Thermally stable uranium dinitrogen complex with siloxide supporting ligands

Article abstract of DOI:10.1021/om4003957

A new dinitrogen adduct of a homoleptic uranium tris(siloxide) complex, [U{OSi(Mes)₃}₃]₂(μ-η²: η²-N₂), is reported. Synthesis of the ¹⁵N-labeled isotopomer and Raman spectroscopy confirm the reductive activation of N₂ to a (N₂)^2- dianion. The ¹⁵N NMR shift of the ¹⁵N₂-labeled isotopomer is also reported. Crystallographic characterization shows a side-on (N ₂)^2- coordinated in either an eclipsed or staggered conformation in different crystals. The U-N₂-U complex is stable to vacuum and shows high thermal stability, retaining the formally reduced dinitrogen at 100 C. The parent three-coordinate uranium(III) [U{OSi(Mes) ₃}₃] could not be isolated in our hands, with N ₂-free syntheses affording only uranium(IV) compounds. The rational synthesis and full characterization of two such U(IV) byproducts, [U{OSi(Mes)₃}{N(SiMe₃)₂}₃] and [U{OSi(Mes)₃}₄], is also reported.

Full text of DOI:10.1021/om4003957

Article
pubs.acs.org/Organometallics
Thermally Stable Uranium Dinitrogen Complex with Siloxide
Supporting Ligands
Stephen M. Mansell, Joy H. Farnaby, Anne I. Germeroth, and Polly L. Arnold*
EaStCHEM School of Chemistry, University of Edinburgh, Joseph Black Building, The King’s Buildings, Edinburgh, EH9 3JJ, U.K.
S
* Supporting Information
ABSTRACT: A new dinitrogen adduct of a homoleptic uranium
tris(siloxide) complex, [U{OSi(Mes)3}3]2(μ-η2:η2-N2), is reported. Syn-
thesis of the 15N-labeled isotopomer and Raman spectroscopy confirm the
reductive activation of N2 to a (N2)2− dianion. The 15N NMR shift of the
15N2-labeled isotopomer is also reported. Crystallographic characterization
shows a side-on (N2)2− coordinated in either an eclipsed or staggered
conformation in different crystals. The U−N2−U complex is stable to
vacuum and shows high thermal stability, retaining the formally reduced dinitrogen at 100 °C. The parent three-coordinate
uranium(III) [U{OSi(Mes)3}3] could not be isolated in our hands, with N2-free syntheses affording only uranium(IV)
compounds. The rational synthesis and full characterization of two such U(IV) byproducts, [U{OSi(Mes)3}{N(SiMe3)2}3] and
[U{OSi(Mes)3}4], is also reported.
synthons for Ln2+ compounds that were too reducing to be
INTRODUCTION
■
isolate otherwise, at the time.32,33 This methodology has even
Uranium(III) complexes are increasingly revealing unique
been extended to Sc with the isolation of {(Cp*)2Sc}2(μ-η2:η2-
small-molecule reactivity based upon the strong reducing
N2).34 Further reduction to an (N2)3− species has also been
potential of the UIII ion.1−3 Arene activation and C−H bond
demonstrated,35,36 and a recent report showed that photo-
borylation exclusively using uranium,4 reductive carbon
chemical activation of heteroleptic cyclopentadienyl derivatives
monoxide coupling,5,6 and carbon dioxide activation7 are but
of the lanthanides also leads to binding and reduction of
a few examples. The study of transition metal dinitrogen
dinitrogen.37
binding is now well advanced8,9 since the initial report of a
The first reported actinide dinitrogen complex was formed
transition metal dinitrogen complex.10,11 Particularly remark-
by exposure of the “naked” UIII[tren′] fragment to N2, affording
able is the catalytic generation of ammonia from N2 achieved by
{U(tren′)}2(μ-η2:η2-N2), A (tren′ = N(CH2CH2NSitBuMe2)3),
Chart 1, from which the N2 readily dissociates under
vacuum.38,39 Weaker N2 binding was observed in the binuclear
uranium-pentalene complex {U(Cp*)(C8H4{SiiPr3-1,4}2)}2(μ-
η2:η2-N2), B,40 and in mononuclear (Cp*)3U(η1-N2), C, in
which the end-on-bound N2 can be maintained only at
dinitrogen pressures above 5.4 atm.41 The heterometallic
complex D was isolated using the combination of UIII and
MoIII amides and showed end-on bridging between the two
metals.42 Most recently, we showed that the U(III) tris-
(aryloxide) U(OTtbp)3 (OTtbp = O-2,4,6-tBu3C6H2) binds N2
in a side-on geometry in {U(OTtbp)3}2(μ-η2:η2-N2), E. This
last complex releases N2 only upon heating a solution to 80 °C,
even in the presence of small O-containing reagents such as
CO2 and CO, whereas the di-tert-butyl aryloxide analogue,
{U(ODtbp)3}2(μ-η2:η2-N2) (Dtbp = 2,6-tBu2C6H3), F, loses
dinitrogen readily, and the majority of the product isolated in
its synthesis is N2-free.43 The use of bulky siloxide ligands in
low-coordinate transition metal44−49 and rare earth50−55
chemistry is well known, and a recent paper has detailed the
use of [OSi(OtBu)3]− in U(III) chemistry.56 We now describe
several molybdenum complexes12−16 with the ultimate goal of
mimicking the huge scales of nitrogen fixation achieved by the
Haber−Bosch process or nature’s use of nitrogenase enzymes.
Today, heterogeneous iron and ruthenium catalysts are used in
the Haber−Bosch process,17 and iron centers are also present
in nitrogenase enzymes;18,19 consequently much research has
focused on molecular iron and ruthenium dinitrogen complexes
and their reaction with dihydrogen.20 However, it was also
observed that heterogeneous uranium catalysts were also
competent for the Haber−Bosch process,17,21 an interesting
observation, as dinitrogen binding is still very rare in f-block
coordination chemistry.22,23 Dinitrogen binding to an f-block
cation was first demonstrated at a highly reducing Sm(II)
center, through side-on binding in {(Cp*)2Sm}2(μ-η2:η2-N2)
(Cp* = C5Me5), although binding was easily reversed, as the N2
complex is in equilibrium with the N2-free complex.24 Since
then, f-block dinitrogen binding has advanced with the
formation of N2 compounds with more reducing Ln2+ centers,
such as Tm, Dy, and Nd,25−27 due to the availability of suitable
Ln2+ halides as starting materials.28−31 A new avenue for
exploring lanthanide reduction chemistry was also pioneered by
the combination of Ln3+ compounds with potassium under
dinitrogen to generate the corresponding Ln23+-(N2)2−
complexes, which were then used as reducing agents and
Received: May 10, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/om4003957 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Chart 1. Uranium Dinitrogen Compounds
Scheme 1
new dinitrogen compounds of uranium, stabilized using bulky
trimesitylsiloxide ligands, which can be boiled in hydrocarbon
solvent without loss of N2.
(III) U{OSi(Mes)3}3 could not be isolated in our hands.
Reactions carried out by vacuum transfer of degassed solvents
onto the reagents UN″3 and HOSi(Mes)3 gave almost identical
1H NMR spectra to reactions performed under argon,
dihydrogen, or carbon monoxide; in each case, the major
product was identified as U{OSi(Mes)3}N″3 2, alongside a
minor byproduct identified as U{OSi(Mes)3}4 3 and small
quantities of other unidentified products.
RESULTS
■
Dinitrogen Binding. The bulky trimesitylsilanol HOSi-
(Mes)3 (Mes = mesityl, 2,4,6-Me3C6H2) has previously been
used to make model complexes for Al2Me6-grafted silica
surfaces.57 Protonolysis reactions between UN″3 (N″ =
N(SiMe3)2)58 and 3 equivalents of HOSi(Mes)3 were carried
out in arene or n-hexane solvent, Scheme 1. In both cases, the
title compound [U{OSi(Mes)3}3]2(μ-η2:η2-N2) (1) was
formed, alongside other new byproducts, as observed by in
The room-temperature 1H NMR spectrum of 1 in d8-toluene
(Figure SI.3) shows 15 different resonances from +10 to −17
ppm, consistent with three distinct mesityl ligand environ-
ments. This could arise either from all Mes groups being
magnetically inequivalent to each other in [OSi(Mes)3]− and
every [OSi(Mes)3]− ligand being equivalent or from three
different [OSi(Mes)3]− ligand environments, in which each
Mes group is magnetically equivalent to the other two in the
same ligand. The chemical shifts for 1 are temperature-
dependent and vary linearly with reciprocal temperature,
consistent with the paramagnetism of the complex, but no
fluxional process was observed between +100 and −80 °C
(Figure SI.1). Resonances consistent with a single ligand
environment were observed by 29Si{1H} (+63.6 ppm) NMR
spectroscopy. 15N{1H} NMR spectroscopy has been used to
observe resonances for 15N2-labeled lanthanide-dinitrogen
complexes,32 and the 15N2 resonance for the 98% 15N2-labeled
complex 1-15N2 can be observed at +4213.5 ppm (Figure SI.2).
This is at higher frequency than those seen previously for
paramagnetic lanthanide complexes (+1001 ppm for Ln = Ce,
1
situ H NMR spectroscopy. The rate of formation of 1 was
increased in n-hexane or by heating at 80 °C for 16 h. Careful
crystallization was required to isolate 1 from the byproducts
formed in any of the syntheses. The most effective techniques
in our hands involve direct crystallization from either hot
benzene reaction mixtures or directly from the synthesis carried
out by layering n-hexane solutions of the two starting materials
UN″3 and HOSi(Mes)3. Both result in modest isolated yields
(up to 36%). The major byproducts are UIV complexes:
U{OSi(Mes)3}N″3 2 predominates in ambient temperature
syntheses, and U{OSi(Mes)3}4 3 in elevated temperature
syntheses. The identities of both have been confirmed by
independent syntheses from the reaction of (N″)2U(κ2-
CH2SiMe2NSiMe3)59 with 1 and 4 equivalents of HOSi(Mes)3,
respectively (see below). The parent three-coordinate uranium-
B
dx.doi.org/10.1021/om4003957 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
4f1, and +2383 ppm for Ln = Pr, 4f2, in {(C5Me4H)2Ln-
(thf)}2(μ-η2:η2-N2)),60 but more importantly confirms that
(N2)2− in 1 is retained when samples are subjected to dynamic
vacuum or stored in solution; the labeled 15N2 was not
observed to exchange with 14N2.
Raman Spectroscopy. Raman spectroscopy was per-
formed on samples of 1-14N2 and 1-15N2 sealed in glass
capillaries (Figure 1). The spectrum of 1-14N2 displays four
Figure 2. Thermal ellipsoid plots of the staggered 1S (left) and
eclipsed 1E (right) isomers of 1, together with Newman projections.
Thermal ellipsoids are at 50% probability. Hydrogen atoms and all
carbon atoms except the ipso carbons of the mesityl groups are omitted
for clarity. Both positions of the disordered dinitrogen ligand in 1E are
shown in the end-on view (bottom right drawing). Green = U; blue =
N; red = O; turquoise = Si; gray = C.
Figure 1. Raman spectra of powdered 1-15N2-labeled (upper) and
1-14N2 (lower).
resonances between 1200 and 1700 cm−1; the resonance at
1437 cm−1 was absent in 1-15N2 and was tentatively assigned as
the υN−N. From reduced mass calculations, the υN−N in 1-15N2 is
then predicted at 1388 cm−1; however there was another
resonance from the complex present in both isotopomers that
was observed at a very similar frequency, which precluded the
definitive assignment of the υN−N in 1-15N2, although this
resonance is seen to increase in intensity in 1-15N2. The value
observed for 1 correlates well to the υN−N seen in the only
other Raman spectroscopic study of a uranium-dinitrogen
compound E (1451 cm−1)43 and to the yttrium N2 compound
[{(Me3Si)2N}2(thf)Y]2(μ-η2:η2-N2) (1425 cm−1),35 where both
compounds have been assigned a NN bond order of 2, and is
consistent with the reduction of N2 to (N2)2− in 1. For the Nd
compound [{C5H2(tBu)3}2Nd]2(μ-η2:η2-N2), a resonance at
1622 cm−1 was observed,33 and these values are, as expected,
significantly lower than free N2 (2331 cm−1).33
Structural Chemistry. X-ray diffraction experiments were
carried out on three different sets of crystals of 1, the first
grown from a room-temperature slow diffusion reaction of the
two reagents in n-hexane, the second from cooling of a heated
benzene solution of the reagents, and the third from crystals of
the benzene solvate placed under vacuum (see below). All three
show dinitrogen bound side-on between two UX3 (X =
OSi(Mes)3) fragments but with the X ligands oriented in a
staggered conformation (1S) in the first (hexane) and eclipsed
conformation (1E) in the latter two (both as a benzene solvate
and after storage under vacuum), Figure 2. The NN bond
lengths are 1.124(12) Å for both N2 positions in 1E and
1.080(11) Å in 1S, shorter than the NN bond lengths of
1.236(5) Å found in E43 and 1.232(10) Å in B40 and in end-on,
bridging bimetallic D (1.232(11) Å).42 The NN distances in
both conformers are identical, within standard uncertainy, to
that in A (1.109(7) Å),38 in terminal-bound C (1.120(14) Å),41
and to the distance in free N2 (1.0975 Å). There are a range of
NN distances observed in lanthanide N2 complexes; for
example [{(Me3Si)2N}2(thf)Y]2(μ-η2:η2-N2) (for which a
Raman-active NN stretch was reported: 1.268(3) Å) and
{(DtbpO)2(thf)2Dy}2(μ-η2:η2-N2) (1.257(7) Å)35 are signifi-
cantly longer than free N2, but the distances in 1E and 1S are
indistinguishable from that in weakly bound {(Cp*)2Sm}2(μ-
η2:η2-N2) (1.088(12) Å).24
No agostic interactions are seen in the structures of 1 unlike
in E. The U−O bond lengths are equivalent, 2.119(3)−
2.128(3) Å for 1E and 2.131(4)−2.134(4) Å for 1S. This is
identical to the U−O distance found in [UCp3(OSiPh3)],
2.135(8) Å.61 The U−O distances for the terminal siloxide
ligands in the UIII dimer [U{OSi(OtBu)3}2{μ-OSi(OtBu)3}]2
were found to be slightly longer (mean 2.193(4) Å), while U−
O siloxide distances in a UIV bridging (OCO2)2− compound are
slightly shorter (mean 2.09(1) Å for the terminal siloxide
ligands).56 The Si−O bonds are also equivalent at 1.658(4)−
1.663(4) Å in 1E and 1.659(3)−1.667(3) Å in 1S and
unchanged from the free ligand (1.669(5) Å)
In order to confirm that N2 is retained when the bulk sample
is placed under vacuum, we carried out powder diffraction on a
sample grown from benzene and subsequently placed under
vacuum (Figure 3). The diffractogram, however, did not
identify the structure grown from either n-hexane (1S) or
benzene (1E), but instead a new phase was observed. The
crystal structure of this phase was identified by single-crystal X-
ray diffraction from a single crystal in the same sample that had
been placed under vacuum, and although the data are of poor
quality and demonstrated connectivity only, it revealed the
same eclipsed conformation of 1 (Figure SI.14) but in a
different space group (C2/c in 1E rather than P21/n in 1Evac).
The structure shows the benzene lattice solvate molecules have
been lost, as the unit cell volume had decreased by 2444 Å3,
similar to the decrease expected from the loss of 24 benzene
molecules (approximately 2592 Å3 using the 18 Å3 per atom as
the approximate volume). This demonstrates the stability of the
C
dx.doi.org/10.1021/om4003957 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
and +100 °C (Figure SI.5), although heating 2 at 100 °C also
leads to further reaction presumably from ligand redistribution,
as evidenced by growing resonances for the metallacycle
[(N″)2U(κ2-CH2SiMe2NSiMe3)] occurring from the thermal
decomposition of transiently formed UN″4. The spectrum for 2
at 100 °C appears to be approaching the high-temperature
regime of fast bond rotations, displaying averaged resonances
for hydrogens of the [OSi(Mes)3]− mesityl and (N″)−
trimethylsilyl groups; however, decreasing the temperature
caused decoalescence of a number of signals in multiple stages,
1
precluding a thorough assignment. The H NMR spectrum of
isolated 3 is much more straightforward with a single mesityl
environment, indicating that the siloxide ligands are equivalent.
1
However, H NMR spectra of the crude reaction mixture and
the spectrum of a previously heated sample of 3 (80 °C in
benzene) are complicated by additional resonances for 8 CH
and 12 CH3 mesityl hydrogens. The 1H VT NMR spectra show
a broadening of all ligand resonances above 70 °C. Additionally,
1H DOSY NMR spectroscopy of these samples showed that all
these ligand resonances have the same diffusion coefficient,
suggestive of a single complex with a size of approximately 2 Å
smaller than that of 1, for which a diffusion coefficient was also
measured (see Supporting Information (SI)). This suggested
that dimerization or aggregation was not occurring at these
temperatures. For these reasons, we suggest that the other
species in the reaction mixtures that complicates the spectra is a
lower symmetry isomer of 3. In complexes of the lanthanides
with (OSiPh3)−, such ligand rearrangements have been
observed to be low-energy processes.63 This would make the
total yield of the reaction, calculated from the molar ratio of
bound siloxide versus HN″ produced, 61%. Complex 3 is
acutely air and moisture sensitive and decomposes in the solid
state and in solution under inert conditions (<3 ppm O2),
which precluded meaningful microanalysis. There is literature
precedent for the thermal decomposition of uranium siloxide
complexes, with the formation of hexamethyldisiloxane cited as
a driving force.64 The UIII dimer [U{OSi(OtBu)3}2{μ-OSi-
(OtBu)3}]2 had limited stability in organic solvents, which is
not surprising, given the reducing nature of UIII. However, the
UIV carbonate complex of the same ligand, [U{OSi-
(OtBu)3}3]2(μ-η1:η2-CO3), was also observed to decompose
to U{OSi(OtBu)3}4 and an unknown carbonate complex over
the course of several days.56 In our case the instability of 3 was
unexpected, given the stability of 1.
Figure 3. Identification of the phase of single crystals of 1E placed
under vacuum, then ground to a powder, by comparison with the
calculated powder pattern from the structure of 1Evac (calculated
using the program Mercury).62
uranium-dinitrogen complex to vacuum in the solid state, as the
dinitrogen ligand has been retained despite the loss of solvent
and gross structural change.
1
Thermal Stability Studies. H NMR spectroscopy was
used to monitor the stability of toluene solutions of 1 at 100 °C
but did not reveal simple dissociation of N2 as previously
observed for U(OTtbp)3.43 Integration of the 1H NMR spectra
against an internal standard showed 92% of the complex intact
after 30 min, but substantial loss of complex after another 30
min (60% remaining). However, heating the sample overnight
(16 h) did not lead to complete loss of 1 (48% remained),
indicating a more complicated decomposition process than the
simple elimination of N2. The products of decomposition
appear to be predominantly UIV byproducts with resonances for
U{OSi(Mes)3}4 3 having been tentatively assigned. This
relatively slow decomposition at 100 °C represents remarkable
stability for a uranium dinitrogen complex.
Uranium(IV) Complexes of the Siloxide Ligand. In
order to facilitate identification of UIV byproducts in the
dinitrogen chemistry and to understand the ligand effect,
hetero- and homoleptic complexes of UIV with the trimes-
itylsiloxide ligand were synthesized, Scheme 2. Reactions of the
1H NMR spectroscopy of the three compounds of the
[OSi(Mes)3]− ligand attest to the sterically congested nature of
the multiple mesityl groups within the siloxide ligand and also
Scheme 2
1
the difficulty in interpretation of H NMR spectra of these
paramagnetic compounds, as assignment is necessarily based on
relative integration rather than chemical shift.
X-ray diffraction experiments were carried out on single
crystals of 2 and 3 grown from n-hexane and a toluene/n-
hexane solution, respectively (Figure 4), and showed
approximately tetrahedral uranium centers bound to four
ligands. The U−O bond lengths (2.161(4) Å for 2,
2.128(3)−2.150(2) Å for 3) are significantly shorter than U−
N bond distances (2.282(5)−2.308(5) Å), as seen generally for
uranium alkoxides versus uranium amides, but equivalent
within standard uncertainties to 1, as are the Si−O distances.
The only structural parameter that does change between the
complexes is the U−O−Si angle, which increases from 174° in
3 to 175.5° in 1 to the almost linear 178.8° in 2. The magnitude
of this angle is greater than that observed for uranium aryloxide
UIV metallacycle [(N″)2U(κ2-CH2SiMe2NSiMe3)]59 with 1
equivalent of HOSi(Mes)3 at room temperature or 4
equivalents of HOSi(Mes)3 at elevated temperature allowed
the synthesis of colorless U{OSi(Mes)3}{N(SiMe3)2}3 2 and
very pale pink U{OSi(Mes)3}4 3 in 40% and 22% isolated
yields, respectively. 1H NMR spectroscopy of benzene solutions
of 2 shows multiple exchange processes occurring between −80
D
dx.doi.org/10.1021/om4003957 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 4. Displacement ellipsoid drawing of 2 (left) and 3 (right) with ellipsoids set at 50% probability. Hydrogen atoms are omitted, and carbon
atoms in 3 are represented without thermal ellipsoids for clarity. Green = U; blue = N; red = O; turquoise = Si; gray = C. Selected bond lengths (Å)
for 2: U1−O1 2.161(4), U1−N3 2.282(5), U1−N1 2.295(5), U1−N2 2.308(5), O1−Si1 1.659(4); selected bond lengths (Å) for 3: O1−U1
2.144(2), O2−U1 2.150(2), O3−U1 2.140(2), O4−U1 2.128(3), O1−Si1 1.657(3), O2−Si2 1.657(3), O3−Si3 1.658(3), O4−Si4 1.668(3).
Figure 5. %Vbur for large, monoanionic ligands ordered in increasing size. Calculated using the SambVca program (see text) and the parameters M−
E = 2.1 Å with E = O, similar to 1; M−E = 2.3 for E = N (U−Nav = 2.295 Å in 2); sphere radius of 3.5 Å.
complexes,65,66 and the combination of a short U−O bond and
a large U−O−C angle has been used as an argument for π-
back-bonding in uranium alkoxides67 and aryloxides.65,66
However, the correlation between bond length and π-bond
order ought not to be assumed as simple without knowing the
extent of the ionic contribution to bonding,68 and the linearity
of the M−O−C bond in alkoxides has been demonstrated to be
predominantly controlled by electrostatic effects in complexes
of the lanthanides.69 In complexes 1−3 the U−O−Si angle
does appear to track with the relative steric bulk of the
complexes (see below).
Estimation of the Steric Bulk of the Siloxide Ligand.
Calculations of the “percentage buried volume” (%Vbur) are
useful to compare the steric demands of N-heterocyclic carbene
and phosphine ligands and readily performed using a simple
Web application.70 We have explored whether this approach
can be useful to determine the relative steric demands of bulky
amide, aryloxide, and siloxide ligands.71 Coordinates for the
ligands of interest were all taken from the UIV compounds
U{N(SiHMe2)2}4 for [N(SiMe2H)2]−,72 [U{OSi(Mes)3}{N-
(SiMe3)2}3] 2 for [N(SiMe3)2]− and [OSi(Mes)3]−,
{(DtbpO)3UOC}2 for [ODtbp]−,43 and {(TtbpO)3UOC}2
for [OTtbp]−.43 The process removes the uranium atom and
places a dummy atom at a set distance away from the
connecting element E (here, O or N) using an axis defined by
two atoms for the planar aryloxides and amides or three atoms
for the cone-shaped siloxide ligand. The intrinsic steric bulk of a
ligand can be taken into account by using the same U−E
distance, but we used two values to take into account the
shorter U−O distances; the standard parameters were U−E =
2.1 Å with E = O, similar to that seen in 1 (2.119−2.134 Å),
and U−E = 2.3 Å for E = N (M−Nav = 2.295 Å in 2), with a
sphere radius of 3.5 Å for calculating the buried volume. The
ligands can be ordered in increasing size, [OSi(Mes)3]−
<
[N(SiHMe2)2]− < [OTtbp]− ≈ [ODtbp]− ≈ [N(SiMe3)2]−
(Figure 5), which indicates that the [N(SiMe3)2]− ligand has
similar steric bulk to the aryloxides, but all three are
significantly larger than the [OSi(Mes)3]− siloxide ligand used
in this study. These were calculated assuming the U−O
distance is 0.2 Å shorter than the U−N distance, although U−E
distances can vary slightly depending on the nature of the other
ligands (for the [OSi(Mes)3]− ligand, average U−O in 3 is
2.141 Å and in 2 it is 2.161(4) Å; for [ODtbp]−, average U−O
in [(DtbpO)3U]2(O2C2) is 2.108 Å, 2.135(4) Å in U(ODtbp)4,
and 2.145(8) Å in U(ODtbp)N″3;66 for [N(SiMe3)2]− in the
same complex, average U−N is 2.285 Å compared to 2.295 Å in
2). U−E distances for UIII homoleptic UL3 compounds are in
general longer due to the larger ionic radius of UIII (1.165 Å for
UIII and 1.03 Å for UIV in six-coordinate octahedral environ-
ments).73 For comparison, the %Vbur calculated values for the
amides at a standard U−E distance (2.1 Å) have also been
calculated (see SI) and produce a much larger value for N″
(28.7%), as would be expected.
DISCUSSION
■
The binding of dinitrogen between two UIII cations is
associated with reduction of the N2 to the (N2)2− diazenide
dianion and concomitant formation of two UIV centers, i.e., U
E
dx.doi.org/10.1021/om4003957 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
5f → N2 π-back-bonding.39,42,43,74,75 Traditionally, a shorter
N−N distance close to, or indistinguishable from, free N2 (1
and A) has been attributed to very weak N2 binding and no
reduction in the bond order from free N2, while a longer N−N
distance (B and E) has been attributed to stronger binding of
the N2 and formal reduction to the (N2)2− dianion.
Investigations of the first uranium-dinitrogen compound A,
acknowledged the discrepancy between the experimental short
N−N bond length and the calculations, which corroborated the
direduced nature of the dinitrogen ligand. The authors
attributed the short N−N distance found to the steric
compression between ligands, which does not allow the
uranium atoms to approach the dinitrogen ligand at an optimal
distance.39 They observed an average U−N distance of 2.419 Å
to the dinitrogen ligand in A, but we observed average U−N
distances of 2.459 Å in 1S, 2.411 Å in 1E, and 2.410 Å in E (the
ligands in this complex are staggered), so the uranium centers
appear to be the same distance from the N2 ligand in these
complexes and also in B (av U−N = 2.412 Å) but have
significantly different N−N distances.40 Also, the U−O
distances, which should shorten on a formal UIII → UIV
electron transfer associated with N2 reduction, change
insignificantly for both 1 and E. We have previously noted
that experimental N−N distances, whose data are based on
electron density, rather than atomic position, can be unreliable,
underestimating what is considered the standard (X-ray)
definition for the interatomic N−N distance,43,76,77 and the
vibrational data are a better indication of the degree of N2
reductive activation. Raman spectroscopy of 1S showed a band
at 1437 cm−1, similar to that seen in E (1451 cm−1) and in
agreement with a two-electron reduction and a N−N bond
order of two.43 In this work, there is a discrepancy between the
short N−N distance observed in the molecular structure and
the experimental evidence of N2 reduction to (N2)2− from
Raman spectroscopy and the high thermal stability of 1.
Interestingly, we have been unable to isolate the N2-free,
three-coordinate complex U{OSi(Mes)3}3 by N2 elimination,
by synthesis under argon, or as a solvato complex using
benzene or thf/benzene or with 2,6-dimethylphenyl-isonitrile as
a co-ligand, in contrast to (THF)U(ODtbp)3 and (THF)U-
(OTtbp)3.43,78 The buried volume calculations have allowed us
to consider the kinetic stabilization of uranium compounds for
several ligands and are in agreement with our previous
experimental finding that the much smaller ligand [N-
(SiHMe2)2]− could not stabilize UIII,72 whereas the homoleptic
UN″3 is a well-known UIII precursor, although we have
previously observed UN″3 reactivity with arenes in a
disproportionation pathway. The lack of solvent binding to
UN″3 (such as thf) was presumed to be a consequence of steric
shielding, but thf is readily and irreversibly bound to
U(ODtbp)3 and U(OTtbp)3 to form four-coordinate com-
plexes. Yet our calculations revealed the three-coordinate UIII
aryloxide complexes to be similarly sterically hindered. In fact,
coordination of an N-heterocyclic carbene to UN″3 has been
observed,79 as well as reactivity with CO.80 The trimesitylsil-
oxide ligand was calculated to be much less sterically bulky than
the aryloxide ligands and [N″]− and is also significantly smaller
than [N(SiHMe2)2]−, for which no homoleptic UIII compound
could be isolated. The trimesitylsiloxide ligand could therefore
also be insufficiently kinetically stabilizing to form the UX3
complex without dinitrogen bound or insufficiently reducing.
The absence of close ligand C−H groups that could provide
agostic-like stabilizing interactions, which must benefit the
isolation of compound E, presumably also contributes to this. A
distinct ligand effect in uranium dinitrogen binding is evident
from the literature, where changes to the ligand environment
have produced drastically different complex stabilities, and is
well illustrated by the two closely related aryloxide complexes,
where U(OTtbp)3 forms a dinitrogen complex, E, stable to 80
°C, whereas the material isolated from the exposure of
U(ODtbp)3 to N2 is predominantly dinitrogen-free. It is
worth noting that the few f-block siloxide complexes known
have mainly used ligands or reagents that contain secondary
oxygen atoms, polyhedral organosilsesquioxanes,81 tris(tert-
butoxy)siloxides,51,56 tetraphenyldisiloxanediols,82,83 or hexa-
methylcyclotrisiloxanes,84 which are both electronically and
sterically different from the trimesitylsiloxide used in this study,
but more importantly perhaps, the additional oxygen atoms
present can act as donors for the metal centers, leading to a
chelate effect, by bridging or otherwise, which stabilizes the
resultant complexes.
CONCLUSION
■
The trimesitylsiloxide [OSi(Mes)3]− ligand has been used for
the first time in uranium chemistry and has allowed the
encapsulation and reduction of N2 to (N2)2− in the most robust
uranium dinitrogen complex yet observed, complex 1. The
ligand is also capable of stabilizing heteroleptic and homoleptic
UIV species, but while the ability of the ligands to interdigitate
and provide a protective environment around the reductively
activated N2 may be beneficial here, it makes spectroscopic
study of the complexes difficult. Reasons for our inability to
isolate a simple homoleptic or solvato-UIII adduct remain
unclear.
EXPERIMENTAL SECTION
■
General Details. All manipulations were carried out under a dry,
oxygen-free, dinitrogen atmosphere using standard Schlenk techniques
or in MBraun Unilab or Vacuum Atmospheres OMNI-lab gloveboxes
unless otherwise stated. THF, toluene, and n-hexane were degassed
and dried using a commercial solvent system. Benzene, deuterated
benzene, and toluene were boiled over potassium, vacuum transferred,
and freeze−pump−thaw degassed three times prior to use. 1H spectra
were recorded on Bruker AVA 400 or 600 MHz NMR spectrometers
at 298 K. 13C NMR spectra were recorded on a 500 MHz
spectrometer equipped with a cryoprobe and referenced to internal
solvent resonances. 15N spectra were recorded on a Bruker 400 MHz
spectrometer referenced to an external sample of neat CH3NO2/
CD3NO2. The experiments were run inverse-gated with decoupling
during acquisition (1 s) but not during the relaxation delay (2 s).
Many experiments of 1024 scans covering 500 ppm each were run
until the signal was located; then a longer collection (4096 scans)
using different O 1p and sweep width settings was used to confirm the
signal, and a line broadening of 10 Hz was applied. Chemical shifts are
reported in parts per million and referenced to residual proton
resonances calibrated against external TMS for 1H, 13C, and 29Si.
Infrared spectra were recorded on Jasco 410 spectrophotometers.
Solutions for UV−vis−NIR spectrophotometry were made in a
nitrogen-filled glovebox, and spectra were recorded in a Teflon-tapped
10 mm quartz cell on a Jasco V-670 spectrophotometer. Raman
spectroscopy was performed using a LabRam instrument equipped
with a 50 mW He−Ne laser of wavelength 632.8 nm. [U{N-
(SiMe3)2}3] was synthesized as previously described in the literature85
and sublimed prior to use. ClSiMes3 (Mes = 2,4,6-Me3C6H2) was
synthesized as previously described in the literature,86 and the
subsequent hydrolysis was a larger scale adaption of a literature
reaction57 using ClSiMes3 (3.03 g), KOH (10.3 g), ethanol (150 cm3),
and water (70 cm3), which were heated to reflux overnight. The
reaction mixture was reduced in volume and extracted with n-hexane,
F
dx.doi.org/10.1021/om4003957 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 1. Crystallographic Data for All Compounds
1S
1E
1Evac
2
3
3
HOSi(Mes)3
C27H34OSi
chemical
formula
C162H198N2O6Si6U2 C162H198N2O6Si6U2·6
C
162H198N2O6Si6U2
C45H87N3OSi7U C108H132O4Si4U·0.31
C
108H132O4Si4U
(C6H6)
(C14H10), 3 (C6H14)
formula mass
cryst syst
a/Å
2913.82
3382.47
2913.82
1120.84
2157.46
1844.60
cubic
402.63
triclinic
monoclinic
38.9299(14)
18.1633(3)
29.0426(9)
90.00
monoclinic
29.2851(19)
17.6856(10)
29.4728(17)
90.00
monoclinic
27.4437(7)
22.2208(6)
20.8071(6)
90.00
triclinic
monoclinic
11.0984(5)
12.1998(4)
17.7889(8)
90.00
16.9787(4)
17.1215(4)
17.9145(4)
109.763(2)
92.606(2)
118.253(2)
4182.4(2)
16.1405(4)
17.1050(4)
25.3526(6)
98.417(2)
95.6393(19)
117.137(2)
6052.6(3)
26.9997(1)
26.9997(1)
26.9997(1)
90
b/Å
c/Å
α/deg
β/deg
γ/deg
120.871(5)
90.00
95.943(6)
90.00
95.759(2)
90.00
90
104.370(4)
90.00
90
unit cell
17626.5(13)
15182.6(16)
12624.6(6)
19682.3(2)
2333.23(17)
volume/Å3
temp/K
170(2)
170(2)
C2/c
4
170(2)
P21/n
4
170(2)
C2/c
8
170(2)
120(2)
170(2)
C2/c
4
space group
P1
P1
Pa3
̅
8
̅
̅
no. of formula
units per
1
2
unit cell, Z
absorption
coeff,
2.025
1.932
2.231
2.733
1.430
1.749
0.116
μ/mm−1
no. of reflns
measd
68 021
19 185
76 838
17 993
79 581
15 834
64 114
11 134
113 175
27 717
308 866
8200
8406
1991
no. of indep
reflns
Rint
0.0748
0.0632
0.0810
0.0424
0.2808
0.2573
0.1058
0.0543
0.0633
0.0478
0.1014
0.0445
0.0463
0.0778
final R1 values
(I > 2σ(I))
final wR(F2)
values (I >
2σ(I))
0.1387
0.1051
0.5214
0.1071
0.1044
0.0962
0.1666
final R1 values
(all data)
final wR(F2)
values (all
data)
0.0821
0.1485
0.0789
0.1125
0.3565
0.5687
0.0946
0.1185
0.0699
0.1141
0.0884
0.1209
0.1094
0.1815
which was dried over MgSO4, and the solvent was removed under
reduced pressure. Recrystallization from n-hexane at −30 °C for 16 h
gave colorless crystals (1.683 g, 58%). [(N″)2U(κ2-
CH2SiMe2NSiMe3)] was synthesized from the reaction between
[UI4(OEt2)2] and 4 KN(SiMe3)2 in toluene at 90 °C for 24 h; the
product was obtained by sublimation of the crude solid obtained after
removal of volatiles at 120 °C and 2 × 10−6 Torr.
resulting solution heated to 80 °C under an atmosphere of dinitrogen
for 16 h. Red crystals of product were observed, and the solution was
left to cool to room temperature. The supernatant was removed, and
the product [U{OSi(Mes)3}3]2(μ-η2:η2-N2) (1) washed with n-hexane
(2 cm3) and dried under vacuum (63 mg, 0.022 mmol, 35%). Single
crystals for X-ray diffraction studies were grown from the hot benzene
reaction as it cooled to room temperature (1E). For a solvated sample:
Anal. Calcd for C162H198N2O6Si6U2·6(C6H6): C, 70.31; H, 6.97; N,
0.83. Found: C, 70.13; H, 6.89; N, 0.93.
Syntheses. [U{OSi(Mes)3}3]2(μ-η2:η2-N2), 1. Method A: In n-
Hexane. A dark purple n-hexane solution of U{N(SiMe3)2}3 (305
mg, 0.424 mmol, 1 equiv, 30 cm3) was layered on top of a colorless n-
hexane solution of HOSiMes3 (512 mg, 1.27 mmol, 3 equiv, 30 cm3),
and the mixture stored under an atmosphere of dinitrogen for 16 h,
during which time a dark orange solution formed. Storage of this at
−30 °C for 3 days yielded dark red crystals of [U{OSi(Mes)3}3]2(μ-
η2:η2-N2), 1, and a pale orange solution, which were separated by
filtration and dried under vacuum (115 mg, 0.039 mmol, 18%). These
crystals were suitable for an X-ray diffraction study (1S). Anal. Calcd
for C162H198N2O6Si6U2·2(C6H14): C, 67.72; H, 7.38; N, 0.91. Found:
C, 67.76; H, 7.27; N, 0.85.
Data for [U{OSi(Mes)3}3]2(μ-η2:η2-N2), 1. 1H NMR (400 MHz, 298
K, C7D8): δ (ppm) 9.59 (1 H, s, meta-H), 9.21 (1 H, s, meta-H), 9.03
(1 H, s, meta-H), 7.68 (1 H, s, meta-H), 6.78 (3 H, br s, Me), 5.03 (1
H, s, meta-H), 4.35 (3 H, s, Me), 4.13 (3 H, s, Me), 3.83 (3 H, s, Me),
2.57 (3 H, s, Me), 1.83 (3 H, s, Me), −0.42 (3 H, s, Me), −0.71 (1 H,
s, meta-H), −3.44 (3 H, br s, Me), −17.16 (3 H, br s, Me). 13C NMR
(125.76 MHz, 298 K, C7D8): δ (ppm) 151.21, 150.89, 149.13, 148.83,
147.54, 146.52, 142.51, 140.89, 139.61, 138.73, 136.93, 135.06, 134.38,
133.65, 132.00, 131.38, 129.64, 120.49, 28.61, 25.08, 24.87, 24.22,
22.74, 22.61, 17.74, 1.53, −51.13. 29Si NMR (79.45 MHz, 298 K,
C7D8): δ (ppm) 63.6. Raman (crystalline sample) cm−1: 1282, 1372,
1437 (N2), 1596. IR (nujol mull) cm−1 (intensity): 1604 (s), 1547
(m), 1406 (m), 1260 (w), 1233 (w), 1064 (m), 1029 (m), 957 (w),
876 (m), 846 (m), 808 (s), 638 (s) . UV−vis−NIR (0.00083 M
toluene solution) λ, nm (ε, M−1 cm−1): 673 (210), 809 (62), 897 (53),
1087 (50), 1153 (51), 1174 (50), 1292 (38), 1367 (32), 1820 (54),
2041 (110).
Method B: In n-Hexane Solvent at 70 °C. An ampule was charged
with U{N(SiMe3)2}3 (52 mg, 0.073 mmol, 1 equiv), HOSi(Mes)3 (88
mg, 0.22 mmol, 3 equiv), and n-hexane (2 cm3), and the resulting
solution heated to 70 °C under an atmosphere of dinitrogen for 16 h.
Red microcrystalline product was observed, and the solution was left
to cool to room temperature. The pale yellow supernatant was
removed, and the product [U{OSi(Mes)3}3]2(μ-η2:η2-N2) (1) dried
1
under vacuum (44 mg, 0.015 mmol, 42%). H NMR spectroscopy
Freeze−pump−thaw degassing had no effect on the 1H NMR
spectroscopic resonances for the dinitrogen compound 1. Reactions of
UN″3 with 3 equivalents of HOSi(Mes)3 performed under argon
revealed a small amount of an impurity, indicating the need for careful
crystallization to separate the desired complex from byproducts.
Method C: In Benzene Solvent at 80 °C. An ampule was charged
with U{N(SiMe3)2}3 (90 mg, 0.125 mmol, 1 equiv), HOSi(Mes)3
(152 mg, 0.376 mmol, 3 equiv), and benzene (2 cm3), and the
1
revealed no H NMR spectroscopic resonances for the N2 compound
1, but resonances for U{OSi(Mes)3}N″3 2 and for U{OSi(Mes)3}4 3
were observed, which increased when heated to 90 °C overnight.
G
dx.doi.org/10.1021/om4003957 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
15N2-Labeled [U{OSi(Mes)3}3]2(μ-η2:η2-15N2): 1-15N2. An ampule was
charged with U{N(SiMe3)2}3 (83 mg, 0.115 mmol, 1 equiv) and
HOSi(Mes)3 (139 mg, 0.345 mmol, 3 equiv) and placed under
vacuum. Benzene (2 cm3) was degassed by the freeze−pump−thaw
method three times and vacuum transferred onto the solids. 15N2
(98.3%, Cambridge isotopes) was then added to the still-frozen
solution, the ampule was sealed, and the resulting solution was heated
to 80 °C for 16 h, during which time red crystals of product were
observed to form. The solution was allowed to cool to room
temperature, the supernatant was removed, and the red crystalline
product [U{OSi(Mes)3}3]2(μ-η2:η2-15N2) (1-15N2) was washed with n-
hexane (2 cm3) and dried under vacuum (62 mg, 0.021 mmol, 36%).
15N NMR (40.52 MHz, 298 K, C7D8): δ (ppm) 4213.5. Raman
(crystalline sample) cm−1: 1282, 1372 (overlapped 15N2 and ligand),
1601.
arene compound N″2U(μ-C6H6)UN″2 in which anthracene had been
added.4
Powder diffraction experiments were carried out on samples flame-
sealed in capillaries using a Bruker D8 Advance diffractometer in
transmittance mode employing Cu radiation from 5° to 40° 2θ in
0.02° increments at room temperature.
ASSOCIATED CONTENT
* Supporting Information
■
S
Additional spectroscopic data and full crystallographic data (cif
format). This material is available free of charge via the Internet
at http://pubs.acs.org. Codes for access to the full crystallo-
graphic data through the CCDC are in Table 1.
U{OSi(Mes)3}N″3 2. [(N″)2U(κ2-CH2SiMe2NSiMe3)] (168 mg,
0.233 mmol, 1 equiv) and HOSi(Mes)3 (94 mg, 0.233, 1 equiv)
were dissolved in toluene (5 cm3), and the solution was stored at room
temperature for 16 h. All of the volatiles were removed under reduced
pressure from the pale yellow solution, and the residues were extracted
with n-hexane (10 cm3). The solution was filtered, reduced in volume
to ca. 5 cm3, and stored at −30 °C for 16 h to afford colorless crystals
of U{OSi(Mes)3}N″3 2, which were isolated by removal of the
supernatant, washing once at −30 °C with n-hexane, and drying under
AUTHOR INFORMATION
Corresponding Author
*E-mail: polly.arnold@ed.ac.uk.
■
Author Contributions
The project was conceived and carried out and the manuscript
written by P.L.A., A.I.G., S.M.M., and J.H.F. All authors have
given approval to the final version of the manuscript.
Notes
1
vacuum (105 mg, 0.094 mmol, 40%). H NMR (400 MHz, 333 K,
The authors declare no competing financial interest.
C7D8): δ (ppm) 12.32 (s, 3 H, Mes-Me), 11.42 (s, 1 H, meta-H), 7.28
(s, 3 H, Mes-Me), 6.68 (s, 1 H, meta-H), 5.51 (vbr s, N(SiMe3)
(SiMe3)), 2.71 (s, 3 H, Mes-Me), −16.34 (vbr s, N(SiMe3) (SiMe3)).
Anal. Calcd for C45H87N3OSi7U: C, 48.22; H, 7.82; N, 3.75. Found: C,
48.45; H, 7.92; N, 3.67.
ACKNOWLEDGMENTS
■
We thank the UK EPSRC, EaStCHEM, and the University of
Edinburgh for funding, Mr. Steven Hunter for assistance with
Raman spectroscopy, and Dr. Mairi Haddow (The University
of Bristol) for help with modeling disorder (the structure of
HOSi(Mes)3) and refinement of diffraction data (crystals of
1Evac).
U{OSi(Mes)3}4 3. [(N″)2U(κ2-CH2SiMe2NSiMe3) ] (61 mg, 0.086
mmol, 1 equiv) and HOSi(Mes)3 (138 mg, 0.340 mmol, 4 equiv) were
dissolved in toluene (5 cm3), and the solution was heated at 100 °C
for 16 h. All of the volatiles were removed under reduced pressure
from the pale orange solution, n-pentane (5 cm3) was added, and the
resultant suspension was sonicated for 10 min. The yellow solution
was filtered, and the solids were washed with a further two times with
n-pentane (5 cm3) and dried under vacuum to yield 3 as a very pale
REFERENCES
■
(1) Arnold, P. L. Chem. Commun. 2011, 47, 9005.
(2) Evans, W. J.; Kozimor, S. A. Coord. Chem. Rev. 2006, 250, 911.
(3) Gardner, B. M.; Liddle, S. T. Eur. J. Inorg. Chem. 2013.
(4) Arnold, P. L.; Mansell, S. M.; Maron, L.; McKay, D. Nat. Chem.
2012, 4, 668.
1
pink solid (31 mg, 0.017 mmol, 22%). H NMR (400 MHz, 298 K,
C6D6): δ (ppm) 6.66 (1 H, s, meta-H), 5.98 (1 H, s, meta-H), 2.42 (3
H, s, Me), 2.09 (3 H, s, Me), −0.97 (3 H, br s, Me). EI-MS m/z =
1844 (100%, M). Anal. Calcd for C108H132O4Si4U: C, 70.32; H, 7.21.
Found: C, 57.81; H, 6.03.
Crystallographic Details. Diffraction experiments on these
samples were carried out on single crystals covered in inert oil cooled
to 170 K on an Oxford diffraction Excalibur four-circle diffractometer
(5) Summerscales, O. T.; Cloke, F. G. N.; Hitchcock, P. B.; Green, J.
C.; Hazari, N. Science 2006, 311, 829.
(6) Arnold, P. L.; Turner, Z. R.; Bellabarba, R. M.; Tooze, R. P.
Chem. Sci. 2011, 2, 77.
(7) Lam, O. P.; Meyer, K. Polyhedron 2012, 32, 1.
(8) Fryzuk, M. D. Chem. Commun. 2013, 49, 4866.
(9) MacKay, B. A.; Fryzuk, M. D. Chem. Rev. 2004, 104, 385.
(10) Allen, A. D.; Senoff, C. V. Chem. Commun. 1965, 621.
(11) Senoff, C. V. J. Chem. Educ. 1990, 67, 368.
(12) Yandulov, D. V.; Schrock, R. R. Science 2003, 301, 76.
(13) Schrock, R. R. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 17087.
(14) Schrock, R. R. Nat. Chem. 2011, 3, 95.
employing Mo Kα radiation (λ = 0.71073 Å) or at 120 K; for 3 in Pa3,
̅
on an Oxford diffraction SuperNova employing Mo Kα radiation. The
structures were solved by direct methods (XS) and refined by least-
squares on weighted F2 values for all reflections (XL).87 All hydrogen
atoms were constrained to ideal geometries and refined with fixed
isotropic displacement parameters except the O−H in the structure of
HOSiMes3, whose position was freely refined with a fixed isotropic
displacement parameter. The crystals of 1E that were placed under
vacuum did not diffract at high angle, and generally the data collected
were poor, so only data out to 44° 2θ were included in the final
refinement, which allowed the anisotropic refinement of only the U
and Si atoms. Extensive use of restraints and constraints was required
to model the OSi(Mes)3 ligands, and only basic connectivity could be
established. Refinement proceeded to give the residuals shown in
Table 1. Complex neutral-atom scattering factors were used.
(15) Arashiba, K.; Miyake, Y.; Nishibayashi, Y. Nat. Chem. 2011, 3,
120.
(16) Nishibayashi, Y. Dalton Trans. 2012, 41, 7447.
(17) Hellman, A.; Baerends, E. J.; Biczysko, M.; Bligaard, T.;
Christensen, C. H.; Clary, D. C.; Dahl, S.; van Harrevelt, R.; Honkala,
K.; Jonsson, H.; Kroes, G. J.; Luppi, M.; Manthe, U.; Norskov, J. K.;
Olsen, R. A.; Rossmeisl, J.; Skulason, E.; Tautermann, C. S.; Varandas,
A. J. C.; Vincent, J. K. J. Phys. Chem. B 2006, 110, 17719.
(18) Howard, J. B.; Rees, D. C. Chem. Rev. 1996, 96, 2965.
(19) Hoffman, B. M.; Dean, D. R.; Seefeldt, L. C. Acc. Chem. Res.
2009, 42, 609.
(20) Rodriguez, M. M.; Bill, E.; Brennessel, W. W.; Holland, P. L.
Science 2011, 334, 780.
(21) Fox, A. R.; Bart, S. C.; Meyer, K.; Cummins, C. C. Nature 2008,
455, 341.
Disordered hexane solvent in 1S, 2, and 3 in Pa3 was modeled
̅
using the SQUEEZE subroutine in the PLATON software suite. 1E is
a benzene solvate, and the N2 ligand was disordered over two positions
90° apart. The structure of HOSi(Mes)3 (see SI) showed the entire
molecule to be disordered over two positions. The structure of
compound 3 was also determined as its n-hexane solvate, which also
contained 0.31 equivalent of anthracene cocrystallized from an
impurity in the starting material. The metallacycle used in this
reaction was formed as a byproduct in the synthesis of the bridging-
(22) Gardiner, M. G. Materials 2010, 3, 841.
H
dx.doi.org/10.1021/om4003957 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(23) Evans, W. J.; Lee, D. S. Can. J. Chem. 2005, 83, 375.
(24) Evans, W. J.; Ulibarri, T. A.; Ziller, J. W. J. Am. Chem. Soc. 1988,
110, 6877.
(25) Evans, W. J.; Allen, N. T.; Ziller, J. W. J. Am. Chem. Soc. 2001,
123, 7927.
(26) Evans, W. J.; Allen, N. T.; Ziller, J. W. Angew. Chem., Int. Ed.
2002, 41, 359.
(27) Evans, W. J.; Zucchi, G.; Ziller, J. W. J. Am. Chem. Soc. 2003,
(60) Evans, W. J.; Rego, D. B.; Ziller, J. W. Inorg. Chem. 2006, 45,
10790.
(61) Porchia, M.; Brianese, N.; Casellato, U.; Ossola, F.; Rossetto, G.;
Zanella, P.; Graziani, R. J. Chem. Soc., Dalton Trans. 1989, 0, 677.
(62) Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.;
McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.; Taylor, R.; van de
Streek, J.; Wood, P. A. J. Appl. Crystallogr. 2008, 41, 466.
(63) McGeary, M. J.; Coan, P. S.; Folting, K.; Streib, W. E.; Caulton,
K. G. Inorg. Chem. 1991, 30, 1723.
(64) Brown, J. L.; Mokhtarzadeh, C. C.; Lever, J. M.; Wu, G.;
Hayton, T. W. Inorg. Chem. 2011, 50, 5105.
(65) Van der Sluys, W. G.; Burns, C. J.; Huffman, J. C.; Sattelberger,
A. P. J. Am. Chem. Soc. 1988, 110, 5924.
125, 10.
(28) Bochkarev, M. N.; Fedushkin, I. L.; Fagin, A. A.; Petrovskaya, T.
V.; Ziller, J. W.; Broomhall-Dillard, R. N. R.; Evans, W. J. Angew.
Chem., Int. Ed. Engl. 1997, 36, 133.
(29) Bochkarev, M. N.; Fagin, A. A. Chem.Eur. J. 1999, 5, 2990.
(30) Evans, W. J.; Allen, N. T.; Ziller, J. W. J. Am. Chem. Soc. 2000,
122, 11749.
(66) Berg, J. M.; Clark, D. L.; Huffman, J. C.; Morris, D. E.;
Sattelberger, A. P.; Streib, W. E.; Vandersluys, W. G.; Watkin, J. G. J.
Am. Chem. Soc. 1992, 114, 10811.
(31) Bochkarev, M. N.; Fedushkin, I. L.; Dechert, S.; Fagin, A. A.;
Schumann, H. Angew. Chem., Int. Ed. 2001, 40, 3176.
(32) Evans, W. J.; Lee, D. S.; Rego, D. B.; Perotti, J. M.; Kozimor, S.
A.; Moore, E. K.; Ziller, J. W. J. Am. Chem. Soc. 2004, 126, 14574.
(33) Jaroschik, F.; Momin, A.; Nief, F.; Le Goff, X. F.; Deacon, G. B.;
Junk, P. C. Angew. Chem., Int. Ed. 2009, 48, 1117.
(34) Demir, S.; Lorenz, S. E.; Fang, M.; Furche, F.; Meyer, G.; Ziller,
J. W.; Evans, W. J. J. Am. Chem. Soc. 2010, 132, 11151.
(35) Evans, W. J.; Fang, M.; Zucchi, G.; Furche, F.; Ziller, J. W.;
Hoekstra, R. M.; Zink, J. I. J. Am. Chem. Soc. 2009, 131, 11195.
(36) Fang, M.; Bates, J. E.; Lorenz, S. E.; Lee, D. S.; Rego, D. B.;
Ziller, J. W.; Furche, F.; Evans, W. J. Inorg. Chem. 2011, 50, 1459.
(37) Fieser, M. E.; Bates, J. E.; Ziller, J. W.; Furche, F.; Evans, W. J. J.
Am. Chem. Soc. 2013, 135, 3804.
(38) Roussel, P.; Scott, P. J. Am. Chem. Soc. 1998, 120, 1070.
(39) Roussel, P.; Errington, W.; Kaltsoyannis, N.; Scott, P. J.
Organomet. Chem. 2001, 635, 69.
(40) Cloke, F. G. N.; Hitchcock, P. B. J. Am. Chem. Soc. 2002, 124,
9352.
(41) Evans, W. J.; Kozimor, S. A.; Ziller, J. W. J. Am. Chem. Soc. 2003,
125, 14264.
(42) Odom, A. L.; Arnold, P. L.; Cummins, C. C. J. Am. Chem. Soc.
(67) Bursten, B. E.; Casarin, M.; Ellis, D. E.; Fragala, I.; Marks, T. J.
Inorg. Chem. 1986, 25, 1257.
(68) Cayton, R. H.; Chisholm, M. H.; Davidson, E. R.; DiStasi, V. F.;
Du, P.; Huffman, J. C. Inorg. Chem. 1991, 30, 1020.
(69) Russo, M. R.; Kaltsoyannis, N.; Sella, A. Chem. Commun. 2002,
2458.
(70) Clavier, H.; Nolan, S. P. Chem. Commun. 2010, 46, 841.
(71) Poater, A.; Cosenza, B.; Correa, A.; Giudice, S.; Ragone, F.;
Scarano, V.; Cavallo, L. Eur. J. Inorg. Chem. 2009, 2009, 1759.
(72) Mansell, S. M.; Perandones, B. F.; Arnold, P. L. J. Organomet.
Chem. 2010, 695, 2814.
(73) Source: WebElements [http://www.webelements.com/].
(74) Kaltsoyannis, N.; Scott, P. Chem. Commun. 1998, 1665.
(75) Cloke, F. G. N.; Green, J. C.; Kaltsoyannis, N. Organometallics
2004, 23, 832.
(76) Hirshfeld, F. L. Acta Crystallogr., Sect. A 1976, 32, 239.
(77) Braga, D.; Koetzle, T. F. Acta Crystallogr., Sect. B 1988, 44, 151.
(78) Van der Sluys, W. G.; Burns, C. J.; Huffman, J. C.; Sattelberger,
A. P. J. Am. Chem. Soc. 1988, 110, 5924.
(79) Nakai, H.; Hu, X. L.; Zakharov, L. N.; Rheingold, A. L.; Meyer,
K. Inorg. Chem. 2004, 43, 855.
(80) Turner, Z. R.; Bellabarba, R.; Tooze, R. P.; Arnold, P. L. J. Am.
Chem. Soc. 2010, 132, 4050.
1998, 120, 5836.
(43) Mansell, S. M.; Kaltsoyannis, N.; Arnold, P. L. J. Am. Chem. Soc.
2011, 133, 9036.
(81) Arnold, P. L.; Blake, A. J.; Hall, S. N.; Ward, B. D.; Wilson, C.
Dalton Trans. 2001, 488.
(44) Krempner, C. Eur. J. Inorg. Chem. 2011, 1689.
(45) Marshak, M. P.; Nocera, D. G. Inorg. Chem. 2013, 52, 1173.
(46) Hulley, E. B.; Bonanno, J. B.; Wolczanski, P. T.; Cundari, T. R.;
Lobkovsky, E. B. Inorg. Chem. 2010, 49, 8524.
(47) Wolczanski, P. T. Polyhedron 1995, 14, 3335.
(48) Kuiper, D. S.; Wolczanski, P. T.; Lobkovsky, E. B.; Cundari, T.
R. J. Am. Chem. Soc. 2008, 130, 12931.
(49) Wolczanski, P. T. Chem. Commun. 2009, 740.
(50) Elvidge, B. R.; Arndt, S.; Spaniol, T. P.; Okuda, J. Dalton Trans.
2006, 0, 890.
(82) Lorenz, V.; Fischer, A.; Jacob, K.; Bruser, W.; Gelbrich, T.;
Jones, P. G.; Edelmann, F. T. Chem. Commun. 1998, 2217.
(83) Lorenz, V.; Fischer, A.; Jacob, K.; Bruser, W.; Edelmann, F. T.
Chem.Eur. J. 2001, 7, 848.
(84) Evans, W. J.; Ulibarri, T. A.; Ziller, J. W. Organometallics 1991,
10, 134.
(85) Clark, D. L.; Sattelberger, A. P.; Andersen, R. A. Inorg. Synth.
1997, 31, 307.
(86) Zigler, S. S.; Johnson, L. M.; West, R. J. Organomet. Chem. 1988,
341, 187.
(87) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112.
(51) Nishiura, M.; Hou, Z. M.; Wakatsuki, Y. Organometallics 2004,
23, 1359.
(52) Fischbach, A.; Eickerling, G.; Scherer, W.; Herdtweck, E.;
Anwander, R. Z. Naturforsch., B: Chem. Sci. 2004, 59, 1353.
(53) Evans, W. J.; Golden, R. E.; Ziller, J. W. Inorg. Chem. 1991, 30,
4963.
(54) Duncalf, D. J.; Hitchcock, P. B.; Lawless, G. A. J. Organomet.
Chem. 1996, 506, 347.
(55) Shao, P. C.; Berg, D. J.; Bushnell, G. W. Inorg. Chem. 1994, 33,
3452.
́ ́
(56) Mougel, V.; Camp, C.; Pecaut, J.; Coperet, C.; Maron, L.;
Kefalidis, C. E.; Mazzanti, M. Angew. Chem., Int. Ed. 2012, 51, 12280−
12284.
(57) Wrobel, O.; Schaper, F.; Brintzinger, H. H. Organometallics
2004, 23, 900.
(58) Andersen, R. A. Inorg. Chem. 1979, 18, 1507.
(59) Simpson, S. J.; Turner, H. W.; Andersen, R. A. Inorg. Chem.
1981, 20, 2991.
I
dx.doi.org/10.1021/om4003957 | Organometallics XXXX, XXX, XXX−XXX