Developing Terminal Phosphinidene Complexes

Developing Terminal Phosphinidene Complexes:

The Quest for Applicability Continues

Halil Aktaş

2009

The investigations described in this thesis were carried out in the Division of

Organic Chemistry, Department of Chemistry and Pharmaceutical Sciences,

Faculty of Sciences, Vrije Universiteit, De Boelelaan 1083, NL-1081 HV

Amsterdam, The Netherlands.

This research was partly supported by a TOP grant to KL (no. 700.52.308) of

the Council for Chemical Sciences of the Netherlands Organization for

Scientific Research (NWO/CW).

Cover and chapter illustrations by Okan Akın

Printed by Wöhrmann Print Service

ISBN: 978-90-9024-839-4

© Halil Aktaş, Zaandam, the Netherlands, 2009

VRIJE UNIVERSITEIT

Developing Terminal Phosphinidene Complexes:

The Quest for Applicability Continues

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad Doctor aan

de Vrije Universiteit Amsterdam,

op gezag van de rector magnificus

prof.dr. L.M. Bouter,

in het openbaar te verdedigen

ten overstaan van de promotiecommissie

van de faculteit der Exacte Wetenschappen

op dinsdag 8 december 2009 om 13.45 uur

in de aula van de universiteit,

De Boelelaan 1105

door

Halil Aktaş

geboren te Karşıyaka, Turkije

promotor:

copromotoren:

prof.dr. K. Lammertsma

dr. A.W. Ehlers

dr. J.C. Slootweg

anneme ve babama

Reading Committee: prof.dr. F.M. Bickelhaupt

dr. B. de Bruin

prof.dr. C.J. Elsevier

prof.dr. E. Hey-Hawkins

dr. C. Müller

Contents

Chapter 1 Nucleophilic Phosphinidene Complexes:

Access and Applicability 11

1.1 General introduction 12

1.2 Transition metal ligation 13

1.3 Generating nucleophilic phosphinidene complexes 14

1.3.1 Salt Metathesis/elimination 14

1.3.2 Insertion/elimination 20

1.3.3 α-Hydrogen migration 22

1.3.4 Oxidation/deprotonation 24

1.3.5 Phosphinidene group-transfer 25

1.3.6 Dehydrohalogenation/addition 25

1.4 Reactivity of nucleophilic phosphinidene complexes 29

1.4.1 Reactive 16-electron intermediates 29

1.4.2 RP-transfer 34

1.4.3 Insertion into the M=P bond 37

1.4.4 Cycloaddition to the M=P bond 38

1.5 Conclusion 41

1.6 Outline of this thesis 42

Chapter 2 N-Heterocyclic Carbene Functionalized Ruthenium

Phosphinidenes: What a Difference a Twist Makes 49

2.1 Introduction 50

2.2 NHC functionalization 50

2.3 Theoretical calculations on [(Ring)M(NHC)=PH] 52

2.4 Reactivity 54

2.5 Transient species 55

2.6 Conclusion 56

2.7 Computational Section 56

2.8 Experimental Section 58

Chapter 3 N-Heterocyclic Carbene Functionalized Group 7–9

Transition Metal Phosphinidene Complexes 69

3.1 Introduction 70

3.2 Results and Discussion 72

3.2.1 Synthesis of NHC-functionalized group 8 phosphinidenes 72

3.2.2 Synthesis of NHC-functionalized group 9 phosphinidenes 74

3.3 Theoretical calculations on [(Ring)M(NHC)=PH] 75

3.3.1 Geometries 76

3.3.2 Energy Decomposition Analysis 77

3.4 Conclusion 82



Chapter 4 Iridium Phosphinidene Complexes: A Comparison with

Iridium Imido Complexes in Their Reaction with Isocyanides 95

4.1 Introduction 96

4.2 Results and Discussion 97

4.3 Transient species 100

4.4 Mechanism 102

4.4.1 Imido complex 102

4.4.2 Phosphinidene complex 103

4.5 Conclusion 105



Chapter 5 ηηηη3-Diphosphavinylcarbene:

A P2 Analogue of the Dötz Intermediate 119

5.1 Introduction 120

5.2 Synthesis of η3-diphosphavinylcarbenes 121

5.3 Synthesis of 1,3-diphospha-3H-indenes 123

5.4 Mechanism 124

5.5 Reactivity 126

5.6 Conclusion 128



Appendix 1 141

Appendix 2 143

Samenvatting 147

Curriculum Vitae 153

List of publications 155

Dankwoord 157

Nucleophilic Phosphinidene Complexes:

Access and Applicability

Halil Aktas,† J. Chris Slootweg,† and Koop Lammertsma*,†

Department of Chemistry and Pharmaceutical Sciences, Vrije Universiteit Amsterdam, De Boelelaan

1083, 1081 HV Amsterdam, The Netherlands

Angew. Chem. 2009, accepted

Abstract:Abstract:Abstract:Abstract: The topic of the research described in this thesis concerns the syntheses, properties, and

reactivities of nucleophilic phosphinidene complexes LnM=P–R. Emphasis is placed on the electronic

tuning of this emerging class of phosphorus reagents, using different ancillary ligands and

coordinatively unsaturated transition metal moieties. The difference in applicability of the

established stable 18-electron and transient 16-electron phosphinidenes is addressed.

Chapter 1Chapter 1Chapter 1Chapter 1

Chapter 1

12

1.1 General introduction

Phosphinidenes[1] (phosphanylidenes,[2] R–P:, A; Scheme 1) are the

phosphorus analogues of carbenes (R2C:)[3] and nitrenes (R–N:).[4]

These exceedingly reactive phosphorus species have been detected

only in the gas phase (MS) and in glassy and cryogenic matrices (EPR,

IR, UV).[5] Their chemistry remains to be fully explored,[6] which

distinguishes them from the explosive growth the carbenes saw.

Terminal transition metal-complexed phosphinidenes LnM=P–R (B;

Scheme 1), which are the phosphorus analogues of the well-

established carbene complexes, appear, however, to be valuable

synthons with rapidly expanding chemistries.[7-10] Enabling were the

discoveries in the 1980s of a transient electrophilic species, (OC)5W=P–

Ph, by the group of Mathey[11] and that of an isolable nucleophilic

phosphinidene complex, Cp2W=P–Mes*, by Lappert’s group.[12]

Illustrative of the ensuing rapid progress in this field are two reviews by

Cowley, one in 1988, entitled “The quest for terminal phosphinidene

complexes”[13a] and the other in 1997 “Terminal phosphinidene and

heavier congeneric complexes. The quest is over”.[13b] Much has

happened since as is emphasized in the present thesis, which focuses

on neutral nucleophilic η1-phosphinidene complexes. Hitherto, these

compounds were considered to have limited applicability, sharply

contrasting the electrophilic ones,[1,8] but their potential is far greater

than believed.

PR PR

MLn

A B

Scheme 1. ‘Free’ (A) and η1–complexed phosphinidene (B).

Nucleophilic Phosphinidene Complexes: Access and Applicability

13

1.2 Transition metal ligation

It is important to recognize the impact of a transition metal group on

the phosphinidene R–P. Terminal complexed phosphinidenes strongly

prefer a singlet ground state, contrasting the uncomplexed species,

and are either nucleophilic (Schrock-type)[14] or electrophilic (Fischer-

type)[15] at the phosphorus atom. An extensive density functional

study[16] on LnM=PH (M = Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W, Fe, Ru, Os, Co,

Rh, Ir and L = CO, PH3, Cp) revealed that the philicity and chemical

reactivity of the phosphinidene complex is influenced mainly by the

metal’s spectator ligand L. Those with strong σ-donor capabilities

increase the electron density on the phosphorus atom, enhancing its

nucleophilicity. Conversely, spectator ligands with strong π-acceptor

capabilities lower the charge concentration on P, causing

electrophilic behavior. Illustrative is the difference between

electrophilic (OC)4Fe=PH and nucleophilic Cp2Cr=PH, which is a

reflection of the different magnitude in which charge is transferred

from the frontier orbital energies of the transition metal fragments to

the phosphorus atom. Indeed all reported phosphinidene complexes

with only CO ligands, like (OC)nM=P–R (M = W, Mo, Cr, n = 5; M = Fe, n

= 4), are known to be transient electrophiles, generated in-situ from

appropriate precursors. Their insertion into σ-bonds, addition to π-

bonds, and coordination to lone pairs is well documented and

reviewed.[1,8] More diversity in ancillary ligands is available in cationic

complexes [LnM=P–R]+ of which stable ones[17] with limited reactivity

have been reported.[17e-f,18] The diversity in ligands and transition

metals is by far the largest for the nucleophilic phosphinidene

complexes, the topic of this chapter. Before advancing, it must be

noted that the M=P bond of all LnM=P–R complexes have genuine

double bond character as established by a DFT bond energy analysis

Chapter 1

14

from which quantitative σ- and π-bond strengths could be

determined.[16] The M=P interaction increases on going from the first to

the second- and third-row transition metals. This chapter emphasizes

first the different methodologies to access these entities and then

addresses their chemical applicability.

1.3 Generating nucleophilic phosphinidene complexes

1.3.1 Salt Metathesis/elimination

The most common route toward nucleophilic phosphinidene

complexes is by combining a metal complex with a halogenated

species under expulsion of M+X–. There are two possibilities, reacting a

Li+ metallocene hydride with a chlorophosphine and conversely

reacting a transition metal halide complex with a lithium phosphide,

but variations on this salt metathesis/elimination theme exist.

Scheme 2. Salt metathesis with dichlorophosphines.

The first stable 18-electron phosphinidene complexes were

synthesized by Lappert and co-workers[12] who reacted lithium

metallocene hydride [Cp2MHLi]4 with dichlorophosphine RPCl2 (R =

Mes*, (Me3Si)2CH) to obtain Cp2M=PR 1a,b (M = Mo (a), W (b)) as

stable, red crystalline material (Scheme 2).[12] The low-field 31P NMR

chemical shifts (R = Mes*; Mo (1a): (δ) 799.5 and W (1b): 661.1) proved

to be characteristic for terminal phosphinidene complexes.[12] The X-

ray crystal structures show a M=PMes* double bond of 2.370(2) Å for

the molybdenum complex[12] and of 2.349(5) Å for the tungsten


15

complex[12] with bent M–P–Mes* angles of 115.8(2)° and 114.8(5)°,

respectively.

The group of Stephan synthesized the first early-transition metal

complex, zirconium phosphinidene Cp2(Me3P)Zr=P–Mes* (4), by salt

metathesis of zirconocene dichloride and lithium supermesityl-

phosphide[19] and from zirconium phosphide [Cp2(Cl)ZrP(H)Mes*]

using an alkali metal base,[20] both in the presence of PMe3.[19,20] The

X-ray crystal structure reveals a short Zr=PMes* double bond of

2.505(4) Å, a Zr–P–Mes* angle of 101.4(1)°, and a long Zr–PMe3 single

bond of 2.741(5) Å, indicating weak bonding of the ancillary ligand.

The 31P NMR chemical shift of the phosphinidene is observed at 792.7

ppm. A superior route with a near quantitative yield is the reaction of

chloro-bis(η5–cyclopentadienyl)methylzirconium (2) with lithium

supermesitylphosphide followed by loss of methane from the incipient

3 in the presence of PMe3 (Scheme 3).[19b] Using a similar procedure,

Protasiewicz and co-workers reported the related phosphinidene

complex Cp2(Me3P)Zr=P–Dmp 5 (Dmp = 2,6-Mes2C6H3, Scheme 3).[21]

Salt metathesis and Lewis base stabilization also enabled the synthesis

of hafnium phosphinidene Cp2(Me3P)Hf=P–Mes* 6,[22] the terminally

bonded phosphanylphosphinidene complex Cp2(PhMe2P)Zr=P–Pt–Bu2

7,[23] and the uranium complex Cp*2(Me3PO)U=P–Mes* 8 (Scheme

3),[24] all of which are bent according to their 31P NMR spectroscopic

data (δ 31P 671 (6), 728 (7), 71 ppm (8)) and solid state structures (M–

P–C/P 115.53(16) (7), 143.7(3) Å (8)).

Salt metathesis was shown by the Lammertsma group to be equally

applicable to the late transition metal iridium, tolerating different

ancillary ligands, such as PPh3 and the N-heterocyclic carbene

IiPr2Me2 together with a Cp* ligand.[25]

Chapter 1

16

Zr P

Mes*

4

PMe3

Cp2ZrLiPHMes*

2

Me

Cl

Zr PH

Mes*

Me

3

PMe3

- CH4

Zr P

Dmp

5

PMe3

Hf P

Mes*

6

PMe3

Zr P

P

7

PPhMe2

8

tBu

tBu

U P

Mes*

O=PMe3

Scheme 3. Metal-complexed phosphinidenes synthesized using salt metathesis and

Lewis base stabilization.

Reacting iridium dichloride complex 9 (a: PPh3; b: IiPr2Me2) with

LiPHMes* provided 10a,b (Scheme 4).[25] NHC-ligated iridium

phosphinidene complex 10b, characterized by an X-ray crystal

structure, strongly resembles that of phosphane complex 10a, both

showing the expected bending (Ir–P–Mes*: 113.73(7)º 10a; 110.76(6)º

10b) for a phosphinidene complex with typical M=P double bonding

(10a: Ir=P 2.2121(5) Å; 10b: Ir=P 2.1959(5) Å).[25] The difference in their 31P NMR resonances (10a: 686.6; 10b: 560.0 ppm) is caused by the

strong σ−donor and moderate π−acceptor capabilities of the NHC

ligand rather than by geometrical differences.[25]

N NN NIr P

Mes*

9

IrCl22 LiPHMes*

- H2PMes*- 2 LiClL

10

L

a L = PPh3b L = IiPr2Me2

IiPr2Me2

Scheme 4. Phosphane (10a) and N-heterocyclic carbene (10b) functionalized iridium

phosphinidenes.


17

A series of tantalum phosphinidenes [N3N]Ta≡P–R 12 ([N3N] =

(Me3SiNCH2CH2)3N; R = t−Bu, Cy, Ph) was reported by Schrock and co-

workers who condensed tantalum dichloride complex 11 with lithium

phosphides (Scheme 5).[26] The large tetradentate triamidoamine

ligand [N3N] with trimethylsilyl groups ensures effective stabilization of

the nucleophilic phosphinidene unit, but narrows the space available

to it. As a result, the Ta–P–R geometry is almost linear (Cy: 170.9º)

enforcing both Ta≡P pseudo-triple bonding (Cy: 2.145(7) Å) and a

high field 31P NMR chemical shift for the phosphinidene (175.1–227.3

ppm). The mechanism by which the phosphinidene complex is

generated is not clear. The Ta≡P multiple bond may be formed via

dehydrohalogenation with a second phosphide acting as base,

whereas a proposed alternative path uses α−proton abstraction from

the tantalum bisphosphide.

-2 LiCl, - RPH2

Ta

N

NN

NMe3Si

ClCl SiMe3Me3Si

Ta

N

NN

NMe3Si

PSiMe3Me3Si

R = t-Bu, Cy, Ph

2 LiPHR

11 12

R

R1 = Me, Bu, SiMe3, SiMe2Ph

for R = Ph

Ta

N

NN

NMe3Si

PLi SiMe3Me3Si

Ta

N

NN

NMe3Si

P SiMe3Me3Si

14

R1 -PhLi

R1X

-LiX

Li0

13 Scheme 5. Linear tantalum phosphinidenes resulting from metathesis and P–C bond

cleavage.

Interestingly, the phosphinidene substituent P–R of 12 is

exchangeable. Reaction of the phenyl derivative with lithium

Chapter 1

18

afforded terminal phosphido complex {[N3N]Ta≡P}– 13 (Scheme 5),[27]

which has a low−field 31P NMR resonance at 575 ppm in concurrence

with a phosphide complex. Subsequent reaction at –35 °C with

organic halides afforded tantalum phosphinidene complexes

[N3N]Ta=P–R 14 (R = Me, n-Bu, SiMe3, SiMe2Ph).

Ar

N

N

Ar

Ti

CH3

CH3

Ar

N

N

Ar

Ti

CH3

CH3

Ar

N

N

Ar

Ti

P

CH3LiPHIsAgOTf

- Ag0OTf

- LiOTf- CH4

1715 16

B(C6F5)3

Ar

N

N

Ar

Ti

P

CB(C6F5)3

H

HH

18

Ar = 2,6-iPr2C6H3Is = 2,4,6-iPr3C6H2

Is

Is

Scheme 6. Titanium phosphinidene formed by salt metathesis followed by α-

hydrogen abstraction and methide elimination.

Mindiola’s group used the sterically hindered β-diketiminate ligand

to react a titanium complex with a lithium phosphide.[28] The process

starts by one electron oxidation of titanium dialkyl complex 15 with

AgOTf, followed by reaction of 16 with LiPHIs (Is = 2,4,6-iPr3C6H2),

presumably giving putative titanium phosphide

(tBunacnac)Ti(Me)2−PHIs and on loss of methane phosphinidene

complex (tBunacnac)(Me)Ti=P–Is 17 (nacnac– =

[Ar]NC(tBu)CHC(tBu)N[Ar]) (Scheme 6). The diagnostic 31P NMR

chemical shift at 231.5 ppm, the large Ti–P–Is angle of 159.95(7)º, and

short Ti=P bond of 2.1644(7) Å reveal a pseudo-linear titanium

phosphinidene complex. Treatment with tris(pentafluorophenyl)-

borane caused methide abstraction to yield the terminal

phosphinidene zwitterion (tBunacnac)Ti≡P–Is{H3CB(C6F5)3} 18 of which


19

the X-ray crystal structure showed a short Ti=P bond (2.1512(4) Å), a

linear Ti–P–Is unit (176.03(5)º), and an essentially departed methide

group (Ti–CH3 2.405(3) Å).

Scheme 7. Niobium phosphinidenes generated via P4 activation.

Another protocol using a transition metal-complexed phosphide

was developed by Cummins and Figueroa. With niobaziridine-hydride

complex [Nb(H)(η2–tBu(H)C=NAr(N[Np]Ar)2] 19 (Np = neopentyl, Ar =

3,5−Me2C6H3)[29a] they activated white phosphorus (P4) to form

bridged diphosphide complex [(µ2:η2,η2–P2){Nb(N[Np]Ar)3}2] 20, which

on treatment with sodium amalgam gave monomeric 21 (Scheme 7).

The formation of this terminal phosphide, having a 31P NMR chemical

shift of 1010 ppm (!), was supported by an X-ray crystal structure that

confirmed its anion-cation separation. Reaction with organic halides

resulted in niobium phosphinidene complexes (Ar[Np]N)3Nb=P–R 22 (R

Chapter 1

20

= SiMe3, SnMe3, PPh2, Pt–Bu2).[29b] The X-ray crystal structure for the

SnMe3 derivative reveals an elongated Nb=P bond (2.2731(8) Å) and

a P−Sn single bond with a length (2.4778(8) Å) that matches the sum

of the covalent radii of phosphorus and tin. The 31P NMR resonances

(401.3−607.0) of the niobium complexes 22 are in accord with a bent

phosphinidene.

1.3.2 Insertion/elimination

Inserting an electron-deficient organometallic fragment into a (R)P=X

(X = CO, CN−Ph) bond is an alternative route to phosphinidene

complexes. Cowley and co-workers[30] prepared tungsten

phosphinidene (MePh2P)2Cl2W(CO)≡P–Mes* 24 by reacting the 16-

electron tetraphosphine complex (MePh2P)4Cl2W 23 with

phosphaketene (Mes*)P=C=O under elimination of two equivalents of

phosphine (Scheme 8). In the phosphinidene product the ketene’s RP

(axial) and CO (equatorial) moieties end up in a syn fashion, but the

mechanism of formation is not known. The short W≡P distance of

2.169(1) Å suggest triple bonding. The large W–P–Mes* angle of

168.2(2)º) is also reflected in the upfield 31P NMR chemical shift of

193.0 ppm. The similar reaction with (Mes*)P=C=N−Ph likely gives the

thermally unstable complex (MePh2P)2Cl2W(C=N−Ph)≡P–Mes*.

24

PMes*

- 2 eq. PMePh2

W

P

Mes*

(Ph2MeP)4Cl2WCO

Cl

PMePh2Cl

CO

Ph2MeP

23

55%

Scheme 8. Tungsten phosphinidene by P=C double bond cleavage.

Another example concerns the insertion into a P−H bond. Oxidative

addition of sterically unhindered phenylphosphine to the ‘electron-


21

poor’ tris-siloxy tantalum complex 25 (Scheme 9) reportedly gives

intermediate phosphide 26 and on 1,2–H2 elimination tantalum

phosphinidene complex (t–Bu3SiO)3Ta=P–Ph 27.[31] The large siloxy

groups ensure kinetic stabilization of the bent phosphinidene. Its X-ray

crystal structure shows a short Ta=P double bond of 2.317(4) Å and

bent Ta–P–Ph moiety of 110.2(4)°.

The preference of a bent over a linear phosphinidene complex was

suggested to originate from O(pπ)–Ta(dπ) backbonding, which

prevents the formation of an otherwise more favorable P(pπ)–Ta(dπ)

interaction.

PTaPh

27

(t-Bu3SiO)3Ta

25

t-Bu3SiO

t-Bu3SiOt-Bu3SiO

PHTa

Pht-Bu3SiO

t-Bu3SiOt-Bu3SiO H

26

H2PPh C6H6

- H2

Scheme 9. Tantalum phosphinidene formed by P–H bond cleavage and subsequent

1,2–H2 elimination.

In situ generation of a transient organometallic precursor is also a

viable option. For example, compound 28 undergoes reductive

elimination and reacts with one equivalent of the primary phosphine

Mes*PH2 in the presence of a threefold excess of PMe3 to afford the

isolable 29 (Scheme 10).[32]

3 PMe3 , H2PMes*Ti

N

tBu3P

Cp

TiN

tBu3P

Cp PMe3

P Mes*

28 29 Scheme 10. Titanium phosphinidene formed by reductive elimination.

Although the structure of 29 could not be confirmed

crystallography, its 31P NMR resonances observed at 769.9, 35.3 and –

Chapter 1

22

10.3 ppm are indicative for the terminal Ti–phosphinidene fragment,

the phosphinimide ligand, and the coordinated PMe3, respectively.

Additional 1H and 13C{1H} spectra were consistent with this formulation.

1.3.3 αααα-Hydrogen migration

α-Hydrogen migration of the initial salt metathesis product is another

route to phosphinidene complexes. The first spectroscopic evidence

for such a process was reported by Niecke’s group for amino-

substituted complex Cp*2M=P–N(H)Mes* 32a,b (M = Mo (a), W (b);

Scheme 11a),[33] which is similar to the complex Cp2M=P–Mes* 1a,b

reported by Lappert.[12] Reaction of metal hydride 30a,b with chloro-

iminophosphine ClP=NMes* is believed to give intermediate 31,

based on the observed 31P NMR resonance at 754 ppm at –40°C for

the tungsten complex. Above this temperature the metal hydride

presumably undergoes a 1,3–hydrogen shift to yield phosphinidene

complex 32, which was characterized by 31P NMR spectroscopy (Mo:

770, W: 663 ppm), but could not be isolated.

A 1,3-shift of an α-hydrogen was used by Mindiola’s group to

prepare the titanium and vanadium phosphinidene complexes

(nacnac)(CH2t–Bu)M=P–R 35a,b (M = Ti (a), V (b); Scheme 11b). Salt

metathesis of titanium alkylidene 33a with LiPHR (R = Cy, Is, Mes*) at

low temperature gave putative neopentylidene-phosphide 34a that

underwent α-hydrogen migration to give phosphinidene

(nacnac)(CH2tBu)Ti=P–R 35a, (Scheme 11b).[34a] The Mes* derivatives

has a short Ti≡P pseudo-triple bond (2.1831(4) Å) and a pseudo-linear

Ti–P–Mes* unit (164.44(5)º),[34b] while solution spectra reveal two 31P

NMR resonances (242 and 216 ppm), suggesting the presence of two

conformers.[34a] Paramagnetic vanadium complexes (nacnac)(CH2t–

Bu)V=P–R 35b (R = Is, Mes*) were synthesized analogously (Scheme


23

11b).[34c] The X-ray crystal structures reveal a distorted tetrahedral

geometry at vanadium, a V=PR bond (R = Is, 2.174(4) Å; Mes*,

2.1602(6) Å) that is considerably shorter than those for the four-

coordinate vanadium phosphides, and a V–P–Cipso angle that

depends on the P-substituent (Is, 136.6(5)°; Mes*, 153.28(6)°).

30

PM

N

M = Mo (a), W (b)

ClP NMes*

31 32

LiP(H)R

M = Ti (a), V (b); Ar = 2,6-iPr2C6H3; R = Is, Mes*; X = I, OTf

68-79%

- LiCl

- LiX

Cp*

Cp*P

M

NMes*

Cp*

Cp*

H

Li

M

Cp*

Cp*

H

1,3

H

Mes*

Ar

N

N

Ar

M

HC

XAr

N

N

Ar

M

HC

P(H)RAr

N

N

Ar

M

H2C

P R

34 3533

PiPr2

PiPr2

Ti

H2C

P Is

36

N

PWMes*

Cp

CO

38

OC H

KOt-BuPWMes*

Cp

OCOC

H

39

79%PWMes*

Cp

OC

37

OCH

Et3N

H -Et3N.HClCl

a)

b)

c)

d)

Scheme 11. Phosphinidene complexes formed by α–hydrogen migration.

The generality of the α-hydrogen migration[28] was further

demonstrated by the synthesis of titanium(IV)phosphinidene 36

(Scheme 11c), bearing the PNP-pincer ligand N[2-P(CHMe2)2-4-

methylphenyl]2–; imide and alkylidene functionalities can be obtained

Chapter 1

24

by the same approach.[34d] Spectroscopic evidence for a base-

induced 1,2-H shift leading to a phosphinidene complex was

provided by the group of Malish (Scheme 11d).[35]

Dehydrohalogenation of phosphine complex 37 with triethylamine

gave phosphenium complex 38, which underwent a 1,2-H shift in the

presence of KOtBu, likely by a deprotonation-reprotonation

sequence, to yield the evidently preferred phosphinidene complex

Cp(CO)2HW=P–Mes* (39). Whereas this product eluded isolation, it

was characterized by its 31P NMR chemical shift at 819.9 ppm (1JPW =

123 Hz) and a hydride signal in the 1H NMR at –10.03 ppm.

1.3.4 Oxidation/deprotonation

One-electron oxidation of paramagnetic nickel(I) phosphido

complex 40 using tropylium hexafluorophosphate was shown by the

Hillhouse group to give cationic complex 41 that can be

deprotonated with a strong base to afford nickel(II) phosphinidene

complex (dtbpe)Ni=P–Dmp 42 (dtbpe = 1,2-bis(di-t-

butylphosphino)ethane, Dmp = 2,6-Mes2C6H3; Scheme 12).[36] The

structure has a short Ni=P bond of 2.0772(9) Å and a bent Ni–P–C unit

with an angle of 130.78°, which is also reflected by its 31P NMR

resonance at 970 ppm ( 2JPP = 134 Hz).

41 4240

R= t-Bu; Dmp = 2,6-Mes2C6H3

PF6-

NaN(TMS)2Ni

P

P

R

RR

R

P

Dmp

H

NiP

P

R

RR

R

P

Dmp

H

NiP

P

R

RR

R

P

Dmp

PF6-

Scheme 12. Nickel phosphinidene complex formed by an oxidation-deprotonation

sequence.


25

1.3.5 Phosphinidene group-transfer

The groups of Mindiola and Protasiewicz jointly demonstrated that

phospha-Wittig reagents of the type Me3P=PAr are effective as PAr

transfer reagents for the synthesis of stable phosphinidene

complexes.[37] The phosphanylidene-σ4-phosphorane reagents

Me3P=PAr (Ar = 2,4,6-tBu3C6H2 and 2,6-Mes2C6H3) can deliver PAr

fragments to low-valent early-transition metal complexes 43 and 44 to

respectively effect oxidation to form terminal Zr(IV) phosphinidene 5

(Scheme 13a) or generate terminal vanadium(V) phosphinidene

complex 45 (Scheme 13b).

Scheme 13. Phosphinidene group-transfer with phospha-Wittig reagents.

1.3.6 Dehydrohalogenation/ligation

Base-induced double dehydrohalogenation of appropriate

precursors in the presence of a suitable donor ligand was used by

Lammertsma et al. to synthesize various Group 8 and 9

phosphinidene complexes in an one-pot procedure (Scheme 14a,b).

Illustrative is the formation of iridium phosphinidene complex

(Cp*)(Ph3P)Ir=P–Mes* 10a by dehydrohalogenating phosphine

Chapter 1

26

complex 46a with two equivalents of DBU (1,8-biazabicyclo-[5.4.0]-

undec-7-ene) and capturing the putative 16-electron [(Cp*)Ir=P–

Mes*] by the donor ligand PPh3 (Scheme 14c).[25a] This mild

procedure proved equally effective for various other donor ligands L

(PH2Mes*, PMe3, P(OMe)3, dppe, AsPh3, tBuNC, XyNC, CO; Scheme

14a).[25a] The X-ray crystal structure of the CO ligated complex

(Cp*)(CO)Ir=P–Mes* (Ir=P 2.1783(8) Å, Ir–P–Mes* 113.77(10)º) shows a Z

conformation for the double bond, which is attributed to the small

size of the CO ligand and differs from that of the PPh3 ligated

complex that has an E conformation but with otherwise similar

structural features. The deshielded 31P NMR resonance for the CO

ligated complex (805 ppm), as compared to the phosphane

analogue (687 ppm), was attributed to the CO π−acceptor

capabilities rather than to geometrical differences.[25] The X-ray

crystal structure for the only isolable cobalt phosphinidene complex

(Cp)(Ph3P)Co=P–Mes* (10c) shows a piano-stool geometry with

bonding properties (Co=P 2.1102(8) Å; ∠Co–P–Mes* 109.00(9)°) that

closely resemble iridium analogue 10a.[38] The CO-ligated Co-

complex Z−(Cp)(CO)Co=P–Mes* could be observed by its

characteristic low-field 31P NMR resonance at 1047 ppm (!).

Illustrative for group 8 phosphinidenes are the ruthenium complexes

(η6–Ar)(L)Ru=P–Mes* 48a (Ar = benzene, p-cymene; L = PPh3, PMe3,

tBuNC; Scheme 14b) with 31P NMR resonances in the range 801–846

ppm, while the non-isolable CO-ligated complex (η6–Bz)(CO)Ru=P–

Mes* reportedly has a chemical shift at lower field (897 ppm).[39]

Ruthenium phosphinidenes (η6–pCymene)(R3P)Ru=P–Mes* 48a (R = Ph,

Cy) are also accessible from (η6–pCymene)RuCl2(PR3) by reaction

with DBU and PH2Mes*.[40] The heavier osmium phosphinidene

complexes (η6–Ar)(L)Os=P–Mes* 48b (Ar = benzene, p-cymene; L =


27

PPh3, PMe3, CO) were obtained equally readily from

dehydrohalogenating/ligating the primary complexed phosphine

(η6–Ar)OsX2(PH2Mes*) 47b.[39] Similar to the third row group 9 transition

metal iridium, also a carbon monoxide ligated osmium complex, (η6–

Ar)(CO)Os=P–Mes* (48b), could be isolated. The E isomers were

shown to be favored for complexes having large PR substituents (e.g.

Mes*) and bulky ligands (e.g. PPh3), the Z isomer for the smaller

carbon monoxide ligand, and a mixture of E/Z products for ligands of

intermediate size, such as PMe3, or P(OMe)3.[25,38]

Scheme 14. Dehydrohalogenation/ligation for Group 8 (a) and 9 (b) transition metal

phosphinidene complexes; c) Synthesis of phosphinidene complex 10a via putative

[(Cp*)Ir=P–Mes*] from iridium precursor 46a.

Dehydrohalogenation/ligation proved also an effective route to

introduce N−heterocyclic carbene ligands, such as IiPr2Me2 (1,3-

Chapter 1

28

diisopropyl-4,5-dimethylimidazol-2-ylidene).[41] For example, the one-

pot reaction of rhodium and iridium precursors 46a,b and ruthenium

and osmium precursors 47a,b with three equivalents of IiPr2Me2

yielded the corresponding Group 8 and 9 NHC-ligated

phosphinidene complexes 49,10b and 50a,b respectively (Scheme

15).[42,43] Because NHC is the stronger base (pKa: IiPr2Me2 24.0 in d6-

DMSO),[44] DBU (pKa: 11.3)[45] cannot be used for the dehydro-

halogenation,[25a,38,39,46] thereby necessitating the use of two

equivalents of NHC as base and one as stabilizing ligand. However,

the imidazolium salt IiPr2Me2⋅HCl that precipitates can be regenerated

to the carbene by deprotonation.[47] The NHC-functionalized

phosphinidene complexes were obtained as colored, air and

moisture sensitive, but thermally stable solids with characteristic 31P

NMR resonances (Ir: 560.0 (10b);[25b] Rh: 745.9 (49);[43] Ru: 751.7

(50a);[42] Os: 557.6 (50b)[43]) that reflect shielding due to σ-donor

capacity of the NHC ligand.

η5-Cp*

PM

49

NN

46, 47

η6-Ar

P

Mes*

M

NN

η5-Cp* Mes*

Toluene

- 2 IiPr2Me2.HCl

η6-Ar

50

M = Ru (a), Os (b);Ar = C6H6, p-Cymene

M = Rh, Ir (10b)

3 IiPr2Me2

Scheme 15. Dehydrohalogenation/ligation using NHCs.

The X-ray crystal structure of rhodium complex

(η5-Cp*)(IiPr2Me2)Rh=PMes* 49 shows a Rh=P bond length of 2.1827(7)

Å and an M−P−Mes* angle of 107.65(4)° that are similar to those for

iridium complex 10b and ruthenium complex


29

(η6-Bz)(IiPr2Me2)Ru=PMes* 50a (Ru=P 2.2222(8) Å; ∠Ru−P−Mes*

105.82(10)°). Both phosphinidene complexes have pronounced M–C

single bonds with lengths of 2.036(2) (Rh) and 2.091(3) Å (Ru) that are

in the typical range for M–NHC complexes.[48]

1.4 Reactivity of nucleophilic phosphinidene complexes

To review the reactivities of the nucleophilic, 18–electron

phosphinidene complexes it is relevant to recognize the impact of

the putative 16–electron [LM=PR]. Whereas their involvement in the

reactions often cannot be ascertained, they are evolving as reactive

entities that we address here separately.

1.4.1 Reactive 16-electron intermediates

Convincing spectroscopic evidence has been presented for the

reactive, 16–electron intermediates [LM=PMes*] (L = η5-Cp(*),η6-Ar).

They presumably form in-situ on dehydrohalogenating primary

phosphine complexes, like 46 and 47, and are than captured by a

ligand to give the discussed 18-electron phosphinidene complexes.

Slowing down the ligation by using the heavily congested carbene

IMes (1,3-dimesityl-imidazol-2-ylidene)[42,43] both as base and as ligand

in the reaction with (η6-pCymene)RuCl2(PH2Mes) 47c gave besides

the expected phosphinidene complex 52 also isomer 53 in which the

pCymene group is replaced by a (d3-)toluene solvent molecule,

indicating the conversion of the 16-electron intermediate

[(η6-pCymene)Ru=PMes] 51a to [(η6-Toluene)Ru=PMes] 51b (Scheme

16a). Solvent stabilization was demonstrated for the putative 16-

electron complex carrying the bulky phosphorus substituent 2,6-

dimesitylphenyl (Dmp). Monitoring by 31P NMR spectroscopy the

reaction of 54 with DBU in dichloromethane (Scheme 16b) showed

Chapter 1

30

the appearance of a characteristic low-field resonance at 672 ppm,

which concurs with the 684 ppm calculated at BP86/TZP for

dichloromethane-solvated phosphinidene 55.[49]

Scheme 16. Solvent-stabilized phosphinidene complexes.

In the absence of a stabilizing donor ligand, Hey−Hawkins and

coworkers reported that on dehydrohalogenating tantalum-

complexed primary phosphine Cp*(Cl4)Ta(PH2−Is), the 14-electron

phosphinidene complex [Cp*(Cl2)Ta=P−Is] (56: 31P δ 488.0 ppm) was

isolated; no mention was made of intermediate products.[50]

Lammertsma’s group showed that dehydrogenation of primary

phosphine complex 46a with the strong phosphazene base t-

butylimino-tri(pyrrolidino)phosphorane (BTPP; pKB ≈ 26) in the absence

of a ligand gave 18-electron complex 57 and dimer ½[Cp*IrCl2]2 58

(Scheme 17a).[51] Phosphinidene 57 was interpreted to result from

[Cp*Ir=PMes*], abstracting PH2Mes* from its precursor. Reaction at low

temperatures showed the intermediate formation of 59 (Scheme 17b),


31

resulting from [2+2]-cycloaddition of [Cp*Ir=PMes*] and

[Cp*(Cl)Ir=P(H)Mes*], as identified by its 31P NMR resonances (366 and

–126 ppm), and suggests that the first dehydrohalogenation is faster

than the second one. The reaction appears sensitive to the size of the

substituent on phosphorus as the smaller Mes group gave the

intermediate dimetallocycle 61, resulting from dimerization of

[Cp*(Cl)Ir=P(H)Mes], and subsequently on dehydrohalogenation

dimer 62 (Scheme 17c).

Scheme 17. Dehydrohalogenation in the absence of a stabilizing ligand.

Stephan and co-workers reported on the generation and reactivity

of the transient, 16-electron phosphinidene [Cp*2Zr=P–R] 64[19]

(Scheme 18) that can be generated from primary phosphide

complex 63 by elimination of phosphane H2PR. Attempts to isolate 64

were unsuccessful. However, it could be detected by 31P NMR

Chapter 1

32

spectroscopy at 537 ppm as an unstable [Cp*2Zr=P–Mes]–LiCl adduct

when prepared from Cp*2ZrCl2 and LiPHMes in dimethoxyethane

(DME).[19a,b] In-situ-generated 64 (R = Mes) is highly reactive and gives

65 by intramolecular C-H insertion and yields metallocycle 66 and

phosphane H2PR by reaction with acetonitrile (Scheme 18).[19] The

reinsertion of phosphane H2PR to the Zr=P bond of 64 is also feasible

and yields complex 67 (and 68 upon reaction with MeCN) irreversibly

with elimination of H2 as the deriving force (Scheme 18).[19]

Scheme 18. Formation and reactivity of [Cp*2Zr=P–R].

Stephan and co-workers reported other examples of the transient

16-electron [Cp(*)2Zr=P–R] 64 (R = SiPh3: 263, Mes*: 478, Cy: 499, Mes:

526, Ph: 579 ppm), all characterized by their 31P NMR chemical

shifts.[19,20] Majoral and co-workers reported 31P NMR chemical shifts for

[Cp(*)2Zr=P–(2,4,6–(MeO)3C6H2)] 69a,b (Cp: 465, Cp*: 438 ppm) using

salt the metathesis approach (Scheme 19a), but dimers and

polymeric forms could not be excluded.[52]

The heavier hafnium congener [Cp*Hf=PPh] 71 was postulated to

be formed from the reaction of precursor compound 70 with

NaN(SiMe3)2 as base (Scheme 19b), based on its 31P NMR resonance

at 376, but it was not isolated or trapped.[53]


33

Cp(*)2ZrLi2PR'

Cl

Cl 69

R' = 2,4,6-(MeO)3C6H2

CpR2Zr P R'a)

b) Cp*2Hf

PHPh

I

NaN(SiMe3)2Cp*

Hf P

Ph

Cp*71

- HN(SiMe3)2- NaI70

Scheme 19. Formation of transient Zr and Hf phosphinidenes.

Attempts to kinetically stabilize the transient lanthanide

phosphinidene complex 73 that was generated from 72 by applying

the bulky phosphine H2PMes* resulted in C−H activation and

formation of phosphaindole 74 (Scheme 20). The sterically less

hindered H2PMes was shown to give isolable lutetium-dimer 75

presumably from dimerization of the putative monomeric complex.[54]

PR2

PR2

Lu

Si

72

N

Si

R= iPr, Ph

PR2

PR2

Lu

73

N PMes*P H

tButBu

- 2 SiMe4

74

H2PMes*

- 2 SiMe4

H2PMes

PR2

PR2

Lu

75

N

P

R2P

R2P

Lu N

P

Mes

Mes

Scheme 20. Postulated lanthanide phosphinidene 73 and isolable Lu-dimer 75.

Chapter 1

34

1.4.2 RP-transfer

Of all the 18-electron phosphinidene complexes, Zr phosphinidene

4 developed by Stephan is the most extensively studied and a variety

of phosphinidene transfer reactions has been developed.[7] The

phospha-Wittig reaction that transfers a PR group is the most applied

reaction of nucleophilic phosphinidene complexes bearing oxo- or

halophilic transition metals such as Zr (Scheme 21). Stephan and co-

workers demonstrated that the reaction of zirconium complex 4 with

ketones and aldehydes yields phosphaalkenes 76 and the insoluble

zirconocene oxide [Cp2ZrO]n, which is easily separated from the

product, together with uncoordinated PMe3 (Scheme 21a).[20b] This

metathesis reaction is thought to proceed via initial decoordination

of PMe3 generating the active 16-electron species [Cp2Zr=P–Mes*].

Subsequent coordination of the carbonyl species to Zr followed by

intramolecular attack of the nucleophilic phosphorus atom gives a 4-

membered intermediate (Scheme 21a), which via retrocyclisation

yields the P=C and Zr=O products. Phosphinidene 4 also undergoes a

metathesis reaction with phenylisothiocyanate to give heteroallene

E−Ph−N=C=P−Mes* 77 and the insoluble zirconocene sulfide dimer

[Cp2Zr(µ-S)]2 (Scheme 21b).[20b] In addition, epoxides can be

converted into the three-membered phosphiranes 78 via a P/O-

exchange (Scheme 21c),[20b] whereas 4 in the presence of gem

dihalides and CHCl3 affords phosphaalkene 76b (Scheme 21d).[20b]

This approach was successfully extended to the synthesis of

phosphirene 79, phospholane 80, and the substituted phosphirane 81

(Scheme 21e,f,g).[20b] The mechanism for formation of phosphirane 79

from Zr-phosphinidene 4 and 1,2-dichloroethane, invoking the 16-

electron [Cp2Zr=P–Mes*] was also addressed computationally.[55]


35

a)

c)

d)

R' = H, Me, Ph

O

R'R'

P

R'R'

Mes*

- [Cp2Zr=O]n- PMe3

- Cp2ZrCl2- PMe3

O

CCHPh

P

HE

Mes*

HPh

P

CCHPh

HPh

R

CHECl2

PhN=C=S N=C=P

Ph

Mes*

b)

76a

78

76b

4

77

- Cp2ZrCl2- PMe3

e)Cl Cl

P

Mes*

P Mes*

P

Mes*

80

79

81

f)

g)

4

4

- Cp2ZrCl2- PMe3

4

- Cp2ZrCl2- PMe3

4

Cl Cl

Cl Cl

E = H, Cl

- PMe3 - [Cp2Zr=O]n

4 Cp

Zr

Cp

P

O R'

R'

Mes*

- PMe3

4 Cp

Zr

Cp

P

S

Mes*

- [Cp2Zr( -S)]2N

Ph

Scheme 21. Phosphinidene transfer reactions of phosphinidene 4.

Schrock’s group reported that tantalum complex 12 reacts with

carbonyl compounds to yield phosphaalkene 76c and Ta=O complex

82 (Scheme 22).[26] As Schrock also showed in earlier work[56] that Ta

alkylidenes and carbonyl compounds yield alkenes and Ta=O species,

the P/C analogy between phosphinidene and carbene complexes is

demonstrated.

Chapter 1

36

Scheme 22. Phosphinidene transfer reaction of tantalum complex 12.

Lammertsma and co-workers showed that the rate of the reaction

of phosphinidenes 10 (M = Co (10c), Rh (10d), Ir (10a)) with

dihalomethanes to afford phosphaalkene 76d (Scheme 23a)

depends on the halogen atom of the substrate, the oxygen/halogen

philicity of the transition metal, and the electronic properties of the

ancillary ligand.[25a,38,42,43] The influence of the stabilizing ligand was

demonstrated by changing the phosphane donor of complex 48a (L

= PPh3) for a NHC carbene ligand in 50a (IiPr2Me2), which accelerates

(40 times) the formation H2C=PMes* 76e (Scheme 23b).[42] It was

demonstrated that for 50a, the relative σ-donor/π-acceptor ability of

the NHC ligands can easily be influenced by a simple substituent-

controlled conformational change.[42]

Scheme 23. Phosphinidene transfer reactions of late transition metal phosphinidenes.


37

1.4.3 Insertion into the M====P bond

The groups of Stephan and Mindiola reported on the insertion of

substrate molecules into the M=P bond. For example, whereas

reaction of Zr-complex 4 with PhCN afforded E/Z imido complex 84 in

a 1:1 ratio (Scheme 24),[20b] that with dicyclohexylcarbodiimine gave

insertion into the Zr=PMes* bond to yield the X-ray crystallographically

characterized phosphaguanidino complex 85 (Scheme 24).[20b]

Zr P

Mes*

4

PMe3

CyN=C=NCyPhC N

- PMe3

Cp2ZrN

NPMes*

Cy

Cy

Cp2Zr N

PMes*Ph

Me3P 85E/Z-84

Scheme 24. Zr=P bond insertion reactions.

The coordinatively unsaturated titanium phosphinidene complex

35a reacts with tBuNC affording the rare η2-(N,C)-phosphaazaallene

complex 86 (Scheme 25).[34b] The two 31P NMR resonances at −8.5 and

−17.6 ppm for 86 indicate the presence of two isomers in solution.

Reaction of 35a with N2CPh2 yielded complex 87, which contains an

uncommon phosphinylimide Ti-ligand.[34b] Both complexes (86 and 87)

are exceedingly reactive and readily decompose in solution and in

the solid state.

Ar = 2,6-iPr2C6H3

Ar

N

N

Ar

Ti

H2C

P Mes*

35a

tBuNC N2CPh2

8786

Ti

CCH2tBu

C

N

NtBuNC

tBu

PMes*

tBu

N

Ar

ArAr

N

N

Ar

Ti

N

N

CPh2

PMes*

CH2tBu

Sc

heme 25. Insertion reactions into the Ti=P bond.

Chapter 1

38

1.4.4 Cycloaddition to the M====P bond

[1+2]- and [2+2]-(retro)cycloadditions are important metal-assisted

bond forming and bond breaking reactions that are well established

for metal alkylidenes,[57] in contrast to the nucleophilic phosphinidene

complexes. Only few examples have been reported, the stepwise

addition of isocyanides being one. In-situ-generated 16-electron

complex [(Cp*)Ir=P−R] (R = Mes, Mes*, Dmp) was shown to react with

an isocyanide to form 18-electron phosphinidene complex

(Cp*)(XyNC)Ir=P−R 10e. Subsequently reaction with ArNC (Ar = Ph, Xy)

gave complex 89, presumably via intermediate 88 as indicated by

DFT calculations (Scheme 26).[49]

η5-Cp*

Ir PR

10e

XyNC

R = Mes, Mes*, Dmp; Ar = Ph, Xy

ArN C

η5-Cp*

Ir P

R

88

XyNC C

η5-Cp*

Ir PR

89

XyNCC

NAr

NAr

Scheme 26. [1+2]-cycloaddition of phosphinidene and isocyanide.

An example of the [2+2]-cycloaddition of C=C and C≡C multiple

bonds to metal phosphinidenes was provided by Stephan et. al..

Zirconium complex 4 reversibly adds to acetylenes to afford

phosphametallacycle 90, which has an indicative 31P NMR resonance

at 55 ppm (Scheme 27a).[58] Loss of PMe3 from 4 is the rate

determining step in this reaction. It was shown that a more

expeditious version of this reaction starts with the spontaneous loss of

methane from Cp2(Me)ZrPH–Mes*.[58]

Hillhouse’s group showed that nickel phosphinidene 42 reacts with

olefins to form phosphirane cyclo-C2H4PDmp in a stereoselective

manner by way of metallacyclobutane 91 (Scheme 27b).[59]


39

a)

b)

c)

d)

- PMe3

42

NiP

P

R

RR

R

P

Dmp

Zr P

Mes*

4

PMe3

[Cp2Zr=PMes*]PhC CR

R = Me, Ph

Cp2ZrCC

PR

Mes*

Ph90

C2H4(dtbpe)Ni

P

Dmp

C2H4

- (dtbpe)Ni(C2H4)P

Dmp

91

42 (dtbpe)NiP

Dmp

92

PhC CH Ph

H

(dtbpe)Ni P

Ph

93

Dmp

H

6-Ar Ru P Mes*

Ar = C6H6, p-Cymene; R = Me, Ph, SiMe3; R' = H, Me

RC CR'

51a

(Ar)RuP

Mes*

94

R'

R

(Ar)RuP

95

R'

R

tBu

tBu

Scheme 27. [2+2] cycloaddition reactions.

In addition, 42 was shown to undergo cycloaddition with alkynes to

give the putative [2+2]-adduct phosphametallacyclobutene 92 that

rearranges to the more stable metallophosphabicyclobutane 93

(Scheme 27c).[60] In-situ-generated ruthenium phosphinidenes 51a

also react with alkynes, as shown by Lammertsma’s group,[39] to give

the stable phosphaallyl complexes 95 (Scheme 27d). It was reasoned

that first [2+2]-cycloadduct 94 is formed, which subsequently

undergoes C−H activation to yield the final product. Analogously,

Menye-Biyogo et al. have reported the formation of the putative

phosphinidenes 51a from the interaction of phosphinidene complex

Chapter 1

40

(η6-pCy)(Cy3P)Ru=PMes* and alkynes by loss of the phosphine

ligand.[61]

Zwitterionic titanium phosphinidene 18 with its labile borate group

was shown to undergo [2+2]-cycloaddition with diphenylacetylene to

generate phosphatitanocyclobutene 96 (Scheme 28a),[28] based on

its characteristic 31P (160.7) and 13C NMR (253.5) chemical shifts.

Complex 18 was demonstrated to be able to function as a

precatalyst in the catalytic hydrophosphination of PhCCPh with

PhPH2. The proposed mechanism (Scheme 28b) involves PAr transfer

of the primary phosphine followed by [2+2]-cycloaddition of

diphenylacetylene to form 97, which generates vinylphosphine

HP(Ph)Ph=C=CHPh 98 upon reaction with phenylphosphine.[28]

Ar

N

N

Ar

Ti

P

CB(C6F5)3

H

HH

18

Is Ar

N

N

Ar

Ti

P

H3C B(C6F5)3

96

Is

Ph

Ph

a)

b)PH2Ph

- PH2Is

PhCCPh[Ti]18 [Ti]

PPh

PPhPh

Ph

PH2Ph

Ph

H

PhHP

Ph

PhCCPh

97

98

Scheme 28. [2+2]-cycloaddition reaction (a) and proposed catalytic PAr transfer in

hydrophosphination reaction (b) for Ti phosphinidene. The β-diketeminate ligand and

the BCH3(C6F5)3 anion in (b) are omitted for clarity.

A diphosphorus analogue of the versatile Dötz intermediate that is

common in the chemistry of complexed carbenes has been reported


41

by the group of Lammertsma. η3-Diphosphavinylcarbene complex

100 resulted on the DBU-induced reaction of Ir and Ru complexed

primary phosphines 46a and 47a with phosphaalkyne Mes*C≡P. The

product obtained with the less congested tBuC≡P was shown to

convert to the 1,3-diphospha-3H-indene complex 101, which

resembles the intermediate of the Dötz benzannulation reaction

(Scheme 29).[62] The reversibility of the phosphaalkyne addition was

demonstrated by the exchange of Mes*C≡P in 100b with PPh3 and

tBuC≡P, yielding phosphinidene complex 48a and 101b, respectively.

LM PH

Cl

Mes*

LM = 46a, η5-Cp*Ir

47a, η6-pCyRu;

DBU

- DBU·HCl

99a,b

, DBU

- DBU·HCl

PRC

R =Mes*, tBu

LM

P

P

tBu

tBu

tBu

R100a,b R = Mes* 101a,b R = tBu

for R = tBu

tBu

tBu

P

PtBu

R

LM

L

M PH2

Mes*

Cl

Cl

Scheme 29. Synthesis and rearrangement of η3-diphosphavinylcarbene 100.

1.5 Conclusion

The emerging applicability of terminal phosphinidene complexes that

are nucleophilic at the phosphorus atom drives the search for novel

reagents and new reactions. The past decades has shown many

openings to evolve this chemistry. Whereas the focus was initially on

stable 18-electron complexes it is evident that the in-situ generated

16-electron analogous are viable reactive intermediates. Much

ground still needs to be covered, but it is clear that the broad

spectrum of reactions that are common place for transition metal-

Chapter 1

42

complexed carbenes are also feasible for the phosphinidenes, such

as [2+2]-cycloadditions to CC multiple bonds, insertions into single

bonds, and phospha-Wittig type reactions. Moreover, the diagonal

relationship between phosphorus and carbon in the Periodic Table

provides opportunities to mimic the carbene complexes by

conducting mechanistic studies that take advantage of the

stabilizing phosphorus atom. Exemplary is the diphospha-Dötz

intermediate. There are many more possibilities. With the increasing

emphasis on the element phosphorus in conducting organic

conversions, in ligands and catalysts, and in the advance of metal-

assisted organophosphorus chemistry much can be expected from

this field.

1.6 Outline of this thesis

In Chapter 2 the synthesis and reactivity of NHC-functionalized Ru

phosphinidene (η6-Bz)(IiPr2Me2)Ru=PMes* 50a is described.[42] The

influence of the stabilizing ligand (PPh3 vs NHC) and the relative σ-

donor/π-acceptor ability of these ligands bonded to the nucleophilic

phosphinidene complexes are investigated computationally.

The scope of the dehydrohalogenation−ligation sequence using

NHCs both as Brønsted base and as stabilizing ligand has been

successfully extended in Chapter 3 by the synthesis of novel

ruthenium, osmium, and rhodium phosphinidene complexes.[43] An

extensive computational analysis of the Group 7-9 transition metal

complexed phosphinidenes revealed Re, Rh, and Ru as the most

reactive transition metals of the Group 7-9 triads.

In Chapter 4 in-situ-generated 16-electron complex [(Cp*)Ir=P−R] (R

= Mes, Mes*, Dmp) were reacted with isocyanide ArN≡C yielding

isolable phosphinidene complexes (Cp*)(ArN≡C)Ir=P−R, which are


43

prone to [1+2] cycloaddition with a second isocyanide Ar’N≡C to

afford novel iridaphosphirane complexes 89 [Cp*(Ar–N≡C) IrPArC┌────┐

=NAr’].[49] For the imido analogue [(Cp*)Ir≡N−tBu] of Bergman a

different mechanism is found with isocyanides.

In Chapter 5 the synthesis, mechanism, and reactivity of the novel

η3-diphosphavinylcarbene complex 100, a diphosphorus analogue of

the versatile Dötz intermediate, is presented.[62] The product obtained

with the less congested tBuC≡P was shown to convert via an

unprecedented rearrangement to the novel 1,3-diphospha-3H-

indene complex 101.

References and Notes

[1] (a) K. Lammertsma, Top. Curr. Chem. 2003, 237, 95–119. (b) J. C. Slootweg, K.

Lammertsma, In Science of Synthesis; Trost, B. M., Mathey, F., Eds.; Georg Thieme

Verlag: Stuttgart, 2009; Vol. 42, pp 15–36.

[2] G. P. Moss, P. A. S. Smith, D. Travenier, Pure Appl. Chem. 1995, 67, 1307−1375.

[3] Isolation of the first carbene complex, (a) E. O. Fischer, A. Maasböl, Angew.

Chem. 1964, 76, 645–645; Angew. Chem. Int. Ed. 1964, 3, 580–581. Isolation of the

first free carbene, which can be classified as a push-push carbene, see: (b) A. J.

Arduengo, III, R. L. Harlow, M. Kline, J. Am. Chem. Soc. 1991, 113, 361–363. For

push-pull carbenes, see: (c) A. Igau, H. Grützmacher, A. Baceiredo, G. Bertrand,

J. Am. Chem. Soc. 1988, 110, 6463–6466. (d) A. Igau, A. Baceiredo, G. Trinquier, G.

Bertrand, Angew. Chem. 1989, 101, 617–618; Angew. Chem., Int. Ed. Engl. 1989,

28, 621–622. For reviews, see: (e) G. Bertrand, Carbene Chemistry (Marcel Dekker,

2002). (f) D. Bourissou, O. Guerret, F. P. Gabbaï, G. Bertrand, Chem. Rev. 2000,

100, 39–92. (g) F. E. Hahn, M. C. Jahnke, Angew. Chem. Int. Ed. 2008, 47, 3122–

3172.

[4] (a) W. A. Nugent, B. L. Haymore, Coord. Chem. Rev. 1980, 31, 123–175. (b) J. K.

Brask, T. Chivers, Angew. Chem. 2001, 113, 4082–4098; Angew. Chem. Int. Ed.

2001, 40, 3960–3976. (c) P. R. Sharp, J. Chem. Soc., Dalton Trans. 2000, 2647–2657.

(d) D. E. Wigley, Prog. Inorg. Chem. 1994, 42, 239–482. (e) L. H. Gade, P.

Mountford, Coord. Chem. Rev. 2001, 216–217, 65–97.

Chapter 1

44

[5] (a) X. Li, S. I. Weissman, T.-S. Lin, P. P. Gaspar, A. H. Cowley, A. I. Smirnov, J. Am.

Chem. Soc. 1994, 116, 7899–7900. (b) G. Bucher, M. L. G. Borst, A. W. Ehlers, K.

Lammertsma, S. Ceola, M. Huber, D. Grote, W. Sander, Angew. Chem. 2005, 117,

3353–3356; Angew. Chem. Int. Ed. Engl. 2005, 44, 3289–3373. (c) J. Glatthaar, G.

Maier, Angew. Chem. 2004, 16, 1314–1317; Angew. Chem. Int. Ed. 2004, 43, 1294–

1296. (d) J. J. Harrison, B. E. Williamson, J. Phys. Chem. A 2005, 109, 1343–1347.

[6] U. Schmidt, Angew. Chem. 1975, 87, 535–540; Angew. Chem. Int. Ed. Engl. 1975,

14, 523–528.

[7] D. W. Stephan, Angew. Chem. 2000, 112, 322–338; Angew. Chem. Int. Ed. 2000,

39, 314–337.

[8] (a) K. Lammertsma, M. J. M. Vlaar, Eur. J. Org. Chem. 2002, 1127–1138. (b) F.

Mathey, N. H. Tran Huy, A. Marinetti, Helv. Chim. Acta. 2001 84, 2938–2957.

[9] L. Weber, Eur. J. Inorg. Chem. 2007, 4095–4117.

[10] F. Mathey, Dalton Trans. 2007, 1861–1868.

[11] a) A. Marinetti, F. Mathey, J. Fischer, A. Mitschler. J. Am. Chem. Soc. 1982, 104,

4484–4485. (b) A. Marinetti, F. Mathey, J. Fischer, A. Mitschler, J. Chem. Soc.,

Chem. Commun. 1982, 667–668.

[12] (a) P. B. Hitchcock, M. F. Lappert, W.-P. Leung, J. Chem. Soc., Chem. Commun.

1987, 1282–1283. (b) R. Bohra, P. B. Hitchcock, M. F. Lappert, W.-P. Leung,

Polyhedron, 1989, 8, 1884.

[13] (a) A. H. Cowley, A. R. Barron, Acc. Chem. Res. 1988, 21, 81–87. (b) A. H. Cowley,

Acc. Chem. Res. 1997, 30, 445–451.

[14] The first reported Schrock complexes: (a) R. R. Schrock, J. Am. Chem. Soc. 1974,

96, 6796−6797. (b) R. R. Schrock, J. Am. Chem. Soc. 1978, 100, 3359−3370.

Selected review: (c) R. R. Schrock, Acc. Chem. Res. 1979, 12, 98−104.

[15] Selected reviews: (a) E. O. Fischer, G. Kreis, C. G. Kreiter, J. Muller, G. Huttner, H.

Lorenz, Angew. Chem. 1973, 85, 618−620; Angew. Chem., Int. Ed. Engl. 1973, 12,

564−565. (b) Transition Metal Carbene Complexes, ed. K. H. Dötz, H. Fischer, P.

Hoffmann, F. R. Kreissl, U. Schubert, K. Weiss, VCH, Weinheim, 1983. (c) K. H. Dötz,

Angew. Chem. 1984, 96, 573−594; Angew. Chem., Int. Ed. Engl. 1984, 23, 587−608.

(d) L. S. Hegedus, in Comprehensive Organometallic Chemistry II, ed. F. W. Abel,

F. G. A. Stone and G. Wilkinson, Pergamon, Oxford, 1995, 12, p. 549. (e) W. D.

Wulff, in Comprehensive Organometallic Chemistry II, ed. F. W. Abel, F. G. A.

Stone and G. Wilkinson, Pergamon, Oxford, 1995, 12, 469.


45

[16] A. W. Ehlers, E. J. Baerends, K. Lammertsma, J. Am. Chem. Soc. 2002, 124, 2831–

2838.

[17] (a) B. T. Sterenberg, K. A. Udachin, A. J. Carty, Organometallics 2003, 22, 3927–

3932. (b) B. T. Sterenberg, A. J. Carty, J. Organomet. Chem. 2001, 617–618, 696–

701. (c) B. T. Sterenberg, K. A. Udachin, A. J. Carty, Organometallics 2001, 20,

2657–2659. (d) B. T. Sterenberg, K. A. Udachin, A. J. Carty, Organometallics 2001,

20, 4463–4465. (e) J. Sánchez-Nieves, B. T. Sterenberg, K. A. Udachin, A. J. Carty, J.

Am. Chem. Soc. 2003, 125, 2404–2405. (f) T. W. Graham, R. P.-Y. Cariou, J.

Sánchez-Nieves, A. E. Allen, K. A. Udachin, R. Regragui, A. J. Carty,

Organometallics 2005, 24, 2023–2026.

[18] (a) T. W. Graham, K. A. Udachin, A. J. Carty, Chem. Commun. 2005, 5890–5892.

(b) B. T. Sterenberg, O. S. Senturk, K. A. Udachin, A. J. Carty, Organometallics 007,

26, 925–937.

[19] (a) Z. Hou, T. L. Breen, D. W. Stephan, Organometallics 1993, 12, 3158–3167. (b) Z.

Hou, D. W. Stephan, J. Am. Chem. Soc. 1992, 114, 10088–10089. (c) J. Ho, Z. Hou,

J. Drake, D. W. Stephan, Organometallics 1993, 12, 3145–3157. (d) J. Ho, D. W.

Stephan, Organometallics 1991, 10, 3001–3003.

[20] (a) J. Ho, R. Rousseau, D. W. Stephan, Organometallics 1994, 13, 1918–1926. (b) T.

L. Breen, D. W. Stephan, J. Am. Chem. Soc. 1995, 117, 11914–11921.

[21] E. Urnezius, K.-C. Lam, A. L. Rheingold, J. D. Protasiewicz, J. Organomet. Chem.

2001, 630, 193–197.

[22] A. T. Termaten, PhD thesis, Vrije Universiteit Amsterdam, the Netherlands, 2004.

[23] J. Pikies, E. Baum, E. Matern, J. Chojnacki, R. Grubba, A. Robaszkiewicz, Chem.

Commun. 2004, 2478–2479.

[24]D. S. J. Arney, R. C. Schnabel, B. C. Scott, C. J. Burns, J. Am. Chem. Soc, 1996, 118,

6780–6781.

[25] (a) A. T. Termaten, T. Nijbacker, M. Schakel, M. Lutz, A. L. Spek, K. Lammertsma,

Organometallics 2002, 21, 3196–3202. (b) A. T. Termaten, M. Schakel, A. W. Ehlers,

M. Lutz, A. L. Spek, K. Lammertsma, Chem. Eur. J. 2003, 9, 3577–3582.

[26] C. C. Cummins, R. R. Schrock, W. M. Davis, Angew. Chem. 1993, 105, 758–761;

Angew. Chem. Int. Ed., 1993, 32, 756–759.

[27] J. S. Freundlich, R. R. Schrock, W. M. Davis, J. Am. Chem. Soc. 1996, 118, 3643–

3655.

[28] G. Zhao, F. Basuli, U. J. Kilgore, H. Fan, H. Aneetha, J. C. Huffman, G. Wu, D. J.

Mindiola, J. Am. Chem. Soc. 2006, 128, 13575–13585.

Chapter 1

46

[29] (a) J. S. Figueroa, C. C. Cummins. J. Am. Chem. Soc. 2003, 125, 4020–4021. (b) J.

S. Figueroa, C. C. Cummins, Angew. Chem. 2004, 116, 1002–1006; Angew. Chem.

Int. Ed. 2004, 43, 984–988.

[30] A. H. Cowley, B. Pellerin, J. L. Atwood, S. G. Bott, J. Am. Chem. Soc. 1990, 112,

6734–6735.

[31] J. B. Bonanno, P. T. Wolczanski, E. B. Lobkovsky, J. Am. Chem. Soc. 1994, 116,

11159–11160.

[32] J. D. Masuda, A. J. Hoskin, T. W. Graham, C. Beddie, M. C. Fermin, N. Etkin, D. W.

Stephan, Chem. Eur. J. 2006, 12, 8696–8707.

[33] E. Niecke, J. Hein, M. Nieger, Organometallics 1989, 8, 2370–2371.

[34](a) F. Basuli, J. Tomaszewski, J. C. Huffman, D. J. Mindiola, J. Am. Chem. Soc. 2003,

125, 10170–10171. (b) F. Basuli, L. A. Watson, J. C. Huffman, D. J. Mindiola, Dalton

Trans. 2003, 4228–4229. (c) F. Basuli, B. C. Bailey, J. C. Huffman, M.-H. Baik, D. J.

Mindiola, J. Am. Chem. Soc. 2004, 126, 1924–1925. (d) B. C. Bailey, J. C. Huffman,

D. J. Mindiola, W. Weng, O. V. Ozerov, Organometallics 2005, 24, 1390–1393.

[35] W. Malisch, U.-A. Hirth, K. Grün, M. Schmeusser, O. Fey, U. Weis, Angew. Chem.

1995, 107, 2717–2719; Angew. Chem. Int. Ed. Engl. 1995, 34, 2500–2502.

[36] R. Melenkivitz, D. J. Mindiola, G. L. Hillhouse, J. Am. Chem. Soc. 2002, 124, 3846–

3847.

[37] U. J. Kilgore, H. Fan, M. Pink, E. Urnezius, J. D. Protasiewicz, D. J. Mindiola, Chem.

Commun. 2009, 4521–4523.

[38] A. T. Termaten, H. Aktas, M. Schakel, A. W. Ehlers, M. Lutz, A. L. Spek, K.

Lammertsma, Organometallics 2003, 22, 1827–1834.

[39] A. T. Termaten, T. Nijbacker, M. Schakel, M. Lutz, A. L. Spek, K. Lammertsma,

Chem. Eur. J. 2003, 9, 2200–2208.

[40]R. Menye-Biyogo, F. Delpech, A. Castel, A. Castel, H. Gornitzka, P. Rivière, Angew.

Chem. 2003, 115, 5768–5770; Angew. Chem. Int. Ed. 2003, 42, 5610–5612.

[41] N. Kuhn, T. Kratz, Synthesis 1993, 561–562.

[42] H. Aktas, J. C. Slootweg, M. Schakel, A. W. Ehlers, M. Lutz, A. L. Spek, K.

Lammertsma, J. Am. Chem. Soc. 2009, 131, 6666–6667.

[43] H. Aktas, J. C. Slootweg, A. W. Ehlers, M. Lutz, A. L. Spek, K. Lammertsma,


[44] R. W. Alder, P. R. Allen, S. J. Williams, J. Chem. Soc., Chem. Comm. 1995, 1267–

1268.


47

[45] The pKa of DBU is 11.3, see: (a) G. Höfle, W. Steglich, H. Vorbruggen, Angew.

Chem. 1978, 90, 602–615; Angew. Chem. Int. Ed. 1978, 17, 569–583. The estimated

pKa value of DBU is 12 (H2O), see: (b) H. Ripin, D. A. Evans.

http://daecr1.harvard.edu/pdf/evans_pKa_table.pdf.

[46] (a) V. Nair, S. Bindu, V. Sreekumar, Angew. Chem. 2004, 116, 5240–5245; Angew.

Chem. Int. Ed. 2004, 43, 5130–5135. (b) A. E. Mattson, A. R. Bharadwaj, A. M. Zuhl,

K. A. Scheidt, J. Org. Chem. 2006, 71, 5715–5724; and references therein.

[47] A. J. Arduengo, III, H. V. R. Dias, R. L. Harlov, M. Kline, J. Am. Chem. Soc. 1992, 114,

5530–5534.

[48]Rh: (a) W. A. Herrmann, C. Köcher, L. J. Gooβen, G. R. J. Artus, Chem. Eur. J. 1996,

2, 1627–1636. (b) W. A. Herrmann, M. Elison, J. Fischer, C. Köcher, G. R. J. Artus,

Chem. Eur. J. 1996, 2, 772–780. (c) N. M. Scott, R. Dorta, E. D. Stevens, A. Correa, L.

Cavallo, S. P. Nolan, J. Am. Chem. Soc. 2005, 127, 3516–3526. (d) M. Viciano, E.

Mas-Marza, M. Sanau, E. Peris, Organometallics 2006, 25, 3063–3069. Ru: (a); (b);

(e) J. Huang, H.-J. Schanz, E. D. Stevens, S. P. Nolan, Organometallics 1999, 18,

2370–2370. (f) W. Baratta, E. Herdtweck, W. A. Herrmann, P. Rigo, J. Schwarz,


[49] H. Aktas, J. Mulder, J. C. Slootweg, M. Schakel, A. W. Ehlers, M. Lutz, A. L. Spek, K.


[50] U. Segerer, R. Felsberg, S. Blaurock, G. A. Hadi, E. Hey−Hawkins, Phosphorus Sulfur

Silicon, 1999, 146, 477–480.

[51] A. T. Termaten, T. Nijbacker, Andreas W. Ehlers, M. Schakel, M. Lutz, A. L. Spek, M.

L. McKee, K. Lammertsma, Chem. Eur. J. 2004, 10, 4063–4072.

[52] A. Mahieu, A. Igau, J.–P. Majoral, Phosphorus Sulfur Silicon 1995, 104, 235–239.

[53] G. A. Vaughan, G. L. Hillhouse, Organometallics 1989, 8, 1760–1765.

[54] J. D. Masuda, K. C. Jantunen, O. V. Ozerov, K. J. T. Noonan, D. P. Gates, B. L.

Scott, J. L. Kiplinger, J. Am. Chem. Soc. 2008, 130, 2408–2409.

[55] T. P. M. Goumans, A. W. Ehlers, K. Lammertsma, J. Organometallic Chem. 2005,

690, 5517–5524.

[56] R. R. Schrock, J. Am. Chem. Soc. 1976, 98, 5399−5400.

[57] R. H. Grubbs, Angew. Chem. 2006, 118, 3845–3850; Angew. Chem. Int. Ed. 2006,

45, 3760–3765.

[58] T. L. Breen, D. W. Stephan, J. Am. Chem. Soc. 1996, 118, 4204–4205.

[59] R. Waterman, G. L. Hillhouse, J. Am. Chem. Soc. 2003, 125, 13350–13351.

[60] R. Waterman, G. L. Hillhouse, Organometallics 2003, 22, 5182–5184.

Chapter 1

48

[61] R. Menye-Biyogo, F. Delpech, A. Castel, V. Pimienta, H. Gornitzka, P. Rivière,


[62] H. Aktas, J. C. Slootweg, A. W. Ehlers, M. Lutz, A. L. Spek, K. Lammertsma, Angew.


N-Heterocyclic Carbene Functionalized Ruthenium

Phosphinidenes: What a Difference a Twist Makes

Halil Aktas,† J. Chris Slootweg,† Marius Schakel,† Andreas W. Ehlers,† Martin

Lutz,‡ Anthony L. Spek,‡ and Koop Lammertsma*,†


1083, 1081 HV Amsterdam, The Netherlands, and Bijvoet Center for Biomolecular Research, Crystal

Structural Chemistry, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands

J. Am. Chem. Soc. 2009, 131, 6666–6667

Abstract:Abstract:Abstract:Abstract: Catalyst tuning by changing ligands is a well established protocol in transition metal

chemistry. N-Heterocyclic carbenes (NHC) and tertiary phosphines (R3P) are the ubiquitous ligand

actors. Here we demonstrate that the relative σ-donor/π-acceptor ability of the NHC ligand itself

can be influenced by a simple substituent-controlled conformational change, thereby directly

impacting the reactivity of the transition metal complex.

Chapter 2

Chapter 2

50

2.1 Introduction

N-heterocyclic carbenes (NHC)[1] are ubiquitous ligands in transition-

metal chemistry and homogeneous catalysis and serve increasingly

often as a replacement for tertiary phosphines (R3P). The two ligand

classes exert often subtle but crucially different electronic influences

on the properties of catalysts.[2] Exemplary is the enhanced activity of

the second-generation Grubbs metathesis catalyst

[(Cy3P)(L)Cl2Ru=CHPh] (Cy = cyclohexyl; L = H2IMes, 1,3-dimesityl-4,5-

dihydroimidazol-2-ylidene) relative to that of the first-generation

catalyst (L = Cy3P), which is caused by the differences in σ-donor/π-

acceptor ability, [3] shape, and symmetry of the ligands.[4]

Does a similar sensitivity apply to the isolobal phosphinidene[5]

complexes? We address here the ligand and conformational

sensitivities for [(η6-C6H6)(L)Ru=PMes*] (L = IiPr2Me2; Mes* = 2,4,6-

tBu3C6H2 (1), L = Ph3P (2)[6]) by examining their solution-phase

chemistry together with their structure-activity parameters modeled

by density functional theory (DFT). We simultaneously demonstrate

the applicability of phosphinidene complexes to the synthesis of

phosphaalkenes (P=C),[5a] which are unique P=ligands[7] and

attractive building blocks for P-functionalized polymers.[8]

2.2 NHC functionalization

The desired novel dark-green crystalline compound 1 (84%) was

obtained by a double dehydrohalogenation–ligation sequence[9] of

the phosphine complex [(η6-C6H6)RuCl2(PH2Mes*)][6] using three

equivalents of IiPr2Me2 in toluene (eq 1). In this reaction, two NHCs act

as Brønsted bases, while the third carbene captures the putative 16-

electron intermediate [(η6-C6H6)Ru=PMes*] (3). The single 31P NMR

resonance of 1 at 751.7 ppm is highly shielded compared to the

N-Heterocyclic Carbene Functionalized Ruthenium Phosphinidenes: What a Difference a Twist Makes

51

known triphenylphosphine analogue 2 (31P NMR δ = 845.9 ppm),[6]

which is attributed to the σ-donor capacity of IiPr2Me2 (Table 1).

P

Mes*

Ru

1

L

2 IiPr2Me2

3

P Mes*RuI iPr2Me2

PH2

Mes*

RuCl

Cl L = IiPr2Me2

(1)-2 IiPr2Me2•HCl

The molecular structure of 1, established unequivocally by a single-

crystal X-ray analysis (Figure 1),[10] has an exact mirror symmetry and

shows a two-legged “piano stool” shape with a characteristic acute

C15–Ru1–P1 angle of 84.88(9)°, a bent phosphinidene complex with a

Ru1–P1–C1 angle of 105.81(9)°, and an E configuration for the

congested Ru1–P1 double bond (2.2222(8) Å). This bond is longer

than that of the “first-generation” phosphinidene 2 (2.1988(6) Å),[6]

whereas its Ru1–Bz(cg) bond is correspondingly shorter (1.7390(12) Å

in 1; 1.7560(12)[6] Å in 2).

Figure 1. Displacement ellipsoid plot (50% probability level) for 1. Only one

conformation of the disordered tert-butyl group is shown. Hydrogen atoms are

omitted for clarity. Bz denotes the centroid position of the benzene ring. Symmetry

operation a: x, 0.5-y, z. Selected bond distances [Å] and angles and torsion angles

[°]: Ru1–P1 2.2222(8), Ru1–C15 2.091(3), Ru1–Bz(cg) 1.7390(12), P1–C1 1.876(3); Ru1–

P1–C1 105.81(9), P1–Ru1–C15 84.88(9), N1–C15–N1a 104.8(3); C2a–C1–C2–C3 –

18.7(4).

Chapter 2

52

Steric congestion is reflected in the 18.7(4)° distortion from planarity

of the Mes* ring and in the restricted rotation of the iso-propyl wingtip

groups of the NHC fragment, indicated by the two 13C NMR

resonances at 21.6 and 21.8 ppm. A striking feature is the orthogonal

relationship between the NHC and Ru=P units, which contrasts with

the in-plane arrangement of the NHC and Ru=C units in the second-

generation Grubbs catalyst.[11]

2.3 Theoretical calculations on [(Bz)M(L)====PPh]

To address the effect of the NHC ligand orientation in 1 and the

impact of the stabilizing ligand (L = NHC vs R3P) on the properties of 1

and 2, we performed BP86/TZP calculations on model structures

(labeled 1', 2', etc.) bearing a P-phenyl group (instead of P–Mes*)

and methyl groups on the NHC (IMe) and phosphine (PMe3) ligands

(Figure 2).

P2 C2

P1

Ru

C1

C2

P1

Ru

C1

C2

P1

Ru

1’-σ 1’-ππππ 2’

Figure 2. BP86/TZP optimized structures of 1’-σσσσ, 1’-ππππ, and 2’ (all Cs symmetry). Selected

bond distances [Å] and angles [°] of 1’-σσσσ: Ru–P1 2.216, Ru–C1 2.071, Ru–Bz(cg) 1.784,

P1–C2 1.859; Ru–P1–C2 108.74, C1–Ru–P1 82.83; 1’-ππππ: Ru–P1 2.204, Ru–C1 2.159, P1–C2

1.862, Ru–Bz(cg) 1.833; Ru–P1–C2 112.39, C1–Ru–P1 91.64; 2’: Ru–P1 2.209, Ru–P2

2.342, Ru–Bz(cg) 1.791, P1–C2 1.859; Ru–P1–C2 109.33, P1–Ru–P2 85.50.

The optimized geometries of 1'-σσσσ, in which the IMe ligand and the

Ru=P bond are orthogonal, and 2' compare well with the

corresponding X-ray structures.[6] But why does the NHC-ligated

structure not prefer a coplanar arrangement of IMe and Ru=P (1'-ππππ)?


53

For the unsubstituted NHC (H instead of Me) the calculations do

indeed show a 2.5 kcal·mol–1 preference for the coplanar form, but

the methyl derivative favors the orthogonal conformation by 12.5

kcal·mol–1. Apparently, the steric congestion induced by the IMe

wingtip groups enforces the “out-of-plane” conformation. This

substituent effect causes a reduction in the π-acceptor capacity of

the IMe fragment (–0.18 → –0.10 e-), making the NHC ligand in 1'-σσσσ

an effective donor (Table 1).

Table 1. Energy Decomposition Analysis for Ru–L bond of 1’-σσσσ, 1’-ππππ, and 2’.a

complex ∆Ea’ ∆Ea” ∆E steric ∆E tot ∆E prep BE

1’-σσσσ –60.8 (0.45, –0.10) –6.4 (–0.01) 17.1 –50.1 9.2 –41.0

1’-ππππ –45.9 (0.41) –13.5 (–0.18) 14.4 –45.0 16.6 –28.4

2’ –50.6 (0.52, –0.10) –10.8 (–0.12) 18.5 –43.0 8.1 –34.9 a in kcal·mol–1; values in parentheses refer to charge transferred (eV) to (positive) and from the metal center.

The Ru–L bond properties impact those of the Ru=P bond, which is

evident from the energy decomposition scheme in ADF (see

Appendix 1). Because the carbene provides less back-bonding than

PMe3, the frontier orbitals of the [(η6-C6H6)(IMe)Ru] fragment (E(dxz) = –

2.27 eV, E(dyz) = –2.49 eV) are higher in energy than those of the Ru-

phosphine fragment (E(dxz) = –2.61 eV, E(dyz) = –2.86 eV; Table 2).

Ru=P bond formation causes transfer of charge from [(η6-C6H6)(L)Ru]

to the 3PPh fragment (E(px) = –4.59 eV, E(py) = –4.86 eV), which is

largest for 1'-σσσσ. Whereas the Ru=P bonds are of similar lengths (2.216

and 2.209 Å for 1'-σσσσ and 2' respectively), the polarity varies with the

phosphorus atom, which carries more charge in 1'-σσσσ (–0.113e) than in

2' (–0.086e).

Chapter 2

54

Table 2. Calculated d-orbital energies ε (eV) for the [(C6H6)(PhP)Ru] fragments.

Ed 1’-σσσσ 1’-ππππ 2’

occ. –4.13 –4.04 –4.47 occ. –4.09 –3.95 –4.44 occ. –3.75 –3.66 –4.14 occ. –2.49 –2.56 –2.86 occ. –2.27 –2.49 –2.61

2.4 Reactivity

The greater Ru=P bond polarity is reflected in the enhanced reactivity

of the NHC-containing phosphinidene 1 (L = IiPr2Me2) over that of

“first-generation” 2 (L = Ph3P) toward diiodomethane (eq 2):[12]

31P NMR monitoring of the reaction of complex 1 showed the

quantitative formation of phosphaalkene H2C=PMes* (6, 94% isolated

yield) within 1 min. at 20 °C (t½ (0 °C, C6D6) = 22 min; five equivalents

of CH2I2). In contrast, the reaction of phosphine ligated complex 2

with CH2I2 is much slower (t½ (20 °C, toluene) = 60 min.; t½ (0 °C, C6D6)

= 925 min.) and also less selective (6, 45%). This difference between 1

and 2 demonstrates that like the difference in catalytic activity of the

Grubbs catalysts, the reactivity of the isolobal nucleophilic 18-

electron phosphinidene complexes can also be readily modified by

changing the ancillary ligands. The applicability of the illustrated

reaction is underscored by the quantitative regeneration of 1 from

the transition metal byproduct [(η6-C6H6)(IiPr2Me2)RuI2] (4) with DBU

and H2PMes*[13] as determined by 31P NMR (63% isolated yield),


55

thereby demonstrating that ruthenium phosphinidene complexes are

viable reagents for the synthesis of phosphaalkenes.

2.5 Transient species

A final aspect to address is the presumed 16-electron phosphinidene

intermediate 3, which could not be detected by 31P NMR

spectroscopy,[14,15] suggesting that if it is indeed formed, it is readily

captured by IiPr2Me2 to yield 1. Increasing the steric bulk by using 1,3-

dimesityl-imidazol-2-ylidene (IMes) to slow the NHC complexation

enough for detection was unsuccessful, but monitoring its ligation

with the less crowded [(η6-C6H6)RuCl2(PH2Mes)], which carries a Mes

instead of a Mes* substituent, did have the anticipated effect.

Besides dark-brown crystalline [(η6-C6H6)(IMes)Ru=PMes] (7a; 31P NMR

δ = 752.5 ppm, 65%), small amounts of the corresponding toluene

adduct [(η6-Tol)(IMes)Ru=PMes] (7b; 31P NMR δ = 736.8 ppm, 3%) were

also observed (eqs 3 and 4).[16]

7a L = IMesbenzenetolueneslippage

8b

P

Mes

Ru

L

IMes

P

Mes

Ru

L

Me

C6H6 P MesRu

C7H8 P MesRuIMes

7b L = IMes

8a

(3)

(4)

The apparent arene exchange is supported by detection of d3-7b

when d3-toluene was used as the solvent. Since no ligand exchange

was observed for the isolated products, it appears that the 16-

electron intermediate is prone to arene exchange. BP86/TZP

calculations support this view. Simplified 16-electron [(η6-C6H6)Ru=PH],

which has an energy minimum, reacts barrier-free with toluene to

Chapter 2

56

form the 5.6 kcal·mol–1 favored [(η2-Tol)(η6-C6H6)Ru=PH] as an initial

adduct in the exchange of the two arene ligands. Associative ring

slippage[17] via [(η4-Tol)(η4-C6H6)Ru=PH] then gives [(η6-Tol)(η2-

C6H6)Ru=PH] (∆E = 1.4 kcal·mol–1), which requires 4.2 kcal·mol–1 to lose

benzene and form the product. Implicitly, this process supports a 16-

electron intermediate that undergoes ligand exchange of aromatic

molecules (8a→8b) before being captured by the carbene ligand to

give 7b.

2.6 Conclusion

Catalyst tuning is generally sought via a change of ligands because

their effect is considered to be constant for a given transition metal

complex. We have now demonstrated that the relative σ-donor/π-

acceptor ability of NHC ligands can easily be influenced by a simple

substituent-controlled conformational change. The sterically imposed

ligand rotation of the NHC fragment in 1 enhances its reactivity and

thereby facilitates the synthesis of phosphaalkene (P=C) building

blocks.

2.7 Computational Section

All DFT calculations have been performed with the parallelized Amsterdam

density functional (ADF) package (version 2005.01 and 2005.01b).[18] The

Kohn-Sham MOs were expanded in a large, uncontracted basis set of Slater-

type orbitals (STOs), of a triple-ζ basis set with polarization functions quality,

corresponding to basis set TZP in the ADF package. The 1s core shell of

carbon and nitrogen and the 1s2s2p core shells of phosphorus were treated

by the frozen core approximation. The transition metal centers were

described by a triple-ζ basis set for the outer ns, np, nd and (n+1)s orbitals,

whereas the shells of lower energy were treated by the frozen core

approximation using a small core. All calculations were performed at the


57

nonlocal exchange self-consistent field (NL-SCF) level, using the local density

approximation (LDA) in the Vosko-Wilk-Nusair parametrization[19] with

nonlocal corrections for exchange (Becke88)[20] and correlation

(Perdew86).[21] All geometries were optimized using the analytical gradient

method implemented by Versluis and Ziegler,[22] including relativistic effects

by the Zeroth Order Regular Approximation (ZORA).[23] All Cartesian

coordinates in Angstroms and energies in kcal.mol–1 for calculated

compounds are available free of charge via the Internet at

htttp://pubs.acs.org.

Bonding Energy Analysis. The bonding interactions of the transition metal

to ligand bonds (Ru–P, Ru–C(NHC), and Ru–Bz) were analyzed with ADF's

established energy decomposition[24] into an exchange (or Pauli) repulsion

(∆E Pauli) between the electrons on the two fragments plus electrostatic

interaction energy part (∆E elst) and an orbital interaction energy (charge

transfer, polarization) part (∆E oi). The energy necessary to convert fragments

from their ground state equilibrium geometries to the geometry and

electronic state they acquire in the complex is represented by a preparation

energy term (∆E prep). The overall bond energy (∆E tot) is formulated as:

∆E tot=∆E Pauli+∆E elst+∆E oi+∆E prep

Note that ∆E tot is defined as the negative of the bond dissociation energy

(BDE), i.e., ∆E tot=E (molecule) – ΣE (fragments), thereby giving negative

values for stable bonds. The orbital interaction term ∆E oi accounts for

interactions between occupied orbitals on one fragment with unoccupied

orbitals on the other fragment, including HOMO-LUMO interactions and

polarization (empty/occupied orbital mixing on the same fragment). The

charge transfer part is the result of both σ-donation from the ligand to the

metal, and π-back-donation from the metal into the unoccupied orbitals of

the ligand. Instead of separating the charge transfer and polarization parts,

we used the extended transition state (ETS) method developed by Ziegler

and Rauk to decompose ∆E oi into contributions from each irreducible

representation of the interacting system.[24] In systems with a clear σ, π-

Chapter 2

58

separation (a' and a”), this symmetry partitioning proves to be most

informative.

2.8 Experimental Section

General. All experiments and manipulations were performed under an

atmosphere of dry nitrogen or argon with rigorous exclusion of air and

moisture using flame-dried glassware. 1H, 13C and 31P NMR spectra were

recorded at 300 K on a Bruker Avance 250 (respectively 250.13, 62.90 and

101.25 MHz) or on a Bruker Avance 400 (respectively 400.13, 100.64 and

162.06 MHz) spectrometer. 1H NMR spectra were internally referenced to

CHCl3 (δ 7.26) or C6D5H (δ 7.16), 13C NMR spectra to CDCl3 (δ 77.16) or C6D6 (δ

128.06) and 31P NMR spectra externally to 85% H3PO4. IR Spectra were

recorded on a Mattson-6030 Galaxy FT-IR spectrophotometer, high-resolution

mass spectra (HRMS) on a Finnigan Mat 900 spectrometer operating at an

ionization potential of 70 eV, and fast atom bombardment (FAB) mass

spectrometry was carried out using a JEOL JMS SX/SX 102A four-sector mass

spectrometer. Melting points were measured on samples in unsealed

capillaries on a Stuart Scientific SMP3 melting point apparatus and are

uncorrected. Elemental analyses were performed by the Microanalytical

Laboratory of the Laboratorium für Organische Chemie, ETH Zürich,

Switzerland.

Reagents. Solvents were distilled from sodium (toluene and n-hexane),

sodium benzophenone (THF), P2O5 (CH2Cl2), or LiAlH4 (pentanes) and kept

under an atmosphere of dry nitrogen. Deuterated solvents were dried over 4

Å molecular sieves (CDCl3, C6D6). All solid starting materials were dried in

vacuo. [(η6-C6H6)RuCl2]2,[25] [(η6-C6Me6)RuCl2]2,[26] 1,3-diisopropyl-4,5-

dimethylimidazol-2-ylidene (IiPr2Me2),[27] 1,3-dimesityl-imidazol-2-ylidene

(IMes),[28] [(η6-C6H6)RuCl2(PH2Mes*)],[6] and Mes*PH2,[29] were prepared

according to literature procedures. MesPH2 was prepared, analogously to

IsPH2,[30] by LiAlH4 reduction of MesPCl2. [(η6-pCy)RuCl2]2 was purchased from

Strem and used as received.


59

Experimental Procedures.

[(ηηηη6-C6H6)RuCl2(PH2Mes)]. MesPH2 (0.117 g, 0.77 mmol) was added to a

dark red suspension of [(η6-C6H6)RuCl2]2 (0.175 g, 0.35 mmol) in CH2Cl2 (15

mL), which was stirred for 1 h at room temperature and filtered to remove

insoluble material. After concentrating the solution to a few mL, pentanes

(30 mL) was added slowly, which resulted in precipitation of brown

microcrystals that were isolated by filtration, washed with pentanes (25 mL),

and dried in vacuo to yield [(η6-C6H6)RuCl2(PH2Mes)] (0.233 g, 0.58 mmol,

83%). Mp ≥ 150 °C (dec). 1H NMR (250.13 MHz, CDCl3, 300 K): δ 2.36 (s, 3 H, p-

CH3), 2.52 (s, 6 H, o-CH3), 5.44 (s, 6 H, C6H6), 5.59 (d, 1JHP = 395.7 Hz, 2 H, PH2);

7.04 (d, 4JHP = 2.7 Hz, 2 H, m-Mes). 13C{1H} NMR (62.90 MHz, CDCl3, 300 K): δ

21.2 (s, p-CH3), 22.2 (d, 3JCP = 8.7 Hz, o-CH3), 87.8 (d, 2JCP = 3.9 Hz, C6H6), 121.9

(d, 1JCP = 44.8 Hz, i-Mes), 129.7 (d, 3JCP = 8.0 Hz, m-Mes), 140.7 (d, 2JCP = 7.6 Hz,

o-Mes), 141.4 (d, 4JCP = 2.5 Hz, p-Mes). 31P NMR (101.3 MHz, CDCl3, 300 K): δ –

44.8 (t, 1JPH = 395.5 Hz, PH2). IR (KBr, cm–1): v = 3088.8 (w), 3061.4 (m), 3019.5

(w), 2967.9 (w), 2930.2 (w), 2365.1 (m, P–H), 2344.4 (m, P–H), 1601.5 (m), 1436.7

(s), 1391.3 (m). HR FAB-MS: calcd for C15H19PCl2Ru: 401.9645; found: 401.9635.

MS m/z (%): 402 (30) [M+], 367 (50) [M+ – Cl], 331 (65) [M+ – 2Cl], 250 (20) [M+ –

H2PMes], 154 (100) [H2PMes].

[(ηηηη6-pCy)RuCl2(PH2Mes)]. Analogous to that described for [(η6-

C6H6)RuCl2(PH2Mes)], [(η6-pCy)RuCl2]2 (0.400 g, 0.653 mmol) and MesPH2

(0.219 g, 1.437 mmol) were used to give [(η6-pCy)RuCl2(PH2Mes)] as an

orange powder (0.561 g, 1.22 mmol, 94%). Mp ≥ 163 °C (dec). 1H NMR (250.13

MHz, CDCl3, 300 K): δ 1.19 (d, 3JHH = 7.0 Hz, 6 H, CH(CH3)2), 1.99 (s, 3 H, pCy-

CH3), 2.34 (s, 3 H, p-CH3), 2.50 (s, 6 H, o-CH3), 2.70 (sp, 3JHH = 7.0 Hz, 1 H,

CH(CH3)2), 5.13 (d, 3JHH = 5.7 Hz, 2 H, C6H4), 5.34 (d, 3JHH = 5.7 Hz, 2 H, C6H4),

5.52 (d, 1JHP = 389.8 Hz, 2 H, PH2); 7.01 (d, 4JHP = 2.3 Hz, 2 H, m-Mes). 13C{1H}

NMR (62.90 MHz, CDCl3, 300 K): δ 18.4 (s, pCy-CH3), 21.6 (d, 5JCP = 1.1 Hz, p-

CH3), 22.5 (s, CH(CH3)2), 22.9 (d, 3JCP = 8.2 Hz, o-CH3), 31.0 (s, CH(CH3)3), 85.5

(d, 2JCP = 5.7 Hz, C6H4), 89.1 (d, 2JCP = 5.3 Hz, C6H4), 97.6 (s, C6H4), 107.7 (d, 2JCP

= 1.0 Hz, C6H4), 122.3 (d, 1JCP = 42.3 Hz, i-Mes), 130.1 (d, 3JCP = 7.8 Hz, m-Mes),

141.4 (d, 2JCP = 7.7 Hz, o-Mes), 141.5 (d, 4JCP = 2.9 Hz, p-Mes). 31P NMR (101.3

Chapter 2

60

MHz, CDCl3, 300 K): δ –48.0 (t, 1JPH = 389.9 Hz, PH2). IR (KBr, cm–1): v = 3043.0

(w), 2967.2 (m), 2913.3 (w), 2864.7 (w), 2387.6 (m, P–H), 2353.7 (m, P–H), 1601.4

(m), 1437.9 (s), 1382.3 (s). HR FAB-MS: calcd for C19H27PCl2Ru: 458.0271; found:

458.0270. MS m/z (%): 458 (70) [M+], 423 (100) [M+ – Cl], 388 (100) [M+ – 2Cl],

271 (35) [M+ – Cl – H2PMes], 154 (90) [H2PMes].

[(ηηηη6-C6Me6)RuCl2(PH2Mes)]. Analogous to that described for [(η6-

C6H6)RuCl2(PH2Mes)], [(η6-C6Me6)RuCl2]2 (1.01 g, 1.50 mmol) and MesPH2

(0.464 g, 3.05 mmol) were used to give [(η6-C6Me6)RuCl2(PH2Mes)] as a yellow

powder (1.29 g, 2.65 mmol, 88%). Mp ≥ 257 °C (dec). 1H NMR (250.13 MHz,

CDCl3, 300 K): δ 1.92 (s, 18 H, C6(CH3)6), 2.29 (s, 3 H, p-CH3), 2.39 (s, 6 H, o-CH3),

5.45 (d, 1JHP = 379.7 Hz, 2 H, PH2), 6.95 (d, 4JHP = 2.1 Hz, 2 H, m-Mes). 13C{1H}

NMR (62.90 MHz, CDCl3, 300 K): δ 15.7 (s, C6(CH3)6), 21.6 (s, p-CH3), 23.4 (d, 3JCP

= 8.0 Hz, o-CH3), 96.1 (d, 3JCP = 3.5 Hz, C6(CH3)6), 119.3 (d, 1JCP = 38.3 Hz, i-Mes),

129.9 (d, 3JCP = 7.6 Hz, m-Mes), 141.1 (d, 4JCP = 2.5 Hz, p-Mes), 142.7 (d, 2JCP =

7.4 Hz, o-Mes). 31P NMR (101.3 MHz, CDCl3, 300 K): δ –53.7 (t, 1JPH = 379.8 Hz,

PH2). IR (KBr, cm–1): v = 3019.1 (w), 2953.2 (m), 2914.4 (m), 2397.3 (m, P–H),

2364.6 (m, P–H), 1601.3 (m), 1447.67 (s), 1383.7 (m). HR FAB-MS: calcd for

C21H31PCl2Ru: 486.0584; found: 486.0588. MS m/z (%): 486 (65) [M+], 451 (50)

[M+ – Cl], 413 (50) [M+ – 2HCl], 299 (60) [M+ – 2HCl – H2PMes], 154 (100)

[H2PMes].

[(ηηηη6-C6H6)(IiPr2Me2)Ru====PMes*] (1). Complex [(η6-C6H6)RuCl2(PH2Mes*)] (0.082

g, 0.154 mmol) was added to a solution of IiPr2Me2 (0.083 g, 0.462 mmol) in

toluene (4 mL) at –78 °C. The mixture was stirred for 3 h and was slowly

allowed to warm up to room temperature. The resulting dark green

suspension was filtered and the salts were extracted with toluene (3 x 1 mL).

After concentration of the solvent, dark green crystals of 1 were obtained at

–20 °C (0.083 g, 0.130 mmol, 84%). Mp ≥ 222 °C (dec). 1H NMR (250.13 MHz,

C6D6, 300 K): δ 1.17 (d, 3JHH = 7.3 Hz, 6 H, CH(CH3)2), 1.32 (d, 3JHH = 7.3 Hz, 6 H,

CH(CH3)2), 1.58 (s, 9 H, p-C(CH3)3), 1.78 (s, 18 H, o-C(CH3)3), 1.89 (s, 6 H, CH3),

4.69 (s, 6 H, C6H6), 5.95 (sp, 3JHH = 7.3 Hz, 2 H, CH(CH3)2), 7.60 (s, 2 H, m-Mes*).

13C{1H} NMR (62.90 MHz, C6D6, 300 K): δ 10.4 (s, CH3), 21.6 (s, CH(CH3)2), 21.8 (s,

CH(CH3)2), 32.0 (s, p-C(CH3)3), 32.5 (d, 4JCP = 7.5 Hz, o-C(CH3)3), 34.9 (s, p-


61

C(CH3)3), 38.6 (s, o-C(CH3)3), 52.3 (s, CH(CH3)2), 79.9 (s, C6H6), 119.4 (s, m-

Mes*), 124.5 (s, =CCH3), 144.6 (s, p-Mes*), 145.6 (s, o-Mes*), 177.0 (d, 1JCP =

100.7 Hz, i-Mes*), 187.9 (s, N2C). 31P{1H} NMR (101.3 MHz, C6D6, 300 K): δ 751.7

(s, Ru=P). HRMS (EI, 70 eV): calcd for C35H55N2PRu: 636.3140; found: 636.3137.

MS m/z (%): 636 (12) [M+], 558 (8) [M+ – C6H6], 536 (8) [M+ – CH(CH3)2 –

C(CH3)3], 319 (20) [M+ – Mes* – NCH(CH3)2 – CH3], 180 (20) [IiPr2Me2]. Anal.

Calcd for C35H55N2PRu: C, 66.10; H, 8.72; N, 4.41. Found: C, 66.32; H, 8.83; N,

4.32. Alternatively, 1 can be prepared from [(η6-C6H6)(IiPr2Me2)RuI2] (4) as

follows: DBU (35.1 µL, 0.235 mmol) was added to a red suspension of complex

4 (0.075 g, 0.1223 mmol) and H2PMes* (0.0334 g, 0.1200 mmol) in toluene (20

mL) at room temperature. After 24 h, 31P{1H} NMR showed full conversion of

H2PMes* into 1. Pentane (5 mL) was added, the resulting dark green

suspension was filtered, and after concentration of the solvent, dark green

crystals of 1 were obtained at –20 °C (0.0471 g, 0.074 mmol, 63%).

[(ηηηη6-C6H6)(IiPr2Me2)RuI2] (4). RuI2-complex 4 was obtained from [(η6-

C6H6)RuCl2]2 and IiPr2Me2 and subsequent anion-exchange using NaI as

follows: a freshly distilled suspension of [(η6-C6H6)RuCl2]2 (0.275 g, 0.55 mmol)

in THF (40 mL) was treated with a solution of IiPr2Me2 (0.194 g, 1.075 mmol) in

THF (10 mL) at room temperature. After 3 days, the solvents were evaporated

and the dark brown residue was purified by column chromatography (silica,

CHCl3) and crystallization from CH2Cl2/pentanes at –20 °C yielded light

brown micro crystals of [(η6-C6H6)(IiPr2Me2)RuCl2] (0.140 g, 0.325 mmol, 30%).

Mp ≥ 193°C (dec). 1H NMR (250.13 MHz, CDCl3, 300 K): δ 1.48 (d, 3JHH = 7.1 Hz,

6 H, CH(CH3)2), 1.55 (d, 3JHH = 7.1 Hz, 6 H, CH(CH3)2), 2.27 (s, 6 H, =CCH3), 5.57

(s, 6 H, C6H6), 5.70 (sp, 3JHH = 7.1 Hz, 2 H, CH(CH3)2). 13C{1H} NMR (62.90 MHz,

CDCl3, 300 K): δ 10.7 (s, =CCH3), 23.0 (s, CH(CH3)2), 23.2 (s, CH(CH3)2), 53.5 (s,

CH(CH3)2), 85.9 (s, C6H6), 127.4 (s, =CCH3), 169.5 (s, N2C). HR FAB-MS: calcd for

[C17H26Cl2N2Ru]: 430.0515; found: 430.0512. MS m/z (%): 430 (5) [M+], 395 (55)

[M+ – Cl], 357 (75), 275 (27), 181 (100) [IiPr2Me2 + H+]. Subsequently, a solution

of [(η6-C6H6)(IiPr2Me2)RuCl2] (0.057 g, 0.135 mmol) and NaI (0.596 g, 3.97

mmol) in acetone (20 mL) was stirred overnight at room temperature. The

resulting dark purple mixture was evaporated and the residue was extracted

Chapter 2

62

with CH2Cl2, filtered and subsequent crystallization from CH2Cl2/pentanes at

–20 °C gave dark purple micro crystals of 4 (0.079 g, 0.128 mmol, 95%). Mp ≥

198 °C (dec). 1H NMR (250.13 MHz, CDCl3, 300 K): δ 1.46 (d, 3JHH = 7.0 Hz, 6 H,

CH(CH3)2), 1.53 (d, 3JHH = 7.0 Hz, 6 H, CH(CH3)2), 2.27 (s, 6 H, =CCH3), 5.67 (s, 6

H, C6H6), 5.94 (sp, 3JHH = 7.0 Hz, 2 H, CH(CH3)2). 13C{1H} NMR (62.90 MHz, CDCl3,

300 K): δ 11.1 (s, =CCH3), 23.4 (s, CH(CH3)2), 23.6 (s, CH(CH3)2), 56.4 (s,

CH(CH3)2), 86.3 (s, C6H6), 128.0 (s, =CCH3), N2C was not observed. HRMS (EI, 70

eV): calcd for [C17H26I2N2Ru] – [HI]: 486.0100; found: 486.0116. MS m/z (%): 486

(100) [M+ – HI], 404 [M+ – HI – 2iPr], 357 (28) [M+ – HI – I].

[(ηηηη6-C6H6)(IMes)Ru====PMes] (7a). A solution of IMes (0.459 g, 1.51 mmol) in

toluene (5 mL) was added to a brown toluene (5 mL) suspension of complex

[(η6-C6H6)RuCl2(PH2Mes)] (0.201 g, 0.50 mmol) at –78 °C. The mixture was

stirred for 1 h and was slowly allowed to warm up to room temperature. The

solvent was evaporated in vacuo and the residue was extracted with

pentanes (40 mL). After filtration (P4 glass filter, Microfibre Teflon and Celite)

and concentration, dark brown crystals of 7a were obtained at room

temperature (0.217 g, 0.343 mmol, 68%). Mp ≥ 190 °C (dec). 1H NMR (250.13

MHz, C6D6, 300 K): δ 2.07 (s, 6 H, o-CH3MesP), 2.14 (s, 6 H, p-CH3MesN), 2.28 (s,

12 H, o-CH3MesN), 2.34 (s, 3 H, p-CH3MesP), 4.43 (s, 6 H, C6H6), 6.42 (s, 2 H, N-

CH=), 6.73 (s, 4 H, m-MesN), 6.89 (s, 2 H, m-MesP). 13C{1H} NMR (100.64 MHz,

C6D6, 300 K): δ 19.6 (s, o-CH3MesN), 21.0 (s, p-CH3MesN), 21.3 (s, p-CH3MesP),

21.5 (s, o-CH3MesP), 79.4 (s, C6H6), 122.1 (s, =CH), 127.6 (s, m-MesP), 128.7 (s,

m-MesN), 131.5 (s, o-MesP), 133.3 (s, p-MesP), 136.2 (s, o-MesN), 138.0 (s, p-

MesN), 138.6 (s, i-MesN), 168.7 (d, 1JCP = 87.3 Hz, i-MesP), 190.4 (s, N2C). 31P{1H}

NMR (101.3 MHz, C6D6, 300 K): δ 752.5 (s, Ru=P). HRMS (EI, 70 eV): calcd for

C36H41N2PRu: 634.2051; found: 634.2054. MS m/z (%): 634 (20) [M+], 556 (22) [M+

– Bz], 484 (40) [M+ – PMes], 322 (100). Crystalline 7a contains, according to 31P

NMR, 2.7% of the corresponding toluene adduct 7b.

[(ηηηη6-pCy)(IMes)Ru====PMes] and [(ηηηη6-Tol)(IMes)Ru====PMes] (7b). A solution of

IMes (0.481 g, 1.415 mmol) in toluene (10 mL) was added to a toluene (5 mL)

suspension of complex [(η6-pCy)RuCl2(PH2Mes)] (0.2125 g, 0.464 mmol) at –78

°C. The mixture was stirred for 0.5 h and was slowly allowed to warm up to


63

room temperature. After two days, the solvent was evaporated in vacuo

and the dark brown solid was extracted with pentanes (20 mL), filtered (P4

glass filter, Teflon microfibre, Celite), and after concentration a mixture of

[(ηηηη6-pCy)(IMes)Ru====PMes] and 7b (5:1 ratio) was obtained as a dark brown

solid (0.180 g, 0.261 mmol, 56%). [(ηηηη6-pCy)(IMes)Ru====PMes]. 1H NMR (400.13

MHz, C6D6, 300 K): δ 0.84 (d, 3JHH = 6.9 Hz, 6 H, pCy-CH(CH3)2), 1.51 (s, 3 H, pCy-

CH3), 1.64 (sp, 3JHH = 6.9 Hz, 1 H, pCy-CH(CH3)2), 2.12 (s, 6 H, o-CH3MesP), 2.25

(s, 6 H, p-CH3MesN), 2.32 (s, 12 H, o-CH3MesN), 2.37 (s, 3 H, p-CH3MesP), 4.38

(d, 3JHH = 5.9 Hz, 2 H, C6H4), 4.65 (d, 3JHH = 5.9 Hz, 2 H, C6H4), 6.35 (s, 2 H, NCH=),

6.73 (s, 4 H, m-MesN), 6.94 (s, 2 H, m-MesP). 13C{1H} NMR (100.64 MHz, C6D6,

300 K): δ 19.5 (s, pCy-CH3), 19.7 (s, o-CH3NMes), 20.9 (s, o-CH3PMes), 21.3 (s, p-

CH3PMes), 21.5 (s, p-CH3NMes), 25.2 (s, pCy-CH(CH3)2), 32.2 (s, pCy-CH(CH3)2),

81.1, 83.1, 87.9, and 100.7 (s, C6H4), 122.3 (s, =CH), 127.5 (s, m-MesP), 128.9 (s,

m-MesN), 132.3 (s, o-MesP), 132.9 (s, p-MesP), 136.3 (s, i-MesN), 138.0 (s, o-

MesN), 138.9 (s, p-MesN), 168.3 (d, 1JCP = 100.0 Hz, i-MesP), 191.7 (s, N2C).

31P{1H} NMR (101.3 MHz, C6D6, 300 K): δ 724.9 (s, Ru=P). HRMS (EI, 70 eV): calcd

for [C40H49N2PRu] – [C9H11P]: 540.2072; found: 540.2091. MS m/z (%): 540 (16)

[M+ – PMes], 302 (10) [M+ – PMes – 2Mes]. 7b: 1H NMR (400.13 MHz, C6D6, 300

K): δ 1.55 (s, 3 H, C6H5CH3), 2.08 (s, 6 H, o-CH3PMes), 2.13 (s, 6 H, p-CH3NMes),

2.28 (bs, 12 H, o-CH3NMes), 2.35 (s, 3 H, p-CH3PMes), 4.28 (d, 3JHH = 5.6 Hz, 2 H,

C6H5CH3), 4.34 (t, 3JHH = 5.6 Hz, 1 H, C6H5CH3), 4.59 (t, 3JHH = 5.6 Hz, 2 H,

C6H5CH3), 6.40 (s, 2 H, NCH=), 6.73 (s, 4 H, m-MesN), 6.90 (s, 2 H, m-MesP).

13C{1H} NMR (100.64 MHz, C6D6, 300 K): δ 19.6 (s, o-CH3NMes), 19.9 (s, p-

CH3PMes), 20.9 (s, CH3), 21.3 (s, p-CH3PMes), 21.5 (s, o-CH3NMes), 74.9, 82.1,

82.2, and 90.5 (s, C6H5), 122.1 (s, =CH), 127.8 (s, m-MesP), 128.9 (s, m-MesN),

131.8 (s, o-MesP), 133.1 (s, p-MesP), 136.1 (s, i-MesN), 138.0 (s, o-MesN), 138.7

(s, p-MesN), 168.8 (d, 1JCP = 80.0 Hz, i-Mes), 191.7 (s, N2C). 31P{1H} NMR (101.3

MHz, C6D6, 300 K): δ 736.8 (s, Ru=P). HRMS (EI, 70 eV): calcd for [C37H43N2PRu] –

[C9H11P]: 498.1603; found: 498.1593. MS m/z (%): 498 (4) [M+ – PMes].

[(ηηηη6-pCy)(IMes)Ru====PMes] and [(ηηηη6-d3-Tol)(IMes)Ru====PMes] (d3-7b). A

solution of IMes (0.164 g, 0.54 mmol) in d3-toluene (0.8 mL) was added to a

d3-toluene (0.2 mL) suspension of complex [(η6-pCy)RuCl2(PH2Mes)] (0.0824 g,

Chapter 2

64

0.180 mmol) at –78°C. The mixture was allowed to warm up to room

temperature overnight and the resulting dark brown reaction mixture was

filtered (P4 glass filter), evaporated in vacuo and crystallized from

toluene/pentanes (respectively 1 and 5 mL) to obtain a mixture of [(ηηηη6-

pCy)(IMes)Ru====PMes] and d3-7b (5:1 ratio) as a dark brown solid at –20 °C

(0.0966 g, 0.142 mmol, 79%). d3-7b: 1H NMR (400.13 MHz, C6D6, 300 K): δ 2.08

(s, 6 H, o-CH3PMes), 2.13 (s, 6 H, p-CH3NMes), 2.28 (bs, 12 H, o-CH3NMes), 2.35

(s, 3 H, p-CH3PMes), 4.28 (d, 3JHH = 5.6 Hz, 2 H, C6H5CH3), 4.34 (t, 3JHH = 5.6 Hz, 1

H, C6H5CH3), 4.59 (t, 3JHH = 5.6 Hz, 2 H, C6H5CH3), 6.40 (s, 2 H, NCH=), 6.73 (s, 4

H, m-MesN), 6.90 (s, 2 H, m-MesP). 13C{1H} NMR (100.64 MHz, C6D6, 300 K): δ

19.5 (s, o-CH3NMes), 19.6 (s, p-CH3PMes), 21.4 (s, p-CH3PMes), 21.5 (s, o-

CH3NMes), 74.0, 82.0, 82.2, and 87.9 (s, C6H5), 122.2 (s, =CH), 127.6 (s, m-MesP),

129.3 (s, m-MesN), 131.8 (s, o-MesP), 133.1 (s, p-MesP), 136.3 (s, i-MesN), 138.0

(s, o-MesN), 138.7 (s, p-MesN), 168.8 (d, 1JCP = 80.0 Hz, i-MesP), 191.8 (s, N2C).

31P{1H} NMR (101.3 MHz, C6D6, 300 K): δ 736.6 (s, Ru=P). HRMS (EI, 70 eV): calcd

for C37H40D3N2PRu: 651.2398; found: 651.2380. MS m/z (%): MS m/z (%): 651 (4)

[M+], 556 (40) [M+ – d3-Tol], 436 (24) [M+ – d3-Tol – Mes], 304 (20).

[(ηηηη6-C6Me6)(IMes)Ru====PMes]. A solution of IMes (0.258 g, 0.85 mmol) in

toluene (1 mL) was added under argon to a toluene (2 mL) suspension of

complex [(η6-C6Me6)RuCl2(PH2Mes)] (0.136 g, 0.28 mmol) at –78°C. The

mixture was allowed to slowly warm up to room temperature and the

resulting dark brown reaction mixture was filtered (P4 glass filter) and

evaporated in vacuo. Extraction with pentanes (5 mL), filtration and

concentration at room temperature yielded [(ηηηη6-C6Me6)(IMes)Ru====PMes] as a

dark brown solid (0.116 g, 0.162 mmol, 58%). Mp ≥ 198 °C (dec). 1H NMR

(400.13 MHz, C6D6, 300 K): δ 1.66 (s, 18 H, C6(CH3)6), 2.14 (s, 6 H, o-CH3PMes),

2.18 (s, 6 H, p-CH3NMes), 2.21 (s, 6 H, o-CH3NMes), 2.36 (s, 3 H, p-CH3PMes),

2.52 (s, 6 H, o-CH3NMes), 6.33 (s, 2 H, =CH), 6.75 (s, 2 H, m-MesN), 6.79 (s, 2 H,

m-MesN), 6.90 (s, 2 H, m-MesP). 13C{1H} NMR (100.64 MHz, C6D6, 300 K): δ 16.5

(s, C6(CH3)6), 19.9 (s, o-CH3NMes), 21.0 (s, p-CH3NMes), 21.1 (s, o-CH3NMes),

21.4 (s, p-CH3PMes), 21.5, and 21.6 (s, o-CH3PMes), 90.9 (s, C6Me6), 122.8 (s,

=CH), 127.6 (s, m-MesP), 128.4, and 129.8 (s, m-MesN), 132.9 (s, p-MesP), 133.6


65

(bs, o-MesP), 135.2 (s, o-MesN), 137.3 (s, o-MesN), 137.5 (s, p-MesN), 139.3 (s, i-

MesN), 165.6 (d, 1JCP = 97.6 Hz, i-MesP), 192.9 (d, 2JCP = 13.0 Hz, N2C). 31P{1H}

NMR (101.3 MHz, C6D6, 300 K): δ 686.4 (s, Ru=P). MS EI m/z (%): 718 (<1) [M+,

correct isotope pattern], 566 (4), 303 (8), 162 (62), 147 (100).

Reactivity of [(ηηηη6-C6H6)(IiPr2Me2)Ru=PMes*] (1).

CH2I2 (0.153 mmol, 12.3 µL) was added to a dark green solution of 1 (19.5 mg,

0.0306 mmol) in C6D6 (0.4 mL) at 20 °C. Within 1.5 min., a brown precipitate

was formed and the quantitative formation of H2C=PMes* 6 was determined

by 31P NMR (t½ = 22 min at 0 °C in d8-toluene). The solvent was evaporated in

vacuo and the residue was extracted with pentanes (2 x 5 mL), subsequent

filtration and evaporation to dryness yielded H2C=PMes*[31] 6 as an off-white

solid (15.0 mg, 94%). Subsequent purification by column chromatography

(silica, CH2Cl2/pentanes) of the dark brown residue and crystallization from

CH2Cl2/pentanes at –20 °C yielded dark purple micro crystals of [(η6-

C6H6)(IiPr2Me2)RuI2] (4) (22.7 mg, 67%).

Reactivity of [(ηηηη6-C6H6)(PPh3)Ru=PMes*] (2).

CH2I2 (0.153 mmol, 12.3 µL) was added to a green solution of 2 (22 mg, 0.0306

mmol) in C6D6 (0.4 mL) at 20 °C. After 5h and 10 min. (t½ = 60 min; t½ = 925 min

at 0 °C in d8-toluene) all phosphinidene reacted to form H2C=PMes* 4

(45.3%), 5 (δ31P 21.4 (53.5%)), and several unidentified byproducts. 31P NMR

(101.3 MHz, C6D6, 293 K): δ 73.3 (s, 22.6%), 38 (s, 3.8%), 35.9 (d, JPP = 56.1 Hz,

7.5%), 25.6 (d, JPP = 56 Hz, 12.6%).

Acknowledgement. This work was partially supported by the Council for

Chemical Sciences of the Netherlands Organization for Scientific Research

(NWO/CW). The assistance from Dr. F. J. J. de Kanter (NMR), Dr. M. Smoluch

(HR EI-MS) and J. W. H. Peeters (HR FAB-MS; University of Amsterdam) is

gratefully acknowledged.

Chapter 2

66


[1] (a) D. Bourissou, O. Guerret, F. P. Gabbaï, G. Bertrand, Chem. Rev. 2000, 100, 39–

92. (b) M. F. Lappert, J. Organomet. Chem. 2005, 690, 5467–5473. (c) W. A.

Herrmann, Angew. Chem. Int. Ed. 2002, 41, 1290–1309. (d) N. M. Scott, S. P. Nolan,

Eur. J. Inorg. Chem. 2005, 1815–1828.

[2] (a) S. Diez-Gonzalez, S. P. Nolan, Coord. Chem. Rev. 2007, 251, 874–883. (b) H.

Jacobsen, A. Correa, A. Poater, C. Costabile, L. Cavallo, Coord. Chem. Rev.

2009, 253, 687–703. (c) S. Würtz, F. Glorius, Acc. Chem. Res 2008, 41, 1523–1533.

(d) N-Heterocyclic Carbenes in Synthesis; S. P. Nolan, Ed.; WILEY-VCH: Weinheim,

2006.

[3] (a) K. Getty, M. U. Delgado-Jaime, P. Kennepohl, J. Am. Chem. Soc. 2007, 129,

15774–15776. (b) M. S. Sanford, M. Ulman, R. H. Grubbs, J. Am. Chem. Soc. 2001,

123, 749–750. (c) M. S. Sanford, J. A. Love, R. H. Grubbs, J. Am. Chem. Soc. 2001,

123, 6543–6554.

[4] (a) C. Adlhart, P. Chen, Angew. Chem. Int. Ed. 2002, 41, 4484–4487. (b) B. F.

Straub, Angew. Chem. Int. Ed. 2005, 44, 5974–5978.

[5] (a) J. C. Slootweg, K. Lammertsma, In Science of Synthesis; B. M.Trost, F. Mathey,

Eds.; Georg Thieme Verlag: Stuttgart, 2009; Vol. 42, pp 15–36. (b) F. Mathey,

Dalton Trans. 2007, 1861–1868.


Chem. Eur. J. 2003, 9, 2200–2208.

[7] (a) F. Mathey, Angew. Chem. Int. Ed. 2003, 42, 1578–1604. (b) P. Le Floch, Coord.

Chem. Rev. 2006, 250, 627–681.

[8] (a) K. J. T. Noonan, D. P. Gates, Angew. Chem. Int. Ed. 2006, 45, 7271–7274. (b) D.

P. Gates, Top. Curr. Chem. 2005, 250, 107–126.

[9] (a) A. T. Termaten, T. Nijbacker, M. Schakel, M. Lutz, A. L. Spek, K. Lammertsma,

Organometallics 2002, 21, 3196–3202. (b) A. T. Termaten, H. Aktas, M. Schakel, A.

W. Ehlers, M. Lutz, A. L. Spek, K. Lammertsma, Organometallics 2003, 22, 1827–

1834.

[10] X-ray crystal structure determination of 1: C35H55N2PRu, Fw = 635.85, black

octahedral block, 0.30 x 0.30 x 0.27 mm3, orthorhombic, Pnma (no. 62), a =

18.9145(7), b = 17.2265(5), c = 10.4262(4) Å, V = 3397.2(2) Å3, Z = 4, Dx = 1.243

g/cm3, µ = 0.53 mm–1. 66347 Reflections were measured on a Nonius Kappa CCD

diffractometer with rotating anode (graphite monochromator, λ = 0.71073 Å) up

to a resolution of (sin θ/λ)max = 0.65 Å–1 at a temperature of 150(2) K. Intensities

were integrated with EvalCCD.[32] Absorption correction and scaling was


67

performed with SADABS[33] (0.71-0.87 correction range). 4041 Reflections were

unique (Rint = 0.035), of which 3557 were observed [I>2σ(I)]. The structure was

solved with Direct Methods using the program SIR-97.[34] The structure was refined

with SHELXL-97[35] against F2 of all reflections. Non hydrogen atoms were refined

with anisotropic displacement parameters. Hydrogen atoms were introduced in

calculated positions and refined with a riding model. The tert-butyl group was

rotationally disordered and refined with two orientations (with 50% occupancy,

each). 208 Parameters were refined with 6 restraints for distances and angles in

the disordered tert-butyl group. R1/wR2 [I > 2σ (I)]: 0.0335 / 0.0817. R1/wR2 [all

refl.]: 0.0408 / 0.0848. S = 1.172. Residual electron density between -0.34 and 0.97

e/Å3. The maximum residual density of 0.97 e/Å3 has a distance of 1.30 Å to P1

and is considered an artifact. Geometry calculations and checking for higher

symmetry was performed with the PLATON program.[36] Crystallographic

Information File (cif) with crystallographic data for compound 1 is available free

of charge via the Internet at http://pubs.acs.org.

[11] S. E. Lehman Jr., K. B. Wagener, Organometallics 2005, 24, 1477–1482.


[13] R. Menye-Biyogo, F. Delpech, A. Castel, H. Gornitzka, P. Rivière, Angew. Chem.

Int. Ed. 2003, 42, 5610–5612.

[14] Only 16-electron zirconium phosphinidenes have been detected by 31P NMR

spectroscopy, see: (a) J. Ho, Z. Hou, R. J. Drake, D. W. Stephan, Organometallics

1993, 12, 3145–3157. (b) A. Mahieu, A. Igau, J.-P. Majoral, Phosphorus, Sulfur

Silicon Relat. Elem. 1995, 104, 235–239.

[15] For complexes with a M≡P triple bond, see: (a) B. P. Johnson, G. Balázs, M.

Scheer, Top. Curr. Chem. 2004, 232, 1–23. (b) G. Balázs, L. J. Gregoriades, M.

Scheer, Organometallics 2007, 26, 3058–3075.

[16] When the para-cymene derivative [(η6-pCy)RuCl2(PH2Mes)] was used, 15% of 7b

was obtained (see the Experimental Section).

[17] (a) E. L. Muetterties, J. R. Bleeke, E. J. Wucherer, T. A. Albright, Chem. Rev. 1982,

82, 499–525. (b) M. A. Bennett, Z. Lu, X. Wang, M. Bown, D. C. R. Hockless, J. Am.

Chem. Soc. 1998, 120, 10409–10415.

[18] ADF2005.01 and ADF2005.01b, SCM, Theoretical Chemistry, Vrije Universiteit,

Amsterdam, The Netherlands, http://www.scm.com/.

[19] S. H. Vosko, L. Wilk, M. Nusair, Can. J. Phys. 1980, 58, 1200–1211.

[20] A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100.

Chapter 2

68

[21] J. P. Perdew, Phys. Rev. B 1986, 33, 8822–8824.

[22] (a) L. Fan, L. Versluis, T. Ziegler, E. J. Baerends, W. Raveneck, Int. J. Quantum

Chem. Quantum Chem. Symp. 1988, S22, 173–181. (b) L. Versluis, T. Ziegler, J.

Chem. Phys. 1988, 88, 322–328.

[23] E. van Lenthe, A. W. Ehlers, E. J. Baerends, J. Chem. Phys. 1999, 110, 8943–8953.

[24] (a) K. Morokuma, Acc. Chem. Res. 1977, 10, 294–300. (b) T. Ziegler, A. Rauk, Inorg.

Chem. 1979, 18, 1755–1759. (c) T. Ziegler, A. Rauk, Theor. Chim. Acta 1977, 46, 1–

10.

[25] R. A. Zelonka, M. C. Baird, Can. J. Chem. 1972, 50, 3063–3072.

[26] M. A. Bennett, T. N. Huang, T. W. Matheson, A. K. Smith, Inorg. Synth. 1982, 21, 74–

78.


[28] A. J. Arduengo, III, H. V. R. Dias, R. L. Harlov, M. Kline, J. Am. Chem. Soc. 1992, 114,

5530–5534.

[29] A. H. Cowley, J. E. Kilduff, T. H. Newman, M. Pakulski, J. Am. Chem. Soc. 1982, 104,

5820–5821.

[30] Y. van den Winkel, H. M. M. Bastiaans, F. Bickelhaupt, J. Organomet. Chem. 1991,

405, 183–194.

[31] R. Appel, C. Casser, M. Immenkeppel, F. Knoch, Angew. Chem. 1984, 96, 905–

906; Angew. Chem. Int. Ed. Engl. 1984, 23, 895–896.

[32] A. J. M. Duisenberg, L. M. J. Kroon-Batenburg, A. M. M. Schreurs, J. Appl. Cryst.

2003, 36, 220-229.

[33] G. M. Sheldrick, 1999, SADABS: Area-Detector Absorption Correction, v2.10,

Universität Göttingen, Germany.

[34] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giacovazzo, A.

Guagliardi, A. G. G. Moliterni, G. Polidori, R. Spagna, J. Appl. Cryst. 1999, 32, 115-

119.

[35] G. M. Sheldrick, Acta Cryst. 2008, A64, 112-122.

[36] A. L. Spek, J. Appl. Cryst. 2003, 36, 7-13.

N-Heterocyclic Carbene Functionalized Group 7–9 Transition

Metal Phosphinidene Complexes

Halil Aktas,† J. Chris Slootweg,† Andreas W. Ehlers,† Martin Lutz,‡ Anthony L.

Spek,‡ and Koop Lammertsma*,†




Organometallics 2009, 28, 5166–5172

Abstract:Abstract:Abstract:Abstract: The N-heterocyclic carbene (NHC) functionalized phosphinidene complexes

[L(NHC)M=PR] (M = Ru, Os, Rh, Ir) of the Group 8-9 transition metals were synthesized. The effect

of the NHC ligand on the electronic properties of the phosphinidene complexes

[(Ring)(NHC)M=PH] 8888–16161616 bearing group 7-9 transition metals and cycloheptatrienyl (Cht+),

benzene, cyclopentadienyl (Cp–) as ancillary ligands were studied by density functional theory. On

going to the right in the Periodic Table, the structures show an increase in M–NHC bond energy

that concurs with the net charge on the phosphorus atom of the M=P bond. The NHC ligand effects

the M=P properties and interacts via the metal with the ancillary ligand L (Cht+, Bz, Cp–).

ChapterChapterChapterChapter 3 3 3 3

Chapter 3

70

3.1 Introduction

Transition metal complexed phosphinidenes display carbene-like

behavior that has enabled the synthesis of a plethora of organo-

phosphorus compounds.[1] Transition metals with strong σ-donor

ancillary ligands render nucleophilic complexes[2] akin to Schrock-

type carbene complexes. Several synthetic approaches to these

isolable species are known, such as the condensation of

lithiumphosphides to halogenated transition metal complexes,[3]

oxidation of Ni-phosphides followed by deprotonation,[4] and addition

of phosphides to alkylidene complexes with a concurrent α-H-

migration.[5] We developed a general route for group 8 (Ru and Os)[6]

and group 9 (Co, Rh,[7] and Ir[8]) complexes entailing the double

dehydrohalogenation of primary phosphine precursors with DBU (1,8-

biazabicyclo-[5.4.0]-undec-7-ene) in the presence of stabilizing

ligands (PR3, P(OR)3 AsR3, CO, RN≡C, dppe). Reacting these

nucleophilic phosphinidenes with gem-dihalogens gives access to

phosphaalkenes (P=C),[9] which are attractive building blocks for P-

functionalized polymers[10] and as catalyst ligands.[11] Stimulated by

the N-heterocyclic carbene (NHC) ligated ruthenium alkylidene

complexes that are efficient catalysts in olefin metathesis,[12,13] we

recently reported the first NHC-functionalized Ru-complexed

phosphinidene [(η6-C6H6)(IiPr2Me2)Ru=PMes*] (1; Mes* = 2,4,6-

tBu3C6H2), which displays enhanced reactivity toward CH2I2 in

comparison to the corresponding phosphine analogue 2 (Scheme 1)

due to the sterically imposed ligand rotation of the NHC fragment.[14]

We also reported the first related iridium complexed phosphinidene

(3) using the lithiumphosphide method (Scheme 1).[3a] In these studies

we showed that NHCs,[15] which are commonly regarded as strong σ-

donor ligands, can also act as π-acceptor ligands, a feat that hitherto

N-Heterocyclic Carbene Functionalized Group 7–9 Transition Metal Phosphinidene Complexes

71

had been related to the magnitude of the charge density on the

metal.[16,17] For 1, we found that the relative σ-donor/π-acceptor

ability of the NHC ligand can be tuned by changing its substituents,

thereby stabilizing either the “in-plane” arrangement of the NHC with

the Ru=P bond for maximum π-overlap or the “out-of-plane”

conformation that makes the NHC an effective σ-donor.[14] This simple

substituent-controlled conformational change of the transition metal

ligand impacts the reactivity of the phosphinidene complex, as was

demonstrated for the synthesis of phosphaalkenes.[14]

Scheme 1. Synthesis of NHC-functionalized phosphinidenes 1 and 3.

In the present study, we report on the synthetic access to NHC-

functionalized phosphinidene complexes of the group 8 and 9 metals

and analyze by DFT calculations the intrinsic electronic properties of a

comprehensive set of model compounds, (Ring)M(NHC)=PH, with a

broader set of transition metals (M = group 7 (Mn, Tc, Re), group 8 (Fe,

Ru, Os), group 9 (Co, Rh, Ir)) and ligands (Ring = cycloheptatrienyl

(Cht), benzene (Bz), cyclopentadienyl (Cp)).

Chapter 3

72

3.2 Results and Discussion

First, we describe the synthesis and characterization of novel

ruthenium phosphinidenes with different η6-arene and NHC ligands

and with different phosphine substituents. Next, we show that the

osmium, rhodium, and iridium complexed phosphinidenes can also

be accessed by the double dehydrohalogenation-ligation sequence

using NHCs. Finally, we perform a detailed theoretical analysis to

unveil the intrinsic electronic properties, reactivities, and stabilities of

these systems using (Ring)M(NHC)=PH as model for a more complete

set of transition metals of groups 7-9.

3.2.1 Synthesis of NHC-functionalized group 8 phosphinidenes

The NHC- and p-Cymene (pCy)-ligated ruthenium-complexed

phosphinidene [(η6-pCy)(IiPr2Me2)Ru=PMes*] 4 (IiPr2Me2 = 1,3-

diisopropyl-4,5-dimethylimidazol-2-ylidene) was obtained, like

ruthenium phosphinidene 1,[14] by a double dehydrohalogenation-

ligation sequence of the primary phosphine [(η6-pCy)RuCl2(PH2Mes*)]

using three equivalents of IiPr2Me2 in toluene at –78 °C (Scheme 2).

Crystallization from n-hexane at –20 °C yielded dark green thermally

stable crystals (60%; 208 °C, dec.) that are air and moisture sensitive.

Indicative for the formation of 4 is the single 31P NMR resonance at

733.5 ppm (1:[14] δ31P 751.7 ppm), identifying a bent phosphinidene

complex,[6-8] and the characteristic 13C NMR resonance for the Ru–

carbene at 188.7 ppm (1:[14] δ13C 187.9 ppm). In addition to the arene

ligand, also the NHC and the P substituent can be varied, as

illustrated by the reaction of the less crowded Ru complexes [(η6-

Ar)RuCl2(PH2Mes)] (Ar = pCy (a), C6Me6 (b)), having mesityl instead of

supermesityl substituents, with the bulkier 1,3-dimesityl-imidazol-2-

ylidene (IMes). The isolated products 5a,b (δ31P 724.9 (a, 56%),[18] 686.4


73

ppm (b, 58%)) are dark brown solids with typical carbene resonances

(δ13C 191.7 (a), 192.9 ppm (b)) and the common E configuration for

the Ru=P bond, as established by NOE experiments (Scheme 2).

P

Mes*

Ru

NN

4

PH2

Mes*

Ru

ClCl

3 IiPr2Me2

- 2 IiPr2Me2.HCl

L

P

Mes

Ru

NN

Mes

Mes

L

PH2

Mes

Ru

ClCl- 2 IMes.HCl

a, L = pCyb, L = C6Me6

5

3 IMes

Scheme 2. Synthesis of NHC-functionalized ruthenium phosphinidenes 4 and 5.

The facile double dehydrohalogenation-ligation sequence, with

two equivalents of NHC acting as a Brønsted base and one

equivalent as a carbene ligand, which affords 18-electron NHC-

functionalized phosphinidenes, can be readily extended to other

transition metals if their primary phosphine precursor complexes are

available.[19] For instance, reaction of the phosphine complex [(η6-

pCy)OsCl2(PH2Mes*)] with three equivalents of IiPr2Me2 in toluene

resulted in the formation of the osmium phosphinidene complex [(η6-

pCy)(IiPr2Me2)Os=PMes*] (6), which was isolated as orange crystals in

85% yield (Scheme 3). The single 31P NMR resonance of 6 at 557.6 ppm

is highly shielded in comparison to its triphenylphosphine analogue

(δ31P 667.5 ppm),[6] in analogy to the corresponding ruthenium

complexes,[14] an may be attributed to the enhanced σ-donor

capacity of the NHC.

Chapter 3

74

3.2.2 Synthesis of NHC-functionalized group 9 phosphinidenes

NHC-functionalized group 9 phosphinidene complexes are equally

accessible by this protocol. Thus, rhodium phosphinidene [(η5-

Cp*)(IiPr2Me2)Rh=PMes*] (7; δ31P 745.9 ppm) was obtained by the

reaction of [(η5-Cp*)RhCl2(PH2Mes*)] with three equivalents of IiPr2Me2

in toluene and isolated as dark pink crystals from n-pentane (88%;

Scheme 3). Likewise, iridium phosphinidene 3[3a] was generated from

[(η5-Cp*)IrCl2(PH2Mes*)] and IiPr2Me2 (60%), thereby offering a facile

alternative to the lithiumphosphide method. Unfortunately, the

attempted synthesis of cobalt phosphinidene [(η5-

Cp)(IiPr2Me2)Co=PMes*] from the readily available precursor [(η5-

Cp)CoI2(PH2Mes*)][7] only led to unidentifiable products.

Scheme 3. General synthesis of NHC-functionalized phosphinidene complexes using

NHCs.

The Cs-symmetrical structure of rhodium phosphinidene 7 was

ascertained by single-crystal X-ray analysis (Figure 1).[20] The geometry

around rhodium shows a two-legged piano stool with a characteristic

acute C18–Rh1–P1 angle (83.94(7)°), a Rh1–P1–C1 angle (110.19(8)°)

that typifies a bent phosphinidene complex, and an E configuration

for the Rh1–P1 double bond (2.1827(7) Å). This latter bond is similar to

that of the triphenylphosphine analogue [(η5-Cp*)(Ph3P)Rh=PMes*]

(2.1903(4) Å),[7] whereas its Ru1–Cp(cg) bond is significantly shorter

(1.8878(10) vs 1.9253(9) Å[7]). The “out-of-plane” conformation of the


75

NHC fragment, like that of 1[14] and 3,[3a] is caused by steric

congestion induced by the isopropyl wingtip groups of the ligand that

prevent the coplanar arrangement.[14]

Figure 1. Molecular structure of 7, displaying displacement ellipsoids at the 50%

probability level. Only one conformation of the disordered p-tert-butyl group is

shown. H-atoms and disordered solvent (pentane) have been excluded for clarity.

Cp(cg) denotes the centroid position of the pentamethylcyclopentadienyl ring.

Selected bond distances (Å) and angles and torsion angles (°): Rh1–P1 2.1827(7),

Rh1–C18 2.036(2), Rh1–Cp(cg) 1.8878(10), P1–C1 1.858(2), N1–C18 1.357(2), N1–C19

1.393(2), C19–C19a 1.342(4); Rh1–P1–C1 110.19(8), P1–Rh1–Cp(cg) 147.47(4), P1–Rh1–

C18 83.94(7), C18–Rh1–Cp(cg) 128.60(7); C2a–C1–C2–C3 –14.9(3). Symmetry

operation (a) x, 0.5 − y, z.

3.3 Theoretical calculations on [(Ring)M(NHC)====PH]

The model compound [(Ring)M(NHC)=PH] was investigated

computationally for the influence of group 7–9 transition metals (M =

Mn (8), Tc (9), Re (10), Fe (11), Ru (12), Os (13), Co (14), Rh (15), Ir (16);

Figure 2) and the stabilizing ligand (Ring = cyclopentadiene anion

(Cp−), neutral benzene (Bz) and cycloheptatriene cation (Cht+)) on

the electronic properties of NHC-functionalized phosphinidenes. We

Chapter 3

76

first discuss their structural properties and then present charge

decomposition analyses.

3.3.1 Geometries.

For each complex, all four possible geometrical isomers were

calculated with the NHC ligand “in-plane” and “out-of-plane” with

both the E and Z arrangements of the M=P double bond (see the

Appendix 2). The lowest energy Cs-symmetric structures of the NHC-

functionalized phosphinidene complexes 8–16 on which we

concentrate here all have similar two-legged piano-stool geometries

with a coplanar, NHC “in-plane” arrangement with the E conformer

of the M=P bond (Figure 2); selected bond lengths and angles are

listed in Table 1. All data for the less stable “out-of-plane” E, “in-

plane” Z, and “out-of-plane” Z isomers are given in the Appendix 2.

M P

N N

C

M P

N N

C

M

P

N N

C

8a M=Mn

9a M=Tc

10a M=Re

11a M=Fe

12a M=Ru

13a M=Os

14a M=Co

15a M=Rh

16a M=Ir

M P

N N

C

M P

N N

C

M

P

N N

C

8a M=Mn

9a M=Tc

10a M=Re

11a M=Fe

12a M=Ru

13a M=Os

14a M=Co

15a M=Rh

16a M=Ir

Figure 2. BP86/TZP optimized “in-plane” E structures (Cs symmetry) for

[(Ring)(NHC)M=PH] (Ring = Cht, Bz, Cp; M = Mn (8a), Tc (9a), Re (10a), Fe (11a), Ru

(12a), Os (13a), Co (14a), Rh (15a), and Ir (16a)).


77

The first-row TM phosphinidenes 8a, 11a, and 14a have rather short

M=P distances (2.091–2.136 Å) and acute ∠M=P–H (104.8–109.8°) and

∠(NHC)C–M=P (85.9–90.8°) angles (Table 1). The calculated M=P

distances follow the size of the TMs (Mn > Fe > Co), as do the (NHC)C–

M bond lengths (2.034–1.884 Å). The second- (2.233–2.199 Å) and

third-row TM (2.255–2.215 Å) phosphinidenes have slightly longer M=P

bond distances. As expected, the (NHC)C–M bonds and M–Ring

distances are shorter for the third-row complexes 10a, 13a, and 16a

than for those of the second row, due to relativistic contraction.

3.3.2 Energy Decomposition Analysis

The metal–NHC bond dissociation energies (BDEs) for the model

complexes (for details see the Computational Section), tabulated in

Table 2, depend on the transition metal and increases on going from

left to right and from top to bottom in the Periodic Table.[21] Thus, the

Mn phosphinidene 8a has the weakest M–NHC bond (34.0 kcal·mol–1)

and Ir complex 16a the strongest bond (60.6 kcal·mol–1; Figure 3). The

same applies to the σ- (∆Ea’) and π-orbital interactions (∆Ea’’) of the

M–NHC bond, which also increase from left to right and from top to

Table 1. Selected BP86/TZP bond lengths (Å) and angles (deg) for model complexes [(Ring)(NHC)M=PH] 8a–16a.

M M=P M–C(NHC) M–Ring M=P–H (NHC)C–M=P

8a Mn 2.136 2.034 1.484 109.8 85.9 9a Tc 2.233 2.141 1.652 111.2 85.3

10a Re 2.255 2.118 1.668 110.2 85.2

11a Fe 2.115 1.952 1.574 106.9 87.5

12a Ru 2.208 2.068 1.782 107.7 85.6

13a Os 2.227 2.061 1.788 107.1 85.9

14a Co 2.091 1.884 1.717 104.8 90.8

15a Rh 2.199 2.022 1.980 104.3 86.9

16a Ir 2.215 1.999 1.944 104.2 87.7

Chapter 3

78

bottom in the Periodic Table (Table 2). The average π-contribution is

~14% of the total orbital interaction (∆E oi) for the M–NHC bond of the

group 7 TMs, whereas this amounts to ~18% for the group 8 and group

9 TMs, indicating that π-backdonation from the metal to the NHC

ligand is substantial for 8a–16a and that it cannot be neglected.[17c]

Mn Tc Re Fe Ru Os Co Rh Ir

BDE (kcal/mol)

65

60

55

50

45

40

35

30

Mn Tc Re Fe Ru Os Co Rh Ir

BDE (kcal/mol)

65

60

55

50

45

40

35

30

Figure 3. M–NHC Bond Dissociation Energies (kcal/mol) for first-row (solid line),

second-row (dashed line), and third-row (dotted line) phosphinidene complexes.

Table 3 gives the data for the metal–phosphinidene bond. The

bond dissociation energies for the first-row metal complexes 8a, 11a

and 14a amount to ~70 kcal·mol–1 and increase for the heavier

analogues, being more pronounced for the group 7 metals than

those for group 8 and 9, e.g., 9a (91.5) > 12a (88.2) > 15a (82.4

kcal·mol–1). The phosphinidene group is a strong ligand in all

[(Ring)M(NHC)=PH] complexes with a high σ-contribution of 74–78% of

the total orbital interaction (∆Eoi), which is in agreement with earlier

findings.[2,3a,7,14] Both the σ- and π-contributions follow the trend noted

for the BDEs. Evidently, the yet unknown phosphinidene complexes 8–

11 have strong M=P double bonds.

N-H

etero

cyclic C

arb

ene Functionaliz

ed G

roup 7

–9 Tra

nsitio

n M

etal Phosp

hinidene C

omplexe

s

79

Table 3

. BP8

6/TZ

P Bo

nd

Dis

soc

iatio

n E

ne

rgie

s a

nd

En

erg

y D

ec

om

po

sitio

n A

na

lysi

s fo

r (R

ing

)(N

HC

)M=PH

do

ub

le b

on

d.

Ene

rgie

s a

re in

kc

al/

mo

l.

M

∆

Ea

’ ∆

Ea

” ∆

E st

eric

∆

E to

t ∆

E p

rep

BDE

8a

Mn

–1

01.4

–3

0.7

55.4

–7

6.7

7.5

69.2

9a

Tc

–120

.9

–41.

7 65

.6

–97.

0 5.

5 91

.5

10a

Re

–1

33.4

–4

6.5

71.9

–1

08.0

7.

6 10

0.4

11a

Fe

–102

.6

–28.

9 57

.0

–74.

5 4.

0 70

.5

12a

Ru

–1

15.3

–3

6.9

60.6

–9

1.6

3.5

88.2

13a

Os

–127

.5

–43.

0 66

.0

–104

.5

5.8

98.7

14a

Co

–9

7.2

–27.

4 50

.4

–74.

2 2.

2 72

.0

15a

Rh

–1

05.5

–3

1.6

52.6

–8

4.5

2.1

82.4

16a

Ir –1

21.0

–3

8.2

60.6

–9

8.6

3.8

94.8

Table 2

. BP8

6/TZ

P Bo

nd

Dis

soc

iatio

n E

ne

rgie

s (in

kc

al/

mo

l) a

nd

En

erg

y D

ec

om

po

sitio

n A

na

lysi

s fo

r LnM

–C(N

HC

) b

on

ds

of

the

Gro

up

7–9

c

om

ple

xes.

Va

lue

s in

pa

ren

the

ses

refe

r to

ch

arg

e (

e– )

tra

nsf

err

ed

to

(p

osi

tive

) a

nd

fro

m (

ne

ga

tive

) th

e t

ran

sitio

n m

eta

l ce

nte

r.

M

∆

Ea

’ ∆

Ea

” ∆

E st

eric

∆

E to

t BD

E

8a

Mn

–4

6.4

(0.5

5)

–7.7

(–0

.09)

6.

0 –4

8.1

34.0

9a

Tc

–52.

7 (0

.46)

–8

.8 (

–0.0

9)

10.3

–5

1.3

42.2

10a

Re

–6

6.6

(0.4

9)

–11.

8 (–

0.11

) 11

.5

–66.

9 49

.9

11a

Fe

–50.

9 (0

.59)

–1

1.6

(–0.

16)

10.8

–5

1.6

42.8

12a

Ru

–5

5.8

(0.4

7)

–11.

7 (–

0.15

) 14

.5

–53.

0 45

.5

13a

Os

–71.

2 (0

.49)

–1

5.1

(–0.

18)

17.1

–6

9.3

55.7

14a

Co

–5

3.8

(0.6

3)

–13.

7 (–

0.18

) 9.

5 –5

8.1

51.6

15a

Rh

–5

8.8

(0.4

8)

–12.

6 (–

0.15

) 14

.5

–56.

9 49

.5

16a

Ir –7

9.4

(0.4

9)

–17.

3 (–

0.19

) 23

.5

–73.

3 60

.6

Chapter 3

80

The reactivity of the phosphinidene complexes toward electron-rich

or electron-poor reagents is directed by the charge, which is

influenced by the transition metal fragment and can be deferred

from the energies of the bond-forming orbitals. The energy for the

singly occupied p orbital of 3PH amounts to –5.5 eV.

The corresponding energies of the [Ring][NHC]M fragments are

collected in Table 4. The major trend for the LnM fragment is the

decreasing energy of the σ-orbital on going down (Mn –2.13 > Tc –

2.24 > Re –2.44 eV) or to the right (Mn –2.13 > Fe –2.23 > Co –2.55) of

the Periodic Table (Figure 4). The π-orbital energies follows the same

trend on going from the first- to the second-row TM but increase

again for the third-row TM, which may be caused by relativistic

effects. Consequently, the net charge on phosphorus (Table 4)

decreases in the row group 7 > group 8 > group 9. The second- and

third-row complexes show an increased charge on P as compared to

the first-row congeners. This effect should be reflected in an

increased nucleophilicity that indeed matches our experimental

observations. The Ru and Rh phosphinidene complexes are the most

reactive ones of the group 8 and 9 triads.[3a,7,14]

We briefly return to the effect of “in-plane” and “out-of-plane”

orientation of the NHC ligand on the electronic properties of the

complexes. In earlier studies, we found that the relative σ-donor/π-

acceptor ability of the NHC ligand in ruthenium complexes can be

tuned by changing its substituents, thereby stabilizing either the “in-

plane” arrangement of the NHC with the Ru=P bond for maximum π-

overlap or the “out-of-plane” conformation that makes the NHC an

more effective σ-donor.[14] Unfortunately, a direct discrimination of

the σ and π orbital contributions is not possible for the “out-of plane”

conformation since both fall into the a’ representation. However,


81

from the amount of charge transferred (see the Appendix 2) it is

apparent that in this case π-backdonation is reduced compared to

the “in-plane” arrangement for the group 8 and 9 complexes while it

is negligible for the NHC-bond to metals for group 7.

Mn Fe Co Tc Ru Rh Re Os Ir

-2.0

-2.1

-2.2

-2.3

-2.4

-2.5

-2.6

-2.7

-2.8

-2.9

-3.0

Orbital Energy (eV)

Mn Fe Co Tc Ru Rh Re Os Ir

-2.0

-2.1

-2.2

-2.3

-2.4

-2.5

-2.6

-2.7

-2.8

-2.9

-3.0

Orbital Energy (eV)

Figure 4. σ- (♦) and π- (●) Orbital energies ε (eV) for the (Ring)(NHC)M fragment of

first-row (solid line), second-row (dashed line), and third-row (dotted line)

phosphinidene complexes.

Table 4. BP86/TZP occupied SOMO energies ε (eV) for the [(Ring)(NHC)M] fragments and calculated Hirshfeld charges of M and P.

M Ed (σ) Ed (π) M P

8a Mn –2.13 –2.35 +0.01 –0.15 9a Tc –2.24 –2.45 +0.04 –0.17

10a Re –2.44 –2.38 +0.13 –0.21

11a Fe –2.23 –2.51 –0.05 –0.12

12a Ru –2.37 –2.62 +0.15 –0.18

13a Os –2.56 –2.50 +0.03 –0.16

14a Co –2.55 –2.78 –0.07 –0.10

15a Rh –2.65 –2.95 +0.11 –0.14

16a Ir –2.78 –2.77 –0.01 –0.13

Chapter 3

82

3.4 Conclusion

The scope of the dehydrohalogenation−ligation sequence using

NHCs both as Brønsted base and as stabilizing ligand has been

successfully extended by the synthesis of novel ruthenium, osmium,

and rhodium phosphinidene complexes. The solid-state structure of

rhodium phosphinidene 7 shows an “out-of-plane” conformation of

the NHC fragment with the (E)-M=P bond and is similar to that

reported for Ir and Ru complexes 1 and 3. The DFT calculations

present insightful details. The orbital energies of the metal–NHC bond

are dominated by a σ-interaction, but the π-interaction is also

substantial (~20%) and thus not negligible. These interactions are

influenced by the ancillary ligand, which was illustrated by a slight

change in σ- and π-capacity due to the Cht+ and Cp– ligands. For the

group 7–9 metals, the BDEs and strength of the σ- and π-interactions

follow the order of the transition metals in the Periodic Table.

According to these theoretical calculations, the rhenium, ruthenium,

and rhodium phosphinidene complexes are expected to be the most

reactive ones of the group 7–9 triad metals.


All DFT calculations have been performed with the parallelized Amsterdam

density functional (ADF) package (version 2006.01).[22] The Kohn–Sham MOs

were expanded in a triple–ζ Slater–type basis set with polarization functions

(TZP). The 1s core shell of carbon and nitrogen, the 1s2s2p core shells of

phosphorus and a ‘small’ core of the TM were treated by the frozen core

approximation. All calculations were performed at the nonlocal exchange

self–consistent field (NL–SCF) level, using the local density approximation

(LDA) in the Vosko–Wilk–Nusair parametrization[23] with nonlocal corrections

for exchange (Becke88)[24] and correlation (Perdew86).[25] All geometries

were optimized using the analytical gradient method implemented by


83

Versluis and Ziegler,[26] including relativistic effects by the Zero Order Regular

Approximation (ZORA).[27]

Bonding Energy Analysis. The bonding interactions of the transition metal

to ligand bonds (M–P, M–C(NHC), and M–Ring) were analyzed with ADF's

established energy decomposition[28] into an exchange (or Pauli) repulsion

(∆E Pauli) between the electrons on the two fragments plus electrostatic

interaction energy part (∆E elst) and an orbital interaction energy (charge

transfer, polarization) part (∆E oi). The energy necessary to convert fragments

from their ground state equilibrium geometries to the geometry and

electronic state they acquire in the complex is represented by a preparation

energy term (∆E prep). The overall bond energy (∆E tot) is formulated as:

∆E tot = ∆E Pauli+∆E elst+∆E oi+∆E prep

Note that ∆E tot is defined as the negative of the bond dissociation energy

(BDE), i.e., ∆E tot=E (molecule) – ΣE (fragments), thereby giving negative

values for stable bonds. The orbital interaction term ∆E oi accounts for

interactions between occupied orbitals on one fragment with unoccupied

orbitals on the other fragment, including HOMO-LUMO interactions and

polarization (empty/occupied orbital mixing on the same fragment). The

extended transition state (ETS) method developed by Ziegler and Rauk was

used to decompose ∆E oi into contributions from each irreducible

representation of the interacting system.[28] In systems with a clear σ, π-

separation (a' and a”), this symmetry partitioning proves to be most

informative.



atmosphere of dry nitrogen with rigorous exclusion of air and moisture using

flame–dried glassware using Schlenk techniques. NMR spectra were

recorded on a Bruker Advance 250 (1H, 13C, 31P; 85% H3PO4) or a Bruker

Advance 400 (1H, 13C, 31P; 85% H3PO4) and referenced internally to residual

solvent resonances (CDCl3: 1H: δ 7.26, 13C{1H}: δ 77.16; C6D6: 1H: δ 7.16,

Chapter 3

84

13C{1H}: δ 128.06). IR Spectra were recorded on a Mattson–6030 Galaxy FT–IR

spectrophotometer, high–resolution mass spectra (HR–MS) on a Finnigan Mat

900 spectrometer operating at an ionization potential of 70 eV. Melting

points were measured on samples in sealed capillaries on a Stuart Scientific

SMP3 melting point apparatus and are uncorrected.

Reagents. Solvents were distilled from sodium (toluene), P2O5 (CH2Cl2), or

LiAlH4 (pentanes, diethyl ether) and kept under an atmosphere of dry

nitrogen. Deuterated solvents were dried over 4 Å molecular sieves (CDCl3,

C6D6). All solid starting materials were dried in vacuo. [(η6-C6Me6)RuCl2]2,[29]

[(η6-pCy)RuCl2(PH2Mes*)],[6] [(η6-pCy)OsCl2(PH2Mes*)],[6] [(η5-

Cp)CoI2(PH2Mes*)],[7] [(η5-Cp*)RhCl2(PH2Mes*)],[7] [(η5-Cp*)IrCl2(PH2Mes*)],[8]

1,3-diisopropyl-4,5-dimethylimidazol-2-ylidene (IiPr2Me2),[30] 1,3-dimesityl-

imidazol-2-ylidene (IMes),[31] and Mes*PH2[32] were prepared according to

literature procedures. MesPH2 was prepared, analogously to IsPH2,[33] by

LiAlH4 reduction of MesPCl2. [(η6-pCy)RuCl2]2 was purchased from Strem and

used as received.

[(ηηηη6-pCy)RuCl2(PH2Mes)]. MesPH2 (0.219 g, 1.437 mmol) was added to a

dark red suspension of [(η6-pCy)RuCl2]2 (0.400 g, 0.653 mmol) in CH2Cl2 (25

mL), which was stirred for 1 h at room temperature and filtered to remove

insoluble material. After concentrating the solution to a few mL, pentanes

(30 mL) was added slowly, which resulted in precipitation of an orange

powder that was isolated by filtration, washed with pentanes (25 mL), and

dried in vacuo to yield [(η6-pCy)RuCl2(PH2Mes)] (0.561 g, 1.22 mmol, 94%).

Mp ≥ 163 °C (dec). 1H NMR (250.13 MHz, CDCl3, 300 K): δ 1.19 (d, 3JHH = 7.0 Hz,

6 H, CH(CH3)2), 1.99 (s, 3 H, pCy-CH3), 2.34 (s, 3 H, p-CH3), 2.50 (s, 6 H, o-CH3),

2.70 (sp, 3JHH = 7.0 Hz, 1 H, CH(CH3)2), 5.13 (d, 3JHH = 5.7 Hz, 2 H, C6H4), 5.34 (d,

3JHH = 5.7 Hz, 2 H, C6H4), 5.52 (d, 1JHP = 389.8 Hz, 2 H, PH2); 7.01 (d, 4JHP = 2.3 Hz,

2 H, m-Mes). 13C{1H} NMR (62.90 MHz, CDCl3, 300 K): δ 18.4 (s, pCy-CH3), 21.6

(d, 5JCP = 1.1 Hz, p-CH3), 22.5 (s, CH(CH3)2), 22.9 (d, 3JCP = 8.2 Hz, o-CH3), 31.0

(s, CH(CH3)3), 85.5 (d, 2JCP = 5.7 Hz, C6H4), 89.1 (d, 2JCP = 5.3 Hz, C6H4), 97.6 (s,

C6H4), 107.7 (d, 2JCP = 1.0 Hz, C6H4), 122.3 (d, 1JCP = 42.3 Hz, i-Mes), 130.1 (d,


85

3JCP = 7.8 Hz, m-Mes), 141.4 (d, 2JCP = 7.7 Hz, o-Mes), 141.5 (d, 4JCP = 2.9 Hz, p-

Mes). 31P NMR (101.3 MHz, CDCl3, 300 K): δ –48.0 (t, 1JPH = 389.9 Hz, PH2). IR

(KBr, cm–1): v = 3043.0 (w), 2967.2 (m), 2913.3 (w), 2864.7 (w), 2387.6 (m, P–H),

2353.7 (m, P–H), 1601.4 (m), 1437.9 (s), 1382.3 (s). HR FAB-MS: calcd for

C19H27PCl2Ru: 458.0271; found: 458.0270. MS m/z (%): 458 (70) [M+], 423 (100)

[M+ – Cl], 388 (100) [M+ – 2Cl], 271 (35) [M+ – Cl – H2PMes], 154 (90) [H2PMes].

[(ηηηη6-C6Me6)RuCl2(PH2Mes)]. Analogous to that described for [(η6-

pCy)RuCl2(PH2Mes)], [(η6-C6Me6)RuCl2]2 (1.01 g, 1.50 mmol) and MesPH2

(0.464 g, 3.05 mmol) were used to give [(η6-C6Me6)RuCl2(PH2Mes)] as a

yellow powder (1.29 g, 2.65 mmol, 88%). Mp ≥ 257 °C (dec). 1H NMR (250.13

MHz, CDCl3, 300 K): δ 1.92 (s, 18 H, C6(CH3)6), 2.29 (s, 3 H, p-CH3), 2.39 (s, 6 H,

o-CH3), 5.45 (d, 1JHP = 379.7 Hz, 2 H, PH2), 6.95 (d, 4JHP = 2.1 Hz, 2 H, m-Mes).

13C{1H} NMR (62.90 MHz, CDCl3, 300 K): δ 15.7 (s, C6(CH3)6), 21.6 (s, p-CH3),

23.4 (d, 3JCP = 8.0 Hz, o-CH3), 96.1 (d, 3JCP = 3.5 Hz, C6(CH3)6), 119.3 (d, 1JCP =

38.3 Hz, i-Mes), 129.9 (d, 3JCP = 7.6 Hz, m-Mes), 141.1 (d, 4JCP = 2.5 Hz, p-Mes),

142.7 (d, 2JCP = 7.4 Hz, o-Mes). 31P NMR (101.3 MHz, CDCl3, 300 K): δ –53.7 (t,

1JPH = 379.8 Hz, PH2). IR (KBr, cm–1): v = 3019.1 (w), 2953.2 (m), 2914.4 (m),

2397.3 (m, P–H), 2364.6 (m, P–H), 1601.3 (m), 1447.67 (s), 1383.7 (m). HR FAB-

MS: calcd for C21H31PCl2Ru: 486.0584; found: 486.0588. MS m/z (%): 486 (65)

[M+], 451 (50) [M+ – Cl], 413 (50) [M+ – 2HCl], 299 (60) [M+ – 2HCl – H2PMes], 154

(100) [H2PMes].

[(ηηηη6-pCy)(IiPr2Me2)Ru====PMes*] (4). A cold solution of IiPr2Me2 (0.119 g, 0.66

mmol) in toluene (4 mL) was added to an orange toluene (5 mL) suspension

of complex [(η6-pCy)RuCl2(PH2Mes*)] (0.117 g, 0.20 mmol) at –78 °C, which

immediately resulted in a color change to green. After the mixture was

stirred for 1 h at –78 °C and warmed up to room temperature, solvents were

removed in vacuo. The precipitates were extracted into n-hexane (10 mL)

and the dark green suspension was filtered. After concentration, green

crystals of 4 were obtained at –20 °C (83 mg, 0.12 mmol, 60%). Mp ≥ 208 °C

(dec). 1H NMR (250.13 MHz, C6D6, 300 K): δ 0.94 (d, 3JHH = 6.9 Hz, 6 H, pCy-

CH(CH3)2), 1.30 (d, 3JHH = 7.1 Hz, 6 H, CH(CH3)2), 1.37 (d, 3JHH = 7.1 Hz, 6 H,

CH(CH3)2), 1.57 (s, 9 H, p-C(CH3)3), 1.64 (s, 3 H, pCy-CH3), 1.82 (s, 18 H, o-

Chapter 3

86

C(CH3)3), 1.89 (s, 6 H, CH3), 2.37 (sp, 3JHH = 6.9 Hz, 1 H, pCy-CH(CH3)2), 4.63 (d,

3JHH = 5.7 Hz, 2 H, C6H4), 4.85 (d, 3JHH = 5.7 Hz, 2 H, C6H4), 5.91 (sp, 3JHH = 7.1 Hz,

2 H, NCH(CH3)2), 7.55 (s, 2 H, m-Mes*). 13C{1H} NMR (100.64 MHz, C6D6, 300 K):

δ 10.5 (s, =CCH3), 18.9 (s, pCy-CH3), 22.0 (s, NCH(CH3)2), 22.2 (s, NCH(CH3)2),

23.9 (s, pCy-CH(CH3)2), 32.2 (s, p-C(CH3)3), 32.2 (s, CH(CH3)2), 32.7 (d, 4JCP =

7.5 Hz, o-C(CH3)3), 34.9 (s, p-C(CH3)3), 38.8 (s, o-C(CH3)3), 54.9 (s, NCH(CH3)2),

76.2, 81.5, 91.7, and 105.9 (s, C6H4), 119.9 (s, m-Mes*), 124.6 (s, =CCH3), 144.9

(s, p-Mes*), 146.3 (s, o-Mes*), 174.1 (d, 1JCP = 111.3 Hz, i-Mes*), 188.7 (s, N2C).

31P{1H} NMR (101.3 MHz, C6D6, 300 K): δ 733.5 (s, Ru=P). HRMS (EI, 70 eV):

calcd for C39H63N2PRu: 692.3766; found: 692.3707. MS m/z (%): 692 (8) [M+],

416 (20) [M+ – PMes*], 283 (8).

[(ηηηη6-pCy)(IMes)Ru====PMes] (5a). A solution of IMes (0.481 g, 1.415 mmol) in

toluene (10 mL) was added to a toluene (5 mL) suspension of complex [(η6-

pCy)RuCl2(PH2Mes)] (0.2125 g, 0.464 mmol) at –78 °C. The mixture was stirred

for 0.5 h and was slowly allowed to warm up to room temperature. After two

days, the solvent was evaporated in vacuo and the dark brown solid was

extracted with pentanes (20 mL), filtered (P4 glass filter, Teflon microfibre,

Celite), and after concentration a mixture of 5a and η6-toluene adduct [(η6-

Tol)(IMes)Ru=PMes]14 (5:1 ratio) was obtained as a dark brown solid (0.180 g,

0.261 mmol, 56%). 5a: 1H NMR (400.13 MHz, C6D6, 300 K): δ 0.84 (d, 3JHH = 6.9

Hz, 6 H, pCy-CH(CH3)2), 1.51 (s, 3 H, pCy-CH3), 1.64 (sp, 3JHH = 6.9 Hz, 1 H, pCy-

CH(CH3)2), 2.12 (s, 6 H, o-CH3MesP), 2.25 (s, 6 H, p-CH3MesN), 2.32 (s, 12 H, o-

CH3MesN), 2.37 (s, 3 H, p-CH3MesP), 4.38 (d, 3JHH = 5.9 Hz, 2 H, C6H4), 4.65 (d,

3JHH = 5.9 Hz, 2 H, C6H4), 6.35 (s, 2 H, NCH=), 6.73 (s, 4 H, m-MesN), 6.94 (s, 2 H,

m-MesP). 13C{1H} NMR (100.64 MHz, C6D6, 300 K): δ 19.5 (s, pCy-CH3), 19.7 (s,

o-CH3NMes), 20.9 (s, o-CH3PMes), 21.3 (s, p-CH3PMes), 21.5 (s, p-CH3NMes),

25.2 (s, pCy-CH(CH3)2), 32.2 (s, pCy-CH(CH3)2), 81.1, 83.1, 87.9, and 100.7 (s,

C6H4), 122.3 (s, =CH), 127.5 (s, m-MesP), 128.9 (s, m-MesN), 132.3 (s, o-MesP),

132.9 (s, p-MesP), 136.3 (s, i-MesN), 138.0 (s, o-MesN), 138.9 (s, p-MesN), 168.3

(d, 1JCP = 100.0 Hz, i-MesP), 191.7 (s, N2C). 31P{1H} NMR (101.3 MHz, C6D6, 300

K): δ 724.9 (s, Ru=P). HRMS (EI, 70 eV): calcd for [C40H49N2PRu] – [C9H11P]:


87

540.2072; found: 540.2091. MS m/z (%): 540 (16) [M+ – PMes], 302 (10) [M+ –

PMes – 2Mes].

[(ηηηη6-C6Me6)(IMes)Ru====PMes] (5b). A solution of IMes (0.258 g, 0.85 mmol) in

toluene (1 mL) was added under argon to a toluene (2 mL) suspension of

complex [(η6-C6Me6)RuCl2(PH2Mes)] (0.136 g, 0.28 mmol) at –78°C. The

mixture was allowed to slowly warm up to room temperature and the

resulting dark brown reaction mixture was filtered (P4 glass filter) and

evaporated in vacuo. Extraction with pentanes (5 mL), filtration and

concentration at room temperature yielded 5b as a dark brown solid (0.116

g, 0.162 mmol, 58%). Mp ≥ 198 °C (dec). 1H NMR (400.13 MHz, C6D6, 300 K): δ

1.66 (s, 18 H, C6(CH3)6), 2.14 (s, 6 H, o-CH3PMes), 2.18 (s, 6 H, p-CH3NMes), 2.21

(s, 6 H, o-CH3NMes), 2.36 (s, 3 H, p-CH3PMes), 2.52 (s, 6 H, o-CH3NMes), 6.33 (s,

2 H, =CH), 6.75 (s, 2 H, m-MesN), 6.79 (s, 2 H, m-MesN), 6.90 (s, 2 H, m-MesP).

13C{1H} NMR (100.64 MHz, C6D6, 300 K): δ 16.5 (s, C6(CH3)6), 19.9 (s, o-

CH3NMes), 21.0 (s, p-CH3NMes), 21.1 (s, o-CH3NMes), 21.4 (s, p-CH3PMes),

21.5, and 21.6 (s, o-CH3PMes), 90.9 (s, C6Me6), 122.8 (s, =CH), 127.6 (s, m-

MesP), 128.4, and 129.8 (s, m-MesN), 132.9 (s, p-MesP), 133.6 (bs, o-MesP),

135.2 (s, o-MesN), 137.3 (s, o-MesN), 137.5 (s, p-MesN), 139.3 (s, i-MesN), 165.6

(d, 1JCP = 97.6 Hz, i-MesP), 192.9 (d, 2JCP = 13.0 Hz, N2C). 31P{1H} NMR (101.3

MHz, C6D6, 300 K): δ 686.4 (s, Ru=P). MS EI m/z (%): 718 (<1) [M+], 566 (4), 303

(8), 162 (62), 147 (100).

[(ηηηη6-pCy)(IiPr2Me2)Os====PMes*] (6): A solution of IiPr2Me2 (0.079 g, 0.438

mmol) in toluene (2 mL) was added under vigorous stirring to a yellow

solution of [(η6-pCy)OsCl2(PH2Mes*)] (0.098 g, 0.146 mmol) in toluene (4 mL),

which resulted in a immediate color change to dark orange. After 16 h at

room temperature, all volatiles were removed in vacuo and the product was

extracted into pentane (15 mL) and filtered through a short Celite column.

After concentration of the pentane solution, orange crystals of 6 were

obtained at room temperature (0.097 g, 0.124 mmol, 85%). Mp ≥ 230 °C

(dec). 1H NMR (400.13 MHz, C6D6, 300 K): δ 0.92 (d, 3JHH = 6.9 Hz, 6 H, pCy-

CH(CH3)2), 1.25 (d, 3JHH = 7.2 Hz, 6 H, NCH(CH3)2), 1.40 (d, 3JHH = 7.2 Hz, 6 H,

Chapter 3

88

NCH(CH3)2), 1.54 (s, 9 H, p-C(CH3)3), 1.70 (s, 3 H, pCy-CH3), 1.87 (s, 6 H,

=CCH3), 1.89 (s, 18 H, o-C(CH3)3), 2.28 (sept, 3JHH = 6.9 Hz, 1 H, pCy-

CH(CH3)2), 4.86 (d, 3JHH = 5.5 Hz, 2 H, C6H4), 5.10 (d, 3JHH = 5.5 Hz, 2 H, C6H4),

5.89 (sept, 3JHH = 7.2 Hz, 2 H, NCH(CH3)2), 7.51 (s, 2 H, m-Mes*). 13C{1H} NMR

(100.64 MHz, C6D6, 300 K): δ 10.4 (s, =CCH3), 19.1 (s, pCy-CH3), 21.7 (s,

NCH(CH3)2), 21.8 (s, NCH(CH3)2), 24.2 (s, pCy-CH(CH3)2), 32.2 (s, p-C(CH3)3),

32.4 (s, pCy-CH(CH3)2), 33.0 (d, 4JCP = 6.8 Hz, o-C(CH3)3), 34.6 (s, p-C(CH3)3),

38.6 (s, o-C(CH3)3), 56.4 (s, NCH(CH3)2), 68.4, 73.6, 85.5, and 101.1 (s, C6H4),

119.8 (s, m-Mes*), 123.9 (s, =CCH3), 144.5 (s, p-Mes*), 147.0 (s, o-Mes*), 174.3

(s, N2C), 175.3 (d, 1JCP = 108.0 Hz, i-Mes*). 31P NMR (101.3 MHz, C6D6, 300 K): δ

557.6 (s, P=Os). IR (KBr, ): v 3055 (w), 2960 (vs), 2900 (s), 2863 (s), 1782 (w), 1712

(w), 1644 (m), 1574 (s), 1476, 1462, and 1444 (m), 1388 (m), 1376 (m), 1354 (s),

1291 (s), 1261 (m), 1238 (m), 1211 (m), 1104 (w), 1084 (w), 1070 (m), 1026 (m),

and 1019 (m), 864 (s), 800 (m), 785 (s), 739 (s), 684 (w), 642 (w), 537 cm–1 (w).

HRMS (EI, 70 eV): calcd for C39H63N2OsP: 782.4344; found: 782.4364. MS m/z

(%): 782 (<1) [M+].

[(ηηηη5-Cp*)(IiPr2Me2)Rh====PMes*] (7): A red solution of [(Cp*)RhCl2(PH2Mes*)]

(0.269 g, 0.457 mmol) in toluene (15 mL) was added dropwise under vigorous

stirring to a solution of IiPr2Me2 (0.247 g, 1.372 mmol) in toluene (5 mL), which

resulted in a immediate color change to dark pink. After 24 h at room

temperature, all volatiles were removed in vacuo and the product was

extracted into pentane (45 mL) and filtered through a short Celite column.

After concentration to few mL, large dark pink crystals of 7�C5H12 were

obtained at –4 °C (0.309 g, 0.403 mmol, 88%). Mp ≥ 204.8 °C (dec). 1H NMR

(400.13 MHz, C6D6, 300 K): δ 0.88 (t, 3JHH = 7.3 Hz, pentane), 1.24 (m, 3JHH = 7.3

Hz, pentane), 1.38 (d, 3JHH = 7.1 Hz, 6 H, NCH(CH3)2), 1.39 (d, 3JHH = 7.1 Hz, 6 H,

NCH(CH3)2), 1.56 (s, 9 H, p-C(CH3)3), 1.61 (s, 15 H, Cp*), 1.77 (s, 18 H, o-

C(CH3)3), 1.88 (s, 6 H, =CCH3), 5.77 (sept, 3JHH = 7.1 Hz, 2 H, NCH(CH3)2), 7.57

(s, 2 H, m-Mes*). 13C{1H} NMR (100.64 MHz, C6D6, 300 K): δ 10.2 (s, =CCH3), 10.4

(s, C5(CH3)5), 14.2 (s, CH3, pentane), 22.5 (s, NCH(CH3)2), 22.7 (s, CH2,

pentane), 23.0 (s, NCH(CH3)2), 31.6 (bs, o-C(CH3)3), 32.1 (s, p-C(CH3)3), 34.4 (s,

CH2, pentane), 34.7 (s, o-C(CH3)3), 38.7 (s, p-C(CH3)3), 54.1 (s, NCH(CH3)2),


89

93.5 (d, 2JCP = 4.2 Hz, C5(CH3)5), 120.6 (s, m-Mes*), 125.0 (bs, =CCH3), 144.0 (s,

o-Mes*), 144.4 (s, p-Mes*), 170.0 (d, 1JCP = 120.1 Hz, i-Mes*), 185.0 (d, 1JCRh =

63.7 Hz, N2C–Rh). 31P NMR (101.3 MHz, C6D6, 300 K): δ 745.9 (d, 1JPRh = 70.4 Hz,

P=Rh). IR (KBr): v 3076 (w), 2955 (s), 2900 (s), 2868 (s), 1738 (w), 1643 (m), 1586

(s), 1462 (s), 1380 (m), 1358 (s), 1290 (s), 1239 (m), 1210 (m), 1189 (m), 1130

(w), 1104 (w), 1068 (w), 1022 (m), 870 (m), 806 (bw), 746 (s), 700 (w), 642 (w),

590 (w), 518 cm–1 (w). HRMS (EI, 70 eV): calcd for C39H64N2PRh: 694.3862;

found: 694.3867. MS m/z (%): 694 (8) [M+], 418 (40) [M+ – PMes*], 376 (100) [M+

– PMes* – C3H6].

[(ηηηη5-Cp*)(IiPr2Me2)Ir====PMes*] (3): Iridium complex [(η5-Cp*)IrCl2(PH2Mes*)]

(0.223 g, 0.33 mmol) was added to a solution of IiPr2Me2 (0.184 g, 1.02 mmol)

in toluene (4 mL) at –78°C. The resulting reaction mixture was allowed to

warm up to room temperature and, after 18 h, the mixture was filtered (P4

glass filter) and evaporated in vacuo. Extraction of the product into

pentanes (5 mL), filtration and removal of all volatiles in vacuo followed by

crystallization from toluene at –20°C yielded 3 as red crystals (0.155 g, 0.198

mmol, 60%) with identical spectroscopic data as previously reported.3a

Acknowledgement. This work was supported by The Netherlands

Foundation for Chemical Sciences (CW) with financial aid from the

Netherlands Organization for Scientific Research (NWO). Dr. M. Smoluch and

J. W. H. Peeters (University of Amsterdam) are acknowledged for measuring

respectively HR EI–MS and HR FAB–MS.

Supporting information Available. 1H NMR spectral data of all novel

compounds, cartesian coordinates (Å) and energies (a.u.) of all stationary

points, and the cif file with crystallographic data for compound 7. This

material is available free of charge on the Internet at http://pubs.acs.org.


[1] (a) J. C. Slootweg, K. Lammertsma, In Science of Synthesis; B. M. Trost, F. Mathey,

Chapter 3

90


Dalton Trans. 2007, 1861–1868. (c) K. Lammertsma, Top. Curr. Chem. 2003, 229,

95–119. (d) K. Lammertsma, M. J. M. Vlaar, Eur. J. Org. Chem. 2002, 1127–1138.

(e) F. Mathey, N. T. Tran Huy, A. Marinetti, Helv. Chim. Acta 2001, 84, 2938–2957.

(f) D. W. Stephan, Angew. Chem. 2000, 112, 322–338; Angew. Chem. Int. Ed.

2000, 39, 314–329. (g) A. H. Cowley, Acc. Chem. Res. 1997, 30, 445–451.

[2] A. W. Ehlers, E. J. Baerends, K. Lammertsma, J. Am. Chem. Soc. 2002, 124, 2831–

2838.

[3] (a) A. T. Termaten, M. Schakel, A. W. Ehlers, M. Lutz, A. L. Spek, K. Lammertsma,

Chem. Eur. J. 2003, 9, 3577–3582. (b) C. C. Cummins, R. R. Schrock, W. M. Davis,

Angew. Chem. 1993, 105, 758–761; Angew. Chem. Int. Ed. 1993, 32, 756–759. (c)

Z. Hou, D. W. Stephan, J. Am. Chem. Soc. 1992, 114, 10088–10089.

[4] R. Melenkivitz, D. J. Mindiola, G. L. Hillhouse, J. Am. Chem. Soc. 2002, 124, 3846–

3847.

[5] (a) F. Basuli, J. Tomaszewski, J. C. Huffman, D. J. Mindiola, J. Am. Chem. Soc.

2003, 125, 10170–10171. (b) F. Basuli, B. C. Bailey, J. C. Huffman, M.-H. Baik, D. J.

Mindiola, J. Am. Chem. Soc. 2004, 126, 1924–1925. (c) G. Zhao, F. Basuli, U. J.

Kilgore, H. Fan, H. Aneetha, J. C. Huffman, G. Wu, D. J. Mindiola, J. Am. Chem.

Soc. 2006, 128, 13575–13585; and references therein.


Chem. Eur. J. 2003, 9, 2200–2208.

[7] A. T. Termaten, H. Aktas, M. Schakel, A. W. Ehlers, M. Lutz, A. L. Spek, K.

Lammertsma, Organometallics 2003, 22, 1827–1834.



[9] (a) T. L. Breen, D. W. Stephan, J. Am. Chem. Soc. 1995, 117, 11914–11921. For

reviews on phosphaalkenes, see: (b) P. Le Floch, Coord. Chem. Rev. 2006, 250,

627–681. (c) F. Mathey, Angew. Chem. 2003, 115, 1616–1643; Angew. Chem. Int.

Ed. 2003, 42, 1578–1604. (d) L. Weber, Eur. J. Inorg. Chem. 2000, 2425–2441. (e) M.

J. Yoshifuji, J. Chem. Soc. Dalton Trans. 1998, 3343–3349. (f) L. Weber, Angew.

Chem. 1996, 108, 292–310; Angew. Chem. Int. Ed. Engl. 1996, 35, 271–288. (g) A.

C. Gaumont, J. M. Denis, Chem. Rev. 1994, 94, 1413–1439. (h) F. Mathey, Acc.

Chem. Res. 1992, 25, 90–96. (i) R. Appel, Multiple Bonds and Low Coordination in

Phosphorus Chemistry (Eds.: M. Regitz, O. J. Scherer), Thieme, Stuttgart, 1990.


91

[10] (a) C.-W. Tsang, M. Yam, D. P. Gates, J. Am. Chem. Soc. 2003, 125, 1480–1481. (b)

D. P. Gates, Top. Curr. Chem. 2005, 250, 107–126. (c) K. J. T. Noonan, D. P. Gates,

Angew. Chem. 2006, 118, 7429–7432; Angew. Chem. Int. Ed. 2006, 45, 7271–7274.

[11] F. Ozawa, M. Yoshifuji, Dalton Trans. 2006, 4987–4995.

[12] (a) T. Weskamp, W. C. Schattenmann, M. Spiegler, W. A. Herrmann, Angew.

Chem. 1998, 110, 2631–2633; Angew. Chem. Int. Ed. 1998, 37, 2490–2493. (b) M.

Scholl, T. M. Trnka, J. P. Morgan, R. H. Grubbs, Tetrahedron Lett. 1999, 40, 2247–

2250. (c) J. Huang, E. D. Stevens, S. P. Nolan, J. L. Petersen, J. Am. Chem. Soc.

1999, 121, 2674–2678. (d) J. Huang, H.–J. Schanz, E. D. Stevens, S. P. Nolan,

Organometallics 1999, 18, 5375–5380. (e) C. W. Bielawski, R. H. Grubbs, Angew.


[13] R. H. Grubbs, Angew. Chem. 2006, 118, 3845–3850; Angew. Chem. Int. Ed. 2006,

45, 3760–3765.


Lammertsma, J. Am. Chem. Soc., 2009, 131, 6666–6667.

[15] M. F. Lappert, J. Organomet. Chem. 2005, 690, 5467–5473.

[16] L. Mercs, G. Labat, A. Neels, A. Ehlers, M. Albrecht, Organometallics 2006, 25,

5648–5656.

[17] (a) C. Boehme, G. Frenking, Organometallics 1998, 17, 5801–5809. (b) S. Diez-

Gonzalez, S. P. Nolan, Coord. Chem. Rev. 2007, 251, 874–883. (c) H. Jacobsen, A.

Correa, A. Poater, C. Costabile, L. Cavallo, Coord. Chem. Rev. 2009, 253, 687–

703.

[18] 5a contains 15% of toluene adduct [(η6-Tol)(IMes)Ru=PMes] (δ31P 736.8), which is

formed by arene exchange with the solvent; see for more details ref.14.

[19] The primary phosphine complexes are unavailable for the Group 7 transition

metals as well as for the iron complex [(η6-Ar)FeCl2(PH2Mes*)].

[20] X-ray crystal structure determination of 7. C39H64N2PRh � 0.66C5H12, Fw = 742.42,

black block, 0.54 x 0.48 x 0.18 mm3, monoclinic, P21/m (no. 11), a = 10.4727(3), b

= 15.1067(4), c = 13.7281(2) Å, β = 95.432(2)°, V = 2162.14(9) Å3, Z = 2, Dx = 1.14



to a resolution of (sin θ/λ)max = 0.65 Å–1 at a temperature of 150(2) K. Intensity

integration was performed with EvalCCD.[34] The SADABS[35] program was used

for absorption correction based on multiple measured reflections (0.62-0.92

transmission range). 5158 Reflections were unique (Rint = 0.021), of which 4627

Chapter 3

92

were observed [I>2σ(I)]. The structure was solved with Direct Methods using the

program SHELXS-97[36]. The structure was refined with SHELXL-97[36] against F2 of all

reflections. Non hydrogen atoms were refined with anisotropic displacement

parameters. All hydrogen atoms were introduced in calculated positions and

refined with a riding model. The p-tert-butyl group was refined with a disorder

model with respect to the mirror plane. The pentane solvent molecule was also

refined with a disorder model and with partial occupancy. 270 Parameters were

refined with 72 restraints concerning the disorder and the solvent molecule.

R1/wR2 [I > 2σ(I)]: 0.0258 / 0.0861. R1/wR2 [all refl.]: 0.0351 / 0.0906. S = 1.053.

Residual electron density between -0.26 and 0.79 e/Å3. Geometry calculations

and checking for higher symmetry was performed with the PLATON program.[37]

[21] G. Frenking, I. Antes, M. Boehme, S. Dapprich, A. Ehlers, V. Jonas, A. Neuhaus, M.

Otto, R. Stegmann, A. Veldkamp, S. F. Vyboyshchikov, in Reviews in

Computational Chemistry; Lipkowitz, K. B.; Boyd, D. B., Eds.; VCH: New York, 1996;

Vol. 8, p 63.

[22] ADF2006.01, SCM, Theoretical Chemistry, Vrije Universiteit, Amsterdam, The

Netherlands, http://www.scm.com/.






Chem. Phys. 1988, 88, 322–328.


[28] (a) K. Morokuma, Acc. Chem. Res. 1977, 10, 294–300. (b) T. Ziegler, A. Rauk, Inorg.

Chem. 1979, 18, 1755–1759. (c) T. Ziegler, A. Rauk, Theor. Chim. Acta 1977, 46, 1–

10.

[29] M. A. Bennett, T. N. Huang, T. W. Matheson, A. K. Smith, Inorg. Synth. 1982, 21, 74–

78.


[31] Arduengo A. J. Arduengo, III, H. V. R. Dias, R. L. Harlov, M. Kline, J. Am. Chem.

Soc. 1992, 114, 5530–5534.

[32] A. H. Cowley, J. E. Kilduff, T. H. Newman, M. Pakulski, J. Am. Chem. Soc. 1982, 104,

5820–5821.


93


405, 183–194.


2003, 36, 220–229.

[35] G. M. Sheldrick, 1999, SADABS: Area-Detector Absorption Correction, v2.10,


[36] G. M. Sheldrick, Acta Cryst. 2008, A64, 112–122.

[37] A. L. Spek, J. Appl. Cryst. 2003, 36, 7–13.

Chapter 3

94

Iridium Phosphinidene Complexes: A Comparison with

Iridium Imido Complexes in Their Reaction with

Isocyanides

Halil Aktas,† Jos Mulder,† Frans J. J. de Kanter,† J. Chris Slootweg,† Marius

Schakel,† Andreas W. Ehlers,† Martin Lutz,‡ Anthony L. Spek,‡ and Koop

Lammertsma*,†




J. Am. Chem. Soc. 2009, 131, 13531–13537

Abstract: 18-Electron nucleophilic, Schrock-type phosphinidene complexes 3

[Cp*(Xy–N≡C)Ir=PAr] (Ar = Mes*, Dmp, Mes) are capable of unprecedented [1+2]-

cycloadditions with one equivalent of isocyanide RNC (R = Xy, Ph) to give novel

iridaphosphirane complexes [Cp*(Xy–N≡C)IrPArC┌────┐

=NR]. Their structures were

ascertained by X–ray diffraction. Density functional theory investigations on model

structures revealed that the iridaphosphirane complexes are formed from the

addition of the isocyanide to 16-electron species [Cp*Ir=PAr] forming first complex 3

that subsequently reacts with another isocyanide to give the products following a

different pathway than its nitrogen analogue [Cp*Ir≡Nt-Bu] 1.

Chapter 4Chapter 4Chapter 4Chapter 4

Chapter 4

96

4.1 Introduction

Ever since stable nucleophilic phosphinidenes were discovered by

Lappert et al. in 1987,[1] the structural properties and intriguing

reactivity of these Schrock-type species continue to fascinate.[2]

Exemplary valuable transfer reactions are those with oxo- and

halophilic transition metal complexes,[3] such as the phospha-Wittig

reaction with carbonyl compounds that yields phosphaalkenes[4,5]

and the P/O-exchange with epoxides that gives the three-

membered phosphiranes.[4] Phosphaalkenes also result on reaction

with geminal dihalides,[4,6,7] while other dihalides give phosphorus

heterocycles.[4] Four-membered phosphametallocycles are

accessible by [2+2]-cycloadditions with alkenes[8] and alkynes,[9]

thereby mimicking the behavior of Schrock-type carbenes.[10] In the

present study we compare the behavior of phosphinidene and imido

complexes.

Inspired by the reported reaction of stable, 18-electron imido

complex [(η5-Cp*)Ir≡Nt-Bu] 1 with isocyanides that gives η2-

coordinated carbodiimide complex 2,[11] we targeted this reaction

for the corresponding phosphinidenes.[13] At the outset there is an

intriguing difference between imido complex 1 and its phosphorus

analogue, namely 16-electron complex [(η5-Cp*)Ir=P–R] is not a

stable species[12] and whereas many 18-electron [(η5-Cp*)(L)Ir=P–

Mes*] (L = PR3, AsR3, dppe, RN≡C, CO, NHC) have been reported,[6,14]

it is not known to be capable of [1+2]-cycloadditions like its well-

established Fischer-type electrophilic counterpart.[15]

Isocyanide-stabilized 18-electron [(η5-Cp*)(Xy–N≡C)Ir=PMes*] 3a

(Mes* = 2,4,6-tBu3C6H2),[6] generated by double dehydrohalogena-

tion[6,16] of [(η5-Cp*)IrCl2(PH2Mes*)] and concomitant ligation with 2,6-

xylyl isocyanide, is an ideal starting point for the present study in

Iridium Phosphinidene Complexes: A Comparison with Iridium Imido Complexes in Their Reaction with Isocyanides

97

which we examine the Ir=P reactivity toward isocyanides and explore

the scope of reaction for the transient, in-situ generated 16-electron

complex [(η5-Cp*)Ir=P–R]. We use DFT calculations to address

differences between the imido and phosphinidene complexes.

4.2 Results and Discussion

Reaction of deep pink iridium phosphinidene [(η5-Cp*)(Xy–

N≡C)Ir=PMes*] 3a (δ 31P 757.1)[6] with a tenfold excess of phenyl or 2,6-

xylyl isocyanide at room temperature resulted in the formation of

yellow crystalline [1+2]-adducts 4a and 5 [(Cp*)(XyNC)IrPMes*C┌───────┐

=N–R]

(R = 4a, Xy; 5, Ph) as sole products in respectively 77% and 87% yield

after crystallization (Scheme 1). The highly shielded resonances in the 31P NMR spectrum at δ = –190.4 (4a) and –190.2 ppm (5) are

diagnostic for three-membered P-rings and indicate the formation of

the desired iridaphosphiranes in analogy to iridaaziridine 2. Single-

crystal X-ray analysis of 4a and 5 established unequivocally the

unique η2-(P,C)-phosphaazaallene moiety[17] and the linear 2,6-xylyl

isocyanide, both coordinated to iridium (Figure 1).[18] The ca. 2.41 Å Ir–

P bond of the IrPC ring is elongated from the reported 2.17–2.21 Å Ir=P

double bond of phosphinidene complexes like 3a (carrying PPh3, CO,

and NHC ligands instead of an isocyanide),[6,14] whereas the ca. 2.03

Å Ir–C bond compares well with the 2.017(9) Å reported for the IrNC

ring in carbodiimide complex 2;[11c] the ca. 1.80 Å P–C bond length is

normal for three-membered P-rings.

Chapter 4

98

Scheme 1. Synthesis of iridaphosphiranes 4a and 5.

Figure 1. Displacement ellipsoid plot (50% probability) of 4a and 5. Hydrogen atoms

are omitted for clarity. Selected bond lengths [Å] and angles [°] for 4a and 5 (in

square brackets): Ir1–P1 2.4147(5) [2.4082(6)], Ir1–C19 2.036(2) [2.029(2)], Ir1–C38[C36]

1.900(2) [1.895(2)], P1–C1 1.8766(19) [1.877(2)], P1–C19 1.805(2) [1.799(2)], N1–C19

1.267(2) [1.265(3)], N1–C20 1.424(3) [1.424(3)], N2–C38[C36] 1.169(3) [1.168(3)], N2–

C39[C37] 1.402(3) [1.396(3)]; Ir1–P1–C19 55.46(6) [55.40(7)], P1–Ir1–C19 46.91(6)

[46.89(7)], Ir1–C38[C36]–N2 177.27(19) [178.3(2)], C19–N1–C20 125.99(18) [123.2(2)],

C38[C36]–N2–C39[C37] 171.7(2) [175.9(2)].

The described reaction illustrates that different isocyanides can be

embedded in the product, something that appears not feasible for

the imido complexes.[10] The question is than whether the reaction

mechanisms are the same for the two systems. Does the distinction lie

in the stability of the 16-electron intermediate or the lack thereof? By

reducing the stability of the 18-electron phosphinidene complex and


99

attempting to synthesize iridaphosphiranes directly via in situ

generated 16-electron iridium phosphinidene complexes, we

examined whether the reaction with isocyanides would mimic more

closely that of imido complex 1. The first step was to reproduce the

formation of 4a. Double dehydrohalogenation[6,16] of orange colored

primary phosphine complex [(η5-Cp*)IrCl2(PH2Mes*)] (6a) with two

equivalents of 1,8-diazabicyclo-[5.4.0]undec-7-ene (DBU) at –78 °C in

the presence of ten equivalents of 2,6-xylyl isocyanide resulted in an

immediate color change to deep purple, indicative for the formation

of 18-electron 3a, and upon warming to room temperature to yellow

to give after crystallization indeed iridaphosphirane 4a in 86% isolated

yield (Scheme 2). This protocol was extended to other phosphine

complexes as illustrated in Scheme 2 for [(η5-Cp*)IrCl2(PH2R)] 6b (R =

Mes) and 6c (R = Dmp = 2,6-dimesitylphenyl).[19]

Scheme 2. Synthesis of iridaphosphiranes 4 via in situ-generated phosphinidenes 3.

Mesityl-substituted 6b undergoes a facile double

dehydrohalogenation-ligation with two equivalents of DBU and an

excess of 2,6-xylyl isocyanide to afford iridacycle 4b (δ 31P –218.7) as

the sole product, which was isolated by crystallization (60%; Scheme

2) and structurally characterized by a single-crystal X-ray structure

determination (Figure 2). In this case, the sterically less shielded

Chapter 4

100

phosphinidene intermediate [(η5-Cp*)(Xy–N≡C)Ir=PMes] 3b could not

be detected by 31P NMR spectroscopy under the reaction conditions;

the accordingly obtained more encumbered Dmp derivative 3c

(deep purple) was observed (δ 31P 768.1) and converted at room

temperature to yellow complex 4c (64 %; δ 31P –216.7).

Figure 2. Displacement ellipsoid plot (50% probability) of 4b. Hydrogen atoms are

omitted for clarity. Selected bond lengths [Å] and angles [°] for 4b: Ir1–P1 2.3952(6),

Ir1–C10 2.039(2), Ir1–C29 1.888(2), P1–C1 1.837(2), P1–C10 1.799(3), N1–C10 1.267(3),

N1–C11 1.429(3), N2–C29 1.172(3), N2–C30 1.398(3); Ir1–P1–C10 56.04(8), P1–Ir1–C10

47.02(7), Ir1–C10–P1 76.94(9), Ir1–C29–N2 179.4(3), C10–N1–C11 123.8(2), C29–N2–C30

161.9(2).

4.3 Transient species

Is the suggested 16-electron [(η5-Cp*)Ir=PR] indeed formed on double

dehydrohalogenation of 6 in the absence of isocyanides (7; Scheme

2)? So far, only 16-electron zirconium phosphinidenes have been

observed in arene solvents by 31P NMR spectroscopy (δ 31P 438–526

ppm)[20] and as unstable Cp*2Zr=PMes*-LiCl adduct in

dimethoxyethane (δ 31P 537 ppm).[21] In contrast to imido analogue 1,

attempted access to Mes- and Mes*-substituted 16-electron [(η5-

Cp*)Ir=PR] (7a,b) only led to dimers.[12] DFT calculations have shown


101

the 16-electron imido and phosphinidene complexes to differ with

the first having a linear IrNH arrangement with an Ir≡N triple bond[22]

and [(η5-Cp)Ir=PH] having a bent conformation (<IrPH 126.4°) with an

Ir=P double bond,[12] a difference with consequences for their

reactivity.

By introducing the sterically demanding Dmp substituent on

phosphorus, we envisioned to stabilize the 16-electron species by

shielding the Ir=P double bond. Double dehydrohalogenation of

primary phosphine complex [(η5-Cp*)IrCl2(PH2Dmp)] (6c) with two

equivalents of DBU in CD2Cl2 at –10 °C showed after warm up to room

temperature a low-field 31P NMR resonance at 672 ppm, suggesting

indeed the formation of a bent phosphinidene. To confirm its identity

the 31P NMR chemical shifts were calculated for the E and Z

conformers of [(η5-Cp*)Ir=PDmp] 7c and their solvent adducts (Figure

3).

Figure 3. Intermediate E- and Z-[7c(CH2Cl2)] calculated at the BP86/TZP level of

theory. Hydrogen atoms are omitted for clarity. Selected bond lengths [Å] and angles

[°] for E-[7c(CH2Cl2)] and Z-[7c(CH2Cl2)] (in square brackets): Ir–P 2.192 [2.206], Ir–Cl1

2.567 [2.368]; Ir–P–Dmp 125.2 [114.6], P–Ir–Cl1 87.5 [102.9].

Chapter 4

102

Those for “solvent-free” E-7c and Z-7c (178.9 and 418.2 ppm,

respectively) are significantly shielded from the experimental one,

but those for the solvent adducts are in the expected range (E-

[7c(CH2Cl2)] 457.2, Z-[7c(CH2Cl2)] 683.6 ppm; Figure 3) with an

excellent match for the more stable Z-conformer (∆E = 0.3 kcal�mol–1).

Therefore, we conclude that the Z-conformer is also a likely

intermediate when applying smaller donor ligands.[6,16a,23]

4.4 Mechanism

The experimental work leads us to conclude that the in situ formed

16-electron phosphinidene coordinates with an isocyanide to the

observable and isolable 18-electron complex 3, which gives a [1+2]-

cycloaddition with another isocyanide molecule to form

iridaphosphirane 4. The reaction with imido complex 1 is different in

that this species is stable and that there are no indications for an

observable isocyanide coordinated 18-electron imido complex. The

question then arises whether this distinction is due to the relative

stabilities of the reactants, thus whether the first or second isocyanide

addition is rate-determining, or to different mechanistic pathways. To

answer this question we resorted to density functional theory at

BP86/TZP to compare the reaction of the 16-electron imido and

phosphinidene complexes 1' and 7' with isocyanide HN≡C using

model structures incorporating only H substituents.

4.4.1 Imido complex

We begin with the imido complex, [(η5-Cp)Ir≡NH] 1'. The reaction

starts with the addition of the isocyanide to give bent 18-electron

imido complex 8' (∆E = –26.1 kcal�mol–1; Figure 4),[24] followed by ring

closure to 16-electron η2-carbodiimide complex 9', requiring 18.4


103

kcal�mol–1 (∆E = –10.9 kcal�mol–1), and subsequent addition of a

second isocyanide to afford the 48.3 kcal�mol–1 more stable product

2'.

Figure 4. Relative BP86/TZP energies (in kcal�mol–1) for the reaction of [(η5-Cp)Ir≡NH] 1'

with HNC. Selected bond lengths [Å] and angles [°] for 8': Ir1–N1 1.876, Ir1–C1 1.859,

C1–N2 1.210, Ir1–N1–H 105.5; TS8'–9': Ir1–N1 1.883, Ir1–C1 1.926, C1–N1 1.900, C1–N2

1.216, Ir1–N1–H 120.1; 9': Ir1–N1 1.913, Ir1–C1 2.040, C1–N1 1.352, C1–N2 1.275, Ir1–N1–

H 153.1; 2': Ir1–N1 2.124, Ir1–C1 2.076, C1–N1 1.337, C1–N2 1.264, Ir1–C2 1.843, C2–N3

1.218, Ir1–N1–H 115.5.

4.4.2 Phosphinidene complex

The mechanism for formation of iridaphosphirane 4 is different.

Namely, addition of HN≡C to [(η5-Cp)Ir=PH] 7' is far more exothermic

in giving the bent 18-electron species ([(η5-Cp)(HNC)Ir=PH] 3' (∆E = –

58.2 kcal�mol–1; Figure 5) that, moreover, is energetically prohibited to

undergo ring closure to 10’ (∆E = 27.1 kcal�mol–1) and instead is

susceptible for attack of a second isocyanide to the phosphorus

center to give η1-(P,C)-phosphaazaallene complex 11' (∆E = –17.6

Chapter 4

104

kcal�mol–1), which rearranges barrierless to the more favorable η2-

coordinated product 4' (∆E = –20.3 kcal�mol–1).

Figure 5. Relative BP86/TZP energies (in kcal�mol–1) for the reaction of [(η5-Cp)Ir=PH] 7'

with HNC. Selected bond lengths [Å] and angles [°] for 3': Ir1–P1 2.223, Ir1–C1 1.858,

C1–N2 1.213, Ir1–P1–H 98.9; 10': Ir1–P1 2.307, Ir1–C1 2.025, P1–C1 1.818, C1–N2 1.268,

Ir1–P1–H 110.7; 11': Ir1–P1 2.241, Ir1–C1 1.834, Ir1–C2 3.403, P1–C2 1.726, C1–N1 1.229,

C2–N2 1.217, Ir1–P1–H 122.9; 4': Ir1–P1 2.439, Ir1–C1 1.850, Ir1–C2 2.066, P1–C2 1.813,

C1–N1 1.216, C2–N2 1.258, Ir1–P1–H 101.2, C2–Ir1–P1 46.6.

The higher nucleophilicity of the imido complex originates from the

lower energies of the ligand orbitals (NH –6.79 eV; PH –5.46 eV) and is

reflected in the calculated charges (–0.27 on N in 8'; –0.04 on P in

3').[25] Two differences that define the dissimilar chemistries of the

phosphinidene and imido complexes are revealed by the calculated

reaction pathways. First, as noted, the initial ligand addition to the 16-

electron complex is far more exothermic for the phosphinidene than

for imido complex 8', which suffers from distortion of the Ir≡N–R unit


105

from linearity. The second difference lies in the electronic structure of

the ring-closed structures of 9' and 10'. An allenic, π-conjugated N–C–

N moiety is formed in Cs-symmetric 9' that binds in a bidentate κ2-

fashion to the metal, but in phosphorus analogue 10' π-conjugation is

less favorable due to the smaller overlap between the C(2p) and

P(3p) orbitals so that the allenic N–C–P unit binds instead to the metal

in an η2-fashion, acting as a 2-electron and not a 4-electron donor as

is the case for 9'. Therefore, nucleophilic attack of the phosphorus of

3' at the isocyanide is prohibited, while a modest barrier is observed

for the corresponding nitrogen attack of 8'. As a consequence,

stable 18-electron phosphinidene complexes can be observed

experimentally whereas the imido complexes are prone to

rearrangements. Formation of the final product 4' occurs by direct

attack of a second isocyanide to the phosphorus, but this process

can be hindered by steric congestion using bulky substituents, which

explains why 3a can be observed at low temperatures by NMR

spectroscopy even in an excess of isocyanide.

4.5 Conclusions

Imido and phosphinidene iridium complexes differ in their properties

and reactivities as established for the reaction with isocyanides. Both

isolated and in-situ generated 18-electron iridium phosphinidene [(η5-

Cp*)(Xy–N≡C)Ir=PMes*] 3 afford iridaphosphirane 4 as sole product.

Despite this apparent resemblance with the stable 16-electron imido

complex 1 that gives iridaazirane 2, the course of events is entirely

different for the two reactions as elucidated by DFT calculations. The

imido complex uses the first isocyanide molecule to construct a 16-

electron η2-carbodiimide complex, whereas for the phosphorus

analogue it is the second isocyanide molecule that induces the ring

Chapter 4

106

closure, thereby giving the unique η2-phosphaazaallene complex. It is

further established that the sterically demanding dimesitylphenyl

substituent enables the detection of the solvent-stabilized 16-electron

phosphinidene intermediate [(η5-Cp*)Ir=PDmp] 7c, generated by

double dehydrogenation of phosphine precursor 6, prior to the

reaction with isocyanides.


All density functional theory calculations have been performed with the

parallelized Amsterdam density functional (ADF) package (version 2005.01b

and 2006.01).[26] The Kohn–Sham MOs were expanded in a large,

uncontracted basis set of Slater–type orbitals (STOs), of a triple–ζ basis set

with polarization functions quality, corresponding to basis set TZP in the ADF

package. The 1s core shell of carbon and nitrogen and the 1s2s2p core shells

of phosphorus were treated by the frozen core approximation. The transition

metal centers were described by a triple–ζ basis set for the outer ns, np, nd

and (n+1)s orbitals, whereas the shells of lower energy were treated by the

frozen core approximation using a small core. All calculations were

performed at the nonlocal exchange self–consistent field (NL–SCF) level,

using the local density approximation (LDA) in the Vosko–Wilk–Nusair

parameterization[27] with nonlocal corrections for exchange (Becke88)[28]

and correlation (Perdew86).[29] All geometries were optimized using the

analytical gradient method implemented by Versluis and Ziegler,[30]

including relativistic effects by the Zeroth Order Regular Approximation

(ZORA).[31] The 31P NMR chemical shift tensors were calculated with ADF’s

NMR program,[32] using single–point calculations with an all–electron basis for

P within the ZORA–approximation on the optimized frozen core structures

(vide supra), using the E-isomer of [Cp*(Me3P)Ir=PMes*][6] as reference (σ –

455.2 ppm) for the total isotropic shielding tensors (δ +629.3 ppm with respect

to 85% H3PO4). All model complexes were calculated without molecular

symmetry.


107



atmosphere of dry nitrogen with rigorous exclusion of air and moisture using

flame–dried glassware using Schlenk techniques. Solvents were distilled from

sodium (toluene), CaCl2 (CH2Cl2), or LiAlH4 (pentanes, diethyl ether) and kept

under an atmosphere of dry nitrogen. Deuterated solvents were purchased

from Aldrich and dried over 4 Å molecular sieves (CD2Cl2, CDCl3, C6D6). All

solid starting materials were dried in vacuo. NMR spectra were recorded on

a Bruker Advance 250 (1H, 13C, 31P; 85% H3PO4) or a Bruker Advance 400 (1H, 13C, 31P; 85% H3PO4) and referenced internally to residual solvent resonances

(CDCl3: 1H: δ 7.26, 13C{1H}: δ 77.16; CD2Cl2: 1H: δ 5.25, 13C{1H}: δ 54.00; C6D6: 1H:

δ 7.16, 13C{1H}: δ 128.06). IR Spectra were recorded on a Mattson–6030

Galaxy FT–IR spectrophotometer, high–resolution mass spectra (HR–MS) on a

Finnigan Mat 900 spectrometer operating at an ionization potential of 70 eV,

and fast atom bombardment (FAB) mass spectrometry was carried out using

a JEOL JMS SX/SX 102A four–sector mass spectrometer (70 eV). Melting points

were measured on samples in sealed capillaries on a Stuart Scientific SMP3

melting point apparatus and are uncorrected. [(η5-Cp*)IrCl2(PH2Mes*)]

(6a),[6] [(η5-Cp*)IrCl2(PH2Mes)] (6b),[6] [(η5-Cp*)(Xy–N≡C)Ir=PMes*] (3a),[6] and

Mes*PH2,[33] were prepared according to literature procedures. PhPH2,

MesPH2, and DmpPH2[34] were prepared analogously to IsPH2,[35] by LiAlH4

reduction of respectively PhPCl2, MesPCl2, and DmpPCl2.[36] Ph–N≡C[37] was

prepared by dehydration of the corresponding formamide with phosphoryl

chloride. 2,6-Xylyl isocyanide (Xy–N≡C) was purchased from Fluka and used

as received.

[(ηηηη5-Cp*)IrCl2(PH2Dmp)] (6c). A mixture of freshly prepared DmpPH2 (1.00 g,

2.89 mmol) and [(η5-Cp*)IrCl2]2 (0.59 g, 0.74 mmol) in CH2Cl2 (50 mL) was

stirred for 30 min. at room temperature. Evaporation to dryness and

chromatography of the residue over silica with CHCl3, followed by

CHCl3/diethyl ether 4:1, as eluent and subsequent crystallization from

CH2Cl2/pentane at –20 °C yielded [(η5-Cp*)IrCl2(PH2Dmp)] (6c) (1.08 g, 1.45

mmol, 98%) as orange crystals. Mp: 258 °C (dec). 1H NMR (400.13 MHz, CDCl3,

Chapter 4

108

300 K): δ 1.32 (d, 4J(H,P) = 3.3 Hz, 15H; C5(CH3)5), 2.15 (s, 12H; o-CH3), 2.34 (s,

6H; p-CH3), 5.49 (d, 1J(H,P) = 378.6 Hz, 2H; PH2), 6.95 (s, 4H; m-MesH), 7.05 (dd, 3J(H,H) = 7.5 Hz, 4J(H,P) = 2.9 Hz, 2H; m-PhH), 7.48 (t, 3J(H,H) = 7.5 Hz, 1H; p-

PhH). 13C{1H} NMR (100.64 MHz, CDCl3, 300 K): δ 7.9 (s; C5(CH3)5), 21.2 (s; p-

CH3), 21.5 (s; o-CH3), 92.1 (d, 2J(C,P) = 3.0 Hz; C5(CH3)5), 121.1 (d, 1J(C,P) = 52.0

Hz; ipso-Ph), 128.9 (s; m-Mes), 129.7 (d, 3J(C,P) = 7.8 Hz; m-Ph), 131.1 (d, 4J(C,P)

= 2.1 Hz; p-Ph), 136.7 (s; p-Mes), 137.3 (s; o-Mes), 138.0 (d, 3J(C,P) = 4.3 Hz;

ipso-Mes), 146.9 (d, 2J(C,P) = 8.8 Hz; o-Ph). 31P NMR (101.3 MHz, CDCl3, 300 K):

δ –82.8 (t, 1J(P,H) = 378.6 Hz, PH2). IR (KBr): ν 3023.9 (w), 2988.2 (m), 2962.1 (m),

2915.9 (s), 2854.1 (w), 2383.6 (m, P–H), 2370.1 (m, P–H), 1610.3 (s), 1565.0 (s),

1448.3 (s), 1376.9 (s), 1027.9 (s), 921.8 (s), 845.6 (s), 806.1 (s), 745.4 (s), 460.9 cm–

1 (s). HR FAB–MS: calcd for C34H42Cl2PIr: 744.2018, found 744.2024. m/z (%): 744

(5) [M]+, 709 (100) [M – Cl]+, 673 (12) [M – Cl – HCl]+, 363 (95) [M – Cl –

DmpPH2]+.

[(ηηηη5-Cp*)IrCl2(PH2Ph)] (6d). A freshly prepared solution of PhPH2 (0.6 M in

Et2O, 1.05 mL, 0.63 mmol) was added to an orange solution of [(η5-Cp*)IrCl2]2

(0.250 g, 0.314 mmol) in CH2Cl2 (20 mL) at room temperature. After 30 min.,

the resulting mixture was filtered over a short silica column and eluted with

CHCl3, after which the orange fractions were combined and evaporated to

dryness. Subsequent crystallization from CH2Cl2/pentane at –20 °C yielded

[(η5-Cp*)IrCl2(PH2Ph)] (6d) as yellow microcrystals, which were washed with

pentane and dried in vacuo (0.268 g, 0.528 mmol, 84%). Mp: 222 °C (dec). 1H

NMR (250.13 MHz, CDCl3, 300 K): δ 1.63 (s, 15H; C5(CH3)5), 5.88 (d, 1J(H,P) =

394.1 Hz, 2H; PH2), 7.44–7.47 (m, 3H; m- and p-PhH), 7.78–7.85 (m, 2H; o-PhH). 13C{1H} NMR (62.90 MHz, CDCl3, 300 K): δ 8.6 (s; C5(CH3)5), 91.9 (d, 2J(C,P) = 3.3

Hz; C5(CH3)5), 122.6 (d, 1J(C,P) = 53.3 Hz; ipso-Ph), 128.8 (d, 2J(C,P) = 10.7 Hz; o-

Ph), 131.5 (d, 4J(C,P) = 2.7 Hz; p-Ph), 133.5 (d, 3J(C,P) = 8.8 Hz; m-Ph). 31P NMR

(101.3 MHz, CDCl3, 300 K): δ –56.3 (t, 1J(P,H) = 394.1 Hz, PH2). IR (KBr): ν 3051.8

(s), 2973.7 (s), 2914.9 (s), 2870.5 (m), 2412.5 (w, P–H), 2389.4 (m, P–H), 1451.2 (s),

1435.8(s), 1378.9 (s), 1156.1 (w), 1070.3 (m), 1031.7 (s), 899.6 (s), 745.4 (s), 697.1

cm–1 (s). HR FAB–MS: calcd for C16H22Cl2PIr: 508.0451, found 508.0459. m/z (%):

508 (8) [M]+, 473 (29) [M – Cl]+, 363 (20) [M – Cl – PhPH2]+.


109

[(Cp*)(Xy–N≡≡≡≡C)IrPMes*C┌───────┐

=NXy] (4a). An orange solution of [(η5-

Cp*)IrCl2(PH2Mes*)] (6a; 0.135 g, 0.20 mmol) in CH2Cl2 (2 mL) was added to a

mixture of DBU (59.8 µL, 0.40 mmol) and Xy–N≡C (0.262 g, 2.0 mmol) in CH2Cl2

(3 mL) at –78 °C, which resulted in a immediate color change to deep

purple. After 1 h, the reaction mixture was allowed to warm up to room

temperature and stirred for an additional hour. After evaporation to dryness,

the yellow residue was washed with pentane (2 x 1 mL) and extracted into

diethyl ether (4 x 15 mL), and the solution was filtered. After concentration of

the solution to a few mL, 4a (0.148 g, 0.171 mmol, 86%) was obtained as

yellow crystals by crystallization at –20 °C. Mp: 143 °C (dec). 1H NMR (400.13

MHz, C6D6, 300 K): δ 1.11 (s, 9H; p-C(CH3)3), 1.60 (s, 15H; C5(CH3)5), 1.92 (s, 9H;

o-C(CH3)3), 1.93 (s, 6H; o-C=NXyCH3), 2.02 (s, 9H; o-C(CH3)3), 2.39 (s, 3H; o-

C≡NXyCH3), 2.45 (s, 3H; o-C≡NXyCH3), 6.69 (d, 3J(H,H) = 7.5 Hz, 2H; m-C=NXy),

6.75 (m, 3J(H,H) = 7.5 Hz, 1H; p-C=NXy), 6.92 (m, 3J(H,H) = 7.0 Hz, 1H; m-C≡NXy),

6.95 (m, 3J(H,H) = 7.0 Hz, 1H; p-C≡NXy), 7.04 (m, 3J(H,H) = 7.0 Hz, 1H; m-C≡NXy),

7.06 (s, 1H; m-Mes*), 7.29 (s, 1H; m-Mes*). 13C{1H} NMR (100.64 MHz, C6D6, 300

K): δ 9.4 (s; C5(CH3)5), 19.5 (s; o-C=NXyCH3), 20.0 and 22.1 (s; o-C≡NXyCH3),

31.2 (s; p-C(CH3)3), 33.9 (d, 4J(C,P) = 11.2 Hz; o-C(CH3)3), 34.3 (s; p-C(CH3)3),

34.6 (d, 4J(C,P) = 6.3 Hz; o-C(CH3)3), 39.8 and 41.0 (s; o-C(CH3)3), 97.1 (s;

C5(CH3)5), 121.9 (s; m-Mes*), 122.7 (s; p-C≡NXy), 123.8 (s; m-Mes*), 125.9 (s; p-

C=NXy), 126.7 (s; o-C≡NXy), 127.8 (s; m-C≡NXy), 127.9 (s; m-C=NXy), 128.8 (s;

m-C≡NXy), 129.7 (s; o-C≡NXy), 130.3 (s; o-C=NXy), 132.1 (d, 1J(C,P) = 92.1 Hz;

C=NXy), 134.1 (s; ipso-C=NXy), 141.5 (s; C≡NXy), 146.1 (s; p-Mes*), 150.6 (d, 4J(C,P) = 12.9 Hz; ipso-C≡NXy), 156.9 (d, 2J(C,P) = 8.4 Hz; o-Mes*), 158.4 (s; o-

Mes*), 180.6 (d, 1J(C,P) = 103.4 Hz; ipso-Mes*). 31P NMR (101.3 MHz, C6D6, 300

K): δ –190.4 (s; PMes*). IR (KBr): v = 3062.4 (w), 2960.2 (s), 2948.6 (s), 2903.3 (s),

2863.8 (s), 2272.7 (very broad w, PC=N), 2072.2 and 2020.1 (s, C≡N), 1637.3

and 1586.2 (s, C=N), 1463.7 (s, C=C), 1390.4, 1381.8 and 1359.6 (s, P–Ar), 1240.0

(s), 1195.7 and 1185.1 (s, P=C), 1124.3 (w), 1090.6 (w), 1025.0 (m, P–Ar), 922.8

(w), 873.6 (w), 773.3 and 745.4 (s, P–C), 708.7 (w), 697.8 (m), 522.6 cm–1 (m).

HR EI–MS: calcd for C37H53IrNP (M – C≡NXy) 735.3548, found 735.3541. m/z (%):

866 (2) [M]+, 735 (100) [M – C≡NXy]+, 590 (16) [M – PMes*]+, 455 (24) [M –

Chapter 4

110

PMes* – Cp*]+. Alternatively, 4a can be prepared from 3a as follows: Xy–N≡C

(0.197 g, 1.50 mmol) was added to a dark purple solution of 3a (0.110 mg,

0.15 mmol) in pentanes (20 mL) at room temperature. After 15 h., the yellow

reaction mixture was evaporated to dryness and the residue was washed

with pentane (2 x 1 mL) and extracted into diethyl ether. After filtration and

concentration to a few mL, yellow crystals of 4a (0.113 g, 0.131 mmol, 87%)

were obtained at –20 °C.

[(Cp*)(Xy–N≡≡≡≡C)IrPMes*C┌───────┐

====NPh] (5). A freshly prepared solution of Ph–N≡C

(0.143 M, 10.5 mL, 1.50 mmol) in CH2Cl2 was added to a dark purple solution

of 3a (0.110 mg, 0.15 mmol) in pentane (10 mL) at room temperature. After

15 h., the yellow reaction mixture was evaporated to dryness and the residue

was extracted into pentane. After removal of a black residue by filtration, all

volatiles were removed in vacuo. The residual yellow solid was extracted into

diethyl ether and after concentration to a few mL, yellow crystals of 5 (0.097

g, 0.116 mmol, 77%) were obtained at –20 °C. Mp: 159 °C (dec). 1H NMR

(400.13 MHz, C6D6, 300 K): δ 0.93 (s, 9H; p-C(CH3)3), 1.62 (s, 15H; C5(CH3)5), 1.94

(s, 15H; o-C(CH3)3 and o-C≡NXyCH3), 2.05 (s, 9H; o-C(CH3)3), 6.63 (d, 3J(H,H) =

7.4 Hz, 2H; m-C≡NXy), 6.70 (t, 3J(H,H) = 7.4 Hz, 1H; p-C≡NXy), 7.00 (t, 3J(H,H) =

7.3 Hz, 1H; p-C=NPh), 7.03 (d, 4J(C,P) = 7.8 Hz, 1H; m-Mes*), 7.18 (d, 3J(H,H) =

7.9 Hz, 2H, m-C=NPh), 7.45 (d, 4J(C,P) = 7.8 Hz, 1H; m-Mes*), 7.82 (d, 3J(H,H) =

7.9 Hz, 2H; o-C=NPh). 13C{1H} NMR (100.64 MHz, C6D6, 300 K): δ 9.4 (s; C5(CH3)5),

19.0 (s; o-C≡NXyCH3), 31.0 (s; p-C(CH3)3), 33.6 (d, 4J(C,P) = 11.3 Hz; o-C(CH3)3),

34.1 (s; p-C(CH3)3), 34.6 (d, 4J(C,P) = 6.8 Hz; o-C(CH3)3), 39.9 and 40.9 (s; o-

C(CH3)3), 96.7 (s; C5(CH3)5), 119.6 (s; m-Mes*), 122.2 (s; m-Mes*), 122.6 (s; o-

C=NPh), 123.8 (s; p-C=NPh), 125.8 (s; p-C≡NXy), 127.9 (s; m-C≡NXy), 128.8 (s;

m-C=NPh), 129.9 (s; ipso-C≡NXy), 131.6 (d, 1J(C,P) = 80.0 Hz; C=NPh), 133.9 (s;

o-C≡NXy), 140.6 (bs, C≡NXy), 146.2 (s; p-Mes*), 152.9 (d, 3J(C,P) = 13.1 Hz;

ipso-C=NPh), 156.9 (d, 2J(C,P) = 8.2 Hz; o-Mes*), 157.7 (s; o-Mes*), 187.6 (d, 1J(C,P) = 101.4 Hz; ipso-Mes*). 31P NMR (101.3 MHz, C6D6, 300 K): δ –190.2 (s;

PMes*). IR (KBr): v = 3069.2 (m), 3047.0 (w), 2966.0 (m), 2945.6 (s), 2913.9 (m),

2898.5 (w), 2866.7 (w), 2851.3 (w), 2283.3 (very broad w, PC=N), 2084.7 and

2037.4 (s, C≡N), 1677.8, 1644.0, 1583.3, and 1544.7 (s, C=N), 1480.1, 1462.8, and


111

1441.5 (s, C=C), 1390.4, 1375.0, and 1359.6 (m, P–Ar), 1315.2 (m), 1261.2 (m),

1198.5 (m, P=C), 1066.4 (m), 1026.0 (m, P–Ar), 898.7 (m), 875.5 (m), 789.7, 763.7,

747.3, and 735.7 (s, P–C), 692.3 and 683.6 (s), 523.6 (m), 496.6 cm–1 (w). HR EI–

MS: calcd for C37H53IrNP (M – C≡NPh) 735.3548, found 735.3517. m/z (%): 838

(2) [M]+, 735 (5) [M – C≡NPh]+, 707 (5) [M – C≡NXy]+, 562 (6) [M – PMes*]+.

[(ηηηη5-Cp*)(Xy–N≡≡≡≡C)IrPMesC┌──────┐

====NXy] (4b). DBU (59.8 µL, 0.40 mmol) was added

to a yellow solution of [(η5-Cp*)IrCl2(PH2Mes)] (6b; 0.110 g, 0.20 mmol) and

Xy–N≡C (0.262 g, 2.0 mmol) in CH2Cl2 (6 mL) at –78 °C, and the mixture was

allowed to warm up to room temperature. After evaporation to dryness, the

residue was washed with pentane (3 mL) and extracted into diethyl ether (2

x 10 mL), and the solution was filtered. After concentration of the solution to

a few mL a yellow solid was obtained, which was recrystallized from diethyl

ether at –20°C to yield 4b (0.089 g, 0.120 mmol, 60%) as yellow crystals. Mp:

122 °C (dec). 1H NMR (400.13 MHz, C6D6, 300 K): δ 1.74 (s, 15H; C5(CH3)5), 1.81

(s, 6H; o-C=NXyCH3), 1.85 (s, 3H; p-MesCH3), 2.46 (s, 6H; o-MesCH3), 2.84 (s, 6H;

o-C≡NXyCH3), 6.44 (bs, 2H; m-Mes), 6.64 (d, 3J(H,H) = 7.3 Hz, 2H; m-C=NXy),

6.71 (m, 3J(H,H) = 7.3 Hz, 1H; p-C=NXy), 6.99 (m, 3J(H,H) = 7.0 Hz, 1H; p-C≡NXy),

7.06 (d, 3J(H,H) = 7.0 Hz, 2H; m-C≡NXy). 13C{1H} NMR (100.64 MHz, C6D6, 300 K):

δ 9.5 (s; C5(CH3)5), 18.8 (s; o-C=NXyCH3), 20.6 (d, 3J(C,P) = 16.5 Hz, o-MesCH3),

20.8 (s; p-MesCH3), 23.8 and 23.9 (s; o-C≡NXyCH3), 96.7 (s; C5(CH3)5), 122.7 (s;

p-C≡NXy), 125.9 (s; p-C=NXy), 127.4 (s; m-C=NXy), 127.5 (s; o-C≡NXy), 128.0 (s;

p-Mes), 128.1 (s; m-C≡NXy), 128.7 (s; m-Mes), 130.2 (s; ipso-C=NXy), 133.7 (s; o-

C=NXy), 135.4 (s; o-Mes), 143.5 (d, 2J(C,P) = 9.7 Hz, C≡NXy), 151.9 (d, 1J(C,P) =

70 Hz, ipso-Mes), C=NXy and ipso-C≡NXy could not be detected. 31P NMR

(101.3 MHz, C6D6, 300 K): δ –218.7 (s; PMes). IR (KBr): v = 3034.6 (w), 2972.8 (w),

2940.0 (w), 2911.0 (m), 2886.0 (m), 2871.5 (m), 2842.6 (w), 2281.4 (very broad

w, PC=N), 2057.7 and 2019.1 (s, C≡N), 1638.2 and 1587.1 (s, C=N), 1459.9 and

1437.7 (s, C=C), 1376.0 (s, P–Ar), 1240.0 (s), 1187.9 (m, P=C), 1025.0 (m, P–Ar),

984.5 (m), 840.8 (s), 774.3 and 761.7 (s, P–C), 678.8 (s), 517.8 cm–1 (s). HR EI–MS:

calcd for C37H44IrN2P 740.2871, found 740.2896. m/z (%): 740 (<1) [M]+, 609 (1)

[M – CNXy]+, 590 (20) [M – PMes]+.

Chapter 4

112

[(Cp*)(Xy–N≡≡≡≡C)IrPDmpC┌──────┐

====NXy] (4c). An orange solution of [(η5-

Cp*)IrCl2(PH2Dmp)] (6c; 0.075 g, 0.10 mmol) in CH2Cl2 (1.5 mL) was added to

a mixture of DBU (29.9 µL, 0.20 mmol) and Xy–N≡C (0.131 g, 1.0 mmol) in

CH2Cl2 (1 mL) at –78 °C and the reaction mixture was allowed to warm up to

room temperature. After evaporation to dryness, the residue was washed

with pentane (2 x 1 mL) and extracted into diethyl ether (2 x 10 mL), and the

solution was filtered. After concentration of the solution to a few mL, 4c

(0.060 g, 0.064 mmol, 64%) was obtained as yellow crystals by crystallization

at –20 °C. Mp: 175 °C (dec). 1H NMR (400.13 MHz, C6D6, 346 K): δ 1.41 (s, 15H;

C5(CH3)5), 1.97 (s, 6H; o-C=NXyCH3), 2.15 (s, 6H; o-MesCH3), 2.20 (s, 6H; o-

MesCH3), 2.23 (s, 3H; p-MesCH3), 2.50 (s, 3H; p-MesCH3), 2.58 (s, 6H; o-

C≡NXyCH3), 6.77 (m, 5H; m-C=NXy, m-Mes, and p-C=NXy), 6.84 (m, 3J(H,H) =

7.4 Hz, 3H; m-PhP and p-C≡NXy), 6.96 (d, 3J(H,H) = 7.4 Hz, 2H; m-C≡NXy), 6.98

(s, 2H; m-Mes), 7.08 (t, 3J(H,H) = 7.4 Hz, 1H; p-PhP). 13C{1H} NMR (100.64 MHz,

C6D6, 300 K): δ 9.1 (d, 3J(C,P) = 2.9 Hz; C5(CH3)5), 19.2 (s; o-C=NXyCH3), 19.8 (s;

p-MesCH3), 21.0 (s; o-MesCH3), 22.0–22.2 (bs; o-MesCH3, o-C≡NXyCH3, o-

C≡NXyCH3, and p-MesCH3), 96.9 (s; C5(CH3)5), 122.4 (s; p-C≡NXy), 126.0 (s, o-

C≡NXy), 126.1 (s; o-C≡NXy), 126.2 (s; p-C=NXy), 127.1 (s; m-C≡NXy), 127.2 (s; p-

PhP), 128.0 (s; ipso-Mes, and m-Mes), 128.5 (s; m-C=NXy), 128.6 (s; m-C≡NXy),

129.0 (s; m-Mes), 129.8 (s; p-Mes), 129.9 (bs; m-PhP), 130.6 (s; ipso-C=NXy),

134.5 (s; o-C=NXy), 136.1 and 136.5 (s; o-Mes), 138.0 (d, 1J(C,P) = 69.5 Hz; ipso-

PhP), 140.8 (s; ipso-Mes), 144.0 (s; C≡NXy), 148.0 (s; ipso-C≡NXy), 148.2 (d, 2J(C,P) = 12.9 Hz; o-Ph), 173.3 (d, 1J(C,P) = 93.5 Hz; P-C=NXy). 31P NMR (101.3

MHz, C6D6, 300 K): δ –216.7 (s; PDmp). IR (KBr): v = 3036.4 (m), 2977.6 (m),

2949.6 (s), 2909.1 (m), 2851.3 (m), 2268.9 (very broad w, PC=N), 2054.8 and

2012.4 (s, C≡N), 1645.0 and 1586.2 (s, C=N), 1457.0 and 1443.5 (s, C=C), 1377.0

(s, P–Ar), 1188.9 (m, P=C), 1088.6 (m), 1030.8 (m, P–Ar), 845.6 (m), 800.3 (m),

761.7 and 745.4 (s, P–C), 672.1 (s), 546.7 (w), 520.0 cm–1 (w). HR EI–MS: calcd

for C52H58IrN2P 934.3967, found 934.3930. m/z (%): 934 (14) [M]+, 803 (100) [M –

C≡NXy]+, 589 (60) [M – DmpPH]+, 454 (16) [M – DmpPH – Cp*]+.

Double dehydrohalogenation of [(ηηηη5-Cp*)IrCl2(PH2Dmp)] (6c). Two

equivalents DBU (8.4 µL, 5.6 µmol) were added to an orange solution of [(η5-


113

Cp*)IrCl2(PH2Dmp)] (6c; 20.9 mg, 2.8 µmol) in CD2Cl2 (0.4 mL) at –10 °C. After

5 min. at room temperature, the reaction mixture turned deep red of which 31P NMR spectroscopy at 263 K showed unreacted [(η5-Cp*)IrCl2(PH2Dmp)] at

δ –81.5 (9.2%) together with phosphinidene [(η5-Cp*)(CD2Cl2)Ir=PDmp)]

(7c(DCM)) at δ 672.1 (11.5%), the mono-dehydrohalogenated product [(η5-

Cp*)(Cl)Ir=PHDmp] at δ 105.8 (d, 1J(P,H) = 369.6 Hz, 50.7%), and two minor

products at δ –17.4 (d, 1J(P,H) = 185.3 Hz, 19.3%), and –67.9 (d, 1J(P,H) = 212.6

Hz, 9.3%).

Attempted synthesis of [(ηηηη5-Cp*)(Xy–N≡≡≡≡C)Ir====PDmp] (3c). An orange

solution of [(η5-Cp*)IrCl2(PH2Dmp)] (6c; 75 mg, 0.10 mmol) in CH2Cl2 (1 mL)

was added to a clear, colorless solution of Xy–N≡C (13 mg, 0.10 mmol) and

DBU (29.8 µL, 0.20 mmol) in toluene (2 mL) at room temperature. After 3 h., a

color changed to dark purple was observed and the phosphinidene [(η5-

Cp*)(Xy–N≡C)Ir=PDmp] (3c) could be detected by 31P NMR spectroscopy at

δ 768.1 (50%) together with another product at –88.6 (d, 1J(P,H) = 202 Hz,

50%).

Acknowledgement. This work was supported by The Netherlands

Foundation for Chemical Sciences (CW) with financial aid from the

Netherlands Organization for Scientific Research (NWO). Dr. M. Smoluch and

J. W. H. Peeters (HR FAB-MS; University of Amsterdam) are acknowledged for

measuring respectively HR EI–MS and HR FAB–MS, and Mr. A.K. Mohamud is

acknowledged for his contribution at the early stage of the study.

Supporting Information. Cartesian coordinates (Å) and energies (au) of all

stationary points. Cif files with crystallographic data and copies of the NMR

spectra of all novel compounds. This material is available free of charge via

the Internet at http://pubs.acs.org.


[1] (a) P. B. Hitchcock, M. F. Lappert, W.–P. Leung, J. Chem. Soc., Chem. Commun.

1987, 1282–1283. (b) R. Bohra, P. B. Hitchcock, M. F. Lappert, Leung, W.–P.,

Chapter 4

114

Polyhedron 1989, 8, 1884.



Dalton Trans. 2007, 1861–1868.

[3] D. W. Stephan, Angew. Chem. 2000, 112, 322–338; Angew. Chem. Int. Ed. 2000,

39, 314–329.


[5] C. C. Cummins, R. R. Schrock, W. M. Davis, Angew. Chem. 1993, 105, 758–761;

Angew. Chem. Int. Ed. Engl. 1993, 32, 756–759.





[8] R. Waterman, G. L. Hillhouse, J. Am. Chem. Soc. 2003, 125, 13350–13351.

[9] (a) T. L. Breen, D. W. Stephan, J. Am. Chem. Soc. 1996, 118, 4204–4205. (b) T. L.

Breen, D. W. Stephan, Organometallics 1996, 15, 5729–5737. (c) G. Zhao, F. Basuli,

U. J. Kilgore, H. Fan, H. Aneetha, J. C. Huffman, G. Wu, D. J. Mindiola, J. Am.

Chem. Soc. 2006, 128, 13575–13585.

[10] R. R. Schrock, Angew. Chem. Int. Ed. 2006, 45, 3748–3759.

[11] (a) D. S. Glueck, F. J. Hollander, R. G. Bergman, J. Am. Chem. Soc. 1989, 111,

2719–2721. (b) D. S. Glueck, J. Wu, F. J. Hollander, R. G. Bergman, J. Am. Chem.

Soc. 1991, 113, 2041–2054. (c) A. A. Danopoulos, G. Wilkinson, T. K. N. Sweet, M. B.

Hursthouse, J. Chem. Soc., Dalton Trans. 1996, 3771–3778.

[12] A. T. Termaten, T. Nijbacker, A. W. Ehlers, M. Schakel, M. Lutz, A. L. Spek, M. L.

McKee, K. Lammertsma, Chem. Eur. J. 2004, 10, 4063–4072.

[13] One example is reported were a titanium phosphinidene reacts with three

equivalents of isocyanide to afford a η2-(N,C)-phosphaazaallene, see: F. Basuli, L.

A. Watson, J. C. Huffman, D. J. Mindiola, Dalton Trans. 2003, 4228–4229.

[14] A. T. Termaten, M. Schakel, A. W. Ehlers, M. Lutz, A. L. Spek, K. Lammertsma, Chem.

Eur. J. 2003, 9, 3577–3582.

[15] (a) K. Lammertsma, Top. Curr. Chem. 2003, 229, 95–119. (b) K. Lammertsma, M. J.

M. Vlaar, Eur. J. Org. Chem. 2002, 1127–1138. (c) F. Mathey, N. H. Tran Huy, A.

Marinetti, Helv. Chim. Acta 2001, 84, 2938–2957.

[16] (a) A. T. Termaten, H. Aktas, M. Schakel, A. W. Ehlers, M. Lutz, A. L. Spek, K.

Lammertsma, Organometallics 2003, 22, 1827–1834. (b) H. Aktas, J. C. Slootweg,


115

A. W. Ehlers, M. Lutz, A. L. Spek, K. Lammertsma, Angew. Chem. 2009, 121, 3154–

3157; Angew. Chem. Int. Ed. 2009, 48, 3108–3111.

[17] Three platinum phosphaazaallene complexes with the general formula PtL2-[η2-

(P,C)-Mes*PCNPh] were characterized spectroscopically, see: M.-A. David, J. B.

Alexander, D. S. Glueck, G. P. A. Yap, L. M. Liable-Sands, A. L. Rheingold,


[18] X-ray crystal structure determination of complexes 4a, 4b, 5. Intensities were

measured at 150(2) K on a Nonius KappaCCD diffractometer with rotating

anode (graphite monochromator, λ = 0.71073 Å) up to a resolution of (sin θ/λ)max

= 0.65 Å–1. Integration was performed with EvalCCD[38] (compounds 4a and 4b)

or HKL2000[39] (compound 5). The program SADABS[40] was used for absorption

correction and scaling. The structures were solved with automated Patterson

methods using the program DIRDIF-99[41] and refined with SHELXL-97[42] against F2

of all reflections. Non hydrogen atoms were refined with anisotropic

displacement parameters. All hydrogen atoms were located in difference Fourier

maps and refined with a riding model. Geometry calculations and checking for

higher symmetry was performed with the PLATON program.[43] Further details are

given in Table 1.

[19] Base-induced double dehydrohalogenation of [(η5-Cp*)IrCl2(PH2Ph)] (6d) in the

presence of ten equivalents of 2,6-xylyl isocyanide gave a mixture of

unidentifiable products.

[20] (a) J. Ho, Z. Hou, R. J. Drake, D. W. Stephan, Organometallics 1993, 12, 3145–3157.

(b) A. Mahieu, A. Igau, J.-P. Majoral, Phosphorus, Sulfur Silicon 1995, 104, 235–239.

[21] (a) Z. Hou, D. W. Stephan, J. Am. Chem. Soc. 1992, 114, 10088–10089. (b) Z. Hou, T.

L. Breen, D. W. Stephan, Organometallics 1993, 12, 3158–3167.

[22] E. W. Jandciu, P. Legzdins, W. S. McNeil, B. O. Patrick, K. M. Smith, Chem.

Commun. 2000, 1809–1810; and references therein.


Chem. Eur. J. 2003, 9, 2200–2208.

[24] The, alternative, direct [1+2]-cycloaddition of HNC to [(η5-Cp)Ir≡NH] 1' to afford

η2-carbodiimide complex 9' was not investigated in detail.

[25] F. L. Hirshfeld, Theoret. Chim. Acta 1977, 44, 129–138.

[26] ADF2005.01b and ADF2006.01, SCM, Theoretical Chemistry, Vrije Universiteit,

Amsterdam, The Netherlands, http://www.scm.com/.



Chapter 4

116




Chem. Phys. 1988, 88, 322–328.


[32] (a) G. Schreckenbach, T. Ziegler, J. Phys. Chem. 1995, 99, 606–611. (b) G.

Schreckenbach, T. Ziegler, J. Quantum Chem. 1997, 61, 899–918. (c) S. K. Wolff, T.

Ziegler, J. Chem. Phys. 1998, 109, 895–905. (d) S. K. Wolff, T. Ziegler, E. van Lenthe,

E. J. Baerends, J. Chem. Phys. 1999, 110, 7689–7698.

[33] A. H. Cowley, J. E. Kilduff, T. H. Newman, M. Pakulski, M. J. Am. Chem. Soc. 1982,

104, 5820–5821.

[34] E. Urnezius, J. D. Protasiewicz, Main Group Chem. 1996, 1, 369–372.


405, 183–194.

[36] C. Overländer, J. J. Tirrée, M. Nieger, E. Niecke, C. Moser, S. Spirk, R. Pietschnig,

Appl. Organometal. Chem. 2007, 21, 46–48.

[37] (a) I. Ugi, R. Meyr, Chem. Ber. 1960, 93, 239–248. (b) R. Obrecht, R. Hermann, I.

Ugi, Synthesis 1985, 400–402.


2003, 36, 220–229.

[39] Z. Otwinowski, W. Minor, In Methods in Enzymology; C.W. Carter, Jr., R.M. Sweet,

Eds.; Academic Press, 1997, Vol. 276, pp 307–326.

[40] Sheldrick, G. M. (1999). SADABS: Area-Detector Absorption Correction, v2.10,


[41] P.T. Beurskens, G. Admiraal, G. Beurskens, W.P. Bosman, S. Garcia-Granda, R.O.

Gould, J.M.M. Smits, C. Smykalla, (1999) The DIRDIF99 program system, Technical

Report of the Crystallography Laboratory, University of Nijmegen, The

Netherlands.




117

Table 1: Details of the X-ray crystal structure determinations.

4a 4b 5

formula C46H62IrN2P C37H44IrN2P C44H58IrN2P

Fw 866.15 739.91 838.09

crystal colour yellow yellow yellow

crystal size [mm3] 0.24x0.21x0.10 0.40x0.15x0.12 0.27x0.04x0.04

crystal system monoclinic monoclinic monoclinic

space group P21/c (no. 14) P21/c (no. 14) P21/c (no. 14)

a [Å] 14.3239(4) 11.8147(2) 11.8029(1)

b [Å] 16.0921(4) 14.3753(4) 16.4042(1)

c [Å] 21.5079(7) 20.8906(4) 20.8175(2)

β [°] 122.067(1) 113.837(2) 96.7501(3)

V [Å3] 4201.2(2) 3245.39(12) 4002.68(6)

Z 4 4 4

Dx [g/cm3] 1.369 1.514 1.391

µ [mm–1] 3.248 4.191 3.407

abs. corr. multi-scan[40] multi-scan[40] multi-scan[40]

abs. corr. range 0.45 – 0.72 0.21 – 0.61 0.47 – 0.87

refl. meas. / unique 69676 / 9645 68819 / 7452 82096 / 9159

param. / restraints 469 / 0 382 / 0 449 / 0

R1/wR2 [I>2σ(I)] 0.0187 / 0.0358 0.0193 / 0.0389 0.0222 / 0.0426

R1/wR2 [all refl.] 0.0313 / 0.0391 0.0277 / 0.0412 0.0334 / 0.0455

S 1.043 1.053 1.043

ρmin/max [e/Å3] -0.54 / 1.05 -0.92 / 1.35 -0.56 / 0.99

Chapter 4

118

η3-Diphosphavinylcarbene:

A P2 Analogue of the Dötz Intermediate

Halil Aktas,† J. Chris Slootweg,† Andreas W. Ehlers,† Martin Lutz,‡ Anthony L.

Spek,‡ and Koop Lammertsma*,†




Angew. Chem. 2009, 121, 3154–3157; Angew. Chem. Int. Ed. 2009, 48, 3108–3111

Abstract:Abstract:Abstract:Abstract: The reaction of in situ generated phosphinidenes with phosphaalkynes is a facile route to

the new η3-diphosphavinylcarbene 5555, which shows facile ligand exchange reactions or undergoes

an unprecedented rearrangement that involves phosphinidene complex 8888 and η3-

phosphaalkenylphosphinidene 9999, the 1,3-isomer of 5555.

Chapter 5

Chapter 5

120

5.1 Introduction

Olefin metathesis[1] constitutes a powerful tool for the construction of

a plethora of unsaturated building blocks, pharmaceuticals and

advanced materials. The principle steps involve a transition-metal

carbene complex (A) that undergoes a [2+2] cycloaddition/cyclo-

reversion protocol with alkenes according to the Chauvin

mechanism.[2] For the widely used Grubbs catalysts, however, support

for the key intermediate of this process, the four-membered

metallacyclobutane, relies on theoretical studies[3] and low-

temperature NMR spectroscopy.[4,5] The unsaturated analogues

provide more insight. For example, reaction of alkynes with the

second-generation Grubbs catalyst gives stable η3-vinylcarbene

complexes (1, Mes = 2,4,6-Me3C6H2) that form after rearrangement of

the initial ruthenium cyclobutenes (B).[6,7] These puckered metal η3-

vinylcarbene complexes[8] (C) are highly potent and represent the

key intermediate in enyne metathesis,[9] alkyne polymerization (via

D),[10] and the versatile Fischer carbene mediated Dötz

(benzannulation) reaction (which involves complexes such as 2).[11,12]

In our quest for novel reactivity patterns and in light of the diagonal

relationship between carbon and phosphorus,[13] we here report on a

diphosphorus analogue of η3-vinylcarbene complex C (P2-C).[14] Our

approach involves the reaction of in situ generated phosphinidene

η3-Diphosphavinylcarbene: A P2 Analogue of the Dötz Intermediate

121

[M=PMes*] (Mes* = (M = 2,4,6-tBu3C6H2;[15] M = (η6-pCy)Ru (pCy =

para-cymene), (η5-Cp*)Ir (Cp* = C5Me5)), the phosphorus analogues

of carbenes, with phosphaalkynes (RC≡P; R = Mes*, tBu).

5.2 Synthesis of ηηηη3-diphosphavinylcarbenes

Treatment of primary phosphine complexes [Cl2M(PH2Mes*)] 3 with

two equivalents of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) in the

presence of one equivalent Mes*C≡P afforded η3-

diphosphavinylcarbenes 5a (P2-C) as the sole products in 89% (Ru)

and 76% (Ir) yield after crystallization (Scheme 1). The synthesis of 5a is

remarkably selective, as evidenced by the single AB spin system in

the 31P NMR spectrum (Ru-5a: δ31P 52.1 and –4.8 ppm, 1JP,P = 431.4 Hz;

Ir-5a: δ31P 55.7 and –21.1 ppm, 1JP,P = 394.0 Hz); the large P,P coupling

indicates the presence of a P–P bond. The highly deshielded

resonances in the 13C NMR spectrum at δ = 306.9 (Ru-5a) and 260.3

ppm (Ir-5a) are diagnostic for metal alkylidenes.[6,16] They

unambiguously support the carbenoid nature of the ring carbon

atom and exclude the diphosphametallacyclobutene (P2-B)

conformation.

Scheme 1. Synthesis and reactivity of η3-diphosphavinylcarbene 5.

Chapter 5

122

X-ray crystal structure determination[17] of dark-green Ru-5a (Figure

1) and deep red crystals of Ir-5a (Figure 2) revealed unequivocally the

puckered structure of the unique η3-diphosphavinylcarbene ligand.

Complex Ru-5a displays short metal-alkylidene and P2–C19 bonds

(1.928(2) and 1.879(2) Å, respectively) and typical single metal–P

(2.4339(5) and 2.4478(6) Å), and P–P bonds 2.1771(8) Å).

Figure 1. Displacement ellipsoid plot of Ru-5a with ellipsoids drawn at the 50%

probability level. Hydrogen atoms and toluene solvent are omitted for clarity.

Selected bond lengths [Å], angles [°] and torsion angles [°]: Ru1–P1 2.4339(5), Ru1–P2

2.4478(6), Ru1–C19 1.928(2), Ru1–pCy(cg) 1.7473(7), P1–C1 1.879(2), P1–P2 2.1771(8),

P2–C19 1.759(2), C19–C20 1.480(3); P2–P1–Ru1 63.84(2), P1–P2–C19 88.26(7), Ru1–C19–

P2 83.06(8); P1–P2–Ru1–C19 111.06(9).

Whereas no intermediates could be detected by variable-

temperature NMR spectroscopy during the formation of Ru-5a,

selective monodehydrohalogenation of iridium precursor 3 occurred

at –80 °C to give syn/anti-Ir-4 (31P NMR δ = 119.5 (1JP,H = 398.9 Hz) and

128.3 ppm (1JP,H = 369.6 Hz)) in a 5:1 ratio. At this temperature, Ir-4 is

unreactive towards Mes*C≡P, suggesting that the second

dehydrohalogenation to afford the transient [M=PMes*] takes place

prior to phosphaalkyne addition. At –60 °C the gradual formation of

Ir-5a was observed.


123

Figure 2. Displacement ellipsoid plot of Ir-5a with ellipsoids drawn at the 50%

probability level. Only the major disorder component is drawn. Hydrogen atoms are

omitted for clarity. Selected bond lengths [Å], angles [°] and torsion angles [°] for the

major disorder component of Ir-5a: Ir1–P1 2.4995(10), Ir1–P2 2.3960(10), Ir1–C1 1.943(4),

P1–C1 1.756(4), P1–P2 2.1961(13), P2–C20 1.920(3), C1–C2 1.461(5); Ir1–P2–P1 65.79(4),

P2–P1–C1 88.62(15), Ir1–C1–P1 84.86(18); Ir1–C1–P1–P2 –52.59(12), P2–P1–Ir1–C1

114.73(18).

5.3 Synthesis of 1,3-diphospha-3H-indenes

Surprisingly, reaction of 3 with the less sterically congested

phosphaalkyne tBuC≡P using two equivalents of DBU resulted, at

room temperature, in yellow crystalline 6b (80% (Ru), 87% (Ir); Scheme

1) instead of the expected carbene 5b, as indicated by a different

set of signals in the 31P NMR spectrum. The smaller P,P couplings are

characteristic (Ru-6b: δ = –1.5 and –92.1 ppm, 2JP,P = 20.7 Hz; Ir-6b: δ =

–32.7 and –113.0 ppm, 2JP,P = 24.5 Hz), as is the absence of a

carbenoid resonance in the 13C NMR spectrum. The molecular

structure of Ru-6b, established by single-crystal X-ray structure

determination,[17] exhibits a number of fascinating features. First and

foremost, 6b contains an unprecedented 1,3-diphospha-3H-indene

moiety,[18] bearing a dearomatized Mes*-substituent (Figure 3). The

striking resemblance of this stable P2-entity with the benzannulation

(Wheland) intermediate of the Dötz reaction is evident.[12c] Secondly,

Chapter 5

124

η4-coordination of the P2C2 moiety to the ruthenium center is favored

over coordination of the all-carbon butadiene fragment.[19] As a result,

the P2C2 unit displays π delocalization, as evidenced by the similar P–

C bond lengths (1.7787(16)–1.7926(15) Å) and an alternating bond

pattern for the dearomatized Mes* ring.

Figure 3. Displacement ellipsoid plot of Ru-6b with ellipsoids drawn at the 30%

probability level. Hydrogen atoms are omitted for clarity. Selected bond lengths [Å]

and torsion angles [°]: Ru1–P1 2.3664(4), Ru1–P2 2.3863(4), Ru1–C1 2.2624(14), Ru1–

C19 2.2055(14), Ru1–pCy(cg) 1.7473(7), P1–C1 1.7926(15), P1–C19 1.7802(15), P2–C6

1.8777(15), P2–C19 1.7787(16), C1–C2 1.486(2), C1–C6 1.547(2), C2–C3 1.351(2), C3–

C4 1.461(2), C4–C5 1.341(2), C5–C6 1.518(2); C1–P1–C19–P2 2.11(9).

5.4 Mechanism

Monitoring the reaction of ruthenium precursor 3 with tBuC≡P by

variable-temperature NMR spectroscopy showed full conversion at –

80 °C to dark-green Ru-5b (31P NMR: δ = –2.0 and –35.7 ppm, 1JP,P =

400.3 Hz) and above –40 °C its subsequent rearrangement to yellow

Ru-6b. Monitoring the corresponding reaction of the iridium congener

of 3 at –80 °C showed, besides the monodehydrohalogenation to

syn/anti-Ir-4 (ca. 3:1 ratio), the formation of two new P2 species (10:1


125

ratio) with sizable P,P and P,H couplings (major isomer, 31P NMR δ =

358.7 and –138.9 ppm, 1JP,P = 164.0 Hz, 1JP,H = 387.8 Hz). We ascribe

these products to the syn and anti adducts of Ir-7b (Figure 4), which

suggests that in this case the addition of the less congested

phosphaalkyne tBuC≡P to Ir-4 is plausible.

Figure 4. Intermediate Ir-7b and model structure Ir-7’, calculated at the B3PW91/6–

31G(d,p) (LANL2DZ for Ir) level of theory. Selected bond lengths [Å] and a torsion

angle [°] of Ir-7’: Ir–P1 2.304, Ir–C1 1.992, P1–P2 2.214, P2–C2 1.707; P1–P2–C1–Ir 4.86.

The intermediacy of Ir-7b is supported by the calculated NMR

parameters[20] of the unsubstituted model structure Ir-7’ (31P NMR δ =

352.6 and –110.8 ppm, 1JP,P = 256.9 Hz; Figure 4). At –40 °C, Ir-7b

underwent the second base-induced dehydrohalogenation to yield

Ir-5b (31P NMR δ = –3.2 and –24.4 ppm, 1JP,P = 371.6 Hz), which upon

warming to room temperature rearranged into Wheland product Ir-

6b, as was evident by the color change from red to yellow.

The remarkable conversion of 5b into 6b was examined by

B3PW91/6–31G(d,p) calculations on model structures containing H

instead of tBu substituents (labeled 5’, 6’, etc.; Figure 5). No simple

direct pathway was found, but instead one that involves

phosphinidene complex 8’ and η3-phosphaalkenyl phosphinidene 9’,

the 1,3-isomer of 5. Interestingly, the planar diphosphametalla-

cyclobutene conformer (P2-B) is not an energy minimum, but

corresponds to the transition-state structure for inversion of puckered

5’ (∆E‡ = 24.4 (Ru), 22.1 (Ir) kcal�mol–1). Isomerization of carbene 5’ to

Chapter 5

126

the less stable phosphinidene 8’ proceeds by P–P bond cleavage (Ru:

∆E = 19.8, ∆E‡ = 20.6 kcal�mol–1). This step is then followed by facile

rotation of the η2-coordinated phosphaalkyne ligand and P–C bond

formation to give regioisomer 9’ (Ru: ∆E = –22.5, ∆E‡ = 4.8 kcal�mol–1).

Finally, the subsequent electrophilic attack of P2 affords the unique

Wheland product 6’ (Ru: ∆E = –0.6, ∆E‡ = 17.7 kcal�mol–1), of which the

experimental analogues, Ru-6b and Ir-6b, can be isolated.

5.5 Reactivity

For an initial assessment of the reactivity of the novel η3-

diphosphavinylcarbenes 5, we treated the ruthenacycle Ru-5a with

one equivalent of tBuC≡P and observed the fast and selective

formation of bicyclic Ru-6b (t½ = 18 min at 0 °C; 93% yield according

to 31P NMR spectroscopy) together with elimination of Mes*C≡P

(Scheme 2).

Scheme 2. Reactivity of Ru-5a towards phosphines.

Addition of PPh3 afforded instead terminal phosphinidene complex

[(η6-pCy)(Ph3P)Ru=Mes*] Ru-10[14a] (t½ = 100 min at 0 °C; quantitative

yield). Both conversions follow an unprecedented pathway for η3-

vinylcarbenes for which an associative ligand exchange at the

phosphinidene stage (8) is proposed, as the dissociative pathway

(Ru-8’ → C6H6Ru=PPh + HC≡P) corresponds to an uphill process (∆E =

28.8 kcal�mol–1).

η3-Diphosphavinylcarbene: A P

2 Analogue of the Dötz Intermediate

127

Figure 5. R

ela

tive

B3P

W91

/6–3

1G(d

,p)

(LA

NL2

DZ

for

Ru

an

d Ir

) e

ne

rgie

s (Z

PE c

orr

ec

ted

, in

kc

al�m

ol–1

) fo

r th

e r

ea

rra

ng

em

en

t o

f 5’

into

6’.

The

re

lativ

e e

ne

rgie

s fo

r th

e η

5 -C

pIr

de

riva

tive

s a

re g

ive

n in

pa

ren

the

ses.

Se

lec

ted

bo

nd

len

gth

s [Å

] fo

r R

u-5’:

Ru

–P1

2.44

3,

Ru

–P2

2.57

1, R

u–C

7 1.

894,

P1–

P2 2

.180

, P2

–C7

1.75

2; TS1:

P1–P

2 2.

876,

P2–

C7

1.63

1; R

u-8’:

Ru

–P1

2.17

7, R

u–P

2 2.

509,

Ru

–C7

2.10

5,

P2–C

7 1.

617;

TS2

: P1

–C7

2.43

9, P

2–C

7 1.

630;

Ru

-9’:

Ru

–P1

2.43

2, R

u–P

2 2.

194,

Ru

–C7

2.24

4, P

1–C

7 1.

784,

P2–

C7

1.77

2; TS3

: R

u–P

1

2.40

4, P

2–C

6 2.

106;

Ru

-6’:

Ru

–P1

2.39

8, R

u–P

2 2.

437,

Ru

–C1

2.21

4, R

u–C

7 2.

199,

P1–

C1

1.80

4, P

1–C

7 1.

787,

P2–

C6

1.88

7, P

2–C

7

1.78

3.

Chapter 5

128

5.6 Conclusion

Phosphorus substitution of metal-bound η3-vinylcarbenes has a

marked influence on their reactivity. Depending on the substituent

pattern, the readily obtained η3-diphosphavinylcarbene complexes 5

show facile ligand exchange reactions or undergo an

unprecedented rearrangement to the stable Wheland product 6.


Density functional theory calculations (B3PW91) were performed with the

Gaussian03 (Revision C.02) suite of programs,[20] using the LANL2DZ basis set

and pseudopotentials for ruthenium and iridium and the 6–31G(d,p) basis set

for all other atoms. The nature of each structure was confirmed by

frequency calculations. Intrinsic reaction coordinate (IRC) calculations were

performed to ascertain the connection between reactant and product.

NMR Calculations. Calculated coordinates were used to calculate the 31P

chemical shielding at the B3PW91/6–311++G(2d,p)//B3PW91/6–31G(d,p

level.[20] These values are relative to a bare phosphorus nucleus and can be

converted to chemical shifts δ relative to an appropriate reference system,

for which we used trimethyl phosphine (σcalcd = 367.8, δexp = –62.0 ppm) to

obtain the relationship: δcalcd(intermediate) = 305.8 ppm –

σcalcd(intermediate). All Cartesian coordinates in Angstroms, energies in

kcal.mol–1 and chemical shielding for calculated compounds are available

free of charge via the Internet at htttp://www. angewandte.org.



atmosphere of dry nitrogen or argon with rigorous exclusion of air and

moisture using flame–dried glassware. All Variable Temperature NMR

measurements were performed under an atmosphere of dry nitrogen in

Wilmad Screw Cap NMR Tubes (Aldrich). 1H, 13C and 31P NMR spectra were

recorded on a Bruker Avance 250 (respectively 250.13, 62.90 and 101.25


129

MHz) or on a Bruker Avance 400 (respectively 400.13, 100.64 and 162.06 MHz)

spectrometer. 1H NMR spectra were internally referenced to CDHCl2 (δ 5.25),

or C6D5H (δ 7.16), 13C NMR spectra to C6D6 (δ 128.06) and 31P NMR spectra

externally to 85% H3PO4. IR Spectra were recorded on a Mattson–6030

Galaxy FT–IR spectrophotometer, high–resolution mass spectra (HR–MS) on a

Finnigan Mat 900 spectrometer operating at an ionization potential of 70 eV,

and fast atom bombardment (FAB) mass spectrometry was carried out using

a JEOL JMS SX/SX 102A four–sector mass spectrometer (70 eV). Melting

points were measured on samples in sealed capillaries on a Stuart Scientific

SMP3 melting point apparatus and are uncorrected. UV spectra were

recorded with a Varian CARY 300 1 Bio UV/Visible spectrophotometer.

Reagents. Solvents were distilled from sodium (toluene), CaCl2 (CH2Cl2), or

LiAlH4 (pentanes, diethyl ether) and kept under an atmosphere of dry

nitrogen. Deuterated solvents were purchased from Aldrich, dried over 4 Å

molecular sieves (CD2Cl2, C6D6), and kept under an atmosphere of dry

nitrogen. All solid starting materials were dried in vacuo. [(η6-

pCy)RuCl2(PH2Mes*)] Ru-3,[14a] [(η5-Cp*)IrCl2(PH2Mes*)] Ir-3,[22] tBu–C≡P,[23] and

Mes*–C≡P,[24] were prepared according to literature procedures.

Experimental Procedures.

Ru-5a. An orange solution of [(η6-pCy)RuCl2(PH2Mes*)] (Ru-3; 0.140 g, 0.239

mmol) in CH2Cl2 (2 mL) was added dropwise to a clear mixture of DBU (71.4

µL, 0.477 mmol) and Mes*–C≡P (0.070 g, 0.24 mmol) in toluene (5 mL) at –

78 °C, which immediately changed the colour of the reaction mixture to

deep green. The mixture was allowed to warm up to room temperature and

was stirred for an additional 30 min. All solvents were evaporated in vacuo

and the residue was extracted into pentanes (3x10 mL). After removal of the

pentanes, Ru-5a was obtained as a dark green solid after crystallization from

diethyl ether at –20 °C (0.170 g, 89%). Suitable dark green crystals for X-ray

crystallography were obtained from a saturated toluene solution at –20 °C.

M.p. ≥ 157 °C (dec). 1H NMR (400.13 MHz, C6D6, 300 K): δ 0.67 (d, 3J(H,H) = 7.5

Hz, 3H; pCy-CH(CH3)2), 0.76 (d, 3J(H,H) = 7.5 Hz, 3H; pCy-CH(CH3)2), 1.04 (s,

Chapter 5

130

9H; o-C(CH3)3), 1.34 (s, 3H; pCy-CH3), 1.39 (s, 9H; p-C(CH3)3), 1.40 (s, 9H; p-

C(CH3)3), 1.79 (s, 9H; o-C(CH3)3), 1.79 (m, 1H; CH(CH3)2), 1.98 (s, 18H; o-

C(CH3)3), 4.71, 4.73, 4.86, 4.88 (AB type, 3J(H,H) = 5.9 Hz, 2H; pCyH), 4.82, 4.83,

4.93, 4.94 (AB type, 3J(H,H) = 5.9 Hz, 2H; pCyH), 7.36 (d, 4J(H,P) = 1.8 Hz, 1H; m-

Mes*H), 7.44 (bs, 3H; m-Mes*H). 13C{1H} NMR (100.64 MHz, C6D6, 300 K): δ 18.8

(s; pCy-CH3), 23.1 (s; pCy-CH(CH3)2), 23.9 (s; pCy-CH(CH3)2), 30.3 (s; pCy-

CH(CH3)2), 31.6 (s; p-C(CH3)3), 31.7 (s; p-C(CH3)3), 31.9 (d, 4J(C,P) = 2.0 Hz; o-

C(CH3)3), 33.3 (s; o-C(CH3)3), 33.4 (s; o-C(CH3)3), 34.3 (br t; o-C(CH3)3), 34.7 (s;

p-C(CH3)3), 34.8 (s; p-C(CH3)3), 37.3 (s; o-C(CH3)3), 38.4 (s; o-C(CH3)3), 39.3 (s;

o-C(CH3)3), 82.8 (s; C6H4), 85.9 (s; C6H4), 92.4 (s; C6H4), 93.1 (s; C6H4), 102.3 (s;

CH3-C6H4), 112.6 (s; (CH3)2CH-C6H4), 120.2 (s; m-Mes*), 121.3 (s; m-Mes*), 121.7

(s; m-Mes*), 137.2 (s; o-Mes*), 140.0 (s; o-Mes*), 146.6 (s; p-Mes*), 146.8 (s; p-

Mes*), 148.2 (d, 1J(C,P) = 113.4 Hz; ipso-Mes*), 154.8 (s; C-ipso-Mes*), 160.0 (d,

3J(C,P) = 5.1 Hz, o-Mes*), 306.9 (CD2Cl2; dd, 1J(C,P) = 97.9 Hz, 2J(C,P) = 16.6 Hz,

RuC). 31P NMR (101.25 MHz, C6D6, 300 K): δ 52.1 (d, 1J(P,P) = 431.4 Hz), –4.8 (d,

1J(P,P) = 431.4 Hz). IR (KBr, cm–1): v = 3098 (w), 2963 (s), 2948 (s), 2907 (s), 2865

(s), 1593 (s), 1476, 1462, and 1445 (s), 1391, and 1361 (s; P–Ar), 1238 (s), 1209

(s), 1117 (w), 1091 (w), 1029 (m; P–Ar), 951 (w), 925 (w), 874 (m), 841 (w), 800

(w), 779 and 745 (s; P–C), 685 (w), 662 (w), 583 (w). HR FAB-MS: calcd for

C47H73P2Ru (M + H): 801.4245, found: 801.4221; m/z (%): 801 (8) [M + H]+, 743

(18) [M – tBu]+, 609 (45) [M – tBu – pCy]+, 493 (60) [M – P2Mes*]+. UV/Vis

(pentanes): λmax(ε) = 211 (59783), 236 (38414), 271 (23674), 319 (17428), 344

(13606), 360 nm (8961).

Ir-5a. Ir-5a was prepared in a similar fashion to that described for Ru-5a by

adding [(η5-Cp*)IrCl2(PH2Mes*)] (Ir-3; 0.280 g, 0.414 mmol) to a toluene

solution of Mes*–C≡P (0.122 g, 0.424 mmol) and DBU (126 µL, 0.828 mmol) to

afford Ir-5a as deep red blocks (0.2814 g, 76%). M.p. 172–174 °C, ≥179 °C

(dec). 1H NMR (400.13 MHz, C6D6, 300 K): δ 1.06 (s, 9H; o-C(CH3)3), 1.23 (s, 15H;

C5(CH3)5), 1.37 (s, 9H; p-C(CH3)3), 1.40 (s, 9H; p-C(CH3)3), 1.75 (s, 9H; o-

C(CH3)3), 1.80 (s, 9H; o-C(CH3)3), 2.12 (s, 9H; o-C(CH3)3), 7.38, 7.38, 7.39, 7.40

(AB type, 4J(H,P) = 1.9 Hz, 2H; m-Mes*H), 7.45 (bs, 1H; m-Mes*H), 7.55 (bs, 1H;

m-Mes*H). 13C{1H} NMR (100.64 MHz, C6D6, 300 K): δ 9.5 (s; C5(CH3)5), 31.7 (s;


131

p-C(CH3)3), 31.8 (s; p-C(CH3)3), 32.3 (d, 5J(C,P) = 2.5 Hz; o-C(CH3)3), 32.4 (d,

5J(C,P) = 4.1 Hz; o-C(CH3)3), 32.8 (s; o-C(CH3)3), 33.0 (s; o-C(CH3)3), 34.7 (s; p-

C(CH3)3), 34.8 (s; p-C(CH3)3), 35.2 (s; o-C(CH3)3), 35.5 (s; o-C(CH3)3), 37.7 (s, o-

C(CH3)3), 38.0 (s, o-C(CH3)3), 38.2 (br s, o-C(CH3)3), 40.1 (br d, 3J(C,P) = 3.9 Hz;

o-C(CH3)3), 95.7 (s; C5(CH3)5), 119.8 (s; m-Mes*), 121.5 (s; m-Mes*), 122.2 (s; m-

Mes*), 134.0 (s; o-Mes*), 141.4 (d, 3J(C,P) = 4.5 Hz; o-Mes*), 146.4 (s; p-Mes*),

146.9 (dd, 1J(C,P) = 121.7 Hz, 2J(C,P) = 8.8 Hz; P-ipso-Mes*), 147.3 (s; p-Mes*),

153.1 (br s; o-Mes*), 156.0 (br s; o-Mes*), 156.3 (br d, 3J(C,P) = 25.2 Hz; ipso-

Mes*), 260.3 (dd, 1J(C,P) = 90.3 Hz, 2J(C,P) = 40.8 Hz; IrC). 31P NMR (101.25 MHz,

C6D6, 300 K): δ 55.7 (d, 1J(P,P) = 394.0 Hz), –21.1 (d, 1J(P,P) = 394.0 Hz). IR (KBr,

cm–1): v = 3081 (w), 2953 (s), 2901 (s), 2864 (s), 1580 (s), 1477, and 1462 (s),

1382, and 1360 (s; P–Ar), 1237, and 1200 (s), 1119 (w), 1070 (w), 1024 (m; P–Ar),

957 (w), 924 (w), 899 (w), 871 (s), 804 (w), 767 and 743 (s; P–C), 706 (w), 654

(w), 640 (w), 604 (w), 581 (w). HR EI-MS (70 eV): calcd for C47H73IrP2: 892.4817,

found: 892.4819; m/z (%): 892 (20) [M]+, 835 (12) [M – tBu]+, 585 (44) [M –

P2Mes*]+. UV/Vis (pentanes): λmax(ε) = 211 (90450), 257 (52785), 344 (20165),

434 nm (6960).

Ru-6b. Ru-6b was prepared in a similar fashion to that described for Ru-5a

by adding [(η6-pCy)RuCl2(PH2Mes*)] (Ru-3; 0.117 g, 0.200 mmol) to a toluene

solution of tBu–C≡P (250 µL, 0.250 mmol, 1M solution in toluene) and DBU

(60.0 µL, 0.400 mmol) to afford Ru-6b as large yellow blocks after

crystallization from a saturated toluene solution at –20 °C (0.098 g, 80%). M.p.

≥149 °C (dec). 1H NMR (400.13 MHz, C6D6, 300 K): δ 0.92 (s, 9H; PC-C(CH3)3),

1.18 (d, 3J(H,H) = 6.9 Hz, 3H; CH(CH3)2), 1.28 (s, 9H; p-C(CH3)3), 1.32 (s, 9H;

P2CC(CH3)3), 1.33 (d, 3J(H,H) = 6.9 Hz, 3H; CH(CH3)2), 1.45 (s, 9 H; o-C(CH3)3),

2.38 (s, 3H; pCy-CH3), 2.77 (septet, 3J(H,H) = 6.9 Hz, 1H; CH(CH3)2), 4.30, 4.32,

4.37, 4.38 (AB type, 3J(H,H) = 5.8 Hz, 2H; C6H4), 4.74, 4.76, 4.79, 4.80 (AB type,

3J(H,H) = 5.8 Hz, 2H; C6H4), 5.87–5.90 (m, 3J(H,P) = 10.3 Hz, 2H; m-Mes*H).

13C{1H} NMR (100.64 MHz, C6D6, 300 K): δ 21.6 (d, 3J(C,P) = 4.9 Hz, pCy-CH3),

24.1 (d, 4J(C,P) = 3.8 Hz; CH(CH3)2), 24.4 (d, 4J(C,P) = 2.2 Hz; CH(CH3)2), 26.6

(dd, 3J(C,P) = 6.7 Hz, 4J(C,P) = 2.5 Hz; PCC(CH3)3), 30.4 (s; p-C(CH3)3), 32.2 (d,

J(C,P) = 12.5 Hz; o-C(CH3)3), 32.8 (s; CH(CH3)2), 34.4 (t, 3J(C,P) = 7.8 Hz;

Chapter 5

132

P2CC(CH3)2), 34.6 (s; p-C(CH3)3), 36.9 (dd, 2J(C,P) = 11.0 Hz, 2J(C,P) = 14.3 Hz;

P2CC(CH3)2), 37.7 (d, 3J(C,P) = 2.2 Hz; o-C(CH3)3), 42.2 (d, 3J(C,P) = 5.4 Hz; PC-

C(CH3)3), 61.4 (dd, 1J(C,P) = 63.3 Hz, 2J(C,P) = 2.3 Hz; PC-C(CH3)3), 75.9 (dd,

1J(C,P) = 18.5 Hz, 2J(C,P) = 4.6 Hz; PCC(t-Bu)P), 84.1 (s; C6H4), 84.1 (s; C6H4),

84.4 (s; C6H4), 88.8 (s; C6H4), 104.0 (s, CH3–C6H4), 116.5 (s, (CH3)2CH–C6H4),

117.0 (dd, 3J(C,P) = 8.3 Hz, 4J(C,P) = 2.5 Hz; m-Mes*), 118.0 (dd, 1J(C,P) = 85.7

Hz, 1J(C,P) = 73.8 Hz; P2C), 121.8 (d, 2J(C,P) = 21.2 Hz; m-Mes*), 144.4 (d,

3J(C,P) = 10.0 Hz, p-Mes*), 154.2 (dd, 2J(C,P) = 14.5 Hz, 3J(C,P) = 2.9 Hz, o-

Mes*). 31P NMR (101.25 MHz, C6D6, 300 K): δ –1.5 (d, 2J(P,P) = 20.7 Hz; CPCtBu),

–92.1 (dd, 2J(P,P) = 20.7 Hz, 3J(P,H) = 11.6 Hz; tBuCPCtBu). IR (KBr, cm–1): v =

2956 (s), 2896 (m), 2862 (w), 1474, 1453, and 1441 (s), 1384, and 1356 (s; P–Ar),

1239 (m), 1192 (m), 1149 (w), 1079 (w), 1049, and 1029 (m; P–Ar), 829 (m), 801

(m), 684 (w). HR EI-MS (70 eV): calcd for C43H63P2Ru (M – tBu): 555.1877,

found: 555.1889; m/z (%): 555 (100) [M – tBu]+. UV/Vis (pentanes): λmax(ε) = 210

(34273), 244 (30424), 379 (3224).

Ir-6b. Ir-6b was prepared in a similar fashion to that described for Ru-5a by

adding [(η5-Cp*)IrCl2(PH2Mes*)] (Ir-3; 0.169 g, 0.250 mmol) to a toluene

solution of tBu–C≡P (312 µL, 0.312 mmol, 1M solution in toluene) and DBU

(74.8 µL, 0.500 mmol) to afford Ir-6b as yellow needles after crystallization

from toluene (0.153 g, 87%). M.p. 173–174 °C. 1H NMR (400.13 MHz, C6D6, 300

K): δ 1.02 (s, 9H; PCC(CH3)3), 1.22 (s, 9H; p-C(CH3)3), 1.38 (s, 9H; P2CC(CH3)3),

1.38 (s, 9H; o-C(CH3)3), 1.17 (s, 15H; C5(CH3)5), 5.57 (dd, 3J(H,P) = 11.3 Hz,

4J(H,H) = 1.3 Hz, 1H; Mes*H), 6.35 (d, 4J(H,H) = 1.3 Hz, 1H; Mes*H). 13C{1H} NMR

(100.64 MHz, C6D6, 300 K): δ 10.3 (s; C5(CH3)5), 26.8 (dd, 3J(C,P) = 6.1 Hz,

4J(C,P) = 2.0 Hz; PC–C(CH3)3), 29.9 (s; p-C(CH3)3), 31.5 (d, 4J(C,P) = 12.4 Hz; o-

C(CH3)3), 32.9 (dd, 3J(C,P) = 8.3 Hz, 3J(C,P) = 6.7 Hz; P2CC(CH3)2), 34.8 (s; p-

C(CH3)3), 35.9 (dd, 2J(C,P) = 13.1 Hz, 2J(C,P) = 10.7 Hz; P2CC(CH3)2), 37.1 (s; o-

C(CH3)3), 46.7 (d, 2J(C,P) = 4.1 Hz; PC-C(CH3)3), 47.9 (d, 1J(C,P) = 61.6 Hz, PC-

C(CH3)3), 75.8 (dd, 1J(C,P) = 14.8 Hz, 2J(C,P) = 4.7 Hz; PCC(t-Bu)P), 92.9 (s;

C5(CH3)5), 101.8 (dd, 1J(C,P) = 81.9 Hz, 1J(C,P) = 72.6 Hz; P2C), 117.1 (dd,

3J(C,P) = 8.0 Hz, 4J(C,P) = 1.9 Hz; m-Mes*), 122.8 (d, 2J(C,P) = 20.0 Hz; m-Mes*),

144.1 (d, 3J(C,P) = 11.0 Hz, p-Mes*), 154.2 (dd, 2J(C,P) = 13.2 Hz, 2J(C,P) = 3.6


133

Hz, o-Mes*). 31P NMR (101.25 MHz, C6D6, 300 K): δ –32.7 (d, 2J(P,P) = 24.5 Hz;

CPCtBu), –113.0 (dd, 2J(P,P) = 24.5 Hz, 3J(P,H) = 11.2 Hz; tBuCPCtBu). IR (KBr,

cm–1): v = 2950 (vs), 2897 (s), 2859 (s), 1640 (w), and 1622 (w), 1462 (s), 1379,

1366, 1356 (s; P–Ar), 1239 (m), 1196 (m), 1114 (w), 1066 (w), 1019 (s), and 990

(w; P–Ar), 835 (m), 797 (m), 689 (w), 633 (w), 588 (w). HR EI-MS (70 eV): calcd

for C29H44P2Ir (M – tBu): 647.2551, found: 647.2573; m/z (%): 647 (100) [M – tBu]+.

UV/Vis (pentanes): λmax(ε) = 211 (60539), 251 (32109), 284 (17354), 379 (4027).

VT-NMR investigations of the formation of 5.

Ru-5a. 2 eq. of DBU (5.6 µL, 37.4 µmol) was slowly added to an orange

solution of [(η6-pCy)RuCl2(PH2Mes*)] (Ru-3; 11.0 mg, 18.7 µmol) and Mes*–

C≡P (5.4 mg, 18.8 µmol) in CD2Cl2 (0.5 mL) at –80 °C, which immediately

changed the colour of the reaction mixture to deep green. At this

temperature, the quantitative conversion into Ru-5a was observed by 31P

NMR spectroscopy and no intermediates could be detected.

Ir-5a. 2 eq. of DBU (4.8 µL, 32.0 µmol) was slowly added to an orange

solution of [(η5-Cp*)IrCl2(PH2Mes*)] (Ir-3; 10.8 mg, 16.0 µmol) and Mes*–C≡P

(4.6 mg, 16.0 µmol) in CD2Cl2 (0.5 mL) at –80 °C, which immediately changed

the colour of the reaction mixture to deep red. At this temperature, selective

(mono-)dehydrohalogenation of Ir-3 was observed by 31P NMR spectroscopy

to give two isomers of intermediate Ir-4 quantitatively (5:1 ratio). This

assignment was confirmed by repeating the experiment using only 1 eq. of

DBU with or without 1 eq. of phosphaalkyne present. 31P NMR (101.25 MHz,

CD2Cl2, 193 K): δ 128.3 (d, 1J(P,H) = 369.6 Hz, 17%), 119.5 (d, 1J(P,H) = 398.9 Hz,

83%). 1H NMR (400.13 MHz, CD2Cl2, 193 K): δ 8.57 (d, 1J(H,P) = 369.6 Hz, 17%),

9.60 (d, 1J(H,P) = 399.2 Hz, 83%). Warming up the reaction mixture to –60 °C

showed the gradual formation of Ir-5a that became the sole product at

room temperature.

VT-NMR investigations of the formation of 6.

Ru-6b. 2 eq. of DBU (6.0 µL, 40.0 µmol) was slowly added to an orange

solution of [(η6-pCy)RuCl2(PH2Mes*)] (Ru-3; 11.7 mg, 20.0 µmol) and tBu–C≡P

Chapter 5

134

(20.0 µmol, 20.0 µL of a 1M solution in toluene) in CD2Cl2 (0.5 mL) at –80 °C,

which immediately changed the colour of the reaction mixture to deep

green. At this temperature, the quantitative conversion into Ru-5b was

observed by 31P NMR spectroscopy and no intermediates could be

detected. 31P NMR (101.25 MHz, CD2Cl2, 193 K): δ –2.0 (d, 1J(P,P) = 400.3 Hz), –

35.7 (d, 1J(P,P) = 400.3 Hz). Warming up the reaction mixture to –40 °C

showed the gradual formation of Ru-6b that became the main product

above 0 °C (> 81%).

Ir-6b. 2 eq. of DBU (6.2 µL, 41.4 µmol) was slowly added to an orange

solution of [(η5-Cp*)IrCl2(PH2Mes*)] (Ir-3; 14.0 mg, 20.7 µmol) and tBu–C≡P

(20.7 µmol, 20.7 µL of a 1M solution in toluene) in CD2Cl2 (0.5 mL) at –80 °C,

which immediately changed the colour of the reaction mixture to deep red.

At this temperature, the two isomers of intermediate Ir-4 were observed by

31P NMR spectroscopy as well as the two isomers of intermediate Ir-7b (10:1

ratio), which are generated via the subsequent reaction with tBu–C≡P, along

with unreacted starting material. 31P{1H} NMR (101.25 MHz, CD2Cl2, 193 K): δ

358.7 (d, 1J(P,P) = 164.0 Hz, 28.5%; Ir-7b-major), 342.4 (d, 1J(P,P) = 185.4 Hz,

2.8%; Ir-7b-minor), 128.7 (s, 1.5%; Ir-4-minor), 119.5 (s, 4.8%; Ir-4-major), –54.0 (t,

1J(P,H) = 392.8 Hz, 15.6%; Ir-3), –70.8 (s, 15.5%; tBu–C≡P), –138.9 (d, 1J(P,P) =

164.0 Hz (1J(P,H) = 387.8 Hz), 28.5%; Ir-7b-major), –163.4 (d, 2J(P,P) = 185.4 Hz

(1J(P,H) = 405 Hz), 2.8%; Ir-7b-minor). Warming up the reaction mixture to –

40 °C showed the formation of product Ir-5b (δ31P: –3.2 (d, 1J(P,P) = 371.6 Hz),

–24.4 (d, 1J(P,P) = 371.6 Hz)), which upon warming up to room temperature

rearranged quantitatively into Ir-6b.

Ligand Exchange Reactions.

Reaction of Ru-5a with PPh3. To a cooled, green solution of Ru-5a (12.0 mg,

15.0 µmol) in CD2Cl2 (0.5 mL) was added 1 eq. of PPh3 (3.9 mg, 15.0 µmol) at

–78 °C, which left the colour of the reaction mixture unchanged.

Subsequently, the reaction was monitored by 31P NMR spectroscopy at 0 °C

to afford [(η6-pCy)(Ph3P)Ru=PMes*] (Ru-10) and Mes*–C≡P quantitatively (t½

(0 °C, CD2Cl2) = 100 min.).


135

Reaction of Ir-5b with PPh3. To a deep red solution of Ir-5a (13.4 mg, 15.0

µmol) in C6D6 (0.5 mL) was added 1 eq. of PPh3 (3.9 mg, 15.0 µmol).

Subsequently, the reaction was monitored by 31P NMR spectroscopy at 50 °C

for 210 h to afford as main products [(η5-Cp*)(Ph3P)Ir=PMes*] (Ir-10; 60.3%)

and Mes*–C≡P (t½ (50 °C, C6D6) = 28 h and 15 min.), along with unidentified

by-products. 31P{1H} NMR (101.25 MHz, C6D6, 273 K): δ 686.6 (d, 2J(P,P) = 102

Hz, 60.3%; Ph3PIr=PMes*), 153.7 (d, J(P,P) = 18.2 Hz, 5.5%), 34.1 (s, 22.2%; Mes*–

C≡P), 24.8 (s, 9.7%), 23.6 (d, 2J(P,P) = 102 Hz, 60.3%; Ph3PIr=PMes*), 14.2 (d,

J(P,P) = 62.7 Hz, 6.6%), –24.9 (d, J(P,P) = 18.2 Hz, 5.7%), –38.9 (s, 4.9%), –79.5 (s,

4.6%), –96.8 (s, 2.7%). The reaction in CD2Cl2 at 20°C gave after 68 hours only

0.8% of Ir-10, whereas in refluxing CD2Cl2 a variety of products were obtained

of which D2C=PMes* could be identified.

Reaction of Ru-5a with tBu–C≡≡≡≡P. To a cooled, green solution of Ru-5a (12.0

mg, 15.0 µmol) in CD2Cl2 (0.5 mL) was added 1 eq. of tBu–C≡P (15.0 µmol,

15.0 µL of a 1M solution in toluene) at 0 °C. Subsequently, the reaction was

monitored by 31P NMR spectroscopy at 0 °C for 2.5 h, during which the

colour of the reaction mixture turned yellow, to afford Ru-6b (≥ 93%) and

Mes*–C≡P (t½ (0 °C, CD2Cl2) = 17 min. and 48 sec.).

Acknowledgement. This work was partially supported by the Council for

Chemical Sciences of the Netherlands Organization for Scientific Research

(NWO/CW). We thank Dr. M. Smoluch (Vrije Universiteit, Amsterdam) and J.

W. H. Peeters (University of Amsterdam) for measuring high-resolution mass

spectra.


[1] (a) R. R. Schrock, Angew. Chem. 2006, 118, 3832–3844; Angew. Chem. Int. Ed.

2006, 45, 3748–3759. (b) R. H. Grubbs, Angew. Chem. 2006, 118, 3845–3850;

Angew. Chem. Int. Ed. 2006, 45, 3760–3765. (c) Handbook of Metathesis (Ed: R. H.

Grubbs), WILEY-VCH, Weinheim, 2003. (d) A. H. Hoveyda, A. R. Zhugralin, Nature

2007, 450, 243–251.

Chapter 5

136

[2] Y. Chauvin, Angew. Chem. 2006, 118, 3824–3831; Angew. Chem. Int. Ed. 2006, 45,

3741–3747.

[3] (a) C. Adlhart, P. Chen, J. Am. Chem. Soc. 2004, 126, 3496–3510. (b) L. Cavallo, J.

Am. Chem. Soc. 2002, 124, 8965–8973. (c) B. F. Straub, Angew. Chem. 2005, 117,

6129–6132; Angew. Chem. Int. Ed. 2005, 44, 5974–5978.

[4] (a) P. E. Romero, W. E. Piers, J. Am. Chem. Soc. 2005, 127, 5032–5033. (b) P. E.

Romero, W. E. Piers, J. Am. Chem. Soc. 2007, 129, 1698–1704. (c) E. F. van der Eide,

P. E. Romero, W. E. Piers, J. Am. Chem. Soc. 2008, 130, 4485–4491. (d) A. G.

Wenzel, R. H. Grubbs, J. Am. Chem. Soc. 2006, 128, 16048–16049.

[5] With the appropriate choice of olefinic substrate, the ruthenium-olefin adduct

can be isolated: J. A. Tallarico, P. J. Bonitatebus, Jr., M. L. Snapper, J. Am. Chem.

Soc. 1997, 119, 7157–7158.

[6] T. M. Trnka, M. W. Day, R. H. Grubbs, Organometallics 2001, 20, 3845–3847; and

references therein.

[7] For the all-carbon cyclobutene to 1,3-butadiene rearrangement, see: J. C.

Slootweg, A. W. Ehlers, K. Lammertsma, J. Mol. Model. 2006, 12, 531–536; and

references therein.

[8] T. Mitsudo, Bull. Chem. Soc. Jpn. 1998, 71, 1525–1538.

[9] S. T. Diver, Coord. Chem. Rev. 2007, 251, 671–701.

[10](a) T. J. Katz, S. J. Lee, J. Am. Chem. Soc. 1980, 102, 422–424. (b) D. E. Schuehler, J.

E. Williams, M. B. Sponsler, Macromolecules 2004, 37, 6255–6257. (c) M. G.

Mayershofer, O. Nuyken, J. Polym. Sci. A: Polym. Chem. 2005, 43, 5723–5747.

[11] (a) J. Barluenga, F. Aznar, A. Martín, S. García-Granda, E. Pérez-Carreño, J. Am.

Chem. Soc. 1994, 116, 11191–11192. (b) J. Barluenga, F. Aznar, I. Gutiérrez, A.

Martín, S. García-Granda, M. A. Llorca-Baragaño, J. Am. Chem. Soc. 2000, 122,

1314–1324.

[12] (a) M. M. Gleichmann, K. H. Dötz, B. A. Hess, J. Am. Chem. Soc. 1996, 118, 10551–

10560. (b) K. H. Dötz, Angew. Chem. 1984, 96, 573–594; Angew. Chem. Int. Ed.

Engl. 1984, 23, 587–608. (c) F. Hohmann, S. Siemoneit, M. Nieger, S. Kotila, K. H.

Dötz, Chem. Eur. J. 1997, 3, 853–859.

[13] For example: (a) E. Niecke, A. Fuchs, M. Nieger, Angew. Chem. 1999, 111, 3213–

3216; Angew. Chem. Int. Ed. 1999, 38, 3028–3031. (b) Y. Canac, D. Bourissou, A.

Baceiredo, H. Gornitzka, W. W. Schoeller, G. Bertrand, Science 1998, 279, 2080–

2082.


137

[14] Previously, only mono-phosphametallacyclobutenes (P1-B) have been observed

or postulated, Ru: (a) A. T. Termaten, T. Nijbacker, M. Schakel, M. Lutz, A. L. Spek,

K. Lammertsma, Chem. Eur. J. 2003, 9, 2200–2208. (b) R. Menye-Biyogo, F.

Delpech, A. Castel, V. Pimienta, H. Gornitzka, P. Rivière, Organometallics 2007, 26,

5091–5101. Zr: (c) T. L. Breen, D. W. Stephan, J. Am. Chem. Soc. 1996, 118, 4204–

4205. (d) T. L. Breen, D. W. Stephan, Organometallics 1996, 15, 5729–5737. Ti: (e) G.

Zhao, F. Basuli, U. J. Kilgore, H. Fan, H. Aneetha, J. C. Huffman, G. Wu, D. J.

Mindiola, J. Am. Chem. Soc. 2006, 128, 13575–13585.


Eds.; Georg Thieme Verlag: Stuttgart, 2009; Vol. 42, pp 15–36. (b) K. Lammertsma,

Top. Curr. Chem. 2003, 229, 95–119.

[16] M. Gandelman, K. M. Naing, B. Rybtchinski, E. Poverenov, Y. Ben-David, N.

Ashkenazi, R. M. Gauvin, D. Milstein, J. Am. Chem. Soc. 2005, 127, 15265–15272.

[17] X-ray crystal structure determination of Ru-5a: C47H72P2Ru � C7H8, Fw = 892.19,

brown block, 0.21 x 0.21 x 0.12 mm3, monoclinic, P21/c (no. 14), a = 16.9441(7), b

= 16.2602(3), c = 17.7916(4) Å, β = 93.347(1)°, V = 4893.5(2) Å3, Z = 4, Dx = 1.211



to a resolution of (sin θ/λ)max = 0.65 Å–1 at a temperature of 150(2) K. Intensities

were integrated with EvalCC[25] using an accurate description of the

experimental setup for the prediction of the reflection contours. The reflections

were scaled and corrected for absorption using the program SADABS[26]

(correction range 0.74-0.95). 11252 Reflections were unique (Rint = 0.0524), of

which 8589 were observed [I>2σ(I)]. The structure was solved with automated

Patterson methods using the program DIRDIF.[27] The structure was refined with

SHELXL-97[28] against F2 of all reflections. Non hydrogen atoms were refined with

anisotropic displacement parameters. All hydrogen atoms were introduced in

calculated positions. Hydrogen atoms H39, H40, H42, and H43 were refined freely

with isotropic displacement parameters; all other hydrogen atoms were refined

with a riding model. 552 Parameters were refined with no restraints. R1/wR2 [I >

2σ(I)]: 0.0356 / 0.0674. R1/wR2 [all refl.]: 0.0613 / 0.0755. S = 1.045. Residual

electron density between -0.69 and 1.17 e/Å3. Geometry calculations and

checking for higher symmetry was performed with the PLATON program.[29]

Ir-5a: C47H73IrP2, Fw = 892.19, red block, 0.30 x 0.24 x 0.12 mm3, monoclinic, P21/c

(no. 14), a = 15.2053(1), b = 18.8033(1), c = 17.5203(1) Å, β = 117.9352(4)°, V =

Chapter 5

138

4425.54(5) Å3, Z = 4, Dx = 1.339 g/cm3, µ = 3.12 mm–1. 95880 Reflections were

measured on a Nonius Kappa CCD diffractometer with rotating anode (graphite

monochromator, λ = 0.71073 Å) up to a resolution of (sin θ/λ)max = 0.65 Å–1 at a

temperature of 150(2) K. Intensities were integrated with HKL2000.[30] The

reflections were scaled and corrected for absorption using the program

SADABS[26] (correction range 0.37-0.69). 10141 Reflections were unique (Rint =

0.0479), of which 9064 were observed [I>2σ(I)]. The structure was solved with

automated Patterson methods using the program DIRDIF.[27] The structure was

refined with SHELXL-97[28] against F2 of all reflections. Non hydrogen atoms were

refined with anisotropic displacement parameters. All hydrogen atoms were

introduced in calculated positions and refined with a riding model. The central Ir-

P-P-C part of the molecule was refined with a disorder model corresponding to a

twofold rotation about this unit and two positions for Ir (occupancy 79.2:20.8%).

511 Parameters were refined with 4 restraints. R1/wR2 [I > 2σ(I)]: 0.0288 / 0.0571.

R1/wR2 [all refl.]: 0.0355 / 0.0586. S = 1.213. Residual electron density

between -0.51 and 0.80 e/Å3. Geometry calculations and checking for higher

symmetry was performed with the PLATON program.[29]

Ru-6b: C33H52P2Ru, Fw = 611.76, yellow plate, 0.42 x 0.30 x 0.15 mm3, triclinic, P 1

(no. 2), a = 10.4153(1), b = 10.7589(1), c = 15.1059(2) Å, α = 92.3448(5), β =

109.5700(5), γ = 94.1699(5)°, V = 1586.88(3) Å3, Z = 2, Dx = 1.280 g/cm3, µ = 0.61

mm–1. 32748 Reflections were measured on a Nonius Kappa CCD diffractometer

with rotating anode (graphite monochromator, λ = 0.71073 Å) up to a resolution

of (sin θ/λ)max = 0.65 Å–1 at a temperature of 150(2) K. Intensities were integrated

with HKL2000.[30] The reflections were scaled and corrected for absorption using

the program SADABS[26] (correction range 0.68-0.91). 7266 Reflections were

unique (Rint = 0.0310), of which 6612 were observed [I>2σ(I)]. The structure was

solved with Direct Methods using the program SIR-97.[31] The structure was refined

with SHELXL-97[28] against F2 of all reflections. Non hydrogen atoms were refined

with anisotropic displacement parameters. All hydrogen atoms were located in

difference Fourier maps. Hydrogen atoms H3, H5, H25, H26, H28, and H29 were

refined freely with isotropic displacement parameters; all other hydrogen atoms

were refined with a riding model. 364 Parameters were refined with no restraints.

R1/wR2 [I > 2σ(I)]: 0.0226 / 0.0526. R1/wR2 [all refl.]: 0.0272 / 0.0544. S = 1.061.

Residual electron density between -0.39 and 0.60 e/Å3. Geometry calculations

and checking for higher symmetry was performed with the PLATON program.[29]


139

CCDC 687878 (Ru-5a), 687879 (Ru-6b), and 687880 (Ir-5a) contain the

supplementary crystallographic data for this paper. These data can be obtained

free of charge from The Cambridge Crystallographic Data Centre via

www.ccdc.cam.ac.uk/data_request/cif.

[18] This coordination mode resembles the known diphospha-cyclobutadiene

complexes, e.g. Ru: (a) D. Himmel, M. Seitz, M. Scheer, Z. Anorg. Allg. Chem. 2004,

630, 1220–1228. Ir: (b) P. B. Hitchcock, M. J. Maah, J. F. Nixon, J. Chem. Soc.,

Chem. Commun. 1986, 737–738. Co: (c) R. Wolf, A. W. Ehlers, J. C. Slootweg, M.

Lutz, D. Gudat, M. Hunger, A. L. Spek, K. Lammertsma, Angew. Chem. 2008, 120,

4660–4663; Angew. Chem. Int. Ed. 2008, 47, 4584–4587.

[19] R. Gleiter, I. Hyla-Kryspin, P. Binger, M. Regitz, Organometallics 1992, 11, 177–181.

[20] Gaussian 03, Revision C.02, 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, 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.

[21] The 1,3-isomer of Ir-7’ (Ir-P=C-P) shows distinct resonances: δ31P 652.8 and –93.0,

2JP,P = 53.4 Hz.



[23] W. Rösch, T. Allspach, U. Bergsträßer, M. Regitz in W. A. Herrmann (ed.), “Synthetic

Methods of Organometallic and Inorganic Chemistry (Herrmann/Brauer)”, Vol. 3,

Thieme, Stuttgart, 1996, p. 11–14.

[24] H. Sugiyama, S. Ito, M. Yoshifuji, Angew. Chem. 2003, 115, 3932-3934; Angew.

Chem. Int. Ed. 2003, 42, 3802–3804.

Chapter 5

140


2003, 36, 220–229.

[26] G. M. Sheldrick, 1999. SADABS: Area-Detector Absorption Correction, v2.10,


[27] P.T. Beurskens, G. Admiraal, G. Beurskens, W.P. Bosman, S. Garcia-Granda, R.O.

Gould, J.M.M. Smits, C. Smykalla, (1999) The DIRDIF99 program system, Technical

Report of the Crystallography Laboratory, University of Nijmegen, The

Netherlands.



[30] Z. Otwinowski, W. Minor, In Methods in Enzymology; C.W. Carter, Jr., R.M. Sweet,

Eds.; Academic Press, 1997, Vol. 276, pp 307–326.

[31] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giacovazzo, A.

Guagliardi, A. G. G. Moliterni, G. Polidori, R. Spagna, J. Appl. Cryst. 1999, 32, 115–

119.

141

Table 3. Bond dissociation energies and energy decomposition analysis for (L1)(PhP)Ru–L2 (L1 = NHC, PMe3; L2 = Bz). Energies are given in kcal/mol.

complex ∆Ea’ ∆Ea” ∆E steric ∆E tot ∆E prep BDE

1’-σσσσ –96.2 –67.9 107.3 –56.8 12.8 44.0 1’-ππππ –98.1 –64.6 108.5 –54.3 21.3 33.0

2’ –94.6 –61.2 98.3 –57.5 11.7 45.8

Table 4. Bond dissociation energies and energy decomposition analysis for (C6H6)(L1)Ru–L2 (L1=NHC, PMe3; L2=PPh). Energies are given in kcal/mol.


1’-σσσσ –114.8 –38.6 67.9 –85.5 5.2 80.3 1’-ππππ –108.9 –39.1 67.6 –80.4 10.6 69.8

2’ –111.6 –38.8 64.8 –85.7 4.8 80.9

Appendix 1 BP86/TZP calculated bond dissociation energies and energy decomposition analysis for L(Bz)Ru=PPh (L = NHC, PMe3) belonging to Chapter 2.

142

143

8 – 10 a b c d

Figure 5. Optimized structures (Cs symmetry) for Group 7 transition metals (M = Mn, 8; Tc = 9; and Re = 10).

11 – 13 a b c d

Figure 6. Optimized structures (Cs symmetry) for Group 8 transition metals (M = Fe, 11; Ru = 12; and Os = 13).

14 – 16 a b c d

Figure 7. Optimized structures (Cs symmetry) for Group 9 transition metals (M = Co, 14; Rh = 15; and Ir = 16).

Appendix 2 BP86/TZP optimized structures and calculated bond dissociation energies and energy decomposition analysis for L(Ring)M=PH belonging to Chapter 3.

M P

N N

C

M

P

N N

C

M P

N N

C

M P

N N

C

M P

N N

C

M P

N N

C

M P

N N

C

M P

N N

C

M P

N N

C

M

P

N N

C

M P

N N

C

M P

N N

C

144

Table 5. Calculated relative energies for complexes 8–16. Energies are relative to geometry a in kcal/mol.

complex

Geometry 8 9 10 11 12 13 14 15 16

a 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

b 4.14 4.70 4.77 3.17 2.84 5.30 1.60 1.34 1.66

c 2.15 1.05 1.14 2.42 2.14 1.42 1.91 1.66 2.15

d 4.01 3.15 –0.06 2.53 2.90 0.04 0.93 1.20 4.01

Table 6. Bond dissociation energies and energy decomposition analysis for LnM–C(NHC) bonds in Group 7–9 complexes. Energies are in kcal/mol. Values in parentheses refer to charge transferred (eV) to (positive) and from the metal center.


8c –56.5 (0.54, –0.10) –4.7 (–0.03) 14.8 –46.4 11.7 34.7 9c –58.3 (0.45, –0.11) –4.6 (–0.01) 13.1 –49.8 9.1 40.7

10c –75.7 (0.48, –0.17) –6.0 (–0.01) 15.8 –66.0 16.9 49.1

11c –58.5 (0.56, –0.12) –5.9 (–0.01) 13.8 –50.6 8.4 42.2

12c –58.9 (0.44, –0.10) –5.5 (–0.01) 12.1 –52.4 9.3 43.1

13c –78.6 (0.47, –0.17) –7.5 (–0.01) 19.1 –67.0 12.9 54.1

14c –61.0 (0.60, –0.11) –6.5 (–0.02) 10.6 –56.8 6.0 50.8

15c –63.6 (0.45, –0.12) –6.0 (–0.02) 13.7 –55.9 8.4 47.5

16c –87.9 (0.48, –0.17) –8.4 (–0.01) 25.0 –71.3 12.5 59.8

Table 7. Bond dissociation energies and energy decomposition analysis for LnM–Ring bonds (Ring = Cht (cycloheptatrienyl), 8–10; Bz (benzene), 11–13; Cp (cyclopentadienyl), 14–16) in Group 7–9 complexes. Energies are in kcal/mol.


8a –190.7 –120.8 77.2 –234.3 19.0 215.4 9a –193.2 –129.8 93.7 –229.4 17.6 211.8 10a –226.9 –145.5 136.1 –236.3 20.1 216.2 11a –109.0 –64.6 105.6 –68.0 14.6 53.4 12a –98.4 –65.6 105.2 –58.7 11.9 46.8 13a –132.4 –79.4 143.6 –68.3 16.4 51.8 14a –88.8 –66.0 –42.5 –197.2 8.6 188.7 15a –72.7 –62.2 –48.0 –182.9 5.6 177.3 16a –99.2 –76.7 –16.9 –192.8 8.5 184.4

145

Table 8. Occupied SOMO energies ε (eV) for the [(Ring)(NHC)M] fragments and calculated Hirshfeld charges of M and P. M Ed (σ) Ed (π) M P

8c Mn –2.10 –2.37 +0.00 –0.17 9c Tc –2.17 –2.40 +0.03 –0.19

10c Re –2.35 –2.27 +0.12 –0.22

11c Fe –2.23 –2.47 –0.06 –0.15

12c Ru –2.38 –2.58 +0.12 –0.20

13c Os –2.52 –2.41 +0.02 –0.18

14c Co –2.54 –2.71 –0.08 –0.12

15c Rh –2.65 –2.91 +0.09 –0.16

16c Ir –2.77 –2.71 –0.03 –0.14

146

147

De ontwikkeling van eindstandige fosfinideencomplexen: de zoektocht naar toepassing gaat door

De toepassing van nucleofiele, eindstandige metaalgecomplexeerde

fosfinidenen [LnM=PR] als reactieve complexen is onderontwikkeld in

vergelijking tot hun electrofiele tegenhangers en de welbekende carbeen

en nitreen analogen [LnM=ER] (E = C,N). Dit proefschrift beschrijft efficiënte

syntheseroutes om nieuwe nucleofiele fosfinideencomplexen te genereren

met een verhoogde reactiviteit en bestudeert de elektronische

eigenschappen hiervan met computerchemie. De empirische diagonale

relatie tussen fosfor en koolstof in het Periodiek Systeem maakt de studie

naar fosfinidenen aannemelijk. In 1975 stelde Schmidt voor dat vrije

fosfinidenen (R-P:) betrokken waren bij de vorming van cyclische

fosforhoudende verbindingen. De chemie en toepassing van vrije

fosfinidenen (R-P:) is betrekkelijk schaars doordat vrije fosfinidenen,

afhankelijk van substituent R, in de meeste gevallen een triplet

grondtoestand prefereren, dat leidt tot niet-(stereo)selectieve reactiviteit.

Complexatie aan een metaal levert de gewenste singlet grondtoestand op

voor het fosfinideen. De eerste voorbeelden van stabiele

fosfinideencomplexen werden gerapporteerd door Lappert en collega’s in

1987 en met röntgendiffractie zijn de karakteristieke structuureigenschap-

pen opgehelderd van Cp2M=PMes* (M = Mo, W). Veruit de meeste

reactiviteitstudies hebben zich beperkt tot het stabiele zirconoceen-

complex Cp2(PMe3)Zr=PMes*. Maar recent zijn ook andere vroege en late

overgangsmetaalcomplexen gerapporteerd met beperkte, maar toch

interessante reactiviteit. De electrofiele fosfinideencomplexen (OC)5M=PR (M

= Cr, Mo, W) zijn kortlevend en hun bestaan berust op de reacties met

geschikte substraten. Toch zijn zij veruit het meest bestudeerd en hebben

Samenvatting

148

een pallet van fosforhoudende verbindingen opgeleverd. Recente

theoretische berekeningen toonden aan dat de mate van filiciteit van een

fosfinideencomplex kan worden toegeschreven aan de liganden die

gebonden zijn aan het overgangsmetaal dat vergelijkbaar is met

carbeencomplexen. Liganden met sterke σ-donor eigenschap veroorzaken

ladingsoverdracht op het fosforatoom van het fosfinideen met als gevolg

een gepolariseerd M=P binding en nucleofiel reactiviteit. In homogene

katalyse worden de tertiaire organofosfines (PR3) en N-heterocyclisch

carbenen (NHC) veelzijdig gebruikt als ligand, waarbij de substituent R niet

alleen de σ-donor en π-acceptor capaciteit ervan bepaalt maar ook voor

de nodige sterische afscherming zorgt. Lang werd aangenomen dat NHC

een PR3 mimetic zou zijn. Echter, door zijn goede σ-donerende capaciteit

vormt NHC sterke M−C(NHC) binding en tenonrechte wordt vaak het π-

backbonding aandeel verwaarloosd. Terwijl de eerste NHC complexen al in

1968 werden ontdekt (onafhankelijk van elkaar door Wanzlick en Schönherr,

en Öfele) duurde het tot in 1991 dat de eerste vrije NHC werd

gerapporteerd, dat is gestabiliseerd door sterisch gehinderde adamantyl

groepen op N-atoom. Maar, mede dankzij de gedeelde Nobelprijs voor de

chemie in 2005 van Grubbs, Schrock en Chauvin voor hun bijdrage in de

metathese reacties, nam het gebruik van NHC als ligand explosief toe. De

tweede generatie carbeencomplexen met NHC als ligand, vernoemd naar

Grubbs, vertonen superieure activiteit in katalytische metathese reacties

dan de corresponderende eerste generatie complexen met PR3 als ligand. Is

een vergelijkbare activiteit ook te verwachten voor de analoge

fosfinideencomplexen met NHC als ligand?

In Hoofdstuk 1 wordt een overzicht gegeven van synthese routes naar alle

bekende nucleofiele fosfinideencomplexen. Met name de laatste

ontwikkelingen in hun toepassing als reactieve deeltjes wordt besproken.

In Hoofdstuk 2 wordt de tweede generatie fosfinideencomplex [(η6-

C6H6)(NHC)Ru=PMes*] 1 gerapporteerd en de reactiviteit wordt vergeleken

met de eerste generatie fosfinideencomplex [(η6-C6H6)(PPh3)Ru=PMes*] 2.

Een verhoogde reactiviteit is bepaald voor 1, welke in een efficiënte

149

éénpotsreactie bereid werd door middel van een selectieve

dehydrohalogeneringsreactie van het primaire fosfine complex 3 met 3

equivalenten NHC. Karakteristieke verschillen in de 31P NMR chemische

verschuivingen tussen 1 en 2 werden toegeschreven aan de sterk σ-donor

eigenschappen van het NHC ligand.

N N3 NHC

-2 NHC•HClP

Mes*

Ru

1

NHC

PH2

Mes*

RuCl

Cl 3 NHC

Inderdaad, complex 1 reageert 40x sneller met CH2I2 dan zijn voorganger

2. Het gevormde fosfaalkeen Mes*P=CH2 is een interessante bouwsteen voor

de vervaardiging van nieuwe fosforhoudende liganden en polymeren.

Recycling van het metaalcomplex 4 tot fosfinideen 1 is mogelijk d.m.v. een

reactie met fosfine H2PMes* en de sterke base 1,8-diazabicyclo[5.4.0]undec-

7-een (DBU). Verhoogde nucleofiliciteit van complex 1 is het gevolg van

sterische invloeden, waardoor het NHC ligand gedwongen ‘uit-het-vlak’

roteert. Hierdoor vindt er effectief meer ladingsoverdracht plaats wat

resulteert in meer nucleofiel gedrag van fosfinideencomplex 1.

RuI2

L

H2C P

Mes*CH2I2P

Mes*

Ru

1 L = NHC2 L = PPh3

L

4 L = NHC5 L = PPh3

H2PMes*, DBU

toluene

toluene

In Hoofdstuk 3 is het gebruik van NHC liganden uitgebreid voor de

synthese van stabiele NHC-gefunctionaliseerde fosfinideencomplexen van

de metalen osmium en rhodium. Van rhodium fosfinideencomplex [(η5-

Cp*)(NHC)Rh=PMes*] 7 werd een kristalstructuur bepaald die nauwelijks

verschilt van het NHC-gefunctionaliseerde iridium fosfinideencomplex [(η5-

Cp*)(NHC)Ir=PMes*] 8. Met behulp van theoretische berekeningen aan

modelstructuren van overgangsmetalen van groep 7-9 met verschillende

150

liganden (cycloheptatriene+, benzeen, cyclopentadienyl−) werd hun invloed

en de rotatie van het NHC ligand op het electronische karakter van de

fosfinideencomplexen onderzocht. Een gunstige metaal-NHC π-interactie (~

20%) ontstaat voor complexen met het NHC ligand ‘in-het-vlak’, als is

getoond voor het berekende complex 7’ zonder sterische hindering,

waarvan de metaal-NHC π-interactie niet mag worden verwaarloosd.

6, LM = (pCy)Os7, LM = (Cp*)Rh8, LM = (Cp*)Ir

L

P

Mes*

M

NN

3 NHC

- 2 NHC.HClPH2

Mes*

MCl

L

Cl

Echter, door sterische afstoting roteert het NHC ligand naar een ‘uit-het-

vlak’ symmetrie, zoals bepaalt is voor de complexen 1,7 en 8, waarbij de σ-

interactie de bijbehorende orbitaal energieën domineert. Verder volgt uit

de theoretische berekeningen dat fosfinideencomplexen met

overgangsmetaal rhenium, ruthenium en rhodium naar verwachting het

meest reactief zijn van alle bestudeerde groep 7-9 overgangsmetaal-

complexen doordat de M=P binding het meest gepolariseerd is.

DFT berekende structuur van 7’ Röntgen kristalstructuur van 7

met NHC ‘in-het-vlak’ met NHC ‘uit-het-vlak’

In Hoofdstuk 4 wordt het CH2Cl2-gestabiliseerde 16-electron

fosfinideencomplex [(η5-Cp*)Ir=PR] Z-9c (R = Dmp) experimenteel

aangetoond en ondersteund met 31P NMR berekeningen. Verder wordt de

reactie met isocyanides (Xy−N≡C) gepresenteerd die via, voor de

nucleofiele fosfinideencomplexen onbekende, [1+2]-cycloadditie leidt tot

nieuwe iridafosfiraancomplexen 11. Eerdere pogingen naar de synthese van

Rh

P

N N

C

151

16-electron fosfinideencomplexen [(η5-Cp*)Ir=PR] 9 (R = Mes, Mes*) zonder

stabiliserende liganden resulteerden, via de voorgestelde dimeren [9]2

structuren, in C-H insertieproducten. Terwijl de nitreen analoog [(η5-

Cp*)Ir≡Nt−Bu] van Bergman en collega’s isoleerbaar is, maar instantaan

reageert met isocyanides tot iridaazirideen complexen, hebben wij

experimenteel en theoretisch aangetoond dat afvangen van fosfinideen 9

met isocyanide eerst stabiele, isoleerbare 18-electron complexen 10 geeft.

Vervolgens ringsluit een tweede isocyanide met het fosfinideen en geeft

iridafosfiraan 11.

Ir P

RCp*

C

N

Xy10

Xy-NCIr P

RCp*

a Mes*b Mesc Dmp

C

N

CN

XyXy

E/Z-9

PH2

R

IrCl

Cp*

Cl

-2 DBU•HCl

Xy-NCIr P

RCp*2 DBU

11

R

Het theoretisch berekende reactiemechanisme van nitreen analoog [(η5-

Cp)Ir≡N−H] 12’ met isocyanide (H−N≡C) geeft eerst een ringsluiting, en pas

dan coördineert een tweede isocyanide als ligand om 13’ te vormen.

H-NCIr NCp

Ir N

HCp

C

NH

H-NC Ir N

HCp

C

NH

CN

H12' 13'

H

In Hoofdstuk 5 worden de eerste difosfor-analoga van de veelzijdige Dötz

intermediairen, η3-difosfavinylcarbene complexen 16, gepresenteerd. Als

uitgangspunt wordt ook hier de in situ gegenereerde reactieve 16-

electronen fosfinidenen [LnM=PR] gebruikt. DBU-geïnduceerde reactie van Ir

en Ru primaire fosfine complexen 14 in aanwezigheid van fosfaalkyn

Mes*−C≡P geeft eerst complex 15 en daarna pas het 16-electronen

152

fosfinideen dat wordt afgevangen door het fosfaalkyn om uitsluitend η3-

difosfavinylcarbeencomplexen 16 te vormen. De kristalstructuur van

complex 16 toont de unieke metallacycle waarvan het koolstofatoom

carbenoid is. In aanwezigheid van PPh3 ondergaat 16 een liganduitwisseling

waarbij het fosfaalkyn Mes*−C≡P wordt vervangen en het stabiele 18-

electronen fosfinideencomplex [(η6-pCy)(PPh3)Ru=PMes*] ontstaat.

LM PH

Cl

Mes*

LM = 14a, η5-Cp*Ir

14b, η6-pCyRu;

DBU

- DBU·HCl

15

, DBU

- DBU·HCl

PRC

R = Mes*, tBu

LM

P

P

tBu

tBu

tBu

R16 R = Mes* 17 R = tBu

for R = tBu

tBu

tBu

P

PtBu

R

LM

L

M PH2

Mes*

Cl

Cl

Wanneer het sterisch minder gehinderde fosfaalkyn tBuC≡P wordt

gebruikt, is 16 niet stabiel en vindt er een omlegging plaats tot complex 17

dat overeenkomt met de Dötz benzannulatie reactie. De stapsgewijze

dehydrohalogeneringsreactie van complex 14 werd m.b.v variabele

temperatuur NMR spectroscopy gevolgd. DFT berekeningen aan

modelstructuren toonden aan dat omlegging van 16 via een reeks

reactiestappen verloopt, met P-P spliting, η2-gecoordineerd fosfaalkyn

rotatie, P-C bond vorming en electrofiele aanval op het aromaat als

belangrijkste.

153

Curriculum Vitae

Halil Aktaş was born in the Karşıyaka district of the province Đzmir in Turkey

on the 20th March 1976. In 1987 he moved together with his brother,

sisters and mother to the Netherlands due to the family reunification. At

the age of 14 he decided to study chemistry and in 2001 he obtained his

B.Sc. in Organic Chemistry at the Hogeschool van Utrecht. His B.Sc. thesis

project concerned the study towards Schrock-type phosphinidenes at

the Vrije Universiteit, Amsterdam. He received his M.Sc. in Organic

Chemistry at the same institute in 2003. Under the supervision of prof.dr.

Romano Orru, Halil’s M.Sc. project thesis consisted of the synthetic study

towards 3’-deoxyribolactones using a hydrolysis-induced lactonization

cascade reaction as synthons to utilize "sugar" part of antibiotic

Mureidomycin A. Then, in September 2003 Halil initiated his Ph.D. project

at the Vrije Universiteit, Amsterdam, under the supervision of prof.dr.

Koop Lammertsma towards the nucleophilic phosphinidene complexes.

His current work as R&D Scientist at Tate and Lyle comprises renewable

ingredients and started in March 2008. His work focuses on developing

food and industrial starches manufactured by means of chemical,

thermal, and physical modification processes. He advices up- and

down-scaling of product manufacturing, provides chemical

understanding of products and the corresponding processes, develops

analytical methodologies, and assists in technical support. Halil is an

amateur guitar, saz, and lute player.

154

155

List of publications ‘Nucleophilic Phosphinidene Complexes – Access and Applicability, Halil Aktas, J. Chris Slootweg, Koop Lammertsma, Angew. Chem. 2009, accepted. ‘Iridium Phosphinidene Complexes: A Comparison with Iridium Imido Complexes in Their Reaction with Isocyanides’, Halil Aktas, Jos Mulder, Frans J. J. de Kanter, J. Chris Slootweg, Andreas W. Ehlers, Martin Lutz, Anthony L. Spek, Koop Lammertsma, J. Am. Chem. Soc. 2009, 131, 13531–13537. ‘N-heterocyclic Carbene Functionalized Group 7-9 Transition Metal Phosphinidene Complexes’, H. Aktas, J. C. Slootweg, A. W. Ehlers, M. Lutz, A. L. Spek, K. Lammertsma, Organometallics 2009, 28, 5166–5172. ‘η3-Diphosphavinylcarbene: A P2 Analogue of the Dötz Intermediate’, H. Aktas, J. C. Slootweg, A. W. Ehlers, M. Lutz, A. L. Spek, K. Lammertsma, Angew. Chem. 2009, 121, 3154–3157; Angew. Chem. Int. Ed. 2009, 48, 3108–3111. ‘N-heterocyclic Carbene Functionalized Ruthenium Phosphinidenes: What a Difference a Twist Makes’, Halil Aktas, Marius Schakel, J. Chris Slootweg, Andreas W. Ehlers, Martin Lutz, Anthony L. Spek, Koop Lammertsma, J. Am. Chem. Soc. 2009, 131, 6666–6667. ‘Synthesis of 3’-Deoxyribolactones using a Hydrolysis-Induced Lactonization Cascade Reaction of Epoxy Cyanohydrins’, Danielle J. Vugts, Halil Aktas, Kanar Al-Mafraji, Renske, Frans J. J. de Kander, Eelco Ruijter, Marinus B. Groen, Romano V. A. Orru, Eur J. Org. Chem. 2008, 1336–1339. ‘Terminal Phosphinidene Complexes Cp[R](L)M=PAr of the Group 9 Transition Metals Cobalt, Rhodium, and Iridium. Synthesis, Structures, and Properties’, Arjan T. Termaten, Halil Aktas, Marius Schakel, Andreas W. Ehlers, Martin Lutz, Anthony L. Spek, Koop Lammertsma, Organometallics 2003, 22, 1827–1834.

156

157

Dankwoord

Eindelijk is het zo ver, mijn proefschrift is af! Het is een mooi moment om

iedereen te bedanken die op welke manier dan ook een bijdrage heeft

geleverd aan de totstandkoming van mijn boekje. Een aantal mensen wil ik

in het bijzonder bedanken. Mijn promotor prof.dr. Koop Lammertsma,

bedankt dat u mij heeft gevraagd en de kans heeft gegeven om aan dit

promotieonderzoek te beginnen. Dat mijn hart in het onderzoek lag, was

snel duidelijk, maar de verleiding om naar de industrie te gaan was ook

groot. Gelukkig hebben we aan het begin van deze rit de nodige discussies

gehad en uiteindelijk kwam alles op zijn pootjes terecht. Wat betreft het

onderzoek kreeg ik vanaf dag 1 alle vrijheid. Ondanks mijn exotische ideeën

liep de chemie echter pas in de tweede helft van mijn Ph.D. Wel bleven we

altijd lachen, ook op momenten wanneer de chemie niet liep. Ondanks uw

drukke agenda, wist u toch tijd vrij te maken. Marius Schakel, bedankt voor

de dagelijkse begeleiding in de experimentele werk en je kritische, maar

heldere blik op de materie. Wanneer experimenten mislukten, wist jij wel een

manier om het toch nog te laten lukken. Andreas Ehlers, bedankt voor al je

hulp op het gebied van computerchemie. Onze gesprekken in de OV en

tijdens de korte wandelstukjes naar de VU waren erg stimulerend. Beste Chris

Slootweg, mede dankzij jouw sterke betrokkenheid vanaf jouw dag 1, je

goede ideeën, je kritische blik op het onderzoek en de correcties aan mijn

manuscripten heeft mij enorm geholpen in het zetten van grote stappen.

Bedankt voor al je hulp en de dagelijkse begeleiding.

I would like to extend my gratitude to the reading committee members

prof.dr. F.M. Bickelhaupt, dr. B. de Bruin, prof.dr. C.J. Elsevier, prof.dr. E. Hey-

Hawkins, and dr. C. Müller for being the referees of this thesis and being part

of the opposition.

Beste Arjan Termaten, mijn dank gaat ook naar jouw enthousiaste

dagelijkse begeleiding die ik tijdens mijn stage periode kreeg. Mede dankzij

jou groeide mijn interesse naar fosfinidenen en gelukkig mocht ik het

158

onderzoek verder oppakken. Tevens wil ik Sander van Assema bedanken

voor de gezelschap op het laboratorium en de buurman Bas de J. voor de

nodige glaswerk. Het was altijd zoeken naar die laatste schone NMR buisje!

Mijn studenten Abdul Karim (laat even weten waar je bent!), en Jos. Jullie

beide wil ik bedanken voor jullie bijdragen, beschreven in Hoofdstuk 4, naar

het oplosmiddel gestabiliseerde 16-elektron deeltje. Frans de Kanter,

bedankt voor je hulp met NMR metingen onder extreme condities en de 2D

verzoekjes. Marek Smoluch, thank you very much for the HRMS

measurements of the compounds, in a relatively short period of time. J.W.

Han Peeters, ook jij bedankt voor de FAB metingen wanneer EI een te

krachtig methode bleek. Martin Lutz, bedankt voor de kristalstructuur

analyses. Vaak bleven de kristallen op hun plek zitten. Ik hield altijd mijn hart

vast en gelukkig klopte de verwachtte structuur als een bus. Rob Schmitz,

bedankt voor al je hulp bij experimentele opstellingen en de vacuüm

pompen.

André en Erik, jullie beide wil ik bedanken voor jullie hulp met Spartan,

Gaussian en ADF tijdens de promotietijd. André, zonder jou zouden we op

lauwe Dommelsch moeten leven en Erik, jij had altijd wel een neus voor een

goed wijn. O ja, nog bedankt voor je hulp met de veldcodes! Also, I’d like to

thank all the international Ph.D. students and Post-docs for the great

company. Nuria Ortega, Federico, Federica, Masahiro, Sebastian Burck,

Hiroshi Naka, and Robert Wolf: thanks.

Verder wil ik emeritus prof.dr. F. Bickelhaupt, prof.dr. Romano Orru, Eelco,

Fedor, Mark B., Bas G., Niels T., Niels E., Helen, Robin, Lisette, Maurice, Rachel,

Amos, Shen, Maarten, Wannes, Anass, alle AIO’s, studenten, en OAC’ers die

ik bij naam ben vergeten bedanken voor de vele onvergetelijke momenten,

de conferentie(s) (16th ICPC Birmingham, NCCC en 4th PhD Workshop on

Phosphorus Chemistry), de borrels, de jaarlijks terugkerende recepties en de

gezellige diners. Judith bedankt voor de onvergetelijke, heerlijke Indische

gerechten, maar ook voor even bijpraten. Miep, jij ook bedankt voor het

regelen van allerlei ‘papier’ zaken, maar ook voor de goede praatjes.

Mijn collega’s en vrienden van Tate and Lyle wil ik hierbij ook bedanken

voor hun interesse in mijn promotieonderzoek. Mark, Nicoline, dr. Jere

159

Koskinen, Serge, Emiel, Sadi, (ha!) Tom (tom) en Coen. Furthermore, I’d like to

thank the people who supported me outside the VUA and T&L and who

shared the great fun during my Ph.D. period. Tommy & Celine, Ana, Micaela,

Nalan and Harry. Thank you for the wonderful moments, drinks, BBQ’s, trips

and parties. Ook de jongens band, dostlarım – waarvan sommige tot de

VUA meubilair behoren – wil ik bedanken voor de vele uitstapjes, heftige

discussies, de etentjes, en de feestjes. Atilla, Ahmet, Bülent, Kaan, Timur,

Volkan, Uğur, Ekrem en “hola dos” Oğuz. Mijn fysiotherapeut arkadaş R.

Meents moet ik ook bedanken. Zonder zijn therapie zou ik de lange

schrijfwerkdagen niet door kunnen komen.

My friends Engin, Şaban, Murat, and Sevda, I’d like to thank you being

great company during the holidays. Thank you for the true friendship, which

started as classmates back in Đzmir. Dostum Okan Akın wil ik speciaal

bedanken. Niet alleen maar voor zijn bijdrage in de omslag en de

binnenwerk, maar ook voor zijn oprechte vriendschap en broederschap.

Succes Mustafa met alles wat je onderneemt! Mijn schatjes Gülay,

Menekşe-Naz, Đpek, en mijn neefje “grote mannetje” Eray. Ik sta altijd voor

jullie klaar. Ook voor jou Emel! Erol, pas goed op haar! Mijn broer Medayim,

en mijn nichtje Emel wil ik bedanken omdat ze mijn paranimf wilden zijn.

Bedankt voor al je steun broer. Je was er altijd wanneer ik je nodig had en ik

heb veel van je geleerd. Mijn zussen Hacer, Yücel, Hazime, en schoonzus

Ebru wil ik bedanken voor alle geduld tijdens het onderzoek en het

schrijfperiode. Ook jullie steunden mij in alles wat ik deed.

Canım hacılarım anneciğim ve babacığım. Üzerimde ödeyemiyeceğim

kadar çok hakkınız var. Nacizhane bu kitabı aileme ve sevdiklerime hediye

ediyorum. Rabbime binlerce şükür ve hamdü senalar olsun.

Nu het boekje eindelijk echt af is, heb ik meer tijd voor mijn dierbaren.

Zaandam, oktober 2009

Halil AKTAŞ

Documents

Developing Terminal Phosphinidene Complexes