intermetallic chemistry : molecular compounds at the

Intermetallic Chemistry: Molecular

Compounds at the Borderline to Cluster Science

Dissertation

M. Sc. Timo Bollermann

2011

Intermetallic Chemistry: Molecular

Compounds at the Borderline to Cluster Science

Dissertation

to obtain the doctorate Dr. rer. nat.

of the Faculty of Chemistry and Biochemistry

at the Ruhr University Bochum

Submitted by

M. Sc. Timo Bollermann

2011

This dissertation is based on the experimental work carried out during the period from June

2008 to June 2011 under the supervision of Prof. Dr. Roland A. Fischer at the Chair of

Inorganic Chemistry II, Organometallics & Materials, Ruhr University Bochum, Germany.

1st

2

Referee: Prof. Dr. Roland A. Fischer

nd

Oral examination: 05.09.2011

Referee: Prof. Dr. William S. Sheldrick

Herewith I declare that I have written this thesis myself and without any other help or sources

which are not specifically and explicitly marked or named in this dissertation.

Furthermore, I declare that I have not applied any other review procedure or submitted this

thesis in the same, similar or different form to another faculty or university as a dissertation.

Timo Bollermann

First of all, I would like to express my special thanks to my supervisor Prof. Dr. Roland A.

Fischer. I am really grateful to have been part of your chair and the ‘Organometallics’

subgroup. Looking back over the three years I was a PhD student in your group, you have

trusted me and my work all the time in a fantastic way. Thank you for supporting my

fellowships of the Fonds der Chemischen Industrie, the Ruhr University Research School and

the Ruth und Gert Massenberg Foundation. Thank you for all your helpfulness in all scientific

matters. In summary, I just want to say THANK YOU! It was a great pleasure for me to be a

part of your group.

I would like to express my special thanks to…

…Dr. Christian Gemel for his unique support and helpfulness, for all the discussions and

brainstorming sessions that contributed to this work. Thank you for asking me nearly every

day: ‘What about news and/or crystals?’ You can be sure that I will miss this question when I

start my research somewhere else. Thank you very much, Christian!

…Dr. Thomas Cadenbach for sparking my interest in organometallic chemistry, your

helpfulness during my Bachelor and Master thesis and during the first steps of my PhD

project. Thanks for fruitful discussions about research and football.

…Kerstin Freitag for several discussions about chemistry and for sharing several frustrating

hours with ‘Carmona´s reagent’. In the course of time, Kerstin has become a great friend of

mine, thus, I want to say THANKS for your friendship, for several nice evenings in the

‘Cotton Club’, for support when I was stressed out and for your helpfulness, especially during

the time of my hospital stay.

...Markus Halbherr for an unforgettable trip to Kiel, your friendship and several hours in your

‘gambling house’. Although our dart games became fewer during writing of my thesis, I

promise you I’ll be back soon.

…Mariusz Molon for his friendship and helpfulness all the time, especially during my time in

hospital. This includes spending lots of time for the calculations of superimpositions.

…Dr. Ganesan Prabusankar for his support in all areas of research and for working with me

on the copper triflate project.

…Dr. Rüdiger W. Seidel for his unique support in crystallographic questions and for spending

lots of time for the solution of single crystal X-ray structures.

…Prof. Dr. Gernot Frenking, Moritz von Hopffgarten and Paul Jerabek for the

‘unprecedented’ collaboration and for performing all the quantum chemical calculations.

…Prof. Dr. Ulrich Zenneck for welcoming me in his group in Erlangen and introducing me to

the world of metal vapour synthesis.

…Prof. Dr. William S. Sheldrick for being the 2nd referee of this thesis.

…Hans Jochen Hauswald, Martin Gartmann and Gregor Barchan for their support in NMR

spectroscopy.

…Manuela Winter and Johannes Schröder for a nice trip to Poland and for their support in X-

ray analysis and measurements.

…Sabine Bendix and the Linden CMS company for their support in mass spectrometry.

…Sabine Pankau for all her support during the last years. You are one of the most pleasant

persons I have ever met in my life and I just want to say THANK YOU for spending so much

time for me.

…Jacinta Essling for her fantastic support in all questions around organizing my workaday

life at university. I am really thankful to you for reading my thesis!

…Denise Zacher, Maike Müller and Malte Hellwig for their friendship and all the funny talks

during the last years. Thanks for all the support and helpfulness!

…Sebastian Henke and Andrian Milanov for an unforgettable trip to Kiel.

…Uschi Herrmann for her unique support during all the years and lots of funny talks and

discussions.

…all the members of ACII and especially the ‘Organometallics’ subgroup!

…the Fonds der Chemischen Industrie for a PhD fellowship.

…the Ruhr University Research School for financial support.

…the Ruth and Gert Massenberg foundation for a travel stipend to attend the 1st

EICC,

Manchester, UK.

I am really thankful to my best friends Katja Preukschat and Uwe Scheel. I met Katja and

Uwe by chance a few years ago and I have never met such lovingly people before. Thanks for

all your support and for your unique friendship.

I would like to thank my parents, Karlheinz and Patrizia Bollermann, my brother Oliver and

my sister Melanie for all their support, love and helpfulness during the last years.

In the end I would like to express my special thanks to Johanna Niesel. I met Johanna about

eight years ago in one of our first lectures. Since then we both coped with our studies in

Bochum and Johanna has become one of the most important people in my life. I am really

thankful for all your support during my PhD studies, for all the funny trips around the Ruhr

area and for your unique friendship. I will never forget what you have done for me during the

last eight years and, most importantly, I will never forget what you mean to me!

„Wir stellen Fragen, ohne uns durch die Antworten irritieren zu lassen.“

Werner Hansch

For Jojo

I I. Table of Contents

I. Table of Contents

1 Motivation .......................................................................................................................... 1

2 Introduction ....................................................................................................................... 5

2.1 Basic Approach to Intermetallic Chemistry: Intermetallics, Zintl-phases and

Clusters ................................................................................................................................... 6

2.1.1 Influences and Parameters Determining the Formation of Intermetallic

Phases ..... ............................................................................................................................. 6

2.1.2 Zintl Compounds: Transition between Intermetallics, Salt-like

Compounds and Clusters .................................................................................................. 13

2.1.3 Cluster Compounds: Transition between Intermetallics and Molecular

Compounds ....................................................................................................................... 16

2.2 Organometallic Chemistry of Low Valent Group 13 Organyls ................................ 19

2.2.1 Theoretical Investigations on the Bonding Characteristics of Low

Valent Group 13 Ligands .................................................................................................. 19

2.2.2 Coordination Chemistry of E(I)Cp* Ligands Towards Reactive

Transition Metal Compounds ........................................................................................... 21

2.2.3 Applications of Low Valent Group 13 Organyls in Material Science ............... 25

2.3 The Renaissance of Zinc Chemistry: Landmark [Zn2Cp*2] ...................................... 26

2.3.1 Theoretical Investigations into Reaction Pathways and Bonding

Situation of [Zn2Cp*2] ...................................................................................................... 27

2.3.2 First Experimental Investigations on the Reactivity of Dizincocene ................. 30

2.3.3 Curiosities in Silico: Fullerene-Dizincocene Hybrids and

Multimetallocenes ............................................................................................................. 32

2.4 Organozinc Ligands in Transition Metal Chemistry: A Brief Overview .................. 33

II I. Table of Contents

3 Results and Discussion .................................................................................................... 40

3.1 Synthesis and Characterisation of Homoleptic and Heteroleptic

Molybdenum and Rhodium GaR (R = Cp*, DDP) Containing Complexes ......................... 40

3.2 First Dinuclear Copper/Gallium Complexes: Supporting Cu(0) and Cu(I)

Centres by Low Valent Organogallium Ligands .................................................................. 52

3.3 Experimental and Theoretical Investigations on the Formation of Zinc-rich

Oligonuclear Cluster Compounds ........................................................................................ 62

3.3.1 Zinc-rich Compounds of Iron and Cobalt: Formation of [Fe2Znx] (x =

2-4) and [Co2Zn3] Cores ................................................................................................... 64

3.3.2 Case Study on the Formation of an Oligonuclear Model System for

Intermetallic Phases: Synthesis, Characterisation and Theoretical Investigations

on the Compound [Pd2Zn6Ga2(Cp*)5(CH3)3] ................................................................... 72

3.4 Experimental and Theoretical Investigations on the Coordination Chemistry

of [Zn2Cp*2] Towards Transition Metal Compounds .......................................................... 82

3.4.1 Trapping Monovalent {ZnZnCp*} at d10 Transition Metal Centres .................. 84

3.4.2 First Reactivity Studies of [Zn2Cp*2] Towards Olefin Containing d10

Transition Metal Centres ................................................................................................... 94

3.4.3 Experimental and Theoretical Investigations on the Formation of a

Novel [PdZn7] Compound: [Zn2Cp*2] as a Source for Stabilized Zn(0) ........................ 105

4 Summary ........................................................................................................................ 117


Molybdenum and Rhodium GaR (R = Cp*, DDP) Containing Complexes ....................... 118

4.2 First Dinuclear Copper/Gallium Complexes: Supporting Cu(0) and Cu(I)

Centres by Low Valent Organogallium Ligands ................................................................ 119

4.3 Experimental and Theoretical Investigations on the Formation of Zinc-rich

Oligonuclear Cluster Compounds ...................................................................................... 122

III I. Table of Contents

4.3.1 Zinc-rich Compounds of Iron and Cobalt: Formation of [Fe2Znx] (x =

2-4) and [Co2Zn3] Cores ................................................................................................. 122

4.3.2 Case Study on the Formation of an Oligonuclear Model System for

Intermetallic Phases: Synthesis, Characterisation and Theoretical Investigations

on the Compound [Pd2Zn6Ga2(Cp*)5(CH3)3] ................................................................. 123

4.4 Experimental and Theoretical Investigations on the Coordination Chemistry

of [Zn2Cp*2] Towards Transition Metal Compounds ........................................................ 124

4.4.1 Trapping Monovalent {ZnZnCp*} at d10 Transition Metal Centres ................ 125

4.4.2 First Reactivity Studies of [Zn2Cp*2] Towards Olefin Containing d10

Transition Metal Centres ................................................................................................. 126

4.4.3 Experimental and Theoretical Investigations on the Formation of a

Novel [PdZn7] Compound: [Zn2Cp*2] as a Source for Stabilized Zn(0) ........................ 128

5 Outlook ........................................................................................................................... 130

6 Experimental Section .................................................................................................... 132

6.1 Materials and Methods ............................................................................................ 132

6.1.1 General Remarks .............................................................................................. 132

6.1.2 Nuclear Magnetic Resonance Spectroscopy (NMR) ....................................... 132

6.1.3 Single Crystal X-Ray Diffraction (XRD) ......................................................... 133

6.1.4 Infrared Spectroscopy (IR) ............................................................................... 133

6.1.5 Mass Spectrometry (MS) ................................................................................. 133

6.1.6 Elemental Analysis (EA) and Atom Absorption Spectroscopy (AAS) ............ 134

6.2 Precursors ................................................................................................................ 134

6.2.1 General Remarks .............................................................................................. 134

6.2.2 Synthesis and Characterisation of Novel Compounds 1-23 ............................. 135

6.3 Computational Details ............................................................................................. 144

IV I. Table of Contents

6.3.1 Computational Details for [Pd2Zn6Ga2(Cp*)5(CH3)3] (14) .............................. 144

6.3.2 Computational Details for [M(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (M = Pd

(15), Pt (17)) .................................................................................................................... 149

6.3.3 Computational Details for [M(ZnCp*)4(ZnZnCp*)4] (M = Pd (16), Pt

(18), Ni (19)) ................................................................................................................... 167

6.3.4 Computational Details for [Pd(ZnCp*)2(ZnMe)2{Zn(tmeda)}] (23) ............... 168

7 References ...................................................................................................................... 174

8 Supplement .................................................................................................................... 189

8.1 Important Crystallographic Data ............................................................................. 189

8.2 Overview of the Novel Compounds ........................................................................ 196

8.3 Publications ............................................................................................................. 204

8.4 Conferences: Oral Presentations and Poster Presentations ...................................... 207

8.5 Curriculum Vitae ..................................................................................................... 209

V II. List of Figures

II. List of Figures

Figure 1. Components of intermetallic chemistry ..................................................................... 5

Figure 2. Influence of atomic size on the solubility in magnesium .......................................... 7

Figure 3. Darken and Gurry map for known solubilities of elements in silver ......................... 9

Figure 4. Equilibrium diagram of the system Cu/Zn ............................................................... 11

Figure 5. Crystal structures of MgCu2, MgZn2 and MgNi2 ..................................................... 12

Figure 6. Left: Overall shape of [Sn9M(CO)3]4- and [Pb9M(CO)3]4- (M = Cr, Mo, W,

shown as a yellow sphere). Right: Structure of [E9Zn-Ph]3-

(E = Si, Ge, Sn, Pb; Zn

shown as a yellow sphere) ........................................................................................................ 14

Figure 7. Stepwise process of cluster manipulations starting with an empty [Ge9]3-

cluster (a) and ending with the Ni-centered Ni(C≡CPh)-substituted species

[Ni@(Ge9Ni-CCPh)]3- (d) ........................................................................................................ 15

Figure 8. Metalloid aluminium and gallium cluster: cut-outs of intermetallic phases ............ 18

Figure 9. Isolobality of E(I)R species and CO ........................................................................ 20

Figure 10. Interactions in M-E(I)R complexes ....................................................................... 20

Figure 11. Reaction pathway of ZnEt2 reacting with decamethylzincocene ........................... 28

Figure 12. Energy correlation diagram for the formation of [Zn2Cp2] from the

fragment {ZnCp} ..................................................................................................................... 29

Figure 13. Molecular structures of [M(ZnR)n] showing the coordination polyhedra

around the central metal M ....................................................................................................... 37

Figure 14. Molecular structure of [{Mo(CO)4}4(Zn)6(µ-ZnCp*)4] ......................................... 39

Figure 15. Molecular structure of 1 in the solid state as determined by single crystal

X-ray diffraction ....................................................................................................................... 43

VI II. List of Figures





Figure 18. Superimposition of the [RhGa5] metal core (grey) and ideal polyhedra

(black). Left: trigonal bi-pyramid, Middle: square pyramid, Right: square pyramid (β

= 105°) ...................................................................................................................................... 48











Figure 24. Molecular structures of 10 (left) and 11 (right) in the solid state as

determined by single crystal X-ray diffraction ........................................................................ 68



Figure 26. Cp* area of the 1 H NMR spectrum (C6D6, rt) of 14 .............................................. 74

Figure 27. Above: Experimental LIFDI-MS spectrum of 14. Below: Calculated

LIFDI-MS spectrum of 14 ........................................................................................................ 75

Figure 28. Left: Modelled structure of [Pd2Zn6Ga2(Cp)5(CH3)3]. Right: Calculated

structure of 14M-3/8 ................................................................................................................ 76

VII II. List of Figures

Figure 29. Above: Molecular structure of 14 in the solid state. Below: Illustration of

the distorted bi-capped trigonal prism ...................................................................................... 77

Figure 30. Left: Molecular graph and contour map of the Laplacian ∇2ρ(r) of 14M-

3/8-H in the molecular plane bisecting the palladium atoms and the bridging ligand

atoms Zn6 and Zn7. Right: Molecular graph and contour map of the Laplacian ∇2

ρ(r)

of 14M-3/8-H in the molecular plane bisecting the palladium atoms and the ligand

atoms Ga3 and Zn4 .................................................................................................................. 79

Figure 31. Orbital shapes of HOMO-18 (left) and HOMO-19 (right) of 14M-3/8-H

(BP86/def2-TZVPP). ................................................................................................................ 80

Figure 32. NMR reaction of ZnCp*2 with GaCp* (1:1 ratio; C6D6, rt). Above: 13C

NMR spectrum. Below: 1 H NMR spectrum. ............................................................................ 86

Figure 33. Assignment of numbers of the Ga/Zn positions in the model system 15H ........... 88



Figure 35. Left: Molecular structure of 16 in the solid state as determined by single

crystal X-ray diffraction. Right: Schematic illustration of the inner [PdZn8] core .................. 93

Figure 36. Experimental LIFDI-MS spectrum of the reaction of [Pt(cod)2] with

[Zn2Cp*2] (ratio 1:8); [M].+ : m/z = 932 [Cp*Pt(ZnCp*)3] (21), m/z = 2061 (18) ................... 97



Figure 38. Molecular orbitals presenting the σ-character of the Zn-Zn interactions. ............. 99

Figure 39. Molecular graph and contour map of the Laplacian ∇2 ρ(r) of 16M .................... 100


X-ray diffraction ..................................................................................................................... 103



VIII II. List of Figures



Figure 43. Left: Superimposition of the [PdZn7] metal core (Pd: blue, Zn: green) and

a trigonal dodecahedron (black). Right: Superimposition of the [PdZn7] metal core

(Pd: blue, Zn: green) and a pentagonal bi-pyramide (black) .................................................. 111

Figure 44. Molecular graph and contour map of the Laplacian ∇2

ρ(r) of 23M in the

molecular plane which contains Pd and the ligand atoms Zn(Cp,ax) and Zn(TMEDA) ....... 113

Figure 45. Synthesis of [Mo(GaCp*)6] (1) ............................................................................ 118

Figure 46. Synthesis of [(Cp*Ga)Cu(µ-GaCp*)3Cu{Ga(CF3SO3)3}] (7) ............................. 120

Figure 47. Synthesis of [(CO)3Co{µ2-Zn(py)2}(µ2-ZnCp*)2Co(CO)3] (12) ......................... 123

Figure 48. Synthesis of [Pd2Zn6Ga2(Cp*)5(CH3)3] (14). ....................................................... 124

Figure 49. Molecular structures of [Pd(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (15) and

[Pd(ZnCp*)4(ZnZnCp*)4] (16) in the solid state. ................................................................... 125

Figure 50. Synthesis of [Ni(ZnCp*)4(ZnZnCp*)4] (19) and [Cp*Ni(ZnCp*)3] (20). ........... 127

Figure 51. Synthesis of [Cp*Pd(ZnCp*)3] (22) and

[Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}] (23). ............................................................................... 128

IX III. List of Schemes

III. List of Schemes

Scheme 1. Reaction schemes of transition metal carbonyl compounds with GaCp* .............. 21

Scheme 2. Synthesis of [M(GaCp*)4] (M = Ni, Pd, Pt) ........................................................... 22

Scheme 3. Synthesis of cationic species [Cp*M(GaCp*)3]m+ (M = Fe, Co) ........................... 23

Scheme 4. Building block synthesis of dimeric d10 metal compounds .................................... 24

Scheme 5. Synthetic routes to oligonuclear transition metal ECp* (E = Al, Ga, In)

compounds ............................................................................................................................... 24

Scheme 6. Use of AlCp* in the formation of intermetallics .................................................... 25

Scheme 7. Dissociation of [Zn(η5-Cp*)(η1-Cp*)] into [Zn(η5-Cp*)]• and [Cp*]•

radicals ..................................................................................................................................... 28

Scheme 8. Derivative structures containing the [Zn2]2+ core .................................................. 30

Scheme 9. Above: Base-stabilised [Zn2]2+ cation. Below: Synthesis of

[Zn2Cp*2(OArMes )(pyr-py)2] and [Zn2Cp*2(µ-OC5Me5)(pyr-py)]2 .......................................... 31

Scheme 10. Left: [Cp2Nb(H)2(ZnCp)]. Right: [Cp’2Ta(H)(ZnCp)2]. ....................................... 34

Scheme 11. Reaction mechanism for the formation of [Ni2Zn4Cp6] ....................................... 35

Scheme 12. Formation of [{Cp*Ru}3(µ-H)3(µ3-ZnR)(µ3-H)] ................................................. 36

Scheme 13. Formation of [{Cp*Rh(ZnCp*)2(ZnMe)(ZnCl)}2] and

[Cp*2Rh][(Cp*Rh)6Zn18Cl12(µ6-Cl)] ........................................................................................ 39

Scheme 14. Synthesis of [Mo(GaCp*)6] (1) and [cis-Mo(GaCp*)2(PMe3)4] (2) .................... 42

Scheme 15. Synthesis of [Rh(GaCp*)5][X] (X = CF3SO3 (3), BArF

(4)) and

[(coe)(toluene)Rh{Ga(DDP)}(CF3SO3)] (5) ............................................................................ 46

Scheme 16. Synthesis of [{(DDP)GaCu(CF3SO3)}2] (6) ....................................................... 53

Scheme 17. Synthesis of [(Cp*Ga)Cu(µ-GaCp*)3Cu{Ga(CF3SO3)3}] (7) ............................ 56

X III. List of Schemes

Scheme 18. Synthesis of [Cu2(GaCp*)3(µ-GaCp*)2][CF3SO3]2 (8) ........................................ 58

Scheme 19. Synthesis of [(CO)4Fe{µ2-Zn(thf)2}2Fe(CO)4] (9), [(CO)3Fe{µ2-

Zn(thf)2}2(µ2-ZnMe)2Fe(CO)3] (10) and [(CO)3Fe{µ2-Zn(py)2}3Fe(CO)3] (11) ..................... 65

Scheme 20. Synthesis of [(CO)3Co{µ2-Zn(L)2}(µ2-ZnCp*)2Co(CO)3] (L = py (12),

thf (13)) .................................................................................................................................... 70

Scheme 21. Synthesis of [Pd2Zn6Ga2(Cp*)5(CH3)3] (14) ........................................................ 73

Scheme 22. Synthesis of [M(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (M = Pd (15), Pt (17))

and [M(ZnCp*)4(ZnZnCp*)4] (M = Pd (16), Pt (18)) .............................................................. 84

Scheme 23. Possible reaction mechanism in the formation of 15-18 ...................................... 87

Scheme 24. Possible reaction mechanism in the formation of 19 and 20 ............................... 96

Scheme 25. Synthesis of [Cp*M(ZnCp*)3] (M = Ni (20), Pt (21)) ....................................... 101

Scheme 26. Synthesis of [Cp*Pd(ZnCp*)3] (22) and

[Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}] (23) ................................................................................ 106

XI IV. List of Tables

IV. List of Tables

Table 1. Example systems for the relative valency effect ......................................................... 8

Table 2. Valence electron counts for Hume-Rothery rules ..................................................... 10

Table 3. Structures and VEC for Cu1-xZnx .............................................................................. 10

Table 4. Examples for Hume-Rothery phases and corresponding VEC ................................. 11

Table 5. Comparison between ideal and experimental angles (°) for

[Rh(GaCp*)5][CF3SO3] (3) ...................................................................................................... 48

Table 6. Comparison of Ga-O bond distances (Å) between [Rh(GaCp*)5][CF3SO3]

(3) and selected reference compounds ..................................................................................... 49

Table 7. NMR spectroscopic data of 6 and 7 (d8-thf, rt) ......................................................... 59

Table 8. 1

H NMR spectroscopic data of compounds 9, 10 (d8-thf, rt) and 11 (d5-

pyridine, rt) ............................................................................................................................... 66

Table 9. Selected distances (Å) and comparison of Fe3Zn10 with 9-11 ................................... 69

Table 10. Selected distances (Å) and comparison of CoZn13 with 12-13 ............................... 71

Table 11. NBO partial charges q of the isomer 14M-3/8 at BP86/def2-TZVPP .................... 80

Table 12. NMR spectroscopic data (C6D6, rt) of 15 and 17 .................................................... 85

Table 13. Relative energies of eight isomers (ΔE ≤ 15kcal/mol) of the model systems

15H (left) and 17H (right) at BP86/def2-TZVPP .................................................................... 88

Table 14. Relative energies of the seven lowest energy isomers of compounds 15

(above) and 17 (below) at different levels of theory ................................................................ 89

Table 15. Important interatomic distances (Å) for 15 and 17 ................................................. 92

Table 16. Important interatomic distances (Å) for 19 ............................................................. 98

XII IV. List of Tables

Table 17. Geometry optimization at BP86-D/SVP (X-Ray structure) and

BP86/TZVPP (16M) ................................................................................................................ 99

Table 18. EDA-NOCV results of 16M (BP86/TZ2P+). Energies in kcal/mol ...................... 101

Table 19. NMR spectroscopic data (C6D6, rt) of 20 and 21 .................................................. 102

Table 20. Important interatomic distances (Å) for 20, 21 and selected reference

compounds ............................................................................................................................. 104

Table 21. 1

H NMR spectroscopic data for the reaction of [PdMe2(tmeda)] with

[Zn2Cp*2] (C6D6) .................................................................................................................... 107

Table 22. Experimental and calculated distances in Å .......................................................... 112

Table 23. Calculated NBO partial charges q for atoms and ligands in 23M ......................... 113

Table 24. EDA results for the Pd-ZnR interactions in 23M .................................................. 114

XIII V. List of Acronyms and Abbreviations

V. List of Acronyms and Abbreviations

[18-crown]-6 1,4,7,10,13,16-Hexaoxacyclooctadecane

AAS Atom Absorption Spectroscopy

Ar* 2,6-(2,4,6-Triisopropylphenyl)-phenyl

AIM Atoms-In-Molecules

av. Average

ArMesOH 2,6-(2,4,6-Me3C6H2)-C6H3OH

BArF B{C6H3(CF3)2}4

tBu tert-Butyl

bipy Bipyridine

Cp* Pentamethylcyclopentadienyl; C5Me5

Cp Cyclopentadienyl; C5H5

Cp’ Metyhl-cyclopentadienyl; C5H4Me

coe Cyclooctene

cht Cycloheptatriene

cod 1,5-Cyclooctadiene

Cy Cyclohexyl

CVMO Cluster Valence Molecular Orbitals

[2.2.2]crypt 4,7,13,16,21,24-Hexaoxa-1,10-diazabicyclo[8.8.8]hexacosane

DDP HC(CMeNC6H3-2,6-i

dppe 1,2-Bis(diphenylphosphino)ethane

Pr2)2

DFT Density Functional Theory

XIV V. List of Acronyms and Abbreviations

dmap 4-(Dimethylamino)-pyridin

dippp 1,3-Bis(diisopropylphosphino)propan

dippe 1,2-Bis(diisopropylphosphino)ethan

EDA Energy Decomposition Analysis

EDA-NOCV Energy Decomposition Analysis (with its) Natural Orbitals for

Chemical Valence variation

EA Elemental Analysis

Et Ethyl

en 1,2-Diaminoethane

e Electron

HOMO Highest Occupied Molecular Orbital

IR Infra-Red (spectroscopy)

LIFDI Liquid Injection Field Desorption Ionization

LUMO Lowest Unoccupied Molecular Orbital

Me Methyl

MS Mass Spectrometry

Mes* 2,6-Bis(2,5,6-triisopropylphenyl)phenyl

MO Molecular Orbital

M Transition metal

NMR Nuclear Magnetic Resonance (spectroscopy)

nbd 2,5-Norbornadiene

Ph Phenyl

py Pyridine

XV V. List of Acronyms and Abbreviations

pyr-py 4-Pyrrolidino-pyridine

F

PNP Bis(4-fluoro-2-(diisopropylphosphino)phenyl)amine

rt Room temperature

tmeda N,N,N',N'-Tetramethyl-ethane-1,2-diamine

thf Tetrahydrofuran

VEC Valence Electron Concentration

VE Valence Electron(s)

VSEPR Valence Shell Electron Pair Repulsion

1 1. Motivation

1 Motivation

The definition of a classical coordination compound as described by Alfred Werner in 1892,

is associated with one of the most important but comparatively less known controversy with

Sophus Mads Jörgensen which became the subject of several publications of historians, most

notably of Georg B. Kauffman.[1-3] The work of both chemists and especially ‘Werner's view

of the two types of linkage, ionizable and non-ionizable, did much to clarify ideas of chemical

bonding a generation before the views of Kossel and Lewis (1916) led to our present concepts

of ionic and covalent bonding’.[1] In principle, Jörgensen represented the opinion of chain-like

structures in today´s known ammonia and halide containing coordination compounds of

transition metals, mainly based on the suggestions of Blomstrand. In contrast, Werner

postulated the existence of a metal centre surrounded by ligand atoms giving rise to

polyhedral coordination geometries as well as associated configurations. Thus, coordination

compounds are such molecules bearing a metal atom or rather metal ion Mz+/0 surrounded by

n ligands L forming complexes of the type [MLn] with n = 1-8.[2, 4-6] Indeed, nearly all

imaginable metal-ligand combinations across the periodic table of elements leading to Werner

type complexes have been prepared in the last three centuries. Herein, even the formation of

multinuclear transition metal complexes has been investigated, but without the existence of

direct metal-metal interactions. Within the definition of a coordination compound, low valent

group 13 organyls E(I)R (E = Al, Ga, In; R = bulky, mono anionic substituents; i.e.

cyclopentadienyl derivatives, alkyl, aryl, dialkylamide, β-diketiminate, guanidinate or

amidinate) can be viewed as a special, often called ‘exotic’, form of ligand, in which the

group 13 metal E is coordinated to the central transition metal atom. Notably, the partial

charges in such complexes are inversed to classical Werner type complexes in which the

metal center is positively charged and the negative charge is placed at the ligator atom.[7-9]

The coordination modes and thus, the structural and electronic features involving enforced

reaction schemes depends on the group 13 metal, as well as to a greater part, on the stabilizing

organic group R. For instance, while Cp* (Cp* = pentamethylcyclopentadienyl) shows

flexible, more or less, soft properties in the stabilization of the low valent group 13 element,

resulting in possible haptotropic shifts (η1→η3→η5), the β-diketiminate DDP (DDP =

HC(CMeNC6H3-2,6-iPr2)2) shows coordinatively rigid, and electronically hard features. Their

reactivity and coordination behavior towards transition metals has been experimentally and

theoretically investigated in detail over the last decades.[10-18] From today´s viewpoint, E(I)R

2 1. Motivation

species can be systematically used in the formation of oligonuclear and polynuclear metalloid

cluster compounds or rather coordination complexes. The term ‘cluster’ has been described in

the 1960s by F. A. Cotton as ‘those containing a finite group of metal atoms held together

entirely, mainly, or at least to a significant extent, by bonds directly between the metal atoms

even though some non-metal atoms may be associated intimately with the cluster’.[19] Later

on, Schnöckel tempted himself to modify Cotton´s suggestion by reason that ‘the definition

chosen by Cotton for a metal cluster is so general…we would like to distinguish them from

such compounds. Metalloid (metal-like) clusters should be epitomized by the property that the

number of metal-metal contacts exceeds the number of metal-ligand contacts and by the

presence of metal atoms which participate exclusively in metal-metal interactions’.[20, 21] The

determination between metal atom clusters and related metal atom containing inorganic

molecules is quite difficult. G. Schmid suggests in his book ‘Clusters and Colloids-From

Theory to Applications’ that the ‘characterizing nature and the extent of the metal-metal

interactions in metal clusters is one of the most challenging problems for theoreticians’.[22]

The most common classification of metalloid cluster compounds found in literature is the

determination in neutral or ionic (1) ligated metal atom clusters such as [Au55(PPh3)12Cl6][23]

which consist of metal cores wrapped into a stabilizing ligand shell (molecular or

organometallic clusters) and (2) naked metal atom clusters, for example [MAu12][24, 25] (M =

Mo, W), in which stabilizing ligands are missing (gas phase clusters). Due to the difficulties

in synthesis and characterisation of naked metal atom clusters, the main focus lies on the

synthesis of ligand-protected metal cores as molecular example systems for the bulk metal

material. First metalloid clusters of the main group elements have been succeeded by the use

of metastable group 13 element halides in solution in the presence of the protecting amide

group [N(SiMe3)2] or rather the all-hydrocarbon ligand [C5Me5] under appropriate

conditions.[26-28] Within this reaction scheme a wide variety of metalloid clusters has been

obtained ranging from one naked aluminum atom in [Al7R6]- [21], over 38 naked atoms in

[Al50C120H180][29] up to 57 in [Al77R12]2- [30]‘in which the arrangement of the Al atoms mimics

the close-packed structure of Al metal’.[31] In addition, the corresponding gallium clusters

have also been investigated under similar conditions with [Ga84(N(SiMe3)2)20]4- as one of the

largest metalloid group 13 metal cluster known so far.[32] In the context of metalloid clusters

and therefore in searching appropriate systems for unifying molecular chemistry and, in a

wider context, material science, the low valent group 13 organyls E(I)R introduced above,

have been proven to be promising reactants or rather ligand systems in this particular field of

research. One part of ECp* chemistry, which nicely presents the feasibility of using such low

3 1. Motivation

valent group 13 species in material science, is the soft chemical synthesis of intermetallic

materials. The preparation of M/E Hume-Rothery phases obtained as colloidal nanoparticles

or rather powder materials, can be achieved in two different ways. On the one hand, using

combinations of all-hydrocarbon precursors, e.g. [{Cu(mesityl)}5] or [Ni(cod)2] and ECp*,[33-

35], and on the other, using single source precursors with already existing M-E bonds of the

desired intermetallic material.[15, 36, 37] In this context, no oligonuclear copper/gallium

complex of the general formula CuaGab has been explored and just three examples containing

Cu-Ga interactions in monomeric complexes have been reported so far.[38, 39] Thus, the

primary aim of this work has been the systematical investigation of CuaGab oligomers by

reaction of readily available copper starting materials and low valent gallium ligands as

potential intermediates or starting materials in soft chemical synthesis of larger M/E

intermetallic materials or nanoparticles. The reaction pathways and consequently the product

formation, namely Cu/Ga ratios, can be easily controlled by the oxidation state of the copper

centre as well as variation of the ligand system of both, the group 11 metal and the E(I)R

species. The successful use of GaCp* containing compounds as precursor materials for the

preparation of nanoparticles has been previously described. The selective hydrogenolysis of

Cp* from GaCp* results in the formation of [Ru2(Ga)(GaCp*)7(H)3] which can be viewed as

an early step on the way to Ru/Ga nanoparticles.[40] The molecular and solid state chemistry

of gallium and zinc includes a couple of similarities. While θ-CuE2 (E = Al, Ga) and Cu1–xAlx

intermetallic phases can be obtained from [CpCuL] and ECp* (L = PMe3, CNtBu), related α-,

β-, γ-Cu/Zn phases have been prepared by reaction of [CpCuL] with ZnCp*2.[33-35, 41]

Subsequently, treatment of GaCp* containing starting materials [M(GaCp*)n] (M = group 6-

10 metal, n = 4-6) with ZnR2 (R = Me, Et) has been developed and permits access to highly

coordinated and metal-rich compounds [M(ZnR)n] (n ≥ 8; M = Mo, Ru, Rh; Ni, Pd, Pt; R =

Me, Et, Cp*).[42-44] The structural features of such compounds show strong similarities to

intermetallic solid state phases of the Hume-Rothery type. Obviously, the success of this

organometallic precursor chemistry for intermetallic materials depends on the knowledge of

the underlying coordination chemistry of E(I)R. The above mentioned reaction schemes are

limited in two ways. Firstly, all these metal-rich zinc compounds have been prepared using

mononuclear transition metal GaCp* compounds and secondly, it is absolutely necessary that

GaCp* is present in the ligand sphere, otherwise the reaction will not work. The transfer of

this reaction scheme in order to aim clusters with higher nuclearity has been successfully

investigated in this work. Treatment of an unsaturated palladium dimer, namely [Pd2(µ-

GaCp*)3(GaCp*)2], with ZnMe2 resulted in the highly coordinated cluster

4 1. Motivation

[Pd2Zn6Ga2(Cp*)5(CH3)3]. The formation of M-ZnR bonds and larger metal-rich compounds

can also be extended to the ZnI/Zn0 pair by using Carmona´s stable Zn(I) compound

[Zn2Cp*2] as obtained by the reaction of [M(GaCp*)4] (M = Pd, Pt). Herein, the unusual

ligand system {ZnZnCp*} with a fully intact Zn-Zn bond is trapped as a ligand coordinated to

the transition metals. In general, nothing has been known about the coordination chemistry of

the Zn(I) dimer until this work. Thus, a major aim of the present work has been to investigate

initial insights into the reactivity of [Zn2Cp*2] towards non GaCp* containing complexes of

the late transition metals which contain substitution labile ligands auch as olefins, alkyl

groups and the N-chelating ligand tmeda.

5 2. Introduction

2 Introduction

As 80% of the chemical elements are metallic it is of no surprise that the chemistry of

compounds containing metal-metal interactions is a very diverse research area. It should be

noted at this point, that the description of intermetallic chemistry includes well-defined,

stoichiometric intermetallic compounds and intermetallic phases which can consist of broad

areas of metal-to-metal constitutions. Thus, the item intermetallics will be used as a general

term for intermetallic compounds as well as intermetallic phases. The first scientific

description of an intermetallic phase was given in 1839 by Karl Karsten, which is todays’

known CuZn alloy (β-brass).[45] At this time it was common to treat alloys with acids to study

their composition. Thus, observable discontinuities by treatment of equiatomic compositions

of copper-zinc alloys with acids lead Karsten to assume that ‘these alloys…are true chemical

compounds’.[45] In the course of time, intermetallic chemistry has become what it is today,

namely one of the most diverse research areas which can be viewed as the staging area of

different scientific fields; mainly chemistry, physics, metallurgy, material science and

engineering (Figure 1).

Figure 1. Components of intermetallic chemistry.

6 2. Introduction

2.1 Basic Approach to Intermetallic Chemistry: Intermetallics, Zintl-

phases and Clusters

The classification of intermetallic phases into the series of inorganic compounds, i.e. salt-like

compounds, clusters, electron-deficient compounds and Zintl-phases, is quite difficult due to

their manifold properties which often converge. A comprehensive overview has been given in

the article ‘Bonding Patterns in Intermetallic Compounds’ by Nesper in 1991, who tried to

integrate intermetallic materials by reference to analogies of well described inorganic

compounds.[46] Nesper classifies intermetallic phases, Zintl-phases and cluster compounds

according to electronic properties and structural analogies. But is it necessary to draw a line

between these areas of research or should it be rather a kind of network? Indeed, the best way

would be a connection of all these formally separate fields of inorganic chemistry. While

Zintl-phases fill the space between the salt-like, ionic structures and intermetallics of the bulk

material, cluster chemistry can be viewed as the transition between molecular compounds and

even the pure solid state intermetallics. The properties of Zintl-phases depend on the used p-

block element and can be either assigned to intermetallics, for example the NaTl[47] Zintl-

phase featuring a high ionic character, or to cluster compounds at which polyanionic

endohedral cluster-like structures are formed, e.g. [Na(crypt)]3[Sb7][48]. As nice as his

overview is, Nesper himself recognized that ‘the definition of the intermetallic compounds as

a uniform family is not easy. A clear definition is almost impossible, in particular owing to

the fluid transition to the semiconductors, the kaleidoscopic variety of chemical bonding, and

the extremely diverse structural chemistry’.[46]

2.1.1 Influences and Parameters Determining the Formation of Intermetallic Phases

The chemical, physical and structural properties of the whole family of intermetallic phases

strongly depend on a multitude of factors from a wide range of scientific areas, most

commonly physical parameters such as temperature, pressure and density, but also factors on

an atomic level, most mentionable electronegativity, electron configurations and atomic radii.

The comprehension, description as well as prediction of the formation of solid solutions and

thus, their existence, can be visualized by several different models.[49-52] One of the most

important founders in the exploration of the formation of intermetallics in dependency of

atomic scale parameters, is William Hume-Rothery, who based his rules on the position of the

7 2. Introduction

element in the periodic table and thus, strictly focused on the atomic sizes and the electronic

behaviour.[53-59] In accordance with the so called Hume-Rothery rules three factors have to be

taken into account if alloy formation is to be favoured or not.[49, 58, 59]

The concept of atomic size-factor.[58, 59] If the atomic sizes of both components differ by

more than 13-14% the size-factor is unfavourable and alloy formation is not expected by more

than a few atomic percent. It should be noted, that the percentage limitations were adjusted a

few years later, up to 15%. The atomic size is defined as the closest distance between the

elements in the crystal structure. This is a very simple view but it has the advantage that

knowledge about the structure of the alloy is not necessary, thus, the atomic diameter can be

viewed as a constant parameter (Figure 2).

Figure 2. Influence of atomic size on the solubility in magnesium.[58]

The electronegative valency effect.

[58, 59] If the electronegativities of both components are

similar, alloy formation will be expected. The more electronegative the element in solution,

and the more electropositive the base solvent, the greater is the tendency to form compounds

with ordered structures instead of the primary solid solution. The existence of an area of

8 2. Introduction

overlap at which both the formation of the primary solid solution as well as compound

formation, is favoured or rather stabilised within the electronegativity difference makes it

quite difficult to define an appropriate value of electronegativity difference above which the

alloy formation is hindered. As suggested by Darken and Gurry, a nearly acceptable value for

hindrance of the formation of an intermetallic is ½ unit on the electronegativity scale

(electronegativity values according to Pauling).

The relative valency effect.[58, 59] Elements of higher valency are more soluble in metals with

a lower valency than vice versa. The validation of this effect works well for high valency

elements which crystallize according to the (8-N) rule with monovalent elements of copper,

silver and gold. For instance, copper dissolves 11.3 at.-% of silicon, but silicon dissolves just

8.5 x 10-5

Table 1. Example systems for the relative valency effect.

at.-% copper. However, the effect of relative valency becomes more unusable and

difficult when considering elements of the same group and transition metals (Table 1).

[59]

Over the years several correlation models have been reported in order to verify the Hume-

Rothery rules and to possibly find direct correlations between different factors assumed by

9 2. Introduction

Hume-Rothery.[49, 60] One correlation has been reported in 1953 by Darken and Gurry who

correlate the electronegativity (ordinate) and the atomic radius (abscissa) for the formation of

magnesium, silver and aluminium alloys which is called ‘the simultaneous effect on size

factor and electronegativity’ (Figure 3).[49, 58, 60] The Darken and Gurry maps place the

elements according to their atomic scale properties with respect to size and electronegativity,

as mentioned above. The ellipsoid which is drawn around the selected element is based on

two definitions: firstly, the electronegativity difference should not exceed ± 0.3% and

secondly, the atomic radius should not differ more than ± 15%.

Figure 3. Darken and Gurry map for known solubilities of elements in silver.[58]

Besides the prediction of solid solution formation, structural features can also be somehow

related to the properties of elements with regard to atomic size and electronegativity. It was

again William Hume-Rothery who found that the stability of solid state structures depends on

the electronic properties of the attended elements giving rise to the valence electron

concentration (VEC) rule.

[49, 58, 59, 61] The valence electron concentration is defined as the

number of valence electrons per atom ne/na. Intermetallics which follow the VEC are called

10 2. Introduction

Hume-Rothery phases or rather electron compounds and are formed between transition metals

and main group elements of the groups 13-15. The valence electron count has been described

as followed:

Table 2. Valence electron counts for Hume-Rothery rules.

Element Valence electrons

Cu, Ag, Au 1

Mg, Zn, Cd, Hg 2

Al, Ga, In 3

Si, Ge, Sn 4

Sb 5

Any other transition metal 0

Hume-Rothery suggested that element combinations which display the same ne/na ratio,

crystallize in equal systems, i.e. VEC ≅ 3/2, β-phase (body-centered cubic, bcc; complex

cubic); VEC ≅ 21/3, γ-phase (complex cubic with 52 atoms/unit cell or superstructures), VEC

≅ 7/4, ε-phase (hexagonal close-packed, hcp).[49]

Table 3. Structures and VEC for Cu1-xZnx.

Phase Example Constitution x VEC Structure type

α Cu(Zn) 0-0.38 1-1.38 fcc

β CuZn 0.45-0.49 1.45-1.49 bcc

γ Cu5Zn8 0.58-0.66 1.58-1.66 complex cubic

ε CuZn3 0.78-0.86 1.78-1.86 hcp

η (Cu)Zn 0.98-1 1.98-2 distorted hcp

Hume-Rothery phases are not stoichiometrically combined but rather consist as broad phases,

thus, structure types can be consistent for several constitutions of intermetallics in one phase

given the valence electron concentration is equal, otherwise rearrangement into the next stable

structure type can occur. Additionaly, there are further intermediate structure types such as the

11 2. Introduction

ς-phase, η-phase and µ-phase (β-Mg). These phases are comparatively rare and exist

sometimes in a very unexpected or rather strange area of high temperatures.

Figure 4. Equilibrium diagram of the system Cu/Zn.

Table 4. Examples for Hume-Rothery phases and corresponding VEC.

[59]

β-phase, VEC ≅ 3/2 = 21/14 γ-phase, VEC ≅ 21/13 ε-phase, VEC ≅ 7/4 = 21/12

CuZn Cu5Zn8 CuZn3

Cu3Al Cu9Al4 CuCd3

Cu5Sn Cu31Sn8 Cu3Sn

AgZn Ag5Zn8 AgZn3

AuZn Au5Zn8 AuZn3

AuMg Au5Cd8 Au5Al3

FeAl Fe5Zn21 Ag3Sn

CoAl Co5Zn21 Au5Al3

12 2. Introduction

The VEC rule has also been adopted in the description of Laves-phases, although these types

of phases are mainly based on size factors which gave them the name size-factor

compounds.[49, 51, 62-64] Laves-phases are composed of the general formula AB2 and consist

mainly of alkaline (earth) metals and transition metals. Requirement for the formation of

Laves-phases is the ideal radius ratio between metal A and metal B rA/rB which should ideally

match 1.225. Notably, the values essentially observed range from 1.05 and 1.7.[49, 62] Although

their existence is dominated by atomic size factors, the valence electron concentration can be

consulted to describe their structural features. The cubic MgCu2 type is predominately formed

for VEC ~ 1.3-1.8, hexagonal MgZn2 for VEC ~ 1.8-2.2 and hexagonal MgNi2 for VEC ~ 1.8-

2.0 (Figure 5).[49]

Figure 5. Crystal structures of MgCu2, MgZn2 and MgNi2. The solid lines between Mg atoms (green) indicate

the connection only and not the real interactions.[61]

Notably, all these rules which are just indications of forming an intermetallic phase or

compound, have been somehow verified and are not applicable for several metal systems.

Reproduced by permission of Prof. Caroline Röhr.

[65,

66] For instance, the concept of atomic size-factor mentioned in the Hume-Rothery rules is a

negative principle.[59] That means, if the atomic diameters differ more than 15% formation of

an intermetallic compound is not expected. However, if they differ less than 15% does not

mean that alloy formation has occurred, only that it is favourable. In general, the above

mentioned parameters are not unique rules and descriptions but instead there are several other

definitions which can be used in prediction of stable intermetallics formation. However,

mostly all these parameters and rules together just build up a framework for expections and

have to be taken into account in equal parts.

13 2. Introduction

2.1.2 Zintl Compounds: Transition between Intermetallics, Salt-like Compounds and

Clusters

Zintl compounds display a very special case of intermetallics as well as cluster compounds.[47,

67-76] In 1939, Zintl reported on intermetallics which did not fit into the scheme assumed by

Hume-Rothery. These were those intermetallics that exist between strong electropositive

alkali metals or alkaline earth metals and more electronegative main group elements of the

group 13 with NaTl as the prototype of Zintl compounds.[47] Zintl observed a curious

arrangement of Na and Tl atoms in the crystal structure which has also been observed in

several other element combinations such as LiAl, LiGa and NaIn. The Tl atoms are arranged

in the diamond-type structure similar to group 14 elements. But Tl has just three valence

electrons, one less than the group 14 elements. He suggested an electron transfer from the

alkaline and alkaline earth metals to the main group elements, in this case Tl. Once Tl has

four valence electrons, the arrangement in the diamond lattice is favoured. The small electron

providing sodium atoms are intercalated in the holes of this diamond framework. Thus, the

characteristic ionic character of such phases, Na+Tl-, gave them the name polar intermetallic

phases. The electron transfer has been further investigated by Klemm and Busmann which led

to todays Zintl-Klemm-Busmann concept in accordance to the (8-N) rule.[74, 77, 78]

Beside these polar intermetallic phases, a second family of compounds has been essentially

discovered by Zintl. In the 19

th century Joannis observed the solubility of lead and antimony

metal in a liquid mixture of sodium and ammonia giving rise to green and red solutions. This

aroused the attention of Zintl who observed similar solubility and colour changes of the

intermetallic phase Na4Pb9 in ammonia and postulated the formation of ionic species due to

potentiometric titration.[70, 79-81] Later on, the observed green or red solutions found by Joannis

and Zintl have been characterized as [Pb9]4- and [Sb7]3- containing liquids. These polyanionic

cluster-like deltahedrons are called Zintl ions. Such anions could not be observed using group

13 elements in accordance with the Zintl line in the periodic table between the group 13 and

group 14. However, the characterisation of Zintl ions was difficult for a long time because of

their highly reactive behaviour and thus low stability to give suitable crystalline material

before decomposition into the corresponding alloy. The modification of the original synthetic

route using ethylendiamin instead of liquid ammonia resulted in some crystalline material

which has been insufficiently described as [Na4(en)7(Sn9)].[68, 82] The breakthrough in

isolating stable Zintl ionic compounds was accomplished in 1975 by Corbett et al by using

14 2. Introduction

the strong multidentate ligand [2.2.2]crypt as an ion-sequestering molecule. The reaction of

NaSb or NaSb3 with [2.2.2]crypt in ethylendiamine resulted in red crystals of the composition

[Na(crypt)]3[Sb7] as determined by single crystal X-ray diffraction.[48] Notably, the use of

crown ethers, e.g. [18-crown]-6, and N,N-dimethylformamid has been recognized as being

suitable by Fässler.[70, 83] Most interestingly, direct connectivity between polar intermetallic

phases and Zintl ions has been reported for Cs4Ge9.[67, 84] The polyanionic Zintl ion [Ge9]4- has

been found to be a structural part of the polar intermetallic phase and in the Zintl compound.

In spite of the reducing properties of the Zintl ions, the reactivity and the use as reagents in

(organometallic) chemistry has been investigated in several reaction schemes[67, 70], e.g.

formation of interconnected chain-like structures[67, 85-87], functionalisation with main group

elements[88] and reactivity towards transition metals and transition metal fragments.[89-98] The

functionalisation with transition metal elements especially aroused great interest for products

for which the term intermetalloid cluster has been suggested.[70] For instance, classical ligand

exchange reactions of [E9]2- (E = Sn, Pb) with reactive transition metal complexes [LM(CO)3]

(L = η6-arene, M = group 6 metal) leads to the substitution of the arene ligand and formation

of [(E9)M(CO)3]4-.[89-93] Similar product formation has been observed using group 12 metal

compounds such as ZnPh2.[94] In both cases, the (transition) metal fragment is attached to the

open side of the square prismatic [E9]4- unit leading to bicapped square antiprismatic structure

of the resultant [E9M]4- cluster (Figure 6).

Figure 6. Left: Overall shape of [Sn9M(CO)3]4- and [Pb9M(CO)3]4- (M = Cr, Mo, W, shown as a yellow sphere).

Right: Structure of [E9Zn-Ph]3- (E = Si, Ge, Sn, Pb; Zn shown as a yellow sphere).[67] Reprinted with permission

from S. C. Sevov and J. M. Goicoechea, Organometallics 2006, 25, 5678-5692. Copyright (2006) American

Chemical Society.

15 2. Introduction

In addition, intercalations of transition metals into the cluster-like framework of the Zintl ions

have been observed for several metals. It has been found that cluster formation strongly

depends on the used group 14 Zintl anion and, for sure, the precursor. The prototype of such

intercalated intermetalloid clusters has been observed via reaction of [Ge9]2- with

[Ni(CO)2(PPh3)2] under appropriate conditions resulting in the formation of

[Ni@(Ge9NiPPh3)]2-.[95, 96] The same reaction with [Sn9]2- results in [Ni@(Sn9Ni(CO))]3-

which obviously contains carbon monoxide attached to nickel and a higher charge.[99] In

addition, starting from [Pb9]4-, naked clusters of the type [M@Pbx]2- (M = Pt, x = 12; M = Ni,

x = 10) have been obtained under suitable conditions.[97, 100] Most interestingly, slight

deviations of reaction conditions allow stepwise functionalisation of the cluster species, nicely

illustrated by Sevov et al using [Ge9]3-.[98] Treatment of the germanium cluster with [Ni(cod)2]

results in the formation of the intercalated cage compound [Ni@Ge9]3-. The second step of the

reaction sequence has been the reaction of [Ni@Ge9]3- with [Ni(CO)2(PPh3)2] yielding

[Ni@(Ge9Ni(CO))]2- and finally, the substitution of the CO ligand with phenylacetylide

(Figure 7).

Figure 7. Stepwise process of cluster manipulations starting with an empty [Ge9]3- cluster (a) and ending with

the Ni-centered Ni(C≡CPh)-substituted species [Ni@(Ge9Ni-CCPh)]3- (d). The first step is insertion of a central

Ni atom by reaction with [Ni(cod)2], which produces [Ni@Ge9]3- (b). Added to this cluster is a capping Ni(CO)

fragment by the reaction of [Ni@Ge9]3- with [Ni(CO)2(PPh3)2], which results in [Ni@(Ge9Ni-CO)]2- (c). Ligand

exchange reaction of the latter with K(C≡CPh) replaces the carbonyl ligand with phenylacetylide in (d).[67]

Reprinted with permission from S. C. Sevov and J. M. Goicoechea, Organometallics 2006, 25, 5678-5692.

Copyright (2006) American Chemical Society.

16 2. Introduction

Comment on electron count and structural features. The electronic properties of Zintl ions

are mainly of electron-deficient character which can be described in accordance with Wades´

rules for classical borane clusters.[67, 69-73] Herein, each B-H unit is substituted by a lone pair

of the group 14 elements giving rise to two electrons of the group 14 element taken into

account for the cluster skeleton bonding interactions which leads to the following structural

features with n = cluster vertices: closo-deltahedra 2n+2 electrons, nido-deltahedra 2n+4

electrons, arachno-deltahedra 2n+6 electrons. This works well for smaller Zintl ions [E4]4-

and [E5]2- but becomes more difficult for nonagon cages [E9]x- (x = 2, 3, 4) because of the

high fluctional behaviour of the cages.[70, 101] While [E9]2- exhibits 20 electrons it should

display 3-capped prism (closo). In comparison, [E9]4- contains 22 electrons leading to a nido

type arrangement and finally, [E9]3- can not count to any Wade-type borane cluster because of

its electron count of 21. Thus, drawing conclusions from skeletal electrons to the structural

behaviour is somehow restricted. In addition, higher charged Zintl ions like [E9]4-

2.1.3 Cluster Compounds: Transition between Intermetallics and Molecular

Compounds

can interact

with the cationic counter parts leading to significant distortion of the polyhedral environment.

The most common definition of metal atom cluster compounds has been described by F. A.

Cotton in the 1960s. He defines clusters as ‘those containing a finite group of metal atoms

held together entirely, mainly, or at least to a significant extent, by bonds directly between the

metal atoms even though some non-metal atoms may be associated intimately with the

cluster’.[19] The simplest molecules in which metal-metal interactions can be observed are of a

dimeric nature. However, enhancement of the nuclearity and thus increasing the possibility of

prevailing metal-metal contacts leads to a new class of compounds. Beside the intrinsic point

of view to develop something ‘new’, the interest in metal cluster compounds aroused from the

fact that ‘metals…offer an exceptional opportunity to study the pathway which leads from the

bulk to the molecular state and finally to mononuclear complexes’.[22] The features providing

the bridge between the bulk intermetallics and smaller molecular units can be of a different

nature, i.e. electronic studies and similarities of structural features. Thus, the interest in metal

atom clusters derives from the perception and interest which can be (a) theoretical description

or rather (b) studies of structural properties of the (small) metal atom clusters and the bulk

material. One special case in cluster chemistry are naked metal atom clusters, which are free

of stabilizing ligand systems, e.g. [MAu12][24, 25] (M = Mo, W). The synthesis via laser

17 2. Introduction

vaporization techniques is quite special and often cluster formation through this method is not

very selective giving rise to many cluster aggregates in different compositions. Moreover, the

characterisation of such gas-phase metal clusters requires special mass spectrometric

analysis, as well as spectroscopic studies in the gas phase at low temperatures. Although the

fascination of such kinds of clusters is undisputed, their application as defined model systems

with regard to the bulk material, i.e. structural features, seems to be not really effectual.

Studies of structure and bonding characteristics and the stability of gas-phase cluster

molecules mainly aroused the attention of physicists and theoreticians. The counter part are

ligated metal atom clusters, containing metal cores wrapped into a stabilizing ligand shell, e.g.

[Mo6Cl8]2+ or [Rh6(CO)16].[19, 102, 103] Nowadays, several molecular cluster compounds are

known, involving a wide variety of combinations between transition metals and main group

elements. Certainly, their chemical and physical properties strongly depend on the ligand

shell, even though this can be accepted as they are helpful opportunities giving us the

possibility to study their properties with suitable analytical equipment under moderate

conditions and with, more or less, conventional (theoretical) descriptions, e.g. classical Wade

rules concerning borane clusters (including electron counting rules) and isolobal analogies.

Several years ago, a novel family of compounds was described as metalloid clusters which are

widely considered to be the most meaningful examples for the linkage between molecular

chemistry and material science in general.[26-28, 104-106] The term metalloid cluster was defined

by Schnöckel in 1999 and is based on the fact that ‘the definition chosen by Cotton for a

metal cluster is so general…we would like to distinguish them from such compounds.

Metalloid (metal-like) clusters should be epitomized by the property that the number of metal-

metal contacts exceeds the number of metal-ligand contacts and by the presence of metal

atoms which participate exclusively in metal-metal interactions’.[20, 21] The first metalloid

clusters reported contained noble transition metals, e.g. [Au55(PPh3)12Cl6][23] (Schmid´scher

gold cluster), [Pt6Ni38(CO)48H]5- [107](6 naked Pt atoms) and [Pd145(CO)60(PEt3)30][108] (55

naked Pd atoms). However, in the recent past metalloid cluster compounds containing main

group 13 metals attracted enormous attention showing the structural features of the bulk metal

material. Herein, the formation of [Al77(N(SiMe3)2)20]2- with 57 naked Al atoms is the largest

metalloid main group 13 cluster compound known today.[30] The preparation procedures are

based on low valent group 13 metal halide solutions E(I)X (E = Al, Ga; X = halide, e.g. Cl,

Br). These meta-stable compounds can be obtained at high temperatures of around 1000°C

and trapped as solvent stabilized solutions at low temperatures in A/B solvent mixtures (A =

e.g. toluene, m-xylene; B = donor solvent, e.g. ethers, amines). Disproportionation of the

18 2. Introduction

trapped EX molecules takes place between -80°C and +80°C into the bulk metal material

Emetal and the corresponding E(III)halide. The addition of bulky ligands R (R = N(SiMe3)3,

C(SiMe3)3, Si(SiMe3)3, Cp* or P(tBu)2) during the warm-up process, results the trapping of

the metalloid clusters which can be viewed as intermediate steps from the meta-stable E(I)X

solution to the bulk material.[26-28] Certainly, the reaction pathways involved are quite

puzzling, leading to several reaction sequences and the mixing of oxidation states, e.g. redox

chemical proceses and X/R substitution reactions. Notably, the cluster size can be easily

controlled changing reaction conditions such as temperature, stoichiometry, solvent, donor

solvent and ligand system as it has been nicely presented for aluminium at which the cluster

series Al7[21]→Al12

[109]→Al14[110]→Al69

[111]→Al77[30] has been obtained under slight deviation

of the reaction conditions. The relationship between the group 13 metalloid clusters and the

bulk material in the solid state structures is shown in Figure 8. Herein, the structural

arrangement of the cluster cores can be relocated in the solid state phase of elemental

aluminium and gallium.

Figure 8. Metalloid aluminium and gallium cluster: cut-outs of intermetallic phases.[27] Reproduced by

permission of John Wiley and Sons.

[a] S. González-Gallardo, T. Bollermann, R. A. Fischer, and R. Murugavel, Chem. Soc. Rev. 2011, submitted.

19 2. Introduction

2.2 Organometallic Chemistry of Low Valent Group 13 Organyls[a]

It has been shown that low valent group 13 metals are suitable starting materials in the

formation of metalloid cluster compounds as highlighted in chapter 2.1.3. Such meta-stable

E(I)X solutions have also been used in stabilizing low valent group 13 metals described by

Schnöckel in the early 1990s. Herein, the first stable Al(I) organyl AlCp* was observed in the

reaction of (AlCl)meta and MgCp*2.[112] Just two years later, in 1993, Schnöckel again reported

on the formation of stable gallium(I) species GaCpR (CpR = Cp*, CptBu, Cp(SiMe3

)3 and

Cp(benzyl)5) which have been obtained via reaction of (GaCl)meta solutions with LiR or rather

MgR2 derivatives.[113] About fourty years before the discovery of stable low valent Al and Ga,

E.O. Fischer and H. Meister reported on the formation of cyclopentadienyl stabilised In(I) and

Tl(I).[114, 115] In contrast to low valent In and Tl, disproportionation tendency of Al(I) and

Ga(I) is favoured. Thus, stabilisation requires substituents with suitable electronic and steric

properties which suppress decay into the bulk material. The same concept of stabilisation

using bulky ligands with appropriate properties has been the assumption for successful

preparation of the metalloid clusters mentioned above. However, since the handling of

(ECl)meta solutions, produced from HCl and Eelemental at 900-1200 °C in toluene or

diethylether, is quite special, these days alternative reaction pathways are used following

procedures reported by Jutzi and Roesky.[116, 117] Herein, quantitative yields can be obtained

via reductive dehalogenation of Cp*EX2 (E = Al, X = Cl; E = Ga, X = I). While all these low

valent group 13 species exhibit oligomeric structural features, e.g. [AlCp*]4 and [GaCp*]6, in

the solid state, the use of bulky organic groups, e.g. ß-diketiminates DDP (HC(CMeNC6H3-

2,6-iPr2)2) and substituted aryl groups such as Mes* (2,6-bis(2,5,6-triisopropylphenyl)phenyl)

resulted in the formation of monomeric E(I) units.[118, 119]

2.2.1 Theoretical Investigations on the Bonding Characteristics of Low Valent Group

13 Ligands

The characteristics of the bonding between group 13 ligands E(I)R and transition metals in

complexes of the general formula [LnM-ER] and [M(ER)4] (E = B, Al, Ga, In, Tl; R = Me,

Cp, Cp*, N(SiH3)2) have been extensively studied mainly by the groups of Frenking and

Cowley around the year 2000 (Figure 9).[7, 8, 120-125]

20 2. Introduction

Low valent group 13 organyls exhibit a free electron-pair located in the σ-orbital (HOMO)

and in addition two unoccupied p-orbitals (LUMO), which are located perpendicular to the E-

R axis. These elementoid ligands feature isolobality to carbon monoxide and phosphanes PR3

(R = alkyl and aryl group).

R E O C

px

py

π∗

Figure 9. Isolobality of E(I)R species and CO.

The term of attractive interactions between E(I)R and transition metals has to be divided into

mainly electrostatic (Coulomb) and covalent (orbital) interactions.

Figure 10. Interactions in M-E(I)R complexes.[7,8] Reprinted with permission from a) J. Uddin and G. Frenking,

J. Am. Chem. Soc. 2001, 123, 1683-1693, b) J. Uddin, C. Boehme and G. Frenking, Organometallics 2000, 19,

571-582. Copyright (2000 and 2001) American Chemical Society.

The differences in electronegativitity of the group 13 element and the transition metal M leads

to the fact that M-E(I)R interactions shall be deemed to be mainly polarized donor-acceptor

bonds, M(δ-)-E(δ+). The σ-donor ability of the group 13 ligand and thus M-E(I)R orbital

interactions result from a combination of the HOMO of the group 13 ligand with the

21 2. Introduction

corresponding dz2 orbital located at the transition metal centre. As a result of the lone pair

effect the M-E bond energy significantly decreases attendant with higher atomic numbers

(B→Tl). However, besides the σ-donor properties discussed above, the free unoccupied p-

orbitals can be generally used for M→ER back-donation. What means generally? The π-

acceptor properties of low valent group 13 ligands directly depend on the group 13 element

itself and on the electronic properties of the organic substituent R. If the organic group R is a

strong π-donor, as is the case for Cp* and other Cp derivatives, which pushes electron density

into the unoccupied p-orbitals of the group 13 metal, the π-acceptor efficiency will be

decreased as a logical consequence. It should be noted that haptotropic shifts η5→η3→η1 of

the Cp* group can be helpful to increase the π-acceptor properties. However, the contrary

situation results when using weaker π-donor ligands such as phenyl groups, alkyl groups as

well as more steric substituents, e.g. Ar* = 2,6-(2,4,6-triisopropylphenyl)-phenyl and

{C(SiMe3)3}.

2.2.2 Coordination Chemistry of E(I)Cp* Ligands Towards Reactive Transition Metal

Compounds

One of the most important and effective synthetic strategy for the preparation of transition

metal-group 13 metal complexes includes ligand substitution reactions at [LnM] fragments

using E(I)R compounds as reagents.

[Fe2(CO)9] GaCp* Ga Fe

CO

CO

CO

CO

Fe Fe

OC

OC

OC

CO

CO

CO

Ga

Ga

[Fe(CO)3(cht)] 3 GaCp*

(cht = cycloheptatrien)Ga

[M(CO)4(nbd)] 2 GaCp*

(M = Cr, Mo; nbd = norbornadiene)

M

Ga

Ga CO

CO

CO

CO

Scheme 1. Reaction schemes of transition metal carbonyl compounds with GaCp*.

22 2. Introduction

At the beginning of the exploration, initial attempts started with binary carbonyls, e.g.

[Fe2(CO)9] or [Co2(CO)8] and mono substituted monomers, e.g. [(nbd)Mo(CO)4] or

[(cht)Fe(CO)3] (nbd = 2,5-norbornadiene, cht = cycloheptatriene) due to the isolobal behavior

of E(I)R towards carbon monoxide, leading to the corresponding substitution products under

liberation of CO (Scheme 1).[116, 126, 127] Already in these first reaction schemes it became

obvious, that the E(I)R ligands exhibit a high synthetic potential acting as two electron

ligands in bridging and terminal positions, forming oligomeric transition metal-main group

metal units and cluster-like structures. In the course of these early studies, an interesting

observation has been made, nicely illustrating the flexible, comparably soft properties of the

Cp* group. In general, Cp* groups are coordinated η5 to the group 13 element. However, at

times facile haptotropic shifts, i.e. η5→η3 and η5→η1, can be observed caused by steric

repulsion.[116, 127]

The successive substitution of CO (or other strong acceptor ligands) is limited by the

coordination of GaCp*. The strong σ-donor abilities of GaCp* increase the over-all electron

density of the transition metal on coordination and thus leads to stronger π-back bonding of

the remaining CO ligands, which then cannot be substituted anymore. Therefore, homoleptic

GaCp* containing complexes can only be obtained by the substitution of labile and very weak

π-acceptor ligands.

4 GaCp*

GaCp*

M*CpGa GaCp*

GaCp*

[M(cod)2]

(M = Ni, Pt)

Pd

N

N Me

Me 5 GaCp*- tmeda- Cp*GaMe2

- 2 cod

(M = Ni, Pd, Pt)

Scheme 2. Synthesis of [M(GaCp*)4] (M = Ni, Pd, Pt).

Thus, [Ni(GaCp*)4], the analogue to [Ni(CO)4], is readily available from [Ni(cod)2] (cod =

1,5-cyclooctadiene) and likewise [M(GaCp*)4] (M = Pd, Pt) can be prepared from [Pt(cod)2]

and [Pd(tmeda)(CH3)2] (tmeda = N,N,N',N'-tetramethyl-ethane-1,2-diamine).[126, 128] The

particular bi-functional feature of E(I)R ligands in contrast to their isolobal carbon or

23 2. Introduction

phosphorus ligator analogues is the combination of reduction (insertion) and coordination

properties which is nicely shown by the synthesis of [Pd(GaCp*)4] from [Pd(CH3)2(tmeda)]

and five equivalents GaCp* with [Cp*Ga(CH3)2] as the stoichometric by-product (Scheme

2).[128] As the first example of a GaCp* adduct to a more electrophilic metal center, the

compound [Zn(GaCp*)4][BArF]2 (BArF = B{C6H3(CF3)2}4) has been synthesized featuring the

homoleptic di-cation [Zn(GaCp*)4]2+ and illustrates the principal stability of cationic

homoleptic GaCp* complexes.[129] Further investigations showed, that electron rich and more

electronegative metal ions, i.e. Cu(I) and Ag(I), form stable cationic homoleptic complexes

[M(GaCp*4)]+ being isoelectronic congeners of the d10 metal compounds discussed above.[39]

Notably, it has been shown that using more electrophilic metal ions leads to competing

reaction sequencies, namely coordination of GaCp* and Cp* transfer reactions yielding

thermodynamically very stable half-sandwich complexes [Cp*M(GaCp*)3][BArF]m (M = Fe,

m= 1; M = Co, m =2). Thus, the reaction of the cationic transition metal acetonitrile complex

[Fe(CH3CN)n]2+ with GaCp* leads to [Cp*Fe(GaCp*)3][BArF] via a redox neutral Cp*

transfer and [Ga2Cp*][BArF] as a by-product, while the formation of

[Cp*Co(GaCp*)3][BArF]2 from [Co(CH3CN)6][BArF]2 is accompanied by oxidation of Co(II)

to Co(III) with GaCp* as the oxidant (Scheme 3).[39]

MH3CCN

H3CCN NCCH3

NCCH3

NCCH3

NCCH3

n+[BArF]-

n

+ 4 GaCp*, C6H5FM

*CpGa GaCp*GaCp*

m+

[BArF]-m

- [Ga2Cp*][BArF]

Scheme 3. Synthesis of cationic species [Cp*M(GaCp*)3]m+ (M = Fe, Co).[39]

However, oligonuclear complexes of the general type [Ma(ECp*)b] (a < b) have been isolated

under optimized conditions and reaction stoichiometries with the right choices of the

transition metal precursors (Scheme 4).[128, 130-132] Two different reaction pathways have to be

considered: the reaction of tetra gallyene coordinated d10 metal complexes [M(GaCp*)4] (M =

Pd, Pt) with [Pt(cod)2] and the immediate addition of GaCp* (building block synthesis)

leading to dinuclear metal-rich compounds, e.g. [PdM(GaCp*)5] (M = Pd, Pt), and the direct

24 2. Introduction

synthesis under suitable reaction conditions leads to kinetically controlled cluster growth, e.g.

[PdM(GaCp*)5] (M = Pd, Pt), [Pd3(ECp*)8] (E = Ga, In).

GaCp*

PtM

GaCp*

Cp*Ga

*CpGa

GaCp*

PtM

GaCp*

Cp*Ga

*CpGa GaCp*MGaCp*

GaCp**CpGa

GaCp*[Pt(cod)2] GaCp*

M = Pt, Pd

xs. GaCp*

Scheme 4. Building block synthesis of dimeric d10 metal compounds.[128] Reprinted with permission from C.

Gemel, T. Steinke, D. Weiss, M. Cokoja, M. Winter and R. A. Fischer, Organometallics 2003, 22, 2705-2710.

Copyright (2003) American Chemical Society.

Notably, while the homoleptic 18 valence electron closed shell compounds [M(GaCp*)4] (M

= Ni, Pd, Pt) are inert against substitution reactions, the oligonuclear compounds [Ma(ECp*)b]

display ligand exchange reactions, leading to substitution products of the general type

[LxMa(ECp*)b-x].[132] However, the variety of these products is strongly limited in terms of

ligand type (AlCp*, GaCp*, CO, phosphanes, isonitriles) and substitution numbers (mostly x

= 1, rarely x = 2) of the incoming ligands L.

[Pd2(dvds)3]

Pd

Cp*In

Pd

InCp*

Cp*In

Pd

InCp*

*CpIn

*CpIn

InCp*

InCp*

Pd Pd

Cp*Ga

GaCp*Ga

Cp*

GaCp**CpGa

Pd

Cp*Ga

Pd

GaCp*

Cp*Ga

Pd

GaCp*

*CpGa

*CpGa

GaCp*

GaCp*

PdPd

*CpAl AlCp*

Pd

Cp*Al

AlCp*

*CpAl AlCp*

8 InCp*, n-hexane, 50°C- 3 dvds

8 GaCp*, n-hexane, - 30°C- 3 dvds

8 GaCp*, toluene, rt- 3 dvds

10 AlCp*, benzene, 60 °C

Scheme 5. Synthetic routes to oligonuclear transition metal ECp* (E = Al, Ga, In) compounds.[132] Reproduced

by permission of John Wiley and Sons.

25 2. Introduction

Notably, ligand exchange reactions using di-nuclear starting materials show no changes of the

over-all linear geometry, whereas the tri-nuclear Pd/In compound [Pd3(GaCp*)4(µ2-GaCp*)4]

reacts with phospanes (dppe and PPh3) under rearrangement of the [Pd3In8] core from linear

to trigonal bi-pyramidal geometry.

2.2.3 Applications of Low Valent Group 13 Organyls in Material Science

The soft and flexible binding modes of the Cp* group as well as the easy split off under mild

conditions declares these ligands as promising starting materials in the formation of nano

materials. Thus, in the recent past it has been shown that ECp* and its transition metal

compounds can be used as single source precursors or rather solid state intermediates in the

formation of intermetallic phases. In general, mixing transition metal precursors containing

substitution labile ligands (at the best all-hydrocarbon ligands to avoid impurities and by-

product formation) with ECp* in inert solvents under hydrogen atmosphere leads to the

formation of the corresponding nanomaterials (Scheme 6).

Cu

PMe3

+ 1/2 [AlCp*]43 bar H2, mesitylene, 150°C

- 2 Cp*H, - Cp, - PMe3

CuAl2(s)

+ 1/4 [AlCp*]43 bar H2, mesitylene, 150°C, 48h

- 2 C8H16β-CoAlCo

Ni + 1/4 [AlCp*]43 bar H2, mesitylene, 150°C, 4d

- 2 C8H16, - Cp*Hβ-NiAl

Scheme 6. Use of AlCp* in the formation of intermetallics.

26 2. Introduction

For instance, in 2006 Fischer et al reported on the preparation of the θ-CuAl2 phase

(Cu0.33Al0.67) via co-hydrogenolysis of AlCp* with [CpCu(PMe3)] in the ratio 1:2.[35]

Additionally, Ni1-xAlx nanoparticles (0.09 ≤ x ≤ 0.50) have been observed via soft chemical

co-hydrogenolysis of [Ni(cod)2] with AlCp* (1:1 ratio) in mesitylene at 150°C.[34] Treatment

of the resulted colloidal solution of intermetallic β-NiAl particles post-synthetically with 1-

adamantanecarboxylic acid (ACA) as a surface capping group, gives nearly monodisperse α-

NiAl colloids. The formation of β-CoAl alloy nanoparticles has been similarly observed from

co-hydrogenolysis of [Co(η4-C8H12)(η3-C8H13)] and AlCp* in mesitylene.[133]

Beside the soft chemical synthesis of M/E Hume–Rothery phases using two-component

reaction pathways, the hydrogenolysis of Cp* has been recognized as a useful method for the

synthesis of building blocks for the formation of intermetallics. In effect, the treatment of

[(η4-cod)Ru(GaCp*)3] with 3 bar H2 in mesitylene at 150°C in the course of seven days leads

to RuGa and RuGa2 particles, while the co-hydrogenolysis of [Ru(η4-cod)(η3-C4H7)2] and

GaCp* in a molar ratio of 1:2 yields RuGa2 and Ru

2.3 The Renaissance of Zinc Chemistry: Landmark [Zn2Cp*2]

particles.[40] Notably, in order to gain

insights into the formation of the above mentioned intermetallic phases, experiments have

been carried out to characterize possible early intermediates. Herein, hydrogenolysis of

[Ru(η4-cod)(η3-C4H7)2] in the presence of GaCp* has been resulted in the formation of

[(Cp*Ga)4(H)Ru(μ2-Ga)Ru(H)2(GaCp*)3] under substitution of Cp*H, cyclooctane and

isobutene. Comparably, [Ru(η2,η2-cod)(η6-cod)] reacts with GaCp* under H2 atmosphere and

excess GaCp* to form the same product. However, under slightly different reaction

conditions, the intermediates [(η4-cod)(η4-cot)Ru(GaCp*)] and [(η4-cod)Ru(GaCp*)3] have

been observed. Thus, [(Cp*Ga)4(H)Ru(μ2-Ga)Ru(H)2(GaCp*)3] can be considered as a

molecular intermediate in the formation of Ru/Ga nanoparticles during a wet chemical

synthesis.

As is taught to every undergraduate student, the most common oxidation state of group 12

elements is +II. For a long time the only group 12 metal known to exhibit +I was mercury.

This fact can be assigned to relativistic effects.[134, 135] The first known low valent group 12

metal has been found in calomel, i.e. Hg2Cl2. However, it took a long time until 1999 in

which the first metal-metal bonded complex containing σ-bonded silyl ligands had been

observed, followed by the formation of organomercury(I) and organocadmium(I) compounds

27 2. Introduction

[E2R2] using the bulky aryl ligand C6H3-2,6-(C6H3-2,6-iPr2)2 by Power et al about eight years

later.[136, 137] In the last few years, synthesis and characterisation of the Cp* stabilized low

valent Zn(I) dimer, [Zn2Cp*2], reported by Carmona in 2004 has been one of the most

impressive examples in the formation of (covalently) bonded metal-metal complexes. The

dimeric Zn(I) dimetallocene has been sythesized using ZnCp*2 and ZnEt2 leading to the

formation of [Zn2Cp*2] as the main product with small traces of the Zn(II) species

[Cp*ZnEt].[138] In 2005, a modified reaction pathway has been reported by the same group.

Herein, the direct reduction of ZnCl2 in the presence of KCp* leads to the dimeric Zn(I)

compound in quantitative yields on the gram scale.[139]

2.3.1 Theoretical Investigations into Reaction Pathways and Bonding Situation of

[Zn2Cp*2]

The first description of the possible reaction mechanism in the formation of [Zn2Cp*2] via

ZnEt2 has been described by Schnepf and Himmel in 2005.[140] The authors postulated

involvement of active zinc, formed in disproportionation steps during the reaction, which

reduces ZnCp*2 to obtain [Zn2Cp*2]. This assumption has been ruled out by Carmona et al. It

has been proven in several experiments that Rieke-zinc is not strong enough to achieve

reduction and this conversion.[139, 141] It has also been shown that the reaction mixture of

[Zn2Cp*2] and [Cp*ZnEt] remains unchanged when stirred at room temperature overnight.

This means, formation of the dimetallocene from the Zn(II) species has not been observed.

Thus, it became obvious that the reaction pathway proceeds via several independent and

competitive reaction sequencies.[142] The favoured perception has been that mainly radical

mechanisms have to be taken into account so that [Zn2Cp*2] is formed due to dimerization of

[Cp*Zn]• radicals. Similar reaction mechanisms and the existence of (intermediate) [RZn]•

species have been noted previously by Boersma et al who reported on the reactions of

reactive transition metal olefin compounds with ZnCp2.[143] However, the availability of

radical mechanistic steps has been confirmed in the course of theoretical investigations

(Figure 11).[142] Herein, two possible reaction pathways were calculated dealing with (1)

neutral charge electrostatic dimetallocene formation and (2) radical dissociation of parent

zincocene. Pathway (1) involves side-on attachment of ZnEt2 which results in splitting off the

sandwich structure and formation of an intermediate. Now, two reaction sequences are

possible: formation of the symmetric product [Cp*ZnEt] via the intermediate #symmetric

(activation barrier: 15.9 kcal/mol) and the formation of the asymmetric product [Zn2Cp*2] via

28 2. Introduction

the intermediate #asymmetric (activation barrier: 64.2 kcal/mol). Although, the stability of the

asymmetric product is greater (-43.0 kcal/mol) than the symmetric product (-20.0 kcal/mol), it

seems to be very unlikely to overcome the activation barrier of 64.2 kcal/mol. Thus, the

results found for (1) clearly argue against the possibility of neutral charge electrostatic

attractions and rearrangement to form [Zn2Cp*2] and [Cp*ZnEt].

Figure 11. Reaction pathway of ZnEt2 reacting with decamethylzincocene (hydrogens removed for clarity). Zinc

atoms appear in red.[142] Reprinted with permission from S. S. Hepperle and Y. A. Wang, J. Phys. Chem. 2008,

112, 9619-9622. Copyright (2008) American Chemical Society.

In addition, the homolytic dissociation of the zincocene, pathway (2), requires 32.3 kcal/mol

which alone can not justify the formation of the dizincocene compared with the much lower

activation energy (15.9 kcal/mol) in the formation of [Cp*ZnEt] (Scheme 7).

Zn 32.3 kcal/molZn

Scheme 7. Dissociation of [Zn(η5-Cp*)(η1-Cp*)] into [Zn(η5-Cp*)]• and [Cp*]• radicals.

29 2. Introduction

As suggested by the authors, the most likely mechanism involves competing reaction

sequences between the formation of symmetric products and homolytic dissociation. Thus,

the partial reaction of ZnCp*2 with ZnEt2 via formation of the intermediate structure #symmetric results in an energy gain of ca. 20 kcal/mol. Since enough energy has been

generated, homolytic dissociation of the remaining zincocene takes place forming [Cp*Zn]•

radicals and [Cp*]• fragments. Finally, the combination of two [Cp*Zn]• radicals results in the

formation of [Zn2Cp*2] which consequently provides the large association energy of 43.0

kcal/mol to stimulate further homolytic dissociation and thus formation of more dizincocene.

Since the discovery of the dimeric Zn(I)-Zn(I) compound reported in 2004 a multitude of

articles have been published concerning theoretical studies.[142, 144-155] The main focus lay on

investigations into the electronic structure, characteristics of the bonding situation and

predictions, or rather, support of spectroscopic data, e.g. IR and Raman properties. The nature

of the Zn-Zn interaction consists mainly of 4s/4s orbital interactions between two ns singly

occupied molecular orbitals (Figure 12).[146]

Figure 12. Energy correlation diagram for the formation of [Zn2Cp2] from the fragment {ZnCp}.[146] Reproduced

by permission of Elsevier.

In addition, four quasi degenerated occupied orbitals have been found to be combinations of

the degenerated e1 orbitals of the Cp* ring with low participation of Zn p-orbitals (< 3%).

30 2. Introduction

Notably, the Zn-Zn bond is very strong (between 66-72 kcal/mol) as determined by energy

decomposition analysis (EDA) whereas the main contribution of the over-all energy comes

from electrostatic interactions (ca. 61 %) and less from orbital interactions (ca. 39 %).[141]

2.3.2 First Experimental Investigations on the Reactivity of Dizincocene

In addition to the wide variety of quantum chemical analysis which have been carried out

over the last years, the synthetic potential of the Cp* stabilized Zn(I) dimer has been the focus

of several research groups. Since stabilizing organic substituents R have a direct influence on

the properties of metal-metal interactions for example the bonding situation, the bond strength

as well as structural features, first reactivity studies aimed at the formation of derivative

structures [Zn2R2] (R ≠ Cp, Scheme 8) because it has been suggested by Carmona that ‘it also

seems plausible that the stabilization of the [Zn-Zn]2+ unit does not require the existence of

Zn-C bonds, which means that classical coordination compounds of the [Zn2]2+ central unit

are reasonable targets for future synthetic and structural studies’.[138]

Zn Zn

N P

PNNP

P N

Ph

Ph

Ph

Ph

Ph

Ph

Ph

Ph

PhPh

PhPh

Zn Zn

N

NN

N

R

R

R

R

Zn Zn

N

NN

N

2,6-iPr2C6H3

2,6-iPr2C6H3

2,6-iPr2C6H3

2,6-iPr2C6H3

R'

R'

R'

R'

Zn Zn

2,6-iPr2C6H3

2,6-iPr2C6H3

2,6-iPr2C6H3

2,6-iPr2C6H3

Zn Zn

N

NN

NZn Zn

N N

BN N

N NHB

N N

N N

N NH

R''R''

Scheme 8. Derivative structures containing the [Zn2]2+ core.

31 2. Introduction

Two common pathways have been found to be suitable, namely ligand exchange reaction of

the parent compound [Zn2Cp*2] with RH and KR (RH = MesnacnacH = [((2,4,6-

Me3C6H2)N(Me)C)2CH]H[156], [{(iPr)2ATI}H] = N-isopropyl-2-

(isopropylamino)troponimine[157], [{4-Br(iPr)2ATI}H] = 4-bromo-N-isopropyl-2-

(isopropylamino)troponimine[157], CH2(Ph2P=NPh)2 = bis(iminodi-

(phenyl)phosphorano)methane[158]; KR = TpMe2 = tris(3,5-

dimetyhylpyrazolyl)hydridoborate[159]) and secondly, reduction of [RZnX]a (R = Ar’ = C6H3-

2,6-(C6H3-2,6-iPr2)2[137, 160], monoanionic α-diimine = [(2,6-iPr2C6H3)NC(Me)]2

- [161],

Dippnacnac = [CH{(CMe)(2,6-iPr2C6H3N)}2][162], dpp-bian = 1,2-bis[(2,6-

diisopropylphenyl)imino]acenaphtalene[163]; X = halogenide; a = 1, 2).

However, comparatively little is known about the reactivity and chemical behaviour in

standard, inorganic reaction pathways. First investigations to understand the reactivity and the

chemical stability beside theoretical predictions have been made in 2009, about five years

after the discovery of the novel Zn-Zn bonded dimer (Scheme 9).

Zn

Zn

xs. dmapZn

Zndmap

dmap

2 [H(OEt2)2][Al{OC(CF3)3}4]

ZnN N

N

ZnN

NN

N

N

N

N

N

N

Zn

Zn

Zn

Zn NO

N

Mes

Mes

N

N Zn

O

Zn

O

Zn

Zn

N

N

N

N

1) 2 pyr-py2) ArMesOH

1) xs. pyr-py2) C5Me5OH

2+

Scheme 9. Above: Base-stabilised [Zn2]2+ cation. Below: Synthesis of [Zn2Cp*2(OArMes)(pyr-py)2] and

[Zn2Cp*2(µ-OC5Me5)(pyr-py)]2.

32 2. Introduction

Reaction with the strong Lewis base dmap (dmap = 4-dimethylaminopyridine) resulted in the

formation of the Lewis acid/base adduct [Zn2Cp*2(dmap)2] with a fully intact Zn-Zn bond.[164]

In addition, protolysis of [Zn2Cp*2(dmap)2] showed the formation of the first Lewis-base

stabilized cation [Zn2]2+.[165] Finally, treatment of [Zn2Cp*2] with aryl and alkyl alcohols in

the presence of a stabilising nitrogen donor leads to either [Zn2Cp*2(OArMes)(pyr-py)2] with a

fully intact Zn-Zn bond or [Zn2Cp*2(µ-OC5Me5)(pyr-py)]2 using the less bulky alkyl alcohol

C5Me5OH.[166, 167]

2.3.3 Curiosities in Silico: Fullerene-Dizincocene Hybrids and Multimetallocenes

There is interdependence between preparatively working chemists and theoretical chemists.

Chemists prepare unusual compounds in their laboratories, break down paradigms or just

discover any other novel molecule of interest. Thus, they need the theoreticians to gain an

understanding of the bonding situation and the reactivity. On the other hand, theoreticians

predict molecules by means of the multitude of programs at which they form compounds

which should be stable and just waiting on their discovery.[135, 168-173] Based on the

experimentally observed fullerene functionalisations leading to C60-FeCp hybrid

materials[174], Gao and Xu studied the possibility of dizincocene functionalised C60 and C70

fullerenes.[168] Their results obtained via quantum chemical calculations showed that

hypothetically, the existence of fullerene-dizincocene hybrid materials with a fully intact Zn-

Zn bond should be possible and thus, giving rise to a challenging target in the laboratory. The

preparation of [Zn2Cp*2], stimulated much interest and, as discussed previously, a wide

variety of articles have been published of theoreticians and coordination chemists dealing

with reactivitystudies, studies about bonding nature and other possible dimetallocenes of

group 12 metals, which have been prepared quite recently by Power et al. The groups of

Frenking and Merino took one step forward and posed the questions: ‘how many atoms Mn

can be sandwiched by two Cp* rings, yielding stable multimetallocenes, [Cp*2Mn]? And, is it

possible that multimetallocene compounds may become synthesized?’.[169] The authors

mentioned that the most important question for coordination chemists is, how stable such

compounds are with respect to loss of inner-trapped metal atoms. Their studies lead to the

conclusion, that multimetallocenes [Cp*MnCp*] with n > 2 are thermodynamically unstable

except for beryllium. Nevertheless, isn´t it a novel challenge to conquer such compounds and

decline these predicted results as Lappert did in 1976 and even Carmona in 2004?

[b] T. Bollermann, C. Gemel, and R. A. Fischer, Coord. Chem. Rev. 2011, submitted.

33 2. Introduction

2.4 Organozinc Ligands in Transition Metal Chemistry: A Brief

Overview[b]

Edward Frankland is recognized today as being one of the main initiators of organometallic

chemistry. Originating from his efforts to isolate ethyl radicals, he succeeded in the isolation

of ZnEt2 in 1849, one of the very first defined and well described organometallic

compounds.[175] Today, organozinc compounds are widely used as reagents in organic

synthesis. The chemical behaviour of ZnR2 is very similar to that of the corresponding

organomagnesium or organolithium compounds, however, the reactivity being significantly

lower due to relatively covalent Zn-C interactions resulting in a less Lewis acidic behaviour.

On one hand this decrease in reactivity is exceedingly desirable due to a high tolerance of

organozinc reagents towards functional groups, on the others their low reactivity to organic

substrates requests addition of catalysts (Ni or Pd complexes in Negishi coupling

reactions)[176-180] or copper reagents (Knochel cuprates)[181-184].

In contrast to the well established use of ZnR2 in organic synthesis, the formation of transition

metal-zinc bonds has been unexplored until the middle of the 20th century. In 1942, Hieber

synthesized the first M-Zn bond in Zn[Co(CO)4]2, which is accessible via the reaction of

CoX2 (X = Br, I) with Zn under CO pressure.[185] Several similar zinc containing carbonyl

complexes have been synthesized since then, very often involving zinc halides as terminal or

bridging ligands, e.g. [(CO)4Fe(ZnCl)2][186] or [C5H5(CO)3MoZnCl x O(C2H5)2]2[187]. In some

cases the electrophilic Zn centers are stabilized by N or O donor ligands, e.g.

[Fe(CO)4Zn(py)3][188], [Fe(CO)4{Zn(tmeda)Cl}2][189] or [CpMo(CO)3{Zn(thf)2Br}][190]. In the

early 1980s, the groups of Boersma and van der Kerk started to extensively and

systematically study the reactions of ZnCp2 with (acidic) transition metal hydrides,

substitution labile olefin complexes and bis(transition metal)zinc compounds of group 6-9

metals, which lead to a wide variety of thermally stable M-ZnR compounds. The selection of

ZnCp2 as the zinc source for the synthesis of M-ZnR compounds has two major advantages:

first, zincocene has a generally high reactivity towards transition metal complexes and

second, the remaining Cp ring stabilizes the electrophilic zinc center in M-ZnCp compounds

and thus suppresses the disproportionation.

34 2. Introduction

The first structurally characterized transition matal complex of ZnR was

[(C5H5Zn)2Co(C5H5)P(C6H5)3], which can be synthesized by the reaction of [HCo(N2)(PPh3)3]

with two equivalents of ZnCp2.[191] In the following years this reaction scheme has been

expanded to early transition metal hydride complexes [L2MH3] (M = Nb, L = C5H5 = Cp; M =

Ta, L = C5H4Me = Cp’).[192, 193] The exact course of the reaction with regard to hydride

abstraction as well as M-Zn bond formation directly depends on the ligand L, the

stoichiometry of the reactants as well as the solvent. The bis(M)zinc compound

(Cp2NbH2)2Zn for example can be obtained using a high stoichiometric amount of the

transition metal hydride, while increasing the amount of ZnCp2 leads to Nb-ZnCp complexes

with various degrees of substitution (Scheme 10).

Nb

H

Zn

H

Ta

Zn

Zn

H

Scheme 10. Left: [Cp2Nb(H)2(ZnCp)]. Right: [Cp’2Ta(H)(ZnCp)2].

Likewise, the treatment of [Cp’2TaH3] with ZnCp2 in a 1:1 ratio yields [Cp’2TaH2(ZnCp)]

which can be converted into the corresponding bis(transition metal)zinc compound

(Cp’2TaH2)2Zn by reaction with excess of [Cp’2TaH3]. Interestingly, the fully substituted

complex [Cp’2Ta(ZnCp)3] can be obtained only from the reaction of the Ta(III) complex

[Cp’2TaH(olefin)] (olefin = C2H4, C3H6) with ZnCp2, but not from [Cp’2TaH3]. The di-

substituted compound is obtained from [Cp’2TaH3] in the presence of excess ZnCp2.

[Cp’2TaH(ZnCp)2] exists in form of two isomers, i.e. the initally at low temperatures formed

kinetic isomer is isomerized to a thermodynamically more stable isomer at higher

temperatures.

Soon after, the reactivity of late transition metal complexes towards ZnR2 became a focus of

interest. [Ni(cod)2] proved to be highly reactive towards ZnCp2 leading to the unexpectedly

metal-rich compound [Ni2Zn4(C5H5)6] (Scheme 11).[143, 194-197]

35 2. Introduction

[Ni(cod)2] [Ni(cod)] + cod[ZnCp2] Ni

Zn

Ni

ZnZn

- Cp.Ni

ZnZn

Zn

Zn Zn

Zn

Ni

Ni

Step 1

Step 3Step 3

[ZnCp2]Step 2

Scheme 11. Reaction mechanism for the formation of [Ni2Zn4Cp6].

A hypothetical mechanism for the formation of [Ni2Zn4(C5H5)6] involves double oxidative

addition of ZnCp2 to Ni(0) and is followed by cleavage of a Cp radical, giving [CpNi(ZnCp)2]

as an intermediate, which finally dimerizes to the product [(NiCp)2(ZnCp)4]. The high

reactivity of the proposed intermediate [CpNi(ZnCp)2] is also reflected in the reaction of the

cod adduct of this intermediate, which dimerizes after cleavage of a [ZnCp]• radical to give

[CpNi(cod)]2. The same intermediate is proposed in the reaction of ZnCp2 with [Ni(cod)2]

(2:1 ratio) in the presence of one equivalent PPh3 leading to [µ-NiCp(PPh3)(µ-Cp)(ZnCp)2].

Soon after the studies of M-ZnR compounds, carried out by Boersma and van der Kerk,

transition metal-organozinc complexes also moved into the focus of application oriented

research. Thus, ZnR groups have been recognized as suitable ligands for stimulating the

reactivity of transition metal centres towards bond activation reactions. M/Zn clusters have

been also discussed as potential hydrogen storage materials. Quite a few transition metal-zinc

complexes bearing organozinc-alkyl ligands have been observed starting from electron-rich,

multi-metallic unsaturated polyhydride complexes of d8 and d9 transition metals. For instance,

the reaction of the rhodium dimers [{(L)Rh]2(µ-H)2}] (L = dippp = 1,3-

bis(diisopropylphosphino)propan, dippe = 1,2-bis(diisopropylphosphino)ethan) with

Zn(CH2Ph)2 yields the monomeric compound [(L)Rh(CH2Ph)] and the tetranuclear rhodium

complex [{(L)Rh}2(µ-H)2{µ-Zn(CH2Ph)}2] with bridging {Zn(CH2Ph)} units.[198, 199] The

same reaction with ZnCp2 leads to bridging ZnCp fragments in [{(dippp)Rh}2(µ-H)2{µ-

ZnCp}2]. The interaction of main group metal alkyl ligands with ruthenium polyhydride

36 2. Introduction

complexes has been studied by the group of Suzuki. In general, Suzuki's polynuclear

polyhydride complexes are active in bond activation reactions under mild conditions, e.g. C-

H, C-C, C=C and Si-H bond cleavage.[200-204]

Ru Ru

RuH H

H

HHZnR2- RH

(R = Me, Et)Ru Ru

RuH H

H

Zn

RH

Scheme 12. Formation of [{Cp*Ru}3(µ-H)3(µ3-ZnR)(µ3-H)].

The reactivity of such cluster compounds directly depends on steric and electronic properties

of the reaction sites. In order to vary the electron density at the transition metal centre and

therefore also the reactivity of the polyhydride clusters, implementation of lewis acidic main

group metals such as Zn has been studied.[205, 206] The reaction of [{Cp*Ru}3(µ-H)3(µ3-H)2]

with an equimolar amount of ZnR2 (R = Me, Et) leads to [{Cp*Ru}3(µ-H)3(µ3-ZnR)(µ3-H)]

under liberation of the corresponding alkane. Also the dinuclear polyhydride complex

[{Cp*Ru}2(µ-H)4] reacts with ZnR2 (R = Me, Et) in a similar manner leading to

[{Cp*Ru}2(µ-ZnR)(µ-H)3] (Scheme 12). The reaction has to be performed in a 1:1 molar ratio

of the reactants in order to obtain a product with on µ3-ZnR capping ligand of the Ru3 core.

Compounds with more than one ZnR caps show reduced reactivity in substrate activation

along the Ru3 plane due to steric reasons.

Quite recently the formation of new metal-rich and highly coordinated molecules [M(ZnR)n]

(n ≥ 8; M = Mo, Ru, Rh; Ni, Pd, Pt; R = Me, Et, Cp*) via treatment of all-gallium coordinated

transition metal complexes with excess of ZnR2 (R = Me, Et) has been described.[42, 43] This

new class of compounds exhibits an interesting bonding situation between metal clusters and

classical coordination compounds, and has been described as molecular models for Hume-

Rothery phases. The structural features directly follow the declarations based on the VSEPR

concept, typical polyhedral geometries are adopted in all cases, namely icosahedron, bicapped

square antiprism, capped square antiprism and dodecahedron (Figure 13).

37 2. Introduction

Figure 13. Molecular structures of [M(ZnR)n] showing the coordination polyhedra around the central metal

M.[43] Reprinted with permission from T. Cadenbach, T. Bollermann, C. Gemel, M. Tombul, I. Fernandez, M.

von Hopffgarten, G. Frenking and R. A. Fischer, J. Am. Chem. Soc. 2009, 131, 16063-16077. Copyright (2009)

American Chemical Society.

All these metal-rich compounds formally fulfill the 18 VE rule basically used in classical

coordination chemistry. Thus, the ZnR units should be viewed as 1e ligands, trapped by

transition metal centers resulting in rather strong covalent M-Zn bonds. Also M/Zn/Ga mixed

metal compounds are available, where GaR units act as 2e ligands. The formation of these

compounds involves redox chemical processes, i.e. reduction of Zn(II) to Zn(I) and oxidation

of Ga(I) to Ga(III). By-products which could be observed in NMR spectroscopic studies were

predominantly [Me2GaCp*] and [GaMe3]. However, the driving force of the reaction pathway

is most likely the oxidation of Ga(I) to Ga(III) in combination of electronic and steric

similarities between mono-valent GaR and ZnR fragments coordinating to the transition metal

centers.

A detailed description of the bonding situation has been studied by means of AIM analysis

(atoms-in-molecules), MO analysis and EDA analysis (energy decomposition analysis).[42, 43]

For instance, the frontier orbitals of the model compound [Mo(ZnH)12], which has been used

as a model for the prototype [Mo(ZnCp*)3(ZnMe)9], suggest an unusual combination of

metal-ligand interactions (Mo-Zn) and attractive interactions in the ligand shell (Zn-Zn). The

MO correlation diagram displays the HOMO-1 orbital (quintuply degenerate hg MO) as a

combination of the five valence 4d orbitals of Mo with the valence 4p orbitals of Zn and the

1s functions of H. The next lower-lying molecular orbital (ag) is produced by a combination

of the Mo 5s orbital with the 4p orbitals of the zinc, as well as the 1s function of H. The shape

of the molecular orbitals suggests that the six valence electrons of Mo are engaged in six

electron-sharing bonds with six valence electrons of the (ZnH)12 fragment. The remaining six

electrons which contribute no value to the Mo-Zn bonds, are delocalized in between the Zn12

cage (HOMO, triply degenerate t1u MO). These results have been complemented by EDA

38 2. Introduction

analysis as well as AIM analysis. The AIM analysis displays twelve critical bond paths

between molybdenum and zinc indicating twelve Mo-Zn bonding interactions. In contrast, no

bond paths between the zinc atoms of the (ZnH)12 cage can be observed. In addition, the EDA

analysis shows total orbital interactions between Mo (5s14d5) and (ZnH)12 in the electronic

septet state of 403.2 kcal/mol which come mainly from the hg orbitals (288.0 kcal/mol,

71.5%) and the ag orbitals (96.4 kcal/mol, 23.9 %) while the Zn-Zn bonding t1u orbitals

contribute only 18.3 kcal/mol (4.5%) to the orbital term. The interaction of the Mo atom with

the (ZnH)12 ligand shell may be best described using a valence bond model in terms of six sd5

hybridized orbitals of the molybdenum metal centre. For a detailed description and an

overview of the bonding situation in [M(ZnR)n] (n ≥ 8; M = Mo, Ru, Rh; Ni, Pd, Pt; R = Me,

Et, Cp*) see the original paper in references 42 and 43.

It should be noted that ternary systems [Mo(M’R)12] and [M(M’R)8] (M: Pd, Pt, Mo; M’: Zn,

Cd; R: Me = CH3, Cp* = pentamethylcyclopentadienyl) have also been reported.[44] For

instance, the reaction of the all-zinc coordinated compound [Pd(ZnMe)4(ZnCp*)4] with

CdMe2 has been afforded the Zn/Cd mixed compound [Pd(CdMe)4(ZnCp*)4]. The bonding

situation of these highly coordinated, metal-rich molecules is similar to those of the all-zinc

containing compounds showing radial interactions M-M’ in the icosahedral compounds which

are best described as classical electron pair sharing covalent bonds. In contrast, the

dodecahedral molecules exhibit mainly metal-ligand donor-acceptor bonds. Notably, it is not

necessary that the ligand metal atoms M’ are of the same type, as long as they donate enough

electrons to fulfil the 18 valence electron rule.

There are two known examples, for which heteroleptic complexes of the type [LaM(GaCp*)b]

(L = inert co-ligand, e.g. CO, Cp*) have been used in the formation of zinc-rich compounds.

Treatment of [Mo(CO)4(GaCp*)2] with ZnMe2 results in the formation of

[{Mo(CO)4}4(Zn)6(µ-ZnCp*)4].[207] The structural feature of [{Mo(CO)4}4(Zn)6(µ-ZnCp*)4]

can be best described as a super-tetrahedron created by four {Mo(CO)4} units. In addition, the

edges of these tetrahedrons are bridged with zinc atoms almost linearly and symmetrically. As

a consequence, the bridging zinc atoms display a distorted Zn6 octahedron. The observed

distortion from an ideal octahedron is a result of additional four ZnCp* ligands bridging four

of the twelve Mo-Zn contacts, which can be viewed as ‘half edges’ of the Mo4 tetrahedron

(Figure 14).

39 2. Introduction

Figure 14. Molecular structure of [{Mo(CO)4}4(Zn)6(µ-ZnCp*)4].[207] Reproduced by permission of John Wiley

and Sons.

The second example of using a heteroleptic GaCp* containing transition metal compound is

the formation of [{Cp*Rh(ZnCp*)2(ZnMe)(ZnCl)}2] and [Cp*2Rh][(Cp*Rh)6Zn18Cl12(µ6-Cl)]

in the reaction of [Cp*Rh(GaCp*)2(GaCl2Cp*)] with ZnMe2.[208] The product formation can

be easily controlled by modification of the reaction conditions with respect to stoichiometry

and temperature (Scheme 13).

Rh

*CpGaGaCp*

GaCl2Cp*

n-hexane, rtRh

ZnCp*

Zn

MeZn

*CpZnCl

ClRh

*CpZn

Zn

ZnMe

ZnCp*

toluene, 60 °C

[Cp*2Rh][(Cp*Rh)6Zn18Cl12(µ6-Cl)]+

trace amounts of [{Cp*Rh(ZnCp*)2(ZnMe)(Zn-µ2-Cl)}2]

4 ZnMe2

9 ZnMe2

Scheme 13. Formation of [{Cp*Rh(ZnCp*)2(ZnMe)(ZnCl)}2] and [Cp*2Rh][(Cp*Rh)6Zn18Cl12(µ6-Cl)].[208]

Reproduced by permission of The Royal Society of Chemistry.

[c] T. Bollermann, T. Cadenbach, C. Gemel, K. Freitag, M. Molon, V. Gwildies, and R. A. Fischer, Inorg. Chem. 2011, 50, 5808-5814. M. Molon, T. Bollermann, C. Gemel, J. Schaumann, and R. A. Fischer, Dalton Trans. 2011, DOI:10.1039/C1DT10583C.

40 3. Results and Discussion

3 Results and Discussion


Molybdenum and Rhodium GaR (R = Cp*, DDP) Containing

Complexes[c]

Abstract

The reaction of Mo(0) and Rh(I) starting materials with the carbenoid group 13 metal ligand

GaR (R = Cp*, DDP; Cp* = pentamethylcyclopentadienyl, DDP = HC(CMeNC6H3-2,6-iPr2)2)

are investigated. The treatment of [Mo(η4-butadiene)3] with GaCp* under hydrogen

atmosphere yields the homoleptic, hexa coordinated and steric crowded complex

[Mo(GaCp*)6] (1). In comparison, reaction of [Mo(N2)(PMe3)5] with excess GaCp* under

moderate conditions yields the heteroleptic complex [cis-Mo(GaCp*)2(PMe3)4] (2) as the

phosphane containing congener to the corresponding carbon monoxide complex. In addition,

[Rh(GaCp*)5][CF3SO3] (3) is prepared by the reaction of GaCp* with the Rh(I) compound

[Rh(coe)2(CF3SO3)]2 (coe = cyclooctene). Subsequent anion exchange results

[Rh(GaCp*)5][BArF] (4) (BArF = B{C6H3(CF3)2}4). The reaction of excess Ga(DDP) with

[Rh(coe)2(CF3SO3)]2 does not result in a high coordinated homoleptic complex but instead

yields [(coe)(toluene)Rh{Ga(DDP)}(CF3SO3)] (5). The common feature of 3 and 5 in the

solid state structure is the presence of short CF3SO2O⋅⋅⋅Ga contacts involving the coordinated

Ga(I).

Introduction

In the coordination chemistry of low valent group 13 metals, special interest has been given to

the Cp* system. The versatile structural and electronic abilities are mainly reflected in

flexible, more or less soft properties in the stabilization of the low valent group 13 metal

centre, resulting in possible haptotropic shifts (η5→η3→η1), as well as selective cleavage of

the Cp* group under mild conditions.[40, 209, 210]


The most important synthetic procedures for the preparation of transition metal E(I)R

complexes are ligand substitutions at [LnM] complexes (L = e.g. olefins, CO, phosphanes,

acetonitrile).[17, 39, 116, 126, 128] The ligand system used directly influences the reaction pathway

and thus the coordination of E(I)R to the transition metal centre. For instance, the substitution

of CO in starting materials is limited by the coordination of GaCp*, since the strong σ-donor

abilities of GaCp* increase the electron density of the transition metal upon coordination and

thus lead to stronger π-back bonding of the remaining CO ligands.[116, 126, 127] Due to this fact,

fully homoleptic GaCp* containing complexes could only be observed by substitution of

labile bonded, weak π-acceptor ligands. The variety of heteroleptic monomeric and

oligomeric complexes [Ma(GaCp*)bLc] found in the literature is somehow restricted in

comparison to their homoleptic opponent. The result of which is that GaCp* compounds of

the type [M(GaCp*)n] show inert properties against substitution reactions, transition metal

complexes containing the desired co-ligand L with Ga(I)R seems to be the most suitable

method to obtain heteroleptic compounds. It could be shown, that E(I)Cp* containing

transition metal complexes are suitable single-source precursors in the formation of M/E

Hume-Rothery phases.[40] Thus it is advantageous to be able to control the reaction pathways

as well as the product formation, in order to be able to deliberately prepare intermetallics with

certain M/E ratios. Descriptive examples of how to control the reaction schemes by means of

ligand choice in the starting materials are presented in the formation of [Mo(GaCp*)6] (1) and

[cis-Mo(GaCp*)2(PMe3)4] (2). While the formation of [Rh(GaCp*)5][X] (X = CF3SO3 (3),

BArF (4)) and [(coe)(toluene){Ga(DDP)}(CF3SO3)] (5) illustrates the dependency of product

formation from the used low valent Ga(I) species GaCp* and Ga(DDP).

Synthesis and Characterisation of [Mo(GaCp*)6] (1) and [cis-Mo(GaCp*)2(PMe3)4] (2)

Treatment of [Mo(η4-butadiene)3] with excess GaCp* in toluene under 3 bar hydrogen

atmosphere at 100°C yields [Mo(GaCp*)6] (1) in yields approx. 50% with n-butane as the by-

product (Scheme 14).[211] The 1H NMR spectrum of pure 1 in C6D6 at room temperature

shows only one sharp signal at 1.96 ppm for six equivalent GaCp* ligands. The Cp* groups

show fluctional behaviour around the [MoGa6] core down to -78°C as determinded by low

temperature 1H NMR measurements. The 13C NMR spectrum reveals no unusual features

with respect to the expected set of signals for the Cp* units. Liquid injection field desorption

ionization mass spectrometry (LIFDI-MS) allowed the identification of the molecular ion

peak [M].+ of 1 at m/z = 1326 as well as one fragment signal at m/z = 1190 for [M-Cp*]+.


[Mo(C4H6)3] Mo*CpGa

*CpGa GaCp*

GaCp*

GaCp*

GaCp*

8 GaCp*, 3 bar H2, toluene, 100°C

- C4H10

(1)

(2)[Mo(N2)(PMe3)5] GaCp*toluene, 65°C

Mo*CpGa

Me3P PMe3

GaCp*

PMe3

PMe3

Scheme 14. Synthesis of [Mo(GaCp*)6] (1) and [cis-Mo(GaCp*)2(PMe3)4] (2).

In comparison, the treatment of [Mo(N2)(PMe3)5] with two equivalents GaCp* in toluene at

60°C leads to the formation of [cis-Mo(GaCp*)2(PMe3)4] (2) in yields around 60 % (Scheme

14). Herein, the substitution of co-ligands is restricted to N2 and one PMe3 ligand. The

reaction conditions, i.e. stoichiometry, temperature and reaction time, do not have any effect

on ligand replacement in the starting material. The 1H NMR spectrum of 2 in C6D6 shows one

signal for the Cp* groups at 2.09 ppm as well as two pseudo-triplets for the PMe3 ligands at

1.28 and 1.35 ppm. The existence of two sets of equivalent phosphane ligands as determined

by 1H NMR spectroscopy indicates a cis configuration in the molecular structure of 2. This

has been affirmed by single crystal X-ray diffraction. The 13C NMR and 31P NMR spectra of

2 display the expected signal pattern. Compounds 1 and 2 dissolve sufficiently well in

aromatic solvents such as benzene, toluene or mesitylene and are stable for several weeks

when stored at -30°C under an inert gas atmosphere in the glove box.

Single Crystal X-Ray Analysis of [Mo(GaCp*)6] (1)

Single crystals of 1 which are suitable for X-ray diffraction studies can be obtained by heating

up a saturated mesitylene solution and storing it at -30°C. Compound 1 crystallizes in the

triclinic space group P-1 (Figure 15). The molybdenum centre of the metal core [MoGa6] is

surrounded by six GaCp* ligands in an almost perfect octahedral fashion as determined by the

continuous shape measure SQ(P) = 0.195.[212, 213] One of the axial GaCp* ligands is slightly

bent towards the equatorial plane (Ga4-Mo1-Ga3 85.12(2)°). The angles between cis-ligands

range from 85.12(2)° to 94.23(2)° (Ga6-Mo1-Ga2), while the trans-ligand angles vary

between 174.35(3)° for Ga4-Mo1-Ga6 and 176.39(3)° for Ga3-Mo1-Ga2. The Mo-Ga bond


distances lie in between 2.3844(6) (Mo1-Ga4) and 2.4930(7) Å (Mo1-Ga2) and have

significantly shorter values compared to [fac-(GaCp*)3Mo(CO)3][127] (2.5228(8) and

2.5188(8) Å) and [cis-Mo(GaCp*)2(CO)4][126] (Mo-Ga 2.554(1) and 2.537(1) Å). The Cp*

groups in compound 1 are asymmetrically coordinated to the gallium centers showing

haptotropic shifts which results in coordination modes from η1 (Ga4) to η5 (Ga1, Ga2, Ga3,

Ga5, Ga6). This situation is likely to be a consequence of the steric crowding situation in the

molecule. The average η5-Cp*centroid-Ga distance (2.069 Å) is somewhat shortened compared

to the free ligand (2.081 Å) (monomer in the gas phase)[214].

Figure 15. Molecular structure of 1 in the solid state as determined by single crystal X-ray diffraction;

displacement ellipsoids (Mo, Ga) are shown at 50% probability level, hydrogen atoms are omitted for clarity.

Selected interatomic distances (Å) and angles (°): Mo1-Ga4 2.384(1), Mo1-Ga1 2.458(1), Mo1-Ga5 2.474(1),

Mo1-Ga3 2.476(1), Mo1-Ga6 2.485(1), Mo1-Ga2 2.493(1), Ga4-C31 2.088(4), Ga1-Cp*centroid 2.127, Ga2-

Cp*centroid 2.091, Ga3-Cp*centroid 2.040, Ga5-Cp*centroid 2.057, Ga6-Cp*centroid 2.024, Ga4-Mo1-Ga1 87.92(2), Ga4-

Mo1-Ga5 88.77(2), Ga4-Mo1-Ga3 85.12(2), Ga1-Mo1-Ga5 175.22(3), Ga4-Mo1-Ga6 174.35(3), Ga3-Mo1-Ga2

176.39(3), Ga1-Mo1-Ga3 90.39(2), Ga5-Mo1-Ga3 92.77(2), Ga1-Mo1-Ga6 91.15(2), Ga5-Mo1-Ga6 92.49(2),

C31-Ga4-Mo1 168.01(12), Ga3-Mo1-Ga6 89.31(2), Ga4-Mo1-Ga2 91.32(2), Mo-Ga1-Cp*centroid 175.21, Mo-

Ga2-Cp*centroid 167.42, Mo-Ga3-Cp*centroid 170.30, Mo-Ga5-Cp*centroid 167.36, Mo-Ga6-Cp*centroid 166.04.


Single Crystal X-Ray Analysis of [cis-Mo(GaCp*)2(PMe3)4] (2)

Orange crystals of 2 suitable for single crystal X-Ray diffraction were obtained from a

saturated toluene solution at -30 °C. Complex 2 crystallizes in the orthorhombic space group

P212121 (Figure 16).


displacement ellipsoids (Mo, Ga) are shown at 50% probability level, hydrogen atoms are omitted for clarity.

Selected interatomic distances (Å) and angles (°): Mo1-Ga1 2.4810(4), Mo1-Ga2 2.4833(4), Mo1-P1 2.4252(8),

Mo1-P2 2.4232(8), Mo1-P3 2.4435(10), Mo1-P4 2.4420(10), Ga1-Cp*centroid 2.114, Ga2- Cp*centroid 2.139, Mo1-

Ga1- Cp*centroid 177.11, Mo1-Ga2- Cp*centroid 171.49, P2-Mo-P1 92.97(3), P2-Mo1-P4 94.89(4), P1-Mo1-P4

88.85(5), P2-Mo1-P3 94.07(4), P1-Mo1-P3 89.02(5), P4-Mo1-P3 170.89(3), P2-Mo1-Ga1 70 172.35(3), P1-

Mo1-Ga1 94.66(2), P4-Mo1-Ga1 85.75(3), P3-Mo1-Ga1 85.60(3), P2-Mo1-Ga2 85.15(3), P1-Mo1-Ga2

178.10(2), P4-Mo1-Ga2 91.00(4), P3-Mo1-Ga2 91.42(4), Ga1-Mo1-Ga2 87.21(1).

Similar to 1, four PMe3 and two GaCp* ligands are coordinated in an almost perfectly

octahedral fashion at the molybdenum centre (continuous shape measure SQ(P) = 0.38).[212, 213]

The angles between cis-ligands range from 85.13(3)° to 94.89(4)°, while the trans-ligand

angles are all between 170.89(3)° and 178.10(2)°. The Mo-Ga bond distances lie between


2.481(1) and 2.483(1) Å and are significantly shorter in comparison to 1 and the heteroleptic

CO containing complexes.[126, 127] The Mo-P distance (av. 2.433 Å) is comparable to the fully

PMe3 substituted compound [Mo(PMe3)6][215, 216] (2.467(2) Å) and within the range of other

phosphane containing complexes of molybdenum found in the literature. Most interestingly,

the equatorial Mo-P bond distances (2.425(1) and 2.423(1) Å) are slightly shortened with

respect to the axial Mo-P distances (2.444(1) and 2.442(1) Å) presumably indicating

Mo→PMe3 back bonding interactions as it has been already described for [cis-

Mo(GaCp*)2(CO)4].[126] Herein, the equatorial Mo-CO distances are significantly shorter with

respect to the axial Mo-CO distances, indicating stronger π-back bonding in CO ligands with

trans-GaCp* units. Due to less steric hindrance in comparison to 1, the Cp* rings in 2 are

almost symmetrically η5-bonded to the gallium atoms displaying small deviations from

linearity for the Mo-Ga-Cp*centroid angles (av. 174°). The Ga-Cp*centroid distances (av. 2.127 Å)

are significantly elongated in comparison to 1 (2.069 Å) and with respect to [fac-

(GaCp*)3Mo(CO)3][127] (1.947 Å) as well as [cis-Mo(GaCp*)2(CO)4][126] (1.930 Å).

Synthesis and Characterisation of [Rh(GaCp*)5][CF3SO3] (3)

Treatment of [Rh(coe)2(CF3SO3)]2 with ten equivalents GaCp* in fluorobenzene at 60°C

results in the formation of [Rh(GaCp*)5][CF3SO3] (3) in good yields of about 80 % (Scheme

15). Compound 3 dissolves well in polar solvents such as fluorobenzene, dichloromethane or

thf and is stable for several weeks when stored under an inert gas atmosphere at -30°C. The 1H NMR spectrum of 3 in CD2Cl2 shows one signal at 1.91 ppm for the equivalent Cp* units.

The fluxional process, which is also present at -78°C and is quite similar to 1, underlies the

fast exchange of the {CF3SO3} group between the Ga centers as well as Cp* exchange

between equatorial and axial positions within the [RhGa5] core. The 13C NMR and 19F NMR

spectrum indicates no unusual features with respect to the expected signal pattern.

Interestingly, LIFDI mass spectrometric measurements show no molecular ion peak of 3.

Instead of that, one peak for the cationic species [Rh(GaCp*)3(Ga)]+ at m/z = 788 occurs. This

observation agrees with the fluctuating behavior of 3 and in addition, the known ligand

properties of naked Ga+ together with the leaving group properties of the neutral Ga(III)

species [(Cp*)2Ga(CF3SO3)]. Furthermore, the rhodocenium cation [RhCp*2]+ (m/z = 373)

was detected in the mass spectrum as a characteristic fragmentation and rearrangement

product of 3. IR spectroscopy can be used to distinguish between (contact) ion pair and

covalent bonding modes of the triflate anion acting as a weak nucleophile. For mainly ionic


existing {CF3SO3} a band near 1280 cm-1 is characteristic and with increasing covalent

character this band is shifted to higher wave numbers around 1380 cm-1.[217] In the case of

compound 3, absorptions have occurred at 1219 cm-1 (strong signal) as well as 1295 and 1373

cm-1 (weaker absorptions). These results could confirm a more ionic than covalent character

in compound 3. It should be noted that the approach of these results are often ambiguous and

should be interpreted with care due to ‘the mixing of CF3 and SO3 vibrational modes and

accidental coincidences of these modes arising particularly in the stretching region’.[217]

GaCp*

C6H5F, 60°C

Rh RhO OS

O CF3

OO SOF3C

*CpGa RhGaCp*

GaCp*

GaCp*

GaCp*

F3CSO2O

RhGaN

NOSO2CF3

(3) (5)

CH2Cl2, rt

*CpGa RhGaCp*

GaCp*

GaCp*

GaCp*

NaBArF

{BArF}-

Ga(DDP) C6H5F, 50°C,Re-crystallisation toluene

(4)

Scheme 15. Synthesis of [Rh(GaCp*)5][X] (X = CF3SO3 (3), BArF (4)) and

[(coe)(toluene)Rh{Ga(DDP)}(CF3SO3)] (5).

Single Crystal X-Ray Analysis of [Rh(GaCp*)5][CF3SO3] (3)

Deep red needle-shape crystals of compound 3 which are suitable for single crystal X-ray

measurements were obtained by slow diffusion of n-hexane into a saturated fluorobenzene

solution at room temperature. Compound 3 crystallizes in the triclinic space group P-1 (Figure

17).



displacement ellipsoids (Rh, Ga) are shown at 50% probability level, hydrogen atoms are omitted for clarity.

Selected interatomic distances (Å) and angles (°): Rh1-Ga1 2.3326(4), Rh1-Ga2 2.3393(4), Rh1-Ga3 2.3799(4),

Rh1-Ga4 2.3800(4), Rh1-Ga5 2.3542(4), Ga2-Cp*centroid 1.963, Ga3-Cp*centroid 1.996, Ga4-Cp*centroid 2.039, Ga5-

Cp*centroid 1.964, Ga1-C5 2.055(3), Ga1-O1 2.079(2), Rh1-Ga2-Cp*centroid 171.87, Rh1-Ga3-Cp*centroid 166.79,

Rh1-Ga4-Cp*centroid 161.73, Rh1-Ga5-Cp*centroid 156.29, O1-Ga1-Rh1 115.94(7), O1-Ga1-C5 92.74(12), C5-Ga1-

Rh1 148.82(10), Ga5-Rh1-Ga2 154.297(18), Ga4-Rh1-Ga3 106.604(16), Ga2-Rh1-Ga3 89.454(16), Ga5-Rh1-

Ga4 100.423(16), Ga1-Rh1-Ga3 143.643(17), Ga1-Rh1-Ga4 109.655(16), Ga1-Rh1-Ga5 83.584(15), Ga1-Rh1-

Ga2 79.725(15).

Notably, the coordination geometry around the rhodium centre of the metal core [RhGa5]

displays major distortions from a regular trigonal bi-pyramid and a regular square pyramid,

the two geometries most often observed for penta coordinated metal centres. This results in

very high continuous shape measures of SQ(P) = 3.76 and 6.01, respectively.[213, 218, 219] Thus,

the geometrical arrangement of the atoms in 3 can be best described as a square pyramid with

an apical angle of β = 105° [220] displaying a comparably low continuous shape measure of

SQ(P) = 0.55 (Figure 18). It becomes obvious that the distortion is caused by steric reasons,

maybe due to coordination of one triflate molecule to one gallium atom in the ligand sphere.

In general, all characteristic angles in compound 3 show strong deviations from an ideal

trigonal bi-pyramid (Table 5).


Figure 18. Superimposition of the [RhGa5] metal core (grey) and ideal polyhedra (black). Left: trigonal bi-

pyramid, Middle: square pyramid, Right: square pyramid (β = 105°).

Table 5. Comparison between ideal and experimental angles (°) for [Rh(GaCp*)5][CF3SO3] (3).

Axial Ligands Equatorial Ligands Axial and Equatorial Ligands

Ideal 180 120 90

3 154.297(18) 106.604(16)°- 143.643(17) 79.725(15)°- 100.423(16)

It should be noted that the known complexes [Rh(GaCp*)4(GaCH3Cp*)] and

[Rh(GaCp*)4(GaCH3)][BArF] which exhibit quite similar coordination geometries display

distinctly weaker distortions than compound 3.[209] The average Rh-Ga distance (av. 2.3572

Å) is significantly longer than the average Rh-Ga distance in

[Rh(GaCp*)4(GaCH3)][BArF][209] (av. 2.2957 Å) but comparable to several known Rh-GaCp*

compounds such as [Cp*Rh(GaCp*)(CH3)2][221] (2.3292(15) Å) and [{Rh(η 2,η2-

nbd)(PCy3)(GaCp*)2}{BArF}][222] (av. 2.408 Å, nbd = 2,5-norbornadiene). The Cp* ligands

on the Ga atoms Ga2-Ga5 are symmetrically η5 coordinated (av. Ga-Cp*centroid distance 1.991

Å), showing significantly shorter values in comparison with free GaCp* (2.081(5) Å,

monomer in the gas phase)[214] and 1 (av. 2.069 Å) as a result of the enhanced electrophilic

character of the coordinated Ga centers. However, η1 coordination can be observed for the

Cp* group coordinated to Ga1 (Ga1-C5 bond distance 2.055(3) Å), most reasonable as a

result of steric over-crowding at the Rh centre and due to weak donor-acceptor interactions

between Ga1 and the oxygen atom of the {CF3SO3} anion. The Ga1-O1 distance (2.079(2) Å)

matches well with other Ga-O distances in {RGa-OSO2CF3} (R = organic substituent)

fragments (Table 6).


Table 6. Comparison of Ga-O bond distances (Å) between [Rh(GaCp*)5][CF3SO3] (3) and selected reference

compounds.

Compound Ga-O

3 2.079(2)

[Ga(thf)4H(CF3SO3)2][Ga(thf)2(CF3SO3)4][223] 1.947(9)-1.988(9)

[{2,6-(Me2NCH2)2C6H3}Ga(CF3SO3)2H][CF3SO3][224] 1.939(8)-1.962(9)

[{(CF3SO3)Bi(GaDDP)}2][225] 2.027(1)

[(TPP)Ga(CF3SO3) x C7H8][226] 1.963

Thus, to avoid interactions between the Lewis acidic Ga(I) center and the counter ion, salt

metatheses reaction with NaBArF has been done to obtain the corresponding compound

[Rh(GaCp*)5][BArF]. Note, that coordination of the [BArF] anion at Lewis acidic gallium

centers has not been observed in the compound [Rh(GaCp*)4(GaCH3)][BArF][209] and even

related cationic species, because of the much weaker nucleophilic character in comparison to

{CF3SO3}. The reaction of compound 3 with a slight excess NaBArF in CH2Cl2 at room

temperature results in the formation of [Rh(GaCp*)5][BArF] (4) as a microcrystalline orange

powder which is stable under inert gas atmosphere at -30°C for several weeks (Scheme 2).

Notably, dissolution of pure microcrystalline 4 in common polar solvents such as CH2Cl2,

fluorobenzene and thf leads to slow decomposition within two hours. 1H NMR spectroscopic

measurements show the elimination of Cp*H and a series of not further assigned broad peaks

in the Cp* area. Most surprisingly, decomposition of 4 takes place even at low temperatures

around -30°C, whereat the initial deep red solution slowly turns to a yellow oily solution

under precipitation of some metallic deposit which makes crystallization of pure 4 not

possible. Nevertheless, successful anion exchange reaction has been proven due to quick

preparation of fresh solutions of 4 in CD2Cl2 and immediate NMR measurements. Herein,

characteristic signals of the GaCp* ligands and the [BArF] anion have been observed (see

Experimental Section for detailed NMR spectroscopic data). Unfortunately, the LIFDI mass

spectrum did neither display the molecular ion peak [M].+ nor the characteristic fragment

[Rh(GaCp*)3(Ga)]+ as it was seen for 3, but rather the rhodocenium cation [RhCp*2]+ (m/z =

373) as the dominant peak. Additionally, the FTIR spectrum displays typical absorptions for


the Cp* unit at 2956, 2895 and 2836 cm-1 as well as one strong band at 1264 cm-1 for the C-F

vibration.

Synthesis and Characterisation of [(coe)(toluene)Rh{Ga(DDP)}(CF3SO3)] (5)

As has already been mentioned in the introduction, reaction schemes and thus product

formation can be easily controlled by the use of suitable Ga(I)R species. The differences in

the reaction behaviour between GaCp* and the more rigid Ga(DDP) can be illustrated on the

basis of the Rh(I) precursor used in the preparation of compound 3. The reaction of

[Rh(coe)2(CF3SO3)]2 with two equivalents Ga(DDP) in fluorobenzene at 50 °C results in the

formation of a yellow crystalline material. Subsequent re-crystallisation from a saturated

toluene solution leads to the formation of the mono gallylene complex

[(coe)(toluene)Rh{Ga(DDP)}(CF3SO3)] (5) (Scheme 2) in yields around ≥ 95 %. The

formation of 5 involves three steps: replacement of one cyclooctene ligand accompanied by

an insertion reaction of gallium into the Rh-O bond as well as coordination of one toluene

molecule during the re-crystallization process. A similar reaction pathway has been reported

in 2006.[227] Herein, reaction of the halide containing starting material [Rh(coe)2Cl]2 with

Ga(DDP) leads to the formation of [(coe)(benzene)Rh{Ga(DDP)}(Cl)]. The NMR

spectroscopic data of 5 show the expected signal pattern and shifts (see Experimental Section

for detailed NMR spectroscopic data).

Single Crystal X-Ray Analysis of [(coe)(toluene)Rh{Ga(DDP)}(CF3SO3)] (5)

Yellow crystals of 5 suitable for single crystal X-ray diffraction have been obtained from a

saturated toluene solution at -30°C. Complex 5 crystallizes in the monoclinic space group

P21/n (Figure 19). The coordination geometry around the Rh centre can be best described as a

piano-stool configuration. The N1-N2-Ga1-Rh1 plane is nearly trigonal planar with an

angular sum of 352.81°. The Rh1-Ga1 bond distance (2.3696(6) Å) is comparable with Rh-Ga

distances in several other complexes.[38, 222, 227] The Ga1-O1 distance (2.113(3) Å) is slightly

longer than in compound 3 but lies well in the range of gallium-oxygen distances found in

{RGa-OSO2CF3} fragments (R = organic substituent) as suggested for 3 (vide supra, Table

6).



displacement ellipsoids (Rh, Ga) are shown at 50% probability level, hydrogen atoms are omitted for clarity.

Selected interatomic distances (Å) and angles (°): Rh1-Ga1 2.3696(6), Ga1-O1 2.113(3), Ga1-N1 1.957(4), Ga1-

N2 1.999(3), Rh1-C9 2.125(4), Rh1-C16 2.136(4), C9-C16 1.409(6), N1-Ga1-N2 94.87(14), N1-Ga1-Rh1

133.39(11), N2-Ga1-Rh1 124.55(10), O1-Ga1-Rh1 107.00(8), N1-Ga1-O1 95.28(14), N2-Ga1-O1 90.37(13).

[d] T. Bollermann, G. Prabusankar, C. Gemel, R. W. Seidel, M. Winter, and R. A. Fischer, Chem.-Eur. J. 2010, 16(29), 8846-8853.


3.2 First Dinuclear Copper/Gallium Complexes: Supporting Cu(0) and

Cu(I) Centres by Low Valent Organogallium Ligands[d]

Abstract

The synthesis and structural characterisation of Ga(I)R containing dinuclear Cu(I) and Cu(0)

complexes are investigated. The reaction schemes include reductive coordination reactions of

GaCp* (Cp* = pentamethylcyclopentadienyl) and Ga(DDP) (DDP = HC(CMeNC6H3-2,6-

iPr2)2, 2-Di-iso-propylphenylamino-4-Di-iso-propyl-phenylimino-2-Pentene) with Cu(II) and

Cu(I) starting materials. The treatment of [Cu(CF3SO3)2] with Ga(DDP) at mild conditions

yields the novel Ga(I)/Cu(I) compound [{(DDP)GaCu(CF3SO3)}2] (6). The single crystal X-

ray structure determination of 6 reveals [(DDP)GaCu(CF3SO3)] dimeric units with a planar

Cu(I)-Ga(I) four membered ring and the shortest Cu(I)••••Cu(I) distance known so far. In

comparison, treatment of [Cu(CF3SO3)2] with GaCp* yields in the formation of the all-

gallium coordinated dinuclear compound [Cu2(GaCp*)(µ-GaCp*)3{Ga(CF3SO3)3}] (7). The

Lewis acidic {Ga(CF3SO3)3} ligand is formed in the course of this redox reaction which

coordinates to one of the electron rich Cu(0) centres. Thus, compound 7 is suggested as the

first case of a structurally characterized Cu(0) complex. Finally, changing the copper source

from Cu(II) to Cu(I), i.e. [{Cu(cod)2}(CF3SO3)] (cod = 1,5-cyclooctadiene), the copper dimer

[Cu2(GaCp*)3(µ-GaCp*)2][CF3SO3]2 (8) is formed, the cationic part of which is related to

previously described isoelectronic dinuclear d10 complexes of the general type [M2(GaCp*)5]

(M = Pd, Pt).

Introduction

In the recent past it has been shown, that low valent group 13 ligand compounds play an

important role in the linkage between coordination and cluster chemistry of metal-rich

molecules and material science. For instance, preparation of zinc-rich compounds such as

[Mo(ZnCp*)3(ZnMe)9][42, 43] and [{(CO)4Mo}4(Zn)6(ZnCp*)4][207] could be obtained from

homoleptic and heteroleptic GaCp* containing starting materials. Most interestingly, the

structural features of these compounds show similarities to structural elements of the Mo/Zn

Hume-Rothery phase MoZn20.44.


In addition, the reaction of all-hydrocarbon precursors of the type [LnM] with E(I)Cp* (E =

Al, Ga) or rather the usage of tailored single source precursors which already consists of the

desired M-E bonds lead to M/E Hume-Rothery phases such as NiAl, NiGa, PtGa, CuAl and

CuGa, as colloidal nanoparticles or as powder material via soft chemical synthetic

pathways.[33-36, 133, 228-230] Obviously, it is of interest to understand and further investigate the

coordination chemistry of low valent group 13 species towards transition metals in order to

gain further insights into the organometallic precursor chemistry for intermetallic systems.

For instance, although Cu/Ga systems are well present in intermetallic chemistry, little is

known about the molecular chemistry of Cu/Ga systems in general and oligonuclear CuaGab

compounds in particular. Before the completion of this work, only three examples exhibiting

terminal Cu-Ga bond interactions had been reported in the literature.[38, 39] Herein, two

different reaction pathways have been reported. Firstly, salt elimination reaction of anionic

Ga(I) heterocycles with suitable copper complexes[38], and secondly, the reaction of a cationic

copper compound which contains weakly bonded acetonitrile ligands with GaCp*.[39] This

chapter points out first investigations in the formation of oligonuclear copper complexes

stabilized by low valent Ga(I) species.

Synthesis and Characterisation of [{(DDP)GaCu(CF3SO3)}2] (6)

The stabilizing effect and simultaneously reducing ability of Ga(DDP) has been used in

several reactions in the recent past. The latest example has been the stabilization of a very

short Bi=Bi bond in [{(CF3SO3)Bi(GaDDP)}2].[225] Accordingly, first reactions to obtain

oligonuclear Cu/Ga compounds have been tried using Ga(DDP). The mild reduction of the

Cu(II) compound [Cu(CF3SO3)2] with two equivalents of Ga(DDP) in fluorobenzene at 60°C

results in the formation of the Cu(I) dimer [{(DDP)GaCu(CF3SO3)}2] (6) through the

reductive elimination of [(CF3SO3)2Ga(DDP)] in yields around 60% (Scheme 16).

GaN

N

Ga N

NCu

Cu

OSO2CF3

OSO2CF3

[Cu(CF3SO3)2] (6)Ga(DDP)

C6H5F, 60°C

Scheme 16. Synthesis of [{(DDP)GaCu(CF3SO3)}2] (6).


The pure crystalline solid of compound 6 is stable under inert atmosphere for several days but

first signs of decomposition can be observed after two weeks. Compound 6 decomposes

immediately when it is re-dissolved in polar coordinating or non-coordinating organic

solvents such as thf, benzene or fluorobenzene, even at low temperatures to produce a grey

solid and [(CF3SO3)2Ga(DDP)], thus no NMR spectroscopic data have been provided for 6.

However, characterisation has been successfully effected by elemental analysis, IR

spectroscopy and single crystal X-ray diffraction. The IR spectroscopic measurement displays

absorptions at 1211 and 1381 cm-1, well consistent of covalently linked triflate to the Cu(I)

centre in compound 6.[217]

Single Crystal X-Ray Analysis of [{(DDP)GaCu(CF3SO3)}2] (6)

Colourless crystals of 6, suitable for single crystal X-ray measurement, have been obtained

from a fluorobenzene/n-hexane mixture at room temperature. Compound 6 crystallizes in the

triclinic space group P-1 (Figure 20). The main structural feature of 6 is the almost perfectly

planar [Cu2Ga2] four-membered ring at which the DDP backbone is located perpendicular to

this ring plane. The Ga1-Cu-Ga1’ (124.90(6)°), Ga1-Cu-O1 (121.7(2)°) and O1-Cu-Ga1’

(113.4(2)°) angles are close to 120° resulting in a sum of angles of 359.98° around each Cu

centre and thus providing trigonal planar geometry. The Cu••••Cu distance in 6 accounts for

2.277(3) Å and is approx. 0.03 Å shorter than in the known Cu(I) dimer, [({N(2,6-iPr2-

C6H3)CH}2C)Cu(H)]2 (2.306(1) Å).[231] Thus, 6 exhibits the shortest Cu••••Cu contact in

molecular compounds known so far. The Cu-Ga bond distances in 6 slightly differ by around

0.08 Å (2.421(1) Å for Ga1-Cu1 and 2.500(1) Å for Ga1-Cu1’) which leads to the suggestion

that compound 6 can be best described by two strongly associated monomeric

[(DDPGa)Cu(CF3SO3)] units in the solid state. The Cu-Ga bond distances in 6 are

significantly elongated in comparison to the few known Cu/Ga compounds at which the GaR

systems occupy only terminal positions: [LCu-GaR][38] (2.307(1) Å and 2.281(1) Å with R =

{N(C6H3-2,6-iPr2)CH}2, L = {N(2,4,6-Me3-C6H2)CH}2C and L = {N(2,6-iPr2-C6H3)CH}2C)

and [Cu(GaCp*)4][BArF][39] (2.352(1) and 2.350(1) Å, BArF = [B{C6H3(CF3)2}4]). The Cu1-

O1 bond distance (1.942(8) Å) indicates the presence of polar covalently linked triflate at the

copper centre, which is consistent with IR spectroscopic results mentioned above. As

expected, coordination of Ga towards the Cu centres results in shorter Ga-N distances (Ga-N1

1.936(10) Å and Ga-N2 1.939(9) Å) and a larger N1-Ga-N2 angle (96.0(4)°) for compound 6


in comparison to free Ga(DDP) (Ga-N, 2.0528(14) and 2.0560(13) Å; N-Ga-N, 87.53(5)°).[227,

232, 233]


displacement ellipsoids (Cu, Ga) are shown at 50% probability level, hydrogen atoms are omitted for clarity.

Selected interatomic distances (Å) and angles (°): Cu1••••Cu1’ 2.277(3), Ga1-Cu1 2.4212(18), Ga1-Cu1’

2.4997(3), Cu1-O1 1.942(8), Ga1-N1 1.936(10), Ga1-N2 1.939(9), Cu1-Ga1-Cu1’ 55.10(6), Cu1’-Cu1-Ga1

64.20(7), Ga1-Cu1-Ga1’ 124.90(6), Cu1’-Cu1-Ga1’ 60.70(6), O1-Cu1-Cu1’ 173.9(2), O1-Cu1-Ga1 121.7(2),

O1-Cu1-Ga1’ 113.4(2), N1-Ga1-N2 96.0(4), N1-Ga1-Cu1 126.5(2), N2-Ga1-Cu1 129.1(4), N1-Ga1-Cu1’

125.2(3), N2-Ga1-Cu1’ 124.5(3).

Synthesis and Characterisation of [Cu2(GaCp*)(µ-GaCp*)3{Ga(CF3SO3)3}] (7)

Using GaCp* (5 eq) instead of Ga(DDP) in the reaction with [Cu(CF3SO3)2] results in the

formation of [(Cp*Ga)Cu(µ-GaCp*)3Cu{Ga(CF3SO3)3}] (7) (Scheme 17). The ability of

GaCp* to stabilize dinuclear compounds of soft cationic d10 coinage metal centres has been

shown through the preparation of the Ag(I) compound [Ag2(GaCp*)3(µ-

GaCp*)2][CF3SO3]2.[39] The 1H NMR spectrum of 7 in d8-thf at room temperature shows only

one signal at 2.03 ppm for the protons of the Cp* rings in terminal and bridging GaCp*


positions. In contrast, the 1H NMR measurement in d8-thf at -60°C shows two signals in a 1:3

ratio at 1.99 and 2.06 ppm which proves a fluxional process of the GaCp* ligands in solution.

The 13C NMR and 19F NMR spectra do not show any unusual features with respect to the

expected signals (Table 7).

GaCp*C6H5F, 60°C, 1h

Cu

Cp*Ga

Cu

GaCp*

[Cu(OTf)2] *CpGa

Cp*Ga

Ga(OTf)3

Scheme 17. Synthesis of [(Cp*Ga)Cu(µ-GaCp*)3Cu{Ga(CF3SO3)3}] (7).

Comment on the oxidation states and mechanisms. Obviously, compound 7 can be most

accurately described as a Lewis acid/base adduct at which the {Ga(CF3SO3)3} ligand acts as

the Lewis acid attaching to the Lewis basic copper centre. This would be in good agreement

with numerous quantum chemical calculations on related transition metal-group 13 metal

complexes revealing more or less polarized covalent donor-acceptor bonds with M(δ-) and

Ga(δ+).[7, 9, 15, 116, 122, 234] These facts lead directly to the suggestion to assign the formal

oxidation states Cu(0) and Ga(I) to the neutral structural fragment [Cu2(GaCp*)4] of 7. It

should be noted, that the assignment of oxidation states in coordination chemistry is mainly

based on heuristic reasoning and needs to be done in a self-consistent way.[235] Possible

mechanistic steps in the formation of 7 can be redox chemical processes in which Ga(I) acts

as the reductant for the Cu(II) starting compound. Herein, gallium is oxidized to its favoured

oxidation state +III found in the fragment {Ga(CF3SO3)3} and copper is reduced to Cu(0).

Certainly, the formation of {Ga(CF3SO3)3} is just one possible by-product of this reaction

sequence. Further by-products can be Cp* containing gallium species produced due to Cp*

transfer reactions. However, the electrophilic Ga(III) centre of {Ga(CF3SO3)3} coordinates at

the vacant pyramidal copper site of the dinuclear fragment [(Cp*Ga)Cu(µ-GaCp*)3Cu],

which is nucleophilic and thus acts as the Lewis base. It should be noted that it seems to be

impossible to obtain the fully homoleptic compound [(Cp*Ga)Cu(µ-GaCp*)3Cu(GaCp*)]

which contains an additional GaCp* donor ligand instead of the {Ga(CF3SO3)3} acceptor. A

very electron rich situation would be created and the stability significantly decreased. This is

consistent with the observation that using an excess of GaCp* in the reaction did not lead to


the hypothetical compound [Cu2(GaCp*)5]. In addition, note the electron count of 32 for such

a hypothetic species [Cu2(GaCp*)5] in comparison to its existing isostructural

[M2(GaCp*)5][128, 130] (M = Pd, Pt) congeners which exhibit an electron count of 30. In

summary, it seems that Lewis acid/base interactions Cu(0)→Ga(III) are stronger than

Cu(I)←Ga(I) and Cu(0)←Ga(I) interactions and thus account for the inaccessibility of

[Cu2(GaCp*)5] as well.

Single Crystal X-Ray Analysis of [(Cp*Ga)Cu(µ-GaCp*)3Cu{Ga(CF3SO3)3}] (7)

Suitable crystals of 7 for single crystal X-ray diffraction have been obtained by slow diffusion

of n-hexane into a saturated fluorobenzene solution at room temperature. Compound 7

crystallizes in the monoclinic space group P21/n (Figure 21).



Selected interatomic distances (Å) and angles (°): Cu1••••Cu2 2.3236(8), Cu1-Ga1 2.2906(8), Cu1-Ga2

2.4011(8), Cu1-Ga3 2.4131(9), Cu1-Ga4 2.4120(9), Cu2-Ga5 2.3268(8), Cu2-Ga2 2.5044(9), Cu2-Ga3

2.4837(9), Cu2-Ga4 2.4883(8), Ga1-O1 1.993(3), Ga1-O4 1.969(3), Ga1-O7 1.964(4), Ga2-Cp*centroid 1.879,

Ga3-Cp*centroid 1.896, Ga4-Cp*centroid 1.902, Ga5-Cp*centroid 1.972, Ga1-Cu1-Cu2 177.17(4), Cu1-Cu2-Ga5

178.91(4), Cu1-Ga4-Cu2 56.59(2), O1-Ga1-Cu1 121.58(11), O1-Ga1-O7 91.96(15), Cu1-Ga4-Cp*centroid 150.95,

Cu2-Ga5-Cp*centroid 177.63.


The central [Cu2] unit is surrounded by three bridging GaCp* ligands and one terminal

bonded GaCp* ligand on one side (Cu2) and one {Ga(CF3SO3)3} group on the other (Cu1) in

an almost linear fashion. The Cu-Cu-Ga angles correlate closely to the ideal value of 180°

(177.17(4)° for Ga1-Cu1-Cu2 and 178.91(4)° for Cu1-Cu2-Ga5). The Ga-Cp*centroid distance

(1.972 Å) of the terminal ligand is slightly elongated in comparison to the bridging Ga-

Cp*centroid units (av. 1.892 Å), contrary to the distances found in the homoleptic dimeric d10

metal compound [Pt2(GaCp*)5].[130] The Cu••••Cu distance (2.3236(8) Å) is distinctly longer

than in compound 6 (2.277(3) Å). The average Cu1-GaCp*bridging bond length (2.409 Å) is

slightly shorter than the average Cu2-GaCp*bridging distance (2.492 Å). In addition, the Cu2-

GaCp*terminal bond distance (2.3268(8) Å) is much shorter than all Cu-GaRbridging (R = Cp* and

DDP) distances in 6 (av. 2.460 Å) and 7 (av. 2.451 Å). All these Cu-Ga bonds are slightly

elongated in comparison to the literature known Cu/Ga compounds mentioned in the

discussion of 6. Most interestingly, the Cu1-Ga1 distance (2.2906(8) Å) involving the

{Ga(CF3SO3)3} fragment is significantly shorter than all the other Cu-GaCp* distances of 7

and in the cation [Cu(GaCp*)4]+ as well. These structural comparisons support the treatment

of 7 as a Lewis acid/base adduct as it has been described above.

Synthesis and Characterisation of [Cu2(GaCp*)3(µ-GaCp*)2][CF3SO3]2 (8)

While the reaction of Ga(DDP) gives no defined products with Cu(I) triflate compounds, i.e.

Cu(OTf)•toluene and [{Cu(cod)2}(CF3SO3)], GaCp* reacts smoothly with the 1,5-

cyclooctadiene ligand stabilized starting material [{Cu(cod)2}(CF3SO3)] in fluorobenzene to

yield [Cu2(GaCp*)3(µ-GaCp*)2][CF3SO3]2 (8) (Scheme 18).

[Cu(cod)2][CF3SO3] GaCp*C6H5F, rt Cu

Cp*Ga

CuGaCp*

*CpGa

*CpGa GaCp*

O3SCF3

CF3SO3-

(8)

Scheme 18. Synthesis of [Cu2(GaCp*)3(µ-GaCp*)2][CF3SO3]2 (8).

The 1H NMR, 13C NMR and 19F NMR spectra at room temperature do not show any unusual

features and indicate a fluxional behaviour of 8 in solution, quite similar as observed for 7 and

the previously described silver analogue [Ag2(GaCp*)3(µ-GaCp*)2][CF3SO3]2 (Table 7).[39]


Compound 8 is highly soluble in several common polar organic solvents like CH2Cl2,

fluorobenzene and thf and stable under an inert gas atmosphere for several weeks at -30°C.

Table 7. NMR spectroscopic data of 6 and 7 (d8-thf, rt). Note: Carbon atoms of the {CF3SO3} groups were not

detected in the 13C NMR spectrum under the standard conditions of the routine measurements.

Compound 1H NMR

(s, 60H, GaCp*)

13C NMR

GaC5Me5 GaC5Me5

19F NMR

7 2.03 9.75 115.46

-78.8

8 2.04 7.10 112.20

-78.4

Single Crystal X-Ray Analysis of [Cu2(GaCp*)3(µ-GaCp*)2][CF3SO3]2 (8)

Crystals of 8 suitable for single crystal X-ray measurements were obtained by slow diffusion

of n-hexane into a fluorobenzene solution at room temperature. Compound 8 crystallizes in

the monoclinic space group P21/n (Figure 22). Compound 8 consists of a central [Cu2] unit,

quite similar to 7, at which the copper atoms are bridged by two GaCp* ligands. In addition,

one copper atom (Cu2) bears to two terminally coordinated GaCp* units, whereas the other

one (Cu1) is coordinated by only one GaCp* ligand as well as one triflate ligand, resulting in

a tetrahedral environment for both copper centres. The Cu••••Cu distance (2.5247(12) Å) is

significantly longer in comparison to compounds 6 (2.277(3) Å) and 7 (2.3238(7) Å). The Cu-

GaCp*bridging distances range between 2.4232(10) Å and 2.4572(11) Å and are expectedly

elongated as compared with the distances Cu-GaCp*terminal (av. 2.389 Å), as it has been also

observed for 7 and in [M2(GaCp*)5] (M = Pd, Pt).[17, 128, 130, 132] As usual, the Ga-Cp*centroid

distance (1.991 Å) of the terminal ligand is slightly elongated in comparison to the bridging

Ga-Cp*centroid units (av. 1.968 Å). In contrast to compound 7, the bond distance of Cu-

GaCp*terminal is slightly longer than the average Cu-Ga bond length in the homoleptic cation

[Cu(GaCp*)4]+ (2.351 Å) and the other known mononuclear Cu-Ga complexes.




Selected interatomic distances (Å) and angles (°): Cu1••••Cu2 2.5247(12), Cu1-Ga3 2.4292(10), Cu2-Ga3

2.4572(11), Cu1-Ga1 2.4232(10), Cu2-Ga1 2.4430(11), Cu1-Ga5 2.3900(12), Cu1-Ga2 2.3886(12), Cu2-Ga4

2.4010(13), Cu2-O1 2.075(5), Ga1-Cp*centroid 1.924, Ga2-Cp*centroid 1.997, Ga3-Cp*centroid 2.011, Ga4-Cp*centroid

1.995, Ga5-Cp*centroid 1.982, Ga4-Cu2-O1 108.25(15), Cu2-Ga3-Cp*centroid 130.29, Cu2-Ga3-Cu1 62.22(3), Ga2-

Cu1-Ga5 108.55(5), Cu1-Ga2-Cp*centroid 171.71, Cu2-Ga4-Cp*centroid 172.89.

Comparison between 7, 8 and [M2(GaCp*)5] (M = Pd, Pt). The hypothetical, dicationic

species [Cu2(GaCp*)5]2+, the neutral fragment [Cu2(GaCp*)4], which has been discussed

above in case of 7, as well as the already known dimeric d10 metal compounds [M2(GaCp*)5]

(M = Pd, Pt) exhibit an electron count of 30. Thus, compound 8 can be viewed as a trapped

intermediate structure of electronically saturated, 30 electron, fluxional [Cu2(GaCp*)5]2+

fragments by coordinating one triflate ligand at one electrophilic Cu(I) site upon

crystallisation from solution. Notably, coordination of triflate at the copper center has been

not observed in case of compound 7 at which the copper centres are more electron rich.

It has been shown that oligomeric CuaGab units can be easily obtained via reaction of

commercially available Cu(I) and Cu(II) starting materials with Ga(I)R species under


moderate reaction conditions. The low valent Ga(I) ligand behaves as both, the selective

reducing agent and the stabilizing unit for the central [Cu2] core. The synthesis of the Cu(I)

dimer [{(GaDDP)Cu(CF3SO3}2] (6) which exhibits a very short Cu••••Cu distance requires the

combination of Ga(DDP) with [Cu(CF3SO3)2]. In addition, the competing reaction pathways

between coordination, insertion and redox processes have been illustrated by the reaction of

[Cu(CF3SO3)2] with GaCp* leading to the formation of the (formal) Cu(0) compound

[(Cp*Ga)Cu(µ-GaCp*)3Cu{Ga(CF3SO3)3}] (7). This Lewis acid/base adduct contains the

fluxional, neutral 30 electron fragment [Cu2(GaCp*)4] and appears to be the first existing

Cu(0) complex or rather cluster. The formation of the Lewis acidic {Ga(CF3SO3)3} fragment

takes place in situ by redox chemical reaction pathways in which Cu(II) is reduced to Cu(0)

and Ga(I) is oxidized to Ga(III). In contrast, the dimeric Cu(I) compound [Cu2(GaCp*)3(µ-

GaCp*)2][CF3SO3]2 (8) is only available with the sterically much less crowded Cp* group and

by choosing the Cu(I) starting compound [{Cu(cod)2}(CF3SO3)] which avoides any Cu/Ga

redox reaction. In general, these results present promising advances to extend the well known

coordination chemistry of GaCp* towards soft cationic precious group 11 metals. Thus,

achievement of oligonuclear cationic or even neutral compounds [Ma(ECp*)b]m+ (m ≥ 0; M ⊃

group ≥ 11; b ≥ a ≥ 2) will be interesting as intermediates or starting precursors for soft

chemical synthesis of larger M/E intermetallic clusters or nanoparticles.

[e] 3.3.1: T. Bollermann, I. Schwedler, M. Molon, K. Freitag, C. Gemel, R. W. Seidel, and R. A. Fischer, Dalton Trans. 2011, submitted. 3.3.2: T. Bollermann, M. Molon, C. Gemel, K. Freitag, R. W. Seidel, M. von Hopffgarten, P. Jerabek, G. Frenking, and R. A. Fischer, J. Am. Chem. Soc. 2011, submitted.


3.3 Experimental and Theoretical Investigations on the Formation of

Zinc-rich Oligonuclear Cluster Compounds[e]

Abstract

The reactions of heteroleptic GaCp*/CO containing complexes of iron and cobalt, namely

[(CO)3M(µ2-GaCp*)nM(CO)3] (M = Fe, n = 3; M = Co, n = 2) and [Fe(CO)4(GaCp*)], with

ZnMe2 in the presence of a coordinating solvent are investigated. The reaction of the iron

complex [Fe(CO)4(GaCp*)] with ZnMe2 in the presence of thf leads to the dimeric compound

[(CO)4Fe{µ2-Zn(thf)2}2Fe(CO)4] (9). In addition, the reaction of [(CO)3Fe(µ2-

(GaCp*)3Fe(CO)3] with ZnMe2 and stoichiometric amounts of thf leads to the formation of

[(CO)3Fe{µ2-Zn(thf)2}2(µ2-ZnMe)2Fe(CO)3] (10) containing {Zn(thf)2} as well as ZnMe

ligands. In contrast, the use of pyridine instead of thf leads to [(CO)3Fe{µ2-

Zn(py)2}3Fe(CO)3] (11) via the replacement of all GaCp* ligands by three {Zn(py)2} groups.

In comparison, the reaction of [(CO)3Co(µ2-GaCp*)2Co(CO)3] with ZnMe2 in the presence of

pyridine or thf leads in both cases to the formation of [(CO)3Co{µ2-Zn(L)2}(µ2-

ZnCp*)2Co(CO)3] (L = py (12) or thf (13)) via the replacement of GaCp* with {Zn(L)2}

units as well as Cp* transfer from the gallium to the zinc center. Compounds 9-13 point out,

that the variations of the transition metal starting compound as well as the solvents used have

an influence on the formation of well-defined compounds with various M/E ratios.

Additionally, the reaction of the homoleptic palladium dimer [Pd2(µ-GaCp*)3(GaCp*)2] with

ZnMe2 leads to the first dimeric cluster compound [Pd2Zn6Ga2(Cp*)5(CH3)3] (14) featuring a

30 valence electron [Pd2Ga2Zn6] core. Compound 14 consits of two Cs symmetric isomers.

The single crystal X-ray analysis of 14 shows a bi-capped trigonal prismatic coordination

environment. Theoretical investigations have been carried out by MO correlations, AIM and

EDA analysis which suggest significant attractive Pd-Pd interactions.

Introduction

Since their discovery in the early 1990s, low valent group 13 elements E(I)R (E = Al, Ga, In;

R = bulky substituent) have attracted much interest in organometallic coordination chemistry

as two electron donor ligands.[10, 17, 18]


Due to their, more or less, isolobal properties to carbon monoxide, first investigations in

possible reaction schemes involved ligand substitution reactions of several binary carbonyls

under liberation of CO, e.g. [Fe2(CO)9] or [Co2(CO)8].[116] In addition, even mono substituted

compounds such as [(nbd)Mo(CO)4] or [(cht)Fe(CO)3] (nbd = 2,5-norbornadiene, cht =

cycloheptatriene) have been used to prepare transition metal GaCp* containing compounds,

i.e. [Mo(CO)4(GaCp*)2] and [Fe2(GaCp*)3(CO)6], respectively.[116, 126] Thus, as discussed

before, the successive substitution of CO is limited by the coordination of GaCp* due to the

strong π-donor abilities of GaCp*. Herein, the over-all electron density of the transition metal

upon coordination of GaCp* is signficicantly increased, leading to stronger π-back bonding to

the remaining CO ligands and thus prohibiting further substitution. As a result of the

electronic properties, homoleptic GaCp* containing complexes can only be obtained by

substitution of labile and very weak π-acceptor ligands such as olefins, alkyl groups or

acetonitrile. For instance, [M(GaCp*)4] (M = Ni, Pt) could be obtained from [M(cod)2] (cod =

1,5-cyclooctadiene) and likewise [Pd(GaCp*)4] has been prepared from

[Pd(tmeda)(CH3)2].[126, 128] In the formation of oligonuclear d10 metal compounds [Ma(ECp*)b]

(a < b), two general reaction schemes have been proven to be successful. On the one hand,

dimeric compounds can be obtained via building block synthesis. Herein, the reaction of

mononuclear homoleptic d10 metal complexes [M(GaCp*)4] (M = Pd, Pt) with substitution

labile compounds such as [Pt(cod)2] and immediate addition of GaCp* leads to

[PdM(GaCp*)5] (M = Pd, Pt).[17, 128] On the other, kinetically controlled cluster growth under

suitable reaction conditions leads to even dinuclear compounds mentioned above as well as

trinuclear clusters such as [Pd3(ECp*)8] (E = Ga, In).[131, 132] However, recently it has been

shown that E(I)Cp* (E = Al, Ga) containing transition metal complexes can be used as

suitable starting materials in the formation of zinc-rich compounds. For instance, homoleptic

metal-rich compounds [M(ZnR)n] (M = Mo, Ru, Rh, Ni, Pd, Pt; n = 8-12) are accesible in

high yields by the reaction of the corresponding GaCp* containing compounds [M(GaCp*)n/2]

with ZnR2 (E = Me, Et).[42, 43] Herein, the formation is based on combined Ga/Zn and Cp*/Me

exchange reactions involving redox chemical processes between gallium and zinc, namely

oxidation of Ga(I) to Ga(III) and simultaneously the reduction of Zn(II) to Zn(I). Thus, the

unusually high coordination numbers at the transition metal centres are based on the

substitution of one 2e donor GaCp* by two 1e donors ZnR. In addition, even heteroleptic

starting materials could be used to achieve compounds of higher nuclearity. The reaction of

[Mo(CO)4(GaCp*)2] with ZnMe2 leads to the formation of the unprecedented compound

[{(CO)4Mo}4(Zn)6(µ-ZnCp*)4].[207] Its structural environment represents a molecular cut-out


of the zinc rich intermetallic phase MoZn20.44 as has also been observed for the ‘homoleptic’

icosahedral compound [Mo(ZnCp*)3(ZnMe)9]. A controlled and predictable synthesis of such

high nuclearity clusters is not known yet and strictly depends on the number of existing co-

ligands. For instance, using [Mo(CO)3(GaCp*)3] did not lead to cluster growth as reported

above, but to the formation of [Mo(CO)3(ZnCp*)3(ZnMe)3].[236] In general, it seems to be of

considerable interest to further investigate the reactivity of ZnMe2 towards heteroleptic

GaCp* containing transition metal compounds as well as dimeric starting materials.

3.3.1 Zinc-rich Compounds of Iron and Cobalt: Formation of [Fe2Znx] (x = 2-4) and

[Co2Zn3] Cores

In order to gain further insights into the influcences, the major questions have been: How do

the transition metals and the nuclearity of the starting materials influence cluster growth?

And, in general, can reaction pathways be easily controlled via suitable GaCp*/CO ratios?

Therefore, mononuclear and dinuclear heteroleptic complexes have been choosen, namely

[Fe(CO)4(GaCp*)], [(CO)3Fe(µ2-GaCp*)3Fe(CO)3] and [(CO)3Co(µ2-GaCp*)2Co(CO)3].

Synthesis and Characterisation of [(CO)4Fe{µ2-Zn(thf)2}2Fe(CO)4] (9), [(CO)3Fe{µ2-

Zn(thf)2}2(µ2-ZnMe)2Fe(CO)3] (10) and [(CO)3Fe{µ2-Zn(py)2}3Fe(CO)3] (11)

The treatment of the monomeric iron complex [Fe(CO)4(GaCp*)] with excess ZnMe2 in

toluene at 90°C for 1 h results in the formation of a slightly yellow solid which is insoluble in

common apolar and non-coordinating organic solvents even at temperatures of around 100°C.

Crystallisation of the formed solid from thf results in the formation of colourless crystals of

[(CO)4Fe{µ2-Zn(thf)2}2Fe(CO)4] (9) in yields around 60% (Scheme 19). Notably, by using

pyridine instead of thf, the monomeric compound [(py)3ZnFe(CO)4] is formed instead, which

has already been synthesized by the reaction of [(NH3)3ZnFe(CO)4] with pyridine and

structurally characterized by Zaugg.[188, 237] In addition, the reaction of dimeric [(CO)3Fe(µ2-

GaCp*)3Fe(CO)3] with excess ZnMe2 in toluene at 90°C for 1 h leads to the formation of an

orange solid. Treatment of the orange solid with thf yields in the formation of [(CO)3Fe{µ2-

Zn(thf)2}2(µ2-ZnMe)2Fe(CO)3] (10) as orange crystals in good yields of about 45%. In

contrast, the uptake of the formed solid in a mixture of toluene/pyridine leads to the formation

of [(CO)3Fe{µ2-Zn(py)2}3Fe(CO)3] (11) in yields around 50%. In the synthesis of 10 and 11

the dimeric structure of the starting material is maintained in the presence of both, thf and


pyridine. Pure crystals of compounds 9-11 are nearly unsolvable in all common polar and

non-polar solvents except of the corresponding coordinated solvents, i.e. thf and pyridine, or

rather toluene mixtures of them. Compounds 9-11 are stable for several weeks when stored at

-30°C under an inert gas atmosphere in the glovebox.

[Fe(CO)4(GaCp*)]1. xs. ZnMe2, toluene, 1 h, 90°C2. thf Zn

ZnFe Fe

COOC

OC COCO

CO

COO

O

O

OOC

(9)

[(CO)3Fe(µ2-GaCp*)3Fe(CO)3]

xs. ZnMe2, toluene, 1 h, 90°Ctoluene/pyridine

Fe Fe

CO

COCO

Zn

ZnZn

OC

OCOC

(11)

NN N

N

N N

thf

(10)

Fe Fe

CO

COCO

Zn

ZnZn

OCOC

Zn

OC

Me

Me O O

OO

Scheme 19. Synthesis of [(CO)4Fe{µ2-Zn(thf)2}2Fe(CO)4] (9), [(CO)3Fe{µ2-Zn(thf)2}2(µ2-ZnMe)2Fe(CO)3] (10)

and [(CO)3Fe{µ2-Zn(py)2}3Fe(CO)3] (11).

At this point it should be noted that the assignment of valences and oxidation states in

coordination chemistry is primarily of heuristic value.[235] That means the electron count has

to be made in a self-consistent way which can be somehow different. The interactions or

electron sharing properties between the metal fragments can be perceived in two common

ways. On the one hand, the [M(CO)n] fragments can be considered as anionic carbonyl

metallates interacting with cationic Zn(II) units, on the other, these fragments can be declared

as neutral [M(CO)n] fragments which interact with neutral Zn(0)L2 or Zn(I)R ligands.

Although the discussion of the bonding situation of carbonyl complexes of the type

[(CO)nM(ER)] and [(CO)nM(ERL2)] justifies the view of anionic carbonyl fragments and


positively charged ECp*, ZnR or ZnLn (n = 2-3), the preferred assignment in this thesis is

based on the neutral fragment approach in accordance with quantum chemical calculations of

the charge distribution in the homoleptic compounds of the type [Ma(ER)b] or [M(ZnR)n].[42-

44] The perception from different points of view does not have any impact on the over-all

electron count and the heuristic concepts of synthesis planning and structure rationalisation

based on the electronic equivalence of one GaCp* to one ZnLn or two ZnR units. The IR

spectroscopic measurements for 9-11 display the expected absorption bands for terminally

bound CO ligands (9: 1899-2020 cm-1, 10: 1849-2028 cm-1, 11: 1728-1846 cm-1) well

comparable with literature known [Fe(CO)n] containing compounds.[116, 237] NMR

spectroscopic data show no unusual features with respect to the coordinated solvent as well as

the ZnMe groups found in 11 (Table 8).

Table 8. 1H NMR spectroscopic data of compounds 9, 10 (d8-thf, rt) and 11 (d5-pyridine, rt).

Compound Zn(thf)2

16H 16H

ZnMe Zn(py)2

9 1.77 3.62 - -

10 1.76 3.61 -0.57 (6H) -

11 - - 8.71 (12H), 7.55 (6H),

7.19 (12H)

Single Crystal X-Ray Analysis of [(CO)4Fe{µ2-Zn(thf)2}2Fe(CO)4] (9), [(CO)3Fe{µ2-

Zn(thf)2}2(µ2-ZnMe)2Fe(CO)3] (10) and [(CO)3Fe{µ2-Zn(py)2}3Fe(CO)3] (11)

The asymmetric unit of 9 comprises two crystallographically independent halves of the

molecule, lying respectively on a crystallographic twofold rotation axis, passing through the

Fe-Fe axis, and a centre of symmetry. Likewise, the crystal structure of 10 comprises two

independent molecules, which sit on two distinct centres of symmetry in this case. Thus,

discussion of angles and distances as determinded by single crystal X-ray analysis display

average values for both crystallographically independent molecules. Compound 9 consists of

two [Fe(CO)4] units which are linked by two bridging {Zn(thf)2} ligands (Figure 23). The

terminally bonded carbonyl ligands are arranged in an eclipsed conformation in relation to the

Fe-Fe axis.



displacement ellipsoids (Fe, Zn) are shown at 50 % probability level, hydrogen atoms are omitted for clarity.

Selected interatomic distances (Å) and angles (°): Fe3-Fe2 4.224(1), Fe1-Fe1 4.2870(8), Zn2-Zn2 2.7341(6),

Zn1-Zn1 2.7378(6), Zn1-Zn1 2.7378(6), Zn2-O8 2.117(2), Zn2-O7 2.096(3), Zn2-O8 2.117(2), Zn2-O7

2.096(3), Zn1-O1 2.105(3), Zn1-O2 2.095(3), Zn1-O1 2.105(3), Zn1-O2 2.095(3), Fe1-Zn1 2.5746(8), Fe1-Zn1

2.5116(7), Fe1-Zn1 2.5116(7), Fe1-Zn1 2.5746(8), Fe3-Zn2 2.5270(9), Fe3-Zn2 2.5270(9), Fe2-Zn2 2.5047(9),

Fe2-Zn2 2.5047(9), Fe1-Zn1-Fe1 114.88(2), Fe1-Zn1-Fe1 114.88(2), Zn1-Fe1-Zn1 65.12(2), Zn1-Fe1-Zn1

65.12(2), Fe3-Zn2-Fe2 114.17(3), Fe3-Zn2-Fe2 114.17(3), Zn2-Fe3-Zn2 65.50(3), Zn2-Fe2-Zn2 66.16(3).

The single crystal X-ray structure determination of 9 reveals a perfectly planar four-

membered ring of the [Fe2Zn2] core (Fe-Zn-Fe av. 114.53°, Zn-Fe-Zn av. 65.59°). The zinc

atoms of the {Zn(thf)2} units each exhibit a pseudo-tetrahedral environment by the oxygen

atoms of the two thf ligands and the two iron atoms. Each of the iron atoms are further

coordinated by four terminal carbonyl ligands. The coordination sphere of the iron atoms is

thus a distorted octahedron, namely [Fe(CO)4Zn2]. Notably, same structural elements have

been found in the related compound [(CO)4Fe{Zn(bipy)}]2.[238] Several compounds exist

consisting of a [Fe2E2] (E = any element) unit such as [(CO)3Fe(µ2-RS)(µ2-

RHgS)Fe(CO)3][239] (R = Me, Et) or [Fe2(CO)8(µ4-Sb)]2[Fe2(CO)6][240]. In general, these

compounds exhibit rather a butterfly configuration than a planar subunit. In comparison, the

molecular structure of 10 features an asymmetrically fourfold capped dimeric iron unit which


contains two bridging tri-coordinated ZnMe and two tetra-coordinated {Zn(thf)2} units, which

results in two [FeZn4] coordinated moieties (Figure 24, left).

Figure 24. Molecular structures of 10 (left) and 11 (right) in the solid state as determined by single crystal X-ray

diffraction; displacement ellipsoids (Fe, Zn) are shown at 50 % probability level, hydrogen atoms are omitted for

clarity. Note: The solid lines between the zinc atoms (10) and iron atoms (11) do not indicate strong covalent Zn-

Zn and Fe-Fe interactions. Selected interatomic distances (Å) and angles (°) for 10: Fe1-Fe1 2.981(2), Fe1-Zn2

2.467(2), Fe1-Zn1 2.433(2), Fe1-Zn2 2.563(1), Fe1-Zn1 2.478(1), Fe1-Zn2 2.563(1), Fe1-Zn1 2.478(1), Fe1-

Zn1 2.433(2), Fe1-Zn2 2.467(2), Zn1-O5 2.031(5), Zn1-O4 2.062(6), Zn1-O4 2.062(6), Zn1-O5 2.031(5), Zn1-

Zn2 2.778(1), Zn1-Zn2 2.778(1), Zn1-Zn2 2.848(1), Zn2-Zn1 2.848(1), Fe2-Fe2 2.972(2), Zn3-O9 2.03(1), Zn3-

O10 2.052(5), Zn3-O10 2.052(5), Zn3-O9 2.03(1), Fe2-Zn3 2.450(2), Fe2-Zn4 2.464(2), Fe2-Zn4 2.557(2), Fe2-

Zn3 2.457(2), Fe2-Zn4 2.464(2), Fe2-Zn3 2.450(2), Fe2-Zn4 2.557(2), Fe2-Zn3 2.457(2), Zn3-Zn4 2.750(2),

Zn3-Zn4 2.872(1), Zn4-Zn3 2.872(1), Zn4-Zn3 2.750(2), Zn1-Zn2-Zn1 87.85(3), Zn2-Zn1-Zn2 92.15(3), Zn1-

Zn2-Zn1 87.85(3), Zn2-Zn1-Zn2 92.15(3), Zn4-Zn3-Zn4 92.07(4), Zn3-Zn4-Zn3 87.93(4), Zn4-Zn3-Zn4

92.07(4), Zn3-Zn4-Zn3 87.93(4). Selected interatomic distances (Å) and angles (°) for 11: Fe2-Fe1 3.1054(9),

Fe2-Zn1 2.4436(7), Fe2-Zn2 2.4318(8), Fe2-Zn3 2.4351(8), Fe1-Zn1 2.4514(8), Fe1-Zn3 2.4564(7), Fe1-Zn2

2.4485(8), Zn1-N1 2.157(4), Zn1-N2 2.120(3), Zn2-N3 2.111(4), Zn2-N4 2.112(3), Zn3-N5 2.206(4), Zn3-N6

2.126(3), Fe2-Zn3-Fe1 78.82(2), Fe1-Zn1-Fe2 78.75(2), Fe2-Zn2-Fe1 79.04(2), Zn3-Fe2-Zn1 87.41(2), Zn3-

Fe1-Zn1 86.77(2), Zn2-Fe2-Zn1 76.08(2), Zn2-Fe1-Zn1 75.63(2).

The structural feature of the inner [Fe2Zn4] core of 10 is best described as a slightly distorted

octahedron containing the [Zn4] unit in the equatorial plane and the iron atoms in axial

positions (Zn1-Zn2-Zn1 av. 87.89°, Zn2-Zn1-Zn2 av. 92.11°). Compound 11 displays a

symmetrically threefold capped iron dimer with three bridging {Zn(py)2} ligands (as the

electronic equivalent to three GaCp* units) and terminally bonded carbonyl ligands staggered

in relation to the Fe-Fe axis in the solid state. The [Fe2Zn3] core can be viewed as a trigonal


bi-pyramide with the iron atoms in axial positions, and is quite similar to the [Fe2Ga3] core as

it has been described for the GaCp* containing starting material [(CO)3Fe(µ2-

GaCp*)3Fe(CO)3][116] and related compounds such as [(CO)3Fe(µ2-GeMe2)3Fe(CO)3][241] and

[CpFe(µ2-SR)3FeCp][242] (R = Et, Ph) which all exhibit 34 valence electrons. The Fe-Fe

distance in 9 (4.26 Å) definitely rules out any attractive bonding interactions. Although, the

distances in 10 and 11 (10: 2.977 Å; 11: 3.105 Å) are significantly shorter and comparable to

the Fe-Fe distance in [(CO)3Fe(µ2-GaCp*)3Fe(CO)3][116] (2.908(6) Å), the description of

direct Fe-Fe interactions seems to be unreasonable. Note that the Fe-Fe distance in

[Fe2(CO)9][243a] (2.522(1) Å) is clearly shorter and even in this classical compound the

assignment of Fe-Fe interactions is still controversially discussed.[243b, 243c] Thus, the Fe-Fe

distance seems to be controlled by steric reasons of the used ligand systems. Distinct

differences can also be observed in Fe-Zn bond distances (9: av. 2.529 Å; 10: av. 2.481 Å, 11:

av. 2.444 Å). All the Fe-Zn distances found in compounds 9-11 are somewhat longer than

those of related compounds such as [(py)3ZnFe(CO)4][188] (2.4017(3) Å), [Zn{Fe(CO)4}2]2- [244] (2.317 Å), [Zn2Cl2Fe(CO)4(tmeda)2][189] (av. 2.488 Å) and [Zn2Cl2Fe(CO)4(thf)2][189]

(2.413(3) Å). The average non-bonding Zn-Zn distance of 9 (av. 2.737 Å) is well comparable

with the distance reported for [(CO)4Fe{Zn(bipy)}]2 (2.788(1) Å).[238] The slightly shorter

value found in 9 can be attributed to steric reasons due to different bulky behavior of the thf

ligands with respect to the more bulky bipyridine. The related Zn-Zn distances in 10 (av.

2.812 Å) and 11 (av. 3.257 Å) are slightly longer in comparison to 9 which is most likely a

direct consequence of the structural features and the reduced steric strain.

Table 9. Selected distances (Å) and comparison of Fe3Zn10 with 9-11.

Compound Fe-Fe Fe-Zn Zn-Zn

Fe3Zn10 av. 2.597 av. 2.621 av. 2.648

9 4.287 av. 2.544 av. 2.737

10 2.972(2) av. 2.485 av. 2.813

11 3.105 av. 2.444 av. 3.257

Notably, all metal-metal distances of 9-11, i.e. Fe-Fe, Fe-Zn and Zn-Zn, show more or less

significant deviations from the distances found in Fe/Zn intermetallic solid state phases, e.g.

Fe3Zn10 (av. 2.597 Å for Fe-Fe, av. 2.621 Å for Fe-Zn, av. 2.648 Å for Zn-Zn).[245] The fact


that only the distances of compound 10 are essentially comparable to Fe3Zn10 clearly points

out the stepwise appearance of structural similarities between molecular Fe/Zn compounds

and the corresponding intermetallic phases with a rising number of Zn atoms coordinated to

Fe (Table 9).

Synthesis and Characterisation of [(CO)3Co{µ2-Zn(L)2}(µ2-ZnCp*)2Co(CO)3] (L = py

(12), thf (13))

The reaction of [(CO)3Co(µ2-GaCp*)2Co(CO)3] with excess ZnMe2 in toluene at 90°C for

1h results in the formation of a colourless solid, which can be crystallised from mixtures

of either toluene/thf or toluene/pyridine to give [(CO)3Co{µ2-Zn(py)2}(µ2-

ZnCp*)2Co(CO)3] (12) and [(CO)3Co{µ2-Zn(thf)2}(µ2-ZnCp*)2Co(CO)3] (13) as yellow

and orange crystals in yields around 60% (Scheme 20). Compounds 12 and 13 are soluble

in the respective coordinating solvent as well as in the corresponding toluene mixtures but

almost insoluble in n-hexane/L mixtures. Both compounds are stable for several weeks

when stored at -30°C under an inert gas atmosphere in the glovebox.

[(CO)3Co(µ2-GaCp*)2Co(CO)3] 1. xs. ZnMe2, toluene, 1 h, 90°C2. L

Co Co

CO

COCO

Zn

ZnZn

OC

OCOC

L L

Cp* Cp*

L = py (12), thf (13)

L = py, thf

Scheme 20. Synthesis of [(CO)3Co{µ2-Zn(L)2}(µ2-ZnCp*)2Co(CO)3] (L = py (12), thf (13)).

The formation of 12 and 13 is based on the substitution of GaCp* by ZnR (R = Cp*, L2)

attended by Cp* transfer from gallium to zinc. The discussion of valence states as well as

electron count seems to be similar as it has been mentioned for compounds 9-11. The 1H

NMR spectrum displays the expected signals for the Cp* groups and the coordinated

solvent ligands (12: 1.95 (s, 30H, C5Me5), 7.57 (s, br, 4H, C5H5N), 7.97 (s, br, 2H,

C5H5N); 13: 1.77 (m, br, 8H, C4H8O), 1.95 (s, 30H, C5Me5), 3.61 (m, br, 8H, C4H8O)).

The 13C NMR spectroscopic data can be found in the Experimental Section. IR

spectroscopic measurements display the expected absorptions for the carbonyl ligands in


the range between 1806-1943 cm-1 for 12 and 1897-1956 cm-1 for 13, respectively.

Single Crystal X-Ray Analysis of [(CO)3Co{µ2-Zn(L)2}(µ2-ZnCp*)2Co(CO)3] (L = py

(12), thf (13))

In analogy to compound 11, the molecular structures of 12 and 13 show a threefold

capped cobalt dimer which consists of three bridging Zn ligands, namely two ZnCp* and

one {ZnL2} ligands (equivalent to the four electrons provided by two GaCp* ligands),

thus the 34 valence electron count is fulfilled. In contrast to 11, the carbonyl ligands in

both cobalt compounds are arranged in an eclipsed conformation in relation to the Co-Co

axis (Figure 25). Obviously, compounds 12 and 13 are the first examples that exhibit a

[Co2E3] core (E = any metal atom), although the structural element occurs in several

compounds in which E consists of a non-metal atom such as [(dppe)Co(µ2-

SPh)3Co(SPh)][246] (dppe = Ph2PCH2CH2PPh2). The Co-ZnL2 (L = py, thf) distances (12:

av. 2.426 Å, 13: av. 2.402 Å) are slightly shorter in comparison to the Co-ZnCp*

distances (12: av. 2.458 Å, 13: av. 2.491 Å). Most interestingly, all Co-Zn distances are

significantly shorter than those of terminal Co-Zn bonds observed in

[(CpZn)2Co(Cp)PPh3] (av. 2.586 Å).[191] The reduced coordination number of two at the

zinc in this latter compound might suggest shorter bonds and the contrary is observed for

12 and 13 with coordination numbers of three and four at Zn (Cp and Cp* treated as

occupying one coordination site). Notably, none of these distances match well with

distances in the intermetallic solid state phase CoZn13 (Table 10).

Table 10. Selected distances (Å) and comparison of CoZn13 with 12-13.

Compound Co-Co Co-Zn Zn-Zn

CoZn13 - av. 2.568 av. 2.680

4 2.949(1) av. 2.442 av. 3.367

5 3.011 av. 2.447 av. 3.233



displacement ellipsoids (Co, Zn) are shown at 50 % probability level, hydrogen atoms are omitted for clarity.

Selected interatomic distances (Å) and angles (°) for 12: Co1-Co2 2.949(2), Co2-Zn1 2.433(1), Co1-Zn1

2.420(2), Co1-Zn3 2.479(2), Co2-Zn3 2.468(2), Co1-Zn2 2.457(1), Co2-Zn2 2.429(2), Zn1-N11 2.069(8), Zn1-

N21 2.147(5), Zn2-Cp*centroid 2.045, Zn3-Cp*centroid 2.013, Co2-Zn2-Co1 74.25(5), Co2-Zn3-Co1 73.19(5), Co2-

Zn1-Co1 74.85(5), Zn2-Co2-Zn3 96.49(5), Zn1-Co1-Zn3 82.04(5), Zn1-Co2-Zn3 82.02(5). Selected interatomic

distances (Å) and angles (°) for 13: Co1-Co1 3.0109(6), Co1-Zn1 2.5092(5), Co1-Zn1 2.4535(5), Co1-Zn1

2.5092(5), Co1-Zn1 2.4535(5), Co1-Zn2 2.4016(5), Co1-Zn2 2.4016(5), Zn2-O4 2.050(2), Zn2-O4 2.050(2),

Zn1-Cp*centroid 2.025, Zn1-Cp*centroid 2.025, Co1-Zn2-Co1 77.64(2), Co1-Zn1-Co1 74.69(1), Co1-Zn1-Co1

74.69(1), Zn2-Co1-Zn1 90.98(2), Zn1-Co1-Zn1 73.27(1), Zn1-Co1-Zn2 92.35(2).

3.3.2 Case Study on the Formation of an Oligonuclear Model System for Intermetallic

Phases: Synthesis, Characterisation and Theoretical Investigations on the

Compound [Pd2Zn6Ga2(Cp*)5(CH3)3]

The synthesis and characterisation of novel heteroleptic d8 and d9 transition metal carbonyl

complexes 9-13 containing several bridging ZnRn (R = Cp*, Me, n = 1; R = py, thf, n = 2)

ligands clearly shows that variation of the transition metal, variation between monomeric and

dimeric starting compounds as well as a change of the solvent used for the crystallisation


processes have a direct influence on the product formation. In general, controllable M/E ratios

can be achieved under suitable conditions which can be helpful in the establishment of well-

defined, stoichiometric M/E Hume-Rothery phases. Unfortunately, crystallisation of the

initially formed solid has not been successful. Nevertheless, it seems that compounds 9-13

can be viewed as trapped, donor stabilized intermediate structures of larger aggregates formed

during the reactions. In order to study the same effects in comparison with mononuclear and

dimeric homoleptic compounds, the palladium dimer [Pd2(µ-GaCp*)3(GaCp*)2] has been

choosen for the first reactivity studies in order to obtain high nuclearity cluster.

Synthesis and Characterisation of [Pd2Zn6Ga2(Cp*)5(CH3)3] (14)

The treatment of [Pd2(µ-GaCp*)3(GaCp*)2] with one equivalent of ZnMe2 in toluene and

heating at 100°C for 1 h leads to a change in the colour of the solution from yellow-orange to

deep red. Red cubic crystals of [Pd2Zn6Ga2(Cp*)5(CH3)3] (14) can be isolated after standard

work-up and re-crystallisation from a saturated toluene solution at -30°C over a period of five

days (Scheme 21). The solid crystalline material is stable for several weeks when stored under

an inert gas atmosphere at -30°C. Pure crystals of compound 14 dissolve poorly in common

organic polar and non-polar solvents such as toluene, n-hexane and thf. During the dissolution

process slight decomposition occurs after a few hours.

Zn(CH3)2toluene, 100°C, 1h

[Pd2(GaCp*)2(µ-GaCp*)3] [Pd2Zn6Ga2(Cp*)5(CH3)3] (14)

Scheme 21. Synthesis of [Pd2Zn6Ga2(Cp*)5(CH3)3] (14).

The 1H NMR spectrum of the isolated crystals of 14 shows various broad peaks in the Cp*

area between 1.90 and 2.21 ppm which are obviously the result of the presence of two

structural isomers 14A and 14B in a ratio of approximately 1:3 (Figure 26). Herein, peaks

assigned to isomer 14A are located at 1.90, 1.94, 2.21 and 2.10 ppm with a ratio of 1:1:2:1,

whereas signals at 1.93, 2.17, 2.15 and 2.20 ppm with a ratio of 1:1:2:1 can be assigned to

isomer 14B. In addition, the CH3 groups give rise to two peaks for each isomer, i.e. at 0.04

(6H) and -0.06 (3H) for the major isomer 14B as well as at 0.49 (6H) and 0.05 (3H) for the

minor isomer 14A. Notably, both isomers show Cs symmetric 1H NMR spectra. The 1H NMR


spectrum shows no significant changes with respect to positions and signal width within a

temperature range of -80°C to +70°C. The 13C NMR spectrum displays various broad peaks in

the typical range of EMe (E = Ga, Zn) as well as ECp* ligands. A precise assignment of all

peaks to certain ligand positions is not possible on simple 1D NMR spectra. As mentioned

above, the stability decreases significantly if pure crystalline samples of compound 14 are re-

dissolved. Thus, all spectroscopic and spectrometric measurements have been carried out with

a minimum of spending time.

Figure 26. Cp* area of the 1H NMR spectrum (C6D6, rt) of 14: Isomer 14A (#), Isomer 14B (*).

Mass Spectrometric Analysis of [Pd2Zn6Ga2(Cp*)5(CH3)3] (14)

Since gallium and zinc are neighbours in the periodic table and show almost identical

diffraction behaviour under routine single crystal X-Ray diffraction conditions, the exact

elemental composition of 14 cannot be assigned only by single crystal X-ray diffraction.

Unfortunately, the elemental analysis and AAS data were rather ambiguous and consistent

with both, sum formulas of [Pd2Ga2Zn6(Cp*)5(CH3)3] for both isomers 14A and 14B and a

mixture of [Pd2Ga4Zn4(Cp*)5(CH3)3] (14A, major component) and [Pd2Ga2Zn6(Cp*)5(CH3)3]


(14B, minor component). Notably, both compositions would be also consistent with the 1H

NMR data (vide supra) found for 14, since several Cs symmetric structures are possible.

However, LIFDI-MS has been used to clearly point out the elemental composition of 14.

Herein, only one molecular ion peak [M].+ of 14 at m/z = 1465.903 (calc. 1465.876) has been

detected which displays an isotopic pattern consistent with the formula

[Pd2Ga2Zn6(Cp*)5(CH3)3] assigned to 14A and 14B (Figure 27). In comparison, the molecular

ion peak [M].+ of [Pd2Ga4Zn4(Cp*)5(CH3)3] would be expected at m/z = 1473.878 resulting in

changes of the observed isotopic pattern.

Figure 27. Above: Experimental LIFDI-MS spectrum of 14. Below: Calculated LIFDI-MS spectrum of 14.


Single Crystal X-Ray Analysis of [Pd2Zn6Ga2(Cp*)5(CH3)3] (14)

The allocation of gallium and zinc positions in the solid state structure of compound 14 has

been supported by quantum chemical calculations because gallium and zinc are not

unambiguously distinguishable by standard X-ray crystallography due to very similar

scattering power as mentioned above. The permutation of the metal ligand atoms in the metal

core [Pd2Zn6Ga2] results in 28 different isomers (for Computational Details see Experimental

Section 6.3.1). From 1H NMR spectroscopy, 20 of these isomers could be ruled out because

they do not exhibit Cs symmetry (Table ES1, 6.3.1). Geometry optimizations without

symmetry constraint of the model system [Pd2Zn6Ga2(Cp)5(CH3)3] (14M) where Cp* is

substituted by Cp were performed for the eight Cs isomers (Figure 28, left).

Figure 28. Left: Modelled structure of [Pd2Zn6Ga2(Cp)5(CH3)3]. The numbers indicate the respective positions of

the Ga/Zn atoms. Right: Calculated structure of 14M-3/8. Experimental bond lengths of 14 are given in

parentheses.

The energetically lowest-lying structure is 14M-3/8. There are two isomers 14M-4/8 (ΔEo =

+4.0 kcal/mol) and 14M-5/8 (ΔEo = +5.1 kcal/mol) which are slightly less stable than 14M-

3/8. Two other forms 14M-3/4 (ΔEo = +10.2 kcal/mol) and 14M-4/5 (ΔEo = +19.2 kcal/mol)

are significantly higher in energy. The geometry optimization of the remaining three isomers

yielded severely distorted structures which no longer have Cs symmetry. The comparison of

the calculated bond lengths for 14M-3/8 with the experimental values for 14 which were

obtained from X-ray diffraction correlate well with the numerical data (Figure 28, right). It

should be noted, that the calculated bond lengths of the energetically higher lying isomers

14M-4/8 and 14M-5/8 contain significantly large deviations from the experimental values

(Figures ES1 and ES2, 6.3.1). Complex 14 crystallizes in the monoclinic spacegroup P21/n. In

the structure refinement two isomers have been considered with 50% occupancy each (14M-


3/8 and 14M-4/8), since it is not clear which of the two isomers represents the major isomer

found in the 1H NMR spectrum. Thus, the crystal structure represents an average structure of

two possible isomers. Therefore the structural discussion is restricted to the overall

coordination geometry of 14 without any discussion on bond lengths. The main focus will be

given to the electronic situation in compound 14.

Figure 29. Above: Molecular structure of 14 in the solid state. Ellipsoids (Pd, E = Zn, Ga) are set at 50%

probability; hydrogen atoms are omitted for clarity. Below: Illustration of the distorted bi-capped trigonal prism.

Note: The solid lines between the Zn atoms are drawn to guide the eye and should not be immediately associated

with (strong) covalent Zn-Zn bonding interactions. Numbering of the atoms has been in accordance to the

theoretical model systems.


The molecular structure of 14 can be best described as a bi-capped trigonal prism in which

one palladium atom (Pd1) is embedded in the center of a Pd/Zn/Ga trigonal prism, with the

other palladium atom (Pd2) as well as one E(1)Me ligand as capping ligands. The

coordination environment of Pd2 is best described as square pyramidal. The Pd-Pd distance

(2.668(1) Å) is slightly longer than in the dimeric starting material [Pd2(GaCp*)2(µ-GaCp*)3]

(2.609(1) Å). The bond distances of the terminal ECp* (E = Ga, Zn) units and the palladium

atoms (Pd2-ECp*terminal 2.383(1) Å, Pd1-ECp*terminal 2.406(1) Å) are significantly shorter than

the Pd-ECp*bridging fragments (average 2.461 Å for Pd1-ECp*bridging, average 2.685 Å for Pd1-

ECp*bridging). Interestingly, this trend is inversed for the Pd-E(CH3) distances (average 2.391

Å for Pd1-E(CH3)terminal, average 2.438 Å for Pd-E(CH3)bridging). The distances between

surrounding E atoms range from 2.837(1) Å to 3.156(1) Å.

Theoretical Investigations on [Pd2Zn6Ga2(Cp*)5(CH3)3] (14). In order to gain an insight

into the bonding situation and metal-ligand interactions of 14M-3/8 theoretical calculations

have been performed. A detailed description of the bonding situation with respect to a

descriptive MO correlation diagram of this 30 valence electron cluster (20e Pd, 6e Zn, 4e Ga)

is not possible due to the complexicity of large systems as is the case for 14. The investigation

of the occupied 23 MOs (46 electrons: 20e from Pd, 12e from Zn, 6e from Ga, 8e from H)

shows that the breakup into 15 CVMOs and 8 E-H MOs is arbitrary. Thus, the primary

investigations focussed on the description of the metal-metal bond in 14 and the description or

rather relations to known 18 valence electron complexes. However, previous studies of highly

coordinated zinc-rich mononuclear species have shown that all such complexes fulfil the 18

valence electron rule if ZnR (R = Cp*, Me) is counted as a 1e donor ligand and GaR as a 2e

donor ligand. Thus, related to compound 14, Pd1 would consist of 18 valence electrons

whereas Pd2 would contain only 17 valence electrons. Now the question arises whether

attractive interactions between both palladium centres would achieve a full electron

configuration of Pd2. Calculations have been made with a reduced model compound 14M-

3/8-H where the optimized geometry of the core atoms (Pd, Zn, Ga) is frozen but the

substituents Cp and Me are replaced by hydrogen atoms as previously discussed for the

mononuclear compounds.[43] First studies on the electronic structure have been carried out

using the AIM method. The AIM analysis shows Pd-Zn bond paths between palladium and

the bridging Zn atoms and a Pd2-Ga8 bond path for the terminal gallium ligand (Figure 30,

left). The bond paths between Pd1 and the terminal Zn atoms do not show up because the

latter ligand atoms are not in the same molecular plane. The most important finding is the


occurrence of a Pd1-Pd2 bond path which supports the assignment of a bonding interaction. It

should be noted that a bond path is not a sufficient criterion for a chemical bond and likewise,

the absence of a bond path does not exclude chemical interactions. However, the molecular

graph and contour map of 14M-3/8-H in the perpendicular molecular plane bisecting the

palladium atoms and the bridging ligand atoms Ga3 and Zn4 (Figure 30, right) shows again a

Pd1-Pd2 bond path as well as three bond paths between the palladium atoms and the ligand

atoms. But there is no Pd2-Zn4 bond path although the Zn4 ligand atom is clearly in a

bridging position between both palladium atoms. The absence of a Pd2-Zn4 bond path

indicates that the distortion of the electronic structure at Pd2, which is caused by the Pd2-Zn4

interactions, is not strong enough to establish a bond path, but it does not exclude significant

attractive interactions.

Figure 30. Left: Molecular graph and contour map of the Laplacian ∇2ρ(r) of 14M-3/8-H in the molecular plane

bisecting the palladium atoms and the bridging ligand atoms Zn6 and Zn7. Right: Molecular graph and contour

map of the Laplacian ∇2ρ(r) of 14M-3/8-H in the molecular plane bisecting the palladium atoms and the ligand

atoms Ga3 and Zn4. Solid lines indicate areas of charge concentration (∇2ρ(r) < 0) while dashed lines show

areas of charge depletion (∇2ρ(r) > 0). The thick solid lines connecting the atomic nulei are the bond paths. The

thick solid lines separating the atomic basins indicate the zero-flux surfaces crossing the molecular plane.

In order to verify the findings of the AIM analysis, the bonding situation of 14M-3/8-H has

been studied with respect to the molecular orbitals analysis. Most interestingly, the inspection

of the shape of the MOs reveals that there exist two orbitals which have dominantly Pd-Pd

bonding character (Figure 31), i.e. HOMO-18 and HOMO-19. Herein, HOMO-18 features a

bonding combination of the dxz AOs of the metals while HOMO-19 is a bonding combination

of the dz2 AOs with additional bonding contributions from the bridging Ga atom. The shape of


the orbitals and the appearance of the bond path strongly support the assignment of a Pd-Pd

bond in 14M-3/8-H and thus, also in the original complex 14.

Figure 31. Orbital shapes of HOMO-18 (left) and HOMO-19 (right) of 14M-3/8-H (BP86/def2-TZVPP).

The calculated atomic partial charges taken from NBO calculations suggest a large negative

charge for Pd1 (-2.99e) and a smaller negative charge for Pd2 (-0.76 e). This indicates that the

former palladium atom can donate electronic charge to the latter Pd which obtains an 18

valence electron configuration (Table 11).

Table 11. NBO partial charges q of the isomer 14M-3/8 at BP86/def2-TZVPP.

Atom q [e]

Pd1 -2.99

Pd2 -0.76

Zn1 (ZnMe) +1.04 (+0.50)

Zn2 (ZnMe) +1.04 (+0.50)

Zn4 (ZnCp) +0.93 (+0.44)

Zn5 (ZnMe) +0.94 (+0.40)

Zn6 (ZnCp) +0.89 (+0.38)

Zn7 (ZnCp) +0.89 (+0.24)

Ga3 (GaMe) +1.18 (+0.72)

Ga8 (GaCp) +0.92 (+0.42)

In summary, the very good agreement of the calculated bond lengths and the energy

calculations strongly support the assignment of the ligand atoms in 14M-3/8 for the major


isomer 14A. The theoretical evidence for the identification of the minor isomer 14B is not as

strong as it is for 14A. The energy calculations suggest that 14M-4/8 is the best candidate for

the ligand assignment of the minor isomer but 14M-5/8 is only slightly higher in energy and

thus, it can not be excluded. There is a Pd-Pd bond in 14M-3/8 which exhibits a bonding

situation where the 18 valence electron rule is fulfilled for both palladium atoms. The

importance of attractive M-M bonding interactions on structural situations and metal

coordinations is known for intermetallics and well represented in compound 14. The

formation of compound 14 shows that cluster growth is possible using dimeric homoleptic

starting materials and represents the first example on the way to more extended structures as

molecular models of intermetallics. The exchange of Ga/Zn positions which can be observed

in solid state phases is illustrated in 14 due to the existence of two isomers determined by

NMR and LIFDI-MS studies.

[f] 3.4.1: T. Bollermann, K. Freitag, C. Gemel, R. W. Seidel, M. von Hopffgarten, G. Frenking, and R. A. Fischer, Angew. Chem. Int. Ed. 2011, 50(3), 772-776. 3.4.2; 3.4.3: T. Bollermann, K. Freitag, C. Gemel, R. W. Seidel, and R. A. Fischer, Organometallics 2011, 30 (15), 4123-4127. 3.4.2: T. Bollermann, K. Freitag, C. Gemel, R. W. Seidel, M. von Hopffgarten, G. Frenking, and R. A. Fischer, Chem.-Eur. J. 2011, manuscript in preparation. 3.4.3: T. Bollermann, K. Freitag, C. Gemel, R. W. Seidel, M. von Hopffgarten, P. Jerabek, G. Frenking, and R. A. Fischer, Inorg. Chem. 2011, submitted.


3.4 Experimental and Theoretical Investigations on the Coordination

Chemistry of [Zn2Cp*2] Towards Transition Metal Compounds[f]

Abstract

This chapter deals with initial investigations on the coordination chemistry of [Zn2Cp*2]

towards a series of transition metal compounds. First reactions are based on the use of

homoleptic GaCp* containing d10 metal complexes [M(GaCp*)4] (M = Pd, Pt) leading to a

product mixture of the hexa coordinated complexes [M(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (M =

Pd (15), Pt (17)) and the octa coordinated complexes [M(ZnCp*)4(ZnZnCp*)4] (M = Pd (16)

Pt (18)). Compounds 15-18 exhibit the novel ligand system {ZnZnCp*} featuring fully intact

Zn-Zn interactions. Possible reaction schemes involve dissociation/association equilibria as

determined by in situ NMR spectroscopy.

In addition, further investigations on the reactivity of [Zn2Cp*2] towards the reactive

transition metal complexes in the absence of GaCp* are investigated. The reactions of

[M(cod)2] (M = Ni, Pt) with eight equivalents [Zn2Cp*2] results in the formation of

[M(ZnCp*)4(ZnZnCp*)4] (M = Pt (18), Ni (19)). In situ NMR spectroscopy shows several

competing reaction steps including the liberation of 1,3-cod, the formation of ZnCp*2 as well

as [Cp*M(ZnCp*)3] (M = Ni (20), Pt (21)) as a rare by-product. The possible reaction

pathway in the formation of 18 and 19 seems to be best described as being quite similar to 15-

18. The selective preparation of [Cp*M(ZnCp*)3] (M = Ni (20), Pt (21)) can be obtained

using two equivalents of [Zn2Cp*2] and lower temperatures. Herein, the Zn(I) dimer acts as a

smooth oxidizing agent for the transition metal centres as well as a natural source in the

formation of ZnCp* ligands.

Finally, the reaction of [PdMe2(tmeda)] with four equivalents [Zn2Cp*2] yields in the

formation of [Cp*Pd(ZnCp*)3] (22) and [Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}] (23) as main

products in a ratio of 1:1. In situ 1H NMR spectroscopy displays several (redox chemical)

reaction sequences and by-product formation of Zn(II) species and [Pd(ZnMe)4(ZnCp*)4].


Compound 23 features the first known compound with an unsupported 2e Zn(0)L donor

ligand as described by theoretical investigations based on AIM and EDA analysis.

Introduction

The field of low valent main group metal chemistry has become one of the most attractive and

challenging fields of inorganic chemistry.[247-250] Since the last decades of the 20th century a

wide variety of uncommon molecules have been synthetically discovered which were thought

to be impossible to prepare.[251-259] For instance, in 1976, Lappert´s findings caused rethinking

about the well accepted position, that multiple bonds between heavier group 14 elements are

impossible as suggested within the double bond rule.[260, 261] The synthesis and

characterisation of the first multiple bond between heavier main group elements succeeded in

the preparation of a stabilized E=E (E = Ge, Sn) double bond found in [E{CH(SiMe3)2}]2 in

the solid state. However, the synthesis of [Zn2Cp*2] by Carmona, which describes the first

molecular compound exhibiting a strong covalent Zn(I)-Zn(I) bond, aroused much interest

and stimulated the research on low coordinated (main group) metal compounds as well.[138]

Besides the multitude of theoretical investigations on the electronic structure of [Zn2Cp*2],

first reactivity studies have been focused on the formation of derivative structures [Zn2R2] (R

≠ Cp*) leading to a wide variety of novel compounds consisting of a [Zn2]2+ core via ligand

exchange reactions and reduction steps starting from [RZnX]a (see chapter 2.3). Quite recently

initial results on the chemistry of [Zn2Cp*2] leading to Lewis acid/base adducts with fully

intact Zn(I)-Zn(I) bonds and a Lewis base stabilized [Zn2]2+ cation were reported. In contrast

to the well established use of ZnR2 in organic synthesis, the formation of transition metal-zinc

bonds remained unexplored until the middle of the 20th century. In 1942, Hieber synthesized

the first M-Zn bond in Zn[Co(CO)4]2.[185] In the 1980s the groups of Boersma and van der

Kerk isolated a series of carbonyl free complexes featuring direct M-ZnR bonds.[143, 191-197] In

2008, the access to very zinc-rich, highly coordinated [M(ZnR)n] compounds (n ≥ 8; M = Mo,

Ru, Rh; Ni, Pd, Pt; R = CH3, Cp*) was reported, which can be viewed as a linking step

between (coordination) compounds, clusters and intermetallic phases of the Hume-Rothery

type.[42, 43] In any case, the preparation starts with mononuclear complexes [M(GaCp*)m] and

ZnR2 (R = Me, Et) and involves two exchange processes, i.e. Ga/Zn and Cp*/R. Mechanistic

insides reveal redox chemical processes in which Zn(II) is reduced to Zn(I) by Ga(I) which in

turn is oxidized to Ga(III). The overall substitution of one 2e GaCp* ligand by two 1e ZnR

ligands in the coordination sphere of the transition metal centre results in such high

coordination numbers.


3.4.1 Trapping Monovalent {ZnZnCp*} at d10 Transition Metal Centres

The coordination chemistry of low valent group 13 metal ligand systems E(I)R (E = Al, Ga,

In; R = Cp* and other bulky substituents) towards transition metal centres and their

application as key compounds in the formation of M-ZnR bonds has been well known. In

contrast, the coordination chemistry of low valent main group metals of the general formula

[M2L2] and, in particular, [Zn2Cp*2] has been considerably less extensively investigated. In

this context it is highly interesting to study the coordination chemistry of Carmona´s

[Zn2Cp*2] towards transition metal centres. Referring to the previously mentioned results in

the chemistry of ZnMe2, first reactivity studies include the use of GaCp* containing d10 metal

compounds [M(GaCp*)4] (M = Pd, Pt).

Synthesis and Characterisation of [M(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (M = Pd (15); M =

Pt (17)) and [M(ZnCp*)4(ZnZnCp*)4] (M = Pd (16); M = Pt (18))

The reaction of [Pd(GaCp*)4] with four equivalents [Zn2Cp*2] in toluene at 95°C over a

period of 2 h leads to the formation of a product mixture containing the hexa-coordinated

complex [Pd(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (15) as well as the octa-coordinated complex

[Pd(ZnCp*)4(ZnZnCp*)4] (16) in a molar ratio of approximately 6:1 as revealed by in situ 1H

NMR spectroscopy. Similarly, [Pt(GaCp*)4] reacts with four equivalents [Zn2Cp*2] in

benzene at 75 °C over a period of 2 h quantitavely to form a mixture of the analogous

platinum compounds [Pt(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (17) and [Pt(ZnCp*)4(ZnZnCp*)4]

(18) (Scheme 22).

[M(GaCp*)4] 4 [Zn2Cp*2] [M(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (M = Pd (15), Pt (17)) + [M(ZnCp*)4(ZnZnCp*)4] (M = Pd (16), Pt (18))

(M = Pd, Pt) Reaction conditions:

15, 16: toluene, 95°C, 2 h17, 18: benzene, 75°C, 2 h

Scheme 22. Synthesis of [M(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (M = Pd (15), Pt (17)) and

[M(ZnCp*)4(ZnZnCp*)4] (M = Pd (16), Pt (18)).


All complexes are stable in solution from room temperature up to higher temperatures (15 and

16: 100°C; 17 and 18: 80°C) over a longer period. In contrast, decomposition occurs rapidly

within a few seconds after separation of pure crystalline material from the supernatant mother

liquor and subsequent drying. Thus, to obtain more stability it is advisable to cover the

crystalline material with a small amount of solvent. Analytically pure compounds can be

obtained by manual separation of crystalline material under the microscope in the glove box

in the presence of effectual amounts of solvent. Nevertheless, the full characterisation of

compounds 16 and 18, which are rare by-products in this reaction pathway, could not be done

because it is not possible to collect substantial amounts of them before decomposition. The 1H

NMR spectroscopic data of 15 and 17 in C6D6 at room temperature displays three sharp

signals with an intensity ratio of 1:1:1 (Table 12). In addition, the 13C NMR spectra show no

unusual features with respect to the expected signal pattern for three non equivalent Cp*

groups assigned to the following three substituents GaCp*, ZnCp* and {ZnZnCp*}. The

Ga/Zn contents of 15 and 17 were obtained by atomic absorption spectroscopy (Gallium

content for 15: calculated 9.6 wt.%, found 9.5 wt.%; Gallium content for 17: calculated 8.5

wt.%, found 9.4 wt.%).

Table 12. NMR spectroscopic data (C6D6, rt) of 15 and 17.

Compound 1H NMR

13C NMR

C5Me5 C5Me5 15 1.87 (s, 30 H)

2.15 (s, 30 H)

2.31 (s, 30 H)

10.17 108.81

10.72 109.66

11.98 113.91

17 1.77 (s, 30 H)

2.10 (s, 30 H)

2.11 (s, 30 H)

10.93 111.72

12.37 114.72

13.39 118.58

Mechanistic studies. To gain an insight into possible reaction mechanisms forming

compounds 15-18, the reaction has been followed via 1H NMR spectroscopy which will be

discussed exemplary for 15 and 16 (reaction conditions: sealed and gastight Low Pressure

and Vacuum J-Young-NMR tubes; C6D6, 75°C, 1.5 h). Beside the signals of pure 15, two


further signals aroused at 2.19 and 2.31 ppm with an intensity ratio of 1:1 which can be

assigned to the ZnCp* and {ZnZnCp*} ligands of compound 16. Note that the assignment of 13C NMR data for 16 failed due to the low concentration of 16 in the product mixture.

Fulvalene species, which could be a result of dimerization of free Cp* radicals were not

observed, in contrast to the reaction schemes discussed for ZnMe2.[42,43] In addition, the 1H

NMR spectrum displays a broad peak at 1.90 ppm which has been assigned to the coalescence

peak of free GaCp* and the known Zn(II) species ZnCp*2 most probably via formation of an

unstable fluxional intermediate such as {Cp*Ga••••ZnCp*2}. This suggestion has been

verified via NMR spectroscopic studies of a 1:1 mixture of GaCp* and ZnCp*2 in C6D6 giving

rise to the same broad signal in the 1H NMR spectrum (Figure 32).

Figure 32. NMR reaction of ZnCp*2 with GaCp* (1:1 ratio; C6D6, rt). Above: 13C NMR spectrum. Below: 1H

NMR spectrum.

Neither free GaCp*, nor ZnCp*2 have been detected in previous studies on the synthesis of

metal rich compounds [M(ZnR)n] as mentioned in the introduction. However, the existence of

free GaCp* and ZnCp*2 as by-products in this reaction scheme is in accordance to the

expectation that redox chemical processes can be disregarded. Thus, the formation of 15 and

16 should be most likely described as dissociation/association mechanisms, i.e. dissociation of


GaCp* from [Pd(GaCp*)4] and trapping of monovalent ZnCp*, at electronically and

coordinatively unsaturated palladium centres.

[M(GaCp*)4][Zn2Cp*2]- GaCp*

[LaM(ZnCp*)b] (L = GaCp*, ZnCp*)

[Zn2Cp*2] [LaM(ZnZnCp*)b] (L = GaCp*, ZnCp*)- [ZnCp*2], GaCp*

Scheme 23. Possible reaction mechanism in the formation of 15-18.

These dissociation/association steps are also in accordance with the evidently low yields of 16

in comparison to 15. The existence of free GaCp* shifts the dissociation equilibrium to the

adduct side (15) which avoids the formation of Pd-Zn bonds with increasing product

concentration. Finally, Cp* transfer from the starting compound [Zn2Cp*2] towards Zn

centres of lower coordinated intermediate species of the type [LaPd(ZnCp*)b] (L = GaCp* or

ZnCp*) leads to the release of ZnCp*2 and coevally the formation of the novel one electron

ligand {ZnZnCp*} at which the Zn-Zn bond seems to be fully intact. Notably, no reaction can

be obtained between [Zn2Cp*2] and GaCp*, neither adduct formation, nor decomposition is

observed by in situ NMR spectroscopy (Scheme 23).

Single Crystal X-Ray Analysis of [M(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (M = Pd (15); M =

Pt (17))

As mentioned before, Ga and Zn are not unambiguously distinguishable by standard X-ray

crystallography due to very similar scattering power. Thus, the allocation of Ga and Zn

positions in the solid state structures of compounds 15 and 17 has been supported by quantum

chemical calculations.[262] Herein, single point calculations at BP86/def2-TZVPP on all

possible positions of isomers 15 and 17 have been performed (for Computational Details see

Experimental Section 6.3.2 and Ref. [262]). The crystal structures of 15 and 17 have been

used with replacement of Cp* by Cp yielding in the model systems 15H and 17H (Figure 33).

Hereafter, the isomer with the lowest energy has been assigned to feature Ga atoms at the

positions 1 and 2 (isomer 1_2), being the trans positions to the {MMCp} ligands, which is in

agreement with the assignment of Ga and Zn atoms in figure 34. This result has been

confirmed by further single point calculations on selected isomers of the full system 15

including Cp* rings and by means of different density functionals. The results show eight

isomers with higher energies by less than 15 kcal/mol in comparison to isomer 1_2, namely

1_3, 1_4, 1_6, 1_8, 2_3, 2_4, 2_6 and 2_8.


Figure 33. Assignment of numbers of the Ga/Zn positions in the model system 15H.

Table 13. Relative energies of eight isomers (ΔE ≤ 15kcal/mol) of the model systems 15H (left) and 17H (right)

at BP86/def2-TZVPP. Isomers are denoted by the Ga-positions, i.e. isomer 1_2 has Ga atoms at positions 1 and

2 as assigned in figure 34. Relative energies ΔE with respect to the lowest energy isomer in kcal/mol.

Isomer

of 15H

ΔE Isomer

of 17H

ΔE

1_2 0.00 1_2 0.00

1_3 9.93 1_3 8.81

1_4 11.98 1_4 12.08

1_6 10.37 1_6 11.62

1_8 9.93 1_8 8.89

2_3 11.11 2_3 9.74

2_4 13.39 2_4 14.30

2_6 12.59 2_6 12.04

2_8 11.11 2_8 9.83


As 15 and 17 are almost Cs symmetric, these eight isomers can be collected as pairs of pseudo

mirror images. These are 1_3 and 2_3, 1_8 and 2_8, 1_4 and 2_6, 1_6 and 2_4. The

comparison of respectively one member of each pair has shown that in all cases isomer 1_2 is

the one with the lowest energy.

Table 14. Relative energies of the seven lowest energy isomers of compounds 15 (above) and 17 (below) at

different levels of theory. Isomers are denoted by the Ga-positions, i.e. isomer 1_2 has Ga atoms at positions 1

and 2 as assigned in figure 34. Relative energies ΔE with respect to the lowest energy isomer in kcal/mol.

Isomer of 15 ΔE [kcal/mol]

BP86/def2-TZVPP

ΔE [kcal/mol]

B3LYP/def2-TZVPP

ΔE [kcal/mol]

M05/def2-TZVPP

ΔE [kcal/mol]

M05-2X/def2-TZVPP

1_2 0.0 0.0 0.0 0.0

1_3 9.9 10.7 10.9 11.1

1_4 15.7 16.1 17.3 14.0

1_6 13.7 14.0 14.3 10.9

1_8 9.9 10.7 10.9 11.1


BP86/def2-TZVPP

ΔE [kcal/mol]

B3LYP/def2-TZVPP

ΔE [kcal/mol]

M05/def2-TZVPP

ΔE [kcal/mol]

M05-2X/def2-TZVPP

1_2 0.00 0.00 0.00 0.00

1_3 8.28 8.69 8.52 9.16

1_4 16.03 16.16 17.09 14.71

1_6 15.13 14.94 14.91 12.05

1_8 8.20 8.60 8.42 9.07

Thus, the assignment revealed from structural data with respect to Pd-Zn and Pd-Ga bond

distances agree well with the corresponding bond distances in other comparable compounds

(vide infra). In addition, the results from DFT calculations strongly support the assignment of

the Ga and Zn positions as given in Figure 34.


Single crystals of 15-18 (15 and 17: orange crystals, 16 and 18: red needle shaped crystals)

suitable for X-ray diffraction studies can be obtained by slow cooling of a saturated toluene

solution (15, 16) down to -30°C or rather cooling a benzene solution (17,18) down to 5°C.

The molecular structures of the [MGa2Zn6] cores of 15 and 17 emphasize the octahedral

arrangement of the six Ga/Zn ligands around the central metal atom. Figure 34 shows the

structure of 15 in the solid state as a representative for both compounds.


displacement ellipsoids (Pd, Ga, Zn) are shown at 50 % probability level, hydrogen atoms are omitted for clarity.

Selected interatomic distances (Å) and angles (°) for 15: Pd1-Ga1 2.359(1), Pd1-Ga2 2.360(1), Pd1-Zn1

2.448(1), Pd1-Zn5 2.375(1), Pd1-Zn2 2.379(1), Zn5-Zn4 2.345(1), Zn2-Zn3 2.346(1), Zn1-Cp*centroid 1.954,

Zn4-Cp*centroid 1.938, Zn3-Cp*centroid 1.941, Ga1-Cp*centroid 1.954, Ga2-Cp*centroid 1.969; Zn1-Pd1-Zn1`129.73(4),

Ga2-Pd1-Ga1 103.76(4), Ga2-Pd1-Zn5 86.39(4), Zn5-Pd1-Zn2 88.29(3), Zn2-Pd1-Ga1 81.56(4), Zn4-Zn5-Pd1

170.31(5), Zn3-Zn2-Pd1 170.45(5), Ga2-Pd1-Zn2 174.68(4), Ga1-Pd1-Zn5 169.85(4), Cp*centroid-Zn4-Zn5

175.62, Cp*centroid-Zn2-Zn3 178.79, Cp*centroid-Zn1-Pd1 169.68, Cp*centroid-Ga1-Pd1 176.10, Cp*centroid-Ga2-Pd1

174.85. Selected interatomic distances (Å) and angles (°) for 17: Pt1-Ga1 2.347(3), Pt1-Ga2 2.360(3), Pt1-Zn1

2.403(3), Pt1-Zn5 2.383(2), Pt1-Zn2 2.4152(16), Zn5-Zn4 2.335(3), Zn1-Zn3 2.359(4), Zn3-Cp*centroid 1.987,

Zn4-Cp*centroid 1.934, Zn2-Cp*centroid 1.983, Ga1-Cp*centroid 1.957, Ga2-Cp*centroid 1.991; Zn2-Pt1-Zn2` 150.09(9),

Ga2-Pt1-Ga1 100.31(13), Ga2-Pt1-Zn5 85.36(10), Zn5-Pt1-Zn1 84.93(9), Zn1-Pt1-Ga1 89.40(12), Zn4-Zn5-Pt1

173.88(14), Zn3-Zn1-Pt1 174.03(14), Ga2-Pt1-Zn1 170.28(10), Ga1-Pt1-Zn5 174.33(12), Cp*centroid-Zn4-Zn5

179.30, Cp*centroid-Zn3-Zn1 167.92, Cp*centroid-Zn2-Pt1 172.28, Cp*centroid-Ga1-Pt1 179.72, Cp*centroid-Ga2-Pt1

168.49.


The coordination geometry around the Pd and Pt atoms in 15 and 17 can be best described as

a distorted octahedron featuring strong deviation from linearity for the axial Zn-M-Zn angles

of the Cp*Zn-M-ZnCp* units (15: 129.73(4)° for Zn1-Pd1-Zn1’; 17: 150.09(9)° for Zn2-Pt1-

Zn2’). The deviations of all other characteristic bond angles from the ideal octahedral

structure are rather small. The Zn-Zn-M units are almost linear (15: 170.31(5)° for Zn4-Zn5-

Pd1 and 170.45(5)° for Zn3-Zn2-Pd1; 17: 174.03(14)° for Zn3-Zn1-Pt1 and 173.88(14)° for

Zn4-Zn5-Pt1). As expected, the Cp*centroid-E-M (E = Ga, Zn) angles show small deviations

from linearity. However, stronger discrepancy is observed for Cp*centroid-Zn1-Pd1 (169.68°) in

15 and Cp*centroid-Ga2-Pt1 (168.49°) in 17 leading to the suggestion, that deviation of the

inner [ME6] cores from the ideal octahedral structure can be basically explained by steric

reasons caused by the bulky Cp* ligands. The Zn-Zn bond distances display values of

2.345(1) and 2.346(1) Å in 15 and 2.359(4) and 2.335(3) Å in 17 and thus matching well with

the Zn-Zn bond length in the parent compound [Zn2Cp*2][138] (2.331 Å). The M-Ga bonds

(15: 2.359(1) and 2.360(1) Å; 17: 2.347(3) and 2.360(3) Å) are comparable with M-Ga bond

distances in the homoleptic starting materials [M(GaCp*)4][128] (M = Pd: 2.354(1) Å; M = Pt:

2.335(2) Å). Notably, significant differences are confirmed within the Pd-Zn bond distances.

The Pd-ZnCp* distances (av. 2.448(1) Å) are comparable to the Pd-Zn bond lengths in the

known compound [Pd(ZnCp*)4(ZnMe)4][43] (2.447(1)-2.459(1)Å), whereas the Pd-ZnZnCp*

distances are distinctly shorter (2.375(1) and 2.379(1) Å). The reasons for this can be (1)

decreased steric demand of the {ZnZnCp*} group with respect to the ZnCp* ligand or (2)

higher π-character of the interaction between the naked Zn atom to the Pd. Notably, these

features are not as pronounced in compound 17. However, the Cp*centroid-E (E = Ga, Zn)

distances (15: 1.954 Ǻ for Cp*centroid-Zn1 and av. 1.962 Ǻ for Cp*centroid-Ga; 17: 1.983 Ǻ for

Cp*centroid-Zn2 and av. 1.974 Ǻ for Cp*centroid-Ga) are almost similar. Both data sets show only

slight deviations from the reference compounds [M(ZnCp*)4(ZnMe)4][43] (M = Pd: Cp*centroid-

Zn 1.934 Å; M = Pt: Cp*centroid-Zn 1.931 Å) and [M(GaCp*)4][128] (M = Pd: 2.019 Å; M = Pt:

2.007 Å) but are otherwise longer than the Cp*centroid-ZnZn distances (15: av. 1.940 Ǻ; 17: av.

1.961 Å). In contrast, the Cp*centroid-ZnZn distances for 15 and 17 are distinctly shorter in

comparison to [Zn2Cp*2] (2.04 Å) which can be assigned as an indirect measurement of the

increased ionic character for coordinated {ZnZnCp*} units.


Table 15. Important interatomic distances (Å) for 15 and 17.

Compound M-ZnCp* M-GaCp* M-ZnZnCp* Zn-Cp*centroid Ga-Cp*centroid ZnZn-Cp*centroid

15 2.448(1) 2.359(1)-

2.360(1)

2.375(1)-

2.379(1)

1.954 1.954-1.969 1.938-1.941

17 2.4152(16) 2.347(3)-

2.360(3)

2.383(2)-

2.403(3)

1.983 1.957-1.991 1.934-1.987

Single Crystal X-Ray Analysis of [Pd(ZnCp*)4(ZnZnCp*)4] (16)

The molecular structure of 16 as determined by single crystal X-ray diffraction is shown in

Figure 35. As it was mentioned above, fully zinc substituted compounds 16 and 18 have only

been observed as rare by-products and as rapidly decomposed crystalline material. Thus,

isolation of substantial amounts suitable for AAS was not successful, so that the existence of

Ga can not be unambiguously ruled out and is mainly based on structural and computational

analysis in addition to heuristic reasons based on electron counting rules known from the

homoleptic and heteroleptic 18 valence electron compounds [M(ZnR)n] as well as

[M(ZnR)n(GaR)m].[43, 44] In addition, only low quality crystals of compound 18 could be

obtained. The structure determination of 18 by the use of single crystal X-ray analysis

displayed a disordered structure which could not be sufficiently refined. Thus, only the

connectivity of the heavy atoms, i.e. Pt and Zn, in the solid state structure could be

satisfactorily assigned. Therefore, discussion of the fully zinc substituted compounds in the

following section is restricted to compound 16.

The polyhedral geometry of the inner [PdZn8] core of 16 can be best described as a slightly

distorted trigonal dodecahedron, well comparable to the [PdZn8] core determined in the

known compound [Pd(ZnCp*)4(ZnMe)4].[43] The Zn-Zn bond distances of the four ZnZnCp*

units (2.347(3)-2.351(3) Å) are in good agreement with the Zn-Zn bond distance found in the

starting compound [Zn2Cp*2][138] (2.331 Å) and compound 15 (2.345(1) and 2.346(1) Å).

Notably, the Pd-ZnZn distances, ranging from 2.473(2) to 2.478(2) Å, are significantly

elongated in comparison to 15 as well as in comparison to the Pd-ZnCp* distances (2.422(2)-

2.433(2) Å). The Zn-Cp*centroid distances are all very similar with average values of 1.993 Å

for ZnCp* and 2.008 Å for the ZnZnCp* units. The Pd-Zn-Zn (Pd1-Zn3-Zn4 179.69(13)°,

Pd1-Zn5-Zn6 177.70(11)°) and Pd-Zn-Cp*centroid (Pd1-Zn1-Cp*centroid 176.53°, Pd1-Zn2-


Cp*centroid 179.19°) bond angles are almost linear. Additionally, linearity can be also observed

for Zn-Zn-Cp*centroid units with an average angle of 178.31°. The interior penta metal atom

chains Zn-Zn-Pd-Zn-Zn (134.40(12)° for Zn5-Pd1-Zn5`and 131.40(11)° for Zn3-Pd1-Zn3`)

display significant deviations from linearity most probably due to the trigonal dodecahedral

geometry found in 16. Molecular compounds, featuring such finite metal chains can be found

quite often in the literature mainly supported by additional ligands in bridging positions and in

coordination polymers of several metal/ligand combinations.[263-268] In contrast, compound 16

can be added to the class of rarely observed oligo (hetero-) metal chains without stabilisation

of the internal atoms as it can be found in some solid state materials, such as [Hg3(AsF6)2] and

[Hg4(AsF6)2].[40, 269, 270]

Figure 35. Left: Molecular structure of 16 in the solid state as determined by single crystal X-ray diffraction;

displacement ellipsoids (Pd, Zn) are shown at 50 % probability level, hydrogen atoms are omitted for clarity.

Right: Schematic illustration of the inner [PdZn8] core of 16. Selected interatomic distances (Å) and angles (°):

Pd1-Zn1 2.422(2), Pd1-Zn2 2.433(2), Pd1-Zn3 2.478(2), Pd1-Zn5 2.473(2), Zn3-Zn4 2.351(3), Zn5-Zn6

2.347(3), Zn1-Cp*centroid 1.981, Zn2-Cp*centroid 2.004, Zn4-Cp*centroid 2.002, Zn6-Cp*centroid 2.013; Pd1-Zn3-Zn4

179.69(13), Pd1-Zn5-Zn6 177.70(11), Zn5-Pd1-Zn5` 134.40(12), Zn3-Pd1-Zn3` 131.40(11), Zn3-Zn4-Cp*centroid

178.00, Zn5-Zn6-Cp*centroid 178.62, Pd1-Zn1-Cp*centroid 176.53, Pd1-Zn2-Cp*centroid 179.19.

Comment on the oxidation states. At this point it is necessary to comment on the oxidation

states. Usually, [Zn2Cp*2] is addressed by the authors as a Zn(I) compound.[138] The Allen

spectroscopic electronegativities of Pd and Zn are equal (1.59) and thus 15 and 16 should be


regarded as Pd(0) complexes. However, Parkin’s arguments point out the low coordination

and the essentially divalent type of bonding of the Zn and Ga atoms in 15 and 16 and thus

avoid including the discussion of formal oxidation states into this description.[235]

Nevertheless, it is clearly determined that the Zn atoms of the {ZnZnCp*} moieties found in

15 and 16 exhibit chemically different surroundings and the presence of ZnCp*2 as a

stoichiometric by-product for each {ZnZnCp*} unit trapped as a ligand to palladium suggests

the reaction be considered as a formal disproportionation of one part of the Zn(I) species into

Zn(0), i.e. Zn(0)Zn(I)Cp*, and Zn(II), i.e. Zn(II)Cp*2.

The formation of the metal rich compounds 15-18 demonstrates the high synthetic potential

going along with the usage of [Zn2Cp*2] in coordination chemistry towards transition metals.

Herein, [Zn2Cp*2] can be viewed as the natural source for two types of one electron zinc

ligands trapped in the ligand sphere of the d10 metal, i.e. the known ZnCp* fragment and the

novel {ZnZnCp*} unit featuring continuity of the fully intact covalent Zn-Zn bond. Beside

the unusual structural features displayed in the solid state structures, i.e. trapping of the novel

ligand system {ZnZnCp*}, these compounds can be viewed as novel steps towards zinc-rich

Hume-Rothery phases starting from coordination compounds with defined M/E polyhedral

geometries. Therefore, these compounds may be useful as building blocks to achieve even

larger molecular units with comparable structural motifs, or rather be suitable starting

materials of intermetallic solid state phases.

3.4.2 First Reactivity Studies of [Zn2Cp*2] Towards Olefin Containing d10 Transition

Metal Centres

In chapter 3.4.1 it has been shown that [Zn2Cp*2] can be used in the formation of zinc rich

compounds of the type [M(GaCp*)2(ZnCp*)2(ZnZnCp*)2] and [M(ZnCp*)4(ZnZnCp*)4] (M

= Pd, Pt). In general, [Zn2Cp*2] can be viewed as a natural source of forming one electron

ZnCp* and {ZnZnCp*} units. Especially the latter ligand system seems to be quite interesting

due to its ‘naked’ Zn atom trapped at the transition metal center. However, one of the main

results discussed before has been, that the formation of the fully zinc substituted compound

featuring a [MZn12] core seemed to be somehow inhibited by the presence of free GaCp*,

abstracted from the d10 metal starting materials [M(GaCp*)4]. Thus, it is interesting to gain

insights into the reactivity of this Zn(I) dimer towards reactive transition metal complexes in

the absence of GaCp*, i.e. [M(cod)2] (M = Ni, Pt). Notably, [Ni(cod)2] proved to be highly


reactive towards ZnCp2 leading to the unexpected metal-rich compound [Ni2Zn4Cp6] as

reported by Boersma and van der Kerk in 1983.[194] A hypothetical mechanism for the

formation of [Ni2Zn4Cp6] involves two-fold oxidative addition of ZnCp2 to Ni(0), followed by

cleavage of a {Cp}• radical, giving [CpNi(ZnCp)2] as an intermediate, which finally dimerizes

to the product [(NiCp)2(ZnCp)4].

Synthesis and Characterisation of [Ni(ZnCp*)4(ZnZnCp*)4] (19)

The reaction of [Ni(cod)2] with eight equivalents [Zn2Cp*2] in toluene at 80°C for 30 h results

in the quantitative formation of [Ni(ZnCp*)4(ZnZnCp*)4] (19), a deep red solid, in yields

around 60%. The 1H NMR spectroscopic data shows two signals in a 1:1 ratio at 2.20 and

2.31 ppm concerning to the different type of zinc ligands namely ZnCp* and {ZnZnCp*}

which have also been observed in the 13C NMR spectroscopic measurement. In addition,

LIFDI mass spectrometry displays the molecular ion peak [M].+ at m/z = 1927.

Mechanistic studies. The reaction of [Zn2Cp*2] with [Ni(cod)2] in a molar ratio of 8:1 has

been monitored by 1H NMR spectroscopy (reaction conditions: sealed and gastight Low

Pressure and Vacuum J-Young-NMR tubes; instant heating of the reaction mixture in C6D6 via

hot air gun for 5 minutes; b.p of C6D6 = 80°C). During the heating process the colour changed

from yellow to deep red and some grey precipitate formed. The 1H NMR spectrum displays

the expected signals of free 1,3-cod as well as unreacted [Zn2Cp*2]. In addition, one signal at

1.87 ppm could be assigned to ZnCp*2. Most importantly, besides the signals of 19, two more

signals in a ratio of 1:3 have been observed in the typical area of coordinated Cp* groups. The

LIFDI-MS spectrum of the reaction mixture in toluene displays two signals at m/z = 1928 for

19 and m/z = 796 for a second by-product in the reaction mixture. At this point it should be

noted that the second minor by-product of this reaction scheme has been identified as

[Cp*Ni(ZnCp*)3] (20). The selective synthesis of 20 will be discussed in the second part of

this chapter. In accordance with the 1H NMR spectroscopic data, the formation of 19 and 20

involves (1) the liberation of the olefin ligands as 1,3-cod, (2) redox chemical processes

attending the formation of Zn(II) as observed in the by-product ZnCp*2, the reduction of

Zn(I) to elemental zinc observed as a grey precipitate in the reaction mixture as well as the

oxidation of Ni(0) to Ni(I) observed in 20 and (3) the formation of novel Cp* ligands. Most

interestingly, these observations and conclusions implicate that the hypothetical 18 valence

electron complex [Ni(ZnCp*)8] seems to be highly unstable due to steric repulsion of the Cp*


groups and thus it is not observed, neither in the 1H NMR spectrum nor in the mass

spectrometric analysis. The structural features in the known complexes [M(ZnCp*)4(ZnMe)4]

(M = Ni, Pd, Pt) indicate the largest possible distances of the ZnCp* fragments to each other

in the coordination sphere around the d10 metal centre leading to an overall D4d symmetry,

respectively.[43] Thus, it seems to be unlikely to obtain products such as [Ni(ZnCp*)8] for

reasons mentioned above. The dissociation/association mechanism described for compounds

15-18 seems to be also adaptive in the formation of 19 (Scheme 24).

[Ni(cod)2][Zn2Cp*2]- 1,3-cod

[LaNi(ZnCp*)b][Zn2Cp*2]

[Ni(ZnCp*)4(ZnZnCp*)4]- [ZnCp*2]

[Zn2Cp*2],- 1,3-cod- Zn(0)

[Cp*Ni(ZnCp*)3]

(a) (b)

(c)

(L = 1,5-cod; ZnCp*)

Scheme 24. Possible reaction mechanism in the formation of 19 and 20.

The reaction starts with the release of 1,3-cod and the trapping of monovalent ZnCp* species

at electronically and coordinatively unsaturated Ni centres (a). The second step includes the

formation of one electron {ZnZnCp*} fragments by Cp* transfer reactions between the

starting compound [Zn2Cp*2] and the Zn centres of lower coordinated intermediate species of

the type [LaNi(ZnCp*)b] (L = 1,5-cod; ZnCp*) which lead to the release of ZnCp*2 as the

second by-product (b). In addition, a simultaneous side reaction takes place yielding

[Cp*Ni(ZnCp*)3] as a rare by-product. Herein, [Zn2Cp*2] acts as a smooth oxidizing agent

for the nickel centre as well as a natural source in the formation of ZnCp* ligands via Zn(I)-

Zn(I) bond cleavage (c).

It should be noted that identical results have been observed with the analogous platinum

compound [Pt(cod)2] determined by in situ LIFDI mass spectrometry (Figure 36). However,

as has been discussed in chapter 3.4.1, this compound is much more unstable than the nickel

analogous compound 19. Thus, further investigations and possible isolation and

characterisation of compound 18 have not been done.


Figure 36. Experimental LIFDI-MS spectrum of the reaction of [Pt(cod)2] with [Zn2Cp*2] (ratio 1:8); [M].+: m/z

= 932 [Cp*Pt(ZnCp*)3] (21), m/z = 2061 (18).

Single Crystal X-Ray Analysis of [Ni(ZnCp*)4(ZnZnCp*)4] (19)

Re-crystallisation of 19 from a saturated toluene solution at -30°C results in the formation of

deep red crystals suitable for single crystal X-ray diffraction. In pure crystalline form, 17 is

nearly insoluble in all common solvents and decomposes slowly within a few hours even

under an inert gas atmosphere, whereas solutions of 19 are stable when stored at -30°C.

Compound 19 crystallizes in the orthorhombic space group Pnna. The molecular structure

determined by single crystal X-ray diffraction is shown in Figure 37. The structural features

of 19 are almost identical to that described for 16 in chapter 3.4.1. Most importantly, the Ni-

ZnZnCp* distances (av. 2.351 Å) and Ni-ZnCp* distances (av. 2.349 Å) are nearly equal and

well comparable to known structures found in the literature.[43, 194] The parity of the M-

ZnZnCp* and M-ZnCp* bond distances found in 19 is very different from compound 16

where significant deviations can be observed. For a detailed comparison of bond distances of

compounds 16, 19 and the parent compound [Zn2Cp*2] see Table 16. As it has been discussed

for 16, the {ZnZnCp*} units are bonded to the metal centre in an almost linear fashion (Ni1-

Zn2-Zn1 178.40(6)°, Ni1-Zn3-Zn4 179.36(6)°, Zn3-Zn4-Cp*centroid 177.64, Ni1-Zn5-


Cp*centroid 178.09). In contrast, the inner Zn-Zn-Ni-Zn-Zn metal chains strongly deviate from

linearity (Zn2-Ni1-Zn2` 127.80(7)°, Zn3-Ni1-Zn3` 128.34(7)°). Thus, compound 19 can be

viewed as one of the rare examples of unsupported metal chains in the same way as 16.

Table 16. Important interatomic distances (Å) for 19.

Compound M-ZnCp* M-ZnZnCp* Zn-Zn Zn-Cp*centroid ZnZn-Cp*centroid

[Zn2Cp*2] - - 2.331 - 1.936

16 2.422(2)-

2.433(2)

2.473(2)-

2.478(2)

2.347(3)-

2.351(3)

1.981-2.004 2.002-2.013

19 2.344(1)-

2.354(1)

2.344(1)-

2.355(1)

av. 2.354 1.946-1.949 1.947


displacement ellipsoids (Ni, Zn) are shown at 50 % probability level, hydrogen atoms are omitted for clarity.

Selected interatomic distances (Å) and angles (°): Ni1-Zn2 2.3435(10), Ni1-Zn3 2.3552(10), Ni1-Zn5

2.3539(12), Ni1-Zn6 2.3442(12), Zn1-Zn2 2.3476(13), Zn3-Zn4 2.3597(13), Zn1-Cp*centroid 1.947, Zn4-

Cp*centroid 1.947, Zn5-Cp*centroid 1.949, Zn6-Cp*centroid 1.946; Ni1-Zn2-Zn1 178.40(6), Ni1-Zn3-Zn4 179.36(6),

Zn2-Ni1-Zn2` 127.80(7), Zn3-Ni1-Zn3` 128.34(7), Zn2-Zn1-Cp*centroid 178.18, Zn3-Zn4-Cp*centroid 177.64, Ni1-

Zn5-Cp*centroid 178.09, Ni1-Zn6-Cp*centroid 179.55.


Theoretical investigations on the bonding situation of 16, 18 and 19. It should be noted

that the following discussions are just temporary results. In the course of the theoretical

calculations it became clear that Cp* groups have a significant effect on bond distances and

angles. The replacement of Cp* with Cp could not be done which makes such calculations

complex. Nevertheless, in the case of palladium, the model compound [Pd(ZnH)4(ZnZnH)4]

shows the right trend and has been used for initial exemplarily calculations (for

Computational Details see Experimental Section 6.3.3). First results show that the M-Zn

distances are shorter than the M-ZnZn bond distances (Table 17 and Table ES6, 6.3.3). This

effect expands for higher elements and is qualitatively well comparable with experimental

data.

Table 17. Geometry optimization at BP86-D/SVP (X-Ray structure) and BP86/TZVPP (16M).

Compound Method d(M-Zn) d(M-ZnZn)

16 Experimental 2.422(2)-2.433(2)

2.473(2)-2.478(2)

19 Experimental 2.344(1)- 2.354(1)

2.344(1)- 2.355(1)

[Ni(ZnCp*)4(ZnZnCp*)4] BP86-D/SVP 2.281-2.302 2.305-2.324

[Pd(ZnCp*)4(ZnZnCp*)4] BP86-D/SVP 2.398-2.412 2.440-2.453

[Pd(ZnH)4(ZnZnH)4] (16M) BP86/TZVPP 2.481 2.500

[Pt(ZnCp*)4(ZnZnCp*)4] BP86-D/SVP 2.427-2.430 2.468-2.470

In order to obtain initial insights into the bonding situation in {ZnZnCp*} containing

compounds, MO correlation and AIM analysis has been carried out. The MO correlation

diagram of 16M shows similar valence molecular orbitals to those found for [Pd(ZnH)8].[43]

Figure 38. Molecular orbitals presenting the σ-character of the Zn-Zn interactions.


Notably, the over-all MO correlation diagram for such a large system becomes more and

more difficult and confusing. Nevertheless, results show that the Zn-Zn interactions in the

{ZnZnH} ligands are mainly of σ-character (Figure 38 and Figure ES4, 6.3.3).

The AIM analysis of the electronic structure of 16M displays bond paths between Pd-ZnH,

Pd-ZnZnH and Zn-Zn of the {ZnZnH} units but no bond paths between Zn atoms of different

ligand systems (Figure 39). These results are comparable with calculations for [M(ZnR)n]

compounds which always showed M-ZnR bond paths while most compounds do not exhibit

any Zn-Zn bond paths.[43]

Figure 39. Molecular graph and contour map of the Laplacian ∇2ρ(r) of 16M. Solid lines indicate areas of

charge concentration (∇2ρ(r) < 0) while dashed lines show areas of charge depletion (∇2ρ(r) > 0). The thick solid

lines connecting the atomic nulei are the bond paths. The thick solid lines separating the atomic basins indicate

the zero-flux surfaces crossing the molecular plane.

The EDA-NOCV results confirm the features mentioned from the MO correlation diagram

and the AIM analysis (Table 18). The interaction energy ΔEint is stronger for ZnH in

comparison to {ZnZnH}. Most interestingly, the Pauli repulsion ΔEPauli seems to be

responsible for the bond weakening of the Pd-ZnZnH interactions. The orbital interaction

term ΔEorb is mainly based on σ interactions (>70 %), but 10% of orbital interactions could

not be assigned.


Table 18. EDA-NOCV results of 16M (BP86/TZ2P+). Energies in kcal/mol.

Pd(ZnH)4(ZnZnH)3 + ZnZnH

Pd(ZnH)3(ZnZnH)4 + ZnH

ΔEint -56.8 -61.2

ΔEPauli +194.5 +161.6

ΔEelstat -156.0 (62.1%) -141.1 (63.3%)

ΔEorb -95.3 (37.9%) -81.7 (36.7%)

ΔE(σ) -68.3 (71.7%) -62.3 (76.3%)

ΔE(π) -10.5 (11.0%) -10.5 (12.9%)

Rest ΔEorb -12.8 (13.4%) -8.9 (10.9%)

Synthesis and Characterisation of [Cp*M(ZnCp*)3] (M = Ni (20), Pt (21))

The formation of [Ni(ZnCp*)4(ZnZnCp*)4] has been carried out with a large excess of

[Zn2Cp*2] over a period of 30 h. Mechanistic studies performed for this reaction displayed the

formation of [Cp*M(ZnCp*)3] (M = Ni (20), Pt (21)) as a minor by-product. Pure compounds

20 and 21 could be obtained by the reaction of [M(cod)2] (M = Ni, Pt) with two equivalents

[Zn2Cp*2], thus providing no excess [Zn2Cp*2] (Scheme 25). After 3 h at 80°C the Cp*

transfer products could be isolated in yields about 60 % (20) and 50 % (21).

[M(cod)2]M

Cp*Zn ZnCp*ZnCp*

- 1,3-cod, Zn

(M = Ni (20), Pt (21))

[Zn2Cp*2], toluene, 80°C, 3h

(M = Ni, Pt)

Scheme 25. Synthesis of [Cp*M(ZnCp*)3] (M = Ni (20), Pt (21)).

The formation of 20 and 21 takes place via loss of 1,3-cod (revealed by 1H NMR

spectroscopic studies) and Zn(I)-Zn(I) bond cleavage to obtain ZnCp* fragments. Final steps

include redox chemical attended Cp* transfer from the ZnCp* units to the transition metal.

The transition metal is oxidized M(0)→M(I) and the Zn(I) reduced to its bulk material leading

to the precipitation of an unsolvable grey solid. Compounds 20 and 21 are quite stable 18

valence electron complexes and can be stored for several weeks at -30°C under inert


atmosphere. Results of the 1H and 13C NMR spectroscopic characterisations show no unusual

features related to the expected signal pattern (Table 19). However, LIFDI-MS shows

molecular ion peaks [M].+ at m/z = 794 (20) and 932 (21), respectively. An additional signal

for [M-Cp*]+ is observed for 21 at m/z = 797.

Table 19. NMR spectroscopic data (C6D6, rt) of 20 and 21.

Compound 1H NMR

ZnCp* MCp*

s, 45H s, 15H

13C NMR

ZnC5Me5 MC5Me5

ZnC5Me5 MC5Me5 20 2.16 1.89 11.98 13.29

111.68 97.98

21 2.17 2.13 11.69 13.08

110.46 102.04

Single Crystal X-Ray Analysis of [Cp*M(ZnCp*)3] (M = Ni (20), Pt (21))

Pure crystals of 20 and 21 suitable for single crystal X-ray diffraction can be obtained from

saturated toluene solutions at -30°C. The crystallisation of compound 21 gave yellow

crystalline material of rather low quality. In addition, compound 21 indeed crystallized with

three independent molecules in the asymmetric unit. However, despite its poor data quality

and low precision, the single crystal X-ray study of compound 21 unambigously evidenced

the three-dimensional structure of the molecule. Compounds 20 and 21 crystallize in the

triclinic space group P-1. Since the structural features are similar for both compounds, only 20

will be discussed in detail (Figure 40). The arrangement of the ligands leads to a pseudo-

tetrahedral piano-stool structure, comparable to the cationic parts of the isoelectronic 18

valence electron complexes [Cp*M(GaCp*)3]m+ (M = Fe, m = 1; M = Co, m = 2).[39] The Zn-

M-Zn bond angles in 20 and 21 vary only slightly from 80° (20: 80.36(4)° (Zn2-Ni1-Zn3) to

82.26(4)° (Zn1-Ni1-Zn3), 21: 81.13(8)° (Zn2-Pt1-Zn3) to 82.61(8)° (Zn1-Pt1-Zn2). Notably,

the Ga-M-Ga bond angles in the isoelectronic iron and cobalt structures mentioned above are

slightly wider with values around 90°. However, qualitative discussion of structural

parameters seems to be restricted, mostly due to different atomic properties and structural

effects, respectively. The Cp*centroid-Zn-M angles (average values: 161.2° (20) and 157.6°


(21)) display significant deviations from linearity. This flexibility of the Cp* ring has been

discussed before for ECp* compounds.[39, 126, 128]


displacement ellipsoids (Ni, Zn) are shown at 50 % probability level, hydrogen atoms are omitted for clarity.

Selected interatomic distances (Å) and angles (°) for 20: Ni1-Zn1 2.289(1), Ni1-Zn2 2.300(1), Ni1-Zn3 2.295(1),

Zn1-Cp*centroid 1.99, Zn2-Cp*centroid 2.04, Zn3-Cp*centroid 2.07, Ni1-Cp*centroid 1.73, Zn1-Ni1-Zn2 80.95(3), Zn1-

Ni1-Zn3 82.26(4), Zn2-Ni1-Zn3 80.36(4), Zn1-Ni1-Cp*centroid 128.6, Zn2-Ni1-Cp*centroid 133.5, Zn3-Ni-

Cp*centroid 131.8, Cp*centroid-Zn1-Ni1 165.2, Cp*centroid-Zn2-Ni1 159.5, Cp*centroid-Zn3-Ni1 158.9. Selected

interatomic distances (Å) and angles (°) for 21: Pt1-Zn1 2.40, Pt1-Zn2 2.380(2), Pt1-Zn3 2.377(2), Pt1-Cp*centroid

1.96, Zn1-Cp*centr 2.00, Zn2-Cp*centroid 1.98, Zn3-Cp*centroid 2.03, Zn1-Pt1-Zn2 82.61(8), Zn1-Pt1-Zn3 82.3, Zn2-

Pt1-Zn3 81.13(8), Pt1-Zn1-Cp*centroid 158.7, Pt1-Zn2-Cp*centroid 163.9, Pt1-Zn3-Cp*centroid 150.0, Zn1-Pt1-

Cp*centroid 130.2, Zn2-Pt1-Cp*centroid 133.1, Zn3-Pt1-Cp*centroid 128.8.

It is an important prerequisite for the coordination of ECp* fragments to transition metal

centres as it allows the lowering of steric repulsion. The Ni-Zn distances found in 20 (av. 2.30

Å) are similar to bond distanes of other Ni-Zn bonds found in the literature, however, it

features the shortest bond distance reported so far.[43, 194] Notably, significantly shortened Pt-

Zn bond interactions (av. 2.38 Å) in comparison to all known literature examples can be

observed for compound 21. The Zn-Cp*centroid distances (average values: 2.03 Å (20), 2.00 Å


(21)) are slightly elongated with respect to several ZnCp* containing reference compounds,

that exhibit values between 1.93 and 1.96 Å.[43, 138] The Zn-Cp* contact may be taken as an

indication of small variations of partial charges at the Zn centre as it is well established for the

related Ga-Cp*centroid contacts in complexes [LnM(GaCp*)] of various kind. However, as

mentioned above, the steric crowding of three facial ZnCp* ligands may also affect the

average Zn-Cp* distances.

Table 20. Important interatomic distances (Å) for 20, 21 and selected reference compounds.

Compound M-ZnCpR Zn-CpRcentroid M-CpR

centroid M-ZnZnCp* M-ZnMe

20 2.289(1)-

2.300(1)

1.99-2.07 1.73 - -

[Ni(ZnCp*)4(ZnZnCp*)4] (19) 2.344(1)-

2.354(1)

1.946-1.949 - 2.344(1)-

2.355(1)

-

[Ni(ZnCp*)4(ZnMe)4][43] 2.351(1)-2.371(1)

1.96 - - 2.313(1)-2.330(1)

[Zn4Ni2(C5H5)4(C5Me5)2][196] 2.370(3)-2.450(3)

η2/η3 mode 1.728 - -

[Cp6Ni2Zn4][194] 2.398(2) 2.02 1.74 - -

21 2.377(2)-

2.400

1.98-2.03 1.96 - -

[Pt(ZnCp*)4(ZnMe)4][43] 2.441(1)-

2.459(1)

1.94 - - 2.402(1)-

2.467(1)

[Pt(GaCp*)2(ZnCp*)2(ZnZnCp*)2]

(17)

2.4152(16) 1.95 - av. 2.39 -

The synthesis and characterisation of compounds 19-21 clearly indicates the high potential of

[Zn2Cp*2] in the preparation of zinc rich transition metal complexes. Compound 19 illustrates

that the formation of {ZnZnCp*} ligands, and thus metal rich [MZn12] cores, depends by no

means on the existence of GaCp* in the starting materials. This point stands directly in

contrast to the formation of zinc-rich highly coordinated compounds via redox chemical

processes by the use of ZnMe2 and GaCp* containing complexes. Herein, no product

formation can be occurred using reactive non-GaCp* starting materials. In addition, the

preparation of compounds 20 and 21 nicely illustrates that reaction schemes are easily


controllable via M/Zn ratios, i.e. stoichiometric amounts of the starting materials and the

reaction conditions. Furthermore, direct correlations between the one electron donor {ZnCp*}

and GaCp* as well as {ZnCp*} and the CH3 group can be drawn. Firstly, the pseudo-

tetrahedral piano-stool geometry obtained for compounds 20 and 21 present analogies

between {ZnCp*} units and GaCp* ligands. Secondly, {ZnCp*} ligands and CH3 seem to be

formally isolobal one electron fragments. Thus, the formation of compound 21 can be

declared as the isostructural {ZnCp*} containing counterpart of [CpRPtL] (CpR = C5H5,

C5Me5; L = Me3, Me2Br, MeBr2).[271, 272] It should be noted that the corresponding Ni

complex [CpRNiL3] is unknown so far, thus it seems to be surprisingly that compound 20

exists as a stable 18 valence electron compound.

3.4.3 Experimental and Theoretical Investigations on the Formation of a Novel

[PdZn7] Compound: [Zn2Cp*2] as a Source for Stabilized Zn(0)

In the previous chapters it has been shown that even [Zn2Cp*2] behaves as a natural source for

the one electron ligand ZnCp* and, surprisingly, for the {ZnZnCp*} fragment. Herein, the

reaction of d10 metal GaCp* complexes [M(GaCp*)4] (M = Pd, Pt) with stoichiometric

amounts [Zn2Cp*2] yields in the formation of the Ga/Zn mixed hexa-coordinated compounds

[M(GaCp*)2(ZnCp*)2(ZnZnCp*)2] as well as the fully zinc substituted octa-coordinated

compound [M(ZnCp*)4(ZnZnCp*)4]. Notably, mechanistic studies revealed reaction

pathways quite different from that of the known ZnR2 reactions, i.e. dissociation/association

steps rather than redox chemical processes. In addition, chapter 3.4.2 describes the synthesis

of [M(ZnCp*)4(ZnZnCp*)4] and [Cp*M(ZnCp*)3] (M = Ni, Pt) by using reactive, substitution

labile d10 metal complexes [M(cod)2] at which it has been shown that product formation can

be controlled via stoichiometry and reaction conditions. In the light of these facts it seemed to

be self-evident to extend the exploration of [Zn2Cp*2] reactivity or rather coordination

chemistry towards a further class of reactive transition metal complexes containing

substitution labile ligands, which can be also act as ‘catching’ or transfer ligand systems.

Herein, [PdMe2(tmeda)] plays an important role. As it has been shown in the case of low

valent group 13 chemistry, [PdMe2(tmeda)] can be used to prepare both homoleptic,

monomeric units and metal-rich compounds such as [Pd3(InCp*)8].[131] In general, redox

chemical reaction schemes have been observed leading to the reduction of Pd(II) to Pd(0) and

the oxidation of E(I) to E(III) resulting in the formation of [Cp*EMe2] (E = Ga, In).


Synthesis and Characterisation of [Cp*Pd(ZnCp*)3] (22) and

[Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}] (23)

Dissolving a mixture of [PdMe2(tmeda)] and four equivalents [Zn2Cp*2] in toluene at 55°C

leads to an immediate colour change of the homogenous solution from colourless to deep red

from which [Cp*Pd(ZnCp*)3] (22) and [Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}] (23) can be

obtained in a 1:1 ratio as orange (22) or rather yellow (23) crystals after specific work-up via

fractionated extraction (Scheme 26). LIFDI-MS displays the molecular ion peak [M].+ of 22 at

m/z = 842 and one signal for [M-Cp*]+ at m/z = 709, the molecular ion peak [M].+ of 23

occurs at m/z = 1252.

[PdMe2(tmeda)]

[Cp*Pd(ZnCp*)3]

[Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}]

[Pd(ZnCp*)4(ZnMe)4]

+ xs. [Zn2Cp*2]toluene, 55 °C

- [RZnCp*](R = (tmeda)Me, Cp*)

(22)

(23)

Scheme 26. Synthesis of [Cp*Pd(ZnCp*)3] (22) and [Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}] (23).

The 1H NMR spectrum of 22 shows the expected pattern of signals for equivalent ZnCp*

ligands and the PdCp* unit at 2.08 (s, 45H) and 2.17 (s, 15H) ppm in a ratio of 3:1. In

addition, the 1H NMR spectrum of 23 at room temperature displays two broad signals in a

ratio of 1:1 at 2.10 (s, br, 30H) and 2.26 (s, br, 30H) ppm, each assigned to two chemically

equivalent ZnCp* ligands, one signal at 0.14 (s, 6H) ppm belonging to two chemically

equivalent ZnCH3 units and finally, signals at 1.62 (s, 4H) and 1.73 (s, 12H) ppm for the

{Zn(tmeda)} ligand. Notably, the observation of broad signals for the ZnCp* ligands indicates

Cp* exchange between axial and equatorial Zn positions at room temperature. Temperature

depended 1H NMR measurements have been carried out and substantiate the fluctional

behaviour. Thus, coalescence of both resonances has been observed at 50°C leading to one

broad peak at 2.17 ppm (s, 60H) and significant line sharpening at 70°C without change of

signal positions. The corresponding 13C NMR spectra of 22 and 23 show no unusual features

with respect to carbon signals for Cp*, CH3 and the tmeda unit.


Mechanistic studies. In order to gain first insights into the reaction pathways, 1H NMR

spectroscopic studies have been carried out (reaction conditions: sealed and gastight Low

Pressure and Vacuum J-Young-NMR tubes; 3 h, 55°C, C6D6; it should be noted that once the

reaction mixture is heated up uncontrollable near the boiling point of C6D6 decomposition

takes place giving rise to a wide variety of undefined signals; b.p of C6D6 = 80°C). The 1H

NMR spectrum of this reaction shows various signals in the typical Cp* area close together.

Thus, determination of ratios and integrals has been done using the line fitting option of

Mestrec® V 4.7.0.0. The assignment of the multitude of peaks found in the reaction mixture is

shown in table 21.

Table 21. 1H NMR spectroscopic data for the reaction of [PdMe2(tmeda)] with [Zn2Cp*2] (C6D6).

Compound PdCp* ZnCp* ZnMe Zn(tmeda)

22 2.17 (s, 15H) 2.08 (s, 45H) - -

23 - 2.10 (s, 30H)

2.26 (s, 30H)

0.14 (s, 6H) 1.62 (s, 4H)

1.73 (s, 12H)

[Pd(ZnCp*)4(ZnMe)4] - 2.08 (s, 60H) -0.04 (s, 12H) -

[ZnCp*2] - 1.87 (s, 30H) - -

[Cp*ZnMe] - 1.98 (s, 15H) -0.68 (s, 3H) -

[(tmeda)Zn(Me)Cp*] - 2.20 (s, 15H) -0.08 (s, 3H) 2.30 (s, 4H)

2.13 (s, 12H)

The ratio of 22 and 23, as revealed by in situ 1H NMR spectroscopic measurements, is ca. 1:1

as it has been described above. [Pd(ZnMe)4(ZnCp*)4] exists only in small traces of about 4 %

in comparison to compounds 22 and 23. Furthermore, ZnCp*2 occurs in a ratio of 3.6:1

reffered to compound 23 and in a ratio of ca. 2.5:1 referred to [Cp*ZnMe]. It should be noted

that the tmeda adduct [(tmeda)Zn(Me)Cp*] has been not reported in the literature so far,

although [(tmeda)ZnMe2] is well known. Additionally, adduct formation of ZnCp*2 with

strong Lewis bases such as dmap has been reported by Schulz in 2009.[164] With respect to the

signal pattern found in the in situ NMR reaction, it is probable that tmeda adducts of ZnCp*2,

[Cp*ZnMe] or rather ZnMe2 are formed during the reaction. Free tmeda displays signals at


2.12 (s, 12H) and 2.36 (s, 4H) ppm in C6D6, which comparable well with the signals found at

2.13 and 2.30 ppm. However, fast exchange or rather dissociation/association mechanisms

should be kept in mind when discussing adduct formations. The evolution of ethane can be

ruled out. Obviously, a variety of competing (redox chemical) reaction pathways have to be

taken into account when [Zn2Cp*2] is combined with substitution labile M(II) complexes such

as [PdMe2(tmeda)]. For instance, [Zn2Cp*2] behaves as a source for Zn(I) due to homolytic

cleavage of the Zn(I)-Zn(I) bond as discussed before. In addition, disproportionation of the

dimeric Zn(I) unit can be generate Zn(0) and Zn(II), the latter observed in several products of

the general formula ZnR2. The reaction sequences are speculative. For example, the Zn(0)

component can undergo insertion reactions into Pd-CH3 bonds leading to Pd-Zn(CH3)

fragments which can be the starting point for further reactions with [Zn2Cp*2]. Furthermore,

Zn(0) can be ‘caught’ by methyl groups and present tmeda to give rather adducts or Pd-

Zn(tmeda) fragments. In any case, once the reaction starts and unsaturated Pd-ZnR fragments

are present in the reaction mixture, they will trap ZnR (R = Me, Cp*, tmeda) units or rather

procede Cp* transfer reactions under redox chemical conditions as it has been reported in the

1980s whereat even ZnCp2 has been used in the formation of Cp transfer products with the

simultaneous formation of M-ZnR bonds.[143, 194]

Single Crystal X-Ray Analysis of [Cp*Pd(ZnCp*)3] (22)

Single crystals of 22 suitable for single crystal X-ray diffraction studies were obtained by

slow cooling of a saturated toluene/n-hexane mixture down to -30 °C (Figure 41). The

structural features are identical with those described for 20 and 21 in chapter 3.4.3 revealing

pseudo-tetrahedral piano-stool geometry with Zn-Pd-Zn bond angles between 77.01(4)° (Zn1-

Pd1-Zn2) and 80.68(4)° (Zn1-Pd1-Zn3). Similar to all known ECp* (E = Ga, Zn) containing

compounds, the Cp*centroid-Zn-Pd angle displays strong deviation from linearity (av. 158.7°) in

order to adjust the sterically crowding environment. The Pd-Cp*centroid bond distance is well

within the range of other PdCp* compounds known in the literature.[273, 274] Finally, the Pd-Zn

bond lengths (av. 2.38 Å) are shorter than those in other Pd-Zn complexes such as

[Pd(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (2.448(1) Å for Pd-ZnCp* and 2.375(1)-2.379(1) Å for

Pd-ZnZnCp*) and [Pd(ZnCp*)4(ZnMe)4][43] (2.447(1)-2.459(1) Å). Notably, the distance

found in 22 is similar to the Pd-Zn distance found in [(FPNP)Pd-Zn-Pd(FPNP)][275] (2.372(1)-

2.379(1) Å) (FPNP = bis(4-fluoro-2-(diisopropylphosphino)phenyl)amine), the only known

compound featuring a covalent 2e-2c Pd-Zn bond. As it has been previously discussed, the


Zn-Cp*centroid distance (av. 1.98 Å) is longer than in other ZnCp* containing compounds such

as [Zn2Cp*2] (1.94 Å). The Zn-Cp* contact may be taken as an indication of small variations

of partial charge at the Zn centre as it is well established for the related Ga-Cp* contacts in

complexes [LnM(GaCp*)] of various kinds. However, the steric crowding of three facial

ZnCp* ligands may also affect the average Zn-Cp* distances.



Selected interatomic distances (Å) and angles (°) for 22: Pd1-Zn1 2.379(1), Pd1-Zn2 2.384(1), Pd1-Zn3

2.368(1), Pd1-Cp*centroid 1.98, Zn1-Cp*centroid 1.99, Zn2-Cp*centroid 1.99, Zn3-Cp*centroid 1.97, Zn1-Pd1-Zn2

77.01(4), Zn1-Pd1-Zn3 80.68(4), Zn2-Pd1-Zn3 78.10(4), Cp*centroid-Pd1-Zn1 135.1, Cp*centroid-Pd1-Zn2 135.2,

Cp*centroid-Pd1-Zn3 128.6, Cp*centroid-Zn1-Pd1 156.7, Cp*centroid-Zn2-Pd1 156.1, Cp*centroid-Zn3-Pd1 163.4.

Single Crystal X-Ray Analysis of [Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}] (23)

Compound 23 crystallizes in the monoclinic space group P21/n. The molecular structure

received by single crystal X-ray diffraction is depicted in Figure 42. The Pd centre is

surrounded by four ZnCp* units, two ZnMe ligands as well as one {Zn(tmeda)} ligand

leading to a [PdZn7] core.




Selected interatomic distances (Å) and angles (°) for 22: Pd1-Zn1 2.389(1), Pd1-Zn2 2.385(1), Pd1-Zn3

2.396(1), Pd1-Zn4 2.454(1), Pd1-Zn5 2.476(1), Pd1-Zn6 2.453(1), Pd1-Zn7 2.480(1), Zn1-N1 2.140(4), Zn1-N2

2.125(4), Zn2-C7 1.972(5), Zn3-C8 1.981(6), Zn4-Cp*centroid 2.049, Zn5-Cp*centroid 2.061, Zn6-Cp*centroid 2.020,

Zn7-Cp*centroid 2.030; Zn1-Zn5-Zn2 104.87(3), Zn7-Zn1-Zn5 115.04(3), Zn1-Pd1-Zn7 66.52(2), Zn3-Pd1-Zn7

75.83(2) Zn4-Pd1-Zn6 136.70(3), Pd1-Zn4-Cp*centroid 174.60, Pd1-Zn5-Cp*centroid 165.72, Pd1-Zn6-Cp*centroid

177.18, Pd1-Zn7-Cp*centroid 168.75.

The coordination geometry of the [PdZn7] metal core seems to be a pentagonal bi-pyramid at

which the ‘equatorial’ Zn atoms, i.e. Zn(1,2,3) and Zn(5,7), stretch the five-membered ring

with angles between 104.87(3)° (Zn2-Zn5-Zn1) and 115.04(3)° (Zn5-Zn1-Zn7) leading to an

angular sum of 539.65° (ideally 540°). The precise description based on continuous shape

measure, which allows a quantitative comparison of coordination polyhedra, clearly displays

major distortion of the suggested pentagonal bi-pyramid (SQ(P) = 4.30).[218, 219, 276] Basically,

both ‘axial’ Zn(4,6)Cp* ligands are strongly bent towards the Zn(2,3)Me units. The almost

coplanar five-membered ring remains untouched as it can be determined by a rather small

continuous shape measure (SQ(P) = 0.32). Lastly, an alternative description of the [PdZn7]

core is based on a deviation from an ideal trigonal dodecahedron with two corners substituted

by one corner in between. This is strongly supported by a very small continuous shape

measure of SQ(P) = 0.07 based on the comparison of the [PdZn7] core with the corresponding


six corners of an ideal trigonal dodecahedron. The trigonal dodecahedral coordination

geometry has also been observed for the [PdZn8] metal core of [Pd(ZnCp*)4(ZnMe)4][43], thus,

the {Zn(tmeda)} ligand (Zn1) takes exactly the place between the ‘missing’ two corners of the

dodecahedron (Figure 43). It should be noted, that two further possibilities for [ML7] units,

namely capped trigonal prismatic and capped octahedral environments, can be ruled out.

Figure 43. Left: Superimposition of the [PdZn7] metal core (Pd: blue, Zn: green) and a trigonal dodecahedron

(black). Right: Superimposition of the [PdZn7] metal core (Pd: blue, Zn: green) and a pentagonal bi-pyramide

(black).

The tmeda backbone of the {Zn(tmeda)} is perpendicularly arranged in relation to the planar

five-membered ring in order to minimize steric strain. As discussed above, the Zn(4,6)Cp*

units located in ‘axial’ positions are significantly bent towards both ZnMe units resulting in

strong deviation from linearity displayed by the Zn4-Pd1-Zn6 angle of 136.70(3)°. The

ZnCp* units feature η5-Cp* coordination mode with rather small deviations from linearity

found for the Pd1-Zn-Cp*centroid angles (axial: Zn4, 174.60°; Zn6, 177.18°; equatorial: Zn5,

165.72°; Zn7, 168.75°). In addition, the Zn-Cp*centroid distances (equatorial: av. 2.045 Å,

axial: av. 2.035 Å) comparable well with the distances of the parent compound [Zn2Cp*2][138]

(av. 2.04 Å). The Zn-N bond lengths at the terminal and tri-coordinated Zn1 (av. 2.133 Å) are

within the range of reference compounds known in the literature.[189, 277] Notably, compound

23 is the first example of a structure containing an unsupported terminally coordinated ZnL2

unit. The differences of Pd-Zn bond distances between all three kinds of Zn ligands (Pd-

Znequatorial: av. 2.478 Å, Pd-Znaxial: av. 2.453 Å, Pd-ZnMe: av. 2.391 Å, Pd-Zn(tmeda):

2.389(1) Å) are negligible, however, they follow the trend Pd-Znequatorial > Pd-Znaxial > Pd-


ZnMe ~ Pd-Zn(tmeda). Notably, all Pd-Zn distances match well with known Pd-Zn

containing compounds but are distintly shorter than the average Pd-Zn distance (2.646 Å)

obtained for the intermetallic solid state phase Pd1Zn1.[278] The Zn-Zn distances of 23 range

between 2.6718(1) Å for Zn1-Zn7 and 2.997(1) Å for Zn3-Zn7 matching very well with the

Zn-Zn contacts in the intermetallic phase Pd1Zn1 (2.899 Å).

Theoretical investigations on the bonding situation of 23. In order to achieve an insight

into the nature of metal-ligand bonding situations and to draw comparisons between all three

different kinds of ZnR ligands, quantum chemical calculations have been carried out on the

BP86/TZVPP level of theory (for Computational Details see Experimental Section 6.3.4).

Herein, the Cp* ligands have been changed into Cp giving rise to the model compound named

23M. The calculated Pd-Zn distances in 23M are comparable with those found in 23,

determined by single crystal X-ray diffraction, which show the same trend in Pd-Zn bond

distances as mentioned above (Table 22).

Table 22. Experimental and calculated distances in Å.

Compound Pd-ZnMe PdZnCpR

axial PdZnCpRequatorial Pd-Zn(tmeda)

23 av. 2.391 av. 2.453 av. 2.478 2.389(1)

23M av. 2.427 av. 2.770 av. 2.485 2.420

In addition, the calculated bond angle between the ‘axial’ ligands Zn4-Pd1-Zn6 (141.0°) in

23M is also in accordance with the experimental bond angle (136.7o). Thus, 23M seems to be

an adequate model system for the analysis of the bonding situation in 23. The theoretical

studies on [M(ZnR)n] systems which contain 1e donor ligands ZnR (R = Me, Cp) have shown,

that they feature strong M-ZnR bonds while the Zn-Zn interactions between ZnR ligands are

only weakly attractive.[42-44] Additionally, the AIM analysis of the electronic structure of

[M(ZnR)n] compounds showed that there are always M-ZnR bond paths while most

compounds do not exhibit any Zn-Zn bond paths. The Laplacian distribution of 23M in the

plane which contains Pd and the ligand atoms Zn(Cp,ax) and {Zn(tmeda)} displays three Pd-

Zn bond paths but no Zn-Zn bond paths as it has been discussed previously. It should be noted

that the same results are obtained for the other Zn ligands in 23M (Figure 44).


Figure 44. Molecular graph and contour map of the Laplacian ∇2ρ(r) of 23M in the molecular plane which

contains Pd and the ligand atoms Zn(Cp,ax) and Zn(TMEDA). Solid lines indicate areas of charge concentration

(∇2ρ(r) < 0) while dashed lines show areas of charge depletion (∇2ρ(r) > 0). The thick solid lines connecting the

atomic nuclei are the bond paths. The thick solid lines separating the atomic basins indicate the zero-flux

surfaces crossing the molecular plane.

Notably, the shape of the Laplacian distribution does not feature any difference between

ZnCp and {Zn(tmeda)} units. However, significant differences are found between the

calculated NBO partial charges of the different ligands.

Table 23. Calculated NBO partial charges q for atoms and ligands in 23M.

Atom/ligand q (NBO)

Pd -3.05

Zn[tmeda]/ (Zn{tmeda}) +0.79 (+1.11)

Zn[Cp,ax]/(ZnCp) +0.92 (+0.33) / +0.89 (+0.30)

Zn[Cp, eq]/(ZnCp) +0.81 (+0.25) / +0.81 (+0.25)

Zn[Me]/(ZnMe) +0.95 (+0.40) / +0.95 (+0.41)


The Pd atom in 23M carries a large negative charge of -3.05e while the Zn atoms of the

ligands are positively charged (Table 23). Interestingly, the Zn atom of {Zn(tmeda)} carries a

positive charge (+0.79e) which is very close to the other Zn atoms (between +0.81 and

+0.95e) and is not very different from the charge of +0.72e calculated for the Zn atoms in

[Zn2Cp*2].[145, 171] However, the overall charge of the {Zn(tmeda)} ligand as a whole is much

higher (+1.11e) than for the other Zn-ligands (between +0.25 and +0.41e). Notably,

significant differences between Pd-ZnR (R = Cp, Me) and Pd-Zn(tmeda) interactions can be

deduced from Energy Decomposition Analysis (Table 24). The overall strength of the Pd-Zn

interactions ∆Eint for the Pd-Zn(tmeda) unit is somewhat larger than for the other Pd-ZnR (R

= Cp, Me) units while the percentaged contributions of the attractive terms ∆Eelstat and ∆Eorb

to the total interactions energy are very similar for both types of ligands. The Pauli repulsion

∆EPauli is much higher for the Pd-Zn(tmeda) bond than for Pd-ZnR (R = Cp, Me). However,

this is compensated by the nearly equally strong increase in the ∆Eelstat and ∆Eorb values for

Pd-Zn(tmeda).

Table 24. EDA results for the Pd-ZnR interactions in 23M. Energies in kcal/mol, distances in Å.

Fragments [Pd] / ZnMe [Pd] / ZnCp(ax) [Pd] / ZnCp(eq) [Pd] / Zn(tmeda)

r(A-B) 2.427 2.471 2.486 2.420

∆Eint -74.5 -65.3 -70.8 -88.0

∆EPauli 182.2 122.7 133.1 231.8

∆Eelstat -159.7 (62.2%) -122.0 (64.9%) -131.4 (64.4%) -211.3 (66.1%)

∆Eorb -97.0 (37.8%) -65.9 (35.1%) -72.5 (35.6%) -108.5 (33.9%)

The structural data of 23 as determined by single crystal X-ray diffraction and further

investigated by continous shape measure as well as the results of the bonding situation of

23M, leads to the suggestion that compound 23 is nothing other than a seven-fold

coordinated variant of [Pd(ZnCp*)4(ZnMe)4] at which two ZnMe ligands are replaced by

one {Zn(tmeda)} ligand. Notably, all compounds of the general types [M(ZnR)n] and

[M(ZnR)n(GaR)m] fulfil the 18 valence electron rule with respect to 1e ZnR ligands and

GaR units regarded as 2e donor ligands whereas the formal oxidation states of the

transition metal centres are assigned as M(0). In addition, the structural features of the

Zn/Ga mixed-metal compounds [M(ZnR)a(GaR)b] can be directly derived from the


homoleptic parent compounds [M(ZnR)n] in the same way as discussed above for

[Pd(ZnCp*)4(ZnMe)4] and 23.[43] Obviously, the {Zn(tmeda)} ligand can be treated as an

electronical equivalent to two ZnR ligands and is quite similar to one GaR ligand. This

reasoning supports a formal assignment of the {Zn(tmeda)} ligand as a strong 2e donor

ligand and determination of the oxidation state Zn(0) supported by the calculated NBO

partial charge of Pd in 23M (-3.05e) which is significantly higher than in [Pd(ZnH)8]

(-1.95e). In addition, the greater donor strength of {Zn(tmeda)} in comparison to ZnCp

and ZnMe, can also be derived from the calculated partial charges of the ligands in 23M.

The Zn atoms in ZnCp and ZnMe are thus formally Zn(I) species while {Zn(tmeda)}

exhibits a Zn(0) atom. The analysis of the electronic structure reveals significant

differences between the 1e and 2e donor ligands. It should be noted at this point, that there

are no correlations between the partial charges and the oxidation state of an atom in

general.

The reaction scheme between [PdMe2(tmeda)] and [Zn2Cp*2] and thus the formation of 22

and especially 23 seems to be the key step towards the selective trapping of Zn(0) atoms in

the ligand sphere of transition metal centres and significantly extends the synthetical

pathways towards metal-rich molecules at the borderline to intermetallics. Theoretical

calculations clearly support {Zn(tmeda)} as a strong 2e donor ligand. Although, the isolation

of 23 leaves it feasible that more than one terminal unit ZnLn can substitute M’R ligands (M’

= Zn, Cd; Al, Ga, In; R = Me, Cp*) in the compounds [M(M’R)n], further substitution of 1e

ZnR ligands by {Zn(tmeda)} has not occurred. Whether it is energetically unfavourable to

obtain hypothetical 18 valence electron species such as [Pd(ZnCp*)4{Zn(tmeda)}2] or

[Pd{Zn(tmeda)}4], which have not been observed as trace products in the reaction scheme,

will be subject of further theoretical studies. However, the dimeric Zn(I) reactant [Zn2Cp*2]

offers a rich chemistry in combination with substitution labile transition metal complexes.

Thus, it has been shown and previously described that reaction pathways can be easily

controlled via stoichiometry and reaction conditions leading to metal-rich compounds such as

[Ni(ZnCp*)4(ZnZnCp*)4] (19) or rather Cp* transfer reactions as it has been shown in the

formation of [Cp*M(ZnCp*)3] (M = Ni (20), Pt (21)). Besides the formation of 1e ligand

fragments ZnCp* and {ZnZnCp*} as well as possible Cp* transfer products, even the

unsupported Zn(0) ligand {Zn(tmeda)} has been trapped in compound 23. In any case,

product formation seems to be thermodynamically controlled following the 18 valence

electron rule as a reliable heuristic guide for rationalization of products and for the planning


of synthesis. These results present a novel, unprecedended concept in the formation of zinc-

rich compounds which offers a multitude of reaction schemes dependent on the starting

material used and the used (strong) Lewis base.

117 4. Summary

4 Summary

This work reports the synthesis and characterisation of novel transition metal compounds

containing low valent zinc and gallium ligands. Over the last years substantial progresses

have been made in the field of coordination chemistry of low valent group 13 elements.[10, 17,

18] Nevertheless, several significant restrictions have been observed in the formation of

homoleptic and heteroleptic GaCp* containing complexes as well as in their applications in

the formation of metal-rich molecules at the borderline to intermetallics. Neither coordination

numbers n > 4 in fully homoleptic compounds [M(GaCp*)n] could be prepared, nor have

oligonuclear compounds of the Cu/Ga systems been reported. In addition, while highly

coordinated compounds of the general type [M(ZnR)n] (n ≥ 8) have been prepared from

mononuclear starting materials, no investigations have been carried out using dimeric starting

compounds in order to achieve cluster growth.[43, 207] The presence of GaCp* in the

coordination sphere of the transition metal is necessary to obtain zinc-rich molecules. This

makes the synthesis of such compounds somehow special and limited with respect to the

choice of starting materials. Finally, quite recently, the preparation of [Zn2Cp*2] aroused great

attention.[138] Initial investigations have been made into the exploration of the reactivity of this

unusual compound considering the formation of derivatives[137, 156, 158, 159], studying the

electronic structure[142, 147, 149] and formation of (classical) Lewis acid/base adducts[164, 166, 167],

but the coordination chemistry of this Zn(I) dimer has not been studied at all.

The focus of this thesis lays on further investigations of low valent Ga(I) coordination

chemistry as well as first experimental and theoretical studies on the coordination chemistry

of [Zn2Cp*2] towards transition metals. The most meaningful results include the preparation

of the first oligonuclear CuaGab compounds in which Cu(I) and Cu(0) centres are stabilized by

low valent Ga(I)R species. Furthermore, initial investigations in controllable cluster growth

using dimeric starting materials have been made. Herein, the formation of donor stabilized,

trapped zinc-rich intermediates as well as the formation of a metal-rich palladium dimer are

reported. Finally, in the course of this thesis a previously unknown chemistry has been

discovered: first results in the coordination chemistry of [Zn2Cp*2] towards transition metals

lead to trapped {ZnZnCp*} ligands as well as the first unsupported Zn(0) unit in the

coordination sphere of d10 transition metals.

118 4. Summary


Molybdenum and Rhodium GaR (R = Cp*, DDP) Containing

Complexes

The highest coordination number of fully homoleptic transition metal GaCp* containing

compounds [M(ER)n] has been restricted to n = 4.[126, 128] Thus, analogues to the classic penta-

or hexa-carbonyl complexes, [M(GaCp*)5] (M = Fe, Ru, Os) and [M(GaCp*)6] (M = Cr, Mo,

W), have not been reported, so far. However, two exceptions have been reported containing

pseudo-homoleptic structural features, namely (1) the C-H activated isomers [M(AlCp*)5][279]

(M = Fe, Ru) as well as (2) the cation [Rh(GaCp*)4(GaCH3)]+.[209] Experimental studies

showed that olefins could not be fully substituted from [M(olefin)x] or [M(olefin)x(PR3)y]

starting materials and heteroleptic products [LnM(GaCp*)m] (L = olefin, PR3) were isolated

instead.[222] Within this thesis first homoleptic compounds [M(GaCp*)n] with n ≥ 5 could be

achieved under suitable conditions. The reaction of the olefin containing Mo(0) compound

[Mo(η4-butadiene)3] with excess GaCp* under hydrogen atmosphere and high temperatures of

around 100°C lead to the formation of the hexa-gallylene compound [Mo(GaCp*)6] (1).

Figure 45. Synthesis of [Mo(GaCp*)6] (1).

119 4. Summary

In comparison, reactivity of the Rh(I) dimer [Rh(coe)2(CF3SO3)]2 towards GaCp* takes place

under mild conditions without the presence of hydrogen, leading to the first fully homoleptic

penta-gallylene compound [Rh(GaCp*)5][CF3SO3] (3). Notably, anion exchange reaction of 3

with NaBArF leads to [Rh(GaCp*)5][BArF] (4) which shows highly unstable character in re-

dissolution processes. These results nicely point out the exceptional position of the Cp* group

in research of this kind. The fluxional behavior and the facile haptotropic shift reduce the

strain of the otherwise steric overcrowded situation in the case of rigid substituents.

Nevertheless, redox chemical processes, which include Cp* transfer reactions, may limit the

stability of compounds of the general type [M(GaCp*)n] (n ≥ 5) as it has been shown for 4.

The importance of co-ligands as well as the selection of the low valent Ga(I) species for these

reaction pathways has been illustrated in the formation of [cis-Mo(GaCp*)2(PMe3)4] (2) and

[(coe)(toluene)Rh{Ga(DDP)}(CF3SO3)] (5). Compound 2 is formed via reaction of

[Mo(N2)(PMe3)5] with GaCp*. Ligand replacement of phosphane groups is limited which is

most likely due to electronic reasons concerning stronger π-back bonding between the

molybdenum centre and the remaining phosphane ligands as it has been discussed for the

corresponding carbonyl containing complexes. Finally, the reaction of [Rh(coe)2(CF3SO3)]2

with Ga(DDP) does not lead to a homoleptic [RhGa5] complex, but rather the heteroleptic

mono gallium complex [(coe)(toluene)Rh{Ga(DDP)}(CF3SO3)] (5) was isolated due to

greater steric bulk and the rigidity of the Ga(I) species. Most interestingly, in compounds 3

and 5 no interaction between the Rh(I) centre and the nucleophilic triflate anion {CF3SO3}

has been observed. The electrophilic character of the Ga centres is significantly increased due

to coordination towards the transition metal. Thus, Lewis acid/base adduct formation can be

observed between electrophilic Ga and the {CF3SO3} counter ion which acts as a weak

nucleophile. These results nicely show the importance of electronic and steric properties of

the two different Ga(I) ligands which effect their reactivity towards substitution labile

transition metal complexes, not only leading to different products but also allowing for

different reaction pathways.

4.2 First Dinuclear Copper/Gallium Complexes: Supporting Cu(0) and

Cu(I) Centres by Low Valent Organogallium Ligands

While the formation of mononuclear and oligomeric d10 metal GaCp* compounds is well

known, fewer investigations have been made for group 11 metal centres.[128, 130-132] For

instance, only three examples exhibiting direct Cu-Ga bond interactions have been reported in

120 4. Summary

the literature.[38, 39] All these structures consist of mononuclear units. In contrast, several

dimeric and polynuclear compounds of Cu(I) or Cu(II) can be found in the literature, at which

the applied ligand system such as N-heterocyclic carbenes as well as bulky ligands such as

phosphines and pyrazolylborates seems to play an important role.[280-282] In the recent past it

has been shown that GaCp* can be employed to stabilize dinuclear compounds of soft

cationic d10 coinage metal centres illustrated by the formation of the Ag(I) dimer

[Ag2(GaCp*)3(µ-GaCp*)2][CF3SO3]2.[39] Additionally, the stabilizing effect and

simultaneously reducing ability of Ga(DDP) has recently been used in several reactions.[225]

In the course of this thesis three dimeric Cu/Ga compounds have been prepared via reductive

coordination reaction of GaCp* and Ga(DDP) with easily available Cu(II) and Cu(I) starting

materials. The reaction of [Cu(CF3SO3)2] with two equivalents of Ga(DDP) results in the

formation of the Cu(I) dimer [{(DDP)GaCu(CF3SO3)}2] (6) via mild reductive pathways

under the elimination of [(CF3SO3)2Ga(DDP)]. Compound 6 features [(DDP)GaCu(CF3SO3)]

dimeric units with a planar four-membered [Cu2Ga2] ring and exhibits the shortest

Cu(I)••••Cu(I) distance known so far. The analogous reaction of [Cu(CF3SO3)2] with five

equivalents GaCp* instead of Ga(DDP) leads to the formation of the unusual [Cu2Ga5]

compound [(Cp*Ga)Cu(µ-GaCp*)3Cu{Ga(CF3SO3)3}] (7).

Figure 46. Synthesis of [(Cp*Ga)Cu(µ-GaCp*)3Cu{Ga(CF3SO3)3}] (7).

121 4. Summary

Herein, several competing reaction sequences have to be taken into account. Firstly, redox

chemical processes at which Ga(I) reduces the Cu(II) centre of the starting compound to

Cu(0) and Ga(I) is oxidized to its favoured oxidation state +III found in the fragment

{Ga(CF3SO3)3}. Secondly, coordination of GaCp* ligands to the present Cu(0) centres giving

rise to the (neutral) 30 valence electron fragment [Cu2(GaCp*)4] and finally, coordination of

the {Ga(CF3SO3)3} ligand to one vacant Cu centre. Thus, 7 can be best described as a Lewis

acid/base adduct at which the Ga(III) ligand acts as the Lewis acid and the Cu centre features

Lewis basic character. Herein, the Lewis acid/base interactions Cu(0)→Ga(III) seem to be

stronger than Cu(I)←Ga(I) and Cu(0)←Ga(I) interactions as it is in good agreement with

structural features found for compound 7. Although, the assignment of formal oxidation states

is based on heuristic reasons, the oxidation states of copper and gallium in the fragment

[Cu2(GaCp*)4] can be best declared as Cu(0) and Ga(I), as mentioned above in course of

redox chemical processes leading to the first known structurally characterised Cu(0) complex

or rather cluster. While Ga(DDP) has been proven to be unusable as reactant with Cu(I)

compounds, GaCp* reacts with [{Cu(cod)2}(CF3SO3)] under the formation of the dimeric

compound [Cu2(GaCp*)3(µ-GaCp*)2][CF3SO3]2 (8). Notably, the (hypothetical) dication

[Cu2(GaCp*)5]2+ described in 8, the (neutral) fragment [Cu2(GaCp*)4] found in compound 7

and the dimeric d10 metal compounds [M2(GaCp*)5] (M = Pd, Pt) all feature an electron count

of 30. Thus, the coordination of one triflate to one Cu(I) centre leads to the suggestion, that

compound 8 can be viewed as a trapped intermediate of electronically saturated, 30 electron

[Cu2(GaCp*)5]2+ fragments. In compound 7, the coordination of triflate to the copper centre

has not been observed due to significantly higher electron density on the formal Cu(0)

centres. However, the ‘naked’ dication [Cu2(GaCp*)5]2+ (M = Cu, Ag) should be a tangible

target for further experimental studies when an appropriate, bulky and very weakly

coordinating anion is chosen. In general, the formation of the first molecular CuaGab units is

an outstanding advancement towards intermediates or starting precursors for soft chemical

synthesis of larger M/E intermetallic clusters or nanoparticles.

122 4. Summary

4.3 Experimental and Theoretical Investigations on the Formation of

Zinc-rich Oligonuclear Cluster Compounds

The reaction of mononuclear transition metal GaCp* compounds with ZnR2 (R = Me, Et)

derived easy access to metal-rich compounds of the general formula [M(ZnR)n] (M = Mo, Ru,

Rh, Ni, Pd, Pt; n = 8-12).[42, 43] While underlying reaction schemes of this class of compounds

have been well investigated, the controlled cluster growth has been proven to be somehow

more puzzling. For instance, [{(CO)4Mo}4(Zn)6(µ-ZnCp*)4][207] is formed from [cis-

Mo(CO)2(GaCp*)4], while the reaction of [fac-Mo(CO)3(GaCp*)3] with ZnMe2 leads to the

mononuclear compound [Mo(CO)3(ZnCp*)3(ZnMe)3].[236] In the course of this thesis, further

experimental and theoretical investigations have been effected to gain more detailed insights

into controllable cluster growth and how it is influenced by the nuclearity of the starting

material, the ratio between inert co-ligand and GaCp* as well as usage of fully homoleptic

dimeric transition metal GaCp* starting compounds.

4.3.1 Zinc-rich Compounds of Iron and Cobalt: Formation of [Fe2Znx] (x = 2-4) and

[Co2Zn3] Cores

The dependency of product formation from the nuclearity, the CO/GaCp* ratio and, most

surprisingly, from the solvent used for crystallisation procedures has been illustrated by the

reactions of heteroleptic mononuclear and dimeric iron and cobalt compounds with ZnMe2.

The reaction of [Fe(CO)4(GaCp*)] with excess ZnMe2 and crystallisation from a saturated thf

solution yields [(CO)4Fe{µ2-Zn(thf)2}2Fe(CO)4] (9). While [(CO)3Fe{µ2-Zn(thf)2}2(µ2-

ZnMe)2Fe(CO)3] (10) has been prepared from [(CO)3Fe(µ2-GaCp*)3Fe(CO)3] under similar

(crystallisation) conditions as has been mentioned for 9. The crystallisation from a

toluene/pyridine mixture leads to the formation of [(CO)3Fe{µ2-Zn(py)2}3Fe(CO)3] (11).

Notably, no influence of the solvent on product formation has been observed in the formation

of [(CO)3Co{µ2-Zn(py)2}(µ2-ZnCp*)2Co(CO)3] (12) and [(CO)3Co{µ2-Zn(thf)2}(µ2-

ZnCp*)2Co(CO)3] (13) prepared from [(CO)3Co(µ2-GaCp*)2Co(CO)3] with excess ZnMe2.

It becomes clear that all novel compounds 9-13 are formed via full replacement of the 2e

donor GaCp*. For this observation two possible explanations have been taken into account.

One possibility is the substitution by formal 2e donor units {Zn0L2} (L = thf, pyridine) and

the second possible explanation is based on replacement by two 1e donor ligands ZnMe or

123 4. Summary

rather ZnCp* based on Cp* transfer reactions from gallium to zinc. Thus, the over-all electron

count between the starting material and the products seems to be unaffected which is in good

agreement with the exchange rate and the 18 valence electron rule predicted for the

mononuclear compounds [M(ZnR)n]. Although, no larger clusters with higher nuclearity have

been observed due to low solutibility of the initially formed solid, compounds 9-13 can be

viewed as trapped, donor stabilized intermediates of greater agglomerates. The new core

structures [Fe2Zn3], [Fe2Zn4] and [Co2Zn3] have not been reported before and are likely to be

not easily accessible by synthesis routes other than the reported one.

Figure 47. Synthesis of [(CO)3Co{µ2-Zn(py)2}(µ2-ZnCp*)2Co(CO)3] (12).

4.3.2 Case Study on the Formation of an Oligonuclear Model System for Intermetallic

Phases: Synthesis, Characterisation and Theoretical Investigations on the

Compound [Pd2Zn6Ga2(Cp*)5(CH3)3]

Controlled cluster growth has been obtained in the reaction of [Pd2(µ-GaCp*)3(GaCp*)2] with

stoichiometric amounts of ZnMe2 leading to the first dimeric cluster compound

[Pd2Zn6Ga2(Cp*)5(CH3)3] (14) featuring a 30 valence electron [Pd2Ga2Zn6] core wrapped into

an all-hydrocarbon shell. Compound 14 consists of two Cs symmetric isomers in a ratio of

approximately 1:3 as determined by NMR spectroscopic measurements. While pure crystals

of 14 are stable for several weeks, dissolution in course of spectroscopic measurements

reveals high instability which leads to undefined decomposition products. The possible

124 4. Summary

existence of Cs symmetric [Pd2Zn4Ga4(Cp*)5(CH3)3] has been ruled out via mass

spectrometry. The structural features determined by singly crystal X-ray diffraction illustrate

a bi-capped trigonal prism. Herein, one palladium atom (Pd1) is embedded in the center of a

Pd/Zn/Ga trigonal prism, with the other palladium atom (Pd2) as well as one EMe unit as

capping ligands. Primary theoretical investigations have been made by MO correlations, AIM

and EDA analysis. They lead to the suggestion of significant attractive Pd-Pd interactions

based on bonding combinations of the dz2 AOs of the metals and bonding combination of the

dxz AOs with additional bonding contributions from the bridging Ga atom. These observations

are verified by calculated atomic partial charges taken from NBO calculations indicating a

large negative charge for Pd1 (-2.99e) and a smaller negative charge for Pd2 (-0.76 e), thus,

donation of electronic charge from Pd1 to the latter Pd2 occurs.

Figure 48. Synthesis of [Pd2Zn6Ga2(Cp*)5(CH3)3] (14).

4.4 Experimental and Theoretical Investigations on the Coordination

Chemistry of [Zn2Cp*2] Towards Transition Metal Compounds

The synthesis and characterisation of [Zn2Cp*2] by Carmona in 2004 has been one of the

latest impressive landmarks in the stabilisation of low valent main group metal centres.[138]

Quite some time after the publication of this Zn(I) dimer, a wide variety of theoretical studies

as well as first investigations on the reactivity have been published, of which the latter has

been mainly based on the synthesis and characterisation of derivative structures [Zn2R2] (R ≠

125 4. Summary

Cp*) and the exploration of Lewis acid/base adducts under the preservation of the intact

Zn(I)-Zn(I) bond. While the coordination chemistry of Cp* stabilized low valent group 13

elements has been well established since their development in the 1990s, nothing was known

about the coordination chemistry of [Zn2Cp*2] towards transition metals. In order to gain

initial insights into the rich chemistry of low valent Zn(I), reactivity studies have been carried

out with suitable transition metal complexes.

4.4.1 Trapping Monovalent {ZnZnCp*} at d10 Transition Metal Centres

First investigations on the reactivity of [Zn2Cp*2] included reactions of homoleptic GaCp*

containing d10 metal complexes [M(GaCp*)4] (M = Pd, Pt) with [Zn2Cp*2] leading to a

product mixture of the hexa-coordinated complexes [M(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (M =

Pd (15), Pt (17)) and the octa-coordinated complexes [M(ZnCp*)4(ZnZnCp*)4] (M = Pd (16)

Pt (18)). Most interestingly, a novel ligand system {ZnZnCp*} could be obtained featuring

fully intact Zn-Zn interactions.

Figure 49. Molecular structures of [Pd(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (15) and [Pd(ZnCp*)4(ZnZnCp*)4] (16)

in the solid state.

Compounds 15-18 show high instability in pure crystalline form even under inert gas

atmosphere, while their stability can be significantly increased when covered with small

amounts of an inert, non-polar solvent such as n-hexane or toluene. In situ NMR

spectroscopic measurements allowed first insights into the possible reaction mechanism

which is mainly based on dissociation/association equilibria rather than redox chemical

126 4. Summary

processes. Herein, GaCp* dissociates from the starting material leading to unsaturated

palladium fragments which trap monovalent ZnCp* ligands. The implementation of

{ZnZnCp*} ligands proceeds most likely via Cp* transfer from the parent compound

[Zn2Cp*2] to ZnCp* under loss of ZnCp*2. The existence of free GaCp* and the Zn(II)

compound ZnCp*2 has been observed as an unstable fluxional intermediate

{Cp*Ga••••ZnCp*2}. In addition, the existence of free GaCp* shifts the dissociation

equilibrium towards the adduct side, so that pure compounds of the fully substituted all-zinc

[MZn12] cores cannot occur. In contrast to the formation of [M(ZnR)n] compounds, no

fulvalene species or Ga(III) by-products have been detected which clearly excludes redox

chemical pathways. One important structural feature of the [M(ZnCp*)4(ZnZnCp*)4] (M = Pd

(16) Pt (18)) compounds is the existence of interior, unsupported penta metal atom chains Zn-

Zn-M-Zn-Zn as they can be found in some solid state intermetallics.

4.4.2 First Reactivity Studies of [Zn2Cp*2] Towards Olefin Containing d10 Transition

Metal Centres

Based on the results obtained from GaCp* containing complexes, further investigations into

the reactivity of this special Zn(I) dimer towards reactive transition metal complexes in the

absence of GaCp* have been carried out. The reaction of [M(cod)2] (M = Ni, Pt) with eight

equivalents [Zn2Cp*2] at 80°C for 30 h results in the formation of [M(ZnCp*)4(ZnZnCp*)4]

(M = Pt (18), Ni (19)). Mechanistic studies derived from NMR spectroscopic measurements

indicate the liberation of 1,3-cod and unreacted [Zn2Cp*2] as well as the formation of ZnCp*2.

In addition, [Cp*M(ZnCp*)3] (M = Ni (20), Pt (21)) could be assigned as a minor by-product.

In summary, the formation of [M(ZnCp*)4(ZnZnCp*)4] (M = Pt (18), Ni (19)) from [M(cod)2]

involves the liberation of 1,3-cod and several redox chemical processes leading to ZnCp*2,

the reduction of Zn(I) to elemental zinc, observed as a grey precipitate and the oxidation of

Ni(0) to Ni(I) attended by Cp* transfer reaction. The reaction sequences in the formation of

18 and 19 are most likely the same as discussed previously: The release of 1,3-cod and the

trapping of monovalent ZnCp* species by unsaturated transition metal centres is followed by

the formation of one electron {ZnZnCp*} fragments. The latter are formed by Cp* transfer

reactions between the starting compound [Zn2Cp*2] and the Zn centres of lower coordinated

intermediate species of the type [LaNi(ZnCp*)b] (L = 1,5-cod; ZnCp*) which lead to the

release of ZnCp*2 as the second by-product. In the formation of [Cp*M(ZnCp*)3] (M = Ni

(20), Pt (21)), [Zn2Cp*2] acts as a smooth oxidizing agent for the transition metal centres as

127 4. Summary

well as a natural source in the formation of ZnCp* ligands via Zn(I)-Zn(I) bond cleavage. The

selective formation of Cp* transfer products 20 and 21 has been successful in the reaction of

[M(cod)2] (M = Ni, Pt) with two equivalents [Zn2Cp*2] at 80°C for 3 h. It could be shown by

NMR spectroscopic measurements, that the reaction pathway includes the loss of 1,3-cod,

Zn(I)-Zn(I) bond cleavage to obtain ZnCp* fragments and, finally, redox chemical processes

leading to Cp* transfer from the ZnCp* fragments to the transition metals, whereas the

transition metal is oxidized M(0)→M(I) and the Zn(I) reduced to its bulk material.

Figure 50. Synthesis of [Ni(ZnCp*)4(ZnZnCp*)4] (19) and [Cp*Ni(ZnCp*)3] (20).

The synthesis and characterisation of compounds 19-21 present the versatile properties of

[Zn2Cp*2] in the formation of transition metal-zinc compounds. Most importantly, significant

differences have occurred in comparison to the preparation of highly coordinated molecules

of the general formula [M(ZnR)n]. While in the latter case M-ZnR bond formation is only

observed from GaCp* containing complexes, product formation in the case of [Zn2Cp*2] can

be obtained from reactive transition metal starting materials in the absence of GaCp*

providing a promising, easier approach to zinc-rich molecules. Furthermore, it has been

shown that reaction pathways are easily controlled by stoichiometry and reaction conditions.

While compounds of the type [M(ZnCp*)4(ZnZnCp*)4] are formed with higher amounts of

[Zn2Cp*2] and over a period of 30 h, the Cp* transfer products [Cp*M(ZnCp*)3] can be

obtained using lower amounts of the Zn(I) dimer and shorter reaction times.

128 4. Summary

4.4.3 Experimental and Theoretical Investigations on the Formation of a Novel

[PdZn7] Compound: [Zn2Cp*2] as a Source for Stabilized Zn(0)

In order to expand the reactivity studies based on substitution labile transition metal starting

materials, [PdMe2(tmeda)] has been chosen due to the ability to form metal-rich compounds.

The reaction of [PdMe2(tmeda)] with four equivalents [Zn2Cp*2] leads to the formation of

[Cp*Pd(ZnCp*)3] (22) and [Pd(ZnCp*)4(ZnMe)2(Zn{tmeda})] (23) as principal products in a

ratio of 1:1. In situ 1H NMR spectroscopic studies revealed several rare by-products such as

[Pd(ZnMe)4(ZnCp*)4] as well as various Zn(II) species. These results indicate competition

between redox chemical reaction pathways, coordination steps of Zn(I)R and Zn(0)L towards

the transition metal centre as well as insertion reactions. The latter can be proceed via

homolytic cleavage of the Zn(I)-Zn(I) or disproportionation of the dimeric Zn(I) unit into

Zn(0) and Zn(II). Continuous shape measurements showed that the solid state structure of 23

can be derived from a trigonal dodechadron as it has been found for [MZn8] cores at which

two corners are replaced by the {Zn(tmeda)} unit. Theoretical calculations based on AIM and

EDA analysis indicate a significant difference between the 1e donor ligands ZnR (R = Cp*,

Me) and the {Zn(tmeda)} ligand which can be best described as an unsupported strong 2e

donor fragment.

Figure 51. Synthesis of [Cp*Pd(ZnCp*)3] (22) and [Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}] (23).

129 4. Summary

Several sections of this thesis have already been published in peer-reviewed journals. These

articles are reproduced in part.

(1) T. Bollermann, T. Cadenbach, C. Gemel, K. Freitag, M. Molon, V. Gwildies, and R.

A. Fischer, ‘Homoleptic Hexa and Penta Gallylene Coordinated Complexes of

Molybdenum and Rhodium’, Inorg. Chem. 2011, 50, 5808-5814.

(2) M. Molon, T. Bollermann, C. Gemel, J. Schaumann, and R. A. Fischer, ‘Mixed

phosphine and group-13 metal ligator complexes [(PR3)aM(ECp*)b] (M = Mo, Ni; E =

Ga, Al; R = Me, C6H5, cyclo-C6H11)’, Dalton Trans. 2011,

DOI:10.1039/C1DT10583C.

(3) T. Bollermann, G. Prabusankar, C. Gemel, R. W. Seidel, M. Winter, and R. A.

Fischer, ‘First dinuclear Copper/Gallium Complexes: Supporting Cu(0) and Cu(I)

centres by low valent Organogallium Ligands’, Chem.-Eur. J. 2010, 16(29), 8846-

8853.

(4) T. Bollermann, K. Freitag, C. Gemel, R. W. Seidel, M. von Hopffgarten, G. Frenking,

and R. A. Fischer, ‘Chemistry of [Zn2Cp*2]: Trapping monovalent .ZnZnCp* in the

metal rich compounds [(Pd, Pt)(GaCp*)a(ZnCp*)4-a(ZnZnCp*)4-a] (a = 0, 2)’, Angew.

Chem. Int. Ed. 2011, 50(3), 772-776.

(5) T. Bollermann, K. Freitag, C. Gemel, R. W. Seidel, and R. A. Fischer, ‘Reactivity of

[Zn2Cp*2] towards Transition Metal Complexes: Synthesis and Characterisation of

[Cp*M(ZnCp*)3] (M = Ni, Pd, Pt)’, Organometallics 2011, 30 (15), 4123-4127.

130 5. Outlook

5 Outlook

The major part of this thesis deals with the successive exploration of possible synthetic routes

for larger cluster units. In this context an important result of the work has been the controlled

cluster growth and trapping of donor stabilized intermediates of zinc-rich molecules using

dimeric heteroleptic and homoleptic GaCp* compounds such as [(CO)3Co(µ2-

GaCp*)2Co(CO)3] and [Pd2(µ-GaCp*)3(GaCp*)2]. It is a promising perspective to extend this

chemistry to homoleptic and heteroleptic oligonuclear compounds of the general formula

[LaMb(GaCp*)c] (c > b > a; b > 2).[40, 116, 128, 130-132] In addition, the used ligand systems can be

helpful initiators. For instance, several heteroleptic mononuclear and dimeric compounds are

known which contain substitution labile ligands such as olefins and carbon monoxide,

meaning that the cluster growth under hydrogenolytic or photolytic conditions in the presence

of ZnMe2 is an important subject of ongoing research.

Additionally, in the course of this thesis initial experimental and theoretical studies on the

coordination chemistry of Carmona´s low valent Zn(I) compound [Zn2Cp*2] towards

transition metal centres have been investigated. The reaction of [Zn2Cp*2] with homoleptic

GaCp* d10 metal complexes, with substitution labile olefin containing compounds and in the

presence of the N-chelating ligand tmeda used in [PdMe2(tmeda)], represents unprecedented

excess to unusual compounds. In this context, structural features such as [MZn12] metal cores

and novel ligand systems, for example {ZnZnCp*} and stabilized Zn(0) in {Zn(tmeda)} have

become possible only within the use of [Zn2Cp*2]. The research most closely related to the

work presented in this thesis will be the reaction of [Zn2Cp*2] with GaCp* containing starting

materials. While the reaction of [M(GaCp*)4] (M = Pd, Pt) with [Zn2Cp*2] leads to hexa and

octa coordinated complexes, larger units should be obtained using monomeric homoleptic

compounds of the general type [M(GaCp*)n] (n ≥ 5) and oligonuclear starting materials. [279]

A second important focus of reseach confirms definitely the usage of substitution labile

transition metal complexes in reactions of [Zn2Cp*2] as has been investigated in this thesis for

[M(cod)2]. Since [Zn2Cp*2] showed its versatile properties as a natural source for several

ligand systems, namely ZnCp*, {ZnZnCp*} and {Zn(tmeda)}, it is auspicious to follow

reactions reported by the groups of Boersma and van der Kerk in the 1980's. They

investigated the reactivity of ZnCp2 towards substitution labile compounds such as early

transition metal hydride complexes and nitrogen containing compounds.[191-194] In addition,

131 5. Outlook

even dimeric and oligomeric hydride compounds have been used in the formation of M-ZnR

bonds.[198, 199, 205, 206] In general, the reaction of [M(cod)2] (M = Ni, Pt) with [Zn2Cp*2] opened

the door to a rich chemistry of reactive transition metal compounds across the periodic table

from the full range of possible reactive starting materials, i.e. from early group 4 transition

metals up to late group 11 elements.

In order to obtain greater cluster units it seems to be also possible to start from pure [MZn12]

fragments. Herein, three common pathways have been found to be suitable in cases of low

valent group 13 metal chemistry and in the formation of Pd-Zn bonds. Firstly, cleavage of the

Cp* group under protolytic conditions, secondly, reactions under hydrogenolytic conditions

and finally, bond formation under photolytical conditions. All these reaction schemes can be

considered as useful tools in the formation of larger agglomerates starting with {ZnZnCp*}

containing compounds.

The reaction of Zintl compounds such as [Ge9]2- and [Sn9]2- with transition metal compounds

lead to intercalation of transition metals into the cluster-like framework as well as capping of

open sites of the cage fragments by [MLn] units.[67, 70] It is a great challenge to prepare similar

products using [Zn2Cp*2]. Herein, intercalation of both zinc sites into Zintl cages would be

something like fullerene-dizincocene hybrid materials mentioned in the introduction.

The attachment of {Zn(tmeda)} to a palladium centre in the compound

[Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}] is the first example of an unsupported 2e Zn(0) donor

ligand. Thus it should be possible that more than one terminal unit ZnLn can substitute M’R

ligands (M’ = Zn, Cd; Al, Ga, In; R = Me, Cp*) in the compounds [M(M’R)n]. This could

lead to species such as [Pd(ZnCp*)4{Zn(tmeda)}2] or even [Pd{Zn(tmeda)}4]. Theoretical

calculations to study the relative bond strength of the 2e donor {Zn(tmeda)} and two 1e

donors ZnR are being performed right now. In the presence of stoichiometric amounts of

Lewis base ligands L (L = e.g. thf, py, bipy, tmeda) zinc-rich compounds could be possible

using reactive and substitution labile transition metal starting materials such as [M(cod)2].

Since this chemistry is new, it is difficult to predict from the compounds prepared so far

where it will lead to. However, one fact seems to be clear: there is plenty of scope for

development of ideas using [Zn2Cp*2] in the coming years.

132 6. Experimental Section

6 Experimental Section

6.1 Materials and Methods

6.1.1 General Remarks

All manipulations, synthesis of compounds and preparations of NMR samples were carried

out in an atmosphere of purified argon using standard Schlenk, vacuum and glove-box

techniques. Herein, purification of argon takes place via a copper catalyst and molecular sieve

(4 Å). Common solvents (n-hexane, pentane, toluene, THF and Et2O) were dried using an

MBraun SPS-Solvent Purification System. Herein, the solvents are pressed out of their

storage vessels by means of argon pressure (300-500 mbar) and passed through two filter

columns to remove the residual moisture from the solvent resulting in purities of about > 99%

and < 5 ppm water as determined by Karl-Fischer titration. The contents of these filter

columns depends on the solvent, so that alumina is used for THF and toluene, molecular sieve

or a copper catalyst for n-hexane and n-pentane. All glassware were silylated with 1,1,1,3,3,3-

Hexamethyldisilazane (0.5-1.0 ml) at ca. 100°C over a period of 2 h or heated by means of a

hot air gun in vacuo. Low temperature experiments were carried out using acetone/dry ice or

rather isopropanol/ dry ice cooling bathes at the desired temperatures. Deuterated solvents

were saturated with purified argon and stored over molecular sieve (4 Å). Benzene,

fluorobenzene, CH2Cl2 and pyridine were dried by standard purification procedures under

argon atmosphere.

6.1.2 Nuclear Magnetic Resonance Spectroscopy (NMR)

NMR spectra were recorded on a Bruker Avance DPX 250 spectrometer (1H, 250.1 MHz; 13C, 62.9 MHz; 19F, 235.3 MHz; 11B, 80.3 MHz; 31P, 101.3 MHz) at 298 K unless otherwise

stated. Chemical shifts δ are reported in ppm (parts per million), given relative to TMS and

were referenced to the residual solvent peak as internal standards as it is the case for 1H and 13C. Signal multiplicities are abbreviated as followed: s = singlet, d = doublet, t = triplet, q =

quartet, sept = septet, m = multiplet, br = broad signal. Coupling constants J are indicated in

Hz as values without appointment of signs. All spectra were analyzed with Mestrec® V

4.7.0.0 or rather MestreNova® V 6.2.0.


6.1.3 Single Crystal X-Ray Diffraction (XRD)

Crystals were coated with a perfluoropolyether, picked up with a glass fiber, and immediately

mounted in the cooled nitrogen stream of the diffractometer. The X-ray diffraction intensities

were collected on an Oxford Diffraction XcaliburTM2 diffractometer with a Sapphire2 CCD.

Absorption corrections were carried out semi-empirically on the basis of multiple-scanned

reflections or empirically using sadabs. The crystal structures were solved by direct methods

using SHELXS-97 and refined with SHELXL-97.[283] Contributions of (strongly) disordered

molecule parts or co-crystallized solvent molecules which could not sufficiently be modeled

were removed from the diffraction data with the SQUEEZE routine of PLATON.[284, 285]

Unique support has been given by Dr. Rüdiger W. Seidel in the solution of several crystal

structures. For detailed crystallographic date see Supplement 8.1.

6.1.4 Infrared Spectroscopy (IR)

IR spectra were measured with a Bruker Alpha-P FTIR spectrometer (KBr pellet) and with a

Bruker Alpha FTIR spectrometer under inert atmosphere in a glove-box (ATR setup).

6.1.5 Mass Spectrometry (MS)

MS measurements were recorded on a Joel AccuTOF GCv spectrometer. Ionization method:

Liquid Injection Field Desorption Ionization (LIFDI). All spectra were analyzed with

MestreNova V 6.2.0.

‘The entire LIFDI sample prep is dipping the capillary into the sample solution for 1-2 s. A

volume of ca. 40 nL is aspirated automatically and forced through the capillary. Upon arrival

at the emitter, osmotic and capillary forces between the dendrites distribute the solution over

the entire emitter. The small volume of solvent is evaporated in the vacuum within seconds,

followed by acquiring spectra at a total sample prep of less than 30 s. The LIFDI emitter is

safe in the ion source for many convenient sample preps for days or even weeks. The LIFDI

emitter cannot be harmed at all by the sample prep’.[286]


6.1.6 Elemental Analysis (EA) and Atom Absorption Spectroscopy (AAS)

Elemental analyses were performed at the Laboratory for Microanalytics of the Ruhr

University Bochum (CHNSO: Vario EL by Elementar Hanau; AAS: AAS 6 Vario by

Analytik Jena) and the University of Duisburg-Essen (CHNSO: EURO EA Elemental

Analyzer by EURO VECTOR; AAS: M-Serie by Thermo Electron).

6.2 Precursors

6.2.1 General Remarks

All precursors used for the preparation of (metal-)organic and inorganic substances were

purchased from commercial sources and used without further purification unless otherwise

stated.

List of synthesized precursors as previously described in the literature

NaBArF [287] [PdMe2(tmeda)][288]

[Pd2(GaCp*)2(µ-GaCp*)3][132] [{Cu(cod)2}(CF3SO3)][289]

GaCp* [116, 290] [Ni(cod)2][291]

Ga(DDP)[119] [Mo(N2)(PMe3)5][292]

[Zn2Cp*2][139] [Mo(η4-butadiene)3][293]

[ZnCp*2][294] [Rh(coe)2(CF3SO3)]2 [295]

[Pd(GaCp*)4][128] [Pt(GaCp*)4][128]

[Pt(cod)2][296] [Fe(CO)4(GaCp*)][116]

[(CO)3Co(µ2-GaCp*)2Co(CO)3][116] [(CO)3Fe(µ2-GaCp*)3Fe(CO)3][116]


6.2.2 Synthesis and Characterisation of Novel Compounds 1-23

[Mo(GaCp*)6] (1). A sample of freshly prepared [Mo(η4-C4H6)3] (0.300 g, 1.162 mmol) was

introduced into a Fischer-Porter bottle and then dissolved in toluene (12 ml). After addition of

GaCp* (1.666 g, 8.129 mmol) the reaction mixture was pressurized to 3 bar dihydrogen. The

orange solution was warmed to 100 °C, whereupon a red microcrystalline precipitate was

formed. After stirring for further 16 h at 80 °C the reaction mixture was transferred into a

schlenk tube. The red crystals were isolated by means of a cannula, washed with a small

amount of n-hexane and dried in vacuo. Re-crystallization from mesitylene gave well formed

dark red needle-shaped single crystals. Yield: 0.785 g (51 %). Elemental analysis (%)

calculated for C60H90Ga6Mo1: C, 54.36; H, 6.84. Found: C, 53.86; H, 6.24. 1H NMR

δH(C6D6), 1.96 (s, 90H, C5Me5). 1H NMR δH(toluene-d8), 1.95 (s, 90H, C5Me5). 1H NMR

δH(toluene-d8, -78°C), 1.99 (s, 90H, C5Me5). 13C{1H} NMR δC{H}(C6D6), 117.11 (C5Me5),

11.91 (C5Me5). IR (ATR, cm-1): 2943 (w), 2877 (m), 2829 (m), 2700 (w), 1478 (w), 1413 (m),

1363 (m), 1285 (w), 1011 (w), 934 (w), 789 (w), 723 (w), 639 (w), 587 (w), 414 (s). MS

(LIFDI, toluene): m/z 1326 [M].+, 1190 [M-Cp*]+.

[cis-Mo(GaCp*)2(PMe3)4] (2). To a solution of [Mo(N2)(PMe3)5] (0.518 g, 1.03 mmol) in

toluene (5 ml) GaCp* (0.464 g, 2.27 mmol) was added. The reaction mixture was heated at

60°C for 1 h, then the solvent was reduced in vacuo, the residue washed with small amounts

of cold n-hexane and dried in vacuo to give an orange-red solid. Re-crystallization of the

crude product from toluene at -30°C gave well formed orange single crystals. Yield: 0.521 g

(62 %) Elemental analysis (%) calculated for C32H62P4Mo1Ga2: C, 47.44; H, 8.21. Found: C,

46.46; H, 8.52. 1H NMR δH(C6D6), 2.09 (s, 30H, C5Me5), 1.35 (t, 18H, PMe3, 2JP-H = 2.4 Hz),

1.28 (t, 18H, PMe3, 2JP-H = 1,7 Hz). 13C{1H} NMR δC{H}(C6D6), 114.73 (s, C5Me5), 38.49 (t,

PMe3, 1JP-C = 7.9 Hz), 33.79 (tt, PMe3, 1JP-C = 9.5 Hz), 11.91 (s, C5Me5). 31P{1H} NMR

δP(C6D6), 4.53 (t, PMe3, 1JC-P = 2.4 Hz), 4.40 (t, PMe3, 1JC-P = 1.7 Hz). IR (cm-1): 2962 (m),

2898 (m), 2849 (m), 1420 (m), 1375 (m), 1289 (m), 1263 (m), 924 (s), 847 (m), 796 (m), 728

(m), 688 (m), 677 (m), 660 (s), 641 (s), 590 (m).

[Rh(GaCp*)5][CF3SO3] (3). To a solution of [Rh(coe)2(CF3SO3)]2 (0.200 g, 0.212 mmol) in

fluorobenzene (5 ml) GaCp* (0.443 g, 2.162 mmol) was added. The reaction mixture was


heated at 60°C for 1 h, then the solvent was reduced in vacuo, the residue washed with n-

hexane and dried in vacuo to give a red-purple solid. Re-crystallization of the crude product

by slow diffusion of n-hexane into a solution of 3 in fluorobenzene gave deep red single

crystals. Yield: 0.435 g (81 %). Elemental analysis (%) calculated for C51H75F3S1O3Rh1Ga5:

C, 48.11; H, 5.94; S, 2.51. Found: C, 47.73; H, 5.83; S, 2.43. 1H NMR δH(CD2Cl2), 1.91 (s,

75H, C5Me5). 1H NMR δH(d8-thf), 1.90 (s, 75H, C5Me5). 1H NMR δH(d8-thf, -78°C), 1.89 (s,

75H, C5Me5). 13C{1H} NMR δC{H}(CD2Cl2), 117.21 (C5Me5), 11.09 (C5Me5). 19F{1H} NMR

δ(CD2Cl2), -78.7 (s, CF3SO3). IR (ATR, cm-1): 2938 (w), 2885 (w), 2834 (w), 1580 (w), 1482

(w), 1409 (w), 1373 (w), 1285 (w), 1219 (s), 1179 (m), 1112 (m), 973 (s), 792 (w), 748 (w),

705 (w), 679 (w), 524 (w), 582 (w), 513 (w), 461 (w), 413 (w). MS (LIFDI, CH2Cl2): m/z 788

[M-{Cp*2Ga(CF3SO3)}]+, 373 [RhCp*2]+.

[Rh(GaCp*)5][BArF] (4). To a solution of 3 (0.120 g, 0.094 mmol) in CH2Cl2 (5 ml) NaBArF

(0.105 g, 0.109 mmol) and further 5 ml of CH2Cl2 were added. The reaction mixture was

stirred at room temperature for 45 min and filtered to remove the generated colorless NaOTf.

The solution was concentrated to ca. 4 ml and n-hexane (12 ml) was added which results in

the precipitation of an orange powder. The powder was isolated, the residue washed with n-

hexane and dried in vacuo to give a slightly orange solid. Yield: 0.184 g (94 %). Elemental

analysis (%) calculated for BC88H96F24Rh1Ga5: C, 49.47; H, 4.40. Found: C, 47.40; H, 4.36

(for deviation of C value see explanation in the main text). 1H NMR δH(CD2Cl2), 1.91 (s,

75H, C5Me5), 7.56 (br, 4H, BArF), 7.72 (br, 8H, BArF). 1H NMR δH(CD2Cl2, -78°C), 1.83 (s,

75H, C5Me5), 7.53 (br, 4H, BArF), 7.72 (br, 8H, BArF). 13C{1H} NMR δC{H}(CD2Cl2), 135.23

ppm (s, ortho CH, BArF), 128.99 (s, ipso to the CF3 units, BArF), 125.03 (q, -CF3, J = 272.3

Hz, BArF), 117.88 (s, para CH, BArF), 117.21 (C5Me5), 11.09 (C5Me5). 19F{1H} NMR

δ(CD2Cl2), -62.9 (s, CF3). 11B{1H} NMR δ(CD2Cl2), -6.6 (s, BArF). IR (ATR, cm-1): 2956

(w), 2895 (w), 2836 (w), 1597 (w), 1411 (w), 1375 (w), 1341 (m), 1264 (s), 1152 (w), 1114

(s), 939 (w), 878 (w), 831 (w), 792 (w), 738 (w), 707 (w), 676 (w), 663 (w), 585 (w), 444 (w).

MS (LIFDI, THF): m/z 373 [RhCp*2]+.

[(coe)(toluene)Rh{Ga(DDP)}(CF3SO3)] (5). [Rh(coe)2(CF3SO3)]2 (0.150 g, 0.159 mmol)

and Ga(DDP) (0.162 g, 0.334 mmol) in fluorobenzene (5 mL) were stirred at 50°C for 2 h.

The solvent was reduced in vacuo, the residue washed with n-hexane and dried in vacuo to


give a yellow solid. Re-crystallization of the crude product by cooling a saturated toluene

solution of 5 to -30°C gave yellow single crystals. Yield: 0.145 g (98 %). Elemental analysis

(%) calculated for C45H63F3S1O3N2Rh1Ga: C, 57.40; H, 6.74; N, 2.97; S, 3.41. Found: C,

58.12; H, 6.32; N, 3.22; S, 3.32. 1H NMR δH(C6D6), 6.99-7.15 (m, 11H, arylic DDP and

C6H5CH3), 5.20 (s, 1H, γ-CH), 3.48 (m, 4H, CH(Me)2, 2.79 (m, 2H, coe), 2.10 (s, 3H,

C6H5CH3), 1.69 (s, 6H, CH3 groups, DDP), 1.48-0.92 (m, 36H, CH(Me)2 and coe). 13C{1H}

NMR δC{H}(C6D6), 168.60 (CN), 143.35 (ar), 137.90 (C1, toluene), 129.30 (CH 2,6; toluene)

127.39 (CH 3,5; toluene), 126.97 (ar), 125.66 (CH 4; toluene), 124.43 (ar), 100.62 (γ-C),

57.16 (d, C=C, JRh-C = 14.1 Hz), 34.90 (coe), 32.40 (coe), 26.46, 25.27 (coe), 24.40, 21.48

(CH3, toluene) (Note: not all carbon signals of the DDP unit could be observed due to low

scan number measurement). 19F{1H} NMR δ(C6D6), -77.1 (s, CF3SO3). IR (ATR, cm-1): 2934

(w), 2902 (w), 2841 (w), 2820 (w), 2020 (w), 1540 (w), 1502 (w), 1449 (w), 1424 (w), 1377

(m), 1349 (w), 1333 (w), 1296 (w), 1276 (w), 1245 (w), 1222 (m), 1191 (m), 1156 (m), 1089

(w), 1047 (w), 1003 (m), 930 (w), 896 (w), 873 (w), 852 (w), 827 (w), 789 (w), 773 (w), 753

(w), 718 (w), 701 (w), 686 (w), 621 (m), 576 (w), 562 (w), 538 (w), 514 (w), 502 (w), 458

(w), 434 (w). MS (LIFDI, toluene): m/z 940 [M].+, 830 [M-coe]+, 791 [M-{CF3SO3]+.

[{(DDP)GaCu(CF3SO3)}2] (6). To a mixture of Ga(DDP) (0.1 g, 0.206 mmol) and

[Cu(CF3SO3)2] (0.037 g, 0.103 mmol), fluorobenzene (2.5 ml) was added under vigorous

stirring at rt. The reaction mixture was heated at 60°C for 1 h during which the clear pale

yellow solution became brown yellow with slight turbid. At this stage the solution was

brought to rt, filtered, layered with n-hexane (0.5 ml) to afford colorless crystals at rt over a

period of 12 h. Yield: 63% (based on [Cu(CF3SO3)2]). Compound 6 slowly decomposes at rt

in the absence of mother liquid under inert atmosphere or when it is dissolved in organic

solvents. Thus, the detailed study of molecule 6 was not successful. Elemental analysis (%)

calculated for C60H82Cu2F6Ga2N4O6S2: C, 51.48; H, 5.90; N, 4.00; S, 4.58. Found: C, 51.73;

H, 5.87; N, 3.87; S, 4.52. IR (KBr pellet, cm-1): 2962, 2926, 2868, 1528, 1460, 1440, 1381,

1362, 1318, 1296, 1255, 1228, 1211, 1164, 1103, 1055, 1019, 936, 863, 798, 757, 631, 573,

531, 515, 440. Note: this compound has been mainly discovered by Dr. G. Prabusankar as

stated in the Acknowledgment Section.

[Cu2(GaCp*)(μ-GaCp*)3{Ga(CF3SO3)3}] (7). To a solution of [Cu(CF3SO3)2] (0.150 g,

0.415 mmol) in fluorobenzene (5 ml) GaCp* (0.357 g, 1.742 mmol) was added. The reaction


mixture was heated at 60°C for 1 h, then the solvent was reduced in vacuo, the residue

washed with n-hexane and dried in vacuo to give a colorless solid. Re-crystallization of the

crude product by slow diffusion of n-hexane into a solution of 7 in fluorobenzene gave

colorless single crystals. Yield: 0.385 g (63 %). Elemental analysis (%) calculated for

C43H60F9S3O9Cu2Ga5: C, 35.40; H, 4.15; S, 6.58. Found: C, 35.65; H, 4.16; S, 6.31. 1H NMR

δH(THF-d8), 2.03 (s, 60H, C5Me5). 13C NMR δC{H}(THF-d8), 115.46 (C5Me5), 9.75 (C5Me5). 19F NMR (THF-d8), -78.8 (CF3SO3). IR (KBr pellet, cm-1): 1593, 1480, 1456, 1417, 1384,

1329, 1308, 1262, 1232, 1189, 1162, 1018, 998, 799, 754, 685, 631, 572, 513.

[Cu2(GaCp*)3(μ-GaCp*)2][CF3SO3]2 (8). To a suspension of [Cu(cod)2][(CF3SO3)] (0.150

g, 0.350 mmol) in fluorobenzene (5 ml) GaCp* (0.388 g, 1.093 mmol) was added whereupon

the suspension becomes a yellow solution. The reaction mixture was stirred at room

temperature for 1 h, then the solvent was reduced in vacuo, the residue washed with n-hexane

and dried in vacuo to give a colorless solid. Re-crystallization of the crude product by slow

diffusion of n-hexane into a solution of 8 in fluorobenzene gave colorless single crystals.

Yield: 0.354 g (70 %). Elemental analysis (%) calculated for C52H75F6S2O6Cu2Ga5: C, 43.21;

H, 5.23; S, 4.43. Found: C, 43.85; H, 5.47; S, 4.14. 1H NMR δH(THF-d8), 2.04 (s, 60H,

C5Me5). 13C NMR δC{H}(THF-d8), 112.2 (C5Me5), 7.1 (C5Me5). 19F NMR (THF-d8), -78.4

(CF3SO3). IR (KBr pellet, cm-1): 1480, 1451, 1420, 1384, 1305, 1268, 1230, 1203, 1157,

1015, 799, 756, 687, 632, 589, 571, 513.

[(CO)4Fe{µ2-Zn(thf)2}2Fe(CO)4] (9). To a suspension of [Fe(CO)4(GaCp*)] (0.085 g, 0.228

mmol) in toluene (5 ml) ZnMe2 (0.6 ml, 0.684 mmol, 1.2 M toluene solution) was added

whereupon the suspension becomes slightly yellow. The reaction mixture was stirred at 90°C

for 1 h. Then the solvent was reduced in vacuo and the residue dried in vacuo to give a

colorless solid which was dissolved in thf (1 ml) and stored at -30°C to give colourless

crystals. Yield: 0.098 g (57 %). Elemental analysis (%) calculated for C24H32Fe2O12Zn2: C,

38.18; H, 4.27; Zn, 17.32. Found: C, 40.44; H, 5.23; Zn, 17.12. 1H NMR δH(THF-d8), 1.77

(m, br, 16H, C4H8O), 3.62 (m, br, 16H, C4H8O). 13C NMR δC{H}(THF-d8), 68.03 (C4H8O),

26.19 (C4H8O),. IR (ATR, cm-1): 2938 (w), 2886 (w), 2838 (w), 2020 (m), 1970 (m), 1925

(m), 1899 (s), 1436 (w), 1401 (w), 1373 (w), 1249 (m), 1081 (m), 1008 (m), 856 (m), 789

(m), 695 (s), 605 (w), 533 (w), 492 (w), 446 (w).


[(CO)3Fe{µ2-Zn(thf)2}2(µ2-ZnMe)2Fe(CO)3] (10). To a suspension of [(CO)3Fe(µ2-

GaCp*)3Fe(CO)3] (0.150 g, 0.168 mmol) in toluene (10 ml) ZnMe2 (1.1 ml, 1.341 mmol, 1.2

M toluene solution) was added whereupon the suspension becomes orange. The reaction

mixture was stirred at 90°C for 1 h. The solvent was then reduced in vacuo and the residue

dried in vacuo to give an orange solid which was dissolved in thf (2 ml) and stored at -30°C to

give orange crystals. Yield: 0.068 g (45 %). Elemental analysis (%) calculated for

C24H38Fe2O10Zn4 x 4 (C4H8O): C, 41.84; H, 6.14; Zn, 22.78. Found: C, 41.46; H, 5.95; Zn,

22.70. 1H NMR δH(THF-d8), -0.57 (s, 6H, ZnMe), 1.76 (m, br, 16H, C4H8O), 3.61 (m, br,

16H, C4H8O). 13C NMR δC{H}(THF-d8), 68.02 (C4H8O), 26.18 (C4H8O), -1.67 (ZnMe). IR

(ATR, cm-1): 2937 (w), 2883 (w), 2837 (w), 2028 (w), 1963 (w), 1894 (w), 1849 (s), 1431

(w), 1401 (w), 1374 (w), 1249 (m), 1195 (w), 1080 (m), 1009 (m), 856 (m), 789 (m), 727 (w),

679 (s), 603 (w), 576 (w), 529 (w), 476 (w).

[(CO)3Fe{µ2-Zn(py)2}3Fe(CO)3] (11). To a suspension of [(CO)3Fe(µ2-GaCp*)3Fe(CO)3]

(0.150 g, 0.168 mmol) in toluene (10 ml) ZnMe2 (1.1 ml, 1.341 mmol, 1.2 M toluene

solution) was added whereupon the suspension becomes orange. The reaction mixture was

stirred at 90°C for 1 h. The solvent was then reduced in vacuo and the residue dried in vacuo

to give an orange solid. The solid was extracted with 3 ml of warm pyridine, the solvent

reduced in vacuo again to give a purple oil which was dissolved in a mixture of

toluene/pyridine (2 ml/2 ml) and stored at -30°C to give orange crystals. Yield: 0.083 g (52

%). Elemental analysis (%) calculated for C36H30Fe2N6O6Zn3: C, 45.49; H, 3.18; N, 8.84; Zn,

20.64. Found: C, 43.11; H, 3.07; N, 8.82; Zn, 20.60. 1H NMR δH(Pyridine-d5), 7.19 (s, br,

12H, C5H5N), 7.55 (s, br, 6H, C5H5N), 8.71 (s, br, 12H, C5H5N). 13C NMR δC{H}(Pyridine-d5),

149.77 (C5H5N), 135.43 (C5H5N), 123.50 (C5H5N). IR (ATR, cm-1): 3471 (w), 3047 (w),

3012 (w), 2975 (w), 1846 (w), 1803 (m), 1746 (s), 1728 (s), 1586 (w), 1558 (m), 1472 (w),

1431 (m), 1345 (m), 1229 (m), 1204 (m), 1142 (m), 1060 (m), 1028 (m), 999 (m), 977 (m),

941 (m), 874 (w), 745 (m), 689 (s), 648 (m), 638 (m), 624 (m), 614 (s), 603 (m), 558 (m), 453

(m), 439 (m), 414 (m).

[(CO)3Co{µ2-Zn(py)2}(µ2-ZnCp*)2Co(CO)3] (12). To a suspension of [(CO)3Co(µ2-

GaCp*)2Co(CO)3] (0.150 g, 0.216 mmol) in toluene (7 ml) ZnMe2 (1.1 ml, 1.341 mmol, 1.2

M toluene solution) was added whereupon the suspension becomes colourless. The reaction



dried in vacuo to give a colourless solid which was dissolved in a mixture of toluene/pyridine

(4 ml/3 ml) and stored at -30°C to give yellow crystals. Yield: 0.083 g (52 %). Elemental

analysis (%) calculated for C36H40Co2N2O6Zn3: C, 47.48; H, 4.63; N, 3.08; Zn, 21.54. Found:

C, 48.91; H, 4.76; N, 3.20; Zn, 17.90. 1H NMR δH(THF-d8), 1.95 (s, 30H, C5Me5), 7.57 (s, br,

4H, C5H5N), 7.97 (s, br, 2H, C5H5N), 8.55 (s, br, 4H, C5H5N). 13C NMR δC{H}(THF-d8),

149.64 (C5H5N), 139.83 (C5H5N), 125.94 (C5H5N), 111.73 (C5Me5), 10.80 (C5Me5). IR

(ATR, cm-1): 1943 (m), 1886 (m), 1837 (w), 1806 (s), 1590 (w), 1560 (m), 1476 (w), 1435

(m), 1347 (m), 1207 (m), 1144 (m), 1061 (m), 1031 (m), 1003 (m), 976 (m), 943 (m), 874

(m), 750 (m), 743 (w), 689 (s), 646 (m), 626 (m), 585 (m), 567 (s), 554 (m), 535 (m), 494

(m), 468 (m), 429 (m), 412 (w).

[(CO)3Co{µ2-Zn(thf)2}(µ2-ZnCp*)2Co(CO)3] (13). To a suspension of [(CO)3Co(µ2-

GaCp*)2Co(CO)3] (0.150 g, 0.216 mmol) in toluene (7 ml) ZnMe2 (1.1 ml, 1.341 mmol, 1.2

M toluene solution) was added whereupon the suspension becomes colourless. The reaction


dried in vacuo to give a colourless solid which was dissolved in a mixture of toluene/thf (2

ml/2 ml) and stored at -30°C to give orange crystals. Yield: 0.129 g (67 %). Elemental

analysis (%) calculated for C34H46Co2O8Zn3 x C4H8O: C, 47.11; H, 5.62; Zn, 20.25. Found: C,

45.14; H, 5.71; Zn, 21.20. 1H NMR δH(THF-d8), 1.77 (m, br, 8H, C4H8O), 1.95 (s, 30H,

C5Me5), 3.61 (m, br, 8H, C4H8O). 13C NMR δC{H}(THF-d8), 111.78 (C5Me5), 68.03 (C4H8O),

26.19 (C4H8O), 10.76 (C5Me5). IR (ATR, cm-1): 2876 (w), 2833 (w), 1965 (m), 1897 (s), 1436

(w), 1365 (w), 1334 (w), 1167 (w), 1055 (w), 1006 (w), 908 (w), 849 (w), 831 (w), 786 (w),

670 (w), 569 (m), 530 (w), 484 (w), 433 (w), 418 (w), 407 (w).

[Pd2Zn6Ga2(Cp*)5(CH3)3] (14). To a solution of [Pd2(GaCp*)2(µ-GaCp*)3] (0.205 g 0.16

mmol) in toluene (6 ml) Zn(CH3)2 (0.08 mL of a 2 M solution in toluene, 0.16 mmol) was

added. The reaction mixture was heated at 100°C for 1 h, then the solvent was reduced in

vacuo and the residue extracted with hot n-hexane (3 x 5 ml). The solvent was again reduced

in vacuo, the residue dissolved in toluene (1 ml) and stored at -30°C to give deep red red

crystals. Yield: 80 mg (34%). Anal. Calcd. for C53H84Ga2Zn6Pd2: C, 43.43; H, 5.78; Ga, 9.51;

Zn, 26.77. Found: C, 42.40; H, 5.58; Ga, 15.90; Zn, 19.59. 1H NMR δH(C6D6, 14A), 0.05 (s,


3H, Me), 0.49 (s, 6H, Me), 1.90 (s, br, 15H, C5Me5), 1.94 (s, br, 15H, C5Me5), 2.10 (s, br,

30H, C5Me5), 2.21 (s, br, 15H, C5Me5). 1H NMR δH(C6D6, 14B), -0.06 (s, 3H, Me), 0.04 (s,

6H, Me), 1.93 (s, br, 15H, C5Me5), 2.15 (s, br, 30H, C5Me5), 2.17 (s, br, 15H, C5Me5), 2.20 (s,

br, 15H, C5Me5). 13C{1H} NMR δC{H}(C6D6), 10.02, 10.11, 10.78, 10.84, 10.93, 11.27, 11.79,

11.93, 12.03, 12.14, 14.84, 109.29, 110.15, 110.27, 110.78, 110.85, 113.35, 114.07, 114.45,

115.07. UV/vis (n-hexane): λmax 250 nm. MS (LIFDI, toluene): m/z 1465 [M].+.

[Pd(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (15) and [Pd(ZnCp*)4(ZnZnCp*)4] (16).

[Pd(GaCp*)4] (0.150 g, 0.161 mmol) and [Zn2Cp*2] (0.266 mg, 0.664 mmol) were dissolved

in a mixture of toluene (5 mL) and hexane (3 ml) and stirred for 2 h at 95°C, whereupon a

dark red solution was formed. The solution was cooled to room temperature and stored at -30

°C. Compounds 15 and 16 crystallized at -30 °C over night. The solution was filtered and the

remaining solvent was removed in vacuo. Yield: 134 mg (58%). 15: Elemental analysis (%)

calculated for C60H90Ga2Zn6Pd: C, 49.71; H, 6.26; Ga 9.6; Zn 27.1. Found: C, 49.41; H 5.93;

Ga 9.5; Zn 24.9.1H NMR δH(C6D6), 1.87 (s, 30H, C5Me5), 2.15 (s, 30H, C5Me5), 2.31 (s, 30H,

C6D6). 13C{1H} NMR δC{H}(C6D6), 113.91 (C5Me5), 109.66 (C5Me5), 108.81 (C5Me5), 11.98

(C5Me5), 10.72 (C5Me5), 10.17 (C5Me5). 16: 1H NMR δH(toluene-d8), 2.19 (s, 60H, C5Me5),

2.30 (s, 60H, C5Me5).

[Pt(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (17) and [Pt(ZnCp*)4(ZnZnCp*)4] (18). [Pt(GaCp*)4]

(0.150 g, 0.148 mmol) and [Zn2Cp*2] (0.177 g, 0.444 mmol) were dissolved in benzene (5

mL). The reaction mixture was stirred for 2 h at 75°C, whereupon a dark red solution was

formed. After two hours the solution was slowly cooled to room temperature without stirring.

At 50°C 17 and 18 began to crystallize as orange and dark red crystals. After cooling down,

the reaction mixture was stored at 7°C. The solution was filtered and the remaining solvent

removed in vacuo. Yield: 0.147 g (65%). 17: Elemental analysis (%) calculated for

C60H90Ga2Zn6Pt x C7H8: C, 49.36; H, 6.06; Ga 8.5; Zn 24.1. Found: C, 49.67; H 6.44; Ga 9.4;

Zn 17.8. 1H NMR δH(C6D6), 1.77 (s, 30H, C5Me5), 2.10 (s, 30H, C5Me5), 2.11 (s, 30H, C6D6). 13C{1H} NMR δC{H}(C6D6), 118.58 (C5Me5), 114.72 (C5Me5), 111.72 (C5Me5), 13.39

(C5Me5), 12.37 (C5Me5), 10.93 (C5Me5). The strong deviation in the Zn value can be

explained by the high reactivity and instability of compound 17. 1H NMR and 13C NMR

spectroscopy as well as elemental analysis could only be done for 18. Due to the fact, that


compound 18 is a rare by-product and the decomposition of the dried product, NMR

measurements for 18 were not successful.

[Ni(ZnCp*)4(ZnZnCp*)4] (19). [Ni(cod)2] (0.038 g, 0.138 mmol) and [Zn2Cp*2] (0.450 g,

1.123 mmol) were dissolved in toluene (6 ml). The reaction mixture was stirred for 30 h at

80°C. The hot solution was filtered, the solvent removed in vacuo up to ca. 3 ml and stored at

-30°C over night. The formed deep red crystals were isolated, washed with cold n-hexane (2 x

2 ml) and dried in vacuo. Yield: 0.151 g (57%). 1H NMR δH(C6D6), 2.19 (s, 60H, C5Me5),

2.30 (s, 60H, C5Me5). 13C NMR δC{H}(C6D6), 111.59 (C5Me5), 109.24 (C5Me5), 12.42

(C5Me5), 10.99 (C5Me5). IR (ATR, cm-1): 2938 (w), 2876 (s), 2832 (s), 2701 (w), 1467 (s),

1421 (s), 1309 (s), 1250 (s), 1012 (s), 790 (s), 720 (s), 687 (s), 548 (s). MS (LIFDI, toluene):

m/z 1927 [M].+.

[Cp*Ni(ZnCp*)3] (20). [Ni(cod)2] (0.150 g, 0.607 mmol) and [Zn2Cp*2] (0.460 g, 1.275

mmol) in toluene (5 ml) were stirred at 80°C for 3 h. The solvent was reduced in vacuo, the

residue washed with n-hexane (4 x 3 ml) and dried in vacuo to give an orange solid. Re-

crystallization of the crude product by cooling a saturated toluene solution (3 ml) of 20 to

-30°C gave orange single crystals. Yield: 0.304 g (63 %). 1H NMR δH(C6D6), 1.89 (s, 15H,

NiC5Me5), 2.16 (s, 45H, ZnC5Me5). 13C{1H} NMR δC{H}(C6D6), 111.68 (ZnC5Me5), 97.98

(NiC5Me5), 13.29 (NiC5Me5), 11.98 (ZnC5Me5). IR (ATR, cm-1): 2944 (w), 2876 (s), 2832

(s), 2700 (w), 1421 (s), 1364 (s), 1233 (w), 1014 (s), 789 (w), 722 (w), 585 (s). MS (LIFDI,

toluene): m/z 794 [M].+.

[Cp*Pt(ZnCp*)3] (21). [Pt(cod)2] (0.106 g, 0.258 mmol) and [Zn2Cp*2] (0.233 g, 0.541

mmol) in toluene (5 ml) were stirred at 100°C for 1.5 h. The solvent was reduced in vacuo,

the residue washed with n-hexane (5 ml) and dried in vacuo to give a yellow solid. Re-

crystallization of the crude product by cooling a saturated toluene solution (3 ml) of 21 to

-30°C gave yellow single crystals. Yield: 0.115 g (48 %). 1H NMR δH(C6D6), 2.13 (s, 15H,

PtC5Me5), 2.17 (s, 45H, ZnC5Me5). 13C{1H} NMR δC{H}(C6D6), 110.46 (ZnC5Me5), 102.04

(PtC5Me5), 13.08 (PtC5Me5), 11.69 (ZnC5Me5). IR (ATR, cm-1): 2941 (w), 2879 (m), 2833


(m), 1425 (w), 1364 (m), 1223 (s), 1115 (m), 975 (s). MS (LIFDI, toluene): m/z 932 [M].+,

797 [M-Cp*]+.

[Cp*Pd(ZnCp*)3] (22). [PdMe2(tmeda)] (0.060 g, 0.237 mmol) and [Zn2Cp*2] (0.390 mg,

0.974 mmol) were dissolved in toluene (6 ml) resulting in a deep red solution. The mixture

was stirred for 1 h at 55°C. The solvent was reduced in vacuo and the yellow-orange residue

washed with n-hexane (3 x 3 ml). The first 3 ml of the filtrate were refused to remove

[Pd(ZnCp*)4(ZnMe)4]. The following 6 ml of the filtrate were collected, the solvent reduced

in vacuo and the orange powder dried in vacuo yielding nearly pure 22. Re-crystallization of

the crude product in a mixture of toluene/n-hexane (3 ml/2 ml) at -30°C over night gave

orange single crystals. Notably, small traces of [Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}] can be

easily removed via crystal separation in the glove box under inert gas atmosphere. Yield:

0.115 g (48 %). 1H NMR δH(CD2Cl2), 2.08 (s, 45H, ZnC5Me5), 2.17 (s, 15H, PdC5Me5). 13C{1H} NMR δC{H}(CD2Cl2), 110.86 (ZnC5Me5), 106.21 (PdC5Me5), 13.37 (PdC5Me5), 11.51

(ZnC5Me5). IR (ATR, cm-1): 2941 (w), 2879 (m), 2833 (m), 1425 (w), 1364 (m), 1223 (s),

1115 (m), 975 (s). MS (LIFDI, toluene): m/z 842 [M].+, 797 [M-Cp*]+.

[Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}] (23). [PdMe2(tmeda)] (0.060 g, 0.237 mmol) and

[Zn2Cp*2] (0.390 mg, 0.974 mmol) were dissolved in toluene (6 ml) resulting in a deep red

clear solution. The mixture was stirred for 1 h at 55°C, the solvent reduced in vacuo, the

yellow-orange residue washed with n-hexane (3 x 3 ml) and dried in vacuo to give a yellow

solid. Re-crystallization of the crude product in a mixture of toluene/n-hexane (3 ml/2 ml) at

-30°C over night gave yellow single crystals. Yield: 105 mg (35%). Anal. Calcd. for

C48H82N2Zn7Pd1: C, 46.07; H, 6.61. Found: C, 46.22; H, 6.96. 1H NMR δH(C6D6), 0.14 (s, 6H,

ZnMe), 1.62 (s, 4H, (CH3)2N-CH2-CH2-N(CH3)2), 1.73 (s, 12H, (CH3)2N-CH2-CH2-N(CH3)2),

2.10 (s, br, 30H, C5Me5), 2.27 (s, br, 30H, C5Me5). 1H NMR δH(C6D6, 50°C), 0.10 (s, 6H,

ZnMe), 1.71 (s, 4H, (CH3)2N-CH2-CH2-N(CH3)2), 1.79 (s, 12H, (CH3)2N-CH2-CH2-N(CH3)2),

2.17 (s, br, 60H, C5Me5). 1H NMR δH(C6D6, 70°C), 0.07 (s, 6H, ZnMe), 1.76 (s, 4H, (CH3)2N-

CH2-CH2-N(CH3)2), 1.83 (s, 12H, (CH3)2N-CH2-CH2-N(CH3)2), 2.17 (s, 60H, C5Me5). 1H

NMR δH(CD2Cl2), -0.41 (s, 6H, ZnMe), 2.01 (s, br, 60H, C5Me5), 2.28 (s, 12H, (CH3)2N-CH2-

CH2-N(CH3)2), 2.67 (s, 4H, (CH3)2N-CH2-CH2-N(CH3)2). 13C{1H} NMR δC{H}( C6D6), 11.91

(ZnC5Me5), 12.18 (ZnC5Me5), 14.22 (ZnMe), 46.23 ((CH3)2N-CH2-CH2-N(CH3)2)), 55.08


((CH3)2N-CH2-CH2-N(CH3)2), 109.91 (ZnC5Me5), 111.10 (ZnC5Me5). IR (ATR, cm-1): 2867

(s), 2829 (s), 2698 (w), 1447 (m), 1427 (w), 1408 (w), 1363 (m), 1277 (w), 1250 (w), 1178

(w), 1152 (w), 1130 (w), 1115 (w), 1088 (w), 1038 (w), 1019 (w), 1001 (m), 941 (w), 786 (s),

757 (w), 722 (m), 688 (w), 633 (w), 583 (w), 523 (s), 476 (w), 460 (w), 429 (w), 403 (w). MS

(LIFDI, toluene): m/z 1252 [M].+.

6.3 Computational Details

All quantum chemical calculations have been performed in the group of Prof. Dr. Gernot

Frenking, Philipps University Marburg.[297] Thus, chapter 6.3 just gives a short overview of

the computational details. For further informations and suggestions please contact Prof. Dr.

Gernot Frenking: [email protected]

6.3.1 Computational Details for [Pd2Zn6Ga2(Cp*)5(CH3)3] (14)

To find the energetically most stable isomer of 14 a model system was used, where Cp* is

replaced by Cp. The energies of every permutation 14M-X/Y of the Gallium positions which

possess Cs symmetry were calculated. The geometries were optimized at BP86/def2-

TZVPP[298-300] with the Gaussian 03, Revision E.01[301] algorithm using energies calculated

with the Turbomole 6.3[302] program package. The RI approximation[303] was applied using

auxiliary basis functions.[304-306] Frequency calculations were used to obtain vibrational

energy contributions. The NBO[307, 308] charges were obtained using the NBO 3.1 program

implemented in Gaussian 03. The AIM[309] analyses were carried out using a modified version

of AIMPAC[310] under the use of BP86/def2-SVP wavefunctions. For the orbital visualization,

the Cp ligands of the energetically lowest lying 14M-3/8 were replaced with hydrogen. The

geometry of the metal cluster was frozen and the hydrogen bonds were set to standard values

to obtain the model system 14M-3/8-H. A BP86/def2-TZVPP Turbomole singlepoint

calculation was performed to generate the orbital plots. The visualization program was

gOpenMol.[311]


Table ES1. Relative energies ΔErel [in kcal/mol] of the calculated isomers of [Pd2Zn6Ga2(Cp)5Me3]. Energies at

BP86/def2-TZVPP. ΔErelvib include vibrational energy contributions.

[Pd2Zn6Ga2(Cp)5Me3] ΔErel ΔErelvib

14M-1/2 (not Cs symmetrical) 25.8 25.4

14M-3/4 (Cs symmetrical but heavily distorted from x-ray structure) 10.2 10.2


14M-3/8 0.0 0.0

14M-4/5 (Cs symmetrical but heavily distorted from x-ray structure) 19.7 19.3

14M-4/8 4.2 4.0

14M-5/8 5.3 5.1


Figure ES1. At BP86/def2-TZVPP optimized structure of isomer 14M-4/8 of [Pd2Zn6Ga2(Cp)5Me3].

Experimental values of the X-ray structure of [Pd2Zn6Ga2(Cp*)5Me3] are given in parentheses.


Figure ES2. At BP86/def2-TZVPP optimized structure of isomer 14M-5/8 of [Pd2Zn6Ga2(Cp)5Me3].

Experimental values of the X-ray structure of [Pd2Zn6Ga2(Cp*)5Me3] are given in parentheses.


Figure ES3. Orbital shapes and eigenvalues in eV of 14M-3/8-H (BP86/def2-TZVPP).

-5.370 (143 a) -5.535 (142 a) -5.923 (141 a)

-6.872 (140 a) -7.096 (139 a) -7.210 (138 a)

-7.257 (137 a) -7.308 (136 a) -7.475 (135 a)

-7.633 (134 a) -7.702 (133 a) -7.784 (132 a)


-8.187 (131 a) -8.194 (130 a) -8.746 (129 a)

-8.758 (128 a) -8.916 (127 a) -8.960 (126 a)

-9.208 (125 a) -9.323 (124 a) -10.144 (123 a)

-11.200 (122 a) -11.538 (121 a)


6.3.2 Computational Details for [M(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (M = Pd (15), Pt

(17))

The single point calculations on all possible isomers of 15 have been performed at BP86/def2-

TZVPP. The crystal structure of 15 has been used where the methyl groups of the Cp*

moieties were replaced by hydrogen atoms (yielding model system 15H; see below for

resulting Cartesian coordinates of this model system) and permuted six Zn atoms and two Ga

atoms over all 8 possible Zn/Ga positions. Table ES2 provides the relative energies of all 28

isomers. The lowest energy isomer has Ga atoms at the positions 1 and 2 (isomer 1_2), being

the trans positions to the M-M-Cp ligands. Single point calculations on selected isomers of

the full system 15 (without replacement of Cp* by Cp; see below for Cartesian coordinates of

this system) with different density functionals were performed to further substantiate these

results. Eight isomers are higher in energy than isomer 1_2 by less then 15 kcal/mol, namely

1_3, 1_4, 1_6, 1_8, 2_3, 2_4, 2_6 and 2_8. As 15 is almost Cs symmetric, these eight isomers

can be collected as pairs of pseudo mirror images. These are 1_3 and 2_3, 1_8 and 2_8, 1_4

and 2_6, 1_6 and 2_4. Thus the energy of only one member of each pair with that of 1_2 was

compared. Results are given in table ES3. In all cases, 1_2 is the lowest energy isomer.

Using the same procedure, the assignment of the lowest energy isomer of the Pt-compound 17

has been done. It turns out, that it is the analogue isomer of 15, were the Ga atoms are in trans

position to the M-M-Cp* ligands.

Table ES2. Relative energies of all 28 isomers of the model system 15H at BP86/def2-TZVPP. Isomers are

denoted by the Ga-positions, i.e. isomer 1_2 has Ga atoms at positions 1 and 2. Relative energies ΔE with

respect to the lowest energy isomer in kcal/mol.

Isomer of 15H

ΔE Isomer of 15H

ΔE

1_2 0.00 3_5 33.50

1_3 9.93 3_6 22.65

1_4 11.98 3_7 32.63

1_5 22.83 3_8 24.55

1_6 10.37 4_5 45.33


1_7 25.12 4_6 20.65

1_8 9.93 4_7 31.61

2_3 11.11 4_8 23.19

2_4 13.39 5_6 31.35

2_5 27.48 5_7 45.11

2_6 12.59 5_8 33.50

2_7 24.31 6_7 43.47

2_8 11.11 6_8 22.65

3_4 23.19 7_8 32.63

Table ES3. Relative energies of the seven lowest energy isomers of the compound 15 at different levels of

theory. Isomers are denoted by the Ga-positions, i.e. isomer 1_2 has Ga atoms at positions 1 and 2. Relative

energies ΔE with respect to the lowest energy isomer in kcal/mol.


BP86/def2-TZVPP

ΔE [kcal/mol]

B3LYP/def2-TZVPP

ΔE [kcal/mol]

M05/def2-TZVPP

ΔE [kcal/mol]

M05-2X/def2-TZVPP

1_2 0.0 0.0 0.0 0.0

1_3 9.9 10.7 10.9 11.1

1_4 15.7 16.1 17.3 14.0

1_6 13.7 14.0 14.3 10.9

1_8 9.9 10.7 10.9 11.1

Table ES4. Relative energies of all 28 isomers of the model system 17H at BP86/def2-TZVPP. Isomers are

denoted by the Ga-positions, i.e. isomer 1_2 has Ga atoms at positions 1 and 2. Relative energies ΔE with

respect to the lowest energy isomer in kcal/mol.

Isomer of 17H

ΔE Isomer of 17H

ΔE

1_2 0.00 3_5 31.58

1_3 8.81 3_6 32.10

1_4 12.08 3_7 21.00

1_5 22.11 3_8 24.86


1_6 11.62 4_5 46.35

1_7 27.06 4_6 32.87

1_8 8.89 4_7 21.56

2_3 9.74 4_8 21.87

2_4 14.30 5_6 31.41

2_5 28.70 5_7 45.55

2_6 12.04 5_8 31.62

2_7 23.60 6_7 46.59

2_8 9.83 6_8 21.07

3_4 21.82 7_8 32.18

Table ES5. Relative energies the seven lowest energy isomers of the compound 17 at different levels of theory.

Isomers are denoted by the Ga-positions, i.e. isomer 1_2 has Ga atoms at positions 1 and 2. Relative energies ΔE

with respect to the lowest energy isomer in kcal/mol.


BP86/def2-TZVPP

ΔE [kcal/mol]

B3LYP/def2-TZVPP

ΔE [kcal/mol]

M05/def2-TZVPP

ΔE [kcal/mol]

M05-2X/def2-TZVPP

1_2 0.00 0.00 0.00 0.00

1_3 8.28 8.69 8.52 9.16

1_4 16.03 16.16 17.09 14.71

1_6 15.13 14.94 14.91 12.05

1_8 8.20 8.60 8.42 9.07

Cartesian Coordinates of Calculated Compounds

Cartesian coordinates of the model system 15H. Numbers of XX nuclei indicate the positions.

Pd 10.665608 4.037050 4.414130

XX1 10.092402 4.037050 6.696694

XX2 13.019700 4.037050 4.417040

XX3 10.081271 1.819741 3.570116


XX4 8.327211 4.037050 3.999396

XX5 5.989273 4.037050 3.975977

XX6 11.020452 4.037050 2.066832

XX7 11.755676 4.037050 -0.153616

XX8 10.081271 6.254359 3.570116

C 8.575213 4.037050 8.363049

C 9.548249 5.148046 8.589982

C 10.720886 4.569941 8.862010

H 7.589935 4.037050 8.192088

H 9.576191 6.146390 8.640268

H 11.484809 5.194730 9.023444

C 15.001960 4.037050 5.479841

C 14.974845 2.887298 4.637572

C 14.923953 3.323300 3.313798

H 15.087934 4.037050 6.476139

H 14.994866 1.941338 4.961238

H 14.911457 2.736930 2.503851

C 8.847539 -0.058134 4.055693

C 10.219390 -0.326194 4.244804

C 10.852964 -0.248682 2.992310

C 9.863975 0.093660 2.038029

C 8.623384 0.208312 2.723191

H 8.142422 -0.090056 4.764066

H 10.674241 -0.590923 5.095116

H 11.803753 -0.465391 2.770867

H 10.016793 0.173087 1.052972

H 7.727727 0.390087 2.317289

C 4.097334 3.324914 5.073982

C 4.037341 2.893757 3.715295

C 4.000997 4.037050 2.894848


H 4.104117 2.733743 5.880500

H 3.965163 1.943219 3.413191

H 3.895900 4.037050 1.900386

C 11.382784 4.752415 -2.283873

C 12.676086 5.183572 -1.908562

C 13.487231 4.037050 -1.669991

H 10.607930 5.331299 -2.537830

H 12.998310 6.129001 -1.860226

H 14.474139 4.037050 -1.508702

C 9.548249 2.926054 8.589982

C 10.720886 3.504159 8.862010

H 9.576191 1.927710 8.640268

H 11.484809 2.879370 9.023444

C 14.974845 5.186802 4.637572

C 14.923953 4.750800 3.313798

H 14.994866 6.132762 4.961238

H 14.911457 5.337170 2.503851

C 4.037341 5.180343 3.715295

H 3.965163 6.130881 3.413191

C 4.097334 4.749186 5.073982

H 4.104117 5.340357 5.880500

C 11.382784 3.321685 -2.283873

C 12.676086 2.890528 -1.908562

H 10.607930 2.742801 -2.537830

H 12.998310 1.945099 -1.860226

C 8.847539 8.132234 4.055693

C 10.219390 8.400294 4.244804

C 10.852964 8.322782 2.992310

C 9.863975 7.980440 2.038029

C 8.623384 7.865788 2.723191


H 8.142422 8.164156 4.764066

H 10.674240 8.665023 5.095116

H 11.803753 8.539491 2.770867

H 10.016792 7.901013 1.052972

H 7.727727 7.684013 2.317289

Cartesian coordinates of compound 15. Numbers of XX nuclei indicate the positions.

Pd 10.665608 4.037050 4.414130

XX1 10.092402 4.037050 6.696694

XX2 13.019700 4.037049 4.417040

XX3 10.081269 1.819741 3.570116

XX4 8.327210 4.037051 3.999396

XX5 5.989272 4.037053 3.975976

XX6 11.020452 4.037050 2.066832

XX7 11.755676 4.037050 -0.153616

XX8 10.081271 6.254359 3.570116

C 8.623384 7.865788 2.723191

C 8.847539 8.132234 4.055693

C 10.219390 8.400294 4.244804

C 10.852964 8.322782 2.992310

C 9.863975 7.980440 2.038029

C 12.676086 2.890528 -1.908562

C 13.487231 4.037050 -1.669991

C 12.676087 5.183572 -1.908562

C 11.382785 4.752415 -2.283873

C 11.382784 3.321685 -2.283873

C 10.219907 5.621189 -2.665003

C 13.160961 6.606229 -1.835827

C 14.955908 4.037050 -1.429966

C 10.219906 2.452913 -2.665004


C 13.160960 1.467871 -1.835827

C 7.777459 8.180678 5.130714

C 10.896380 8.794310 5.510390

C 12.277021 8.647361 2.660640

C 10.090778 7.862559 0.576060

C 7.310534 7.599343 2.128221

C 4.097334 4.749190 5.073982

C 4.097333 3.324918 5.073982

C 4.037340 2.893761 3.715295

C 4.000997 4.037054 2.894847

C 4.037341 5.180347 3.715295

C 4.107597 2.430307 6.294472

C 3.928208 1.456572 3.258521

C 3.843572 4.037054 1.405238

C 3.928211 6.617536 3.258521

C 4.107598 5.643799 6.294472

C 10.720885 3.504158 8.862010

C 10.720886 4.569940 8.862011

C 9.548250 5.148046 8.589983

C 8.575213 4.037051 8.363050

C 9.548248 2.926054 8.589982

C 7.158377 4.037052 8.117206

C 9.590284 6.649829 8.665627

C 11.994388 5.611497 9.131130

C 9.590280 1.424271 8.665626

C 11.994386 2.462599 9.131129

C 14.923953 4.750798 3.313798

C 14.923952 3.323298 3.313798

C 14.974844 2.887295 4.637572

C 15.001960 4.037047 5.479841


C 14.974845 5.186799 4.637572

C 15.133014 4.037047 6.998544

C 15.004988 1.463024 5.124895

C 14.905369 2.451295 2.109310

C 15.004991 6.611070 5.124896

C 14.905371 5.622801 2.109310

C 8.623381 0.208313 2.723191

C 8.847536 -0.058133 4.055693

C 10.219387 -0.326194 4.244804

C 10.852961 -0.248682 2.992310

C 9.863972 0.093660 2.038029

C 7.777456 -0.106577 5.130714

C 10.896376 -0.720211 5.510390

C 12.277017 -0.573262 2.660640

C 10.090774 0.211541 0.576060

C 7.310531 0.474759 2.128221

H 9.282932 6.976671 9.669972

H 8.914729 7.077367 7.910090

H 10.617417 6.992716 8.472166

H 12.108242 5.778310 10.212430

H 11.794627 6.570779 8.631256

H 12.918767 5.172573 8.727559

H 16.046054 1.109310 5.157576

H 14.422628 0.830299 4.438953

H 14.566673 1.412264 6.132517

H 15.937329 2.257531 1.781418

H 14.353624 2.953437 1.300959

H 14.413315 1.498424 2.354096

H 7.332299 -1.112094 5.158460

H 6.997107 0.635193 4.905252


H 8.229560 0.120088 6.107558

H 10.850758 -1.813304 5.624699

H 10.389419 -0.241614 6.361236

H 11.947459 -0.397314 5.479340

H 12.361555 -1.641522 2.412295

H 12.916259 -0.344838 3.526198

H 12.599258 0.029931 1.799061

H 9.915234 -0.764000 0.099085

H 11.127227 0.530499 0.391580

H 9.396764 0.954683 0.156429

H 6.857011 -0.472752 1.801812

H 7.429528 1.141551 1.261487

H 6.659662 0.953450 2.874695

H 3.074343 2.252753 6.627428

H 4.581288 1.470283 6.041550

H 4.673968 2.916070 7.102715

H 2.869191 1.160912 3.225840

H 4.469551 0.806478 3.961603

H 4.367258 1.355141 2.255053

H 10.254700 5.827623 -3.744899

H 10.273709 6.569258 -2.109782

H 9.280109 5.104659 -2.420101

H 13.548033 6.912569 -2.818848

H 13.962150 6.684191 -1.086148

H 12.326496 7.263618 -1.550329

H 9.282928 1.097429 9.669971

H 8.914724 0.996734 7.910088

H 10.617412 1.081383 8.472165

H 12.108239 2.295786 10.212429

H 11.794623 1.503318 8.631255


H 12.918765 2.901523 8.727558

H 16.046057 6.964784 5.157577

H 14.422632 7.243796 4.438955

H 14.566676 6.661830 6.132518

H 15.937331 5.816563 1.781418

H 14.353626 5.120660 1.300959

H 14.413318 6.575672 2.354096

H 2.869194 6.913196 3.225841

H 4.469554 7.267629 3.961604

H 4.367261 6.718966 2.255054

H 3.074345 5.821355 6.627429

H 4.581291 6.603823 6.041550

H 4.673970 5.158036 7.102716

H 10.254698 2.246479 -3.744899

H 10.273707 1.504844 -2.109783

H 9.280108 2.969444 -2.420102

H 13.548032 1.161531 -2.818848

H 13.962149 1.389909 -1.086148

H 12.326496 0.810482 -1.550329

H 7.332303 9.186196 5.158460

H 6.997110 7.438909 4.905252

H 8.229563 7.954013 6.107558

H 10.850762 9.887403 5.624700

H 10.389422 8.315713 6.361236

H 11.947462 8.471413 5.479340

H 12.361559 9.715620 2.412296

H 12.916263 8.418937 3.526199

H 12.599261 8.044167 1.799061

H 9.915238 8.838100 0.099085

H 11.127230 7.543601 0.391581


H 9.396767 7.119418 0.156428

H 6.857015 8.546854 1.801812

H 7.429530 6.932550 1.261487

H 6.659664 7.120652 2.874695

H 16.196983 3.996128 7.274760

H 14.683620 4.954953 7.405353

H 14.613009 3.160059 7.411446

H 15.485046 4.037050 -2.394337

H 15.234138 4.935196 -0.859070

H 15.234138 3.138904 -0.859070

H 2.773690 4.037055 1.149597

H 4.320710 4.935200 0.986104

H 4.320709 3.138908 0.986104

H 6.619806 4.037052 9.076342

H 6.885760 4.935198 7.543609

H 6.885759 3.138906 7.543609

Cartesian coordinates of the model system 17H. Numbers of XX nuclei indicate the positions.

Pt 10.679964 12.436425 4.513688

XX1 9.996244 12.436425 6.758524

XX2 13.024653 12.436425 4.787179

XX3 10.341452 10.103020 3.990286

XX4 8.374628 12.436425 3.837021

XX5 6.054155 12.436425 3.411720

XX6 11.146233 12.436425 2.177009

XX7 11.844572 12.436425 -0.051380

XX8 10.341452 14.769830 3.990286

C 8.752543 8.465060 3.554360

C 9.502910 8.685599 2.304075

C 10.821259 8.347328 2.536957


C 10.923041 7.899617 3.810530

C 9.754356 7.969261 4.465511

H 7.768138 8.551296 3.707691

H 9.098127 9.042614 1.462237

H 11.539041 8.363872 1.840885

H 11.791598 7.568998 4.179718

H 9.542453 7.625149 5.380215

C 11.458761 13.144472 -2.162891

C 12.765835 13.588867 -1.807747

C 13.552282 12.436425 -1.577776

H 10.688069 13.734534 -2.403432

H 13.093078 14.533073 -1.770508

H 14.535771 12.436425 -1.396804

C 8.486127 13.111308 8.354638

C 9.760950 13.563994 8.734527

C 10.590130 12.436425 8.960131

H 7.710456 13.701342 8.130596

H 10.116949 14.490862 8.853611

H 11.554180 12.436425 9.225850

C 14.969546 13.134523 3.657701

C 15.009293 13.587209 5.006961

C 15.085545 12.436425 5.763827

H 14.928482 13.743566 2.865627

H 15.044991 14.528324 5.343158

H 15.155499 12.436425 6.761377

C 11.458761 11.728378 -2.162891

C 12.765835 11.283983 -1.807747

H 10.688069 11.138316 -2.403432

H 13.093078 10.339777 -1.770508

C 8.486127 11.761542 8.354638


C 9.760950 11.308856 8.734527

H 7.710456 11.171508 8.130596

H 10.116949 10.381988 8.853611

C 14.969546 11.738327 3.657701

C 15.009293 11.285641 5.006961

H 14.928483 11.129284 2.865628

H 15.044991 10.344526 5.343158

C 4.065982 12.050067 4.606696

C 4.028140 11.225946 3.469941

C 4.045561 12.051725 2.333188

C 4.093151 13.386568 2.768384

C 4.106918 13.386568 4.172953

H 4.060318 11.740126 5.557435

H 3.994177 10.226524 3.468966

H 4.024080 11.744146 1.381908

H 4.111719 14.196095 2.181596

H 4.137924 14.194614 4.761256

C 8.752543 16.407790 3.554360

C 9.502910 16.187251 2.304075

C 10.821259 16.525522 2.536957

C 10.923041 16.973233 3.810530

C 9.754356 16.903589 4.465511

H 7.768138 16.321554 3.707691

H 9.098126 15.830236 1.462238

H 11.539040 16.508978 1.840885

H 11.791598 17.303852 4.179719

H 9.542452 17.247701 5.380215


Cartesian coordinates of compound 17. Numbers of XX nuclei indicate the positions.

Pt 10.679964 12.436425 4.513688

XX1 9.996244 12.436425 6.758524

XX2 13.024653 12.436425 4.787179

XX3 10.341452 10.103020 3.990286

XX4 8.374628 12.436425 3.837021

XX5 6.054155 12.436425 3.411720

XX6 11.146233 12.436425 2.177009

XX7 11.844573 12.436425 -0.051380

XX8 10.341452 14.769830 3.990286

C 8.752543 8.465060 3.554360

C 9.502910 8.685599 2.304075

C 10.821259 8.347328 2.536957

C 10.923041 7.899617 3.810530

C 9.754356 7.969261 4.465511

C 7.276099 8.594399 3.784331

H 6.850210 8.965833 2.986710

H 7.115824 9.191347 4.548475

H 6.896303 7.712242 3.979370

C 8.897533 9.219536 1.045058

H 7.941300 9.363799 1.176054

H 9.039879 8.576159 0.318757

H 9.328521 10.070188 0.813631

C 11.900411 8.372201 1.490444

H 11.541152 8.750269 0.660802

H 12.213949 7.458539 1.324516

H 12.646974 8.922720 1.804835

C 12.299592 7.375629 4.395646

H 12.990289 7.430349 3.702822

H 12.198741 6.443726 4.683837


H 12.560691 7.927806 5.162701

C 9.429643 7.441957 5.867169

H 8.492070 7.635964 6.079673

H 10.010295 7.878060 6.523604

H 9.574628 6.471915 5.893380

C 11.458762 13.144472 -2.162891

C 12.765836 13.588867 -1.807747

C 13.552283 12.436425 -1.577776

C 10.302226 14.029946 -2.523858

H 10.570840 14.968481 -2.440893

H 9.551625 13.849203 -1.919820

H 10.027597 13.849203 -3.446652

C 13.264671 15.028176 -1.750982

H 12.524470 15.638390 -1.944565

H 13.973645 15.154198 -2.416151

H 13.618460 15.215551 -0.855842

C 15.086810 12.436425 -1.295407

C 8.486127 13.111308 8.354638

C 9.760950 13.563994 8.734527

C 10.590130 12.436425 8.960131

C 7.322064 13.996782 8.018415

H 7.586129 14.935317 8.102835

H 6.577244 13.812723 8.634097

H 7.028157 13.821014 7.098532

C 10.304855 14.980088 8.916466

H 9.594857 15.630099 8.714150

H 11.057284 15.121035 8.309518

H 10.598017 15.101136 9.842172

C 12.037048 12.436425 9.358942

C 14.969546 13.134523 3.657701


C 15.009293 13.587209 5.006961

C 15.085545 12.436425 5.763827

C 14.908726 14.036578 2.484559

H 14.932538 14.968481 2.787305

H 14.078898 13.874075 1.991140

H 15.675814 13.862468 1.899443

C 15.063385 15.013252 5.516390

H 14.982398 15.633415 4.762435

H 15.916588 15.164148 5.974876

H 14.325802 15.164148 6.143716

C 15.198841 12.436425 7.379445

H 15.248015 11.512813 7.711302

H 14.411803 12.875845 7.768067

H 16.006590 12.920616 7.658904

C 4.065982 12.822783 4.606696

C 4.028140 13.646904 3.469941

C 4.045561 12.821125 2.333187

C 4.093151 11.486282 2.768384

C 4.106918 11.486282 4.172953

C 4.057527 13.285418 6.025820

H 4.094633 12.507727 6.622580

H 3.246955 13.799457 6.200482

H 4.847093 13.852519 6.185927

C 3.977426 15.139275 3.468486

H 3.960395 15.465938 2.544236

H 4.771560 15.492469 3.922606

H 3.172877 15.439407 3.937161

C 4.013483 13.280444 0.912607

H 3.984839 14.260434 0.886407

H 3.225503 12.915642 0.467220


H 4.824186 12.968704 0.452665

C 4.120877 10.277462 1.892165

H 4.157564 9.471581 2.446716

H 4.920111 10.310625 1.317238

H 3.321428 10.257563 1.331794

C 4.153175 10.280778 5.050626

H 4.153616 10.559354 5.990887

H 4.969785 9.771714 4.862865

H 3.371288 9.718652 4.875965

C 11.458762 11.728378 -2.162891

C 12.765836 11.283983 -1.807747

C 10.302226 10.842904 -2.523858

H 10.570840 9.904369 -2.440893

H 9.551625 11.023647 -1.919820

H 10.027597 11.023647 -3.446652

C 13.264671 9.844674 -1.750982

H 12.524470 9.234460 -1.944565

H 13.973645 9.718652 -2.416151

H 13.618460 9.657299 -0.855842

H 15.394212 13.358379 -1.174599

H 15.270540 11.915753 -0.491964

H 15.554313 12.035143 -2.062461

C 8.486127 11.761542 8.354638

C 9.760950 11.308856 8.734527

C 7.322064 10.876068 8.018415

H 7.586129 9.937533 8.102835

H 6.577244 11.060127 8.634097

H 7.028157 11.051836 7.098532

C 10.304855 9.892762 8.916466

H 9.594857 9.242751 8.714150


H 11.057284 9.751815 8.309518

H 10.598017 9.771714 9.842172

H 12.353095 11.512813 9.446272

H 12.566319 12.892427 8.666118

H 12.146283 12.904035 10.210416

C 14.969546 11.738327 3.657701

C 15.009293 11.285641 5.006961

C 14.908727 10.836272 2.484560

H 14.932539 9.904369 2.787306

H 14.078899 10.998774 1.991140

H 15.675815 11.010382 1.899444

C 15.063385 9.859598 5.516390

H 14.982398 9.239435 4.762435

H 15.916588 9.708702 5.974876

H 14.325802 9.708702 6.143716

C 8.752543 16.407790 3.554360

C 9.502910 16.187251 2.304075

C 10.821259 16.525522 2.536957

C 10.923041 16.973233 3.810530

C 9.754356 16.903589 4.465511

C 7.276099 16.278451 3.784331

H 6.850210 15.907017 2.986710

H 7.115824 15.681503 4.548475

H 6.896303 17.160608 3.979370

C 8.897532 15.653314 1.045058

H 7.941299 15.509051 1.176054

H 9.039878 16.296691 0.318757

H 9.328520 14.802662 0.813631

C 11.900411 16.500649 1.490444

H 11.541152 16.122581 0.660802


H 12.213949 17.414311 1.324516

H 12.646974 15.950130 1.804835

C 12.299592 17.497221 4.395646

H 12.990289 17.442501 3.702822

H 12.198741 18.429124 4.683837

H 12.560691 16.945044 5.162701

C 9.429643 17.430893 5.867169

H 8.492070 17.236885 6.079673

H 10.010295 16.994789 6.523604

H 9.574627 18.400934 5.893380

6.3.3 Computational Details for [M(ZnCp*)4(ZnZnCp*)4] (M = Pd (16), Pt (18), Ni

(19))

Table ES6. Geometry optimization at different levels of theory.

Compound Method d(M-Zn) d(M-ZnZn) α(M-Zn-Zn) ΔE

[Pd(ZnCp*)4(ZnZnCp*)4] X-Ray 2.422-2.433 2.473-2.478

[Pd(ZnCp)4(ZnZnCp)4] B3LYP/TZVPP 2.527-2.528 2.494-2.495 176.0-176.1

[Pd(ZnCp)4(ZnZnCp)4] B3LYP-D/TZVPP ~168

[Pd(ZnCp)4(ZnZnCp)4] PBE/TZVPP 2.501-2.502 2.468-2.469 177.8-177.9

[Pd(ZnCp)4(ZnZnCp)4] PBE-D/TZVPP 2.483-2.485 2.459-2.460 158.9-160.9

[Pd(ZnCp)4(ZnZnCp)4] PBE0/TZVPP 2.467 2.439-2.440 173.8-174.1

[Pd(ZnCp)4(ZnZnCp)4] TPSS/TZVPP 2.482-2.483 2.449-2.450 178.2-178.5

[Pd(ZnCp)4(ZnZnCp)4] TPSS-D/TZVPP ~158

[Pd(ZnCp)4(ZnZnCp)4] TPSSh/TZVPP 2.477 2.451 177.1-177.2

[Pd(ZnCp)4(ZnZnCp)4] BP86/TZVPP 2.507-2.508 2.467-2.468 179.5-179.9

[Pd(ZnCp)4(ZnZnCp)4] BP86-D/TZVPP ~155

[Pd(ZnCp*)4(ZnZnCp*)4] BP86/SVP 2.474-2.275 2.477-2.479 177.3-179.4


[Pd(ZnCp*)4(ZnZnCp*)4] BP86-D/SVP 2.398-2.412 2.440-2.453 169.5-176.1

[Pd(ZnCp*)4(ZnZnCp*)4] BP86/I[a] 2.502 2.483-2.484 179.1-179.7

[Pd(ZnCp*)4(ZnZnCp*)4] BP86-D/I[a] 2.413-2.445 2.422-2.471 168.2-175.1

[Pd(ZnCp)4(ZnZnCp)4] BP86/TZVPP 2.426 2.483-2.484 177.6-177.7 +2.1

[Pd(ZnCp)4(ZnZnCp)4] BP86/TZVPP 2.506-2.507 2.476 179.2-179.4 +0.0

[a] Basis set I: def2-TZVPP for Pd and Zn; def2-SVP for C and H

Green entries were held fixed in optimization

Figure ES4. Selection of occupied valence molecular orbitals of 16M.

6.3.4 Computational Details for [Pd(ZnCp*)2(ZnMe)2{Zn(tmeda)}] (23)

The geometry of the model compound 23M where Cp* is replaced by Cp was optimized at

BP86/def2-TZVPP[298-300] with the Gaussian 03, Revision E.01[301] algorithm using energies


calculated with the Turbomole 6.3[302] program package. The RI approximation[303] was

applied using auxiliary basis functions.[304-306]

Energy-decomposition analyses (EDA) were carried out at BP86/TZ2P using the

ADF(2009.01) program package.[312] Uncontracted Slater-type orbitals (STOs) were

employed as basis functions in self-consistent field (SCF) calculations.[313] Triple-zeta-quality

basis sets were used which were augmented by two sets of polarization functions, that is, p

and d functions for the hydrogen atom and d and f functions for the other atoms. An auxiliary

set of s, p, d, f and g STOs was used to fit the molecular densities and to represent the

Coulomb and exchange potentials accurately in each SCF cycle.[314] Scalar relativistic effects

were considered using the zero-order regular approximation (ZORA).[315-319]

Within the EDA, bond formation between the interacting fragments is divided into three

steps: In the first step, the fragments which are calculated with the frozen geometry which

they possess in the entire molecule, are superimposed without electronic relaxation to yield

the quasiclassical electrostatic attraction ΔEelstat. In the second step, the product wave function

becomes antisymmetrized and renormalized, which gives the Pauli repulsion ΔEPauli. The third

step consists of the relaxation of the molecular orbitals to their final form to yield the

stabilizing orbital interactions ΔEorb. The sum of the three terms ΔEelstat+ ΔEPauli+ ΔEorb gives

the total interaction energy ΔEint.

The NBO[307, 308] charges were obtained using the NBO 3.1 program implemented in

Gaussian03. The AIM[309] analyses were carried out using a modified version of AIMPAC[310]

using a BP86/def2-SVP wavefunction.


Figure ES5. Calculated geometry of 23M. Bond lengths in Å, angles in degree. Experimental values for 23 are

given in parentheses.

Cartesian Coordinates of the Calculated Compound

Cartesian coordinates of the model system 23M.

Pd 9.976783885 6.763683231 7.287112075

Zn 12.139292929 5.717149246 7.575178749

Zn 7.610431029 6.408484205 7.691184096

Zn 8.976248637 8.908024207 6.750118487

Zn 8.963583859 6.400735791 5.066497100

Zn 9.819863124 4.386737046 7.997764084

Zn 9.532738269 7.895605326 9.438372340

Zn 11.852288890 8.107280263 6.365751227

N 13.616895241 5.671009837 9.268200197

N 13.678739022 4.378605015 6.621986714

C 13.046878722 4.856279347 10.361238967

H 12.832004887 3.839632843 10.013592655


H 12.100212215 5.305646830 10.682911603

H 13.741035032 4.808763702 11.219820978

C 13.871175984 7.041657553 9.758260505

H 14.249962139 7.663492574 8.938754180

H 14.600723955 7.040697837 10.587993143

H 12.926789771 7.475606101 10.108533076

H 15.501149377 4.684507301 9.522305191

C 14.846141250 5.060493102 8.714852799

H 15.404343220 5.848768931 8.192367440

C 14.523089737 3.920997877 7.749659812

H 13.970428856 3.129618985 8.274218947

H 15.463270896 3.470829284 7.380454206

C 13.025337599 3.233441550 5.955664222

H 12.404546073 2.690578582 6.677444929

H 13.772386895 2.545548264 5.520452247

H 12.375135666 3.606077596 5.155986222

C 14.464179677 5.136824745 5.626916307

H 14.933135095 6.012608102 6.087243045

H 13.792584749 5.495865757 4.839409351

H 15.246588518 4.500276586 5.175065813

C 5.691202600 6.116333707 8.056167800

H 5.114139583 7.006827397 7.771396435

H 5.519145772 5.923020738 9.123569128

H 5.309929487 5.261109531 7.482013845

C 8.160539387 10.640992113 6.264457547

H 7.067969283 10.583990595 6.366828949

H 8.394816178 10.908371788 5.225356739

H 8.523799989 11.443101360 6.921394334

C 8.264073397 7.270277878 2.948840749

C 7.150899084 6.564599754 3.484427689


C 7.517867683 5.195712061 3.621012438

C 8.863638084 5.052763055 3.169220044

C 9.325005232 6.338464662 2.751879691

H 8.299145000 8.332324239 2.726250256

H 6.189544395 6.995449483 3.748068272

H 6.882722828 4.398546583 3.995237154

H 9.425673509 4.124581197 3.121731386

H 10.303663128 6.562513722 2.338664715

C 9.562881915 2.051065389 7.631131020

C 8.295211282 2.624003834 7.932213961

C 8.325123444 3.071137914 9.289796101

C 9.606958871 2.772207744 9.823261432

C 10.373550837 2.143933848 8.801458545

H 9.847050854 1.599199587 6.685305630

H 7.447763245 2.690457818 7.257152430

H 7.504437059 3.543665860 9.820672114

H 9.937844102 2.981097029 10.836400321

H 11.382066596 1.752226026 8.910391068

C 10.077700712 8.057878138 11.721772021

C 8.679874870 7.779882988 11.661346536

C 8.037672452 8.890497660 11.047548812

C 9.032330625 9.856627153 10.726320066

C 10.295187253 9.345414683 11.142593073

H 10.834251634 7.423027574 12.174287108

H 8.192342839 6.882105802 12.028094288

H 6.973119479 8.983381513 10.854187995

H 8.859742127 10.817049325 10.249626934

H 11.249533672 9.857704102 11.057008353

C 12.355470085 9.684465193 4.753020223

C 12.245617063 10.413712300 5.977864647


C 13.329787828 10.034524796 6.815068914

C 14.108130820 9.071294428 6.118878515

C 13.506739164 8.851172536 4.842510737

H 11.692463206 9.769436464 3.897811087

H 11.478359641 11.144965295 6.214428590

H 13.523650188 10.416270401 7.813652603

H 15.026296993 8.612685875 6.477023322

H 13.879515887 8.197633484 4.059580567

174 7. References

7 References

[1] G. B. Kauffman, J. Chem. Educ. 1959, 36, 521-527.

[2] G. B. Kauffman, Platinum Met. Rev. 1997, 41, 34-40.

[3] H. Kragh, BJHS 1997, 30, 203-219.

[4] A. Werner, Zeit. anorg. Chem. 1895, 8, 153-188 and 189-197.

[5] A. Werner, Zeit. anorg. Chem. 1895, 9, 382-417.

[6] A. Werner and A. Miolati, Zeit. physikal. Chem. 1896, 21, 225-238.

[7] J. Uddin and G. Frenking, J. Am. Chem. Soc. 2001, 123, 1683-1693.

[8] J. Uddin, C. Boehme and G. Frenking, Organometallics 2000, 19, 571-582.

[9] J. Weiss, D. Stetzkamp, B. Nuber, R. A. Fischer, C. Boehme and G. Frenking, Angew.

Chem., Int. Ed. 1997, 36, 70-72.

[10] M. Asay, C. Jones and M. Driess, Chem. Rev. 2011, 111, 354-396.

[11] C. Dohmeier, D. Loos and H. Schnöckel, Angew. Chem., Int. Ed. 1996, 35, 129-149.

[12] R. M. N. Sudheendra, H. W. Roesky and G. Anantharaman, J. Organomet. Chem. 2002,

646, 4-14.

[13] H. W. Roesky, Inorg. Chem. 2004, 43, 7284-7293.

[14] H. W. Roesky and S. S. Kumar, Chem. Commun. 2005, 4027-4038.

[15] R. A. Fischer and J. Weiss, Angew. Chem., Int. Ed. 1999, 38, 2831-2850.

[16] G. Linti and H. Schnöckel, Coord. Chem. Rev. 2000, 206-207, 285-319.

[17] C. Gemel, T. Steinke, M. Cokoja, A. Kempter and R. A. Fischer, Eur. J. Inorg. Chem.

2004, 4161-4176.

[18] P. W. Roesky, Dalton Trans. 2009, 1887-1893.

[19] F. A. Cotton, Q. Rev., Chem. Soc. 1966, 20, 389-401.

[20] A. Schnepf, G. Stoesser and H. Schnöckel, J. Am. Chem. Soc. 2000, 122, 9178-9181.

[21] A. Purath, R. Koppe and H. Schnöckel, Angew. Chem., Int. Ed. 1999, 38, 2926-2928.

[22] G. Schmid, Clusters and Colloids: From Theory to Applications, VCH, 1994.

[23] G. Schmid, R. Pfeil, R. Boese, F. Brandermann, S. Meyer, G. H. M. Calis and J. W. A.

van der Velden, Chem. Ber. 1981, 114, 3634-3642.

[24] P. Pyykko and N. Runeberg, Angew. Chem., Int. Ed. 2002, 41, 2174-2176.

[25] X. Li, B. Kiran, J. Li, H.-J. Zhai and L.-S. Wang, Angew. Chem., Int. Ed. 2002, 41, 4786-

4789.

[26] H. Schnöckel, Dalton Trans. 2008, 4344-4362.

175 7. References

[27] A. Schnepf and H. Schnöckel, Angew. Chem., Int. Ed. Engl 2002, 41, 3532-3552.

[28] H. Schnöckel, Dalton Trans. 2005, 3131-3136.

[29] J. Vollet, J. R. Hartig and H. Schnöckel, Angew. Chem., Int. Ed. 2004, 43, 3186-3189.

[30] A. Ecker, E. Weckert and H. Schnöckel, Nature 1997, 387, 379-381.

[31] S. Schulz, Chem.- Eur. J. 2010, 16, 6416-6428.

[32] A. Schnepf and H. Schnöckel, Angew. Chem., Int. Ed. 2001, 40, 712-715.

[33] M. Cokoja, B. R. Jagirdar, H. Parala, A. Birkner and R. A. Fischer, Eur. J. Inorg. Chem.

2008, 3330-3339.

[34] M. Cokoja, H. Parala, A. Birkner, O. Shekhah, M. W. E. van den Berg and R. A. Fischer,

Chem. Mater. 2007, 19, 5721-5733.

[35] M. Cokoja, H. Parala, M.-K. Schroeter, A. Birkner, M. W. E. Van den Berg, W. Gruenert

and R. A. Fischer, Chem. Mater. 2006, 18, 1634-1642.

[36] R. A. Fischer, H. D. Kaesz, S. I. Khan and H. J. Muller, Inorg. Chem. 1990, 29, 1601-

1602.

[37] M. A. Malik and P. O'Brien, Top. Organomet. Chem. 2005, 9, 173-204.

[38] S. P. Green, C. Jones, D. P. Mills and A. Stasch, Organometallics 2007, 26, 3424-3430.

[39] T. Bollermann, A. Puls, C. Gemel, T. Cadenbach and R. A. Fischer, Dalton Trans. 2009,

1372-1377.

[40] T. Cadenbach, C. Gemel, R. Schmid, M. Halbherr, K. Yusenko, M. Cokoja and R. A.

Fischer, Angew. Chem., Int. Ed. 2009, 48, 3872-3876.

[41] M. Cokoja, H. Parala, M. K. Schroeter, A. Birkner, M. W. E. van den Berg, K. V.

Klementiev, W. Gruenert and R. A. Fischer, J. Mater. Chem. 2006, 16, 2420-2428.

[42] T. Cadenbach, T. Bollermann, C. Gemel, I. Fernandez, M. von Hopffgarten, G. Frenking

and R. A. Fischer, Angew. Chem., Int. Ed. 2008, 47, 9150-9154.

[43] T. Cadenbach, T. Bollermann, C. Gemel, M. Tombul, I. Fernandez, M. von Hopffgarten,

G. Frenking and R. A. Fischer, J. Am. Chem. Soc. 2009, 131, 16063-16077.

[44] T. Bollermann, T. Cadenbach, C. Gemel, M. von Hopffgarten, G. Frenking and R. A.

Fischer, Chem.- Eur. J. 2010 16, 13372-13384.

[45] J. H. Westbrook, Intermetallic Compounds, Wiley, 1966.

[46] R. Nesper, Angew. Chem. 1991, 103, 805-834.

[47] E. Zintl, Angew. Chem. 1939, 52, 1-6.

[48] J. D. Corbett, D. G. Adolphson, D. J. Merryman, P. A. Edwards and F. J. Armatis, J. Am.

Chem. Soc. 1975, 97, 6267-6268.

[49] R. Ferro and A. Saccone, Intermetallic Chemistry, Elsevier, 2007.

176 7. References

[50] W. E. Wallace and H. Eyring, Annu. Rev. Phys. Chem. 1964, 15, 109-130.

[51] G. Sauthoff, Intermetallics, VCH, 1995.

[52] G. P. Tiwari and R. V. Ramanujan, J. Mater. Sci. 2001, 36, 271-283.

[53] F. Jona and P. M. Marcus, J. Phys.: Condens. Matter 2002, 14, 1275-1284.

[54] L. M. Hoistad and S. Lee, J. Am. Chem. Soc. 1991, 113, 8216-8220.

[55] U. Mizutani, T. Noritake, T. Ohsuna and T. Takeuchi, Philos. Mag. 2010, 90, 1985-2008.

[56] J. P. Gaspard and R. Ceolin, Solid State Commun. 1992, 84, 839-842.

[57] S. Tanigawa and M. Doyama, Phys. Lett. A 1973, 43, 17-18.

[58] L. S. Darken and R. W. Gurry, Physical Chemistry of Metals, McGraw-Hill Book Co.,

1953.

[59] W. Hume-Rothery, R. E. Smallman and C. W. Haworth, The Structure of Metals and

Alloys. 5th ed., Metals and Metallurgy Trust, 1969.

[60] J. A. Alonso and S. Simozar, Phys. Rev. B 1980, 22, 5583-5589.

[61] http://ruby.chemie.uni-freiburg.de/Vorlesung/intermetallische_0.html.

[62] K. A. Gschneidner, Jr. and V. K. Pecharsky, Z. Kristallogr. 2006, 221, 375-381.

[63] R. L. Johannes, R. Haydock and V. Heine, Phys. Rev. Lett. 1976, 36, 372-376.

[64] R. L. Johnston and R. Hoffmann, Z. Anorg. Allg. Chem. FIELD 1992, 616, 105-120.

[65] K. Miyajima, N. Fukushima, H. Himeno, A. Yamada and F. Mafune, J. Phys. Chem. A

2009, 113, 13448-13450.

[66] U. Mizutani, R. Asahi, H. Sato, T. Noritake and T. Takeuchi, J. Phys.: Condens. Matter

2008, 20, 275228/275221-275228/275211.

[67] S. C. Sevov and J. M. Goicoechea, Organometallics 2006, 25, 5678-5692.

[68] J. D. Corbett, Angew. Chem., Int. Ed. 2000, 39, 670-690.

[69] J. D. Corbett, Chem. Rev. 1985, 85, 383-397.

[70] S. Scharfe, F. Kraus, S. Stegmaier, A. Schier and T. F. Faessler, Angew. Chem., Int. Ed.

2011, 50, 3630-3670.

[71] T. F. Faessler, Coord. Chem. Rev. 2001, 215, 347-377.

[72] T. F. Faessler and S. D. Hoffmann, Angew. Chem., Int. Ed. 2004, 43, 6242-6247.

[73] J. D. Corbett, Inorg. Chem. 2011, 49, 13-28.

[74] H. Schaefer, B. Eisenmann and W. Mueller, Angew. Chem. 1973, 85, 742-760.

[75] C. Knapp, B. Zhou, M. S. Denning, N. H. Rees and J. M. Goicoechea, Dalton Trans.

2010, 39, 426-436.

[76] C. Belin and M. Tillard-Charbonnel, Coord. Chem. Rev. 1998, 178-180, 529-564.

[77] W. Klemm, Angew. Chem. 1950, 62, 133-142.

177 7. References

[78] W. Klemm and E. Busmann, Z. Anorg. Allg. Chem. 1963, 319, 297-311.

[79] E. Zintl, J. Goubeau and W. Dullenkopf, Z. physik. Chem. 1931, 154, 1-46.

[80] E. Zintl and A. Harder, Z. physik. Chem. 1931, 154, 47-91.

[81] E. Zintl, A. Harder and S. Neumayr, Z. physik. Chem. 1931, 154, 92-96.

[82] L. Diehl, K. Khodadadeh, D. Kummer and J. Straehle, Chem. Ber. 1976, 109, 3404-3418.

[83] T. F. Faessler and R. Hoffmann, Dalton Trans. 1999, 3339-3340.

[84] V. Queneau and S. C. Servov, Angew. Chem., Int. Ed. 1997, 36, 1754-1756.

[85] L. Xu and S. C. Sevov, J. Am. Chem. Soc. 1999, 121, 9245-9246.

[86] R. Hauptmann and T. F. Faessler, Z. Anorg. Allg. Chem. 2003, 629, 2266-2273.

[87] C. Downie, Z. Tang and A. M. Guloy, Angew. Chem., Int. Ed. 2000, 39, 338-340.

[88] A. Ugrinov and S. C. Sevov, Chem.-Eur. J. 2004, 10, 3727-3733.

[89] B. W. Eichhorn, R. C. Haushalter and W. T. Pennington, J. Am. Chem. Soc. 1988, 110,

8704-8706.

[90] B. Kesanli, J. Fettinger and B. Eichhorn, Chem.-Eur. J. 2001, 7, 5277-5285.

[91] B. W. Eichhorn and R. C. Haushalter, Chem. Commun. 1990, 937-938.

[92] J. Campbell, H. P. A. Mercier, H. Franke, D. P. Santry, D. A. Dixon and G. J.

Schrobilgen, Inorg. Chem. 2002, 41, 86-107.

[93] L. Yong, S. D. Hoffmann and T. F. Faessler, Eur. J. Inorg. Chem. 2005, 3663-3669.

[94] J. M. Goicoechea and S. C. Sevov, Organometallics 2006, 25, 4530-4536.

[95] D. R. Gardner, J. C. Fettinger and B. W. Eichhorn, Angew. Chem., Int. Ed. 1997, 35,

2852-2854.

[96] E. N. Esenturk, J. Fettinger and B. Eichhorn, Polyhedron 2006, 25, 521-529.

[97] E. N. Esenturk, J. Fettinger, Y.-F. Lam and B. Eichhorn, Angew. Chem., Int. Ed. 2004,

43, 2132-2134.

[98] J. M. Goicoechea and S. C. Sevov, J. Am. Chem. Soc. 2006, 128, 4155-4161.

[99] B. Kesanli, J. Fettinger, D. R. Gardner and B. Eichhorn, J. Am. Chem. Soc. 2002, 124,

4779-4786.

[100] E. N. Esenturk, J. Fettinger and B. Eichhorn, Chem. Commun. 2005, 247-249.

[101] J. Rosdahl, T. F. Faessler and L. Kloo, Eur. J. Inorg. Chem. 2005, 2888-2894.

[102] S. Rossi, K. Kallinen, J. Pursiainen, T. T. Pakkanen and T. A. Pakkanen, J. Organomet.

Chem. 1991, 419, 219-232.

[103] B. F. G. Johnson, Coord. Chem. Rev. 1999, 190-192, 1269-1285.

[104] A. Schnepf, Chem. Soc. Rev. 2007, 36, 745-758.

[105] H. Schnöckel and H. Kohnlein, Polyhedron 2002, 21, 489-501.

178 7. References

[106] R. B. King and G. H. Robinson, J. Organomet. Chem. 2000, 597, 54-60.

[107] A. Ceriotti, F. Demartin, G. Longoni, M. Manassero, M. Marchionna, G. Piva and M.

Sansoni, Angew. Chem. 1985, 97, 708-710.

[108] N. T. Tran, D. R. Powell and L. F. Dahl, Angew. Chem., Int. Ed. 2000, 39, 4121-4125.

[109] A. Purath, R. Koppe and H. Schnöckel, Chem. Commun. 1999, 1933-1934.

[110] H. Kohnlein, G. Stosser, E. Baum, E. Mollhausen, U. Huniar and H. Schnöckel, Angew.

Chem., Int. Ed. 2000, 39, 799-801.

[111] H. Koehnlein, A. Purath, C. Klemp, E. Baum, I. Krossing, G. Stoesser and H.

Schnöckel, Inorg. Chem. 2001, 40, 4830-4838.

[112] C. Dohmeier, C. Robl, M. Tacke and H. Schnöckel, Angew. Chem. 1991, 103, 594-595.

[113] D. Loos and H. Schnöckel, J. Organomet. Chem. 1993, 463, 37-40.

[114] H. Meister, Angew. Chem. 1957, 69, 533-534.

[115] E. O. Fischer and H. P. Hofmann, Angew. Chem. 1957, 69, 639-640.

[116] P. Jutzi, B. Neumann, G. Reumann and H.-G. Stammler, Organometallics 1998, 17,

1305-1314.

[117] S. Schulz, H. W. Roesky, H. J. Koch, G. M. Sheldrick, D. Stalke and A. Kuhn, Angew.

Chem. 1993, 105, 1828-1830.

[118] C. Cui, H. W. Roesky, H.-G. Schmidt, M. Noltemeyer, H. Hao and F. Cimpoesu,

Angew. Chem., Int. Ed. 2000, 39, 4274-4276.

[119] N. J. Hardman, B. E. Eichler and P. P. Power, Chem. Commun. 2000, 1991-1992.

[120] W. Uhl, M. Benter, S. Melle, W. Saak, G. Frenking and J. Uddin, Organometallics

1999, 18, 3778-3780.

[121] C. Boehme and G. Frenking, Chem.-Eur. J. 1999, 5, 2184-2190.

[122] C. L. B. Macdonald and A. H. Cowley, J. Am. Chem. Soc. 1999, 121, 12113-12126.

[123] D. Weiss, T. Steinke, M. Winter, R. A. Fischer, N. Froehlich, J. Uddin and G. Frenking,

Organometallics 2000, 19, 4583-4588.

[124] M. Doerr and G. Frenking, Z. Anorg. Allg. Chem. 2002, 628, 843-850.

[125] G. Frenking, K. Wichmann, N. Frohlich, C. Loschen, M. Lein, J. Frunzke and V. M.

Rayon, Coord. Chem. Rev. 2003, 238-239, 55-82.

[126] P. Jutzi, B. Neumann, L. O. Schebaum, A. Stammler and H.-G. Stammler,


[127] M. Cokoja, T. Steinke, C. Gemel, T. Welzel, M. Winter, K. Merz and R. A. Fischer, J.

Organomet. Chem. 2003, 684, 277-286.

179 7. References

[128] C. Gemel, T. Steinke, D. Weiss, M. Cokoja, M. Winter and R. A. Fischer,


[129] A. Kempter, C. Gemel, T. Cadenbach and R. A. Fischer, Inorg. Chem. 2007, 46, 9481-

9487.

[130] D. Weiss, M. Winter, R. A. Fischer, C. Yu, K. Wichmann and G. Frenking, Chem.

Commun. 2000, 2495-2496.

[131] T. Steinke, C. Gemel, M. Winter and R. A. Fischer, Angew. Chem., Int. Ed. 2002, 41,

4761-4763.

[132] T. Steinke, C. Gemel, M. Winter and R. A. Fischer, Chem.-Eur. J. 2005, 11, 1636-1646.

[133] M. Cokoja, H. Parala, A. Birkner, R. A. Fischer, O. Margeat, D. Ciuculescu, C. Amiens,

B. Chaudret, A. Falqui and P. Lecante, Eur. J. Inorg. Chem. 2010, 11, 1599-1603.

[134] E. Carmona and A. Galindo, Angew. Chem., Int. Ed. 2008, 47, 6526-6536.

[135] M. Kaupp and H. G. von Schnering, Inorg. Chem. 1994, 33, 4179-4185.

[136] D. Bravo-Zhivotovskii, M. Yuzefovich, M. Bendikov, K. Klinkhammer and Y. Apeloig,

Angew. Chem., Int. Ed. 1999, 38, 1100-1102.

[137] 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, 10847-10857.

[138] I. Resa, E. Carmona, E. Gutierrez-Puebla and A. Monge, Science 2004, 305, 1136-

1138.

[139] D. del Rio, A. Galindo, I. Resa and E. Carmona, Angew. Chem. Int. Ed. 2005, 44, 1244-

1247.

[140] A. Schnepf and H.-J. Himmel, Angew. Chem., Int. Ed. 2005, 44, 3006-3008.

[141] A. Grirrane, I. Resa, A. Rodriguez, E. Carmona, E. Alvarez, E. Gutierrez-Puebla, A.

Monge, A. Galindo, D. Del Rio and R. A. Andersen, J. Am. Chem. Soc. 2007, 129, 693-703.

[142] S. S. Hepperle and Y. A. Wang, J. Phys. Chem. 2008, 112, 9619-9622.

[143] B. Fischer, J. Boersma, G. Van Koten and A. L. Spek, New J. Chem. 1988, 12, 613-620.

[144] D. del Rio, I. Resa, A. Rodriguez, L. Sanchez, R. Koppe, A. J. Downs, C. Y. Tang and

E. Carmona, J. Phys. Chem. 2008, 112, 10516-10525.

[145] J. W. Kress, J. Phys. Chem. 2005, 109, 7757-7763.

[146] Y. Kan, J. Mol. Struct. 2007, 805, 127-132.

[147] M. R. Philpott and Y. Kawazoe, J. Mol. Struct. 2006, 773, 43-52.

[148] M. R. Philpott and Y. Kawazoe, Chem. Phys. 2006, 327, 283-290.

[149] K. K. Pandey, J. Organomet. Chem. 2007, 692, 1058-1063.

[150] Z.-Z. Xie and W.-H. Fang, Chem. Phys. Lett. 2005, 404, 212-216.

180 7. References

[151] M. R. Philpott and Y. Kawazoe, J. Mol. Struct. 2006, 776, 113-123.

[152] Z.-Z. Liu, W. Q. Tian, J.-K. Feng, G. Zhang and W.-Q. Li, J. Mol. Struct. 2006, 758,

127-138.

[153] Y. Xie, H. F. Schaefer, III and R. B. King, J. Am. Chem. Soc. 2005, 127, 2818-2819.

[154] M. R. Philpott and Y. Kawazoe, Chem. Phys. 2007, 333, 201-207.

[155] S. L. Richardson, T. Baruah and M. R. Pederson, Chem. Phys. Lett. 2005, 415, 141-145.

[156] S. Schulz, D. Schuchmann, U. Westphal and M. Bolte, Organometallics 2009, 28,

1590-1592.

[157] H. P. Nayek, A. Luehl, S. Schulz, R. Koeppe and P. W. Roesky, Chem.-Eur. J. 2011,

17, 1773-1777.

[158] S. Schulz, S. Gondzik, D. Schuchmann, U. Westphal, L. Dobrzycki, R. Boese and S.

Harder, Chem. Commun. 2010, 46, 7757-7759.

[159] S. Gondzik, D. Blaeser, C. Woelper and S. Schulz, Chem.-Eur. J. 2010, 16, 13599-

13602.

[160] Z. Zhu, R. J. Wright, M. M. Olmstead, E. Rivard, M. Brynda and P. P. Power, Angew.

Chem., Int. Ed. 2006, 45, 5807-5810.

[161] Y. Liu, S. Li, X.-J. Yang, P. Yang, J. Gao, Y. Xia and B. Wu, Organometallics 2009,

28, 5270-5272.

[162] 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, 11944-11945.

[163] I. L. Fedushkin, A. A. Skatova, S. Y. Ketkov, O. V. Eremenko, A. V. Piskunov and G.

K. Fukin, Angew. Chem., Int. Ed. 2007, 46, 4302-4305.

[164] D. Schuchmann, U. Westphal, S. Schulz, U. Florke, D. Blaser and R. Boese, Angew.

Chem., Int. Ed. 2009, 48, 807-810.

[165] S. Schulz, D. Schuchmann, I. Krossing, D. Himmel, D. Blaeser and R. Boese, Angew.

Chem., Int. Ed. 2009, 48, 5748-5751.

[166] M. Carrasco, R. Peloso, I. Resa, A. Rodriguez, L. Sanchez, E. Alvarez, C. Maya, R.

Andreu, J. J. Calvente, A. Galindo and E. Carmona, Inorg. Chem. 2011, 50, 6361-6371.

[167] M. Carrasco, R. Peloso, A. Rodriguez, E. Alvarez, C. Maya and E. Carmona, Chem.-

Eur. J. 2010, 16, 9754-9757.

[168] G. Gao, X. Xu and H. S. Kang, J. Comput. Chem. 2009, 30, 978-982.

[169] A. Velazquez, I. Fernandez, G. Frenking and G. Merino, Organometallics 2007, 26,

4731-4736.

[170] J.-C. Guo and J.-G. Li, J. Mol. Struct. 2010, 942, 43-46.

181 7. References

[171] H. S. Kang, J. Phys. Chem. 2005, 109, 4342-4351.

[172] Q. S. Li and Y. Xu, J. Phys. Chem. 2006, 110, 11898-11902.

[173] Z.-Z. Liu, W. Q. Tian, J.-K. Feng, G. Zhang, W.-Q. Li, Y.-H. Cui and C.-C. Sun, Eur.

J. Inorg. Chem. 2006, 2808-2818.

[174] M. Toganoh, Y. Matsuo and E. Nakamura, J. Am. Chem. Soc. 2003, 125, 13974-13975.

[175] E. Frankland, Justus Liebigs Ann. Chem. 1849, 171-213.

[176] J. A. Casares, P. Espinet, B. Fuentes and G. Salas, J. Am. Chem. Soc. 2007, 129, 3508-

3509.

[177] Y. Zhao, H. Wang, X. Hou, Y. Hu, A. Lei, H. Zhang and L. Zhu, J. Am. Chem. Soc.

2006, 128, 15048-15049.

[178] Q. Liu, Y. Lan, J. Liu, G. Li, Y.-D. Wu and A. Lei, J. Am. Chem. Soc. 2009, 131,

10201-10210.

[179] E.-i. Negishi, G. Wang, H. Rao and Z. Xu, J. Org. Chem. 2010, 75, 3151-3182.

[180] A. O. King, N. Okukado and E. Negishi, Chem. Commun. 1977, 683-684.

[181] B. Kranke and H. Kunz, Org. Biomol. Chem. 2007, 5, 349-354.

[182] P. Knochel, X. Yang and N. Gommermann, Handb. Funct. Organomet. 2005, 2, 379-

395.

[183] A. M. Caporusso, S. Filippi, F. Barontini and P. Salvadori, Tetrahedron Lett. 2000, 41,

1227-1230.

[184] A. M. Caporusso, A. Zampieri, L. A. Aronica and D. Banti, J. Org. Chem. 2006, 71,

1902-1910.

[185] W. Hieber and U. Teller, Z. Anorg. Allg. Chem. 1942, 249, 43-57.

[186] J. M. Burlitch and R. C. Winterton, Inorg. Chem. 1979, 18, 2309-2311.

[187] J. M. Burlitch, S. E. Hayes and G. E. Whitwell, II, Organometallics 1982, 1, 1074-

1083.

[188] B. E. Zaugg, T. Kolb, A. M. Arif and R. D. Ernst, J. Chem. Crystallogr. 2010, 40, 778-

782.

[189] O. Fuhr and D. Fenske, Z. Anorg. Allg. Chem. 2000, 626, 1822-1830.

[190] D. E. Crotty, T. J. Anderson, M. D. Glick and J. P. Oliver, Inorg. Chem. 1977, 16,

2346-2350.

[191] P. H. M. Budzelaar, J. Boersma, G. J. M. Van der Kerk, A. L. Spek and A. J. M.

Duisenberg, Inorg. Chem. 1982, 21, 3777-3780.

[192] P. H. M. Budzelaar, K. H. Den Haan, J. Boersma, G. J. M. Van der Kerk and A. L.

Spek, Organometallics 1984, 3, 156-159.

182 7. References

[193] P. H. M. Budzelaar, A. A. H. Van der Zeijden, J. Boersma, G. J. M. Van der Kerk, A. L.

Spek and A. J. M. Duisenberg, Organometallics 1984, 3, 159-163.

[194] P. H. M. Budzelaar, J. Boersma and G. J. M. Van der Kerk, Angew. Chem. 1983, 95,

335-336.


Duisenberg, Organometallics 1985, 4, 680-683.

[196] B. Fischer, H. Kleijn, J. Boersma, G. Van Koten and A. L. Spek, Organometallics 1989,

8, 920-925.


Duisenberg, J. Organomet. Chem. 1985, 287, C13-C17.

[198] M. D. Fryzuk, D. H. McConville and S. J. Rettig, Organometallics 1990, 9, 1359-1360.

[199] M. D. Fryzuk, D. H. McConville and S. J. Rettig, Organometallics 1993, 12, 2152-

2161.

[200] H. Suzuki, H. Omori, D. H. Lee, Y. Yoshida, M. Fukushima, M. Tanaka and Y. Moro-

oka, Organometallics 1994, 13, 1129-1146.

[201] H. Suzuki, H. Omori and Y. Moro-Oka, Organometallics 1988, 7, 2579-2581.

[202] A. Inagaki, T. Takenori, M. Tanaka and H. Suzuki, Angew. Chem., Int. Ed. 2000, 39,

404-406.

[203] Y. Ohki, T. Kojima, M. Oshima and H. Suzuki, Organometallics 2001, 20, 2654-2656.

[204] T. Takao, A. Inagaki, E. Murotani, T. Imamura and H. Suzuki, Organometallics 2003,

22, 1361-1363.

[205] M. Ohashi, K. Matsubara, T. Iizuka and H. Suzuki, Angew. Chem., Int. Ed. 2003, 42,

937-940.

[206] M. Ohashi, K. Matsubara and H. Suzuki, Organometallics 2007, 26, 2330-2339.

[207] T. Cadenbach, C. Gemel and R. A. Fischer, Angew. Chem., Int. Ed. Engl. 2008, 47,

9146-9149.

[208] M. Molon, T. Cadenbach, T. Bollermann, C. Gemel and R. A. Fischer, Chem. Commun.

2010, 46, 5677-5679.

[209] T. Cadenbach, C. Gemel, D. Zacher and R. A. Fischer, Angew. Chem., Int. Ed. 2008,

47, 3438-3441.

[210] M. Halbherr, T. Bollermann, C. Gemel and R. A. Fischer, Angew. Chem., Int. Ed. 2010,

49, 1878-1881.

[211] The structural and spectroscopic characterisation of [Mo(GaCp*)6] has been also

previously described: a) T. Cadenbach, T. Bollermann, C. Gemel, I. Fernandez, M. von

183 7. References

Hopffgarten, G. Frenking and R. A. Fischer, Angew. Chem., Int. Ed. 2008, 47, 9150-9154, b)

T. Cadenbach, Dissertation, Ruhr Universität Bochum, 2009.

[212] M. G. B. Drew, Coord. Chem. Rev. 1977, 24, 179-275.

[213] J. Cirera, E. Ruiz and S. Alvarez, Organometallics 2005, 24, 1556-1562.

[214] D. Loos, E. Baum, A. Ecker, H. Schnöckel and A. J. Downs, Angew. Chem., Int. Ed.

1997, 36, 860-862.

[215] F. G. N. Cloke, K. P. Cox, M. L. H. Green, J. Bashkin and K. Prout, Chem. Commun.

1982, 393-394.

[216] M. Brookhart, K. Cox, F. G. N. Cloke, J. C. Green, M. L. H. Green, P. M. Hare, J.

Bashkin, A. E. Derome and P. D. Grebenik, Dalton Trans. 1985, 423-433.

[217] G. A. Lawrance, Chem. Rev. 1986, 86, 17-33.

[218] H. Zabrodsky, S. Peleg and D. Avnir, J. Am. Chem. Soc. 1993, 115, 8278-8289.

[219] M. Pinsky and D. Avnir, Inorg. Chem. 1998, 37, 5575-5582.

[220] L. Sacconi, Pure Appl. Chem. 1968, 17, 95-127.

[221] T. Cadenbach, C. Gemel, R. Schmid, S. Block and R. A. Fischer, Dalton Trans. 2004,

3171-3172.

[222] T. Cadenbach, C. Gemel, T. Bollermann and R. A. Fischer, Inorg. Chem. 2009, 48,

5021-5026.

[223] G. Linti and A. Seifert, Z. Anorg. Allg. Chem. 2008, 634, 1312-1320.

[224] A. H. Cowley, F. P. Gabbaie, D. A. Atwood, C. J. Carrano, L. M. Mokry and M. R.

Bond, J. Am. Chem. Soc. 1994, 116, 1559-1560.

[225] G. Prabusankar, C. Gemel, P. Parameswaran, C. Flener, G. Frenking and R. A. Fischer,

Angew. Chem., Int. Ed. 2009, 48, 5526-5529.

[226] G. DiPasquale Antonio and M. Mayer James, J. Am. Chem. Soc. 2008, 130, 1812-1813.

[227] A. Kempter, C. Gemel, N. J. Hardman and R. A. Fischer, Inorg. Chem. 2006, 45, 3133-

3138.

[228] R. A. Fischer, D. Weiss, M. Winter, I. Mueller, H. D. Kaesz, N. Froehlich and G.

Frenking, J. Organomet. Chem. 2004, 689, 4611-4623.

[229] D. Weiss, M. Winter, K. Merz, A. Knufer, R. A. Fischer, N. Frohlich and G. Frenking,

Polyhedron 2002, 21, 535-542.

[230] B. Fraser, L. Brandt, W. K. Stovall, H. D. Kaesz, S. I. Khan and F. Maury, J.

Organomet. Chem. 1994, 472, 317-328.

[231] N. P. Mankad, D. S. Laitar and J. P. Sadighi, Organometallics 2004, 23, 3369-3371.

184 7. References

[232] A. Kempter, C. Gemel, T. Cadenbach and R. A. Fischer, Organometallics 2007, 26,

4257-4264.

[233] A. Kempter, C. Gemel and R. A. Fischer, Chem.-Eur. J. 2007, 13, 2990-3000.

[234] Q. Yu, A. Purath, A. Donchev and H. Schnöckel, J. Organomet. Chem. 1999, 584, 94-

97.

[235] G. Parkin, J. Chem. Educ. 2006, 83, 791-799.

[236] T. Cadenbach, Dissertation, Ruhr Universität Bochum, 2009.

[237] R. D. Ernst and T. J. Marks, Inorg. Chem. 1978, 17, 1477-1484.

[238] R. J. Neustadt, T. H. Cymbaluk, R. D. Ernst and F. W. Cagle, Jr., Inorg. Chem. 1980,

19, 2375-2381.

[239] C. Chieh, D. Seyferth and L. C. Song, Organometallics 1982, 1, 473-476.

[240] A. M. Arif, A. H. Cowley and M. Pakulski, Chem. Commun. 1987, 622-623.

[241] M. Elder and D. Hall, Inorg. Chem. 1969, 8, 1424-1427.

[242] Y. Chen, Y. Zhou and J. Qu, Organometallics 2008, 27, 666-671.

[243] a) F. A. Cotton and J. M. Troup, Dalton Trans. 1974, 800-802, b) J. H. Hang, J. G. Lee,

Y. Xie and H. F. Schaefer III, J. Phys. Chem. 1998, 102, 5298-5304, c) Y. Xie, H. F. Schaefer

III and R. B. King, J. Am. Chem. Soc. 2000, 122, 8746-8761.

[244] C. G. Pierpont, R. G. Shong and B. A. Sosinsky, Inorg. Chem. 1982, 21, 3247-3248.

[245] J. K. Brandon, R. Y. Brizard, P. C. Chieh, R. K. McMillan and W. B. Pearson, Acta

Crystallogr., Sect. B 1974, 30, Pt. 6, 1412-1417.

[246] G. Wei, H. Liu, Z. Huang, L. Huang and B. Kang, Chem. Commun. 1989, 1839-1840.

[247] F. A. Cotton, C. A. Murillo and R. A. Walton, Mult. Bonds Met. At. (3rd Ed.) 2005, 1-

21.

[248] G. Merino, K. J. Donald, J. S. D'Acchioli and R. Hoffmann, J. Am. Chem. Soc. 2007,

129, 15295-15302.

[249] G. Frenking, Science 2005, 310, 796-797.

[250] E. Maslowsky, Jr., Chem. Rev. 1971, 71, 507-524.

[251] S. P. Green, C. Jones and A. Stasch, Science 2007, 318, 1754-1757.

[252] S. P. Green, C. Jones and A. Stasch, Angew. Chem., Int. Ed. 2008, 47, 9079-9083.

[253] Y. Wang, Y. Xie, P. Wei, R. B. King, H. F. Schaefer, III, P. v. R. Schleyer and G. H.

Robinson, Science 2008, 321, 1069-1071.

[254] Y. Wang, B. Quillian, P. Wei, C. S. Wannere, Y. Xie, R. B. King, H. F. Schaefer, III, P.

v. R. Schleyer and G. H. Robinson, J. Am. Chem. Soc. 2007, 129, 12412-12413.

185 7. References

[255] Y. Wang, B. Quillian, P. Wei, Y. Xie, C. S. Wannere, R. B. King, H. F. Schaefer, III, P.

v. R. Schleyer and G. H. Robinson, J. Am. Chem. Soc. 2008, 130, 3298-3299.

[256] M. Y. Abraham, Y. Wang, Y. Xie, P. Wei, H. F. Schaefer, III, P. v. R. Schleyer and G.

H. Robinson, Chem.-Eur. J. 2010, 16, 432-435.

[257] A. Sidiropoulos, C. Jones, A. Stasch, S. Klein and G. Frenking, Angew. Chem., Int. Ed.

2009, 48, 9701-9704.

[258] Y. Wang, Y. Xie, P. Wei, R. B. King, H. F. Schaefer, III, P. v. R. Schleyer and G. H.

Robinson, J. Am. Chem. Soc. 2008, 130, 14970-14971.

[259] B. Quillian, P. Wei, C. S. Wannere, P. v. R. Schleyer and G. H. Robinson, J. Am. Chem.

Soc. 2009, 131, 3168-3169.

[260] P. J. Davidson and M. F. Lappert, Chem. Commun. 1973, 317.

[261] D. E. Goldberg, D. H. Harris, M. F. Lappert and K. M. Thomas, Chem. Commun. 1976,

261-262.

[262] For Computational Details see: T. Bollermann, K. Freitag, C. Gemel, R. W. Seidel, M.

von Hopffgarten, G. Frenking and R. A. Fischer, Angew. Chem., Int. Ed. 2011, 50, 772-776.

[263] B. Lippert and D. Neugebauer, Inorg. Chem. 1982, 21, 451-452.

[264] E. Goto, R. A. Begum, S. Zhan, T. Tanase, K. Tanigaki and K. Sakai, Angew. Chem.,

Int. Ed. 2004, 43, 5029-5032.

[265] C.-C. Wang, W.-C. Lo, C.-C. Chou, G.-H. Lee, J.-M. Chen and S.-M. Peng, Inorg.

Chem. 1998, 37, 4059-4065.

[266] C. Yin, G.-C. Huang, C.-K. Kuo, M.-D. Fu, H.-C. Lu, J.-H. Ke, K.-N. Shih, Y.-L.

Huang, G.-H. Lee, C.-Y. Yeh, C.-h. Chen and S.-M. Peng, J. Am. Chem. Soc. 2008, 130,

10090-10092.

[267] E. J. Fernandez, A. Laguna, J. M. Lopez-de-Luzuriaga, M. E. Olmos and J. Perez,

Dalton Trans. 2004, 1801-1806.

[268] C. Yamamoto, H. Nishikawa, M. Nihei, T. Shiga, M. Hedo, Y. Uwatoko, H. Sawa, H.

Kitagawa, Y. Taguchi, Y. Iwasa and H. Oshio, Inorg. Chem. 2006, 45, 10270-10276.

[269] B. D. Cutforth, C. G. Davies, P. A. W. Dean, R. J. Gillespie, P. R. Ireland and P. K.

Ummat, Inorg. Chem. 1973, 12, 1343-1347.

[270] B. D. Cutforth, R. J. Gillespie, P. Ireland, J. F. Sawyer and P. K. Ummat, Inorg. Chem.

1983, 22, 1344-1347.

[271] S. Roth, V. Ramamoorthy and P. R. Sharp, Inorg. Chem. 1990, 29, 3345-3349.

[272] S. D. Robinson and B. L. Shaw, J. Chem. Soc. 1965, 1529-1530.

186 7. References

[273] N. M. Boag, D. Boucher, J. A. Davies, R. W. Miller, A. A. Pinkerton and R. Syed,


[274] H. J. Kraus, H. Werner and C. Krueger, Z. Naturforsch. 1983, 38B, 733-737.

[275] C. M. Fafard, C.-H. Chen, B. M. Foxman and O. V. Ozerov, Chem. Commun. 2007,

4465-4467.

[276] H. Zabrodsky, S. Peleg and D. Avnir, J. Am. Chem. Soc. 1992, 114, 7843-7851.

[277] V. Riera, M. A. Ruiz, A. Tiripicchio and M. Tiripicchio-Camellini, Organometallics

1993, 12, 2962-2972.

[278] H. Nowotny and H. Bittner, Monatsh. Chem. 1950, 81, 679-680.

[279] T. Steinke, M. Cokoja, C. Gemel, A. Kempter, A. Krapp, G. Frenking, U. Zenneck and

R. A. Fischer, Angew. Chem., Int. Ed. 2005, 44, 2943-2946.

[280] G. V. Goeden, J. C. Huffman and K. G. Caulton, Inorg. Chem. 1986, 25, 2484-2485.

[281] P. Hofmann, I. V. Shishkov and F. Rominger, Inorg. Chem. 2008, 47, 11755-11762.

[282] C. Mealli, C. S. Arcus, J. L. Wilkinson, T. J. Marks and J. A. Ibers, J. Am. Chem. Soc.

1976, 98, 711-718.

[283] G. M. Sheldrick, Acta Crystallogr. 2008, A64, 112-122.

[284] A. L. Spek, Acta Crystallogr. 2009, D65, 148-155.

[285] P. Van der Sluis and A. L. Spek, Acta Crystallogr. 1990, A46, 194-201.

[286] http://www.linden-cms.de/

[287] D. L. Reger, T. D. Wright, C. A. Little, J. J. S. Lamba and M. D. Smith, Inorg. Chem.

2001, 40, 3810-3814.

[288] W. De Graaf, J. Boersma, W. J. J. Smeets, A. L. Spek and G. Van Koten,


[289] R. G. Salomon and J. K. Kochi, J. Am. Chem. Soc. 1973, 95, 1889-1897.

[290] P. Jutzi and L. O. Schebaum, J. Organomet. Chem. 2002, 654, 176-179.

[291] D. J. Krysan and P. B. Mackenzie, J. Org. Chem. 1990, 55, 4229-4230.

[292] E. Carmona, J. M. Marin, M. L. Poveda, J. L. Atwood and R. D. Rogers, J. Am. Chem.

Soc. 1983, 105, 3014-3022.

[293] W. Gausing and G. Wilke, Angew. Chem. 1981, 93, 201-202.

[294] R. Blom, J. Boersma, P. H. M. Budzelaar, B. Fischer, A. Haaland, H. V. Volden and J.

Weidlein, Acta Chem. Scand. 1986, A40, 113-120.

[295] H. Werner, M. Bosch, M. E. Schneider, C. Hahn, F. Kukla, M. Manger, B. Windmuller,

B. Weberndorfer and M. Laubender, Dalton Trans. 1998, 3549-3558.

187 7. References

[296] M. Green, J. A. K. Howard, J. L. Spencer and F. G. A. Stone, Dalton Trans. 1977, 271-

277.

[297] http://www.uni-marburg.de/fb15/fachgebiete/theorie/frenking

[298] A. D. Becke, Phys. Rev. 1988, 38, 3098-3100.

[299] J. Perdew, Phys. Rev. 1986, 33, 8822-8824.

[300] F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297-3305.

[301] Gaussian 03, Revision E.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,

M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C. Burant,

J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.

Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,

M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P.

Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O.

Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G.

A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.

Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz,

Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P.

Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A.

Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C.

Gonzalez, and J. A. Pople, Gaussian, Inc., Wallingford CT, 2004.

[302] R. Ahlrichs, M. Baer, M. Haeser, H. Horn and C. Koelmel, Chem. Phys. Lett. 1989,

162, 165-169.

[303] R. Ahlrichs, Phys. Chem. Chem. Phys. 2004, 6, 5119-5121.

[304] K. Eichkorn, O. Treutler, H. Oehm, M. Haeser and R. Ahlrichs, Chem. Phys. Lett. 1995,

242, 652-660.

[305] K. Eichkorn, F. Weigend, O. Treutler and R. Ahlrichs, Theor. Chem. Acc. 1997, 97,

119-124.

[306] F. Weigend, Phys. Chem. Chem. Phys. 2006, 8, 1057-1065.

[307] A. E. Reed, L. A. Curtiss and F. Weinhold, Chem. Rev. 1988, 88, 899-926.

[308] A. E. Reed, R. B. Weinstock and F. Weinhold, J. Chem. Phys. 1985, 83, 735-746.

[309] R. F. W. Bader, Atoms in Molecules: A Quantum Theory, University of Oxford Press,

1990.

[310] AIMPAC: http://www.chemistry.mcmaster.ca/aimpac

[311] gOpenMol: http://www.csc.fi/english/pages/g0penMol

188 7. References

[312] ADF2009.01, SCM, Theoretical Chemistry, Vrije Universiteit, Amsterdam,

http://www.scm.com.

[313] J. G. Snijders, P. Vernooijs and E. J. Baerends, At. Data Nucl. Data Tables 1981, 26,

483-509.

[314] J. Krijn, E. J. Baerends, Internal Report: Fit Functions in the HFSMethod, Vrije

Universiteit Amsterdam, 1984.

[315] C. Chang, M. Pelissier and P. Durand, Phys. Scr. 1986, 34, 394-404.

[316] J. L. Heully, I. Lindgren, E. Lindroth, S. Lundqvist and A. M. Maartensson-Pendrill, J.

Phys. 1986, 19, 2799-2815.

[317] J. G. Snijders and A. J. Sadlej, Chem. Phys. Lett. 1996, 252, 51-61.

[318] E. van Lenthe, E. J. Baerends and J. G. Snijders, J. Chem. Phys. 1993, 99, 4597-4610.

[319] E. Van Lenthe, R. Van Leeuwen, E. J. Baerends and J. G. Snijders, Int. J. Quantum

Chem. 1996, 57, 281-293.

189 8. Supplement

8 Supplement

8.1 Important Crystallographic Data

1 2 3

Empirical formula C87H126Ga6Mo C32H66P4Ga2Mo C51H75F3S1O3Rh1Ga5

Molecular weight 1686.14 810.11 2648.45

Temperature (K) 113(2) 111(2) 113(2)

Wavelength Mo-Kα (Å) 0.71073 0.71073 0.71073

Crystal size (mm) 0.25 × 0.2 × 0.15 0.30 × 0.28 × 0.20 0.45 × 0.35 × 0.15

Crystal system, space group Triclinic, P-1 Orthorhombic, P212121 Triclinic, Pī

a (Å) 12.2106(9) 11.6503(2) 12.1447(3)

b (Å) 14.7410(12) 14.6989(4) 12.5678(3)

c (Å) 21.1396(17) 23.1578(5) 20.3119(5)

α (°) 79.893(7) 90 87.634(2)

β (°) 87.157(7) 90 81.140(2)

γ (°) 77.266(7) 90 64.001(2)

Cell volume (Å3) 3653.6(5) 3965.70(15) 2752.06(12)

Z 2 4 1

Density ρcalc. (g cm-3) 1.533 1.357 1.598

Absorption coefficient µ (mm-1) 2.391 1.841 2.796

F (000) 1752 1688 1345

Θ range for data collection (°) 2.61- 27.53 3.17 - 25.00 2.97 - 25.00

Index ranges

-15<=h<=15,

-19<=k<=19,

-27<=l<=27

-13<=h<=13,

-13<=k<=17,

-27<=l<=27

-14<=h<=14,

-14<=k<=14,

-24<=l<=24

Reflexions collected 69567 29247 30380

Reflexions unique 16576 [R(int) = 0.0984] 16576 [Rint = 0.0308] 9677 [Rint = 0.0329]

Refinement method Full-matrix least-squares

on F2

Full-matrix least-squares

on F2


on F2

Data/restraints/Parameters 16576 / 0 / 766 6978 / 81 / 399 9677 / 0 / 613

Absorption correction Empirical Semi-Empirical Empirical

Goodness-of-fit on F2 (GOF) 0.756 0.990 1.044

Final R indices [I>2σ(I)] R1 = 0.0429, wR2 = 0.0733 R1 = 0.0272, wR2 = 0.0603 R1 = 0.0302, wR2 = 0.0721

R indices (all data) R1 = 0.1231, wR2 = 0.0828 R1 = 0.0411, wR2 = 0.0637 R1 = 0.0389, wR2 = 0.0763

Largest difference peak and hole (e Å-3) 1.149 and -0.972 0.784 and -0.326 0.716 and -0.538

190 8. Supplement

5 6 7

Empirical formula C45H63F3S1O3N2Rh1Ga C60H82Cu2F6Ga2N4O6S2 C43H60Cu2F9Ga5O9S3


Temperature (K) 113(2) 110(2) 113(2)


Crystal size (mm) 0.41 × 0.28 × 0.14 0.30 × 0.25 × 0.20 0.45 × 0.35 × 0.15

Crystal system, space group Monoclinic, P21/n Triclinic, Pī Monoclinic, P21/n

a (Å) 11.5810(5) 12.0897(11) 14.2074(4)

b (Å) 21.9442(10) 12.4723(17) 21.4730(8)

c (Å) 18.6298(9) 13.1793(17) 20.2482(8)

α (°) 90 79.644(8) 90

β (°) 96.623(4) 71.134(10) 97.744(3)

γ (°) 90 83.296(10) 90

Cell volume (Å3) 4702.9(4) 1846.2(4) 6120.9(4)

Z 4 1 4



F (000) 1960 724 2928

Θ range for data collection (°) 2.87 - 25.00 3.33 - 25.00 3.02 - 27.67

Index ranges

-13<=h<=13,

-26<=k<=26,

-22<=l<=22

-14<=h<=6,

-14<=k<=14,

-10<=l<=14

14<=h<=6,

-14<=k<=14,

-10<=l<=14


Reflexions unique 8265 [Rint = 0.1104] 4666 [Rint = 0.0648] 14168 [Rint = 0.0685]


on F2


on F2


on F2


Absorption correction Empirical Empirical Empirical





191 8. Supplement

8 9 10 x 2 thf

Empirical formula C58H80Cu2F7Ga5O6S2 C24H32Fe2O12Zn2 C24H38Fe2O10Zn4 ⋅ 2 C4H8O


Temperature (K) 103(2) 108(2) 113(2)


Crystal size (mm) 0.41 × 0.28 × 0.14 0.35 × 0.23 × 0.19 0.22 × 0.17 × 0.06

Crystal system, space group Monoclinic, P21/n Monoclinic, C2/c Monoclinic, P21/c

a (Å) 11.6265(5) 14.7039(6) 20.416(5)

b (Å) 28.6899(17) 12.3522(5) 14.321(3)

c (Å) 19.9662(11) 31.5375(11) 14.392(3)

α (°) 90 90 90

β (°) 91.444(4) 93.005(3) 110.34(3)

γ (°) 90 90 90

Cell volume (Å3) 6657.9(6) 5720.1(4) 3945.4(16)

Z 4 8 4



F (000) 3136 3072 2056


Index ranges

14<=h<=6,

-14<=k<=14,

-10<=l<=14

-18<=h<=18,

-15<=k<=11,

-38<=l<=39

-24<=h<=24,

-17<=k<=17,

-17<=l<=17




on F2


on F2


on F2


Absorption correction Empirical Semi-Empirical Semi-Empirical





192 8. Supplement

11 12 x 2 toluene 13 x thf

Empirical formula C36H30Fe2N6O6Zn3 C36H40Co2N2O6Zn3 C34H46Co2O8Zn3 ⋅C4H8 O


Temperature (K) 113(2) 110(2) 113(2)


Crystal size (mm) 0.17 × 0.11 × 0.09 0.19 × 0.15 × 0.13 0.21 × 0.20 × 0.19

Crystal system, space group Orthorhombic, Pna21 Monoclinic, P21/n Orthorhombic, Pbcn

a (Å) 20.0442(8) 17.2422(18) 16.7762(6)

b (Å) 10.0158(4) 14.6947(13) 17.0143(6)

c (Å) 18.6317(7) 19.774(2) 14.2531(5)

α (°) 90 90 90

β (°) 90 113.370(12) 90

γ (°) 90 90 90

Cell volume (Å3) 3740.5(3) 4599.0(8) 4068.3(2)

Z 4 4 4



F (000) 1912 2248 1992

Θ range for data collection (°) 2.88 - 27.50 2.77 – 25.00 3.04 - 27.50

Index ranges

-25<=h<=25,

-12<=k<=13,

-23<=l<=23

-20<=h<=20,

-17<=k<=17,

-23<=l<=23

-21<=h<=21,

-22<=k<=22,

-18<=l<=18




on F2


on F2


on F2


Absorption correction Semi-Empirical Semi-Empirical Semi-Empirical





193 8. Supplement

14 15 16

Empirical formula C53H84Ga2Pd2Zn6 x C7H8 C60H90Zn6Ga2Pd x C6 H14 C80H120Zn12Pd x 2 C6H14


Temperature (K) 110(2) 294(2) 105(2)


Crystal size (mm) 0.14 × 0.09 × 0.08 0.40 × 0.31 × 0.30 0.15 × 0.08 × 0.02

Crystal system, space group Monoclinic, P21/n Monoclinic, P21/m Monoclinic, C2/c

a (Å) 12.7115(7) 14.5796(5) 31.608(11)

b (Å) 26.1406(16) 16.1503(5) 12.601(3)

c (Å) 19.2456(10) 14.7729(5) 25.942(8)

α (°) 90 90 90

β (°) 98.754(5) 98.472(3) 111.11(4)

γ (°) 90 90 90

Cell volume (Å3) 6320.6(6) 3440.6(2) 9638(5)

Z 4 2 4



F (000) 3144 1576 4424


Index ranges

-15<=h<=15,

-31<=k<=22,

-22<=l<=22

-9<=h<=17,

-10<=k<=19,

-17<=l<=17

-33<=h<=37,

-14<=k<=14,

-30<=l<=30




on F2


on F2


on F2







194 8. Supplement

17 19 20

Empirical formula C60H90Zn6Ga2Pt x C7H8 C100H150NiZn12 C40H60NiZn3


Temperature (K) 110(2) 112(2) 110(2)


Crystal size (mm) 0.30 × 0.25 × 0.20 0.28 × 0.17 × 0.14 0.23 × 0.20 × 0.17

Crystal system, space group Monoclinic, P21/m Orthorhombic, Pnna Triclinic, Pī

a (Å) 14.5430(13) 17.5266(11) 11.0041(7)

b (Å) 16.5819(12) 32.7410(16) 11.0846(8)

c (Å) 14.6737(11) 19.0262(11) 16.8992(12)

α (°) 90 90 94.025(6)

β (°) 97.289(8) 90 92.167(6)

γ (°) 90 90 113.805(7)

Cell volume (Å3) 3510.0(5) 10918.0(11) 1876.3(2)

Z 2 4 2



F (000) 1640 4552 836


Index ranges

-17<=h<=17,

-19<=k<=19,

-17<=l<=17

-20<=h<=20,

-38<=k<=38,

-22<=l<=22

-13<=h<=13,

-13<=k<=13,

-20<=l<=20




on F2


on F2


on F2







195 8. Supplement

21 22 23

Empirical formula C40H60PtZn3 C40H60PdZn3 C58H97N2PdZn7


Temperature (K) 110(2) 113(2) 113(2)


Crystal size (mm) 0.13 × 0.12 × 0.09 0.35 × 0.30 × 0.20 0.30 × 0.25 × 0.20

Crystal system, space group Triclinic, Pī Triclinic, Pī Monoclinic, P21/n

a (Å) 11.0373(5) 11.0117(11) 12.3193(4)

b (Å) 17.2860(9) 11.0198(9) 23.4719(9)

c (Å) 30.5800(15) 17.2737(13) 21.3947(8)

α (°) 89.836(4) 91.914(6) 90

β (°) 84.948(4) 92.360(7) 90.851(4)

γ (°) 87.393(4) 113.629(9) 90

Cell volume (Å3) 5805.7(5) 1915.9(3) 6185.8(4)

Z 6 2 4



F (000) 2808 872 2860


Index ranges

-13<=h<=13,

-20<=k<=15,

-36<=l<=36

14<=h<=6,

-14<=k<=14,

-10<=l<=14

-15<=h<=15,

-29<=k<=29,

-26<=l<=26




on F2


on F2


on F2


Absorption correction Semi-Empirical Semi-Empirical Empirical





196 8. Supplement

8.2 Overview of the Novel Compounds

[Mo(GaCp*)6] (1)

Discussion: Chap. 3.1, p. 41

Experimental Section: Chap. 6.2.2, p. 135

Crystallographic Data: Chap. 8.1, p. 189

[cis-Mo(GaCp*)2(PMe3)4] (2)




[Rh(GaCp*)5][CF3SO3] (3)




197 8. Supplement

[Rh(GaCp*)5][BArF] (4)



Crystallographic Data: no data available

[(coe)(tol)Rh{Ga(DDP)}(CF3SO3)] (5)




[{(DDP)GaCu(CF3SO3)}2] (6)




198 8. Supplement

[Cu2(GaCp*)4{Ga(CF3SO3)3}] (7)




[Cu2(GaCp*)3(μ-GaCp*)2][CF3SO3]2 (8)




[(CO)4Fe{μ2-Zn(thf)2}2Fe(CO)4] (9)

Discussion: Chap. 3.3.1, p. 64



199 8. Supplement

[{(CO)3Fe}2{μ2-Zn(thf)2}2(μ2-ZnMe)2 (10)




[(CO)3Fe{μ2-Zn(py)2}3Fe(CO)3] (11)




[{(CO)3Co}2{μ2-Zn(py)2}(ZnCp*)2] (12)




200 8. Supplement

[{(CO)3Co}2{μ2-Zn(thf)2}(µ-ZnCp*)2] (13)




[Pd2Zn6Ga2(Cp*)5(CH3)3] (14)




[Pd(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (15)




201 8. Supplement

[Pd(ZnCp*)4(ZnZnCp*)4] (16)




[Pt(GaCp*)2(ZnCp*)2(ZnZnCp*)2] (17)




[Pt(ZnCp*)4(ZnZnCp*)4] (18)



Crystallographic Data: no data available

202 8. Supplement

[Ni(ZnCp*)4(ZnZnCp*)4] (19)




[Cp*Ni(ZnCp*)3] (20)




[Cp*Pt(ZnCp*)3] (21)




203 8. Supplement

[Cp*Pd(ZnCp*)3] (22)




[Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}] (23)




204 8. Supplement

8.3 Publications

• T. Bollermann, K. Freitag, C. Gemel, M. Molon, R. W. Seidel, R. A. Fischer,

‘Reactivity of [Zn2Cp*2] towards Transition Metal Complexes: Synthesis and

Characterisation of [Cp*M(ZnCp*)3] (M = Ni, Pd, Pt)’, Organometallics 2011, 30

(15), 4123-4127.

• T. Bollermann, T. Cadenbach, C. Gemel, K. Freitag, M. Molon, V. Gwildies, R. A.

Fischer, ‘Homoleptic Hexa and Penta Gallylene Coordinated Complexes of

Molybdenum and Rhodium’, Inorg. Chem. 2011, 50(12), 5808-5814.

• T. Bollermann, K. Freitag, C. Gemel, R. W. Seidel, M. von Hopffgarten, G. Frenking,

R. A. Fischer, ‘The Reactivity of [Zn2Cp*2]: Trapping Monovalent .ZnZnCp* in the

Metal-Rich Compounds [(Pd,Pt)(GaCp*)a(ZnCp*)4-a(ZnZnCp*)4-a] (a=0, 2)’, Angew.

Chem. Int. Ed. 2011, 50(3), 772-776.

• M. Molon, T. Bollermann, C. Gemel, J. Schaumann, R. A. Fischer, ‘Mixed phosphine

and group-13 metal ligator complexes [(PR3)aM(ECp*)b] (M = Mo, Ni; E = Ga, Al; R

= Me, C6H5, cyclo-C6H11)’, Dalton Trans. 2011, DOI:10.1039/C1DT10583C.

• M. Molon, T. Cadenbach, T. Bollermann, C. Gemel, R. A. Fischer, ‘One electron

organozinc ligands in metal rich molecules by Ga/Zn exchange: from Cp*Rh(GaCp*)2

to Cp*Rh(ZnR)4 units’, Chem. Commun. 2010, 46(31), 5677-5679.

• M. Halbherr, T. Bollermann, C. Gemel, R. A. Fischer, ‘Selective Oxidative Cleavage

of Cp* from Coordinated GaCp*: Naked Ga+ in [GaNi(GaCp*)4]+ and [(µ2-

Ga)nM3(GaCp*)6]n+’, Angew. Chem. Int. Ed. 2010, 49(10), 1878-1881.

• T. Bollermann, G. Prabusankar, C. Gemel, R. W. Seidel, M. Winter, R. A. Fischer,

‘First Dinuclear Copper/Gallium Complexes: Supporting Cu0 and CuI Centres by

Low-Valent Organogallium Ligands’, Chem.-Eur. J. 2010, 16(29), 8846-8853.

205 8. Supplement

• T. Bollermann, T. Cadenbach, C. Gemel, M. von Hopffgarten, G. Frenking, R. A.

Fischer, ‘Molecular Alloys: Experimental and Theoretical Investigations on the

Substitution of Zinc by Cadmium and Mercury in the Homologous Series

[Mo(M'R)12] and [M(M'R)8] (M=Pd, Pt; M'=Zn, Cd, Hg)’, Chem.-Eur. J. 2010,

16(45), 13372-13384.

• T. Cadenbach, C. Gemel, T. Bollermann, R. A. Fischer, ‘Synthesis and crystal

structures of ruthenium and rhodium olefin complexes containing GaCp’, Inorg.

Chem. 2009, 48(11), 5021-5026.

• T. Cadenbach, T. Bollermann, C. Gemel, R. A. Fischer, ‘Synthesis and structure of

electron rich ruthenium polyhydride complexes and clusters containing AlCp* and

GaCp*’, Dalton Trans. 2009, 2, 322-329.

• T. Cadenbach, T. Bollermann, C. Gemel, M. Tombul, I. Fernandez, M. von

Hopffgarten, G. Frenking, R. A. Fischer, ‘Molecular alloys, linking organometallics

with intermetallic Hume-Rothery phases: the highly coordinated transition metal

compounds [M(ZnR)n] (n ≥ 8) containing organo-zinc ligands’, J. Am. Chem. Soc.

2009, 131(44), 16063-16077.

• T. Bollermann, A. Puls, C. Gemel, T. Cadenbach, R. A. Fischer, ‘Reactions of

cationic transition metal acetonitrile complexes [M(CH3CN)n]m+ with GaCp*: novel

gallium complexes of iron, cobalt, copper and silver’, Dalton Trans. 2009, 8, 1372-

1377.

• T. Cadenbach, C. Gemel, T. Bollermann, I. Fernandez, G. Frenking, R. A. Fischer,

‘Organometallic chemistry of Ga+: Formation of an Unusual Gallium Dimer in the

Coordination Sphere of Ruthenium’, Chem.-Eur. J. 2008, 14(34), 10789-10796.

• T. Cadenbach, T. Bollermann, C. Gemel, I. Fernandez, M. von Hopffgarten, G.

Frenking, R. A. Fischer, ‘Twelve one-electron ligands coordinating one metal center:

structure and bonding of [Mo(ZnCH3)9(ZnCp*)3]’, Angew. Chem. Int. Ed., 2008,

47(47), 9150-9154.

206 8. Supplement

Submitted

• T. Bollermann, K. Freitag, C. Gemel, M. Molon, R. W. Seidel, M. von Hopffgarten,

P. Jerabek, G. Frenking, R. A. Fischer, ‘The rich chemistry of [Zn2Cp*2]: Trapping

three different types of zinc ligands in the PdZn7 complex

[Pd(ZnCp*)4(ZnMe)2{Zn(tmeda)}]’, Inorg. Chem. 2011, submitted.

• T. Bollermann, I. Schwedler, M. Molon, K. Freitag, C. Gemel, R. W. Seidel, R. A.

Fischer, ‘Zinc-rich Compounds of Iron and Cobalt: Formation of [Fe2Znn] (n = 2-4)

and [Co2Zn3] Cores’, Dalton Trans. 2011, submitted.

• T. Bollermann, M. Molon, C. Gemel, K. Freitag, R. W. Seidel, M. von Hopffgarten,

P. Jerabek, G. Frenking, R. A. Fischer, ‘On the way to oligonuclear molecular models

of intermetallic phases: A case study on [Pd2Zn6Ga2(Cp*)5(CH3)3]’, J. Am. Chem.

Soc. 2011, submitted.

• T. Bollermann, C. Gemel, R. A. Fischer, ‘Organozinc Ligands in Transition Metal

Chemistry’, Coord. Chem. Rev. 2011, submitted.

• S. González-Gallardo, T. Bollermann, R. A. Fischer, R. Murugavel, ‘Cyclopentadiene

Based Low-valent Group 13 Metal Compounds: Ligands in Coordination Chemistry

and Link between Metal Rich Molecules and Intermetallic Materials’, Chem. Rev.

2011, submitted.

Manuscripts in preparation

• T. Bollermann, K. Freitag, C. Gemel, R. W. Seidel, M. von Hopffgarten, P. Jerabek,

G. Frenking, R. A. Fischer, ‘Coordination Chemistry of [Zn2Cp*2] to Transition

Metals: Experimental and Theoretical Investigations on the Formation of {ZnZnCp*}

Containing Compounds [M(ZnCp*)4(ZnZnCp*)4] (M = Ni, Pd, Pt)’, Chem.-Eur. J.

2011, manuscript in preparation.

• A. Puls, T. Bollermann, C. Gemel, R. A. Fischer, ‘Molecular Building Blocks for

Nanostructures: Preparation of [Mo(AuPPh3)8(AuCl)2(EMe)2] (E = Ga/Zn),

[Ni(AlCp*)2(AuPPh3)4(AuCl)3] and [Pt(AuPMe3)11(AuCl)][GaCl4]3’, Nature

Chemistry 2011, manuscript in preparation.

207 8. Supplement

8.4 Conferences: Oral Presentations and Poster Presentations

11. – 14.04. 2011 T. Bollermann, T. Cadenbach, C. Gemel, M. Molon, K.

Freitag, A. Puls, M. von Hopffgarten, G. Frenking and R.

A. Fischer, ‘Synthesis and Characterisation of Metal

Rich Molecules – Bridging the Gap between Molecular

Compounds and Intermetallic Phases’, 1st EuCheMS

Inorganic Chemistry Conference (EICC), Manchester,

UK (Oral)



A. Fischer, ‘Coordination Chemistry of [Zn2Cp*2]:

Formation of Novel {ZnZnCp*} Ligands Coordinating to

Transition Metals’, 1st EuCheMS Inorganic Chemistry

Conference (EICC), Manchester, UK (Poster)



A. Fischer, ‘A Novel Family of Metal Rich Molecules -

Embryonic Alloys Wrapped into an All-Hydrocarbon

Shell’, 24th International Conference on Organometallic

Chemistry (ICOMC), Taipeh, Taiwan (Oral)

18. – 23.07. 2010 T. Bollermann, K. Freitag, C. Gemel, M. von

Hopffgarten, G. Frenking and R. A. Fischer, ‘Formation

of Novel Metal Rich Coordination Compounds

Containing [ZnZnCp*] as One-Electron Ligands’, 24th

International Conference on Organometallic Chemistry

(ICOMC), Taipeh, Taiwan (Poster)

18.02. 2010 T. Bollermann and R. A. Fischer, ‘Formation of metal-

rich cluster compounds as molecular example systems

for intermetallic phases’, VCI Fellows Meeting 2010,

Heinrich Heine University Düsseldorf (Oral)

208 8. Supplement

06.11. 2009 T. Bollermann, C. Gemel, T. Cadenbach and R. A.

Fischer, ‘Formation of metal-rich cluster compounds as

molecular example systems for intermetallic phases’,

Section Days, Ruhr University Bochum (Poster)

209 8. Supplement

8.5 Curriculum Vitae

Personal Data

First name: Timo

Surname: Bollermann

Private address: Semperstr. 117, App. 18

44801 Bochum

Germany

Phone number: +49-234-7089093

E-Mail adress: [email protected]

Date of birth: 08.07.1982

Family status: Unmarried

Nationality: German

General Teaching and Supervision Experiences

2011/2012 Co-Supervisor of the lecture Organometallics & Materials,

Ruhr University Bochum

2008 – 2011 Supervision of several (advanced) practical courses, Ruhr

University Bochum

2008 – 2011 Supervision of several Bachelor and Master thesis, Ruhr

University Bochum

2006/2007 Supervision of the Chemical practical course for medical

students, Ruhr University Bochum

Professional Training and Research Visits

05/2010 – 06/2010 Metal Vapour Synthesis (MVS), Prof. Dr. U. Zenneck, Friedrich

Alexander University Erlangen-Nürnberg

01/2010 – 02/2010 Metal Vapour Synthesis (MVS), Prof. Dr. U. Zenneck, Friedrich

Alexander University Erlangen-Nürnberg

Since 06/2008 Research assistant, Chair of Inorganic Chemistry II, Ruhr

University Bochum

07/2007 – 05/2008 Student research assistant, Chair of Inorganic Chemistry II, Ruhr

University Bochum; Task area: Organometallic synthesis

210 8. Supplement

10/2006 – 06/2007 Student research assistant, Chair of Inorganic Chemistry I, SFB

558, Ruhr University Bochum; Task area: Synthesis of

modified catalysts for methanol oxidation

Academic and Vocational Education

10/2006 – 03/2008 Master studies (Chemistry), Ruhr University Bochum, main

focus: Inorganic Chemistry, Final Degree: Master of Science

(Chemistry), Final grade: 1.2

10/2003 – 09/2006 Bachelor studies (Chemistry), Ruhr University Bochum, Final

Degree: Bachelor of Science (Chemistry), Final grade: 2.0

09/2003 – 10/2003 Pre-university studies (mathematics and chemistry)

06/2003 – 07/2003 Employee in the educational service of the Westphalian Institute

for Child and Adolescent Psychiatry and Psychotherapy, Hamm

08/2002 – 05/2003 Civial Service: Westphalian Institute for Child and Adolescent

Psychiatry and Psychotherapy, Hamm

08/1993 – 06/2002 Secondary School Hammonense Hamm

Final Degree: University-Entrance Diploma

08/1989 – 07/1993 Primary School Hellweg, Bönen

Fellowships

04/2011 Travel stipend of the Ruth and Gert Massenberg Foundation,

Germany (1st EICC, Manchester, UK)

03/2009 – 12/2010 PhD fellowship of the Fonds der Chemischen Industrie,

Germany

Since 10/2008 Member of the Ruhr University Research School (GSC 98,

Exzellenzinitiative)

Further Educations and Workshops

26. – 28.10. 2010 Advanced software training, Oxford Diffraction, Wroclaw,

Poland

23. – 25.09. 2009 Scientific Writing: Getting Research Published, Ruhr University

Bochum

211 8. Supplement

19.06. 2009 Science College; Simulation and Modelling: Chances and

Limitations of Analyses and Prognoses, Ruhr University

Bochum

30.03. – 01.04. 2009 Scientific Presentation, Ruhr University Bochum

Language Skills

• German: Mother tongue

• Englisch: Fluent in spoken and written

Memberships

Since 04/2004 Member of the Gesellschaft Deutscher Chemiker (GDCh)

intermetallic chemistry : molecular compounds at the

Documents