2
G.K. Rao et al. / Polyhedron xxx (2017) xxx–xxx
The chemistry of rhenium(I) carbonyl complexes is dominated
by complexes derived from mono- and disubstitution of Re
CO) X (X = halo ligand). It has been known for nearly 50 years that
the common substitution pattern with Re(CO) X is sequential
replacement of up to two CO groups cis to the halo group thus lead-
ing to compounds with a fac-tri(carbonyl)Re core [37–40]. It is
both interesting and significant that even when potentially triden-
tate ligands were employed, the formation of bidentate coordina-
tion to facial tricarbonyl isomers are the only reported products
of X = Cl, this reaction led to the formation of compound 1 in 75%
yield. Consistent with this formulation as a mer-tricarbonyl species
was the appearance of three CO stretching bands at 1911, 1961,
(
5
ꢀ1
5
2065 cm in the infrared spectrum of the isolated product. The
3
1
1
P{ H} NMR spectrum of 1 exhibited a characteristic singlet at
68.8 ppm indicating equivalent P centers within 1 and this chemi-
cal shift was reminiscent of that observed for the PNP complex of
3
2,6-bis(di-tert-butylphosphinomethyl)-pyridine,
cis-[Re(
j
-2,6-
t
13
1
2 2 2 5 3 2
{ Bu PCH } NC H )(CO) Cl] [51]. The C{ H}NMR of 1 displayed
[
41–43]. One of these reports provides tantalizing spectroscopic
the anticipated two CO signals at d 190.7 and 198.5 ppm. Finally,
2
evidence of the conversion of the bidentate fac-[Re(
(
heating at to 270 °C in the solid state [43]. There is a unique report
for the facial tridentate coordination of 2,6-bis(pyrazolylmethyl)
pyridine (‘‘BPz”, Fig. S1) to yield the ionic species fac-[Re(
(
BPz)X(CO)
The recent reports of structurally and spectroscopically charac-
terized complexes of tridentate ‘‘NNN” ligand frames, such as ter-
pyridine and bis(imino)pyridine, have stimulated an increasing
interest in targeting pincer supported Re(I) carbonyl complexes
j
-terpy)Br
the proton NMR spectra showed a characteristic triplet for the
3
3
CO)
3
] complex into the cis-[Re(
j
-terpy)Br(CO)
2
] with CO loss by
para-CH for the pyridine at d 6.85 ppm ( JHH=8.4 Hz). This spectro-
3
scopic data pointed to the formation of the complex, mer-[Re(
j
-
+
ꢀ
2 2 5 3 3
2,6-{Ph PNH} NC H )(CO) ] Cl (1) formed by loss of two carbonyl
3
j
-BPz)
ligands and autoionization to liberate the chloro ligand.
+
ꢀ
2
CO)
3
] X starting from the more typical bidentate fac-[Re(
j
-
Fortunately, X-ray quality crystals of 1 could be reproducibly
obtained and used to confirm the proposed identity and structural
features for 1 and these results are summarized in the structural
3
] (X = Cl, Br, I) complex [44].
diagram presented in Fig. 1 (Tables S1, S2 and Fig. S2). The struc-
3
ture of 1 confirmed our proposition for the cation as mer-[Re(
j
-
+
2,6-{Ph
pseudo-octahedral Re(I) cationic species supported by the triden-
tate PN P pincer ligand and with the coordination environment
2
PNH}
2
NC
5
H )(CO)
3
3
]
and revealed
a
six-coordinate
[
45–50]. A common feature of these reports is these syntheses
required high temperature reactions in order to get tridentate
coordination.
The applications of pyridine-based ‘‘PNP” pincer ligands with Re
are particularly rare and only recently has the Re(I) chemistry been
3
completed by three meridional CO ligands. There are two different
environments for the CO ligands with one of them laying in the
pseudo-equatorial site of the ligand plane (C(31)–O(2)) and the
other two residing in axial sites and trans to each other. The chlo-
ride counterion, that originated from the Re starting material, was
well-separated from the Re-centered cation.
investigated. Given the strong historical precedent for bidentate
t
ligand coordination, the reaction of Re(CO)
butylphosphinomethyl)pyridine to yield cis-[Re(
PCH NC )(CO) Cl] under rather mild conditions (80 °C/THF)
is noteworthy [51]. The non-innocent nature of this ligand frame
was exploited for reactions with CO [52]. In that case, the dearo-
matized ligand, generated by deprotonation of a CH group, was
able to participate in reversible CO binding and hydrogenation
chemistry. Only three reports for higher oxidation state Re com-
plexes with ‘‘PNP” ligands have appeared. A Re(III) complex of this
ligand family, mer-[Re(
structurally characterized [53] as have three mixed valent, multi-
5
Cl with 2,6-bis(di- -
3
t
j
2
-2,6-{ Bu -
2
}
2
H
5 3
2
Similar reactivity was observed when the bromo analog, Re
(CO)
5
Br, was employed in the analogous reaction scheme with L1
3
2
as shown in Scheme 1 to yield the ionic compound 2, [Re(
2 2 5 3 3
{Ph PNH} NC H )(CO) ] Br . Not surprisingly, compound 2 yielded
j
-2,6-
+
ꢀ
2
2
NMR and IR spectroscopic signatures very similar to 1 which sup-
ported our initial formulation. Again, we obtained crystals of 2
which were analyzed by single crystal X-ray analysis and which
gave definitive confirmation of a nearly identical connectivity of
the cationic Re complex with the non-coordinated anion replace
by bromide (Tables S1, S2 and Fig. S3). Complexes 1 and 2 not only
3
j
2
-2,6-{Cy PCH
2
}
2
NC
5
H )Cl
3 3
]
has been
5+
ple bonded dirhenium complexes in which the [Re
2
]
core has
NC
ligand [54]. Finally, some rhenium(V) oxo complexes bearing
3
one of the Re centers coordinated to a
j
-2,6-{Ph
2
PCH
2
}
2
5
H
3
3
represent the first complexes of the ‘‘PN P” ligand family, they also
provide structures that significantly contrast with reported pincer
t
different bis(R
2
phosphinomethyl)pyridine (R = Cy, Bu, Ph) ligands
species. In particular, the autoionization of the Re-halo group and
+
have been reported and the non-innocent behavior of the ‘‘PNP”
ligands discussed [55].
more specifically the formation of the mer-Re(CO)
3
core.
In terms of ligand variation, we anticipated that the change of
the N-H of the ligand to an N-Me moiety to give L2 would intro-
duce a steric load to the ligand, may provide stronger electron
donation to a coordinated metal center and remove the potentially
reactive NH group [28–30,35,36]. Under the same conditions
Replacement of one of the PR
NEt gives ‘‘PNN” ligand, 2-[(diethylamino)methyl]-6-
(diphenylphosphino)methyl]pyridine. This ligand reacted
smoothly with Re(CO) X at 110 °C in toluene to yield the Re(I) pin-
2
arms of the ‘‘PNP” frame with
2
a
[
5
cer complexes cis-[Re(PNN)(CO)
2
X] (X = Cl, Br) analogs of the ‘‘PNP”
employed for the reaction with L1, the reaction of Re(CO)
5
Cl was
complex [56].
carried out with L2 to yield complex 3 as shown in Scheme 2.
0
31
1
Herein, we report the first Re complexes of N,N -bis
The single P{ H} resonance d = 87.1 ppm suggested a symmetri-
cal coordination environment. The IR spectrum of 3 displayed only
0
(
(
diphenylphosphino)-2,6-aminopyridine
(L1)
and
N,N -bis
ꢀ1
diphenylphosphino)-2,6-(methylamino)pyridine (L2). Of particu-
P frameworks yielded tridentate complexes
under rather mild conditions compared to the recently reported
‘NNN” frameworks. The formation of meridional-tricarbonyl com-
two mCO at 1834 and 1954 cm . These carbonyl carbons appeared
1
3
1
lar note is that the PN
3
in the C{ H} NMR at d 187.8 and 194.5 ppm.
The spectroscopic divergence between 3 and 1 suggested a
structural difference for 3. Colorless crystals of 3 were readily
obtained from dichloromethane and the results of a single crystal
X-ray analysis are displayed in Fig. 2 (Tables S3, S4 and Fig. S4).
The distorted pseudo-octahedral Re coordination environment is
constructed from the meridionally coordinated pincer ligand, two
cis-oriented carbonyl ligands, and a chloro ligand. The chloro group
‘
plexes was dominant and the products of these reactions showed
a dependence on the identity of the ligand N-R group as well as
the halo group bonded to Re starting material.
2
. Results and discussion
3
is cis to the Npy and perpendicular to the PN P coordination plane.
The two CO ligands reside in different environments with one
(C(33)–O(2)) being along the same axis and trans to the chloro
ligand and the other (C(32)–O(1)) residing in an equatorial site.
Complex 3 displays a coordination environment reminiscent of
0
Reaction of soluble N,N -bis(diphenylphosphino)-2,6-aminopy-
ridine ligand (L1) with the Re(I) starting material, Re(CO)
X = Cl, Br), followed the routes outlined in Scheme 1. In the case
5
X
(