Direct synthesis of an anionic 13-vertex closo-cobaltacarborane cluster

Article abstract of DOI:10.1039/c9dt03111a

Reaction of 1,2-bis(diphenylphosphino)-ortho-carborane (L) with [K(thf){(^MesBIAN)Co(η⁴-cod)}] (1, ^MesBIAN = bis(mesityliminoace-naphthene)diimine, cod = 1,5-cyclooctadiene) affords an anionic 13-vertex closo-cobaltacarborane cluster (2) in one step. The mechanism of this transformation has been studied by experimental and quantum chemical techniques, which suggest that a series of outer-sphere electron transfer and isomerisation processes occurs. This work shows that low-valent metalate anions are promising reagents for the synthesis of anionic metallacarborane clusters.

Full text of DOI:10.1039/c9dt03111a

Dalton
Transactions
View Article Online
View Journal | View Issue
COMMUNICATION
Direct synthesis of an anionic 13-vertex
closo-cobaltacarborane cluster†
Cite this: Dalton Trans., 2019, 48,
15772
Thomas M. Maier, a Peter Coburger, a,b Nicolaas P. van Leest,
c
Received 30th July 2019,
Accepted 26th September 2019
a
Evamarie Hey-Hawkins *b and Robert Wolf
*
DOI: 10.1039/c9dt03111a
rsc.li/dalton
Reaction of 1,2-bis(diphenylphosphino)-ortho-carborane (L) with the cluster with the cobalt phosphane complex simultaneously
[K(thf){(MesBIAN)Co(η4-cod)}] (1, MesBIAN = bis(mesityliminoace- acting as the reducing agent and metal source.14
naphthene)diimine, cod = 1,5-cyclooctadiene) affords an anionic
As shown by several groups in electrochemical investi-
13-vertex closo-cobaltacarborane cluster (2) in one step. The gations, neutral 13-vertex cobaltacarboranes can be reduced
mechanism of this transformation has been studied by experi- reversibly.3,8,9,15 Such reduced species were used in situ as
mental and quantum chemical techniques, which suggest that a nucleophilic intermediates for the synthesis of 14-vertex
series of outer-sphere electron transfer and isomerisation pro- bimetallacarboranes.12,16–18 Despite their synthetic value, only
cesses occurs. This work shows that low-valent metalate anions a few anionic 13-vertex cobaltacarboranes have been well
are promising reagents for the synthesis of anionic metallacarbor- characterised. One important example is the homoleptic
ane clusters.
complex anion [4,4′-Co-(1,6-closo-C2B10H12)2]− (A) reported as
its tetraethylammonium salt by Hawthorne in 1973.3 Shortly
afterwards, Hawthorne published the fascinating complex B,
where the B–H moieties in the 5-position of both cluster units
in A were formally substituted by cyclopentadienyl cobalt frag-
ments.19 Homoleptic complex C shown in Fig. 1, an isomer of
A, was synthesised and characterised by X-ray crystallography
by Welch and co-workers.8
Supraicosahedral metallacarboranes are of interest due to
various applications in catalysis, materials science, and biome-
dicine.1 13-Vertex metallacarboranes (MC2B10) dominate this
area. The first member of this class of compounds, [CpCo-
(C2B10H12)] (Cp
=
cyclopentadienyl), was reported by
Hawthorne and co-workers in 1971.2 In this seminal work, the
synthesis of the 13-vertex cobaltacarborane was achieved by
reduction of the parent 1,2-dicarba-closo-dodecaborane
(C2B10H12) with elemental sodium and subsequent treatment
with NaCp, CoCl2 and air. This synthetic scheme was called
“polyhedral expansion”,
a
term coined by Hawthorne.
Subsequently, a large variety of 13-vertex cobaltacarborane deriva-
tives was synthesised following this approach in the next
decades.3–13 Notably, Stone, Welch and co-workers established
another synthetic route towards such species in 1984 by treating
the parent 12-vertex carborane with the low-valent cobalt complex
[Co(PEt3)4]. Here a Co(PEt3)2 fragment was directly inserted into
aUniversity of Regensburg, Institute of Inorganic Chemistry, 93040 Regensburg,
Germany. E-mail: robert.wolf@ur.de; https://www.uni-regensburg.de/chemistry-phar-
macy/inorganic-chemistry-wolf/index.html
bLeipzig University, Institute of Inorganic Chemistry, Johannisallee 29, 04103
Leipzig, Germany. E-mail: hey@uni-leipzig.de; https://research.uni-leipzig.de/hh/
cVan’t Hoff Institute for Molecular Sciences (HIMS), University of Amsterdam (UvA)
Science Park 904, 1098 XH Amsterdam, The Netherlands
†Electronic supplementary information (ESI) available: For details of the syn-
thesis, structural, spectroscopic and computational data. CCDC 1943845,
1943847 and 1943848. For ESI and crystallographic data in CIF or other elec-
tronic format see DOI: 10.1039/c9dt03111a
Fig. 1 Examples of anionic 13-vertex cobaltacarboranes. For clarity,
only the 4,1,6-isomer of D is shown, the respective 4,1,8-, 4,1,10- and
4,1,12-isomers were described in the same work.9
15772 | Dalton Trans., 2019, 48, 15772–15777
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Communication
Further detailed electrochemical and spectroscopic studies
have shown that a series of anionic 13-vertex indenyl cobalta-
carboranes with 4,1,6-, 4,1,8-, 4,1,10- and 4,1,12-CoC2B10 archi-
tectures, e.g. D, are accessible by the reduction of neutral
precursors.9 These anions were characterised in situ by UV-vis
and EPR spectroscopy.
Here we report the synthesis and characterisation of a new
anionic 13-vertex cobaltacarborane, 2, which contains a redox-
active bis(iminoacenapthene) (BIAN) ligand at cobalt and two
phosphanyl moieties attached to the cluster carbon atoms.
Similar to Stone’s approach, this complex is obtained in a
facile one-pot procedure from 1,2-bis(diphenylphosphino)-
closo-carborane L and a highly reduced cobaltate anion 1
(Scheme 1).20,21 We describe the structural and spectroscopic
characteristics of 2, and we propose a mechanism for its
formation based on model reactions, NMR spectroscopic
monitoring and quantum chemical calculations.
As part of a research program studying the chemistry of
phosphanyl-substituted carborane compounds,22 it was pre-
sumed that bis-phosphane L might be a suitable chelate
ligand for the [Co(BIAN)]− fragment due the strongly electron-
withdrawing properties of L.20,21 In order to test this hypoth-
esis, the 1 : 1 reaction of 1 with L was performed in THF as
shown in Scheme 1. Upon mixing 1 and L in THF, a deep
purple reaction mixture is formed immediately. In contrast to
Fig. 2 Solid-state structure of 2 with thermal ellipsoids at 40% prob-
ability. Selected bond lengths [Å] and angles [°]: N1–C1 1.332(2), N2–C2
1.325(2), C1–C2 1.420(2), N1–Co1 1.960(1), N2–Co1 1.935(1), Co1–C3
2.090(1), C3–B1 1.549(2), K1⋯B5 3.409(1), K1⋯B6 3.409(1), C3–P1
1.837(1), C4–P2 1.867(1), N1–Co1–N2 82.20(5). Hydrogen atoms are
omitted for clarity.
initial expectations, the desired adduct 2′ is not observed at 13-vertex cluster, the carbon atoms occupy the positions 1 and
any stage of the reaction; instead, polyhedral expansion 8.23 Thus, 2 can be classified as a 4,1,8-cobaltacarborane
occurs, resulting in a 13-vertex cluster anion. The deep purple according to common numbering schemes of such clusters.5
crystalline potassium salt 2 can be isolated in 59% yield after Such an isomerisation is frequently observed for related
work-up and crystallisation of the crude product from DME/ 13-vertex metallacarboranes.1,24 The potassium cation K1
n-hexane. 31P{1H} NMR spectroscopic monitoring shows that interacts with two B–H moieties (K1-B5/B6 3.409(1) Å) of the
the reaction takes approximately 40 h at 50 °C to go to com- carborane cluster. Cobalt is additionally coordinated by the
pletion. Several reaction intermediates can be observed MesBIAN ligand with unexceptional Co–N distances of 1.960(1) Å
(vide infra), yet the formation of the final product is very selec- and 1.935(1) Å. Key bond lengths within the MesBIAN framework
tive when the reaction is finished.
of this α-diimine ligand suggest that it is present in its monoa-
Crystals of 2 suitable for single-crystal X-ray diffraction nionic form. In particular, the C1–C2 bond (1.420(2) Å) is
(XRD) were obtained by slow diffusion of n-hexane into a DME slightly longer than in the structures of closely related dianionic
solution. The molecular structure is shown in Fig. 2. A contact BIAN ligands (1.402(4) Å for Na2[DippBIAN], DippBIAN = bis(2,6-
ion pair between [K(dme)3]+ and a 13-vertex cobaltacarborane diisopropylphenyliminoacenaphthene)diimine), and the N1–C1
cluster anion is observed, in which the cobalt atom caps a six- (1.332(2) Å), and N2–C2 bonds (1.325(2) Å) are also shorter than
membered CB5 ring with typical Co–B distances of 2.149(2)– expected for a dianionic ligand (1.387(4) Å for Na2[DippBIAN]).25
2.197(2) Å and Co1–C3 of 2.090(1) Å.4,5,7,23 During the course
Multinuclear NMR spectra of 2 in THF-d8 are consistent
of the reaction an isomerisation takes place, and in the final with the crystallographically determined molecular structure.
The 31P{1H} NMR spectrum shows two singlets at 28.9 ppm
and 24.1 ppm (cf. a shift of 7.8 ppm for the starting material L),
while the 1H and 13C{1H} NMR data show that the ortho-
methyl substituents of the mesityl groups are diastereotopic
presumably due to hindered rotation around the C(mesityl)–N
bond. These methyl groups show four resonances in the
13C{1H} NMR spectrum. Two of them split into a doublet due
to through-space coupling to the phosphorus atom P1, proven
by a 13C{1H,31P} NMR experiment. The hydrogen atoms of the
carborane unit give rise to a very broad signal from 4.0 to
0.0 ppm. The 11B and 11B{1H} NMR spectra are typical for a
metallacarborane framework and show four broad signals in
Scheme 1 Synthesis of 2 and initially attempted synthesis of 2’.
the range of 13.3 to −20.9 ppm.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 15772–15777 | 15773

View Article Online
Communication
Dalton Transactions
Due to the presence of three potentially redox-active sites
(cobalt atom, MesBIAN and carborane ligand), the redox pro-
perties of 2 were of interest. A cyclic voltammogram recorded
in THF/nBu4NPF6 shows a reversible wave at E1/2 = −0.6 V vs.
Fc/Fc+, which may be assigned to the oxidation of anion 2 by
one electron to a neutral compound (see the ESI†). In addition,
two irreversible waves are observed at +0.5 V and −2.5 V vs.
Fc/Fc+, which can be assigned to a second oxidation and a
reduction process, respectively. The UV-vis spectrum of deep
purple 2 shows intense absorptions at 320 nm and 504 nm (ε =
51 000
L ) with a shoulder at 605 nm. A
mol−1 cm−1
calculated spectrum obtained with TD-DFT methods qualitat-
ively agrees with the experimental spectrum (see Fig. S24,
ESI†) and shows that the intense absorptions observed in the
visible region arise from MLCT transitions involving cobalt
and the MesBIAN ligand (see the ESI†) as in other main group
and transition metal BIAN complexes.26 The electronic
structure of 2 was additionally probed by CASSCF calculations
(see the ESI† for details). These calculations reveal the
strongly covalent nature of the bonding between cobalt and
the BIAN ligand, making the assignment of an oxidation
state to the cobalt centre somewhat ambiguous. An MO ana-
lysis clearly shows that
a BIAN-centred π* orbital is
Fig. 3 31P{1H} NMR spectra (161 MHz, 300 K, C6D6) of a 1 : 1 reaction of
1 and L in THF at 25 °C (a–c); reaction NMR at 50 °C in THF (d); signal at
−15.1 ppm = P2Ph4.
partially occupied and thus indicates a redox-non-innocent
behaviour of the ligand (see the ESI† for details). This is in
line with the solid state molecular structure of 2, which
suggests that
a monoanionic BIAN ligand is present
(vide supra).
To gain insight into the reaction mechanism, a 1 : 1 reac- in the closely related Ni complex [(DippBIAN)Ni(η4-cod)].27
tion between cobaltate 1 and L was monitored by 31P{1H} NMR These values suggest the presence of
a monoanionic
spectroscopy at room temperature, where the rate of the reac- diimine ligand and, consequently, a +I oxidation state for
tion was slow enough to detect reaction intermediates (Fig. 3). cobalt.25
After 30 min, a signal at 17.9 ppm was observed (see Fig. 3a),
which can presumably be assigned to 1,2-bis(diphenyl- netic moment of µeff = 1.8(1) µB determined by the Evans
phosphanyl)-1,2-dicarba-nido-dodecaborate(12) (nido-L2−
as method in C6D6 solution. An EPR spectrum of Int-A in toluene
[(MesBIAN)Co(η4-cod)] (Int-A) is paramagnetic with a mag-
)
independently confirmed by the detection of the protonated glass at 20 K is slightly axial (nearly isotropic) with simulated g
derivative (nido-HL−) in electrospray ionisation mass values of 2.01 (g∥) and 2.00 (g⊥) (see the ESI, Fig. S10†) indicat-
spectra (vide infra).‡ After work-up and crystallisation of the ing a ligand-centred radical. This is also supported by DFT cal-
crude product from n-hexane, a mixture of dark orange crystals culations, which show that the spin density mainly resides on
of [(MesBIAN)Co(η4-cod)] (Int-A) and a few crystals of a second the MesBIAN ligand (see the ESI, Fig. S11†). Fig. 4b shows the
complex [(MesBIAN)Co(L)] (3) were isolated.§ 3 is likely formed single-crystal XRD structure of 3. Both the MesBIAN ligand and
by reaction of Int-A and remaining L. An isolated yield of the ortho-carboranyl bis-phosphane L act as bidentate chelate
17% can be estimated for Int-A assuming that it is by far ligands, while the cobalt atom shows a distorted tetrahedral
the major species in this mixture. Complex 3 was not coordination environment (83.7° between the planes
detected by 31P{1H} NMR spectroscopy due to its paramagnetic P1–Co1–P2 and N1–Co1–N2) (Fig. 4b). The bite angle of the
nature. The molecular structure of Int-A was determined by bis-phosphane P1–Co1–P2 (90.1°) is similar to our previously
single-crystal XRD and is shown in Fig. 4a. A square-planar reported cobalt(III)-phosphanido (90.7°) complex.28 The C–N
cobalt(I) complex is observed, which contains an and C–C distances in the diimine unit of the MesBIAN frame-
η4-coordinated 1,5-cyclooctadiene molecule. In addition, a work (C1–N1 1.340(3) Å, C2–N2 1.322(3) Å, and C1–C2 1.432(3)
MesBIAN ligand binds via the two nitrogen atoms. Using the Å) indicate the presence of a MesBIAN− monoanion similar to
midpoints of the CvC bonds of cod, the dihedral angle Int-A.25 Assuming that the bis-phosphane ligand L remains
between the planar MesBIAN moiety and the cod ligand was neutral, this would result in the +I oxidation state for cobalt,
determined to be 5.3°, which shows that the coordination geo- for which a square-planar coordination is normally preferred.
metry for cobalt is essentially square planar. The C–C and C–N A tetrahedral structure is probably observed due to steric repul-
bond lengths of the diimine fragment (C1–C2 1.428(5) Å, sion between the bulky mesityl groups and the phenyl substi-
C1–N1 1.316(5) Å, and C2–N2 1.332(5) Å) are similar to those tuents on L. Further spectroscopic characterisation of 3 was
15774 | Dalton Trans., 2019, 48, 15772–15777
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Communication
unfortunately not possible due to the small amount of isolated
crystalline material, which could not be successfully separated
from the major species Int-A.
When the 1 : 1 reaction between cobaltate 1 and L is carried
out for a longer reaction time, the formation of additional
intermediates can be observed. As shown in Fig. 3b and c, a
new, broad singlet (20.9 ppm, Δν1/2 = 74 Hz) was detected by
31P{1H} NMR spectroscopy after several hours. The chemical
shift is very similar to the final product 2, thus, this signal can
likely be assigned to the symmetric 13-vertex cobaltacarborane
Int-C shown in Scheme 2. After four days at room temperature,
the signals of nido-L2− and Int-C are still present, while signifi-
cant quantities of 2 are formed at this point. As shown by
Fig. 2d, 2 is by far the dominant species observed by 31P{1H}
NMR spectroscopy after heating at 50 °C for 40 h, showing that
the reaction is eventually very selective towards the formation
of the final product 2. The mechanism shown in Scheme 2 can
be proposed based on the experimental observations discussed
above in conjunction with quantum chemical calculations.
According to DFT calculations at the M06-D3(0)/def2-TZVP
CPCM(THF) level of theory (see the ESI†),29 it is conceivable
that the ortho-carboranyl bis-phosphane ligand L is reduced by
1 in the first step of the reaction, forming ortho-L2− and Int-A.
The calculations suggest that this step is exothermic by
−8.4 kcal mol−1. In addition, such a redox reaction is in agree-
ment with the reduction and oxidation potentials determined
by cyclic voltammetry for 1 (E1/2 = −1.72 V vs. Fc/Fc+ in THF)
and for L (E1/2 = −1.9 V vs. Fc/Fc+ in MeCN).30 ortho-L2− was
not directly observed, because it rapidly isomerises to nido-L2−
(step 2) in an exothermic process (−7.7 kcal mol−1). The latter
species was detected by 31P{1H} NMR spectroscopy
Fig. 4 Solid-state structures of (a) [(MesBIAN)Co(η4-cod)] (Int-A) and (b)
[(MesBIAN)Co(L)] (3) with thermal ellipsoids at 40% probability. Selected
bond lengths [Å] and angles [°]: For Int-A: N1–C1 1.316(5), N2–C2
1.332(5), C1–C2 1.428(5), N1–Co1 1.963(3), N2–Co1 1.976(3), C31–C32
1.391(6), N1–Co1–N2 83.0(1); For 3: N1–C1 1.340(3), N2–C2 1.322(3),
C1–C2 1.432(3), N1–Co1 2.042(2), Co1–P1 2.169(6), Co1–P2 2.198(1),
P1–C31 1.902(2), P2–C32 1.908(2), C31–C32 1.697(3). P1–Co1–P2
90.08(2), N1–Co1–N2 83.16(8). Hydrogen atoms are omitted for clarity.
Scheme 2 Proposed mechanism based on experimental and quantum mechanical methods. Cations are omitted for clarity. Reaction energies were
calculated on the M06-D3(0)def2-TZVP CPCM(THF) level of theory.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 15772–15777 | 15775

View Article Online
Communication
Dalton Transactions
(vide supra). Moreover, the protonated species nido-HL− was
also detected in electrospray ionisation mass spectra of the
reaction mixture (see the ESI†). A substitution of 1,5-cycloocta-
diene in Int-A by nido-L2− (step 3) followed by concomitant oxi-
dation of the resulting intermediate Int-B yields the
symmetric 13-vertex metallacarborane Int-C, also detected by
Notes and references
‡The possible formation of nido-L2− as a reaction intermediate is additionally
corroborated by an independent experiment, where diphosphane L was reduced
with two equivalents of potassium graphite (KC8, see the ESI†).
§Note that the CHN analysis for Int-A is not in agreement with the calculated
values for C38H40N2Co [found (calc.): C: 76.26 (78.20), H: 6.95 (6.91), N: 4.09
31P{1H} NMR spectroscopy. Taken together, steps 3 and 4 are (4.80)]. The observed discrepancies are likely due to the contamination of Int-A
exothermic by −7.4 kcal mol−1. An isomerisation to the
with a minor amount of 3.
¶The chelate complex 2′ was not observed in any of our NMR spectroscopic
investigations. In fact, this complex lies higher in energy than 2 by 5.7 kcal mol−1
unsymmetrical complex 2 is the final step of this sequence
(step 5, again exothermic by −13.3 kcal mol−1). Such an iso-
at the M06-D3(0)/def2-TZVP CPCM(THF) level of theory.
merisation is commonly observed for related 13-vertex
metallacarboranes.1,24 The mechanism proposed in Scheme 2
is thus in line with all NMR spectroscopic and crystallographic
observations. The proposed intermediates Int-A–Int-C are
viable species based on the DFT calculations, and the (mostly
exothermic) reaction steps add up to a total reaction energy of
−36.8 kcal mol−1.¶
1 (a) R. N. Grimes, Coord. Chem. Rev., 2000, 200–202, 773–
811; (b) R. N. Grimes, Carboranes, Elsevier, Saint Louis,
3rd edn, 2016; (c) V. I. Bregadze, Chem. Rev., 1992, 92, 209–
223.
2 M. F. Hawthorne, G. B. Dunks and M. M. McKown, J. Am.
Chem. Soc., 1971, 93, 2541–2543.
3 D. F. Dustin, G. B. Dunks and M. Frederick Hawthorne,
J. Am. Chem. Soc., 1973, 95, 1109–1115.
4 M. R. Churchill and B. G. DeBoer, Inorg. Chem., 1974, 13,
1411–1418.
Conclusions
5 D. Ellis, M. E. Lopez, R. McIntosh, G. M. Rosair,
A. J. Welch and R. Quenardelle, Chem. Commun., 2005,
1348–1350.
6 S. Zlatogorsky, D. Ellis, G. M. Rosair and A. J. Welch, Chem.
Commun., 2007, 2178–2180.
7 L. Deng, H.-S. Chan and Z. Xie, Inorg. Chem., 2007, 46,
2716–2724.
8 D. Ellis, R. D. McIntosh, S. Esquirolea, C. Viñas,
G. M. Rosair, F. Teixidor and A. J. Welch, Dalton Trans.,
2008, 1009–1017.
9 G. Scott, A. McAnaw, D. McKay, A. S. F. Boyd, D. Ellis,
G. M. Rosair, S. A. Macgregor, A. J. Welch, F. Laschi,
F. Rossi and P. Zanello, Dalton Trans., 2010, 39, 5286–
5300.
10 D. Ellis, G. M. Rosair and A. J. Welch, Chem. Commun.,
2010, 46, 7394–7396.
11 A. McAnaw, M. E. Lopez, G. Scott, D. Ellis, D. McKay,
G. M. Rosair and A. J. Welch, Dalton Trans., 2012, 41,
10957–10969.
The anionic 13-vertex closo-cobaltacarborane cluster was syn-
thesised by a direct route using α-diimine cobaltate [K(thf)-
{(MesBIAN)Co(η4-cod)}] (1) and isolated in a high yield of 59%.
The mechanism of this transformation was studied by experi-
mental techniques (single-crystal XRD, multinuclear NMR
spectroscopy and ESI-MS) and through DFT calculations.
An intermediate [(MesBIAN)Co(η4-cod)] (Int-A) was isolated,
while ESI-MS data and DFT investigations hint at the
formation of a dianionic nido-carborane nido-L2− as a key
intermediate en route to the final cluster 2. These results
suggest a redox mechanism that is initiated by an electron
transfer from
1 to the 1,2-bis(diphenylphosphino)-ortho-
carborane ligand L. Based on the work presented here, the syn-
thesis of new anionic metallacarborane derivatives by reaction
of further carborane derivatives with low-valent metalate
anions31 should be an appealing subject for future
investigations.
12 A. McAnaw, M. E. Lopez, D. Ellis, G. M. Rosair and
A. J. Welch, Dalton Trans., 2012, 42, 671–679.
13 D. Mandal, W. Y. Man, G. M. Rosair and A. J. Welch, Acta
Crystallogr., Sect. C: Struct. Chem., 2015, 71, 793–798.
14 G. K. Barker, M. P. Garcia, M. Green, F. G. A. Stone and
A. J. Welch, J. Chem. Soc., Chem. Commun., 1983, 137–139.
15 Z. Qiu, K.-H. Wong and Z. Xie, J. Organomet. Chem., 2012,
721–722, 97–103.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
We thank Prof. B. de Bruin (University of Amsterdam, the 16 J. S. Ward, H. Tricas, G. Scott, D. Ellis, G. M. Rosair and
Netherlands) for access to the EPR instrument and valuable A. J. Welch, Organometallics, 2012, 31, 2523–2525.
assistance and MSc Axel Straube (Leipzig University) for 17 W. J. Evans and M. F. Hawthorne, J. Chem. Soc., Chem.
measurement of 13C{1H,31P} NMR spectra. In addition, we
Commun., 1974, 38–39.
gratefully acknowledge financial support by the Fonds der 18 A. McAnaw, M. E. Lopez, D. Ellis, G. M. Rosair and
Chemischen Industrie (fellowship to T. M. M.), the A. J. Welch, Dalton Trans., 2014, 43, 5095–5105.
Studienstiftung des deutschen Volkes (fellowship to P. C.) and 19 D. F. Dustin and M. F. Hawthorne, J. Am. Chem. Soc., 1974,
the Graduate School BuildMoNa (Leipzig University).
96, 3462–3467.
15776 | Dalton Trans., 2019, 48, 15772–15777
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Communication
20 (a) R. P. Alexander and H. Schroeder, Inorg. Chem., 1963, 2,
1107–1110; (b) A. R. Popescu, F. Teixidor and C. Viñas,
Coord. Chem. Rev., 2014, 269, 54–84.
K. J. Dalby, G. M. Rosair and A. J. Welch, Dalton Trans.,
2005, 10, 1842–1846; (e) L. Deng, H.-S. Chan and Z. Xie,
J. Am. Chem. Soc., 2006, 128, 5219–5230.
21 (a) S. Pelties, T. Maier, D. Herrmann, B. de Bruin, 24 J. A. Doi, E. A. Mizusawa, C. B. Knobler and
C. Rebreyend, S. Gärtner, I. G. Shenderovich and R. Wolf, M. F. Hawthorne, Inorg. Chem., 1984, 23, 1482–1484.
Chem. – Eur. J., 2017, 23, 6094–6102; (b) C. P. G. Ziegler, 25 For characteristic C–C and C–N bond lengths see e.g.:
T. M. Maier, S. Pelties, C. Taube, F. Hennersdorf,
A. W. Ehlers, J. J. Weigand and R. Wolf, Chem. Sci., 2019,
10, 1302–1308; (c) T. M. Maier, S. Sandl, I. G. Shenderovich,
I. L. Fedushkin, A. A. Skatova, V. A. Chudakova and
G. K. Fukin, Angew. Chem., Int. Ed., 2003, 42, 3294–3298,
(Angew. Chem., 2003, 115, 3416–3420).
A. Jacobi von Wangelin, J. J. Weigand and R. Wolf, Chem. – 26 (a) J. Wang, R. Gaguly, L. Yongxin, J. Diaz, H. S. Soo and
Eur. J., 2019, 25, 238–245.
22 Selected examples: (a) S. Stadlbauer, R. Frank, I. Maulana,
P. Lönnecke, B. Kirchner and E. Hey-Hawkins, Inorg.
F. Garcia, Inorg. Chem., 2017, 56, 7811–7820;
(b) P. A. Papanikolaou and N. V. Tkachenko, Phys. Chem.
Chem. Phys., 2013, 15, 13128–13136.
Chem., 2009, 48, 6072–6082; (b) S. E. Lyubimov, 27 M. J. Srgo and D. W. Stephan, Dalton Trans., 2010, 39,
V. A. Davankov, K. N. Gavrilov, T. B. Grishina, 5786–5794.
E. A. Rastorguev, A. A. Tyutyunov, T. A. Verbitskaya, 28 P. Coburger, S. Demeshko, C. Rödl, E. Hey-Hawkins and
V. N. Kalinin and E. Hey-Hawkins, Tetrahedron Lett., 2010,
51, 1682–1684; (c) A. Kreienbrink, P. Lönnecke,
R. Wolf, Angew. Chem., Int. Ed., 2017, 56, 15871–15875,
(Angew. Chem., 2017, 129, 16087–16091).
M. Findeisen and E. Hey-Hawkins, Chem. Commun., 2012, 29 (a) S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem.
48, 9385–9387; (d) P. Coburger, J. Schulz, J. Klose,
B. Schwarze, M. B. Sárosi and E. Hey-Hawkins, Inorg.
Chem., 2017, 56, 292–304; (e) P. Coburger, G. Kahraman,
A. Straube and E. Hey-Hawkins, Dalton Trans., 2019, 48,
9625–9630; and references therein.
Phys., 2010, 132, 154104; (b) F. Weigend and R. Ahlrichs,
Phys. Chem. Chem. Phys., 2005, 7, 3297–3305; (c) Y. Zhao
and D. G. Truhlar, Theor. Chem. Acc., 2008, 120, 215–241;
(d) Y. Zhao and D. G. Truhlar, Chem. Phys. Lett., 2011, 502, 1–13;
(e) M. Cossi, N. Rega, G. Scalmani and V. Barone, J. Comput.
Chem., 2003, 24, 669–681; (f) A. V. Marenich, C. J. Cramer and
D. G. Truhlar, J. Phys. Chem. B, 2009, 113, 6378–6396.
23 (a) M. R. Churchill and B. G. DeBoer, J. Chem. Soc., Chem.
Commun., 1972, 24, 1326–1327; (b) K. J. Donaghy,
P. J. Carroll and L. G. Sneddon, J. Organomet. Chem., 1998, 30 R. Núñez, M. Tarrés, A. Ferrer-Ugalde, F. F. de Biani and
550, 77; (c) A. Burke, R. McIntosh, D. Ellis, G. M. Rosair F. Teixidor, Chem. Rev., 2016, 116, 14307–14378.
and A. J. Welch, Collect. Czech. Chem. Commun., 2002, 31 (a) J. E. Ellis, Inorg. Chem., 2006, 45, 3167–3186;
67(7), 991; (d) R. McIntosh, D. Ellis, J. Gil-Lostes,
(b) J. E. Ellis, Dalton Trans., 2019, 48, 9538–9563.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans., 2019, 48, 15772–15777 | 15777