Alkaline-Earth-Metal-Induced Liberation of Rare Allotropes of Elemental Silicon and Germanium from N-Heterocyclic Metallylenes

Article abstract of DOI:10.1021/acs.inorgchem.5b01643

The synthesis and striking reactivity of the unprecedented N-heterocyclic silylene and germylene ("metallylene") alkaline-earth metal (Ae) complexes of the type [(η⁵-C₅Me₅)₂Ae←:E(N^tBuCH)₂] (3, 4, and 7-9; Ae = Ca, E = Ge 3; Ae = Sr, E = Ge 4; Ae = Sr, E = Si 7; Ae = Ba, E = Si 8; Ae = Ba, E = Ge 9) are reported. All complexes have been characterized by spectroscopic means, and their bonding situations investigated by density functional theory (DFT) methods. Single-crystal X-ray diffraction analyses of examples revealed relatively long Si-Ae and Ge-Ae distances, respectively, indicative of weak E:→Ae (E = Si, Ge) dative bonds, further supported by the calculated Wiberg bond indices, which are rather low in all cases (～0.5). Unexpectedly, the complexes undergo facile transformation to 1,4-diazabuta-1,3-diene Ae metal complexes of the type [(η⁵-C₅Me₅)₂Ae(κ²-{N^tBu = CHCH = N^tBu})] (Ae = Sr 10, Ae = Ba 11) or in the case of calcium to the dinuclear complex [(η⁵-C₅Me₅)₂Ca←:N(^tBu) = CHCH = (^tBu)N:→Ca(η⁵-C₅Me₅)₂] (12) under concomitant liberation of elemental silicon and germanium. The formation of elemental silicon and germanium is proven by inductively coupled plasma atomic emission spectroscopy, transmission electron microscopy, selected area electron diffraction, and energy dispersive X-ray spectroscopy. Notably, the decomposition of the Si(II)→Ba complex 8 produces allo-silicon, a rare allotropic form of elemental silicon. Similarly, the analogous Ge(II)→Ba complex 9, upon decomposition, forms tetragonal germanium, a dense and rare allotrope of elemental germanium. The energetics of this unprecedented alkaline-earth-metal-induced liberation of elemental silicon and germanium was additionally studied by DFT methods, revealing that the transformations are pronouncedly exergonic and considerably larger for the N-heterocyclic germylene complexes than those of the corresponding silicon analogues. (Figure Presented).

Full text of DOI:10.1021/acs.inorgchem.5b01643

Article
pubs.acs.org/IC
Alkaline-Earth-Metal-Induced Liberation of Rare Allotropes of
Elemental Silicon and Germanium from N‑Heterocyclic Metallylenes
Burgert Blom,*,† Amro Said,† Tibor Szilvasi,‡ Prashanth W. Menezes,† Gengwen Tan,†
́
Judith Baumgartner,§ and Matthias Driess*,†
†Department of Chemistry: Metalorganics and Inorganic Materials, Technische Universitat Berlin, Strasse des 17. Juni 135, Sekr. C2,
̈
D-10623 Berlin, Germany
‡Department of Inorganic and Analytical Chemistry, Budapest University of Technology and Economics (BUTE), Szent Geller
́ ́
t ter 4,
1111 Budapest, Hungary
§Institut fur Chemie, Universitat Graz, Stremayrgasse 9, 8010 Graz, Austria
̈
̈
S
* Supporting Information
ABSTRACT: The synthesis and striking reactivity of the
unprecedented N-heterocyclic silylene and germylene (“metal-
lylene”) alkaline-earth metal (Ae) complexes of the type [(η5-
C5Me5)2Ae←:E(NtBuCH)2] (3, 4, and 7−9; Ae = Ca, E = Ge
3; Ae = Sr, E = Ge 4; Ae = Sr, E = Si 7; Ae = Ba, E = Si 8; Ae =
Ba, E = Ge 9) are reported. All complexes have been
characterized by spectroscopic means, and their bonding
situations investigated by density functional theory (DFT)
methods. Single-crystal X-ray diffraction analyses of examples
revealed relatively long Si−Ae and Ge−Ae distances,
respectively, indicative of weak E:→Ae (E = Si, Ge) dative
bonds, further supported by the calculated Wiberg bond
indices , which are rather low in all cases (∼0.5). Unexpectedly, the complexes undergo facile transformation to 1,4-diazabuta-1,3-
diene Ae metal complexes of the type [(η5-C5Me5)2Ae(κ2-{NtBuCHCHNtBu})] (Ae = Sr 10, Ae = Ba 11) or in the case of
calcium to the dinuclear complex [(η5-C5Me5)2Ca←:N(tBu)CHCH(tBu)N:→Ca(η5-C5Me5)2] (12) under concomitant
liberation of elemental silicon and germanium. The formation of elemental silicon and germanium is proven by inductively
coupled plasma atomic emission spectroscopy, transmission electron microscopy, selected area electron diffraction, and energy
dispersive X-ray spectroscopy. Notably, the decomposition of the Si(II)→Ba complex 8 produces allo-silicon, a rare allotropic
form of elemental silicon. Similarly, the analogous Ge(II)→Ba complex 9, upon decomposition, forms tetragonal germanium, a
dense and rare allotrope of elemental germanium. The energetics of this unprecedented alkaline-earth-metal-induced liberation of
elemental silicon and germanium was additionally studied by DFT methods, revealing that the transformations are pronouncedly
exergonic and considerably larger for the N-heterocyclic germylene complexes than those of the corresponding silicon analogues.
electrons (resonance form II in Chart 1). This description
has subsequently been largely abandoned in contemporary
literature and differs substantially from the conventionally
accepted bonding description (structure I in Chart 1), where
the group 14 element atom is formally in the oxidation state
+II, bearing only one lone pair of electrons. More recently, the
existence of so-called ylidones (structure III, Chart 1) has been
predicted by Frenking and co-workers7 and could subsequently
be synthetically obtained, bearing “authentic” zerovalent group
14 atoms, with two lone pairs of electrons residing at the group
14 element atoms.8 The latter class of compounds offer the
potential of being possible molecular precursors for zerovalent
Ge and Si atoms, respectively, by ligand dissociation.
INTRODUCTION
■
Since the landmark discoveries of the first N-heterocyclic
germylene (NHGe)1 and the corresponding N-heterocyclic
silylene (NHSi)2 shortly thereafter, a plethora of d-block metal
complexes bearing NHSis3 and, to a lesser extent, NHGes as
supporting ligands4 have been reported. Both classes of
complexes represent more electron-rich analogues of traditional
N-heterocyclic carbene (NHC) complexes, offering pronoun-
cedly larger σ-donor capacities and, concomitantly, different
reactivities.5 Early investigations dating back to 1994 on first
NHSi and NHGe by photoelectron spectroscopy (PES)6
revealed that on descent from carbon to its heavier
homologues, the group 14 element atom tends to adopt
zerovalent character as the HOMOs of the NHSi and NHGe
become more centered on the group 14 element atom, which
itself can be thought of as being chelated by a neutral 1,4-
diazabuta-1,3-diene ligand and bearing two lone pairs of
Received: July 25, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.5b01643
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
rather strong donor−acceptor interactions generally persist
between the Si and metal centers and in some cases even
exhibit WBIs greater than unity.13
Chart 1. Donor−Acceptor Description of N-Heterocyclic
a
Metallylenes I↔II and Metallylones III (E = Si, Ge)
In the present report, the first examples of N-heterocyclic
germylene alkaline-earth-metal (Ae) complexes (Ae = Ca, Sr,
and Ba) are reported. In addition, the hitherto unprecedented
NHSi complexes of Sr and Ba are also reported, with a
particular focus on the trends in the bonding interaction
between Si or Ge and the Ae metals. Strikingly, the
coordination of the NHSi and NHGe to Ae metals induces
an unprecedented facile transformation process, affording
elemental silicon and germanium and the corresponding 1,4-
diazabuta-1,3-diene Ae metal complexes as decomposition
products. Our results confirm for the first time that NHSis and
NHGes can serve as molecular precursors for elemental silicon
and germanium, respectively.
a
The resonance forms II and III bear zero-valent silicon and
germanium atoms, respectively.
In recent years, our group has explored a variety of NHSis
and NHGes and their capacity to coordinate to a variety of
metals to afford novel complexes with a keen interest in
harnessing their potential in catalysis or other transformations.9
As part of these efforts, we recently reported facile entry to
novel s-block metal complexes bearing three- and four-
coordinate NHSi ligands, namely, [(η5-C5Me5)2Ca←:Si-
(NtBuCH)2] and [(η5-C5Me5)2Ca←:Si(O−C6H4-2-tBu)-
{(NtBu)2CPh}], respectively (Chart 2).10 On the basis of
RESULTS
■
Following a similar synthetic approach applied previously to
access [(η5-C5Me5)2Ca←:Si(NtBuCH)2] and [(η5-
C5Me5)2Ca←:Si(O−C6H4-2-tBu){(NtBu)2CPh}], respec-
tively,10 the reaction of [Ca(η5-C5Me5)2] (1) with Ge-
(NtBuCH)2 (2)1 in toluene at room temperature affords the
desired [(η5-C5Me5)2Ca←:Ge(NtBuCH)2] adduct (3) in
quantitative yields as a colorless solid (Scheme 1). Complex
3 represents the first example of an s-block NHGe complex.
Extension to the heavier homologues of Ca (Sr and Ba) follows
in a similar manner, and under analogous reaction conditions
the complex [(η5-C5Me5)2Sr←:Ge(NtBuCH)2] (4) could also
be isolated upon reaction of [Sr(η5-C5Me5)2] (5)14 with NHGe
2. The reaction of the base-stabilized (η5-C5Me5)2Sr(OEt2) (5·
OEt2)15,16 complex withSi(NtBuCH)2 (6) affords [(η5-
C5Me5)2Sr←:Si(NtBuCH)2] (7).
Chart 2. First Examples of s-Block N-Heterocyclic Silylene
Complexes10
Notably, the ether donor molecule in the starting material (5·
OEt2) cannot be replaced by the NHGe but by the NHSi ligand
due to the weaker σ-donor character of Ge(II) vs Si(II). Finally,
the barium analogues [(η5-C5Me5)2Ba←:E(NtBuCH)2] (E = Si
8; E = Ge 9) could also be prepared reacting the base-free
Ba(η5-C5Me5)2 (5) with the NHSi 6 and NHGe 2, respectively
(Scheme 1).
The high-resolution mass spectra (HR-ESI or HR-APCI) did
not show the parent (M + H+) peak and generally only the free
NHSi or NHGe could be observed. This is very likely due to
reactivity and DFT investigations it was shown that the Si:→Ca
bonding interaction in both complexes is rather weak and
characterized by a strong ionic component, readily seen in
Bader’s atoms in molecules (AIM) analysis.11 Moreover, rather
low Wiberg bond indices (WBIs)12 of ∼0.5 are found for both
complexes, revealing only minor covalent character in the
donor−acceptor Si:→Ca bond. These findings contrast
strongly with the plethora of d-block NHSi complexes where
Scheme 1. Facile Entry to the Alkaline-Earth-Metal NHSi and NHGe Complexes 3, 4, and 7−9 by Simple Coordination or
Ligand Displacement Reactions
B
DOI: 10.1021/acs.inorgchem.5b01643
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1
Table 1. Some Selected 29Si and H NMR Shifts in Complexes 3, 4, and 7−9 Compared with the “Free” NHGe 2 or NHSi 6
1
1
1
complex
29Si, ppm
δ (CHCH), H, ppm
δ (C(CH3)3), H, ppm
δ (C5(CH3)5), H, ppm
3
6.99
7.05
6.64
6.74
7.05
7.05
6.75
1.40
1.43
1.34
1.40
1.42
1.43
1.40
1.98
2.12
2.12
2.06
2.06
4
7
80.7
8
78.5 (Δν1/2 = 11 Hz)
9
NHGe 2
NHSi 6
78.1
the fragility of the E:→Ae coordination (E = Si, Ge; Ae = Ca,
Sr, Ba), reminiscent of the calcium NHSi complexes reported
earlier.10 Microanalyses were attempted on the complexes, but
owing to their lability, the results were not repeatable and large
variations in the C, H values were observed between
measurements. Even enclosing the samples in two tin capsules
did not work, likely due to their extreme sensitivity. Similar
problems have been reported for alkaline-earth-metal com-
plexes by Burns et al.18
complex 3 over a longer period (>1 week) was carried out. This
procedure however afforded red crystals with a data set of
higher precision, but contained two molecules in the
asymmetric unit, namely, complex 3 and a decomposition
product: the dinuclear 1,4-diaza-1,3-butadiene complex [(η5-
t
C5Me5)2Ca←:NtBuCHCH BuN:→Ca(η5-C5Me5)2] (12)
(Figures 1 and 2).
In the multinuclear (1H, 13C{1H}, 29Si{1H}) NMR spectra of
all isolated complexes, their chemical shifts are only very
marginally shifted from the starting materials or even identical
(Table 1), indicative of a very weak coordination, akin to what
was earlier observed for the complexes [(η5-C5Me5)2Ca←
:Si(NtBuCH)2] and [(η5-C5Me5)2Ca←:Si(O−C6H4-2-tBu)-
{(NtBu)2CPh}], respectively.10 Accordingly, the 29Si NMR
chemical shift of the Sr complex 7 occurs at roughly the same
value as that of its calcium analogue reported earlier, indicating
no or a very slow exchange process on the NMR time scale of
the measurement. In contrast, the 29Si NMR shift of the barium
complex 8 is broadened (Δν1/2 = 11 Hz) and at roughly at the
same chemical shift position of the free NHSi, suggestive of a
detectible exchange process on the time scale of the 29Si NMR
measurement.17 This situation is similar to what was observed
for a series of [(η5-C5Me5)2Ae←:PEt3] complexes (Ae = Ca, Sr,
Ba) reported by Burns et al.,18 where a detectible exchange
process was found for both Sr and Ba, suggestive of slightly
weaker coordination in solution, compared with its calcium
analogue.
Figure 1. ORTEP representation of 3 in the solid state as one of the
molecules in the asymmetric unit alongside the decomposition product
12. Thermal ellipsoids are set at the 50% probability level; H atoms are
omitted for clarity. Selected distances [Å]: Ca(1)−Ge(1) 3.1174(11),
Ge(1)−N(1) 1.844(4), Ge(1)−N(2) 1.843(4). Selected angles [deg]:
N(1)−Ge(1)−Ca(1) 134.76(15), N(2)−Ge(1)−Ca(1) 138.66(11),
N(2)−Ge(1)−N(1) 85.51(18).
In order to further probe this exchange phenomenon, NMR-
scale experiments were conducted where two equivalents of the
NHSi were added to one equivalent of the [(η5-C5Me5)2Ae]
1
(Ae = Ca, Ba) or [(η5-C5Me5)2Sr(OEt2)], respectively, and H
and 29Si spectra were recorded at room temperature. In all
cases, in the 1H NMR spectra, only one set of sharp resonance
signals was detected, despite the presence of two equivalents of
the NHSi, roughly in between the resonance signals of the free
NHSi and that of the isolated complex, signifying rapid
exchange (see Supporting Information). Moreover, in the 29Si
NMR spectra only one resonance signal was observed in all
three cases, also demonstrating exchange dynamics in solution.
These results collectively and conclusively show that the
coordination of the NHE (E = Si, Ge) to the alkaline-earth
metal is labile and has the propensity to readily undergo facile
exchange in solution.
Obtaining crystals suitable for single-crystal X-ray structural
investigations of the complexes proved very difficult. In the case
of 3 suitable colorless crystals could be obtained, and the
structure solution revealed two independent molecules of 3 in
the asymmetric unit (see Figure S13, Supporting Information).
This data set was however of moderate precision (R1 all data =
0.3195; wR2 all data = 0.3534), so another crystallization of
Figure 2. ORTEP representation of 12 in the solid state. Thermal
ellipsoids are set at the 50% probability level; H atoms omitted for
clarity. Selected bond lengths [Å]: Ca(2)−N(3) 2.610(4), N(3)−
C(51) 1.262(7), C(51)−C(51i) 1.471(9).
C
DOI: 10.1021/acs.inorgchem.5b01643
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. ORTEP representation of 7 in the solid state (left). Thermal ellipsoids are set at the 50% probability level; H atoms omitted for clarity.
Optimized structure of complex 7 derived by DFT methods B97-D/def2-TZVP[C,N,H:6-31G(d)]. Calculated Si−Sr bond length: 3.272 Å (gold =
Sr; cyan = Si; dark blue = N; gray = C.).
Complex 3 is the first example of an s-block NHGe complex,
which is structurally analogous with the silicon analogue
reported earlier.10 The complex features tilted Cp* rings with
the NHGe ligand occupying a free coordination site. The Ge−
Ca distance of 3.1174(11) Å is rather long, akin to what was
observed for the Si analogue and longer than the values
observed for existing calcium germanide complexes, which are
exceptionally rare in their own right.19 The complex [Ca-
(thf)3{Ge(TMS)3}2] (TMS = trimethylsilyl), by Teng and
Ruhlandt-Senge, for example, features an average Ca−Ge
distance of 3.022 Å. The notably long bond length supports the
notion that the bonding situation between Ge and Ca is likely a
very weak donor−acceptor interaction. The Ge center in 3 is
almost trigonal planar with the sum of bond angles around the
Ge atom of 358.93°.
As mentioned above the asymmetric unit also contained the
decomposition product [(η5-C5Me5)2Ca←:N(tBu)
CHCH(tBu)N:→Ca(η5-C5Me5)2], 12. Remarkably, the
mode of decomposition from 3 to 12 clearly involves
unprecedented liberation of elemental germanium (see below
for detailed discussion). The structure of 12 is shown in Figure
2.
Compound 12 possesses a mirror plane between C51 and
C51i in the asymmetric unit. The N3−C51 distance is 1.262(7)
Å and, hence, represents a double bond. According to the
metric parameters, 12 is a dicalcium(II) complex with a
bridging neutral 1,4-diazabuta-1,3-diene ligand. The Ca−N
bond length is comparable to, albeit slightly longer than, the
related ansa-metallocene complex from Edelmann and co-
workers, [Ca{Me4C2-(η5-C5H4)}(κ2-{NtBuCHCH
NtBu}]: Ca−N = 2.503(6) and 2.564(6) Å, respectively.20
Compound 12 was synthesized independently by the reaction
of 1 with the respective 1,4-diazabuta-1,3-diene ligand, tBuN
C(H)−C(H)NtBu, in the molar ratio of 2:1 (see below).
For complex 7, substantial difficulties arose in trying to
obtain suitable crystals for a single-crystal X-ray diffraction
analysis. Because the crystal quality of 7 turned out to be
moderate and led only to a medium precision of metric
parameters, DFT calculations at the B97-D/def2-TZVP-
[C,N,H:6-31G(d)] level of theory were performed (see
below). The connectivity of 3 as determined by an X-ray
diffraction analysis is in accordance with the results by DFT
calculations (Figure 3). The calculated Si−Sr bond length is
3.272 Å, which again falls outside the sum of the single-bond
covalent radii of both elements (Sr = 0.185 Å; Si = 0.116 Å),22
indicative of a weak donor−acceptor interaction. To our
knowledge, only two examples of molecular complexes bearing
Sr−Si bonds were reported by Teng and Ruhland-Senge:
[Sr(thf)3{Si(TMS)3}2], where the average Sr−Si bond length is
3.196 Å, and [Sr(tmeda) (thf){Si(TMS)3}2] (tmeda =
N,N,N′,N′-tetramethylethylenediamine), which features an
average Sr−Si bond length of 3.223 Å.21
In close analogy with the NHGe→Ca complex 3, complex 7
also undergoes repeatable decomposition to a 1,4-diazabuta-
1,3-diene Sr(II) complex upon prolonged standing in solution
(>1 week) to give [(η5-C5Me5)2Sr(κ2-{NtBuCHCH
NtBu})], 10. This corresponds with concomitant extrusion of
elemental silicon, akin to what was observed for 3, and
highlights the apparently similar lability of complexes 3 and 7.
In contrast to the decomposition product of 3, which is
dinuclear, complex 10 is mononuclear and bears only one
chelating 1,4-diazabuta-1,3-diene ligand in κ2-fashion. The
probable reason for the difference in coordination mode of
the ligand is the increased ion radius of Sr(II) vs Ca(II),22
which facilitates κ2-coordination toward Sr(II). The molecular
structure of 10 is shown in Figure 4. Noteworthy is the rather
acute bite angle of the chelating C2N2 ligand (62.36°). The Sr−
Figure 4. ORTEP representation of 10. Thermal ellipsoids are set at
the 50% probability level; H atoms are omitted for clarity. Selected
bond lengths [Å]: Sr(1)−N(1) 2.7451(13), Sr(1)−N(2) 2.7538(13).
Selected bond angle [deg]: N(1)−Sr(1)−N(2) 62.36(4).
D
DOI: 10.1021/acs.inorgchem.5b01643
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 2. Direct Syntheses of 10−12
Figure 5. Boundary surface plots of the frontier orbitals (from left to right) in complex 4: (HOMO, −3.64 eV); (LUMO, −1.69 eV) and 8:
(HOMO, −3.53 eV); (LUMO, −1.56 eV). Complexes were optimized at the B97-D/def2-TZVP[C,N,H:6-31G(d)] level of theory.
N1 and Sr−N2 bond lengths (2.7451(13), 2.7538(13) Å,
respectively) are slightly longer in comparison to the only
existing complex with which a meaningful comparison can be
made: Sr(η5-C5Me5)2(κ2-{Bipy}) (Bipy = bipyridine), where
Sr−N = 2.624(3) and 2.676(3) Å, respectively.23
DFT analyses at the B97-D/def2-TZVP[C,N,H:6-31G(d)]
level of theory were carried out for complexes 3, 4, 7, 8, and 9
in order to gain insights into the bonding situation between the
group 14 element and the alkaline-earth metal; in addition, the
silicon and germanium liberation processes with corresponding
formation of the respective 1,4-diazabuta-1,3-diene Ae com-
plexes were investigated computationally.
In the NHE:→Ae complexes (3, 4, 7−9) the HOMO is
localized on the Cp* rings, with some delocalization over the
metal center (Figure 5 and Supporting Information). The
LUMO is in all cases located on the coordinated metallylene
ligand. This picture resembles the situation of the calcium
complexes studied previously.10
A donor−acceptor interaction from the divalent group 14
site to the Ae cation is clearly visible in each complex and
represented by the HOMO−5 of complex 3, where some
orbital contribution from the calcium center is also involved
(Figure 6 and Supporting Information). Surprisingly, the
absolute energies of the HOMOs−5 are very similar across
Attempts at growing crystals of the NHGe→Sr complex 4
were unsuccessful and in each instance afforded exclusively
compound 10 along with formation of elemental germanium.
Similarly, for the barium complexes 8 and 9, we were unable to
obtain suitable crystals for a single-crystal X-ray diffraction
analysis. In both instances, even growing crystals at −78 °C, we
obtained single crystals of the Ba analogue of 10, [(η5-
C5Me5)2Ba(κ2-{NtBuCHCHNtBu})]; its connectivity
could be proven by X-ray diffraction analysis, but the poor
crystal quality precludes a discussion of metric parameters.
These results collectively show that the NHE:→Ae
complexes (Ae = Ca, Sr, Ba) are thermodynamically unstable
and prone to decomposition, which is in line with results from
DFT (density functional theory) calculations (see below).
All of the respective 1,4-diazabuta-1,3-diene Ae decom-
position complexes 10−12 could readily be synthesized by
t
direct reactions of 1,4-diazabuta-1,3-diene BuNCHCH
NtBu with the respective Cp*2Ae (Ae = Ca, Sr, Ba) precursors.
While the reaction of tBuNCHCHNtBu with [(η5-
C5Me5)2Ca] in the molar ratio of 1:1 afforded a mixture of
products, the reaction in the ratio of 1:2 cleanly afforded 12.
This result fits with the decomposition mode of [(η5-
C5Me5)2Ca←:Ge(NtBuCH)2] (3) to give merely 12. In
t
contrast, the reactions of BuNCHCHNtBu with one
molar equivalent of 5·OEt2 afforded complex 11 selectively,
Figure 6. Boundary surface plot of the HOMO−5 of complex 3
showing the donor−acceptor interaction between Ge and Ca. For
complexes 4, 7, 8, and 9 a very similar picture emerges (see Supporting
Information); the complexes were optimized at the B97-D/def2-
TZVP[C,N,H:6-31G(d)] level of theory.
with concomitant OEt2 elimination, and in analogy, the
t
reaction of BuNCHCHNtBu with [(η5-C5Me5)2Ba] in
the molar ratio of 1:1 yielded complex 11 quantitatively
(Scheme 2).
E
DOI: 10.1021/acs.inorgchem.5b01643
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
all the complexes, and Table 2 provides the computed energies
of the HOMOs−5 and the calculated E:→Ae (E = Si, Ge; Ae =
Ca, Sr, Ba) distances.
analogues. Noteworthy is that within each couple (NHSi vs
NHGe Ca, Sr, and Ba complexes, respectively) the NHSi
complex exhibits a slightly higher WBI compared to the NHGe
analogue, which is a consequence of the enhanced σ-donor
ability of Si(II) vs Ge(II), resulting in a slightly stronger Si:→
Ae interaction.
Table 2. Energies of the HOMOs−5 and Calculated E:→Ae
Distances (in Å) for 3, 4, 7, 8, and 9
According to weak E:→Ae bonding interactions, the
calculated bond dissociation energies (Table 5) are very low
energy HOMO−
d(E:→Ae)
calculated
d(E:→Ae) by X-ray
complex
5 (eV)
analysis
Ge:→Ca (3)
Ge:→Sr (4)
Si:→Sr (7)
Si:→Ba (8)
Ge:→Ba (9)
−6.21
−6.36
−6.06
−6.04
−6.36
3.166
3.319
3.272
3.491
3.565
3.1174(11)
3.224(3)
Table 5. Bond Dissociation Energies (BDE) of Complexes 3,
4, and 7− 9 into NHSi/NHGe and [(η5-C5Me5)2]Ae (in kcal
mol−1)
complex
BDE
Ge:→Ca (3)
Si:→Sr (7)
Ge:→Sr (4)
Si:→Ba (8)
Ge:→Ba (9)
19.7
22.9
20.8
24.4
21.7
The NBO analyses of the E:→Ca (E = Si, Ge) bonds
revealed high s-character with a heavily polarized bonding
situation toward the group 14 atom (Table 3). In the case of
Table 3. NBO Analyses of the σ-Donation from E(II):→Ae
a
in the Complexes 3, 4, and 7−9
for all complexes (19.7−22.9 kcal mol−1). Again, for each Ae
(Ca, Sr, Ba) complex, the NHSi exhibits a slightly larger BDE
compared with the respective NHGe analogue. It is of interest
to note that while there appears to be a decrease in the
covalency on descending from Ca to Ba, the BDE on average
increases. This might be due to a preponderance of increased
ionic contributions, which result in a more stable bonding
interaction.
s-
p-
d-
atom polarization character character character
Ca−Ge
(3)
σ-bond
Ca
Ge
Si
16.50%
83.50%
14.24%
59.82%
74.25%
35.63%
40.06%
25.69%
50.13%
0.10%
0.05%
Sr−Si (7) lone
pair
Sr−Ge
(4)
lone
pair
Ge
Si
82.43%
74.59%
82.72%
78.95%
86.40%
17.52%
25.36%
17.25%
21.01%
13.58%
0.05%
0.04%
0.03%
0.04%
0.02%
Most striking, however, is the propensity of complexes 3, 4,
and 7−9 to undergo a different decomposition process induced
by the Ae metal with concomitant liberation of elemental
silicon and germanium and the formation of the respective 1,4-
diazabuta-1,3-diene Ae complexes 10−12 (Scheme 3). The
liberation of neutral group 14 elements from metallylenes has
merely been reported for N-heterocyclic stannylenes and
germylenes by indirect means,24 but to date no such facile
process is reported for NHSis. To exlude the possibility that the
decomposition is perhaps triggered by the presence of trace
amounts of O2 or H2O, control experiments were conducted.25
The liberation of Si(0) and Ge(0) was calculated for all the
complexes. In all cases it was found that the main driving force
of the decomposition is the formation of crystalline silicon and
germanium (Table 6). In particular, on descending from Ca to
Ba, the extrusion process becomes considerably more
thermodynamically favorable, which explains the facile
formation of 10−12. Moreover, the free-energy change ΔG
for the NHGe complexes is considerably more negative than
those of the corresponding NHSi complexes, which is likely due
to the increased propensity of the heavier group 14 elements to
undergo this [1+4] cycloreversion process24 caused by the
zerovalent nature of a heavier group 14 atom in such
metallylenes.6 Notably, the liberation process of silicon and
germanium from the “free” NHSi (−2.4 kcal·mol−1) and NHGe
(−15.5 kcal·mol−1) is considerably less thermodynamically
favored, respectively. This underlines the fact that the process is
facilitated by coordination of the metallylene ligand to the Ae
center and likely driven by the high stability of the resulting 1,4-
diazabuta-1,3-diene Ae complexes.
Ba−Si
(8)
lone
pair
Ba−Ge
(9)
lone
pair
Ge
Si
NHSi
lone
pair
NHGe
lone
pair
Ge
a
Note that for complexes 4 and 7−9 the NBO analysis dissects the
bond into lone pairs on the E(II) centers and empty orbitals on Ae
since the bonds are very polarized (>90% E, <10% Ae).
the Sr and Ba complexes, the bond is polarized in such a way
that the NBO analysis dissects it into a donor lone pair
localized on the group 14 atom and a vacant acceptor orbital at
the Ae center. This result shows that on descent from calcium
to its heavier homologues, the donor−acceptor strength of the
bonds decreases, a result that is also shown by a subtle decrease
in the WBI (see below).
Inspection of Table 4 readily shows that for all complexes a
rather low WBI exists for the donor−acceptor E:→Ae
interaction. Remarkably, the WBIs of the calcium complexes
are somewhat higher than the corresponding Sr and Ba
Table 4. Wiberg Bond Index (WBI) and Mayer Bond Order
(MBO) of the E:→Ae Interactions and Charges of the Ae
Elements (Ae = Ca, Sr, Ba) and E(II) Centers (E = Si, Ge) in
the Complexes 3, 4, and 7−9
compound
WBI
MBO
charge of Ae element
charge of E
Ge:→Ca (3)
Si:→Sr (7)
Ge:→Sr (4)
Si:→Ba (8)
Ge:→Ba (9)
0.53
0.44
0.41
0.44
0.42
0.16
0.18
0.14
0.17
0.16
+0.84
+1.14
+1.15
+1.11
+1.09
+1.05
+1.03
+0.92
+1.03
+0.92
Furthermore, we were curious about the nature of elemental
silicon and germanium produced from the molecular precursors
8 and 9, respectively. The thermal decomposition of 8 afforded
lustrous, large, grayish-black samples of elemental silicon
(Figures S6−S9 in the Supporting Information), whereas 9
F
DOI: 10.1021/acs.inorgchem.5b01643
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 3. Decomposition of 3, 4, and 7−9 to Afford the Corresponding 1,4-Diazabuta-1,3-diene Ae Complexes 10−12 with
Concomitant Liberation of Elemental Silicon and Germanium
Table 6. Calculated Free Energy Change (ΔG in kcal mol−1)
electron diffraction (SAED), and energy dispersive X-ray
for the Facile Decomposition of Complexes 3, 4, and 7−9 to
the Corresponding 1,4-Diazabuta-1,3-diene Ae Complexes
10−12 and Elemental Silicon and Germanium (Si(0) and
Ge(0))
spectroscopy (EDX) (Figures S11 and S12 in the Supporting
Information).
The samples turned out to be only partially crystalline, as
shown by X-ray diffraction analyses; unfortunately, due to weak
reflections, the determination of the unit cell was not possible.
However, an SAED investigation of the silicon samples yielded
from 8 could be carried out, and from the unique d(hkl) values,
the data correspond to the rare metastable silicon allotrope
named allo-silicon (Figure 7, JCPDS 41-1111).26 The allo-Si
form was first reported as a result of the decomposition of the
silicon-rich silicide Li3NaSi6 in a solution of benzophenone in
dry THF over several weeks.26 Since then, exhaustive strategies
have been applied to determine its structure by a combination
of experimental and quantum chemical methods, but the
structure still remains elusive.26,27 Although a structural model
for allo-Si has been proposed, it has escaped experimental
elucidation.28
decomposition process
ΔG
[(η5-C5Me5)2Ca←:Si(NtBuCH)2] → 12 + Si(0)
−23.7
−36.7
−26.7
−40.4
−35.4
−48.7
−2.4
3 → 12 + Ge(0)
4 → 10 + Ge(0)
7 → 10 + Si(0)
8 → 11 + Si(0)
9 → 11 + Ge(0)
t
:Si(NtBuCH)2 → Si(0) + BuNCHCHNtBu
:Ge(NtBuCH)2 → Ge(0) + BuNCHCHNtBu
−15.5
t
produced small particles of elemental germanium (Figure S10
in the Supporting Information), as confirmed by inductively
coupled plasma atomic emission spectroscopy (ICP-AES),
transmission electron microscopy (TEM), selected area
Remarkably, the analysis of the SAED pattern on elemental
germanium yielded from 9 revealed that the dense form of
Figure 7. Selected area electron diffraction (SAED) pattern on elemental silicon and germanium samples synthesized from the molecular precursors
8 (formation of allo-silicon, left) and 9 (tetragonal-germanium, right), respectively. The d(hkl) values clearly indicate the formation of allo-silicon, a
rare allotrope of silicon and a dense form of solid tetragonal germanium (see also the Supporting Information).
G
DOI: 10.1021/acs.inorgchem.5b01643
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
tetragonal-germanium (JCPDS 72-1089, Figure 7), also known
as Ge-III or an ST-12 structure, was formed. The latter
alloptrope represents a somewhat rare metastable form of
germanium, which was first reported in 1963 by compressing
diamond-like cubic germanium at high pressures and reducing
the pressure back to atomospheric.29 Theoretical calculations
predicted that this Ge allotrope is a semiconductor with a direct
gap of only 1.47 eV30 and, therefore, has been considered very
attractive for device applications, especially in the area of
optoelectronics and photovoltaics. Given the potential use of
this allotrope, its facile preparation reported here is of obvious
interest.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: burgert.blom@tu-berlin.de.
*E-mail: matthias.driess@tu-berlin.de.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful to the Cluster of Excellence “UniCat”
(sponsored by the Deutsche Forschungsgemeinschaft and
administered by the TU Berlin) for financial support of this
research and Dr. C. Gobel for TEM measurements.
̈
SUMMARY AND CONCLUSIONS
■
The first complete series of N-heterocyclic germylene and N-
heterocyclic silylene alkaline-earth-metal complexes have been
prepared and characterized spectroscopically, in selected cases
by X-ray diffraction analysis, and investigated by DFT methods.
All complexes feature very weak donor−acceptor interactions
with limited covalency in the E:→Ae bond (E = Si, Ge; Ae =
Ca, Sr, Ba). The complexes are thermodynamically unstable
and prone to decomposition, affording in the case of calcium
the dinuclear complex [(η5-C5Me5)2Ca←:NtBu
REFERENCES
■
(1) Herrmann, W. A.; Denk, M.; Behm, J.; Scherer, W.; Klingan, F. −
R.; Bock, H.; Solouki, B.; Wagner, M. Angew. Chem., Int. Ed. Engl.
1992, 31, 1485.
(2) Denk, M.; Lennon, R.; Hayashi, R.; West, R.; Belyakov, A. V.;
Verne, H. P.; Haaland, A.; Wagner, M.; Metzler, N. J. Am. Chem. Soc.
1994, 116, 2691.
(3) For recent reviews see: (a) Blom, B.; Gallego, D.; Driess, M.
Inorg. Chem. Front. 2014, 1, 134. (b) Blom, B.; Stoelzel, M.; Driess, M.
Chem. - Eur. J. 2013, 19, 40. (c) Blom, B.; Driess, M. Struct. Bonding
(Berlin) 2014, 156, 85.
t
CHCH BuN:→Ca(η5-C5Me5)2] (12) and in the cases of Sr
and Ba the chelate 1,4-diazabuta-1,3-diene complexes [(η5-
C5Me5)2Ae(κ2-{NtBuCHCHNtBu})] (Ae = Sr 10 and Ae
= Ba 11). This decomposition process is accompanied by the
unexpected facile liberation of Si(0) and Ge(0), in accordance
with what was proposed more than 20 years ago on the N-
heterocyclic metallylenes based on photoelectron spectrosco-
py.6 Most notably, the thermal decomposition of 8 and 9 led to
rare allotropes of the respective elements: allo-silicon and
tetragonal germanium, a dense form of solid germanium, could
be isolated and characterized by several methods. The results
presented here highlight new possibilities in the generation of
rare and otherwise inaccessible allotropes of silicon and
germanium from readily accessbile molecular precursors
under facile reaction conditions. Not only due to the superior
electronic properties of elemental silicon and germanium in
technological applications but also from a fundamental point of
view, the route highlighted in this report is of general
interest.31,32 We are currently investigating how different
reaction conditions and modifications in pressure might enable
access to other hitherto unknown allotropes of silicon and
germanium.
(4) See as selected examples: (a) Jones, C.; Rose, R. P.; Stasch, A.
Dalton Trans. 2008, 2871. (b) Matioszek, D.; Katir, N.; Saffon, N.;
Castel, A. Organometallics 2010, 29, 3039. (c) Cabeza, J. A.; García-
́
Alvarez, P.; Perez-Carreno, E.; Polo, D. Inorg. Chem. 2014, 53, 8735.
́
̃
(d) Gallego, D.; Inoue, S.; Blom, B.; Driess, M. Organometallics 2014,
33, 6885. (e) Matioszek, D.; Saffon, N.; Sotiropoulos, J.-M.; Miqueu,
K.; Castel, A.; Escudie, J. Inorg. Chem. 2012, 51, 11716. (f) Cabeza, J.
́
́
A.; García-Alvarez, P.; Polo, D. Dalton. Trans. 2013, 42, 1329.
́
(g) Cabeza, J. A.; García-Alvarez, P.; Per
- Eur. J. 2014, 20, 8654. (h) Nagendran, S.; Sen, S. S.; Roesky, H. W.;
́
ez-Carreno, E.; Polo, D. Chem.
̃
Koley, D.; Grubmuller, H.; Pal, A.; Herbst-Irmer, R. Organometallics
̈
2008, 27, 5459.
(5) See as leading references of NHC compounds and their
relevance: (a) Herrmann, W. A. Angew. Chem., Int. Ed. 2002, 41, 1290.
(b) Nair, V.; Bindu, S.; Sreekumar, V. Angew. Chem., Int. Ed. 2004, 43,
5130. (c) Lin, J. C. Y.; Huang, R. T. W.; Lee, C. S.; Bhattacharyya, A.;
Hwang, W. S.; Lin, I. J. B. Chem. Rev. 2009, 109, 3561. (d) Díez-
Gonzalez, S.; Marion, N.; Nolan, S. P. Chem. Rev. 2009, 109, 3612.
́
(e) Poyatos, M.; Mata, J. A.; Peris, E. Chem. Rev. 2009, 109, 3677.
(f) Kantchev, E. A. B.; O’Brien, C. J.; Organ, M. G. Angew. Chem., Int.
Ed. 2007, 46, 2768. (g) N-Heterocyclic Carbenes in Synthesis; Nolan, S.
P., Ed.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany,
2006.
(6) Arduengo, A. J., III; Bock, H.; Chen, H.; Denk, M.; Dixon, D. A.;
Green, J. C.; Herrmann, W. A.; Jones, N. L.; Wagner, M.; West, R. J.
Am. Chem. Soc. 1994, 116, 6641.
ASSOCIATED CONTENT
■
S
* Supporting Information
(7) (a) Tonner, R.; Frenking, G. Angew. Chem., Int. Ed. 2007, 46,
8695. (b) Tonner, R.; Frenking, G. Chem. - Eur. J. 2008, 14, 3260.
(c) Tonner, R.; Frenking, G. Chem. - Eur. J. 2008, 14, 3273.
(d) Tonner, R.; Frenking, G. Pure Appl. Chem. 2009, 81, 597.
(e) Takagi, N.; Shimizu, T.; Frenking, G. Chem. - Eur. J. 2009, 15,
3448. (f) Takagi, N.; Shimizu, T.; Frenking, G. Chem. - Eur. J. 2009,
15, 8593. (g) Takagi, N.; Tonner, R.; Frenking, G. Chem. - Eur. J.
2012, 18, 1772.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.5b01643.
Experimental details and procedures and details on X-ray
crystal structure analyses, selected NMR spectra, addi-
tional details on the characterization of elemental silicon
and germanium, as well as the Cartesian coordinate NBO
analyses, optimized structures, and additional MOs of all
complexes (PDF)
Crystallographic data for complexes 3 and 12 (CIF)
Crystallographic data for complex 5 (CIF)
Crystallographic data for complex 7 (CIF)
Crystallographic data for complex 10 (CIF)
(8) (a) Mondal, K. C.; Roesky, H. W.; Schwarzer, M. C.; Frenking,
G.; Niepotter, B.; Wolf, H.; Herbst-Irmer, R.; Stalke, D. Angew. Chem.,
̈
Int. Ed. 2013, 52, 2963. (b) Xiong, Y.; Yao, S.; Tan, G.; Inoue, S.;
Driess, M. J. Am. Chem. Soc. 2013, 135, 5004. (c) Ishida, S.; Iwamoto,
T.; Kabuto, C.; Kira, M. Nature 2003, 421, 725. (d) Kira, M. Chem.
Commun. 2010, 46, 2893. (d) Kira, M.; Iwamoto, T.; Ishida, S.;
Masuda, H.; Abe, T.; Kabuto, C. J. Am. Chem. Soc. 2009, 131, 17135.
(e) Iwamoto, T.; Masuda, H.; Kabuto, C.; Kira, M. Organometallics
H
DOI: 10.1021/acs.inorgchem.5b01643
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
2005, 24, 197. (f) Xiong, Y.; Yao, S.; Inoue, S.; Epping, J. D.; Driess,
M. Angew. Chem., Int. Ed. 2013, 52, 7147.
(9) As recent examples see: (a) Tan, G.; Enthaler, S.; Inoue, S.; Blom,
B.; Driess, M. Angew. Chem., Int. Ed. 2015, 54, 2214. (b) Blom, B.;
Pohl, M.; Gallego, D.; Tan, G.; Driess, M. Organometallics 2014, 33,
5272. (c) Tan, G.; Blom, B.; Gallego, D.; Driess, M. Organometallics
(27) Zeilinger, M.; Jantke, L.-A.; Scherf, L. M.; Kiefer, F. G.;
Neubuser, G.; Kienle, L.; Karttunen, A. J.; Konar, S.; Haussermann, U.;
̈
̈
Fassler, T. F. Chem. Mater. 2014, 26, 6603.
̈
(28) Conesa, J. C. J. Phys. Chem. B 2002, 106, 3402.
(29) (a) Kasper, J. S.; Richards, S. M. Acta Crystallogr. 1964, 17, 752.
(b) Bundy, F. P.; Kasper, J. S. Science 1963, 139, 340.
(30) Joannopoulos, J. D.; Cohen, M. L. Phys. Rev. B: Solid State 1973,
7, 2644.
(31) Szczech, J. R.; Jin, S. Energy Environ. Sci. 2011, 4, 56.
(32) Okamoto, H.; Sugiyama, Y.; Nakano, H. Chem. - Eur. J. 2011,
17, 9864.
2014, 33, 363. (d) Gallego, D.; Bruck, A.; Irran, E.; Meier, F.; Kaupp,
̈
M.; Driess, M.; Hartwig, J. F. J. Am. Chem. Soc. 2013, 135, 15617.
(e) Blom, B.; Enthaler, S.; Inoue, S.; Irran, E.; Driess, M. J. Am. Chem.
Soc. 2013, 135, 6703. (f) Stoelzel, M.; Prasang, C.; Blom, B.; Driess, M.
̈
Aust. J. Chem. 2013, 66, 1163. (g) Bruck, A.; Gallego, D.; Wang, W.;
̈
Irran, E.; Driess, M.; Hartwig, J. F. Angew. Chem., Int. Ed. 2012, 51,
11478. (h) Blom, B.; Driess, M.; Gallego, D.; Inoue, S. Chem. - Eur. J.
2012, 18, 13355. (i) Wang, W.; Inoue, S.; Enthaler, S.; Driess, M.
Angew. Chem., Int. Ed. 2012, 51, 6167. (j) Wang, W.; Inoue, S.; Irran,
E.; Driess, M. Angew. Chem., Int. Ed. 2012, 51, 3691. (k) Stoelzel, M.;
Prasang, C.; Inoue, S.; Enthaler, S.; Driess, M. Angew. Chem., Int. Ed.
̈
2012, 51, 399.
(10) Blom, B.; Klatt, G.; Gallego, D.; Tan, G.; Driess, M. Dalton
Trans. 2015, 44, 639.
(11) (a) Bader, R. F. W. Chem. Rev. 1991, 91, 893. (b) Bader, R. F.
W.; Essen, H. J. Chem. Phys. 1984, 80, 1943.
(12) Wiberg, K. A. Tetrahedron 1968, 24, 1083.
(13) See for examples ref 9e and 9h.
(14) Burns, C. J.; Andersen, R. A. J. Organomet. Chem. 1987, 325, 31.
(15) Surprisingly complex 5·OEt2 has escaped characterization by
single-crystal X-ray diffraction analysis, despite being known since
1987. Since we fortuitously obtained suitable crystals thereof, we
carried out such an investigation. Complex 5·OEt2 features bent Cp*
groups, which is typical and unremarkable for complexes of the type
[(η5-C5Me5)2Ae←:L] (Ae = Ca, Sr, Ba; L = neutral ligand), with the
OEt2 moiety occupying one of the coordination sites. The complex
features a Sr−O bond length of 2.5598(14) Å, and the distance from
one of the H atoms (H20a) of the methyl group is in close proximity
to the Sr center (2.885 Å, dotted line, Figure S14), compared with the
other methyl H atoms. This shorter bond distance might be indicative
of an agostic interaction to the Sr center (see Figure S12 in the
Supporting Information).
(16) Agostic interactions are known for alkaline earth metal
complexes. As a recent example, see: Sarazin, Y.; Rosç a, D.; Poirier,
V.; Roisnel, T.; Silvestru, A.; Maron, L.; Carpentier, J. − F.
Organometallics 2010, 29, 6569.
(17) For an explanation on how measuring frequency affects the
NMR time scale see: Bryant, R. G. J. Chem. Educ. 1983, 60,
933.10.1021/ed060p933
(18) See ref 14: A detectible exchange process is also observed for
the PEt3 analogues.
(19) Teng, W.; Ruhlandt-Senge, K. Organometallics 2004, 23, 952.
(20) Rieckhoff, M.; Pieper, U.; Stalke, D.; Edelmann, F. T. Angew.
Chem. 1993, 105, 1102.
(21) Teng, W.; Ruhlandt-Senge, K. Organometallics 2004, 23, 2694.
(22) Pyykko, P.; Atsumi, M. Chem. - Eur. J. 2009, 15, 186.
̈
(23) Kazhdan, D.; Hu, Y.- J.; Kokai, A.; Levi, Z.; Rozenel, S. Acta
Crystallogr., Sect. E: Struct. Rep. Online 2008, E64, 1134.
(24) (a) Gans-Eichler, T.; Gudat, D.; Nattinen, K.; Nieger, M. Chem.
̈
- Eur. J. 2006, 12, 1162. (b) Gans-Eichler, T.; Gudat, D.; Nieger, M.
Angew. Chem., Int. Ed. 2002, 41, 1888.
(25) Samples of Cp*2Ae←:E(NtBuCH)2 were tested against
decomposition with O2 and H2O, respectively, in control experiments
to exclude the possibility that the formation of the diazabutadiene
complexes is due to decomposition by O2 or H2O. In both cases (O2
or H2O addition), a color change to yellow is observed along with
decomposition to an intractable array of products; however, none of
these correspond with the 1,4-diazabuta-1,3-diene Ae complexes 10−
12. This rules out that the decomposition is triggered by trace
amounts of O2 or H2O.
(26) von Schnering, H. G.; Schwarz, M.; Nesper, R. J. Less-Common
Met. 1988, 137, 297.
I
DOI: 10.1021/acs.inorgchem.5b01643
Inorg. Chem. XXXX, XXX, XXX−XXX