Experimental and theoretical studies on arene-bridged metal-metal-bonded dimolybdenum complexes

Article abstract of DOI:10.1002/chem.201400236

The bis(hydride) dimolybdenum complex, [Mo₂(H) ₂{HC(N-2,6-iPr₂C₆H₃) ₂}₂(thf)₂], 2, which possesses a quadruply bonded Mo₂^II core, undergoes light-induced (365 nm) reductive elimination of H₂ and arene coordination in benzene and toluene solutions, with formation of the Mo^I₂ complexes [Mo₂{HC(N-2,6-iPr₂C₆H₃) ₂}₂(arene)], 3·C₆H₆ and 3·C₆H₅Me, respectively. The analogous C ₆H₅OMe, p-C₆H₄Me₂, C ₆H₅F, and p-C₆H₄F₂ derivatives have also been prepared by thermal or photochemical methods, which nevertheless employ different Mo₂ complex precursors. X-ray crystallography and solution NMR studies demonstrate that the molecule of the arene bridges the molybdenum atoms of the Mo^I₂ core, coordinating to each in an η² fashion. In solution, the arene rotates fast on the NMR timescale around the Mo₂-arene axis. For the substituted aromatic hydrocarbons, the NMR data are consistent with the existence of a major rotamer in which the metal atoms are coordinated to the more electron-rich C-C bonds. Quintuply bound [Mo₂{HC(N-2,6-iPr ₂C₆H₃)₂}₂(arene)] complexes (arene=C₆H₆, C₆H₅Me, C₆H₅OMe, p-C₆H₄Me₂, C₆H₅F, and p-C₆H₄F₂ and Dipp=2,6-iPr₂C₆H₃) have been prepared and structurally characterized by experimental and theoretical methods (see figure). Coordination of the arene to the quintuple Mo^I-Mo^I bond results in an effective metal-metal bond order substantially lower than five as a consequence of a strong electronic interaction between δ and δ* dimetal orbitals and the π and π* orbitals of the arene.

Full text of DOI:10.1002/chem.201400236

DOI: 10.1002/chem.201400236
Full Paper
&
Quintuple Bonds
Experimental and Theoretical Studies on Arene-Bridged Metal–
Metal-Bonded Dimolybdenum Complexes
Mario Carrasco,[a] Natalia Curado,[a] Eleuterio ꢀlvarez,[a] Celia Maya,[a] Riccardo Peloso,[a]
Manuel L. Poveda,[a] Amor Rodrꢁguez,[a] Eliseo Ruiz,[b] Santiago ꢀlvarez,*[b] and
Ernesto Carmona*[a]
Abstract: The bis(hydride) dimolybdenum complex,
[Mo2(H)2{HC(N-2,6-iPr2C6H3)2}2(thf)2], 2, which possesses
a quadruply bonded Mo2 core, undergoes light-induced
nevertheless employ different Mo2 complex precursors. X-ray
crystallography and solution NMR studies demonstrate that
the molecule of the arene bridges the molybdenum atoms
of the MoI2 core, coordinating to each in an h2 fashion. In so-
lution, the arene rotates fast on the NMR timescale around
the Mo2-arene axis. For the substituted aromatic hydrocar-
bons, the NMR data are consistent with the existence of
a major rotamer in which the metal atoms are coordinated
to the more electron-rich CꢀC bonds.
II
(365 nm) reductive elimination of H2 and arene coordination
in benzene and toluene solutions, with formation of the
MoI2 complexes [Mo2{HC(N-2,6-iPr2C6H3)2}2(arene)], 3·C6H6
and 3·C6H5Me, respectively. The analogous C6H5OMe, p-
C6H4Me2, C6H5F, and p-C6H4F2 derivatives have also been pre-
pared by thermal or photochemical methods, which
Introduction
namely the bis(amidinate) complexes [Mo2{RC(N-2,6-
iPr2C6H3)2}2] (R=H, Ph), which were obtained by reduction with
[8]
ꢁ
Dimolybdenum complexes containing multiple bonds between
metal atoms were first uncovered almost five decades ago.[1]
Despite the time that has elapsed and the spectacular devel-
opment experienced by the field,[2] interesting discoveries con-
tinue to be made. The quadruple MoꢀMo bond has been in-
tensively investigated.[2–5] Its dimetal unit with a s2p4d2 elec-
tron configuration is redox active and suitable to study mixed
valency.[4] However, even if a two-electron reduction is expect-
ed to lead to a s2p4d4 electron configuration and hence to five-
fold MoꢀMo bonding, this was not experimentally achieved
until recently. Following the discovery by the Power group of
the first quintuple bond in an isolable molecule, Ar’CrCrAr’,
(Ar’=C6H3-2,6-(C6H3-2,6-iPr2)2)[6] and the subsequent burgeon-
ing of the field,[7] in 2009 prominent research from the group
of Tsai allowed isolation of the first molecules featuring
a formal bond order of five between molybdenum atoms,
potassium graphite of suitable MoꢁMo precursors. Since
then, other compounds with CrꢀCr[9a] and MoꢀMo[9b] quintuple
bonds have been characterized. Reactivity studies of the quin-
tuple MꢀM bond have been reported too.[7i,9a,10,11]
Some years ago our group started to study the reactions of
the dimolybdenum tetracarboxylates [Mo2(O2CR)4] (R=H, CH3,
CF3) with the lithium derivatives of bulky terphenyl, amidinate,
and aminopyridinate ligands, aiming to obtain suitable precur-
ꢁ
sors for further reduction to the corresponding MoꢁMo com-
plexes. In collaboration with the group of Power, we reported
recently the formation of some mono- and bis(terphenyl) di-
molybdenum compounds of unusual unsaturated struc-
tures.[12a] Latterly, our group also communicated that the bis-
(hydride)
bis(amidinate)
derivative,
[Mo2(H)2{HC(N-2,6-
iPr2C6H3)2}2(thf)2], 2, with a MoII2 core, and Tsai’s complex,[8]
[Mo2{HC(N-2,6-iPr2C6H3)2}2], 4, interconvert readily by formal di-
nuclear reductive elimination and oxidative addition of H2.[12b]
This work also demonstrated that interesting MoI2·arene struc-
tures, compounds 3·arene, could be generated starting from
either 2 or 4, under suitable conditions. Multiply bonded M2
compounds with bridging arene ligands are scarce; the first ex-
ample of a quadruply bonded MoII2 species of this kind was
described recently by Masuda and co-workers.[13,14] Accordingly,
we extended our initial study that was limited to C6H6 and
C6H5Me, to other arenes, including C6H5OMe, p-C6H4Me2 and
the fluoroarenes C6H5F and 1,4-C6H4F2. Herein, we give a full
description of this work that encompasses reactivity, structural
(solid-state and solution) and theoretical studies of the
MoI2·arene motif.
[a] Dr. M. Carrasco, N. Curado, Dr. E. ꢀlvarez, Dr. C. Maya, Dr. R. Peloso,
Prof. Dr. M. L. Poveda, Dr. A. Rodrꢁguez, Prof. Dr. E. Carmona
Instituto de Investigaciones Quꢁmicas
Departamento de Quꢁmica Inorgꢂnica Universidad
de Sevilla-Consejo Superior de Investigaciones Cientꢁficas
Avenida Amꢃrico Vespucio 49, 41092 Sevilla (Spain)
Fax: (+34)954460565
E-mail: guzman@us.es
[b] Prof. Dr. E. Ruiz, Prof. Dr. S. ꢀlvarez
Departament de Quꢁmica Inorgꢄnica and
Institut de Quꢁmica Teꢅrica i Computacional
Universitat de Barcelona, Diagonal 647, 08028 Barcelona (Spain)
E-mail: santiago@qi.ub.es
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201400236.
Chem. Eur. J. 2014, 20, 6092 – 6102
6092
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1525 cmꢀ1 due to n(MoꢀH), which disappeared upon deutera-
tion of a solid sample investigated by diffuse reflectance infra-
red spectroscopy (DRIFT). The corresponding n(MoꢀD) band,
which would be expected around 1090 cmꢀ1 on the assump-
tion of a n(MoꢀH)/n(MoꢀD) ratio of ꢂ1.4, could not be as-
signed with certainty because of the presence in the 1060–
1120 cmꢀ1 region of the spectrum of other bands originated
by vibrations of the amidinate ligands. In the 1H NMR spec-
trum, a singlet at d=5.67 ppm, with relative intensity corre-
sponding to 2H, can be confidently attributed to the two
equivalent hydrides. This signal vanished upon treatment with
D2, when it became clearly discernable in the 2H NMR spectrum
of the deuterated sample. The chloroform test provided further
support for this assignment (see the Supporting Information).
In this regard, it should be noted that even if the MoꢀH reso-
nance of 2 appears at a higher frequency than is commonly
observed for transition-metal hydrides, metal–metal-bonded
molybdenum hydrides included,[17] our observation is not un-
common for MoꢀH compounds, and in general for early transi-
tion-metal hydrides.[18–20] The structure proposed for 2 was fur-
ther substantiated by a single-crystal X-ray study (Figure 1). De-
spite the caveats of defining hydride positions by X-ray diffrac-
tion methods, the two hydride ligands were located in the dif-
ference Fourier maps and were ascertained to be coplanar
with the two Mo atoms. The MoꢀH distance of 1.71 ꢃ is in the
1.55–1.75 ꢃ range expected for terminal MoꢀH bonds. On the
other hand the MoꢀMo bond length of 2.077(1) ꢃ is in accord-
ance with a quadruple MoꢀMo bond formulation.[21]
Results and Discussion
Bis-(carboxylate) and -(hydride) complexes of the bis(amidi-
nate) fragment [Mo2{HC(N-2,6-iPr2C6H3)2}2]
The MoI2·arene complexes described in this paper emanated
originally from the bis(acetate) bis(amidinate) dimolybdenum
compound, 1a, that exhibits a quadruple MoꢀMo bond. This
complex, and the formate analogue 1b, were generated by
treatment of the lithium amidinate, Li[HC(N-2,6-iPr2C6H3)2] with
the corresponding [Mo2(O2CR)4] precursor, under the condi-
tions shown in Scheme 1. Both are yellow, air-sensitive crystal-
line solids, of low solubility in common organic solvents (Et2O,
THF or aromatic hydrocarbons). Solid-state magnetic suscepti-
bility, NMR and other characterization data for the two com-
plexes are given in the Supporting Information. Their solid-
state molecular structures were determined by X-ray crystallog-
raphy (see the Supporting Information, Figure S5 and S6). The
two compounds feature a MoꢀMo quadruple bond of length
ꢂ2.10 ꢃ. Interestingly, the formate derivative 1b, which was
crystallized from a hot, saturated solution in tetrahydrofuran,
contains in the axial positions two weakly bound molecules of
the solvent, with a MoꢀO separation of 2.75 ꢃ. As in other
ꢁ
MoꢁMo compounds with axial ligands, the coordination of
these extra molecules has a minimal effect on the MoꢀMo
bond,[2] which at 2.1047(7) ꢃ is only marginally longer than the
2.0892(8) ꢃ value found for 1a.
Attempts to reduce THF solutions of these compounds with
Na, K, KC8 or KH, either at room temperature or at 608C,
proved fruitless. However, acetate 1a reacted cleanly with
LiMe with displacement of the two acetate groups and forma-
tion of the “ate” complex Li[Mo2(CH3)3{HC(N-2,6-iPr2C6H3)2}2-
(thf)], which underwent readily loss of LiMe with production of
the neutral species [Mo2(CH3)2{HC(N-2,6-iPr2C6H3)2}2]. These and
other related complexes with metal–metal quadruple bonding
will be described in a forthcoming publication. Meanwhile,
action of H2 on the latter complex under ambient conditions
readily eliminated CH4 and produced the bis(hydride)
[Mo2(H)2{HC(N-2,6-iPr2C6H3)2}2(thf)2], 2, as represented in
Scheme 1.
Synthesis and solid-state structure of arene-bridged adducts
[Mo2{HC(N-2,6-iPr2C6H3)2}2(arene)] (3·arene)
To induce reductive elimination of dihydrogen from the bis-
(hydride) complex 2, and form the known binuclear species
[Mo2{HC(N-2,6-iPr2C6H3)2}2] (4), with
a
MoꢀMo quintuple
bond,[8] we performed the photolysis of a benzene solution of
2. After stirring for 24 h at room temperature under UV light
(365 nm), clean conversion to a new dark-red complex, identi-
fied as [Mo2{HC(N-2,6-iPr2C6H3)2}2(C6H6)] (3·C6H6; Scheme 2A)
was observed. Spectroscopic and X-ray studies demonstrated
that a molecule of C6H6 bridges the Mo atoms of the electron-
rich, formally MoI2 central unit, coordinating to each in a h2
fashion through the C1ꢀC2 and C4ꢀC5 bonds.
Complex 2 is a rare example of a formally quadruply
bonded dimolybdenum hydride species.[15–17] Its MoII2(H)2 core
is characterized by a medium intensity IR absorption at
ꢁ
Scheme 1. Syntheses of bis(amidinate) MoꢁMo complexes with carboxylate (1) and hydride ligands (2).
Chem. Eur. J. 2014, 20, 6092 – 6102 www.chemeurj.org
6093
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ꢁ
the synthesis and structural characterization of a new MoꢁMo
complex supported by 1,4-diazadiene-derived diamido ligands,
with a molecule of benzene bound to each metal center in an
h2 manner, also through the C1ꢀC2 and C4ꢀC5 bonds. Subse-
quently, the same group extended these results to the toluene
and anisole analogues.[14] Nonetheless, despite extensive reac-
tivity studies on the quintuple metal–metal bond,[7i,9a,10,11] its
interaction with aromatic hydrocarbons has not been investi-
gated. This is surprising, particularly if one considers that in
landmark dichromium molecules by the Power group, that is,
Ar’CrCrAr’,[6,7a] each chromium atom exhibits a CrꢀCarene interac-
tion with the ipso carbon of a flanking aryl ring belonging to
the terphenyl ligand bonded to the other. This so-called secon-
dary arene interaction was later shown by quantum mechani-
cal calculations to cause only a slight weakening of the CrꢀCr
quintuple bond, whereas for cobalt and iron complexes strong
h6-arene coordination precluded significant metal–metal bond-
ing.[25]
Figure 1. Molecular structure of 2 with anisotropic displacement parameters
drawn at the 50% level and all hydrogen atoms omitted for clarity, except
H1. This atom was determined by difference Fourier synthesis and refined,
along with its isotropic thermal parameters, by least-square procedures.
On these grounds, it was thought worthwhile to expand our
knowledge of these molecules, and in this vein this study fo-
cuses on a range of arenes, which in addition to benzene and
toluene,[12b] includes anisole, 1,4-C6H4Me2, and the fluoroarenes
C6H5F and 1,4-C6H4F2. The complexes 3·arene were synthesized
by one or more of the pathways A--C in Scheme 2. Routes A
and B are photolytic (365 nm), whereas C involves direct coor-
dination of the arene to the
Transition-metal arenes are a prominent family of organome-
tallic compounds.[22—24] However, molecules that possess a mul-
tiple metal–metal bond coordinated to an aromatic hydrocar-
bon that actually bridges the two metal atoms are very rare.
As mentioned briefly, in 2009 the group of Masuda[13] reported
MoꢀMo quintuple bond in Tsai’s
complex.[8] Alternatively, these
compounds may be produced
thermally by arene exchange re-
actions (see below), but experi-
mentally this latter procedure is
less convenient as prolonged
heating at 1208C in the dark is
required. Accordingly, the C6H6,
C6H5CH3, and 1,4-C6H4(CH3)2 ad-
ducts were quantitatively ob-
tained starting from the bis-
(hydride) compound 2 (route A
in Scheme 2) and this was also
a
suitable method to isolate
3·anisole, which was additionally
generated through procedure B.
However, because of the insolu-
bility of bis(hydride) 2 in C6H5F
and p-C6H4F2, corresponding
fluoarenes,
3·C6H5F
and
3·C6H4F2, were best prepared
either by route B or C. Unfortu-
nately, we did not succeed in
isolating analogous complexes
containing hexafluorobenzene,
pyridine or anthracene, as bridg-
ing aromatic molecules.
All arene complexes 3 are very
air-sensitive both in solution and
in the solid state. They were iso-
Scheme 2. Photochemical and thermal generation of complexes 3·arene (arene=C6H6, C6H5CH3, C6H5OCH3, 1,4-
C6H4(CH3)2, C6H5F, and 1,4-C6H4F2).
Chem. Eur. J. 2014, 20, 6092 – 6102
www.chemeurj.org
6094
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
lated as dark-red diamagnetic crystalline solids. In solution,
they exhibit a strong absorption in the higher energy region of
the Vis spectrum (ca. 435 nm; e ca. 3000 molꢀ1 Lcmꢀ1), along
with other lower energy and intensity bands in the region
580–600 nm (e ca. 700 molꢀ1 Lcmꢀ1).
leans toward a m-h3:h3 binding mode. There is a perceptible
(albeit relatively small) arene distortion upon coordination that
leads, among other variations, to ring expansion. For the four
compounds studied by using X-ray crystallography, the molyb-
denum-bound CꢀC bonds are the longest (average distances
between 1.45 and 1.47 ꢃ). Then, in 3·C6H6 and 3·C6H4Me2 there
are two intermediate bonds (ca. 1.42–1.44 ꢃ) and two slightly
shorter ones of length ꢂ1.40 ꢃ, the latter being identical to
the CꢀC bonds in free benzene. Somewhat differently, in
3·C6H5Me and 3·C6H4F2 the non-coordinated CꢀC bonds are
almost equal and average ꢂ1.41 ꢃ. Taking 3·C6H6 as a represen-
tative example, ring expansion upon coordination leads to
a total sum of CꢀC distances of ꢂ8.55 ꢃ, which is 0.21 ꢃ
longer than in free C6H6. In our view, this is not unexpected as
two CꢀC bonds bind to the electron-rich MoI2 unit. For exam-
ple, in the also electron-rich Mo(0) complex [Mo(C2H4)2(PMe3)4],
the coordinated CꢀC bonds are 0.06 ꢃ longer than in C2H4.[27]
In fact, this is a common observation in transition-metal com-
plexes with low formal oxidation states. For instance, the ethyl-
ene CꢀC distance in Zeise’s salt is very close to that of free eth-
ylene (1.34 ꢃ), whereas in the Pt(0) complex [Pt(C2H4)(PPh3)2]
the CꢀC bond length increases to 1.43 ꢃ.[28]
In addition to these changes, there is some deviation from
planarity. For example, for 3·C6H4F2 the dihedral angle between
the C(2)-C(1)-C(6) and C(3)-C(4)-C(5) planes is of 3.58, whereas
for 3·C6H4Me2 the corresponding angle is of 15.58. The C6H6
and C6H5Me adducts exhibit intermediate values of 8.9 and
6.88, respectively. Therefore, the
X ray crystallographic studies of some of the arene com-
plexes 3 were accomplished. Their solid-state molecular struc-
tures are presented in Figure 2. Despite substantial differences
in the donor–acceptor properties of the aromatic molecules in-
vestigated, their Mo2-arene linkages display similar structural
parameters. Thus, the MoꢀMo distances have almost the same
value within experimental error, as they range from 2.105(1) ꢃ
in 3·C6H5Me to 2.117(1) in 3·C6H4F2. This bond length is longer
than in Tsai’s complex by nearly 0.1 ꢃ, and even if it lies in the
upper part of the 2.02–2.12 ꢃ interval calculated theoretically
for fivefold MoꢀMo bonding,[26] it is clearly signaling that the
effective metal–metal bond order in compounds 3 must be
substantially lower than five (see below).
Coordination of the CꢀC bonds to the Mo atoms is symmet-
ric and it is characterized by MoꢀC distances of about 2.22 ꢃ
in all compounds. Instead, distances between each Mo atom
and the two remaining arene carbon atoms are significantly
longer (between 2.63 and 2.74 ꢃ, approximately) and may be
judged as non-bonding. There is one exception to this trend
that appears in 3·C6H4Me2, in which each Mo atom features
a MoꢀC separation shorter than the other (2.596(2) ꢃ vs.
2.726(2) ꢃ), so that in the solid state p-xylene coordination
aromatic ring is nearly planar in
3·C6H4F2 but becomes signifi-
cantly folded in 3·C6H4Me2. Like-
wise, arene substituents undergo
some bending out of the ring
plane (1.4, F; 3.5 and 13.68 for
the Me groups in 3·C6H5Me and
3·C6H4Me2, respectively).
In all these complexes the aryl
groups of the amidinate ligands
pair up in a form that minimizes
their steric interactions. Interest-
ingly, in complex 3·p-C6H4Me2
the para CꢀMe substituents are
not symmetrically oriented rela-
tive to the MoꢀMo bond but
adopt a distribution that reduces
steric strain with the amidinate
aryl substituents.
In summary, as a general rule,
coordination of arenes to transi-
tion metals in low formal oxida-
tion states causes an elongation
of the six aromatic CꢀC bonds,
deviation from planarity and
a bending of the arene substitu-
ents with respect to the ring pla-
narity. These distortions may be
induced by charge-transfer from
Figure 2. ORTEP drawing of 3·arene complexes (arene=C6H6, C6H5Me, p-C6H4Me2, and p-C6H4F2) in which each
atom is represented by its ellipsoid of thermal vibration to enclose 50% of the electron density.
Chem. Eur. J. 2014, 20, 6092 – 6102
www.chemeurj.org
6095
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
the electron-rich metal center to the arene. However, since the
distortions ascertained for compounds 3 are comparatively
smaller than in C6H6 complexes formulated as benzenide di-
anions,[29,30] we conclude that such charge-transfer is modest in
the Mo2·arene derivatives 3. In accordance with this premise,
we formulate complexes 3 as MoI2 species with a coordinated
neutral molecule of the arene, a proposal that is supported by
the solution behavior of these complexes, in particular by their
arene substitution chemistry that will be discussed below.
Solution structure and reactivity of complexes 3·arene
To disclose the structural characteristics of the MoI2·arene link-
age of complexes 3·arene in solution, we inspected their NMR
spectroscopic properties. The analysis included variable tem-
perature (ꢀ80 to 808C) 1H and 13C{1H} NMR experiments, as
well as NOESY and other 2D studies. All 3·arene compounds
exhibit dynamic behavior in solution due to fast rotation of
the arene around the Mo2-arene axis. With reference, for the
sake of simplicity, first to the benzene complex 3·C6H6, eight
doublet and four septet resonances appear in the 1H NMR
spectrum due respectively to the methyl and methine protons
of the eight iPr ligand substituents, in accordance with C2 mo-
lecular symmetry. This observation clearly indicates that in so-
lution the amidinate ligands maintain the conformation ob-
served in the solid-state structure and furthermore that rota-
tion of the non-equivalent aryl substituents (A and B in
Figure 3) is slow on the NMR timescale (500 MHz, 258C). Inter-
estingly, the high-temperature NOESY spectrum of 3·C6H6
(C6D6, 500 MHz, 708C) revealed the exchange of two of the
four pairs of iPr groups. A reasonable explanation is that this
results from rotation around the NꢀCar bond of two C2-related
aryl substituents (either A or B in Figure 3). The six CꢀH groups
of the m-benzene ring appear as isochronous at room tempera-
Figure 3. Schematic representations of the solution structures of complexes
3·arene. In all cases the rotation of the arene around the MoI2 core is fast on
the NMR timescale at 258C (500 MHz).
In the remaining 3·arene complexes, fast rotation of the
arene engenders effective C2 symmetry and pairs up the 2,6-
iPr2C6H3 amidinate substituents. As a consequence, their NMR
spectroscopic features are similar to those of the benzene
complex analogue. Similarly to 3·C6H6, two of the four pairs of
iPr groups undergo exchange due to rotation around the Nꢀ
Car bonds, but the movement is now discernible in the NOESY
experiment at 258C. Considering the electronic diversity of
substituents (Me, OMe, and F) the somewhat more facile rota-
tion around the NꢀCar bonds may be due to the steric pressure
exerted by the coordinated molecule of the arene.
1
ture and yield only one resonance in the H (d=3.87 ppm) and
13C{1H} NMR spectra (d=71.2 ppm). Cooling at ꢀ808C causes
a significant broadening of the signals but does not permit
separation of those corresponding to the four molybdenum-
bound and the two non-bonded CꢀH units. These data imply
that rotation around the MoI2 core is a low-barrier process that
exchanges the C(H)–C(H) bonds fast in the NMR timescale.
From the variable temperature NMR data, a crude estimation
¼
of DG =(7.5ꢃ2.6) kcalmol-1 can be made for this process.
This is in contrast with the results reported by the group of
Masuda for a MoII2·C6H6 complex, in which signals of the coor-
dinated and non-coordinated C(D)–C(D) bonds were observed
The NMR parameters of the arene ligand are very informa-
tive with regard to the nature of the main rotamer present in
solution (only one out of the three possible rotameric struc-
tures). For instance, in 3·C6H5F all the arene CꢀH units are non-
equivalent and feature d values that support the structure rep-
resented in Figure 3 (bottom left) for the major rotamer, in
which the CꢀF and its para CꢀH carbon atoms are uncoordi-
nated. Thus, the four Mo-bound CꢀH groups resonate at con-
siderably lower frequencies relative to free C6H5F (by ca. d=4
2
in the room temperature H NMR spectrum of the C6D6 isoto-
1
pologue.[13] In 3·C6H6 both the H and 13C chemical shifts have
moved by about d=3.3 and 56 ppm, respectively, to lower fre-
quency relative to uncoordinated benzene (evidently, the ob-
served d values are the average of those corresponding to four
coordinated and two non-coordinated CꢀH units). Such shifts
are common in transition-metal olefin, diene, and arene p-
complexes and are a consequence of the nature of the MꢀL(p)
bonding.[31] No significant variation was measured in the value
1
and 77 ppm, for the H and 13C nuclei, respectively), whereas
the 13CꢀF signal remains essentially unchanged and those of
the para CꢀH nuclei move to lower frequency, but only by
about d=1 (1H) and 20 ppm (13C). Similar NMR patterns and
chemical shift variations are found in the NMR spectra of
1
of the averaged JCꢀH (ca. 164 Hz) indicating very little change
in hybridization upon coordination.
Chem. Eur. J. 2014, 20, 6092 – 6102
www.chemeurj.org
6096
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
3·C6H4F2 and 3·C6H5OMe, underpinning analogous solution
structures, which for the former complex is coincident with the
solid-state structure.
the reaction was disclosed by 1H NMR spectroscopic studies
(C6D6) of the crude product obtained after evaporation of the
arene solvent mixture (see the Supporting Information,
Table S1). For experiments in which C6H6 was one of the com-
ponents, 3·C6H6 was the exclusive reaction product (arene-2:
toluene, anisole, C6H5F and C6H4F2). Similarly the p-C6H4F2/p-
C6H4Me2 mixture resulted in 3·C6H4F2 as the only detectable
arene complex. Interestingly, the pairs C6H5Me/p-C6H4Me2 and
C6H5F/p-C6H4F2 yielded about a 3:1 ratio in favor of the mono-
substituted arene, whereas closer ratios of ꢂ3:2 were ascer-
tained for the C6H5Me/C6H5F and C6H5Me/p-C6H4F2 couples. As
A rotamer with a different disposition of the coordinated
arene with respect to the MoI2 linkage predominates in solu-
tion for the toluene and p-xylene complexes (Figure 3). For the
latter, the coordination of the arene is identical to that found
in the solid state, as revealed by the following observations:
the two coordinated quaternary C(Me) nuclei move to d=
71.4 ppm and the coordinated CꢀH to d=52.9 ppm (more
than d=60 and 75 ppm respectively, to lower frequency rela-
tive to the uncoordinated arene), whereas the non-bonded Cꢀ indicated below, these results reveal the importance of steric
effects.
An analogous approach to that summarized in Scheme 3
H carbon nuclei resonate with d=103.5 ppm (a variation of
about d=25 ppm). Comparable features are observed in the
1H NMR spectrum. With regards to 3·C6H5Me, NMR chemical
shifts of the coordinated arene are in accordance with a similar
structure, with a C(Me)ꢀC(H) bond bound to one of the Mo
atoms, and therefore dissimilar to the solid-state structure.
Thus, in the 13C{1H} NMR spectrum the C(Me) nucleus has
a chemical shift of d=70.1 ppm; then, there are three, compa-
ratively low frequency, CꢀH signals in the interval d=59–
65 ppm, and two higher frequency ones around d=92 ppm. In
summary, the NMR data recorded for the Mo2·arene linkage of
complexes 3·arene evidence that in all cases studied the most
favorable coordination of the molecule of the arene in solution
is that in which the best C-donor atoms are bonded to the Mo
atoms of the central Mo2 core. Inductive and resonance effects
offer a simple explanation to this observation. Under this as-
sumption, the coordination of the molecule of toluene through
four CH carbon atoms found in the solid-state structure of
3·C6H5Me may be due to crystal-packing effects.
was employed for the kinetic competition experiments, al-
though now Tsai’s complex [Mo2{HC(N-2,6-iPr2C6H3)2}2], 4, was
treated with the same large excess of the 1:1 arene solvent
mixture indicated above. Reactions were carried out at room
temperature in the dark for 24 h, since under these conditions
arene-exchange reactions are negligible. The results (see the
Supporting Information for details, Table S2) were strikingly
similar to those described above for the experiments per-
formed under conditions of thermodynamic control. Therefore,
in comparative terms the less bulky arenes not only furnish
thermodynamically more stable complexes 3·arene but they
also feature faster reaction rates toward the quintuple MoꢀMo
bond of 4. A good correlation exists between our experimental
observations and the van der Waals radii[32] of the arene sub-
stituents: 1.0 and 1.47 ꢃ for aromatic H and F, respectively;[32b]
1.50 ꢃ for oxygen;[32b] and 2.0 ꢃ for CH3.[32a]
Solutions of complexes 3·arene are light-sensitive with
regard to substitution of the coordinated aromatic hydrocar-
bon. Thus, under sunlight, solutions of 3·C6H6 in toluene, or of
3·C6H5Me in benzene, underwent complete arene exchange
after only ꢂ0.5 h. This contrasts with the inertness that these
complexes exhibit in the dark. Heating 3·C6H6 in C6D6 at 808C
for 24 h in the absence of light resulted in scarcely any substi-
tution; in reality, formation of the isotopologue 3·C6D6 re-
quired prolonged heating at 1208C (ca. 24 h) for completion.
To complement these studies we developed some competi-
tion experiments that clarified both the kinetic and the ther-
modynamic preferences of the aromatic molecules, toward the
MoI2 core. With reference to the latter, complex 3·C6H6 was dis-
solved in about 0.6 mL of an equimolar mixture of two arene
solvents (arene-1 and arene-2 in Scheme 3) and the resulting
solution heated at 1208C for 24 h in the dark. The outcome of
Computational studies
The computational studies developed in our earlier communi-
cation[12b] for the benzene complex 3·C6H6 have now been ex-
tended to other 3·arene adducts (arene=C6H5Me, p-C6H4Me2,
and p-C6H4F2) and their rotameric preferences. Geometry opti-
mization yields in all cases structures that are in excellent
agreement with the experimental ones in the MoꢀMo and
MoꢀC distances, in the CꢀC distance distribution pattern of
the aromatic ring, and in the N-Mo-Mo-N hinge angles, as well
as in the orientation of the substituted arene rings. The Moꢀ
Mo bond seems to be practically unaffected by the nature of
the coordinated arene, with all four complexes showing the
same distance (2.134 ꢃ) in the calculations within 0.001 ꢃ, simi-
lar to the experimental results (2.110ꢃ0.007 ꢃ). The four short
MoꢀC distances vary also in less than 0.02 ꢃ from one complex
Scheme 3. Arene competition experiments performed under thermodynamic control. Dipp=2,6-iPr2C6H3.
Chem. Eur. J. 2014, 20, 6092 – 6102 www.chemeurj.org
6097
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
to another. The interaction energies between the arene and
the Mo2L2 fragment are around 60 kcalmolꢀ1. The difluoro de-
the two B2 orbitals are occupied and the two A2 ones are
empty (the Supporting Information, Figure S13), in the Mo2ꢀ
C6H6 adduct the antibonding B2 combination is empty whereas
the bonding orbital with A2 symmetry is occupied. As a conse-
quence, the MoꢀMo bonding d(B2) orbital is partially depopu-
lated and its antibonding counterpart d*(A2) is partially occu-
pied, resulting in a decreased MoꢀMo bond order.[12b] Similarly,
the CꢀC bonding p(B2) orbital is in part depopulated and the
antibonding p*(A2) is partly oc-
rivative is somewhat more strongly bound (ꢀ62 kcalmolꢀ1
)
and the benzene, toluene and xylene adducts have similar in-
teraction energies (ꢀ57, ꢀ56, and ꢀ55 kcalmolꢀ1, respectively).
The orbitals of the dimolybdenum and arene fragments that
are responsible for the bonding between the quintuple bond
and the aromatic hydrocarbon are shown in Figure 4, in which
cupied, ensuing in a weakening
of the CꢀC bonds. From the va-
lence-bond point of view one
could construct several Lewis
structures with different occupa-
tions of these fragment orbitals
that imply two-electron excita-
tions of either fragment and
cross-donation of two electron
pairs between Mo2 and C6H6, or
two-electron transfer from one
fragment to the other (the Sup-
porting Information, Scheme S1).
Such a symmetry-imposed elec-
tronic rearrangement[12b] results
in a weakening of both the Moꢀ
Mo and benzene CꢀC bonds,
which is neatly reflected in the
experimental
structures.
A
lengthening of the MoꢀMo
bond is indeed appreciated
when comparing the structures
of the quintuply bonded Mo2L2
compound and its benzene
adduct, whereas a ring expan-
sion due to the weakening of
the CꢀC bonds can be appreciat-
ed by increases in the sum of
Figure 4. Occupied (lower three rows) and empty (upper three rows) orbitals that participate in bonding between
the Mo2 and benzene fragments, classified according to their symmetry representations in the C2v point group.
they have been classified in accordance with their symmetry in
the idealized C2v symmetry point group. These comprise the p
system of the arene (represented in Figure 4 by C6H6), the in-
phase and out-of-phase combinations of the acceptor sp2
hybrid orbitals of the Mo2 unit, and one of the two pairs of d
and d* orbitals (with symmetry representations B2 and A2, re-
spectively). The other d/d* orbital pair is not shown because
these orbitals have the same symmetry as the combinations of
the sp2 hybrids (see the Supporting Information, Figure S12)
and play a minor role in the Mo2-arene interaction. Disregard-
ing in a first approximation the high lying p orbitals of ben-
zene of the A1 and B2 symmetry representations (depicted in
gray in Figure 4), we are left with a pair of orbitals for each
symmetry representation to build up four Mo2-benzene bond-
ing orbitals and their four antibonding counterparts. The six
electrons of benzene and the two d electrons therefore
occupy fully the four bonding MOs.
the six experimental bond distances of 0.22 ꢃ or higher upon
coordination (0.24 ꢃ computationally).
As a consequence of the back donation from the d(B2) orbi-
tal to the p system, the coordinated arene becomes negatively
charged, according to a natural population analysis (NPA).
Whereas the differences between benzene, toluene, and
xylene are minimal (ꢀ0.21, ꢀ0.20, and ꢀ0.19, respectively), di-
fluorobenzene bears a significantly higher negative charge
(ꢀ0.30). This differential behavior is due to the interaction of
the p orbitals of the fluorine atoms with those of the phenyl
ring. In the uncoordinated difluorobenzene molecule there is
a three-orbital interaction of the in-phase combination of the
fluorine p lone pair orbital with one occupied and one empty
p orbitals of the phenyl ring, all belonging to the A1 symmetry
species (the Supporting Information, Figure S14). The result is
a localization of the occupied p orbital at the unsubstituted
carbon atoms of the phenyl ring, which may explain the pref-
erence of difluorobenzene for an orientation that is not the
most favorable one from a steric point of view (see the Sup-
porting Information for a more detailed discussion of steric ef-
Focusing the discussion on the orbitals with A2 and B2 sym-
metries that are responsible for the back-bonding interac-
tions,[12b] despite the fact that in the independent fragments
Chem. Eur. J. 2014, 20, 6092 – 6102
www.chemeurj.org
6098
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
fects). This orientation is, however, consistent with a high posi-
tive charge at the substituted carbon atoms (+0.32 and +0.45
in the uncoordinated and coordinated molecule, respectively),
which contrasts with practically neutral carbon atoms in tolu-
ene and xylene, and with a negative charge in benzene
(ꢀ0.21). The above observations are also in agreement with
the marked NMR shieldings observed for the CH units involved
in the coordination to the Mo2 core that was considered
before.
data (see above). For toluene the predominant rotamer in solu-
tion is different from the one found in the crystal structure,
but those two structures are coincident with the calculated
minima, which differ by only 2.5 kcalmolꢀ1 and are separated
by a barrier of 14 kcalmolꢀ1. The existence of only two minima
for the 3·arene complexes (Figure 5) can be understood by an-
alyzing the topology of the cavity in which the p-coordinated
arene sits surrounded by the aryl groups of the bidentate ami-
dinate ligands. This is shown in Figure S15 (left, in the Support-
ing Information); additional details for this structural problem
are provided in the Supporting Information.
To complete this investigation we carried out a computation-
al study of the rotational isomers and the rotational barriers
for the benzene, p-xylene, p-difluorobenzene, and toluene ad-
ducts, with the results summarized in Figure 5 (see the Sup-
Conclusion
We have demonstrated that the bis(hydride) complex
[Mo2(H)2{HC(N-2,6-iPr2C6H3)2}2(thf)2], 2, which features a MoꢀMo
quadruple bond, experiences photolytic reductive elimination
of dihydrogen in different reaction solvents to form bis(amidi-
nate) derivatives of the MoI2 core. The use of cyclohexane per-
mits isolation of quintuply bonded molecules of [Mo2{HC(N-
2,6-iPr2C6H3)2}2], 4, known since 2009 thanks to the pioneering
work of Tsai and co-workers,[8] whereas employing benzene or
other arene solvents, corresponding adducts of the MoI2 cen-
tral unit are obtained instead. These 3·arene complexes can
also be accessed by the direct reaction of 4 with the arene.
The experimental and computational studies reported
herein support the formulation of the arene complexes as MoI2
species resulting from a Lewis acid/base interaction between
the quintuple MoꢀMo bond of 4 and the aromatic hydrocar-
bon. This proposal finds reinforcement in the reversibility of
the reactions of 4 with the investigated arenes and is addition-
ally underpinned by the arene substitution chemistry discussed
above. In accordance with these arguments, throughout this
paper compounds 3·arene have been represented as MoI2 de-
rivatives, thereby exhibiting a “formal” MoꢀMo bond order of
five. However, the strong electronic interaction that exists be-
tween a pair of d/d* orbitals of the dimolybdenum unit and
the p system of the arene results in the partial depopulation
of the bonding d(B2) orbital and the partial occupation of the
antibonding d*(A2) orbital. Hence, despite the formal quintuple
bond representation, the calculated Meyer bond order is of
about 3[12b] (ca. 3.6 in Power’s Ar’CrCrAr’ complexes)[6], which
makes a quadruple MoꢀMo bond a more realistic description
of the electronic structure of the MoI2 central unit of 3·arene
complexes.
Figure 5. The experimentally found rotamers in the solid state and in solu-
tion (first column), and calculated free energies for different energy minima
and transition states found computationally. The dashed line indicates the
relative position of the Mo2 unit.
porting Information for additional information details). It can
be seen that in the first three cases the lowest energy mini-
mum has the same geometry as found experimentally both in
the solid state and in solution. With reference once more to
3·C6H6 as model complex, the exchange of the coordinated
and non-coordinated CꢀC bonds, which explains the already
discussed solution dynamic behavior of these complexes,
occurs through a m-h3:h3 transition state that lies about 9.9 kcal
Experimental Section
General considerations
All manipulations were carried out using standard Schlenk and glo-
vebox techniques, under an atmosphere of argon and of high
purity nitrogen, respectively. All solvents were dried and degassed
prior to use, and stored over 4 ꢃ molecular sieves. Cyclohexane
(C6H12), toluene (C7H8), n-pentane (C5H12), and anisole (C6H5OMe)
were distilled under nitrogen over sodium. Tetrahydrofuran (THF)
was distilled under nitrogen over sodium/benzophenone.
[D6]Benzene and [D8]THF were distilled under argon over sodium/
molꢀ1 above the ground state. This energy barrier is in good
¼
agreement with the DG value of ꢂ(7.5ꢃ2.6) kcalmolꢀ1
,
which can be estimated from variable temperature 1H NMR
Chem. Eur. J. 2014, 20, 6092 – 6102
www.chemeurj.org
6099
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
3·C6H6 and 3·C6H5Me (Method A)[12b]
benzophenone; [D8]toluene and [D12]cyclohexane were distilled
under argon over sodium. Fluorobenzene and 1,4-difluorobenzene
were dried over sodium during 4 and 12 h, respectively. The quad-
ruply bonded dimolybdenum complexes [Mo2(O2CR)4] (R=H,
Me)[33] as well as the formamidinate ligand H{HC(N-2,6-iPr2C6H3)2},[34]
and its lithium salt [Li{HC(N-2,6-iPr2C6H3)2}][THF]2,[35] were prepared
according to literature methods. [Mo2(O2CMe)4] was washed with
toluene at 1008C to remove any acidic residue. [Mo2(O2CR)2{HC(N-
2,6-iPr2C6H3)2}2] (R=Me, 1a and R=H, 1b), bis(hydride) complex
[Mo2(H)2{HC(N-2,6-iPr2C6H3)2}2(thf)2] (2) and the quintuply bonded
dimolybdenum complex [Mo2{HC(N-2,6-iPr2C6H3)2}2] (4)[8] were pre-
pared by the synthetic methods communicated recently by our re-
search group.[12b] All other compounds were commercially available
and were used as received. Photochemical reactions were carried
out using a medium pressure mercury lamp, model 3010 (Photo-
chemical Reactors Ltd). Solution NMR spectra were recorded on
Bruker AMX-300, DRX-400 and DRX-500 spectrometers. The reso-
nance of the solvent was used as the internal standard, chemical
shifts are reported relative to TMS and the NMR signals of fluori-
nated derivatives are reported relative to CFCl3. UV/Visible spectra
were recorded on a PerkinElmer Lambda 750 spectrometer. For el-
emental analyses a LECO TruSpec CHN elementary analyzer was
utilized and infrared experiments were carried out on a Bruker
Vector 22 and Tensor 27 spectrometer. For details on the computa-
tional study see ref. 12b.
The two compounds were obtained by UV irradiation (365 nm) of
stirred solutions of complex 2 (ca. 0.40–0.50 g) in benzene or tolu-
ene (10 mL) for 24 h. Green/yellow solutions resulted from these
transformations that were evaporated to dryness. Solids were
washed with pentane at 08C giving products sufficiently pure for
reactivity studies in yields of about 70%. Crystallization of the
crude solids from light-protected solutions in hexane with small
amounts of benzene and ether (in the case of 3·C6H6), or toluene
plus pentane (for 3·C6H5Me), provided microanalytically pure com-
pounds.
Compound 3·C6H6
1H NMR (500 MHz, C6D6, 258C): d=0.04, 0.76, 0.98, 1.05, 1.06, 1.12,
3
1.19, 1.51 (d, 6 H each, JHH =6.7 Hz; CHMe2), 2.54, 2.89, 3.39, 3.59
3
(sept, JHH =6.7 Hz, 2H each; 2CHMe2), 3.87 (s, 6H, Mo2-C6H6), 6.92
2
3
(dd, JHH =2.7 Hz, JHH =6.7 Hz, 2H; m-DippB), 6.95–7.00 (m, 6H; m-
DippB, p-DippB, m-DippA), 7.09 (t, 3JHH =7.6 Hz, 2H; p-DippA), 7.20
(dd, 2JHH =1.4 Hz, 3JHH =7.6 Hz, 2H; m-DippA), 7.56 ppm (s, 2H;
NC(H)N); 13C{1H} NMR (125 MHz, C6D6, 258C): d=23.3, 24.3, 24.4,
24.7, 25.3, 25.6 (CHMe2), 27.1 (CHMe2 and CHMe2), 27.6, 27.8
(CHMe2), 27.9 (CHMe2), 28.3 (CHMe2), 71.7 (Mo2-C6H6), 123.0, 123.4,
124.7, 124.9 (m-Dipp), 125.0, 125.1 (p-Dipp), 141.7, 143.4, 143.9,
144.3 (o-Dipp), 144.8, 147.4 (ipso-Dipp), 161.1 ppm (NC(H)N); Mag-
netic susceptibility: c=ꢀ15.6ꢄ10ꢀ4 cm3 molꢀ1. UV/Vis (C6H6); lmax
(e) 435 (3050), 595 nm (545 mꢀ1 cmꢀ1); elemental analysis calcd (%)
for C56H76Mo2N4: C 67.45; H 7.68; N 5.62; found: C 67.8; H 8.0; N
5.9.
Syntheses
[Mo2(H)2{HC(N-2,6-iPr2C6H3)2}2(thf)2] (2)
This synthesis involves compound 1a and LiMe and produces ini-
tially a trimethyl dimolybdate lithium derivative, by substitution of
the two acetate ligands of 1a. Loss of LiMe from the “ate” complex
occurs upon heating in a mixture of hydrocarbon solvents, yielding
a neutral dimethyl bis(amidinate) dimolybdenum species, which
upon reaction with H2 affords the title compound. Details of this
three-step, one-flask procedure are given below.
Compound 3·C6H5Me
1H NMR (500 MHz, C6D6, 258C): d=0.05, 0.85 (d, 3JHH =6.8 Hz, 6H
each; CHMe2), 0.99–1.02 (m, 12H; 2CHMe2), 1.06 (d, 3JHH =6.8 Hz,
6H; CHMe2), 1.17, 1.20 (d, JHH =6.8 Hz; 6H each, CHMe2), 1.50 (d,
3JHH =6.8 Hz; 6H, CHMe2), 1.60 (s, 3H; C6H5Me), 2.64 (sept, JHH
3
3
=
3
6.8 Hz, 2H; 2CHMe2), 3.10 (sept, JHH =6.8 Hz, 2H; 2CHMe2), 3.31–
3.37 (m, 4H; C3H, C4H and 2CHMe2), 3.40 (br. d, JHH =5.6 Hz, 1H;
3
A suspension of 1a (4.0 g, 3.86 mmol) in THF (80 mL) was cooled
at ꢀ108C and LiMe (8.3 mL, 1.6m solution in Et2O) was added
slowly with a syringe. The reaction mixture was stirred overnight,
allowing it to reach slowly to room temperature, to give a red sus-
pension that was centrifuged and taken to dryness. A mixture of
toluene/hexane (45/15 mL) was added and the solution was
heated at 1008C for 5 h, after which time it was cooled to 258C
and centrifuged. The solvent was removed under vacuum and the
residue dissolved in toluene (20 mL) and treated with dihydrogen
for one day (1 bar, 208C, with stirring). Removal of the solvent
under vacuum, washing of the crude product with cold pentane
(08C) and crystallization from THF at ꢀ208C for 2 days, provided
yellow crystals of the desired product that were dried under
3
C6H), 3.58 (sept, JHH =6.8 Hz, 2H; 2CHMe2), 4.88 (m, 1H; C2H), 5.14
(t, 3JHH =5.6 Hz, 1H; C5H), 6.91 (br. d, 3JHH =7.4 Hz, 2H; m-DippA),
3
6.96 (t, JHH =7.4 Hz, 2H; p-DippA), 7.00 (m, 2H; m-DippB), 7.06 (m,
2H; m-DippA), 7.11 (t, JHH =7.6 Hz, 1H, p-DippB), 7.21 (br. d, JHH
=
3
3
7.6 Hz, 2H; m-DippB), 7.57 ppm (s, 2H; NC(H)N); 13C{1H} NMR
(125 MHz, C6D6, 258C): d=17.3 (C6H5Me), 23.5, 24.3, 24.7, 24.8, 25.5,
25.7, 27.3 (CHMe2), 27.4, 27.6, 27.8 (CHMe2), 27.9 (CHMe2), 28.3
(CHMe2), 59.4 (C4), 62.3 (C3), 65.1 (C6), 70.1 (C1), 91.6 (C2), 92.5 (C5),
123.1, 123.6, 124.8, 124.9, 125.0, 125.1 (metha and para aromatics
DippA and DippB), 142.0, 143.1, 144.4, 144.7 (o-Dipp), 144.9, 147.6
(ipso-DippA, ipso-DippB), 161.3 ppm (NC(H)N); UV/Vis (C6H6); lmax (e)
430 (2160), 595 nm (680 mꢀ1 cmꢀ1); elemental analysis calcd (%) for
C57H78Mo2N4: C 67.71; H 7.78; N 5.54; found: C 68.0; H 8.1; N 5.7.
1
vacuum for 2 h (yield: 1.85 g, 45%). H NMR (500 MHz, C6D6, 258C):
3
d=1.02, 1.09 (d, JHH =6.8 Hz, 12H each; 2 CHMe2), 1.27 (m, 8H; O-
CH2CH2), 1.35, 1.48 (d, 3JHH =6.8 Hz, 12H each; 2CHMe2), 3.40 (m,
8H; O-CH2CH2), 3.80, 4.46 (sept, 3JHH =6.8 Hz, 4H each; 4CHMe2),
5.67 (s, 2H; Mo-H), 7.01–7.07 (m, 12H; aromatics), 8.45 ppm (s, 2H;
NC(H)N); 13C{1H} NMR (125 MHz, C6D6, 258C): d=24.9, 25.0 (CHMe2),
25.6 (O-CH2CH2), 25.7, 27.1 (CHMe2), 28.3, 28.5 (CHMe2), 69.2 (O-
CH2CH2), 123.7, 123.9 (m-Dipp), 126.0 (p-Dipp), 144.9, 143.9 (o-
Dipp), 145.9 (ipso-Dipp), 161.6 ppm (NC(H)N); UV/Vis (benzene so-
3·C6H4Me2 (Method A)
This complex was prepared following the procedure described
above. Dark-red crystals were obtained from a p-xylene solution of
this complex stored at 58C for 48 h and were dried under vacuum
1
(yield: 70%). H NMR (500 MHz, C6D12, 258C): d=ꢀ0.18, 0.80, 0.98,
3
1.02, 1.12, 1.14, 1.27 (d, JHH =6.7 Hz, 6H each; CHMe2), 1.31 (s, 6H;
3
lution 10ꢀ4 m):
l
(e)=348 (5300), 425 (2000), 548 nm (1200
p-C6H4Me2), 1.46 (d, 3JHH =6.7 Hz, 6H; CHMe2), 2.60 (dd, JHH
=
M-1 cmꢀ1); magnetic susceptibility: c=ꢀ12.7ꢄ10ꢀ4 cm3 molꢀ1; ele-
mental analysis calcd (%) for C58H88Mo2N4O2: C 65.40; H 8.33; N
5.26; found: C 65.1; H 7.9; N 5.0.
5.5 Hz, JHH =1.7 Hz, 2H; C3,6 H, AA’BB’ system), 2.65, 3.02, 3.42, 3.55
4
3
3
4
(sept, JHH =6.7 Hz, 2H each; 2CHMe2), 5.49 (dd, JHH =5.5 Hz, JHH
=
1.7 Hz, 2H; C2,5 H, AA’BB’ system), 6.77 (brd, 3JHH =6.8 Hz, 2H; m-
Chem. Eur. J. 2014, 20, 6092 – 6102
www.chemeurj.org
6100
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
3
3
DippA), 6.86 (t, JHH =7.6 Hz, 2H; p-DippA), 6.97 (d, JHH =7.6 Hz, 2H;
analysis calcd (%) for C56H75FMo2N4: C 66.26; H 7.45; N 5.52; found:
C 65.9; H 7.9; N 5.7.
m-DippA), 7.01 (br. d, 3JHH =7.6 Hz, 2H; m-DippB), 7.08 (t, JHH
=
3
7.6 Hz, 2H; p-DippB), 7.21 (brd, 3JHH =7.6 Hz, 2H; m-DippB),
7.38 ppm (s, 2H; NC(H)N); 13C{1H} NMR (125 MHz, C6D12, 258C): d=
15.8 (p-C6H4Me2), 24.4, 24.7, 25.4, 26.0, 26.1, 26.7 (CHMe2), 27.8,
27.9, 28.5, 28.6 (CHMe2), 28.9, 29.0 (CHMe2), 52.9 (C3,6 H), 71.4 (Cq1,4),
103.5 (C2,5 H), 123.5 (m-DippA), 124.4 (m-DippB), 125.2 (m-DippA, p-
DippA, m-DippB), 125.7 (p-DippB), 143.0, 143.1, 144.1 (o-Dipp), 145.2
(ipso-Dipp), 145.4 (o-Dipp), 148.4 (ipso-Dipp), 161.6 ppm (NC(H)N);
UV/Vis (C6H6); lmax (e) 435 (3280), 595 nm (500 mꢀ1 cmꢀ1); elemental
analysis calcd (%) for C58H80Mo2N4: C 67.95; H 7.87; N 5.47; found:
C 68.3; H 8.4; N 5.9.
Compound 3·C6H4F2
1H NMR (400 MHz, C6D6, 258C): d=0.07, 0.85, 0.99, 1.12 (d, JHH
=
3
6.5 Hz, 6H each; CHMe2), 1.21 (m, 12H; 2CHMe2), 1.27, 1.49 (d,
3JHH =6.5 Hz, 6H each; CHMe2), 2.98 (m, 2H; C2,5H or C3,6H), 3.21 (m,
6H; CHMe2, C2,5H or C3,6H), 3.31, 3.51 (sept, JHH =6.5 Hz, 2H each;
3
2CHMe2), 6.95–7.25 (m, 12H; aromatics DippA and DippB), 7.67 (s,
2H; NC(H)N); 13C{1H} NMR (125 MHz, C6D6, 258C): d=23.9, 24.5,
25.5, 25.9, 26.3, 26.4 (CHMe2), 27.5 (CHMe2), 27.7 (CHMe2), 28.2, 28.3
(CHMe2), 28.4 (CHMe2), 28.8 (CHMe2), 37.9, 46.2 (dd, 3JCF =20 Hz,
4JCF =13 Hz, C2,5H or C3,6H), 123.2, 124.1 (m-DippA each), 125.5 (p-
DippA), 125.7 (p-DippB), 125.8, 125.9 (m-DippB each), 142.9, 143.8,
144.0, 144.5 (o-Dipp), 145.1, 147.7 (ipso-Dipp), 159.5 (d, JCF
270 Hz; C1,4-F), 162.2 (NC(H)N); 19F NMR (125 MHz, C6D6, 258C): d=
ꢀ114.3 ppm. UV/Vis (C6H6); lmax (e) 445 (1900), 560 nm
(620 mꢀ1 cmꢀ1); elemental analysis calcd (%) for C56H74F2Mo2N4: C
65.10; H 7.22; N 5.42; found: C 65.17; H 7.16; N 5.02.
3·C6H5OMe (Method B)
1
=
This compound was obtained by UV irradiation (365 nm) of stirred
solutions of complex 3·C6H6 (ca. 0.35 g) in anisole (5 mL) for 12 h.
A dark-red solution was evaporated to dryness. The solid was
washed with pentane at 08C giving a product sufficiently pure for
1
reactivity studies in yields of about 60%. H NMR (500 MHz, C6D6,
3
258C): d=ꢀ0.17 (brs, 6H; CHMe2), 0.65, 1.01, 1.05 (d, JHH =6.7 Hz,
3
6H each; CHMe2), 1.18 (m, 12H; 2CHMe2), 1.21, 143 (d, JHH
=
Acknowledgements
6.7 Hz, 6H each; CHMe2), 2.91 (d, 3JHH =5.4 Hz, 1H; C2H or C6H),
3.08–3.15 (m, 4H; C2H or C6H, C3H or C5H, CHMe2), 3.22 (m, 4H;
3
C6H5OMe, C3H or C5H), 3.37 (m, 4H; 4CHMe2), 3.51 (sept, JHH
=
Financial support (FEDER contribution and Subprogramas Juan
de la Cierva) from the Spanish Ministry of Science and Innova-
tion (Projects CTQ2010–15833, CTQ2011–23862-C02-01 and
Consolider-Ingenio 2010 CSD2007-00006), the Generalitat de
Catalunya (Project 2009SGR-1459), and the Junta de Andalucꢁa
(Grant FQM-119 and Project P09-FQM-5117) is gratefully ac-
knowledged. M. C. and N. C. thank the Spanish Ministry of Edu-
cation (AP-4193) and the Spanish Ministry of Science and Inno-
vation (BES-2011-047643) for research grants. We also thank Dr.
J. Lꢅpez-Serrano for assistance and discussions on NMR spec-
troscopy, and the Centre de Supercomputaciꢅ de Catalunya
(CESCA) for the allocation of computational resources.
6.7 Hz, 2H; 2CHMe2), 6.05 (t, JHH =5.4 Hz, 1H; C4H), 6.94–7.22 (m,
12H; aromatics DippA and DippB), 7.57 (s, 2H; NC(H)N);
13C{1H} NMR (125 MHz, C6D6, 258C): d=23.5–28.4 (12C; CHMe2 and
CHMe2), 42.8, 46.4 (C2 or C6), 52.1 (C3 or C5), 54.8 (C6H5OMe), 56.8
(C3 or C5), 100.0 (C4), 123.1–125.4 (6 C; meta and para aromatics
DippA and DippB), 141.7–147.7 (6C; ortho and ipso aromatics DippA
and DippB), 161.5 ppm (NC(H)N); UV/Vis (C6H6); lmax (e) 430 (2745),
560 nm (870 mꢀ1 cmꢀ1); elemental analysis calcd (%) for
C57H78Mo2N4O: C, 66.25; H 7.65; N 5.45; found: C 66.3; H 7.9; N 5.4.
3
3·C6H5F and 3·C6H4F2 (Method C)
The compounds 3·C6H5F and 3·C6H4F2 were obtained by stirring
a solution of complex 4 (ca. 0.20–0.25 g) in fluorobenzene or 1,4-
difluorobenzene (5–8 mL) of at room temperature for 24 h in the
dark. The red filtered solution was concentrated and stored at 58C
for 48 h. For 3·C6H4F2, dark-red crystals were separated from the
solution and dried under vacuum (yield: 55%). In the case of
3·C6H4F, the solid was separated from the red solution and dried in
vacuo (yield: 50%).
Keywords: arenes
molybdenum · multiple bonds · NMR spectroscopy
·
density functional calculations
·
[1] D. Lawton, R. Mason, J. Am. Chem. Soc. 1965, 87, 921–922.
[2] F. A. Cotton, C. A. Murillo, R. A. Walton, Multiple Bonds Between Metal
Atoms, 3rd ed., Springer, New York, 2005.
[3] a) M. H. Chisholm, Proc. Natl. Acad. Sci. USA 2007, 104, 2563–2570;
b) M. H. Chisholm, B. J. Lear, Chem. Soc. Rev. 2011, 40, 5254–5265.
[4] a) M. H. Chisholm, A. M. Macintosh, Chem. Rev. 2005, 105, 2949–2976;
b) M. H. Chisholm, Coord. Chem. Rev. 2013, 257, 1576–1583; c) M. H.
Chisholm, Phil. Trans. R. Soc. A. 2008, 366, 101–112; d) J. Hicks, S. P.
Ring, N. J. Patmore, Dalton Trans. 2012, 41, 6641–6650; e) L. A. Wilkin-
son, L. McNeill, A. J. H. M. Meijer, N. J. Patmore, J. Am. Chem. Soc. 2013,
135, 1723–1726; f) L. A. Wilkinson, L. McNeill, P. A. Scattergood, N. J.
Patmore, Inorg. Chem. 2013, 52, 9683–9691.
[5] Y.-C. Tsai, Y.-M. Lin, J.-S. K. Yu, J.-K. Hwang, J. Am. Chem. Soc. 2006, 128,
13980–13981.
[6] T. Nguyen, A. D. Sutton, M. Brynda, J. C. Fettinger, G. J. Long, P. P. Power,
Science 2005, 310, 844–847.
[7] a) R. Wolf, C. Ni, T. Nguyen, M. Brynda, G. J. Long, A. D. Sutton, R. C.
Fischer, J. C. Fettinger, M. Hellman, L. Pu, P. P. Power, Inorg. Chem. 2007,
46, 11277–11290; b) C. Ni, P. P. Power, Struct. Bonding (Berlin) 2010, 136,
59–112; c) K. A. Kreisel, G. P. A. Yap, O. Dmitrenko, C. R. Landis, K. H.
Theopold, J. Am. Chem. Soc. 2007, 129, 14162–14163; d) A. Noor, F. R.
Wagner, R. Kempe, Angew. Chem. 2008, 120, 7356–7359; Angew. Chem.
Int. Ed. 2008, 47, 7246–7249; e) Y.-C. Tsai, C.-W. Hsu, J.-S. K. Yu, G.-H.
Lee, Y. Wang, T.-S. Kuo, Angew. Chem. 2008, 120, 7360–7363; Angew.
Compound 3·C6H5F
1H NMR (500 MHz, C6D6, 258C): d=0.04, 0.80, 0.99, 1.03, 1.10 (d,
3JHH =6.8 Hz, 6H each; CHMe2), 1.17 (m, 12H; 2CHMe2), 1.48 (d,
3JHH =6.8 Hz, 6H; CHMe2), 2.85 (m, 3H; 2CHMe2, C3H or C5H), 3.02
(m, 2H; C2H or C6H, C3H or C5H), 3.11 (m, 3H; CHMe2, C2H or C6H),
3
3.26, 3.52 (sept, JHH =6.8 Hz, 2H each; 2CHMe2), 5.77 (m, 1H; C4H),
6.93–7.21 (12H; aromatics DippA and DippB), 7.58 ppm (s, 2H;
NC(H)N); 13C{1H} NMR (125 MHz, C6D6, 258C): d=23.8, 24.5, 25.2,
25.9, 26.0, 27.5 (CHMe2), 27.7 (CHMe2), 27.9 (br. s, CHMe2), 28.2 (br.
2
s, CHMe2), 28.7 (CHMe2), 39.2 (d, JCF =21.1 Hz; C2H or C6H), 48.2 (d,
3
2JCF =19.5 Hz; C2H or C6H), 48.4 (d, JCF =7.3 Hz; C3H or C5H), 58.1
3
(d, JCF =10.3 Hz, C3H or C5H), 102.7 (C4H), 123.2, 123.9, 125.3, 125.4
(m-Dipp), 125.5, 125.6 (p-Dipp), 142.5, 143.7, 144.1, 144.6 (o-Dipp),
1
145.0, 147.7 (ipso-Dipp), 161.8 (NC(H)N), 162.3 ppm (d, JCF
=
274 Hz; C1-F); 19F NMR (125 MHz, C6D6, 258C): d=ꢀ105.4 ppm; UV/
Vis (C6H6); lmax (e) 430 (1900), 580 nm (410 Mꢀ1 cmꢀ1); elemental
Chem. Eur. J. 2014, 20, 6092 – 6102
www.chemeurj.org
6101
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Chem. Int. Ed. 2008, 47, 7250–7253; f) C.-W. Hsu, J.-S. K. Yu, C.-H. Yen,
G.-H. Lee, Y. Wang, Y.-C. Tsai, Angew. Chem. 2008, 120, 10081–10084;
Angew. Chem. Int. Ed. 2008, 47, 9933–9936; g) A. Noor, G. Glatz, R.
Mꢆller, M. Kaupp, S. Demeshko, R. Kempe, Z. Anorg. Allg. Chem. 2009,
635, 1149–1152; h) F. R. Wagner, A. Noor, R. Kempe, Nat. Chem. 2009, 1,
529–536; i) A. Noor, G. Glatz, R. Mꢆller, M. Kaupp, S. Demeshko, R.
Kempe, Nat. Chem. 2009, 1, 322–325.
[19] a) T. E. Burrow, A. Hills, D. L. Hughes, J. D. Lane, R. H. Morris, R. L. Ri-
chards, J. Chem. Soc. Dalton Trans. 1991, 1813–1820; b) T. E. Burrow,
A. J. Lough, R. H. Morris, A. Hills, D. L. Hughes, J. D. Lane, R. L. Richards,
J. Chem. Soc. Dalton Trans. 1991, 2519–2526; c) T. E. Burrow, A. Hills,
D. L. Hughes, J. D. Lane, N. J. Lazarowych, M. J. Maguire, R. H. Morris,
R. L. Richards, J. Chem. Soc. Chem. Commun. 1990, 1757–1759.
[20] a) D. F. Schafer II, P. T. Wolczanski, J. Am. Chem. Soc. 1998, 120, 4881–
4882; b) R. R. Schrock, K.-Y. Shih, D. A. Dobbs, W. M. Davis, J. Am. Chem.
Soc. 1995, 117, 6609–6610; c) H. W. Turner, R. R. Schrock, J. D. Fellmann,
S. J. Holmes, J. Am. Chem. Soc. 1983, 105, 4942–4950.
[8] Y.-C. Tsai, H.-Z. Chen, C.-C. Chang, J.-S. K. Yu, G.-H. Lee, Y. Wang, T.-S.
Kuo, J. Am. Chem. Soc. 2009, 131, 12534–12535.
[9] a) Y.-L. Huang, D.-Y. Lu, H.-C. Yu, J.-S. K. Yu, C.-W. Hsu, T.-S. Kuo, G.-H.
Lee, Y. Wang, Y.-C. Tsai, Angew. Chem. 2012, 124, 7901–7905; Angew.
Chem. Int. Ed. 2012, 51, 7781–7785; b) S.-C. Liu, W.-L. Ke, J.-S. K. Yu, T.-S.
Kuo, Y.-C. Tsai, Angew. Chem. 2012, 124, 6500–6503; Angew. Chem. Int.
Ed. 2012, 51, 6394–6397.
[21] F. A. Cotton, L. M. Daniels, E. A. Hillard, C. A. Murillo, Inorg. Chem. 2002,
41, 2466–2470.
[22] C. Ni, B. D. Ellis, J. C. Fettinger, G. J. Long, P. P. Power, Chem. Commun.
2008, 1014–1016.
[23] a) E. L. Muetterties, J. R. Bleeke, E. J. Wucherer, T. A. Albright, Chem. Rev.
1982, 82, 499–525; b) H. Wadepohl, Angew. Chem. 1992, 104, 253–268;
Angew. Chem. Int. Ed. Engl. 1992, 31, 247–262; c) E. J. Maslowsky Jr, J.
Chem. Educ. 1993, 70, 980–984.
[10] a) A. Noor, R. Kempe, Chem. Rec. 2010, 10, 413–416; b) A. Noor, T.
Bauer, T. K. Todorova, B. Weber, L. Gagliardi, R. Kempe, Chem. Eur. J.
2013, 19, 9825–9832; c) A. K. Nair, N. V. S. Harisomayajula, Y. Tsai, Dalton
Trans. 2014, 43, 5618.
[24] For an informative and enjoyable account of the history of the sand-
wich complexes see: H. Werner, Landmarks in Organotransition Metal
Chemistry, Springer, New York, 2009, Chapter 5.
[25] G. La Macchia, L. Gagliardi, P. P. Power, M. Brynda, J. Am. Chem. Soc.
2008, 130, 5104–5114.
[26] G. Merino, K. J. Donald, J. S. D’Acchioli, R. Hoffmann, J. Am. Chem. Soc.
2007, 129, 15295–15302.
[27] E. Carmona, J. M. Marin, M. L. Poveda, J. L. Atwood, R. D. Rogers, J. Am.
Chem. Soc. 1983, 105, 3014–3022.
[28] J. P. Collman, L. S. Hegedus, Principles and Applications of Organotransi-
tion Metal Chemistry, Oxford University Press, 1987, Chapter 3, p. 150
and references therein.
[29] S. M. Hubig, S. V. Lindeman, J. K. Kochi, Coord. Chem. Rev. 2000, 200–
202, 831–873.
[30] a) A. Sekiguchi, K. Ebata, C. Kabuto, H. Sakurai, J. Am. Chem. Soc. 1991,
113, 1464–1465; b) M. C. Cassani, D. J. Duncalf, M. F. Lappert, J. Am.
Chem. Soc. 1998, 120, 12958–12959; c) A. Sattler, G. Parkin, J. Am.
Chem. Soc. 2012, 134, 2355–2366.
[31] C. Elschenbroich, Organometallics, 3rd ed., Wiley-VCH, Weinheim, 2006.
[32] a) L. Pauling, The Nature of the Chemical bond, 3rd. ed., Cornell Universi-
ty Press: Ithaca, 1960, Chapter 7–12; b) A. Bondi, J. Phys. Chem. 1964,
68, 441–451.
[11] a) C. Ni, B. D. Ellis, G. J. Long, P. P. Power, Chem. Commun. 2009, 2332–
2334; b) J. Shen, G. P. A. Yap, J.-P. Werner, K. H. Theopold, Chem.
Commun. 2011, 47, 12191–12193; c) A. Noor, E. S. Tamne, S. Qayyum, T.
Bauer, R. Kempe, Chem. Eur. J. 2011, 17, 6900–6903; d) C. Schwarzmaier,
A. Noor, G. Glatz, M. Zabel, A. Y. Timoshkin, B. M. Cossairt, C. C. Cum-
mins, R. Kempe, M. Scheer, Angew. Chem. 2011, 123, 7421–7424;
Angew. Chem. Int. Ed. 2011, 50, 7283–7286; e) P.-F. Wu, S.-C. Liu, Y.-J.
Shieh, T.-S. Kuo, G.-H. Lee, Y. Wang, Y.-C. Tsai, Chem. Commun. 2013, 49,
4391–4393; f) H.-G. Chen, H.-W. Hsueh, T.-S. Kuo, Y.-C. Tsai, Angew.
Chem. 2013, 125, 10446–10450; Angew. Chem. Int. Ed. 2013, 52, 10256–
10260; g) E. S. Tamne, A. Noor, S. Qayyum, T. Bauer, R. Kempe, Inorg.
Chem. 2013, 52, 329–336.
[12] a) M. Carrasco, M. Faust, R. Peloso, A. Rodrꢁguez, J. Lꢅpez-Serrano, E. ꢀl-
varez, C. Maya, P. P. Power, E. Carmona, Chem. Commun. 2012, 48,
3954–3956; b) M. Carrasco, N. Curado, C. Maya, R. Peloso, A. Rodrꢁguez,
E. Ruiz, S. ꢀlvarez, E. Carmona, Angew. Chem. 2013, 125, 3309–3313;
Angew. Chem. Int. Ed. 2013, 52, 3227–3231.
[13] Y. Kajita, T. Ogawa, J. Matsumoto, H. Masuda, Inorg. Chem. 2009, 48,
9069–9071.
[14] T. Ogawa, Y. Kajita, H. Masuda, Jpn. J. Appl. Phys. 2011, 50, 01AK05-1-
01AK05-5.
[15] K. W. Chiu, R. A. Jones, G. Wilkinson, A. M. R. Galas, M. B. Hursthouse, J.
Chem. Soc. Dalton Trans. 1981, 1892–1897.
[16] a) F. A. Cotton, P. C. W. Leung, W. J. Roth, A. J. Schultz, J. M. Williams, J.
Am. Chem. Soc. 1984, 106, 117–120; b) F. A. Cotton, J. P. Donahue, P.
Huang, C. A. Murillo, D. Villagrꢇn, Z. Anorg. Allg. Chem. 2005, 631, 2606–
2612.
[33] a) F. A. Cotton, J. G. Norman Jr., B. R. Stults, T. R. Webb, J. Coord. Chem.
1976, 5, 217–223; b) F. A. Cotton, Inorg. Synth. 1972, 13, 87–88.
[34] E. L. Kolychev, I. A. Portnyagin, V. V. Shuntikov, V. N. Khrustalev, M. S. Ne-
chaev, J. Organomet. Chem. 2009, 694, 2454–2462.
[35] M. L. Cole, A. J. Davies, C. Jones, P. C. Junk, J. Organomet. Chem. 2004,
689, 3093–3107.
[17] M. E. Garcꢁa, A. Ramos, M. A. Ruiz, M. Lanfranchi, L. Marchiꢈ, Organome-
tallics 2007, 26, 6197–6212.
[18] P. S. Pregosin, NMR in Organometallic Chemistry, Wiley-VCH, Weinheim,
2012.
Received: January 20, 2014
Published online on April 11, 2014
Chem. Eur. J. 2014, 20, 6092 – 6102
www.chemeurj.org
6102
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim