inﬂuence on the ﬁnal product even with prolonged heating, and only monoinsertion products are isolated.22 (Trimethylsilyl)diazomethane, Me3SiCHN2, is not metalated by the metallocene allyl complexes Cp*2Ln(C3H5) but instead inserts to form the lanthanide hydrazonato complexes Cp*2Ln[Z2(N,N 0 )-CH2QCHCH2NNQCHSiMe3] [16] (Ln=Sm, La, Y). Although the La, Y, and Sm complexes are isomorphous, the double bond in the allyl substituent is oriented toward La and away from Y and Sm.23 Insertions were also used toward materials synthesis. Two organometallic precursor approaches leading to the hitherto-unknown dioxo monocarbodiimides (Ln2O2CN2) of the late lanthanides Ho, Er, and Yb as well as yttrium were outlined. The ﬁrst involved insertion of CO2 into the Ln–N bond of [(Cp2ErNH2)2] to give the single-source precursor [Er2(O2CN2H4)Cp4]. Ammonolysis of this amorphous compound at 7001 aﬀords Er2O2CN2. In the second approach, the molecular carbamato complex [Cp4Ho2{m-Z1:Z2OC(OBut)NH}] [17] was prepared and found to display an unusual bonding mode of the tert-butylcarbamate ligand, which acts as both a bridging and sideon chelating group. Ammonolysis of this compound also yielded Ln2O2CN2.24 N(SiMe3)2 N(SiMe3)2

t

N

O

N

C N3

La

H C

La

N3

NH SiMe3

La N3

Bu

N

O

Ho

Ln

Ho

O

N

NH

C

O

N

CH2

N(SiMe3)2

[15]

[16]

t

Bu

[17]

Organomet. Chem., 2010, 36, 121–147 | 127

through hydride bridges with trans-accommodated terminal aryloxide groups. These compounds initiate co-polymerizationn of CO2 and cyclohexene oxide under mild conditions to aﬀord polymers with modest molecular weights and high carbonate linkages. Insight into these reactions comes from reactions with CO2 that generate the mixed formate/carbonate complexes [Cp 0 Ln(m-Z1:Z1-O2CH)(m-Z1:Z1-O2COAr)]2. The two Cp 0 Ln fragments in these complexes are bridged by the formate and carbonate species, respectively, to form two square-pyramidal geometries around the metal centers.25 Compounds Cp2Ln[k3-[4-NH(C8N2H4)](2-NH2C6H4)] [Ln=Er, Y] were synthesized by the reaction of Cp2LnNiPr2(THF) with anthranilonitrile in THF, indicating a novel organolanthanide-mediated intermolecular nucleophilic addition/cyclization of anthranilonitrile. A multi-step mechanism was proposed. To trap the intermediate o-Cp2LnNHC6H4CN, a probe reaction of Cp2ErNiPr2(THF) with anthranilonitrile and carbodiimide iPrNQCQNiPr also was studied. This reaction gave both the same product described above and [Cp2Er(m-Z1:Z2-NCC6H4N(H)C(NHiPr)Q NiPr)]2, indicating that nucleophilic additions of carbodiimide and anthranilonitrile with the intermediate are competitive.26 Dicationic Me complexes of the rare-earth metals [LnMe(THF)n][BAr4]2 (Ln=Sc, Y, La-Nd, Sm, Gd-Lu; Ar=Ph, C6H4F-4) were synthesized, by protonolysis of either the ate complexes [Li3LnMe6(THF)n] (Ln=Sc, Y, Gd-Lu) or the tris(tetramethylaluminates) [Ln(AlMe4)3] (Ln=La-Nd, Sm, Dy, Gd) with ammonium borates [NR3H][BAr4] in THF. The number of coordinated THF ligands varied from n=5 (Ln=Sc, Tm) to n=6 (Ln=La, Y, Sm, Dy, Ho). The conﬁguration of representative examples was established by X-ray diﬀraction studies and conﬁrmed by DFT calculations. The highly polarized bonding between the Me group and the rareearth metal center results in the reactivity pattern dominated by the carbanionic character and the pronounced Lewis acidity. The dicationic Me complex [YMe(THF)6]2 þ inserted benzophenone as an electrophile to give the alkoxy complex [Y(OCMePh2)(THF)5]2 þ . Nucleophilic addition of the soft anion X (X =I , BH4 ) led to the monocationic Me complexes [YMe(X)(THF)5] þ .27 The reaction of the lanthanide alkyls with 1-methylalk-2-ynes CH3CCR (R=Me, Et, Pr, tBu, SiMe3, Ph, C6H4Me-2, C6H3Me2-2,6, C6H3iPr2-2,6, C6F5) aﬀords the corresponding Z3-propargyl/allenyl complexes Cp*2LnCH2CCR and Me2Si(Z5-C5Me4)2CeCH2CCR via propargylic metalation. Hydride complexes [Cp*2Ln(m-H)]2 (Ln=Y, Ce, La) react rapidly with these products to produce mixtures of insertion and propargylic metalation products, and the relative rate of these processes depends on the metal and alkyne substituent. Further reactions of selected Z3-propargyl/allenyl products with Brønsted acids, such as alcohols and acetylenes, aﬀord the corresponding substituted allenes (RCHQCQ CH2) and 1-methylalk-2-ynes (CH3CCR) as organic products, and reactions with Lewis bases such as py or THF gave the corresponding base adducts.28 The tris-alkyl complex YR30 (THF)2 reacts with 1,2,3-trimethyl-1Hcyclopenta[l]phenanthrene (PCp*H) to give (PCp*)YR20 (THF). The THF is 128 | Organomet. Chem., 2010, 36, 121–147

labile and can be displaced by bipy to give (PCp*)YR20 (bipy). Reactions with CO2, Me3SiNCO, and iPrNQCQNiPr aﬀord the expected insertion products while Me3SiCCH aﬀords a terminal bis(acetylide) (PCp*)Y(CCSiMe3)2(THF) that dimerizes in the solid state to {[(PCp*)Y(CCSiMe3)(THF)]2(m-CCSiMe3)2}.29

3.

Redox chemistry

The chemistry of lanthanides in unusual oxidation states continues to twist and turn in unexpected directions as a number of workers have begun to explore ligand combinations with steric demands that would previously have been been considered laughable. In a surprising example of spontaneous reduction, deprotonation of the perarylated cyclopentadiene [(4-nBu-C6H4)5C5H, (CpBIGH)] by the chelating [(Me2Nbenzyl)3Sm] in benzene at 601 gave the divalent samarocene [(CpBIG)2Sm] [18] in moderate yield. The exceptional stability of such sterically congested metallocenes is explained by a merry-go-round C–H?C(p) hydrogen-bond network. Similar chemistry was observed for the more reducible Yb complex, and crystal structures of these and the isomorphous calcium compound were determined.30 Striking solvent dependence was noted to the redox-transmetalation ligand-exchange reaction of Yb or Ca metal with 2 equivalents of pentaphenylcyclopentadiene (C5Ph5H) induced with 1 equivalent of HgPh2. In THF this aﬀorded the solvent-separated ion pairs (SSIPs) [M(THF)6][C5Ph5]2 (M=Yb, Ca). Addition of toluene to the isolated SSIPs led to the precipitation of the homoleptic sandwich complexes [M(C5Ph5)2] (M=Yb, Ca). In the reaction of Ba metal with C5Ph5H and HgPh2 the corresponding SSIP was observed in situ, and only the sandwich complex [Ba(C5Ph5)2] could be isolated. Highly symmetric structures with two parallel cyclopentadienyl ligands were found. Oxidation and metal-ligand exchange reactions were studied for the divalent Yb complexes.31 The reductive reactivity of lanthanide hydride ligands in [Cp*2LnH]x complexes (Ln=Sm, La, Y) was examined to see if these hydrides would mimic actinide hydrides. Each lanthanide hydride complex reduces PhSSPh to give [Cp*2Ln(m-SPh)]2 in very high yield. [Cp*2SmH]2 reduces phenazine and anthracene to make [Cp*2Sm]2(m-Z3:Z3-C12H8N2) and [Cp*2Sm]2(m-Z3:Z3-C10H14), respectively. However, the analogous [Cp*2LaH]x and [Cp*2YH]2 reactions are more complicated. All three lanthanide hydrides reduce COT to aﬀord Cp*Ln(COT) and Cp*3Ln, a reaction that constitutes yet another synthetic route to Cp*3Ln complexes. In the reaction of [Cp*2YH]2 with C8H8, two unusual byproducts were obtained. In benzene, a Cp*Y[(Z5-C5Me4CH2-C5Me4CH2-Z3)] complex forms in which two Cp* rings are linked to make a new type of ansa-allyl-cyclopentadienyl dianion that binds as a pentahapto-trihapto chelate. In cyclohexane, on the other hand, a Cp*2Y(m-Z8:Z1-C8H7)YCp* [19] complex forms in which a (COT)2 ring is metalated to form a bridging (C8H7)3 trianion.32 Organomet. Chem., 2010, 36, 121–147 | 129

Ar-t-Biu tBuAr

Ar-t-Biu Ar-t-Biu

tBuAr tBuAr

Sm

Y

H

Ar-t-Biu

tBuAr

Y

Ar-t-Biu Ar-t-Biu

[18]

[19]

The limits of steric crowding in tris(pentamethylcyclopentadienyl) complexes were redeﬁned by isolation of Cp*3LaL2. Addition of tBuCN to Cp*3La produced the ﬁrst crystallographically characterized Cp*3M adduct, Cp*3La(NCtBu)2 [20]. Although neither THF nor Ph3PO formed crystallographically characterizable Cp*3LaLx complexes, these Cp*3La/Lx mixtures displayed enhanced reactivity compared with Cp*3La towards substrates such as COT and CO2. Attempts to use adamantyl azide, AdN3, as a ligand, led to the ﬁrst example of azide insertion into a metal cyclopentadienyl linkage to generate Cp*2La[Z2-(N,N 0 )-(Cp*)NN 0 N 0 0 Ad] [21].33 Ad N N t

Bu

C

N

La

N

C tBu

N

La

N

N N Ad

[20]

4.

[21]

Catalysis

Hydroamination processes continues to attract attention. Highly constrained, axially chiral atropos diamines and their tropos analogs were evaluated as ligands for ytterbium-catalyzed intramolecular hydroamination and compared to the ligand 1,1 0 -binaphthyl-2,2 0 -bis(benzylamine). They aﬀorded highly active catalysts for the cyclization of aminopentenes and aminohexenes with up to 58% ee.34 New chiral binaphthylamido yttrium and ytterbium -ate complexes with lithium and potassium counterions have been synthesized and characterized. X-ray structures have been obtained for isomorphous [Li(thf)4][Ln{(R)-C20H12(NC5H9)2}2] (Ln=Yb, Y) and [K(thf)5][Yb{(R)-C20H12(NCH2CMe3)2}2]. The eﬃciency of these complexes for the enantioselective intramolecular hydroamination was examined and the role of the counter cation was explored. The most active catalyst of this series, [Li(thf)4][Yb{(R)-C20H12(NCH2CMe3)2}2], was successfully used for the cyclization of aminopentenes with internal double bonds.35 A new neutral tridentate ligand, 1,4,6-trimethyl-6-pyrrolidin-1-yl-1,4diazepane (L) has been used to make the hexacoordinate complexes (L)MR 0 3 (M=Sc, Y) and (L)M(CH2Ph)3 (M=Sc, La) [22]. Cationic 130 | Organomet. Chem., 2010, 36, 121–147

complexes were then prepared by careful protonation in the absence of THF. The neutral complexes and their cationic derivatives were studied as catalysts for the hydroamination/cyclization of 2,2-diphenylpent-4-en-1amine and N-methylpent-4-en-1-amine and compared with corresponding ligand-free Sc, Y, and La neutral and cationic catalysts. The most eﬀective catalysts in the series were the cationic L-yttrium catalyst for 2,2-diphenylpent-4-en-1-amine and the cationic lanthanum systems for N-methylpent-4-en-1-amine. For the La catalysts, evidence was obtained for release of L from the metal during catalysis.36 In related work, reaction of (L)Sc(CH2Ph)2 with two equivalents of phenylacetylene aﬀords the monomeric dialkynyl complex (L)Sc(CCPh)2, while reactions of (L)M(CH2Ph)2 (M=Y, La) give the dimeric dialkynyl complexes [(L)M(CCPh)(mCCPh)]2. Catalysts for the Z-selective linear head-to-head dimerization of phenylacetylenes were also identiﬁed. Cationic Y and neutral La systems were the most eﬀective catalysts in the series while the Sc analogues showed poor activity and selectivity.36 In what looks like useful chemistry, a variety of solvated divalent lanthanide complexes with the formula LnL2(THF)n (Ln=Sm, Eu, Yb; L=N00 , MeCp, ArO=[2,6-(tBu)2-4-MeC6H2])), were found to be excellent catalyst precursors for the addition of various primary and secondary amines to carbodiimides, eﬃciently providing the corresponding guanidine derivatives with a wide range of substrates under solvent-free conditions. The reaction showed good functional group tolerance. The work also revealed some excellent precatalysts for addition of terminal alkynes to carbodiimides yielding a series of propiolamidines. For both reactions, activity was found to be YboEuoSm for metal and MeCpoArOoN00 for ligand. The ﬁrst step in both reactions is believed to be the formation of a bimetallic bisamidinate species formed by reductive coupling of carbodiimide. The active species is proposed to be a lanthanide guanidinate or amidinate.37 A related patent has been registered for a catalytic method for the conversion of amines and carbodiimides to guanidines using divalent lanthanides with a variety of ancillary ligands.38

N N

N Ln Me3Si

PPh2

Ln Me2Si N

Ph2P

Ln

N SiMe2

SiMe3 SiMe3

[22]

[23]

The acid-base reactions between the rare-earth metal tris(o-N,N-dimethylaminobenzyl) complexes [Ln(CH2C6H4NMe2-o)3] with one equivalent of the geometry-constraining ligand (C5Me4H)SiMe2NH(C6H2Me3-2,4,6) aﬀorded the corresponding half-sandwich aminobenzyl complexes [{SiMe2(C5Me4H)N(C6H2Me3-2,4,6))}Ln(CH2C6H4NMe2-o)(THF)] (Ln= Y, La, Pr, Nd, Sm, Gd, Lu). These complexes were found to be catalyst Organomet. Chem., 2010, 36, 121–147 | 131

precursors for the catalytic addition of various phosphine P-H bonds to carbodiimides to form phosphaguanidine derivatives with excellent tolerance to aromatic C-halogen bonds. Activity was highest for the larger metals. The reaction of the La complex with Ph2PH yielded the corresponding phosphido complex [{Me2Si(C5Me4)(NC6H2Me3-2,4,6)} La(PPh2)(THF)2], which, on recrystallization from benzene, gave the dimeric analog [{Me2Si(C5Me4)(NC6H2Me3-2,4,6)}La(PPh2)]2 [23] Addition of iPrNQCQNiPr to such phosphide complexes gave phosphaguanidinate complexes.39 Finally, new chiral binaphthylamido alkyl ate complexes [(R)-C20H12 (NC5H9)2]Y[(m-Me)2Li(THF)2(m-Me)Li(THF)] and [(R)-C20H12(NC5H9)2] Ln[(m-Me)2Li(TMEDA)(m-Me)Li(OEt2)] (Ln=Y, Yb), and the neutral complex [(R)-C20H12(NC5H9)2]YR 0 (DME) (C20H12(NHC5H9)2=2,2-bis (cyclopentylamino)-1,1 0 -binaphthyl) were prepared. Both types of complexes can be easily prepared in a one-pot procedure starting from LnCl3 and used in situ. They proved to be very eﬃcient catalysts for enantioselective intramolecular hydroamination of aminopentenes or aminohexene at room temperature with enantiomeric excesses up to 83%.35 5.

Organolanthanides in organic synthesis

A practical method for the synthesis of gem-2,2-disubstituted tertiary amines from the corresponding lactams (or amides) has been reported. It is based on the reaction of thioiminium ions, easily prepared from lactams and amides with primary alkylcerium reagents.40 Z-selective cross-coupling of terminal alkynes with isocyanides to exclusively yield (Z)-1-aza-1,3-enyne products has been achieved for the ﬁrst time using a constrained geometry catalysts [Y(C5Me4Si(Me2)NR)(Alk) (THF)x] (Alk=R 0 , dimethylaminobenzyl; R=tBu, Ph, 2,4,6-Me3C6H2. The intermediate was shown to be the acetylide-bridged dimer [Y(C5Me4Si (Me2)NR)(m-CCPh)]2.41 [24] Stoichiometric reactions of mesityl azide (MesN3, Mes=2,4,6-C6H2Me3) with amino-phosphine ligated rare-earth metal alkyl, LLn(CH2SiMe3)2(THF) (L=(2,6-C6H3Me2)NCH2C6H4PPh2; Ln=Lu, Sc), amide, LLu(NH(2,6C6H3iPr2))2(THF) and acetylide at room temperature gave the amino-phosphazide ligated rare-earth metal bis(triazenyl) complexes, [L(MesN3)]Ln [(MesN3)(CH2SiMe3)]2 (Ln=Lu; Sc), bis(amido) complex [L(MesN3)]Lu [NH(2,6-C6H3iPr2)]2, and bis(alkynyl) complex (L(MesN3)Lu(CCPh)2)2, respectively The triazenyl group coordinates to the metal ion in a rare Z2-mode via Nb and Ng atoms, generating a triangular metallocycle. The aminophosphazide ligand, L(MesN3) chelates to the metal ion in a Z3-mode via Na and Ng atoms. In the presence of excess phenylacetylene, [L(MesN3)]Ln [(MesN3)(CH2SiMe3)]2 isomerized and the triazenyl group coordinates to the metal ion in a Z3 mode via Na and Ng atoms. These complexes showed an unprecedented catalytic activity towards the cycloaddition of organic azides and aromatic alkynes to aﬀord 1,5-disubstituted 1,2,3-triazoles selectively.42 The sterically crowded, monoanionic scorpionate ligand tris(3-tert-butyl5-methylpyrazol-1-yl)borate (TptBu,Me) provides a unique environment for the isolation of discrete rare-earth metal complexes with YMe{AlMe4} and 132 | Organomet. Chem., 2010, 36, 121–147

La{CH2(AlMe3)2} [25] moieties very similar to the scandium species such as [8]. The latter shows promising reactivity as a Tebbe reagent analog, reacting with 9-ﬂuorenone to give 9-methylideneﬂuorene and (9-methylﬂuorenoxy)dimethylaluminum.43 Ph R

C

N

C

Y

Me2Si N R

Me Me

Y

SiMe2

H

C

B

N

N

N Me

C

t Bu Bu N Me t

N

Me Al

Me

CH2 Yb N Me Al Me t

Bu

Me

Ph

[24]

6.

[25]

Organolanthanides in materials synthesis

Atomic layer deposition (ALD) continues to be used for the preparation of high-permittivity (k) rare-earth oxide ﬁlms for advanced gate stack applications. In one study, transmission electron microscopy measurements in the high-resolution mode coupled with electron energy loss spectroscopy experiments were used to probe nanometric scale interface layer issues for ALD-grown La2O3/Si stacks. Complementary results from electron and X-ray diﬀraction measurements indicate that La2O3 ﬁlm reactivity with a Si surface can be controlled up to a certain extent by appropriately choosing the ALD precursor combination, with LaCp3 þ O3 scheme better than LaCp3 þ H2O for depositing La2O3 ﬁlms.44 Yttrium oxide ﬁlms were also deposited on silicon using Y(C5H4Et)3 with water vapor as the oxygen source. Film growth kinetics were examined with respect to reactor conditions and concentrations. PES analysis of the Y2O3 product indicated no evidence of carbon contamination, and glancing incidence X-ray diﬀraction data suggests the ﬁlm structure to be polycrystalline.45 Er2O3 ﬁlms were deposited by low-pressure metalorganic CVD (MOCVD) also plasma assisted (RP-MOCVD), using Er(C5H4iPr)3 and O2 on Si(100), Si(111), and Corning glass substrates. The RP-MOCVD approach produced highly (100)-oriented, dense and mechanically stable Er2O3 ﬁlms with a columnar structure, while ﬁlms with (111) texture were deposited by MOCVD. Physical properties suggested potential applications as antireﬂective and protective coatings or as high-k dielectrics in CMOS devices.46,47 A minor correction to this work was also published.48 For heterometallic materials, lanthanum hafnium oxide thin ﬁlms continue to attract attention, with Hf(N(Me)(Et))4 serving as the Hf source, and either La(C5H4Et)3 49 or La(C5H4iPr)350 serving as the La source. 7.

Polymer chemistry

7.1

Apolar monomers

Allyl complexes continue to ﬁnd uses in polymerization catalysis. The new isoprene-co-1,3,7-octatriene [P(IP-co-OT)] copolymers with controlled Organomet. Chem., 2010, 36, 121–147 | 133

composition, molecule weight and microstructure have been prepared using a catalyst system [Nd(allyl)2Cl(MgCl2)2(THF)4/MAO]. The pendant vinyl moieties of the polymers were then converted to phosphonate groups by hydrophosphorylation in the presence of homogeneous rhodium phosphine catalysts.51 Patents have been reported for a series of bimetallic metallocene catalysts based on lanthanide and group IV metals. The bridging ligand is a substituted ethyl-linked ﬂuorenyl indenyl bearing a substituent of varying length [26]. The complexes are reported to act as good oleﬁn polymerization catalysts in the presence of MAO.52,53 A series of aryldiimine NCN-pincer ligated rare earth metal dichlorides (2,6-(2,6-C6H3R2NQCH)2-C6H3)LnCl2(THF)2 (Ln=Y, R=Me, Et, iPr; R=Et, Ln=La, Nd, Gd, Sm, Eu, Tb, Dy, Ho, Yb, Lu) were prepared by metathesis. Combining these with AlR3 and [Ph3C][B(C6F5)4] established a homogeneous Ziegler-Natta catalyst system, which exhibited high activities and excellent cis-1,4 selectivities for the polymerizations of butadiene and isoprene. Remarkably, such high cis-1,4 selectivity remained at elevated polymerization temperatures up to 80 1C and did not vary with the type of the central lanthanide element. Reactivity was inﬂuenced by the orthosubstituent of the N-aryl ring of the ligands and the bulkiness of the aluminum alkyls. The Ln-Al bimetallic cations were considered as the active species. These results shed new light on improving the catalytic performance of the conventional Ziegler-Natta catalysts for the speciﬁc selective polymerization of dienes.54 The ansa-metallocene complex (CpCMe2Flu)Nd(C3H5)(THF) is an eﬀective single-component catalyst for the production of syndiotactic styrene-rich co-polymer materials modiﬁed by isoprene and/or ethylene. The properties of the ﬁnal polymer could be tuned by altering the monomer ratio.55 The utility of aminobenzyl complexes as starting materials continues to burgeon. Hou and coworkers report that reaction of [Ln(CH2C6H4 NMe2-o)3] with pyrroles gives cationic complexes bearing mono(pyrrolyl) ligands which coordinate either through the nitrogen [27] or face on in Z5 fashion [28]. The coordination mode of the pyrrolyl ligands has a signiﬁcant inﬂuence on the polymerization of styrene.56 Broad patents have been applied for these systems.57

R

Cl Cl

t

Ti

NMe2

N NMe2

N

t

Bu

Sc NMe2

Sc

Ln Cl

Bu

NMe2

[26]

[27]

[28]

Building on their recent work with group 4 elements using their Cp-pendant pyrazolide hybrid ligands, Otero and coworkers have now turned their attention to rare earths. Simple methathesis of [Li(bpzcp)(THF)] [bpzcp=2,2134 | Organomet. Chem., 2010, 36, 121–147

bis(3,5-dimethylpyrazol-1-yl)-1,1-diphenylethylcyclopentadienyl], [Li(bpztcp) (THF)] [bpztcp=2,2-bis(3,5-dimethylpyrazol-1-yl)-1-tert-butylethylcyclopentadienyl], and the in situ-generated Li(bpzpcp) [bpzpcp=2,2-bis(3,5dimethylpyrazol-1-yl)-1-phenylethylcyclopentadienyl] with MCl3(THF)3 (M=Sc, Y) aﬀorded the group 3 halide compounds [MCl2(bpzcp)(THF)] (M=Sc Y, [MCl2(bpztcp)(THF)] (M=Sc; Y), and [MCl2(bpzpcp)(THF)] (M=Sc; Y). An aquo adduct of [YCl2(bpztcp)(H2O)] was isolated by deliberately leaving one of their compounds to stand in air. The hydrocarbyl complexes, [MR’2(bpzcp)] (M=Sc, Y) [29] undergo protonolysis to give the bis(aryloxide) derivatives [M(OAr)2(bpzcp)] (M=Sc, Y OAr=2,6-dimethylphenoxide; M=Y, OAr=3,5-dimethylphenoxide). Reaction of an aryloxide complex with moisture gave the tetranuclear cluster [{Y(bpzcp)} (m-OH)2(m3-OH){Y(OAr)2}]2. VTNMR experiments show that the aryloxides undergo rapid ﬂuxional exchange between coordinated and noncoordinated pyrazolyl rings. Interestingly, the starting chlorides can act as styrene polymerization catalysis in the presence of MAO.58 Ph Ph Ln N

N

N

N

CH2SiMe3 CH2SiMe3

[29]

Amines of varying Brønsted acidity and steric bulk have been investigated as chain-transfer agents to functionalize polyoleﬁns via organolanthanide-mediated oleﬁn polymerization processes. Ethylene polymerizations giving narrow product polydispersities were carried out using Cp*2LnR00 (Ln=La, Sm, Y, Lu) precatalysts in the presence of a variety of amines. Primary amines were the most inert toward Cp20 La-mediated polymerizations, aﬀording no detectable insertion products, while disec-butylamine and HN00 are marginally eﬃcient and produce monoethylene insertion products. In contrast, N-tert-butyl(trimethylsilyl)amine and di-isopropylamine aﬀord amine-capped oligoethylenes, while dicyclohexylamine is the most eﬃcient chain-transfer agent investigated, producing high molecular weight amine-terminated polyethylenes. In all of the above systems, protonolysis appears to be the dominant chain-transfer pathway, and this was shown to be well-behaved for NHCy2.59 Several series of half-sandwich complexes with diﬀerent Cp ligands have been prepared by protonolysis with [Ln(AlMe4)3] to give [Ln(AlMe4)2(CpR)]. For Cp 0 and Cp 0 0 , the reaction proceeds easily for the entire Ln(III) cation size range (Ln=Lu, Y, Sm, Nd, La), while the less reactive CpBu3 only reacted with larger metal centers Sm, Nd, and La and required heating. Treatment with Me2AlCl leads to the formation of varying amounts of partially and fully exchanged products [{Ln(AlMe4)(m-Cl)(CpR)}2] and [{Ln(m-Cl)2(CpR)}n]. Complexes [{Y(AlMe4)(m-Cl)Cp 0 }2] and [{Nd(AlMe4)-(m-Cl){1,2,4-(tBu)3C5H2}}2] have been characterized Organomet. Chem., 2010, 36, 121–147 | 135

structurally. All of the chlorinated half-sandwich complexes are inactive in isoprene polymerization. On the other hand, activation of the AlMe4 species with boron-containing cocatalysts, such as [Ph3C][B(C6F5)4], [PhNMe2H][B(C6F5)4], or B(C6F5)3, produces initiators leading to trans1,4-polyisoprene. Living systems were obtained. 60 In related work, reaction of [Cp*La(AlMe4)2] with B(C6F5)3 generates [Cp*La(AlMe3(C6F5) (m-(AlMe2(C6F5)2)]2. The authors are keen to point out the explosive nature of (AlMe2(C6F5)2) to keep the competition at bay.61 Metathesis of the borohydride ligand for cyclopentadienyl in a reaction with KCp or diorganomagnesium compounds gives half sandwich borohydride complexes [Mg(THF)6][Cp 0 Ln(BH4)3] (Cp 0QCp, Cp*C5H2Ph3, Ln=La, Nd) that are stable to redistribution. Combined with dialkylmagnesium, the Nd complexes provide useful catalysts for stereospeciﬁc isoprene polymerization.62 7.2

Polar monomers

Room temperature polymerization of MMA initiated by [Ln(BH4)3(thf)3] (Ln=Nd, Sm) or [Cp*2Sm(BH4)(thf)] proceeds at ambient temperature to give reasonably syndiotactic PMMA with high molecular weight but broad polydispersity. DFT calculations on a Cp model metallocene showed that in the reaction of [Cp2Eu(BH4)] with MMA, the borate [EuCp2{(OBH3) (OMe)CQC(Me)2}] is a key intermediate that is calculated to be exergonic and is the most likely of all of the possible products, other structures being kinetically disfavoured. Calculations suggested that other hydride and borohydrido complexes react by similar pathways.63 Two similar DFT studies are reported in Section 9. The organolanthanide(II) complexes with tetrahydro-2H-pyranyl- or N-piperidineethyl-functionalized ﬂuorenyl ligands [Z5:Z1-C5H9OCH2C13H8]2 Ln and [Z5:Z1-C5H10NCH2CH2C13H8]2Ln (Ln=Eu, Yb) were prepared and used in both CL and LA polymerization experiments.64 8. 8.1

Synthesis and characterization of new compounds Compounds with boron

Boron-based ancillaries continue to be used, and in particular the chemistry of neutral borabenzene has generated a nice little surprise this year. Reaction of the phosphine adduct of borabenzene, C5H5B-PMe3, with YbN002 in toluene gave crystals of an ansa-heteroborabenzene lanthanide amide [30] which results from an unexpected C-H bond cleavage believed to result from deprotonation of a methyl group of PMe3. Addition of a carbodiimide to the product gave the corresponding insertion product.65 Several solvent-free boratabenzene yttrium chlorides have also been made. [(C5H5B–E)2YCl]2 (E=NPh2, NEt2) can be prepared by reaction of K(C5H5B–E) with YCl3. Reaction with KNR002 gives the expected boratabenzene yttrium amides which react further with KNR002, resulting in p-ligand displacement and isolation of [Y(NR002)3]. The boratabenzene yttrium amides showed good catalytic activities for intramolecular hydroamination.66 In analogous chemistry the divalent Yb boratabenzene 136 | Organomet. Chem., 2010, 36, 121–147

complex (C5H5B–NPh2)2Yb(THF)2 was synthesized. It reacts by electron transfer with the diimine PhNC(Me)C(Me)NPh in toluene to aﬀord a trivalent Yb complex, (C5H5B–NPh2)2[PhNC(Me)C(Me)NPh]. The redox process is solvent sensitive and reversible.67 Using Cp-like substituted azaborolyl ligands, complexes of Y have been reported. Reaction of YCl3(THF)3 with 2 equivalents of LLi (L=1-methyl2-phenyl-1,2-azaborolyl) in THF gave L2YCl as a mixture of meso and racemic isomers in a 1:3 molar ratio. Subsequent alkylation with LiR 0 gave rise to YL2R 0 (THF) (meso:racemic=1:3). It was possible to isolate the pure racemic isomer by repeated recrystallization from a hydrocarbon solvent. A protonolysis route was also reported, though with a diﬀerent ratio of isomers. The hydrocarbyls were found to initiate MMA living polymerization yielding syndiotactic-rich PMMA.68 In continuing work, s-caboranyl-Z5-indenyl constrained geometry lanthanide complexes have been prepared and structurally characterized. Di-deprotonation of 1-(3-indenyl)-o-carborane 1-(3-Ind)-1,2-C2B10H11 and reaction with LnCl3 in THF generated the ionic lanthanum complex [K(THF)6][[Z5:s-(Ind)C2B10H10]2La(THF)]. Smaller, late lanthanides and yttrium gave chloride-bridged dipotassium-dilanthanide complexes [[Z5:s-(Ind)C2B10H10]Ln(THF)2(m-Cl)2K(THF)2]2 (Ln=Y, Gd, Er, Yb). Metathesis reactions with NaCp or KCH2C6H4-o-NMe2 aﬀorded the corresponding salt metathesis products [Z5:s-(Ind)C2B10H10]Ln(Z5-Cp) (THF)2 (Ln=Y, Gd) or [[Z5:s-(Ind)C2B10H10]Ln(CH2C6H4-o-NMe2) (DME)] (Ln=Y, Er), [31] respectively. The latter were also isolated via protonolysis of 1-indenyl-1,2-carborane with Ln(CH2C6H4-o-NMe2)3.69

B Me P Me

O

SiMe3 Yb

N SiMe3

B

C

Ln

O

C NMe2

[30]

8.2

[31]

Compounds with S, P donor functional groups

Cp* supporting ligands were used to prepare Ln and U compounds that served to address the nature of the Ln-S and U-S bond. The lanthanide compounds [K(THF)2Ln(Cp*)(SBT)3] were obtained by treating [Ln(BH4)3 (THF)3] with KSBT and KCp*; isomorphous crystals of [K(15-crown5)2][Ln(Cp*)(SBT)3] [Ln=La, Ce, Nd] were formed upon addition of 15crown-5. Comparison with isostructural U(III) analogs show that the U-N bonds are shorter than expected, given metals with identical ionic radii. A DFT analysis suggests that the structural diﬀerences in the series of [M(Cp*)(SBT)3] anions are related to the uranium 5f orbital-ligand mixing, which is greater than the lanthanide 4f orbital-ligand mixing. Moreover, the consideration of the corresponding bond orders and the analysis of the bonding energy bring to light a strong and speciﬁc interaction between the uranium and apical nitrogen atoms.70 Organomet. Chem., 2010, 36, 121–147 | 137

The eight coordinate phospholyl compounds complexes [M(Z5-P3C2 But2)2(Z2-P3C2But2)] (M=Sc, Y, Tm, and U) [32] were prepared in moderate yields. In the solid state the complexes, which are all iso-structural, display an interesting assembly of ligands comprising one Z2-(bent) and two Z5-ligated triphospholyl rings. NMR studies indicate that the complexes are ﬂuxional in solution.71 Similarly, the 1,2,4-diazaphospholide (dp-) and and 3,5-R2-2,4-diazaphospholide (R2dp, R=Ph, t-Bu) samarium complexes including [Sm(Ph2dp)3(THF)3], [Sm(t-Bu2dp)3(THF)2], [Sm4(Ph2dp)8Cl2O] [33] and [Sm2(Ph2dp)6] were prepared. The chemistry is broadly analogous to pyrazolide chemistry reported in this series in previous years.72 8.3

Ln-aromatic interactions

In chemistry that harks back to the early experiments by Cotton, two new lanthanum(III) chloroaluminate complexes with neutral arenes, [La(Z6C6H5Me)(AlCl4)3] and [La(Z6-C6Me6)(AlCl4)3], as well as the ﬁrst lanthanide chlorogallate complex [La(Z6-C6Me6)(GaCl4)3] were prepared. The coordination polyhedron around the central lanthanum ions are distorted pentagonal bipyramids formed by six Cl-atoms and an Z6-bound arene.73 Another example was observed in one of the clusters [6] reported by Anwander discussed in Section 2.1.1.6 8.4

Miscellaneous compounds

In ‘‘simple’’ hydrocarbyl chemistry [Sc(CH2Ph)3(THF)3] was shown to adopt a distorted fac-octahedral geometry in the solid state. One of the coordinated THF molecules could be removed by trituration with toluene to give [Sc(CH2Ph)3(THF)2]. In contrast, at –1961 ScCl3 and KCH2Ph led to a two-dimensional coordination polymer [Sc(CH2Ph)5K2(THF)3]n, in which the Sc atom is surrounded by ﬁve benzyl groups in a trigonal bipyramidal fashion. With the larger LuCl3 as the starting material, [Lu(CH2Ph)3(THF)3] was obtained, which also lost a coordinated THF ligand to give [Lu(CH2Ph)3(THF)2].74 This chemistry should be compared with that explored by Hou with the dimethylaminobenzyl ligand which appears to be a more convenient starting material.39 Phosphonium bis(ylide) rare-earth complexes Cp*M(byl)2 (M=Sc, Y, Dy, Yb and Lu; byl=[Ph2P(CH2)2] ) were synthesized. The structure of three complexes (M=Dy, Yb and Lu) showed the chelating coordination mode for byl with the two smaller Ln and as a bridging ligand in dimeric [Cp*Dy(byl)2]2 [34].75 A tridentate [N,N,O] Schiﬀ base [3,5-But 2-2-(OH)-C6H2CHQN-8C9H6N] (LH) was prepared, and the corresponding Na and Yb complexes were synthesized and characterized. Reaction of LH with NaH in THF at room temperature aﬀorded the Na salt of the Schiﬀ base as a dimer [{LNa(THF)}2] that reacted with YbCl3 in THF to give the monomeric Yb Schiﬀ base dichloride complex, [LYbCl2(DME)]. Reaction of this compound with NaMeCp in THF gave the expected product [LYb (MeCp)Cl(THF)] in a good isolated yield and with 2 equivalents of ArONa 138 | Organomet. Chem., 2010, 36, 121–147

(ArO=OC6H3-But2-2,6) in THF to form the desired solvent-free Yb aryloxide [LYb(OAr)2].76 In some rather neat organometallic chemistry, new Sc and Y tetramethyl aluminate complexes supported by a ferrocene 1,1 0 diamide ligand, [35] have been shown not only to ring-open and link methylimidazole groups, but also to couple pairs of pyridines to generate bipy complexes. The chemistry proceeds by a series of s-bond metathesis steps.77 Treatment of the imido/nitrido titanium complex [{TiCp*(m-NH)}3(m3N)] with Y and Er halide complexes [MCl3(THF)3.5] and [MCpCl2(THF)3] gives cubic clusters [Cl3M{(m3-NH)3Ti3Cp*3(m3-N)}] and [CpCl2M{(m3NH)3Ti3Cp*3(m3-N)}]. An analogous reaction with [{MCp2Cl}2] in toluene aﬀords [Cp3M(m-Cl)ClCpM{(m3-NH)3Ti3Cp*3(m3-N)}] (M=Y, Er) [36]. Imido groups are present but have little to do with the rare earth.78 Ph

But

P P

P But P

But P

N

P

P

But

Bu

P

t

N

N

P

[32]

Cl

Ph

Sm

[33]

Ph

P Ph

Ph

Ph

Dy

Ph

Ph

P N

N Ph P

P

P

N

N

N

Ph P

Dy

Ph

Ph

N

Ph

P

Sm

O

P Ph

Ph N

Cl

N

Ph

N

Sm

Sm

Ph

Ph

N

N Ph

Ph

N

P

N

N

Ph P

But

P

Ph

P

Ph

Ph

[34]

A Ln-C interaction was noted involving the central carbon in the backbone of the ancillary ligand in the cationic b-diphosphiniminate complexes [[Ph2P(=NSiMe3)CHPPh2(=NSiMe3)-kN,kN 0 ,kC]Eu(THF)3][BPh4] [37], [(L)Eu(THF)3] and [(L)Eu(THF)[N(PPh2)2-kN,kP]].79 N SitBuMe2

NH

N

Me Fe

Ti

Ti

Sc

Me

Al

Me N

Me

NH

HN Ln Cl

SitBuMe2

[35]

SiMe3

Ti

[36]

Ph2P

N

H

Cl

THF Eu THF

Ph2P

Ln

N

+ BPh4–

THF SiMe3

[37]

Organomet. Chem., 2010, 36, 121–147 | 139

The reaction of the pentalene salt C8H4{SiiPr3-1,4}2[K]2 with [SmCp*(m-I) (THF)2]2 yields the unanticipated Sm(III) sandwich complexes [Sm(Z8C8H4{SiiPr3-1,4}2)(Z5-Cp*)] and [Sm(Z8-C8H4{SiiPr3-1,4}2)(Z5-C8H5{SiiPr31,4}2)] [38] and the mixed-valence cluster [Cp*6Sm6(OMe)8O][K(THF)6] [39] via solvent activation of THF. The Sm(III) sandwich compound incorporates an Z8-pentalene ligand and an Z5-hydropentalenyl ligand. The Sm(II)/ Sm(III) mixed-valence cluster compound contains a centrosymmetric hexanuclear array of Cp*Sm units, bridged by face-centered m3-methoxy groups, with a central oxo unit.80 An unsymmetrical Pr sandwich complex containing the donor-functionalized 9-(2-methoxyethyl)ﬂuorenyl ligand, (COT)Pr(C13H8CH2CH2OMe)(THF) was prepared by treatment of dimeric [(COT)Pr(m-Cl)(THF)2]2 with KC13H8CH2CH2OMe.81 The reactions between (CpBu3)2CeH and several hydroﬂuorobenzene derivatives have been investigated. Aryl derivatives were the primary products, (CpBu3)2Ce(C6H5-xFx) (x=1,2,3,4), with the majority decomposing at diﬀerent rates to give (CpBu3)2CeF and a ﬂuorobenzyne. The latter is trapped by either solvent when C6D6 was used or by a CpBu3 ring when C6D12 was the solvent. The aryl derivatives were generated cleanly by reaction of the tucked-in metallacycle, [(CpBu3)(tBu)2C5H2C(Me2)CH2))Ce] [40], with a hydroﬂuorobenzene. The thermodynamic isomer has been shown to be the one in which the CeC bond is ﬂanked by two ortho-CF bonds. This orientation is suggested to arise from the negative charge that is localized on the ipso-C atom due to C(d þ )F(d-) polarization. The more thermodynamically stable regioisomer is formed in the solid-state at 251 Bu3 over two months, converting CpBu3 2 Ce(2,3,4,5-C6HF4) to Cp2 Ce (2,3,4,6-C6HF4), an isomerization that involves a CeC(ipso) for C(ortho)F site exchange.82

SiiPr3 OH

Sm

OH

CMe3

OH i

(Pr )3Si (Pri)3Si

Sm

Sm

Sm

Me3C

O H

SiiPr3

[38]

8.5

OH

OH

Sm OH

OH

Sm OH

Me3C

CMe2

Ce CH2 CMe3

Me3C

Sm

[39]

[40]

Gas phase chemistry

The gas-phase reactivity of Ln2 þ with alkanes (methane, ethane, propane, n-butane) and alkenes (ethene, propene, 1-butene) was studied by Fourier transform ICR mass spectrometry. The reaction products consisted of different combinations of doubly charged organometallic ions-adducts or 140 | Organomet. Chem., 2010, 36, 121–147

species formed via metal ion induced H, dihydrogen, alkyl, or alkane eliminations from the hydrocarbons and singly charged ions that resulted from electron, hydride, or methide transfers from the hydrocarbons to the metal ions. The only lanthanide dications capable of activating the hydrocarbons to form doubly charged organometallic ions were La, Ce, Gd, and Tb, which have ground-state or energetically accessible d1 electronic conﬁgurations. Lu, with an accessible d1 electronic conﬁguration but a rather high electron aﬃnity, reacted only through transfer channels. The remaining Ln reacted via transfer channels or adduct formation. The diﬀerent accessibilities of d1 electronic conﬁgurations and the range of electron afﬁnities of the Ln2 þ ions allowed for a detailed analysis of the trends for Ln2 þ reactivity and the conditions for occurrence of bond activation, adduct formation, and electron, hydride, and methide transfers.83 Laser-ablated La atoms were codeposited at 4 K with acetylene in excess Ar. The products La(C2H2), LaCCH2, HLaCCH, and La2(C2H2), were all characterized using IR spectroscopy. DFT calculations gave calculated vibrational frequencies, relative absorption intensities, and isotopic shifts that supported the identiﬁcation of these molecules from the matrix IR spectra.84 Finally, the perennial multilayer lanthanide-COT organometallic clusters, Lnn(COT)m (Ln=Eu, Tb, Ho, Tm; n=1–7; m=n 1, n, n þ 1) were produced by a laser vaporization synthesis method and the magnetic deﬂections of these organometallic sandwich clusters were measured. Most of the sandwich species displayed one-sided deﬂection, while some of smaller clusters showed symmetric broadening without or with only very small (or absent) net high-ﬁeld deﬂection. In general, the total magnetic moments, calculated from the magnitude of the beams deﬂections, increase with the number of lanthanide atoms (i.e., with increasing sandwich layers); however for Tb, Ho, and Tm clusters with nW3, a suppression of the magnetic moments was observed, possibly through antiferromagnetic interactions. For Eu a linear increase of the magnetic moments with the number of Eu atoms up to n=7 was noted, with moments similar to those displayed by conventionally synthesized mononuclear EuC8H8 complexes. These results suggest that Eun(C8H8)nþ 1 is a promising candidate for a high-spin, onedimensional building block in organometallic magnetic materials.85 9.

Theory

Ab initio calculations have been used to explore how the magnetization direction in organic magnetic molecules can be changed by altering their oxidation state. One study considered the molecule Eu2(COT)3 which is predicted to show strong ferromagnetism due to a hole-mediated exchange mechanism. In spite of the air instability of these systems, exciting new applications were predicted by the authors.86 There have been a couple of studies on triple decker benzene complexes of scandium and related transition metals M2(C6H6)3 (M=Sc, Ti, V). SternGerlach measurements have shown that the magnitudes of their magnetic moments are reduced from their spin-only values by a factor of 1/4 for M=Sc and Ti, and 3/5 for V.87 Ab initio calculations have given insight into Organomet. Chem., 2010, 36, 121–147 | 141

how the magnetic structure of such compounds can be tailored to favour ferro- or antiferromagnetic behaviour.88 In mechanistic work, DFT (B3PW91) calculations were carried out to understand the nature of the mechanism of the hydromethylation of propene and isobutene using H2Si(C5H4)2ScCH3 as a model. The calculations suggest that the hydromethylation of isobutene would be catalytic, but not that of propene. Since oleﬁns bind more strongly than methane, propene would prevent coordination of methane, especially in the case of the more open ansa complexes. Oleﬁns have greater coordination strengths than methane unless steric factors dominate. Isobutene on the other hand allows methane to compete.89 The mechanisms of polymerization of e-caprolactone (ECL) initiated by either the rare-earth hydride [Cp2Eu(H)] or the borohydrides [Cp2Eu(BH4)] or [(N2NN 0 )Eu(BH4)] have been studied at the DFT level (N2NN 0 = (2-C5H4N)CH2N(CH2CH2NMe)2). For all compounds the reaction proceeds in two steps: hydride transfer from the initiator to the carbonyl carbon of the lactone, followed by ring-opening of the monomer. In the last step a diﬀerence is observed between the hydride and borohydride complexes, because for the latter the ring-opening is induced by an additional B-H bond cleavage leading to a terminal -CH2OBH2 group.90 A particularly thoughtful computational study of chain transfer mechanisms in oleﬁn polymerization catalysis suggests that two distinct pathways exist for b-hydrogen of a hydrogen from the growing chain to a second oleﬁn, the classical path involving a M?H interaction and a direct transfer in which the metal does not participate directly. Whether a catalyst will display one or the other mechanism is determined by subtle eﬀects that are discussed in detail in the paper.91 10.

Spectroscopy/electronic structure

As part of a continuing series, the absorption spectra (in the IR/NIR/Vis/ UV range) of Ln(C5Me4H)3 (Ln=La, Ce) were analyzed in gory detail. The diﬀerence between the experimental energies of the barycenters of CF levels of the multiplets 2F7/2 and 2F5/2 is larger than in the gaseous free Ce3 þ ion is referred to as an ‘‘anti’’-relativistic nephelauxetic eﬀect, and is explained by coupling eﬀects of these multiplets via the CF, resulting in lower spin-orbit coupling parameters.92 Related work on Pr(C5Me4H)3 has also been carried out. Here, polarized optical measurements of oriented single crystals were made at room temperature. To separate ‘‘cold’’ and ‘‘hot’’ f-f-transitions and nC H combination vibrations, the absorption spectra of KBr pellets were obtained at temperatures down to 77 K. Additionally, MCD and polarized Raman spectra were recorded as well allowing the usual ﬁt to an empirical Hamiltonian. The data were compared with previous Xa calculations.93 Synthetic routes leading to two series of (COT)lanthanide(III) scorpionate ‘‘mixed sandwich’’ complexes were reported. The early lanthanide derivatives (COT)Ln(Tp) and (COT)Ln(TpMe2) (Ln=Ce, Pr, Nd, Sm) were obtained by reacting the dimeric halide precursors [(COT)Ln(m-Cl)(THF)]2 with KTp or KTpMe2. For small lanthanide ions (COT)Ln(Tp) (Ln=Er, 142 | Organomet. Chem., 2010, 36, 121–147

Lu) were made by the reaction of (Tp)LnCl2(THF)1.5 with K2(COT). Several structures were obtained. Optical spectra were run at room and low temperatures to reveal the underlying crystal ﬁeld splitting patterns of complexes and these were compared with previous studies on actinides.94 The curious story of metallocenes bound to non-innocent ligands continues to develop. Work has now been extended to samarium with Cp*2Sm(tpy) (tpy=2,2 0 :6 0 ,2 0 0 -terpyridine) and its 1-electron oxidized congener [Cp*2Sm(tpy)]PF6 having been prepared. Data for the neutral complex indicates that the ground state electronic conﬁguration is Sm(III) tpyd i.e. [(4f)5-(p*)1] similar to that found previously for the Yb analogue, and there are few signiﬁcant structural implications. The redox potentials are also consistent with established trends. The optical spectra are also very similar, but there does appear to be a red shift (B400 cm 1) for the Sm complex relative to Yb, suggesting slightly greater stabilization of the p* level(s) in the Sm(III) complex. The magnetism is consistent with this and is otherwise unremarkable.95 Acknowledgements JGB acknowledges support from the US National Science Foundation (CHE-0747165). Abbreviation Ln R0 R00 Cp MeCp Cp* Cp 0 Cp00 Cpw Cp* 0 CpBu CpBu2 CpBu3 Ind Flu tmp COT COT00 TMEDA TMS DAD HMPA DME MMA MAO DFT

lanthanide CH2SiMe3 CH(SiMe3)2 C5H5 C5H4Me Cp* C5H4(SiMe3) 1,3–C5H3(SiMe3)2 1,2,4–C5H2(SiMe3)3 C5Me4SiMe3 C5H4But 1,3–C5H3But2 1,2,4-But3C5H2 Z–C9H7, indenyl Z–C13H8, ﬂuorenyl Z-C4Me4P C8H8 1,4-C8H6(SiMe3)2 tetramethylethylenediamine(1,2-bis(dimethylamino)ethane) SiMe3 diazabutadiene OP(NMe2)3 CH3OCH2CH2OCH3 methylmethacrylate methylaluminoxane density functional theory Organomet. Chem., 2010, 36, 121–147 | 143

References 1 B. Liu, X. Liu, D. Cui and L. Liu, Organometallics, 2009, 28, 1453–1460. 2 K. D. Conroy, W. E. Piers and M. Parvez, J. Organomet. Chem., 2008, 693, 834–846. 3 D. M. Lyubov, G. K. Fukin, A. V. Cherkasov, A. S. Shavyrin, A. A. Trifonov, L. Luconi, C. Bianchini, A. Meli and G. Giambastiani, Organometallics, 2009, 28, 1227–1232. 4 J. Scott, F. Basuli, A. R. Fout, J. C. Huﬀman and D. J. Mindiola, Angew. Chem., Int. Ed., 2008, 47, 8502–8505. 5 C. S. Tredget, E. Clot and P. Mountford, Organometallics, 2008, 27, 3458– 3473. 6 L. C. H. Gerber, E. Le Roux, K. W. Tornroos and R. Anwander, Chem.–Eur. J., 2008, 14, 9555–9564. 7 J. Scott, H. Fan, B. F. Wicker, A. R. Fout, M.-H. Baik and D. J. Mindiola, J. Am. Chem. Soc., 2008, 130, 14438–14439. 8 P. L. Arnold, S. Zlatogorsky, N. A. Jones, C. D. Carmichael, S. T. Liddle, A. J. Blake and C. Wilson, Inorg. Chem., 2008, 47, 9042–9049. 9 Y. Yang, P. M. Gurubasavaraj, H. Ye, Z. Zhang, H. W. Roesky and P. G. Jones, J. Organomet. Chem., 2008, 693, 1455–1461. 10 H.-M. Sommerfeldt, C. Meermann, M. G. Schrems, K. W. Toernroos, N. A. Froystein, R. J. Miller, E.-W. Scheidt, W. Scherer and R. Anwander, Dalton Trans., 2008, 1899–1907. 11 A. Venugopal, A. Hepp, T. Pape, A. Mix and N. W. Mitzel, Dalton Trans., 2008, 6628–6633. 12 R. Qi, B. Liu, X. Xu, Z. Yang, Y. Yao, Y. Zhang and Q. Shen, Dalton Trans., 2008, 5016–5024. 13 C. T. Carver, M. J. Monreal and P. L. Diaconescu, Organometallics, 2008, 27, 363–370. 14 S. Bambirra, F. Perazzolo, S. J. Boot, T. J. J. Sciarone, A. Meetsma and B. Hessen, Organometallics, 2008, 27, 704–712. 15 F. Han, J. Zhang, Y. Han, Z. Zhang, Z. Chen, L. Weng and X. Zhou, Inorg. Chem., 2009, 48, 1774–1781. 16 W. J. Evans, E. Montalvo, D. J. Dixon, J. W. Ziller, A. G. DiPasquale and A. L. Rheingold, Inorg. Chem., 2008, 47, 11376–11381. 17 J. D. Masuda, K. C. Jantunen, B. L. Scott and J. L. Kiplinger, Organometallics, 2008, 27, 1299–1304. 18 J. D. Masuda, K. C. Jantunen, B. L. Scott and J. L. Kiplinger, Organometallics, 2008, 27, 803–806. 19 T. Shima and Z. Hou, Chem. Lett., 2008, 37, 298–299. 20 M. V. Butovskii, O. L. Tok, F. R. Wagner and R. Kempe, Angew. Chem., Int. Ed., 2008, 47, 6469–6472. 21 W. J. Evans, E. Montalvo, T. M. Champagne, J. W. Ziller, A. G. DiPasquale and A. L. Rheingold, J. Am. Chem. Soc., 2008, 130, 16–17. 22 R. Liu, P. Zheng, L. Weng, X. Zhou and C. Liu, J. Organomet. Chem., 2008, 693, 1614–1620. 23 W. J. Evans, E. Montalvo, T. M. Champagne, J. W. Ziller, A. G. DiPasquale and A. L. Rheingold, Organometallics, 2008, 27, 3582–3586. 24 M. Zeuner, S. Pagano and W. Schnick, Chem.–Eur. J., 2008, 14, 1524–1531. 25 D. Cui, M. Nishiura, O. Tardif and Z. Hou, Organometallics, 2008, 27, 2428–2435. 26 J. Zhang, Y. Han, F. Han, Z. Chen, L. Weng and X. Zhou, Inorg. Chem., 2008, 47, 5552–5554. 144 | Organomet. Chem., 2010, 36, 121–147

27 M. U. Kramer, D. Robert, S. Arndt, P. M. Zeimentz, T. P. Spaniol, A. Yahia, L. Maron, O. Eisenstein and J. Okuda, Inorg. Chem., 2008, 47, 9265– 9278. 28 V. F. Quiroga Norambuena, A. Heeres, H. J. Heeres, A. Meetsma, J. H. Teuben and B. Hessen, Organometallics, 2008, 27, 5672–5683. 29 J. Sun, D. J. Berg and B. Twamley, Organometallics, 2008, 27, 683–690. 30 C. Ruspic, J. R. Moss, M. Schuermann and S. Harder, Angew. Chem., Int. Ed., 2008, 47, 2121–2126. 31 G. B. Deacon, C. M. Forsyth, F. Jaroschik, P. C. Junk, D. L. Kay, T. Maschmeyer, A. F. Masters, J. Wang and L. D. Field, Organometallics, 2008, 27, 4772–4778. 32 W. J. Evans, B. M. Schmiege, S. E. Lorenz, K. A. Miller, T. M. Champagne, J. W. Ziller, A. G. Di Pasquale and A. L. Rheingold, J. Am. Chem. Soc., 2008, 130, 8555–8563. 33 W. J. Evans, T. J. Mueller and J. W. Ziller, J. Am. Chem. Soc., 2009, 131, 2678–2686. 34 I. Aillaud, K. Wright, J. Collin, E. Schulz and J.-P. Mazaleyrat, Tetrahedron: Asymmetry, 2008, 19, 82–92. 35 I. Aillaud, D. Lyubov, J. Collin, R. Guillot, J. Hannedouche, E. Schulz and A. Trifonov, Organometallics, 2008, 27, 5929–5936. 36 S. Ge, A. Meetsma and B. Hessen, Organometallics, 2008, 27, 5339–5346. 37 Z. Du, W. Li, X. Zhu, F. Xu and Q. Shen, J. Org. Chem., 2008, 73, 8966–8972. 38 Q. Shen, Z. Du, X. Chen and X. Zhu, Application: CN Pat., 2008-10019946 101264456, 2008 39 W.-X. Zhang, M. Nishiura, T. Mashiko and Z. Hou, Chem.–Eur. J., 2008, 14, 2167–2179. 40 A. Agosti, S. Britto and P. Renaud, Org. Lett., 2008, 10, 1417–1420. 41 W.-X. Zhang, M. Nishiura and Z. Hou, Angew. Chem., Int. Ed., 2008, 47, 9700–9703. 42 B. Liu and D. Cui, Dalton Trans., 2009, 550–556. 43 R. Litlaboe, M. Zimmermann, K. Saliu, J. Takats, K. W. Toernroos and R. Anwander, Angew. Chem., Int. Ed., 2008, 47, 9560–9564. 44 S. Schamm, P. E. Coulon, S. Miao, S. N. Volkos, L. H. Lu, L. Lamagna, C. Wiemer, D. Tsoutsou, G. Scarel and M. Fanciulli, J. Electrochem. Soc., 2009, 156, H1–H6. 45 P. Majumder, G. Jursich, A. Kueltzo and C. Takoudis, J. Electrochem. Soc., 2008, 155, G152–G158. 46 M. Losurdo, M. M. Giangregorio, P. Capezzuto, G. Bruno, G. Malandrino, I. L. Fragala, L. Armelao, D. Barreca and E. Tondello, J. Electrochem. Soc., 2008, 155, G44–G50. 47 M. M. Giangregorio, A. Sacchetti, M. Losurdo, P. Capezzuto and G. Bruno, J. Non-Cryst. Solids, 2008, 354, 2853–2857. 48 M. Losurdo, M. M. Giangregorio, P. Capezzuto, G. Bruno, R. G. Toro, G. Malandrino, I. L. Fragala, L. Barreca, E. Tondello, A. A. Suvorova, D. Yang and E. A. Irene, Adv. Funct. Mater., 2008, 18, 10. 49 W.-S. Kim, T.-S. Kim, B.-W. Kang, M.-G. Ko, S.-K. Park and J.-W. Park, J. Vac. Sci. Technol., B: Microelectron. Nanometer Struct.--Process., Meas., Phenom., 2008, 26, 1588–1591. 50 W. He, D. S. H. Chan, S.-J. Kim, Y.-S. Kim, S.-T. Kim and B. J. Cho, J. Electrochem. Soc., 2008, 155, G189–G193. 51 N. Ajellal, E. Guillevic, C. M. Thomas, R. Jackstell, M. Beller and J.-F. Carpentier, Adv. Synth. Catal., 2008, 350, 431–438. Organomet. Chem., 2010, 36, 121–147 | 145

52 53 54 55 56 57 58

59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82

Q. Ban, CN Pat., 2007-10015195, 101121732, 2008 Q. Ban, CN Pat., 2007-10015194, 101117342, 2008 W. Gao and D. Cui, J. Am. Chem. Soc., 2008, 130, 4984–4991. A.-S. Rodrigues, E. Kirillov, B. Vuillemin, A. Razavi and J.-F. Carpentier, Polymer, 2008, 49, 2039–2045. M. Nishiura, T. Mashiko and Z. Hou, Chem. Commun. (Cambridge, U. K.), 2008, 2019–2021. M. Nishiura, T. Masuko and Z. Hou, JP Pat., 2007-60121, 2008222780, 2008 A. Otero, J. Fernandez-Baeza, A. Lara-Sanchez, A. Antinolo, J. Tejeda, E. Martinez-Caballero, I. Marquez-Segovia, I. Lopez-Solera, L. F. SanchezBarba and C. Alonso-Moreno, Inorg. Chem., 2008, 47, 4996–5005. S. B. Amin, S. Seo and T. J. Marks, Organometallics, 2008, 27, 2411–2420. M. Zimmermann, K. W. Toernroos and R. Anwander, Angew. Chem., Int. Ed., 2008, 47, 775–778. M. Zimmermann, K. W. Tornroos, H. Sitzmann and R. Anwander, Chem.-Eur. J., 2008, 14, 7266–7277. M. Visseaux, P. Zinck, M. Terrier, A. Mortreux and P. Roussel, J. Alloys Compd., 2008, 451, 352–357. N. Barros, M. Schappacher, P. Dessuge, L. Maron and S. M. Guillaume, Chem.–Eur. J., 2008, 14, 1881–1890. Y. Wei, Z. Yu, S. Wang, S. Zhou, G. Yang, L. Zhang, G. Chen, H. Qian and J. Fan, J. Organomet. Chem., 2008, 693, 2263–2270. P. Cui, Y. Chen, G. Li and W. Xia, Angew. Chem., Int. Ed., 2008, 47, 9944–9947. Y. Yuan, Y. Chen, G. Li and W. Xia, Organometallics, 2008, 27, 6307–6312. P. Cui, Y. Chen, G. Wang, G. Li and W. Xia, Organometallics, 2008, 27, 4013–4016. X. Fang, Y. Deng, Q. Xie and F. Moingeon, Organometallics, 2008, 27, 2892–2895. H. Shen, H.-S. Chan and Z. Xie, Organometallics, 2008, 27, 5309–5316. M. Roger, L. Belkhiri, T. Arliguie, P. Thuery, A. Boucekkine and M. Ephritikhine, Organometallics, 2008, 27, 33–42. G. K. B. Clentsmith, F. G. N. Cloke, M. D. Francis, J. R. Hanks, P. B. Hitchcock and J. F. Nixon, J. Organomet. Chem., 2008, 693, 2287–2292. C. Pi, L. Wan, Y. Gu, W. Zheng, L. Weng, Z. Chen and L. Wu, Inorg. Chem., 2008, 47, 9739–9741. A. S. Filatov, A. Y. Rogachev and M. A. Petrukhina, J. Mol. Struct., 2008, 890, 116–122. N. Meyer, P. W. Roesky, S. Bambirra, A. Meetsma, B. Hessen, K. Saliu and J. Takats, Organometallics, 2008, 27, 1501–1505. K. A. Rufanov and A. Spannenberg, Mendeleev Commun., 2008, 18, 32–34. B.-Y. Li, Y.-M. Yao, Y.-R. Wang, Y. Zhang and Q. Shen, Polyhedron, 2008, 27, 709–716. C. T. Carver and P. L. Diaconescu, J. Am. Chem. Soc., 2008, 130, 7558–7559. J. Caballo, M. Garcia-Castro, A. Martin, M. Mena, A. Perez-Redondo and C. Yelamos, Inorg. Chem., 2008, 47, 7077–7079. M. Wiecko and P. W. Roesky, Organometallics, 2009, 28, 1266–1269. O. T. Summerscales, D. R. Johnston, F. G. N. Cloke and P. B. Hitchcock, Organometallics, 2008, 27, 5612–5618. A. Zaeni, F. Olbrich, A. Fischer and F. T. Edelmann, J. Organomet. Chem., 2008, 693, 3791–3796. E. L. Werkema and R. A. Andersen, J. Am. Chem. Soc., 2008, 130, 7153–7165.

146 | Organomet. Chem., 2010, 36, 121–147

83 J. Marcalo, M. Santos, A. Pires de Matos, J. K. Gibson and R. G. Haire, J. Phys. Chem. A, 2008, 112, 12647–12656. 84 Y.-L. Teng and Q. Xu, J. Phys. Chem. A, 2008, 112, 10274–10279. 85 K. Miyajima, M. B. Knickelbein and A. Nakajima, J. Phys. Chem. A, 2008, 112, 366–375. 86 N. Atodiresei, P. H. Dederichs, Y. Mokrousov, L. Bergqvist, G. Bihlmayer and S. Blugel, Phys. Rev. Lett., 2008, 100, 117207/117201–117207/117204. 87 A. Goto and S. Yabushita, Chem. Phys. Lett., 2008, 454, 382–386. 88 H. Weng, T. Ozaki and K. Terakura, J. Phys. Soc. Jpn., 2008, 77, 064301/ 064301–064301/064308. 89 N. Barros, O. Eisenstein, L. Maron and T. D. Tilley, Organometallics, 2008, 27, 2252–2257. 90 N. Barros, P. Mountford, S. M. Guillaume and L. Maron, Chem.–Eur. J., 2008, 14, 5507–5518. 91 G. Talarico and P. H. M. Budzelaar, Organometallics, 2008, 27, 4098–4107. 92 H.-D. Amberger and H. Reddmann, Z. Anorg. Allg. Chem., 2008, 634, 173–180. 93 H.-D. Amberger and H. Reddmann, Z. Anorg. Allg. Chem., 2008, 634, 1542–1554. 94 H.-D. Amberger, F. T. Edelmann, J. Gottfriedsen, R. Herbst-Irmer, S. Jank, U. Kilimann, M. Noltemeyer, H. Reddmann and M. Schaefer, Inorg. Chem., 2009, 48, 760–772. 95 J. M. Veauthier, E. J. Schelter, C. N. Carlson, B. L. Scott, R. E. Da Re, J. D. Thompson, J. L. Kiplinger, D. E. Morris and K. D. John, Inorg. Chem., 2008, 47, 5841–5849.

Organomet. Chem., 2010, 36, 121–147 | 147

Alkali/coinage metals – organolithium, organocuprate chemistry Volker Engelsa and Andrew E. H. Wheatleya DOI: 10.1039/9781847559616-000148

In Part 1 of this chapter, the alkali metal coordination compounds are reviewed, starting with mixed-metal ZnLi species, the interest in which has grown out of their synthetic importance as bases, and metallocene systems, including cyclopentadienyl derivatives. Cases of extreme interest notwithstanding, discussion is limited to compounds that contain at least one carbon-alkali metal interaction. Part 2 provides an overview of the latest developments in coinage metal organometallic chemistry. Aiming to reﬂect the balance between diﬀerent areas of the most recent research, a review of copper-based metal-organic frameworks and coordination polymers is followed by compounds of more general interest. Similarly, for silver and gold, polymeric structures and coordination frameworks are described, along with carbene complexes and, for gold, phosphine complexes. As for Part 1, the emphasis is placed on systems that contain at least one carbon-metal interaction. The analytical discussion focuses on solid-state investigations and, where appropriate, applications are mentioned along with structural results.

1. 1.1

The alkali metals Lithium, sodium and potassium compounds

Previous investigations into the eﬀectiveness of bimetallic dialkyl-amido zincates in chemoselective aromatic deprotonation and zincation reactions have been extended recently, with Armstrong and co-workers synthesizing Me2(hmds)ZnLi from dimethylzinc and (hmds)Li in hexane.1 X-ray crystallography conﬁrmed the planarity of the polymeric zincate’s four-membered LiNZnC rings. Subsequent chelation by pmdeta leads to the adduct Me(hmds)Zn(m-Me)Lipmdeta as the ﬁrst dialkyl-amido zincate in which the amide occupies a terminal position exclusively bonded to zinc. The same author also obtained novel chiral sodium amido-bis-tert-butyl zincates through the treatment of tmp(tBu)2ZnNatmeda with either diisopropylamine or hexamethyldisilazane in hexane.2 In a similar vein, Clegg and co-workers focused on the chemistry of mixed metal diisopropylamide zincates and have characterised (tBu)Zn(da)(tBu)Litmeda from tmeda, lda and tBu2Zn mixtures. Analogously, the tmeda complex (tBu)Zn(da) (tBu)Natmeda was yielded by the same reaction undertaken with sodium diisopropylamide.3 Subsequent addition of phenylacetylene lead to formation of the corresponding acetylide complexes [(m-tmeda)0.5 (tBu)Zn(CCPh)2Litmeda]2 and [(tBu)Zn(CCPh)2Natmeda]2, respectively. While the lithium acetylide complex shows a pseudodimeric structure with two LiCZnC rings, the sodium analogue can be seen as a distorted cubane with alternating PhCC and Na/Zn corners. Contrastingly, reaction by Garcı´ a-A´lvarez of the mixed-metal tris(diisopropyl) a

Department of Chemistry, University of Cambridge, Lensﬁeld Road, Cambridge CB2 1EW, UK

148 | Organomet. Chem., 2010, 36, 148–167 c

The Royal Society of Chemistry 2010

amides (iPr2N)3MgM (M=Li, Na) with phenylacetylene, lead to the heteroanionic complexes [(PhCC)(iPr2N)2MgLi]2 and [(iPr2N) 4 Mg(PhCC)2Natmeda]2, respectively. Whereas the Li compound adopts a cationic eight membered [(MgNLiN)2]2 þ ring, in the latter case the Na analogue was found to adopt a tetranuclear NaMgMgNa near-linear chain arrangement, the structural integrity of which was attributable to the action both of acetylido and of amido bridges. Utilizing potassium in conjunction with zinc, the co-complexation of (tmp)K with pmdeta and diethylzinc leads to the isolation of (Et)Zn(Et)(tmp)Kpmdeta, the feasibility of which as a zincating reagent has been demonstrated by yielding {[2-Zn(Et)2-4-Me2NC5H3N]Kpmdeta}2 and catena-{[3-Zn(Et)2-4-MeO-C5H3N]Kpmdeta}n from 4-(dimethylamino)pyridine and 4-methoxypyridine, respectively.5 More fundamentally, investigation by Strohmann has revealed the structures of lithium diamine complexes using tmeda, teeda and (R,R)-tecda. In the complex (iPrLitmeda)2, Li–C bond lengths of 2.255(8) and 2.236(8) A˚ lie in the average range for known dimeric alkyl lithium structures. In (iPrLi)3(teeda)2, the central dimeric ring is accompanied by another linked iPrLi unit. The bond distance between the latter metal centre and the bridging carbon was shown, at 2.095(4) A˚, to be uncommonly short.6 A recent focus in group 1 organometallic chemistry has been on use of the cyclopentadienyl ligand and its heterocyclic analogues. By reacting 2,2 0 dilithiobiphenyl with nickelocene, Buchalski and co-workers have yielded a 9-nickelaﬂuorenyllithium complex, in which one Li atom remains bonded (its complexation by two molecules of diethyl ether notwithstanding) to the biphenyl ring, while maintaining a relatively weak interaction with the Ni centre (d(Li–Ni)=2.513(5) A˚).7 Using dimethylamino-substituted ferrocenes, researchers have generated lithio[3]ferrocenophane systems with a central, four membered Li-C ring. In this case, crystallography indicates that the lithium centres, apart from being bonded to ferrocenyl carbon centres, experience additional stabilization through the Cp-dimethylamino N-donor atoms.8 In a similar vein, the formation of a tetracyclic lithium zirconimidate based on zirconocene has been found to result from reaction of the corresponding chloride Cp*2Zr(Me)Cl with tBuNHLi.9 In the context of research into alkene cycloaddition catalysts, Fu¨rstner has lately published a variety of ferrocene-derived ferrate complexes, in which the lithium counterion plays a stabilizing role as an additional anchor for the alkene moiety.10 The complexes, in which both the iron and the lithium centres were shown to be in close proximity, prove to be active in catalyzing cycloaddition and cycloisomerization reactions of polyunsaturated substrates. The same author has reported on a range of other heterobimetallic iron-lithium complexes, including a mixed-metal pseudocubane achieved through the combination of FeCl3 and methyllithium.11 In the complex [(Me4Fe)(MeLi)][LiOEt2]2, the iron centre can be viewed as capping a triangle of tetrahedrally coordinated lithium centres, which in this case avoids the necessity for the iron atom bearing any additional stabilizing ligands, the Z1-bonded methyl moieties notwithstanding. The product can, therefore, be seen as the ﬁrst iron alkyl-ate complex to have been fully characterized. Substituting the cyclopentadienyl anion for a heterocyclic analogue, Ly et al. have prepared Li, Na and K complexes using a range of stronger Organomet. Chem., 2010, 36, 148–167 | 149

p-coordinating, substituted 1,2-diaza-3,5-diborolyl ring systems.12 Similarly, by using N-heterocyclic bis(carbene)borates in conjunction with lda in thf it has proved possible to achieve a dimeric structure, in which the two bidentate ligands span the two lithium centres.13 In a recent paper by Eisler, the crystal structure of a system consisting formally of a phosphide-stabilised, 6p-stannylene dianion, [C6H4P2Sn]2 , that acts as a counterion to two tmeda stabilised Li moieties, has been described. The latter ions each coordinate to a P atom and to the stannylene ring in Z5 fashion, respectively, with one cationic unit exhibiting bridging activity to form a polymer in the solid-state.14 As part of an investigation into the inﬂuence of internal vs. external Li coordination in phenylallylic compounds on the stereochemistry of the aromatic group, Fraenkel has been able to report that the phenyl groups in the internally coordinated complexes 2-(bis(2-methoxyethyl))aminomethyl)-1-phenylallyllithium and 2-(bis(2-methoxyethyl)aminomethyl)-1,3-diphenylallyllithium favour an endo structure. Conversely, the structures of the externally coordinated complexes 1-phenylallyllithium(tmpH)4 and 1-phenylallyllithium(thf)8 were found to incorporate exo phenyl conformations. They concluded that for internal coordination, the alkali metal participates in the mechanism of phenyl ring rotation.15 By reducing [Me(tBu)P(C5Me4H)2]I with KH in thf, it proved possible to isolate [Me(tBu)P(C5Me4)2]Kthf as a coordination polymer with the alkali metal being Z3,Z5 coordinated between two C5Me4 rings. The latter exhibit noncoplanar relative geometries with an angle of 147.71 between the planes.16 Alkali metal 1,2,4-diazaphospholide complexes have been obtained by reacting H(3,5-Ph2dp) with either nBuLi or KH. The products were shown to be dimeric in the Li complex [(Z2,Z1-3,5-Ph2dp)Li(thf)2]2, or polymeric for the potassium product {(Z2:Z4-3,5-Ph2dp)K(OEt2)2}n.17 In the polymeric scandium complex [Sc(CH2Ph)5K2(thf)3]n, the alkali metal is Z6-coordinated to the benzyl groups and links the polymeric subunits.18 Lastly, in the iron complex [(L)Fe–O(H)–K(OEt2)] (L=[2-N(4-tBuPh)H-C6H4]3N), obtained by C¸elenligil-C¸etin and co-workers by deprotonation of the corresponding trisamido-amine ligand, followed by the addition of FeCl2 and exposure to O2, a bridging Fe–O(H)–K þ moiety constitutes part of a ﬁve-membered ring that involves K þ -(Z6 -arene) interaction.19 Several attempts have been made to exploit sterically encumbered ligands in order to investigate unusual alkaline metal coordination. Using the imine nitrogen donor 2,6-iPr2C6H3-NC(SiMe3)2, Cheng et al. have characterized the unusual eight membered ring structure that resulted from reaction with LiAlH4 in thf. In this ring, hydride moieties bridge between Li and the N-coordinated Al centres, the result being a trihydroaluminate dimer.20 In the context of investigations into the sterically protective function of the 1,8diphenylcarbazol-9-yl ligand, a (1,8-diphenyl-3,6-dimethylcarbazol-9-yl)potassium dimer has been characterized. Conformational ﬂexibility inherent in diphenyl substituted ligands means that this species reveals secondary metal-arene interactions and a centrosymmetric dimer structure with outof-plane bending of the phenyl substituents. The authors suggest that replacing the phenyl rings with mesityl groups might render a low coordinate metal system.21 In the case of {[(tBuN)2SCH2]Li2thf}2, Deuerlein et al. have successfully synthesized a cubane structured sulfur ylide,22 whilst Guino-o 150 | Organomet. Chem., 2010, 36, 148–167

et al. have reported that the simple co-complexation of benzylcalcium with benzyllithium leads to the formation of a heterobimetallic benzyl calciate (PhCH2)4CaLi2(tmeda)2 in which one benzylate moiety bridges between the tmeda-complexed lithium atom and the calcium centre.23 In the cases of biphenyl- and terphenyl-substituted triazenes (tph)2N3H, it has proved possible to characterize the range of potassium complexes KN3(tph)2, KN3(dmp)(mph), KN3(dmp)(tph) and KN3[Me4 (ter)]2 by reacting metallic potassium in n-heptane.24 Moreover, Loron˜oGonzales obtained the potassium coordination polymer poly[{m4-dihydrobis(pyrazol-1-yl)borato-k2N,N 0 }potassium] through treatment with MeMgCl in thf, a two-dimensional network resulting, with each potassium ion being coordinated by four pyrazolyl-N donors, while weak (m-BH)-K þ interactions additionally stabilize the structure.25 By reaction of the corresponding pentaarylboroles with potassium or potassium graphite in thf, so giving the borole anions [Ph4C4BAr]2 (Ar=p-Me3SiC6H4, Ph), for which potassium counterions are found to Z5 coordinate between the planar borole C4B rings.26 Using [(Z2,Z2-nbd)RuCl2]n and [(Z2,Z2-cod)RuCl2]n in conjunction with methyllithium, researchers have prepared [(Z2,Z2nbd)RuMe4][Litmeda]2 and [(Z3,Z2-C8H11)RuMe3][Litmeda]2, respectively.27 Remaining with heterometal systems, a mixed-metal yttrium complex (2,6-iPr2C6H3NSiiPr3)Y(CH2SiMe3)3Li(thf)2 was achieved using Me3SiCH2Li, the product revealing a distorted tetrahedrally coordinated lithium that is coordinated to Si, with a methylene unit bridging to the transition metal.28 Experiments on cyclic allenes have led to the isolation of a novel cyclic allenic lithium tetraﬂuoroborate adduct.29 Meanwhile, metallated allyl silanes incorporating Li and K have been shown to exhibit Z3 coordination of the metal by the allyl groups in dimeric and polymeric products, respectively.30 Matsumoto recently described the structure of a dme-stabilized lithium which is Z5 bonded to a substituted silabenzene radical anion,31 the product having been synthesized using 1-{2,4,6-tris [bis(trimethylsilyl)methyl]phenyl}silabenzene and lithium naphthalenide in dme. For early transition metal heterometallic complexes, Morse and coworkers have synthesized [HfEt4(C2H4)][Litmeda]2, [TaHEt(C2H4)3] [Litmeda]3(tmeda)0.5 and [WH(C2H4)4][Litmeda]3 by the ethyllithium alkylation of HfCl4, TaCl2(OMe)3 and WCl3(OMe)3, respectively. For both the trianionic Ta and W complexes, the transition metal centres have been noted to adopt distorted square-pyramidal geometries with the hydride group in the axial position.32 Several trialkylsilyl-substituted benzyllithium complexes have been synthesized by Schumann and co-workers; Me3SiPhCHLitmeda, tBuMe2SiPhCHLitmeda, PhMe2SiPhCHLitmeda, Me3Si (3,5-Me2C6H3)CHLitmeda, and tBuMe2Si(3,5-Me2C6H3)CHLitmeda all having been prepared through the treatment of the corresponding benzyland 3,5-dimethylbenzylsilanes with butyllithium and tmeda.33 The trimethylsilylmethyllithium-based structures of diﬀerent donor solvents, (Me3SiCH2Litmeda)2, [Me3SiCH2Li(–)-sparteine]2 and Me3SiCH2Li pmdeta, were noted by Stalke.34 Remaining with the trimethylsilylmethyl ligand, the possibility that this ligand might demonstrate potentially powerful applicability in dimetallation reactions in conjunction with bimetallic sodium-manganese or sodium-magnesium bases has been Organomet. Chem., 2010, 36, 148–167 | 151

presented. Toluene has been shown to be regioselectively metallated in the 3,5 positions through the use of the inverse crown complexes (3,5Mn2C6H3CH3)Na4(tmp)6 and (3,5-Mg2C6H3CH3)Na4(tmp)6.35 The same authors reported (tmp)Mn(o-C6H4OMe)(tmp)Natmeda and (Me3SiCH2) Mn{o-[C(O)N(iPr)2C6H4](tmp)Natmeda from reaction of (tmp)Mn(Me3SiCH2)(tmp)Natmeda with anisole or N,N-diisopropylbenzamide, respectively.36 Mono- and geminal bimetallic alkali metal complexes {[Ph2P (NSiMe3)]2C}2M4 where M4=Na4, Li2Na2, LiNa3, Li2K2, Na2K2 and Na3K have been prepared by Hull using a variety of methods.37 Whereas in the homometallic products, the authors found an octahedral arrangement of the L2M4 core, a rhombic geometry results for the heterometallic complexes owing to the presence of diﬀering metal-carbanion interactions. Meanwhile, the use of NaH reduction of the bis(iminopyridine) iron complex {2,6-[2,6-(iPr)2PhN=C(CH3)]2(C5H3N)}FeCl2 has led to a variety of sodium-coordinating structures.38 Apart from this, the heterometallic potassium barate complex [(Me3Si)2N]2Ba(N3Ar2)K(thf)4 displays an unusual bridging mode by the triazenide ligand. It was reported by Barrett et al., who revealed that the N3Ar2 ligand bonds via both its terminal and internal nitrogen donors.39 Using KOH and 1,2-phenylenediacetic acid in methanol, researchers have obtained a coordination polymer based on K-m2-[2-(carboxymethyl)phenyl]acetate building blocks, in which each K centre is coordinated by six O atoms and a K–phenyl distance of 3.14 A˚ is noted.40 Through the addition of Et2O to a mixture of KPPh2 and Mg(PPh2)2 in thf, Westerhausen has been able to report both the strandstructured complex [(Ph2P)3Mg(thf)]KOEt2 and layer-structured [(Ph2P)4Mg]K2(OEt2)x(thf)y. A similarly layered structure resulted upon recrystallization of the latter complex from hot 1,4-dioxane. This could be identiﬁed as [(Ph2P)4Mg]K2(dioxane)2. In this case, the central units were found to consist of eight-membered K2Mg2P4 rings that were interconnected by P-K-P bridges.41 Several rhenium calixarenes [Re(O)(PPh3)cax(O)4M(NCMe)2]4MeCN (M=Na, K) have been reported recently. In these the alkali metals were found to reside within an elliptical ligand conformation.42 In mixed-metal Cu scorpionate complexes, potassium has been noted to act, together with copper, as a bridge between two pyrazolyl moieties and to thus form a ring-like structure.43 1.2

Rubidium and cesium compounds

For both rubidium and cesium, Atwood has noted three-dimensional networks of nanocapsules from p-carboxylatocalix[4]arene-O-methyl ether. In these capsules, the calixarene can be found in a 1,3-alternate conformation with the metal centres accounting for the top and bottom lock in a cage-like structure.44 The reaction of the tetrapyrrolic ligand 2,2 0 -bis-dipyrrin with iron chloride and subsequent alkali metal halide metathesis by Bro¨ring and coworkers resulted in iron pyrrole complexes, in which Li, Na and Cs have all been found to serve as adequate counterions for the distorted pentacoordinate iron-based cationic unit.45 In all products, the iron-bonded bis-dipyrrin ligand was found to reside in a helical conformation. Furthermore, coordination studies with a novel set of tetramethoxyresorcinarene tribenzo-bis-crown 152 | Organomet. Chem., 2010, 36, 148–167

ethers have revealed a favourable eﬀect on alkali metal coordination of ligand aromaticity, with an aﬃnity maximum noted for Cs and the meta-tetramethoxyresorcinarene tribenzo-bis-crown-6 ligand.46 The authors ascribe these dependencies between the cation-p interaction aﬃnity of the alkali metals and their hosts to the morphological preorganization of the ligands. 2. 2.1

Group 11 Copper compounds

Working in the ﬁeld of copper organometallic compounds, Wang has obtained a series of homometallic Cu-oleﬁn coordination polymers of formulation [CuX(C9H9N5)]n (X=Cl, Br) through the solvothermal treatment of 3-(2-allyl-2H-tetrazol-5-yl)pyridine and 4-(2-allyl-2H-tetrazol-5-yl)pyridine with the corresponding copper(I) halides.47–50 While 3-substituted pyridine ligands led to polymers, in which the Cu atoms were linked by the allylic function and the pyridine N-donors, 4-(2-allyl-2H-tetrazol-5-yl)pyridines lead to the formation of polymers based on dinuclear Cu2Br2 rings bridged by two ligands. A variety of other straightforward Cu based coordination polymers have been published.51–54 Hence, researchers have characterized several tetrazolate-based polymers and found diﬀerent polynuclear copper iodide species to be entrapped in the products en route to their formation.54 Upon use of 5-(3-cyanophenyl)tetrazolate in conjunction with cuprous iodide, the resulting [Cu2(m3-I)(m5-cpta)]n showed two varieties of tetrahedrally coordinated copper centres, which are coordinated by iodine, itself constituting the cationic (Cu4I2)2 þ subunit. On the other hand, [Cu5(m4-I)(m4-mtta)3(CN)]n, obtained by an analogous procedure from 5-methyltetrazolate, exhibits macrocyclic subunits linked by cyanide groups. In this case, three crystallographically independent copper atoms can be found. Among them, two are tetrahedrally and one, in a CN linking function, trigonally coordinated. Finally, [Cu5(m6-I)(m2-I)(m4-mtta)3]n shows a bilayered structure with a (Cu10I4)6 þ , ribbon-shaped unit. All copper atoms show a tetrahedral environment and coordination by m2-bridging iodine atoms. The coordination polymer [Cu7(CN)7(L)2]n (L=4-(6-amino2-pyridyl)-1,2,4-triazole),52 was obtained from the hydrothermal reaction of CuCN with K3[Fe(CN)6] and shows hexagonal channels with L in a tridentate bridging function. In this product, four independent copper atoms display a distorted trigonal-pyramidal, trigonal planar and near-linear coordination. The analogous reaction with CuSCN in acetonitrile aﬀorded the polymer [Cu2(SCN)2(L)]n, in which the [Cu(SCN)]n moieties form a ladderlike structure. In the latter, two-dimensional polymer, two crystallographically independent copper atoms can be found, with both being in distorted tetrahedral coordination geometry. The work of Li et al. has aﬀorded a range of interesting mixed-metal coordination polymers, namely [Et4N][WS3.5O0.5Cu3(CN)0.5(LH0.5)], [Et4N][WS4Cu3(L)] and [Et4N]2 [MoS4Cu4(CN)(LH0.5)2] (L=1,3,4-thiadiazole-2,5-dithiolate). While [Et4N] [WS3.5O0.5Cu3(CN)0.5(LH0.5)] can be rationalized as consisting of a nest-shaped WO(S)S3Cu3 and half of a CN group as anionic units, and a [Et4N] þ cation accompanied by one L ligand, the WS4Cu3 unit in [Et4N][WS4Cu3(L)] forms a T-shaped structure. Bonding to eight equivalent Organomet. Chem., 2010, 36, 148–167 | 153

units via four L moieties, an anionic two-dimensional network is formed. In the case of [Et4N]2[MoS4Cu4(CN)(LH0.5)2], the corresponding MoS4Cu4 core adopts a saddle shaped conformation. The coordination geometries of copper are distorted tetrahedral, tetrahedral/trigonal and distorted tetrahedral, respectively. Lastly, Agustı´ and co-workers obtained the mixedmetal polymer {Fe(pmd)2[Cu(CN)2]2}n. In the iron spin-crossover polymer, two-dimensional layers of iron and copper were observed with copper in distorted trigonal coordination at 180 K. Upon cooling to 90 K, a cell volume contraction of 18 A˚3 was observed with the bond lengths around the octahedrally coordinated iron all shortening, but the Cu–Cu bond lengths remaining largely unaﬀected.53 Copper cyanide has been extensively employed as a building block for the synthesis of coordination polymers and metal-organic frameworks by virtue of the rigid linear structure of the cyanide group. In conjunction with a wide variety of ancillary ligands, each of which was selected in order to enforce speciﬁc steric constraints, diverse structures have been achieved. For example, the solvothermal reaction of 4,4 0 -dimethyl-2,2 0 -bipyridine with CuCN in acetonitrile led to the polymer [Cu3(CN)3(C12H12N2)2]n, in which two bipyridine N,N 0 -chelated, tetrahedrally coordinated Cu atoms are linked by a cyanide bridge.55 In conjunction with molybdenum, several interesting heterobimetallic cage structures have been obtained. Accordingly, [Et4N]2[MoOS3(CuCN)] has been used to source the polymeric clusters {[Et4N]2[MoOS3Cu2(m-CN)]2(aniline)2}n, which displays a butterﬂy-like MoOS3Cu2 fragment, and {[Et4N]4[MoOS3Cu3CN(m 0 -CN)]2(mCN)2}n, with its cubanelike MoOS3Cu3 core.56 Using thioethers and CuCN units as bridging functionalities, the [Cu2(CN)2bppe]n metal-organic framework has been prepared by Zhang and co-workers. It is based on a centrosymmetric ligand coordinating to the metal in a bis-monodentate fashion.57 A further attempt to use CuCN for the synthesis of framework structures was included in the experiments of Liu and Guo, who obtained a layer-structured cyanide, Cu(CN)(CH3CN)(18-crown-6)0.5, from the reaction of CuCN and 18-crown-6 in acetonitrile.58 Employing [WS4Cu3] þ as a heterobimetallic precursor, together with (Bu4N)CN in isopropanol, has led to the isolation of {[Bu4N][WS4Cu3(CN)2]}n, a complex with a distorted diamond like topology that has been shown to display nonlinear optical behaviour.59 A unique Cu4Cl4 crown core element with two distorted trigonal-planar and two two-coordinate metal atoms are incorporated in the core of the solvothermally synthesized complex [(Me4N)(H3O)] [(Cu4Cl4)][Cu(CN)2]2 – the ﬁrst two-dimensional halogeno(cyano)cuprate, achieved by Liu and Guo.60 A diﬀerent approach was adopted by Stollenz et al., who obtained a metal-organic framework consisting of mesitylbridged [(MesCu)4(m4-O)]2 cuprate anions, the pocket-like structure of which stems from the pseudo-tetrahedral arrangement of the (m4-O)Cu4 nuclei.61 In the same way, various amine-ligands, such as substituted piperazines62 and 9,10-bis{[3,4-bis(methylthio)phenyl]ethynyl}anthracene, have been used in conjunction with CuCN to form partly thermochromic frameworks with non-linear optical properties and a network showing interesting alkynyl-p overlap,63 respectively. With a similar aim, and in order to investigate the apparent dependence of framework optical 154 | Organomet. Chem., 2010, 36, 148–167

properties on the conjugated chain-length of the spacer ligands, Zhang synthesized the honeycomb-like frameworks {[(NO3)(NMe4)3][WOS3Cu3 (CN)3]}n, {[(NH4)(dmf)2][W2O2S6Cu6(CN)3(bipy)4]} and {[(NH4)(dmf)2] [W2O2S6Cu6(CN)3(bpee)4]} via the self-assembly of [NH4]2WOS3, CuCN, and either [NMe4]NO3, 4,4 0 -bipy, or 1,2-(E)-bis(4-pyridyl)ethene, respectively, at room temperature.64 Ferroelectric properties and three crystallographically independent Cu centres are displayed by a framework obtained from the solvothermal CuCN treatment of N-4-cyanobenzyl quinidinium bromide.65 Using a CuSO4 precursor and the exotridentate multipyridyl ligand 2,4,6-tris(4-pyridyl)-1,3,5-triazine in conjunction with 1,2,4,5-benzenetetracarboxylic acid piperidine, [Cu(tpt)(H2btec)0.5]n was formed, which was found to incorporate Cu, trigonal-pyramidally coordinated by three tpt-pyridyl nitrogen donors and axially coordinated by the carboxylate oxygen. By these means, the tpt-Cu layers form a two-dimensional, bilayered network incorporating H2btec2-bridges.66 Amazing examples of copper frameworks [Cu12(CN)11(SCN)4][{M(m-L)3}2{M3 (m-OH)}] (M=Ni2 þ , Zn2 þ ; LH=3,5-bis(2-pyridyl)pyrazole) have been published. These reveal anionic, helical copper units that entrap cationic nickel or zinc helicates, and are synthesized by reaction of [{M (m-L)3}2{M3(m-OH)}](SCN)36H2O(M=Ni2 þ , Zn2 þ ) with CuSCN.67 The isostructural nickel and zinc complexes can be rationalized as [{Ni (m-L)3}2{Ni3(m-OH)}]3 þ cluster helicates that are encapsulated by an anionic {[Cu12(CN)11(SCN)4]3 }n moiety. This anionic part was found to consist of tubular, intertwined CuSCN/CuCN helical units built around the helical core unit. Therefore, a helical transfer has taken place resulting in a double helix with either single strand exhibiting the same helicity. Further examples of cyano-bridged copper metal-organic frameworks were published by Li and Pagola,68,69 with a systematic study on the tuning of framework morphology by ligand elongation having also been conducted.70 Among the copper(I)-tetrazole coordination frameworks obtained by Li et al., {[Cu2(m3-mtta)2(CN)][Na(NCMe)]}n was found to consist of a coplanar Cu2(azole)2 subunit, in which the bidentate azole moieties are bridged by two tetrahedrally coordinated copper atoms, while CN acts as a linker for framework formation. Two crystallographically independent copper atoms can be located in the tetranuclear product {[Cu4 (m3-Hmtta)2(CN)3](OH)}n. These can be found in a distorted trigonal environment (with bonding to one tetrazole moiety and two CN bridging units) and in trigonal coordination (to two tetrazole N atoms and one CN ligand), respectively. Pagola described the framework [Cu(m-CN)(m-L)]n (L=2-aminopyrazine), in which a tetrahedrally coordinated copper centre is ligated by two CN and two L units with [Cu–CN] chains being formed along the c-axis. Finally, as part of an investigation into coordination framework formation between copper(I) halides and the thioether ligands bis(2-pyrimidinylthio)methane and 1,2-bis(2-pyrimidinylthio)ethane, it proved possible to obtain four metal-organic frameworks, for which they found a dimensional increase from chains, to two-dimensional layers, to networks depending on the ligand chain lengths and spacers. Parallel to these characteristics, macrocycle size of was found to increase from 16- to 42-membered rings. Eventually, in the framework [Cu6I2(L)4] Organomet. Chem., 2010, 36, 148–167 | 155

(L=1,2-bis(2-pyrimidinylthio)ethane), the large macrocyclic size leads to the interpenetration of the diamondoid nets and a host-guest arrangement of the ligands within the cycles. A Cu p-complex, consisting of an anionic structure of stacked [CuCl2 (HOCH2CCCH2OH)]– units with polymeric [Na þ . (OH2)2]n counterions, results from the reaction of 2-butyne-1,4-diol with CuCl in concentrated saline. The metal atoms adopt a trigonal planar geometry in conjunction with the butyne triple bond.71 Reaction of the carbene ligand 2,10-di-tert-butyldipyrido[1,2-c;2 0 ,1 0 -e]imidazolin-6-ylidene with CuI and tBuOK in thf leads to a trimeric complex with two Cu-Cu bonds in the solid state.72 A linear, two-coordinate copper atom has been noted in an imidazol copperCF3 complex. This species was found to be an eﬀective triﬂuoromethylating reagent for arylhalides.73 Bruce et al. obtained a tetranuclear Ru2Cu2 cluster, Ru2Cu2(C2Ph)5H2(Cl)(PPh3)Cp*2, by heating RuCl(PPh3)2Cp* and (CuCCPh)n to reﬂux in benzene. The product had a chain-like Ru-Cu-Cu-Ru core structure.74 During the course of a systematic investigation into the Cu coordination modes of multidentate sulfur ligands, it was found that tetranuclear clusters were formed when CuI was treated with ligands such as bis(2pyridylthio)methane in acetonitrile, while precursors such as CuCl2 mostly aﬀorded polymeric products.75 A Cu2I2 motif results from reaction of the novel adamantyl ligand 1,3-bis[1-(1-adamantyl)-3-phenyl-1,2,4-triazol-5-yliden-4-yl]benzene with CuI in acetonitrile/toluene. In this instance the metallic centres, chelated by the triazole carbon atoms, show clear asymmetries in terms of their bonding activity with respect to iodine.76 Heteromultimetallic complexes have been investigated. In the trimetallic compound [(Z5C5H5)(dppf)Ru-CC-2-bipy-5{[Ti](m-s,p-CCSiMe3)2}Cu]PF6, a pseudotetrahedral arrangement of copper exists within the {[Ti](m-s,p-C CSiMe3)2}Cu] þ cationic subunit.77 Lastly, the multinuclear complex {Pd[CH(CMeNPh)2](L)2}2(m-Cu6Cl8) (L=N-methyl-4,5-diphenylimidazole) has been obtained by Hadzovic and Song.78 The Cu6Cl8 core of this species can be rationalized as constituting two stacked m2-Cl linked Cu3Cl3 rings, with the metal exhibiting a trigonal-planar coordination geometry. 2.2

Silver compounds

Silver coordination polymers have been synthesized from N-phenyl-N 0 cyano-formamidine with diﬀerent silver salts.79 Branzea obtained several copper polymers based on the formulation (CuL)K(LCu)[Ag(CN)2] from the reaction of N,N 0 -propylene-bis-(3-methoxysalicylideneiminato copper(I) with K[Ag(CN)2],80 while Steiner used allylamino-substituted cyclotriphosphazene ligands,81 Rukiah has described a poly(m2-2,2-dimethylpropane-1,3-diyldiisocyanide)-complex82 and Li, the silver sulfonate polymer of Ag(C6H5NHC6H4SO3).83 These examples notwithstanding, Busetto described the Boc-protected structure [(nhc-NHBoc)2Ag4I4]n, in which the Ag4I4 motif is responsible for the adoption of a ladder morphology.84 Concerning coordination frameworks, relevant structures have been obtained from the reaction of AgClO4 with anthracene-9,10-dicarboxylic acid and hexamethylenetetramine.85 The resulting complex, {[Ag8(L)3 156 | Organomet. Chem., 2010, 36, 148–167

(m4-hmt)2(H2O)6](ClO4)2}n (L=anthracene-9,10-dicarboxylate), displays a bimodal, 4-connected (4363)2(426282)3 topology by virtue of the inﬂuence of the anthracene p-system. Z2 Ag–C bonds between four-coordinate silver atoms and phenyl-C atoms were shown to be present in polymeric [Ag2 (m8-L)]n (H2L=4-[(4-hydroxyphenyl)sulfonyl]-1-benzenol).86 Using 3,5diethynylpyridine in conjunction with CF3CO2Ag and AgNO3, Zhang obtained the frameworks [Ag2(3,5-C2pyC2)(CF3CO2Ag)4(H2O)4] and [Ag2(3,5-C2pyC2) . (AgNO3)3 . H2O], respectively,87 whereas the employment of 2,2 0 -bis(prop-2-ynyloxy)biphenyl as a ligand led to structures with a variety of unusual metal-p interactions, which are presumably essential for the morphological stabilization of the framework.88 Lastly, Zhao has obtained several diﬀerent silver(I) complexes in which p-p stacked, columnar Ag-ethynide structures are revealed.89 Chen has obtained oligomeric alkynyl cluster complexes. One such species displays a structure that can be rationalized as an octasilver cation resulting from the fusion of four Ag3 triangles in the complex [Ag8(2-bipy)6(CCtBu)4](BF4)4.90 In {(AgCN)2 [P(NC4H8NMe)3]}n, the phosphorous/nitrogen donor ligand tris(4-methylpiperazin-1-yl)phosphane is structure-determining, with the formation of a zigzag structure resulting from the bridging of [{P(NC4H8NMe)3}Ag] þ units by the anionic [NCAgCN] moieties.91 Macrocyclic structures have been obtained by Salazar-Mendoza, using the 5-(4-cyanophenyl)dipyrrin ligand to give structures containing the metal in [AgNC] þ bridging moieties,92 Dias, who synthesized trinuclear triazolate macrocycles from both silver and copper,93 and Winkelmann, with a 1,3bis(2,4,6-trimethylphenyl)imidazol-2-ylidene derived silver metallocycle.94 Furthermore, Cacciapaglia and co-workers have obtained diverse cyclophane-based macrocycles.95 Using paracyclophane formaldehyde acetals, the authors prepared a C2-symmetric (CF3SO2)2NAg complex, in which the cationic silver moiety is accompanied by a [(CF3SO2)2N]– counterion. The metal is coordinated by two opposite oxygen atoms and by two opposite sets of Z3 Ag þ -C bonds from aromatic rings. Skonieczny et al. synthesized a four-coordinate silver(III) carbacorrole using AgBF4 and iso-carbacorrole in CH2Cl2.96 The formation of this structure was rationalized by invoking a combination of both tetrahedral-trigonal and trigonal-tetrahedral carbon rearrangements to account for the formation of a M(III)-C sp3 bond. The metal is rendered four-coordinate, interacting with three ligand N-centres and one sp3 carbon atom. Owing to their enhanced catalytic activity in numerous organic transformations and their use as carbene transfer reagents, N-heterocyclic silver carbene complexes have formed the basis of much recent research. One such example, the Ag(I) complex [AgBr(L)] (L=1,3-dibenzhydrylbenzimidazolium), has been reported and has proved to be useful as a carbene-transfer reagent.97 Using the pincer pyridine dicarbene ligand 2,6-bis-[(dipp)imidazol-2-ylidene]-3,5-dimethylpyridine, research has yielded a dimeric cycle with two-coordinate silver bridging the imidazol rings. In this system the [Ag6I8]2 counterion is found to adopt the rare arrangement of a planar, 6-membered Ag ring.98 Following the same rationale, another group obtained an imidazolium-based, trinuclear complex Ag3Cl2(m-Cl)(m-L)2, employing the 3-methyl-1-(1-ethyl-2-methylbenzoyl)imidazolin-2-ylidene Organomet. Chem., 2010, 36, 148–167 | 157

ligand (L), in which only one metal atom, bonded to two (imidazole-N)coordinated Ag centres, shows carbene coordination.99 In contrast, the ﬁrst example of a linear Ag3 arrangement has been published, with [Ag3(L)2 (MeCN)][PF6]3 (L=2,7-bis(alkylimidazolylidenyl)naphthyridine), with X-ray analysis suggesting the presence of argentophilic interactions.100 In comparison to the corresponding Au–Au contact of 3.2042(2) A˚, shorter argentophilic interactions, of 3.1970(12) A˚, have been seen in the complex {[1-(benzyl)-3-(N-tert-butylacetamido)imidazol-2-ylidene]MCl}2 (M=Ag, Au).101 Dit Dominique and co-workers found novel Ag(I) carbene species to result from the treatment of diﬀerent diimidazolium salts with Ag2O in dmso. The cyclic nature of the ligands led to a dimeric product with near square-planar, crossed metal coordination and weak Ag-Ag interactions. Upon ion-exchange of the [TsO] counterion for [PF6] , considerable geometric rearrangement was observed. In the latter case, the ligand was found to form a hydrophilic pocket around the anion, while the metal slightly deviates from its originally linear arrangement.102 The pyrazolefunctionalized imidazolium ligands 3,5-bis(N-benzylimidazoliumyl)pyrazole and 3,5-bis[N-(2,4,6-trimethylphenyl)imidazoliumyl]pyrazole have been used by researchers to generate the tetranuclear complexes [Ag4L2]X2, in which either [PF6] or [BF4] serve as counterion X.103 In either product, two ligands bridge four silver atoms, the latter being bicoordinated by one pyrazolate nitrogen and one carbene carbon atom. Using N-alkyl substituted bis(benzimidazoliumylmethyl)durene halides, the complexes durene(CH2bimyEtAgBr)2 and durene(CH2bimynBuAgCl)2 have been synthesized. Whereas, upon use of the n-butyl substituted ligand, a dinuclear product with simple carbene-AgCl coordination is formed, employment of the ethyl ligand leads to the formation of a macrocyclic product, with bromide-bridging of the two metal centres and aromatic p-p stacking interactions between the benzimidazol moieties.104 The same author described the anthracenyl complex [1-(9-anthracenylmethyl)-3-butylbimy]2AgPF6 as a mononuclear product in which the metal is bonded by two ligand carbene carbon atoms.105 Dyson and co-workers obtained phenoxyimine-incorporating products106 and, for the purpose of synthesizing Pd catalysts active in Heck coupling, researchers have described the silver carbene-transfer complexes [AgL2]PF6 and [Ag2L2][PF6]2 wherein L=1-n-butyl-3-(2-pyrimidyl)imidazolylidene and 1-(2-picolyl)-3-(2-pyrimidyl)imidazolylidene, respectively.107 In both cases, the pyrimidine groups remain uncoordinated. 2.3

Gold compounds

Similar to the abovementioned silver nhc coordination compounds, carbene chemistry has also been dominant in the ﬁeld of gold organometallic chemistry. Noteworthy examples include a Au(PPh3)-compound derived from tetraaminoallene, that can be rationalised in terms of a dicarbene with ylide character and which, owing to the electron-rich character of the central carbon atom, oﬀers the potential for dimetallation products.108 Nonactivated allenes and alkynes have been found by Lavallo to be readily aminated by cationic carbene gold complexes.109 For this purpose, a 2,6diisopropylphenyl functionalized cyclic alkylaminocarbene gold(I) complex 158 | Organomet. Chem., 2010, 36, 148–167

was reacted with NH3, 3-hexyne, and deuterated benzene to yield 3-iminohexane. Further investigations revealed the necessity of a cationic gold centre being present for the compounds to exhibit catalytic activity. The related Au-NH3 complex was identiﬁed as the resting state of the catalyst and the related silver complex was found to be inactive in the same reaction. With AuClL, [AuL2]BF4 and trans-[AuI2L2]BF4, the isolation has been reported both of Au(I) and of Au(III) complexes derived from the nhc ligand 1,3-diisopropylbenzimidazolin-2-ylidene (L).110 While the ﬁrst product, obtained from reaction of AuCl(SMe2) with AgClL in methylene chloride, was shown to display linear gold coordination with no aurophilic interactions, the metal is coordinated by two ligand units in [AuL2]BF4, which was derived from the reaction of AuClL with the benzimidazolium salt LH þ BF4 in the presence of K2CO3 in acetone. Furthermore, and in contrast to AuClL, an upﬁeld shift of the isopropyl CH resonance in the 1H NMR spectrum indicates free rotability of the N-isopropyl groups and, therefore, the absence of CH-M interactions. Lastly, trans-[AuI2L2]BF4 was obtained by the reaction of [AuL2]BF4 with iodine in methylene chloride. In the product, which is reported as the ﬁrst Au(III) benzimidazolin-2-ylidene complex, 1H NMR indicates an even stronger shielding of the isopropyl protons and the square-planar coordinated metal centre is ligated by two iodide and two nhc ligands. Both the L-Au-L and I-Au-I bonds exhibit near linear arrangements, while the carbene ring planes are in orientations perpendicular to the AuC2I2 plane, there being no indication that aurophilic or p-p-interactions are manifest. A mixed-metal N-heterocyclic Au carbene has been published with bis{1-[(E)-2-butenyl]-3-(4-ferrocenylphenyl)2H-imidazol-2-ylidene}gold(I) tetraﬂuoroborate being derived from the combination of 1-[(E)-2-butenyl]-3-(4-ferrocenylphenyl)imidazolium tetraﬂuoroborate, Ag2O and chloro(dimethylsulﬁde)gold(I).111 The resulting complex, in which the metal centre is linearly coordinated by two carbene carbon atoms, showed cytotoxic activity towards selected lines of cancer cells. Workers observed an unusual 1,7-bromination in gold carbenes in N-aryl imino substituted AuCl complexes.112 The authors assumed the imino-mediated conjugation between both sites to be responsible for the observed reactivity. Meanwhile, the group of Bertrand have focused on the structural aspects of a series of gold complexes bearing cyclic alkylamino carbene ligands.113 In the linearly carbene carbon-coordinated gold complexes, the authors found monocarbene products to be favoured by steric bulk in the ligands, while cationic dicarbenes resulted from reaction with ligands displaying less steric hindrance. For ortho amine-substituted pyridylcarbenes, the NH-pyridyl moiety has been reported to be capable of forming both intra- and intermolecular hydrogen-bonds.114 The intramolecular bonds observed between the amine hydrogen and the pyridyl nitrogen were also found to be preserved in acetone solution. For the conjugate addition of Grignard reagents to 3-methyl- and 3-ethylcyclohexenones, Matsumoto et al. found that chiral 2-methoxyphenyl substituted imidazolium gold carbene complexes in particular can induce the formation of high enantiomeric excesses (of up to 80%).115 This can be rationalized in terms of the conformation of the aromatic moieties around the metal centre. Lastly, Olmstead has reported dependencies between the blue and green Organomet. Chem., 2010, 36, 148–167 | 159

luminescence of the complex [Au{C(NHMe)2}2]AsF6 and the crystallographic order or disorder of their stacking pattern.116 Gold phosphine complexes have continued to constitute a focus of investigation recently, by virtue of their known catalytic potential in organic synthesis. Shapiro synthesized a structure in which either silver or gold monoalkyne substituted triphenylphosphine is incorporated in a dimeric structure that exhibits both metal-P and metal-alkyne Z2-coordination.117 In [Au3(PPh2(C6H4)2PPh2)3{Au6Cu6(C2Ph)12}][PF6]3, A bimetallic cluster that features a central Au6Cu6 containing fragment which is wrapped by an outer sphere of {Au[PPh2(C6H4)2PPh2]}3 through direct Au–Au linkages has been described.118 Meanwhile, a heterotetrametallic compound containing ferrocenyl-coordinated iron, phosphine-bonded gold, bis(Z2-alkynyl)-bonded copper and a titanocene derived moiety has resulted from experiments by Packheiser et al.119 Furthermore, the tris(azolyl)phosphine ligand system has been studied.120 In comparing the thus yielded complexes R3PAuCl (R=1methylimidazol-2-yl, thiazol-2-yl, 4-methylthiazol-2-yl, 4,5-dimethylthiazol2-yl), the corresponding thiazol-2-yl- and 4-methylthiazol-2-yl-substituted gold phosphine compounds were found to display aurophilic interactions of varying strengths, whereas reaction of the methylimidazol complex with C6F5Au(tht) led, among other species, to the complex bis(pentaﬂuorophenyl)-m-[tris(1-methylimidazol-2-yl)phosphine-k2P,N]digold(I)OCMe2, in which p-stacking of the aromatic moieties accompanies a strong aurophilic interaction between the two metal centres. Conversely, hydrogen bonding dominates the dimeric interactions within the terpyridine-based structures n Bu3PAu(LH) and (LH)Au(m-dppe)Au(LH) (L=HCCCH2Otpy).121 In the former compound, a trans conformation for the tpy unit and a linear C–CC–Au–P moiety were noted by Constable. Likewise, a transoid arrangement of the tpy moieties in the latter complex avoids intramolecular Au–Au contacts. In a related vein, using diphenylphosphanide and 2,6diphenylpyridine as a coordinating environment, Li and co-workers have achieved square-planar metal bonding within a dinuclear complex involving weak p-stacking and hydrogen bonding.122 Other reported phosphine-derived gold complexes obtained by Khairul and Osawa have incorporated 1,2diaryl acetylenes and 2-naphthyl moieties, respectively.123,124 In the ﬁrst case, various monometallic gold phosphines, Au(CCC6H4CCC6H4X)(PPh3) (X=Me, OMe, CO2Me, NO2, CN), were obtained from the reaction of the corresponding 1,2-diaryl acetylene derivatives and NaOH in methanol. The reaction of these complexes with Ru3(CO)10(m-dppm) in thf has aﬀorded the corresponding mixed-metal clusters Ru3(m-AuPPh3)(m-Z1,Z2-C2C6H4C CC6H4X)(CO)7(m-dppm). Whereas, in the monometallic products, the metal shows a linear, two-coordinate arrangement with p-p-interactions between the 1,2-diaryl acetylene moieties in the solid state, the bimetallic structures can be described in terms of a triangular Ru3 cluster with a Au(PPh3) moiety acting as a bridging function. In these cases, p-p stacking is not observed to play a role in the solid-state structure, although there are indications of weak hydrogen bonding activity. Lastly, the phosphorescent structures LAu(PPh3) and [m-L{Au(PPh3)}2]ClO4 (L=2-naphthyl) were obtained by Osawa through the reaction of 2-lithionaphthalene with Au(PPh3)Cl and treatment of the obtained LAu(PPh3) with HClO4, 160 | Organomet. Chem., 2010, 36, 148–167

respectively. While in the former product the metal is nearly linearly coordinated by the naphthyl-C2 centre and one phosphorous atom, the latter species displays a binuclear structure with the gold atoms describing a trigonal, naphthyl-bridged arrangement. Ferna´ndez has obtained one-dimensional, anionic structures of the type {NBu4[Tl2(AuR2)3]}n and {NBu4[Tl(AuR2)2]}n (R=2-C6BrF4, 2-C6F4I).125 Two-dimensional, heterobimetallic framework structures have been obtained, with [Pb(en)][Au(CN)2]2 displaying an unusual bonding interaction between the metal centres,126 and Agustı´ having described a cyanide-bridged Au-Fe network.127 A three-dimensional framework consisting of both hydrogen bonded and aurophilically linked Au(CN)-containing subunits has been obtained by Qu.128 Lastly, self-assembled monolayers derived from 1,1 0 -disubstituted ferrocenes that result in a 3,4-diaura-[6]ferrocenophane structure that incorporates aurophilic interactions have been reported.129 A systematic study into the eﬀect of pyridine and bipyridine ligand chain lengths on intermolecular aurophilic interactions has been presented. It found that Au–Au interactions were favoured in the complex (Ph3P)Au(CC)L(CC)Au(PPh3) (L=2,5-pyridyl), in which the chain length is relatively short and the ancillary phosphine groups are not overly bulky.130 Similar studies have been reported by Ferrer.131 Trigonal-planar gold coordination was reported in the halogen-substituted triazapentadienyl complex [N{(C3F7)C(2,6-Cl2C6H3)N}2]Au(C2H4).132 The reaction of a series of compounds of the type AuBr(CN)2L, incorporating diﬀerent bipyridine and phenanthroline ligands (L), with K[AuCl4] has been reported. These transformations yielded salts with square-planar Au coordination.133 Moving to heterometal systems, a mixed-metal [Ru(L)(NCMe)2][Au(CN)2] complex (L=N,N 0 -bis(salicylidene)-o-phenylenediamine) has been published.134 Bimetallic Au(III)Cu(III) and Au(III)Rh(I) complexes were obtained by Mori and Osuka using a pentaﬂuorophenyl-polysubstituted, cyclic [26]hexaphyrin(1.1.1.1.1.1) ligand, the products revealing planar Cu, but out-of-plane Rh, coordination due to aromaticity.135 The (aza-15-crown-5) dithiocarbamate ligand was shown to stabilize mono-, di-, and hexanuclear gold complexes.136 A remarkable result was obtained by Dias, who reacted AuCl with Ag[SbF6] and achieved the ﬁrst isolation of the gold ethylene complex [Au(C2H4)3][SbF6], in which the metal has a trigonal-planar coordination geometry.137 The alkyne complex Au(CCL)(PPh3) (L=phenyl4-(2,2 0 :6 0 ,2 0 0 -terpyridine-4-yl), obtained from Au(acac)(PPh3) and HCCL, is one of several obtained by Jones, in which intramolecular CH–Au interactions determine the ligand arrangement around the two-coordinate metal centre.138 Further examples include the dimethylgold(III) complexes with salicylaldimine Schiﬀ bases and carboxylate complexes with triﬂuoroacetate, trimethylacetate and benzoate ligands reported by Bessonov.139,140 Abbreviations acac bimy 2-bipy 2-bipy-5

acetylacetonato benzimidazol-2-ylidene 2,2 0 -bipyridine 2,2 0 -bipyridine-5-yl Organomet. Chem., 2010, 36, 148–167 | 161

4-bipy Boc bpee bppe C2pyC2 cax cod cpta da dipp dme dmf dmp dmso dppe dppf en Hdp H2sb H4btec hmds hmt lda mes mph mtta nbd nhc pmd pmdeta t Bu (R,R)-tecda teeda ter thf tht tmeda tmp tph tpt tpy

4,4 0 -bipyridine tert-butoxycarbonyl 1,2-(E)-bis(4-pyridyl)ethene 1,2-bis[4-(3-pyridyl)pyrimidin-2-ylsulfanyl]ethane 3,5-diethynylpyridine tert-butylcalix[4]-areneH4 cyclooctadiene 5-(3-cyanophenyl)tetrazolate diisopropylamide 2,6-diisopropylphenyl dimethoxyethane dimethylformamide 2,6-(2,4,6-Me3C6H2)2C6H3 dimethylsulfoxide bis(diphenylphosphino)ethane 1,1 0 -bis(diphenylphosphino)ferrocene ethylenediamine 1H-1,2,4-diazaphosphole 4-[(4-hydroxyphenyl)sulfonyl]-1-benzenol 1,2,4,5-benzenetetracarboxylic acid 1,1,1,3,3,3-hexamethyldisilazide hexamethylenetetramine lithium diisopropylamide mesitylene 2-(2,4,6-Me3C6H2)C6H4 5-methyltetrazolate norbornadiene N-heterocyclic carbene pyrimidine N,N,N 0 ,N 0 0 ,N 0 0 -pentamethyldiethylenetriamine tert-butyl (1R,2R)-tetraethylcyclohexane-1,2-diamine N,N,N 0 ,N 0 -tetraethylethylenediamine terphenyl tetrahydrofuran tetrahydrothiophene tetramethylethylenediamine 2,2,6,6-tetramethylpiperidide 2-(2,4,6-iPr3C6H2)C6H4 2,4,6-tris(4-pyridyl)-1,3,5-triazine 2,2 0 :6 0 ,2 0 0 -terpyridine

References 1 D. R. Armstrong, E. Herd, D. V. Graham and A. R. Kennedy, Dalton Trans., 2008, 1323. 2 D. R. Armstrong, W. Clegg, S. H. Dale, J. Garcı´ a-A´lvarez, R. W. Harrington, E. Hevia, G. W. Honeyman, A. R. Kennedy, R. E. Mulvey and C. T. O’Hara, Chem. Commun., 2008, 187. 162 | Organomet. Chem., 2010, 36, 148–167

3 W. Clegg, J. Garcy´a-Alvarez, P. Garcı´ a-A´lvarez, D. V. Graham, R. W. Harrington, E. Hevia, A. R. Kennedy, R. E. Mulvey and L. Russo, Organometallics, 2008, 27, 2654. 4 J. Garcı´ a-A´lvarez, D. V. Graham, E. Hevia, A. R. Kennedy and R. E. Mulvey, Dalton Trans., 2008, 4, 1481. 5 B. Conway, D. V. Graham, E. Hevia, A. R. Kennedy, J. Klett and R. E. Mulvey, Chem. Commun., 2008, 2638. 6 C. Strohmann, V. H. Gessner and A. Damme, Chem. Commun., 2008, 234, 3381. 7 P. Buchalski, I. Grabowska, E. Kaminska and K. Suwinska, Organometallics, 2008, 27, 2346. 8 C. Chen, R. Fro¨hlich, G. Kehr and G. Erker, Organometallics, 2008, 27, 3248. 9 M. Chiu, H. M. Hoyt, F. E. Michael and R. G. Bergman, Angew. Chem. Int. Ed., 2008, 47, 6073. 10 A. Fu¨rstner, K. Majima, R. Martin, H. Krause, E. Kattnig, R. Goddard and C. W. Lehmann, J. Am. Chem. Soc., 2008, 130, 1992. 11 A. Fu¨rstner, R. Martin, H. Krause, G. Seidel, R. Goddard and C. W. Lehmann, J. Am. Chem. Soc., 2008, 130, 8773. 12 H. V. Ly, J. Konu, M. Parvez and R. Roesler, Dalton Trans., 2008, 3454. 13 I. Nieto, R. P. Bontchev and J. M. Smith, Eur. J. Inorg. Chem., 2008, 2476. 14 D. J. Eisler, R. J. Less, V. Naseri, J. M. Rawson and D. S. Wright, Dalton Trans., 2008, 2382. 15 G. Fraenkel, X. Chen, J. Gallucci and Y. Ren, J. Am. Chem. Soc., 2008, 130, 4140. 16 E. D. Brady, S. C. Chmely, K. C. Jayaratne, T. P. Hanusa and V. G. Young, Organometallics, 2008, 27, 1612. 17 L. Wan, C. Pi, L. Zhang, W. Zheng, L. Weng, Z. Chen and Y. Zhang, Chem. Commun., 2008, 2266. 18 N. Meyer, P. W. Roesky, S. Bambirra, A. Meetsma, B. Hessen, K. Saliu and J. Takats, Organometallics, 2008, 27, 1501. 19 R. C¸elenligil-C¸etin, P. Paraskevopoulou, R. Dinda, N. Lalioti, Y. Sanakis, A. M. Rawashdeh, R. J. Staples, E. Sinn and P. Stavropoulos, Eur. J. Inorg. Chem., 2008, 1, 673. 20 X. Cheng, J. Zhang, H. Song and C. Cui, Organometallics, 2008, 27, 678. 21 N. D. Coombs, A. Stasch, A. Cowley, A. L. Thompson and S. Aldridge, Dalton Trans., 2008, 332. 22 S. Deuerlein, D. Leusser, U. Flierler, H. Ott and D. Stalke, Organometallics, 2008, 27, 2306. 23 M. A. Guino-o, C. F. Campana and K. Ruhlandt-senge, Chem. Commun., 2008, 1692. 24 H. S. Lee, S. Hauber, D. Vindus and M. Niemeyer, Inorg. Chem., 2008, 47, 4401. 25 D. J. Loron˜o-Gonzalez, Acta Crystallogr. C, 2008, 64, m228. 26 C. So, D. Watanabe, A. Wakamiya and S. Yamaguchi, Organometallics, 2008, 27, 3496. 27 P. M. Jeﬀries, R. E. Ellenwood and G. S. Girolami, Inorg. Chim. Acta, 2008, 361, 3165. 28 B. Shen, L. Ying, J. Chen and Y. Luo, Inorg. Chim. Acta, 2008, 361, 1255. 29 V. Lavallo, C. A. Dyker, B. Donnadieu and G. Bertrand, Angew. Chem. Int. Ed., 2008, 47, 5411. 30 S. A. Solomon, C. A. Muryn and R. A. Lay, Chem. Commun., 2008, 3142. 31 T. Matsumoto, T. Sasamori, K. Sato, T. Takui and N. Tokitoh, Organometallics, 2008, 27, 305. Organomet. Chem., 2010, 36, 148–167 | 163

32 P. M. Morse, Q. D. Shelby, D. Y. Kim and G. S. Girolami, Organometallics, 2008, 27, 984. 33 H. Schumann, D. M. Freckmann and S. Dechert, Z. Anorg. Allg. Chem., 2008, 634, 1334. 34 T. Tatic, H. Ott and D. Stalke, Eur. J. Inorg. Chem., 2008, 2, 3765. 35 V. L. Blair, L. M. Carrella, W. Clegg, B. Conway, R. W. Harrington, L. M. Hogg, J. Klett, R. E. Mulvey, E. Rentschler and L. Russo, Angew. Chem. Int. Ed., 2008, 47, 6208. 36 V. L. Blair, W. Clegg, B. Conway, E. Hevia, A. Kennedy, J. Klett, R. E. Mulvey and L. Russo, Chem. Eur. J., 2008, 14, 65. 37 K. L. Hull, I. Carmichael, B. C. Noll and K. W. Henderson, Chem. Eur. J., 2008, 14, 3939. 38 J. Scott, I. Vidyaratne, I. Korobkov, S. Gambarotta and P. H. Budzelaar, Inorg. Chem., 2008, 47, 896. 39 A. G. Barrett, M. R. Crimmin, M. S. Hill, P. B. Hitchcock, G. Kociok-Ko¨hn and P. A. Procopiou, Inorg. Chem., 2008, 47, 7366. 40 R. Garcı´ a-Zarracino, M. Rangel-Marro´n, H. Tlahuext and H. Ho¨pﬂ, Acta Crystallogr. E, 2008, 64, m1626. 41 M. Ga¨rtner, H. Go¨rls and M. Westerhausen, Inorg. Chem., 2008, 47, 1397. 42 C. Redshaw, X. Liu, S. Zhan, D. L. Hughes, H. Baillie-Johnson, M. R. Elsegood and S. H. Dale, Eur. J. Inorg. Chem., 2008, 2698. 43 K. Ruth, S. Tu¨llmann, H. Vitze, M. Bolte, H. Lerner, M. C. Holthausen and M. Wagner, Chem. Eur. J., 2008, 14, 6754. 44 S. J. Dalgarno, K. M. Claudio-Bosque, J. E. Warren, T. E. Glass and J. L. Atwood, Chem. Commun., 2008, 1410. 45 M. Bro¨ring, S. Ko¨hler, S. Link, O. Burghaus, C. Pietzonka, H. Kelm and H. Kru¨ger, Chem. Eur. J., 2008, 14, 4006. 46 K. Salorinne and M. Nissinen, Tetrahedron, 2008, 64, 1798. 47 W. Wang, Acta Crystallogr. E, 2008, 64, m759. 48 W. Wang, Acta Crystallogr. E, 2008, 64, m900. 49 W. Wang, Acta Crystallogr. E, 2008, 64, m902. 50 W. Wang, Acta Crystallogr. E, 2008, 64, m930. 51 Z. Li, P. Lin and S. Du, Polyhedron, 2008, 27, 232. 52 S. Liang, M. Li, M. Shao and X. He, J. Mol. Str., 2008, 875, 17. 53 G. Agustı´ , A. L. Thompson, A. B. Gaspar, M. C. Mun˜oz, A. E. Goeta, J. A. Rodrı´ guez-Velamaza´n, M. Castro, R. Burriel and J. A. Real, Dalton Trans., 2008, 642. 54 M. Li, Z. Li and D. Li, Chem. Commun., 2008, 3, 3390. 55 S. Lin, Y. Yang and S. W. Ng, Acta Crystallogr. E, 2008, 64, m1076. 56 W. Zhang, Y. Song, Y. Zhang and J. Lang, Crystallogr. Growth Des., 2008, 8, 253. 57 Y. Zhang, H. Dong and L. Cheng, Acta Crystallogr. E, 2008, 64, m868. 58 X. Liu and G. Guo, Aust. J. Chem., 2008, 481. 59 J. Zhang, Y. Song, J. Yang, M. G. Humphrey and C. Zhang, Crystallogr. Growth Des., 2008, 8, 387. 60 X. Liu and G. Guo, Crystallogr. Growth Des., 2008, 8, 776. 61 M. Stollenz, C. Grosse and F. Meyer, Chem. Commun., 2008, 1744. 62 M. J. Lim, C. A. Murray, T. A. Tronic, K. E. deKraﬀt, A. N. Ley, J. C. deButts, R. D. Pike, H. Lu and H. H. Patterson, Inorg. Chem., 2008, 47, 6931. 63 Y. Sun, C. Tsang, Z. Xu, G. Huang, J. He, X. Zhou, M. Zeller and A. D. Hunter, Crystallogr. Growth Des., 2008, 8, 1468. 64 C. Zhang, Y. Cao, J. Zhang, S. Meng, T. Matsumoto, Y. Song, J. Ma, Z. Chen, K. Tatsumi and M. G. Humphrey, Adv. Mater., 2008, 20, 1870. 164 | Organomet. Chem., 2010, 36, 148–167

65 D. Fu, H. Ye, Q. Ye, K. Pan and R. Xiong, Dalton Trans., 2008, 874. 66 M. Li, Z. Miao, M. Shao, S. Liang and S. Zhu, Inorg. Chem., 2008, 47, 4481. 67 J. Hou, M. Li, Z. Li, S. Zhan, X. Huang and D. Li, Angew. Chem. Int. Ed., 2008, 47, 1711. 68 Z. Li, M. Li, S. Zhan, X. Huang, S. W. Ng and D. Li, Crystallogr. Eng. Comm., 2008, 10, 978. 69 S. Pagola, R. D. Pike, K. deKraﬀt and T. A. Tronic, Acta Crystallogr. C, 2008, 64, m134. 70 W. Shi, C. Ruan, Z. Li, M. Li and D. Li, Crystallogr. Eng. Comm., 2008, 10, 778. 71 Y. Slyvka, V. Kinzhybalo, T. Lis, B. Mykhalitchko and M. Myœkiv, Z. Anorg. Allg. Chem., 2008, 634, 626. 72 M. Nonnenmacher, D. Kunz and F. Rominger, Organometallics, 2008, 27, 1561. 73 G. G. Dubinina, H. Furutachi and D. A. Vicic, J. Am. Chem. Soc., 2008, 130, 8600. 74 M. I. Bruce, N. N. Zaitseva, B. W. Skelton and A. H. White, J. Organomet. Chem., 2008, 693, 1400. 75 C. R. Samanamu´, P. M. Lococo, W. D. Woodul and A. F. Richards, Polyhedron, 2008, 27, 1463. 76 A. V. Knishevitsky, N. I. Korotkikh, A. H. Cowley, J. A. Moore, T. M. Pekhtereva, O. P. Shvaika and G. Reeske, J. Organomet. Chem., 2008, 693, 1405. 77 R. Packheiser, P. Ecorchard, B. Walfort and H. Lang, J. Organomet. Chem., 2008, 693, 933. 78 A. Hadzovic and D. Song, Organometallics, 2008, 27, 1290. 79 L. Chiang, C. Yeh, Z. Chan, K. Wang, Y. Chou, J. Chen, J. Wang and J. Y. Lai, Crystallogr. Growth Des., 2008, 8, 470. 80 D. G. Branzea, A. Guerri, O. Fabelo, C. Ruiz-Pe´rez, L. Chamoreau, C. Sangregorio, A. Caneschi and M. Andruh, Crystallogr. Growth Des., 2008, 8, 941. 81 P. I. Richards, J. F. Bickley, R. Boomishankar and A. Steiner, Chem. Commun., 2008, 1656. 82 M. Rukiah and M. Al-Ktaifani, Acta Crystallogr. C, 2008, 64, m170. 83 F. Li, J. Ma, J. Yang and Y. Liu, J. Chem. Crystallograllogr., 2008, 38, 525. 84 L. Busetto, M. Cristina Cassani, C. Femoni, A. Macchioni, R. Mazzoni and D. Zuccaccia, J. Organomet. Chem., 2008, 693, 2579. 85 C. Liu, Z. Chang, J. Wang, L. Yan, X. Bu and S. R. Batten, Inorg. Chem. Commun., 2008, 11, 889. 86 R. Bashiri, K. Akhbari, A. Morsali and M. Zeller, J. Organomet. Chem., 2008, 693, 1903. 87 T. Zhang, J. Kong, Y. Hu, X. Meng, H. Yin, D. Hu and C. Ji, Inorg. Chem., 2008, 47, 3144. 88 S. Zang, L. Zhao and T. C. Mak, Organometallics, 2008, 27, 2396. 89 L. Zhao, X. Chen and T. C. Mak, Organometallics, 2008, 27, 2483. 90 M. Chen, X. Xu, Z. Cao and Q. Wang, Inorg. Chem., 2008, 47, 1877. 91 C. Ganesamoorthy, J. T. Mague and M. S. Balakrishna, Eur. J. Inorg. Chem., 2008, 596. 92 D. Salazar-Mendoza, S. A. Baudron and M. W. Hosseini, Inorg. Chem., 2008, 47, 766. 93 H. V. Dias, S. Singh and C. F. Campana, Inorg. Chem., 2008, 47, 3943. 94 O. Winkelmann, C. Na¨ther and U. Lu¨ning, J. Organomet. Chem., 2008, 693, 923. Organomet. Chem., 2010, 36, 148–167 | 165

95 R. Cacciapaglia, S. Di Stefano, L. Mandolini, P. Mencarelli and F. Ugozzoli, Eur. J. Org. Chem., 2008, 186. 96 J. Skonieczny, L. Latos-Grazyn˜ski and L. Szterenberg, Chem. Eur. J., 2008, 14, 4861. 97 Y. Han, Y. Hong and H. V. Huynh, J. Organomet. Chem., 2008, 693, 3159. 98 D. Pugh, A. Boyle and A. A. Danopoulos, Dalton Trans., 2008, 1087. 99 F. Li, S. Bai and T. S. Hor, Organometallics, 2008, 27, 672. 100 J. Ye, S. Jin, W. Chen and H. Qiu, Inorg. Chem. Commun., 2008, 11, 404. 101 L. Ray, M. M. Shaikh and P. Ghosh, Inorg. Chem., 2008, 47, 230. 102 F. Jean-Baptiste dit Dominique, H. Gornitzka and C. Hemmert, J. Organomet. Chem., 2008, 693, 579. 103 Y. Zhou, X. Zhang, W. Chen and H. Qiu, J. Organomet. Chem., 2008, 693, 205. 104 Q. Liu, X. Zhao, X. Wu, L. Yin, J. Guo, X. Wang and J. Feng, Inorg. Chim. Acta, 2008, 361, 2616. 105 Q. Liu, L. Yin, X. Wu, J. Feng, J. Guo and H. Song, Polyhedron, 2008, 27, 87. 106 G. Dyson, J. Frison, S. Simonovic, A. C. Whitwood and R. E. Douthwaite, Organometallics, 2008, 27, 281. 107 J. Ye, W. Chen and D. Wang, Dalton Trans., 2008, 4015. 108 A. Fu¨rstner, M. Alcarazo, R. Goddard and C. W. Lehmann, Angew. Chem. Int. Ed., 2008, 47, 3210. 109 V. Lavallo, G. D. Frey, B. Donnadieu, M. Soleilhavoup and G. Bertrand, Angew. Chem. Int. Ed., 2008, 47, 5224. 110 R. Jothibasu, H. V. Huynh and L. L. Koh, J. Organomet. Chem., 2008, 693, 374. 111 U. E. Horvath, G. Bentivoglio, M. Hummel, H. Schottenberger, K. Wurst, M. J. Nell, C. E. van Rensburg, S. Cronje and H. G. Raubenheimer, New J. Chem., 2008, 32, 533. 112 M. K. Samantaray, K. Pang, M. M. Shaikh and P. Ghosh, Dalton Trans., 2008, 4893. 113 G. D. Frey, R. D. Dewhurst, S. Kousar, B. Donnadieu and G. Bertrand, J. Organomet. Chem., 2008, 693, 1674. 114 C. Bartolome´, M. Carrasco-Rando, S. Coco, C. Cordovilla, J. M. Martı´ nAlvarez and P. Espinet, Inorg. Chem., 2008, 47, 1616. 115 Y. Matsumoto, K. Yamada and K. Tomioka, J. Org. Chem., 2008, 73, 4578. 116 D. Rios, D. M. Pham, J. C. Fettinger, M. M. Olmstead and A. L. Balch, Inorg. Chem., 2008, 47, 3442. 117 N. D. Shapiro and F. D. Toste, Proc. Nat. Acad. Sci., 2008, 105, 2779. 118 I. O. Koshevoy, L. Koskinen, M. Haukka, S. P. Tunik, P. Y. Serdobintsev, A. S. Melnikov and T. A. Pakkanen, Angew. Chem. Int. Ed., 2008, 47, 3942. 119 R. Packheiser, A. Jakob, P. Ecorchard, B. Walfort and H. Lang, Organometallics, 2008, 27, 1214. 120 C. E. Strasser, W. F. Gabrielli, C. Esterhuysen, O. B. Schuster, S. D. Nogai, S. Cronje and H. G. Raubenheimer, New J. Chem., 2008, 32, 138. 121 E. C. Constable, C. E. Housecroft, M. Neuburger, S. Schaﬀner and E. J. Shardlow, Polyhedron, 2008, 27, 65. 122 X. Li, J. Mo, S. Zhang, L. Yuan and J. Liu, Acta Crystallogr. E, 2008, 64, m1126. 123 W. M. Khairul, D. Albesa-Jove´, D. S. Yuﬁt, M. R. Al-Haddad, J. C. Collings, F. Hartl, J. A. Howard, T. B. Marder and P. J. Low, Inorg. Chim. Acta, 2008, 361, 1646. 124 M. Osawa, M. Hoshino and D. Hashizume, Dalton Trans., 2008, 2248.

166 | Organomet. Chem., 2010, 36, 148–167

125 E. J. Ferna´ndez, A. Laguna, T. Lasanta, J. M. Lo´pez-de-Luzuriaga, M. Montiel and M. E. Olmos, Organometallics, 2008, 27, 2971. 126 M. J. Katz, V. K. Michaelis, P. M. Aguiar, R. Yson, H. Lu, H. Kaluarachchi, R. J. Batchelor, G. Schreckenbach, S. Kroeker, H. H. Patterson and D. B. Leznoﬀ, Inorg. Chem., 2008, 47, 6353. 127 G. Agustı´ , M. C. Mun˜oz, A. B. Gaspar and J. A. Real, Inorg. Chem., 2008, 47, 2552. 128 J. Qu, W. Gu and X. Liu, J. Coord. Chem., 2008, 61, 618. 129 T. Weidner, N. Ballav, M. Zharnikov, A. Priebe, N. J. Long, J. Maurer, R. Winter, A. Rothenberger, D. Fenske, D. Rother, C. Bruhn, H. Fink and U. Siemeling, Chem. Eur. J., 2008, 14, 4346. 130 P. Li, B. Ahrens, A. D. Bond, J. E. Davies, O. F. Koentjoro, P. R. Raithby and S. J. Teat, Dalton Trans., 2008, 1635. 131 M. Ferrer, A. Gutie´rrez, L. Rodrı´ guez, O. Rossell, J. C. Lima, M. Font-Bardia and X. Solans, Eur. J. Inorg. Chem., 2008, 2899. 132 J. A. Flores and H. V. Dias, Inorg. Chem., 2008, 47, 4448. 133 B. Pitteri, M. Bortoluzzi and V. Bertolasi, Transition Met. Chem., 2008, 33, 649. 134 W. Man, H. Kwong, W. W. Lam, J. Xiang, T. Wong, W. Lam, W. Wong, S. Peng and T. Lau, Inorg. Chem., 2008, 47, 5936. 135 S. Mori and A. Osuka, Inorg. Chem., 2008, 47, 3937. 136 J. Arias, M. Bardajı´ and P. Espinet, Inorg. Chem., 2008, 47, 1597. 137 H. V. Dias, M. Fianchini, T. R. Cundari and C. F. Campana, Angew. Chem. Int. Ed., 2008, 47, 556. 138 J. Vicente, J. Gil-Rubio, N. Barquero, P. G. Jones and D. Bautista, Organometallics, 2008, 27, 646. 139 A. A. Bessonov, N. B. Morozova, N. V. Kurat, I. A. Baidina, N. V. Gel and I. K. Igumenov, Russ. J. Coord. Chem., 2008, 34, 73. 140 A. A. Bessonov, N. B. Morozova, N. V. Gelfond, P. P. Semyannikov, I. A. Baidina, S. V. Trubin, Y. V. Shevtsov and I. K. Igumenov, J. Organomet. Chem., 2008, 693, 2572.

Organomet. Chem., 2010, 36, 148–167 | 167

Group 2 (Be-Ba) and Group 12 (Zn-Hg) Thomas H. Bullock,a Rebecca L. Melena and Dominic S. Wrighta DOI: 10.1039/97818147559616-00168

1.

Scope and organisation of the review

This review presents a perspective of the important structural and synthetic studies reported in 2008. The strict deﬁnition of an organometallic compound as one containing at least one C-metal bond or contact has been used throughout the literature survey. As with previous years this review is not intended to be comprehensive, although it is based on a comprehensive search. Individual topics are highlighted in bold in the text in order to facilitate rapid access to a particular area of the literature. Research in 2008 on Group 2 and 12 organometallics was dominated by structural studies of novel types of compounds. This is reﬂected in the focus of the review for this year. 2.

Group 2

Structural studies of r-bonded organometallics of Mg were relatively sparse in 2008.1–8 One of the highlights in this area has continued to be studies of heterometallic complexes.1–4 A recent study has shown for the ﬁrst time that synergic bases can be tuned to metalate arenes selectively at diﬀerent positions depending on the identity of the basic alkyl component.1 In situ generation of the mixed-metal bases [Na4M2(tmp)6(CH2SiMe3)2][M=Mn (1), Mg (2); tmp=2,2,6,6-tetramethylpiperidide] from a 4:6:2 ratio of BuNa/Htmp/Mn(CH2SiMe3)2 in n-hexane solution followed by stoichiometric reaction with toluene both aﬀord the same 3,5-deprotonation of the aromatic ring. The structurally-characterised products, [(tmp)6Na4(3,5M2C6H3CH3)] [M=Mn (3), Mg (4)], have identical inverse-crown arrangements in the solid state (Fig. 1). This outcome is in stark contrast with the result of deprotonation of toluene with the related heterometallic base [Na(nBu)(tmp)Mg(tmp)] (5), which gives the inverse-crown [(tmp)6Na4(2,5Mg2C6H3CH3)] (6) in which 2,5-deprotonation is observed.9 3 and 4 undergo electrophilic reactions with iodine to give 3,5-diiodotoluene. The key to this diﬀerent selectivity appears to be the switch from the less sterically-demanding nBu in 5 to the more sterically-demanding -CH2SiMe3 group in 1 and 2. These results have important implications to a broad range of aromatic substitution reactions in particular. The reactivity of the new mixed-metal base [(tmeda)Na(CH2SiMe3) (tmp)Mg(tmp)] (7)2 (Fig. 2a) with furan is also of interest in respect to that of the previously reported reagent [(tmeda)Na(nBu)(tmp)Mg(tmp)] (8).10 Reaction of furan with 7 leads to metallation at the 2-position of the aromatic ring within the isolated product [{(thf)3Na2}{(tmeda)Mg2}(2-C4H3O)6}N] (9),10 whereas 2,5-dimetallation as well as 2-monometallation are observed in the highly elaborate structure of [{(tmeda)3Na6Mg3(CH2SiMe3)(2,5-C4H2O)3 (2-C4H3O)5}2] (10) (obtained from reaction of 7 with furan).2 The bowl-shaped, a

Department of Chemistry, University of Cambridge, Lensﬁeld Road, Cambridge, CB2 1EW, UK

168 | Organomet. Chem., 2010, 36, 168–181 c

The Royal Society of Chemistry 2010

N Na

Na CH3

N

N

M

M

N

N Na

Na N

3 (M = Mn), 4 (M = Mg) Fig. 1

(a)

(b)

N

Mg

N

Mg

O

Mg

Na

O

O

N CH2 SiMe3

Mg

7 Fig. 2

metallacyclic [Mg(2,5-C4H2O)}3] fragments present in 10 (Fig. 2b) closely resemble the arrangement found in subporphyrins. Metallacylic arrangements are also found in the ﬁrst structurally characterised magnesium organocuprates derived from Grignard reagents.4 Reaction of [PhMgI] with [CuMes] (Mes=2,4,6-Me3C6H2] in Et2O gives [(Ph2Cu)(MgI Et2O)2]2 (11), in which (a crown-like arrangement is formed by the association of two [Ph2Cu] ions with two [IMg.Et2O] þ cations [similar to the previously reported lithium complex [(Ph2Cu)(Li.Et2O)2]2 (12)].11 However, a uniquely larger metallacyclic structure is formed in the reaction of [CuMes] with [MesMgBr] in Et2O. The product is [{Mes3Cu2}(MgBr)]2 (13), formed from the association of [Mes3Cu2] anions with an [Mg(m-Br)]22 þ unit (Fig. 3).4 The applications of new types of s-organometallic ligands in Mg chemistry, like the amido-dicarbene 14, are also worthy of note5 (Fig. 4). Very few studies of r-bonded organometallics of the heavier Group 2 metals were reported in 2008.12–15 A recent review in this area has provided a Organomet. Chem., 2010, 36, 168–181 | 169

R Cu

Cu

R

R Br Mg

Mg Br

R

R Cu

Cu R

13

R = Mes Fig. 3

Mes

Mes

N

N

N

N

N

14 Fig. 4

comprehensive overview of structural and synthetic developments in both s- and p-bonded organometallics of Ca, Sr and Ba and their reactivity.14 Of the recent s-bonded examples, the benzylate complex [(tmeda.Li)2 {(PhCH2)4Ca}] (15) provides a valuable addition to this still rare class of heavier group 2 compounds.13 p-Complexes of the Group 2 elements in 2008 were dominated by the cyclopentadienyl family.16–22 In a particularly interesting recent study it was shown that the reaction of solid [Ba(Z5-Cp*)2] with excess [In(Z5-Cp*)] results in C-H bond activation of the barocene unit and the formation of the new barium metallocene [Ba(Z5-Cp*){Z5-C5Me5(CH2C5Me4)}] (16) (Fig. 5).16 Two mechanistic pathways to 16 were proposed, a concerted mechanism involving the insertion of [In(Z5-Cp*)] into a C-H bond of [Ba(Z5-Cp*)2] followed by a formal extrusion of InH, or a two-step radical mechanism in which [In(Z5-Cp*)] ﬁrstly cleaves homolytically into In metal and the [C5Me5]. radical. The latter radical then abstracts a proton from a CH3 group of [Ba(Z5-Cp*)2] with the resulting [Ba(Z5-Cp*){Z5-C5Me4(CH2)}]. radical combining with a [C5Me5]. radical to give 16 (Scheme 1). The superbulky ligand [(4-nBu-C6H4)5C5] (CpBIG)17,18 has recently proved remarkably eﬀective at the stabilization of highy-reducing, unstable metal oxidation states. The reactions of the benzyl lanthanide metal complexes [{2-Me2N-1-CH2-C6H4}3LnIII] (17) with CpBIGH give the metallocenes [LnII(Z5-CpBIG)2] (18), in which spontaneous reduction of LnIII to LnII 170 | Organomet. Chem., 2010, 36, 168–181

M

Fig. 5

[In( 5-Cp*)] Ba

Ba

16 Scheme 1

has occurred. The stabilizing eﬀect of the CpBIG ligand is explained by the presence of short, intramolecular C-H?arene interactions between the ortho-H and ortho-C atoms of the 4-nBu-C6H4 groups of the two CpBIG ligands in the structurally characterized Ca, Sr, Ba, Yb and Sm (=M) sandwich compounds [2.68–2.89 A˚] (Fig. 5).17,18 These interactions are responsible for an atypical bending of the 4-nBu-C6H4 groups downwards towards the coordinated metal centres, resulting in shorter Cp?metal distances than expected and providing further steric shielding. Studies of p-complexes containing ligands other than the Cp family of ligands have also appeared in the literature in 2008.23–25 Of particular note is the synthesis of Mg and Ca complexes of the tetrakis(trimethylsilyl)cyclobutadiene dianion, [Mg(thf)3][(Me3Si)4C4] (19) and [Ca(thf)n][(Me3Si)4C4] (20), via the reduction of tetrakis(trimethylsilyl)cyclobutadiene [(Me3Si)4C4] with metallic Mg or Ca. The structure of the half-sandwich 19 shows the presence of an Z4-bonded [(Me3Si)4C4]2 6p-dianion to the Mg centre.23 As in previous years, the applications of Grignard reagents in organic synthesis have been a major area of research in 2008. Recent reports can be divided broadly into fundamental mechanistic studies,26,27 metal-catalysed coupling, addition and substitution reactions,28–33 the exploration of novel types of reactions involving Grignards34–40 and the development of new Organomet. Chem., 2010, 36, 168–181 | 171

types of Grignard reagents for synthesis.41–47 Very few studies involving the applications of the heavier Group 2 organometallics have been reported.48 This area is far too extensive to allow comprehensive coverage in this review. However, some key studies are worthy of mention. The recent measurement of the relative rates of Br/Mg exchange of substituted aromatic bromides is relevant to developments in the direct functionalisation of aryl bromides with Grignard-based reagents of the type i PrMgCl/LiCl in the past few years,27 and should be of value in facilitating targeted use of reagents like this with polyfunctional aromatics. In this context, it has recently been shown that the N,N,N 0 ,N 0 -tetramethylphosphorodi-amidate group [(Me2N)2P(O)O-] is a very strong directing group for magnesiation and that it may override the eﬀects of other substituents present on the aromatic substrate.35 Metallation of para- (21) or meta- (22) substituted precursors with the reagent [(tmp)2Mg.2LiCl] (23) followed by addition of electrophiles. The [(Me2N)2P(O)O-] groups can then be readily substituted by nucleophiles to give para/ortho (24) or ortho/para (25) functionalized products. The nature of the directed metallation stems from the coordination of 25 by the [(Me2N)2P(O)O-] group, resulting in activation of the N-Mg bonds by inducing ’ate character (Scheme 2). The related reagent [(tmp)MgCl.LiCl] (26) has also been employed in the development of a new methodology for the selective meta- and paradifunctionalization of arenes via sulfoxide-magnesium exchange,36 whilst 23 and 26 have also been shown to be enable the full functionalization of protected uracils and thiouracils in good to excellent yields.37

FG

FG

FG

FG

O P O para- (21) meta- (22)

E

-tmpH

23 O NMe2 NMe2

tmp

P

tmp

O

NMe2 NMe2

Mg tmp

P O

E

24

FG

O

H Mg

Nu Nu

NMe2 NMe2 E Nu

25

Scheme 2

In a novel use of Grignard reagents it has been shown that their addition to [60]fullerene takes place smoothly in the presence of dimethylformamide to produce (organo)(hydro)[60]fullerenes, C60R1H (27) (Scheme 3).38 The hydrofullerene was then deprotonated to generate the corresponding anion, [C60R1] , which can then be alkylated to obtain 58p-electron di(organo)[60]fullerenes, C60R1R2. This two-step methodology provides access to a wide variety of 1,4-di(organo)[60] fullerenes. In a novel application of Grignard reagents (RMgX) it has also been shown that their reactions with Au nanoparticles, bearing a mixed 172 | Organomet. Chem., 2010, 36, 168–181

MgCl

H

R1

R1

R1MgX/ DMF

NH4Cl

27 Scheme 3

monolayer of the alkane thiol ligands S(CH2)9Me and S(CH2)10 (CQO)N(Me)(OMe), can be employed in nanoparticle surface modiﬁcation.49 The N(Me)(OMe) groups rapidly react with RMgX or RLi, to give (CQO)R terminated groups. Demonstrating the utility of these strong organometallic reagents opens the door to a large class of reactions that are under-utilized within the ﬁeld of nanomaterials. 3.

Group 12

Structural studies of r-bonded Zn organometallics were particularly numerous in 2008.50–84 In comparison, the structural chemistry of Cd organometallic was far less studied.84–86 Like organomagnesium described earlier, many of the most important studies of organozinc complexes have concerned heterobimetallic s-block/Zn compounds and their applications as regioselective reagents for organic synthesis.50–55 The structurally characterized bimetallic (R,R)-[(tmeda)NaZn(tBu)2{N(CH2Ph)(C*H(CH3)Ph)}] (28), featuring a chiral amido group, represents an interesting development in this area with the potential for chiral metallation of organic substrates.50 Metallation of m-tolunitrile (1-Me-3-CN)-C6H4) or 1-cyano-naphthalene (1-CN)-C10H7) with the mixed-metal base [(tmeda)Na(tBu)(tmp)Zn(tBu)] (29) in excess tmeda results in zincation at a position ortho to the CN group (leaving the CN functionality in tact).53A related regioselective metallation of phenyl dimethylcarbamate (30) with the reagent [(thf)nLi(tmp)ZnEt2] (31) gives the ortho-metallated product [C6H4{OC(QO)NMe2}{Zn(mtmp)Et}Li]2 (32) (a dimer in the solid state) (Scheme 4), without the normal anionic Fries rearrangement occurring using organolithium bases.55 NMe2

NMe2 O

O

O

N 2 (thf)nLi

Li Zn

N

-EtH

Et + 2

Zn Et

Et 30

O

2

32

31 Scheme 4

Organomet. Chem., 2010, 36, 168–181 | 173

R Zn

H H

Zn Zn

H

H

R

Me3Si

R

H Zn H

C

Me3Si R=

N Me2Si

Zn

N

R

N

33 Fig. 6

Of particular structural interest is the new zinc hydride complex 33 (Fig. 6), obtained by reduction of [RZnBr] with NaH. 33 can be regarded as an encapsulated [ZnH4]2 dianion.56 A study of the reaction of ZnMe2 with [iPrNQCQNiPr] (34) is also worthy of note.57 This reaction does not lead to addition of a methyl group to the C atom of 34 but to a series of tetranuclear Zn4 cages (35, 36 and 37, Fig. 7) in which extensive C-C bond formation has occurred between the amidinate units. Interestingly, the anion units in the structurally characterized complexes 36 and 37 are isomeric, with calculations showing that the unit of 36 is about 16.6 kcal mol 1 more stable than that of 37. The formation of C-C bonds in this reaction has important implications for the addition reactions of other organometallics to carbodiimides like 34.

N N

N

N

Zn N

N

Zn

Zn

N

Zn

N

N

N

N

N

N

Zn

36

N

Zn

N N N

Zn

N

N Zn

35

Zn

N

N

N

N Zn

37

Fig. 7

Reports on the reactions of molecular O2 with organozinc compounds have shed new light of the mechanisms of C-Zn insertion and oxo-ligand incorporation.58,60 A detailed structural and NMR spectroscopic study strongly suggests that the reaction of O2 with the mononuclear organozinc species [{(CH)N(tBu)}2ZnMe2] (38) occurs via the peroxide 39, which then rearranges in the peroxo-cubane 40a before ultimately forming the unusual double-cubane 41 (Scheme 5).58 The observation of C-N bond cleavage in the other ﬁnal product of this reaction 40b is relevant to the reactivity of 34 with ZnMe2 (noted above), a radical mechanism seeming most likely in both of these studies. The reaction of O2 with the hexamer [EtZn(O2CPh)]6 (41) also occurs via an organometallic peroxide, before an O2 dianion is introduced into the centre of the ﬁnal product [Zn4(m4-O)(O2CPh)6] (42). 174 | Organomet. Chem., 2010, 36, 168–181

Me

Me

Me t

O Bu N

t

t

Bu N

Me Zn

t

Me

Me t

N Bu t

Bu N N H t N Bu N Zn t Zn Bu Me Me O t

Bu

38

Bu

Zn N

OOMe

N t

Zn O

40a

Zn

Bu

Zn

t

N

Me

N

O Zn

N Me t Bu

Bu

O2

Me O Zn O Zn O

39 t

Bu

Bu

N Me Zn Zn

Me Me O Zn

O

O

Bu N

Zn

Me

Zn t

O Zn

Bu

O

Zn O

t

40b

Zn

O Me

O

N

Me

Me

Zn 41

N t

Bu Scheme 5

Related reaction of 41 with S8 at room temperature gives the suﬁdo-cage [Zn6(m3-S)2-(O2Ph)8(thf)2] (43).60 Once again structural studies of r-bonded Hg organometallics have again been numerous in 2008.84,87–103 Much of the structural interest in these species stems from the presence of Hg?Hg and Hg?donor interactions and the manner in which these can inﬂuence molecular and supramolecular structure. Particularly interesting is the observation of Hg?Pd bonding within the structures of the complexes [Pd(salophen)2Hg(C5F5)2] (44) (salophen=N,N 0 -disalicylidene-o-phenylenediaminate) and [Pd(N6C)(OAc)2 Hg(C5F5)2] (45) [N6C=2-2-pyridyl)phenyl-C,N] (Fig. 8).87 The presence of a mononuclear Pd donor in 44 results in a longer Pd?Hg interaction [3.2841(2) A˚] since the ﬁlled dz2 orbital of PdII is too high in energy to form a stronger donor interaction with HgII. In 45, however, the two PdII centres are forced together by the bridging AcO ligands, resulting in the formation of ﬁlled s and s* Pd MOs. The presence of the ﬁlled s* orbital that is Organomet. Chem., 2010, 36, 168–181 | 175

Me Me O

Pd

N

N

C 6F 5

O

O O

Pd

Hg

Pd

N

C6F5

N

Pd

O O

O O Me

Me 45 Fig. 8

orientated towards the HgII centre and is higher in energy than the dz2 orbital in 44 results in an increase in the basicity and a shorter Pd?Hg bond [3.1085(8) A˚]. Overall, the Pd?Hg interactions in 44 and 45 originate from favourable dispersion forces complemented, at least for 45, by a PdII?HgII donor-acceptor component. Only a few reports of p-complexes of Zn,104,105 Cd and Hg102,103 have appeared, with two reviews highlighting important new advances in the area of metal-metal bonded s- and p-complexes also appearing in 2008.106,107 Of these studies, the synthesis and structure of [{Mo(CO)4}4(Zn)6(ZnCp*)4] (46) is worthy of speciﬁc mention (Fig. 9).104 Also of interest in the p-arene complex [(Z3-C6Me6)(C6F5)Zn(m-Cl)]2 (47).105

Cp*Zn (CO)4Mo Zn (CO)4Mo

ZnCp* Zn

Mo(CO)4

Zn Zn

Zn

Zn

Mo(CO)4

Cp*Zn

ZnCp* 46 Fig. 9

Like organomagnesium compounds discussed above, applications of organozinc compounds in organic synthesis continues to grow.108–132 Far fewer reports of Cd and Hg organometallics have appeared in 2008.133 Studies of organozinc reagents have concerned the development of new types of reagents,108–114 further developments in the synthesis and applications of polyzinc reagents,115,116 transition metal-mediated coupling reactions using organozincs,117–129 new deprotonation reagents,130,131 and 176 | Organomet. Chem., 2010, 36, 168–181

applications in polymerization.132 A report of a general method for the expedient synthesis of salt-free diorganozinc reagents using zinc methoxide is also worthy of special mention (Scheme 6).134

-Mg(OMe)X

Zn(OMe)2 + < 2 RMgX

Et2O

ZnR2

Scheme 6

References 1 V. L. Blair, L. M. Carrella, W. Clegg, B. Conway, R. W. Harrington, L. M. Hogg, J. Klett, R. E. Mulvey, E. Rentschler and L. Russo, Angew. Chem., Int. Ed., 2008, 47, 6208. 2 V. L. Blair, A. R. Kennedy, J. Klett and R. E. Mulvey, Chem. Commun., 2008, 5426. 3 J. Garcı´ a-A´lvarez, D. V. Graham, E. Hevia, A. R. Kennedy and R. E. Mulvey, Dalton Trans., 2008, 1481. 4 R. Bomparola, R. P. Davies, S. Hornauer and A. J. P. White, Angew. Chem., Int. Ed. Engl., 2008, 47, 5812. 5 P. L. Arnold, I. S. Edworthy, C. D. Carmichael, A. J. Blake and C. Wilson, Dalton Trans., 2008, 3739. 6 I. L. Fedushkin, A. G. Morozov, M. Hummert and H. Schumann, Eur. J. Inorg. Chem., 2008, 1584. 7 S. A. Solomon, C. A. Muryn and R. A. Layﬁeld, Chem. Commun., 2008, 3142. 8 M. Ga¨rtner, H. Go¨rls and M. Westerhausen, Inorg. Chem., 2008, 47, 1397. 9 D. R. Armstrong, A. R. Kennedy, R. E. Mulvey and R. B. Rowlings, Angew. Chem. Int. Ed., 1999, 38, 131. 10 D. V. Graham, E. Hevia, A. R. Kennedy, R. E. Mulvey, C. T. O’Hara and C. Talmard, Chem. Commun., 2006, 417. 11 N. P. Lorenzen and E. Weiss, Angew. Chem., Int. Ed. Engl., 1990, 29, 300. 12 M. Ga¨rtner, H. Go¨rls and M. Westerhausen, J. Organomet. Chem., 2008, 693, 221. 13 M. A. Guino-o, C. F. Campana and K. Ruhlandt-Senge, Chem. Commun., 2008, 1692. 14 M. Westerhausen, Coord. Chem. Rev., 2008, 252, 1516. 15 A. G. M. Barrett, M. R. Crimmin, M. S. Hill, P. B. Hitchcock, S. L. Lomas, M. F. Mahon, P. A. Procopiou and K. Suntharalingam, Organometallics, 2008, 27, 3600. 16 M. Wiecko, C. Eidamshaus, R. Ko¨ppe and P. W. Roesky, Dalton Trans., 2008, 4837. 17 C. Ruspic, J. R. Moss, M. Schu¨rmann and S. Harder, Angew. Chem., Int. Ed. Engl., 2008, 47, 2121. 18 L. Orzechowski, D. F.-J. Piesik, C. Ruspic and S. Harder, Dalton Trans., 2008, 4742. 19 E. D. Brady, S. C. Chmely, K. C. Jayaratne, T. P. Hanusa and V. G. Young Jr., Organometallics, 2008, 27, 1612. 20 D. Kazhdan, Y.-J. Hu, A. Kokai, Z. Levi and S. Rozenel, Acta Cryst., 2008, E64, m1134. 21 A. Jaenschke, J. Paap and U. Behrens, Z. Anorg. Allg. Chem., 2008, 634, 461. 22 M. Huber and H. Schno¨ckel, Inorg. Chim. Acta, 2008, 361, 457. Organomet. Chem., 2010, 36, 168–181 | 177

23 K. Takanashi, A. Inatomi, V. Ya. Lee, M. Nakamoto, M. Ichinohe and A. Sekiguchi, Eur. J. Inorg. Chem., 2008, 1752. 24 T. M. Cameron, C. Xu, A. G. Dipasquale and A. L. Rheingold, Organometallics, 2008, 27, 1596. 25 L. He, J. Cheng, T. Wang, C. Li, Z. Gong, H. Liu, B.-B. Zeng, H. Jiang and W. Zhu, Chem. Phys. Letts., 2008, 462, 45. 26 F. Bickelhaupt, M. Newcomb, C. B. DeZutter and H. J. R. de Boer, Eur. J. Org. Chem., 2008, 6225. 27 L. Shi, Y. Chu, P. Knochel and H. Mayr, Angew. Chem., Int. Ed., 2008, 47, 202. 28 C. Studte and B. Breit, Angew. Chem. Int. Ed., 2008, 47, 5451. 29 L. Hintermann, L. Xiao and A. Labonne, Angew. Chem., Int. Ed., 2008, 47, 8246. 30 J. Terao and N. Kambe, Acc. Chem. Res., 2008, 41, 1545. 31 K. Murakami, K. Hirano, H. Yorimitsu and K. Oshima, Angew. Chem. Int. Ed., 2008, 47, 5833. 32 T. Robert, J. Velder and H.-G. Schmalz, Angew. Chem. Int. Ed., 2008, 47, 7718. 33 K. B. Selim, K.-I. Yamada and K. Tomioka, Chem. Commun., 2008, 5140. 34 M. S. Maji, T. Pfeifer and A. Studer, Angew. Chem., Int. Ed., 2008, 47, 9547. 35 C. J. Rohbogner, G. C. Clososki and P. Knochel, Angew. Chem., Int. Ed., 2008, 47, 1503. 36 C. B. Rauhut, L. Melzig and P. Knochel, Org. Lett., 2008, 10, 3891. 37 M. Mosrin, N. Boudet and P. Knochel, Org. Biomol. Chem., 2008, 6, 3237. 38 Y. Matsuo, A. Iwashita, Y. Abe, C.-Z. Li, K. Matsuo, M. Hashiguchi and E. Nakamura, J. Am. Chem. Soc., 2008, 130, 15429. 39 T. Ramnial, S. A. Taylor, M. L. Bender, B. Gorodetsky, P. T. K. Lee, D. A. Dickie, B. M. McCollum, C. C. Pye, C. J. Walsby and J. A. C. Clyburne, J. Org. Chem., 2008, 73, 801. 40 G. Brunner, L. Eberhard, J. Oetiker and F. Schro¨der, J. Org. Chem., 2008, 73, 7543. 41 V. A. D’yakonov, A. A. Makarov, A. G. Ibragimov, L. M. Khalilov and U. M. Dzhemilev, Tetrahedron, 2008, 64, 10188. 42 L. Huang, S. Wu, Y. Qu, Y. Geng and F. Wang, Macromolecules, 2008, 41, 8944. 43 H. Todo, J. Terao, H. Watanabe, H. Kuniyasu and N. Kambe, Chem. Commun., 2008, 1332. 44 F. M. Piller, P. Appukkuttan, A. Gavryushin, M. Helm and P. Knochel, Angew. Chem., Int. Ed., 2008, 47, 6802. 45 A. Metzger, F. M. Piller and P. Knochel, Chem. Commun., 2008, 5824. 46 C. Despotopoulou, R. C. Bauer, A. Krasovskiy, P. Mayer, J. M. Stryker and P. Knochel, Chem. Eur. J., 2008, 14, 2499. 47 C. B. Rauhut, V. A. Vu, F. F. Fleming and P. Knochel, Org. Lett., 2008, 10, 1187. 48 T. M. A. Al-Shboul, H. Go¨rls and M. Westerhausen, Inorg. Chem. Commun., 2008, 11, 1419. 49 C. J. Thode and M. E. Williams, Langmuir, 2008, 24, 5988. 50 D. R. Armstrong, W. Clegg, S. H. Dale, J. Garcı´ a-A´lvarez, R. W. Harrington, E. Hevia, G. W. Honeyman, A. R. Kennedy, R. E. Mulvey and C. T. O’Hara, Chem. Commun., 2008, 187. 51 D. R. Armstrong, C. Dougan, D. V. Graham, E. Hevia and A. R. Kennedy, Organometallics, 2008, 27, 6063. 178 | Organomet. Chem., 2010, 36, 168–181

52 D. R. Armstrong, E. Herd, D. V. Graham, E. Hevia, A. R. Kennedy, W. Clegg and L. Russo, Dalton Trans., 2008, 1323. 53 W. Clegg, S. H. Dale, E. Hevia, L. M. Hogg, G. W. Honeyman, R. E. Mulvey, C. T. O’Hara and L. Russo, Angew. Chem., Int. Ed., 2008, 47, 731. 54 B. Conway, D. V. Graham, E. Hevia, A. R. Kennedy, J. Klett and R. E. Mulvey, Chem. Commun., 2008, 2638. 55 F. Garcı´ a, M. McPartlin, J. V. Morey, D. Nobuto, Y. Kondo, H. Naka, M. Uchiyama and A. E. H. Wheatley, Eur. J. Org. Chem., 2008, 644. 56 M. P. Coles, S. M. El-Hamruni, J. D. Smith and P. B. Hitchcock, Angew. Chem., Int. Ed., 2008, 47, 10147. 57 M. Mu¨nch, U. Flo¨rke, M. Bolte, S. Schulz and D. Gudat, Angew. Chem., Int. Ed. Engl., 2008, 47, 1512. 58 J. Lewin´ski, K. Suwa"a, M. Kubisiak, Z. Ochal, I. Justyniak and J. Lipkowski, Angew. Chem., Int. Ed. Engl., 2008, 47, 7888. 59 E. Jaime, A. N. Kneifel, M. Westerhausen and J. Weston, J. Orgnaomet. Chem., 2008, 693, 1027. 60 J. Lewin´ski, W. Bury, M. Dutkiewicz, M. Maurin, I. Justyniak and J. Lipkowski, Angew. Chem. Int. Ed. Engl., 2008, 47, 573. 61 S. Miki, K. Nakamoto, J. Kawakami, S. Handa and S. Nuwa, Synthesis, 2008, 409. 62 J. F. Greco, M. J. McNevin, R. K. Shoemaker and J. R. Hagadorn, Organometallics, 2008, 27, 1948. 63 S. Javed and D. M. Hoﬀman, Inorg. Chem., 2008, 47, 11984. 64 M. R. J. Elsegood and C. Redshaw, Chem. Eur., J., 2008, 14, 3530. 65 S. Jana, R. Fro¨hlich and N. W. Mitzel, Z. Anorg. Allg. Chem., 2008, 634, 1477. 66 K. Kitano, N. Kuwamura, R. Tanaka, R. Santo, T. Nishioka, A. Ichimura and I. Kinoshita, Chem. Commun., 2008, 1314. 67 S. Datta, M. T. Gamer and Peter. W. Roesky, Dalton Trans., 2008, 1761. 68 I. L. Fedushkin, A. N. Tishkina, G. K. Fukin, M. Hummert and H. Schumann, Eur. J. Inorg., Chem., 2008, 483. 69 W.-C. Hung, S.-L. Laib and C.-C. Lin, Acta Cryst., 2008, E64, m129. 70 M. S. Khalaf, M. P. Coles and P. B. Hitchcock, Dalton Trans., 2008, 4288. 71 C. Koch, M. Kahnes, M. Schulz, H. Go¨rls and M. Westerhausen, Eur. J. Inorg. Chem., 2008, 1067. 72 J. Krahmer, R. Beckhaus, W. Saak and D. Haase, Z. Anorg. Allg. Chem., 2008, 634, 1696. 73 A. Lennartson and M. Ha˚kansson, Acta Cryst., 2008, C64, m10. 74 E. Martin, C. Spendley, A. J. Mountford, S. J. Coles, P. N. Horton, D. L. Hughes, M. B. Hursthouse and S. J. Lancaster, Organometallics, 2008, 27, 1436. 75 C. Neuha¨user, D. Domide, J. Mautz, E. Kaifer and H.-J. Himmel, Dalton Trans., 2008, 1821. 76 S. Suh, L. A. Mıˆ nea, S. Javed and D. M. Hoﬀman, Polyhedron, 2008, 27, 513. 77 W. Clegg, J. Garcı´ a-A´lvarez, P. Garcı´ a- A´lvarez, D. V. Graham, R. W. Harrington, E. Hevia, A. R. Kennedy, R. E. Mulvey and L. Russo, Organometallics, 2008, 27, 2654. 78 J. D. Farwell, P. B. Hitchcock, M. F. Lappert, G. A. Luinstra, A. V. Protchenko and X.-H. Wei, J. Organomet. Chem., 2008, 693, 1861. 79 F. A. Akkerman, R. Kickbusch and D. Lentz, Chem. Asian J., 2008, 3, 719. 80 C. Alonso-Moreno, A. Garce´s, L. F. Sa´nchez-Barba, M. Fajardo, J. Ferna´ndez-Baeza, A. Otero, A. Lara-Sa´nchez, A. Antin˜olo, L. Broomﬁeld, M. Isabel Lo´pez-Solera and A. M. Rodrı´ guez, Organometallics, 2008, 27, 1310. Organomet. Chem., 2010, 36, 168–181 | 179

81 Z.-Y. Chai, C. Zhang and Z.-X. Wang, Organometallics, 2008, 27, 1626. 82 M. Dochnahl, K. Lo¨hnwitz, J.-W. Pissarek, P. W. Roesky and S. Blechert, Dalton Trans., 2008, 2844. 83 W. Yao, Y. Mu, A. Gao, W. Gao and L. Ye, Dalton Trans., 2008, 3199. 84 P. Ferna´ndez, A. Sousa-Pedrares, J. Romero, J. A. Garcı´ a-Va´zquez, A. Sousa and P. Pe´rez-Lourido, Inorg. Chem., 2008, 47, 2121. 85 A. L. Johnson, N. Hollingsworth, G. Kociok-Ko¨hn and K. C. Molloy, Inorg. Chem., 2008, 47, 9706. 86 S. Jana, R. Fro¨hlich, A. Hepp and N. W. Mitzel, Organometallics, 2008, 27, 1348. 87 M. Kim, T. J. Taylor and F. P. Gabbaı¨ , J. Am. Chem. Soc., 2008, 130, 6332. 88 W. Henderson, B. K. Nicholson, S. M. Devoy and T. S. A. Hor, Inorg. Chim. Acta, 2008, 361, 1908. 89 Z. Li, G. Li, L. Wang, J. Wu and Z. Zhang, Inorg. Chem. Commun., 2008, 11, 691. 90 S. J. Sabounchei, H. Nemattalab, S. Salehzadeh, M. Bayat, H. R. Khavasi and H. Adams, J. Organomet. Chem., 2008, 693, 1975. 91 S. J. Sabounchei, A. Dadrass, F. Akhlaghi, Z. B. Nojini and H. R. Khavasi, Polyhedron, 2008, 27, 1963. 92 S. J. Sabounchei, H. Nemattalab, S. Salehzadeh, S. Khani, M. Bayat and H. R. Khavasi, Polyhedron, 2008, 27, 2015. 93 J. Seo, S. S. Lee, W.-T. Gong and K. Hiratani, Tet. Letts., 2008, 49, 3770. 94 W.-Y. Wong and Y.-H. Guo, J. Mol. Struct., 2008, 890, 150. 95 M. Akkurt, K. Karami, S. ¸ P. Yalc¸ına and O. Bu¨yu¨kgu¨ngo¨rc, Acta Cryst., 2008, E64, m612. 96 J. S. Casas, A. Castin˜eiras, M. D. Couce, M. Garcı´ a-Vega, M. Rosende, A. Sa´nchez, J. Sordo, J. M. Varela and E. M. V. Lo´pez, Polyhedron, 2008, 27, 2436. 97 K.-T. Chen, F.-A. Yang, J.-H. Chen, S.-S. Wang and J.-Y. Tung, Polyhedron, 2008, 27, 2216. 98 Q.-X. Liu, L.-N. Yin, X.-M. Wu, J.-C. Feng, J.-H. Guo and H.-B. Song, Polyhedron, 2008, 27, 87. 99 J. G. Melnick, K. Yurkerwich, D. Buccella, W. Sattler and G. Parkin, Inorg. Chem., 2008, 47, 6421. 100 M. Nolte, I. Pantenburg and G. Meyer, Z. Anorg. Allg. Chem., 2008, 634, 362. 101 C. Peng, F.-A. Yang, J.-H. Chen, S.-S. Wang and J.-Y. Tung, Polyhedron, 2008, 27, 23. 102 C. L. Dorsey, P. Jewula, T. W. Hudnall, J. D. Hoefelmeyer, T. J. Taylor, N. R. Honesty, C.-W. Chiu, M. Schulte and F. P. Gabbaı¨ , Dalton Trans., 2008, 4442. 103 T. J. Taylor, O. Elbjeirami, C. N. Burress, M. Tsunoda, M. I. Bodine, M. A. Omary and F. P. Gabbaı¨ , J. Inorg. Organomet. Polym., 2008, 18, 175. 104 T. Cadenbach, C. Gemel and R. A. Fischer, Angew. Chem. Int. Ed. Engl., 2008, 47, 9146. 105 Y. Sarazin, J. A. Wright, D. A. J. Harding, E. Martin, T. J. Woodman, D. L. Hughes and M. Bochmann, J. Organomet. Chem., 2008, 693, 1494. 106 A. Grirrane, I. Resa, A. Rodrı´ guez and E. Carmona, Coord. Chem. Rev., 2008, 252, 1532. 107 E. Carmona and A. Galindo, Angew. Chem. Int. Ed. Engl., 2008, 47, 6526. 108 W. R. Dolbier, P. Xie, L. Zhang, W. Xu, Y. Chang and K. A. Abboud, J. Org. Chem., 2008, 73, 2469. 109 M. Gilani and R. Wilhelm, Tet. Asymm., 2008, 19, 2346. 110 M. D. Helm, P. Mayer and P. Knochel, Chem. Commun., 2008, 1916. 111 K. Kobayashi, M. Ueno, H. Naka and Y. Kondo, Chem. Commun., 2008, 3780. 180 | Organomet. Chem., 2010, 36, 168–181

112 K. Kobayashi, H. Naka, A. E. H. Wheatley and Y. Kondo, Org. Lett., 2008, 10, 3375. 113 A. Metzger, M. A. Schade and P. Knochel, Org. Letts., 2008, 10, 1107. 114 M. Uchiyama, Y. Kobayashi, T. Furuyama, S. Nakamura, Y. Kajihara, T. Miyoshi, T. Sakamoto, Y. Kondo and K. Morokuma, J. Am. Chem. Soc., 2008, 130, 472. 115 A. Metzger, M. A. Schade, G. Manolikakes and P. Knochel, Chem. Asian J., 2008, 3, 1678. 116 A. Metzger, F. M. Piller and P. Knochel, Chem. Commun., 2008, 5824. 117 F. A. Akkerman, R. Kickbusch and D. Lentz, Chem. Asian J., 2008, 3, 719. 118 M. Amatore and C. Gosmini, Chem. Commun., 2008, 5019. 119 C. C. Bausch and J. S. Johnson, J. Org. Chem., 2008, 73, 1575. 120 G. Dunet, P. Mayer and P. Knochel, Org. Lett., 2008, 10, 117. 121 Z.-H. Guan, Z.-H. Ren, L.-B. Zhao and Y.-M. Liang, Org. Biomol. Chem., 2008, 6, 1040. 122 T. Kawamoto, S. Ejiri, K. Kobayashi, S. Odo, Y. Nishihara and K. Takagi, J. Org. Chem., 2008, 73, 1601. 123 G. Manolikakes, M. A. Schade, C. M. Hernandez, H. Mayr and P. Knochel, Org. Lett., 2008, 10, 2765. 124 H. Ochiai, M. Jang, K. Hirano, H. Yorimitsu and K. Oshima, Org. Lett., 2008, 10, 2681. 125 I. Rilatt and R. F. W. Jackson, J. Org. Chem., 2008, 73, 8694. 126 S. Sase, M. Jaric, A. Metzger, V. Malakhov and P. Knochel, J. Org. Chem., 2008, 73, 7380. 127 S. W. Smith and G. C. Fu, J. Am. Chem. Soc., 2008, 130, 12645. 128 G. Wang, Z. Huang and E. Negishi, Org. Letts., 2008, 10, 3223. 129 C. S. Yeung and V. M. Dong, J. Am. Chem. Soc., 2008, 130, 7826. 130 S. Wunderlich and P. Knochel, Chem. Commun., 2008, 6387. 131 S. Wunderlich and P. Knochel, Org. Lett., 2008, 10, 4705. 132 J. Ejﬂer, S. Szafert, K. Mierzwicki, L. B. Jerzykiewicz and P. Sobota, Dalton Trans., 2008, 6556. 133 C. Bonini, M. Campaniello, L. Chiummiento and V. Videtta, Tetrahedron, 2008, 64, 8766. 134 A. Coˆte and A. B. Charette, J. Am. Chem. Soc., 2008, 130, 2771.

Organomet. Chem., 2010, 36, 168–181 | 181

Organo-transition metal cluster complexes Mark G. Humphreya and Marie P. Cifuentesa DOI: 10.1039/9781847559616-00182

1.

Introduction

This chapter covers the chemistry of transition metal carbonyl and organometallic clusters containing three or more metal atoms. The treatment is in Periodic Group order, homometallic compounds being followed by heterometallic clusters. Ligands are not shown for high-nuclearity clusters, emphasis being placed on core geometry. 2.

Reviews

The preparation and chemistry of basal-edge-bridged square-pyramidal hexaruthenium complexes has been reviewed.1 3.

Theory

DFT studies of a series of Fe/Ru/Os trinuclear carbonyl clusters have permitted a structural comparison to be made; major diﬀerences are suggested for the unsaturated ruthenium and osmium derivatives M3(CO)n (n=9–11), diﬀerent isomers corresponding to diﬀering carbonyl ligation being predicted.2 The possible structures of a series of tetraosmium carbonyl clusters have been investigated using DFT. Os4(CO)n (n=14–16) are suggested to aﬀord multiple structures of similar energy with varying arrangements of bridging and semi-bridging carbonyl ligands. The unsaturated tetrahedral clusters Os4(CO)n (n=12–13) are predicted to have short metal-metal bonds consistent with multiple bonding, and with the global minimum for Os4(CO)n possessing an all-terminal CO ligand disposition.3 DFT studies on the carbido clusters [Co6(m6-C)(CO)n]2 (n=12–16) show that although the trigonal prismatic geometry is energetically favoured for n=15 and 16, the latter is thermodynamically unstable and loses CO readily. The distorted octahedral geometry predicted for n=14 is also unstable, with disproportionation giving the n=15 and octahedral n=13 clusters. For the n=12 case, a D3d structure is predicted, containing a C-centred Co6 puckered hexagon in the chair form.4 4.

Spectroscopy

Ultrafast X-ray scattering has been used to identify the presence of a Ru3(CO)10 intermediate with an all-terminal CO ligand distribution in the photolysis products of Ru3(CO)12, in addition to the two previously-identiﬁed species with bridging COs. This technique did not reveal the presence of one of the species with bridging COs, emphasizing its complementarity with ultrafast spectroscopy.5 The kinetics of the formation of [Rh6(m6-N) (CO)15] from Rh6(CO)16 and NO2 have been examined. Six kinetically a

Research School of Chemistry, Australian National University, Canberra ACT 0200, Australia

182 | Organomet. Chem., 2010, 36, 182–205 c

The Royal Society of Chemistry 2010

distinguishable steps have been identiﬁed, and another four have been identiﬁed speculatively based on mechanistic information from analogous systems.6 5. 5.1

High-nuclearity clusters Homonuclear high-nuclearity clusters

Microwave irradiation of Os3(CO)12 in diglyme for 1 h aﬀords the known decaosmium carbido cluster dianion [Os10(m6-C)(CO)24]2 in 57% yield.7 [Ni6(CO)12]2 reacts with CCl4 to give [HNi25(Z2-C2)4(CO)32]3 and [Ni22(Z2-C2)4(CO)28Cl]3 . Both trianionic complexes have metal frameworks based on the square-orthobicupola geometry and containing interstitial Ni(Z2-C2)4 and Ni2(m-Z2-C2)4 units.8 Thermolysis of Pd10(CO)12{P(pC6H4Me)3)12 in thf over 30 min aﬀords Pd37(CO)28{P(p-C6H4Me)3}12 in good yield. The complex consists of an unusual non-spheroidal Pd37-atom polyhedron possessing a molecular two-fold axis (on omission of the p-C6H4Me substituents). A similar reaction with the analogous Pd10(CO)12(PPh3)12 affords the known dodecanuclear cluster Pd12(CO)12(PPh3)6.9 [Pt15(CO)30]2 has been used in the MicroInject Moulding In Capilliaries (MIMIC) technique to form nanowires with an electrical conductivity four orders of magnitude higher than the raw material.10 AgCCBut reacts with CF3CO2Ag in the presence of TMEDA to form Ag19(CCBut)11(CF3CO2)7Cl and {Ag4(CCBut)(CF3CO2)3(TMEDA)2}n. The former consists of a square antiprism and pentacapped pentagonal prism fused through a square face and contains a templating chloride atom within the large cage. The latter is comprised of tetrasilver butterﬂy units arranged in a chain structure. The same reaction in the absence of TMEDA aﬀords {Ag16(CCBut)8(CF3CO2)8(MeOH)3}n, with a columnar structure. Attempts to prepare analogous bromide complexes were not successful.11 5.2

Heteronuclear high-nuclearity clusters

Treatment of Mo(CO)4(GaCp*)2 with four equivalents of ZnMe2 aﬀords tetradecanuclear 1, consisting of an expanded tetrahedral arrangement of metal atoms deﬁned by the four Mo(CO)4 units. The reaction involves reduction of the ZnII to ZnI and Zn0 with concomitant oxidation of GaCp* (Cp*= Z-C5Me5), suggesting a radical mechanism.12 A series of iron-gold carbonyl clusters has been isolated from the oxidation of [Fe3(CO)11]2 with [AuCl4] salts, including {Fe(CO)4}4Au5, [{Fe(CO)4}10Au21]5 , [{Fe(CO)4}12Au22]6 , [{Fe(CO)3}4{Fe(CO)4}10Au28]8 and [{Fe(CO)3}6{Fe(CO)4}8Au34]8 . Crystallographic studies show that even small changes in composition result in large structural changes, with C5 symmetry dominating.13 Reaction of [Ni9C (CO)17]2 with CdCl2 aﬀords bimetallic carbide clusters [H6-nNi30C4(CO)34 (m5-CdCl)2]n (n=3–6), crystallographic studies of [H2Ni30C4(CO)34 (CdCl)2]4 , [HNi30C4(CO)34(CdCl)2]5 and [Ni30C4(CO)34(CdCl)2]6 showing that they contain almost identical structures.14 Treatment of Au2(CCPh)2(m-4,4 0 -Ph2PC6H4C6H4PPh2) with þ [Cu(NCMe)4] aﬀords [Au3(Ph2PC6H4C6H4PPh2)3{Au6Cu6(CCPh)12}]3 þ (2) containing a central Au6Cu6 unit surrounded by a {Au(Ph2PC6H4C6 Organomet. Chem., 2010, 36, 182–205 | 183

H4PPh2)}3 ‘‘belt’’. The central Au6Cu6(CCPh)12 unit (3) can be prepared from reaction between AuCl(SC4H8) and [Cu(NCMe)4] þ . The complexes are highly luminescent.15

6.

Group 7

H3Re3(CO)12 has been shown to preferentially bind to the Ti3 þ site defects on hydroxylated titania; the clusters are deprotonated on absorption, with concomitant removal of O and O2 species.16 Treatment of [Re6(m3Se)8X(PEt3)5] þ and [Re6(m3-Se)8X2(PEt3)4] (cis- and trans-isomers) (X=I) with CO in the presence of AgSbF6 aﬀords the corresponding carbonyl clusters (X=CO). The large diﬀerence in the IR spectra and oxidation potentials of the isomers suggests an electronic diﬀerence between the cisand trans- sites, indicating signiﬁcant inﬂuence of the Se atoms in the CO backbonding interactions.17 7.

Group 8

Reaction of Fe3(CO)12 with 3-pentyn-1-ol gives the hydrido cluster 4, formed by coupling of an allenylidene unit with a coordinated carbonyl ligand and methoxy group. In contrast, reaction with the isomeric alkyne 2-methyl-3-butyn-2-ol aﬀords a binuclear iron complex.18 Treatment of Fe3(m3-S)2(CO)9 with Me3NO aﬀords the vinylferrocene complex 5, whereas the similar reaction with Fe3(m3-Te)2(CO)9 gives no cluster products.19 184 | Organomet. Chem., 2010, 36, 182–205

Addition of acetylene to a solution of isomers of [(RuCp*)3(m3-Z2:Z2:Z2C3H2Me)(m3-CH)(m-H)]2 þ gives the m3-Z2:Z2-ruthenacyclohexadienyl complex 6, which converts to the m3-Z3:Z2- cluster 7 on thermolysis. Treatment of 6 with excess CF3SO3H aﬀords an equilibrium mixture of 6 and the dicationic hydrido complex 8. Treatment of the [(RuCp*)3(m3-Z2:Z2:Z2C3H2Me)(m3-CH)(m-H)]2 þ mixture with nitriles RCN (R=Me, Et, Ph) results in insertion into the Ru–C bond and hydride migration to give 9.20

The diyne cluster 10 is formed in good yield from the low-temperature photolysis of Ru3(CO)12 with FcCCCCFc, whereas reaction at slightly higher temperatures aﬀords diruthenium products coupled with two diyne fragments.21

Organomet. Chem., 2010, 36, 182–205 | 185

The triruthenium allyl clusters 11–13 are formed from reaction of Ru3(CO)12 and the conjugated dienes 1,3-cyclooctadiene, 1,3-cyclohexadiene and cis-1-methoxybutadiene. Cluster symmetry is maintained on protonation of 11 with tetraﬂuoroboric acid to give [Ru3(m-H){m-Z3CC8H13(m3-Z2-HNNMe2)}(m-CO)2(CO)6] þ , with the hydride ligand bridging the same Ru–Ru bond as the allyl unit, whereas hydride addition occurs at the allyl ligand, producing cyclooctene and resulting in cluster decomposition. Reaction with tert-butylisocyanide occurs at elevated temperatures to give 11 (R=CNBut), and addition of diphenylacetylene results in loss of diene to give the known diphenylethenyl derivative 14, suggesting reaction occurs via a hydrido diene intermediate.22

The alkenyl cluster 14 reacts with diynes RCCCCR (R=Ph, Me) to give 15 and subsequently 16–17; only one of the possible isomers was isolated in the case of 15, whereas 16 and 17 are isomeric products resulting from coupling of the incoming diyne with the pendant alkyne moiety of 15.23

186 | Organomet. Chem., 2010, 36, 182–205

The reaction of Ru3(CO)10(NCMe)2 with HSnPh3 has aﬀorded a number of triruthenium clusters incorporating terminal and bridging triphenyltin ligands. The reaction is solvent dependent; use of hexane aﬀords products resulting from two oxidative additions to the cluster, whereas a tris-complex can be isolated from reaction in dichloromethane. Heterometallic RuPtSn complexes have also been reported, with a slight preference observed for addition of the Pt(PBut3) moiety to the Ru–Ru bond over the Ru–Sn bond.24,25 Ru3(CO)12 reacts with Ph3XSPh (X=Sn, Ge) to give open triangular cluster 18, whereas reactions of Os3(CO)10(NCMe)2 and Ru3(CO)10(dppm) with Ph3SnSPh give 19 and 20, respectively, the latter containing an intact SnPh3 group. The unusually short Ru–Sn bond in 18 is consistent with a stannylene-ruthenium donor-acceptor interaction.26

The group 8 cluster chemistry of N-heterocyclic carbenes (NHCs) has come under considerable scrutiny recently. The inﬂuence of the N–bound R group on the reactivity of ImMe2, MeOx (N-methyloxazol-2-ylidene), ImDipp2 (1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene) and ImMes2 (1,3-dimesitylimidazol-2-ylidene) towards M3(CO)12 (M=Ru, Os) has been assessed. Reaction with Ru3(CO)12 with ImMe2 occurs faster than with MeOx, aﬀording the monosubstituted product Ru3(NHC)(CO)11 in each case. In contrast, use of the bulkier ImDipp2 results in a very slow reaction in thf at room temperature and gives the hexanuclear cluster [ImDipp2H]2 [Ru6(m3-CO)2(m-CO)2(CO)14] as the only product in low yield after 4 h, a quantitative yield being obtained when a drop of water is added to the reaction mixture. DFT calculations show that the monosubstitution product is a potential energy minimum, so the product distribution is determined primarily by the reaction kinetics in this case. Treatment of Ru3(CO)12 with ImMes2 in similar conditions aﬀords a mixture of the two possible products, suggesting the ImMes2 is small enough to promote reaction but at a rate Organomet. Chem., 2010, 36, 182–205 | 187

that allows some formation of the competitive product. Treatment of the generally less reactive Os3(CO)12 with this series of NHCs aﬀords Os3(ImMe2)(CO)11 as the only product, and is the ﬁrst example of CO substitution on this cluster at room temperature.27 Thermolysis of Ru3(m-H)2 (MeImCH)(CO)9 (21, M=Ru) (ImMe2=1,3-dimethylimidazol-2-ylidene) in toluene aﬀords a 1:1 mixture of 22 and 23, showing the transformation of an N-methyl group into a carbyne and then a carbide ligand.28

Similarly, thermolysis of the triosmium NHC cluster Os3(ImMe2)(CO)11 aﬀords the edge-bridged (24, M=Os) and derivatives (21, M=Os), resulting from sequential oxidative addition of one and two C–H bonds, respectively; the ruthenium analogue 21 is prepared from a similar reaction, which aﬀords as minor products the pentanuclear clusters 25 and 26 containing m4-Z2-CO ligands.29

188 | Organomet. Chem., 2010, 36, 182–205

The abnormal carbene complex 27 (bonded through C3) is formed from the reaction between M3(CO)12 (M=Ru, Os) and the bulky NHC ImAd2 (1,3-di(adamantyl)imidazol-2-ylidene); the reaction with the ruthenium precursor occurs readily in thf at room temperature, whereas the osmium reaction requires heating at 70 1C. Thermolysis of 27 aﬀords 28.30

The reactivity of the carbene transfer agent ImMes2.AgCl towards a variety of osmium clusters (Os3(m-H)2(CO)10, Os3(CO)10(NCMe)2, Os4(mH)4(CO)12) and Ru4(m-H)4(CO)12 has been examined, and a number of new NHC complexes resulting from C–H and C–N bond activation reported (29– 34), including examples containing coordinated silver and chloride atoms.31

The strong basic character of the NHC group promotes oxidative addition in these clusters, resulting in a wide variety of products. Thermolysis of 32 (M=Os) gives a mixture of products (35–38), including the highernuclearity carbide clusters 35 and 36. The pentanuclear complex 36 is formed via C–N activation leading to loss of one of the mesitylene groups and the carbide atom originating from a coordinated CO ligand, while 37 is Organomet. Chem., 2010, 36, 182–205 | 189

an example of an abnormal carbene (bonded through C3) of osmium and probably results from the decrease in steric strain on isomerization of the NHC ligand. The unusual tetranuclear cluster 38 contains a six-membered metallacycle made up of a bridging carbene atom, formed from the dehydrogenation of the C–H bonds on one of the methyl groups of the ImMes2 ligand, coupled to a coordinated solvent benzene molecule. The analogue 38 (R=Me) is isolated as a 1:1 mixture of regioisomers from a similar reaction in toluene.31

Reaction between Ru3(CO)12 and three equivalents of the NHCs ImMes2 and ImDipp2 give quantitative yields of mononuclear Ru(CO)4(NHC), whereas sulﬁdo clusters 39 and 40 are isolated from the reaction with the NHC precursor SQImMes2. A similar reaction with the smaller ImMe4 (1,3,4,5-tetramethylimidazol-2-ylidene) aﬀords the trinuclear cluster 41, a result that most likely stems from the diﬀerence in steric bulk around the nitrogen.32

190 | Organomet. Chem., 2010, 36, 182–205

Reaction of Ru3(CO)12 with the functionalized indenes 3-RC9H6 aﬀords cluster products 42 (R=CMe2CH2NC5H4) and 43, along with diruthenium complexes (R=But, CMe2CH2Ph, CMe2CH2NC5H3Me, CMe2CH2NC9H6); the formation of the syn-Z5:Z6-bonded product seems to be favoured by the presence of bulky substituents on the indenyl ring. The reaction is solvent dependent, with ionic clusters (containing [Ru6(m6-H)(CO)18] as the anion) being formed from similar reactions using heptane instead of xylene.33

The kinetics of nitrile hydration by H4Ru4(m4-O)(m3-OH)(m2-OH)(PCy3)4 (CO)4 has been examined; the nitrile-containing tetranuclear cluster 44 was reported to have four times the catalytic activity of its precursor in the hydration of benzonitrile.34

The reactivity of the methylidyne cluster 45, formed along with 46 from Ru3(m-H)(m-COMe)(CO)10 and the diphosphine ligand 4,5-bis(diphenylphosphino)-4-cyclopenten-1,3-dione, has been reported. Mild thermolysis Organomet. Chem., 2010, 36, 182–205 | 191

in 1,2-dichloroethane aﬀords 47, which takes up hydrogen to give the trihydride cluster 48, or reacts with PR3 (R=Ph, Me) to give the monophosphine substituted 47. Further reaction of 47 (L=CO, L 0 =PMe3) with trimethylphosphine results in Ru–Ru bond opening to give 49, which readily loses CO to reform the triangular metal structure 50. The facecapping P(Ph)CQC(PPh2)C(O)CH2C(O) ligand thus functions as a 7- or 9-electron donor ligand in these complexes.35

Thermolysis of Ru3(m-dppm)(CO)9{P(C4H3E)3} (E=S, O) in the presence of Me3NO results in C–H and C–P bond activation to give 51 (L=CO) as an isomeric mixture containing m3-Z2- thiophyne and furyne ligands, respectively. Thermolysis of 51 (L=CO, E=S) at 80 1C results in ring-opening of the thiophyne ligand to give the open triruthenium complex 52, containing a m3Z2-1-thia-1,3-butadiene ligand, whereas a similar reaction with the furyne analogue aﬀords the phosphinidene cluster 53. Triphenylphosphine addition to 51 (L=CO) aﬀords 51 (L=PPh3) and reaction with HBr gives 54, with terminal and bridging bromine atoms and an unusual unsymmetric m3-alkynyl ligand with one carbon acting as a bridging alkylidene and the other as a terminally bound Fischer carbene, suggesting the presence of a C–C single bond.36

192 | Organomet. Chem., 2010, 36, 182–205

The phosphine thiol HPPhCH2CH2SH reacts with Ru3(m-dppm)(CO)10 to give the chelated product Ru3(m-H)(m-Z2-SCH2CH2PPhH)(m-dppm)(CO)7 and the linked cluster Ru3(m-H)(m-dppm)(CO)8(m-Z2-SCH2CH2PPhH)Ru3(m-dppm) (CO)9, whereas a similar reaction with Ru3(CO)12 aﬀords the tetranuclear cluster 55. Ru3(m-dppm)(CO)10 reacts with HPMe{C6H4(CH2OMe)} to give 56 and 57, the latter resulting from the presence of some adventitious phosphine oxide.37

Reaction of [Os2(m-H)5Cp*2] þ with H5OsCp* aﬀords the triangular cluster cation [{OsCp*}3(m-H)6] þ , which converts to {OsCp*}3(m3-H)3(mH)2 on treatment with n-butyllithium.38 Reaction of the unsaturated triosmium clusters 58 with ButNC results in C–H bond activation to give 59, whereas similar treatment with 60 (L=CO) leaves the cluster bonding intact giving the substituted product 60 (L=CNBut).39 The multifunctional ligand {Ph2P(2-C6H4)CH=NCH2CH2}3N reacts with three equivalents of Os3(CO)10(NCMe)2, aﬀording a ‘‘triple cluster’’ and then 61 via C–H and C–N bond activation; the ligand acts as a 7 electron donor.40 Os3(m-H) (m-OH)(CO)10 and Os3(CO)10(NCMe)2 have been reacted with a wide range _ of bifunctional ligands HE E 0 H (E, E 0 =O, S, CO2) giving Os3(m-H) _ (m-EE 0 H)(CO)10 as the major product, the functional groups showing preferential cluster binding as SHWCO2HWOH; increasing the chain length _ increases the chance of forming linked clusters {Os3(m-H)(CO)10}2(m-E E 0 ).41

Organomet. Chem., 2010, 36, 182–205 | 193

A series of thianthrene-containing triosmium clusters has been prepared (62–64). The electron deﬁcient complex 63 (L=CO) reversibly adds CO to give 64 and reacts with PPh3 to give 63 (L=PPh3).42

A facile route into Fe, Ru, Os, and Fe2O3 nanoparticles has been reported that involves thermolysis of cluster precursors in ionic liquids; the nanoparticles are in the range 1.5–2.5 nm, and require no extra stabilizers or capping molecules.43 8.

Group 9

Tricobalt carbonyl clusters have been used as end-capping groups for carbon chains containing up to m-C26 units, and including examples linked via a central Hg atom.44 A series of complexes containing Ru(dppe)Cp 0 (Cp 0 =Cp*, Cp) and Co3(m-dppm)(CO)7 linked with C7 chains have been prepared; addition of tcne (tetracyanoethene), tcnq (7,7,8,8-tetracyano-1,4quinodimethane) and Fe2(CO)9 across the central CC bond, and Ni replacement for Co in the Co3 cluster proceed as expected, suggesting the length of the carbon chain prevents any signiﬁcant interactions between the end groups.45 The octaphosphine polypodal compound 1,2,4,5C6H2{SCH2CH2N(PPh2)2}4 has been used as a central core unit coordinated to four Co3(m3-CCl)(CO)7 or four Co4(m-PPh2XPPh2)(m-CO)3(CO)5 clusters (X=NH, CH2) to form centrosymmetric metal-rich macromolecules.46 Addition of pyridine to Rh4(CO)12 under a CO atmosphere aﬀords the pentanuclear anionic rhodium salt cis-[Rh(CO)2(py)2][Rh5(CO)15] initially; 194 | Organomet. Chem., 2010, 36, 182–205

further addition of pyridine results in CO substitution to give cis[Rh(CO)2(py)2][Rh5(CO)15-x(py)x] (x=1, 2). The same reaction with bipyridine gives cis-[Rh(CO)2(bipy)][Rh5(CO)15(bipy)] as the only product. In contrast, reaction of Rh4(CO)12 with pyridine under N2 gives Rh6 (CO)16-x(py)x (x=1, 2). NMR spectroscopic studies show that the CO substitution occurs solely on the apical rhodium atom.47 The hexarhodium hydrido cluster [Rh6H12(PPri3)6]2 þ reversibly adds water to give the hydroxy hydride derivative [Rh6H11(m-OH)(PPri3)6]2 þ .48 The mono- and bis-diphosphine tetrairidium clusters 65 and 66 have been reported; 66 readily loses CO to give the orthometalated 67.49

9.

Group 10

Treatment of trans-PtCl2(PHBut2)2 with NaBH4 aﬀords a mixture of products which slowly convert to the spectroscopically-characterized 68 on standing in ethanol; the complex contains both terminal and bridging hydride ligands.50

A study of the reactivity of [Pt6(m-PBut2)4(CO)6]2 þ has shown that the two carbonyls on the edge-bridging Pt atoms are most easily substituted or attacked by nucleophiles. A range of derivatives have been reported (69-70), Organomet. Chem., 2010, 36, 182–205 | 195

including halo-substituted clusters obtained from treatment with [Bun4N]Cl or with an excess of halide salts.51

10.

Group 11

The polymeric 71 (L=Et2O) is formed from reaction of [Au(C6F5)2] with equimolar AgClO4 in CH2Cl2/Et2O; stirring in thf for 10 mins aﬀords the thf adduct (71, L=thf) which was shown to contain tetranuclear units linked via aurophilic contacts, forming a 1D polymer. Both complexes, L=Et2O, thf, together with the previously reported L=NCMe, Me2CO, show characteristic bright colours, ranging from orange through to yellow and green. A comparative study of their vapochromic behaviour shows that L=Et2O loses two molecules of coordinated solvent, whereas the other three examples lose less deﬁned volatile organic compounds and ﬂuorinated ligands; in each case, a 1:1 Ag/Au product is obtained. Reaction of solid L=Et2O complex with the vapours of the organic solvents used in the preparation of the analogues aﬀords complete substitution in the order NCMeWMe2COWthfWEt2O, and a perceptible change in colour. All complexes show luminescence at room temperature and at 77 K in the solid state, assigned to p–p* transitions in the pentaﬂuorophenyl rings.52

11. 11.1

Mixed-metal clusters Group 6

The unsaturated methyl complex Mo2(m-Z1:Z2-Me)(m-PCy2)(CO)2 has a weak agostic interaction and undergoes facile dehydrogenation in the presence of metal carbonyl reagents, giving methylidyne-bridged products 196 | Organomet. Chem., 2010, 36, 182–205

72 when reacted with Mo(CO)6 under UV-irradiation, or the mixed-metal tetranuclear 73 with Fe2(CO)9. The latter is an unsaturated 60-electron cluster containing a butterﬂy arrangement of metal atoms.53

Reaction of the dimolybdenum nitrile complex 74 with Ru3(CO)12 in reﬂuxing toluene results in CN bond cleavage to give the m4- and m5nitrido clusters 75 and 76, respectively, in low yield. The metal geometry of 76 consists of a distorted bicapped square pyramid with a Ru atom spike, the nitrogen atom lying below the basal Mo2Ru2 plane.54

Heating Cp(CO)(m-CO)2Mo=NiCp* with excess allyldiphenylphosphine aﬀords the trinuclear cluster 77; this 46-electron cluster contains a very short Mo–Mo bond distance, suggesting some multiple-bond character.55

Organomet. Chem., 2010, 36, 182–205 | 197

The hexanuclear cluster cation 78 is formed from reaction of 79 (L=Cl) with TlPF6, or directly from a solution of 79 (L=tetrahydrothiophene) at room temperature over 48 h, and consists of an unusual almost planar arrangement of metal atoms and strong p–p-stacking interactions between the cations. Interestingly, the structure is maintain in solution.56

11.2

Group 8

Ligand substitution on the 10-vertex metallocarborane [8,10-{Ir(m-PPh2) (Ph)(L)(PPh3)}-8-(m-H)-6,6,6,10,10-(CO)5-closo-6,10,1-Fe2CB7H7]n (80, L=CO, n=0) with [NEt4]CN aﬀords the anionic complex 80 (L=CN, n=1), which can be treated with cationic metal reagents {CuCl(PPh3)}4 and AuCl(PPh3) to give derivatives with four metal centres (80, L=CN {Cu(PPh3)2}, CN{Au(PPh3)}.57 Reaction of tris(N,N-diethyldithiocarbamato)cobalt with Ru3(CO)12 or Co2(CO)8 aﬀords the thiocarboxamide clusters 81 or 82, containing di- and trimetalated sulfur(diethylamino)carbene ligands, respectively.58 The phosphine thiol cluster RuCo3(m3-H)(CO)11(Ph2PCH2CH2SH) has been anchored to a Au(111) surface, and characterized using scanning tunneling microscopy. The cluster was then decarbonylated by controlled thermal annealing to give metallic products containing Ru:Co, in the expected 1:3 ratio, which were characterized by X-ray photoelectron spectroscopy. The disulﬁdelinked octanuclear complex {RuCo3(m3-H)(CO)11(Ph2PCH2CH2S)}2 was also isolated as a by-product during puriﬁcation.59

Triosmium clusters Os3(m-H)2(CO)10 and Os3(CO)10(NCMe)2 react with {IrCp*Cl2}2 to give a series of mixed-metal Os–Ir clusters 83–90.60,61 198 | Organomet. Chem., 2010, 36, 182–205

Reaction of Os3(CO)12 with Pd(PBut3)2 in reﬂuxing octane aﬀords 91, containing Pd(PBut3) groups capping all the triangular faces of a central tetrahedral Os4(CO)12 cluster. The complex is electronically unsaturated and reacts readily with H2 to give Os4(m-H)4(CO)12.62 Reaction of Organomet. Chem., 2010, 36, 182–205 | 199

Ru3(CO)12 with Pt(PBut3)2 at room temperature aﬀords 92 as the main product, whereas reaction in reﬂuxing hexane aﬀords small amounts of octanuclear cluster 93 and tetranuclear 94. The capped-pentagonal bipyramidal 93 contains 104 cluster electrons, or six fewer than predicted by the Polyhedral Skeletal Electron Pair Theory. Solutions of 93 slowly decompose to give a small amount of 93, together with Pt3(CO)3(PBut3)3. No platinum homologue of 91 has yet been isolated.63

The 68 electron unsaturated pentanuclear cluster Os3Pt2(CO)10(PBut3)2 (95) is obtained along with three other products (96–98) from the reaction of Os3(CO)10(NCMe)2 and Pt(PBut3)2. Conversion of 95 to 98 occurs readily at 68 1C with a 50% yield; the transformation involves metallation of one of the methyl groups of the Pt-coordinated But unit. Complex 95 adds hydrogen reversibly at 0 1C, giving the di- and tetra-hydrido clusters 99 and 100 with an Os–Pt bond cleavage on each successive addition of H2. Prolonged exposure of 95 to hydrogen results in degradation of the cluster via loss of one of the Pt(PBut3) units, giving 101–103.64 200 | Organomet. Chem., 2010, 36, 182–205

The reactivity of the unsaturated cluster Os3Pt(m4-CHCMeCH) (m-PBut2)(CO)7(PBut3) towards hydrogen, GeHPh3 and PhC2H has been investigated, and three new clusters (104–106) isolated. The oxidative addition of H2 occurs at the unsaturated Pt atom and is reversible under mild conditions, whereas the GeHPh3 adds to an osmium atom. Addition of phenylacetylene results in loss of the Pt atom to give 106, with the m3-Z5bridging organic ligand acting as a seven-electron donor.65

Organomet. Chem., 2010, 36, 182–205 | 201

Reaction of the selenium-capped triiron cluster [Fe3(m3-Se)(CO)9]2 with 1-3 equivalents of CuX (X=Cl, Br, I) aﬀords clusters incorporating one and two Cu units, 107–108, and the linked clusters [{Fe3(m3-Se)(CO)9}2 (Cu4X2)]2 (X=Cl, Br). The addition of the CuX to the iron core has been shown to produce a signiﬁcant anodic shift in the iron oxidation potential. The synthesis and electrochemical properties of these clusters have been examined in detail using DFT theory.66

Reaction of [Fe3(m3-Te)(CO)9]2 with equimolar [Cu(NCMe)4] þ at 0 1C aﬀords the polymeric anionic complex 109, consisting of a zig-zag chain structure, whereas reaction with two equivalents in the presence of 4,4 0 dipyridine aﬀords 110, containing linked cluster units, and reaction with a 1:3 ratio gives the neutral cluster 111, shown to be the precursor to 109 and 110. The polymeric chains possess signiﬁcant semiconducting properties.67

Abbreviations bipy Cp Cp* Cy DFT dppe dppm ImAd2 ImDipp

2,2 0 -bipyridyl Z5-cyclopentadienyl Z5-pentamethylcyclopentadienyl cyclohexyl density functional theory 1,2-bis(diphenylphosphino)ethane bis(diphenylphosphino)methane 1,3-di(adamantyl)imidazol-2-ylidene 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene

202 | Organomet. Chem., 2010, 36, 182–205

ImMe2 ImMe4 ImMes2 MeOx Me2phen MO NHC py tcne tcnq thf Th TMEDA

1,3-dimethylimidazol-2-ylidene 1,3,4,5-tetramethylimidazol-2-ylidene 1,3-dimesitylimidazol-2-ylidene N-methyloxazol-2-ylidene 2,9-dimethyl-1,10-phenanthroline molecular orbital N-heterocyclic carbene pyridyl tetracyanoethene 7,7,8,8-tetracyano-1,4-quinodimethane tetrahydrofuran 3-thiophenyl N,N,N 0 ,N 0 -tetramethylethylenediamine

References 1 J. A. Cabeza and P. Garcı´ a-A´lvarez, Organometallics, 2008, 27, 2878. 2 B. Peng, Q.-S. Li, Y. Xie, R. B. King and H. F. Schaefer III, Dalton Trans., 2008, 6977. 3 B. Xu, Q.-S. Li, Y. Xie, R. B. King and H. F. Schaefer III, Dalton Trans., 2008, 1366. 4 J. Zhao, J. Xu and R. B. King, Inorg. Chem., 2008, 47, 9314. 5 Q. Kong, J. H. Lee, A. Plech, M. Wulﬀ, H. Ihee and M. H. J. Koch, Angew. Chem. Int. Ed., 2008, 47, 5550. 6 C. Badij, D. H. Farrar, A. J. Poe¨ and S. P. Tunik, Dalton Trans., 2008, 5922. 7 K. D. Johnson and G. L. Powell, J. Organomet. Chem., 2008, 693, 1712. 8 C. Femoni, M. C. Iapalucci, G. Longoni and S. Zacchini, Chem. Commun., 2008, 3157. 9 E. G. Mednikov and L. F. Dahl, J. Am. Chem. Soc., 2008, 130, 14813. 10 P. Greco, M. Cavallini, P. Stoliar, S. D. Quiroga, S. Dutta, S. Zacchini, M. C. Iapalucci, V. Morandi, S. Milita, P. G. Merli and F. Biscarini, J. Am. Chem. Soc., 2008, 130, 1177. 11 S.-D. Bian and Q.-M. Wang, Chem. Commun., 2008, 5586. 12 T. Cadenbach, C. Gemel and R. A. Fischer, Angew. Chem. Int. Ed., 2008, 47, 9146. 13 C. Femoni, M. C. Iapalucci, G. Longoni, C. Tiozzo and S. Zacchini, Angew. Chem. Int. Ed., 2008, 47, 6666. 14 A. Bernardi, C. Femoni, M. C. Iapalucci, G. Longoni, F. Ranuzzi, S. Zacchini, P. Zanello and S. Fedi, Chem. Eur. J., 2008, 14, 1924. 15 I. O. Koshevoy, L. Koskinen, M. Haukka, S. P. Tunik, P. Y. Serdobintsev, A. S. Melnikov and T. A. Pakkanen, Angew. Chem. Int. Ed., 2008, 47, 3942. 16 K. Suriye, R. J. Lobo-Lapidus, G. J. Yeagle, P. Praserthdam, R. D. Britt and B. C. Gates, Chem. Eur. J., 2008, 14, 1402. 17 P. J. Orto, G. S. Nichol, N. Okumura, D. H. Evans, R. Arratia-Pe´rez, R. Ramirez-Tagle, R. Wang and Z. Zheng, Dalton Trans., 2008, 4247. 18 F. Bertolotti, G. Gervasio, D. Marabello, E. Sappa and A. Secco, J. Organomet. Chem., 2008, 693, 2673. 19 P. Mathur, A. K. Singh, J. R. Mohanty, S. Chatterjee and S. M. Mobin, Organometallics, 2008, 27, 5094. 20 T. Takao, M. Moriya and H. Suzuki, Organometallics, 2008, 27, 1044. 21 P. Mathur, A. Das, S. Chatterjee and S. M. Mobin, J. Organomet. Chem., 2008, 693, 1919. Organomet. Chem., 2010, 36, 182–205 | 203

22 J. A. Cabeza, I. del Rı´ o, M. Gille, M. C. Goite and E. Pe´rez-Carren˜o, Organometallics, 2008, 27, 609. 23 J. A. Cabeza, I. del Rı´ o, L. Martı´ nez-Me´ndez and E. Pe´rez-Carren˜o, J. Organomet. Chem., 2008, 693, 97. 24 R. D. Adams, B. Captain and E. Trufan, J. Organomet. Chem., 2008, 693, 3593. 25 R. D. Adams and E. Trufan, Organometallics, 2008, 27, 4108. 26 S. E. Kabir, A. K. Raha, M. R. Hassan, B. K. Nicholson, E. Rosenberg, A. Sharmin and L. Salassa, Dalton Trans., 2008, 4212. 27 J. A. Cabeza, I. del Rı´ o, D. Miguel, E. Pe´rez-Carren˜o and M. G. Sa´nchez-Vega, Organometallics, 2008, 27, 211. 28 J. A. Cabeza, I. del Rı´ o, D. Miguel and M. G. Sa´nchez-Vega, Angew. Chem. Int. Ed., 2008, 47, 1920. 29 J. A. Cabeza, I. del Rı´ o, D. Miguel, E. Pe´rez-Carren˜o and M. G. Sa´nchez-Vega, Dalton Trans., 2008, 1937. 30 M. R. Crittall, C. E. Ellul, M. F. Mahon, O. Saker and M. K. Whittlesey, Dalton Trans., 2008, 4209. 31 C. E. Cooke, M. C. Jennins, M. J. Katz, R. K. Pomeroy and J. A. C. Clyburne, Organometallics, 2008, 27, 5777. 32 M. I. Bruce, M. L. Cole, R. S. C. Fung, C. M. Forsyth, M. Hilder, P. C. Junk and K. Konstas, Dalton Trans., 2008, 4118. 33 D. Chen, S. Xu, H. Song and B. Wang, Eur. J. Inorg. Chem., 2008, 1854. 34 C. S. Yi, T. N. Zeczycki and S. V. Lindeman, Organometallics, 2008, 27, 2030. 35 S. G. Bott, H. Shen, S.-H. Huang and M. G. Richmond, J. Organomet. Chem., 2008, 693, 2327. 36 Md. N. Uddin, N. Begum, M. R. Hassan, G. Horgath, S. E. Kabir, Md. A. Miah, E. Nordlander and D. A. Tocher, Dalton Trans., 2008, 6219. 37 L. T. Byrne, N. S. Hondow, G. A. Koutsantonis, B. W. Skelton, A. A. Torabi, A. H. White and S. B. Wild, J. Organomet. Chem., 2008, 693, 1738. 38 H. Kameo and H. Suzuki, Organometallics, 2008, 27, 4248. 39 A. K. Raha, S. Ghosh, M. M. Karim, D. Tocher, N. Begum, A. Sharmin, E. Rosenberg and S. E. Kabir, J. Organomet. Chem., 2008, 693, 3613. 40 W.-Y. Yeh and M.-J. Yu, J. Organomet. Chem., 2008, 693, 2392. 41 C. Li and W. K. Leong, J. Organomet. Chem., 2008, 693, 1292. 42 A. K. Raha, M. R. Hassan, S. E. Kabir, M. M. Karim, B. K. Nicholson, A. Sharmin, L. Salassa and E. Rosenberg, J. Cluster Sci., 2008, 19, 47. 43 J. Kra¨mer, E. Redel, R. Thomann and C. Janiak, Organometallics, 2008, 27, 1976. 44 M. I. Bruce, N. N. Zaitseva, B. K. Nicholson, B. W. Skelton and A. H. White, J. Organomet. Chem., 2008, 693, 2887. 45 M. I. Bruce, M. L. Cole, C. R. Parker, B. W. Skelton and A. H. White, Organometallics, 2008, 27, 3352. 46 M. Rodriguez-Zubiri, V. Gallo, J. Rose´, R. Welter and P. Braunstein, Chem. Commun., 2008, 64. 47 K. J. Bradd, B. T. Heaton, J. A. Iggo, C. Jacob, J. T. Sampanthar and S. Zacchini, Dalton Trans., 2008, 685. 48 T. M. Douglas, S. K. Brayshaw, P. R. Raithby and A. S. Weller, Inorg. Chem., 2008, 47, 778. 49 W. H. Watson, G. Wu and M. G. Richmond, J. Organomet. Chem., 2008, 693, 1439. 50 P. Mastrorilli, Dalton Trans., 2008, 4555. 51 C. Bonaccorsi, F. F. de Biani, P. Leoni, F. Marchetti, L. Marchetti and P. Zanello, Chem. Eur. J., 2008, 14, 847.

204 | Organomet. Chem., 2010, 36, 182–205

52 E. J. Fe´rnandez, J. M. Lo´pez-de-Luzuriaga, M. Monge, M. E. Olmos, R. C. Puelles, A. Laguna, A. A. Mohamed and J. P. Fackler Jr, Inorg. Chem., 2008, 47, 8069. 53 M. A. Alvarez, D. Garcı´ a-Vivo´, M. E. Garcı´ a, M. E. Martı´ nez, A. Ramos and M. A. Ruiz, Organometallics, 2008, 27, 1973. 54 B. Li, S. Xu, H. Song and B. Wang, J. Organomet. Chem., 2008, 693, 87. 55 S. Clapham, P. Braunstein, N. M. Boag, R. Welter and M. J. Chetcuti, Organometallics, 2008, 27, 1758. 56 M. A. Alvarez, I. Amor, M. E. Garcia and M. A. Ruiz, Inorg. Chem., 2008, 47, 7963. 57 A. Franken, T. D. McGrath and F. G. A. Stone, Organometallics, 2008, 27, 148. 58 J.-M. Yang, B. Hu and C.-G. Xia, J. Cluster Sci., 2008, 19, 615. 59 A. Naitabdi, O. Toulemonde, J. P. Bucher, J. Rose´, P. Braunstein, R. Welter and M. Drillon, Chem. Eur. J., 2008, 14, 2355. 60 Y.-B. Lee and W.-T. Wong, J. Cluster Sci., 2008, 19, 133. 61 Y.-B. Lee and W.-T. Wong, J. Organomet. Chem., 2008, 693, 1528. 62 R. D. Adams, E. M. Boswell and B. Captain, Organometallics, 2008, 27, 1169. 63 R. D. Adams, E. M. Boswell, B. Captain and L. Zhu, J. Cluster Sci., 2008, 19, 121. 64 R. D. Adams, B. Captain and L. Zhu, J. Organomet. Chem., 2008, 693, 819. 65 R. D. Adams, E. M. Boswell, M. B. Hall and X. Yang, Organometallics, 2008, 27, 4938. 66 M. Shieh, C.-Y. Miu, C.-J. Lee, W.-C. Chen, Y.-Y. Chu and H.-L. Chen, Inorg. Chem., 2008, 47, 11018. 67 M. Shieh, C.-H. Ho, W.-S. Sheu, B.-G. Chen, Y.-Y. Chu, C.-Y. Miu, H.-L. Liu and C.-C. Shen, J. Am. Chem. Soc., 2008, 130, 14114.

Organomet. Chem., 2010, 36, 182–205 | 205

Organometallic Chemistry, Volume 36 (Specialist Periodical Reports)

Organometallic Chemistry, Volume 31 (Specialist Periodical Reports)

Heterocyclic Chemistry: v.4 (Specialist Periodical Reports)

Organosphophorus Chemistry (Specialist Periodical Reports) (v. 4)

Photochemistry, Volume 37 (Specialist Periodical Reports)

Catalysis, Volume 22 (RSC Specialist Periodical Reports)

Organophosphorus Chemistry (Volume 18) Specialist Periodical Report

Heterocyclic Chemistry: v.3 (Specialist Periodical Reports) (Vol 3)

Catalysis: v. 18 (Specialist Periodical Reports)

Catalysis: Vol. 19 (Specialist Periodical Reports)

Theoretical Chemistry,vol 2 (Specialist Periodical Reports) (v. 2)

Alkaloids Volume 4: Specialist Periodical Rep (Specialist Periodical Reports) (v. 4)

Catalysis: Vol 17 (Specialist Periodical Reports)

Photochemistry (Specialist Periodical Reports) (Vol 13)

Annual Reports in Medicinal Chemistry, Volume 36

Nuclear Magnetic Resonance: Volume 37 (Specialist Periodical Reports)

Terpenoids and Steroids: Volume 12 (Specialist Periodical Reports)

Terpenoids and Steroids: Volume 3 (Specialist Periodical Reports)

Dielectric and Related Molecular Processes: Volume 1 (Specialist Periodical Reports)

Organometallic chemistry, Volume 27

Organometallic Chemistry, Volume 29

Organometallic Chemistry, Volume 32

Inorganic Chemistry of the Transition Elements, Volume 2 (Specialist Periodical Reports) (v. 2)

Chemical Modelling: Vol. 2: Applications and Theory (Specialist Periodical Reports)

Medicinal Organometallic Chemistry (Topics in Organometallic Chemistry, Volume 32)

Annual Reports on NMR Spectroscopy, Volume 36

Inorganic Chemistry of the Transition Elements: v. 4: A Review of Chemical Literature (Specialist Periodical Reports)