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
3.5 Computational Section 82
3.6 Experimental Section 83
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
4.6 Computational Section 106
4.7 Experimental Section 107
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
5.7 Computational Section 128
5.8 Experimental Section 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
Nucleophilic Phosphinidene Complexes: Access and Applicability
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.
Nucleophilic Phosphinidene Complexes: Access and Applicability
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
Nucleophilic Phosphinidene Complexes: Access and Applicability
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-
Nucleophilic Phosphinidene Complexes: Access and Applicability
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
Nucleophilic Phosphinidene Complexes: Access and Applicability
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.
Nucleophilic Phosphinidene Complexes: Access and Applicability
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 =
Nucleophilic Phosphinidene Complexes: Access and Applicability
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
Nucleophilic Phosphinidene Complexes: Access and Applicability
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),
Nucleophilic Phosphinidene Complexes: Access and Applicability
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]
Nucleophilic Phosphinidene Complexes: Access and Applicability
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]
Nucleophilic Phosphinidene Complexes: Access and Applicability
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.
Nucleophilic Phosphinidene Complexes: Access and Applicability
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]
Nucleophilic Phosphinidene Complexes: Access and Applicability
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
Nucleophilic Phosphinidene Complexes: Access and Applicability
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
Nucleophilic Phosphinidene Complexes: Access and Applicability
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.
Nucleophilic Phosphinidene Complexes: Access and Applicability
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,
Organometallics 2009, 28, 5166–5172.
[44] R. W. Alder, P. R. Allen, S. J. Williams, J. Chem. Soc., Chem. Comm. 1995, 1267–
1268.
Nucleophilic Phosphinidene Complexes: Access and Applicability
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,
Organometallics 2002, 21, 2101–2106.
[49] H. Aktas, J. Mulder, J. C. Slootweg, M. Schakel, A. W. Ehlers, M. Lutz, A. L. Spek, K.
Lammertsma, J. Am. Chem. Soc. 2009, 131, 13531–13537.
[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,
Organometallics 2007, 26, 5091–5101.
[62] 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,† J. Chris Slootweg,† Marius Schakel,† Andreas W. Ehlers,† Martin
Lutz,‡ Anthony L. Spek,‡ and Koop Lammertsma*,†
Department of Chemistry and Pharmaceutical Sciences, Vrije Universiteit Amsterdam, De Boelelaan
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)?
N-Heterocyclic Carbene Functionalized Ruthenium Phosphinidenes: What a Difference a Twist Makes
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),
N-Heterocyclic Carbene Functionalized Ruthenium Phosphinidenes: What a Difference a Twist Makes
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
N-Heterocyclic Carbene Functionalized Ruthenium Phosphinidenes: What a Difference a Twist Makes
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.
N-Heterocyclic Carbene Functionalized Ruthenium Phosphinidenes: What a Difference a Twist Makes
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-
N-Heterocyclic Carbene Functionalized Ruthenium Phosphinidenes: What a Difference a Twist Makes
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
N-Heterocyclic Carbene Functionalized Ruthenium Phosphinidenes: What a Difference a Twist Makes
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
N-Heterocyclic Carbene Functionalized Ruthenium Phosphinidenes: What a Difference a Twist Makes
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
References and Notes
[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.
[6] A. T. Termaten, T. Nijbacker, M. Schakel, M. Lutz, A. L. Spek, K. Lammertsma,
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
N-Heterocyclic Carbene Functionalized Ruthenium Phosphinidenes: What a Difference a Twist Makes
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.
[12] T. L. Breen, D. W. Stephan, J. Am. Chem. Soc. 1995, 117, 11914–11921.
[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.
[27] N. Kuhn, T. Kratz, Synthesis 1993, 561–562.
[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*,†
Department of Chemistry and Pharmaceutical Sciences, Vrije Universiteit Amsterdam, De Boelelaan
1083, 1081 HV Amsterdam, The Netherlands, and Bijvoet Center for Biomolecular Research, Crystal
Structural Chemistry, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands
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
N-Heterocyclic Carbene Functionalized Group 7–9 Transition Metal Phosphinidene Complexes
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
N-Heterocyclic Carbene Functionalized Group 7–9 Transition Metal Phosphinidene Complexes
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)).
N-Heterocyclic Carbene Functionalized Group 7–9 Transition Metal Phosphinidene Complexes
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,
N-Heterocyclic Carbene Functionalized Group 7–9 Transition Metal Phosphinidene Complexes
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.
3.5 Computational Section
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
N-Heterocyclic Carbene Functionalized Group 7–9 Transition Metal Phosphinidene Complexes
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.
3.6 Experimental Section
General. All experiments and manipulations were performed under an
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,
N-Heterocyclic Carbene Functionalized Group 7–9 Transition Metal Phosphinidene Complexes
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]:
N-Heterocyclic Carbene Functionalized Group 7–9 Transition Metal Phosphinidene Complexes
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),
N-Heterocyclic Carbene Functionalized Group 7–9 Transition Metal Phosphinidene Complexes
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.
References and Notes
[1] (a) J. C. Slootweg, K. Lammertsma, In Science of Synthesis; B. M. Trost, F. Mathey,
Chapter 3
90
Eds.; Georg Thieme Verlag: Stuttgart, 2009; Vol. 42, pp 15–36. (b) F. Mathey,
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.
[6] A. T. Termaten, T. Nijbacker, M. Schakel, M. Lutz, A. L. Spek, K. Lammertsma,
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.
[8] A. T. Termaten, T. Nijbacker, M. Schakel, M. Lutz, A. L. Spek, K. Lammertsma,
Organometallics 2002, 21, 3196–3202.
[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.
N-Heterocyclic Carbene Functionalized Group 7–9 Transition Metal Phosphinidene Complexes
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.
Chem. 2000, 112, 3025–3028; Angew. Chem. Int. Ed. 2000, 39, 2903–2906.
[13] R. H. Grubbs, Angew. Chem. 2006, 118, 3845–3850; Angew. Chem. Int. Ed. 2006,
45, 3760–3765.
[14] 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.
[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
g/cm3, µ = 0.46 mm–1. 28979 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. 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/.
[23] S. H. Vosko, L. Wilk, M. Nusair, Can. J. Phys. 1980, 58, 1200–1211.
[24] A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100.
[25] J. P. Perdew, Phys. Rev. B 1986, 33, 8822–8824.
[26] (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.
[27] E. van Lenthe, A. W. Ehlers, E. J. Baerends, J. Chem. Phys. 1999, 110, 8943–8953.
[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.
[30] N. Kuhn, T. Kratz, Synthesis 1993, 561–562.
[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.
N-Heterocyclic Carbene Functionalized Group 7–9 Transition Metal Phosphinidene Complexes
93
[33] Y. van den Winkel, H. M. M. Bastiaans, F. Bickelhaupt, J. Organomet. Chem. 1991,
405, 183–194.
[34] A. J. M. Duisenberg, L. M. J. Kroon-Batenburg, A. M. M. Schreurs, J. Appl. Cryst.
2003, 36, 220–229.
[35] G. M. Sheldrick, 1999, SADABS: Area-Detector Absorption Correction, v2.10,
Universität Göttingen, Germany.
[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*,†
Department of Chemistry and Pharmaceutical Sciences, Vrije Universiteit Amsterdam, De Boelelaan
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, 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
Iridium Phosphinidene Complexes: A Comparison with Iridium Imido Complexes in Their Reaction with Isocyanides
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
Iridium Phosphinidene Complexes: A Comparison with Iridium Imido Complexes in Their Reaction with Isocyanides
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
Iridium Phosphinidene Complexes: A Comparison with Iridium Imido Complexes in Their Reaction with Isocyanides
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
Iridium Phosphinidene Complexes: A Comparison with Iridium Imido Complexes in Their Reaction with Isocyanides
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.
4.6 Computational Section
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.
Iridium Phosphinidene Complexes: A Comparison with Iridium Imido Complexes in Their Reaction with Isocyanides
107
4.7 Experimental Section
General. All experiments and manipulations were performed under an
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]+.
Iridium Phosphinidene Complexes: A Comparison with Iridium Imido Complexes in Their Reaction with Isocyanides
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
Iridium Phosphinidene Complexes: A Comparison with Iridium Imido Complexes in Their Reaction with Isocyanides
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-
Iridium Phosphinidene Complexes: A Comparison with Iridium Imido Complexes in Their Reaction with Isocyanides
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.
References and Notes
[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.
[2] (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.
[3] D. W. Stephan, Angew. Chem. 2000, 112, 322–338; Angew. Chem. Int. Ed. 2000,
39, 314–329.
[4] T. L. Breen, D. W. Stephan, J. Am. Chem. Soc. 1995, 117, 11914–11921.
[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.
[6] A. T. Termaten, T. Nijbacker, M. Schakel, M. Lutz, A. L. Spek, K. Lammertsma,
Organometallics 2002, 21, 3196–3202.
[7] 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.
[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,
Iridium Phosphinidene Complexes: A Comparison with Iridium Imido Complexes in Their Reaction with Isocyanides
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,
Organometallics 1997, 16, 378–383.
[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.
[23] A. T. Termaten, T. Nijbacker, M. Schakel, M. Lutz, A. L. Spek, K. Lammertsma,
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/.
[27] S. H. Vosko, L. Wilk, M. Nusair, Can. J. Phys. 1980, 58, 1200–1211.
[28] A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100.
Chapter 4
116
[29] J. P. Perdew, Phys. Rev. B 1986, 33, 8822–8824.
[30] (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.
[31] E. van Lenthe, A. W. Ehlers, E. J. Baerends, J. Chem. Phys. 1999, 110, 8943–8953.
[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.
[35] Y. van den Winkel, H. M. M. Bastiaans, F. Bickelhaupt, J. Organomet. Chem. 1991,
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.
[38] A. J. M. Duisenberg, L. M. J. Kroon-Batenburg, A. M. M. Schreurs, J. Appl. Cryst.
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,
Universität Göttingen, Germany.
[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.
[42] G. M. Sheldrick, Acta Cryst. 2008, A64, 112–122.
[43] A. L. Spek, J. Appl. Cryst. 2003, 36, 7–13.
Iridium Phosphinidene Complexes: A Comparison with Iridium Imido Complexes in Their Reaction with Isocyanides
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*,†
Department of Chemistry and Pharmaceutical Sciences, Vrije Universiteit Amsterdam, De Boelelaan
1083, 1081 HV Amsterdam, The Netherlands, and Bijvoet Center for Biomolecular Research, Crystal
Structural Chemistry, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands
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.
η3-Diphosphavinylcarbene: A P2 Analogue of the Dötz Intermediate
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
η3-Diphosphavinylcarbene: A P2 Analogue of the Dötz Intermediate
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.
5.7 Computational Section
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.
5.8 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. 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
η3-Diphosphavinylcarbene: A P2 Analogue of the Dötz Intermediate
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;
η3-Diphosphavinylcarbene: A P2 Analogue of the Dötz Intermediate
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
η3-Diphosphavinylcarbene: A P2 Analogue of the Dötz Intermediate
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.).
η3-Diphosphavinylcarbene: A P2 Analogue of the Dötz Intermediate
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.
References and Notes
[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.
η3-Diphosphavinylcarbene: A P2 Analogue of the Dötz Intermediate
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.
[15] (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) 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
g/cm3, µ = 0.42 mm–1. 79778 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 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]
η3-Diphosphavinylcarbene: A P2 Analogue of the Dötz Intermediate
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.
[22] A. T. Termaten, T. Nijbacker, M. Schakel, M. Lutz, A. L. Spek, K. Lammertsma,
Organometallics 2002, 21, 3196–3202.
[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
[25] A. J. M. Duisenberg, L. M. J. Kroon-Batenburg, A. M. M. Schreurs, J. Appl. Cryst.
2003, 36, 220–229.
[26] G. M. Sheldrick, 1999. SADABS: Area-Detector Absorption Correction, v2.10,
Universität Göttingen, Germany.
[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.
[28] G. M. Sheldrick, Acta Cryst. 2008, A64, 112–122.
[29] A. L. Spek, J. Appl. Cryst. 2003, 36, 7–13.
[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.
complex ∆Ea’ ∆Ea” ∆E steric ∆E tot ∆E prep BDE
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.
complex ∆Ea’ ∆Ea” ∆E steric ∆E tot ∆E prep BDE
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.
complex ∆Ea’ ∆Ea” ∆E steric ∆E tot ∆E prep BDE
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