Solid-state isomerisation reactions of (η5 -C5H4R)M(CO)2 (PR3′) I (M=W, Mo; R=tBu, Me; R′=Ph, Oi Pr3)

Article abstract of DOI:10.1016/j.jorganchem.2004.04.009

The cis and trans monosubstituted cyclopentadienyl tungsten and molybdenum complexes (η⁵-C₅ H₄R)M(CO)₂(L)I (1) (M=W, R=Me, ^tBu, L=P(OⁱPr)₃, PPh₃; M=Mo, R=Me, L=PPh₃) have been synthesised and fully characterised by elemental analysis and IR and NMR spectroscopy. It was found that 1 nderwent a thermal solid-state ligand isomerisation reaction and that the favoured direction of the isomerisation reaction is related to the melting points of the cis and trans isomers, i.e., with intermolecular forces in the solid state. No obvious relationship between the melting point and the metal, the ring-substituent or the ligand was observed. Crystal structure determinations of the cis and trans isomers of (η⁵-C₅H₄Me)W(CO)₂ (PPh₃)I reveal that a limited amount of isomer conversion can be accommodated in the unit cell of the trans isomer, prior to crystal fragmentation. The rearrangement of the molecules within the unit cell, during isomerisation, also leads to disorder in the crystal.

Full text of DOI:10.1016/j.jorganchem.2004.04.009

Journal of Organometallic Chemistry 689 (2004) 2207–2215
www.elsevier.com/locate/jorganchem
0
Solid-state isomerisation reactions of (g5-C5H4R)M(CO)2(PR3 )I
(M ¼ W, Mo; R ¼ tBu, Me; R0 ¼ Ph, OiPr3)
Olalere G. Adeyemi, Uche B. Eke, Lin Cheng, Leanne M. Cook, David G. Billing,
*
Bhekie B. Mamba, Demetrius C. Levendis, Neil J. Coville
School of Chemistry, Molecular Sciences Institute, University of the Witwatersrand, 1 Jan Smuts Ave., Private Bag 3,
WITS 2050 Johannesburg, South Africa
Received 21 February 2004; accepted 6 April 2004
Abstract
The cis and trans monosubstituted cyclopentadienyl tungsten and molybdenum complexes (g5-C5H4R)M(CO)2(L)I (1) (M ¼ W,
R ¼ Me, tBu, L ¼ P(OiPr)3, PPh3; M ¼ Mo, R ¼ Me, L ¼ PPh3) have been synthesised and fully characterised by elemental analysis
and IR and NMR spectroscopy. It was found that 1 underwent a thermal solid-state ligand isomerisation reaction and that the
favoured direction of the isomerisation reaction is related to the melting points of the cis and trans isomers, i.e., with intermolecular
forces in the solid state. No obvious relationship between the melting point and the metal, the ring-substituent or the ligand was
observed. Crystal structure determinations of the cis and trans isomers of (g5-C5H4Me)W(CO)2(PPh3)I reveal that a limited amount
of isomer conversion can be accommodated in the unit cell of the trans isomer, prior to crystal fragmentation. The rearrangement of
the molecules within the unit cell, during isomerisation, also leads to disorder in the crystal.
Ó 2004 Elsevier B.V. All rights reserved.
Keywords: Solid state; Substitution reaction; Molybdenum; Tungsten
1. Introduction
separation and study of cis and trans (lat and diag)
isomers of (g5-C5H5)M(CO)2LX (M ¼ Cr, Mo, W;
X ¼ halide) and (g5-C5H5)Re(CO)2Br2 have provided a
frame-work for many later studies [3]. With few excep-
tions these studies have been performed in the solution
state.
Recently we have undertaken a study of the isom-
erisation reactions of a range of (g5-C5H5)ML4 com-
plexes in the solid state [4]. Most of our earlier work
focussed on Re complexes, where we observed that so-
lution and solid-state isomerisation reactions could be
different [4]. Indeed studies in the solid state revealed a
possible alternative isomerisation reaction pathway to
those found in solution.
We have now extended our studies to (g5-C5H5)ML4
(M ¼ Mo, W) type complexes. In an earlier study we
noted that (g5-C5H4Me)Mo(CO)2[P(OiPr)3]I underwent
a bidirectional cis/trans isomerisation reaction in the
solid state that favoured the trans isomer (trans:cis ¼ 70/
30) [5]. Herein we report an extension of this earlier
study to the solid-state cis/trans isomerisation reactions
Isomerisation reactions in organometallic chemistry
have a long and rich history [1]. Many systems have
been studied over the last five decades and mechanisms
entailing ligand rearrangements around a central metal
atom have been described. Isomerisation at a ligand that
is attached to a metal in an organometallic complex has
also been reported [2]. In general as the number of li-
gands around a central metal atom increases the ease of
ligand rearrangement around the metal increases, i.e.,
the activation energy for the rearrangement (isomerisa-
tion) reaction decreases. Consequently, isomerisation
reactions of psuedo-seven coordinate complexes of the
type (g5-C5H5)ML4 (g5-C5H5 occupies three sites)
generally require modest energy input and have been
well studied. For example, early pioneering work on the
* Corresponding author. Tel.: +27-11-717-6738; fax: +27-11-717-
6749.
E-mail address: ncoville@aurum.chem.wits.ac.za (N.J. Coville).
0022-328X/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.04.009

2208
O.G. Adeyemi et al. / Journal of Organometallic Chemistry 689 (2004) 2207–2215
2.1. Preparation of trans- and cis-(g5-C5H4Me)-
M(CO)2(PPh3)I (M ¼ W, Mo) and trans- and cis-(g5-
C5H4 Bu)W(CO)2 (PR3)I (R ¼ Ph, OiPr)
t
A mixture of the cis and trans isomers were prepared
in good yield (55–75%) by reaction of (g5-
C5H4R)M(CO)3I with a 5–10-fold excess of ligand L
and trimethylamine N-oxide in dichloromethane at
room temperature [9]. The reaction proceeded slowly
when only a stoichiometric amount of ligand or tri-
methylamine N-oxide was used. Isomer separation was
achieved by dissolving the crude material in CH2Cl2
followed by mixing with a small quantity of silica gel.
The yellow powder remaining after removal of CH2Cl2
was chromatographed on a silica gel column (2 ꢀ 60)
with a 1:2 benzene/hexane mixture. For example (g5-
C5H4Me)Mo(CO)2(PPh3)I was obtained in 58% total
yield (cis 32%, trans 26%). The pure isomers were char-
acterised by mp, CHN analysis, IR and 1H NMR
spectroscopy (Tables 1 and 2) [10]. The complexes
were stable at room temperature in the solid state and
had good solubility in organic solvents, including
hexane.
Fig. 1. The solid-state cis–trans isomerisation reactions of (g5-
C5H4R)M(CO)2(PR03)I.
of (g5-C5H4Me)M(CO)2(PPh3)I (M ¼ Mo, W) and (g5-
t
C5H4R)W(CO)2[P(OiPr)3]I (R ¼ Me, Bu) in order to
evaluate the effect of ring substituents, metals and li-
gands on the solid-state reaction (Fig. 1). To our
knowledge only two other reports on solid-state isom-
erisation reactions of similar Mo complexes have ap-
peared in the literature [6,7].
2. Experimental
(g5-C5H4Me)W(CO)3I and (g5-C5H4Me)Mo(CO)3I
were prepared by the standard procedures used to syn-
thesise other ring-substituted analogues [8]. Trimethyl-
amine N-oxide dihydrate (Aldrich) was used as received.
All reactions were carried out using standard Schlenk
techniques under nitrogen. Solvents were dried by con-
ventional methods, distilled under nitrogen and used
immediately. Melting points were recorded on a Kofler
hot stage melting point apparatus. Infrared spectra were
measured on a Midac FTIR spectrometer, usually in
KBr cells (solutions). DSC profiles were measured on a
Du Pont 911 DSC instrument and TGA data were
measured on a Perkin–Elmer Pyris 1 thermal analyser.
Powder X-ray diffraction data were collected on a
Phillips PW 1710 diffractometer using a Cu tube anode.
NMR spectra were measured on a Bruker AC200
spectrometer operating at 200 MHz. Microanalysis was
carried out at the CSIR, Pretoria, South Africa.
2.2. Solid-state isomerisation of trans- and cis-(g5-C5
H4Me)M(CO)2(PPh3)I (M ¼ Mo, W)
About 5–10 mg of solid cis- or trans-(g5-C5H4Me)-
M(CO)2(PPh3)I was placed in a NMR tube under ni-
trogen and heated while totally immersed in an oil bath
for 1–15 h at a temperature at least 15 °C below the
lower melting point of the two isomers. The color and
shape of the starting materials showed no visible change
and no obvious decomposition during the heating pro-
cess. The identification of the cis and trans products was
achieved by means of TLC, and solution IR and NMR
spectroscopy on the products after the reaction.
The isomerisation direction in the solid state is given in
Table 1.
Table 1
0
Solid-state isomerisation reactions of cis and trans-(g5-C5H4R)M(CO)2(PR3 )I complexes
Complex
m.p.a (°C)
>141b
>149b
Reaction temperature (°C)
100–120
Isomerisation directionc
trans ! cis
cis-(g5-C5H4Me)Mo(CO)2PPh3I
trans-(g5-C5H4Me)Mo(CO)2PPh3I
cis-(g5-C5H4Me)W(CO)2[P(OiPr)3]I
trans-(g5-C5H4Me)W(CO)2[P(OiPr)3]I
cis-(g5-C5H4Me)W(CO)2PPh3I
trans-(g5-C5H4Me)W(CO)2PPh3I
cis-(g5-C5H4t Bu)W(CO)2[P(OiPr)3]I
trans-(g5-C5H4t Bu)W(CO)2[P(OiPr)3]I
128–130
144–146
179–181
177–179
135–137
127–129
113–115
cis ! trans
150–155
trans ! cis
100–105
trans ! cis
a Melting point determined by melting point apparatus.
b Darkened.
c Determined by 1H NMR spectroscopy.

O.G. Adeyemi et al. / Journal of Organometallic Chemistry 689 (2004) 2207–2215
2209
2.3. Powder XRD study
The trans isomer (40 mg) was packed into an NMR
tube and immersed completely into an oil bath (120 °C)
for 80 min. Powder diffraction data were then recorded
on the solid mixture. Powder diffraction data were also
recorded on pure cis and trans isomers.
2.4. Single crystal XRD study of cis- and trans-(g5-
C5H4Me)W(CO)2(PPh3)I
The cis and trans isomers of (g5-C5H4Me)-
W(CO)2(PPh3)I complexes were synthesised as
described above and recrystallised from a mixture
of hexane/dichloromethane under nitrogen at room
temperature.
For both structures, intensity data were collected on a
Bruker SMART 1K CCD area detector diffractometer
with graphite monochromated Mo Ka radiation (50 kV,
30 mA). The collection method involved x-scans of
width 0.3°. Data reduction was carried out using the
program SAINT+ [11] and absorption corrections were
made using the program SADABS [11]. The structures
were solved by standard Patterson procedures and re-
fined by least-squares methods based on F 2. The SHELX
[12] suite of programs, as incorporated into WINGX [13],
were used for all crystallographic computations. In the
final stages of refinement hydrogen atoms for both cis
and trans isomers were placed in geometrically calcu-
lated positions and refined in riding mode with dis-
placement parameters linked to those of their connected
carbon atoms.
Numerical data pertaining to the experimental mea-
surement and details of the structure analyses are given
in Table 3.
Significant disorder was observed in the trans isomer.
Both the cyclopentadienyl methyl group and the car-
bonyl/iodine positions are disordered, but not to the
same extent. The cyclopentadienyl methyl group is dis-
ordered over two positions. In the refinement, the oc-
cupancy of the two sites was initially allowed to vary,
converging to a site occupancy factor of ꢁ0.49:0.51 for
each site. This was fixed to 0.50 for both sites in the final
cycles of the refinement. The iodine was also disordered,
with about 11% of the molecules in the crystal having
Iodine (I1b) occupying the position of the trans-car-
bonyl site (C(1)O(1)). The corresponding carbonyl in
the I1a site (11% cis) could not be unambiguously
identified.
The crystal data (Table 3) for both isomers are, to a
very large extent, similar. Further, both structures have
a monoclinic space group. The atomic numbering
scheme for the trans isomer and the cis isomer are shown
in the ORTEP diagrams [14] (Figs. 2 and 3). The
packing diagrams (line drawings) of both isomers,
viewed down the a-axis, are shown in Figs. 4 and 5.

2210
O.G. Adeyemi et al. / Journal of Organometallic Chemistry 689 (2004) 2207–2215
Table 3
Crystal data and structure refinement for the cis and trans isomers of (g5-C5H4Me)W(CO)2(PPh3)I
trans isomer
cis isomer
Empirical formula
Formula weight
Temperature (K)
C26H22IO2PW
708.16
296(2)
C26H22IO2PW
708.16
296(2)
ꢀ
Wavelength (A)
0.71073
Monoclinic, P21=n
0.71073
Monoclinic, P21=n
Crystal system, space group
Unit cell dimensions
ꢀ
ꢀ
a ¼ 10:4696ð5Þ A, a ¼ 90°
a ¼ 9:0325ð5Þ A, a ¼ 90°
ꢀ
ꢀ
b ¼ 16:0379ð9Þ A, b ¼ 94:493ð1Þ°
b ¼ 15:0013ð8Þ A, b ¼ 99:802ð1Þ°
ꢀ
ꢀ
c ¼ 14:6428ð8Þ A, c ¼ 90°
c ¼ 17:9727ð10Þ A, c ¼ 90°
3
ꢀ
Volume (A )
2451.1(2)
4, 1.919
2399.7(2)
4, 1.960
Z, Dcalc (mg mꢂ3
)
Absorption coefficient (mmꢂ1
F ð000Þ
)
6.057
1344
6.187
1344
Crystal size
h range for data collection (°)
Index ranges
0.36 mm ꢀ 0.22 mm ꢀ 0.20 mm
1.89–28.28
0.70 mm ꢀ 0.40 mm ꢀ 0.18 mm
1.78–28.25
)10 6 h 6 13, )16 6 k 6 21, )18 6 l 6 19
14232/5451 ½Rint ¼ 0:0434]
99.3
Full-matrix least-squares on F 2
5451/12/290
1.138
)11 6 h 6 11, )19 6 k 6 19, )23 6 l 20
14750/5438 [Rint ¼ 0:0271]
99.6
Full-matrix least-squares on F 2
5438/0/269
1.188
Reflections collected/unique
Completeness to 2h ¼ 25° (%)
Refinement method
Data/restraints/parameters
Goodness-of-fit on F 2
Final R indices [I > 2rðIÞ]
R1 ¼ 0:0776, wR2 ¼ 0:1546
R1 ¼ 0:1091, wR2 ¼ 0:1670
3.222 and )2.140
R1 ¼ 0:0281, wR2 ¼ 0:0648
R1 ¼ 0:0316, wR2 ¼ 0:0665
0.617 and )1.309
R indices (all data)
Largest difference peak and hole (e Aꢂ3
)
Fig. 2. An ORTEP plot of trans-(g5-C5H4Me)W(CO)2(PPh3)I.
Fig. 3. An ORTEP plot of cis-(g5-C5H4Me)W(CO)2(PPh3)I.
ꢀ
The complexes were characterised by elemental analy-
ses, and IR and NMR spectroscopy. The cis and trans
isomers could readily be identified by NMR spectros-
copy [3a,10] and were also characterised by single crystal
X-ray crystallography which confirmed the spectro-
scopic results (see below).
Selected bond lengths (A) and bond angles (°) are given
in Table 4.
Crystallographic data were deposited with the Cam-
bridge Data Base (Ref. # CCDC 231709 and 231710)
3. Results and discussion
3.2. Isomerisation direction and melting points
3.1. Synthesis
The isomerisation reactions for the new complexes
were studied and the general trend was for the isom-
erisation reaction to occur from the trans to the cis
The synthetic strategy used to make the cis and trans
Mo and W complexes followed standard procedures.

O.G. Adeyemi et al. / Journal of Organometallic Chemistry 689 (2004) 2207–2215
2211
dependent on electronic factors associated with ligand
orientation effects.
3.3. Heating studies on (g5-C5H4Me)Mo(CO)2(PPh3)I
TGA profiles for the cis and trans isomers are shown
in Fig. 6. The data reveal that significant mass loss only
occurs above 140 °C. 31P NMR spectra recorded on the
samples heated above 100 °C revealed the presence of
PPh3O (d ¼ 29:84 ppm) in the samples.
The DSC profiles for cis-(g5-C5H4Me)Mo-
(CO)2(PPh3)I at different heating rates are shown in
Fig. 7. Exotherms can be seen at T > 120 °C. Similar
profiles for trans-(g5-C5H4Me)Mo(CO)2(PPh3)I, at dif-
ferent heating rates, are shown in Fig. 8, but here a
further broad exotherm (denoted by *) is noted at about
90 °C.
Fig. 4. A packing diagram of cis-(g5-C5H4Me)W(CO)2(PPh3)I viewed
down a-axis.
In a separate study trans samples were heated in the
DSC (heating rate ¼ 10 °C minꢂ1) and the heating was
discontinued when the temperature reached 80, 100, 120
and 130 °C, respectively. NMR spectra were then re-
corded on the samples. From this study a 70/30 cis/trans
isomer ratio of (g5-C5H4Me)Mo(CO)2(PPh3)I was
Fig. 6. TGA profiles for the cis and trans isomers of (g5-
C5H4R)Mo(CO)2(PPh3)I.
Fig. 5. A packing diagram of trans-(g5-C5H4Me)W(CO)2(PPh3)I
viewed down a-axis.
Table 4
ꢀ
Selected bond lengths (A) and angles (°) for the trans and cis isomers
trans isomer
cis isomer
W(1)–C(1)
W(1)–C(2)
W(1)–P(1)
W(1)–I(1)
1.992(1)
1.926(11)
2.465(3)
2.838(1)
1.976(1)
1.941(1)
2.513(1)
2.855(1)
C(2)–W(1)–C(1)
C(2)–W(1)–P(1)
C(1)–W(1)–I(1)
C(2)–W(1)–I(1)
C(1)–W(1)–I(1)
P(1)–W(1)–I(1)
111.7(4)
78.7(1)
80.3(1)
78.9(1)
74.3(1)
136.7(1)
72.81(1)
77.67(2)
122.40(1)
121.59(1)
74.26(1)
81.22(2)
isomer. However, for (g5-C5H4Me)W(CO)2[P(OiPr)3]I,
a cis–trans reaction was observed. This clearly indicates
that the direction of the solid-state reaction is not
Fig. 7. DSC profile of cis-(g5-C5H4Me)Mo(CO)2(PPh3)I as a function
of scanning rate (°C minꢂ1).

2212
O.G. Adeyemi et al. / Journal of Organometallic Chemistry 689 (2004) 2207–2215
Me)W(CO)2[P(OiPr)3]I complex isomerised into
the corresponding trans isomer during the DSC
measurement.
The data show that the isomerisation reaction (g5-
C5H4Me)W(CO)2[P(OiPr)3]I proceeds both in the solid
state and in the melt.
3.5. Kinetic study on trans- and cis-(g5-C5H4Me)-
Mo(CO)2(PPh3)I
The above studies suggested that a kinetic investiga-
tion of the solid-state isomerisation reaction was possi-
ble in the temperature regime 80–120 °C.
Reactions were performed on both the solid isomers
of (g5-C5H4Me)Mo(CO)2(PPh3)I in NMR tubes as
described in Section 2. Because both isomers had
similar red colors, the solid-state isomerisation reac-
tions of the molybdenum (or tungsten) complexes
could not be monitored visually and were only moni-
tored after appropriate time periods by silica gel TLC,
IR and NMR spectroscopy (1H, 31P). The kinetic data
for isomerisation of the trans isomer (110–120 °C) are
shown in Fig. 9. If the reaction temperature was
lowered to 90 or 100 °C, then very long reaction times
(days) are required for significant reaction. At higher
temperature (>135 °C), decomposition of the reactant
was observed. Analysis of the reactants revealed that
the cis isomer is thermally unstable at high tempera-
ture and shows about 30% conversion to the trans
isomer, triphenylphosphine oxide and an unidentified
product.
The same equilibrium mixture of cis/trans isomers is
reached from either isomer and the isomer ratio does not
vary significantly in the temperature range studied. The
data thus reveal a bi-directional isomer reaction. Rate
constants for the reactions were determined, assuming
opposing first-order reactions [15], and the k values for
the reactions were used in an Arrhenius plot to give an
activation energy (<160 kJ molꢂ1) for the reaction.
Some curvature was noted in the Arrhenius plot. This
could have arisen from a number of sources: (i) diffu-
Fig. 8. DSC profile of trans-(g5-C5H4Me)Mo(CO)2(PPh3)I as a
function of scanning rate (°C minꢂ1).
obtained only for samples heated above 100 °C. Thus,
the broad exotherm at 90 °C (Fig. 8) represents the
exotherm for the trans/cis isomerisation reaction.
Although expected, no equivalent exotherm was de-
tected in the DSC for the cis isomer, a finding that may
relate to the low% isomer conversion (30%).
A third exotherm (shown by both cis and trans iso-
mers) occurs above 180 °C and corresponds to the ma-
terial decomposition temperature, consistent with the
TGA data (see above).
3.4. Heating studies on (g5-C5H4Me)W(CO)2-
(P(OiPr)3)I
The solid-state isomerisation reaction of (g5-
C5H4Me)W(CO)2[P(OiPr)3]I unexpectedly proceeds
from the cis–trans isomer at T < 105 °C. This reaction
was investigated further by thermomicroscopy and
DSC. Under
a
microscope the red cis-(g5-C5H4
Me)W(CO)2[P(OiPr)3]I crystals, on heating, first melted
at 128–130 °C, then quickly resolidified and remelted at
141–142 °C, the latter temperature being close to that of
the melting point of the trans isomer (m.p. 144–146 °C).
Silica gel TLC and IR and NMR spectroscopy con-
firmed that the cis isomer had converted to the trans
product after the melting point process. Indeed, when
cis-(g5-C5H4Me)W(CO)2[P(OiPr)3]I was heated at 132
°C (in the molten state) for 15 min, 70% of the trans
isomer was produced. Further, a DSC spectrum re-
corded on the material after this procedure revealed that
the endothermic melting peak of cis-(g5-C5H4R)-
W(CO)2[P(OiPr)3]I was no longer observed (heating
rate: 1.0 °C minꢂ1). The endothermic peak at 143 °C
corresponding to the melting point of the trans
isomer was still noted. Silica gel TLC, and IR and
NMR spectroscopy confirmed that the cis-(g5-C5H4
Fig. 9. The kinetic data for isomerisation of the trans isomer (110–
120 °C).

O.G. Adeyemi et al. / Journal of Organometallic Chemistry 689 (2004) 2207–2215
2213
sional effects, (ii) decomposition effects and (iii) the
possibility that the reaction follows a more complex
mechanism. The activation data are to be compared
with the value for the solvent phase reaction ((g5-
C5H5)Mo(CO)2(PPh3)I ¼ 108 kJ molꢂ1 in o-dichloro-
benzene [10]) and for the corresponding solid-state
reaction for (C5H4Me)Mo(CO)2[P(OiPr)3]I (>68
kJ molꢂ1) [5].
in the two isomers. Further, the methyl groups on the
cyclopentadienyl ring of the respective isomers are also
seen to be in similar positions. This feature suggests
that, provided a low energy barrier exists between the
two isomers, the possibility of isomer interconversion
should be trivial. Thus, a simple rotation of a CO and I
ligand through 180 o could relate the two structures [4c].
3
ꢀ
The cell volume of the trans isomer (2451.1 A ) is
slightly larger (2% different) than that of the cis isomer
3
ꢀ
3.6. Powder XRD analysis
(2399.7 A ). Notwithstanding the cell volume change no
unusual non-bonded distances in either of the isomers
were noted.
The XRD profiles of the pure trans- and cis-(g5-
C5H4Me)Mo(CO)2(PPh3)I isomers are quite distinctive
and provide a possible means of monitoring the isom-
erisation reaction in the solid state (data not shown).
The profiles show that after 80-min reaction (120 °C) the
trans isomer had partially been converted to the cis
isomer and confirmed the analysis of the reaction
monitored by NMR spectroscopy (30/70, trans/cis ratio,
similar reaction conditions). Attempts were made to
obtain kinetic data from the analysis of the powder
diffraction data, but quantification of the results was not
achieved.
If the triphenylphosphine part of the molecule is
treated as the ‘‘tail’’ of the molecule and the cyclopen-
tadienyl ring as the ‘‘head’’, a ‘‘tail to tail’’ and ‘‘head to
head’’ packing can be observed for the cis isomer in
Fig. 5. A centre of symmetry relates two columns of the
isomer. The packing of the trans isomer in a unit cell
shows a ‘‘head to tail’’ arrangement and a ‘‘tail to head’’
arrangement. Noteworthy is the observation that if the
molecules of the trans isomer in the packing diagram
undergo a translation by half a unit cell, then a ‘‘tail to
tail’’ and ‘‘head to head’’ arrangement similar to that of
the cis isomer would result. This could provide a means
of generating the required new arrangement of mole-
cules in the unit cell after a trans ! cis isomerisation
reaction of the compound on heating.
3.7. Single crystal XRD analysis
Single crystals of pairs of isomers were only obtained
for trans- and cis-(g5-C5H4Me)W(CO)2(PPh3)I. Single
crystal analysis was thus performed on these two iso-
mers. Relevant bond length/angle data are given in Ta-
ble 4. The ORTEP plots of the cis and trans isomers of
(g5-C5H4Me)W(CO)2(PPh3)I are shown in Figs. 2 and
3. The bond length/angle data for both isomers are, to a
very large extent, similar. The analysis confirmed the
basic structure of the isomers; no unexpected bond
lengths or angles were noted.
A search of the Cambridge Data File [16] revealed
that no piano-stool cyclopentadienyl tungsten complexes
of the type (g5-C5H4R)W(CO)2LI have previously been
reported although the structure of trans-(g5-C5H5)-
Mo(CO)2(PPh3)I [17] has been documented. Only slight
differences in the Mo and W structures are expected, and
were found, and these related to the slightly different
sizes of the metal atoms.
Refinement of the trans isomer revealed that the
structure was disordered. One of the CO groups and the
iodine atom were interchangeable, i.e., the trans isomer
contained 11% of the cis-isomer in the crystal structure
after refinement. A second disorder was also observed.
The methyl group attached to the cyclopentadienyl ring
of the trans isomer refined in two separate positions
(50% occupancy).
Packing diagrams for the two isomers are shown in
Figs. 4 and 5. If one considers the cyclopentadienyl ring
and the PPh3 groups as ‘‘anchors’’, then the geometry of
the three remaining groups are almost superimposable
3.8. Solution state isomerisation reactions of cis and
trans-(g5-C5H4R)M(CO)2(L)I
The cis- and trans-(g5-C5H4R)M(CO)2(L)I com-
plexes are stable in solution (e.g., CHCl3, benzene) at
room temperature for a period of hours. However, when
refluxed in benzene they quickly isomerised to a ther-
modynamic equilibrium mixture. Thus, when cis-(g5-
C5H4Me)Mo(CO)2(PPh3)I was refluxed in benzene (or
CHCl3) for 4 h, a [trans]/[cis] ratio, 70/30, was achieved.
The corresponding trans isomer reached the same
equilibrium ratio in 12 h in refluxing CH2Cl2. This result
shows that the solution state isomerisation of (g5-
C5H4R)M(CO)2(L)I is also bidirectional. Similar equi-
librium studies on (g5-C5H5)Mo(CO)2(L)X (L ¼ PPh3,
PBu3; X ¼ Br, I) [6] and (g5-C5H4R)M(CO)2(L)I
t
(R ¼ Me, Bu, SiMe3; L ¼ isonitrile, phosphine, phos-
phite) [9b] in refluxing benzene have also been reported.
It is noteworthy that, even in the solution state, PPh3O
1
was observed in the H and 31P NMR spectra.
3.9. Mechanism
Earlier work suggested that the direction of the re-
action related to a trans–cis geometry preference and
that this could be associated with intramolecular effects
[18]. The current data suggest that not even this influ-
ence determines the isomerisation direction. Thus, the

2214
O.G. Adeyemi et al. / Journal of Organometallic Chemistry 689 (2004) 2207–2215
choice of the metal M, the ring-substituent R and the
ligand L do not appear to be the factors that by them-
selves determine the direction of the reaction, nor whe-
ther the reaction will proceed or not [19]. The only
property that correlates with the direction of the solid-
state isomerisation reaction of (g5-C5H4R)M(CO)2(L)I
(1) (M ¼ W, Mo) is the melting point of the cis and trans
isomers [4b]. The favoured reaction direction is always
from the low melting point isomer to the high melting
point isomer; when the differences are small, isomer
mixtures are formed. This suggests that the intermolec-
ular interactions cannot simply relate to cis/trans inter-
molecular polar effects in the solid state. Even this
melting point information is no guarantee that a reac-
tion will proceed [4b], which suggests that the ‘shape’ of
the molecule and the way in which it sterically interacts
with its nearest neighbours will determine the preferred
solid-state isomer.
4. Conclusions
The trans and cis isomers of the tungsten and mo-
lybdenum complexes (g5-C5H4R)M(CO)2(L)I (M ¼ W,
t
R ¼ Me, Bu, L ¼ P(OiPr)3, PPh3; M ¼ Mo, R ¼ Me,
L ¼ PPh3) have been synthesised and fully characterised
by elemental analysis and IR and NMR spectroscopy.
Unlike the rhenium complexes (g5-C5H4R)Re(CO)-
(L)X2, the tungsten and molybdenum complexes (g5-
C5H4R)M(CO)2(L)I undergo a bidirectional thermal
solid-state isomerisation reaction. A key finding was the
discovery that the direction of the reaction could not be
correlated with a particular isomer. The only correlation
noted is that the direction of the isomerisation reaction
is determined by melting points, i.e., the reaction pref-
erentially goes from the low melting point isomer to the
high melting point isomer.
The intermolecular forces that dominate in the solid
phase could thus be both steric (phase rebuilding [20],
topotatic [21]) or electronic in nature. Thus, an under-
standing of the reaction is possible from solid state
packing information [22]. However, no significant inter-
molecular interactions were observed between the mol-
ecules in the unit cells.
Acknowledgements
We thank the NRF (GUN #2059698), THRIP and
the University for financial support.
References
In earlier studies we determined that the cis–trans
isomerisation reaction of Re complexes was consistent
with a Turnstile type mechanism (this could involve
rotation of two or three of the small ligands (CO, X)).
The same could apply here. The single crystals became
opaque as the isomerisation reaction proceeded, even-
tually disintegrating into small crystallites.
[1] (a) G. Wilkinson, F.G.A. Stone, E.W. Abel (Eds.), Comprehen-
sive Organometallic Chemistry, Pergamon Press, New York, 1982;
(b) E.W. Abel, F.G.A. Stone, G. Wilkinson (Eds.), Comprehen-
sive Organometallic Chemistry, Pergamon Press, New York, 1994.
[2] Y. Ohashi, Acc. Chem. Res. 21 (1988) 268.
[3] (a) D.L. Beach, K.W. Barnett, J. Organomet. Chem. 97 (1975)
C27;
However, the structural analysis provided informa-
tion via the disorder content. Part of this disorder re-
lated to the presence of cis isomer (11%) in the trans
isomer. This would suggest that at low conversions, the
cis isomer is accommodated within the unit cell of the
trans isomer. This phenomenon of a major structural
isomer determining the structure of a minor isomer has
precedent in the work of Sacconi et al. [23] and has
been referred to as ‘structure mimicry’. Disorder in the
cyclopentadienyl ring is indicative of the strain that
must be built up (50% disorder) as the reaction
progresses.
(b) R.B. King, R.H. Reimann, Inorg. Chem. 12 (1976) 183;
(c) R.B. King, R.H. Reimann, D.J. Darensbourg, J. Organomet.
Chem. 93 (1975) C23.
[4] (a) L. Cheng, N.J. Coville, J. Organomet. Chem. 571 (1998) 149;
(b) N.J. Coville, D.C. Levendis, Eur.J. Inorg. Chem. (2002) 3067;
(c) R. Bogadi, D.C. Levendis, N.J. Coville, J. Am. Chem. Soc.
124 (2002) 1104.
[5] O.G. Adeyemi, M.A. Fernandes, L. Cheng, U.B. Eke, D.C.
Levendis, N.J. Coville, C. R. Chimie 5 (2002) 387.
[6] D.L. Beach, M. Dattilo, K.W. Barnett, J. Organomet. Chem. 140
(1977) 47.
[7] A.C. Filipou, J.G. Winter, M. Feist, G. Kociok, I. Hinz,
Polyhedron 17 (1998) 1103.
[8] R.B. King (Ed.), Organometallic Synthesis, vol. 1, Academic
Press, New York, NY, 1955.
Further, even though both unit cells are monoclinic,
the arrangement of atoms within the unit cells of the two
isomers is different.
[9] (a) D.J. Blumer, K.W. Barnett, T.L. Brown, J. Organomet.
Chem. 71 (1979) 173;
(b) N.J. Coville, M.S. Loonat, L. Carlton, S. Afr. J. Chem. 45
(1992) 63;
The proposed mechanism suggests that in the early
stages of reaction, a cis–trans isomerisation reaction
occurs prior to inter-molecule rearrangement. Eventu-
ally the rearrangement of molecules will result in inter-
molecular stresses in the unit cell and subsequent break
up of the crystalline phase. Whether indeed the cis iso-
mer formed in the trans–cis reaction is a poymorph of
the cis isomer formed by crystallisation procedures
cannot be addressed by the data at hand.
(c) M.O. Albers, N.J. Coville, Coord. Chem. Rev. 53 (1984)
227.
[10] F.W. Faller, A.S. Anderson, J. Am. Chem. Soc. 92 (1970) 5852.
[11] Bruker, SAINT+, Version 6.02 (includes XPREP and SADABS),
Bruker AXS Inc., Madison, WI, 1999.
[12] G.M. Sheldrick, SHELX 97 (includes SHELXS 97, SHELXL 97 and
CIFTAB), Programs for Crystal Structure Analysis (Release 97-2),
€
University of Gottingen, Germany, 1997.
[13] L.J. Farrugia, J. Appl. Cryst. 32 (1999) 837.

O.G. Adeyemi et al. / Journal of Organometallic Chemistry 689 (2004) 2207–2215
2215
[14] L.J. Farrugia, J. Appl. Cryst. 30 (1977) 565.
M.S. Adsetts, L.M. Cook, J.C.A. Boeyens, D.C. Levendis,
Organometallics 19 (2000) 2597.
[15] D.J. Darensbourg, R.L. Gray, Inorg. Chem. 23 (1984) 2993.
[16] F.H. Allen, Cambridge Structural Database, Version 5.25,
November 2003, Acta Cryst., Sect. B 58 (2002) 380.
[20] G. Kaupp, in: J.L. Atwood, J.E.D Davies, D.D. MacNicol, F.
Votgle (Eds.), Comprehensive Supramolecular Chemistry, vol. 8,
Pergamon Press, New York, 1996 (Chapter 9).
[21] G.M.J. Schmidt, Pure Appl. Chem. 27 (1971) 647.
[22] D. Braga, Chem. Rev. 92 (1992) 633.
[17] M.A. Bush, A.D.U. Hardy, Lj. Manojlovic-Muir, G.A. Sim, J.
Chem. Soc. (A) (1971) 1003.
[18] J.E. Smith, N.J. Coville, Organometallics 15 (1996) 3388.
[19] (a) L. Cheng, N.J. Coville, Organometallics 15 (1996) 867;
(b) J.M. Smith, L. Cheng, N.J. Coville, J. Schulte, P.S. Dimpe,
[23] L. Sacconi, M. Ciampolini, G.P. Speroni, J. Am. Chem. Soc. 87
(1965) 3102.