Manganese complexes of the pentaphenylcyclopentadienyl ligand

Article abstract of DOI:10.1016/j.poly.2005.08.055

The preparation and characterisation of [Mn(η⁵-C₅Ph₅)(CO)₃] (1), [Mn(η⁵-C₅Ph₅)(CO)₂(PMe₂Ph)] (2), and [Mn(η⁵-C₅(p-tol)₅)(CO

Full text of DOI:10.1016/j.poly.2005.08.055

Polyhedron 25 (2006) 1498–1506
www.elsevier.com/locate/poly
Manganese complexes of the pentaphenylcyclopentadienyl ligand
*
Leslie D. Field, Taian He, Paul Humphrey, Anthony F. Masters , Peter Turner
School of Chemistry and Centre for Heavy Metals Research, The University of Sydney, Sydney, NSW 2006, Australia
Received 27 July 2005; accepted 22 August 2005
Available online 30 January 2006
Abstract
The preparation and characterisation of [Mn(g5-C5Ph5)(CO)3] (1), [Mn(g5-C5Pꢀh5)(CO)2(PMe2Ph)] (2), and [Mn(g5-C5(p-tol)5)(CO)3]
(3) are described; these complexes are the first reported fully characterised C5Ar5 derivatives of Group 7. Compounds (1) and (3) are
reversibly oxidised at +0.98 and +0.89 V, respectively (versus Fc+/0), significantly more positive than the oxidation of the analogous
[Mn(g5-C5H5)(CO)3]. Single crystal X-ray diffraction studies of (1) and (2) (refined to R1(F) 0.0967, wR2(F2) 0.2971 and R1(F) 0.0472,
wR2(F2) 0.1356, respectively) are also reported.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
[Mn(g5-C5Ph5)(CO)2(PMe2Ph)] (2), as well as the analo-
gous penta-p-tolylcyclopentadienyl manganese complex, (g5-
penta(p-tolyl)cyclopentadienyl)tricarbonylmanganese(I),
[Mn(g5-C5(p-tol)5)(CO)3] (3). Surprisingly, interpretations
of the molecular structure of [Mn(g5-C5H5)(CO)3] have
been the subject of some ambiguity, the crystal structure
of this simple molecule having been reported three times
in the literature [5–7].
The cyclopentadienyl ligand has become one of the most
widely used ligands in organometallic chemistry, since the
discovery and characterisation of ferrocene were reported
in the early 1950s. Cyclopentadienyl complexes are known
for all of the transition metal and main group metals and
metalloids, as well as all of the lanthanoid and the more
common actinide metals. Since the synthesis of the ligand
precursors in 1925 [1,2] and the report of the first metal
complex in 1964 [3], pentaphenylcyclopentadienyl deriva-
tives of most elements of the Periodic Table have been
reported. Although main group, first, second and third
row transition metals and lanthanide complexes have been
reliably reported, fully characterised complexes of Groups
5 and 7 are notably absent. [Mn(g5-C5H5)(CO)3] is one of
the prototypical ‘‘half-sandwich’’ complexes [4], and its
derivatives have been used as gasoline additives and have
some industrial catalytic applications.
Ph
p-tol
p-tol
Ph
Ph
Ph
Ph
p-tol
p-tol
OC
p-tol
Mn
Mn
OC
L
CO
CO
CO
3
1
2
L = CO
L = PMe2Ph
2. Results and discussion
We report here, the synthesis and characterisation of
three pentaphenylcyclopentadienyl complexes of manganese,
(g5-pentaphenylcyclopentadienyl)tricarbonylmanganese(I),
[Mn(g5-C5Ph5)(CO)3] (1) (dimethylphenylphosphine)-
(g5-pentaphenylcyclopentadienyl)dicarbonylmanganese(I),
Pale yellow crystals of [Mn(g5-C5Ph5)(CO)3] (1) were
prepared in yields of 38–68% by reaction of LiC5Ph5 with
[Mn(CO)3(CH3CN)3]Br, or [Mn(CO)5Br], or of C5Ph5Br
with [Mn(CO)3(CH3CN)2Br]/Zn in refluxing tetrahydrofu-
ran. Yellow [Mn(g5-C5(p-tol)5)(CO)3] (3) was prepared
similarly. The synthetic routes to (1) and (3) are remarkably
different to the original syntheses of [Mn(g5-C5H5)(CO)3],
*
Corresponding author. Fax: +61 2 9351 3329.
E-mail address: a.masters@chem.usyd.edu.au (A.F. Masters).
0277-5387/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2005.08.055

L.D. Field et al. / Polyhedron 25 (2006) 1498–1506
1499
in which [Mn(g5-C5H5)2] was reacted with carbon monox-
ide at elevated temperatures and pressures [4,8]. Reactions
analogous to those used for the preparations of [Mn(g5-
C5Me5)(CO)3] and [Mn(g5-C5Bz5)(CO)3] were also not
suitable due to the very low solubility of C5Ph5H [9–12].
Table 1
Selected average interatomic distances and angles in [Mn(g5-C5Ph5)(CO)3]
(1)
Atoms
Average dimension
˚
Mn–C (C5)
Mn–CO
C–O
2.16(1) A
˚
ꢀ
1.78(1) A
For instance, the purple radical, C5Ph5, generated from
˚
1.17(1) A
the mixture of C5Ph5Br, zinc dust and Mn2(CO)10 in reflux-
ing tetrahydrofuran, was not sufficiently reactive to cleave
the Mn–Mn bond even though the reaction was continued
for four days. In most reactions involving Mn2(CO)10 as
a precursor, metal–metal bond cleavage has been shown
to be the initial step [13]. [Mn(CO)3Br], [Mn(CO)3-
(CH3CN)3]Br and [Mn(CO)3(CH3CN)2Br] are readily pre-
pared and are the most commonly used precursors for
substitution reactions as a consequence of the lability of
the CH3CN and Brꢀ ligands.
Both (1) and (3) are air stable in solid and in solution.
Their inertness parallels that of other substituted cyclopen-
tadienyl complexes. The kinetic inertness of [Mn(g5-C5RH4)-
(CO)3] (R = H, Me) is well known; no carbonyl exchange
with 14CO occurs in toluene at 32 ꢁC over several weeks,
no reaction occurs between PPh3 and [Mn(g5-C5H5)-
(CO)3] even in refluxing decalin and substitution of CO is
only possible either by ultraviolet irradiation or by using
Me3NO to actively remove CO [13].
˚
C–C (C5 ring)
C(C5)–ipso-C (C6)
CO–Mn–CO
Mn–C–O
1.42(2) A
˚
1.50(2) A
90.3(5)ꢁ
175(1)ꢁ
C–C–C (C5 ring)
107.7(1)ꢁ
reversible oxidation at ca. 0.29 V [19]. In general, the value
of the redox couple for a pentaphenylcyclopentadienyl
complex is shifted 0.25–0.30 V more positively than that
of its cyclopentadienyl analogue [20,21].
In addition, [Mn(g5-C5H5)(CO)3] exhibits a second irre-
versible oxidation at 1.48 V detected in CF3CO2H at a plat-
inum electrode [19], an electrochemically reversible
reduction, coupled to a follow-up chemical reaction at
ꢀ2.46 V [22] and a second irreversible reduction at
ꢀ3.32 V observed in tetrahydrofuran at a platinum elec-
trode [23]. The first reduction has been shown by prepara-
tive electrolysis to generate red [Mn(g5-C5H5)(CO)3]ꢀ.
However, chemical reduction of [Mn(g5-C5H5)(CO)3] in
HMPA yields predominantly first [Mn(CO)5]ꢀ and then
[Mn(CO)4]3ꢀ by poorly defined ligand redistribution reac-
tions [24]. We have observed no additional oxidations,
nor corresponding reductions of [Mn(g5-C5Ar5)(CO)3].
Similarly, [Mn(g5-C5Me5)(CO)3] undergoes monosub-
stitution either under ultraviolet irradiation or by oxidative
substitution [12,14]. There are many reports of cyclopenta-
dienyl manganese derivatives with one or more ligands
substituted for CO [13,15]. Recently, a study on the mono-
substitution of [Mn(g5-C5R5)(CO)3] (R = H, Me or Et) in
n-heptane solution under ultraviolet irradiation showed
that the bulkier ligand, C5Et5, better stabilised the forma-
tion of the solvated intermediate, [Mn(g5-C5R5)(CO)2-
(n-heptane)] than did the other two ligands [16]. Substitution
of one CO in [Mn(g5-C5Ph5)(CO)3] (1) by PMe2Ph was
readily achieved by refluxing a THF solution in the pres-
ence of Me3NO to give [Mn(g5-C5Ph5)(CO)2(PMe2Ph)]
(2). Using an analogous procedure, it was possible to intro-
duce a range of other substituents (including phosphines
and isonitriles).
2.1. Crystal structure determinations
2.1.1. [Mn(g5-C5Ph5)(CO)3] (1)
An air-stable crystal of [Mn(g5-C5Ph5)(CO)3] (1) was
grown by slow vapour diffusion of pentane into a dichloro-
methane solution of (1) and the crystal structure was
determined at room temperature. A single crystal X-ray
diffraction study of (1) was undertaken for comparison
with that of (2). Compound (1) crystallises in the mono-
clinic space group P21/n (#14). The crystal structure of
(1) consists of discrete molecules in the solid state and
no intermolecular contacts of significance were observed
in the solid. The closest intermolecular contact is
[Mn(g5-C5Ph5)(CO)3] (1) is not oxidised by reactants
such as Br2, I2 or NOBF4, even in the presence of CH3CN.
Recently, it has been found that monosubstitution of a car-
bonyl ligand proceeded smoothly when [Ru(g5-C5Ph5)
(CO)2Br] or [Os(g5-C5Ph5)(CO)2Br] were reacted with
other ligands in the presence of Me3NO [17,18].
However, compounds (1) and (3) each undergo a revers-
ible, one-electron cyclic voltametric oxidation in CH2Cl2 at
0.98 V for (1) and 0.89 V for (3) (both versus Fc+/0).
˚
3.46 A between C(33) on molecule (x,y,z) and C(34)
on molecule (ꢀx,ꢀy,ꢀz). The closest intermolecular
non-bonding approaches to manganese are greater than
˚
4 A. Final atomic positional coordinates, with estimated
standard deviations, bond lengths and angles and aniso-
tropic thermal parameters are deposited.1 Selected inter-
atomic bond distances and angles, with standard
½Mnðg5-C5Ar5ÞðCOÞ3ꢁ ꢁ ½Mnðg5-C5Ar5ÞðCOÞ3ꢁþ þ eꢀ
The reversibility of these oxidations confirms that com-
plexes of the penta-p-tolylcyclopentadienyl ligand,
C5ðp-tolÞ5ꢀ, are easier to oxidise than are those of the
1
The atomic coordinates and thermal parameters have been submitted
as supplementary material for deposition at the Cambridge Crystallo-
graphic Data Centre with deposition number CCDC 284608 (see
Deposition of crystallographic data).
ꢀ
C5Ph5 ligand. These oxidation potentials are much higher
than that for [Mn(g5-C5H5)(CO)3] which undergoes a

1500
L.D. Field et al. / Polyhedron 25 (2006) 1498–1506
Fig. 1. ORTEP [39] illustration of [Mn(g5-C5Ph5)(CO)3] (1), viewed perpendicular to the C5 ligand plane, showing the atom numbering scheme.
deviations derived from the refinement, are given in
Table 1. The atomic nomenclature is defined in Fig. 1.
Figs. 1 and 2 illustrate the molecular geometry viewed
along, and perpendicular to, the normal to the C5 plane
[{Os3(l-H)(CO)8(PPh3)2}(l-O2C-{(g5-C5H4)Mn(CO)3}]2)
[25], that of [Mn(g -C5H5)(CO)3] is 1.794 A.
5
˚
The C5 nucleus of the pentaphenylcyclopentadienyl
˚
ligand is planar to within 0.014 A. The C–C bond lengths
ꢀ
˚
of the C5Ph5 ligand, respectively.
of the C5 ring range from 1.39(2) to 1.46(1) A, the average
˚
C–C bond length being 1.42 A, which is consistent with
2.1.2. The [Mn(g5-C5Ph5)(CO)3] molecule
C5Ph5 rings in other metal complexes. The ipso-carbon
Coordination by the g5-pentaphenylcyclopentadienyl
group and the three carbonyl groups creates a ‘‘three-
legged piano stool’’ environment about the manganese
atoms of each of the phenyl rings are displaced slightly
˚
from the C5 plane by between 0.09 and 0.20 A, with the
˚
average distance being 0.16 A. The ipso-carbon atoms of
˚
atom. The Mn–(C5 centroid) distance is 1.784 A and
all phenyl rings are on the opposite side of the C5 plane
to the manganese atom. The phenyl rings are all also pla-
˚
the average Mn–C distance of 2.16 A may be compared
˚
with the mean of nearly 147 structures of [Mn(C5R5)(CO)3]
nar to within a mean deviation from the plane of 0.02 A
˚
compounds of 1.775(1) and 2.142(2) A. The Mn–CO dis-
and are canted at between 46.2ꢁ and 61.9ꢁ (average 51.6ꢁ)
to the C5 ring in a paddle wheel arrangement. The average
angle between the planes of adjacent phenyl rings of the
˚
tances (1.74(1), 1.77(1) and 1.83(1) A) are indistinguishable
and may be compared, for example, with the mean bond
length of 1.784(2) in the [Mn(C5R5)(CO)3] compounds.
The C@O bond lengths in (1) are the same within exper-
imental error. The Mn–(C5 centroid) distance is relatively
long; there being only 42 compounds with Mn–(C5 cen-
ꢀ
C5Ph5 ligand is 54.9ꢁ.
2.1.3. [Mn(g5-C5Ph5)(CO)2(PMe2Ph)] (2)
[Mn(g5-C5Ph5)(CO)2(PMe2Ph)] (2) also crystallises in the
monoclinic space group P21/n (#14). The crystal structure of
(2) consists of discrete molecules in the solid state and no
˚
troid) distances of 1.785 A or longer. There appears to
be no simple correlation between Mn–(C5 centroid) dis-
tance and other parameters – whilst many of these 42
compounds do have bulky cyclopentadienyl ligands and
the longest Mn–(ring centroid) distance reported to date
2
This distance is a considerable outlier in Mn–(C5 centroid) distances,
d, of [Mn(g5-C5R5)(CO)3] compounds – the next 7 longest distances are
in the range 1.812 6 d 6 1.823 A.
˚
˚
is the exceptionally long (at R = 0.0979) 1.912 A of

L.D. Field et al. / Polyhedron 25 (2006) 1498–1506
1501
Fig. 2. ORTEP [39] illustration of [Mn(g5-C5Ph5)(CO)3] (1), viewed parallel to the C5 ligand plane, showing the atom numbering scheme.
Table 2
intermolecular contacts of significance were observed in the
solid. The closest intermolecular contact is 3.332(5) A
Selected average interatomic distances and angles in [Mn(g5-C5Ph5)-
(CO)2PMe2Ph] (2)
˚
between O(1) on molecule (x,y,z) and C(41) on molecule
(1/2 ꢀ x,1/2 + y,1/2 ꢀ z). The closest intermolecular non-
Atoms
Average dimension
˚
˚
Mn–C (C5)
Mn–P
Mn–CO
2.178(4) A
bonding approaches to manganese are greater than 4 A.
˚
2.264(1) A
Final atomic positional coordinates, with estimated stan-
dard deviations, bond lengths and angles and anisotropic
thermal parameters are deposited.3 Selected interatomic
bond distances and angles, with standard deviations derived
from the refinement, are given in Table 2. The atomic
nomenclature is defined in Fig. 3. Figs. 3 and 4 illustrate
the molecular geometry viewed along, and perpꢀendicular
to, the normal to the C5 plane of the C5Ph5 ligand,
respectively.
˚
1.765(4) A
˚
C–O
1.158(4) A
˚
C–C (C5 ring)
C(C5)–ipso-C (C6)
CO–Mn–CO
Mn–C–O
1.427(5) A
˚
1.492(5) A
91.2(2)ꢁ
178.9(3)ꢁ
108.0(3)ꢁ
C–C–C (C5 ring)
for [Mn(g5–C5R5)(CO)2(PX3)] (X ¼ M, P) compounds.
As in (1), the C@O bond lengths in (2) are the same within
˚
The Mn–(C5 centroid) distance is 1.809 A and the
˚
average Mn–C(C5) distance of 2.178(4) A are both
˚
experimental error. The Mn–P distance of 2.264(1) A is
slightly longer than those of (1). The Mn–(C5 centroid)
distance is at the upper end of such distances in
[Mn(C5R5)(CO)3] compounds, with only 8 compounds
with longer Mn–(C5 centroid) distances. It is easily the
longest such distance in [Mn(g5-C5R5)(CO)2(PX3)]
(X ¼ M, P) compounds, as are, therefore, the Mn–
C(C5) distances.4 That is, on average, the Mn–cyclopenta-
dienyl distance of [Mn(C5R5)(CO)3] compounds contracts
on phosphine substitution, but has increased in this case.
again easily the longest in the 31 [Mn(g5-C5R5)(CO)2-
(PX3)] (X ¼ M, P) compounds reported to date (range
˚
2.126 6 d 6 2.359, mean 2.215 A). These distances easily
exceed those of the [Mn(g5-C5R5)(CO)2(PX3)]+ (X ¼ M,
P) cations, which are included in the above analyses.
The C5 nucleus of the pentaphenylcyclopentadienyl
˚
ligand is planar to within 0.006 A. The C–C bond lengths
of the C5 ring range from 1.415(5) to 1.432(5) A, the
average C–C bond length being 1.427(5) A, which is
consistent with C5Ph5 rings in other metal complexes.
The ipso-carbon atoms of each of the phenyl rings are
displaced slightly from the C5 plane by between 0.158
˚
˚
˚
The Mn–CO distances (1.757(4), 1.773(4) A) are indistin-
guishable and similar to those in (1), but relatively long
3
˚
˚
and 0.268 A, with the average distance being 0.202 A.
The atomic coordinates and thermal parameters have been submitted
This is a greater displacement than in [Mn(g5-C5Ph5)
as supplementary material for deposition at the Cambridge Crystallo-
graphic Data Centre with deposition number CCDC 284609 (see
Deposition of crystallographic data).
˚
(CO)3] (mean 0.161 A). The ipso-carbon atoms of all
phenyl rings are on the opposite side of the C5 plane
to the manganese atom. The phenyl rings are all also
planar to within a mean deviation from the plane of
4
41 observables from 31 compounds, ranges: Mn–(C5 centroid)
˚
1.744 6 d 6 1.791, mean 1.774 A; Mn–C(C5) 2.113 6 d 6 2.211, mean
2.140 A.
˚

1502
L.D. Field et al. / Polyhedron 25 (2006) 1498–1506
Fig. 3. ORTEP [39] illustration of [Mn(g5-C5Ph5)(CO)2PMe2Ph] (2), viewed perpendicular to the C5 ligand plane, showing the atom numbering scheme.
Fig. 4. ORTEP [39] illustration of [Mn(g5-C5Ph5)(CO)2PMe2Ph] (2), viewed parallel to the C5 ligand plane, showing the atom numbering scheme.
2.1.4. The [Mn(g5-C5Ph5)(CO)2(PMe2Ph)] molecule
A similar ‘‘three-legged piano stool’’ environment about
the manganese atom results from coordination by the g5-
pentaphenylcyclopentadienyl group, the two carbonyl
groups and the phosphine.
˚
0.012 A and are canted at between 47.06ꢁ and 56.87ꢁ
(average 51.5ꢁ) to the C5 ring in a paddle wheel, or pro-
peller, arrangement. The average angle between the
ꢀ
planes of adjacent phenyl rings of the C5Ph5 ligand is
52.6ꢁ.

L.D. Field et al. / Polyhedron 25 (2006) 1498–1506
1503
The most remarkable feature of the structure is the near
parallel orientations of the C5 and the PMe2Ph phenyl
planes. The angle between the phosphine phenyl and C5
planes is only 7.55ꢁ, a consequence of the need to minimise
intramolecular interferences between the phosphine methyl
and cyclopentadienyl phenyl substituents. This steric pres-
sure is also reflected in the slightly longer Mn-(C5 centroid)
distance in (2) by comparison with (1), and the consider-
ably longer dimensions than in all [Mn(g5-C5R5)
(CO)2(PX3)] (X ¼ M, P) compounds reported to date.
was obtained from Merck. Melting points were recorded
on a Reichert hot platform in air and are uncorrected.
1H and 13C NMR spectra were recorded on Bruker
AC200F (1H NMR 200.13 MHz, 13C NMR 50.33 MHz)
or Bruker AMX400 (1H NMR 400.21 MHz) spectrome-
ters. The spectra were referenced internally to TMS or to
1
residual solvent resonances (CD2Cl3, H d 5.30 ppm, 13C
d 53.8 ppm). 31P NMR spectra were recorded on a Bruker
AMX400 spectrometer operating at 162.0 MHz and were
referenced to external neat trimethylphosphite (taken to
be d 140.85 ppm from 85% H3PO4). Electron impact mass
spectra were obtained using a Kratos MS 9 geometry mass
spectrometer with a direct insertion probe, a 280 ꢁC source
temperature, 70 eV ionisation voltage and 4 kV accelera-
tion voltage.
Melting points were determined using a Gallenkamp
melting point apparatus in air and are uncorrected.
Elemental analyses were carried out by one of the fol-
lowing departments: the Australian Microanalytical Ser-
vice in Melbourne, the Microanalytical Service at the
Australian National University or at the JEAF in The Uni-
versity of Sydney.
ꢀ
With the chiral C5Ph5 phenyl propeller, and the chiral-
ity at the manganese atom, molecule (2) is diastereotopic in
the solid state. However, the phenyl propellers and the
phosphine substituents appear to rotate freely on the
NMR time-scale at room temperature in solution, as there
are no unexpected multiplicities in the 1H, 13C or 31P NMR
spectra, and, unlike [Ru(g5-C5Ph5)(CO)2(PPh3)Br] [26], the
diastereomers are not resolved in solution.
3. Conclusions
ꢀ
Manganese derivatives of C5Ar5 ligands are readily
accessible in high yield from easily available starting mate-
rials. The compounds [Mn(g5-C5Ar5)(CO)3] undergo
reversible one-electron oxidations at potentials more posi-
tive than the cyclopentadienyl analogue. Substitution of
one of the CO ligands by P- and C-donor ligands is conve-
niently effected via the intermediacy of Me3NO. The Mn-
(C5 centroid) distances of [Mn(g5-C5Ph5)(CO)3] and
[Mn(g5-C5Ph5)(CO)2(PMe2Ph)] are relatively long for
compounds of these types. In particular, the Mn-ligand dis-
tances in [Mn(g5-C5Ph5)(CO)2(PMe2Ph)] are considerably
longer than those reported for [Mn(g5-C5R5)(CO)2(PX3)]
(X ¼ M, P) compounds to date. Although the phosphine
and cyclopentadienyl phenyl substituents of [Mn(g5-
C5Ph5)(CO)2(PMe2Ph)] adopt configurations minimising
steric interferences in the solid state, these substituents
appear to be freely rotating (on the NMR time-scale) in
solution at room temperature.
Infrared spectra were recorded on a Digilab FTS-40
infrared spectrophotometer either as a solution or as a sus-
pension in a potassium bromide matrix. The resolution
used was 4 cmꢀ1 and the number of scans was 16.
Electrochemical experiments were performed using a
BAS 100W Electrochemical Analyser. The sample cell
was fitted with a plastic top that could be sealed almost
air-tight. A three-electrode configuration was used. The
working electrode was a BAS glassy carbon disc (3 mm
diameter) which was polished with an alumina (0.25 mm
grade) suspension on a velvet polishing cloth, then rinsed
with distilled water and acetone, and then dipped in the
solvent of use for one minute before each cyclic voltammo-
gram was recorded. The auxiliary electrode was a
platinum wire (0.5 mm diameter, 4 cm length) which did
not need cleaning between voltammograms but was
cleaned by rinsing with 5 M aqueous HNO3, distilled water
and the solvent used prior to each set of experiments. The
reference electrode was a BAS Ag/AgCl/NaCl (3 M) elec-
trode with a vycor frit. All samples were thoroughly
degassed with high purity argon (CIG) which was passed
through a scrubber and a drying tower to remove any
traces of oxygen and water, respectively. All potentials
are quoted relative to the ferrocenium/ferrocene redox
couple, which was observed at ca. 500 mV against the
reference electrode [Ag/AgCl/NaCl (3 M)]. Samples were
prepared as approximately 1 mM solutions with tetra-
(n-butyl)ammonium tetrafluoroborate (TBATFB ca. 0.1 M)
as the supporting electrolyte. TBATFB was prepared either
from aqueous HBF4 (35% v/v, Merck) and (n-C4H9)NOH
(40% w/w, Aldrich) or by recrystallisation of TBATFB
(Aldrich) three times from ethyl acetate/diethyl ether in
air, then drying in vacuo. Full iR (100%) compensation
was employed in each scan. Except as otherwise stated,
all measurements were carried out in dichloromethane
4. Experimental
All manipulations were performed at atmospheric pres-
sure under an atmosphere of dinitrogen by using conven-
tional Schlenk techniques. Chlorobenzene (Merck) was
dried, distilled and stored over calcium hydride. Tetrahy-
drofuran (thf) (Merck) was pre-dried over sodium wire
and distilled from sodium benzophenone ketyl. n-Hexane
(BDH) was distilled from sodium wire. Dichloromethane
(Ajax) was distilled from calcium hydride. C5Ph5Br [27],
and C5(p-MeC6H4)5Br [28] were prepared as described pre-
viously. Mn2(CO)10 (Aldrich), PMe2Ph (Fluka) and tBuNC
(Aldrich) were used as received. [Mn(CO)5Br],
[Mn(CO)3(CH3CN)Br]2 and [Mn(CO)3(CH3CN)3]Br were
prepared according to literature procedures [29–31]. Tri-
methylamine N-oxide dihydrate (Aldrich) was dehydrated
by sublimation prior to use. Flash silica (240–400 mesh)

1504
L.D. Field et al. / Polyhedron 25 (2006) 1498–1506
solution with 0.1 M TBATFB as the supporting electrolyte
and a scan rate of 100 mV sꢀ1
31P{1H} NMR (CDCl3): d 45.86 (s) ppm. I.R. (KBr): m
1850 (vs), 1918 (vs) cmꢀ1
.
.
4.1. Preparation of [Mn(g5-C5Ph5)(CO)3] (1)
4.3. Preparation of [Mn(g5-C5(p-tol)5)(CO)3] (3)
4.1.1. Method A: reaction of [Mn(CO)3(CH3CN)3]Br
with LiC5Ph5
4.3.1. Method A: reaction of [Mn(CO)3(CH3CN)3]Br
with LiC5(p-tol)5
A solution of LiC5Ph5 (1.0 mmol) in tetrahydrofuran
(10 mL) was added to a solution of [Mn(CO)3(CH3CN)3]Br
(0.34 g, 1.0 mmol) in tetrahydrofuran (10 mL). The solution
was refluxed with stirring for 16 h, then the solvent was
removed under reduced pressure. The orange/yellow solid
was extracted into CH2Cl2, and chromatographed on a
hexane/alumina column under N2. Elution with CH2Cl2/
hexane (1:3) gave a light yellow band which was collected.
The solvent was removed under vacuum to yield [Mn-
(g5-C5Ph5)(CO)3] (1) as a pale yellow solid (0.40 g, 68%).
A solution of LiC5(p-tol)5 (0.87 mmol) in tetrahydrofu-
ran (10 mL) was added to a solution of [Mn(CO)3-
(CH3CN)3]Br (0.30 g, 0.86 mmol) in tetrahydrofuran
(10 mL). The resultant solution was refluxed with stirring
for 16 h, then the solvent was removed under reduced pres-
sure. The orange/yellow solid was extracted into CH2Cl2,
and chromatographed on a hexane/alumina column under
N2. Elution with CH2Cl2/hexane (1:5) gave a light yellow
band which was collected. The solvent was then removed
to yield [Mn(g5-C5(p-tol)5)(CO)3] (3) as a pale yellow solid
(0.19 g, 34%).
4.1.2. Method B: reaction of [Mn(CO)5Br] with LiC5Ph5
A solution of LiC5Ph5 (1.0 mmol) in tetrahydrofuran
(10 mL) was added to a solution of [Mn(CO)5Br] (0.27 g,
1.0 mmol) in tetrahydrofuran (10 mL) and the resultant
solution was refluxed with stirring for 16 h. The product
was purified chromatographically as in Method A (0.22 g,
38%).
4.3.2. Method B: reaction of [Mn(CO)3(CH3CN)3]Br with
C5(p-tol)5Br and Zn
Tetrahydrofuran (15 mL) was added to a solid mixture
of [Mn(CO)3(CH3CN)3]Br (0.34 g, 1.0 mmol), C5(p-tol)5Br
(0.59 g, 1.0 mmol) and zinc dust (0.075 g, 1.1 mmol). The
resultant solution was refluxed with stirring for 16 h. The
product was purified chromatographically as in Method
A (0.45 g, 69%).
4.1.3. Method C: reaction of [Mn(CO)3(CH3CN)2Br]
with C5Ph5Br and Zn
M.p. 255–257 ꢁC. Anal. Calc. for C43H35MnO3: C,
1
A
mixture of [Mn(CO)3(CH3CN)2Br] (0.15 g,
78.89; H, 5.39. Found: C, 78.1; H, 5.2%. H NMR (ppm,
0.50 mmol), C5Ph5Br (0.26 g, 0.50 mmol) and zinc dust
(0.033 g, 0.50 mmol) in tetrahydrofuran (15 mL) was
refluxed with stirring for 16 h. The product was purified
chromatographically as in Method A (0.17 g, 59%). M.p.
287–289 ꢁC. Anal. Calc. for C38H25MnO3: C, 78.08; H,
200 MHz, CD2Cl2): d 7.09–6.83 (m, 20H, Ar-H) 2.26–
2.17 (m, 15H, CH3). IR mmax (cmꢀ1, KBr): 3020 w, 2922
w, 2008 vs, 1932 vs, 1926 vs, 1520 m, 1187 w, 1020 w,
838 w, 820 w, 805 w, 728 w, 666 m, 630 m, 538 m sh.
MS (EI) (m/z%): 654 (M+, 4.4), 598 (M+ ꢀ 2CO, 2), 570
(M+ ꢀ 3CO, 100), 526 (C5Ph5H+, 34).
1
4.31. Found: C, 78.0; H, 4.3%. H NMR (ppm, 200 MHz,
CD2Cl2): d 7.06 (br, m, 25 H, Ar-H). IR mmax (cmꢀ1, KBr):
3060 w, 2007 vs, 1934 vs, 1923 vs, 1503 w, 1445 w, 746 w,
4.4. Structure determinations
739 w, 699 m sh, 665 m, 636 m, 559 w, mmax (CO) (cmꢀ1
,
CH2Cl2): 2015 s, 1934 s. MS (EI) (m/z%): 584 (M+, 12),
528 (M+ ꢀ 2CO, 17), 500 (M+ ꢀ 3CO, 100), 446
(C5Ph5H+, 33).
[Mn(g5-C5Ph5)(CO)3] (1): a yellow blade like crystal
having approximate dimensions of 0.47 · 0.20 · 0.12 mm
was attached to a thin glass fibre, and mounted on an
Enraf-Nonius CAD4 diffractometer employing graphite
monochromated Mo Ka radiation. Primitive monoclinic
cell constants were obtained from a least-squares refine-
ment using the setting angles of 25 reflections in the range
16.02 < 2h < 24.24ꢁ. Diffraction data were collected at a
temperature of 21 1 ꢁC using x scans to a maximum 2h
value of 49.9ꢁ. The intensities of three representative reflec-
tions measured every 60 min, decreased by 53.2% and a
polynomial correction factor was accordingly applied to
the data. An analytical absorption correction was applied
and the data were also corrected for Lorentz and polarisa-
tion effects.
4.2. Preparation of [Mn(g5-C5Ph5)(CO)2(PMe2Ph)] (2)
PPhMe2 (0.22 mL, 0.70 mmol) and trimethylamine oxide
(0.045 g, 0.60 mmol) were added to
a solution of
[Mn(C5Ph5)(CO)3] (0.295 g, 0.51 mmol) in tetrahydrofuran
(30 mL) and the resulting reaction mixture was heated at
reflux for 16 h. The solvent was removed in vacuo to afford
an orange oil and the crude product was purified by column
chromatography on alumina eluting with CH2Cl2/hexane
(20/80 v/v). The product was recrystallised from CH2Cl2/
hexane to give [Mn(g5-C5Ph5)(CO)2(PMe2Ph)] (2) as an
orange crystalline solid (0.285 g, 81%). M.p. 287–289 ꢁC.
Anal. Calc. for C45H36MnOP: C, 77.80; H, 5.22. Found: C,
All calculations were undertaken with the TEXSAN [32]
crystallographic software package. Neutral atom scattering
factors were taken from Cromer and Waber [33]. Anoma-
lous dispersion effects were included in Fcalc [34] and the
1
77.4; H, 4.9%. H NMR (CDCl3): d 7.55–6.75 (br m, 30H,
(C6H5)C5 and P(C6H5)), 1.59 (br s, 6H, 2 · PCH3) ppm.

L.D. Field et al. / Polyhedron 25 (2006) 1498–1506
1505
values for Df0 and Df00 were those of Creagh and McAuley
[35]. The values for the mass attenuation coefficients were
those of Creagh and Hubbell [36]. The structure was solved
in the space group P21/n (#14) by heavy-atom Patterson
methods [37] and expanded using Fourier techniques [38].
In general non-hydrogen atoms were modelled anisotropi-
cally, however C(3), C(5), C(17), C(18), C(30) and C(35)
were refined with isotropic thermal parameters. Hydrogen
atoms were included in the model at calculated positions
with group thermal parameters. ORTEP projections of
the molecule are provided in Figs. 2 and 3 [39].
in the space group P21/n (#14) by direct methods [40] and
expanded using Fourier techniques [38]. Non-hydrogen
atoms were modeled with anisotropic thermal parameters
and hydrogen atoms were included in the model at calcu-
lated positions with group thermal parameters. ORTEP
projections of the molecule are provided in Figs. 1 and 2
[39].
4.6. Crystal data for [Mn(g5-C5Ph5)(CO)2(PMe2Ph)]
(2)
Formula C45H33MnO2P, M 691.62, monoclinic, space
4.5. Crystal data for [Mn(g5-C5Ph5)(CO)3] (1)
group P21/n(#14), a 15.251(4), b 13.804(3), c 17.682(4) A,
˚
3
b 106.39(2), V 3571.2(15) A , Dc 1.286 g cmꢀ3, Z 4, crystal
˚
Formula C38H25MnO3, M 584.52, monoclinic, space
size 0.28 by 0.25 by 0.11 mm, colour yellow, habit blade,
˚
˚
group P21/n (#14), a 16.929(4), b 8.928(2), c 20.054(4) A, b
temperature 294(2) K, k(Cu Ka) 1.5418 A, l(Cu Ka)
3
106.04(2), V 2913.0(11) A , Dc 1.333 g cmꢀ3, Z 4, crystal size
3.714 mmꢀ1
, T(Analytical)min,max 0.420, 0.698, 2hmax
˚
0.47 by 0.20 by 0.12 mm, colour yellow, habit blade, temper-
130.20, hkl range 0 17, 0 16, ꢀ20 19, N 6028, Nind 6027
(Rmerge 0.05469), Nobs 4016(I > 2r(I)), Nvar 443, residuals6
R1(F) 0.0472, wR2ꢀ(F32) 0.1356, GoF(all) 1.189, Dqmin,max
˚
ature 294(2) K, k(Mo Ka) 0.71073 A, l(Mo Ka)
0.4900 mmꢀ1, T (Analytical)min,max 0.8984, 0.9489, 2hmax
49.98, hkl range ꢀ20 19, ꢀ1 10, ꢀ1 10, N 4061, Nind 3167-
(Rmerge 0.0738), Nobs 2058(I > 2r(I)), Nvar 344, residuals5
R1(F) 0.0967, wR2ꢀ(F32) 0.2971, GoF(all) 1.274, Dqmin,max
ꢀ
˚
ꢀ0.383, 0.495 e A
.
Acknowledgement
ꢀ
˚
ꢀ1.516, 1.455 e A
.
[Mn(g5-C5Ph5)(CO)2(PMe2Ph)] (2): a yellow blade like
crystal having approximate dimensions of 0.28 · 0.25 ·
0.11 mm was attached to a thin glass fibre, and mounted
on a Rigaku AFC7R diffractometer employing graphite
monochromated Cu Ka radiation from a rotating anode
generator. Primitive monoclinic cell constants were
obtained from a least-squares refinement using the setting
angles of 21 reflections in the range 68.01 < 2h < 91.99ꢁ.
Diffraction data were collected at a temperature of
21 1 ꢁC using x–2h scans to a maximum 2h value of
130.2ꢁ. Omega scans of several intense reflections made
prior to data collection, had an average width at half-
height of 0.18ꢁ, and scans of (1.73 + 0.35 tan h)ꢁ were made
at a speed of 16.0ꢁ/min (in omega). The weak reflections
(I < 15.0r(I)) were rescanned up to 10 times. Stationary
background counts were recorded on each side of the
reflection, with a 2:1 ratio of peak to background counting
time. The intensities of three representative reflections
measured every 150 reflections, did not change significantly
during the data collection. An analytical absorption correc-
tion was applied and the data were also corrected for
Lorentz and polarisation effects.
We thank the Australian Research Council for financial
support.
References
[1] K. Ziegler, B. Schnell, Justus Liebigs Ann. Chem. 445 (1925) 266.
[2] K. Ziegler, F.A. Fries, Ber. 59B (1926) 242.
[3] W. Hubel, R. Merenyi, J. Organomet. Chem. 2 (1964) 213.
¨
[4] T.S. Piper, F.A. Cotton, G. Wilkinson, J. Inorg. Nucl. Chem. 1
(1955) 165.
[5] A.F. Berndt, R.E. Marsh, Acta Crystallogr. 16 (1963) 118.
[6] P.J. Fitzpatrick, Y. Le Page, J. Sedman, I.S. Butler, Inorg. Chem. 20
(1981) 2852.
[7] J. Cowie, E.J.M. Hamilton, J.C.V. Laurie, A.J. Welch, J. Organomet.
Chem. 394 (1990) 1.
[8] E.O. Fischer, R. Jira, Z. Naturforsch. 9B (1954) 618.
[9] M.D. Rausch, W.-M. Tsai, J.W. Chambers, R. Rogers, H.G. Alt,
Organometallics 8 (1989) 816.
[10] R.B. King, A. Efraty, J. Am. Chem. Soc. 94 (1972) 3773.
[11] R.B. King, M.Z. Iqbal, A.D. King, J. Organomet. Chem. 171 (1979) 53.
[12] I. Bernal, J.D. Korp, W.A. Herrmann, R. Serrano, Chem. Ber. 117
(1984) 434.
[13] G. Wilkinson, F.G.A. Stone, E.W. Abel (Eds.), Comprehensive
Organometallic Chemistry, vol. 4, Pergamon, 1982, p. 1.
[14] R.B. King, A. Efraty, W.M. Douglas, J. Organomet. Chem. 56 (1973)
345.
All calculations were undertaken with the TEXSAN [32]
crystallographic software package. Neutral atom scattering
factors were taken from Cromer and Waber [33]. Anoma-
lous dispersion effects were included in Fcalc [34] and the
values for Df0 and Df00 were those of Creagh and McAuley
[35]. The values for the mass attenuation coefficients were
those of Creagh and Hubbell [36]. The structure was solved
[15] G. Wilkinson, F.G.A. Stone, E.W. Abel (Eds.), Comprehensive
Organometallic Chemistry, vol. 6, Pergamon, 1995, p. 109.
[16] F.P.A. Johnson, M.W. George, V.N. Bagratashvili, L.N. Vereshcha-
gina, M. Poliakoff, Mendeleev Commun. (1991) 26.
[17] N.G. Connelly, I. Manners, J. Chem. Soc., Dalton Trans. (1989)
283.
[18] L.D. Field, T.W. Hambley, P.A. Humphrey, A.F. Masters, P.
Turner, Polyhedron 17 (1998) 2587.
P
P
P
P
P
P
2
2
5
6
R1 ¼ kF oj ꢀ jF ck= jF oj for F o > 2rðF oÞ; wR2 ¼ ð wðF 2o ꢀ F c2Þ =
R1 ¼ kF oj ꢀ jF ck= jF oj for F o > 2rðF oÞ; wR2 ¼ ð wðF 2o ꢀ F c2Þ =
P
P
2
1=2
2
2
1=2
2
ðwF 2c Þ Þ all reflections w ¼ 1=½r2ðF 2oÞ þ ð0:10PÞ þ 15:0Pꢁ where P ¼
ðwF 2cÞ Þ all reflections w ¼ 1=½r2ðF 2oÞ þ ð0:05PÞ þ 1:0Pꢁ where P ¼
ðF 2o þ 2F 2c Þ=3.
ðF 2o þ 2F 2c Þ=3.

1506
L.D. Field et al. / Polyhedron 25 (2006) 1498–1506
[19] C.J. Pickett, D. Pletcher, J. Chem. Soc., Dalton Trans. (1976)
636.
[20] K. Broadley, G.A. Lane, N.G. Connelly, W.E. Geiger, J. Am. Chem.
Soc. 105 (1983) 2486.
[21] N.G. Connelly, W.E. Geiger, G.A. Lane, S.J. Raven, P.H. Rieger, J.
Am. Chem. Soc. 108 (1986) 6219.
[32] Molecular Structure Corporation, TEXSAN: Single Crystal Structure
Analysis Software, Version 1.6, Molecular Structure Corporation,
The Woodlands, TX, USA, 1993.
[33] D.T. Cromer, J.T. Waber, in: J.A. Ibers, W.C. Hamilton (Eds.),
International Tables for X-Ray Crystallography, vol. IV, Kynock
Press, Birmingham, England, 1974, p. 71.
[22] S.M. Sawtelle, R.F. Johnston, C.C. Cook, Inorg. Chim. Acta 221
(1994) 85.
[23] T. Wurminghausen, D. Sellmann, J. Organomet. Chem. 199 (1980)
77.
[34] J.A. Ibers, W.C. Hamilton, Acta Crystallogr. 17 (1964) 781.
[35] D.C. Creagh, W.J. McAuley, in: A.J.C. Wilson (Ed.), International
Tables for Crystallography, vol. C, Kluwer Academic Publishers,
Boston, 1992, p. 219.
[24] J.E. Ellis, R.A. Faltynek, S.G. Hentges, J. Organomet. Chem. 120
(1976) 389.
[36] D.C. Creagh, J.H. Hubbell, in: A.J.C. Wilson (Ed.), International
Tables for Crystallography, vol. C, Kluwer Academic Publishers,
Boston, 1992, p. 200.
[25] N.V. Podberezskaya, V.A. Maksakov, I.V. Slovokhotova, S.P.
Babailov, A.V. Virovets, V.P. Kirin, J. Struct. Chem. (Translation
of Zhurnal Strukturnoi Khimii) (2000) 40 (1999) 914.
[26] H. Adams, N.A. Bailey, A.F. Browning, J.A. Ramsden, C. White, J.
Organometal. Chem. 387 (1990) 305.
[27] L.D. Field, K.M. Ho, C.M. Lindall, A.F. Masters, A.G. Webb,
Aust. J. Chem. 43 (1990) 281.
[28] L.D. Field, A.F. Masters, M. Gibson, D.R. Latimer, T.W. Hambley,
I.E. Buys, Inorg. Chem. 32 (1993) 211.
[37] P.T. Beurskens, G. Admiraal, G. Beurskens, W.P. Bosman, S. Garcia-
Granda, R.O. Gould, J.M.M. Smits, C. Smykalla, PATTY The DIRDIF
program system. Technical Report of the Crystallography Laboratory,
University of Nijmegen, The Netherlands, Version Nijmegen, 1992.
[38] P.T. Beurskens, G. Admiraal, G. Beurskens, W.P. Bosman, R. de
Gelder, R. Israel, J.M.M. Smits, The DIRDIF-94 Program System.
Technical Report. Crystallography Laboratory, University of Nijme-
gen, The Netherlands, DIRDIF-94, Nijmegen, 1994.
[29] M.H. Quick, R.J. Angelici, in: D.F. Shriver (Ed.), Inorganic
Syntheses, vol. 19, Wiley, New York, 1979, p. 160.
[30] J.G. Dunn, D.A. Edwards, J. Organomet. Chem. 27 (1971) 73.
[31] R.H. Reimann, E. Singleton, J. Chem. Soc., Dalton Trans. (1974)
808.
[39] C.K. Johnson, ORTEPII, a Thermal Ellipsoid Plotting Program, Report
ORNL-5138, Version Oak Ridge National Laboratories, Oak Ridge
TN, 1976.
[40] A. Altomare, M. Cascarano, C. Giacovazzo, A. Guagliardi, J. Appl.
Crystallogr. 26 (1993) 343.