Lithiation studies of [PhP(CH2Ph)3]Cl; X-ray crystal structure of the phosphoniodiylide [(TMEDA)Li(PhCH)2PPh(CH2Ph)] and its rhodium and chromium complexes

Article abstract of DOI:10.1016/S0022-328X(01)00674-X

Treatment of the phosphonium salt [PhP(CH₂Ph)₃]Cl with three molar equivalents of ⁿBuLi gives the dilithiated species [Li₂PhP(CHPh)₃]. The crystal structure of the monolithiated species as its TMEDA a

Full text of DOI:10.1016/S0022-328X(01)00674-X

Journal of Organometallic Chemistry 625 (2001) 236–244
www.elsevier.nl/locate/jorganchem
Lithiation studies of [PhP(CH2Ph)3]Cl; X-ray crystal structure of
the phosphoniodiylide [(TMEDA)Li(PhCH)2PPh(CH2Ph)] and its
rhodium and chromium complexes
Philip J. Bailey *, Tony Barrett, Simon Parsons
Department of Chemistry, Uni6ersity of Edinburgh, The King’s Buildings, West Mains Road, Edinburgh EH9 3JJ, UK
Received 7 November 2000; received in revised form 3 January 2001; accepted 3 January 2001
Abstract
Treatment of the phosphonium salt [PhP(CH2Ph)3]Cl with three molar equivalents of nBuLi gives the dilithiated species
[Li2PhP(CHPh)3]. The crystal structure of the monolithiated species as its TMEDA adduct has been determined. Treatment of the
dilithiate species with [((h-C5Me5)MX2]2 (M=Cr, X=Br; M=Rh, X=Cl) provides [(h-C5Me5)M{h2-PhP(CH2Ph)(CHPh)2}X]
containing chelating phosphoniodiylide ligands which are thought to be the hydrolysis products of the target complexes
[{PhP(CHPh)3}M(C5Me5)]. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: Lithiation; X-ray crystal structures; Phosphoniodiylide; Pentamethylcyclopentadienyl
applications of TMM as a dianionic replacement for
cyclopentadienyl in a-alkene polymerisation applica-
tions has recently been discussed and investigated [3].
We realised some time ago that complexes containing
ligands of the TMM type with a single substituent on
each arm of the ligand framework could adopt chiral
conformations with C3 symmetry [5], and thus belong
to a small class of ligands which occupy three fac
coordination sites, and provide three equivalent chiral
(homotopic) environments for substrate coordination
opposite the ligand, in an octahedral complex (Fig. 1)
[6]. The potential for application of such complexes in
enantioselective catalytic processes, which involve octa-
hedral intermediates is clear and has been the subject of
a recent review article [7]. Such ligands are complemen-
tary to the well-established chiral C2 symmetric ligands,
most commonly diphosphines, which serve a similar
purpose in square planar geometry [8].
1. Introduction
A
large number of complexes containing the
trimethylenemethane (TMM) ligand [C(CH2)3] have
now been characterised [1–3]. Until quite recently they
were almost exclusively restricted to those of the middle
and late transition metals, research in this area being
fuelled partly by the application of Pd-TMM complex
intermediates in [3+2] cycloaddition reactions [4].
However, an increasing number of high oxidation state
early transition metal examples are now being reported
[2,3], and the ligands appear well suited for coordina-
tion to these more electropositive centres. Potential
Although trimethylenemethane complexes are readily
available by
a number of routes, the synthetic
difficulties involved in the preparation of complexes of
trisubstituted derivatives [C(CHR)3] mean that this
class of ligand is unlikely to be widely applicable and
we have therefore sought alternative ligand systems
with the same symmetry properties. Our initial ap-
proach was to exploit the generally more facile synthe-
Fig. 1. C3 symmetric arrangement of groups around trisubstituted
trimethylenemethane and related ligands.
* Corresponding author.
E-mail address: philip.bailey@ed.ac.uk (P.J. Bailey).
0022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(01)00674-X

P.J. Bailey et al. / Journal of Organometallic Chemistry 625 (2001) 236–244
237
ses of nitrogen based systems in an investigation of the
nitrogen analogues of TMM, [C(NR)3]2− [5]. However,
although these proved to be interesting chelating and
bridging ligands [9,10], the distortion from planarity
required for the h3-coordination of such Y-shaped lig-
ands is thought to be precluded by the energy associ-
ated with the in-plane p-bonding in such systems. This
energy is increased in comparison to that for the
[C(CH2)3]2− system by the replacement of the methyl-
ene carbon atoms by the more electronegative nitrogen.
In order to address this problem we have begun an
investigation of ligands in which the required distortion
is built-in to the ligand before coordination, and to this
end we are employing ligands centred on tetrahedral
rather than trigonal planar atoms. We have previously
reported initial studies on the potentially tridentate
nitrogen donor system [RP(NR)3]2− which was found
to be a very powerful p-donating, chelating ligand, but
were unable to demonstrate the coordination of the
third nitrogen [11]. Here we report of investigations of
the carbon-donor system [RP(CHR)3]2− which is inca-
pable of p-donation in this fashion. Mono- and dilithia-
tion of the phosphonium salt [PhP(CH2Ph)3]Cl and the
X-ray crystal structure of the monolithiate
[(TMEDA)Li(PhCH)2PPh(CH2Ph)] are reported, along
with those of its complexes with Cp*Rh(III) and
Cp*Cr(III) moieties.
phosphorous, resulting in a large degree of delocalisa-
tion within the PꢀC bonds. Repeating this experiment
using three molar equivalents of nbutyllithium gives
spectra that indicate the formation of the symmetrical
dilithiated species [Li2PhP(CHPh)3] (2) and which ex-
hibit a 31P-13CHPh coupling constant of 105 Hz.
Crystallisation of the dilithiate species 2 with the
addition of various quantities of TMEDA or HMPA to
the THF solution to complex the lithium ions was
attempted, however, spectroscopic data suggests that
the crystalline material isolated in all cases contains the
monolithiate 1 rather than the dilithiate 2. An X-ray
crystal structure determination of [1·TMEDA] was un-
dertaken and the molecular structure is shown in Fig. 3
with significant bond lengths and angles detailed in
Table 1. The lithium is four coordinate, chelated by
both the TMEDA and the phosphoniodiylide ligand in
a
distorted tetrahedral geometry. The phospho-
niodiylide system is coordinated not via both P-CHPh
carbon atoms as might be anticipated, but rather
through one of the CHPh carbon atoms and the ipso–
ortho bond of the phenyl ring of the second CHPh
group. The phosphorous bonds to the CHPh carbon
,
atoms [P(1)ꢀC(3) 1.712(3), P(1)ꢀC(2) 1.743(3) A] are
significantly shorter than that to the uncoordinated
,
benzylic methylene [P(1)ꢀC(1) 1.845(3) A], and the
bonding in this unit is best interpreted in terms of a
P(1)ꢀC(3) double bond and a P(1)ꢀC(2) single bond
shortened by its polar nature. The chelate angle ob-
tained by TMEDA at the lithium centre is as would be
expected at 88.3(2)°, compared to that of the ligand
bite-angle of 98.9(3)°, highlighting the extremely dis-
torted tetrahedral environment of the lithium. The dif-
ference in the bond lengths of the two coordinated
2. Results and discussion
2.1. Lithiation studies of [PhP(CH2Ph)3]Cl
Treatment of phenyl tribenzylphosphonium chloride
n
,
[PhP(CH2Ph)3]Cl with three molar equivalents of BuLi
ligand units [LiꢀC(2) 2.158(6), LiꢀC(36) 2.639(6) A]
in THF results in the removal of three benzylic protons
and formation of the orange dilithiated species
[Li2PhP(CHPh)3]. Triple deprotonation was established
by the observation that treatment of this species with
excess MeI provides the trimethylated phosphonium
salt [PhP(CHMePh)3]I. This salt was characterised by
mass spectrometry, with the successive loss of each
methylated benzyl being observed, however NMR data
was complicated by the presence of a mixture of
diastereomers, although the phenyl and methyl signal
integrals show the expected 20:9 ratio.
emphasises the difference between the two interactions.
The difference between the 31P chemical shift observed
for 1 (−1.7 ppm) and 1.TMEDA (35.8 ppm) is note-
worthy and arises possibly as a result of the coordina-
tion of the Li by TMEDA rather than THF which will
alter the Lewis acidity of this centre. The possibility
that the Li is coordinated in a different manner in 1, via
the ylidic PꢁCHPh unit for example, cannot be ex-
cluded and the large chemical shift difference suggests
that this may be the case.
Addition of two equivalents of nbutyllithium to a
suspension of finely ground [PhP(CH2Ph)3]Cl in THF-
d8 provides a solution whose NMR spectra indicate
formation of the monolithiated (phosphoniodiylide)
species [LiPhP(CHPh)2(CH2Ph)] (1) (Fig. 2). The 13C-
31P coupling constants observed in this system are of
interest with coupling to the CH carbon atoms being
much larger (124 Hz) than to the CH2 carbon atom (44
Hz), a feature which we attribute to the increased
interaction of the unsaturated carbon centres with the
2.2. Synthesis and characterisation of transition metal
complexes
Treatment of the chloro-bridged dimer [(h-
C5Me5)RhCl2]2 with the dilithiate 2 in THF provides an
orange solution from which a species, initially assumed
to be [(h-C5Me5)Rh{h3-(CHPh)3PPh}], was obtained.
However, spectroscopic data show that the complex
isolated contains a mono- rather than a dianionic lig-
and and as such, may be identified as the phospho-

238
P.J. Bailey et al. / Journal of Organometallic Chemistry 625 (2001) 236–244
Fig. 2. NMR spectra of [LiPhP(CHPh)2(CH2Ph)] (1) and [Li2PhP(CHPh)3] (2) in THF-d8.
niodiylide complex [(h-C5Me5)Rh{h2-(CHPh)2(CH2-
Ph)PPh}Cl] (3). Although this complex may be formed
directly from the monolithiate 1, we feel that, in light of
the demonstrated formation of the ligand dianion
above, its formation in the current instance is the result
of the hydrolysis of the putative [(h-C5Me5) Rh{h3-
(CHPh)3PPh}] followed by the re-coordination of chlo-
ride, presumably present as LiCl which has a small but

P.J. Bailey et al. / Journal of Organometallic Chemistry 625 (2001) 236–244
239
Fig. 3. Molecular structure of [(TMEDA)LiPhP(CHPh)2(CH2Ph)]
(1·TMEDA).
Fig. 4. Molecular structure of [(h-C5Me5)Rh{h2-(CHPh)2-
PPh(CH2Ph)}Cl] (3).
significant solubility in THF, to the resulting cationic
species. This assumption is supported by the fact that,
although lithiation of 3 may be effected by BuLi, as
evidenced by a colour change from red to green, all
attempts at isolation of the resulting species lead only
to the recovery of 1.
An X-ray crystal structure determination of 3 was
undertaken and the molecular structure is shown in
Fig. 4 and significant bond lengths and angles are
provided in Table 1. The diylide ligand chelates the
n
rhodium with
a
bite-angle of 77.1(3)°, and the
Table 1
,
Selected bond lengths (A) and angles (°) for 1·TMEDA, 3 and 4
1·TMEDA
3
4
Bond lengths
LiꢀN(54)
LiꢀN(51)
LiꢀC(2)
LiꢀC(36)
LiꢀC(53)
PꢀC(3)
PꢀC(2)
PꢀC(41)
PꢀC(1)
2.055(6)
2.126(6)
2.158(6)
2.639(6)
2.730(6)
1.712(3)
1.743(3)
1.827(3)
1.845(3)
1.506(4)
1.458(4)
1.421(4)
Rh(1)ꢀC(1)
Rh(1)ꢀC(2)
Rh(1)ꢀCl(1)
P(1)ꢀC(1)
P(1)ꢀC(2)
P(1)ꢀC(3)
2.166(7)
2.184(7)
2.4507(18)
1.762(7)
1.7775(18)
1.824(7)
1.806(7)
Cr(1)ꢀC(3)
Cr(1)ꢀC(2)
Cr(1)ꢀBr(1)
P(1)ꢀC(1)
P(1)ꢀC(2)
P(1)ꢀC(3)
P(1)ꢀC(41)
2.203(9)
2.169(8)
2.4606(19)
1.818(9)
1.802(8)
1.776(9)
1.821(9)
P(1)ꢀC(11)
C(1)ꢀC(11)
C(2)ꢀC(21)
C(3)ꢀC(31)
Bond angles
N(54)ꢀLiꢀN(51)
C(2)ꢀLiꢀC(36)
N(51)ꢀLiꢀC(36)
N(54)ꢀLiꢀC(2)
C(3)ꢀPꢀC(2)
C(3)ꢀPꢀC(41)
C(2)ꢀPꢀC(41)
C(3)ꢀPꢀC(1)
C(2)ꢀPꢀC(1)
C(41)ꢀPꢀC(1)
PꢀC(2)ꢀLi
88.3(2)
98.9(2)
106.3(2)
129.9(3)
116.79(14)
103.46(13)
103.46(13)
102.84(14)
112.74(15)
107.49(13)
108.9(2)
C(1)ꢀRh(1)ꢀC(2)
C(1)ꢀP(1)ꢀC(2)
C(1)ꢀP(1)ꢀC(11)
C(2)ꢀP(1)ꢀC(11)
C(1)ꢀP(1)ꢀC(3)
C(2)ꢀP(1)ꢀC(3)
C(11)ꢀP(1)ꢀC(3)
77.1(3)
100.0(3)
113.7(3)
115.6(3)
111.6(3)
111.2(3)
104.9(3)
C(3)ꢀCr(1)ꢀC(2)
C(1)ꢀP(1)ꢀC(2)
C(1)ꢀP(1)ꢀC(41)
C(2)ꢀP(1)ꢀC(41)
C(1)ꢀP(1)ꢀC(3)
C(2)ꢀP(1)ꢀC(3)
C(41)ꢀP(1)ꢀC(3)
78.7(3)
110.6(4)
105.1(4)
114.6(4)
109.9(4)
101.6(4)
115.2(4)
PꢀC(3)ꢀC(31)
126.2(2)

240
P.J. Bailey et al. / Journal of Organometallic Chemistry 625 (2001) 236–244
phosphoniodiylide ligand do not differ significantly
C(1)ꢀP(1)ꢀC(2) angle at phosphorus is 100.0(3)°. With
the exception of C(11)ꢀP(1)ꢀC(3) [104.9(3)°], the re-
maining angles at phosphorus are correspondingly in-
creased from tetrahedral and range from 111.2(3) to
115.6(3)°. These distortions are less severe than those
observed for a ruthenium complex of a nitrogen donor
analogue of this ligand which we have recently reported
[11], reflecting the greater atomic size of carbon. The
phosphorus bonds to the metal coordinated carbon
,
[CrꢀC(2)=2.169(8), CrꢀC(3)=2.203(9) A] and all PꢀC
bond lengths are likewise not distinguishable on the
basis of crystallographic data, ranging from 1.776(9) to
1.818(9). The diylide chelates the chromium with a bite
angle of 78.7(3)° and the C(2)ꢀP(1)ꢀC(3) angle at phos-
phorous is 101.6(4)°, with the remaining angles around
the phosphorous ranging from 105 to 115°, indicating a
minor distortion from tetrahedral geometry.
Phosphoniodiylide ligands [R2P(CR2)2]− exhibit a
preference for a (m2-h2-) bridging mode of coordination
and have a particular affinity for Au(I) [12], although
examples of the bridging of other metal dimers by
phosphoniodiylides are also known, e.g. [Mo2{(m2-h2-
CH2)2PMe2}4] [13]. In comparison, relatively few exam-
,
atoms [P(1)ꢀC(1) 1.762(7), P(1)ꢀC(2) 1.775(8) A] are
significantly shorter than that to the uncoordinated
,
benzylic methylene [P(1)ꢀC(3) 1.824(7) A] indicating the
heteroallylic nature of the coordinated unit. Signifi-
cantly, the phenyl groups bonded to the coordinated
carbon atoms are arranged trans with respect to the
C(1)ꢀC(2) vector as would be required in a C3 symmet-
ric h3-coordinated ligand, and furthermore the uncoor-
dinated benzyl group adopts the correct orientation for
such a coordination symmetry.
ples
of
complexes
containing
chelating
phosphoniodiylide ligands have been characterised [14],
and the Rh and Cr species reported here are, to our
knowledge, the first such complexes for these metals
[15].
Treatment of the bromo-bridged dimer [(h-
C5Me5)CrBr2]2 with 2 in THF provides a deep green
solution. Removal of the THF under vacuum leaves a
green residue, which is readily redissolved in ether
resulting in the formation of a white precipitate of
LiBr. Cooling of this solution to −30°C for several
days provides a crop of blue crystals from the green
solution. X-ray crystal structure analysis reveals the
structure to be that of the phosphoniodiylide complex
[(h-C5Me5)Cr{h2-(CHPh)2PPh(CH2Ph)}Br] (4). The
molecular structure of 4 is shown in Fig. 5 and signifi-
cant bond lengths and angles are given in Table 1. The
cyclopentadienyl ring is bound in a symmetrical h5-
fashion with CrꢀCring distances in the range of [2.240–
In an attempt to obtain the h3-coordinated ligand,
the chelate complex 4 was dissolved in THF to obtain
n
a blue solution and addition of 1 equivalent of BuLi
was carried out at −78°C. On warming the solution
slowly to room temperature a deep green colour re-
sulted, indicating that the third arm of the ligand has
been successfully deprotonated. However, only a blue
precipitate could be obtained from the resulting solu-
tion, which elemental analysis confirmed to be the
starting phosphoniodiylide complex 4, and it must
therefore be assumed that the coordination of the third
benzylic carbon, if it occurs at all, is only weak.
Our lithiation–methylation and NMR studies have
shown that [PhP(CHPh)3Li2] (2) is formed on treatment
of the phosphonium salt with 3 molar equivalents of
nBuLi, and the isolation of phosphoniodiylide com-
plexes from the reaction of this species with the halide
complexes [(h-C5Me5)RhCl2]2 and [(h-C5Me5)CrBr2]2
must therefore be assumed to be due to the hydrolytic
instability of the desired species [(h-C5Me5)M{h3-
(CHPh)3PPh}] due to the inability of this system to act
as an effective tridentate ligand. In our studies of the
analogous nitrogen donor ligand system [(NR)2PPh-
(NHR)]−, for which we found a similar inability to
coordinate in an h3-mode, we were able to attribute this
failure to the electronic stabilisation of the unsaturated
metal centre by strong p-donation by the chelating
ligand, thus destabilising the h3-mode relative to the
chelating mode of coordination [11]. However, the car-
bon donors of the chelating [(CHPh)2PPh(CH2Ph)]−
phosphoniodiylide ligand are incapable of p-donation
and such an explanation is thus not valid for this
ligand. We must therefore attribute our inability to
isolate the complexes containing h3-(CHPh)3PPh lig-
ands by removal of HX from [(h-C5Me5)M{h2-
PhP(CH2Ph)(CHPh)2}X] to the angle strain involved in
,
2.280 A]. The two CrꢀC bond distances to the chelating
Fig. 5. Molecular structure of [(h-C5Me5)Cr{h2-(CHPh)2PPh-
(CH2Ph)}Br] (4).

P.J. Bailey et al. / Journal of Organometallic Chemistry 625 (2001) 236–244
241
forming the three M-CꢀPꢀC 4-membered rings neces-
sary in such a process. However, electronic factors must
also be important as such angle strain does not prevent
the trimethylenemethane ligand [C(CH2)3]2− from
forming stable complexes in an h3-coordination mode.
spectra were recorded from the resultant clear
solutions.
Data for 1: H-NMR (250 MHz, THF-d8): l 5.8–7.9
1
(mult, 20H, Ph), 3.55 (d, JHꢀP=13 Hz, 2H, CH2), 2.29
(d, JHꢀP=17 Hz, 2H, CH); 13C-NMR (62.9 MHz,
THF-d8): l 151.4 (d, quat, JCꢀP=32.0 Hz, PꢀPh), 138.0
(d, quat, JCꢀP=15.0 Hz, CHꢀPh), 136.7 (d, quat,
3. Experimental
JCꢀP=4.0 Hz, CH2ꢀPh), 9 other peaks between 108–
141 (Ph), 34.6 (d, JCꢀP=124 Hz, CH), 34.3 (d, JCꢀP
=
3.1. General
45.0 Hz, CH2); 31P-NMR (101.2 MHz, THF-d8) l
−1.7.
All reactions were carried out under an atmosphere
of dry, oxygen free nitrogen using standard Schlenk
techniques and solvents which were dried and distilled
under nitrogen immediately prior to use. The phospho-
nium salt [PhP(CH2Ph)3]Cl was prepared by treatment
of phenyl dibenzyl phosphine (from PhPCl2 and benzyl
magnesium chloride) with benzyl chloride, while [{(h-
C5Me5)RhCl2}2] and [{(h-C5Me5)CrBr2}2] were pre-
pared by the literature procedures [16,17]. NMR
spectra were recorded on a Bruker AC 250 spectrome-
ter and infrared spectra on a Perkin–Elmer Paragon
1000 spectrometer from samples as KBr discs. Mass
spectra were recorded on a Kratos MS50 TC instru-
ment in positive ion FAB mode using 3-nitrobenzyl
alcohol as matrix and CsI as calibrant. Elemental
analyses were conducted by the microanalytical service
of this department.
Data for 2: 1H-NMR (250 MHz, THF-d8): l 5.9
(mult, Ph), 6.7 (mult, Ph), 7.2 (mult, Ph), 8.1 (mult, Ph),
2.6 (d, JHꢀP=11.2 Hz, 3H, CH). 13C-NMR (62.9 MHz,
THF-d8): l 142.3 (d, JCꢀP=61.0 Hz, PꢀPh), 117.8 (d,
JCꢀP=12.0 Hz, PꢀPh), 130.7 (d, JCꢀP=9.5 Hz, PꢀPh),
126.1 (d, JCꢀP=8.0 Hz, PꢀPh), 151.7 (d, JCꢀP=4.0 Hz,
CHꢀPh), 127.2 (CHꢀPh), 108.8 (CHꢀPh), 105.1
(CHꢀPh), 40.4 (d, JCꢀP=105 Hz, CH). 31P-NMR
(101.2 MHz, THF-d8) l −3.9.
3.4. Synthesis of [(TMEDA)Li(CH2Ph)(CHPh)2PPh]
(1·TMEDA)
To a suspension of [PhP(CH2Ph)3]Cl (0.20 g, 0.481
n
mmol) in ether, BuLi (1.6 M solution in hexane) (0.90
cm3, 1.442 mmol) and TMEDA (0.15 cm3, 0.962 mmol)
were added at −78°C. The reaction mixture was al-
lowed to warm to r.t. and stirred for 2 h. The bright red
solution obtained was filtered through celite and the
resultant clear filtrate reduced to half its original vol-
ume. After leaving this solution for 10 h at −30°C,
yellow crystals of 1 were obtained in 78% yield. Ele-
mental analysis: Anal. Found C, 79.13; H, 7.69; N,
5.92; C33H40N2PLi requires C, 78.85; H, 8.04; N, 5.58%.
1H-NMR (250 MHz, CDCl3): l 7.1–7.6 (mult, 19H,
Ph), 3.3 (d, JHꢀP=14.0 Hz, 4H, CH2ꢀP), 2.4 (4H,
CH2ꢀN), 2.2 (12 H, CH3ꢀN), 2.1 (d, JHꢀP=12.0 Hz,
1H, CHꢀP).); 13C-NMR (62.9 MHz, CDCl3): l 125–
135 (Ph), 45.6 (CH3ꢀN), 57.4 (CH2ꢀN), 29.5 (quat,
PhꢀLi), 37.2 (d, JPꢀC=63.5 Hz, PꢀCH2), 326.2 (d,
3.2. Synthesis of [PhP(CHMePh)3]I
Finely ground [PhP(CH2Ph)3]Cl (0.50 g, 1.20 mmol)
was suspended in THF (25 cm3) with rapid magnetic
n
stirring and BuLi (2.25 cm3 of a 1.6 M solution in
hexane, 3.6 mmol) was slowly added by syringe under
nitrogen. After stirring for 1 h, methyl iodide (0.25 cm3,
4.01 mmol) was added by syringe. After a further 15
min 50 cm3 of water was added and the mixture ex-
tracted with CH2Cl2 (3×20 cm3). Crystallisation pro-
vided the product as a colourless microcrystalline
1
material (0.46 g, 70%). H-NMR (CDCl3): complex set
of overlapping multiplets due to
a
mixture of
J
PꢀC=64Hz, CHꢀP); 31P-NMR (101.2 MHz, CDCl3): l
35.8.
diastereoisomers; MS (+FAB): 423 (M+), 319 (M+−
CHMePh), 213 (M+−2CHMePh), 105 (CHMePh).
3.5. Synthesis of
3.3. Preparation of [Li(CHPh)2(CH2Ph)PPh] (1) and
[Li2(CHPh)3PPh] (2) for NMR studies
[(p-C5Me5)Rh{p2-(CHPh)2PhP(CH2Ph)}Cl] (3)
Finely ground [PhP(CH2Ph)3]Cl (0.825 g, 1.98 mmol)
was suspended in THF (25cm3) with rapid magnetic
The solvent was removed from the required volume
of butyllithium in hexane (1.6 M) in vacuo in a 5 mm
n
n
stirring and BuLi (3.72 cm3 of a 1.6 M solution in
NMR tube. Deuterated THF (0.5 cm3) was then added
to the resultant powder at −30°C before addition of
either 0.5 (for 1) or 0.33 (for 2) molar equivalents of
[PhP(CHPh)3]Cl. The NMR tube was sealed and the
mixture was allowed to warm to room temperature
(r.t.) for 30 min with occasional shaking. The NMR
hexane, 5.95 mmol) was slowly added by syringe under
nitrogen. After stirring for 1 h [(h-C5Me5)RhCl2]2
(0.611 g, 0.99 mmol) was added to the orange solution
and allowed to stir overnight upon which the reaction
mixture had become deep red in colour. The solvent
was removed in vacuo and the red residue redissolved

242
P.J. Bailey et al. / Journal of Organometallic Chemistry 625 (2001) 236–244
Table 2
Crystal data for 1·TMEDA, 3 and 4
1·TMEDA
3
4
Empirical formula
Formula weight
Crystal system
Space group
C35.25H45.25LiN2P
534.89
Monoclinic
P21/c
C
635.01
Monoclinic
P21/c
37H39ClPRh
C37H39CrPBr
646.56
Monoclinic
P21/c
,
a (A)
9.6393(9)
16.9065(15)
19.852(2)
92.382(8)
3232.4(5)
4
10.685(6)
19.349(7)
15.194(4)
93.67(3)
3135(2)
10.645(4)
19.381(10)
15.325(7)
94.04(6)
3154(3)
,
b (A)
,
c (A)
i (°)
V (A )
3
,
Z
4
4
Temperature (K)
Crystal habit
Crystal size (mm)
220(2)
200(2)
150(2)
Orange block
0.58×0.39×0.23
1.54184
5–140
1.099
Red block
0.23×0.23×0.23
0.71073
5–50
1.384
Blue plate
0.57×0.23×0.08
1.54184
5–140
1.326
,
Wavelength (A)
2q range (°)
Dcalc (Mg m−3
)
v(Mo–Ka) (mm−1
)
0.919
0.705
5.115
Transmission range
Number of parameters
0.311–0.417
474
0.834–0.864
367
0.355–0.761
366
R1 (Number of data with F\4|(F)]
0.0655 (4219)
0.1906 (5933)
−0.35 to +0.45
0.0632 (3097)
0.1299 (5547)
−0.81 to +0.61
0.0929
0.2852
−1.64 to +2.05
wR2 (all data)
Final difference map (eA
−3
,
)
in ether, filtered through celite and passed through a
dry flash column using EtOAc as eluent. The solvent
was then allowed to evaporate slowly resulting in dark
red crystals of compound 1 (32%). Elemental analysis:
Anal. Found: C, 68.7; H, 6.01; C37H39ClPRh requires
ing to r.t. provided a green solution. Removal of the
solvent in vacuo and dissolution of the resulting residue
in ether gave a similar deep green solution with a small
amount of white precipitate which was identified as
LiBr by AgNO3 and flame tests. Filtration of this
solution through a pad of celite, reduction in the vol-
ume of ether and storing of the solution at −30°C
overnight resulted in the formation of blue crystals
(0.160 g, 52%). These were found to be of insufficient
quality for X-ray diffraction analysis. The solution was
then re-filtered through celite and the resulting solution
was stored at 5°C. After one week the solution afforded
a crop of blue crystals suitable for X-ray diffraction
analysis. Elemental analysis: Anal. Found C, 68.61; H,
6.13; N, 0; C37H39PClCr requires C, 68.72; H, 6.09; N,
0%.
1
C, 68.0; H, 5.97%. H-NMR (250 MHz, CDCl3): l 1.1
2
(s, 15H, Cp*CH3), 2.8 [dd, 1H, HCHPh, J(1H-31P)=
13.5 Hz, 2J(1H-1H)=10.0 Hz], 3.2 [dd, 1H, RhCH,
2
1J(1H-31P)=8.5, J(1H-103Rh)=3.0 Hz], 3.3 [dd, 1H,
RhCH, 1J(1H-31P)=7.0, 2J(1H-103Rh)=3.0 Hz], 5.7
2
2
[dd, 1H, HCHPh, J(1H-31P)=14.0, J(1H-1H)=12.0
Hz], 6.5–8.0 (m, 20H, ArH). 13C-NMR (62.89 MHz,
CDCl3): l −7.1 [dd, RhCP, J(13C-31P)=43.0, J(13C-
1
2
103Rh)=23.0 Hz], −3.7 [dd RhCP, J(13C-31P)=52.5,
1
2J(13C-103Rh)=18.5 Hz], 7.9 (Cp*CH3), 31.7 [d, CH2,
1J(13C-31P)=40.0 Hz], 94.0 [d, Cp*, quatC, 1J(13C-
103Rh)=6.5 Hz], 135.2, 135.7 (RhCHPh, quatC), 140.5
[d, PPh, quatC, J(13C-31P)=6.5 Hz], 144.2 (CH2Ph,
1
quatC). 31P-NMR (101.26 MHz, CDCl3): l 37.8 [d,
2J(31P-103Rh)=27.0 Hz). MS (+FAB): m/z=617 (M+
), 447, 380, 289, 237, 154, 77.
4. X-ray crystallography
All data collections were performed on a Stoe Stadi-4
diffractometer equipped with an Oxford Cryosystems
low-temperature device. Structures 1·TMEDA and 3
were solved by direct methods (SIR92 [18]), structure 4
was solved by Patterson methods (DIRDIF [19]); all
refinements were performed against F2 (SHELXL97 [20]).
The absorption correction for 1·TMEDA was per-
formed using „-scan data; those for 3 and 4 were
performed by Gaussian integration following refine-
ment of the crystal shape and dimensions against a set
of „-scans (Stoe X-Shape [21]). Crystal and refinement
3.6. Synthesis of
[(p-C5Me5)Cr{p2-(CHPh)2PhP(CH2Ph)}Br] (4)
Finely ground [PhP(CH2Ph)3]Cl (0.20 g, 0.480 mmol)
was suspended in THF (25 cm3) and BuLi (0.90 cm3,
n
1.44 mmol, 1.6 M solution in hexane) was added at
−78°C. The solution was allowed to warm to r.t. and
then recooled to −78°C. To the resulting red solution
was added [Cp*CrBr2]2 (0.166 g, 0.240 mmol), rewarm-

P.J. Bailey et al. / Journal of Organometallic Chemistry 625 (2001) 236–244
243
data are given in Table 2. The structure of 1·TMEDA
References
contained a disordered hexane molecule in the region of
a crystallographic inversion centre. The H atoms incor-
porated in the phosphoniodiylide were located in a
difference map and refined freely, methyl groups were
treated as rotating rigid bodies and all other H atoms are
in idealised positions. All non-H atoms with the exception
of the minor solvent component were refined anisotrop-
ically. Refinement of the crystal structure of 3 presented
no major problems. H atoms attached to C1, C2 and C3
were located in difference maps, but are included in
idealised positions, CH3 groups were treated as rotating
rigid bodies, and all other H positions were calculated.
Allnon-Hatomsweremodelledwithanisotropicdisplace-
ment parameters. Crystals of 4 were of rather low quality,
being mostly small and clustered. The crystal selected was
a plate with another small crystal attached to the
prominent growth face (100), but it was not possible to
dissect the crystal clusters without completely fragment-
ing the sample. Cu–Ka radiation was used for data
collection for its higher intensity and scattering efficiency
relative to Mo–Ka radiation, although absorption effects
were fierce. The resulting crystal structure is therefore of
low precision, and while it confirms connectivity (note the
isostructural relationship with 3) it is not of sufficient
quality for detailed analysis. H positions were all calcu-
lated; all non-H atoms were refined anisotropically.
Twinning via 180° rotation about (001) would predom-
inantly affect data where 0.1 l is near integral; omitting
these data reduces the final difference map residuals to
[1] (a) M.D. Jones, R.D.W. Kemmitt, Adv. Organomet. Chem. 27
(1987) 279 and references therein. (b) R.S. Lokey, N.S. Mills,
Organometallics, 8 (1989) 1803. (c) G.E. Herberich, U. Englert,
L. Wesemann, P. Hofmann, Angew. Chem. Int. Ed. Engl. 30
(1991) 313. (d) G.E. Herberich, T.P. Spaniol, J. Chem. Soc.,
Chem. Commun. (1991) 1457. (e) G.E. Herberich, T.P. Spaniol,
J. Chem. Soc., Dalton Trans. (1993) 2471. (f) I.-Y.Wu, M.-C.
Cheng, Y.-C. Lin, Y. Wang, Organometallics, 12 (1993) 1686.
[2] G.E. Herberich, C. Kreuder, U Englert, Angew. Chem. Int. Ed.
Engl. 33 (1993) 2465.
[3] (a) G.C. Bazan, G. Rodriguez, B.P. Cleary, J. Am. Chem. Soc.
116 (1994) 2177. (b) G. Rodriguez, G.C. Bazan, J. Am. Chem.
Soc. 117 (1995) 10155.
[4] B.M. Trost, Angew. Chem. Int. Ed. Engl. 25 (1986) 1 and
references therein.
[5] P.J. Bailey, A.J. Blake, M.J. Kryszczuk, S. Parsons, D. Reed, J.
Chem. Soc. Chem. Commun. (1995) 1647.
[6] (a) M.J. Burk, R.L. Harlow, Angew. Chem Int. Ed. Engl. 29
(1990) 1462. (b) T.R. Ward, L.M. Venanzi, A. Albinati, F.
Lianza, T. Gerfin, V. Gramlich, G.M.R. Tombo, Helv. Chim.
Acta 74 (1991) 983. (c) M.J. Burk, J.E. Feaster, R.L. Harlow,
Tet. Asym. 2 (1991) 569. (d) W.A. Nugent, J. Am. Chem. Soc.
114 (1992) 2786. (e) C.J. Tokar, P.B. Kettler, W.B. Tolman,
Organometallics 11 (1992) 2737. (f) H. Adolfsson, K. Warnmark,
C. Moberg, J. Chem. Soc.Chem. Commun. (1992) 1054. (g) D.D.
LeCloux, W.B. Tolman, J. Am. Chem. Soc. 115 (1993) 1153. (h)
D.D. LeCloux, C.J. Tokar, M. Osawa, R.P. Houser, M.C. Keys,
W.B. Tolman, Organometallics 13 (1994) 2855. (i) W.A. Nugent,
R.L. Harlow, J. Am. Chem. Soc. 116 (1994) 6142. (j) B.S.
Hammes, D. Ramos-Maldonado, G.P.L. Yap, L. Liable-Sands,
A.L. Rheingold, V.G. Young Jr., A.S. Borovik, Inorg. Chem. 36
(1997) 3210. (k) C. Bolm, D. Kadereit, M. Valacchi, Syn. Let.
(1997) 687.
[7] C. Moberg, Angew. Chem. Int. Ed. Engl. 37 (1998) 248.
[8] J.K. Whitesell, Chem. Rev. 89 (1989) 1581.
91.3 eA−3, but only reduces R1 to 8.8%. Improvements
,
[9] (a) P.J. Bailey, L.A. Mitchell, S. Parsons, J. Chem. Soc. Dalton.
Trans. (1996) 2839. (b) P.J. Bailey, R.O. Gould, C.N. Harmer, S.
Pace, A. Steiner, D.S. Wright, Chem. Commun. (1997) 1161. (c)
M.B. Dinger, W. Henderson, J. Chem. Soc. Chem. Commun.
(1996) 211. (d) S.D. Robinson, A. Sahajpal, J. Chem. Soc.
Dalton Trans. (1997) 3349. (e) J.R. da, S. Maia, P.A. Gazard,
M. Kilner, A.S. Batsanov, J.A.K. Howard, J. Chem. Soc. Dalton
Trans. (1997) 4625. (f) P.J. Bailey, S.F. Bone, L.J. Yellowlees,
L.A. Mitchell, S. Parsons, K.J. Taylor, Inorg. Chem. 36 (1997)
867. (g) P.J. Bailey, S.F. Bone, L.J. Yellowlees, L.A. Mitchell, S.
Parsons, K.J. Taylor, Inorg. Chem. 36 (1997) 5420.
of a similar order of magnitude could be obtained using
a calculated absorption correction (SHELXA), and so the
high R factor and noisy difference map could also be due
to residual absorption errors. Both these effects are
neglected in the data presented here.
5. Supplementary material
[10] P.J. Bailey, S. Pace, Coord. Chem. Rev. 214 (2001) 91.
[11] P.J. Bailey, K.J. Grant, S. Parsons, Organometallics 17 (1998)
551.
[12] (a) H. Schmidbaur, Acc. Chem. Res. 8 (1975) 113. (b) H.
Schmidbaur, Angew. Chem. Int. Ed. Engl. 15 (1976) 728. (c) H.
Schmidbaur, Pure Appl. Chem. 50 (1978) 19.
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC Nos. 151 738, 151 739 and 151 740
for compounds 1·TMEDA, 3 and 4, respectively. Copies
of this information may be obtained free of charge from
The Director, CCDC, 12 Union Road, Cambridge CB2
1EZ, UK (fax: +44-1223-336033 or e-mail: deposit@
ccdc.cam.ac.uk or www: http://www.ccdc.cam.ac.uk).
[13] F.A. Cotton, B.E. Hanson, W.H. Ilsley, G.W. Rice, Inorg.
Chem. 18 (1979) 2713.
[14] (a) D.J. Brauer, C. Kruger, P.J. Roberts, Y.-H. Ssay, Chem. Ber.
107 (1974) 3706. (b) J.-M. Bassett, J.R. Mandt, H. Schmidbaur,
Chem. Ber. 113 (1980) 1145. (c) H. Schmidbaur, A. Moetl, B.
Zimmer-Gasser, Chem. Ber. 114 (1981) 3161. (d) G.W. Rice,
G.B. Ansell, M.A. Modrick, S. Zentz, Organometalics 2 (1983)
154. (e) J.P. Fakler, B. Trzcinska-Bancroft, Organometallics 4
(1985) 1891. (f) E.J. More, B.D. Santarsiero, Acta Crystallogr.
Sect. C 44 (1988) 1911. (g) M. Gomez, G. Jimenez, P. Royo,
M.A. Pellinghelli, A. Tiripicchio, J. Organomet. Chem. 439
(1992) 309. (h) M.J. Hostetler, M.D. Butts, R.G. Bergman,
Inorg. Chim. Acta 198 (1992) 377.
Acknowledgements
We thank the Engineering and Physical Science Re-
search Council (EPSRC) for a project studentship (to
T.B.) and provision of a 4-circle diffractometer.

244
P.J. Bailey et al. / Journal of Organometallic Chemistry 625 (2001) 236–244
[15] Cambridge Structural Database search, issue 20, October 2000.
[16] C. White, A. Yates, P.M. Maitlis, Inorg. Synth. 29 (1992)
228.
Garcia-Granda, R.O. Gould, R. Israel, J.M.M. Smits, Crystal-
lography Laboratory, University of Nijmegen, The Netherlands.
1996.
[17] F.H. Kohler, J. Lachmann, G. Muller, H. Zeh, H. Bruneer, J.
Wachter, J. Organomet. Chem. 365 (1989) C15.
[18] (a) A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, J.
Appl. Crystallogr. 26 (1993) 343. (b) Stoe X-Shape–Stoe & Cie,
X-Shape, Darmstadt, Germany, 1998.
[20] G. M. Sheldrick, SHELX 97 & SHELXA Programs for Crystal
Structure Analysis (Release 97-2), Institu¨t fu¨r Anorganische
Chemie der Universita¨t, Tammanstrasse 4, D-3400 Go¨ttingen,
Germany, 1998.
[21] Stoe X-Shape–Stoe & Cie, X-Shape, Darmstadt, Germany,
1998.
[19] P.T. Beurskens, G. Beurskens, W.P. Bosman, R. de Gelder, S.
.