Oxo-rhenium(V) complexes with 8-hydroxyquinoline derivatives

Article abstract of DOI:10.1139/v05-063

Compounds of the types ReOCl₂(L)(PPh₃) and ReOCl(L)₂ were prepared by reacting ReOCl₃(PPh ₃)₂ with 8-hydroxyquinoline (HL) and its 2-methyl, 2-chloro, 5-chloro, 5-nitro, 5,7-dichloro, 5,7-dibromo, and 5,7-diiodo derivatives. With the bulky 2-phenyl-8-hydroxyquinoline, only ReOCl ₂(L)(PPh₃) could be isolated, whereas the still bulkier 2-tert-butyl derivative did not react. For ReOCl₂(L)(PPh ₃), the coordination of the quinoline oxygen trans to the Re=O bond and the cis-dichloro arrangement in the equatorial plane were established from crystallographic studies on the 2-chloro and the 5,7-dibromo complexes. From the combined data for these various derivatives, the ¹H NMR signals could be fully assigned. With both series of compounds, a complex d-d absorption pattern is observed in the visible spectra, corresponding to the excitation of a d electron from the interaxial d orbital in the equatorial plane to the empty d_xz and d_yz orbitals, which are inequivalent in these low-symmetry systems. Deconvolution revealed the presence of two very weak low-energy components (～10 000 and ～12 000 cm^-1), which are assigned to the two expected singlet-triplet transitions, whereas two stronger bands at higher energy (～14 000 and ～17 000 cm_-1) originate from the two singlet-singlet transitions. These bands are not substantially displaced by substitution on the 8-hydroxyquinoline rings.

Full text of DOI:10.1139/v05-063

4
60
Oxo-rhenium(V) complexes with 8-
hydroxyquinoline derivatives
Olivier Sigouin and André L. Beauchamp
Abstract: Compounds of the types ReOCl (L)(PPh ) and ReOCl(L) were prepared by reacting ReOCl (PPh ) with 8-
2
3
2
3
3 2
hydroxyquinoline (HL) and its 2-methyl, 2-chloro, 5-chloro, 5-nitro, 5,7-dichloro, 5,7-dibromo, and 5,7-diiodo deriva-
tives. With the bulky 2-phenyl-8-hydroxyquinoline, only ReOCl (L)(PPh ) could be isolated, whereas the still bulkier 2-
2
3
tert-butyl derivative did not react. For ReOCl (L)(PPh ), the coordination of the quinoline oxygen trans to the Re=O
2
3
bond and the cis-dichloro arrangement in the equatorial plane were established from crystallographic studies on the 2-
1
chloro and the 5,7-dibromo complexes. From the combined data for these various derivatives, the H NMR signals
could be fully assigned. With both series of compounds, a complex d–d absorption pattern is observed in the visible
spectra, corresponding to the excitation of a d electron from the interaxial d orbital in the equatorial plane to the empty
dxz and dyz orbitals, which are inequivalent in these low-symmetry systems. Deconvolution revealed the presence of two
–
1
very weak low-energy components (~10 000 and ~12 000 cm ), which are assigned to the two expected singlet–triplet
–
1
transitions, whereas two stronger bands at higher energy (~14 000 and ~17 000 cm ) originate from the two singlet–
singlet transitions. These bands are not substantially displaced by substitution on the 8-hydroxyquinoline rings.
Key words: rhenium, quinoline, crystal structure, d–d electron transitions.
Résumé : Des composés de types ReOCl (L)(PPh ) et ReOCl(L) ont été préparés par réaction de ReOCl (PPh ) avec
2
3
2
3
3 2
la 8-hydroxyquinoline (HL) et ses dérivés 2-méthyl, 2-chloro, 5-chloro, 5-nitro, 5,7-dichloro, 5,7-dibromo et 5,7-diiodo.
Avec le ligand encombré 2-phényl-8-hydroxyquinoline, seul ReOCl (L)(PPh ) a pu être isolé, alors que le dérivé 2-tert-
2
3
butyl encore plus encombré n’a pas réagi. Pour ReOCl (L)(PPh ), la coordination de l’oxygène de la quinoline en trans
2
3
de la liaison Re=O et l’arrangement cis-dichloro dans le plan équatorial ont été déterminés au moyen d’études cristallo-
graphiques sur les complexes des ligands 2-chloro et 5,7-dibromo. L’ensemble des données pour ces différents dérivés
1
a permis de faire une attribution complète des signaux RMN H. Pour les deux séries de composés, un motif d’absorp-
tion d–d compliqué est observé dans le spectre visible, qui correspond à l’excitation d’un électron d depuis l’orbitale d
interaxiale dans le plan équatorial vers les orbitales vacantes dxz et d , qui ne sont pas équivalentes dans ces systèmes
yz
de basse symétrie. La déconvolution révèle la présence de deux composantes très faibles à basse énergie (~10 000 et
–
1
~
12 000 cm ), qui sont attribuées aux deux transitions singulet–triplet attendues, tandis que les bandes plus fortes à
–
1
énergie supérieure (~14 000 et ~17 000 cm ) proviennent de deux transitions singulet–singulet. Ces bandes ne sont pas
déplacées appréciablement par substitution sur le cycle de la 8-hydroxyquinoline.
Mots clés : rhénium, quinoline, structure cristalline, transitions électroniques d–d. Sigouin and Beauchamp 470
Introduction
sess a readily accessible high-spin excited state that could be
involved in spin crossover processes.
Systems giving rise to electronic transitions between low-
spin and high-spin states at low energy cost present consid-
erable interest, since they may provide a basis for develop-
ing molecular switches and devices for efficient information
In a previous study (1), the HOMO–LUMO gap was
found to be relatively small in ReOCl(P~O) compounds
2
–
containing bidentate mixed P~O ligands. To better under-
stand the factors affecting the HOMO–LUMO gap, we
recently examined various ReOX (dppe) and ReO(OR)X -
2
storage. The d oxo-rhenium(V) core is a stable unit in
3
2
which the ground state is low-spin and diamagnetic, since
(dppe) complexes, where the coordination sphere remained
the strong π bonds formed via the oxygen p and p orbitals
close to that of the ReOCl(P~O) species and the equatorial
x
y
2
raise the d and d orbitals (LUMO) above the other inter-
cis-P,P arrangement was provided by bis(diphenylphosphino)–-
xz
yz
axial d orbital (HOMO) in which the two d electrons are
coupled. However, if the HOMO–LUMO gap could be re-
duced and become comparable with the pairing energy, the
molecule could adopt a high-spin triplet ground state or pos-
ethane(dppe) (2). The presence of a strong π-donor RO
group trans to the Re=O bond in ReO(OR)X (dppe) was
2
found to produce an energy gap greater than those of the
ReOX (dppe) compounds, where the trans position is occu-
3
Received 6 October 2004. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 12 May 2005.
1
O. Sigouin and A.L. Beauchamp. Département de chimie, Université de Montréal, C.P. 6128, Succursale Centre-ville, Montréal,
QC H3C 3J7, Canada.
1Corresponding author (e-mail: andre.beauchamp@umontreal.ca).
Can. J. Chem. 83: 460–470 (2005)
doi: 10.1139/V05-063
© 2005 NRC Canada

Sigouin and Beauchamp
461
pied by a halogen. The present study was intended to deter-
mine to what extent the energy gap is sensitive to π
interactions with donor atoms in the equatorial plane. To this
end, we selected the 8-hydroxyquinoline ligand. Although
little is known about the influence of this ligand on d–d tran-
sitions, it was hoped that the Re(V) complexes could pro-
vide information about the interaction of its aromatic π
system with the interaxial d orbitals: when coordinated in
the equatorial plane, the π system should interact with the
Preparation of 2-chloro-8-hydroxyquinoline
This ligand was prepared by reacting POCl with 2,8-
dihydroxyquinoline obtained from Hqn via the N-oxide (5–
7).
In glacial acetic acid (25 mL) and 30% H O (10 mL)
were dissolved 4.00 g (27.6 mmol) of Hqn. The mixture was
heated at 70 °C for 16 h and after cooling to room tempera-
3
2
2
ture (RT), pH was adjusted to 12 with NH OH. The brown
4
solid was collected by filtration and extracted with a
hexane–acetone (10:1) mixture. Solvent evaporation yielded
yellow crystals of Hqn-N-oxide (1.47 g, yield 33%); mp
138 °C (lit. value (5) mp 138–140 °C).
d
and d orbitals, whereas coordination parallel to the
xz
yz
Re=O direction would generate interactions with the third
interaxial orbital.
We are reporting here the preparation and characterization
of ReOCl (L)(PPh ) and ReOCl(L) compounds with a se-
The Hqn-N-oxide (3.85 g, 23.9 mmol) was dissolved in
acetic anhydride and heated at 140 °C for 5 h. The solvent
was removed under reduced pressure and the dark-gray solid
2
3
2
ries of variously substituted deprotonated 8-hydroxyquinolines
–
(
L ) (I). The low-energy UV–vis spectra were found to be
was washed with ethanol to give light-yellow 8-acetoxy-2-
1
rather insensitive to ring substitution, but in contrast with the
dppe compounds studied earlier, all four expected low-
energy d–d bands could be unraveled from the complicated
spectral pattern observed.
hydroxyquinoline (1.90 g, yield 39%). H NMR (CDCl ,
3
ppm) δ: 10.17 (br s, 1H, OH), 7.80 (d, J = 9.5 Hz, H4), 7.44
(dd, J = 7.9 Hz, J = 1.2 Hz, H5), 7.37 (dd, J = 7.9 Hz,
1
2
1
J2 = 1.2 Hz, H7), 7.21 (t, J = 7.9 Hz, H6), 6.68 (d, J =
.6 Hz, H3), 2.51 (s, CH COO). Anal. calcd. for C H NO
3
9
3
11
9
H4
H5
(%): C 65.02, H 4.46, N 6.89; found: C 64.89, H 4.56, N
.83.
The 8-acetoxy-2-hydroxy derivative was dissolved in
methanol with 1 equiv. of K CO and the reaction mixture
6
H3
H6
2
3
was stirred at RT for 1 h. The solvent was evaporated, the
brown product was dissolved in water, and a 10% HCl solu-
tion was slowly added. Colorless 2,8-dihydroxyquinoline
H2
N
H7
1
was collected by filtration (1.40 g, yield 93%). H NMR
-
(CD OD, ppm) δ: 7.99 (d, J = 9.5 Hz, H4), 7.18 (dd, J =
3
1
O
7
.9 Hz, J = 1.2 Hz, H5), 7.12 (t, J = 7.8 Hz, H6), 7.02 (dd,
2
I
J = 7.7 Hz, J = 1.2 Hz, H7), 6.66 (d, J = 9.5 Hz, H3).
1 2
Anal. calcd. for C H NO (%): C 67.07, H 4.38, N 8.69;
9
7
2
found: C 66.31, H 4.22, N 8.40.
The dihydroxy compound was dissolved in excess POCl3
Experimental section
(
10 mL) and heated on a steam bath for 1 h. The mixture
Reactants and methods
was slowly transferred into a mixture of ice (75 g) and
NH OH (37 mL). The white solid was filtered and dissolu-
2
8
-Hydroxyquinoline (Hqn), most of its derivatives (2-
4
methyl, 5-monosubstituted and 5,7-disubstituted), KReO4,
and all other reactants were obtained from Aldrich and used
without further purification. 2-Phenyl- and 2-tert-butyl-8-
hydroxyquinoline were prepared as described by Delapierre
et al. (3). The preparation of 2-chloro-8-hydroxyquinoline is
described in the following. ReOCl (PPh ) (4) was prepared
tion in concd. HCl (75 mL) gave a clear yellow solution,
which was heated on a steam bath for 1 h. A Na CO solu-
2
3
tion (1.8 mol/L, 200 mL) was then added to precipitate the
light-brown solid, which was collected by filtration and puri-
fied by chromatography on silica gel with hexane – ethyl ac-
etate (80:20) as eluent. After solvent evaporation, the solid
was dissolved in methanol, water was added to cloudiness,
and the mixture was placed in the freezer for a few hours.
3
3 2
by the literature method.
IR spectra were recorded on PerkinElmer Spectrum One
–
1
The white crystalline solid (0.68 g, yield 41%) was filtered
(
4000–400 cm , KBr pellets) or Bio-Rad Excalibur series
1
–
1
and dried in vacuo. H NMR (CDCl ): see Table 1. Anal.
FTS 3000FX (4000–200 cm , CsI pellets). The NMR spec-
tra were measured with Bruker AV-300, AMX-300, AV-400,
3
calcd. for C H ClNO (%): C 60.19, H 3.37, N 7.80; found:
9
6
1
C 59.81, H 3.17, N 7.80.
or ARX-400 spectrometers. For the H spectra, the residual
solvent signal (DMSO-d (2.50 ppm), acetone-d (2.17 ppm),
6
6
CDCl (7.26 ppm), CD OD (3.31 ppm)) was used as internal
Preparation of the ReOCl (L)(PPh ) compounds
3
3
2
3
reference. H PO was used as an external reference (δ = 0)
These compounds were prepared following the method of
Mazzi et al. (8), but benzene was replaced by THF as the
solvent and reaction time was adjusted for each system.
In a typical run, ReOCl (PPh ) (0.12 mmol) was stirred
in a refluxing solution of the ligand (0.12 mmol) in THF
(10 mL). The colorless or yellow solution turned green first
3
4
3
1
for the P spectra. Electronic spectra were recorded as
DMSO, chloroform, or dichloromethane solutions in quartz
cells with a UV–vis–near-IR Cary 5E spectrometer. Elemen-
tal analyses were run at the Laboratoire d’Analyse
Élémentaire de l’Université de Montréal.
3
3 2
2
–
The neutral ligand is represented by Hqn and the coordinated deprotonated form by qn . The various derivatives are identified by an appro-
–
priate prefix showing the positions of the substituents: for instance, 2-Cl-qn for the 2-chloro-8-hydroxyquinoline anion.
©
2005 NRC Canada

4
62
Can. J. Chem. Vol. 83, 2005
and then became very dark. When the reaction was stopped,
the solvent was evaporated to dryness, the solid was
redissolved in dichloromethane (acetone for the 2-Cl-qn
compound), precipitated with diethyl ether, filtered, washed
with acetone and diethylether, and dried in vacuo. For the
less soluble Hqn complex, a precipitate was present at the
end of the reaction and solvent evaporation was unneces-
sary.
Details for each compound are provided in the following.
1
In all cases, the H NMR spectra showed multiplets between
7
.2 and 7.5 ppm for coordinated PPh . The remaining sig-
3
nals are listed in Table 1.
ReOCl (qn)(PPh )
2
3
The brown precipitate was filtered from the reaction mix-
–
1
ture after 19 h. Yield: 87%. IR (CsI, cm ): 976 (s) ν(Re=O),
3
1
3
–
47 (w), 293 (w) ν(Re-Cl). P NMR (DMSO-d ) (ppm) δ:
6
12.5 (s). Anal. calcd. for C H Cl NO PRe (%): C 47.72,
2
7
21
2
2
H 3.11, N 2.06; found: C 47.57, H 3.02, N 2.13.
ReOCl (2-Me-qn)(PPh )
2
3
Reaction time: 24 h. Green-brown solid. Yield: 77%. IR
–
1
31
(
CsI, cm ): 972 (s) ν(Re=O), 314 (w), 284 (w) ν(Re-Cl). P
NMR (DMSO-d , ppm) δ: –17.4 (s). Anal. calcd. for
6
C H Cl NO PRe (%): C 48.49, H 3.34, N 2.02; found: C
2
8
23
2
2
4
8.37, H 3.42, N 2.01.
ReOCl (2-Cl-qn)(PPh )
2
3
Reaction time: 2 h. Recrystallization in acetone gave a
green solid containing one lattice acetone molecule per for-
–
1
mula. Yield: 63%. IR (CsI, cm ): 977 (s) ν(Re=O), 324 (w),
3
1
2
81 (w) ν(Re-Cl). P NMR (DMSO-d , ppm) δ: –18.0 (s).
6
Anal. calcd. for C H Cl NO PRe (%): C 46.67, H 3.39, N
3
0
26
3
3
1
.81; found: C 46.71, H 3.28, N 1.75.
ReOCl (2-Ph-qn)(PPh )
2
3
Reaction time: 23 h. Green-brown solid. Yield: 42%. IR
–
1
31
(
CsI, cm ): 974 (s) ν(Re=O), 328 (w), 279 (w) ν(Re-Cl). P
NMR (DMSO-d , ppm) δ: –16.7 (s). Anal. calcd. for
6
C H Cl NO PRe (%): C 50.03, H 3.27, N 1.74; found: C
3
3
25
2
2
4
9.82, H 3.31, N 1.79.
ReOCl (5-Cl-qn)(PPh )
2
3
Reaction time: 24 h. Brown solid. Yield: 91%. IR
–
1
31
(
CsI, cm ): 980 (s) ν(Re=O), 345 (w), 296 (w) ν(Re-Cl). P
NMR (DMSO-d , ppm) δ: –12.1 (s). Anal. calcd. for
6
C H Cl NO PRe (%): C 45.42, H 2.82, N 1.97; found: C
2
7
20
3
2
4
5.21, H 2.75, N 2.02.
ReOCl (5,7-Cl -qn)(PPh )
2
2
3
Reaction time: 21 h. Brown solid. Yield: 85%. IR
–
1
31
(
CsI, cm ): 980 (s) ν(Re=O), 311 (w), 282 (w) ν(Re-Cl). P
NMR (DMSO-d , ppm) δ: –12.4 (s). Anal. calcd. for
6
C H Cl NO PRe (%): C 43.33, H 2.56, N 1.87; found: C
2
7
19
4
2
4
3.03, H 2.57, N 1.78.
ReOCl (5,7-Br -qn)(PPh )
2
2
3
Reaction time: 21 h. Brown solid. Yield: 83%. IR
–
1
31
(
CsI, cm ): 984 (s) ν(Re=O), 324 (w), 279 (w) ν(Re-Cl). P
NMR (DMSO-d , ppm) δ: –10.8 (s). Anal. calcd. for
6
C H Br Cl NO PRe (%): C 38.73, H 2.29, N 1.67; found:
2
7
19
2
2
2
C 37.43, H 2.07, N 1.66.
©
2005 NRC Canada

Sigouin and Beauchamp
463
1
ReOCl (5,7-I -qn)(PPh )
matography on silica gel. The H NMR data for these 2:1
species are included in Table 1.
2
2
3
Reaction time: 21 h. Brown solid. Yield: 93%. IR
–
1
31
(
CsI, cm ): 984 (s) ν(Re=O), 324 (w), 290 (w) ν(Re-Cl). P
NMR (DMSO-d , ppm) δ: –10.9 (s). Anal. calcd. for
6
Crystallographic measurements and structure
determination
Dark-green crystals of ReOCl (2-Cl-qn)(PPh ) were ob-
C H Cl I NO PRe (%): C 34.82, H 2.06, N 1.50; found: C
2
7
19 2 2
2
3
4.87, H 1.75, N 1.49.
2
3
tained by slow evaporation of an acetone solution. A crystal
was mounted on a Bruker SMART CCD 2K diffractometer
operating with graphite-monochromatized CuKα radiation
and controlled by the SMART software (10). Sets of 30 os-
cillation 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 orientations of
the crystal and detector so as to cover at least 95% of the re-
ciprocal sphere. At the end, the first 101 frames were
remeasured and the intensities of the 907 reflections did not
show any decomposition of the crystal. The frames were an-
alyzed with the SAINT software (11), which determined the
intensity and the position of each spot. Accurate cell param-
eters were obtained by least-squares refinement over the po-
sitions of the whole data set. Irregular crystal shapes
precluded the calculation of an analytical correction for ab-
sorption, but an empirical correction was applied using
SADABS (12).
ReOCl (5-NO -qn)(PPh )
2
2
3
Prepared according to Chen et al. (9). Deep-green solid.
–
1
Yield: 81%. IR (CsI, cm ): 977 (s) ν(Re=O), 324 (w), 285
3
1
(
w) ν(Re-Cl). P NMR (DMSO-d , ppm) δ: –9.3 (s). Anal.
6
calcd. for C H Cl N O PRe (%): C 44.76, H 2.78, N 3.87;
2
7
20
2
2
4
found: C 44.71, H 2.78, N 3.87.
Preparation of the ReOCl(L) compounds
2
These compounds were obtained according to Mazzi et al.
8), but the solvent used was THF instead of benzene. In a
(
typical run, ReOCl (PPh ) (500 mg, 0.60 mmol) was added
3
3 2
to a refluxing solution of the ligand (1.3 mmol) and Et N
3
(
0.2 mL, 1.44 mmol) in THF (50 mL). The mixture was
stirred and the dark-brown or greenish-brown solid was fil-
tered, washed with diethylether, rinsed with water to remove
(
Et NH)Cl, and finally dried in vacuo. Individual details are
3
1
provided in the following and the H NMR data are given in
Table 1.
The SHELXTL system (13) was used for all calculations
and drawings. The coordinates of the Re atoms were deter-
mined by direct methods and the positions of all other non-
hydrogen atoms were found by the standard Fourier method.
ReOCl(qn)2
Reaction time: 20 h. Brown solid. Yield: 82%. IR
–
1
(
CsI, cm ): 968 (s) ν(Re=O), 312 (w) ν(Re-Cl). Anal. calcd.
2
for C H ClN O Re (%): C 41.11, H 2.30, N 5.33; found: C
The structures were refined on Fo using all reflections. The
1
8
12
2
3
4
0.49, H 2.20, N 5.26.
H atoms were fixed using a riding model with Uiso = 1.2 ×
Ueq of the atom to which they are bonded. Crystal data are
3
ReOCl(2-Me-qn)2
listed in Table 2.
Reaction time: 20 h. Brownish-green solid. Yield: 56%.
Brown crystals of ReOCl (5,7-Br -qn)(PPh ) were
2
2
3
–
1
IR (CsI, cm ): 970 (s) ν(Re=O), 320 (w), ν(Re-Cl). Anal.
obtained by recrystallization in CH Cl –pentane. The crys-
2 2
calcd. for C H ClN O Re (%): C 43.36, H 2.91, N 5.06;
tallographic study was carried out with an Enraf-Nonius
CAD-4 diffractometer operating with CuKα radiation under
the control of the CAD-4 software (14). The reduced cell
was initially determined from 20–25 spots located on a pre-
liminary rotation photograph and centered in the detector ap-
erture. Accurate cell parameters were then determined from
the positions of 25 high-angle reflections centered with the
SETANG and DETTH procedures. The intensities were re-
corded by ω/2θ scan. Possible crystal decomposition was
monitored by measuring five standard reflections every hour.
All data in the whole sphere were collected at room temper-
ature. An absorption correction based on the crystal geome-
try was applied. The data were finally corrected for the
effects of Lorentz and polarization, and averaged to provide
the basic two-octant set.
2
0
16
2
3
found: C 42.72, H 2.88, N 4.97.
ReOCl(2-Cl-qn)2
Reaction time: 3 h. Green solid. The compound recry-
stallized from CH Cl contained 0.8 lattice solvent mole-
2
2
–
1
cules per formula. Yield: 57%. IR (CsI, cm ): 968 (s) ν(Re=O),
24 (w), ν(Re-Cl). Anal. calcd. for C Cl N O Re
%): C 34.07, H 1.76, N 4.23; found: C 34.19, H 1.91, N
.17.
3
(
4
H
1
8.8 11.6 4.6
2
3
ReOCl(5-Cl-qn)2
Reaction time: 21 h. Dark-brown solid. Yield: 31%. IR
–
1
(
CsI, cm ): 976 (s) ν(Re=O), 324 (w) ν(Re-Cl). Anal. calcd.
for C H Cl N O Re (%): C 36.34, H 1.69, N 4.71; found:
1
8
10
3
2
3
C 36.29, H 1.74, N 4.34.
The non-hydrogen atoms of the complex molecule were
located and refined as stated previously. In the subsequent
∆F map, two regions centered on the inversion centers at (0,
0, 1/2) and (0, 0, 0), respectively, contained a number of
small peaks owing to severely disordered solvent. No consis-
ReOCl(5-NO -qn) and ReOCl(5,7-X -qn) for X = Cl, Br, I
2
2
2
2
The dark-brown raw solids contained a small amount of
ReOCl (L)(PPh ) that could not be totally removed by chro-
2
3
3
Supplementary data for this article are available on the Web site or may be purchased from the Depository of Unpublished Data, Document
Delivery, CISTI, National Research Council Canada, Ottawa, ON K1A 0S2, Canada. DUD 3653. For more information on obtaining mate-
rial refer to http://cisti-icist.nrc-cnrc.gc.ca/irm/unpub_e.shtml. CCDC 251781 and 251782 contain the crystallographic data for this manu-
script. 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).
©
2005 NRC Canada

4
64
Can. J. Chem. Vol. 83, 2005
Table 2. Crystal data.
ReOCl (2-Cl-qn)(PPh )
ReOCl (5,7-Br -qn)(PPh )
2 2 3
2
3
Formula
C H Cl NO PRe
C H Br Cl NO PRe–(CH Cl ) (C H )
27 19 2 2 2 2 2 0.75 5 12 0.40
2
7
20
3
2
Formula weight
Crystal shape
T (K)
713.96
Dark-green blocks
220(2)
929.91
Brown platelets
220(2)
Diffractometer
Crystal system
Space group
a (Å)
b (Å)
c (Å)
Smart CCD
Monoclinic
Cc (No. 9)
9.8861(2)
15.0186(3)
17.4254(3)
90
96.144(1)
90
2572.4(1)
4
CAD-4
Monoclinic
P2 /c (No. 14)
1
16.967(8)
10.168(3)
18.366(6)
90
96.88(3)
90
3146(2)
4
1.963
α (°)
β (°)
γ (°)
3
Volume (Å )
Z
dcalcd (g/cm3)
1.844
Crystal dimensions (mm)
0.40 × 0.27 × 0.14
CuKα, 1.541 78
129.1
10 492
4060 (Rint = 0.052)
0.46 × 0.29 × 0.04
CuKα, 1.541 78
138.5
22 767
5972 (Rint = 0.102)
Radiation (λ, Å)
–
1
µ (cm )
Rfls. measured
Rfls. independent
Rfls. obs. (I > 2σ)
3982
4579
a
R (I > 2σ)
0.0477
0.1203
1.056
0.0455
0.1080
0.965
a
wR2 (I > 2σ)
a
S (I > 2σ)
a
2
2
2
2
2
1/2
2
2
2
1/2
R = Σ||F | – |F ||/Σ|F |; wR2 = {Σ[w(F – F ) ]/Σw(F ) } ; S = {Σ[w(Fo – F ) ]/(N – Nparam)}
.
o
c
o
o
c
o
c
obs
tent models for CH Cl or pentane molecules could be as-
bulky tert-butyl substituent was present α to the N atom,
whereas only the 1:1 complex was isolated with the phenyl-
substituted ligand.
2
2
sembled from these peaks. Therefore, this part of the
structure was modeled by using the SQUEEZE procedure of
the PLATON software (15), which indicated the presence of
3
two cavities of 310 Å , each occupied by 95 electrons. Con-
IR and NMR spectroscopies
Beside the large number of vibrations because of the
sidering that CH Cl and pentane both contain 42 electrons
2
2
3
and that their volumes are 106 and 191 Å , respectively,
each cavity should contain, on average, 0.8 pentane and
quinolinate and PPh ligands, all compounds show strong IR
3
–1
bands in the 968–984 cm range for the Re=O stretching
mode. Weaker features at ~325 and ~285 cm are likely due
1
.5 CH Cl molecules. The contribution of the disordered
–1
2
2
solvent was calculated with the BYPASS procedure of van
to Re-Cl stretching, bands in this region being commonly
found for similar systems (17).
2
der Sluis and Spek (16) and a new hkl/Fo list bypassing the
solvent contribution was generated. The final model consist-
ing of the ordered part only, without the disordered solvent
contribution, was refined against this new data list, which is
the one corresponding with the CIF file. Refinement con-
verged normally to R = 0.0455. The solvent molecules were
taken into account in the calculation of the derived quanti-
ties given in Table 2.
1
The H NMR signals of the free ligands are listed in Ta-
ble 1 and identified following the numbering scheme I. Ex-
cept for the 2-substituted compounds, a typical doublet of
doublets is observed for H-3 at 7.4–7.9 ppm with couplings
of 4 to 5 and 8 to 9 Hz, respectively. From the coupling con-
stants and the COSY spectrum of Hqn, the adjacent protons
are identified at 8.1–8.6 ppm (J = 8 to 9 Hz) and 8.8–
9
.0 ppm (J = 4 to 5 Hz). The latter signal must originate
from H-2, since it disappears for the 2-substituted deriva-
tives, whereas the remaining doublets for H-3 and H-4 show
the larger coupling. Systematically, the H-2 and H-4 protons
appear downfield from H-3, as expected from the mesomeric
effect in the six-membered nitrogen heterocycle.
For the 5,7-disubstituted compounds, the remaining C-H
signal is a singlet obviously due to H-6. When the phenol
ring is not substituted, H-6 usually appears as a pseudo-
triplet at ~7.4 ppm, resulting from roughly equal couplings
(7–9 Hz) to two adjacent protons at ~7.1 and ~7.35 ppm, re-
spectively. The latter signal is assigned to H-5, since it
Results and discussion
The chemistry used by Mazzi et al. (8) to prepare
ReOCl (qn)(PPh ) and ReOCl(qn) is applied here to obtain
2
3
2
a series of complexes with various substituted 8-hydroxy-
quinolines. Refluxing equivalent amounts of the ligand and
ReOCl (PPh ) leads to ReOCl (L)(PPh ) compounds, in
3
3 2
2
3
which one phosphine and the chlorine atom trans to the
Re=O bond are displaced. By reacting 2 equiv. of ligand in
the presence of 2 equiv. of Et N, the ReOCl(L) products
3
2
were obtained. No complexes could be prepared when a
©
2005 NRC Canada

Sigouin and Beauchamp
465
1
a
Table 3. H NMR chemical shifts (ppm) and coupling constants (Hz) for the quinolinate signals in the complexes.
H-2
H-3
H-4
H-5
H-6
H-7
ReOCl (L)(PPh )
2
3
b
b
b
b
b
b
b
b
qn
8.37 (d, 5.0)
7.70 (dd, 8.3, 5.2)
7.70 (d, 8.0)
7.89 (d, 8.6)
8.21 (d, 8.3)
8.11 (d, 8.1)
8.26 (d, 8.6)
8.28 (d, 8.3)
8.23 (d, 8.3)
8.26 (d, 8.4)
8.16 (d, 8.4)
7.99 (d, 8.2)
8.49 (d, 8.8)
6.79 (d, 7.7)
6.62 (d, 7.1)
6.81 (d, 7.1)
6.70 (7.1)
2
2
2
5
5
5
5
5
-Me-qnc
-Cl-qn
-Ph-qn
-Cl-qn
2.87 (s, CH3)
7.04–7.10 (m, Ph)
8.51 (d, 5.0)
8.61 (d, 4.9)
8.59 (d, 5.0)
8.52 (d, 5.1)
8.63 (d, 5.0)
7.68 (d, 8.4)
7.85 (dd, 8.4, 5.2)
7.89 (dd, 8.4, 5.3)
7.90 (dd, 8.4, 5.2)
7.84 (dd, 8.5, 5.2)
8.00 (dd, 8.7, 5.2)
7.57 (d, 8.3)
7.79 (s)
7.99 (s)
8.21 (s)
8.88 (d, 8.8)
6.82 (d, 8.3)
,7-Cl -qn
2
,7-Br -qn
2
,7-I -qn
2
-NO -qn
6.90 (d, 8.8)
2
ReOCl(L) d
2
qn
8.88 (d, 4.7)
8.00 (dd, 8.2, 5.2)
7.71 (dd, 8.2, 5.2)
7.96 (d, 8.5)
7.55 (d, 8.4)
8.14 (d, 8.6)
8.27 (m, H-2′/H-4)
8.83 (d, 8.2)
8.18 (d, 8.4)
8.48 (d, 8.4)
8.41 (d, 8.6)
8.77 (d, 8.8)
8.34 (d, 8.3)
8.90 (d, 8.6)
8.41 (d, 8.5)
8.96 (d, 7.4)
8.32 (d, 8.6)
8.88 (d, 9.3)
8.16 (d, 8.4)
8.70 (d, 8.2)
8.59 (d, 8.9)
9.04 (d, 8.8)
7.42 (d, 8.1)
7.64 (d, 8.0)
7.31 (m)
7.47 (d, 8.3)
7.45 (m)
7.51 (t, 7.9)
7.86 (t, 8.0)
7.31 (m)
7.74 (t, 8.0)
7.45 (m)
7.85 (t, 8.0)
7.62 (d, 8.6)
8.06 (d, 8.4)
7.94 (s)
8.34 (s)
8.14 (s)
8.55 (s)
7.99 (s)
6.61 (d, 7.6)
7.62 (d, 7.4)
6.40 (dd, 6.5, 2.2)
7.49 (d, 7.9)
6.67 (dd, 6.7, 2.1)
7.66 (d, 8.4)
6.62 (d, 8.4)
8
.27 (m, H-2′/H-4)
3.56 (s, CH3)
.34 (s, CH3)
2
2
5
5
5
5
5
-Me-qn
-Cl-qn
-Cl-qn
2
7
.81 (d, 8.6)
7.64 (d, 8.7)
8.99 (d, 4.8)
.48 (d, 4.9)
9.02 (d, 4.3)
.62 (d, 5.1)
9.01 (d, 4.7)
.57 (d, 5.1)
8.93 (d, 5.0)
.40 (d, 5.0)
9.11 (d, 4.3)
.77 (d, 4.1)
8.15 (dd, 8.6, 5.1)
7.83 (dd, 8.5, 5.1)
8.19 (dd, 8.5, 5.2)
7.88 (dd, 8.5, 5.2)
8.19 (dd, 8.5, 5.2)
7.88 (dd, 8.5, 5.2)
8.14 (dd, 8.4, 5.1)
7.90 (dd, 8.6, 5.0)
8.32 (dd, 8.9, 5.2)
8.04 (dd, 8.9, 5.1)
8
7.66 (d, 8.6)
,7-Cl -qn
2
8
,7-Br -qn
2
8
,7-I -qn
2
8
8.74 (s)
8.95 (d, 8.9)
9.53 (d, 8.9)
-NO -qn
6.73 (d, 8.9)
7.77 (d, 8.8)
2
8
a
In DMSO-d , except when otherwise stated.
6
b
c
Masked by PPh multiplets.
3
In acetone-d6.
d
For the ReOCl(L) complexes, the signals in the first line are assigned to the axial ligand.
2
shows a proximity to H-4 in the NOESY spectrum. For the
-substituted molecules, the doublet at 7.0–7.2 ppm is as-
with the structure (II) observed by Chen et al. (9) for
5
ReOCl(NO -qn)(NH -qn). From the COSY spectrum of
2
2
signed to H-7, by comparison with the unsubstituted rings.
As expected, the remaining H-6 resonance is found to be
very sensitive to the presence of the adjacent electron-
attracting substituents: for instance, introducing a 5-nitro
group induces a large 1.8 ppm downfield shift. In the free
molecules, small couplings of 1 to 2 Hz are generally ob-
served between nonadjacent protons on either ring. Also, in
most cases, a broad resonance appears at low field for the
OH proton.
ReOCl(qn) , two sets of connected H-2/H-3/H-4 signals
2
were defined, in which the H-2 and H-4 doublets could be
identified individually from their respective coupling con-
stants. Two H-5/H-6/H-7 sets could also be assembled, but
the similar coupling constants made it impossible to identify
H-5 and H-7 individually at this stage. This could be done,
however, from a NOESY spectrum, where each H-4 proton
showed a cross peak with a proton in the other ring, which
identified the latter proton as H-5 and established the con-
nection between the H-2/H-3/H-4 and the H-5/H-6/H-7 sets
belonging to the same ligand.
In the ReOCl (L)(PPh ) compounds, the phosphine gives
2
3
1
31
rise to H multiplets in the 7.2–7.5 ppm range and to a
P
singlet lying between –9 and –12.5 ppm, except for the
O
2
–
-substituted ligands, where this signal is shifted to ca.
17 ppm. The signals for the coordinated quinolinate are
O
Cl
N
listed in Table 3. Assignments in the pyridine ring are
straightforward, based on multiplicity and couplings. For the
phenolate ring, the lowest-field signal, whose position is
very substituent-sensitive, originates from H-6. When the
ring is not substituted, two of the protons are masked by
Re
O
N
II
those of PPh , and the only resolved signal at 6.6–6.8 ppm is
3
assigned to H-7 by comparison with the 5-substituted ligands.
1
The ReOCl(L) compounds exhibit two sets of H signals,
There is no direct evidence to associate a given complete
set to either the axial or the equatorial ligand. The tentative
2
showing that the two ligands are inequivalent, in agreement
©
2005 NRC Canada

4
66
Can. J. Chem. Vol. 83, 2005
identification proposed in Table 3 is based on the assump-
tion that the phenol ring signals for the axial ligand in
Fig. 1. ORTEP drawing of the ReOCl2(5,7-Br2-qn)(PPh3) mole-
cule. Ellipsoids correspond to 30% probability.
ReOCl(L) should be similar to those of the equally axial
2
ligand in ReOCl (L)(PPh ), since this ring should be simi-
2
3
larly influenced by the very large electronic effect of the oxo
ligand via the trans O=Re-O unit and it should not be greatly
affected by the anisotropy or other features of the equatorial
ligands. With the assignment proposed, the average δ[axial
ReOCl(L) ] – δ[ReOCl (L)(PPh )] difference is 0.13 ppm
2
2
3
for H-6 and –0.18 ppm for H-7, whereas these values would
be 0.56 and 0.85 ppm, respectively, for the alternate possi-
bility. In contrast, the pyridine ring signals of these axial lig-
ands show larger differences: 0.45 ppm for H-2, 0.29 ppm
for H-3, and 0.12 ppm for H-4. Since these differences are
systematically positive, they may reflect the anisotropy ef-
fect of the PPh rings in ReOCl (L)(PPh ), which is no lon-
3
2
3
ger present in ReOCl(L) . Also consistent with this
2
interpretation is the fact that the H-2 or 2-CH protons of the
3
equatorial ligand appear at a much higher field than those of
the axial one. For instance, the chemical shifts of H-2 in
ReOCl(qn) are 8.27 ppm (equatorial) vs. 8.88 ppm (axial),
2
whereas those of the 2-CH groups in ReOCl(2-Me-qn) are
3
2
2
.34 ppm (equatorial) vs. 3.56 ppm (axial). This could be as-
cribed to the fact that these protons lie in the cone of aniso-
tropy of the axial pyridine ring (III). Thus, these various
effects are consistent with the structure proposed for
ReOCl(L) , which should be the one favored electronically,
2
since two π-donor ligands lie trans to weakly π-accepting
pyridine rings.
Fig. 2. ORTEP drawing of the ReOCl (2-Cl -qn)(PPh ) molecule.
Ellipsoids correspond to 40% probability.
2
2
3
III
X-ray diffraction studies
The ReOCl (5,7-Br -qn)(PPh ) and ReOCl (2-Cl-qn)(PPh )
2
2
3
2
3
molecules are shown in Figs. 1 and 2, respectively. In both
compounds, the rhenium atom has a distorted octahedral co-
ordination, created by an oxo ligand, two chlorine atoms, a
phosphine, and a N,O-coordinated quinolinate. In agreement
with a general trend in mono-oxo Re(V) chemistry, the
quinolinate oxygen donor lies trans to the Re=O bond,
which results in the quinolinate plane being roughly parallel
to the Re=O direction. The PPh group is found to be cis to
the nitrogen atom in the equatorial plane. This arrangement
3
for Re—OR bonds (2), reflect the constraint imposed by the
five-membered chelate ring.
has also been reported for ReOCl (5-nitro-qn)(PPh ) (9).
2
3
The oxo ligand tends to produce a general displacement of
the adjacent bonds in the opposite direction, which results in
the Re atom lying above the plane defined by the four equa-
torial donor atoms. The distances of rhenium to the PNCl2
plane are 0.23 and 0.28 Å for the 5,7-Br-qn and 2-Cl-qn
complexes, respectively. Additional distortion is introduced
by the bidentate coordination of the quinolinate ligand,
whose O(2)-Re-N bite angle is ~76°. This small angle af-
Selected distances and angles are presented in Table 4.
The Re=O bond lengths (1.656(8) and 1.671(4) Å) are in
good agreement with the value expected for Re(V) mono-
oxo complexes (18). The trans influence of PPh makes the
3
opposite Re—Cl(1) bond 0.05 Å longer than the other in
both complexes. The Re—O(2) distances (1.999(6) and
2
.021(4) Å), which are ~0.1 Å longer than the typical value
©
2005 NRC Canada

Sigouin and Beauchamp
Table 4. Selected distances (Å) and angles (°) in the ReOCl (L)(PPh ) compounds.
467
2
3
ReOCl (2-Cl-qn)(PPh )
ReOCl (5,7-Br -qn)(PPh )
2
3
2
2
3
Bond distances (Å)
Re=O(1)
Re—O(2)
Re—N
Re—Cl(1)
Re—Cl(2)
Re—P
1.656(8)
1.999(6)
2.183(8)
2.390(2)
2.334(2)
2.485(2)
1.671(4)
2.021(4)
2.130(6)
2.371(2)
2.327(2)
2.469(2)
Bond angles (°)
O(1)-Re-O(2)
O(1)-Re-N
163.2(3)
93.2(3)
160.2(2)
86.7(2)
O(2)-Re-N
76.2(3)
75.5(2)
O(1)-Re-Cl(1)
O(2)-Re-Cl(1)
N-Re-Cl(1)
O(1)-Re-Cl(2)
O(2)-Re-Cl(2)
N-Re-Cl(2)
Cl(1)-Re-Cl(2)
O(1)-Re-P
O(2)-Re-P
N-Re-P
Cl(1)-Re-P
Cl(2)-Re-P
100.0(3)
92.5(2)
87.1(2)
102.3(3)
89.5(2)
164.0(2)
86.42(10)
86.3(3)
99.8(2)
87.24(13)
84.76(14)
105.67(17)
92.79(13)
167.15(14)
89.59(7)
91.18(16)
81.86(13)
94.35(14)
168.92(6)
88.98(7)
119.4(4)
127.2(5)
115.4(4)
81.1(2)
91.3(2)
173.58(8)
93.49(8)
118.7(5)
131.1(8)
112.0(6)
Re-O(2)-C(6)
Re-N-C(1)
Re-N-C(5)
fects mainly the O(1)-Re-O(2) unit, which is definitely non-
Fig. 3. UV–vis spectra of ReOCl (qn)(PPh ) in DMSO (full
2 3
line). Dashed lines correspond to the four components obtained
by deconvolution.
linear: 163.2(3)° for 2-Cl-qn and 160.2(2)° for 5,7-Br -qn.
2
The Re=O bond is displaced away from the chloro ligands,
as evidenced from the Cl-Re=O angles (100°–106°) being
systematically greater than the N-Re=O and P-Re=O angles
(
87°–94°).
The 2-chloro substituent enhances the distortion of the
octahedron. This group further displaces the Re=O bond to-
ward PPh , as evidenced from the O(1)-Re-P angle of
3
8
6.3(3)°, compared to 91.2(2)° in the 5,7-Br -qn complex.
2
The angle between the Cl(2)-Re-O(1) plane and the mean
plane through the quinolinate ligand increases from 10.5° for
the dibromo complex to 22.2° here. The distortion is also il-
lustrated by the sum of the deviations of the Cl, P, and N at-
oms from their least-squares plane: these atoms are virtually
coplanar for the dibromo complex, but the mean deviation is
0
.69 Å for ReOCl (2-Cl-qn)(PPh ).
2 3
In both complexes, intramolecular π-stacking interactions
exist between one of the phosphine phenyl rings (C(31)-
C(36)) and the quinolinate ligand, which are roughly parallel
(
dihedral angles of 11.7° and 12.9° for the 5,7-Br -qn and 2-
2
Cl-qn complexes, respectively) and separated by ~3.42 Å.
The molecules are packed individually in the unit cell,
with normal van der Waals contacts, including some
intermolecular π stacking.
erally in CHCl . For a few less soluble complexes, DMSO
was used to improve the quality of the spectra. The DMSO
solutions were stable and the spectral features were similar
3
to those of the weaker spectra in CHCl , showing that the
3
UV–vis spectroscopy
The low-energy (8 000–20 000 cm ) region of the UV–
vis absorption spectra was recorded for all complexes, gen-
same species was present in both solvents. The spectra of
ReOCl (qn)(PPh ) and ReOCl(qn) are illustrated in Figs. 3
–
1
2
3
2
and 4, respectively. In each case, four individual compo-
©
2005 NRC Canada

4
68
Can. J. Chem. Vol. 83, 2005
Table 5. Transition energy (cm–1 × 10 ) and absorption coefficient (M cm , within parentheses) of the
–3
a
–1
–1
low-energy components of the electronic spectra.
[
S → T*]
[S → T**]
[S → S*]
[S → S**]
b
ReOCl (qn)(PPh )
9.6 (29)
12.1 (150)
12.6 (73)
12.6 (29)
12.1 (149)
11.8 (195)
11.8 (142)
11.8 (166)
11.8 (160)
11.6 (120)
12.2 (185)
12.5 (153)
11.7 (51)
14.3 (381)
14.8 (235)
14.9 (97)
16.2 (299)
16.4 (241)
16.6 (99)
16.3 (310)
17.5 (430)
16.6 (290)
16.6 (312)
16.5 (320)
17.7 (181)
17.7 (360)
18.3 (226)
—
2
3
b
c
ReOCl (2-Me-qn)(PPh )
10.0 (17)
2
3
b
c
ReOCl (2-Cl-qn)(PPh )
10.9 (8)
2
3
d
ReOCl (5-Cl-qn)(PPh )
9.8 (18)
14.3 (407)
14.2 (544)
14.3 (389)
14.4 (430)
14.2 (424)
13.7 (243)
14.4 (441)
14.6 (346)
13.8 (111)
2
3
e
c
ReOCl (5-NO -qn)(PPh )
10.0 (22)
2
2
3
d
ReOCl (5,7-Cl -qn)(PPh )
9.8 (16)
2
2
3
d
c
ReOCl (5,7-Br -qn)(PPh )
10.0 (18)
2
2
3
d
c
ReOCl (5,7-I -qn)(PPh )
10.0 (16)
2
2
3
ReOCl(qn) d
9.8 (24)
9.5 (38)
—
2
ReOCl(2-Me-qn) d
2
ReOCl(2-Cl-qn) d
2
ReOCl(5-Cl-qn) d
9.4 (6)
—
2
ReOCl(5-NO -qn)2d
12.0 (230)
c
14.5 (466)
c
17.5 (970)
c
2
ReOCl(5,7-Cl -qn)2d
—
12.0 (263)
c
14.0 (480)
c
17.5 (754)
c
2
ReOCl(5,7-Br -qn)2d
—
12.0 (140)
c
14.0 (283)
c
17.0 (460)
c
2
ReOCl(5,7-I -qn)2d
—
12.0 (433)
c
14.0 (780)
c
17.0 (1270)
c
2
a
The ε values listed are those obtained at the indicated wavelength from the nondeconvoluted spectra.
b
c
In DMSO.
Peak position estimated visually.
d
In CHCl3.
In CH Cl .
e
2
2
Fig. 4. UV–vis spectra of ReOCl(qn) in DMSO (full line).
Dashed lines correspond to the four components obtained by
deconvolution.
whereas the related band is present as an unresolved
component at slightly lower energy for the ReOCl (L)(PPh )
2
2
3
compounds. These absorptions are assigned to the spin-
allowed [S → S*] and [S → S**] transitions, respectively.
On the low-energy side, there is a distinct dissymmetry in
the main peak and a rather long tail before intensity falls off
to zero. Two weak features are revealed by deconvolution,
which are assigned to the two anticipated singlet–triplet
transitions. These transitions being spin-forbidden, their ex-
tinction coefficients are expected to be ~10, and this is in-
2
1
1
deed the case for the first transition (d ) → (d ) (d )
xy
xy
yz
–
1
[
(
S → T*] at ~10 000 cm . The coefficient is much higher
2 1 1
~100) for the second transition (d ) → (d ) (d ) [S → T**]
xy
xy
xz
–1
at ~12 000 cm , but enhanced intensity could be due to mix-
ing between the higher-energy triplet state and the lowest-
energy singlet states (20), as a result of strong spin-orbit
couplings commonly found for third-row transition metals.
The separations (Table 6) between the two triplet states, the
two singlet states, and the second triplet – first singlet states
–
1
(
roughly 2000 cm ) are in good agreement with the values
calculated by Da Re and Hopkins (20).
Comparison of the data within each series reveals that
ring substitution does not introduce dramatic effects on the
spectra: the strongly electron-attracting 5-nitro substituent
seems to slightly increase the [S → S**] transition energy in
nents were obtained by deconvolution using ORIGIN (19).
The data are listed in Table 5.
Since these complexes have no symmetry, the dyz and dxz
orbitals, which are degenerate under D4h symmetry, become
inequivalent. For consistency with the recent study of Da Re
and Hopkins (20) on the isoelectronic oxo-Mo(IV) core, dyz
the ReOCl (L)(PPh ) series, but the effect is modest and no
2
3
recognizable trend can be detected. Thus, the HOMO–
LUMO gap is not substantially affected by ligand substitu-
tion in these systems.
will be assumed to lie at lower energy than d . The five low-
xz
est-energy electronic states expected are depicted in Fig. 5,
in increasing order of energy.
The spin-allowed [S → S*] and [S → S**] bands in both
series are observed at somewhat lower energy than for the
Both series of complexes show a strong maximum be-
ReO(OR)X (dppe) compounds studied earlier, where they
were found at ~16 500 and 20 000 cm , respectively (2).
This can be explained by the lower electron-donating ability
2
–
1
–1
tween 13 700 and 14 900 cm . For the ReOCl(L) com-
2
–1
pounds, another strong peak is visible at 17 000–18 300 cm ,
©
2005 NRC Canada

Sigouin and Beauchamp
469
2
Fig. 5. Low-energy electron configurations for the Re(V) d systems.
Table 6. Energy separation (cm–1 × 10 ) between the low-energy states from the electronic spectra.
–3
∆
[T* → T**]
∆[S* → S**]
∆[T** → S*]
∆[T* → S**]
ReOCl (qn)(PPh )
2.5
2.6
1.7
2.3
1.8
2.0
1.8
1.8
1.8
2.7
—
1.9
1.6
1.7
2.0
3.3
2.3
2.3
2.2
2.3
4.0
3.3
3.7
3.0
3.5
3.0
3.0
2.2
2.2
2.3
2.2
2.4
2.5
2.6
2.4
2.1
2.2
2.1
2.1
2.5
2.0
2.0
2.0
6.6
6.4
5.7
6.5
7.5
6.8
6.6
6.5
7.9
8.2
—
2
3
ReOCl (2-Me-qn)(PPh )
2
3
ReOCl (2-Cl-qn)(PPh )
2
3
ReOCl (5-Cl-qn)(PPh )
2
3
ReOCl (5-NO -qn)(PPh )
2
2
3
ReOCl (5,7-Cl -qn)(PPh )
2
2
3
ReOCl (5,7-Br -qn)(PPh )
2
2
3
ReOCl (5,7-I -qn)(PPh )
2
2
3
ReOCl(qn)2
ReOCl(2-Me-qn)2
ReOCl(2-Cl-qn)2
ReOCl(5-Cl-qn)2
2.3
—
—
ReOCl(5-NO -qn)2
—
2
ReOCl(5,7-Cl -qn)2
—
—
2
ReOCl(5,7-Br -qn)2
—
—
2
ReOCl(5,7-I -qn)2
—
—
2
is crucial for the eventual development of optically or mag-
netically tunable materials (1, 21).
of the quinolinate oxygen compared to alkoxo groups, con-
sidering that donation from the ligand trans to the Re=O
bond was previously identified as a very sensitive factor.
However, this reduction is not large enough to bring the en-
Acknowledgments
ergies below those of the ReOX (dppe) compounds
3
–
1
We wish to thank M. Palardy for his participation in the
preparative work, Dr. M. Simard and F. Bélanger-Gariépy
for assistance in the interpretation of the X-ray diffraction
data, and the Natural Sciences and Engineering Research
Council of Canada (NSERC) for financial support.
(
~12 000 and ~16 400 cm ).
Concluding remarks
Homologous series of ReOCl (L)(PPh ) and ReOCl(L)
2
2
3
compounds were prepared with various 8-hydroxyquinolines
bearing substituents with different steric and electronic prop-
erties. Comparisons between these related compounds al-
References
1
. 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).
1
lowed us to fully interpret their H NMR spectral features.
The positions of the UV–vis bands used to estimate the
HOMO–LUMO gap failed to reveal a large effect of ring
substitution on electronic transitions, but it confirmed that
the donor ability of the ligand trans to the Re=O band plays
a major role. On the other hand, the fact that the spin-
forbidden singlet–triplet bands were readily observed makes
these systems attractive for future in-depth spectroscopic
studies to improve our understanding of the low-lying ex-
cited states and the spin-crossover processes, whose control
2
3
4
. F. Baril-Robert and A.L. Beauchamp. Can. J. Chem. 81, 1326
(
2003).
. G. Delapierre, J.M. Brunel, T. Constantieux, and G. Buono.
Tetrahedron Asymmetry, 12, 1345 (2001).
. J. Chatt and G.A. Rowe. J. Chem. Soc. 4019 (1962).
5. G.R. Pettit and M.V. Kalnins. J. Org. Chem. 25, 1365 (1960).
6. J.J. Plattner, R.D. Gless, and H. Rapoport. J. Am. Chem. Soc.
94, 8613 (1972).
©
2005 NRC Canada

4
70
Can. J. Chem. Vol. 83, 2005
7
8
. W.C. Fleming and G.R. Pettit. J. Org. Chem. 36, 3490 (1971).
. U. Mazzi, E. Roncari, R. Rossi, V. Bertolasi, O. Traverso,
and L. Magon. Transition Met. Chem. 5, 289 (1980).
. X. Chen, F.J. Femia, J.W. Babich, and J. Zubieta. Inorg. Chim.
Acta, 308, 80 (2000).
15. A.L. Spek. PLATON. July 1995 version [computer program].
University of Utrecht, Utrecht, The Netherlands. 1995.
16. P. van der Sluis and A.L. Spek. Acta Crystallogr. A46, 194
(1990).
17. A.M. Lebuis and A.L. Beauchamp. Can. J. Chem. 71, 441
(1993).
9
1
1
1
1
1
0. SMART. Release 5.059 [computer program]. Bruker AXS Inc.,
Madison, Wisconsin, USA. 1999.
1. SAINT. Release 6.06 [computer program]. Bruker AXS Inc.,
Madison, Wisconsin, USA. 1999.
2. G.M. Sheldrick. SADABS [computer program]. Bruker AXS
Inc., Madison, Wisconsin, USA. 1996.
3. G.M. Sheldrick. SHELXTL. Release 5.10 [computer program].
Bruker AXS Inc., Madison, Wisconsin, USA. 1997.
4. CAD-4. Version 5.0 [computer program]. Enraf-Nonius, Delft,
The Netherlands. 1989.
18. Cambridge Structural Database. Version 5.4, November 2002
release with February 2003 updates. Cambridge University,
Cambridge, UK.
19. ORIGIN 6.1, 6.1052 (B232) [computer program]. OriginLab
Corporation, Northampton, UK. 2000.
20. R.E. Da Re and M.D. Hopkins. Inorg. Chem. 41, 6973 (2002).
21. C. Savoie and C. Reber. J. Am. Chem. Soc. 122, 844 (2000).
©
2005 NRC Canada