UIII-CN versus UIV-NC coordination in tris(silylamide) complexes

Article abstract of DOI:10.1021/acs.inorgchem.5b00034

Treatment of the metallacycle [UN?₂(N,C)] [N? = N(SiMe₃)₂; N,C = CH₂SiMe₂N(SiMe₃)] with [HNEt₃][BPh₄], [HNEt₃]Cl, and [pyH][OTf] (OTf = OSO₂CF

Full text of DOI:10.1021/acs.inorgchem.5b00034

Article
pubs.acs.org/IC
UIII−CN versus UIV−NC Coordination in Tris(silylamide) Complexes
Alexandre Herve,† Yamina Bouzidi,‡ Jean-Claude Berthet,*,† Lotfi Belkhiri,*,‡ Pierre Thuery,†
́
́
Abdou Boucekkine,*,§ and Michel Ephritikhine*,†
†CEA, IRAMIS, UMR 3685 NIMBE, CEA/CNRS NIMBE, CEA/Saclay, 91191 Gif-sur-Yvette, France
‡URCHEMS, Dep
́ ́
artement de chimie, Universite Constantine 1, 25000 Constantine, Algeria
§Institut des Sciences Chimiques de Rennes, UMR 6226 CNRS - Universite
France
́
de Rennes 1, Campus de Beaulieu, 35042 Rennes Cedex,
S
* Supporting Information
ABSTRACT: Treatment of the metallacycle [UN*2(N,C)] [N* = N(SiMe3)2; N,C = CH2SiMe2N(SiMe3)] with
[HNEt3][BPh4], [HNEt3]Cl, and [pyH][OTf] (OTf = OSO2CF3) gave the cationic compound [UN*3][BPh4] (1) and the
neutral complexes [UN*3X] [X = Cl (3), OTf (4)], respectively. The dinuclear complex [{UN*(μ-N,C)(μ-OTf)}2] (5) and its
tetrahydrofuran (THF) adduct [{UN*(N,C)(THF)(μ-OTf)}2] (6) were obtained by thermal decomposition of 4. The
successive addition of NEt4CN or KCN to 1 led to the formation of the cyanido-bridged dinuclear compound [(UN*3)2(μ-
CN)][BPh4] (7) and the mononuclear mono- and bis(cyanide) complexes [UN*3(CN)] (2) and [M][UN*3(CN)2] [M = NEt4
(8), K(THF)4 (9)], while crystals of [K(18-crown-6)][UN*3(CN)2] (10) were obtained by the oxidation of [K(18-crown-
6)][UN*3(CN)] with pyridine N-oxide. The THF adduct of 1, [UN*3(THF)][BPh4], and complexes 2−7, 9 and 10 were
characterized by their X-ray crystal structure. In contrast to their UIII analogues [NMe4][UN*3(CN)] and [K(18-crown-
6)]2[UN*3(CN)2] in which the CN anions are coordinated to the metal center via the C atom, complexes 2 and 9 exhibit the
isocyanide U−NC coordination mode of the cyanide ligand. This UIII/UIV differentiation has been analyzed using density
functional theory calculations. The observed preferential coordinations are well explained considering the electronic structures of
the different species and metal−ligand bonding energies. A comparison of the different quantum descriptors, i.e., bond orders,
NPA/QTAIM data, and energy decomposition analysis, has allowed highlighting of the subtle balance between covalent, ionic,
and steric factors that govern the U−CN/NC bonding.
prepared the corresponding UIV compounds [UN*3(CN)] (2),
INTRODUCTION
■
[(UN*3)2(μ-CN)][BPh4] (7), and [M][UN*3(CN)2] [M =
NEt4 (8), K(THF)4 (9)], thus extending the series of [UN*3X]
derivatives, which constitute one of the most popular families of
complexes in uranium chemistry.2 These compounds could be
synthesized either by the oxidation of [UN*3] and the
aforementioned uranium(III) cyanide complexes or by the
addition of MCN to the novel cationic UIV precursor
[UN*3][BPh4] (1). However, the main objective of this
study was, from solid-state structural comparisons on uranium-
(III) and uranium(IV) cyanide counterparts, to demonstrate
the influence of the electronic configuration, 5f3 or 5f2, on the
ligation mode of the CN ligand. The X-ray crystal structures
revealed the preferential coordination of the cyanide (−CN)
and isocyanide (CN−) ions in the uranium(III) and uranium-
(IV) tris-N* complexes, respectively, and this difference was
Recently, we reported on the series of uranium(III) cyanide
complexes [M][(UN*3)2(μ-CN)], [M][UN*3(CN)],
[M]2[UN*3(CN)2], and [M]2[UN*2(CN)3] [M = K(18-
crown-6) or NR4; N* = N(SiMe3)2], which have been
n
synthesized by the addition of NR4CN (R = Me, Et, Bu) or
KCN in the presence of 18-crown-6 to the tris(silylamide)
precursor [UN*3].1 Structural comparisons of these com-
pounds with the CeIII analogues revealed the distinct
coordination modes of the CN group, through the C or N
atom to the U or Ce metal center, respectively, and this
differentiation was related to the better energy matching
between the 6d/5f U and ligand orbitals, leading to a
nonnegligible covalent character of the U−CN bond. In
order to extend the variety of soluble molecular cyanide
complexes of the actinides and lanthanides, in different
oxidation states, which could serve as valuable building blocks
for the synthesis of novel clusters and coordination polymers
with interesting physicochemical properties, we have then
Received: January 6, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Synthesis of the Complexes
analyzed using relativistic density functional theory (DFT)
calculations.
N,C metallacycle, which was already noted in the insertion
reactions of small molecules into the U−C bond.5,9
Importantly, the reaction did not proceed in toluene and the
reaction time in THF must be limited (not to exceed 30 min)
because of degradation of the cation and the formation of
intractable products. Nevertheless, pale-green crystals of the
THF adduct [UN*3(THF)][BPh4] [1(THF)] were formed by
the slow diffusion of diethyl ether (Et2O) into a THF solution
of 1, and this adduct was converted back to 1 upon exposure to
a vacuum. The 1H NMR spectrum of 1 in benzene-d6 shows, in
addition to the N* signal integrating for 54 H at δ −6.33, three
phenyl resonances in the intensity ratio of 4:8:8 at δ 2.91, 2.82,
and 1.60, suggesting that 1 adopts a zwitterionic structure in the
solid state or in noncoordinating solvents, with phenyl groups
of BPh4 coordinated to the U atom, as previously observed with
[U(NEt2)3][BPh4]10 and [U(Cp*)2][BPh4].11 The treatment
of 1 with 0.9 mol equiv of NEt4CN in THF led to formation of
the pale-pink powder 2, which was obtained in 75% yield after
extraction in Et2O. Pink crystals of 2 were formed by
crystallization from THF.
The metallacycle [UN*2(N,C)] was also useful for the
synthesis of [UN*3X] derivatives, and its protonolysis reactions
with 1 mol equiv of [HNEt3]Cl and [pyH][OTf] (OTf =
OSO2CF3) gave the chloride and triflate complexes 3 and
[UN*3(OTf)] (4) in good yield. Complex 4 was also obtained
by the oxidation of [UN*3] with Ph3COTf, but it was difficult
to separate it from Gomberg’s dimer (Ph3CCH(C4H4)C
CPh2), which was formed as a soluble byproduct.12 The same
difficulty was encountered in the synthesis of 3 by the oxidation
of [UN*3] with Ph3CCl.13 The pyridinium triflate was already
used for the synthesis of triflate complexes by protonolysis of
amide or alkyl precursors and, in particular, [U(OTf)4(py)] was
obtained by treating [UN*2(N,C)] with 4 mol equiv of
[pyH][OTf].14 The chloride 3 was previously synthesized by
the reaction of UCl4 with NaN*,15 the treatment of
[UN*2(N,C)] by HCl,9 and the oxidation of [UN*3] with
CuCl3 or Ph3CCl,13 but its crystal structure has not been
RESULTS AND DISCUSSION
■
The syntheses of the complexes are summarized in Scheme 1.
[UN*3X] Complexes (X = Cl, OSO2CF3, CN). Taking as an
example the reaction of [UN*3] with CuCl, which afforded
[UN*3(Cl)] (3) in good yield,3 our first attempts at the
preparation of the mono(cyanide) complex 2 consisted of the
oxidation of [UN*3] with CuCN. The reaction was performed
in benzene and unexpectedly gave the metallacycle
[UN*2(N,C)] [N,C = CH2SiMe2N(SiMe3)] as the sole
uranium compound.4−6 Using other oxidizing reagents
[AgCN, RCN (R = Me, Ph, Me3Si), benzoquinone, Me3NO]
and/or [M][UN*3(CN)] [M = NR4 or K(18-crown-6)] as the
starting product gave unidentified compounds. In one reaction
of [K(18-crown-6)][UN*3(CN)] and pyridine N-oxide, pink
crystals of the bis(cyanide) complex [K(18-crown-6)]-
[UN*3(CN)2] (10) were isolated and characterized by X-ray
diffraction analysis (vide infra). The synthesis of 2 from the UIV
precursor 3 was then considered, but in contrast to
[U(Cp*)2UX2] (Cp* = η-C5Me5; X = I, OSO2CF3), which
reacted with KCN or NEt4CN to give [U(Cp*)2(CN)2],7 3 did
not undergo straightforward substitution with cyanide ligands
but was transformed into the addition product [NEt4]-
[UN*3(Cl)(CN)].
These unsuccessful attempts led us to consider the synthesis
of 2 by the simple addition of MCN to the cationic precursor 1,
with the easy elimination of the MBPh4 salt. While attempts to
prepare in THF the UIV cation [UN*3]+ by one-electron
oxidation of [UN*3] led to the rapid catalytic polymerization of
the solvent and the formation of [UN*4] in low yield,8 complex
1 was readily obtained by treating the metallacycle
[UN*2(N,C)] with 1 mol equiv of [HNEt3][BPh4] in THF
and, after evaporation of the solvent, isolated as a beige powder
in 86% yield. The selectivity of the protonolysis reaction is
related to the larger nucleophilic character of the C atom of the
B
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 2. Thermal Evolution of Complex 4
determined. Brown crystals of 3 and pinkish crystals of 4
suitable for X-ray diffraction analysis were formed at −35 °C by
crystallization from pentane and a pentane−THF mixture,
respectively. Complex 4 was converted in refluxing THF to the
metallacycle [{UN*(μ-N,C)(μ-OTf)}2] (5), isolated as brown
crystals in 42% yield after crystallization from Et2O, and yellow
platelets of the THF adduct [{UN*(N,C)(THF)(μ-OTf)}2]
(6) crystallized from this solvent. The synthesis of 5 could
occur via the concerted elimination of HN*, which proceeds
under thermal conditions (Scheme 2).4,8
The distinct ligation modes of the U(N,C) metallacycles,
bridging in 5 and terminal in 6, are clearly differentiated by the
1H NMR signals of the methylene groups, which are visible at δ
−520.1 and −228.7, respectively.
Views of one of the two independent and very similar cations
of 1(THF) and of the cyanide, chloride, and triflate compounds
2−4 are presented in Figures 1−4, respectively, and selected
bond lengths and angles are listed in Table 1. The complexes
adopt a pseudotetrahedral structure with the exception of the
cyanide 2, which is in a trigonal-pyramidal configuration, with
the U1 and N atoms of the N* ligands being almost coplanar,
as shown by the sum of the N−U1−N angles equal to 358.5°.
The U1 and U2 atoms of the cations [UN*3(THF)]+ in
1(THF), which are at a distance of 0.3843(12) and 0.5250(12)
Å from the plane of the three N atoms, exhibit γ U···H−C
agostic-type interactions [U1···C1 = 2.892(3) Å and U2···C23
= 3.050(3) Å, with C1 and C23 being in the trans position with
respect to the O atom of the THF ligand]. The distance
between the metal center and the plane of the amide N atoms
in complexes [UN*3X] increases with X in the order F
[0.1158(9) Å]16 < CN [0.157(4) Å] < Cl [0.406(2) Å] < I
Figure 2. View of complex 2. H atoms are omitted. Displacement
ellipsoids are drawn at the 50% probability level. Symmetry codes: i, 2
− y, x − y + 1, z; j, y − x + 1, 2 − x, z.
Figure 3. View of complex 3. H atoms are omitted. Displacement
ellipsoids are drawn at the 50% probability level. Symmetry codes: i, 1
− y, x − y + 1, z; j, y − x, 1 − x, z.
(0.456 Å)17 < OTf [0.6574(12) Å], reflecting the increase of
the steric interactions between X and the N* ligands with the
size of X. Compounds 2 and 3, which crystallize in the trigonal
system, are isomorphous with [UN*3I],17 [MN*3Cl] (M =
Pb,18 Ti, Zr, Hf19), and [CeN*3X] (X = Cl, Br,18 I20). The U−
N(N*) distances, with a mean value of 2.23(4) Å for all of the
complexes, are identical with that of 2.238(4) Å in [UN*3I],17
very similar to the Ce−N(N*) distances in [CeN*3X]
complexes [2.22(1) Å], and are ca. 0.13 Å smaller than the
U−N(N*) distances in the UIII analogues, 2.362(3) Å in the
carbene complex [UN*3(C{NMeCMe}2)]21 and 2.350(6) Å in
Figure 1. View of one of the two independent cations in 1(THF). H
atoms are omitted. The agostic bond is shown as a dashed line.
Displacement ellipsoids are drawn at the 30% probability level.
C
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
its UIII analogue has been analyzed by DFT calculations (vide
infra).
Views of the metallacycles 5 and 6 are shown in Figures 5
and 6, respectively, and selected bond lengths and angles are
listed in Table 2. In both centrosymmetric complexes, the metal
centers are bridged by two triflate ligands, with mean U1−O
distances of 2.425(9) and 2.526(2) Å in 5 and 6, respectively.
These values can be compared to those measured in the
binuclear complexes [{U(OTf)2(py)2(μ-OTf)}2(μ-O)]
[2.394(4) Å],25 [{U(C8H8)(Cp*)(μ-OTf)}2] [2.50(1) Å],26
and [{U(Cp*)2(Me)(μ-OTf)}2]27 and [{U(Cp*)(C5Me4H)-
(μ-OTf)}2(μ-O)] [2.51(2) Å].28 However, the U1···U1i
distance of 3.8146(3) Å in 5 is much smaller than that of
6.4491(7) Å in 6, a shortening that is likely due to the bridging
position of the CH2 group of the N,C metallacycle. Such M2(μ-
N,C) linkages were previously encountered in polynuclear
complexes, resulting from deprotonation of [UN*2(Cl)2],29
such as [U{(μ-N,C)]2[μ-Li(DME)}]2, where the U1···U1i
distance is equal to 3.6161(12) Å, and in the binuclear
compounds [{YN*(THF)(μ-N,C)}2],30 [{MN*(NMe2)(μ-
N,C)}2] (M = Zr, Hf),31 [{VN*(μ-N,C)}2],32 and [{Cr(C5H5)-
(μ-N,C)}2].33 The atoms U1, C1, U1i, and C1i are coplanar and
form a quite perfect square [C1−U1−C1i = 87.15(10)°], which
is almost orthogonal, with a dihedral angle of 86.96(9)°, to the
plane defined by the metal centers and the O1 and O2 atoms of
the bridging triflate ligands. The U−C distances of the bridging
metallacycles in 5, which average 2.63(4) Å, are expectedly
larger than that of 2.400(8) Å of the terminal U(N,C) moiety in
6, which is itself at the lower limit of the range of U−C bond
lengths for such metallacycles (typically equal to 2.50 Å).5 The
U−N distances, which vary from 2.190(5) to 2.256(6) Å, and
the U−O(THF) distance of 2.468(5) Å are unexceptional.
Cyanido-Bridged and Bis(cyanide) Complexes
[(UN*3)2(μ-CN)] and [M][UN*3(CN)2]. As in the case of the
mono(cyanide) [M][UN*3(CN)], oxidation of the UIII
complexes [M][(UN*3)2(μ-CN)] and [M]2[UN*3(CN)2] [M
= NR4 or K(18-crown-6)] with a variety of oxidizing reagents
(CuCl, CuI, AgI, AgOTf, TlBPh4, C5H5NO) did not permit
isolation of a pure product. Here again, the cyanido-bridged
and bis(cyanide) UIV complexes were readily synthesized by the
addition of MCN to 1. Thus, treatment of 1 with 0.45 mol
equiv of NEt4CN in THF afforded after the usual workup an
off-white powder of 7 in 65% yield, which was recrystallized as
the yellow solvate 7·0.5Et2O from Et2O. A similar reaction with
2 mol equiv of NEt4CN gave the bis(cyanide) [NEt4]-
Figure 4. View of complex 4. H atoms are omitted. Displacement
ellipsoids are drawn at the 30% probability level.
[NMe4][UN*3(CN)],1 in agreement with the variations of the
ionic radii of the metal centers.22 However, the striking feature
of the structure of 2 is coordination of the CN group through
the N atom, in contrast to the cyanide ligation mode observed
in the UIII counterpart [NMe4][UN*3(CN)]. The U1−N(NC)
bond with a length of 2.378(9) Å is short by comparison with
the U−C(CN) distances in the anionic UIII analogue
[2.455(15) and 2.604(14) Å] and the UIV complex [NEt4]-
[UN*2(N,C)(CN)] [2.559(4) Å].6 To the best of our
knowledge, this U−NC linkage is unprecedented in UIV
compounds, but it is noteworthy that the X-ray crystal structure
23
t
of the ThIV complex [Th(C5 Bu3H2)2(NC)(OSiMe3)]
revealed the Th−NC bonding of the CN ligand while the
U−CN linkage was characterized in the UIV analogue
24
t
[U(C5 Bu3H2)2(CN)(OSiMe3)], a difference that has not
been commented on. In these two compounds, the Th−N
[2.454(4) Å] and U−C [2.415(6) Å] distances are quite
similar. The distinct ligation modes of the CN ligand in these
pairs of isocyanide and cyanide complexes can be explained by
the metal ion in the former being harder than that in the latter
in the hard and soft acids and bases classification and having a
greater affinity for the harder N end of the CN ligand. The
distinct nature of the UIV−NC and UIII−CN bonding in 2 and
Table 1. Selected Bond Lengths (Å) and Angles (deg) in Complexes 1(THF) and [UN*3X] with X = CN (2), Cl (3), and OTf
(4)
a
b
b
1(THF)
2.3998(18)
2
3
4
U1−O1
U1−C1
U1−N1
U1−N2
U1−N3
U2−O2
U2−C23
U2−N4
U2−N5
U2−N6
2.4310(18)
3.050(3)
U1−N1
U1−N2
2.378(9)
2.234(4)
U1−Cl1
U1−N1
2.593(2)
2.239(2)
U1−O1
2.3315(18)
2.892(3)
2.2371(19)
2.229(2)
2.230(2)
U1−N1
U1−N2
U1−N3
2.232(2)
2.218(2)
2.210(2)
2.213(2)
2.2141(19)
2.2159(19)
N1−U1−N2
N2−U1−N3
N1−U1−N3
O1−U1−N1
O1−U1−N2
O1−U1−N3
111.17(7)
110.14(7)
129.70(7)
98.41(7)
118.39(7)
86.00(7)
N4−U2−N5
N5−U2−N6
N4−U2−N6
O2−U2−N4
O2−U2−N5
O2−U2−N6
109.49(7)
124.60(8)
109.39(7)
126.49(7)
85.00(7)
N2−U1−N2i
N1−U1−N2
119.51(3)
94.02(10)
N1−U1−N1i
Cl1−U1−N1
116.79(4)
100.44(6)
N1−U1−N2
N2−U1−N3
N1−U1−N3
O1−U1−N1
O1−U1−N2
O1−U1−N3
102.14(8)
118.28(8)
114.10(7)
112.43(7)
102.65(7)
106.81(7)
101.68(7)
a
b
Values for the two independent cations. Symmetry code for 2: i, 2 − y, x − y + 1, z. Symmetry code for 3: i, 1 − y, x − y + 1, z.
D
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
yield, and green crystals of [K(THF)4][UN*3(CN)2] (9) were
formed by crystallization from THF. Not surprisingly, the
[UN*3(CN)2]− anion was also obtained by the addition of
CN− to the mono(cyanide) 2. Rapid exchange of the CN
ligand between the mono- and bis(cyanide) complexes was
1
revealed by the H NMR spectra to be mixtures of these
complexes in THF-d8, which exhibited average resonances at δ
between −3.07 and −1.40, values that correspond to the N*
ligands of 2 and 8, respectively.
Attempts at the synthesis of uranium(V) cyanide complexes
by the oxidation of 2, 8, or 9 were unsuccessful. In one reaction
of 9 with CuBr2, red crystals of [UN*3(Br)2] were isolated and
1
characterized by H NMR spectroscopy and X-ray diffraction
analysis.34 It is also noteworthy that [UN*3(CN)2] could not
be synthesized by the substitution of [UN*3X2] (X = F, Cl, Br)
with cyanide ligands, a difficulty that was explained by
unfavorable thermodynamic factors.35 Only three uranium(V)
cyanide complexes were reported: [Nn Bu4 ]2[U-
(Cp*)2(CN)5],36 [Na(15-crown-5)][UN*(N,O)2(CN)] [N,O
= OC(CH2)SiMe2N(SiMe3)]6 and [NEt4][UON*3(CN)].35
Views of the cation of the cyanide-bridged complex 7 and the
anion of the bis(cyanide) 9 are shown in Figures 7 and 8,
respectively, and selected bond distances and angles are listed
in Table 3; the structural parameters of 10, which was obtained
from the oxidation of [K(18-crown-6)][UN*3(CN)] with
pyridine N-oxide (vide supra), are also given in Table 3, and
a view of this compound is presented in Figure 9. The
structures of 7 and 9 resemble those of the anionic fragments in
the corresponding UIII complexes [K(18-crown-6)]-
[(UN*3)2(μ-CN)] and [K(18-crown-6)]2[UN*3(CN)2].1 The
U1 and U2 atoms in 7 are in a pseudotetrahedral configuration,
at a distance of 0.550(4) and 0.698(3) Å from the plane of the
nitrogen amide ligands, while the U1 atom in 9 lies in this
plane, being in a quite perfect trigonal-bipyramidal environ-
ment. The U−N(N*) distances, which average 2.204(9) and
2.252(3) Å in 7 and 9, respectively, are smaller than those of
2.367(10) and 2.386(10) Å in the UIII counterparts, in line with
variation of the radii of the UIV and UIII ions. The mean U−C/
N(CN) distance of 2.562(13) Å in 7 is larger than that in the
U−N(CN) distance of 2.493(4) and 2.468(3) Å in 9 and 10,
respectively. In the latter two complexes, the K−C distances
average 2.930(5) Å in 9 and 2.916(13) Å in 10. However, the
most striking feature in the structures of the uranium(III) and
Figure 5. View of complex 5. H atoms are omitted. Displacement
ellipsoids are drawn at the 30% probability level. Symmetry code: i, 1
− x, −y, 1 − z.
Figure 6. View of complex 6. H atoms are omitted. Displacement
ellipsoids are drawn at the 30% probability level. Symmetry code: i,
−x, 1 − y, 1 − z.
Table 2. Selected Bond Lengths (Å) and Angles (deg) in
Complexes 5 and 6
5
6
U1−C1
U1−C1i
U1−N1
U1−N2
U1−O1
U1−O2i
2.661(3)
2.604(3)
2.190(3)
2.254(2)
2.418(3)
2.432(3)
U1−C1
U1−O4
U1−N1
U1−N2
U1−O1
U1−O2i
2.400(8)
2.468(5)
2.200(6)
2.256(6)
2.528(5)
2.525(5)
C1−U1−N1
C1i−U1−N1
C1−U1−N2
C1−U1−C1i
N1−U1−N2
O1−U1−O2i
U1−C1−U1i
70.59(9)
C1−U1−N1
C1−U1−N2
C1−U1−O2i
N1−U1−N2
N1−U1−O4
N2−U1−O1
O1−U1−O2i
73.2(3)
154.37(10)
161.53(12)
87.15(10)
106.97(9)
146.66(7)
92.86(10)
92.5(3)
159.0(2)
110.9(2)
155.1(2)
159.9(2)
72.71(18)
Symmetry code for 5: i, 1 − x, −y, 1 − z. Symmetry code for 6: i, −x, 1
− y, 1 − z.
[UN*3(CN)2] (8), which was isolated as a pink powder in 58%
yield. The reaction of 1 with an excess of KCN led to the
formation of a greenish powder of K[UN*3(CN)2] in 63%
Figure 7. View of the cation in 7·0.5Et2O. H atoms are omitted.
Displacement ellipsoids are drawn at the 30% probability level. One
particular position of the disordered cyanide is represented.
E
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 8. View of complex 9. H atoms are omitted. Displacement
Figure 9. View of complex 10. H atoms are omitted. Displacement
ellipsoids are drawn at the 20% probability level. Only one position of
the disordered parts is represented. Symmetry codes: i, x, y, z + 1; j, x,
y, z − 1.
ellipsoids are drawn at the 20% probability level. Symmetry codes: i, 1
3
3
3
− x, y, /2 − z; j, /2 − x, /2 − y, 2 − z.
uranium(IV) bis(cyanide) compounds is, as in the case of the
mono(cyanide) complexes, the distinct ligation mode of the
CN ligand, UIII−CN versus UIV−NC (Scheme 3). The U−
N(NC) distance of 2.493(4) Å in 9 is 0.11 Å larger than in the
mono(cyanide) 2, reflecting variation in the coordination
number and charge of the complexes. The CN− ligand in 10
has been refined as a mixture of cyanide and isocyanide, but the
structure determination is of lower quality because of
substantial disorder (see the Experimental Section), so that
the modeling of this anion must be taken with caution. In
addition, the ∼0.8 Å difference between the UIII−C and UIV−
N(CN) distances in the mono(cyanide) [UN*3(CN)]q
complexes (q = 1−, 0)1 is in the range of expected values by
changing the oxidation state of the metal center by one unit in
an ionic model.
Scheme 3. U−NC versus U−CN Coordination Mode in the
Complexes [UIIIN*3(CN)2]2−, [UIVN*3(NC)2]−,
[UIVN*(N,N)(CN)2]−, and UIVN*3(NC)
those of the mononuclear UIV compounds [NEt4][U-
The IR spectra of the mono- and bis(cyanide) complexes 2
and 8 show strong absorption bands assigned to the ν(CN)
stretching frequencies at 2044 and 2058 cm−1, respectively,
while the IR vibrational frequency of the cyanide ion in
NEt4CN is 2050 cm−1. These values can be compared with
(Cp*)2(CN)3] (2053 and 2188 cm−1),7 [NR4]3[U-
36
(Cp*)2(CN)5] (2091 cm−1), [U(C5 Bu3H2)2(CN)(OSiMe3)]
(2040 cm−1),24 and [NEt4][U(C8H8)2(CN)] (2073 cm−1),37
and they suggest the absence of π back-bonding from the U4+
ion to the cyanide ligand.
t
Table 3. Selected Bond Lengths (Å) and Angles (deg) in Complexes 7·0.5Et2O, 9, and 10
a
a
7·0.5Et2O
2.549(8)
9
10
U1−C1a
U2−N1a
2.574(7)
U1−N1
K−C1
2.493(4)
2.930(5)
U1−N1a
K−C1a
U1−N2a
Ki−C2a
U1−N3
U1−N4
U1−N5
2.466(9)
2.929(11)
2.471(9)
2.904(11)
2.316(16)
2.201(14)
2.246(7)
U1−N2
U1−N3
U1−N4
2.210(6)
2.219(6)
2.193(6)
U2−N5
U2−N6
U2−N7
2.197(5)
2.205(6)
2.206(6)
U1−N2
U1−N3
2.254(3)
2.248(5)
N2−U1−N3
N3−U1−N4
N2−U1−N4
C1a−U1−N2
C1a−U1−N3
C1a−U1−N4
107.9(2)
126.7(2)
107.0(2)
112.5(2)
106.7(2)
95.5(2)
N5−U2−N6
N6−U2−N7
N5−U2−N7
N1a−U2−N5
N1a−U2−N6
N1a−U2−N7
113.5(2)
110.7(2)
107.1(2)
110.4(2)
105.0(2)
110.2(2)
N2−U1−N3
N2−U1−N2i
118.55(9)
N3−U1−N4
N4−U1−N5
N3−U1−N5
N1a−U1−N3
N1a−U1−N4
N1a−U1−N5
N2a−U1−N3
N2a−U1−N4
N2a−U1−N5
118.2(3)
122.9(7)
118.8(7)
95.7(4)
88.5(5)
89.8(3)
93.5(5)
84.5(6)
88.4(3)
170.3(5)
122.89(18)
N1−U1−N2
N1−U1−N2i
N1−U1−N3
N1−U1−N1i
89.91(13)
89.15(13)
90.99(10)
178.0(2)
N1a−U1−N2a
a
3
Symmetry code for 9: i, 1 − x, y, /2 − z. Symmetry code for 10: i, x, y, z − 1.
F
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
The ν(CN) frequencies of the cyano-bridged complexes 7
and 9 at 2115 and 2078 cm−1, respectively, are expectedly larger
than those corresponding to the terminal cyanide ligands in 2
and 8. Indeed, the ν(CN) frequencies of cyanide bridges M−
CN−M′ are generally shifted to higher energies than those of
terminal cyanides M−CN.38 The larger ν(CN) values for 9
likely reflect the presence of the U−CN−K linkages, with these
being absent in 8, which involves the noncoordinating NEt4
cation.38 The values for the UIV complexes 7 and 9 are also
larger than those measured for the UIII counterparts [NEt4]-
[(UN*3)2(μ-CN)] (2096 cm−1) and [K(18-crown-
6)]2[UN*3(CN)2] (2063 cm−1).1 This is most probably to be
accounted for by the increase of the ν(CN) frequency generally
observed with the increasing oxidation state of the metal ion
rather than by the slightly more covalent character of the UIII−
CN bonding, which is devoid of any UIII-to-cyanide π* back-
donation according to DFT studies on trivalent [UIIIN*3X2]2−
compounds.1 In contrast, the ν(CN) stretching frequencies of 2
and 8 at 2044 and 2058 cm−1 are smaller than or equal to those
of the UIII analogues [NEt4][UN*3(CN)] (2057 cm−1) and
[NnBu4]2[UN*3(CN)2] (2058 cm−1).1 This feature, difficult to
explain, could reflect the distinct U−CN and U−NC ligation
modes in the corresponding UIII and UIV complexes, although
these U−ligand bonds are predominantly ionic (vide infra).
It is important to note that the UIII−CN versus UIV−NC
coordination mode observed in the tris(silylamide) UN*3
complexes, as well as the CeIII/UIII differentiation previously
studied,1 cannot be considered as a general feature. Indeed, it
seems strongly related to the electronic structure of the
complexes and the coordination geometry of the ligands. For
example, the M−C bonding mode was unambiguously
determined in the series of bis(Cp*) complexes [NnBu4]-
[UIV(Cp*)2(CN)3] and [NnBu4]2[MIII(Cp*)2(CN)3] (M = U,
Ce),7,36,39 which are much more electron-rich than the tris(N*)
compounds. In the UVI compound [UO2(CN)5]3−, the lower
charge of the uranyl ion and the trianionic charge of the whole
complex render the metal softer, thus favoring U−C
coordination of the cyanide ion.40 Moreover, whereas the
N,N metallacyclic bis(cyanide) complex [NEt4][UN*(N,N)-
considered here have been computed at this level of theory.
The computed UIII−CN, UIV−NC, U−N*, and C−N bond
lengths for the actual mono- and bis(cyanide/isocyanide)
complexes, respectively, are reported in Tables 4 and 5. Their
hypothetical UIII−NC and UIV−CN analogues were also
computed and considered for comparison.
The optimized structures of the [UN*3X]q (q = 1−, 0) and
[UN*3X2]q (q = 2−, 1−) complexes are depicted in Figures 10
and 11, respectively, and can be compared with the available X-
ray crystallographic data of the actual UIII−CN and UIV−NC
complexes 2 and 9. First, the computed geometries of the actual
UIII−CN and UIV−NC systems are in good agreement with
available X-ray data with a slight discrepancy, reaching ca. 0.06
Å, for the metal−ligand bond distance in the case of the
UIV(NC)2 complex (Table 5). This good agreement shows
once again the reliability of the ZORA/BP86/TZP method in
computing molecular f complex geometries.1,42 The shortening
of the metal−ligand bond distances when passing from UIII to
UIV, in line with uranium ionic radii variation,22 is well
predicted by computation. Moreover, the C−N bond length
undergoes no significant change in the UIII and UIV complexes,
suggesting mainly cyanide-to-metal σ donation, with no π-back-
donation effects, in agreement with the experimental findings.
In addition, the computed U−C−N and U−N−C bond angles
are indicative of linear coordination with an unperturbed
multiple C−N bond. It is also noteworthy that the computed
N*−U−X (X = CN, NC) bond angles are significantly smaller
in the UIV complexes than in the UIII analogues (92.8 vs
100.8°). This trend is also observed experimentally for the
actual systems, where the N*−UIII−CN and N*−UIV−NC
bond angles are equal to 101.1 and 94.0°, respectively,
differences that could result from a slightly stronger donation
ability of CN over NC ligands toward UIII, while the trend is
converse toward UIV metal centers. Electronic structure analysis
will shed light on these points.
The optimized structure of the metallacyclic uranium(IV)
bis(cyanide) complex [UN*(N,N)(CN)2]− is shown in Figure
12. As aforementioned, the striking difference with the
[UIVN*3X2]− complex is the UIV−CN coordination. The
computed bond lengths (Table 5) compare very well with
the X-ray data.6 The computed N*−U−N* angles, equal to
114.2−130.1°, deviate significantly from 120°, in agreement
with the experimental data.6 This structural feature is also
predicted for the isocyanide hypothetical analogue [UN*-
(N,N)(NC)2]−. The striking difference in coordination
between the [UN*(N,N)(CN)2]− complex and the other UIV
species could not be directly related to these angle variations.
Electronic Structures of the Complexes. As previously
stated,43 the cyanide or isocyanide CN−/NC− anions are strong
( C N )
]
( 1 1 ) [ N , N
=
( M e S i ) N S i -
2
3
Me2CH2CH2SiMe2N(SiMe3)]6 is a close analogue of 10 with
similar structure (Scheme 3), the CN ligands are also
unambiguously coordinated through the C atoms. This striking
and unexpected difference, which is apparently related to slight
variations in the geometrical parameters, in particular the N−
U−N angles of the amido ligand set, which deviate from 120°
by 4−10°, was explained by DFT analysis.
DFT Molecular Geometry Optimization of the Com-
plexes. To investigate further the cyanide coordination
preference to the UIII or UIV metal center, the two mono-
and bis(cyanide/isocyanide) complexes [UN*3X]q (q = 1−, 0)
and [UN*3X2]q (q = 2−, 1−) (U3+/4+; X = CN, NC) were first
considered. Recently, comparative studies on actinide com-
pounds using various DFT-based techniques41a demonstrated
that generalized gradient approximation functionals (e.g., BP86
or PBE; see the Computational Details) give results with
comparable reliability to that of hybrid ones (e.g., B3LYP)
regarding molecular geometries.41b For our part, we carried out
computations in the framework of the relativistic zeroth-order
regular approximation (ZORA) using the BP86 functional and
the polarized triple-ζ Slater basis set (TZP). As previously
stated,1,42 ZORA/BP86/TZP-computed geometries are in
good agreement with X-ray structures. Thus, all structures
Table 4. ZORA/BP86/TZP Relevant Optimized Averaged
Metal−Ligand Bond Distances (Å) and Angles (deg) of the
Mono(cyanide/isocyanide) Complexes [UN*3X]q (q = 1−,
0) (U3+/4+; X = CN/NC) with Available X-ray Data
CN/NC structure
⟨U−X⟩
2.500
⟨C−N⟩
1.177
⟨U−N*⟩ N*−U−C/N
UIIIN*3−CN
2.349
100.8
X-ray1 for the UIII 2.455(15)
complex
1.17(2)
2.348(5)
101(1)
UIVN*3−CN
UIIIN*3−NC
UIVN*3−NC
X-ray for 2
2.462
1.172
2.252
91.3
2.384
1.187
2.350
100.7
2.324
1.186
2.256
92.8
2.378(9)
1.118(15)
2.234(4)
94.02(10)
G
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 5. ZORA/BP86/TZP Relevant Optimized Averaged Metal−Ligand Bond Distances (Å) and Angles (deg) of the
Bis(cyanide/isocyanide) Complexes [UN*3X2]q (q = 2−, 1−) and [UN*(N,N)X2]−(X = CN, NC) with Available X-ray Data
CN/NC structure
⟨U−X⟩
⟨C−N⟩
⟨U−N*⟩
⟨N*−U−N*⟩
120.0
1
UIIIN*3−(CN)2
2.604
1.180
2.378
X-ray1 for the UIII complex
2.62(6)
2.576
1.168(3)
1.173
2.39(1)
2.280
119.9(4)
120.1
UIVN*3−(CN)2
1
UIIIN*3−(NC)2
2.483
1.185
2.389
119.9
UIVN*3−(NC)2
X-ray for 9
2.429
1.183
2.293
120.1
2.493(4)
2.575
1.154(7)
1.174
2.252(3)
2.241
120.5(5)
114.2−130.1
114.7(2)−130.3(2)
114.3−130.0
UIVN*(N,N)−(CN)2
X-ray6 for 11
2.50(2)
2.441
1.14(1)
1.183
2.266(1)
2.250
UIVN*(N,N)−(NC)2
toward UVI (d0f0) in uranyl systems UO22+ was investigated by
Bursten et al.,43a who concluded that metal-based orbitals
match energetically better with the C-localized σ orbital of the
cyanide ligand, making it a much more effective donor than the
N-localized σ orbital. Our previous work on cyanide/isocyanide
coordination related to LnIII/AnIII differentiation revealed
distinct coordination modes of the CN group.1 It appeared
that the better energy matching between 6d/5f U and ligand
orbitals plays a significant role in the metal−ligand coordination
preference of the cyanide/isocyanide ligands toward the UIII/
CeIII pair in the [MN*3X2]2− complexes (M = Ce, U; X = CN,
NC), and the U−CN versus Ce−NC coordination was well
corroborated by consideration of the binding energies of these
ligands to the metal ions and by natural population analysis
(NPA), the quantum theory of atoms in molecules (QTAIM),
and bond order analyses. Thus, the distinct coordination of the
CN versus NC ligands toward the UIII/UIV pair can also be
related to the energy matching and stronger σ-donating ability
of the cyanide and isocyanide ligands.
Figure 10. ZORA/BP86/TZP-optimized structures of the mono-
(cyanide/isocyanide) complexes [UN*3X]q (q = 1−, 0; X = CN, NC).
The computed Mayer44 and Nalewajski−Mrozek (NM)45
bond indices (see the Computational Details) for the U−C/N
and C−N bonds are reported in Table 6. As expected, the NM
approach, which accounts for ionic and covalent contributions,
gives greater metal−ligand bond orders (up to 2 times) than
the Mayer approach. For the C−N/N−C bonds, the ionic
character is low, so that the NM and Mayer indices get closer. It
is worth noting that, for all systems, Mayer analysis consistently
gives larger U−X bond orders (X = CN, NC) for the UIV
complexes than for their UIII congeners, in agreement with the
computed UIII to UIV bond shortening. Interestingly, in the case
of the mono(cyanide/isocyanide) [UN*3X]q (q = 1−, 0)
systems, systematically larger Mayer bond orders for cyanide
UIII/IV−CN coordination than for isocyanide UIII/IV−NC
coordination are observed. On the contrary, NM bond orders
are in line with the different coordination preferences of the UIII
and UIV ions. Indeed, a significantly higher bond order is
obtained for UIII−CN, 1.176 versus 0.994 for UIII−NC, likely
indicating a significantly stronger coordination preference of
cyanide than isocyanide toward UIII. Such a good correlation
between NM bond orders and the coordination preference has
already been observed considering the CeIII/UIII pair,1 thus
confirming the reliability of the NM approach.46
Figure 11. ZORA/BP86/TZP-optimized structures of the bis-
(cyanide/isocyanide) complexes [UN*3X2]q (q = 2−, 1−; X = CN,
NC).
Moreover, the reverse is obtained for UIV systems because
the cyanide UIV−CN NM bond order is significantly smaller
than the isocyanide UIV−NC bond order (1.082 vs 1.245),
correlating well with the observed coordination. Furthermore,
in the bis(cyanide/isocyanide) [UN*3X2]q (q = 2−, 1−)
complexes, the computed NM bond orders are larger for the
UIII(CN)2 coordination than for the UIII(NC)2 one (0.814 vs
Figure 12. ZORA/BP86/TZP-optimized structure of the metallacyclic
uranium(IV) bis(cyanide) complex [UN*(N,N)(CN)2]−.
σ donors. This originates from σ donation of either the C-
localized upper orbital of CN− or the lower N-localized orbital
of NC− with N major character. Their coordination preference
H
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 6. ZORA/BP86/TZP NM and Mayer Average Bond Orders for the Mono- and Bis(cyanide/isocyanide) Complexes
a
[UN*3X]q (q = 1−, 0), [UN*3X2]q (q = 2−, 1−), and [UN*(N,N)X2]− (X = CN, NC)
atom−atom bond orders
Mayer
⟨C−N/N−C⟩
NM
CN/NC structure
⟨d⟩ (Å)
⟨U−X⟩
⟨U−X⟩
⟨C−N/N−C⟩
UIIIN*3−CN
UIVN*3−CN
UIIIN*3−NC
UIVN*3−NC
Q
T
Q
T
Q
T
Q
T
T
T
2.500
2.462
2.384
2.324
2.604
2.576
2.483
2.429
2.575
2.441
0.591
0.612
0.486
0.605
0.614
0.628
0.439
0.467
0.610
0.496
2.801
2.833
2.463
2.357
2.814
2.864
2.602
2.487
2.873
2.552
1.176
1.082
0.994
1.245
0.814
0.896
0.778
1.037
0.939
0.820
2.935
3.019
2.981
2.857
3.095
3.016
3.034
2.891
2.899
3.006
1
UIIIN*3−(CN)2
UIVN*3−(CN)2
1
UIIIN*3−(NC)2
UIVN*3−(NC)2
UIVN*(N,N)−(CN)2
UIVN*(N,N)−(NC)2
a
Q = quartet. T = triplet.
0.778)1 and smaller for UIV(CN)2 than for UIV(NC)2 (0.896 vs
1.037). These results show that, for the systems under study,
the observed UIII−CN versus UIV−NC coordination prefer-
ences can be deduced from the highest NM bond orders.
To get a clearer picture of covalency in U−CN/NC
coordination, electronic structure analyses including NPA47a
and QTAIM47b descriptors have been considered. NPA and
QTAIM approaches have proved to be more reliable than the
Mulliken population analysis.47c QTAIM bond critical point
(BCP) properties (electron and energy densities ρc and Hc)
have also been calculated using the ADF program.47k,l In
particular, for f-element complexes, these two topological
approaches were used to probe the covalency and gave results
in good agreement with the experimental trends.47d−f The
results of NPA and QTAIM calculations, computed at the
ZORA/BP86/TZP level for both the U−CN and U−NC
coordination modes in the mono- and bis(cyanide/isocyanide)
[UN*3X]q (q = 1−, 0) and [UN*3X2]q (q = 2−, 1−)
complexes, are reported in Table 7. The natural metal charge as
well as the metallic spin population ρU are presented. The latter
property is computed as the difference between the total α- and
β-spin electronic populations of the metal. The QTAIM
descriptors, i.e., electron (ρc) and energy density (Hc) data, at
the metal−ligand BCPs for UIII/IV−CN/NC bonds are also
given. As expected, inspection of the NPA results for the
[UN*3X]q (q = 1−, 0; X = CN/NC) complexes shows
significantly smaller natural metal charges than the formal value
3 of the UIII ion, independent of the U−CN or U−NC
coordination (for instance, 2.13/2.15 for UIII instead of 3),
indicating a strong σ donation of the ligand. Interestingly, the
metal spin population of the UIII complex, which is a 5f3
species, is lower than 3, whereas this population is higher than 2
in the case of the UIV 5f2 complex. This means that for the latter
complexes a small negative spin density is spread over the
ligands as noted previoulsy.40 Moreover, the comparison
between mono- and bis(cyanide/isocyanide) systems shows
that metallic net charges become smaller for the latter
complexes (2.13 vs 2.07 for UIII−CN and 2.46 vs 2.40 for
UIV−NC). It appears also that, for the UIII/UIV [UN*3X]q (q =
1−, 0) pairs, the NPA charges show only a slight difference
between the cyanide and isocyanide coordination (e.g., 2.13 vs
2.15 for UIII−CN/NC). Nevertheless, as suggested by bond
order analysis (Table 6), the UIII−CN coordination is
significantly stronger than the UIII−NC one, and the opposite
is obtained for the UIV complexes, in agreement with the
observed coordination preference. Furthermore, the stronger
donation ability and covalent bonding of the cyanide over
isocyanide toward UIII is also highlighted by the QTAIM
approach. Indeed, as given by the BCPs, the computed electron
density (ρc) is slightly higher for the UIII−CN coordination
than for the UIII−NC one (0.075 vs 0.070), and the energy
density (Hc) data are meaningfully more negative (−0.052 vs
−0.034). Thus, the covalent contribution appears to be more
important for the cyanide UIII−CN coordination than for the
isocyanide UIII−NC coordination. Moreover, in the UIV case, as
corroborated by bond orders, the UIV−NC electron density ρc
is higher than that in UIV−CN (e.g., 0.042 vs 0.037), indicating
a slightly more covalent character in the former. More
interestingly, the critical electron density (ρc) is slightly larger
for the actual uranium(IV) bis(isocyanide) complex
[UN*3(NC)2]− than for its hypothetical bis(cyanide) congener
[UN*3(CN)2]− (0.067 vs 0.059). These results correlate well
with the stronger isocyanide NC σ-donation ability toward UIV
and its experimentally observed coordination preference.
A comparison of all of the uranium(III) and uranium(IV)
mono(cyanide/isocyanide) pairs shows a decrease in the
electron density ρc (e.g., 0.075 vs 0.037 from UIII−CN to
UIV−CN), suggesting also a decrease in the covalency. The
same trend is noted with the isocyanide UIII/IV−NC complexes
Table 7. ZORA/BP86/TZP NPA and QTAIM Results for
the Mono- and Bis(cyanide/isocyanide) Complexes
[UN*3X]q (q = 1−, 0), [UN*3X2]q (q = 2−, 1−), and
[UN*(N,N)X2]− (X = CN, NC)
QTAIM U−C/N
natural spin
population ρU
NPA net
ρc (e/
Hc
(au)
U−(CN/NC) structure
charge qU bohr3)
UIIIN*3−CN
UIVN*3−CN
UIIIN*3−NC
UIVN*3−NC
2.91
2.12
2.92
2.12
2.88
2.10
2.85
2.09
2.03
2.01
2.13
2.41
2.15
2.46
2.07
2.34
2.18
2.40
2.28
2.36
0.075 −0.052
0.037 −0.013
0.070 −0.034
0.042 −0.026
0.027 −0.026
0.059 −0.023
0.025 −0.018
0.067 −0.043
0.032 −0.021
0.029 −0.013
1
UIIIN*3−(CN)2
UIVN*3−(CN)2
1
UIIIN*3−(NC)2
UIVN*3−(NC)2
UIVN*(N,N)−(CN)2
UIVN*(N,N)−(NC)2
I
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 13. ZORA/BP86/TZP interaction diagram for the [UN*3]q (q = 1, 0) molecular fragments and the cyanide CN− ligand. In this diagram, the
UIII [UN*3] and UIV [UN*3]+ levels are displayed on the right and left, respectively, and the cyanide ion MOs are given in the middle.
(i.e., 0.070 vs 0.042 for ρc). As suggested by Kaltsoyannis et
al.,47d,h this phenomenon can be related to the better energy
matching between the metallic 6d/5f and ligand orbitals. In
fact, the decrease of the covalency (weaker orbital mixing)
when passing from UIII−CN to UIV−CN is attributed to the
smaller σ-donation ability of the C-localized upper orbital
toward a deeper UIV 6d/5f shell. The stronger σ-donation
ability of isocyanide in the case of UIV−NC than that of cyanide
in UIV−CN is also explained by the better energy matching
between UIV 6d/5f and N-localized orbitals. This stronger
interaction leads to a slight increase in the covalency factor, i.e.,
the electron density ρc (0.037 for UIV−CN vs 0.042 for UIV−
NC). This explains the observed coordination preference of
CN/NC toward the UIII/UIV pair in the actual [UN*3X]q (q =
1−, 0) systems. However, in the bis(cyanide/isocyanide)
[UN*3X2]2− compounds, the opposite trend is observed
when passing from UIII to UIV. Indeed, a comparison between
the UIII(CN)2 and UIV(CN)2 complexes shows a significant
increase in the electron density ρc (0.027 vs 0.059), suggesting
a more covalent contribution in the latter. The same trend is
noted with the bis(isocyanide) UIII/IV(NC)2 derivatives (0.025
vs 0.067).
Overall, it appears that the stronger σ-donation ability (due
to better energy matching) correlates with the increase in the
covalent contribution (orbital mixing), which accounts for a
significant part of the UIII/IV−CN/NC bonding. It is also worth
noting that QTAIM data clearly highlight the slight role of
covalency in the observed cyanide CN versus isocyanide NC
coordination preference toward the UIII/UIV pair. In fact, as
noted in previous works on f complexes,47g,h the BCP values
provided from the QTAIM method are small and suggest weak
covalency, thus indicating dominant ionic metal−ligand
bonding.
Turning back to the [UN*(N,N)X2]− complex, the major
variations of the Mayer and NM U−C/N bond indices between
the cyanide and isocyanide derivatives [UN*(N,N)X2]− are
similar to those obtained for the corresponding [UN*3X2]− (X
= CN, NC) complexes (Table 6). Here again, the observed
cyanide coordination for the metallacyclic complex is well
predicted with a metal−ligand NM bond order larger for U−
CN than for U−NC (0.939 vs 0.820). Indeed, to further assess
this point, the QTAIM data (Table 7) considering the distinct
coordination modes of the cyanide ligand in the two actual
compounds indicate that the electron and energy densities ρc
J
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 14. ZORA/BP86/TZP α-spin MO diagram for the complexes [UN*3(CN)]− and [UN*3(NC)]. Cutoff used, 0.03 e/bohr3.
and Hc of the metallacycle U−CN bond are slightly larger than
those in the isocyanide counterpart.
interact more with UIV 5f2 fragment orbitals, leading to the
UIV−NC coordination preference.
Molecular Orbital (MO) Analysis of the Mono-
(cyanide/isocyanide) Complexes. To further assess the
CN/NC coordination preference in the [UN*3X]q (q = 1−, 0)
complexes, an interaction diagram between the UIII/IV
[UN*3]0/+ and CN− orbitals has been computed and is
displayed in Figure 13. The better energy matching between
the UIII 5f3 metallic fragment orbitals and the upper C-localized
cyanide CN− orbitals is in favor of UIII−CN coordination.
Similarly, the deeper N-localized isocyanide NC− orbitals
MO frontier diagrams of the [UN*3X]q (q = 1−, 0; X = CN/
NC) complexes are presented in Figure 14. For the sake of
simplicity, only the α-spin MOs are displayed; the UIII−CN and
UIV−NC MOs are shown on the right and left sides of the
diagram, respectively. The percentages % (UN*3/X) represent
the contributions of the UIII/IV metallic fragment and the CN/
NC ligands to the frontier MOs. As displayed in this diagram,
two different but significant sets of MOs appear. For all
complexes, the diagram shows that the highest occupied α-spin
K
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 15. ZORA/BP86/TZP α-spin MO diagram for the metallacyclic complexes [UN*(N,N)X2]− (X = CN/NC). Cutoff used, 0.03 e/Bohr3.
orbitals, i.e., SOMO, SOMO−1, and SOMO−2 of
[UIIIN*3(CN)]− 5f3 and SOMO and SOMO−1 of
[UIVN*3(NC)] 5f2 complexes, are essentially metallic, with a
dominant 5f orbital character. In line with QTAIM and NM
analyses, a stronger UIII−CN bonding results from σ donation
as shown by an important orbital mixing, namely, 72.0% of CN
and 21.7% of UIIIN*3 for MO 96A. Comparatively, the
isocyanide UIV−NC coordination is ensured by σ donation
with an orbital mixing of 30% UN*3/63.1% NC for MO 95A
and 4.1% UN*3/89.3% NC for MO 99A. It is also worth noting
that the σ donation of MO 96A for UIII−CN cyanide bonding
is located at a much higher energy (−4.10 eV) than its
isocyanide UIV−NC MOs 96A and 99A (∼−6.5 eV) counter-
parts, confirming the crucial energy matching rule for CN/NC
coordination preference toward the UIII/UIV pair, in good
agreement with previous theoretical works.1,47 As previously
noted,1 the UIII/IV−CN/NC coordination mode preference
originates from both the σ-donating ability of ligands toward
the UIII/IV pair (best energy matching) and the orbital mixing
(covalency) between the 6d/5f metal orbitals and the C-
localized cyanide or N-localized isocyanide MO. We shall study
below the stabilities of the actual cyanide and isocyanide
systems.
percentages %(UN(N,N)/X2) represent the contributions of
the UIV metallic molecular fragment and the bis(CN/NC)
ligands to the frontier MOs. The diagrams show that the
highest occupied α-spin orbitals, i.e., the SOMOs for the triplet
state (5f2) complexes, are essentially metallic, with dominant 5f
orbital character. Moreover, no metal-to-ligand π* back-
donation is observed in these frontier MOs. The main
difference between the cyanide and isocyanide species is
noted considering the occupied MOs relative to the σ donation
of the ligands. Indeed, these levels are significantly lower in
energy for the cyanide than for the isocyanide coordination, as
exemplified with the σ-donation MOs 97A versus 102A energy
levels (i.e., −4.35 vs −3.54 eV). This energy difference is
certainly partly responsible for the preferred U−(CN)2
coordination in the [UN*(N,N)X2]− complex. This can also
be illustrated by the higher mixing orbital and the energy
splitting of the two σ-donating MO levels, which is significantly
larger for the cyanide complex than for the isocyanide complex.
Energy Decomposition Analysis. In order to investigate
the energetic factors driving the cyanide/isocyanide coordina-
tion preference toward UIII and UIV ions, the total bonding
energies TBEfrag between the metal and cyanide ligand have
been computed. Molecular fragments have been considered as
follows:
Turning back to the metallacyclic UIV[UN*(N,N)X2]−
complex, we now consider the MO diagrams of the
bis(cyanide/isocyanide) structures, depicted in Figure 15. The
L
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 8. Energy Decomposition Analysis at the ZORA/BP86/TZP Level for the Mono- and Bis(cyanide/isocyanide)
Complexes [UN*3X]q (q = 1−, 0), [UN*3X2]q (q = 2−, 1−) and [UN*(N,N)X2]− (X = CN, NC)
X = CN/NC
Est (eV)
Eorb(eV)
TBEfrag(eV)
relative stability ΔE (kcal mol−1
)
UIIIN*3−X
+0.391/+0.038
−3.068/−3.134
+0.349/+0.517
−6.492/−6.103
−6.582/−6.264
−3.976/−3.246
−2.966/−3.469
−5.846/−5.630
−5.255/−5.715
−5.089/−5.354
−3.584/−3.208
−6.035/−6.603
−5.497/−5.112
−11.747/−11.817
−11.671/−11.618
0.0/8.7
13.1/0.0
0.0/8.9
1.6/0.0
0.0/1.4
UIVN*3−X
1
UIIIN*3−X2
UIVN*3−X2
UIVN*(N,N)−X2
[UN*3]0/+ + X− → [UN*3UX]q
(X = CN/NC; q = 1 − , 0)
leading to a higher polarization effect for the isocyanide ligand
and then to a more important effect on the orbital energies.47i
This partly explains the significantly larger orbital term (in
absolute value) for the NC versus CN coordination.
As previously stated,1,42g,47d the orbital term Eorb (stabilizing
energy partly due to orbital mixing) correlates with TBEfrag
satisfactorily in predicting the CN/NC coordination preference
toward UIII/IV, except for the UIV metallacyclic complex
[UN*(N,N)(CN)2]. Interestingly, the comparison between
the cyanide UIII−CN and isocyanide UIV−NC coordination
modes shows a decrease of the orbital term Eorb (−3.976 vs
−3.469 eV). The fact that TBEfrag is significantly larger for UIV−
NC binding than for UIII−CN binding is mainly due to the
[UN*3]0/+ + 2X− → [UN*3UX2]q
(X = CN/NC; q = 2 − , 1 − )
TBEfrag values have been computed at the spin-unrestricted
ZORA/BP86/TZP level (see the Computational Details). The
results presented in Table 8 are in good agreement with the
experimental findings, namely, that the complex exhibiting the
highest TBEfrag is the most stable one. Indeed, considering the
[UN*3UX]q (q = 1−, 0) complexes, it can be seen that TBEfrag
for the UIII−CN coordination is larger (in absolute value) than
that for the UIII−NC one (i.e., −3.584 vs −3.208 eV). The
opposite is obtained for the UIV complex, for which the most
stable coordination is the UIV−NC one (−6.035 vs −6.603 eV).
Similarly, for the bis(cyanide/isocyanide) complexes
[UN*3UX2]q (q = 2−, 1−), our previous work1 revealed that
the cyanide UIII−CN coordination is more stable than the UIII−
NC one (−5.497 vs −5.112 eV), whereas the opposite is
obtained for the tetravalent UIV complexes, where the UIV−NC
bonding is predicted to be slightly stronger than the UIV−CN
one (−11.747 vs −11.817 eV). The calculated TBEfrag values
thus predict correctly the observed coordination.
We consider now the different energy terms, i.e., the steric
Esteric and orbital Eorb energy terms composing TBEfrag: TBEfrag
= Esteric + Eorb (see the Computational Details). These two
terms (Table 8) drive the CN/NC coordination preference
toward the UIII/IV pair. First, it is worth remembering that the
orbital Eorb part includes both a polarization term due to the
rearrangement of the metal and ligand charge distribution with
complexation and some possible covalence if their orbitals
overlap.47g,i,j Unfortunately, these two terms cannot be
evaluated separately.47i Moreover, the repolarization compo-
nent is likely not the same for the C or N coordination modes
so that differences in Eorb cannot be directly related to
differences in covalency. For the UIII mono(CN/NC)
complexes, because the steric part of TBE is positively small,
the UIII−CN coordination preference is largely favored by the
Eorb term, which is more important (in absolute value) than
that for the UIII−NC coordination (−3.976 vs −3.246 eV). In
the same way, the UIV case reveals a more negative Eorb term for
UIV−NC than for UIV−CN (−3.469 vs −2.966 eV). On the
other hand, the steric Esteric term, which is summed from the
Pauli repulsion (repulsion between electron pairs) and
stabilizing electrostatic interactions, shows that the cyanide/
isocyanide UIII−CN/NC binding mode is sterically unfavorable
compared to UIV−CN/NC. The difference is due to the larger
electrostatic term because of the differently charged [N*3U]0/+
species, whereas Pauli repulsions do not vary significantly. It is
also notable that, in the case of the UIV monocoordination, the
UIV−NC bond distance is shorter than the UIV−CN one,
higher stabilization provided by the negative steric term Esteric
,
namely, −3.134 versus 0.391 eV, respectively. The same
conclusion can be drawn for the bis(cyanide/isocyanide)
[UIIIN*3(CN)2]2− and [UIVN*3(NC)2]− complexes. Concern-
ing the metallacyclic [UN*(N,N)(CN)2]− complex, which, in
contrast to [UN*3(NC)2]−, exhibits the U−CN coordination
mode, the energy difference between the actual cyanide and
hypothetical isocyanide complexes is small. Looking at the
energy decomposition, it can be seen that, although the
computed orbital energy Eorb term is larger for U−NC
coordination than for U−CN coordination (−5.354 vs
−5.089), the Esteric term acts more importantly toward the
observed U−CN coordination. Finally, we can conclude that
the U−CN/NC coordination preference toward the UIII/UIV
pair is related to the subtle balance between steric, covalent,
and ionic factors that govern U−CN/NC bonding. The energy
decomposition calculations combined with QTAIM analysis
indicated a predominantly ionic U−CN/NC bonding but also a
significant contribution of the orbital interaction to the
bonding. Theoretical investigation of triatomic cyanides M−
CN and isocyanides M−NC of first-row transition metals
already brought to light the delicate balance between
electrostatic and orbital effects that drive the chosen
coordination.48 These subtle effects are also at work
considering the TiIV complexes [Ti(CN)n]4−n (n = 1−6),
leading to the Ti−NC coordination for n = 1−5 and Ti−CN
for n = 6.49 It is also worth noting that NM bond orders
correlate nicely with the observed U−CN/NC coordination
preferences of the complexes under consideration. The
computed TBEfrag allows in all cases to predict the observed
coordination mode.
CONCLUSION
■
Uranium(IV) cyanide complexes in the tris(silylamide) UN*3
series could not be obtained by the oxidation of trivalent
precursors but were easily isolated from addition reactions of
CN− to the cationic complex 1, which was prepared by
protonolysis of the metallacycle [UN*2(N,C)] [N,C =
CH2SiMe2N(SiMe3)] with [HNEt3][BPh4]. The cyanido-
bridged dinuclear compound 7 and the mononuclear mono-
M
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
[NEt4][UN*3(Cl)(CN)], whose structure could not be determined
with accuracy because of their poor quality for X-ray diffraction
analysis.
and bis(cyanide) complexes 2 and [M][UN*3(CN)2] [M =
K(THF)4 and K(18-crown-6)] were characterized by their X-
ray crystal structure. In contrast to their UIII analogues
[NMe4][UN*3(CN)] and [K(18-crown-6)]2[UN*3(CN)2] in
which the CN anions are coordinated to the metal center via
the C atom, the UIV derivatives exhibit the isocyanide U−NC
coordination mode of the cyanide ligand. The observed
coordination preference of the cyanide and isocyanide ligands
toward UIII and UIV complexes has been computationally
investigated. Consideration of the actual complexes and their
hypothetical counterparts [UN*3X]q (q = 1−, 0) and
[UN*3X2]q (q = 2−, 1−) (X = CN, NC) shows that the
stronger σ-donating ability of cyanide and isocyanide toward
the UIII/UIV pair is governed by the best energy matching
between 6d/5f metal and ligand orbitals and covalency
contribution (orbital mixing). This latter effect seems to play
a more significant role for the observed UIII−CN coordination
than for the UIV−NC coordination. A comparison of the
different quantum descriptors, i.e., bond orders, NPA/QTAIM
data, and energy decomposition analysis, has allowed high-
lighting of the subtle balance between covalent, ionic, and steric
factors that govern the U−CN/NC bonding. Furthermore, the
UIII−CN versus UIV−NC coordination mode observed in the
tris(silylamide) UN*3 complexes cannot be considered as a
general feature because the structure of the metallacyclic
bis(cyanide) complex [NEt4][UIVN*(N,N)(CN)2] revealed the
UIV−C ligation mode of the CN ligand. These distinct
coordination modes remain difficult to explain because of the
small energy difference between the cyanide and isocyanide
complexes, but in all cases, DFT computations afforded good
predictions of the observed U−CN/NC coordination prefer-
ences.
Synthesis of [UN*3][BPh4] (1). A 50 mL flask was charged with
[UN*2(N,C)] (1000 mg, 1.4 mmol) and [HNEt3][BPh4] (578 mg,
1.37 mmol), and THF (20 mL) was distilled in it under reduced
pressure at −78 °C. After stirring for 30 min, the pale-brown solution
was evaporated to dryness, and the beige powder of 1 was washed with
toluene (20 mL) and pentane (2 × 15 mL) and dried under vacuum
(1300 mg, 86%). Anal. Calcd for C42H74BN3Si6U: C, 48.58; H, 7.18;
1
N, 4.05. Found: C, 48.94; H, 7.03; N, 3.65. H NMR (THF-d8): δH
5.49 (t, J = 7.1 Hz, 4H, p-Ph), 5.20 (t, J = 7.1 Hz, 8H, m-Ph), 4.80 (m,
8H, o-Ph), −3,54 (s, 54H, SiMe3). 1H NMR (pyridine-d5): δH 7.44 (m,
8H, Ph), 6.78 (m, 12H, Ph), −3.98 (s, 54H, SiMe3). 1H NMR
(benzene-d6): δH 2.91 (m, 4H, p-Ph), 2.82 (m, 8H, Ph), 1.60 (m, 8H,
BPh4), −6.33 (s, 54H, SiMe3). 13C{1H} NMR (THF-d8): δC 162.9 (q,
JC−B = 49.4 Hz, ipso-Ph), 133.8 (Ph), 124.1 (q, JC−B = 2.7 Hz, o-Ph),
119.4 (Ph), −4.7 (SiMe3). 13C{1H} NMR (benzene-d6): δC 130.5
(Ph), 120.4 (Ph), 116.7 (Ph), −6.1 (SiMe3). The signal of the ipso-C
atoms is not visible. Pale-green platelets of 1(THF) were obtained by
the slow diffusion of Et2O into a THF solution of 1.
Synthesis of [UN*3(CN)] (2). A 50 mL flask was charged with 1
(300 mg, 0.29 mmol) and NEt4CN (42.8 mg, 0.26 mmol), and THF
(20 mL) was distilled in it under reduced pressure at −78 °C. After
stirring for 15 h at 20 °C, precipitation of [NEt4][BPh4] was observed.
The solvent was evaporated off, and 2 was extracted in Et2O by
Soxhlet extraction and isolated as a pink powder after drying under
vacuum (145 mg, 75%). Anal. Calcd for C19H54N4Si6U: C, 30.62; H,
7.30; N, 7.51. Found: C, 30.66; H, 7.27; N, 7.50. 1H NMR (THF-d8):
1
δH −3.07 (s, w1/2 = 49 Hz, 54H, SiMe3). H NMR (benzene-d6): δH
−2.81 (s, w1/2 = 83 Hz, 54H, SiMe3). 13C{1H} NMR (THF-d8): δC
−13.6 (SiMe3). 13C{1H} NMR (benzene-d6): δC −14.5 (SiMe3). IR
(Nujol): ν(CN) 2044 cm−1. Pale-pink crystals of 2 were obtained by
crystallization from a 1:5 mixture of pentane and THF at −35 °C.
Synthesis of [UN*3(Cl)] (3). An NMR tube was charged with
[UN*2(N,C)] (20.0 mg, 0.028 mmol) and [HNEt3][Cl] (3.99 mg,
0.028 mmol) in THF-d8 (0.5 mL). After 24 h at 20 °C, the spectrum
showed the complete conversion of the metallacycle into 3,
characterized by a singlet at δ −2.48. The solvent was evaporated
off, and pentane (0.5 mL) was added. Cooling the solution at −35 °C
led to the formation of brown crystals of 3.
EXPERIMENTAL SECTION
■
General Procedure. All reactions were carried out under argon
with the rigorous exclusion of air and water (<5 ppm oxygen or water)
using standard Schlenk-vessel and vacuum-line techniques or in a
glovebox. Solvents were thoroughly dried by standard methods and
distilled immediately before use. The commercial reagents (Fluka,
Aldrich) KCN, CuCN, CuBr2, NEt4CN, NnBu4CN, [HNEt3]Cl, and
pyridine-N-oxide have been used as received. [pyH][OTf] was
precipitated by the addition of pyridine into a solution of TfOH in
Et2O, and [HNEt3][BPh4] was obtained by mixing [HNEt3]Cl and
NaBPh4 in water. 18-crown-6 (Fluka) was dried under vacuum before
use. [UN*3],50 [UN*2(N,C)],9 and [K(18-crown-6)][UN*3(CN)]1
were prepared according to literature procedures. IR samples were
prepared as Nujol mulls between KBr round cell windows and the
spectra recorded on a PerkinElmer FT-IR 1725X spectrometer. The
1H and 13C{1H} NMR spectra were recorded on a 200 instrument at
20 °C when not otherwise specified and referenced internally using the
residual protio solvent resonances relative to tetramethylsilane (δ 0).
The signals corresponding to the CN ligands were not visible on the
13C{1H} NMR spectra. Elemental analyses were performed by
Analytische Laboratorien at Lindlar (Germany) or by Medac Ltd. at
Chobham (UK).
Synthesis of [UN*3(OTf)] (4). (a) A 50 mL flask was charged with
[UN*2(N,C)] (310 mg, 0.43 mmol) and [PyH][OTf] (98.5 mg, 0.43
mmol), and THF (20 mL) was distilled in it under reduced pressure at
−78 °C. After stirring for 2 h at 20 °C, the pale-brown solution was
evaporated to dryness, leaving the pale-brown powder of 4, which was
washed with pentane (20 mL) and then Et2O (20 mL) and dried
under vacuum (321 mg, 86%). Anal. Calcd for C19H54F3N3O3SSi6U:
1
C, 26.28; H, 6.27; N, 4.84. Found: C, 26.66; H, 5.93; N, 4.76. H
1
NMR (THF-d8): δH −1.45 (s, 54 H, SiMe3). H NMR (benzene-d6):
δH −1.34 (s, 54 H, SiMe3). 13C{1H} NMR (THF-d8): δC −5.0
(SiMe3). 13C{1H} NMR (benzene-d6): δC −4.9 (SiMe3). IR (Nujol):
1196, 1162, 992 cm−1. Pinkish crystals of 4 were formed at −35 °C by
crystallization in a pentane−THF mixture.
(b) An NMR tube was charged with [U(N*)3] (10.0 mg, 0.014
mmol) and Ph3COTf (5.48 mg, 0.014 mmol) in benzene-d6 (0.5 mL).
The violet color of the solution turned immediately orange, and the
spectrum showed the formation of 4 as the sole uranium complex (δH
−1.34) and of Gomberg’s dimer Ph3CCH(C4H4)CCPh2.
Reaction of [UN*3] and CuCN. An NMR tube was charged with
[UN*3] (10.0 mg, 0.014 mmol) and CuCN (6.25 mg, 0.07 mmol) in
benzene-d6 (0.5 mL). After 48 h at 20 °C, the spectrum of the brown
solution showed the presence of [UN*2(N,C)] as the sole uranium
product.
Reaction of [UN*3(Cl)] (3) and NEt4CN. An NMR tube was
charged with 3 (20.0 mg, 0.027 mmol) and NEt4CN (13.46 mg, 0.053
mmol) in THF-d8 (0.5 mL). The tube was immersed in an ultrasound
bath, and after 24 h, the spectrum of the pale-brown solution showed a
large signal of the N* ligands at δ −3.16 (w1/2 ∼ 700 Hz). Cooling the
THF solution led to the formation of pale-pink platelets of
Synthesis and Crystals of [{UN*(μ-N,C)(μ-OTf)}2] (5) and
[{UN*(N,C)(THF)(μ-OTf)}2] (6). A 50 mL flask was charged with 4
(124.5 mg, 0.14 mmol), and THF (20 mL) was distilled in it under
reduced pressure at −78 °C. After stirring for 2 days at 70 °C, the
brown solution was evaporated to dryness, leaving a brown powder
that was crystallized in Et2O (5 mL) at −33 °C. The brown crystals of
5 were filtered off and dried under vacuum (42.5 mg, 42%). Anal.
Calcd for C26H70F6N4O6S2Si6U2: C, 23.00; H, 5.20; N, 4.13. Found: C,
22.25; H, 4.87; N, 4.01. 1H NMR (THF-d8): δH 35.03 (s, 18H, SiMe3),
15.84 (s, 12H, SiMe2), −0.72 (s, 36H, SiMe3), −228.72 (s, 4H, CH2).
1H NMR (benzene-d6): δH 34.48 (s, 12H, SiMe2), 17.37 (s, 18H,
N
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 9. Crystal Data and Structure Refinement Details
1(THF)
2
3
4
5
chemical formula
C50H90BN3O2Si6U
1182.63
C19H54N4Si6U
745.23
trigonal
R3c
C18H54ClN3Si6U
754.66
trigonal
R3c
C19H54F3N3O3SSi6U
868.28
C26H70F6N4O6S2Si8U2
1413.76
M (g mol−1
)
cryst syst
triclinic
triclinic
triclinic
space group
a (Å)
P1
̅
P1
̅
P1
̅
16.9269(7)
18.7867(6)
19.4285(7)
107.7551(18)
93.7460(18)
91.4929(19)
5864.5(4)
4
18.4684(6)
18.4684(6)
16.9719(10)
90
18.4359(3)
18.4359(3)
16.9152(4)
90
11.9323(5)
12.1577(8)
12.7144(8)
87.839(3)
87.957(4)
89.746(4)
1841.98(18)
2
11.2771(6)
11.4316(5)
11.6871(7)
80.495(3)
88.841(2)
62.631(3)
1316.96(13)
1
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
Z
90
90
120
120
5013.3(5)
6
4978.9(2)
6
Dcalcd (g cm−3
)
1.339
1.481
1.510
1.566
1.783
μ(Mo Kα) (mm−1
)
2.928
5.085
5.198
4.697
6.460
F(000)
2432
2232
2256
864
684
reflns collcd
indep reflns
348431
35783
42748
3391
40042
3370
102157
11241
9896
69151
8033
obsd reflns [I > 2σ(I)]
24518
2255
2885
6809
Rint
0.049
0.047
0.018
0.038
0.064
params refined
1169
98
95
342
255
R1
wR2
S
0.034
0.038
0.028
0.027
0.031
0.072
0.079
0.074
0.066
0.067
0.947
0.936
1.064
0.993
0.984
Δρmin (e Å−3
)
−1.15
0.65
−1.24
2.05
−0.86
0.45
−1.30
0.88
−1.46
1.50
Δρmax (e Å−3
)
6
7·0.5Et2O
9
10
chemical formula
C34H86F6N4O8S2Si8U2
1557.97
C63H133BN7O0.5Si12U2
1820.71
C36H86KN5O4Si6U
1098.77
monoclinic
C2/c
C32H78KN5O6Si6U
1074.66
monoclinic
P21
M (g mol−1
)
cryst syst
triclinic
triclinic
space group
a (Å)
P1
̅
P1
̅
11.8587(10)
11.9910(10)
12.5171(9)
92.187(5)
104.462(5)
114.001(4)
1554.7(2)
1
15.5570(10)
16.0199(10)
19.1374(17)
81.057(5)
89.273(5)
67.531(4)
4348.1(5)
2
19.4829(10)
11.9358(7)
24.8150(9)
90
11.4734(9)
18.469(2)
13.0501(14)
90
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
Z
108.710(3)
90
110.476(7)
90
5465.6(5)
4
2590.6(5)
2
Dcalcd (g cm−3
)
1.664
1.391
1.335
1.378
μ(Mo Kα) (mm−1
)
5.482
3.923
3.213
3.391
F(000)
764
1842
2256
1096
reflns collcd
indep reflns
85800
191955
16485
101389
8338
109242
9801
5913
obsd reflns [I > 2σ(I)]
5280
11815
6050
8022
Rint
0.045
0.060
0.042
0.057
params refined
300
827
251
571
R1
wR2
S
0.048
0.051
0.044
0.058
0.128
0.114
0.107
0.142
1.089
1.027
1.007
1.048
Δρmin (e Å−3
)
−1.49
3.10
−2.07
1.75
−1.08
1.17
−1.53
1.48
Δρmax (e Å−3
)
SiMe3), −18.08 (s, 36 H, SiMe3), −520.11 (s, 4H, CH2). Yellow
platelets of 6 were obtained by crystallization of 5 from THF.
Synthesis of [(UN*3)2(μ-CN)][BPh4] (7). A 50 mL flask was
charged with 1 (150 mg, 0.14 mmol) and NEt4CN (10.70 mg, 0.065
mmol), and THF (15 mL) was distilled in it under reduced pressure at
−78 °C. After stirring for 15 h at 20 °C, the yellow solution was
filtered and evaporated to dryness. The yellow solid was extracted in
Et2O (15 mL), and after evaporation of the solvent, the yellow powder
of 7 was washed with pentane (2 × 20 mL) and dried under vacuum
(74 mg, 65%). Anal. Calcd for C61H128BN7Si12U2: C, 41.08; H, 7.23;
N, 5.49. Found: C, 39.00; H, 6.50; N, 4.72. H NMR (THF-d8): δH
1
6.90, 6.66, 6.63 (m, 20H, BPh4), −6.75 (s, w1/2 = 74 Hz, 108H,
SiMe3). 13C{1H} NMR (THF-d8): δC 165.0 (q, JC−B = 50.2 Hz, ipso-
Ph), 137.0 (o-Ph), 125.5 (m-Ph), 121.6 (p-Ph), −9.2 (SiMe3). IR
(Nujol): ν(CN) 2115 cm−1. Yellow crystals of 7·0.5Et2O were
obtained by crystallization of 7 from Et2O at −33 °C.
Synthesis of [NEt4][UN*3(CN)2] (8). A 50 mL flask was charged
with 1 (307 mg, 0.30 mmol) and NEt4CN (97 mg, 0.60 mmol), and
O
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
THF (15 mL) was distilled in it under reduced pressure at −78 °C.
After stirring for 3 h at 20 °C, a white precipitate of [NEt4][BPh4] was
formed, and the pinkish solution was filtered and evaporated to
dryness, leaving a pale-pink powder. Complex 8 was extracted in a 1:5
mixture of THF and Et2O (15 mL) and isolated as a pink powder after
evaporation of the solution and drying under vacuum (156 mg, 58%).
Anal. Calcd for C28H74N6Si6U: C, 37.31; H, 8.27; N, 9.32. Found: C,
37.02; H, 8.21; N, 9.63. 1H NMR (THF-d8): δH 1.56 (s, 8H,
NCH2CH3), −0.16 (s, 12H, NCH2CH3), −1.40 (br s, w1/2 = 193 Hz,
solvent molecule is disordered around an inversion center, and it has
been refined with an occupancy factor of 0.5 and restraints on the
bond lengths and displacement parameters.
10. The three N* ligands are disordered over two positions each,
with N atoms common to both positions and also some common C
atoms. These two sets of positions correspond to the two orientations
of the helical arrangement, and they were given occupancy factors of
0.5. The exact location of the C and N atoms in the cyanide/
isocyanide anions cannot be determined from the values of the
displacement parameters, and both sites were refined with a mixture of
C and N, with a refined occupancy factor and constraints on
coordinates and displacement parameters. With the quality of the data
being quite low, many restraints were necessary, particularly in the
disordered parts.
1
54H, SiMe3). H NMR (pyridine-d5): δH 2.07 (q, J = 7.2 Hz, 8H,
NCH2CH3), 0.21 (t, J = 7.2 Hz, 12H, NCH2CH3), −0.39 (br s, w1/2
∼
200 Hz, 54H, SiMe3). 13C{1H} NMR (THF-d8): δC 50.7 (s,
NCH2CH3), 5.6 (s, NCH2CH3), −17.7 (s, SiMe3). IR (Nujol):
ν(CN) 2058 cm−1.
Crystal data and structure refinement parameters are given in Table
Synthesis of K[UN*3(CN)2] and Crystals of [K(THF)4]-
[UN*3(CN)2] (9). A 50 mL flask was charged with 1 (500 mg, 0.48
mmol) and KCN (313.5 mg, 4.8 mmol), and THF (15 mL) was
distilled in it under reduced pressure at −78 °C. After stirring for 15 h
at 20 °C, the yellow reaction mixture was filtered, and the solvent was
evaporated off to give a pale-green powder. Extraction in a 1:5 mixture
of THF and Et2O (15 mL), followed by evaporation to dryness,
afforded a pale-green powder, which was recrystallized from THF to
give pale-green crystals of 9. The latter were transformed into
K[UN*3(CN)2] upon drying under vacuum (243 mg, 63%). Anal.
9. The molecular plots were drawn with ORTEP-3.54
Computational Details. The methods and procedures were
described in our previous paper on U−CN versus Ce−NC
coordination in trivalent complexes derived from MN*3 (M = Ce, U).1
ASSOCIATED CONTENT
■
S
* Supporting Information
Tables of crystal data, atomic positions, and displacement
parameters, anisotropic displacement parameters, and bond
lengths and bond angles in CIF format, as well as DFT-
optimized geometries. This material is available free of charge
via the Internet at http://pubs.acs.org.
Calcd for C20H54KN5Si6U: C, 29.64; H, 6.72; N, 8.64. Found: C,
1
30.91; H, 6.88; N, 7.67. H NMR (THF-d8): δH −0.47 (br s, w1/2
=
=
1
186 Hz, 54H, SiMe3). H NMR (pyridine-d5): δH 0.32 (br s, w1/2
184 Hz, 54H, SiMe3). IR (Nujol): ν(CN) 2078 cm−1.
Formation of [UN*3(Br)2] by Oxidation of 9 with CuBr2. An
NMR tube was charged with K[UN*3(CN)2] (10.0 mg, 0.012 mmol)
and CuBr2 (3.60 mg, 0.016 mmol) in THF-d8 (0.5 mL). After 30 min
at 20 °C, the spectrum of the red solution showed the signals of
[U(N*)3(Br)2], and red crystals of this compound suitable for X-ray
diffraction were formed upon cooling of the solution at −33 °C.34
Crystals of [K(18-crown-6)][UN*3(CN)2] (10). An NMR tube was
charged with [UN*3] (20.0 mg, 0.028 mmol), KCN (9.40 mg, 0.14
mmol), and 18-crown-6 (7.39 mg, 0.028 mmol) in Et2O (0.5 mL).
The tube was immersed in an ultrasound bath for 6 h, and pyridine N-
oxide (2.66 mg, 0.028 mmol) was added to the dark-blue solution of
[K(18-crown-6)][UN*3(CN)],1 which immediately turned brown.
After 1 h at 20 °C, the brown precipitate in the red solution was
filtered off, and recrystallization in a 1:1 mixture of THF and Et2O
afforded pink crystals of 10 suitable for X-ray diffraction.
Crystallography. The data were collected at 150(2) K with a
Nonius Kappa CCD area detector diffractometer51 using graphite-
monochromated Mo Kα radiation (λ = 0.71073 Å). The crystals were
introduced into glass capillaries with a protective coating of Paratone-
N oil (Hampton Research). The unit cell parameters were determined
from 10 frames and then refined on all data. The data (combinations
of φ and ω scans with a minimum redundancy of 4 for 90% of the
reflections) were processed with HKL2000.52 Absorption effects were
corrected empirically with the program SCALEPACK.52 The structures
were solved by direct methods, expanded by subsequent difference
Fourier synthesis, and refined by full-matrix least squares on F2 with
SHELXL-97.53 All non-H atoms were refined with anisotropic
displacement parameters. In the case of compounds 2 and 9, it has
been possible to determine unambiguously the location of the C and
N atoms by selecting the solution giving the most satisfying refined
displacement parameters (i.e., close to one another for a bridging
cyanide or giving the most regular progression from metal to terminal
atom for a monodentate cyanide). In compounds 7 and 10, mixing of
the two cyanide orientations was considered (see below). The H
atoms were introduced at calculated positions (except for those bound
to C1 in 5, which were found on a difference Fourier map) and treated
as riding atoms with an isotropic displacement parameter equal to 1.2
times that of the parent atom (1.5 for CH3, with optimized geometry).
Special details are as follows:
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: jean-claude.berthet@cea.fr.
*E-mail: lotfi.belkhiri@umc.edu.dz.
*E-mail: abdou.boucekkine@univ-rennes1.fr.
*E-mail: michel.ephritikhine@cea.fr.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors are grateful to GENCI-IDRIS and GENCI-CINES
for an allocation of computing time (Grant 2013/2014-
080649).
REFERENCES
■
(1) Herve,
́
A.; Bouzidi, Y.; Berthet, J. C.; Belkhiri, L.; Thuer
́
y, P.;
Boucekkine, A.; Ephritikhine, M. Inorg. Chem. 2014, 53, 6995.
(2) (a) Bradley, D. C.; Chisholm, M. H. Acc. Chem. Res. 1976, 9, 273.
(b) Lappert, M. F.; Power, P. P.; Sanger, A. R.; Srivastava, R. C. C.
Metal and Metalloid Amides; Ellis Horwood: Chichester, England,
1980. (c) Edelmann, F. In Comprehensive Organometallic Chemistry;
Abel, E. W., Stone, F. G. A., Wilkinson, G., Eds.; Pergamon: Oxford,
U.K., 1995; Vol. 4, Chapter 2, p 11. (d) Burns, C. J.; Eisen, M. S. In
The Chemistry of the Actinides and Transactinide Elements; Morss, L. R.,
Edelstein, N. M., Fuger, J., Eds.; Springer: Dordrecht, The Nether-
lands, 2006; Vol. 5, Chapter 25, p 2799. (e) Lappert, M. F.;
Protchenko, A. V.; Power, P. P.; Seeber, A. L. Metal Amide Chemistry;
Wiley: Chichester, England, 2009.
(3) Thomson, R. K.; Graves, C. R.; Scott, B. L.; Kiplinger, J. L. Inorg.
Chem. Commun. 2011, 14, 1742.
(4) (a) Simpson, S. J.; Turner, H. W.; Andersen, R. A. Inorg. Chem.
1981, 20, 2991. (b) Karl, M.; Harms, K.; Seybert, G.; Massa, W.; Fau,
S.; Frenking, G.; Dehnicke, K. Z. Anorg. Allg. Chem. 1999, 625, 2055.
(c) Han, F.; Zhang, J.; Yi, W.; Zhang, Z.; Yu, J.; Weng, L.; Zhou, X.
Inorg. Chem. 2010, 49, 2793.
7·0.5Et2O. The cyanide ligand is disordered over the two possible
positions, which were given occupancy factors of 0.5 and refined with
constraints on coordinates and displacement parameters. The Et2O
(5) Ben
Chem. 2010, 49, 8117.
́
aud, O.; Berthet, J. C.; Thuer
́
y, P.; Ephritikhine, M. Inorg.
P
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(6) Ben
Chem. 2011, 50, 12204.
(7) Maynadie, J.; Berthet, J. C.; Thuer
Organometallics 2007, 26, 4585.
(8) Lewis, A. J.; Williams, U. J.; Carroll, P. J.; Schelter, E. J. Inorg.
Chem. 2013, 52, 7326.
(9) Dormond, A.; El Bouadili, A.; Aaliti, A.; Moise, C. J. Organomet.
Chem. 1985, 288, C1.
(10) Berthet, J. C.; Boisson, C.; Lance, M.; Vigner, J.; Nierlich, M.;
Ephritikhine, M. J. Chem. Soc., Dalton Trans. 1995, 3019.
(11) Evans, W. J.; Nyce, G. W.; Forrestal, K. J.; Ziller, J. W.
Organometallics 2002, 21, 1050.
(12) (a) Gomberg, M. Ber. Dtsch. Chem. Ges. 1900, 33, 3150.
(b) Lankamp, H.; Nauta, W. T.; MacLean, C. Tetrahedron Lett. 1968,
9, 249.
(13) Arnold, P. L.; Turner, Z. R.; Kaltsoyannis, N.; Pelekanaki, P.;
Bellabarba, R. M.; Tooze, R. P. Chem.Eur. J. 2010, 16, 9623.
(14) Berthet, J. C.; Nierlich, M.; Ephritikhine, M. Eur. J. Inorg. Chem.
2002, 2002, 850.
́
aud, O.; Berthet, J.-C.; Thuer
́
y, P.; Ephritikhine, M. Inorg.
(39) Maynadie,
Organometallics 2007, 26, 2623.
(40) Berthet, J.-C.; Thuery, P.; Ephritikhine, M. Chem. Commun.
2007, 604.
(41) (a) Wu, Q. Y.; Wang, C. Z.; Lan, J. H.; Xiao, C. L.; Wang, X. K.;
Zhao, Y. L.; Chai, Z. F.; Shi, W. Q. Inorg. Chem. 2014, 53, 9607.
(b) Averkiev, B.; Mantina, M.; Valero, R.; Infante, I.; Kovacs, A.;
Truhlar, D.; Gagliardi, L. Theor. Chem. Acc. 2011, 129, 657.
́
J.; Berthet, J. C.; Thuer
́
y, P.; Ephritikhine, M.
́
́
́
y, P.; Ephritikhine, M.
́
(42) (a) Roger, M.; Belkhiri, L.; Arliguie, T.; Thuery, P.; Boucekkine,
A.; Ephritikhine, M. Organometallics 2008, 27, 33. (b) Gaunt, A. J.;
Reilly, S. D.; Enriquez, A. E.; Scott, B. L.; Ibers, J. A.; Sekar, P.; Ingram,
K. I. M.; Kaltsoyannis, N.; Neu, M. P. Inorg. Chem. 2008, 47, 29.
(c) Gaunt, A. J.; Reilly, S. D.; Enriquez, A. E.; Scott, B. J.; Ibers, J. A.;
Sekar, P.; Ingram, K. I. M.; Kaltsoyannis, N.; Neu, M. P. Inorg. Chem.
2008, 47, 29. (d) Roger, M.; Belkhiri, L.; Thuer
Boucekkine, A.; Ephritikhine, M. Inorg. Chem. 2009, 48, 221.
(e) Meskaldji, S.; Belkhiri, L.; Arliguie, T.; Fourmigue, M.;
́
y, P.; Bouaoud, S. E.;
́
Ephritikhine, M.; Boucekkine, A. Inorg. Chem. 2010, 49, 3192.
(f) Meskaldji, S.; Zaiter, A.; Belkhiri, L.; Boucekkine, A. Theor.
Chem. Acc. 2012, 131, 1151. (g) Zaiter, A.; Boudersa, A.; Bouzidi, Y.;
Meskaldji, S.; Belkhiri, L.; Boucekkine, A.; Ephritikhine, M. Inorg.
Chem. 2014, 53, 4687−4697.
(15) Turner, H. W.; Andersen, R. A.; Zalkin, A.; Templeton, D. H.
Inorg. Chem. 1979, 18, 1221.
(16) (a) Williams, U. J.; Robinson, J. R.; Lewis, A. J.; Carroll, P. J.;
Walsh, P. J.; Schelter, E. J. Inorg. Chem. 2014, 53, 27. (b) Stewart, J. L.
Tris[bis(trimethylsilyl)amido]uranium: Compounds with Tri-, Tetra-, and
Pentavalent Uranium; Lawrence Berkeley Laboratory: Berkeley, CA,
1988.
(43) (a) Sonnenberg, J. L.; Hay, P. J.; Martin, R. L.; Bursten, B. E.
Inorg. Chem. 2005, 44, 2255. (b) Pae
́
z-Hernan
́
dez, D.; Ramírez-Tagle,
R.; Codorniu-Hernandez, E.; Montero-Cabrera, L. A.; Arratia-Per
́
́
ez, R.
Polyhedron 2010, 29, 975. (c) Neidig, M. L.; Clark, D. L.; Martin, R. L.
Coord. Chem. Rev. 2013, 257, 394.
(17) Fortier, S.; Brown, J. L.; Kaltsoyannis, N.; Wu, G.; Hayton, T.
W. Inorg. Chem. 2012, 51, 1625.
(44) Mayer, I. Chem. Phys. Lett. 1983, 7, 270.
(45) (a) Nalewajski, R. F.; Mrozek, J. Int. J. Quantum Chem. 1994, 51,
187. (b) Nalewajski, R. F.; Mrozek, J.; Michalak, A. Int. J. Quantum
Chem. 1997, 61, 589.
(18) Hitchcock, P. B.; Hulkes, A. G.; Lappert, M. F. Inorg. Chem.
2004, 43, 1031.
(19) Airoldi, C.; Bradley, D. C.; Chudzynska, H.; Hursthouse, M. B.;
Abdul Malik, K. M.; Raithby, P. R. J. Chem. Soc., Dalton Trans. 1980,
2010.
(20) Williams, U. J.; Carroll, P. J.; Schelter, E. J. Inorg. Chem. 2014,
53, 6338.
(21) Nakai, H.; Hu, X.; Zakharov, L. N.; Rheingold, A. L.; Meyer, K.
Inorg. Chem. 2004, 43, 855.
(22) Shannon, R. D. Acta Crystallogr., Sect. A 1976, 32, 751.
(23) Ren, W.; Zi, G.; Fang, D.-C.; Walter, M. D. J. Am. Chem. Soc.
2011, 133, 13183.
(24) Zi, G.; Jia, L.; Werkema, E. L.; Walter, M. D.; Gottfriedsen, J. P.;
Andersen, R. A. Organometallics 2005, 24, 4251.
(25) Natrajan, L.; Mazzanti, M.; Bezombes, J.-P.; Pec
Chem. 2005, 44, 6115.
(26) Evans, W. J.; Takase, M. K.; Ziller, J. W.; Rheingold, A. L.
Organometallics 2009, 28, 5802.
(27) Kiplinger, J. L.; John, K. D.; Morris, D. E.; Scott, B. L.; Burns, C.
(46) (a) Michalak, A.; De Kock, R. L.; Ziegler, T. J. Phys. Chem. A
2008, 112, 7256. (b) Curley, J. J.; Piro, N. A.; Cummins, C. C. Inorg.
Chem. 2009, 48, 9599. (c) Mills, D. P.; Cooper, O. J.; Tuna, F.;
McInnes, E. J. L.; Davies, E. S.; McMaster, J.; Moro, F.; Lewis, W.;
Blake, A. J.; Liddle, S. T. J. Am. Chem. Soc. 2012, 134, 10047. (d) Patel,
D.; McMaster, J.; Lewis, W.; Blake, A. J.; Liddle, S. T. Nat. Commun.
2013, 4, 2323. (e) King, D. M.; Tuna, F.; McInnes, E. J. L.; McMaster,
J.; Lewis, W.; Blake, A. J.; Liddle, S. T. Nat. Chem. 2013, 5, 482.
(47) (a) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988,
88, 899. (b) Bader, R. F. W. Atoms in Molecules: A Quantum Theory;
OUP: Oxford, U.K., 1990. (c) Mulliken, R. S. J. Chem. Phys. 1955, 23,
1833. (d) Mountain, A. R. E.; Kaltsoyannis, N. Dalton Trans. 2013, 42,
13477. (e) Jones, M. B.; Gaunt, A. J.; Gordon, J. C.; Kaltsoyannis, N.;
Neu, M. P.; Scott, B. L. Chem. Sci. 2013, 4, 1189. (f) Schnaars, D. D.;
Gaunt, A. J.; Hayton, T. W.; Jones, M. B.; Kirker, I.; Kaltsoyannis, N.;
May, I.; Reilly, S. D.; Scott, B. L.; Wu, G. Inorg. Chem. 2012, 51, 8557.
(g) Vlaisavljevich, B.; Miro, P.; Cramer, C. J.; Gagliardi, L.; Infante, I.;
Liddle, S. T. Chem.Eur. J. 2011, 17, 8424. (h) Arnold, P. L.; Turner,
Z. R.; Kaltsoyannis, N.; Pelekanaki, P.; Bellabarba, R. M.; Tooze, R. P.
Chem.Eur. J. 2010, 16, 9623. (i) Petit, L.; Adamo, C.; Maldivi, P.
Inorg. Chem. 2006, 45, 8517. (j) Manna, D.; Mula, S.; Bhattacharyy, A.;
Chattopadhyay, S.; Ghanty, T. K. Dalton Trans. 2015, 44, 1332.
́
aut, J. Inorg.
J. Organometallics 2002, 21, 4306.
(28) Evans, W. J.; Walensky, J. R.; Furche, F.; Ziller, J. W.;
DiPasquale, A. G.; Rheingold, A. L. Inorg. Chem. 2008, 47, 10169.
(29) Korobkov, I.; Gambarotta, S. Inorg. Chem. 2010, 49, 3409.
(30) Niemeyer, M. Inorg. Chem. 2006, 45, 9085.
(31) Yu, X.; Bi, S.; Guzei, I. A.; Lin, Z.; Xue, Z.-L. Inorg. Chem. 2004,
43, 7111.
(32) Berno, P.; Minhas, R.; Hao, S.; Gambarotta, S. Organometallics
1994, 13, 1052.
(33) Messere, R.; Spirlet, M. R.; Jan, D.; Demonceau, A.; Noels, A. F.
Eur. J. Inorg. Chem. 2000, 1151.
(34) Lewis, A. J.; Nakamaru-Ogiso, E.; Kikkawa, J. M.; Carroll, P. J.;
(k) Rodríguez, J. I.; Koster, A. M.; Ayers, P. W.; Santos-Valle, A.; Vela,
̈
A.; Merino, G. J. Comput. Chem. 2009, 30, 1082. (l) Rodríguez, J. I.;
Bader, R. F. W.; Ayers, P. W.; Michel, C.; Gotz, A. W.; Bo, C. Chem.
̈
Phys. Lett. 2009, 472, 149.
(48) Rayon
J. Phys. Chem. A 2007, 111, 6334.
(49) Rayon, V. M.; Redondo, P.; Valdes
J. Chem. Phys. 2009, 131, 094507.
́
, V. M.; Redondo, P.; Valdes
́
, H.; Barrientos, C.; Largo, A.
́
́
, H.; Barrientos, C.; Largo, A.
Schelter, E. J. Chem. Commun. 2012, 48, 4977.
(35) Lewis, A. J.; Mullane, K. C.; Nakamaru-Ogiso, E.; Carroll, P. J.;
Schelter, E. J. Inorg. Chem. 2014, 53, 6944.
(50) (a) Andersen, R. A. Inorg. Chem. 1979, 18, 1507. (b) Monreal,
M. J.; Thomson, R. K.; Cantat, T.; Travia, N. E.; Scott, B. L.; Kiplinger,
J. L. Organometallics 2011, 30, 2031.
(51) Hooft, R. W. W. COLLECT; Nonius BV: Delft, The
Netherlands, 1998.
(52) Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307.
(53) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112.
(54) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
(36) Maynadie,
Ephritikhine, M. Angew. Chem., Int. Ed. 2007, 46, 2010.
́ ́
J.; Barros, N.; Berthet, J. C.; Thuery, P.; Maron, L.;
(37) Berthet, J. C.; Thuer
2008, 27, 1664.
(38) Herve,
́
y, P.; Ephritikhine, M. Organometallics
́
A.; Thuer
́
y, P.; Ephritikhine, M.; Berthet, J.-C.
Organometallics 2014, 33, 2088.
Q
DOI: 10.1021/acs.inorgchem.5b00034
Inorg. Chem. XXXX, XXX, XXX−XXX