The Br and I analogues of ReCl(H2)(PMePh2)4; crystal structures of ReBR(H2)(PMePh2)4 and the [ReO2(Py)4]+ cation; possible solution state evidence of Re-H···H···N(Py) interactions as evident in T1 measurements

Article abstract of DOI:10.1016/S0277-5387(01)00691-X

The syntheses of the Br and I analogues of ReCl(H₂)(PMePh₂)₄ are reported together with their characterizations by NMR, and in the case of the Br analogue, FABMS and a single crystal X-ray diffraction study. The minimum T₁, for the metal bonded hydrogen atoms in the Br and I analogues was determined to be 62 and 101 ms at 400 MHz, respectively. A decrease of 9 ms in the room temperature T₁ measurement of ReCl(H₂)(PMePh₂)₄ was observed upon addition of pyridine to the NMR tube. This is suggestive of an interaction between the dihydrogen ligand and the lone pair of electrons on the N-atom in py. The crystal structure of ReBr(H₂)(PMePh₂)₄ revealed that the structure consisted of a trans dihydrogen to halide arrangement and was disordered with respect to the locations of the dihydrogen ligand and the halide in the crystal. The crystal structure of the trans-[ReO₂(Py)₄]⁺ cation is also reported.

Full text of DOI:10.1016/S0277-5387(01)00691-X

Polyhedron 20 (2001) 773–782
www.elsevier.nl/locate/poly
The Br and I analogues of ReCl(H2)(PMePh2)4; crystal structures
of ReBr(H2)(PMePh2)4 and the [ReO2(Py)4]+ cation; possible
solution state evidence of Re–H···H···N(Py) interactions as evident
in T1 measurementsꢀ,ꢀꢀ
Rudy L. Luck *, Ryan S. O’Neill
Department of Chemistry, Michigan Technological Uni6ersity, 1400 Townsend Dri6e, Houghton, MI 49931, USA
Received 4 April 2000; accepted 14 August 2000
Abstract
The syntheses of the Br and I analogues of ReCl(H2)(PMePh2)4 are reported together with their characterizations by NMR, and
in the case of the Br analogue, FABMS and a single crystal X-ray diffraction study. The minimum T1 for the metal bonded
hydrogen atoms in the Br and I analogues was determined to be 62 and 101 ms at 400 MHz, respectively. A decrease of 9 ms
in the room temperature T1 measurement of ReCl(H2)(PMePh2)4 was observed upon addition of pyridine to the NMR tube. This
is suggestive of an interaction between the dihydrogen ligand and the lone pair of electrons on the N-atom in py. The crystal
structure of ReBr(H2)(PMePh2)4 revealed that the structure consisted of a trans dihydrogen to halide arrangement and was
disordered with respect to the locations of the dihydrogen ligand and the halide in the crystal. The crystal structure of the
trans-[ReO2(Py)4]+ cation is also reported. © 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Br and I analogues; Single crystal X-ray diffraction study; Crystal structures
bidentate phosphine analogues ReCl(H2)(dppe)2 and
ReCl(H2)(dppee)2 have also been previously synthesized
and characterized [4]. These complexes have similar T1
(min) times (43 ms at −40°C and 29 ms at 26°C, by
200 MHz NMR) and JHP values (20 Hz for both) to
ReCl(H2)(PMePh2)4, but have greatly lessened reactiv-
ity to H2 displacement reactions.
The purpose of this work is to determine the effect
that changing the identity of the halogen trans to the
dihydrogen ligand in ReCl(h2-H2)(PMePh2)4 (1) would
have on the (H–H) interaction, as it is anticipated that
the nature of the dihydrogen ligand will change as the
trans halogen is varied, [5]. This should allow us to
understand more about the steric and electronic condi-
tions which lead to classical and nonclassical behavior.
An additional goal of this work is to place the dihydro-
gen ligand in circumstances which may be suitable to
such an ‘end-on’ arrangement and observe the molecule
via NMR and X-ray crystallography. This could be a
useful endeavor because an h1-H2 ligand could display
novel hydrogen activation properties. Recently, there
1. Introduction
Complexes of the form ReX(H2)(PMePh2)4 are of
interest because the results of X-ray crystallography
data on the Cl analog [1] suggest that the dihydrogen
ligand [2] may be bound in an unsymmetrical manner.
ReCl(H2)(PMePh2)4 (1) has also been characterized in
the past as having a nonclassical stretched (H2) ligand.
This assignment was based on T1 NMR studies [3]
which placed the spin-lattice relaxation time minimum
well within the accepted range (25 ms at 200 MHz,
−50°C) of stretched molecular dihydrogen behavior
and suggested an H···H interaction on the order of 1.08
,
,
A (fast rotation) to 1.37 A (slow rotation) [4]. The
ꢀ This article was presented at the Second Contemporary Inorganic
Chemistry Conference (CIC-II) held at Texas A&M University, Col-
lege Station, TX, USA, on March 12–15, 2000.
ꢀꢀ This paper is based on Ryan O’Neill’s Master’s thesis.
* Corresponding author. Tel.: +1-906-4872309; fax: +1-906-
4872061.
E-mail address: rluck@mtu.edu (R.L. Luck).
0277-5387/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0277-5387(01)00691-X

774
R.L. Luck, R.S. O’Neill / Polyhedron 20 (2001) 773–782
have been reports of long-range interactions involving
metal hydrides with H–NR and H–CR fragments,
[6–8] and, alcohols [9,10].
C52H54BrP4Re·SiO2: C, 55.12; H, 4.80. Found: C, 55.18,
H, 5.28. Spectroscopic data for 3: H NMR (C6D6): l
6.6–7.8 (m, 40 H, (C6H5)2MePRe), 2.2 (br., 12H,
CH3Ph2PRe), −9.4 (q., 2H, JHP=20.5 Hz, H2Re). The
1H NMR data also indicated the presence of silicone
grease in the final sample sent off for elemental analysis
(singlet at l 0.29).
1
2. Experimental
2.1. General considerations
2.3. Dihydrogen ligand interaction studies
All manipulations were carried out under either an
argon or nitrogen atmosphere. Methanol was dried
with magnesium methoxide, and THF, benzene, and
hexane were dried with finely divided sodium metal.
Standard Schlenck and drybox techniques were used in
all cases. ReCl5 and PMePh2 were used as purchased
from Strem Chemicals, Inc. and all other reagents were
obtained from the Aldrich Chemical Company. Mass
spectral data were obtained at the Michigan State
University Mass Spectrometry Facility which is sup-
ported, in part, by a grant (DRR-00480) from the
Biotechnology Research Technology Program, Na-
tional Center for Research Resources, National Insti-
In an attempt to synthesize complexes containing
dihydrogen ligands that exhibit bonding patterns of the
geometry shown in Fig. 1, 1 was exposed to organic
bases, namely pyridine and 2,4,6-collidine. Essentially,
the lone pair of electrons on the nitrogen atom in the
base could interact with the (H2) ligand and potentially
lend stability to an asymmetrical dihydrogen relative to
the metal.
NMR samples were prepared consisting of 1, ben-
zene-d6 as the solvent, and one of the following added:
one equivalent of pyridine, an excess of pyridine, 1
equiv. of 2,4,6-collidine, or an excess of 2,4,6-collidine.
Each sample was subjected to 1H NMR T1 inversion-re-
covery experiments at room temperature. In attempts
to obtain crystals featuring this interaction, 1 was dis-
solved in benzene in a 1/4%% I.D. tube and layered with
a mixture of pyridine and MeOH.
1
tutes of Health. H data were recorded using a Varian
XL-400 spectrometer. The microanalysis was done by
Galbraith® Laboratories, Inc., Knoxville, TN.
2.2. Syntheses
ReCl(H2)(PMePh2)4 (1) was synthesized by the
method reported by Cotton and Luck with no modifi-
cation [3]. ReBr(H2)(PMePh2)4 (2) and ReI(H2)-
(PMePh2)4 (3) were synthesized by dissolving 1 in a 4:1
mixture of THF+MeOH with 50 equiv. of KBr or KI,
respectively, at room temperature. The mixture was
stirred under argon for three days. The solvents were
then removed under vacuum and benzene was added.
The solid salts were separated by filtration. The remain-
ing solution was frozen under vacuum and the benzene
sublimed off overnight. 2 was recrystallized from ben-
zene+MeOH (obtained in 50% yield overall from
starting 1), but 3 could not be separated efficiently. The
nature of the dihydrogen ligands in 1, 2 and 3 were
ascertained by variable temperature T1 NMR measure-
ments using the inversion-recovery method. The
fluoride analogue of the above molecules could not be
synthesized by the methods used for 2 or 3. Spectro-
scopic data for 2: 1H NMR (C6D6): l 6.6–7.8 (m, 40 H,
(C6H5)2MePRe), 2.1 (br., 12H, CH3Ph2PRe), −9.1 (q.,
2.4. HD exchange reaction
HD exchange for the dihydrogen ligand was per-
formed on 1 by dissolving the compound in benzene
and stirring overnight under HD gas generated by
reacting sodium hydride with deuterated water.
2.5. X-ray structural analysis
Crystals of 2·acetone were obtained from a solution
of 2 in acetone layered with hexane. Crystals of trans-
dioxo-[ReO2(Py)4][OH]·1.75H2O (4·1.75H2O) were ob-
tained by placing 1 in a solution of 30% benzene, 20%
pyridine, 20% water, and 30% hexane. This was done in
an attempt to obtain crystals displaying pyridine inter-
acting with a dihydrogen ligand. Unfortunately, we
were unaware of the presence of water in the py used
and only after the structure was completed made this
discovery and subsequent experiments were conducted
with dry pyridine.
2H,
J
HP=20.0 Hz, H2Re). Anal. Calc. for
In both cases, crystals were removed from the mother
liquor to a microscope slide where they were stabilized
by immersion in a mixture of the mother liquor and
mineral oil. The crystal was then rolled in epoxy and
mounted on the head of a thin glass fiber on a go-
niometer mounting pin. The pin mounted crystal is then
inserted into the goniometer head of the X-ray diffrac-
tometer and centered in the beam path. Standard
Fig. 1. (H2)–lone pair interaction.

R.L. Luck, R.S. O’Neill / Polyhedron 20 (2001) 773–782
775
Table 1
Crystallographic data for 2 and 4
2
4
Chemical formula
Color
C
yellow
52H54BrP4Re·C3H6O
[(C5H5N)4O2Re][OH]·1.75H2O
red
Habit
prism
oblong block
0.40×0.05×0.05
loses moisture to atmosphere
30.758(4)
9.149(3)
18.035(5)
90
107.967(19)
90
0.71073
monoclinic
Size (mm3)
Behavior
0.30×0.30×0.15
oxygen sensitive
11.706(2)
14.227(2)
17.304(4)
76.499(17)
71.486(17)
83.346(13)
0.71073
,
a (A)
,
b (A)
,
c (A)
h (°)
i (°)
k (°)
,
u (A)
Crystal system
triclinic
(
Space group
Z
P1
2
C2/c
8
Type of diffractometer
Method of data collection
Total number of data collected
Rint value
Nonius Turbo CAD4 diffractometer
ꢀ–2q
7331
0.012
Nonius Turbo CAD4 diffractometer
ꢀ
4310
0.022
Number observed (cut-off parameter)
2q value
6453 with \2|(I)
45
2630 with \2|(I)
50
Absorption correction a
psi scans
0.032
psi scans
0.053
b
R
c
Rw
0.096 d
554
0.175 e
286
Number of parameters
Form of refinement
F2
F2
Treatment of hydrogens
fixed if refined
fixed if refined
a Ref. [11].
b R=ꢀ(Fo−Fc)/ꢀ(Fo).
c Rw=[ꢀ[w(Fo2−Fc2)2]/ꢀ[w(F2o)2]]1/2
.
d w=1/[|2(Fo2)+(0.1106P)2+25.78P], where P=(max(Fo2,0)+2F2c)/3.
e w=1/[|2(Fo2)+(0.0493P)2+8.98P], where P=(max(Fo2,0)+2F2c)/3.
CAD4 centering, indexing, and data collection pro-
grams were utilized. Twenty-five reflections between 10
and 15° in q were located by a random search pattern,
centered, and used in indexing. The cell constants and
orientation matrix were thus obtained and refined by a
least-squares fit. The scan width for each reflection
and the scan rate were dependent on the intensity of
the observed diffraction signal. During data collection,
three intensity standards and three orientation stan-
dards were measured at regular intervals to measure
the rate of decay of the crystal and to accommodate
for crystal movement.
In both cases, data were first reduced and corrected
for absorption using psi-scans [11] and then solved
using the program DIRDIF-96 [12]. The models were
then refined using SHELXL-97 [13]. The refinement of 2
was straightforward and it was clear from the outset
that there was a disorder between the Br and the H2
ligands. Evidence for an acetone molecule of solvation
was also present in the original solution. The Br disor-
der was accounted for by constraining the sum of the
occupancies at both sites to be unity and allowing a
free refinement of the occupancies to derive the final
quantities. In the final model there was no evidence of
electron density that would represent a disordered Cl
atom bonded to the Re, thus, the crystal used was
pure 2.
With 4, it was clear that apart from the two Re-
containing cations, there were several regions of elec-
tron density that could be defined as either water
molecules or hydroxide anions. These atoms were
refined first with isotropic and then some with an-
isotropic thermal parameters to convergence. It was
clear due to elongated thermal ellipsoids that some of
these water molecules were split. Indeed two of these
atoms were refined this way. In the final cycles of
refinement, H-atoms (apart from those on the O
atoms belonging to the hydroxide anions and the wa-
ter molecules) were fixed at ideal positions with com-
mon isotropic displacement parameters (Uiso=0.05
2
,
A ) and this was the final model. The crystallographic
data for 2 and 4 are listed in Table 1.

776
R.L. Luck, R.S. O’Neill / Polyhedron 20 (2001) 773–782
3. Results
JHP has been suggested to be associated with a more
hydridic nature in the H2 ligand [14] and this would
imply that going from Cl to I results in a more hydridic
nature in the H···H ligand on complexes 1, 2 and 3.
3.1. Synthesis
The synthesis of ReBr(H2)(PMePh2)4 (2) was accom-
plished by a metathesis type reaction of 1 with a
massive excess of KBr in a 4:1 vol/vol mixture of
THF+MeOH over a period of 3 days under argon at
room temperature. The pale green product was a mix-
ture of 1 and 2 with the bromide analogue dominating
by more than 80%. Attempts to increase the yield of the
reaction by varying the solvent and temperature were
unsuccessful. The synthesis of ReI(H2)(PMePh2)4 (3)
was performed similarly to that of 2 by exposing 1 to a
large excess of KI in a 4:1 mixture of THF+MeOH
for 3 days under argon at room temperature. 3 domi-
nated the products of this reaction by roughly 70%
judging by the relative intensity of the NMR signals in
the hydride region. Variance in the above reaction
conditions did not change the product distribution in
favor of the bromo or iodo analogues.
3.3. HD exchange reactions
Hydrogen–deuterium coupling constants can be a
source of information about how the hydrogens in a
dihydrogen ligand interact with each other. Larger cou-
pling constants have been suggested to indicate greater
interaction and thus more non-classical behavior. This
has been shown to be the case for those complexes
whose neutron diffraction structures are known [2]. HD
exchange of the H2 ligand in 1 produced mixed samples
of ReCl(H2)(PMePh2)4 and ReCl(HD)(PMePh2)4. This
exchange was accomplished by producing HD gas from
the reaction of NaH and D2O and allowing the gas to
pass through a liquid nitrogen trap and then react with
1 dissolved in benzene.
1
Fig. 2 displays the H NMR spectrum of a sample of
1 upon which partial HD exchange has been per-
formed. The H2 signal from 1 is visible as a quintet at
l −8.55 and the HD resonance, also a quintet, is
shifted to the right and centered at l −8.60.
3.2. NMR studies
The signal in the 1H NMR hydride region shifts
steadily to the right as the size of the halogen trans to
the dihydrogen ligand is increased and the electronega-
tivity of said halogen decreases. JHP displays a slight
increasing trend as shown in Table 2. An increase in
The HD quintet appears broad in the above spectra
and it was thought that the H2 quintet could be mask-
ing the HD coupling pattern. In an effort to resolve the
1
complexity of the HD peaks, the H{31P} NMR spec-
Table 2
JHP for ReCl(H2)(PMePh2)4, ReBr(H2)(PMePh2)4, ReI(H2)(PMePh2)4
Compound
Chemical shift in acetone (ppm)
Coupling constant (Hz)
Pauling a electronegativity
ReCl(H2)(PmePh2)4
ReBr(H2)(PmePh2)4
ReI(H2)(PMePh2)4
−8.55
−9.1
−9.35
19.2
20.0
20.5
3.0
2.8
2.5
a For the trans halogen [15].
Fig. 2. 400 MHz 1H NMR of ReCl(H2)(PMePh2)4 and ReCl(HD)(PMePh2)4 in benzene-d6.

R.L. Luck, R.S. O’Neill / Polyhedron 20 (2001) 773–782
777
Fig. 3. 400 MHz 1H{31P} NMR of ReCl(H2)(PMePh2)4 and ReCl(HD)(PMePh2)4 in benzene-d6.
trum was collected and is shown below in Fig. 3. The
H2 and HD peaks appear at l −8.55 and −8.60
again, respectively. The HD peak is still broad here and
partially obscured by the dihydrogen signal. The broad-
ness of this peak may be suggestive of some HD
coupling, but this is not resolved in our spectra. The
analogous arsine compound, ReCl(HD)(AsMePh2)4,
(where the HD signal is also to the right of the HH
complex) displays a JHD of 4 Hz [16].
The difference between the left system and metal-
bound dihydrogens forming H-bonds with amines may
only be one of degree, however, to our knowledge, the
interaction on the right has never been observed previ-
ously (and certainly not by the T1 technique). To test
this hypothesis, pyridine and 2,4,6-collidine were re-
acted with 1 to observe their interactions. Figs. 5 and 6
1
display H NMR spectra of 5 with one equivalent and
an excess of pyridine added, respectively.
Attempts to separate the two peaks by lowering the
temperature caused the signals to coalesce. Further
experiments were performed in which the delay time
between the 90° pulse and the FID collection was
varied so that only the species that relaxed more slowly
[2] (i.e. the HD compound as D is not as effective as H
in the T1 relaxation mechanism) would be observed.
This experiment was not successful, as the peak as-
signed to the HD complex appeared to be decreasing in
intensity relative to the other peak which would mean
that it was relaxing faster than that assigned as the
H–H complex. At this point it is not known if this was
a complication due to inadequacies in our NMR probe
design or if it is a meaningful experiment. Further,
measurements at lower temperatures were complicated
by the fact that the two resonances merged which
hindered obtaining cleanly resolved spectra.
The normal dihydrogen quintet can be observed at l
−8.55 ppm along with what appears to be a doublet of
triplets centered at l −7.6 ppm. As more pyridine is
added, the new signals centered at l −7.6 ppm, grow
in intensity and gradually replace the dihydrogen quin-
tet completely when a sufficient excess is added. With
excess pyridine, the resonance at l −7.6 resolves into a
pair of triplets located at l −7.88 and −7.97. Each
hydrogen is split by two phosphorous atoms with a
different coupling constant for each splitting (14.2 and
15.2 Hz). This may be attributed to the displacement of
two phosphine ligands with pyridine. One way of ac-
counting for these resonances is by having the hydro-
gens on the metal in differing electronic environments
with each forming a triplet from splitting by the two
remaining phosphorus atoms and no (or very small)
coupling between the different H-atoms. A drawing for
this hypothetical complex, which we could not isolate
or crystallize, is shown in Fig. 7.
3.4. NMR study of metal-bonded dihydrogen to lone
pair interactions
Transition metal hydrides have been shown to form
unconventional intermolecular bonds with amine and
alcohol hydrogens [6–10,17]. A pictorial representation
of this for cases involving N atoms is displayed in Fig.
4.
Fig. 4. Intermolecular hydrogen bonding involving metal-hydrides.

778
R.L. Luck, R.S. O’Neill / Polyhedron 20 (2001) 773–782
Fig. 5. 400 MHz 1H NMR of ReCl(H2)(PMePh2)4 with 1 equiv. pyridine in benzene-d6.
Fig. 6. 400 MHz 1H NMR of ReCl(H2)(PMePh2)4 with excess pyridine in acetone-d6.
very weakly with the metal-bound hydrogens of 1 albeit
sufficient enough to slow down the rate of rotation. It
is also possible that the addition of pyridine resulted in
the formation of an unidentified paramagnetic species
which reduced the T1 time for the H2 ligand in 1.
However, 5, in the same NMR tube, displayed
markedly greater relaxation times for both triplets indi-
When 2,4,6-collidine is added to 1, no changes in
1
either the H (hydride region) or 31P spectra are ob-
served. This suggests that no displacement of the phos-
phine ligands occurred.
Room temperature T1 data for the metal bound
hydrogens of 1, 1 interacting with pyridine, 1 interact-
ing with 2,4,6-collidine, ReCl(H)2(PMePh2)2(Py)2 (5),
ReH3(PMePh2)4 (6) [18], and 6 interacting with pyridine
are shown in Table 3.
Collidine had no effect on the dihydrogen ligand of
1, while pyridine decreased the T1 time for the ligand.
Curiously, the dihydrogen quintet did not change loca-
1
tion in the H NMR spectrum, removing the possibility
of phosphine ligand displacement as an explanation of
the decrease in T1 time. This would imply that the
nitrogen lone pair of the pyridine may be interacting
Fig. 7. ReCl(H)2(PMePh2)2(Py)2.

R.L. Luck, R.S. O’Neill / Polyhedron 20 (2001) 773–782
779
Table 3
T1 NMR data for 1, 1 w/pyridine, 1 w/collidine, 5, 6 and 6 w/py in
acetone-d6 at 25°C
Compound
T1 value (ms)
Error (ms)
ReCl(H2)(PMePh2)4
ReCl(H2)(PMePh2)4
w/pyridine
147.0
138.4
1.8
1.2
ReCl(H2)(PMePh2)4
w/collidine
147.1
1.8
ReCl(H)2(PMePh2)2(Py)2
186.8 (l −7.88),
183.4 (l −7.97)
218.7
2.1, 2.3
Fig. 8. Theoretical ion distribution of the molecular ion
ReH3(PMePh2)4
ReH3(PMePh2)4
w/pyridine
5.1
6.6
[ReBr(H2)(PMePh2)4]+
.
216.8
halogen trans to the dihydrogen ligand in ReX(H2)-
(PMePh2)4, the ability of the halogen to donate s-elec-
tron density to the metal increases with the steric bulk
of the halogen [19]. The data suggest that the electronic
effects may be in competition with steric crowding
effects in these molecules. 2, by virtue of the lowest T1
(min) at 62 ms, shows itself to possess greater molecular
dihydrogen character than the parent complex. The
greater size of the bromide ligand compared to the
chloride may be forcing the hydrides closer together in
spite of greater electron donation on the part of the
bromide. This trend is reversed where 3 is concerned.
Based on the T1 data, ReI(H2)(PMePh2)4 displays more
classical H2 ligand behavior than does ReCl(H2)-
(PMePh2)4. The iodide ligand’s ability to donate elec-
tron density appears to outweigh steric considerations
in this case and thus more extensive s-donation by
iodide results in an increased d–p* metal–dihydrogen
interaction.
cating a transition towards a classical dihydrogen lig-
and. Therefore, the ligand substitution of two
methyldiphenylphosphine ligands by pyridine appears
to be unrelated to the decrease in T1 NMR time for the
quintet at l −8.55 ppm. ReH3(PMePh2)4, a known
classical polyhydride [18], was also evaluated for possi-
ble interaction with pyridine, but no change in relax-
ation time was observed. This removes from
consideration classical hydrogen to pyridine interac-
tions as well as interactions with the tertiary phosphine
ligands.
3.5. Variable temperature T1 (min) of 1, 2 and 3
T1 (min) data for 1, 2 and 3 can be found in Table 4.
As a means of testing the current method and instru-
mentation, the T1 (min) value for ReCl(H2)(PMePh2)4
was re-determined to be ꢀ88 ms, which was slightly
lower than the T1 (min) of 92 ms reported previously
[3].
3.6. Mass spectral analysis
From a graph of the above data, one can by interpo-
lation conclude that the T1 (min) for 2 is 62 ms and 3
is 101 ms at 400 MHz. Given the difficulties with
scaling for rhenium complexes, these values may indi-
cate that 2 has a less classical H2 ligand and 3 possesses
a more classical H2 ligand than 1. In changing the
FABMS was performed on a sample of compound 2
and Fig. 8 displays the theoretical ion distributions
(TID) for a molecular ion of this complex.
TIDs are determined by the relative abundance of
isotopes of the component elements of the expected
sample. The peak profiles are then compared to the
Table 4
Variable temperature T1 values for 1, 2 and 3
Temperature
(°C)
T1 (ms) for ReCl(H2)(PMePh2)4
in CD2Cl2
T1 (ms) for ReBr(H2)(PMePh2)4
in CD2Cl2
T1 (ms) for ReI(H2)(PMePh2)4
in acetone-d6
25
5
−5
−15
−25
−35
−45
−55
−75
123.7
107.0
105.9
74.08
63.46
62.28
66.29
78.51
88.53
173.1
133.7
89.91
80.36
96.89
113.2
100.9
101.3
106.1
121.5

780
R.L. Luck, R.S. O’Neill / Polyhedron 20 (2001) 773–782
Fig. 9. FABMS of ReBr(H2)(PMePh2)4 in acetone and 3-nitrobenzyl alcohol.
,
FABMS output to see if they are a good match at the
mass points of interest. A section of a FABMS analysis
of 2 is displayed in Fig. 9. The peaks located around
1017 are due to contamination of the sample with
complex 1.
The molecular ion pattern for 2 is centered at 1068
m/z as expected and thus we can conclude that the
molecular ion for 2 under these conditions is simply the
ionized version of the molecule, i.e. [ReBr(H2)-
(PMePh2)4]+.
in 3 of 2.707(8) and 2.807(10) A. These data were
obtained in structural studies on a solid state solution
of 1 and 3 [21]. Unfortunately, final figures of merit
worthy of publication were not obtained with this
determination, despite collecting data on crystals grown
from different sources. The point here is that it is not
clear how the different halide ligands would be better
able to s-donate electron density if the distance be-
tween them and the metal increases as is apparently the
case with 1, 2 and 3 [19].
The
compound
trans-dioxotetrakis(pyridine)-
rhenium(V) hydroxide (4), Fig. 11, was obtained in
crystalline form with approximately 2.0 water molecules
per Re-containing cation. Two trans-dioxotetrakis-
(pyridine)rhenium cations (with the Re atoms on inver-
sion points and one O atom and two pyridine molecules
per Re atom forming a complete cation), two O atoms
3.7. Structural results
Crystals of 2 were obtained in crystalline form with
one molecule of acetone per Re-containing molecule in
the unit cell. The geometry of the molecule is distorted
octahedral with two trans phosphine ligands lying
above and two below the equatorial plane as depicted
on special positions and three other
O atoms
,
in Fig. 10. The average Re–P distance was 2.443 A,
which was somewhat longer than reported for 1 [1]
,
(2.423 A) in the same solvent, see Table 5. There was a
disorder in the packing of the bromide ligand with
halogens appearing trans to each other and perpendicu-
lar to the plane of the phosphine ligands with one
bromide refined at 44% occupancy and the other at
56% occupancy. The P–Re–P (trans phosphines) angles
were 168.36 and 169.29° for 2 compared to 169.18 and
163.36° for 1 in earlier studies [1].
It is also noteworthy that the two Re–Br distances in
,
2 of 2.6592(15) and 2.688(3) A are significantly longer
than those for the Re–Cl ligand in 1 at 2.568 (3) and
,
2.587(3) A [1]. These are data from a conformer of 1
where the phosphine ligands adopted a conformation
that resulted in a disordered packing arrangement.
These distances can be compared to the Re–I distance
Fig. 10. ORTEP3 [20] representation of ReBr(H2)(PMePh2)4 (2) with-
out the acetone solvate.

R.L. Luck, R.S. O’Neill / Polyhedron 20 (2001) 773–782
781
Table 5
tween the dihydrogen ligand in 1 and the lone pair on
the pyridine molecule. Unfortunately, despite NMR
evidence that such an interaction may take place under
stoichiometric quantities of 1 and pyridine, under the
conditions of this particular experiment, ligand substi-
tution of PMePh2 by pyridine, oxidation of the Re
atom, reaction with water and chloride and dihydrogen
molecule displacement must have taken place to pro-
duce 4.
There was a previous study of this cation obtained
with a chloride counter anion and as a dihydrate [22].
Here the atoms in the Re-containing cation were lo-
cated on general positions and it is noteworthy that the
unit cell dimensions (monoclinic, Cc, a=13.577 (4),
,
Selected bond distances (A) and bond angles (°) for ReBr(H2)-
(PMePh2)4·(C3H6O)
Bond lengths
Re1–P2
Re1–P3
Re1–P4
Re1–P1
Re1–Br2
Re1–Br1
P1–C121
P1–C111
P1–C11
2.4377 (16)
2.4421 (16)
2.4450 (15)
2.4488 (16)
2.6592 (15)
2.688 (3)
P2–C21
1.837 (2)
1.841 (6)
1.849 (6)
1.837 (2)
1.842 (6)
1.860 (6)
1.836 (6)
1.853 (2)
1.854 (6)
P2–C221
P2–C211
P3–C31
P3–C321
P3–C311
P4–C411
P4–C41
1.838 (6)
1.853 (6)
1.854 (2)
P4–C421
Bond angles
P2–Re1–P3
P2–Re1–P4
P3–Re1–P4
P2–Re1–P1
90.33 (5)
169.29 (5)
90.87 (5)
90.80 (5)
168.36 (5)
90.16 (5)
94.91 (5)
83.88 (5)
95.80 (5)
84.47 (5)
85.02 (6)
96.29 (6)
84.27 (6)
95.35 (6)
179.81 (6)
99.4 (3)
C11–P1–Re1
C21–P2–C221
C21–P2–C211
113.43 (8)
101.4 (2)
96.3 (2)
,
b=11.951 (3), c=15.498 (4) A, i=116.3 (1)°) are very
different from those obtained in this study. However
the density at 1.79 (1) g cm−3 is similar to that ob-
tained in this study at 1.718 g cm−3 and this indicates
a similar tight packing arrangement. The average of the
C221–P2–C211 100.2 (3)
P3–Re1–P1
P4–Re1–P1
C21–P2–Re1
C221–P2–Re1
C211–P2–Re1
C31–P3–C321
C31–P3–C311
C321–P3–C311
C31–P3–Re1
C321–P3–Re1
C311–P3–Re1
C411–P4–C41
113.52 (8)
122.0 (2)
119.1 (2)
103.1 (2)
96.1 (2)
P2–Re1–Br2
P3–Re1–Br2
P4–Re1–Br2
P1–Re1–Br2
P2–Re1–Br1
P3–Re1–Br1
P4–Re1–Br1
P1–Re1–Br1
Br2–Re1–Br1
C121–P1–C111
C121–P1–C11
C111–P1–C11
C121–P1–Re1
C111–P1–Re1
,
Re–O distances at 1.764 (13) A, see Table 6, obtained
in that study are similar to those of Re1–O1 and
99.7 (3)
,
Re2–O2 at 1.767 (8) and 1.765 (9) A, respectively, in
113.52 (8)
120.79 (19)
119.73 (19)
102.7 (2)
this study. Further, the range of Re–N distances at
,
2.139 (9)–2.166 (14) A in the dihydrate study is also
similar to those obtained in this report at 2.131 (11)–
C411–P4–C421 100.0 (3)
,
2.150 (11) A.
C41–P4–C421
C411–P4–Re1
C41–P4–Re1
C421–P4–Re1
96.08 (19)
122.3 (2)
113.23 (4)
118.4 (2)
102.2 (2)
96.72 (19)
121.15 (19)
119.90 (19)
4. Conclusions
The synthesis and characterization of ReBr(H2)-
(PMePh2)4 and ReI(H2)(PMePh2)4 was accomplished
with this work. The T1 (min) times for these com-
pounds were assessed and found to be 62 and 101 ms,
respectively. These data would suggest
a more
Table 6
,
Selected Bond distances (A) and bond angles (°) for [(C5H5N)4O2Re]-
[OH]·1.75H2O
Bond lengths
Re1–O1
Re1–N1
Re1–N2
Re2–O2
Re2–N3
Re2–N4
N1–C11
1.767 (8)
2.140 (10)
2.142 (11)
1.765 (9)
2.131 (11)
2.150 (11)
1.313 (17)
N1–C15
N2–C25
N2–C21
N3–C35
N3–C31
N4–C41
N4–C45
1.339 (18)
1.319 (18)
1.337 (17)
1.39 (2)
1.41 (2)
1.320 (18)
1.341 (16)
Fig. 11. ORTEP3 [20] representation of the two [ReO2(py)4]+ cations
in 4.
assigned as water molecules constitute the asymmetric
unit. We were unable to determine which O atoms
represented the hydroxide anions. The geometry
around both Re atoms is octahedral with the four N
atoms of the pyridine molecules and the Re atom
defining the equatorial plane. There is no significant
difference in the two Re–O bond lengths for the two
cations.
Bond angles
O1–Re1–N1
O1–Re1–N2
N1–Re1–N2
O2–Re2–N3
O2–Re2–N4
N3–Re2–N4
C11–N1–C15
C11–N1–Re1
C15–N1–Re1
89.9 (4)
90.1 (4)
90.8 (4)
89.5 (5)
89.9 (4)
91.3 (4)
117.3(12)
122.4 (9)
120.2 (10)
C25–N2–C21
C25–N2–Re1
C21–N2–Re1
C35–N3–C31
C35–N3–Re2
C31–N3–Re2
C41–N4–C45
C41–N4–Re2
C45–N4–Re2
116.1 (13)
122.7 (10)
121.0 (10)
119.1 (15)
120.0 (11)
120.8 (12)
118.6 (12)
120.3 (9)
4 was obtained in an experiment which was designed
to produce material that would exhibit H-bonding be-
121.0 (9)

782
R.L. Luck, R.S. O’Neill / Polyhedron 20 (2001) 773–782
classical dihydrogen ligand for the iodo complex and a
less classical dihydrogen ligand for the bromo complex
relative to the parent complex, ReCl(H2)(PMePh2)4.
The JHP values for the compounds increase as the
halogen is changed from Cl at 19.2 Hz, Br at 20.0 and
I at 20.5 Hz. and the JHD values for the HD exchanged
version of the chloro analog could not be observed. The
JHP results suggest that the H2 ligand may be adopting
a more hydridic nature and this is at variance with the
T1 min results. Our studies here considering the values
for the T1 minimum and JHP suggest that the Br com-
plex 2 is not following the expected trends.
posit@ccdc.cam.ac.uk or www: http://www.ccdc.cam.
ac.uk).
Acknowledgements
We thank Michigan Technological University for
supporting this research and Jerry Lutz for assistance
with the NMR studies.
References
Some evidence suggestive of an h1-H2 ligand was
obtained by novel T1 studies aimed at investigating the
interaction of the dihydrogen ligand on 1 with the lone
pair on the N-atom of pyridine and 2,4,6-collidine. The
more hindered base displayed no interaction with the
rhenium compound while pyridine showed two effects.
First, the pyridine slightly lowered the T1 value of the
dihydrogen ligand (original quintet) and second, it dis-
placed phosphine ligands on the parent compound per-
haps to form ‘ReClH2(PMePh2)2(Py)2’. These effects
are separate and not related. The T1 time reduction
may be indicative of a hydrogen–lone pair interaction
between the metal–dihydrogen complex and the nitro-
gen of the organic base. Clearly more studies with other
dihydrogen compounds are needed in order to substan-
tiate that conclusion. Perhaps the effect only pertains to
stretched dihydrogen compounds.
The relationship between the halogen trans to a
dihydrogen ligand and the nature of the dihydrogen
ligand would appear to be more complex than had been
previously reported [23]. The electronegativity, orbital
overlap, and steric bulk of the atoms involved all
combine to produce a pattern of behavior that defies
simplistic models. These competing effects merit further
halogen substitution studies with more rigid bidentate-
diphosphine ligands [4]. Our results here are also at
odds with earlier work on complexes of the series
IrX(H2)H2(PR3)2 [23–25]. It must be noted that the
iridium complexes have a much less crowded coordina-
tion sphere than does the system studied herein.
[1] F.A. Cotton, R.L. Luck, Inorg. Chem. 30 (1990) 767.
[2] P.G. Jessop, R.H. Morris, Coord. Chem. Rev. 121 (1992) 155.
[3] F.A. Cotton, R.L. Luck, J. Chem. Soc., Chem. Commun. (1988)
1277.
[4] M. Kohli, D.J. Lewis, R.L. Luck, J.V. Silverton, K. Sylla, Inorg.
Chem. 33 (1994) 879.
[5] E. Rocchini, A. Mezzetti, H. Ru¨egger, U. Burckhardt, V. Gram-
lich, A. Del Zotto, P. Martinuzzi, P. Rigo, Inorg. Chem. 36 (1997)
711.
[6] K. Abdur-Rashid, D.G. Gusev, A.J. Lough, R.H. Morris,
Organometallics 19 (2000) 834 and refs. therein.
[7] J. Wessel, J.C. Lee, E. Peris, G.P.A. Yap, J.B. Fortin, J.S. Ricci,
G. Sini, A. Albinati, T.F. Koetzel, O. Eisenstein, A.L. Rheingold,
R.H. Crabtree, Angew. Chem., Int. Ed. Engl. 34 (1995) 2507.
[8] B.P. Patel, W. Yao, G.P.A. Yap, A.L. Rheingold, R.H. Crabtree,
J. Chem. Soc., Chem. Commun. (1996) 991.
[9] E.S. Shubina, N.V. Belkova, A.N. Krylov, E.V. Vorontsov, L.M.
Epstein, D.G. Gusev, M. Niedermann, H. Berke, J. Am. Chem.
Soc. 118 (1996) 1105.
[10] Y. Guari, J.A. Ayllon, S. Sabo-Etienne, B. Chaudret, B. Hessen,
Inorg. Chem. 37 (1998) 640.
[11] A.C.T. North, D.C. Phillips, F.S. Mathews, Acta Crystallogr.,
Sect. A 24 (1968) 351.
[12] P.T. Beurskens, G. Beurskens, W.P. Bosman, R. de Gelder, S.
Garcia-Granda, R.O. Gould, R. Israel, J.M.M. Smitts, The
DIRDIF-96 program system, Technical report of the Crystallogra-
phy Laboratory, University of Nijmegen, The Netherlands, 1996.
[13] G.M. Sheldrick, SHELXL-97, Program for crystal structure refine-
ment, University of Go¨ttingen, Germany, 1997.
[14] P.A. Maltby, M. Schlaf, M. Steinbeck, A.J. Lough, R.H. Morris,
W.T. Klooster, T.F. Koetzle, R.C. Srivastava, J. Am. Chem. Soc.
118 (1996) 5396.
[15] R. Chang, Chemistry, 4th edn., McGraw Hill, New York, 1991,
p. 364.
[16] R.L. Luck, E.J. Schelter, Inorg. Chem. (2001), accepted pending
revisions.
[17] P.C. Junk, J.W. Steed, J. Organomet. Chem. 587 (1999) 191.
[18] F.A. Cotton, R.L. Luck, Inorg. Chem. 28 (1989) 2181.
[19] F.A. Cotton, G. Wilkinson, Advanced Inorganic Chemistry, 5th
edn., Wiley, New York, 1988, p. 1300.
5. Supplementary material
[20] L.J. Farrujia, J. Appl. Cryst. 30 (1997) 565.
[21] R.L. Luck, R.S. O’Neill, Unpublished results.
[22] C.J.L. Lock, G. Turner, Acta Crystallogr., Sect. B 34 (1978) 923.
[23] T. Le-Husebo, C.M. Jensen, Inorg. Chem. 32 (1993) 3797.
[24] M. Mediati, G.N. Tachibana, C.M. Jenson, Inorg. Chem. 29
(1990) 3.
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC nos. 142490 for 2 and 142491 for
4. Copies of this information may be obtained from
The Director, CCDC, 12 Union Road, Cambridge,
CB2 1EZ, UK (fax: +44-1233-336033; e-mail: de-
[25] M. Mediati, G.N. Tachibana, C.M. Jenson, Inorg. Chem. 31
(1992) 1827.
.