Facile access to homo- and heteroleptic, triply bonded dimolybdenum hexaalkoxides with unsaturated alkoxide ligands

Article abstract of DOI:10.1002/ejic.201001236

A series of monodentate, triply bonded Mo₂(OR)₆ complexes [R = MBE (1) (MBE = 2-methylbut-3-ene-2-yl), MMP (2) (MMP = 1-methoxy-2-methylpropane-2-yl), Terp (3) [Terp = 2-(4-methylcyclohex-3-enyl) propane-2-yl], which exhibit C-C doub

Full text of DOI:10.1002/ejic.201001236

FULL PAPER
DOI: 10.1002/ejic.201001236
Facile Access to Homo- and Heteroleptic, Triply Bonded Dimolybdenum
Hexaalkoxides with Unsaturated Alkoxide Ligands
Sebastian Krackl,[a] Jian-Gong Ma,[a] Yilmaz Aksu,[a] and Matthias Driess*[a]
Keywords: Multiple bonds / O ligands / Heteroleptic alkoxides / Anisotropy / Molybdenum
A series of monodentate, triply bonded Mo2(OR)6 complexes
[R = MBE (1) (MBE = 2-methylbut-3-ene-2-yl), MMP (2)
(MMP = 1-methoxy-2-methylpropane-2-yl), Terp (3) [Terp =
2-(4-methylcyclohex-3-enyl)propane-2-yl], which exhibit C–
C double bonds or an ether function in the ligand sphere,
leads to a different magnetic environment for the alkoxide
ligands due to the magnetic anisotropy of the Mo–Mo triple
bond. 95Mo NMR studies of the complexes 1–8 show that the
resonance strongly depends on the substitution pattern of the
alkoxide and that a shift to higher field is observed when
going from the tertiary to the primary alkoxides. The molecu-
lar structures for 4–8 were determined by single-crystal X-
ray diffraction, and all of the complexes show a staggered
conformation as well as an asymmetric ligand distribution,
which results in unequal Mo–O bond lengths. For the hetero-
leptic complexes 4–7, the RO ligands (R = tBu, MBE, MMP
and Terp, respectively) exhibit the longest bond lengths,
which suggests that the position of the ligand strongly de-
pends on the steric congestion of the α-carbon atom of the
alkoxide ligand. In 6, the methoxy function enables an intra-
molecular OǞMo coordination as is indicated by a Mo–O
distance of 2.2464(7) Å. This fact is supported by a lengthen-
ing of the Mo–Mo triple bond.
were synthesized and characterized by multinuclear (1H, 13
C
and 95Mo) NMR studies. The partial alcoholysis of the latter
complexes with neopentyl alcohol (neopentOH) led to the
heteroleptic alkoxides Mo2(OR)n(Oneopent)6–n (4–7) {n = 2
[for R = tBu (4), MBE (5), MMP (6)], 4 [for R = Terp (7)]}. This
concept was further applied to the synthesis of Mo2(O2DMH)2-
(OtBu)2 (8) (DMH = 2,5 dimethylhexyl) by starting from the
Mo2(OtBu)6 precursor. The 1H NMR spectra for the hetero-
leptic complexes 4–8 show signals that are significantly
shifted to a higher field for the RO ligand protons compared
to those of their homoleptic analogues. This is the result of a
change in the spatial position of the alkoxide ligands (RO) in
the homoleptic compared to the heteroleptic complexes that
molecule,[15] that are not yet fully understood. Their struc-
tural characterization can be difficult[15,16] due to the disor-
der of the Mo2 entity.[17] In addition, the heteroleptic, triply
bonded dimolybdenum complexes of the structure
Mo2A2B4 are known[18] and prefer the conformation b with
both ligands equally distributed at the Mo atoms (Fig-
ure 1). A high kinetic barrier between the conformers a and
b prevents transmutational processes.
Introduction
The first complex with a metal–metal multiple bond,
K2[Re2Cl8]·H2O, was described by Cotton et al. in 1965.[1]
This discovery stimulated an increased interest in synthesiz-
ing other multiply bonded metal–metal complexes and has
led to numerous well-defined systems with various metals.[2]
Among these systems, the dinuclear, triply bonded MoIII
compounds of the type Mo2R6 (R = organyl) represent a
fascinating group.[3] The synthesis of the first example,
Mo2(CH2SiMe3)6, as described by Wilkinson et al.,[4] was
extended by the seminal studies of Chisholm et al. on vari-
ous systems with alkyl, dialkylamino[5] and alkoxido ligands
at the molybdenum atoms.[6] The latter type of complexes,
i.e. dimolydenum(III) hexaalkoxides, show an intriguingly
versatile reactivity that included the addition of Lewis
bases,[7] oxidative addition reactions,[8,9] the addition of al-
kynes,[10,11] CO2 insertion[12] and the reversible addition of
CO.[13,14] Furthermore, these complexes display interesting
structural features, i.e. an asymmetric ligand distribution
that resulted in different Mo–O bond lengths in the same
Figure 1. Possible configurations for the heteroleptic Mo2A2B4
complexes. A and B can be various ligands, e.g., amido, halogenido,
alkyl, alkoxido, etc.
Various Mo2A2B4 complexes with different ligands A
and B (e.g., amido, halogenido and alkyl) have been de-
scribed and have even been employed for the synthesis of
unique heterodimetallic clusters.[19] Unexpectedly, hetero-
leptic complexes that contain only alkoxido ligands and
homoleptic complexes with unsaturated alkoxido ligands
are currently unknown. This prompted us to introduce dif-
[a] Institute of Chemistry, Metalorganics and Inorganic Materials,
Technische Universität Berlin,
Strasse des 17. Juni 135, Sekr. C2, 10623 Berlin, Germany
E-mail: Matthias.Driess@tu-berlin.de
Eur. J. Inorg. Chem. 2011, 1725–1732
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1725

S. Krackl, J.-G. Ma, Y. Aksu, M. Driess
FULL PAPER
ferent alkoxide groups at the triply bonded Mo–Mo system. pentOH = neopentyl alcohol) and bidentate DMHOH
These alkoxide groups bear additional functionalities that (DMH = 2,5 dimethylhexyl) towards Mo2(OtBu)6 in a hex-
could be useful for subsequent modifications such as immo- ane solution in a molar ratio of 4:1 and 2:1, respectively. In
bilization on supports and self-assembly.
both cases we obtained the desired heteroleptic derivatives
Mo2(OtBu)2(Oneopent)4 (4) and Mo2(OtBu)2(ODMH)2 (8)
(Scheme 2) in 65% and 89% yield, respectively. When a
smaller molar ratio of the alcohol was used, 4 and 8 were
formed, and Mo2(OtBu)6 remained; but when an excess of
the alcohol was used, the respective homoleptic analogues
were formed. Furthermore, we applied the same protocol
for the synthesis of heteroleptic and additionally function-
alized alkoxide compounds, starting from the precursors 1–
3. The conversion of 1 and 3 with neopentOH in a molar
ratio of 1:4 afforded the desired complexes Mo2(OMBE)2-
(Oneopent)4 (5) and Mo2(OMMP)2(Oneopent)4 (6) in 89%
and 76% yield, respectively. However, the partial
alcoholysis of complex 2 in a molar ratio of 1:4 did not
result in the replacement of the four TerpO ligands, but
led to complex Mo2(OTerp)4(Oneopent)2 (7) in 85% yield
Results and Discussion
Syntheses
Homoleptic Mo2(OR)6 (1–3) (R = MMP, Terp and
MBE)
We chose to introduce the functional alcohols MBEOH
(MBE = 2-methylbut-3-ene-2-yl), MMPOH (MMP = 1-
methoxy-2-methylpropane-2-yl) and TerpOH [Terp = 2-(4-
methylcyclohex-3-enyl)propane-2-yl], which possess either a
C–C double bond or a methoxy group, to the dimolybde-
num triple bond by direct alcoholysis of the precursor
Mo2(OtBu)6.[15] Tertiary alcohols were employed exclu-
sively due to their steric congestion in order to prevent
intermolecular oligomerization to [Mo(OR)3]x. Complexes
Mo2(OMMP)6 (1) and Mo2(OTerp)6 (2) were obtained as
orange powders after the addition of an excess of the corre-
1
(Scheme 2), as proven by H NMR spectroscopy. Interest-
ingly, when an excess of neopentOH was used in the reac-
tions with 1–3, complete alkoxido exchange did not occur
to give the homoleptic analogue Mo2(Oneopent)6. Instead,
a mixture of products with varying ligand distributions was
observed. The new complexes 4–8 were isolated as yellow,
crystalline solids by means of recrystallization or sublima-
tion (100–120 °C, 10–3 mbar).
sponding alcohol at room temperature to a pentane solu-
[20]
tion of Mo2(OtBu)6
(Scheme 1) in 74% and 67% yield,
respectively. Unfortunately, the addition of MBEOH under
the same conditions afforded only a mixture of products.
However, the reaction of Mo2Cl6(DME)2 with in situ gener-
ated LiOMBE in DME yielded the desired complex
Mo2(OMBE)6 (3) in 79% yield.[20]
Scheme 2. Partial alcoholysis of Mo2(OtBu)6 and complexes 1–3
with neopentOH to afford the heteroleptic complexes 4–7, as well
as the partial alcoholysis of Mo2(OtBu)6 with DMHOH to afford
the heteroleptic complex 8.
Scheme 1. Synthesis of functionalized Mo2(OR)6 complexes 1–3.
To the best of our knowledge, no transition metal com-
plex bearing the (–)-p-terpineolate ligand has been reported
to date. The metathesis and alcoholysis reactions led to al-
most pure products (1H NMR) that were purified by means
of recrystallization. Complexes 1 and 3 were also isolated
by sublimation (100–120 °C, 10–3 mbar), whereas complex
2 decomposed under these conditions without subliming.
NMR Studies
1
Complexes 1–8 were characterized by H, 13C and 95Mo
1
NMR spectroscopy. In the H NMR spectra the peaks as-
signed to the alkoxido ligand protons for the homoleptic
complexes 1–3 are significantly shifted to a lower field com-
pared to the signals of the corresponding alcohols and lith-
ium alcoholates, respectively, which confirms the successful
Heteroleptic Mo2(OR)n(Oneopent)6–n (4–7) [R = tBu,
MBE, MMP (n = 2), Terp (n = 4)] and Mo2(ODMH)2-
(OtBu)2 (8)
In addition, we probed the suitability of Mo2(OtBu)6 as a coordination of the ligands to the Mo2 entity (Table 1).[2]
1
precursor to the heteroleptic, triply bonded dimolybdenum The signal shift in the H NMR spectra for compound 1
alkoxides through partial alkoxido ligand exchange. Thus, and its corresponding lithium alcoholate is exemplified in
we probed the reactivity of monodentate neopentOH (neo- Figure 2.
1726
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1725–1732

Homo- and Heteroleptic, Triply Bonded Dimolybdenum Hexaalkoxides
1
Table 1. Selected H and 13C NMR data for 1–8; δ values in ppm;
For the heteroleptic complexes 4–8, we observed the res-
onances that corresponded to both of the alkoxido ligands
in the expected integral ratios (Table 1).
R represents the respective alcohol.
1H NMR[a]
13C NMR
However, when comparing the signals of the homoleptic
complexes 1–3 with the signals of the corresponding hetero-
leptic analogues 4–7, we observed an interesting phenome-
1[a] 3.44 (R2CH2OCH3)
3.18 (R2CH2OCH3)
1.60 (RCH3)
81.7 (R3CO)
78.9 (R2CH2OCH3)
58.2 (R2CH2OCH3)
29.0 (RCH3)
133.6 (R2C=CHR)
121.8 (R2C=CHR)
83.6 (R3C–O)
1
non. In the H NMR spectra for the complexes 4–7, the
2[a] 5.53 (R2C=CHR)
1.70 (RCH3)
resonances that correspond to the functional ligands
(MMPO) are shifted to higher field, and the resonances
that correspond to the neopent ligand are shifted to lower
field; this is exemplified in Figure 3 by compound 6. This
effect can be explained by taking into account the diamag-
netic anisotropy of the dimolybdenum triple bond.[2,21,22] In
general, the protons that are closer to the axis of the Mo–
Mo triple bond are more shielded, and the protons that are
equatorial to it are more deshielded. This effect, which was
originally proposed in 1972 by San Filippo,[23] was first ob-
served by means of low-temperature NMR measurements
2.47–1.46[b]
1.25 (R–CH3)
47.0, 31.7, 29.6, 25.7, 23.8, 23.1[b]
147.6 (RHC=CH2)
111.0 (RHC=CH2)
81.2 (R3CO)
3[a] 6.26 (RHC=CH2)
5.24 (RHC=CH2)
4.93 (RHC=CH2)
1.62 (RCH3)
28.8 (RCH3)
88.62 (R3CO)
4[a] 1.17 (RCH3)
5.25 (OCH2R)
1.13 (RCH3)
34.27 (RCCH3)
76.67 (OCH2R)
32.22 (R3CCH2O)
26.36 (RCH3)
146.8 (RHC=CH2)
110.1 (RHC=CH2)
80.5 (R3CO)
28.0 (RCH3)
80.5 (OCH2R)
34.7 (R3CCH2O)
28.0 (RCH3)
87.9 (R3CO)
5[a] 5.70 (RHC=CH2)
4.92 (RHC=CH2)
4.66 (RHC=CH2)
1.62 (RCH3)
[5]
for Mo2(NMe2)6 and resulted in a splitting of the singlet
signal observed at room temperature into two different res-
onances for the magnetically nonequivalent methyl groups.
In the present case we assume that the different resonances
are a result of a change in the spatial position of the at-
tached ligands relative to the Mo–Mo triple bond on going
from the homoleptic to the heteroleptic complexes, which is
due to the change in the geometric environment caused by
the different adjacent ligands.
5.32 (OCH2R)
1.07 (RCH3)
6[a] 2.68 (RCH2OCH3)
2.55 (RCH2OCH3)
1.33 (RCH3)
80.5 (RCH2OCH3)
57.4 (RCH2OCH3)
26.5 (RCH3)
5.38 (OCH2R)
1.20 (RCH3)
77.9 (OCH2R)
34.3 (R3CCH2O)
27.6 (RCH3)
7[a] 5.31 (R2C=CHR)
1.61 (RCH3)
133.2 (R2C=CHR)
121.2 (R2C=CHR)
83.5 (R3CO)
2.14–1.26[b]
1.19 (RCH3)
5.52 (OCH2R)
1.17 (RCH3)
46.4, 34.4, 29.2, 26.3, 24.4, 23.2[b]
77.3 (OCH2R)
31.3 (R3CH2O)
27.5 (RCH3)
82.4 (R3CO)
37.6 (RCH2R)
32.4 (RCH2R)
8[c] 1.77 (RCH2R)
1.27 (RCH2R)
1.11 (RCH3)
27.6 (RCH3)
68.2 (R3CO)
31.3 (RCH3)
1
Figure 3. Comparison between the H NMR spectra for 6 (blue),
[a] The chemical shifts were measured in [D6]benzene at 25 °C. [b]
The respective resonances overlapped and could not be assigned
unambiguously. [c] The chemical shifts were measured in [D1]-
chloroform at 25 °C.
its corresponding homoleptic derivative 3 (red) and Mo2(Oneop-
ent)6 (green). The spectra were measured in [D6]benzene.
A similar effect was observed for complex 8. Here, the
expected set of signals was observed, but the resonance that
corresponds to the ethylene bridge in the DMH ligand was
strongly shifted to lower field. Once again we attributed this
shift to the diamagnetic anisotropy of the Mo–Mo triple
bond. If we assume, as previously reported, that the Mo–
Mo triply bonded complexes are invariant to conforma-
tional changes in solution, we can approximately compare
the crystallographic data with the obtained NMR chemical
shifts (Figure 9). In complex 8, the DMH ligand bridges
the Mo–Mo triple bond, and the ethylene unit is located
across the axial position. Owing to the aforementioned an-
isotropy, the related protons should be strongly deshielded.
1
Figure 2. Comparison between the 1H NMR spectra for 1 (blue)
and its corresponding lithium alcoholate (red).
In fact, the corresponding H NMR resonance appears at
low field (δ = 1.77 ppm).
Eur. J. Inorg. Chem. 2011, 1725–1732
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1727

S. Krackl, J.-G. Ma, Y. Aksu, M. Driess
FULL PAPER
95Mo NMR measurements for 1–8 were performed in or-
der to confirm the presence of the Mo–Mo triple bond,
which gives rise to significant resonances at lower field. This
enabled us to easily distinguish the dimolybdenum(III) alk-
oxides from the mononuclear MoIII species and the other
species with Mo in different oxidation states (Figure 4).[24]
Figure 4. 95Mo NMR spectrum for 1, which was dissolved in [D6]-
benzene.
Figure 5. ORTEP presentation of the heteroleptic compound 4.
Hydrogen atoms and equal ligands are omitted for clarity.
Although the resonances are relatively broad and low in
intensity, which is due to the quadrupole moment and the
relatively low natural abundance of the 95Mo nuclei, we
were able to confirm that the triple bond of the Mo2 sub-
unit is retained in solution (Table 2).
The tBuO ligands exhibit the longest Mo–O bond
[1.906(3) Å] in the molecular structure of 4. They are con-
nected to different molybdenum atoms and are thus trans
to one another. In addition, they are oriented differently in
space relative to the Mo–Mo triple bond compared to the
neopentO ligands.
Table 2. 95Mo NMR data for 1–8.[a]
Compound
δ [ppm]
Compound
δ [ppm]
Mo2(OMBE)2(Oneopent)4 (5)
1
2
2632
2720
2645
2667
2593
5
2611
2602
2481
2689
2445
The Mo–Mo bond length for complex 5 is 2.2219(14) Å,
which is practically the same as that for 4 and for the related
homoleptic analogues (Table 3).[24] As described for the
tBuO ligands in 4, the MBE ligands are located differently
from the residual neopentO ligands relative to the Mo–Mo
triple bond and display the longest Mo–O bond
[1.920(4) Å] in the molecular structure of 5 (Figure 6).
6
[24]
Mo2(OtBu)6
3
4
7
8
[24]
Mo2(neopent)6
[a] The chemical shifts were measured in [D6]benzene at 25 °C.
As previously reported, the resonances depend on the
substitution pattern at the Mo–Mo triple bond, and the
shielding of the 95Mo nuclei increases from the tertiary to
the secondary and the primary alkoxides.[24] Interestingly,
the synthesized heteroleptic complexes gave rise to reso-
Table 3. Selected bond lengths [Å] and angles [°] for 4–8.
4[a]
5[a]
6
7[a]
8[b]
Mo1–Mo1Ј
2.2217(8)
1.906(3)
1.896(3)
1.883(3)
2.2219(14) 2.2416(11) 2.2464(7) 2.246(1)
1.920(4) 1.929(5) 1.880(3) 1.887(4)
1.879(4) 1.889(5) 1.912(3) 1.887(4)
1.887(4) 1.901(4) 1.872(3) 1.924(6)
115.47(17) 115.7(2) 109.82(12) 115.8(2)
116.60(18) 118.1(2) 116.36(12) 111.4(2)
115.47(17) 114.7(2) 116.22(11) 115.8(2)
nances that are the average of those observed for the corre- Mo1–O2
sponding homoleptic derivatives.
Mo1–O3
Mo1–O1
O1–Mo1–O2 114.45(14)
O2–Mo1–O3 115.89(14)
O3–Mo1–O1 115.44(14)
Molecular Structures
[a] The selected data corresponds to only one of the two molecules
in a single unit cell. [b] O2 and O3 are identical due to the crystal
symmetry in complex 8.
Complexes 4–8 were investigated by single-crystal X-ray
diffraction analysis (Table 4). In general the complexes
show a typical “ethane-like” constitution and an asymmet-
ric ligand distribution around the dimolybdenum triple
bond (Figure 10), as reported for their homoleptic
M2(OR)6 (M = Mo, W) analogues.
Interestingly, the C–C double bond is turned towards the
Lewis-acid site of the triple bond (average distance
3.416 Å), which could indicate π–metal interactions. How-
ever, no lengthening of the Mo–Mo triple bond is observed.
Thus, this interaction appears to be very weak, and the ori-
entation of the ligand may be induced by the crystal pack-
ing.
Mo2(OtBu)2(Oneopent)4 (4)
The crystal structure of 4 (Figure 5) revealed an Mo–Mo
bond length of 2.2217(8) Å, which is similar to the value
reported for the homoleptic complex Mo2(Oneopent)6
[2.222(2) Å]. The alkoxido ligands are located in a staggered
Mo2(OMMP)2(Oneopent)4 (6)
In complex 6, which was formally obtained by the ex-
conformation and exhibit Mo–O bond lengths of 1.906(3), change of the C–C double bond in complex 5 with a meth-
1.896(3) and 1.883(3) Å. oxy group, the Mo–Mo distance is significantly elongated
1728
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1725–1732

Homo- and Heteroleptic, Triply Bonded Dimolybdenum Hexaalkoxides
Figure 6. ORTEP presentation of the heteroleptic complex 5. Hy-
drogen atoms and equal ligands are omitted for clarity.
Figure 8. ORTEP presentation of the homoleptic complex 7. Hy-
drogen atoms and equal ligands are omitted for clarity.
to 2.2416(11) Å (Figure 7). Once again, the MMPO ligands
exhibit the longest Mo–O bond [1.929(5) Å] in the molecu-
lar structure of 6, and they are oriented differently in space
from the neopentO ligands. Similarly to 5, the oxygen atom
of the methoxy functionality points towards the Lewis-acid
site of the Mo–Mo triple bond, which led to a relatively
short intramolecular O4ǞMo distance of 2.665 Å and thus
potential Mo–O interactions. This was further supported
by the lengthening of the Mo–Mo distance for 6 (2.242 vs
2.221 Å in 5).
ably caused by the introduction of the bidentate ligand, as
already reported for other complexes.[15] Once again, un-
equal Mo–O distances are observed. In this case, the tBuO
ligands exhibit the longest Mo–O bond [1.924(6) Å] in the
molecular structure of 8 and a different orientation relative
to the dimolybdenum triple bond.
Figure 9. ORTEP presentation of the homoleptic complex 8. Hy-
drogen atoms and equal ligands are omitted for clarity.
Figure 7. ORTEP presentation of the heteroleptic complex 6. Hy-
drogen atoms and equal ligands are omitted for clarity.
In Figure 10, the general pattern for the asymmetric dis-
tribution of the alkoxide ligands in the complexes 4–8 is
diagrammed.
Mo2(OTerp)4(Oneopent)2 (7)
The crystal structure for 7 confirmed that only two,
rather than four, of the TerpO ligands were exchanged by
the partial alcoholysis of the Mo2(OtBu)6 precursor. The
Mo–Mo bond [2.2464(7) Å] is significantly longer than
those for complexes 4 and 5 (Figure 8), which is most likely
because of the steric congestion of the TerpO ligands.
This complex shows unequal Mo–O bond lengths in a
similar pattern to the other complexes described herein. The
longest Mo–O bond [1.912(3) Å] in the molecular structure
for 7 is exhibited by one pair of the TerpO ligands.
Figure 10. Schematic presentation of the asymmetric ligand distri-
bution in the heteroleptic alkoxides 4–8.
Mo2(ODMH)2(OtBu)2 (8)
The Mo–Mo bond length for complex 8 is 2.246(1) Å
In the heteroleptic complexes 4–7, the tertiary alcohol
(Figure 9). The slightly longer bond compared to those of always exhibits the differing ligand position (R), which has
the unfunctionalized monodentate analogues is most prob- the longest Mo–O bond and exhibits the different orienta-
Eur. J. Inorg. Chem. 2011, 1725–1732
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1729

S. Krackl, J.-G. Ma, Y. Aksu, M. Driess
FULL PAPER
sured in a cold stream of N2. The data for 4–8 were collected with
a Bruker-AXS SMART CCD diffractometer (Mo-Kα radiation, λ
= 0.71707 Å, ω-scan). The structures were solved by direct meth-
ods. The refinements were carried out with the SHELXL-97 pack-
age.[25] All of the thermal displacement parameters were refined
anistropically for non-H atoms and isotropically for H atoms. All
of the refinements were carried out by full-matrix least-squares re-
finement on F2. Details are listed in Table 4. CCDC-764328 (for
5), -764329 (for 6), -764330 (for 4), -764331 (for 7) and 783446
(for 8) contain the supplementary crystallographic data for this
tion in the corresponding molecule. Although the reason
for this favoured asymmetric distribution around the dimo-
lybdenum triple bond is unknown, the data shows a clear
correlation between the steric congestion of the α-carbon
atom of the alkoxido ligand and the resulting distribution
in the heteroleptic, monodentate complexes. Based on the
literature and on our own observations,[2] we assumed that
the main factor that determines the arrangement of the al-
koxido ligands in the heteroleptic complexes is the provision
of the best possible metal–metal interaction and thus a paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
strong Mo–Mo triple bond. Owing to the different steric
demand of the alkoxido ligands, the tertiary alkoxides pre-
fer a distinct ligand position in order to minimize the intra-
molecular repulsion (Figure 10).
Mo2(OMMP)6 (1): A solution of MMPOH (416 mg, 4 mmol) in
hexane (4 mL) was slowly added at room temperature to a solution
of Mo2(OtBu)6 (400 mg, 0.64 mmol) in hexane (10 mL). The solu-
tion was stirred overnight, and an almost pure, orange solid was
obtained after evaporation of the volatiles. The product was puri-
fied by recrystallization, using very small volumes of hexane, and
obtained as an orange powder. The yield was 74% (384 mg,
0.47 mmol). C30H66Mo2O12 (810.7): calcd. C 44.44, H 8.21; found
Conclusions
In this paper we have presented the successful introduc-
tion of different functional alkoxide groups into triply
bonded Mo–Mo complexes. The alkoxides have a C–C
double bond and a methoxy group in their ligand back-
bone. Furthermore, partial alcoholysis was applied in order
to synthesize previously unknown heteroleptic, mono-
dentate alkoxides. The resulting complexes 1–8 show that
1
C 43.81, H 7.93. H NMR (C6D6, 298 K): δ = 3.44 (s, 12 H), 3.18
(s, 18 H), 1.60 (s, 36 H) ppm. 13C NMR (C6D6, 298 K): δ = 81.7
(s), 78.9 (s), 58.2 (s), 29.0 (s) ppm. 95Mo NMR (C6D6, 298 K): δ =
2632 ppm.
Mo2(OTerp)6 (2): A solution of (–)-p-terpineol (308 mg, 2 mmol)
in hexane (2 mL) was slowly added at 0 °C to a solution of
Mo2(OtBu)6 (200 mg, 0.32 mmol) in hexane (5 mL). The solution
was warmed to room temperature and stirred overnight. An almost
pure, orange solid was obtained after evaporation of the volatiles.
The product was purified by recrystallization, using very small vol-
umes of hexane, and obtained as an orange, crystalline solid. The
yield was 67% (234 mg, 0.21 mmol). C60H102Mo2O6 (1111.3):
calcd. C 64.85, H 9.25; found C 64.53, H 9.18. 1H NMR (C6D6,
298 K): δ = 5.53 (br. s, 6 H), 1.70 (br. s, 36 H), 1.46–2.47 (br. m, 7
H), 1.25 (br. s, 3 H) ppm. 13C NMR (C6D6, 298 K): δ = 133.6 (s),
121.8 (s), 83.6 (s), 47.0 (s), 31.7 (s), 29.6 (s), 25.7 (s), 23.8 (s), 23.1
(s) ppm. 95Mo NMR (C6D6, 298 K): δ = 2720 ppm.
1
the H NMR chemical shifts of the attached alkoxide li-
gands are substantially influenced by the magnetic anisot-
ropy of the Mo–Mo triple bond. All of the new complexes
were investigated structurally and reveal a general pattern
in the distribution of the unequal Mo–O bond lengths. The
new functionalized complexes could be useful building
blocks for inorganic–organic hybrid polymers and/or for
heterogenization of dimolybdenum complexes on solid sup-
ports. The investigations into the respective applications are
currently in progress.
Mo2(OMBE)6 (3): MBEOH (176 mg, 2.04 mmol) was dissolved in
hexane (5 mL), and 1 equiv. of BuLi (1.3 mL of a 1.6 m hexane
solution) was added at –78 °C. After the reaction mixture had been
allowed to warm to room temperature, the volatiles were evapo-
rated in vacuo. Hexane (5 mL) was added, followed by the slow
addition of Mo2Cl6(DME)2 (200 mg, 0.34 mmol) at –20 °C. After
the solution had been stirred overnight, it was filtered through Ce-
lite, and the filter cake was washed with hexane (2ϫ2 mL). The
solvent was evaporated from the combined filtrates to give an
orange-brown residue. This solid was extracted with hexane
(2ϫ5 mL), and the extracts were filtered to give a dark orange
solution. Evaporation of the volatiles led to an orange, analytically
pure solid. The yield was 79% (186 mg, 0.27 mmol). C30H54Mo2O6
Experimental Section
General Methods and Starting Materials: All of the reactions were
performed under anaerobic conditions by using standard Schlenk
techniques. The solvents and alcohols were purchased from Sigma
Aldrich, TCI Europe and Alfa Aesar, heated at reflux in the pres-
ence of an appropriate drying agent, distilled, degassed and N2-
saturated prior to use. The starting materials Mo2Cl6(DME)2 and
[20]
Mo2(OtBu)6 were synthesized as described in the literature with
only slight variations.
Instrumentation and Physical Measurements: The 1H and 13C NMR
spectra (TMS standard) were recorded with Bruker ARX 200 (1H,
200 MHz; 13C, 50 MHz) and ARX 400 (1H, 400 MHz; 13C,
100.64 MHz) spectrometers at ambient temperature. The 95Mo
NMR spectra (Na2MoO4 as standard) were recorded with a Bruker
DRX 600 (95Mo, 39.2 MHz) spectrometer, equipped with a 5 mm
BBO probe head and ATM unit by using self-shielded gradients.
The experiments were performed without proton decoupling and
with a repetition rate of 4 scans per second, a 200 ms relaxation
time and a 50 ms acquisition time. The elemental analyses were
performed with a Perkin–Elmer Series II CHNS/O Analyzer 2400
instrument.
1
(702.6): calcd. C 51.28, H 7.75; found C 50.74, H 7.71. H NMR
3
3
(C6D6, 298 K): δ = 6.26 (q, JH,H = 17.28 Hz, JH,H = 10.80 Hz, 6
3
2
H), 5.24 (dd, JH,H = 17.28 Hz, JH,H = 1.50 Hz, 12 H), 4.93 (dd,
3JH,H = 10.80 Hz, JH,H = 1.50 Hz, 1 H), 1.62 (s, 36 H) ppm. 13C
2
NMR (C6D6, 298 K): δ = 147.6 (s), 111.0 (s), 81.2 (s), 28.8 (s) ppm.
95Mo NMR (C6D6): δ = 2667 ppm.
Mo2(OtBu)2(Oneopent)4 (4): A solution of neopentOH (114 mg,
1.3 mmol) in hexane (2 mL) was added dropwise at room tempera-
ture to a solution of Mo2(OtBu)6 (200 mg, 0.32 mmol) in hexane
(5 mL). After the solution had been stirred for 3 h, the volatiles
were evaporated in vacuo, which resulted in a yellow solid. The
Single-Crystal X-ray Structure Determination: The single-crystals
were mounted on a glass capillary in perfluorinated oil and mea-
1730
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1725–1732

Homo- and Heteroleptic, Triply Bonded Dimolybdenum Hexaalkoxides
Table 4. Single-crystal X-ray data and refinement parameters for 4–8.
4
5
6
7
8
Empirical formula
M [g/mol]
T [K]
Crystal system
Space group
a [Å]
b [Å]
c [Å]
α [°]
β [°]
C28H62Mo2O6
686.66
150(2)
C30H62Mo2O6
710.68
150(2)
C30H66Mo2O8
746.71
150(2)
C50H90Mo2O6
979.10
150(2)
monoclinic
P21/c
10.0608(3)
20.0873(7)
13.5462(5)
90
111.202(4)
90
2552.30(15)
2
1.274
1044
C24H50Mo2O6
626.52
150(2)
orthorhombic
Pnnm
8.3134(7)
9.7943(10)
18.142(2)
90
90
90
1477.2(3)
2
1.409
triclinic
triclinic
triclinic
¯
¯
¯
P1
P1
P1
11.9298(3)
12.0255(4)
15.0270(5)
78.301(3)
71.871(3)
62.217(3)
1808.32(10)
2
11.8740(6)
12.1842(6)
15.6584(8)
75.568(4)
69.980(5)
61.083(5)
1853.44(16)
2
10.0485(8)
11.0771(6)
20.0787(15)
97.100(5)
99.637(6)
114.841(6)
1952.0(2)
2
γ [°]
V [Å3]
Z
Dcalcd. [g/cm3]
F(000)
1.261
724
1.273
748
1.270
788
652
Crystal size [mm]
θ range [°]
Index ranges
0.21ϫ0.14ϫ0.07
2.99–25.00
–14ՅhՅ14
–13ՅkՅ14
–17ՅlՅ17
14533
0.21ϫ0.09ϫ0.07
3.35–25.00
–14ՅhՅ14
–14ՅkՅ14
–18ՅlՅ18
13876
0.27ϫ0.18ϫ0.07
3.05–25.00
–11ՅhՅ11
–13ՅkՅ13
–23ՅlՅ23
15728
0.37ϫ0.18ϫ0.10
2.97–25.00
–11ՅhՅ11
–18ՅkՅ23
–13ՅlՅ16
11693
0.41ϫ0.17ϫ0.16
3.40–25.00
–7ՅhՅ9
–11ՅkՅ10
–21ՅlՅ19
5514
Reflections collected
Independent reflections
Rint
6349
0.0275
6520
0.0716
6844
0.0567
4357
0.0410
1350
0.0399
Completeness to θ = 25.00° [%] 99.7
99.8
0.9520, 0.8652
359
99.7
0.9539, 0.8375
379
97.0
0.9485, 0.8266
271
99.7
0.8720, 0.7142
100
Relative transmission factors
Parameters
0.9510, 0.8626
343
GOF
1.044
0.853
1.047
1.211
1.298
Final R indices [IϾ2σ(I)][a,b]
R1 = 0.0450
wR2 = 0.0950
R1 = 0.0665
wR2 = 0.1012
R1 = 0.0521
wR2 = 0.0679
R1 = 0.1393
wR2 = 0.0826
R1 = 0.0718
wR2 = 0.1595
R1 = 0.1049
wR2 = 0.1741
R1 = 0.0553
wR2 = 0.0920
R1 = 0.0791
wR2 = 0.0975
R1 = 0.0565
wR2 = 0.1221
R1 = 0.0623
wR2 = 0.1240
R indices (all data)[a,b]
[a] R1 =∑||Fo| – |Fc||/∑Fo. [b] wR2 = {∑[w(Fo – Fc )]/[w(Fo2)2]} .
2
2
½
product was further purified by sublimation at 100–120 °C or by
recrystallization from a hexane solution at –20 °C in 65% yield
(147 mg, 0.21 mmol). Single crystals, suitable for X-ray diffraction,
were obtained from the recrystallization. C28H62Mo2O6 (687.7):
calcd. C 48.98, H 9.10; found C 48.62, H 9.36. 1H NMR (C6D6,
298 K): δ = 5.25 (br., 8 H), 1.17 (s, 18 H), 1.13 (s, 36 H) ppm. 13C
NMR (C6D6, 298 K): δ = 88.62 (s), 76.67 (s), 34.27 (s), 32.22 (s),
26.36 (s) ppm. 95Mo NMR (C6D6, 298 K): δ = 2593 ppm.
C-3), 57.4 (s), 34.3 (s), 27.6 (s), 26.5 (s) ppm. 95Mo NMR (C6D6,
298 K): δ = 2602 ppm.
Mo2(OTerp)4(Oneopent)2 (7): The complex was synthesized analo-
gously to compound 4, except that the complex could not be puri-
fied by sublimation. Mo2(OTerp)6 (200 mg, 0.18 mmol) was used
and afforded 7 as yellow crystals in a 85% yield (159 mg,
0.16 mmol), which were suitable for single-crystal X-ray diffraction.
C50H90Mo2O6 (979.1): calcd. C 61.33, H 9.26; found C 60.73, H
8.71. 1H NMR (C6D6, 298 K): δ = 5.52 (br. s, 4 H), 5.31 (br., 4 H),
1.61 (br. s, 24 H), 1.26–2.14 (br. m, 7 H), 1.19 (br. s, 12 H), 1.17
(br. s, 12 H) ppm. 13C NMR (C6D6, 298 K): δ = 133.2 (s), 121.2
(s), 89.0 (s), 83.5 (s), 46.4 (s), 34.4 (s), 31.3 (s), 29.2 (s), 27.5 (s),
26.3 (s), 24.4 (s), 23.2 (s) ppm. 95Mo NMR (C6D6, 298 K): δ =
2481 ppm.
Mo2(OMBE)2(Oneopent)4 (5): The complex was synthesized analo-
gously to compound 4. Mo2(OMBE)6 (200 mg, 0.29 mmol) was
used and afforded 5 as yellow crystals in 89% yield (183 mg,
0.26 mmol), which were suitable for single-crystal X-ray diffraction.
C30H62Mo2O6 (710.7): calcd. C 50.70, H 8.79; found C 50.33: H
8.69. 1H NMR (C6D6, 298 K): δ = 5.70 (q, 3JH,H = 17.35 Hz, 3JH,H
= 10.75 Hz, 2 H), 5.32 (br., 8 H), 4.92 (dd, 3JH,H = 17.35 Hz, 2JH,H
Mo2(O2DMH)2(OtBu)2 (8): A solution of DMH(OH)2 (93 mg,
0.64 mmol) in hexane (2 mL) was added dropwise at room tem-
perature to a solution of Mo2(OtBu)6 (200 mg, 0.32 mmol) in hex-
ane (5 mL). After the solution had been stirred for 3 h, a yellow
solid was obtained, which was filtered and washed with a small
quantity of hexane. The product was further purified by recrystalli-
zation from dichloromethane at –20 °C in 89% yield (179 mg,
0.29 mmol). The single crystals obtained from the recrystallization
were suitable for X-ray diffraction. C24H50Mo2O6 (626.5): calcd. C
3
2
= 1.34 Hz, 2 H), 4.66 (dd, JH,H = 10.75 Hz, JH,H = 1.34 Hz, 2
H), 1.62 (s, 36 H), 1.07 (s, 36 H) ppm. 13C NMR (C6D6, 298 K): δ
= 146.8 (s), 110.1 (s), 80.5 (s), 34.7 (s), 28.0 (s) ppm. 95Mo NMR
(C6D6, 298 K): δ = 2611 ppm.
Mo2(OMMP)2(Oneopent)4 (6): The complex was synthesized anal-
ogously to compound 4 or with Mo2(Oneopent)6 as the starting
material. Mo2(Oneopent)6 (264 mg, 0.37 mmol) and MMPOH
(77 mg, 0.74 mmol) were used and afforded 6 as yellow crystals in
a 76% yield (210 mg, 0.28 mmol), which were suitable for single-
crystal X-ray diffraction. C30H66Mo2O8 (746.7): calcd. C 48.25, H
1
46.01, H 8.04; found C 45.82, H 8.08. H NMR ([D1]chloroform,
298 K): δ = 1.77 (s, 8 H), 1.27 (s, 18 H), 1.11 (s, 24 H) ppm. 13C
NMR ([D1]chloroform, 298 K): δ = 82.4 (s), 68.2 (s), 37.6 (s), 32.4
(s), 31.3 (s), 27.6 (s) ppm. 95Mo NMR (C6D6, 298 K): δ =
2689 ppm.
1
8.91; found C 48.39, H 8.24. H NMR (C6D6, 298 K): δ = 5.38 (q,
8 H), 2.68 (s, 4 H), 2.55 (s, 6 H), 1.33 (s, 12 H), 1.20 (s, 36 H) ppm.
13C NMR (C6D6, 298 K): δ = 87.9 (s, C-6), 80.5 (s, C-2), 77.9 (s,
Eur. J. Inorg. Chem. 2011, 1725–1732
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1731

S. Krackl, J.-G. Ma, Y. Aksu, M. Driess
FULL PAPER
[10]
[11]
M. H. Chisholm, J. Organomet. Chem. 1990, 400, 235–253.
M. H. Chisholm, K. Folting, J. C. Huffman, I. P. Rothwell, J.
Am. Chem. Soc. 1982, 104, 4389–4399.
M. H. Chisholm, F. A. Cotton, M. W. Extine, W. W. Reichert,
J. Am. Chem. Soc. 1978, 100, 1727–1734.
M. H. Chisholm, F. A. Cotton, M. W. Extine, R. L. Kelly, J.
Am. Chem. Soc. 1979, 101, 7645–7650.
M. H. Chisholm, R. L. Kelly, F. A. Cotton, M. W. Extine, J.
Am. Chem. Soc. 1978, 100, 2256–2257.
T. M. Gilbert, C. B. Bauer, A. H. Bond, R. D. Rogers, Polyhe-
dron 1999, 18, 1293–1301.
M. H. Chisolm, F. A. Cotton, C. A. Murillo, W. W. Reichert,
Inorg. Chem. 1977, 16, 1801–1808.
R. H. Cayton, M. H. Chisholm, Inorg. Chem. 1991, 30, 1422–
1425.
F. A. Cotton, S. Kitagawa, Polyhedron 1988, 7, 463–470.
J. Ma, Y. Aksu, L. J. Gregoriades, J. Sauer, M. Driess, Dalton
Trans. 2010, 39, 103–106.
Acknowledgments
We thank the Cluster of Excellence “Unifying Concepts in Cataly-
sis“ (sponsored by the Deutsche Forschungsgemeinschaft and ad-
ministered by the Technische Universität Berlin) for financial sup-
port. We also thank Dr. Peter Schmieder at the Leibniz-Institut für
molekulare Pharmakologie in Berlin (Robert-Rössle-Str. 10, 13125
Berlin) for recording the 95Mo NMR spectra. S. K. thanks the
Fonds der Chemischen Industrie for financial support (Kekulé
scholarship) as well as the Berlin International Graduate School
for Science and Engineering (BIG-NSE) and Dr. A. Company for
help.
[12]
[13]
[14]
[15]
[16]
[17]
[1] F. A. Cotton, C. B. Harris, Inorg. Chem. 1965, 4, 330–333.
[2] F. A. Cotton, C. A. Murillo, R. A. Walton, Multiple Bonds be-
tween Metal Atoms, 3rd ed., Springer Science and Business Me-
dia Inc., New York, 2005.
[18]
[19]
[3] M. H. Chisholm, Pure Appl. Chem. 1991, 63, 665–680.
[4] F. Huq, W. Mowat, A. Shortland, A. C. Skapski, G. Wilkinson,
J. Chem. Soc., Chem. Commun. 1971, 1079–1080.
[5] M. H. Chisholm, F. A. Cotton, B. A. Frenz, W. W. Reichert,
L. W. Shive, B. R. Stults, J. Am. Chem. Soc. 1976, 98, 4469–
4476.
[6] M. H. Chisholm, F. A. Cotton, C. A. Murillo, W. W. Reichert,
Inorg. Chem. 1977, 16, 1801–1808.
[7] M. H. Chisholm, Polyhedron 1983, 2, 681–721.
[8] M. H. Chisholm, K. Folting, J. C. Huffman, C. C. Kirkpatrick,
Inorg. Chem. 1984, 23, 1021–1037.
[20]
[21]
T. M. Gilbert, A. M. Landes, R. D. Rogers, Inorg. Chem. 1992,
31, 3438–3444.
T. P. Blatchford, M. H. Chisholm, J. C. Huffman, Polyhedron
1987, 6, 1677–1680.
M. J. McGlinchey, Inorg. Chem. 1980, 19, 1392–1394.
J. S. Filippo, Inorg. Chem. 1972, 11, 3140–3143.
C. G. Young, E. M. Kober, J. H. Enemark, Polyhedron 1987, 6,
255–259.
G. M. Sheldrick, SHELXL-97, University of Göttingen, Ger-
many, 1997.
[22]
[23]
[24]
[25]
[9] M. H. Chisholm, C. C. Kirkpatrick, J. C. Huffman, Inorg.
Chem. 1981, 20, 871–876.
Received: November 24, 2010
Published Online: February 25, 2011
1732
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1725–1732