Preparation and electronic properties of rhenium(V) complexes with bis(diphenylphosphino)ethane

Article abstract of DOI:10.1139/v03-143

Complexes of bis(diphenylphosphino)ethane (dppe) of the types ReOX ₃(dppe) and ReO(OR)X₂(dppe) (with X= Cl or Br, and R = Me, Et, Pr, Ph, cyclohexyl (Cy), or -CH₂CH₂OH) were prepared to evaluate the influence of ligand changes on the low-energy d-d transitions in these Re(V) low-spin d² systems. Arylimido compounds Re(NR)Cl ₃(dppe) (with R = Ph and p-ClC₆H₄) were also obtained. X-ray diffraction studies on ReO(OPr)Cl₂(dppe), ReO(OPh)Br₂(dppe), and ReO(OCy)Cl₂(dppe) confirmed the presence of the trans O=Re-OR unit and showed that the structural characteristics of the Re-O-R segment are not affected by the presence of a bulky cyclohexyl substituent or an aromatic phenyl group. The structure of the arylimido complex Re(p-ClC₆H₄N)Cl₃(dppe) was also determined. The electronic absorption spectra of the ReOX₃(dppe) compounds include two low-energy components at ～11 500 and ～16 000 cm^-1, assigned to the two spin-allowed d-d transitions expected for these low-symmetry systems. Substitution of the oxo ligand by an arylimido group has little effect on the lower-energy component, but moves the two components closer together. Replacing the halogen trans to the Re=O bond by an alkoxo group shifts the whole system to higher energies. These variations were found to correlate well with the energies of the frontier orbitals determined from DFT calculations. While attempting to prepare ReOCl₃(dppe) from ReOCl₃(PPh₃)₂, the Re(III) compound fac-ReCl₃(PPh₃){Ph₂PC₂H ₄PPh₂(O)} was obtained, in which one end of dppe had become a phosphine oxide. Upon standing in DMSO, this compound gave the octahedral compound ReCl₄{Ph₂PC₂H ₄PPh₂(O)}, in which the formation of a chelate ring involving a phosphine and a phosphine oxide was ascertained by X-ray diffraction.

Full text of DOI:10.1139/v03-143

1326
Preparation and electronic properties of rhenium(V)
complexes with bis(diphenylphosphino)ethane1
François Baril-Robert and André L. Beauchamp
Abstract: Complexes of bis(diphenylphosphino)ethane (dppe) of the types ReOX3(dppe) and ReO(OR)X2(dppe) (with
X= Cl or Br, and R = Me, Et, Pr, Ph, cyclohexyl (Cy), or -CH2CH2OH) were prepared to evaluate the influence of
ligand changes on the low-energy d–d transitions in these Re(V) low-spin d2 systems. Arylimido compounds
Re(NR)Cl3(dppe) (with R = Ph and p-ClC6H4) were also obtained. X-ray diffraction studies on ReO(OPr)Cl2(dppe),
ReO(OPh)Br2(dppe), and ReO(OCy)Cl2(dppe) confirmed the presence of the trans O=Re-OR unit and showed that the
structural characteristics of the Re-O-R segment are not affected by the presence of a bulky cyclohexyl substituent or
an aromatic phenyl group. The structure of the arylimido complex Re(p-ClC6H4N)Cl3(dppe) was also determined. The
electronic absorption spectra of the ReOX3(dppe) compounds include two low-energy components at ~11 500 and
~16 000 cm–1, assigned to the two spin-allowed d–d transitions expected for these low-symmetry systems. Substitution
of the oxo ligand by an arylimido group has little effect on the lower-energy component, but moves the two compo-
nents closer together. Replacing the halogen trans to the Re=O bond by an alkoxo group shifts the whole system to
higher energies. These variations were found to correlate well with the energies of the frontier orbitals determined
from DFT calculations. While attempting to prepare ReOCl3(dppe) from ReOCl3(PPh3)2, the Re(III) compound fac-
ReCl3(PPh3){Ph2PC2H4PPh2(O)} was obtained, in which one end of dppe had become a phosphine oxide. Upon stand-
ing in DMSO, this compound gave the octahedral compound ReCl4{Ph2PC2H4PPh2(O)}, in which the formation of a
chelate ring involving a phosphine and a phosphine oxide was ascertained by X-ray diffraction.
Key words: rhenium, crystal structure, DFT calculations, d–d electron transitions.
Résumé : Des complexes du bis(diphénylphosphino)éthane (dppe) de types ReOX3(dppe) et ReO(OR)X2(dppe) (où X =
Cl ou Br, et R = Me, Et, Pr, Ph, cyclohexyl (Cy) ou -CH2CH2OH), ont été préparés afin d’évaluer l’influence de chan-
gements de ligands sur les transitions d–d de basse énergie dans ces systèmes d2 de Re(V) à spin faible. Des composés
arylimido Re(NR)Cl3(dppe) (où R = Ph et p-ClC6H4) ont aussi été obtenus. Les études par diffraction des rayons X sur
ReO(OPr)Cl2(dppe), ReO(OPh)Br2(dppe) et ReO(OCy)Cl2(dppe) ont confirmé la présence du motif trans O=Re-OR et
montré que les particularités structurales de l’ensemble Re-O-R ne sont pas affectées par la présence du substituant vo-
lumineux cyclohexyle ou du groupe aromatique phényle. La structure du complexe arylimido Re(p-ClC6H4N)Cl3(dppe)
a également été determinée. Les spectres d’absorption électroniques des composés ReOX3(dppe) comportent deux com-
posantes de basse énergie à ~11 500 et ~16 000 cm–1, attribuées respectivement aux deux transitions d–d permises de
spin, attendues pour ces systèmes de basse symétrie. La substitution du ligand oxo par un groupe arylimido a peu
d’influence sur la composante de plus basse énergie, mais elle rapproche les deux composantes l’une de l’autre. Le
remplacement de l’halogène trans au lien Re=O par un group alkoxo déplace l’ensemble du système vers des énergies
supérieures. Ces variations sont en bonne corrélation avec les énergies des orbitales frontières déterminées par calculs
DFT. En tentant de préparer ReOCl3(dppe) à partir de ReOCl3(PPh3)2, le composé de Re(III) fac-ReCl3(PPh3){Ph2-
PC2H4PPh2(O)} a été obtenu, dans lequel l’une des extrémités du dppe est devenue un oxyde de phosphine. En laissant
reposer dans le DMSO, ce composé a conduit au complexe octaédrique ReCl4{Ph2PC2H4PPh2(O)}, où la formation de
chélate impliquant une phosphine et un oxyde de phosphine a été confirmée par diffraction des rayons X.
Mots clés : rhénium, structure cristalline, calculs DFT, transitions électroniques d–d. Baril-Robert and Beauchamp
1340
Introduction
tions in radiopharmacy as labeling or therapeutic agents (1–
4). Our contributions to this field include compounds with
amino acids (5, 6), nitrogen heterocycles (7–12), and various
bidentate ligands containing a soft phosphine group and a
hard oxygen donor (13–18). Multiple-bonded Re(V) com-
Over the past decades, many research groups have been
examining various rhenium(V) compounds containing the
Re=O3+, Re=NR3+, and ReϵN2+ cores for eventual applica-
Received 5 September 2003. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 24 October 2003.
F. Baril-Robert and A.L. Beauchamp.2 Département de chimie, Université de Montréal, C.P. 6128, Succursale Centre-ville,
Montréal, QC H3C 3J7, Canada.
1This article is part of a Special Issue dedicated to Professor John Harrod.
2Corresponding author (e-mail: andre.beauchamp@umontreal.ca).
Can. J. Chem. 81: 1326–1340 (2003)
doi: 10.1139/V03-143
© 2003 NRC Canada

Baril-Robert and Beauchamp
1327
pounds are of interest in many other fields of chemistry and
biochemistry. For instance, their role in oxygen transfer and
other catalytic processes is under active investigation at the
moment (19–23). The optical properties associated with the
multiple bonds have produced a large number of publica-
tions and are still extensively studied (24–28).
ReOCl3(PPh3)2 was prepared by a literature method (30).
Deuterated solvents were purchased from Aldrich or CDN
Isotopes.
IR spectra were recorded on Perkin-Elmer 1600 (4000–
400 cm–1, KBr pellets) or 1430 spectrophotometers (4000–
200 cm–1, CsI pellets). The NMR spectra were measured
with Bruker AMX-300, ARX-400, or AV-300 spectrometers.
In the perspective of developing devices for storage of in-
formation at the molecular level, these simple diamagnetic
d2 centers would be potentially useful systems if a low-
energy paramagnetic high-spin state can be readily accessed.
The oxygen or nitrogen px and py orbitals of the multiple-
bonded ligand form strong π bonds with the dxz and dyz
orbitals, which become antibonding and higher in energy
than the other interaxial d orbital (dxy or dx2 2 , depending
on point group symmetry). Occupancy of th−ey latter orbital
by the two d electrons results in a diamagnetic ground state.
However, in a previous study (29), we showed that the
HOMO–LUMO gap is relatively small in certain oxo com-
pounds (I) containing O-P ligands. If the gap becomes com-
parable with the pairing energy, the molecule could possess
a high-spin excited state accessible at low energy cost via a
spin–crossover process. Such systems may be of interest as
catalysts and materials with unusual optical or magnetic
properties.
1
For the H spectra, the residual solvent signal (DMSO-d6,
2.50 ppm) was used as internal reference and the chemical
shifts are expressed in ppm vs. SiMe4. H3PO4 was used as
an external reference (δ = 0) for the 31P{1H} spectra. Elec-
tronic spectra were recorded in quartz cells with a UV–vis–
NIR Cary 5E spectrometer. The solution spectra were taken
as DMSO solutions. Spectra of crystalline samples were ob-
tained with a tungsten-halogen lamp, the CCD detector of a
Renishaw 3000 Raman imaging microscope, and a 10% neu-
tral density filter to avoid blinding the detector. FAB+ mass
spectra in NBA were measured at the Centre de Spectro-
métrie de Masse de l’Université de Montréal. Elemental
analyses were run at the Laboratoire d’Analyse Élémentaire
de l’Université de Montréal.
Preparation of ReOX3(dppe) (X = Cl, Br)
Modifications were introduced in the procedure originally
described by Chatt and Rowe (30) for ReOCl3(dppe) to
eliminate impurities of MX salts and reduced complexes.
The K+ ion of KReO4 was first replaced by H+ on a
cation-exchange resin (Dowex 50W-X8, 20–50 mesh). The
column (2 cm diameter) was filled with 10 g of resin, which
was activated with 15 mL of 6mol/L HCl to obtain the H+
form and washed with boiling water (75 mL) to neutrality.
KReO4 (1.5 g, 5.2 mmol) dissolved in a minimum amount
(ca. 18 mL) of boiling water was poured into the hot column
and the elution was done with boiling water. The HReO4 so-
lution was collected as long as it remained acidic. Water was
then removed with a rotating evaporator and a viscous light
yellow-green liquid containing HReO4 and Re2O7 was ob-
tained. To this oily material was added concentrated HX
(5 mL) and the mixture was refluxed for 2 min. A 25 mL
fraction of a chloroform solution of dppe (3.90 g, 9.80 mmol
in 150 mL) was first added and the mixture darkened imme-
diately. The rest of the dppe solution was then added in
small portions over a period of 4 h. The mixture was cooled
to room temperature (RT) and the precipitate (green for Br,
blue for Cl) was filtered, washed with acetone (to remove
water), chloroform, and diethyl ether. The compound was
recrystallized from acetonitrile.
To cast some light on the factors leading to small
HOMO–LUMO gaps, we first examined a series of simple
compounds retaining an equatorial cis-P,P arrangement per-
pendicular to an (axial) Re=O or Re=NR bond. The common
bis(diphenylphosphino)ethane (dppe) ligand was chosen to
provide this arrangement. The strongly π-donor trans-
O=Re=O+ and ReϵN2+ cores were not considered at this
point, since their energy gaps had been found to be substan-
tially higher (29). ReOCl3(dppe) was originally prepared by
Chatt and Rowe (30), and X-ray work showed that it indeed
corresponds to the structural type considered (31). The cor-
responding phenylimido compound Re(NPh)Cl3(dppe) was
obtained by Cotton et al. (32). In the present study, the
chloro and bromo compounds ReOX3(dppe) were prepared,
together with trans-ReO(OR)X2(dppe) derivatives with vari-
ous R groups. A few arylimido compounds were also in-
cluded. In addition to the preparation and characterization of
these compounds, we are reporting preliminary electronic
spectra showing how absorption is affected by ligand substi-
tution. DFT calculations are used to rationalize these results.
ReOCl3(dppe)
Yield: 82%. IR (CsI, cm–1): 980 (s) ν(Re=O); 320 (w),
1
290 (w) ν(Re-Cl). H NMR (DMSO-d6) (ppm) δ: 8.01 (m,
8H, Ho), 7.48 (s br, 12H, Hm,p), 3.79 (m, 2H, ethylene), 3.24
(m, 2H, ethylene). 31P NMR (DMSO-d6) (ppm) δ: 5.7 (s).
MS FAB+ (m/z): 706 [M]+; 671 [M – Cl]+. Anal. calcd. for
C26H24OP2Cl3Re (%): C 44.17, H 3.42; found: C 44.24, H
3.34.
Experimental section
ReOBr3(dppe)
1
Yield: 76%. IR (KBr, cm–1): 971 (s) ν(Re=O). H NMR
Reactants and methods
KReO4, dppe, and all other reactants were obtained
from Aldrich and used without further purification.
(DMSO-d6) (ppm) δ: 8.03 (m, 8H, Ho), 7.48 (s br, 12H,
H
m,p), 3.90 (m, 2H, ethylene) 3.17 (m, 2H, ethylene). 31P
© 2003 NRC Canada

1328
Can. J. Chem. Vol. 81, 2003
NMR (DMSO-d6) (ppm) δ: –1.0 (s). MS FAB+ (m/z): 840
ReOBr2(OPr)(dppe)
[M]+, 760 [M – Br]+, 680 [M – 2Br]+.
1
Yield: 74%. IR (KBr, cm–1): 937 (s) ν(Re=O). H NMR
(DMSO-d6) (ppm) δ: 8.13 (m, 8H, Ho), 7.49 (s br, 12H,
H
m,p), 3.38 (m, 2H, ethylene, near water peak), 3.08 (m, 2H,
Preparation of ReOX2(OR)(dppe) (X = Cl, Br)
3
ethylene), 2.06 (t, 2H, J = 6.5 Hz, propoxo α-CH2), 0.38
3
(m, 2H, propoxo β-CH2), 0.11 (t, 3H, J = 7.4 Hz, propoxo
Method A (R = Me, Et, Pr)
CH3). 31P NMR (DMSO-d6) (ppm) δ: 17.4 (s). Anal. calcd.
for C29H31O2P2Br2Re (%): C 42.50, H 3.81; found: C 42.78,
H 3.87.
For ReOCl2(OEt)(dppe), a suspension of ReOCl3(dppe)
(0.31 g, 0.44 mmol) in 35 mL of ethanol was refluxed for
3 days, during which time the blue solid progressively
turned violet. It was filtered hot, washed with ethanol and
diethyl ether, and recrystallized in acetonitrile.
Method B (R = cyclohexyl, -CH2CH2OH)
A suspension of ReOCl2(OEt)(dppe) (0.20 g, 0.28 mmol)
in 25 mL of an acetone–cyclohexanol (or –ethylene glycol)
(4:1, v/v) mixture was refluxed for 3 days. The color of the
solid did not change appreciably during the reaction. The vi-
olet precipitate was filtered, washed with ethanol and diethyl
ether, and recrystallized in acetonitrile. For the soluble ethyl-
ene glycol complex, the final mixture was evaporated to dry-
ness, diethyl ether was added, and the violet product was
filtered.
This method was used to introduce alkoxo groups corre-
sponding to low-boiling alcohols (methanol, ethanol, n-
propanol). Reaction time varied (2–6 days) from one system
to another. The reaction was stopped ~24 h after the reaction
mixture has become clear violet. When there was no color
change after 3 days, adding fresh alcohol helped to acceler-
ate the reaction.
ReOCl2(OEt)(dppe)
Yield: 90%. IR (KBr, cm–1): 950 (s) ν(Re=O), 913 (s)
ReOCl2(OCy)(dppe)
1
δ(O-CH2). H NMR (DMSO-d6) (ppm) δ: 8.14 (m, 8H, Ho),
1
Yield: 72%. IR (KBr, cm–1): 926 (s) ν(Re=O). H NMR
7.48 (s br, 12H, Hm,p), ~3.31 (m, 2H, ethylene, near water
(DMSO-d6) (ppm) δ: 8.27 (m, 4H, Ho), 8.16 (m, 4H, Ho),
7.47 (s br, 12H, Hm,p), ~3.45 (m, 2H, ethylene, in the tail of
the water peak), ~3.18 (m, 2H, ethylene, near water peak);
cyclohexyl signals assigned from a 2D spectrum: 2.33 (m,
1H, α-CyO), 1.03 (m, 3H, γ/δ-CyO), 0.56 (m, 2H, β-CyO),
0.38 (m, 3H, γ/δ-CyO), –0.11 (m, 2H, β-CyO). 31P NMR
(DMSO-d6) (ppm) δ: 16.9 (s).
3
peak), 3.14 (m, 2H, ethylene), 2.28 (q, 2H, J = 7.0 Hz,
3
ethoxo CH2), –0.07 (t, 3H, J = 6.9 Hz, ethoxo CH3). 31P
NMR (DMSO-d6) (ppm) δ: 16.5 (s). MS FAB+ (m/z): 716
[M]+, 681 [M – Cl]+, 671 [M – OEt]+, 637 [M – Cl – OEt]+.
Anal. calcd. for C28H29O2P2Cl2Re (%): C 46.93, H 4.08;
found: C 46.66, H 3.94.
ReOCl2(OMe)(dppe)
ReOCl2(OCH2CH2OH)(dppe)
1
Yield: 79%. IR (KBr, cm–1): 944 (s) ν(Re=O). H NMR
Yield: 85%. IR (KBr, cm–1): 948 ν(Re=O), 907 (s) δ(O-
(DMSO-d6) (ppm) δ: 8.11 (m, 8H, Ho), 7.49 (s br, 12H,
1
CH2). H NMR (DMSO-d6) (ppm) δ: 8.10 (m, 8H, Ho), 7.49
H
m,p), ~3.23 (m, 2H, ethylene, near water peak), 3.15 (m,
(s br, 12H, Hm,p), 3.43 (m, 2H, ethylene, near water peak),
2H, ethylene), 1.91 (s, 3H, methoxo CH3). 31P NMR
3.11 (m, 2H, ethylene), 2.42 (t, 2H, 3J = 6.2 Hz, β-
(DMSO-d6) (ppm) δ: 16.6 (s).
3
CH2CH2OH), 2.17 (t, 2H, J = 6.3 Hz, α-CH2CH2OH). 31P
NMR (DMSO-d6) (ppm) δ: 19.1 (s). MS FAB+ (m/z): 732
[M]+, 697 [M – Cl]+, 671 [M – OC2H4OH]+, 661 [M –
2Cl]+, 636 [M – Cl – C2H4OH]+.
ReOCl2(OPr)(dppe)
Yield: 85%. IR (KBr, cm–1): 938 (s) ν(Re=O). H NMR
1
(DMSO-d6) (ppm) δ: 8.16 (m, 8H, Ho), 7.48 (s br, 12H,
H
m,p), 3.34 (m, 2H, ethylene), 3.15 (m, 2H, ethylene), 1.99
Method C (R = Ph)
3
(t, 2H, J = 6.5 Hz, propoxo α-CH2), 0.34 (m, 2H, propoxo
A solution of phenol (5.5 g, 59 mmol) in 70 mL of ben-
zene was heated to the boiling point and ReOBr2(OPr)(dppe)
(0.49 g, 0.60 mmol) was added. The violet mixture turned
orange quickly. After 1 day, the solvent was evaporated, the
raw product was added to 50 mL of diethyl ether, and
refluxed for 30 min to dissolve the unreacted phenol. The
dark-brown precipitate was filtered hot, washed with diethyl
ether, and recrystallized in acetonitrile.
β-CH2), 0.11 (t, 3H, J = 7.3 Hz, propoxo CH3). 31P NMR
3
(DMSO-d6) (ppm) δ: 15.8 (s).
ReOBr2(OEt)(dppe)
Yield: 61%. IR (KBr, cm–1): 951 (s) ν(Re=O), 913 (s)
1
δ(O-CH2). H NMR (DMSO-d6) (ppm) δ: 8.12 (m, 8H, Ho),
7.50 (m br, 12H, Hm,p), 3.39 (m, 2H, ethylene), 3.07 (m, 2H,
3
ethylene), 2.35 (q, 2H, J = 6.8 Hz, ethoxo CH2), –0.03 (t,
3H, J = 6.9 Hz, ethoxo CH3). 31P NMR (DMSO-d6) (ppm)
3
ReOBr2(OPh)(dppe)
Yield: 83%. IR (KBr, cm–1): 959 (s) ν(Re=O). H NMR
δ: 18.2 (s).
1
3
(DMSO-d6) (ppm) δ: 8.13 (m, 4H, Ho), 8.00 (dd, 4H, J =
3
ReOBr2(OMe)(dppe)
7.6 Hz, JH-P = 10.9 Hz, Ho), 7.53 (s br, 6H, Hm,p), 7.35 (t,
1
3
3
Yield: 75%. IR (KBr, cm–1): 936 (s) ν(Re=O). H NMR
2H, J = 7.0 Hz, Hp), 7.26 (t, 4H, J = 7.5 Hz, Hm), 6.62 (m,
3
(DMSO-d6) (ppm) δ: 8.12 (m, 4H, Ho), 8.01 (m, 4H, Ho),
7.51 (s br, 12H, Hm,p), ~3.35 (m, 2H, ethylene, near water
peak), 3.08 (m, 2H, ethylene), 2.02 (s, 3H, methoxo CH3).
31P NMR (DMSO-d6) (ppm) δ: 18.8 (s).
3H, phenoxo Hm,p), 5.67 (d, 2H, J = 8.0 Hz, phenoxo Ho),
3.55 (m, 2H, ethylene), 3.23 (m, 2H, ethylene). 31P NMR
(DMSO-d6) (ppm) δ: 17.3 (s). MS FAB+ (m/z): 854 [M]+,
773 [M – Br]+, 761 [M – OPh]+, 680 [M – Br – OPh]+.
© 2003 NRC Canada

Baril-Robert and Beauchamp
1329
Anal. calcd. for C32H29O2P2Br2Re (%): C 45.03, H 3.42;
found: C 45.09, H 3.39.
Anal. calcd. for C45.5H42Cl6OP3Re (%): C 49.83, H 3.86;
found: C 49.68, H 3.98.
ReOCl2(OPh)(dppe)
X-ray diffraction studies
1
Yield: 90%. IR (KBr, cm–1): 960 (s) ν(Re=O). H NMR
Three data sets were collected with a Bruker Smart 2K
CCD diffractometer controlled by the SMART software
(33), using the graphite-monochromatized CuKα radiation.
Sets of 30 oscillation frames of 0.3° over a range of 9° were
recorded in four regions of the reciprocal space, from which
the reduced cell was determined by least-squares refinement.
For data collection, frames were recorded for different orien-
tations of the crystal and detector so as to cover at least 95%
of the reciprocal sphere. At the end, the first 101 frames
were remeasured to check for crystal decomposition. The
frames were analyzed with the SAINT software (34), which
determined the intensity and the position of each spot. Accu-
rate cell parameters were obtained by least-squares refine-
ment over the positions of the whole data set.
For one compound, the X-ray work was done with an
Enraf-Nonius CAD-4 diffractometer using graphite-
monochromatized CuKα radiation under the control of the
CAD-4 software (35). The reduced cell was initially deter-
mined from 20 to 25 spots located on a preliminary rotation
photograph and centered in the detector aperture. Accurate
cell parameters were then determined from the positions of
25 high-angle reflections centered with the SETANG and
DETTH procedures. The intensities were recorded by ω/2θ
scan. Possible crystal decomposition was monitored by mea-
suring five standard reflections every hour.
3
(DMSO-d6) (ppm) δ: 8.13 (m, 4H, Ho), 8.00 (dd, 4H, J =
3
7.7 Hz; JH-P = 11.1 Hz, Ho), 7.51 (s br, 6H, Hm,p), 7.30 (m,
2H, Hp), 7.23 (m, 4H, Hm), 6.59 (m, 3H, phenoxo Hm,p),
3
5.63 (d, 2H, J = 7.9 Hz, phenoxo Ho), 3.57 (m, 2H, ethyl-
ene), 3.32 (m, 2H, ethylene, near water peak).31P NMR
(DMSO-d6) (ppm) δ: 16.3 (s).
Preparation of Re(NPh)Cl3(dppe)
The compound was prepared following the literature
method (32). ReOCl3(dppe) (0.15 g, 0.21 mmol) was added
to 30 mL of a refluxing aniline–benzene (1:5, v/v) mixture.
The blue solution was refluxed for 36 h, during which time
it turned green. The light-green precipitate was filtered, and
washed with benzene and diethyl ether. It recrystallized in
acetonitrile as light-green fibers.
Re(NPh)Cl3(dppe)
1
Yield: 75%. H NMR (DMSO-d6) (ppm) δ: 8.16 (m, 4H,
Ho), 7.96 (m, 4H, Ho), 7.52 (t, 1H, 3J = 8.2 Hz, phenylimido
Hp), 7.41 (s br, 6H, Hm,p), 7.16 (s br, 6H, Hm,p), 6.79 (t, 2H,
3J = 8.5 Hz; phenylimido Hm), 6.15 (d, 2H, J = 8.7 Hz,
3
phenylimido Ho), ~3.59 (m, 2H, ethylene), 3.33 (m, 2H, eth-
ylene). 31P NMR (DMSO-d6) (ppm) δ: 11.3 (s).
Another data set was collected with the Bruker P-4
diffractometer of the Université du Québec à Montréal,
using the graphite-monochromatized MoKα radiation. The
procedures to determine the cell parameters and collect the
intensity data (XSCANS software) (36) were similar to those
described above for the CAD-4 system. Three standard re-
flections were measured every 97 reflections during data
collection.
The data were interpreted with the SHELXTL software
(37). The XPREP procedure (38) was used to apply to the
raw intensities an absorption correction based on crystal
morphology and to determine the Laue symmetry, system-
atic absences, and space group. The structure was solved by
the direct methods or the heavy-atom method with SHELXS
(39). The heavy atoms were initially found and the remain-
ing atoms were then located from structure-factor calcula-
tions and ∆F maps with SHELXL (40). The structure was
refined by least squares on F2. The non-hydrogen atoms
were generally refined anisotropically. The hydrogens were
first placed at idealized positions with the standard (C,N,O)-
H distances of SHELXL and allowed to ride on the support-
ing atom. Their isotropic temperature factors (Uiso) were
fixed at values related to the equivalent temperature factor
(Ueq) of the supporting atom by Uiso = k × Ueq (where k =
1.5 (methyl) or 1.2 (others)). The ORTEP diagrams (40%
probability ellipsoids) were produced with the XP routine of
SHELXTL.
Preparation of Re(NC6H4Cl)Cl3(dppe)
This variation of the previous method was used to intro-
duce a phenylimido group derived from a solid aniline. A
solution of p-chloroaniline (5.56 g, 43.6 mmol) in 25 mL of
benzene was heated to its boiling point and ReOCl3(dppe)
(0.15 g, 0.21 mmol) was added. The solution was refluxed
for 1 day, during which time the blue solution turned dark
green. The solvent was evaporated completely, the solid ob-
tained was placed in 50 mL of diethyl ether, and the solution
was refluxed for 30 min to dissolve unreacted p-
chloroaniline. The green precipitate was filtered hot, washed
with diethyl ether, and recrystallized in acetonitrile.
Re(NC6H4Cl)Cl3(dppe)
1
Yield: 78%. H NMR (DMSO-d6) (ppm) δ: 8.16 (m, 4H,
Ho), 7.96 (m, 4H, Ho), 7.42 (s br, 6H, Hm,p), 7.19 (s br, 6H,
3
H
m,p), 6.88 (d, 2H, J = 8.8 Hz, arylimido Hm), 6.13 (d, 2H,
3J = 8.6 Hz, arylimido Ho), 3.56 (m, 2H, ethylene), 3.33 (m,
2H, ethylene, near water peak). 31P NMR (DMSO-d6) (ppm)
δ: 11.1 (s).
Preparation of ReCl3(PPh3)(dppe=O)
ReOCl3(PPh3)2 (0.112 g, 0.13 mmol) was added to a solu-
tion of dppe (0.311 g, 0.79 mmol) in 25 mL CH2Cl2 under a
dynamic inert atmosphere. The resulting lime-green solution
was stirred at room temperature for 1 h. The beige precipi-
tate formed during the reaction was filtered, washed with
chloroform and diethyl ether. Yield: 51%. It corresponded to
the 1.5CH2Cl2 solvate. IR (KBr, cm–1): 1058 (s), 1125 (s).
1H NMR (DMSO-d6): see text. MS FAB+ (m/z): 968 [M]+,
933 [M – Cl]+, 706 [M – PPh3]+, 635 [M – 2Cl – PPh3]+.
The crystals of Re(NC6H4Cl)Cl3(dppe) and the three
ReOX2(OR)(dppe) compounds were obtained by slow crys-
tallization of concentrated acetonitrile solutions, whereas
those of ReCl4(dppe=O) appeared in a DMSO solution of
ReCl3(PPh3)(dppe=O) kept at room temperature for
© 2003 NRC Canada

1330
Can. J. Chem. Vol. 81, 2003
© 2003 NRC Canada

Baril-Robert and Beauchamp
1331
3 months. In all cases, the intensities of the standard reflec-
tions revealed no crystal decomposition during data collec-
tion. Space groups were defined unambiguously from the
Laue symmetry and systematic absences, except for the
triclinic compound ReCl4(dppe=O)•Me2SO. Crystal data are
collected in Table 1.
Results and discussion
Preparation of ReOX3(dppe) and ReOX2(OR)(dppe)
The procedure originally described to prepare
ReOCl3(dppe) (30) required a tedious series of recrystalli-
zations to remove the MX salt and various reduction prod-
ucts. By starting with a HReO4–Re2O7 solution and using
chloroform as solvent, we obtained good yields of pure
chloro and bromo compounds. The IR spectra show the
strong ν(Re=O) band at 970–980 cm–1 expected for a mono-
oxo system (42). Clean isotopic patterns are found in FAB+
spectra for the parent ion and the [M – X]+ fragment. The
presence of only one 31P{1H} NMR singlet confirms the for-
mation of the fac isomer, since the P donors would be
inequivalent in the mer isomer. Although a “frozen” puck-
ered chelate ring contains four inequivalent aliphatic
protons, a fast flip-flop motion across the equatorial plane
generates a vertical mirror plane (including the axial Re=O
bond and bisecting the P-Re-P angle) on the NMR
timescale. In the “average” molecule, the pair of protons on
the same side of the equatorial plane become equivalent. Ac-
cordingly, two complex multiplets are found between 3 and
4 ppm, one of which is often obscured by the residual water
peak in DMSO. Because of the anisotropic effect of the
Re—oxo bond, the protons on the oxo side are believed to
be the more deshielded (43, 44). Similarly, two distinct sets
of phenyl signals are expected. However, only two complex
peaks (2:3 intensity ratio) are found. The weaker one at
~8.0 ppm is due to the eight ortho protons, whereas the other
at ~7.5 ppm includes the meta and para protons of both
types of rings. This assignment is based on a comparison
with ReOX2(OPh)(dppe), where one set of phosphine aro-
matic signals is displaced by the anisotropic effect of the
phenoxo ring.
The ReOX2(OR)(dppe) compounds were obtained by con-
trolled alcoholysis of ReOX3(dppe). For R = Me, Et, and Pr,
refluxing for several days in the neat alcohol gave good
yields of pure products. For the higher-boiling alcohols
cyclohexanol and ethylene glycol, decomposition took place
upon refluxing in the pure liquid and the reaction was car-
ried out in a 4:1 acetone–alcohol mixture. With phenol, a
solid, the reaction was run in benzene.
In all cases, halide substitution occurs only at the position
trans to the Re=O bond. This is supported by the presence of
only one singlet in the 31P{1H} spectra, downfield from that
of the corresponding trihalo complex. The strong ν(Re=O)
vibration appears at ~940 cm–1. The ethoxo and 2-
hydroxyethoxo compounds also give a very strong band at
~910 cm–1 involving an O-CH2 deformation mode (30). The
FAB+ spectra generate characteristic patterns for the parent
ion and other fragments obtained by removing X or OR. The
1H aromatic regions of the alkoxo compounds are similar to
those of the trihalo systems, except that the ortho signals are
displaced downfield slightly (8.1–8.2 ppm). For
ReOX2(OPh)(dppe), one of the two overlapping ortho sig-
nals is displaced upfield (~8.00 ppm) by the anisotropy of
In most cases, disorder was observed in the alkoxo group,
phosphine phenyl rings, or lattice solvent molecules. It was
modeled by means of two individual orientations, whose oc-
cupancies were first refined with a constraint applied to nor-
malize the sum of the occupancies to unity. For the final
cycles, these occupancies were rounded off to the nearest
0.05 and then fixed. The non-hydrogen atoms were usually
refined anisotropically, but the thermal parameters were re-
stricted (ISOR) so as to obtain reasonable ellipsoids. In se-
vere cases, the interatomic distances were simultaneously
constrained with DFIX. Details are provided in the CIF files.3
For ReOCl2(OPr)(dppe), the propyl unit adopted two con-
formations in a 80:20 ratio (Fig. S-1, supplementary mate-
rial).3 The major conformer (C61-C63) was stretched,
whereas the other (C81-C83) was bent. DFIX constraints
based on the average data found for propoxo groups in the
Cambridge Structural Database (41) (C-O (1.42 Å), C-C
(1.45 Å), O•••Cβ (2.39 Å), Cα•••Cγ (2.41) Å) were applied.
Data were also collected on crystals of the related
ReOBr2(OPr)(dppe) complex. It is not isostructural with the
chloro compound, since it belongs to the triclinic space
group P1 and the unit cell (a = 9.848(3) Å, b = 17.602(6) Å,
c = 18.082(6) Å, α = 78.83(2)°, β = 89.98(3)°, γ = 89.03(2)°)
contains four molecules. The two independent molecules are
almost identical and differ very little from the chloro analog.
Even the packing patterns of the two compounds are very
close. However, the R factor could not be reduced below
0.094 and large correlations between the two independent
molecules led to high instability in the refinement. Subtle
twinning is suspected to be present and the structure of the
bromo compound is not described in detail here.
In the case of ReOBr2(OPh)(dppe), rotational disorder
about a P—C bond of dppe led to two orientations (60:40 ratio)
for one of its phenyl groups (Fig. S-2).3 With ReOCl2-
(OCy)(dppe), the cyclohexyl group in a chair conformation
was disordered over two orientations (60:40 ratio, Fig. S-2).3
In these cases, only ISOR restrictions were applied.
For Re(NC6H4Cl)Cl3(dppe), the crystal shape was difficult
to describe and the absorption correction was not totally sat-
isfactory. Nevertheless, the structure solved and refined nor-
mally.
The data for the triclinic ReCl4(dppe=O)•(CH3)2SO com-
pound were interpreted in the centric P1 space group, which
turned out to be the correct choice. The ReCl4(dppe=O) mole-
cule itself showed no disorder, but the lattice DMSO mole-
cule was found to be disordered over two interlocked
orientations (0.65:0.35, Fig. S-3).3 DFIX constraints based
on the average distances found for lattice DMSO in the
Cambridge Structural Database (41) (S=O (1.50 Å), C-S
(1.76 Å), C•••O (2.61 Å), C•••C (2.65 Å)) were applied.
3 Supplementary data may be purchased from the Directory of Unpublished Data, Document Delivery, CISTI, National Research Council
Canada, Ottawa, ON K1A 0S2, Canada (http://www.nrc.ca/cisti/irm/unpub_e.shtml for information on ordering electronically). CCDC
210020–210024) contain the crystallographic data for this manuscript. These data can be obtained, free of charge, via
www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ,
U.K.; fax +44 1223 336033; or deposit@ccdc.cam.ac.uk).
© 2003 NRC Canada

1332
Can. J. Chem. Vol. 81, 2003
1
Fig. 1. H NMR signals of the aromatic protons of the phos-
the phenoxo ligand and it can be safely assigned to ortho
protons, since JH-P couplings of ~11 Hz are observed (45).
3
phine groups in ReCl3(PPh3)(dppe=O).
Anisotropy also displaces the meta and para signals upfield:
they become resolved at ~7.3 (para) and ~7.2 ppm (meta),
whereas the meta and para multiplets for the rings on the
oxo side remain unresolved at ~7.5 ppm.
The alkoxo protons are shielded considerably compared
with those of the corresponding alcohol. This is consistent
with the alkoxo group lying trans to the oxo ligand, since it
was noticed (13) that these protons are deshielded when they
occupy a position cis to the oxo. For the 2-hydroxyethoxo
compound, the protons α to the coordinated oxygen are as-
sumed to be more shielded than the β protons, whose chemi-
cal shift returns to a normal value (3.34 ppm) (46) for free
ethylene glycol. The phenoxo ligand gives a doublet at
~5.65 ppm for the ortho protons and a multiplet including
the meta and para protons at ~6.60 ppm. Therefore, with re-
spect to uncoordinated phenol, whose signals occur at 6.77
(ortho and para) and 7.16 ppm (meta), coordination induces
upfield shifts whose amplitude decreased with the distance
from the O atom.
Preparation of Re(NR)Cl3(dppe)
of the bidentate Ph2PC2H4PPh2(O) ligand (dppe=O). The
phosphine oxide unit is believed to be derived from dppe in-
stead of PPh3, since the fragmentation products observed in
the FAB+ spectra correspond to the loss of PPh3, not Ph3PO.
Furthermore, upon aging, this compound gives a Re(IV)
compound ReCl4(dppe=O), whose crystal structure is de-
scribed hereafter. Even though the complex is paramagnetic,
Cotton et al. (32) first prepared the phenylimido com-
pound by reacting ReOCl3(dppe) with aniline. A similar pro-
cedure was used to obtain the p-chlorophenylimido complex
here. The original stereochemistry of the oxo complex is re-
tained, as evidenced from X-ray crystallography (see below)
and from NMR spectroscopy, namely the presence of only
one singlet in the 31P{1H} spectrum.
1
its H NMR spectrum could be fully analyzed.
1
We previously reported H NMR data of arylphosphine
Isolation of reduced species
complexes of Re(III) and Re(IV) (50, 51). In these com-
plexes, the aromatic phosphine signals usually occur be-
tween 10 and 20 ppm for the ortho protons, and between 7
and 11 ppm for the meta and para protons. As shown in
Attempts were made to obtain ReOCl3(dppe) by displac-
ing the phosphines of ReOCl3(PPh3)2. When the reaction
was run in CH2Cl2 at room temperature for 1 h, a beige solid
precipitated. The parent peak and fragmentation pattern in
the FAB+ spectrum were consistent with the presence of one
PPh3, one dppe, a ReCl3 unit, and one extra oxygen. The ab-
3
Fig. 1 and Table 2, this generalization is obeyed here. J
couplings typical of aromatic rings are observed and proton
positions around the rings are readily identified from multi-
plicities and integrations. A 2D experiment allowed us to
distribute the nine peaks between 12 and 7.5 ppm into three
groups of three, corresponding to PPh3 and to the two
inequivalent phenyl rings at the PPh2 end of the dppe=O
ligand. Two extra sets of three coupled signals are also
found at higher field for the aromatic protons of two phenyl
rings in the phosphine oxide group. In the latter sets, the
ortho protons appear at relatively high field (1.8 and –3.0 ppm)
and they are somewhat broadened. Finally, broad multiplets
with suitable intensities are observed for the aliphatic pro-
tons, two of which occur at extreme chemical shifts (40.5
and –8.5 ppm, respectively). These multiplets, which were
already complex for diamagnetic systems, showed extra
broadening because of the paramagnetism. They did not give
clear cross-peaks in the 2D spectrum and could not be as-
signed to individual protons.
1
sence of the strong IR ν(Re=O) band and the presence of H
NMR signals extending far outside the standard 0–10 ppm
range ruled out the possibility of a diamagnetic oxo-
rhenium(V) species. Therefore, rhenium-to-phosphorus oxy-
gen transfer had probably occurred, generating a Re(III) cen-
ter and a phosphine oxide, as observed before (47, 48). This
was supported by the strong IR bands appearing at 1058 and
1125 cm–1, one of which could be assigned to a ν(P=O) vi-
bration (49). In the structure II proposed, the oxygen is part
These available data and the crystallographic results
on ReCl4(dppe=O) suggest that the metal center is coordi-
nated to six ligands: three Cl, one PPh3, and one
Ph2PC2H4PPh2(O). Of the three possible isomers for an octa-
hedral ReX3L(L′-L′′) compound, only the fac isomer (II) is
consistent with the presence of four inequivalent phenyl
© 2003 NRC Canada

Baril-Robert and Beauchamp
1333
1
Table 2. H NMR chemical shifts (ppm) and coupling constants (Hz) for ReCl3(PPh3)(dppe=O) (in
DMSO-d6).
ortho
meta
para
3
3
3
PPh3
11.62 (d, 6H) J = 7.7
8.25 (t, 6H) J = 7.7
7.51 (t, 3H) J = 7.4
3
3
3
PPh2 (A)
PPh2 (B)
O=PPh2 (A)
O=PPh2 (B)
11.70 (d, 2H) J = 7.4
8.64 (t. 2H) J = 7.4
9.94 (t, 1H) J = 7.4
3
3
3
11.44 (d, 2H) J = 7.4
8.69 (t, 2H) J = 7.5
9.01 (t, 1H) J = 7.3
3
3
1.80 (br, 2H)
7.76 (t, 2H) J = 7.5
6.74 (t, 1H) J = 7.5
3
3
3
–2.99 (d, 2H) J = 7.9
3.69 (t, 2H) J = 7.9
4.99 (t, 1H) J = 7.5
C2H4
–8.47 (m br, 1H), 2.45 (m br, 1H), 4.81 (m br, 1H), 40.47 (m br, 1H)
1
rings in the dppe=O ligand, as revealed by H NMR spec-
troscopy.
since the corresponding distances are normal in the
ReOX3(dppe) complexes (Re—Cl = 2.38 Å, Re—Br =
2.52 Å) (31). The lengthening of the Re—X bonds may be
ascribed to the good π donor character of the end-on coordi-
nated OR group, whose donation into dxz and dyz makes
these orbitals less electrophilic and reduces their ability to
accept π-electron density from the equatorial halides. This
pattern is also observed for the trans bis-PPh3 complexes:
the cis Re—Cl bonds in ReOCl3(PPh3)2 (56) are shorter
(2.397(1) Å) than in the related oxo-alkoxo complexes
(2.41 Å) (31, 57, 58).
The oxo ligand tends to repel the adjacent bonds, resulting
in a displacement of the Re atom from the “equatorial” P2X2
plane on the oxo side, as evidenced from the mean of the
O=Re-Lcis angles being ~93° compared with ~88° for the
RO-Re-Lcis angles. The puckered conformation of the ring
creates systematic differences in the coordination geometry
of the two ends of dppe. The trans-XA-Re-PA units (involv-
ing the PA atom bonded to the methylene group on the oxo
side of the P2X2 plane) are more linear (alkoxo (~177°), halo
(~174°) (31)) than the trans-PB-Re-XB unit (alkoxo (~172°),
halo (~164°)). Also, our Re-XA distances are systematically
~0.025 Å longer than Re-XB, whereas the opposite trend is
noted, to a lesser extent, for the Re-P distances. This pattern
of bond lengths is not shared by the ReOX3(dppe) com-
pounds, however (31).
Crystal structures of the ReOX2(OR)(dppe) compounds
Since the Re-O-R angle in the known structures of this
type spans a broad range (125°–180°, based on 33 structures
in the Cambridge Structural Database) (41) and the oxygen-
to-rhenium π interactions could be sensitive to this factor,
structures with three different OR groups were studied: the
n-propoxo unit was considered as a “normal” group, the
cyclohexyl substituent was expected to be sterically demand-
ing, whereas the phenoxo group was used to probe the influ-
ence of an extended π system.
The three ReOX2(OR)(dppe) molecules (Fig. 2) consist of
octahedral molecules containing the trans-oxo-alkoxo unit
observed for the bis(diphenylphosphino)propane complex
(52). Selected distances and angles are listed in Table 3.
In the three compounds, the O=Re-O unit shows a signifi-
cant, but relatively small, deviation from linearity (171°–
173°), corresponding to small displacements of the two oxy-
gen donors to the dppe side. Despite the fact that the three R
groups have different sizes and (or) electronic properties, the
OR groups are all roughly end-on coordinated (Re-O-C ~
160°). Our Re=O and Re-OR distances in the two alkoxo
compounds are in good agreement with the mean values of
1.70(3) and 1.89(5) Å, respectively, obtained from the litera-
ture (41). For the phenoxo complex, our distances of
1.686(6) and 1.900(6) Å are not very different from those re-
ported for ReOCl3(PPh3)(L) with L = 2-diethylaminomethyl-
4-methyl-phenol (1.669(3) and 1.915(3) Å, respectively),
where the Re-O-C angle (166.0(3)°) is similar to ours (16).
In the two other structures published so far with simple
phenolates, the distances tend to be greater (1.666(6) and
1.969(5) Å, 1.703(6) and 1.989(6) Å, respectively) (53, 54),
which could be correlated with the pronounced bent in the
Re-O-C region (146.5(5)° and 137.7(6)°, respectively).
In the three known structures of 1:1 Re(V)-dppe com-
pounds (31, 55), the mean Re-P distance is 2.45(2) Å, the
bite angle P-Re-P is 82.9(6)°, and the P-C-C-P torsion angle
in the chelate ring is 58(4)°. Our results are in good agree-
ment with these data and the phenyl rings also adopt very
similar orientations. The Re-Cl (2.428–2.457 Å, σ =
0.001 Å) and Re-Br (2.570(1), 2.599(1) Å) distances are ap-
preciably greater than the mean values from the literature for
Re—X bonds cis to Re=O (Re—Cl = 2.38(3) Å, Re—Br =
2.51(3) Å; based on 130 and 14 structures, respectively)
(41). Although these bonds could be weakened by the phos-
phine trans influence, this cannot be the controlling factor,
The molecules are packed in the unit cell with normal van
der Waals contacts (Figs. S-4 and S-5).3
Crystal structure of Re(NC6H4Cl)Cl3(dppe)
The Re(NC6H4Cl)Cl3(dppe) molecule adopts the structure
shown in Fig. 3. The octahedral coordination of Re includes
a multiple-bonded 4-chlorophenylimido group with the trans
site filled by a chloro ligand, leading to the fac configuration
observed for the corresponding bis(diphenylphosphino)ferro-
cene complex (59). Selected interatomic distances and bond
angles are given in Table 4.
The Re=NR distance (1.707(5) Å) and the small departure
from linearity in the Re=N-C unit (171.5(5)°) correspond
well with the available data on arylimido complexes (mean
values: Re=N (1.716(15) Å), Re=N-C (172(5)°); based
on 42 structures) (41). A detailed comparison with the re-
lated compounds Re(NPh)Cl3(PPh3)2 (trans-P,P) (60),
Re(NPh)Cl3{bis(diphenylphosphino)ferrocene} (cis-P,P) (59),
and Re(NPh)Cl3(PMe3)2 (cis-P,P) (61) is provided in the
supplementary material (Table S-1).3 For the dppe ligand,
the P-Re-P “bite” angle and the ring conformation are close
to those of the above oxo-alkoxo compounds, but the Re—P
© 2003 NRC Canada

1334
Can. J. Chem. Vol. 81, 2003
Fig. 2. ORTEP drawings of ReOCl2(OPr)(dppe) (A), ReOBr2(OPh)(dppe) (B), and ReOCl2(OCy)(dppe) (C). For simplicity, only the
major conformations are shown and the hydrogens are omitted.
bond lengths are ~0.01 Å shorter here. The Re-Cl distances
range from 2.417(2) to 2.438(2) Å, in agreement with the
average value of 2.40(3) Å from the literature. As noted for
oxo-alkoxo complexes, the metal is displaced from the P2X2
plane on the nitrogen side by ~0.20 Å. Steric interactions
with the dppe phenyl rings probably explain that the depar-
ture from linearity in the trans RN=Re-Cl unit (168.0(2)°) is
slightly greater than commonly found (mean for 16 struc-
tures ~174(4)°). The phenyl rings at the PA end of dppe are
oriented in the same way (Re-PA-Ci-Co = 28° and –71°) as
in the oxo-alkoxo complexes, but those of the PB phosphine
group adopt a different orientation corresponding to the op-
posite pattern of torsion angles (Re-PB-Ci-Co = –67° and
15°, respectively). This is likely due to stacking interactions
between the C41-C46 dppe group and the phenylimido ring,
roughly parallel to the Re—PA bond. This orientation, also
observed in the phosphinoferrocene complex, positions the
rings nearly parallel to each other (dihedral angle =
24.3(4)°) with a relatively small mean inter-ring distance of
3.6 Å. The complexes are packed individually in the unit
cell (Fig. S-6, supplementary material),3 where the aromatic
rings participate in various stacking interactions.
The Re(NC6H4Cl)Cl3(dppe) molecule is compared to the
isoelectronic ReOCl3(dppe) compound in Table 4. In the
imido complex, the Re—Clcis bonds are ~0.05 Å longer,
whereas the Re—P bonds are ~0.04 Å shorter. This is con-
sistent with the fact that the donation of the imido group
π
into the metal dxz and dyz orbitals is greater that for the oxo
ligand. This should strengthen the bonds with the π-acceptor
phosphine, but weaken the Re—X bonds, since the halogens
are competing with the oxo or imido ligands for π donation
into the same metal orbitals. Interestingly, the orientations of
the dppe rings in ReOCl3(dppe) do not follow the pattern
found for Re(NC6H4Cl)Cl3(dppe) and the ReO(OR)Cl2(dppe)
compounds. This conformational change has no appreciable
effect on the coordination geometry.
ReCl4(dppe=O)•(CH3)2SO
In the ReCl4(dppe=O) complex shown in Fig. 4, the
Re(IV) center possesses an octahedral coordination. The
© 2003 NRC Canada

Baril-Robert and Beauchamp
Table 3. Distances (Å) and angles (°) in the ReOX2(OR)(dppe) compounds.a
1335
R
X
Propyl
Cl
Cyclohexyl
Cl
Phenylb
Br
Bond distances (Å)
Re=O
Re—OR
1.698(3)
1.885(3)
2.4260(10)
1.700(2)
1.877(2)
2.4305(9)
1.686(6)
1.900(6)
2.432(2)
Re—PA
Re—PB
Re—XA
Re—XB′
2.4320(10)
2.4475(10)
2.4280(11)
2.4364(8)
2.4570(10)
2.4296(9)
2.446(3)
2.5990(11)
2.5700(11)
Bond angles (°)
PA-Re-PB
83.26(3)
87.57(4)
83.19(3)
87.11(3)
83.59(9)
87.41(4)
XA-Re-XB
O=Re-OR
PA-Re-XA
172.57(14)
176.37(4)
172.87(11)
177.03(3)
171.8(3)
176.81(7)
PB-Re-XB
PA-Re-XB
PB-Re-XA
O=Re-PA
O=Re-PB
O=Re-XA
O=Re-XB
RO-Re-PA
RO-Re-PB
RO-Re-XA
RO-Re-XB
Re-O-C
170.61(4)
95.96(4)
172.93(3)
95.45(3)
94.09(3)
86.27(9)
91.52(9)
95.02(9)
95.32(9)
87.95(8)
83.65(8)
90.57(8)
89.37(8)
172.65(7)
94.92(6)
93.82(7)
86.7(2)
93.13(4)
86.29(10)
92.49(11)
96.80(11)
94.17(10)
87.89(9)
90.1(2)
95.2(2)
97.0(2)
87.50(19)
83.47(19)
90.36(19)
89.28(18)
162.3(5)
82.24(10)
88.38(10)
91.36(10)
160.8(5) (80%)c
159.5(9) (20%)c
58.9(3)
153.6(3) (60%)c
157.0(6) (40%)c
57.1(3)
PA-C-C-PB
C-C-PA-Ci
57.9(9)
–168.3(3)
–168.7(2)
–171.7(7)
79.7(3)
81.6(3)
80.2(3)
82.3(3)
76.7(7)
84.3(8)
–166.8(9) (60%)c
–185.4(7) (40%)c
14.2(10)
C-C-PB-Ci
–169.2(3)
–169.2(2)
Re-PA-Ci-Co
Re-PB-Ci-Co
34.3(5)
26.6(4)
–60.7(4)
–16.5(4)
–64.8(3)
–22.0(4)
–57.5(7)
–27.3(10)
75.8(5)
69.0(4)
88(2) (60%)c
59.0(2) (40%)c
0.1012(10)
Re•••PPXX
0.1053(6)
0.0882(4)
aPA bonded to the CH2 group on the oxo side of the P2X2 plane, PB bonded to the CH2 group on the OR side, XA
trans to PA, XB trans to PB.
bTorsion angles calculated for the (–x, –y, –z) molecule for consistency.
cValues corresponding to the two disordered orientations, occupancies within round brackets.
bidentate ligand is actually Ph2PC2H4PPh2(O), which is a
dppe ligand in which one end has been oxidized to phos-
phine oxide. A similar arrangement was observed for
ReCl4(dppm=O). Selected interatomic distances and bond
angles are given in Table 5.
The angles in the octahedron are close to ideality. The
largest deviation, found for the mutually trans Re—Cl
bonds, reflects a small displacement for the chlorines to-
wards the chelate ring. The Re-ligand distances compare
well with the literature results. For mutually trans Re—Cl
bonds, the distances observed here are close to the mean
value 2.336(14) Å determined from 21 Re(IV) structures
(41). When the chloro ligand sits trans to a different donor
atom, the mean value from the literature (23 structures) is
similar (2.325(30) Å), but it covers a broader range, mainly
because it is sensitive to the trans influence of the opposite
ligand. In our structure, the Re—Cl bond trans to oxygen
(2.319(2) Å) is appreciably shorter than the one trans to the
phosphine group (2.335(2) Å), whose trans influence is
large.
© 2003 NRC Canada

1336
Can. J. Chem. Vol. 81, 2003
Fig. 3. ORTEP drawing of fac-Re(NC6H4Cl)Cl3(dppe). The hy-
Table 4. Distances (Å) and angles (°) in fac-Re(NC6H4Cl)Cl3(dppe)
drogens are omitted for simplicity.
and ReOCl3(dppe).
NC6H4Cla
Ob
Bond distances (Å)
Re=NR
Re—Cltrans
1.707(5)
2.420(1)
1.680
2.429
Re—PA
Re—PB
Re—ClA
Re—ClB
2.421(1)
2.414(2)
2.438(2)
2.417(2)
2.463
2.445
2.377
2.380
Bond angles (°)
PA-Re-PB
83.37(5)
85.92(6)
168.0(2)
174.96(5)
165.56(6)
96.56(5)
92.92(5)
90.5(2)
83.3
ClA-Re-ClB
RN=Re-Cltrans
PA-Re-ClA
PB-Re-ClB
PB-Re-ClA
PA-Re-ClB
N-Re-PA
86.3
163.0
173.2
164.5
92.9
95.9
84.3
N-Re-PB
90.6(2)
91.4
N-Re-ClA
94.6(2)
101.4
103.9
82.2
N-Re-ClB
103.4(2)
83.92(5)
78.19(5)
Fig. 4. ORTEP drawing of ReCl4(dppe=O). The hydrogens are
omitted for simplicity.
Cltrans-Re-PA
Cltrans-Re-PB
76.8
Cltrans-Re-ClA
Cltrans-Re-ClB
91.13(6)
87.56(6)
91.5
87.7
Re=N-C
P-Re•••Ci-Co (imido)
171.5(5)
3.9(6)
—
—
PA-C-C-PB
C-C-PA-Ci
55.7(6)
59.8
–162.9(5)
–163.7
89.2(5)
77.9(5)
81.8
76.4
C-C-PB-Ci
–175.1(5)
28.3(6)
–173.7
54.3
Re-PA-Ci-Co
Re-PB-Ci-Co
Re•••PPXX
–70.8(6)
–67.0(6)
–58.0
–28.9
15.1(8)
0.2018(7)
74.2
0.226
aTorsion angles for Re(NC6H4Cl)Cl3(dppe) calculated for the
(–x, –y, –z) molecule for consistency with Table 3.
bReference 31.
In the unit cell (Fig. S-7),3 the molecules form layers par-
allel to the a axis and oriented along the cell diagonal. Be-
tween these layers, the superposition of DMSO molecules
fills channels extending along the cell a axis.
ternal strain. The very small angle for the dppm=O complex
reflects the high constraint for a five-membered chelate ring.
One of the Re-Cl distances (2.215 Å) reported for the
dppm=O complex is extremely small. This bond, trans to the
phosphine oxide group, seems to be ~0.1 Å shorter than the
corresponding bond in our dppe=O complex. Partial substi-
tution of the Re(IV) complex ReCl4(dppm=O) in the crystal
by Re(V) molecules mer-ReOCl3(dppm=O) (63, 64) could
explain this “apparent” shortening, since the oxo ligand
would be trans to the phosphine oxide in the latter species.
Such solid solutions of “quasi-isostructural” oxo-Re(V) and
chloro-Re(IV) complexes have already been detected (65)
and in the resulting average structure, the “Re—Cl(O)” bond
Comparisons are made in Table 5 with ReCl4(dppm=O)
(62), which differs from our compound by the size of the
chelate ring. The O-Re-P angle in the six-membered ring
with dppe=O (89.2(2)°) is much greater than that of the
dppm=O five-membered ring (80.7°). This leaves less space
for the opposite chlorines, which get closer together: the
Cl-Re-Cl angle is 93.46(9)° here, compared to 97.7° for the
dppm=O complex. Another difference is the Re-O=P angle,
which is 140.5(3)° here, but 130.1° in ReCl4(dppm=O). This
angle is known to be very flexible: it is ~155° for
monodentate O=PR3 ligands, but it can adjust readily to ex-
© 2003 NRC Canada

Baril-Robert and Beauchamp
1337
Fig. 5. Typical electronic spectra of the ReEX2(Y)(dppe) com-
Table 5. Distances (Å) and angles (°) in ReCl4(dppe=O) and
ReCl4(dppm=O).
plexes in DMSO.
ReCl4(dppe=O)
ReCl4(dppm=O)a
Bond distances (Å)
Re—O(1)
Re—P(2)
Re—Cl(1)
Re—Cl(2)
Re—Cl(3)
Re—Cl(4)
O(1)—P(1)
2.039(5)
2.512(2)
2.319(2)
2.353(2)
2.312(2)
2.339(2)
1.517(6)
2.087(7)
2.467(3)
2.215(5)
2.353(3)
2.340(3)
2.361(3)
1.523(8)
Bond angles (°)
Cl(3)-Re-Cl(4)
P(2)-Re-Cl(1)
O(1)-Re-Cl(2)
O(1)-Re-P(2)
173.61(8)
176.37(7)
178.02(15)
89.2(2)
173.8
169.1
173.2
80.7(2)
97.7(1)
88.6(2)
87.1(2)
86.7(2)
93.0(1)
86.8(1)
92.6(1)
95.3(2)
90.9(2)
90.1(1)
89.4(1)
130.1(4)
—
Fig. 6. Ground and low-energy excited states for the Re(V) d2
systems: singlet ground state (S), first triplet excited state (T*),
first singlet excited state (S*), second singlet excited state (S**).
Cl(1)-Re-Cl(2)
O(1)-Re-Cl(2)
O(1)-Re-Cl(3)
O(1)-Re-Cl(4)
P(2)-Re-Cl(1)
93.46(9)
87.2(2)
86.3(2)
88.1(2)
90.16(7)
88.69(7)
87.98(8)
91.76(9)
93.70(9)
90.90(8)
92.08(9)
140.5(3)
–78.8(6)
15.1(6)
134.8(5)
–106.3(6)
–1.2(3)
117.9(3)
–116.0(3)
55.0(7)
P(2)-Re-Cl(3)
P(2)-Re-Cl(4)
Cl(3)-Re-Cl(1)
Cl(1)-Re-Cl(4)
Cl(3)-Re-Cl(2)
Cl(4)-Re-Cl(2)
Re-O(1)-P(1)
P(1)-C(11)-C(12)-P(2)
Re-O(1)-P(1)-C(11)
Re-O(1)-P(1)-C(21)
Re-O(1)-P(1)-C(31)
O(1)-Re-P(2)-C(12)
O(1)-Re-P(2)-C(41)
O(1)-Re-P(2)-C(51)
Re-P(2)-C(41)-Cortho
generate under D4h symmetry, become inequivalent. For this
point group symmetry, the interbond orbital in the xy plane
–2.5
–123.8
115.1
22.8
139.2
is labeled dx2
(instead of dxy), and for consistency with
− y2
the recent study on isoelectronic oxo-molybdenum(IV) sys-
tems by Da Re and Hopkins (66), the dyz will be assumed to
lie at lower energy than dxz. The two absorption features in
–96.5
–9.0
Fig. 5 are assigned to the two spin-allowed singlet–singlet
2
transitions (dx
) → (dx2 )1(dyz)1 [S → S*] and (d
)2 →
x2 − y2
2 − y2
− y2
Re-P(2)-C(51)-Cortho
O(1)-P(1)-C(21)-Cortho
O(1)-P(1)-C(31)-Cortho
69.5(7)
–8.2(7)
–71.9
63.3
5.6
(dx2 )1(dxz)1 [S → S**], respectively (Fig. 6). Therefore,
− y2
the position of these bands can be used to estimate the two
HOMO–LUMO energy gaps. Singlet–triplet processes [S →
T*] could in principle take place, but there is no evidence
for these much weaker spin-forbidden transitions in the
spectra recorded here.
–23.8(8)
aReference 62 esds not calculated for all entries.
seems to be shortened with respect to a normal Re—Cl
bond, but to be stretched when compared to an Re=O bond.
Two well-differentiated components (12 000 and
16 400 cm–1) are visible for ReOCl3(dppe). Replacing Cl by
Br produces similar red shifts on both (11 300 and
15 500 cm–1). Upon substitution of the oxo ligand by a
phenylimido group in Re(NPh)Cl3(dppe), the two compo-
nents are not equally displaced: the low-energy component
undergoes a very small red shift of ~900 cm–1, whereas the
shift on the other is greater (~3000 cm–1). This could be con-
nected with the different symmetries of these two ligands.
For the cylindrical oxo ligand, the px and py orbitals are
equally available for π interacting with the metal dxz and dyz
orbitals. In contrast, for the imido group, the phenyl ring
renders these orbitals inequivalent, since only one is partici-
pating in the aromatic π system. Consequently, π interactions
with the metal dxz and dyz orbitals being no longer equal, the
transitions to these orbitals are anticipated to be differently
affected.
Electronic spectra and DFT calculations
The low-energy portions (8000–25 000 cm–1) of the ab-
sorption spectra of representative compounds are shown in
Fig. 5. Spectra taken with crystalline ReOX3(dppe) and
ReOX2(OR)(dppe) compounds confirmed that the species in
the solid is retained in DMSO. In all cases, this spectral re-
gion shows two major components. The data are listed in Ta-
ble 6.
Strictly speaking, the molecules examined here have no
symmetry, but if chelate ring puckering and other minor dis-
tortions are ignored,
a mirror plane containing the
Re=O(NR) bond bisects the P-Re-P angle, so that the metal
environment has an approximate Cs symmetry. Under this
low symmetry, the antibonding dxz/dyz orbitals, which are de-
© 2003 NRC Canada

1338
Can. J. Chem. Vol. 81, 2003
Table 6. [S → S*] and [S → S**] of the low-energy components of the electronic spectra.
X = Cl
X = Br
[S → S*]
12.0(13)
17.0(11)
16.4(12)
16.5(10)
~15.1(sh)a
11.1(118)
11.5(115)
[S → S**]
[S → S*]
[S → S**]
ReOX3(dppe)
16.4(25)
20.1(23)
20.0(21)
20.0(22)
—
11.3(13)
16.8(14)
16.4(15)
16.6(16)
—
15.5(21)
20.2(34)
20.0(33)
20.0(35)
—
ReO(OMe)X2(dppe)
ReO(OEt)X2(dppe)
ReO(OPr)X2(dppe)
ReO(OPh)X2(dppe)
Re(NPh)X3(dppe)
Re(NC6H4Cl)X3(dppe)
13.4(74)
—
—
13.5(70)
—
—
aThe position of the shoulder produced by the lower-energy component was estimated; the higher energy component
was completely masked by a phenoxo strong band.
Table 7. Comparison of the experimental transition energies with the DFT-calculated values (cm–1 × 10–3) for the E=ReX2Y(dppe)
compounds.
[S → S*]
[S → S**]
Experimental
(cm–1 × 10–3)
Calculated
Experimental
(cm–1 × 10–3)
Calculated
E
X
Y
(cm–1 × 10–3)
(cm–1 × 10–3)
O
O
O
O
Cl
Cl
Cl
Cl
OEt
OPh
Cl
16.4
15.1
12.0
—
9.42
8.41
6.78
5.95
20.0
—
16.4
—
10.21
8.71
8.25
6.63
OC6F5
O
O
Br
I
Cl
Cl
Br
I
Cl
Cl
11.3
—
11.1
11.5
6.63
6.15
5.91
5.91
15.5
—
13.4
13.5
7.66
7.09
7.55
6.94
NPh
NC6H4Cl
NC6F5
Cl
Cl
—
4.74
—
6.54
Substitution of the trans halogen of ReOX3(dppe) by an
-OR group has more drastic effect. In the
Fig. 7. Variation of the DFT-calculated energy gap as a function
of the transition energy obtained experimentally from the spectra
for each of the two transitions.
a
ReOX2(OR)(dppe) compounds, both components undergo
relatively large blue shifts to ~16 600 and 20 000 cm–1, re-
spectively, and the separation (~3000 cm–1) is a little smaller
than in the tris-halogen complexes (~4000 cm–1). Again,
these unequal variations can be ascribed to the noncylin-
drical symmetry of the bent alkoxo group. One of the oxy-
gen p orbitals (perpendicular to the Re-O-C plane) remains
suitably oriented for optimal π interacting with one of the
d orbitals, but partial hybridization will decrease the partici-
pation of the other p orbital (in the Re-O-C plane) in π inter-
actions.
To understand the behavior of these oxo- and arylimi-
dorhenium(V) systems and orient our research strategies,
theoretical calculations were performed with the
PCSPARTAN PRO software (67). A simplified model was
used to reduce computation time. Structure optimization on
molecules containing dppe itself were time consuming, be-
cause adjusting the orientation of the four phenyl rings re-
quired many refinement cycles and convergence was not
always reached. Since the comparisons to be made always
involve the same diphosphine, the calculations were run us-
ing bis(dimethylphosphino)ethane (dmpe), in which the sub-
stituents are “cylindrical” methyl groups. On the other hand,
since PCSPARTAN does not include rhenium parameters for
DFT calculations, technetium was used in the model. The
geometry and orbital energies were obtained by DFT calcu-
lations based on the generalized gradient approximation uti-
lizing the functionals proposed by Becke and Perdew (67).
The BP86 (self-consistent model) and DN* (double numeri-
cal polarization) options were used. The calculated energies
are listed in Table 7 and they are plotted against the experi-
mental results in Fig. 7.
For each of the two transitions, there is an approximate
linear relationship between the energies determined from the
spectra and those predicted from the calculations. This sug-
gests that the calculations reflect in a satisfactory manner the
factors affecting the energy levels in these systems. The re-
© 2003 NRC Canada

Baril-Robert and Beauchamp
1339
duction of the transition energy upon substitution of the oxo
ligand by an arylimido moiety and the energy increase by re-
placing the trans halogen with an -OR group are reproduced
well by the calculations. Even the small red shift resulting
from the substitution of chlorine by bromine is consistent
with the DFT predictions. However, this approach was un-
able to explain various fine features, for instance the fact
that the difference between the two transition energies does
not vary smoothly. The calculations did not reveal any sub-
stantial effect of the orientation of the noncylindrical
arylimido or -OR groups on the transition energies and it is
not clear whether orientation is actually unimportant or
masked by model simplification or other weaknesses in our
approach. The fact that the calculations were actually carried
out on Tc complexes could introduce some distortions when
the results are applied to Re compounds, the more so that
spin-orbit coupling, which cannot be taken into account in
PCSPARTAN, could play a role. Another weakness of our
results is that significant geometrical distortions, namely
nonlinearity in the RN(O)=Re-L unit and displacement of
the Re atom above the equatorial plane, were not accurately
reproduced in the converged model and this could introduce
small distortion in the orbital energy pattern. Nevertheless,
the method is believed to reproduce the general absorption
features sufficiently well to be useful in the orientation of
future studies.
possible strategy to reduce further transition energy. Pre-
parative work is underway to test these predictions.
Acknowledgments
We wish to thank the Natural Sciences and Engineering
Research Council of Canada (NSERC) for financial support,
Dr. M. Simard and F. Bélanger-Gariépy for assistance in the
interpretation of the X-ray diffraction data, C. Tessier and
Professor F. D. Rochon for collecting an X-ray data set on
the diffractometer at the Université du Québec à Montréal,
and M. Martinho and M. Dussault for their participation in
the preparative work.
References
1. M. Nicolini and U. Mazzi. Technetium, rhenium and other
metals in chemistry and nuclear medicine 5. Servizi Grafici
Editoriali, Padova, Italy. 1999.
2. J.R. Dilworth and S.J. Parrott. Chem. Soc. Rev. 27, 43 (1998).
3. S.S. Jurisson and J.D. Lydon. Chem. Rev. 99, 2205 (1999).
4. S. Liu and D.S. Edwards. Chem. Rev. 99, 2235 (1999).
5. C. Tessier, A.L. Beauchamp, and F.D. Rochon. Inorg. Chem.
41, 6527 (2002).
6. C. Tessier, A.L. Beauchamp, and F.D. Rochon. J. Inorg.
Biochem. 85, 77 (2001).
7. S. Fortin and A.L. Beauchamp. Inorg. Chem. 40, 105 (2001).
8. S. Fortin and A.L. Beauchamp. Inorg. Chem. 39, 4886 (2000).
9. S. Fortin and A.L. Beauchamp. J. Chem. Soc. Dalton Trans.
3520 (2001).
Concluding remarks
10. S. Bélanger and A.L. Beauchamp. Inorg. Chem. 36, 3640
(1997).
11. S. Bélanger, S. Fortin, and A.L. Beauchamp. Can. J. Chem.
75, 37 (1997).
12. S. Bélanger and A.L. Beauchamp. Inorg. Chem. 35, 7836
(1996).
On the basis of the two low-energy UV–vis absorptions
observed for the series of ReEX2Y(dppe) compounds (with
E = O or NR, X = Cl or Br, Y = Cl, Br, OR), it was estab-
lished that the energies of the spin-allowed transitions
between the nonbonding interbond orbital (dx2 ) and the
− y2
π-antibonding dxz/dyz orbitals follow the order RN=Re-X <
O=Re-X < O=Re-OR. Replacing chlorine by bromine re-
duces the energies slightly, whereas the lengthening of the
alkyl chain in the -OR group (R = Me, Et, Pr, -C2H4OH) had
no detectable effect. We expected the bulky cyclohexyl
substituent would force the Re-O-R segment to be more lin-
ear, but these expectations did not materialize (see crystal
structure) and the absorption spectrum of this compound
does not differ from those of compounds with a smaller R
group. The aromatic group in ReOX2(OPh)(dppe) seems to
reduce the first transition energy slightly, but this is difficult
to evaluate, because the weak d–d transitions are masked
here by a strong allowed transition involving the phenoxo
group.
Although the DFT calculations cannot explain all the fine
details present in the data, the general trends observed
experimentally are reproduced sufficiently well to attribute
some predictive capability to this approach. Transition ener-
gies calculated for a few promising systems are included in
Table 7. The calculations suggest that incorporating strong
electro-attracting features into either the phenylimido or the
phenoxo group should shift the transitions to lower energies.
On this basis, perfluorophenoxo complexes ReOX2-
13. X. Couillens, M. Gressier, R. Turpin, M. Dartiguenave, Y.
Coulais, and A.L. Beauchamp. J. Chem. Soc. Dalton Trans.
914 (2002).
14. X. Couillens, M. Gressier, M. Dartiguenave, S. Fortin, and
A.L. Beauchamp. J. Chem. Soc. Dalton Trans. 3032 (2002).
15. F. Connac, Y. Lucchèse, M. Dartiguenave, and A.L.
Beauchamp. Inorg. Chem. 36, 256 (1997).
16. F. Connac, Y. Lucchese, M. Dartiguenave, Y. Coulais, and
A.L. Beauchamp. Inorg. Chim. Acta, 295, 209 (1999).
17. F. Connac, Y. Lucchese, M. Gressier, M. Dartiguenave, and
A.L. Beauchamp. Inorg. Chim. Acta, 304, 52 (2000).
18. F. Loiseau, F. Connac, Y. Lucchese, M. Dartiguenave, S.
Fortin, A.L. Beauchamp, and Y. Coulais. Inorg. Chim. Acta,
306, 94 (2000).
19. J.B. Arterburn and S.L. Nelson. J. Org. Chem. 61, 2260
(1996).
20. J.B. Arterburn and M.C. Perry. Tetrahedron Lett. 37, 7941
(1996).
21. J.B. Arterburn, M.C. Perry, S.L. Nelson, B.R. Dible, and M.S.
Holguin. J. Am. Chem. Soc. 119, 9309 (1997).
22. H.Q.N. Gunaratne, M.A. McKervey, S. Feutren, J. Finlay, and
J. Boyd. Tetrahedron Lett. 39, 5655 (1998).
23. X. Shan, A. Ellern, I.A. Guzei, and J.H. Espensen. Inorg.
Chem. 42, 2362 (2003).
(OC6F5)(dppe)
or
perfluorophenylimido
compounds
24. J.R. Winkler and H.B. Gray. Inorg. Chem. 24, 346 (1985).
25. C. Savoie, C. Reber, S. Bélanger, and A.L. Beauchamp. Inorg.
Chem. 34, 3851 (1995).
Re(NC6F5)X3(dppe) could be interesting candidates. Re-
placing chlorine or bromine by iodine also seems to be a
© 2003 NRC Canada

1340
Can. J. Chem. Vol. 81, 2003
26. C. Savoie and C. Reber. Coord. Chem. Rev. 171, 387 (1998).
27. C. Savoie and C. Reber. J. Am. Chem. Soc. 122, 844 (2000).
28. J. Landry-Hum, V. Tessier, M. Ernzerhof, and C. Reber.
Coord. Chem. Rev. 233–234, 63 (2002).
29. U. Oetliker, C. Savoie, S. Stanislas, C. Reber, F. Connac, A.L.
Beauchamp, F. Loiseau, and M. Dartiguenave. J. Chem. Soc.
Chem. Commun. 657 (1998).
30. J. Chatt and G.A. Rowe. J. Chem. Soc. 4019 (1962).
31. V.S. Sergienko, M.A. Porai-Koshits, V.E. Mistryukov, and K.V.
Kotegov. Koord. Khim. 8, 230 (1982).
32. F.A. Cotton, N.F. Curtis, and W.R. Robinson. Inorg. Chem. 4,
1696 (1965).
47. S. Banerjee, S. Bhattacharyya, B.K. Dirghangi, M. Menon, and
A. Chakravorty. Inorg. Chem. 39, 6 (2000).
48. R.R. Conry and J.M. Mayer. Inorg. Chem. 29, 4862 (1990).
49. K. Nakamoto. Infrared and Raman spectra of inorganic coordi-
nation compounds. 4th ed. John Wiley and Sons, New York.
1986. p. 341.
50. C. Pearson and A.L. Beauchamp. Can. J. Chem. 75, 220 (1997).
51. C. Pearson and A.L. Beauchamp. Inorg. Chim. Acta, 237, 13
(1995).
52. C. Kremer, M. Rivero, E. Kremer, L. Suescun, A.W. Mombru,
R. Mariezcurrena, S. Dominguez, A. Mederos, S. Midollini,
and A. Castineiras. Inorg. Chim. Acta, 47, 294 (1999).
53. J.R. Dilworth, D.V. Griffiths, S.J. Parrott, and Y. Zheng. J.
Chem. Soc., Dalton Trans. 2931 (1997).
33. SMART. Release 5.059 [computer program]. Bruker AXS Inc.,
Madison, Wis. 1999.
34. SAINT. Release 6.06 [computer program]. Bruker AXS Inc.,
Madison Wis. 1999.
35. CAD-4 software. Version 5.0 [computer program]. Enraf-
Nonius, Delft, The Netherlands. 1989.
54. K.J.C.V. Bommel, W. Verboom, H. Kooijman, A.L. Spek, and
D.N. Reinhoudt. Inorg. Chem. 37, 4197 (1998).
55. R. Chiozzone, R. Gonzalez, C. Kremer, G. De Munno, and J.
Faus. Inorg. Chim. Acta, 325, 203 (2001).
36. XSCANS [computer program]. Bruker Analytical X-Ray Sys-
tems, Madison, Wis. 1995.
56. A.M. Lebuis and A.L. Beauchamp. Can. J. Chem. 71, 441
(1993).
37. G.M. Sheldrick. SHELXTL. Release 5.10 [computer program].
Bruker AXS Inc., Madison, Wis. 1997.
57. R. Graziani, U. Casellato, R. Rossi, and A. Marchi. J.
Crystallogr. Spectrosc. Res. 15, 573 (1985).
38. XPREP. Release 5.10 [computer program]. Bruker AXS Inc.,
Madison, Wis. 1997.
58. S. Abram, U. Abram, E. Schulz-Lang, and J. Strähle. Acta
Crystallogr. C51, 1078 (1995).
39. G.M. Sheldrick. SHELXS-97 [computer program]. University
of Göttingen, Göttingen, Germany. 1997.
59. S.W. Lee and N.-S. Choi. Acta Crystallogr. Sect. C: Cryst.
Struct. Commun. 55, 2018 (1999).
40. G.M. Sheldrick. SHELXL-96 [computer program]. University
of Göttingen, Göttingen, Germany. 1996.
60. B.-G. Park, Y.-W. Kim, G.-S. Jung, H.-S. Park, and S.W. Lee.
Bull. Korean Chem. Soc. 16, 835 (1995).
41. Cambridge Structural Database. Version 5.4, November 2002
release with February 2003 updates. Cambridge Crystallo-
graphic Data Centre, Cambridge, U.K.
42. B. Jezowska-Trzebiatowska, J. Hanuza, and M. Baluka.
Spectrochim. Acta, 27A, 1753 (1971).
61. Y.-W. Kim, J.-H. Jung, H.-S. Park, and S.W. Lee. Bull. Korean
Chem. Soc. 15, 891 (1994).
62. R. Rossi, A. Marchi, L. Marvelli, L. Magno, M. Peruzzini, U.
Casellato, and R. Graziani. Inorg. Chim. Acta, 204, 63 (1993).
63. K.V. Katti and C.L. Barnes. Inorg. Chem. 31, 4231 (1992).
64. K.-Y. Shih, P.E. Fanwick, and R.A. Walton. Inorg. Chim. Acta,
212, 23 (1993).
43. J.P. O’Neil, S.R. Wilson, and J.A. Katzenellenbogen. Inorg.
Chem. 33, 319 (1994).
44. M.S. Papadopoulos, M. Pelecanou, I.C. Pirmettis, D.M.
Spyriounis, C.P. Raptopoulou, A. Terzis, C.I. Stassinopoulou,
and E. Chioellis. Inorg. Chem. 35, 4478 (1996).
45. E. Vincent, L. Verdonck, and G.P.V.D. Kelen. Spectrochim.
Acta, 36A, 699 (1980).
65. K. Yoon, G. Parkin, and A.L. Rheingold. J. Am. Chem. Soc.
114, 2210 (1992).
66. R.E. Da Re and M.D. Hopkins. Inorg. Chem. 41, 6973 (2002).
67. PCSPARTAN PRO 1.0.6 [computer program]. Wavefunction
Inc., Irvine Calif. 2001.
46. H.E. Gottlieb, V. Kotlyar, and A. Nudelman. J. Org. Chem. 62,
7512 (1997).
© 2003 NRC Canada