Synthesis of phosphorus(V)-stabilized geminal dianions. the cases of mixed P=X/PBH3 (X = S, O) and P=S/SiMe3 derivatives

Article abstract of DOI:10.1021/om300954a

The monodeprotonation of [CH₂(PPh₂BH ₃)(PPh₂=E)] (E = S (6), O (7)) afforded [CH(PPh ₂BH₃)(PPh₂=E)]^- (E = S (6 ^-), O (7^-)), whose structures were confirmed by X-ray crystallography. The kinetics of the second deprotonation appeared to be crucial in efficient synthesis of the corresponding dianions. Thus, the double deprotonation of 6 only led to 6^2-; the analogous reaction with 7 was slower and resulted only in the partial formation of 7^2-. Double deprotonation of the compound [CH₂(SiMe₃)(PPh ₂=S)] (8) also resulted in the partial formation of [C(SiMe ₃)(PPh₂=S)]^2- (8^2-), whose structure was confirmed by X-ray crystallography. The rare monomeric Mg carbene compound [MgC(PPh₂BH₃)(PPh₂=S)] (9) was obtained by the reaction of 6 with Mg(nBu)₂. The X-ray structure of 9 is presented.

Full text of DOI:10.1021/om300954a

Article
pubs.acs.org/Organometallics
Synthesis of Phosphorus(V)-Stabilized Geminal Dianions. The Cases
of Mixed PX/P→BH3 (X = S, O) and PS/SiMe3 Derivatives
Hadrien Heuclin,† Marie Fustier-Boutignon,† Samuel Ying-Fu Ho,†,‡ Xavier-Frederic Le Goff,†
́
́
Sophie Carenco,† Cheuk-Wai So,*,‡ and Nicolas Mezailles*,†,§
́
†Laboratoire “Het
́ ́ ́ ́
eroelements et Coordination”, Ecole Polytechnique, CNRS, Route de Saclay, 91128 Palaiseau Cedex, France
‡Division of Chemistry and Biological Chemistry, School of Physical and Mathematical Sciences, Nanyang Technological University,
637371 Singapore
S
* Supporting Information
ABSTRACT: The monodeprotonation of [CH2(PPh2→
BH3)(PPh2E)] (E = S (6), O (7)) afforded [CH(PPh2→
BH3)(PPh2E)]− (E = S (6‑), O (7−)), whose structures
were confirmed by X-ray crystallography. The kinetics of the
second deprotonation appeared to be crucial in efficient
synthesis of the corresponding dianions. Thus, the double
deprotonation of 6 only led to 62−; the analogous reaction
with 7 was slower and resulted only in the partial formation of
72−. Double deprotonation of the compound [CH2(SiMe3)-
(PPh2S)] (8) also resulted in the partial formation of
[C(SiMe3)(PPh2S)]2‑ (82−), whose structure was confirmed by X-ray crystallography. The rare monomeric Mg carbene
compound [MgC(PPh2→BH3)(PPh2S)] (9) was obtained by the reaction of 6 with Mg(nBu)2. The X-ray structure of 9 is
presented.
involving Nd carbene complexes.10 In 2006 and 2008, the
INTRODUCTION
■
Henderson group published the synthesis of geminal dianions
The groundbreaking syntheses of electrophilic carbene
incorporating other alkali metals,11,12 and in 2009, Harder and
complexes by Fischer in 19641 and of nucleophilic carbene
co-workers reported the synthesis of a bis(cesium) derivative of
complexes by Schrock roughly 10 years after2 have opened the
1d.13 This strategy is currently limited because of (i) the very
way for a considerable number of studies.3 The use of carbene
small number of known geminal dianions (A−E; Scheme 2)
complexes in organic synthesis, in stoichiometric as well as in
catalytic processes, was then developed extensively. In
particular, among the many processes involving carbene
complexes as catalysts, the alkene metathesis reaction has
and (ii) a lack of an efficient access to them (12− to 42−) as,
needless to say, these dianionic species are extremely water
sensitive and cannot be readily purified. In this respect, it is
noteworthy that double deprotonation of the ligand in the
coordination sphere of the metal has been used in some
seen a tremendous development over the past decades, leading
to applications in various fields ranging from polymer science to
total synthesis.4 The almost infinite variations of both the
instances as an alternative strategy to yield the same carbene
complexes.14
substitution scheme of the carbene fragment “CR1R2” and the
The underlying reason for the paucity of geminal dianions
lies in the required efficient stabilization of two charges on the
metal fragment allow for a very fine tuning of the properties of
the carbene complex, ranging from nucleophilic to electrophilic.
In the past decade, a novel strategy relying on the use of
geminal dianions to bring formally the four electrons of the
MC bond has been devised (Scheme 1).
same carbon atom. We have shown using DFT calculations in
the thiophosphinoyl case, 22−, that these two lone pairs interact
in donor−acceptor type interactions with empty antibonding
orbitals of appropriate energy and symmetry. Upon coordina-
This chemistry has been mainly developed with three
tion of the dianions with metal fragments, an electron transfer
geminal dilithiated compounds for which the carbon atom is
from the carbon center to the metal center occurs to lead to a
substituted symmetrically by an iminophosphorane
formal MC interaction. The extent of this electron transfer
(PPh2NSiMe3), a thiophosphinoyl (PPh2S), or a phosphonate
moiety (P(OiPr)2O) (1a2−, 22−, and 32−; Scheme 2).5,6 These
depends on two factors: (i) the energy match between the
orbitals of the dianion and those of the metal fragment and (ii)
species have allowed the synthesis of a large variety of carbene
complexes of transition metals, rare earths, and uranium.7 In
the orbital overlap. The electron transfer to the metal center is
obviously in competition with the electron transfer from the
2006, Le Floch et al. developed a general method allowing the
introduction of other substituents at the nitrogen atom of the
8,9
2−
iminophosphorane moiety (1b‑e
)
and subsequently proved
Received: October 16, 2012
Published: January 11, 2013
the influence of the nitrogen substituent in a catalytic process
© 2013 American Chemical Society
498
dx.doi.org/10.1021/om300954a | Organometallics 2013, 32, 498−508

Organometallics
Article
Scheme 1. Strategies for the Synthesis of Carbene Complexes
very complex matter based not only on thermodynamic
stabilization but also on kinetic factors.
Scheme 2. Geminal Dianions Reported in the Literature
RESULTS AND DISCUSSION
■
We have shown previously that two Ph2PS substituents at C
allow a very efficient stabilization of the dianion 22−. In a first
approach, it was thus interesting to keep at least one of them in
conjunction with an a priori less stabilizing moiety such as
PPh2→BH3 (target 62−). The monoanion derived from the bis-
borane adduct of dppm (bis(diphenylphosphino)methane) has
been successfully synthesized and used in coordination.17 The
corresponding dianion with group 1 metals has not been
reported yet, although Harder and co-workers have obtained
the dianionic calcium species.18 In parallel, the related
phosphine oxide was targeted (compound 72−), allowing a
direct comparison of the stabilizing power of PS and PO.
Moreover, the only PO moieties reported to stabilize
dianions are phosphonate derivatives (Scheme 2).
Compound 6 was synthesized in a one-pot fashion by simple
borylation of ((diphenylphosphino)methyl)diphenylphosphine
sulfide. The latter was prepared according to a procedure
reported by Mitchell and Grim (Scheme 4, eq 1).19 Formation
central C to the P substituents which is necessary for the
stability of the dianion. A fine tuning of the MC interaction,
and therefore of the reactivity of the complexes, is to be
expected from a modification of the substituents at the carbon
center. Our goals here are manifold. First, we investigate
syntheses for novel types of dianions bearing two different
substituents at C (Scheme 3). Several examples of such
Scheme 4. Synthesis of Compounds 6 and 7
Scheme 3. Targets of This Study
“unsymmetrical geminal dianions” are known (Scheme 2; A−E,
42−, and 52−) but only 42−, reported in 2009 by So and co-
of the intermediate compound was verified by 31P NMR (AB
system at δP 40.1 ppm and δP −28 ppm with JP−P = 75 Hz)
before reaction with the borane−dimethyl sulfide. Compound 6
was obtained in decent yield (51%) as a white crystalline solid
after purification. Another strategy was used for compound 7.
Monobromation of dppm by 1 equiv of bromine in
2
workers, and 52−, reported by Gessner and Schroter very
̈
recently have been obtained efficiently.15,16 On the way to the
desired dianions, the monoanionic species were synthesized
and characterized. The results of the present study clearly
demonstrate that the efficient synthesis of geminal dianions is a
499
dx.doi.org/10.1021/om300954a | Organometallics 2013, 32, 498−508

Organometallics
Article
dichloromethane20 followed by hydrolysis of the reaction
yielded the mono-oxidized species, which was reacted with
BH3·SMe2 to yield compound 6 after workup in 50% yield
(Scheme 4, eq 2).
Scheme 5. Synthesis of the Anions 6− and 7−
In the 31P NMR spectrum both compounds are characterized
by an AX system at δP 34.4 ppm (d, 2JP−P = 9 Hz) and δP 14.6
2
ppm (bd) for compound 6 and at δP 23.2 ppm (d, JP−P = 15
Hz) and δP 13.9 ppm (br) for compound 7, the broadening of
the last doublet being characteristic of phosphine−borane
1
compounds. In the H NMR spectrum in CD2Cl2, the central
CH2 protons are both seen as a doublet of doublets at δH 3.86
ppm (2JP−H = 13 Hz, 2JP−H = 10 Hz) for 6 and at δH 3.37 ppm
(2JP−H = 10.5 Hz, JP−H = 12 Hz) for 7, the corresponding
2
at δP 40.7 ppm (d, JP−P = 41 Hz) and δP 8.7 ppm (br) for 6−
and at δP 34.6 ppm (br) and δP 5.9 ppm (br) for 7− . Both
anions could be isolated as pale yellow solids after evaporation
of the solvent. They were further characterized by multinuclear
NMR spectroscopy. In both compounds the central CH proton
is shifted upfield at δH 1.15 and 1.44 ppm (for 6− and 7−,
respectively, in deuterated THF) in comparison to the neutral
compound (δH 3.86 and 3.37 ppm for 6 and 7). Similarly, the
bridging carbons were found at δC 13.6 and 14.0 ppm (in
comparison to δC 29.6 ppm in 6 and δC 30.4 ppm in 7). X-ray
diffraction analysis gave definitive proof of the successful
deprotonation of the neutral compounds. Single crystals of 6−
and 7− were grown by slow diffusion of pentane into a
concentrated solution in THF and diethyl ether, respectively,
and views of the structures of 6− and 7− are given in Figures 2
and 3, respectively.
2
1
carbon atoms being found at δC 29.6 ppm (dd, JP−C = 26 Hz,
1
1JP−C = 47 Hz) for 6 and at δC 30.4 ppm (dd, JP−C = 62 Hz,
1JP−C = 26 Hz) for 7. Table 1 summarizes the main NMR data
Table 1. Significant NMR Data for All Compounds
δP(P(X)) δP(P(B)) δH(PC(H)P) δC(PCP)
solvent
6 (X = S)
34.4
40.7
25.5
23.2
34.6
42
14.6
8.7
4.8
13.9
5.9
5
3.86
1.15
29.6
13.6
CD2Cl2
THF-d8
ether-d10
CD2Cl2
THF-d8
Tol/Et2O
CDCl3
6−
62−
7 (X = O)
3.37
1.44
30.4
14.0
7−
72−
8 (X = S)
8−
82−
37.7
44.5
2.00
0.50
21.1
11.7
C6D6
Both compounds crystallize as dimers. The lithium atoms are
not directly coordinated to the central carbons but are
stabilized by coordination with two sulfur or oxygen atoms,
respectively, one BH3 moiety, and one solvent molecule (THF
or diethyl ether). Compound 6− features an elongation of the
P−S bond (2.0323(8) Å vs 1.963(2) Å in 6) and a shortening
of the two P−C bonds (1.699(3) and 1.722(3) Å vs 1.839(2)
and 1.830(2) Å in 6). Surprisingly, the P−B bond is not
significantly affected by the deprotonation. The same pattern is
seen in 7−, with an elongation of the P−O bond length
(1.523(1) Å vs 1.489(3) Å in 7) and a subsequent shortening
of the two P−C bonds (1.703(2) and 1.725(2) Å vs 1.819(3)
and 1.829(3) Å in 7).
for 6 and 7. X-ray crystal structures were obtained for both
compounds. Single crystals suitable for X-ray diffraction analysis
were obtained by slow evaporation of concentrated solutions of
6 and 7 in dichloromethane. Views of 6 and 7 are given in
Figure 1. Bond lengths and angles are completely standard in
the two compounds (note that the structure of 6 features a
statistical repartition of boron and sulfur atoms on the two
positions).
Single deprotonation of 6 and 7 was easily achieved by
reaction with respectively 2 equiv of butyllithium or
methyllithium at −78 °C in THF (Scheme 5).
Evidence of the complete conversion of the neutral
compounds into the corresponding anions 6− and 7− was
seen by 31P NMR, both anions exhibiting two sets of doublets
The synthesis and characterization of monoanion 8− have
already been reported by Gessner,21 and we have reported its
Figure 1. ORTEP plots (50% thermal ellipsoids) of the X-ray crystal structures of compounds 6 (left) and 7 (right). Selected bond lengths (Å) and
angles (deg) for 6: C(1)−P(1) = 1.839(2), C(1)−P(2) = 1.830(2), P(1)−S(1) = 1.963(2), P(1)−B(2) = 1.93(2), P(2)−S(2) = 1.963(2), P(2)−
B(1) = 1.95(2), P(1)−C(2) = 1.821(2), P(1)−C(8) = 1.815(2), P(2)−C(14) = 1.817(2), P(2)−C(20) = 1.804(2); P(1)−C(1)−P(2) = 117.8(1).
Selected bond lengths (Å) and angles (deg) for 7: C(1)−P(1) = 1.819(3), C(1)−P(2) = 1.829(3), P(1)−O(1) = 1.489(2), P(2)−B(1) = 1.920(4),
P(1)−C(2) = 1.810(3), P(1)−C(8) = 1.805(3), P(2)−C(14) = 1.822(3), P(2)−C(20) = 1.812(3); P(1)−C(1)−P(2) = 114.6(2).
500
dx.doi.org/10.1021/om300954a | Organometallics 2013, 32, 498−508

Organometallics
Article
Scheme 6. Synthesis of Dianions 62− and 72−
purified. This solid was soluble in 1,2-dimethoxyethane and
diethyl ether and could be characterized by multinuclear NMR
spectroscopy. The 31P NMR confirmed the total conversion of
6 into a single new product characterized by a doublet at δP
25.5 ppm (2JP−P = 8 Hz) and a broad doublet at 4.8 ppm. In the
1H NMR spectrum, only aromatic protons are observed and no
central proton could be located. The 13C NMR spectrum
suffered poor resolution, and no significant resonance could be
located for the central carbon. This new compound was
therefore identified as the dianion 62−.
Figure 2. ORTEP plot (50% thermal ellipsoids) of the X-ray crystal
structure of compound 6−. Hydrogen atoms on the phenyl rings and
on the solvent molecules are omitted for clarity. Selected bond lengths
(Å) and angles (deg): C(1)−P(1) = 1.699(3), C(1)−P(2) = 1.722(3),
P(1)−S(1) = 2.0323(8), P(2)−B(1) = 1.933(3), S(1)−Li(1) =
2.478(5), S(1)−Li(2) = 2.466(5), P(1)−C(14) = 1.827(3), P(1)−
C(20) = 1.831(3), P(2)−C(2) = 1.830(3), P(2)−C(8) = 1.835(3),
P(3)−C(27) = 1.833(3), P(3)−C(33) = 1.824(2), P(4)−C(39) =
1.835(2), P(4)−C(45) = 1.832(2); P(1)−C(1)−P(2) = 126.4(2).
No resonance corresponding to a cosolvent was seen in the
NMR spectrum of 62−, which was confirmed by elemental
analysis. We can therefore presume that 62− forms polymeric
structures in toluene and its precipitation drives the equilibrium
toward its complete formation. To confirm that double
deprotonation of 6 did occur, a trapping experiment with
D2O was carried out. After addition of an excess of D2O to a
suspension of 62−, stirring for 1 h, and purification the bis-
1
deuterated derivative of 6 was isolated (6-D2). Indeed, the H
NMR spectrum lacks the signal of the central CH2 protons and
the bridging carbon signal is found as a highly coupled
multiplet at δC 29.6 ppm in the 13C NMR spectrum. Single
crystals of 62− grown by diffusion of pentane into a
concentrated solution in DME allowed the determination of
a structure for 62−. A view of 62− is given in Figure 4.
Interestingly, compound 62− crystallizes as a monomer in
DME. The two lithium atoms are coordinated to the central
carbon, their coordination spheres being completed by one
solvent molecule. The hydrogen atoms on the BH3 group have
been geometrically fixed; thus, no discussion is possible on the
exact nature of the Li−BH3 interaction. In 62− the P(B)−C
bonds are even shorter than in 6 or 6− (1.701(3) Å in 62− vs
1.722(3) Å in 6−). In contrast, the P−B bond once again
remains unchanged. A summary of the evolution of some bond
lengths in 6, 6−, and 62− is given in Table 2.
Figure 3. ORTEP plot (50% thermal ellipsoids) of the X-ray crystal
structure of compound 7−. Hydrogen atoms on the phenyl rings and
carbon atoms of the solvent molecules (diethyl ether, O2, and O2′)
are omitted for clarity. Selected bond lengths (Å) and angles (deg):
C(1)−P(1) = 1.703(2), C(1)−P(2) = 1.725(2), P(1)−O(1) =
1.523(1), P(2)−B(1) = 1.937(2), O(1)−Li(1) = 1.900(3), O(1)−
Li(1′) = 1.940(3), B(1′)−Li(1) = 2.651(3), P(1)−C(2) = 1.821(2),
P(1)−C(8) = 1.811(2), P(2)−C(14) = 1.831(2), P(2)−C(20) =
1.831(2); P(1)−C(1)−P(2) = 126.5(1).
With these encouraging results, similar reactions were
attempted with both 7 and 8. When 2 equiv of methyllithium
or butyllithium was added to a solution of 7 in toluene at −78
°C, followed by warming to room temperature, a yellow
solution containing exclusively the monoanionic species 7− was
obtained within 5 min, as proved by 31P and 1H NMR
spectroscopy (Scheme 6). After 12 h of stirring, a mixture of
only two products was observed by 31P NMR, the monoanion
7− (doublet at δP 44.1 ppm and broad signal centered at δP 4.5
ppm in toluene) and a another species with very close chemical
shifts: δP 42.0 ppm and a broad signal at ca. δP 5 ppm
overlapping with the signal of the anion 7−. More interesting
crystal structure.22 Its NMR shifts are included in Table 1 for
comparison.
Encouraged by these results, and on the basis of the excellent
stability of the three monoanionic species (over several days in
solution under strict air- and water-free conditions), we then
attempted to synthesize the corresponding dianions. It was
proven that dianionic species of the type 1−4 (Scheme 2) are
unstable in many solvents (including THF) and therefore
double deprotonation of 6 was attempted in toluene. Addition
of 2 equiv of butyllithium to a suspension of 6 in toluene
resulted in a yellow reaction mixture and the solubilization of
the starting material (Scheme 6). 31P NMR spectroscopy
confirmed the rapid formation of 6−. After 2 h, a yellow solid
precipitated concomitant with the loss of all signals in 31P
NMR. After 3 h, this solid was extracted by filtration and
1
information was gathered from the H NMR spectrum, which
presented only one signal for a P−CH−P moiety. Moreover, it
showed two kinds of aromatic signals: a set of sharp, well-
resolved multiplets corresponding to the aromatic protons of
501
dx.doi.org/10.1021/om300954a | Organometallics 2013, 32, 498−508

Organometallics
Article
P).13,24 The same procedure was then carried out on
compound 8. Addition of 2 equiv of methyllithium in a
mixture of diethyl ether and hexane at −78 °C resulted in the
instant formation of 8− upon slow warming to room
temperature, as shown by 31P NMR spectroscopy.21,22 After
several days of stirring at room temperature, several products
were seen in 31P NMR spectroscopy, together with 8−. Few
yellow crystals deposited spontaneously from the reaction
media. X-ray diffraction analysis confirmed the existence of
dianion 82−, as shown in Figures 5 and 6. Here also, one
Figure 4. ORTEP plot (50% thermal ellipsoids) of the X-ray crystal
structure of compound 62−. Hydrogen atoms on the phenyl rings and
carbon atoms of the solvent molecules are omitted for clarity. Selected
bond lengths (Å) and angles (deg): C(1)−P(1) = 1.699(3), C(1)−
P(2) = 1.701(3), P(1)−S(1) = 2.033(1), P(2)−B(1) = 1.943(3),
C(1)−Li(1) = 2.225(6), C(1)−Li(2) = 2.162(6), P(1)−C(14) =
1.833(3), P(1)−C(20) = 1.834(3), P(2)−C(2) = 1.840(3), P(2)−
C(8) = 1.835(3); P(1)−C(1)−P(2) = 126.4 (2).
Figure 5. ORTEP plot (50% thermal ellipsoids) of the X-ray crystal
structure of compound 82−. Phenyl rings (except ipso carbon atoms),
methyl groups of the TMS moieties, hydrogen atoms, and solvent
molecules are omitted for clarity. Selected bond lengths (Å): C(1)−
P(1) = 1.633(8), S(1)−P(1) = 2.093(2), Si(1)−C(1) = 1.778(8),
P(1)−C(8) = 1.827(8), P(1)−C(2) = 1.831(6), C(1)−Li(2′) =
2.09(2), C(1)−Li(1′) = 2.18(1), S(1)−Li(3) = 2.454(8), S(1)−Li(1)
= 2.57(1), S(1)−Li(2) = 2.45(1), S(1)−Li(1′) = 2.55(1), O(1) =
Li(1′′) 1.90(1), O(1)−Li(1) = 1.90(1), O(1)−Li(1′) = 1.90(1),
Si(2)−O(1) = 1.601(6).
7− in toluene and a set of broad multiplets that belonged to the
new species detected in 31P NMR. Because of the lack of a CH
proton for this second species, we proposed it to be the
expected dianion 72−. The same trapping experiment with D2O
was carried out, proving our hypothesis right. Indeed, the
neutral species 7HD and 7D2 were formed, as shown by the
integration of the signal corresponding to the P−CH(D)−P
moiety in comparison to the integration of the aromatic
protons in the 1H NMR spectrum (see the Supporting
Information). Thus, after 12 h, ca. 70% of the desired dianion
72− had formed (7HD/7D2 = ca. 3/7) from the monoanion 7−.
Unfortunately, the double-deprotonation reaction did not result
in the expected full transformation of the anion into the
dianion. Rather, several species started to appear after 1 day and
after 1 week crystals had deposited from the crude mixture. X-
ray analysis of these crystals revealed the formation of the
SiMe3OLi byproduct that crystallized with the monoanion (see
the Supporting Information).23 This result, although dis-
appointing, proved that the kinetics of the deprotonation is
crucial for the formation of the desired dianions. If the second
deprotonation at the “central” carbon center is too slow, then
side reactions such as deprotonation at the ortho position of a
Ph group followed by nucleophilic attack on the opposite
phosphorus substituent occur, preventing the clean formation
of the desired dianion. These competitive reactions depend on
many factors (nature of the cation, solvent, and substituents at
Figure 6. View of one molecular fragment of 82−.
anionic μ3-OSiMe3 fragment which lies on the C3 axis is present
in the X-ray structure. We propose that this fragment originates
from the slow reaction of the strong base with the silicon
grease.23 Each carbon atom of the parent ligand is bound to two
lithium atoms, and each sulfur atom coordinates four lithium
atoms. These lithium atoms are of two types. The inner ones
are coordinated by two sulfur atoms, one carbon atom, and the
Table 2. Summary of Significant Bond Lengths (Å) in Compounds 6−8 and Related Anions
bond
6
6−
62−
7
7−
8
8−
82−
PX−C
PB−C
P−X
P−B
Si−C
1.839(2)
1.830(2)
1.963(2)
1.94(2)*
1.699(3)
1.722(3)
2.0323(8)
1.933(3)
1.699(3)
1.701(3)
2.033(1)
1.943(3)
1.819(3)
1.829(3)
1.489(2)
1.920(4)
1.703(2)
1.725(2)
1.523(1)
1.937(2)
1.803(2)
1.701(1)
1.633(8)
1.9628(14)
1.896(2)
2.0263(4)
1.827(1)
2.093(2)
1.778(8)
502
dx.doi.org/10.1021/om300954a | Organometallics 2013, 32, 498−508

Organometallics
Article
μ3-oxygen atom from the OSiMe3 fragment. The others are
tricoordinated and bound to one sulfur atom, one carbon atom,
and one solvent molecule (not shown in Figure 5). Compound
82− features a highly elongated P−S bond (2.093(2) Å in 82− vs
1.9628(14) Å in 8). This elongation is of much greater
amplitude than for compound 6. As seen for compounds 6 and
7, the P−C bond length shortens with the deprotonation of
compound 8, from 1.803(2) to 1.633(8) Å. This trend is also
observed for the Si−C bond. Further characterization of 82−
was hampered by the amount of crystals gathered from this
reaction. Nevertheless, these two experiments clearly show that,
although the dianions 72− and 82− are thermodynamically
stable, their efficient synthesis is highly challenging and
appropriate conditions have not been found so far.
In order to gain insights into the electronic structure of these
new species, a theoretical study has been carried out. We
focused our attention on the description of dianion 62−, for
which three states have been considered: neutral, monoanionic,
and dianionic (VI, VI·Li, and VI·Li2, respectively). Optimiza-
tions have been carried out on model compounds in which the
solvent molecules (THF or DME) have been replaced by
dimethyl ether (see the Supporting Information).
Table 5. NBO Analysis of the Hyperconjugation in VI·Li and
VI·Li2
donor
LP1
acceptor
E(2)
ΔEij
Fij
VI·Li
σ*(P−B)
σ*(P−S)
10.7
17.7
0.29
0.35
0.053
0.073
σ*(PS-CPh)
σ*(PB-CPh)
σ*(PB-CPh)
σ*(PB-CPh)
11.8
9.9
0.37
0.37
0.062
0.058
total
50.1
1.9
VI·Li2
LP1
σ*(P−B)
0.48
0.029
σ*(P−S)
σ*(PS−CPh)
σ*(PB−CPh)
σ*(PB−CPh)
σ*(PB−CPh)
σ*(PS−CPh)
15.8
10.1
4.9
0.43
0.44
0.44
0.075
0.061
0.043
4.2
0.43
0.039
total
36.9
LP2
σ*(P−B)
σ*(P−S)
19.9
18.5
13.6
0.28
0.36
0.35
0.071
0.077
0.066
σ*(PB−CPh)
σ*(PB−CPh)
σ*(PB−CPh)
σ*(PS−CPh)
σ*(PS−CPh)
total
As seen in Table 3, the evolution of the different bond
lengths and angles upon deprotonation is very well reproduced.
7.7
5.1
0.34
0.34
0.048
0.039
Table 3. Evolution of Characteristic Bond Lengths and
Angles in VI and Its Derivatives
64.8
101.7
total LP1 + LP2
bond
VI
VI·Li
VI·Li2
Δ
VI−VI·Li2(DFT)
Δ
VI−VI·Li2(RX)
PS−C
PB−C
P−S
1.86
1.85
1.96
1.93
1.72
1.74
2.03
1.94
1.71
1.73
2.06
1.95
0.15
0.12
0.10
0.02
0.14
0.11
0.06
Table 4). Similarly, the charge at the sulfur atom increases in
absolute value upon deprotonation steps as it ranges from
−0.55 in VI to −0.80 in VI·Li2, but the magnitude is much
smaller than for C. This indicates, however, an electronic
transfer from the central carbon to the sulfur atom. Both
phosphorus atoms bear a positive charge in all species, and the
charges only vary to a small extent. One notes that the charges
of both P atoms increase from VI to VI·Li but decrease
between VI·Li and VI·Li2. Surprisingly, the charge at the boron
atom does not vary much (−0.70 in average), which is coherent
with a weak electron transfer from the central carbon to this
heteroatom. As seen from the different X-ray diffraction
analyses of the structures of the different derivatives of
compound 6, the deprotonation process causes geometrical
P−B
In particular, the unexpected fact that the P−B bond length is
hardly affected is also observed in the calculations. A NBO
analysis was then performed on the three model compounds
(see Tables 4 and 5). In the dianionic species, the C−Li
interaction is mostly ionic (q(Li) = 0.87 and 0.86 and n(C−Li)
< 0.1) and the central carbon bears a greater negative charge
than in both the anionic and the neutral derivatives with a total
difference of −0.6 e between the neutral and dianionic forms
(−1.76 in VI·Li2 compared to −1.41 in VI·Li and −1.16 in VI;
Table 4. Summary of NBO Data for VI, VI·Li, and VI·Li2
Wiberg bond index
n(C−PS)
n(C−PB)
n(P−S)
n(P−B)
n(PS−CPh)
n(PB−CPh)
n(C−Li)
VI
0.81
1.08
1.21
0.84
1.08
1.22
1.44
1.18
1.12
0.99
0.97
0.96
0.83, 0.84
0.84, 0.80
0.80, 0.77
0.90, 0.88
0.86, 0.85
0.80, 0.80
VI·Li
VI·Li2
0.052, 0.033
NBO charge
qC
qLi
0.86
qP(S)
qP(B)
qS
qB
VI
−1.16
−1.41
−1.76
1.37
1.45
1.41
1.40
1.42
1.35
−0.55
−0.76
−0.80
−0.70
−0.76
−0.70
VI·Li
VI·Li2
0.87, 0.86
lone pair
LP1
LP2
hybridization
occupation
hybridization
p1.00
occupation
1.63
VI·Li
VI·Li2
sp23.06
sp2.75
1.72
1.73
503
dx.doi.org/10.1021/om300954a | Organometallics 2013, 32, 498−508

Organometallics
Article
Figure 7. Map of the electronic density for a value of 0.052 e Å−3 for the two highest occupied molecular orbitals of VI·Li2.
Figure 8. Electron density map for the donor (plain) and acceptor (shaded) NBO orbitals describing the hyperconjugation interaction resulting in
the stabilization of lone pairs in VI·Li2.
changes in the ligand. The P−C bonds between the central
carbon and the phosphorus atoms are shortened, and the P−S
bond length is increased, but the P−B bond length remains
surprisingly unchanged. The Wiberg bond indexes calculated
for the different models are in agreement with these
observations. Upon deprotonation, the bond index for the
two P−C bonds increases (0.81 to 1.08 and then to 1.21, and
0.84 to 1.08 and then to 1.22 on going from VI to VI·Li2),
which indicates a strengthening of these bonds. In contrast, the
Wiberg bond index calculated for the P−S bond decreases
(1.44 to 1.12), which is in accordance with the elongation of
this bond upon successive deprotonations. On the other hand,
the bond order for the P−B bond remains almost constant
(0.99 to 0.96).
In order to explain these evolutions or lack thereof, we
focused our attention on the stabilization of the negative
charges at the central carbon atom. An NBO analysis was
carried out on both VI·Li and VI·Li2. It shows that the carbon
bears a lone pair in the anion, as expected from deprotonation,
and that it is an almost pure p orbital (sp23.1). It also confirms
that two orthogonal lone pairs are found in VI·Li2 (denoted
LP1 and LP2), one of sp2.75 character (LP1) and the other
being a pure p orbital (LP2). These two orbitals have an
important participation in the two highest orbitals of dianion
VI·Li2 (LP2 is seen in the HOMO and LP1 in the HOMO-1;
see Figure 7).
The two occupied lone pairs are stabilized by electronic
donation into the neighboring empty antibonding orbitals of
appropriate symmetry and energies, namely the σ*(P−S),
σ*(P−B), and σ*(P−CPh) orbitals. Table 5 summarizes these
results. LP1 in VI·Li is stabilized by negative hyperconjugation
into the neighboring σ*(P−CPh) orbitals and σ*(P−S) orbitals.
The difference in energy (ΔEij) is rather small, because LP1 is
almost p pure, and the overall stabilization is thus quite good,
with a total of ca. 50 kcal/mol.
In VI·Li2, LP1 is stabilized mainly by negative hyper-
conjugation into the neighboring σ*(P−CPh) orbitals (Figure
8A). Interestingly, the σ*(P−B) orbital does not participate
much in the stabilization of LP1 (E(2) = 1.9 kcal/mol), though
it has the right symmetry to do so (note that because of
504
dx.doi.org/10.1021/om300954a | Organometallics 2013, 32, 498−508

Organometallics
Article
symmetry considerations, σ*(P−S) cannot participate in the
stabilization of LP1). Because of an increased s character in this
LP1 in comparison to the analogous lone pair in VI·Li, the
energy difference is higher between donor and acceptor
orbitals, and the total energy associated with these transfers is
ca. 37 kcal/mol, smaller than in VI·Li. On the other hand, the
stabilization of LP2 is much stronger (total stabilization energy
of ca. 65 kcal/mol). LP2 is highly stabilized by donation into
the σ*(P−S) bond (E(2) = 19.9 kcal/mol, Figure 8B) and by
donation into the σ*(P−C) orbitals. Here also, the σ*(P−B)
bond is not involved in a major interaction (for symmetry
reasons). When all the contributions are added, the total
stabilization energy of LP2 is 101.7 kcal/mol. LP2 is a better
electron donor than LP1, mainly because of its higher energy
that narrows the gap between LP2 and the accepting orbitals
σ*(P−C) (ΔE ≈ 0.35 for LP2 and ΔE ≈ 0.44 for LP1). These
results help us understand the geometrical changes in 6 upon
double deprotonation. The stabilization of the lone pairs
involves populating the antibonding σ*(P−S) bond, which
weakens the bond and results in its lengthening. The same
applies for the different P−Cphenyl bonds. As a result of these
donor−acceptor interactions the two P−C bonds of the P−C−
P bridge are shortened. As these interactions are much weaker,
almost negligible, with the σ*(P−B), the P−B bond length
remains nearly unchanged.
Finally, the synthesis of a group 2 derivative with the most
rewarding reagent, 6, was attempted. In 2010, Leung et al. had
reported the synthesis of the magnesium carbene
[(PPh2S)2CMg]2 by the reaction of the neutral derivative of
compound 22− with dibutylmagnesium.25 The X-ray crystal
structure shows that the magnesium carbene is dimerized in a
head to head manner. We also reported the synthesis of the
magnesium carbenes [(PPh2NSiMe3)2CMg]2 and
[(PPh2NSiMe3)(PPh2S)CMg]2 derived from 1a2− and 42−,
respectively.26 Unlike [(PPh2S)2CMg]2, they are dimerized in a
head to tail manner in the solid state. The fact that the
structures of the Mg carbenes are affected by the compositions
of the geminal dianions prompted our interest in investigating
the reaction of compound 6 with dibutylmagnesium. The
reaction was performed in a Et2O/THF (1/1) mixture under
reflux for 12 h, and compound 9 was isolated (Scheme 7).
The X-ray crystal structure of 9 shows that it is monomeric
(Figure 9). The magnesium atom coordinates to the C(1) and
Figure 9. ORTEP plot (50% thermal ellipsoids) of the X-ray crystal
structure of compound 9. Mg(1)−C(1) = 2.113(4), Mg(1)−O(1) =
2.053(3), Mg(1)−O(2) = 2.038(3), Mg(1)−O(3) = 2.103(3), C(1)−
P(1) = 1.703(4), C(1)−P(2) = 1.666(4), P(1)−B(1) = 1.938(4),
P(2)−S(1) = 2.0308(13), Mg(1)−S(1) = 2.6907(16); P(1)−C(1)−
P(2) = 132.6(2), P(2)−C(1)−Mg(1) = 99.44(17), C(1)−Mg(1)−
S(1) = 76.17(11), Mg(1)−S(1)−P(2) = 74.24(5), S(1)−P(2)−C(1)
= 107.98(13).
S(1) atoms of the ligand. The coordination sphere of the
magnesium atom is further supplemented by coordination with
three THF molecules. The PPh2→BH3 substituent is
uncoordinated. The result is in contrast to that for
[(PPh2BH3)2CCa]2, in which the BH3 substituents coordinate
to the Ca2+ atom in a bidentate fashion.17 The C(1)−Mg(1)
bond (2.113(4) Å) is comparable with that in [(PPh2S)2CMg]2
(2.156(5) Å).25
CONCLUSION
■
In conclusion, we have presented here three novel examples of
rare geminal dianions. Although the three species are
thermodynamically stable, the kinetics of the double deproto-
nation appears to be crucial for their efficient formation.
Eventually, only the mixed dianion 62−, where the two lone
pairs are stabilized by a PPh2S and a PPh2→BH3 moieties, has
been obtained selectively, opening ways for its use as a carbene
precursor. Interestingly, the stabilization by the PPh2→BH3
moiety appears to be weaker, which in turn should lead to
increased donation of the lone pairs to a metal center. The
synthesis of the rare Mg derivative 9 was also achieved with the
same precursor, 6. DFT calculations are currently being
performed in order to quantify and compare the stabilizing
properties of the different C substituents of the geminal
dianions. In parallel, work in our laboratories using the novel
compounds 62− and 9 for the synthesis of carbene complexes is
being pursued.
Scheme 7. Synthesis of Complex 9
Compound 9 was isolated as a highly air and moisture
sensitive yellow crystalline solid which is soluble in hydro-
carbon solvents. It has been characterized by NMR spectros-
1
copy and X-ray crystallography. The H NMR spectrum of 9
displays resonances for the BH3, THF, and phenyl protons. It is
noteworthy that the 13C{1H} NMR signal for the carbenic
carbon in 9 is not observed. Similarly, there is no 13C{1H}
NMR resonance for the carbenic carbon in [(PPh2NSiMe3)-
(PPh2S)CMg]2 and [(PPh2NSiMe3)2CMg]2.25 The 31P{1H}
NMR spectrum of 9 displays a doublet at δP 37.9 ppm (2JP−P
56.3 Hz) for the PS moiety and a broad signal at δP 3.41 ppm
for the PB moiety. The 11B NMR spectrum shows one broad
signal at δB −30.2 ppm.
EXPERIMENTAL SECTION
General Procedures. All reactions were routinely performed
under an inert atmosphere of argon or nitrogen by using Schlenk and
glovebox techniques and dry deoxygenated solvents. Dry tetrahy-
■
505
dx.doi.org/10.1021/om300954a | Organometallics 2013, 32, 498−508

Organometallics
Article
drofuran, pentane, and 1,2-dimethoxyethane were obtained by
distillation from Na/benzophenone. Dry dichloromethane was
distilled on P2O5 and dry toluene on Na. Nuclear magnetic resonance
spectra were recorded on a Bruker AC-300 SY spectrometer operating
temperature and stirred for 1 h. Solvents were evaporated under
vacuum, and the remaining pale yellow solid was washed with 2 × 5
mL of pentane and dried (200 mg, 95%). 31P NMR (121.5 MHz, thf-
1
d8): δP 40.4 (d, JP−P = 41 Hz, PS), 8.7 (bd, PBH3). H NMR (300
1
at 300.0 MHz for H, 75.5 MHz for 13C, and 121.5 MHz for 31P.
MHz, thf-d8): δH 7.96−7.89 (m, 4H, Harom), 7.81−7.74 (m, 4H,
Harom), 7.21−7.13 (m, 12H, Harom), 1.15 (dd, JP−H = 4 Hz, JP−H = 6
Hz, 1H, PC(H)P), 1.3−0.23 (bs, BH3). 13C NMR (75.5 MHz, thf-d8):
δC 143.9 (dd, JP−C = 4 Hz, JP−C = 85 Hz, Cipso), 141.5 (dd, JP−C = 4 Hz,
1
Solvent peaks are used as internal reference relative to Me4Si for H
and 13C chemical shifts (ppm); 31P chemical shifts are relative to a
85% H3PO4 external reference. Coupling constants are given in hertz.
The following abbreviations are used: s, singlet; bs, broad singlet; br,
broad signal; d, doublet; bd, broad doublet; dd, doublet of doublets; t,
triplet; m, multiplet. Triphenylthiophosphine was obtained by reaction
of elemental sulfur with triphenylphosphine in THF. All other reagents
and chemicals were obtained commercially and used as received.
CCDC 893679−893685 contain crystallographic data for this paper.
These data can be obtained free of charge from the Cambridge
Crystallographic Data Center via www.ccdc.cam.a-c.uk/data_request/
cif.
JP−C = 59 Hz, Cipso), 132.6 (d, JP−C = 10 Hz, CHarom), 132.1 (d, JP−C
=
10 Hz, CHarom), 128.9 (d, JP−C = 3 Hz, CHpara), 128.5 (d, JP−C = 3 Hz,
CHmeta), 127.6 (d, JP−C = 10 Hz, CHarom), 127.3 (d, JP−C = 11 Hz,
CHarom), 13.6 (dd, JP−C = 78 Hz, JP−C = 112 Hz, PC(H)P). 11B NMR
(89.0 MHz, CD2Cl2): δB −37.3 (bs). Anal. Calcd for C29H32P2BLiOS:
C, 68.52; H, 6.35. Found: C, 68.35; H, 6.65.
Synthesis of Compound 7−. To a solution of 7 (66.2 mg, 0.16
mmol) in THF (8 mL) was added methyllithium (1.6 M in Et2O, 0.1
mL, 0.16 mmol) at −78 °C dropwise. The resulting pale yellow
solution was warmed to room temperature and stirred for 1 h. Drying
under vacuum allowed the isolation of 6− as a yellow solid (69 mg,
88%). 31P NMR (121.5 MHz, thf-d8): δP 34.6 (br, PO), 5.9 (br, PB).
1H NMR (300 MHz, thf-d8): δH 7.80−7.72 (m, 8H, Harom), 7.26−7.17
(m, 12H, Harom), 3.62 (br, CH2 of thf), 1.76 (br, CH2 of thf), 1.44 (dd,
JP−H = 7.0 Hz, JP−H = 11.6 Hz, PCHP), BH3. 13C NMR (75.5 MHz,
thf-d8): δC 142.5 (dd, JP−C = 6 Hz, JP−C = 108.3 Hz, Cipso), 141.6 (dd,
JP−C = 5.3 Hz, JP−C = 58.0 Hz, Cipso), 131.3 (d, JP−C = 9.4 Hz, CHmeta),
131.2 (d, JP−C = 9.4 Hz, CHmeta), 128.5 (d, JP−C = 2.6 Hz, CHpara),
127.7 (d, JP−C = 2.3 Hz, CHpara), 127.0 (d, JP−C = 10.9 Hz, CHortho),
126.9 (d, JP−C = 12.8 Hz, CHortho), 67.2 (s, C2 of thf), 25.4 (s, C1 of
thf), 14.0 (dd, JP−C = 76.6 Hz, JP−C = 142.7 Hz, PCHP). Anal. Calcd
for C29H34P2O2BLi: C, 70.47; H, 6.93. Found: C, 70.18; H, 7.14.
Synthesis of Compound 62−. To a suspension of 6 (688.8 mg,
1.6 mmol) in toluene (10 mL) was added butyllithium (1.6 M in
hexanes, 2 mL, 3.2 mmol) at −78 °C. The solution was warmed to
room temperature and stirred for 3 h. The reaction mixture
progressively turned yellow, and a yellow solid precipitated. The
solid was isolated by centrifugation and washed once with toluene (6
mL) then with 2 × 6 mL of pentane and dried under vacuum. The
desired compound was obtained as a pale yellow solid (700 mg, 99%).
31P NMR (300 MHz, ether-d10): δp 25.5 (d, JP−P = 8 Hz, PS), 4.8 (bd,
PB). 1H NMR (300.0 MHz, ether-d10): δH 7.94−7.85 (m, 1H, Harom),
7.76−7.62 (m, 4H, Harom), 7.49−7.44 (m, 4H, Harom), 7.27−6.95 (m,
11H, Harom), 1.40−0.34 (m, BH3). 11B NMR (89.0 MHz, ether-d10):
δB −29.6 (bs). Anal. Calcd for C25H23P2SBLi2: C, 67.85; H, 5.24.
Found: C, 67.84; H, 5.35.
Synthesis of Compound 6. To a solution of triphenylthiophos-
phine (3.76 g, 12.77 mmol) in THF (40 mL) was added methyllithium
(1.6 M in Et2O, 8 mL, 12.8 mmol) at −78 °C. The solution was
warmed to room temperature and stirred overnight. Upon reaction its
changed from colorless to dark red. This solution was then added
dropwise to a solution of chlorodiphenylphosphine (2.30 mL, 12.8
mmol) in THF (15 mL) at 0 °C. The reaction mixture was warmed to
·
room temperature and stirred for 12 h, and BH3 SMe2 (1.2 mL, 12.8
mmol) was added. The reaction mixture was stirred for 2 h. Volatiles
were evaporated under vacuum, and dichloromethane (30 mL) was
added. Insoluble impurities were discarded by filtration, and the
remaining solution was concentrated. Pentane (40 mL) was added,
which allowed the precipitation of a white solid that was isolated by
filtration and washed twice with pentane (10 mL). The title compound
was obtained as a white solid after drying under vacuum (2.8 g, 51%).
31P NMR (121.5 MHz, CD2Cl2): δP 34.1 (d, JP−P = 8.9 Hz, PS), 14.8
1
(bd, PBH3). H NMR (300 MHz, CD2Cl2) δH 8.04−7.93 (m, 8H,
Harom), 7.76−7.56 (m, 12H, Harom), 3.86 (dd, JP−H = 10.2 Hz, JP−H
=
13.1 Hz, 2H, PCH2P), BH3. 13C NMR (75.5 MHz, CD2Cl2): δC
133.08 (d, JP−C = 10 Hz, CHortho), 132.9 (dd, JP−C = 2 Hz, JP−C = 83
Hz, Cipso), 131.6 (d, JP−C = 3 Hz, CHpara), 131.4 (d, JP−C = 3 Hz,
CHpara), 131.0 (d, JP−C = 11 Hz, CHortho), 128.6 (d, JP−C = 10 Hz,
CHmeta), 128.4 (d JP−C = 10 Hz, CHmeta), 128.1 (dd, JP−C = 3 Hz, JP−C
= 57 Hz, Cipso), 29.6 (dd, JP−C = 26 Hz, JP−C = 48 Hz, PCH2P). 11B
NMR (89.0 MHz, CD2Cl2) δB −38.0 (bs). MS (HRMS EI): m/z
430.1244 (calcd for C25H25BP2S 430.1245). Anal. Calcd for
C25H25P2SB: C, 69.75; H, 5.86. Found: C, 69.84; H, 5.95.
Synthesis of Compound 7. To a solution of dppm (960.9 mg, 2.5
mmol) in 60 mL of dichloromethane was added bromine (129 μL, 2.5
mmol) dropwise at −78 °C, under a nitrogen atmosphere. The
solution turned yellow and changed to colorless upon warming to
room temperature. DABCO (281.8 mg, 2.5 mmol) was then added
under a nitrogen flux followed by degassed water (45 μL, 2.5 mmol).
After 1 h of stirring, the precipitate was removed by filtration. The
solution was dried under high vacuum, and 60 mL of THF was added
under an inert atmosphere. BH3·SMe2 (480 μL, 5 mmol) was added at
−78 °C, and the solution was warmed to room temperature. Volatiles
were removed under vacuum, and the title compound was obtained as
a white solid after column chromatography in 50% yield. 31P NMR
(121.5 MHz, CD2Cl2): δP 23.2 (d, JP−P = 15.5 Hz, PO), 13.9 (b,
PB).1H NMR (300 MHz, CD2Cl2) δH 7.87 (m, 6H, Harom), 7.61 (m,
6H, Harom), 7.39 (m, 8H, Harom), 3.37 (dd, JP−H = 10.5 Hz, JP−H = 12.0
Hz, 2H, PCH2P), 1.56−0.34 (br, 3H, BH3). 13C NMR (75.5 MHz,
CD2Cl2): δC 135.5 (dd, JP−C = 103.5 Hz, JP−C = 1.5 Hz, Cipso), 135.3
(d, JP−C = 10.2 Hz, CHortho), 134.0 (d, JP−C = 2.5 Hz, CHpara), 133.6
(d, JP−C = 2.5 Hz, CHpara), 132.7 (d, JP−C = 9.5 Hz, CHortho), 130.8 (d,
JP−C = 7.7 Hz, CHmeta), 130.7 (d, JP−C = 6.3 Hz, CHmeta), 130.3 (dd,
Synthesis of Compound 6-D2. To a suspension of 6 (68.8 mg,
0.16 mmol) in toluene (4 mL) was added butyllithium (1.6 M in
hexanes, 0.2 mL, 0.32 mmol) at −78 °C. The solution was warmed to
room temperature and stirred for 3 h. Excess D2O was added, and the
reaction mixture was then stirred for 2 h. Solvents were evaporated
under vacuum. Dichloromethane (5 mL) was added, and insoluble
impurities were discarded by filtration. The remaining solution was
dried over MgSO4 and filtered. Evaporation of the solvent afforded the
title compound as a white solid (65 mg, 94%). 31P NMR (121.5 MHz,
1
CD2Cl2): δP 34.1 (d, JP−P = 8.9 Hz, PS), 14.8 (bd, PBH3). H NMR
(300 MHz, CD2Cl2): δH 7.81−7.69 (m, 8H, Harom), 7.62−7.31 (m,
12H, Harom), 1.49−0.25 (PBH3). 13C NMR (75.5 MHz, CD2Cl2): δC
133.4 (d, JP−C = 10 Hz, CHarom) 133.2 (dd, JP−C = 2 Hz, JP−C = 84 Hz,
Cipso), 131.9 (d, JP−C = 3 Hz, CHpara), 131.7 (d, JP−C = 3 Hz, CHpara),
131.3 (d, JP−C = 10 Hz, CHarom), 128.8 (d, JP−C = 10 Hz, CHarom),
128.7 (d, JP−C = 12 Hz, CHarom), 128.5 (dd, JP−C = 3 Hz, JP−C = 53 Hz,
Cipso), 29.6 (m, PCD2P).
Deuterolysis of Compound 72−. Two equivalents of methyl-
lithium in diethyl ether (1.6 M, 0.2 mL, 0.32 mmol) was added
dropwise to a solution of compound 7 (66 mg, 0.16 mmol) in toluene
(4 mL) at −78 °C with vigorous stirring. The solution was brought
slowly to room temperature and the pressure frequently equilibrated.
After 24 h, excess deuterated water was added. The solution was
stirred for 6 h before the solvents were removed under vacuum.
Organic products were extracted in dichloromethane and dried over
JP−C = 56.8 Hz, JP−C = 2.7 Hz, CHipso), 30.4 (dd, JP−C = 26 Hz, JP−C
=
62 Hz, PCH2P). Anal. Calcd for C25H25P2OB: C, 72.49; H, 6.08.
Found: C, 72.28; H, 6.14.
Synthesis of Compound 6−. To a solution of 6 (206.5 mg, 0.48
mmol) in THF (5 mL) was added methyllithium (1.6 M in Et2O, 0.3
mL, 0.48 mmol) at −78 °C. The yellow solution was warmed to room
506
dx.doi.org/10.1021/om300954a | Organometallics 2013, 32, 498−508

Organometallics
Article
MgSO4. Compound 7, as a mixture of the H and D isotopes, was
isolated as a white powder (64 mg, 95%).
Stendel, J., Jr. Chem. Rev. 2009, 109, 3227. (g) Herndon, J. W. Coord.
Chem. Rev. 2012, 256, 1281.
Many variations in the experimental conditions were attempted
without success to obtain quantitative formation of 72−, among which
were using other solvents (DME or Et2O) and bases (n-BuLi) and
grease-free glassware. Using TMEDA to increase the kinetics of the
deprotonation was also unsuccessful.
(4) (a) Furstner, A. Angew. Chem., Int. Ed. 2000, 39, 3012. (b) Trnka,
̈
T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18. (c) Nicolau, K. C.;
Bulger, P. G.; Sarlah, D. Angew. Chem., Int. Ed. 2005, 44, 4490.
(d) Vougioukalakis, G. C.; Grubbs, R. H. Chem. Rev. 2010, 110, 1746.
(5) (a) Ong, C. M.; Stephan, D. W. J. Am. Chem. Soc. 1999, 121,
2939. (b) Kasani, A.; Babu, R. P. K.; McDonald, R.; Cavell, R. G.
Angew. Chem., Int. Ed. 1999, 38, 1483.
Synthesis of Compound 9. Mg(nBu)2 (1 M in heptane, 2.2 mL,
2.20 mmol) was added dropwise to 6 (0.86 g, 2.00 mmol) in THF/
Et2O (1/1, 20 mL) at room temperature. The reaction mixture was
then refluxed for 12 h, while the solution changed from colorless to
pale yellow. The reaction mixture was filtered, and the filtrate was
concentrated to afford 9 as yellow crystals (0.60 g, 44.6%). Mp: 233
(6) (a) Cantat, T.; Mez
Angew. Chem., Int. Ed. 2004, 43, 6382. (b) Cantat, T.; Ricard, L.; Le
Floch, P.; Mezailles, N. Organometallics 2006, 25, 4965.
́
ailles, N.; Ricard, L.; Jean, Y.; Le Floch, P.
́
(7) (a) Cavell, R. G.; Babu, R. P. K.; Kasani, A.; McDonald, R. J. Am.
Chem. Soc. 1999, 121, 5805. (b) Kasani, A.; McDonald, R.; Cavell, R.
G. Chem. Commun. 1999, 1993. (c) Cantat, T.; Jaroschik, F.; Nief, F.;
2
°C dec. 31P NMR (160.3 MHz, C6D6): δP 37.9 (d, JP−P = 56.27 Hz,
PS), 3.41 (bs, PBH3). 1H NMR (395.9 MHz, C6D6): δH 7.82 (m, 4H,
Ph), 7.65 (m, 4H, Ph), 6.89 (m, 12H, Ph), 3.63 (m, 12H, THF), 1.85
Ricard, L.; Mez
(d) Cantat, T.; Demange, M.; Mez
Floch, P. Organometallics 2005, 24, 4838. (e) Cantat, T.; Jaroschik, F.;
Ricard, L.; Le Floch, P.; Nief, F.; Mezailles, N. Organometallics 2006,
25, 1329. (f) Cantat, T.; Ricard, L.; Mezailles, N.; Le Floch, P.
́
ailles, N.; Le Floch, P. Chem. Commun. 2005, 5178.
(bs, 3H, BH3), 1.38 (m, 12H, THF). 13C NMR (99.5 MHz, C6D6): δC
́
ailles, N.; Ricard, L.; Jean, Y.; Le
3
145.9 (d, JP−C = 66.2 Hz, Cipso), 144.3 (dd, JP−C = 42.6 Hz, JP−C
6.64 Hz, Cipso), 133.4 (d, JP−C = 8.67 Hz, CHarom), 132.5 (d, JP−C
10.9 Hz, CHarom), 128.0 (d, 4JP−C = 2.70 Hz, CHpara), 127.5 (d, 4JP−C
1.97 Hz, CHpara), 127.1 (d, JP−C = 8.67 Hz, CHarom), 126.9 (d, JP−C
=
=
=
=
́
́
Organometallics 2006, 25, 6030. (g) Orzechowski, L.; Jansen, G.;
Harder, S. J. Am. Chem. Soc. 2006, 128, 14676. (h) Cantat, T.; Arliguie,
10.9 Hz, CHarom), 68.2 (s, THF), 26.4 (s, THF). 11B NMR (127.0
MHz, C6D6): δB −30.2 (bs). Anal. Calcd for C37H47BMgO3P2S: C,
66.44; H, 7.08. Found: C, 66.12; H, 6.87.
́ ́
T.; Noel, A.; Thuery, P.; Ephritikhine, M.; Le Floch, P.; Mezailles, N. J.
̈
Am. Chem. Soc. 2009, 131, 963. (i) Foo, C.; Lau, K.-C.; Yang, Y.-F.; So,
C.-W. Chem. Commun. 2009, 6816. (j) Cooper, O. J.; McMaster, J.;
Lewis, W.; Blake, A. J.; Liddle, S. T. Dalton Trans. 2010, 39, 5074.
ASSOCIATED CONTENT
* Supporting Information
■
́
(k) Heuclin, H.; Grunstein, D.; Le Goff, X.-F.; Le Floch, P.; Mezailles,
̈
S
N. Dalton Trans. 2010, 39, 492. (l) Ma, G.; Ferguson, M. J.; Cavell, R.
G. Chem. Commun. 2010, 46, 5370. (m) Tourneux, J.-C.; Berthet, J.-
Figures, tables, and CIF files giving crystallographic data, NMR
spectra, and details of the theoretical study. This material is
available free of charge via the Internet at http://pubs.acs.org.
C.; Thuer
Trans. 2010, 39, 2494. (n) Fustier, M.; Le Goff, X. F.; Le Floch, P.;
Mezailles, N. J. Am. Chem. Soc. 2010, 132, 13108. (o) Ma, G.;
́ ́
y, P.; Mezailles, N.; Le Floch, P.; Ephritikhine, M. Dalton
́
AUTHOR INFORMATION
Corresponding Author
Ferguson, M. J.; McDonald, R.; Cavell, R. G. Inorg. Chem. 2011, 50,
6500. (p) Cooper, O. J.; Mills, D. P.; McMaster, J.; Moro, F.; Davies,
E. S.; Lewis, W.; Blake, A. J.; Liddle, S. T. Angew. Chem., Int. Ed. 2011,
50, 2383. (q) Mills, D. P.; Moro, F.; McMaster, J.; van Slageren, J.;
Lewis, W.; Blake, A. J.; Liddle, S. T. Nat. Chem. 2011, 3, 454.
■
*E-mail: mezailles@chimie.ups-tlse.fr (N.M.); cwso@ntu.edu.
sg (C.-W.S.).
Present Address
(r) Heuclin, H.; Cantat, T.; Le Goff, X. F.; Le Floch, P.; Mez
Eur. J. Inorg. Chem. 2011, 2540. (s) Tourneux, J.-C.; Berthet, J.-C.;
Cantat, T.; Thuery, P.; Mezailles, N.; Le Floch, P.; Ephritikhine, M.
Organometallics 2011, 30, 2957. (t) Tourneux, J.-C.; Berthet, J.-C.;
Cantat, T.; Thuery, P.; Mezailles, N.; Ephritikhine, M. J. Am. Chem.
Soc. 2011, 133, 6162. (u) Fustier-Boutignon, M.; Heuclin, H.; Le Goff,
X. F.; Mezailles, N. Chem. Commun. 2012, 48, 3306. (v) Mills, D. P.;
́
ailles, N.
§Laboratoire “Het
́ ́ ́
ochimie Fondamentale et Appliquee”,
er
́
Paul Sabatier, CNRS, 118 Route de Narbonne,
Universite
31062 Toulouse Cedex, France.
́
́
́
́
Notes
The authors declare no competing financial interest.
́
Cooper, O. J.; Tuna, F.; McInnes, E. J. L.; Davies, E. S.; McMaster, J.;
Moro, F.; Lewis, W.; Blake, A. J.; Liddle, S. T. J. Am. Chem. Soc. 2012,
134, 10047. For reviews see: (w) Heuclin, H.; Fustier, M.; Auffrant,
ACKNOWLEDGMENTS
■
The CNRS, the Ecole Polytechnique, and the MOE (AcRF
Tier 1, RG 57/11) are thanked for supporting this work. H.H.
and M.F.-B. thank the Ecole Polytechnique for fellowships.
S.Y.-F.H. gratefully acknowledges Nanyang Technological
University and the Ecole Polytechnique for a joint Ph.D.
scholarship. N.M. and C.-W.S. thank the Merlion Program
(Project ID: 4.05.10) for logistical support.
A.; Mez
Wooles, A. J. Chem. Soc. Rev. 2011, 40, 2164.
(8) Demange, M.; Boubekeur, L.; Auffrant, A.; Mez
L.; Le Goff, X.; Le Floch, P. New J. Chem. 2006, 30, 1745.
(9) Cooper, O. J.; Wooles, A. J.; McMaster, J.; Lewis, W.; Blake, A. J.;
Liddle, S. T. Angew. Chem., Int. Ed. 2010, 49, 5570.
́
ailles, N. Lett. Org. Chem. 2010, 7, 596. (x) Mills, D. P.;
́
ailles, N.; Ricard,
(10) Buchard, A.; Platel, R. H.; Auffrant, A.; Le Goff, X. F.; Le Floch,
P.; Williams, C. K. Organometallics 2010, 29, 2892.
(11) Hull, K. L.; Noll, B. C.; Henderson, K. W. Organometallics 2006,
25, 4072.
(12) Hull, K. L.; Carmichael, I.; Noll, B. C.; Henderson, K. W. Chem.
Eur. J. 2008, 14, 3939.
DEDICATION
■
We dedicate this article to the memory of the late Prof. P. Le
Floch, an inspiring chemist and a missed friend.
(13) Orzechowski, L.; Jansen, G.; Harder, S. Angew. Chem., Int. Ed.
2009, 48, 3825.
REFERENCES
■
(14) (a) Babu, R. P. K.; McDonald, R.; Decker, S. A.; Klobukowski,
M.; Cavell, R. G. Organometallics 1999, 18, 4226. (b) Kasani, A.;
Ferguson, M.; Cavell, R. G. J. Am. Chem. Soc. 2000, 122, 726. (c) Babu,
R. P. K.; McDonald, R.; Cavell, R. G. Chem. Commun. 2000, 481.
(d) Aparna, K.; Babu, R. P. K.; McDonald, R.; Cavell, R. G. Angew.
Chem., Int. Ed. 2001, 40, 4400. (e) Leung, W.-P.; Wang, Z.-X.; Li, H.-
W.; Mak, T. C.W. Angew. Chem., Int. Ed. 2001, 40, 2501.
(f) Orzechowski, L.; Harder, S. Organometallics 2007, 26, 5501.
(1) (a) Fischer, E. O.; Maasbol, O. Angew. Chem., Int. Ed. 1964, 3,
580. (b) Fischer, E. O.; Maasbol, O. Chem. Ber. 1967, 100, 2445.
(2) Schrock, R. R. J. Am. Chem. Soc. 1974, 96, 6796.
(3) (a) Bruce, M. I. Chem. Rev. 1991, 91, 197. (b) Doyle, M. P.;
Forbes, D. C. Chem. Rev. 1998, 98, 911. (c) Enders, D.; Niemeier, O.;
Henseler, A. Chem. Rev. 2007, 107, 5606. (d) Hahn, F. E.; Jahnke, M.
C. Angew. Chem., Int. Ed. 2008, 47, 3122. (e) de Frem
N.; Nolan, S. P. Coord. Chem. Rev. 2009, 253, 862. (f) Dotz, K. H.;
̈
̈
́
ont, P.; Marion,
̈
507
dx.doi.org/10.1021/om300954a | Organometallics 2013, 32, 498−508

Organometallics
Article
(g) Orzechowski, S.; Harder, S. Organometallics 2007, 26, 2114.
(h) Liddle, S. T.; McMaster, J.; Green, J. C.; Arnold, P. L. Chem.
Commun. 2008, 1747. (i) Mills, D. P.; Cooper, O. J.; McMaster, J.;
Lewis, W.; Liddle, S. T. Dalton Trans. 2009, 4547. (j) Mills, D. P.;
Wooles, A. J.; McMaster, J.; Lewis, W.; Blake, A. J.; Liddle, S. T.
Organometallics 2009, 28, 6771. (k) Leung, W.-P.; Wan, C.-L.; Kan, K.-
W.; Mak, T. C. W. Organometallics 2010, 29, 814. (l) Wooles, A. J.;
Cooper, O. J.; McMaster, J.; Lewis, W.; Blake, A. J.; Liddle, S. T.
Organometallics 2010, 29, 2315. (m) Wooles, A. J.; Mills, D. P.; Lewis,
W.; Blake, A. J.; Liddle, S. T. Dalton Trans. 2010, 39, 500. (n) Leung,
W.-P.; Mak, Y.-C.; Mak, T. C. W. Inorg. Chem. 2011, 50, 10517.
(15) Schroter, P.; Gessner, V. H. Chem. Eur. J. 2012, 18, 11223.
̈
(16) Chen, J.-H.; Guo, J.; Li, Y.; So, C.-W. Organometallics 2008, 28,
4617.
(17) (a) Blug, M.; Grunstein, D.; Alcaraz, G.; Sabo-Etienne, S.; Le
̈
Goff, X.-F.; Le Floch, P.; Mez
́
ailles, N. Chem. Commun. 2009, 4432.
(b) Westerhausen, M.; Gorls, H.; Wimmer, K.; Langer, J. Dalton Trans.
̈
2009, 2951.
(18) Orzechowski, L.; Jansen, G.; Lutz, M.; Harder, S. Dalton Trans.
2009, 2958.
(19) Grim, S. O.; Mitchell, J. D. Synth. React. Inorg. Met.-Org. Chem.
1977, 4, 221.
(20) Boubekeur, L.; Ricard, L.; Mez
Organometallics 2005, 24, 1065.
(21) Gessner, V. H. Organometallics 2011, 30, 4228.
(22) Heuclin, H.; Carenco, S.; Le Goff, X.-F.; Mezailles, N. Eur. J.
́
ailles, N.; Le Floch, P.
́
Inorg. Chem. 2012, 1453.
(23) Haiduc, I. Organometallics 2004, 23, 3.
(24) Yang, Y.-F.; Foo, C.; Ganguly, R.; Li, Y.; So, C.-W.
Organometallics 2012, 31, 6538.
(25) Leung, W.-P.; Wan, C.-L.; Mak, T. C. W. Organometallics 2010,
29, 1622.
(26) Guo, J.-Y.; Lee, J.-S.; Foo, M.-C.; Lau, K.-C.; Xi, H.-W.; Lim, K.
H.; So, C.-W. Organometallics 2010, 29, 939.
508
dx.doi.org/10.1021/om300954a | Organometallics 2013, 32, 498−508