Stability and decomposition of copper(i) boryl complexes: [(IDipp)Cu-Bneop], [(IDipp?)Cu-Bneop] and copper clusters

Article abstract of DOI:10.1039/d0nj03166f

The NHC copper boryl complexes [(IDipp)Cu-Bneop] (2) and [(IDipp?)Cu-Bneop] (3) (IDipp = 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene, IDipp? = 1,3-bis(2,6-(diphenylmethyl)-4-methylphenyl)imidazol-2-ylidene, neop = (OCH2)2CMe2) were synthesised and fu

Full text of DOI:10.1039/d0nj03166f

NJC
View Article Online
View Journal
PAPER
Stability and decomposition of copper(I) boryl
complexes: [(IDipp)Cu–Bneop], [(IDipp*)Cu–Bneop]
and copper clusters†‡
Cite this:DOI: 10.1039/d0nj03166f
Wiebke Drescher, Corinna Borner and Christian Kleeberg
*
The NHC copper boryl complexes [(IDipp)Cu–Bneop] (2) and [(IDipp*)Cu–Bneop] (3) (IDipp = 1,3-
bis(2,6-diisopropylphenyl)imidazol-2-ylidene, IDipp* = 1,3-bis(2,6-(diphenylmethyl)-4-methylphenyl)imidazol-
2-ylidene, neop = (OCH2)2CMe2) were synthesised and fully characterised including single crystal X-ray
structure determinations. The comparably rapid decomposition of 2, as reported previously by Sadighi and
co-workers, was corroborated and can be attributed to the properties of the Bneop ligand in combination
with the steric properties of the IDipp ligand. The decomposition of 2 leads predominantly to the free
NHC, the diborane(4) B2neop2 (1) and, presumably, elemental copper. Minor products of this reductive
decomposition are the unprecedented low-valent copper clusters [(IDipp)6Cu55] and [(IDipp)12Cu179] that
were characterised by single crystal X-ray structure determinations. It is concluded that an insight in the
decomposition pathways and products of copper boryl complexes is essential for the understanding of
copper boryl complexes and therefor for the further development of the flourishing field of copper
mediated borylation reactions.
Received 23rd June 2020,
Accepted 27th July 2020
DOI: 10.1039/d0nj03166f
rsc.li/njc
B2neop2 (1) (neop = (OCH2)2CMe2).1 The latter one, whilst
Introduction
already introduced in 1994 by Marder and co-workers,9 has
Copper(I) borly complexes are recognised as central intermediates in been somewhat in the shadow of B2pin2 and B2cat2 but is
copper catalysed borylation reactions with diboranes(4) as boron readily available nowadays.1
sources.1 Since the initial report on those reactions by Miyaura and
Consistently, the chemistry of copper(I) Bneop complexes is
co-workers in 2001 the formation of a copper boryl complex from a comparably little developed with [(IDipp)Cu–Bneop] (2) being
suitable precursor complex and a diborane(4) derivative has been the only isolated complex.8 This complex was reported by
considered central in those reactions.2 The first isolated and Sadighi et al. in 2018 but found to be stable for only less than
fully characterized copper boryl complex [(IDipp)Cu–Bpin] (pin = 20 min at room temperature in solution, as a consequence the
(OCMe2)2, IDipp = 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene) complex was only partly characterised (Scheme 1).
was reported in 2005 by Sadighi and co-workers and is since
This apparent low stability of 2 is unexpected in the light of
considered as the prototypic copper boryl complex.3,4 Nonetheless, the generally higher stability of copper boryl complexes with
in the recent years a number of new copper boryl complexes derived the IDipp ligand (or closely related ligands such as ClIDipp =
from diboranes(4) have been reported, e.g. with sterically less 1,3-bis(2,6-diisopropylphenyl)-4,5-dichloro-imidazol-2-ylidene and
demanding NHC or phosphine ligands and a variety of dialkoxy
and diamino boryl ligands, by our group,5,6 as well as with sterically
more demanding NHC ligands by Sadighi et al. and others.7,8
The three synthetically most broadly employed diborane(4)
reagents are undoubtedly B2pin2, B2cat2 (cat = 1,2-O2C6H4) and
¨
¨
Institut fur Anorganische und Analytische Chemie, Technische Universitat Braunschweig,
Hagenring 30, 38106 Braunschweig, Germany. E-mail: ch.kleeberg@tu-braunschweig.de
† Dedicated to Prof. Dr Todd B. Marder on occasion of his 65th birthday for his
outstanding contributions in boron chemistry and in particular transition metal
catalysed borylation reactions.
‡ Electronic supplementary information (ESI) available: Additional analytical and
structural data. CCDC 2008213–2008218. For ESI and crystallographic data in CIF
or other electronic format see DOI: 10.1039/d0nj03166f
Scheme 1 Formation of [(IDipp)Cu–Bneop] (2) from [(IDipp)Cu–OtBu]
and B2neop2 (1) as reported by Sadighi et al.8
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
SIDipp = 1,3-bis(2,6-diisopropylphenyl)-imidazolidin-2-ylidene) 1,3-bis(tert-butyl)imidazol-2-ylidene or 1,3-bis(iso-propyl)imidazol-
and boryl ligands such as Bpin, Bcat, Bdmab (dmab = 1,2- 4,5-dimethyl-2-ylidene, that we have successfully employed to
(NMe)2C6H4) or Bdbab (dbab = 1,2-(NBn)2C6H4).3,5,7 In fact the obtain related copper(I) boryl complexes.6b
low stability of 2 is more reminiscent to copper boryl complexes
For 2 at room temperature a very broad singlet signal at
with sterically less demanding NHC ligands.6
40 ppm (Dw = 3510 Hz) was observed in the 11B{1H} NMR
1
2
Given our work on reactive, in particular, sterically little spectrum, whereas at ꢀ40 1C no signal was observed due to the
demanding NHC copper boryl complexes and their decomposition excessive line broadening.11 For 3 however no 11B NMR signal
pathways we endeavoured to (re-)investigate 2 and its sterically was unambiguously identified even at room temperature. Which is
more demanding congener [(IDipp*)Cu–Bneop] (3) (IDipp* = 1,3- rationalised with the reduced transverse relaxation time (T2), and
bis(2,6-(diphenylmethyl)-4-methylphenyl)imidazol-2-ylidene).
therefore even larger linewidths than observed for 2 for less flexible
and mobile 3.
For 2, however, the NMR data are in line with the 11B NMR
chemical shifts observed for related mononuclear NHC copper
boryl complexes such as [(IDipp)Cu–Bpin] 41.7 ppm,3 [(SIDipp)Cu–
Results and discussions
Synthesis and characterisation
Bcat] 45 ppm (Dw = 1700 Hz),5b and [(IDipp)Cu–Bdmab] 44.1 ppm
1
25a
The reaction of [(IDipp)Cu–OtBu] with B2neop2 (1) at low
temperatures following Sadighi’s original procedure furnished
[(IDipp)Cu–Bneop] (2) (Scheme 1).8 An alternative, slightly
modified protocol (ꢀ35 1C vs. r.t./ꢀ40 1C, different handling
and work-up) was found to be equally suitable (vide infra).
However, both procedures led in our hands to varying yields
and purities of isolated 2. Nonetheless, 2 was isolated repeatedly
as its PhMe solvate 2(PhMe)x (x = 0.22–0.90). Whilst the X-ray
structure determination (vide infra) corroborates the solvate
2(PhMe), elemental analysis and NMR data suggest a gradual
removal of the co-crystallised PhMe upon drying in vacuo at
ambient temperature, possibly accompanied by decomposition.
Despite the low stability of 2 in solution at ambient tem-
perature, as reported by Sadighi and co-workers, we were able to
characterise 2(PhMe)x (x = 0.75–0.85) by 1H and 13C NMR
spectroscopy at ꢀ40 1C as well as at room temperature, although
the latter was accompanied by severe decomposition (vide infra).
We concluded that increasing the steric demand of the NHC
ligand should increase the stability of the NHC copper Bneop
complex. We chose the established complex [(IDipp*)Cu–OtBu] as
the starting material, essentially replacing the iso-propyl groups
of the IDipp ligand by diphenylmethyl groups (Scheme 2).10e
Indeed, the complex [(IDipp*)Cu–Bneop] (3) can be isolated
upon reaction of [(IDipp*)Cu–OtBu] with 1 reproducibly in 70%
isolated yield and is much more stable than its leaner congener
2 (Fig. S3, ESI‡).13 In line with that, so far no boryl complexes
of Bneop could be isolated using other NHC ligands such as
1
(Dw = 1020 Hz). The same is true for the carbene carbon
2
13C NMR shifts of 186.5 ppm and 187.0 ppm of 2 and 3,
respectively, that compare well with those of [(IDipp)Cu–Bpin]
187.2 ppm,3 and [(IDipp)Cu–Bdmab] 187.1 ppm.5a
Solid-state structures
Single crystals of 2 were obtained from PhMe/n-pentane and
from THF/n-pentane at ꢀ40 1C as the solvates 2(PhMe) and
2(THF), respectively. 2(PhMe) forms colourless plates and com-
prises two independent molecules in the asymmetric unit in a
0
%
triclinic unit cell in P1 (Z = 4, Z = 2). Whereas 2(THF) forms
colourless needles and comprises only one molecule in the
asymmetric unit in the monoclinic unit cell in P21/n (Z = 4, Z0 = 1)
(Fig. S6, ESI‡).13 In 2(PhMe) the two independent molecules A
and B of 2 form chains along the b axis, with Hꢁꢁ ꢁO and Hꢁ ꢁ ꢁp
interactions as the most pronounced intermolecular interactions
(Fig. 1). A Hirshfeld surface analysis reveals that besides the
ubiquitous Hꢁ ꢁ ꢁH contacts with minimal di and de of 0.9–1.1 Å
(surface area: A 88.8%, B 80.1%) the 2D fingerprint plots exhibit
two distinct features (Fig. 1).12 Most pronounced are the Hꢁ ꢁ ꢁO
contacts (surface area: A 5.6%, B 4.2%) with very distinct
sharp features in the 2D fingerprint plots with minimal di + de
of 2.1–2.2 Å. Whilst generally analogous, the Hꢁ ꢁ ꢁO interactions
realised in the two independent molecules are slightly different
as evidenced by the unsymmetrical 2D fingerprint plots as well as
slightly different di and de distances (vide infra).13 Less pronounced,
but in agreement with the data reported for polycyclic aromatic
hydrocarbons are the HꢁꢁꢁC contacts (surface area: A 11.6%, B
12.6%) with the minimal di + de around 2.6 Å associated with the
Hꢁꢁꢁp interaction H2ꢁꢁꢁCt1A and H3A0ꢁꢁꢁCt1 (Fig. S4, ESI‡).12a,13
The Hꢁ ꢁ ꢁO interactions are also illustrated by the short
Hꢁ ꢁ ꢁO distance of H3ꢁ ꢁ ꢁO2A 2.24 Å and H2Aꢁ ꢁ ꢁO10 2.35 Å as
well as the angles C3ꢁ ꢁ ꢁH3ꢁ ꢁ ꢁO2A 172.91 and C2Aꢁ ꢁ ꢁH2Aꢁ ꢁ ꢁO10
161.11 (Fig. 1). Although, these data must be considered with
some care due to the use of a riding model in the refinement of
the hydrogen atoms, it can certainly be stated that the distances
are short compared to the sum of the respective van-der-Waals radii
of 2.64 Å.14 The intramolecular CCarbene–Cu and B–Cu distances and
the B–Cu–CCarben angles (Fig. 1) are in line with the respective data
for [(IDipp)Cu–Bpin] (C–Cu 1.937(2) Å, B–Cu 2.002(3) Å, C–Cu–B
168.1(2)1) and [(SIDipp)Cu–Bcat] (C–Cu 1.930(1) Å, B–Cu 1.984(2) Å,
Scheme 2 Formation of [(IDipp*)Cu–Bneop] (3) from [(IDipp*)Cu–OtBu]
and B2neop2 (1).
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
Fig. 1 Section of the solid state structure of [(IDipp)Cu–Bneop] (2) from 2(PhMe) (left) and Hirshfeld surface analysis data (right); ellipsoids at the 50%
probability level, selected H and C atoms shown with arbitrary radius. Selected distances (Å) and angles (1): C1–Cu1 1.932(4), B1–Cu1 2.007(6), H3ꢁ ꢁ ꢁO2A
2.240(4), C3ꢁ ꢁ ꢁO2A 3.185(6), H2ꢁ ꢁ ꢁCt1A 3.1819(4), B1–Cu1–C1 174.8(2), C3ꢁ ꢁ ꢁH3ꢁ ꢁ ꢁO2A 172.9(2), +(N,C,N,Cu)/(O,B,O,Cu) 47.4(3); C1A–Cu1A 1.928(5),
B1A–Cu1A 2.000(5), H2AꢁꢁꢁO10 2.351(4), C2AꢁꢁꢁO2A 3.265(6), H3AꢁꢁꢁCt10 2.816(3), B1–Cu1–C1 174.6(2), C2AꢁꢁꢁH2AꢁꢁꢁO10 161.1(3), +(N,C,N,Cu)/(O,B,O,Cu) 45.5(8).
C–Cu–B 172.52(7)1).3,5b The interplanar angles +(N,C,N,Cu)/(O,B,O,Cu)
,
however, exhibit much more variation amongst these three
complexes ([(IDipp)Cu–Bpin] 40.01, [(SIDipp)Cu–Bcat] 71.9(2)1)
(Fig. 1).3,5b
The THF solvate 2(THF) exhibits comparable geometrical
properties as the PhMe solvate. In particular, an analogous
chain packing motif with distinct Hꢁ ꢁ ꢁO and Hꢁ ꢁ ꢁp interactions
is observed; albeit the inter chain interactions are different
(Fig. S6 and S7, ESI‡).13
Single crystals of 3 were obtained from an ethereal solution
by slow evaporation at room temperature under an inert atmo-
sphere. 3 forms colourless laths with one molecule in the
asymmetric unit in an orthorhombic unit cell in Pbca (Z = 8,
Z0 = 1).13,15 The Hirshfeld surface analysis reveals that Hꢁ ꢁ ꢁH
contacts with minimal di and de of 0.9 Å (surface area: 72.8%)
and Hꢁ ꢁ ꢁC contacts (surface area: 24.4%) are the most pro-
nounced interactions (Fig. 2 and Fig. S5, ESI‡).13 Whereas, as
an effect of the increased steric shielding of the ligand and
in contrast to the structures of 2(PhMe) and 2(THF), inter-
molecular Hꢁ ꢁ ꢁO contacts are only marginal (surface area:
1.3%, di + de 4 2.5 Å).13 The molecular geometry with respect
Fig. 2 Molecular structure of [(IDipp*)Cu–Bneop] (3) and Hirshfeld surface
analysis data (insets); ellipsoids at the 50% probability level, H atoms are
omitted for clarity. Selected distances (Å) and angles (1): C1–Cu1 1.935(2),
B1–Cu1 2.020(2), B1–Cu1–C1 171.36(8), +(N,C,N,Cu)/(N,B,N,Cu) 82.6(6).
to the CCarbene–Cu and B–Cu distances and the B–Cu–CCarben and one with an IDipp scaffold. The corresponding 11B NMR
angles are, however, in line with the respective data for 2 and data exhibit however, two signals. Comparison with an authentic
other IDipp/SIDipp boryl complexes (vide supra). The interpla- sample of the diborane(4) 1 indicates that 1 is the neop based
nar angle +(N,C,N,Cu)/(O,B,O,Cu) is with 82.61, in contrast to the decomposition product detected by 1H NMR spectroscopy and is
IDipp/SIDipp boryl complexes (vide supra), almost rectangular, also identified in the 11B{1H} NMR spectrum (28.8 ppm,
1
certainly an effect of the increased steric encumbrance of 3.
Dw = 240 Hz). Moreover, 1 has occasionally been deposited as
2
single crystals from solution of decomposed 2, as verified by unit
cell determination.
Stability and decomposition
The decomposition of 2 is readily followed by in situ 1H and
1
However, the second signal at 17.7 ppm (Dw = 130 Hz)
2
11B{1H} NMR spectroscopy (Fig. 3 and Fig. S1a, ESI‡).13 As indicates the presence of a second boron containing species,
reported for other NHC copper boryl complexes the decom- possibly a four coordinate boric acid derivative, e.g. the spiro
position is quite selective.6b The 1H NMR spectra show the borate [Bneop2]ꢀ (vide infra). The second species in the 1H NMR
formation of essentially two species, one with a neop scaffold spectrum is not as readily assigned, the NMR signals of an
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
nicely with our earlier findings that sterically little encumbered NHC
copper boryl complexes are extremely unstable.6b Furthermore, the
observed packing pattern in the solid-state structures of 2
including distinct intermolecular Hꢁ ꢁ ꢁO interactions suggests
a certain accessibility of the Bneop moiety and is in line with a
higher reactivity/lower stability of this complex.
Additional decomposition products of 2
During the studies on 2 three different additional decomposition
products [(IDipp)2Cu][Bneop2], [(IDipp)6Cu55] and [(IDipp)12Cu179
]
were repeatedly, although not very reproducibly obtained as a
few single crystals. The cationic bis-NHC copper(I) complex
[(IDipp)2Cu][Bneop2], was obtained performing the reaction of
[(IDipp)Cu–OtBu] with 1 in THF. The formation of spiroborates
as a side-product of reactions of diboranes(4) has been observed
earlier, though, in particular with B2cat2.5b,16
Fig. 3 Decomposition of 2 monitored by NMR spectroscopy (C6D6, r.t.).11
As this complex is a copper(I) complex and as such likely not
associated with the reductive decomposition of 2, it is not
discussed here any further.13
authentic sample of the free carbene IDipp does not quite fit to
the signals of the decomposition product. However, considering the
very small deviations and the observation of the free NHC ligand as
decomposition product of other boryl complexes it may be con-
cluded that the small deviations are caused by concentration effects
and/or weak dynamic coordination to one of the boron containing
decomposition products. The kinetics of the decomposition appears
to be in agreement with the report of substantial decomposition
The remarkable metalloid cluster [(IDipp)6Cu55] was repeatedly
obtained from reaction mixtures of 1 and [(IDipp)Cu–OtBu] in
PhMe or C6D6 at room temperature, whereas [(IDipp)12Cu179] was
repeatedly obtained from THF/n-pentane or Et2O/n-pentane
solution at room temperature. It has to be emphasised that both
clusters were obtained only in miniscule amounts and only in
about 30% of the – apparently identical – experiments, and are
solely characterised by single crystal X-ray determinations.
Nonetheless, both complexes have been obtained repeatedly
and even from two different solvents. Whilst ongoing work is
directed towards a preparative access to these clusters, the
structures of these clusters may be discussed here due to their
relevance for the decomposition of 2.
1
after 20 min by Sadighi et al.8 The in situ H NMR data of the
decomposition suggests a half-life of 2 in C6D6 solution at room
temperature of about 1.5 h (Fig. S1b, ESI‡).13 The decomposition of
2 is further accompanied by the formation of a dark precipitate.
Though this could not be isolated, it may be concluded to
be elemental copper, in line with the findings with other
NHC copper boryl complexes.6b A powder diffraction pattern
recorded of the dark residue obtained after drying of a solution
of in situ generated and decomposed 2 in vacuo supports the
formation of the free carbene IDipp and 1 as the major
decomposition products. However, no clear evidence for the
formation of elemental copper or low-valent copper clusters
(vide infra) was obtained (Fig. S2, ESI‡).13
Finally it may be concluded, that the decomposition of 2 is
surprisingly fast in comparison to the stability of the congeners
[(IDipp)Cu–Bpin], [(ClIDipp)Cu–Bpin], [(SIDipp)Cu–Bcat], [(IDipp)Cu–
Bdmab] and [(IDipp)Cu–Bdbab].3,5,7 The decomposition pathway,
however, appears generally to be comparable to the one observed
for sterically little encumbered NHC copper boryl complexes,
leading to the symmetrical diborane(4) derivative 1 as oxidative
coupling product and, presumably, the free NHC and elemental
copper.
The copper atoms in the cluster [(IDipp)6Cu55] adopt –
approximately – icosahedral symmetry (Fig. 4). It may be
dissected in an inner shell comprising of a centred – slightly
distorted – icosahedron of copper atoms (Cu13) and an outer
shell comprising of 42 copper atoms.13 The outer shell itself
comprises an icosahedron of copper atoms (Cu12) with additional
copper atoms in the middle of all icosahedron edges (Cu30) (Fig. 4,
right). Crystallographically the cluster does not exhibit icosahedral
symmetry but is situated only on a centre of inversion in a
0
1
%
spacegroup of the type P1 (Z = 1, Z = 2).
However, in several instances additional decomposition
products were obtained in minuscule amounts but in single
crystalline form (vide infra).
In contrast to 2, the steric more encumbered complex 3 does
not exhibit any significant decomposition within 16 h at room
temperature (Fig. S3, ESI‡).13 Hence, the stability of copper boryl
complexes depends on both the steric properties of the auxiliary,
NHC, ligand as well as the properties of the boryl ligand. This agrees
Fig. 4 Left: Views of Ball and Stick model of [(IDipp)6Cu55]. Right: Ball and
Stick of the outer shell of [(IDipp)6Cu55]. NHC only represented by their
carbene carbon atoms (grey).13
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
The carbene carbon atoms of the six, pairwise equivalent, centre of inversion present, the middle pentagon Cu1w–Cu5w is
IDipp ligands form an approximate octahedron, coordinating disordered over two positions and so is this entire Cu19 motif.
non-icosahedron copper atoms. The coordinated copper atoms Consequently, the entire Cu179 cluster is not centrosymmetric
are slightly displaced from their idealised positions on the and possess only one approximate five-fold axis (vide infra). This
middle of icosahedron edges. The Cuꢁ ꢁ ꢁCu distances are disorder than propagates through the 2nd (52 Cu atoms) and 3rd/
between 2.39 Å, found between the inner and outer icosahedra outer (107 Cu atoms) shell of this cluster (Fig. 5 and Fig. S10, S11,
along the approximate fivefold axis, and 2.72 Å involving copper ESI‡).13 This account so far for 178 copper atoms, one additional
atoms coordinated by an IDipp ligand. Nonetheless, the aver- copper atom is found disordered over 10 pairwise symmetry
age Cuꢁ ꢁ ꢁCu distance around 2.55 Å (Fig. S9, ESI‡)13 agrees well equivalent positions on the surface of the Cu178 motif.21 Finally,
with that in elemental copper of 2.56 Å.17 The CNHC–Cu the 12 NHC ligands coordinate 12 copper atoms of the outer
distances are with a range of 1.94–1.98 Å slightly longer than shell, obeying inversion symmetry. The structure of this copper
those in IDipp and SIDipp copper boryl complexes (vide supra), cluster as discussed here is one interpretation of this disorder
where typically distances of around 1.93 Å are found.3,5 The and is disputable; for a detailed reasoning leading to this
structure of [(IDipp)6Cu55] is related to that one of the mixed structure see the ESI‡ provided.13 It is also noted that the Cu55
NHC/phosphine copper cluster [(Me2IiPr)10Cu23(PMe3)2] in so cluster motif as found in [(IDipp)6Cu55] is also found as a partial
far, as both clusters comprise centred copper icosahedra as motif in [(IDipp)12Cu179] (around Cu54, Fig. S12, ESI‡).13
central motif. The latter one, however, comprises 10 additional
The Cuꢁ ꢁ ꢁCu distances in the cluster [(IDipp)12Cu179] cover a
copper atoms above the 10 trigonal equatorial faces.6b More- range from 2.35 to 2.70 Å with a maxima around 2.55 Å
over, both the Cuꢁ ꢁ ꢁCu and the CNHC–Cu distances in the latter and thus are comparable with those found in [(IDipp)6Cu55]
cluster are comparable with those in [(IDipp)6Cu55].6b It should (Fig. S13, ESI‡).13 The CNHC–Cu distances that are in a range of
be further mentioned that centred icosahedra are also realised 1.93–1.97 Å and, hence, also comparable to those in [(IDipp)6Cu55].
in the low-valent copper hydrido clusters [Cu25H22(PPh3)12]Cl
In summary, it is stated that the decomposition of 2 leads to
and [Cu29Cl4H22(Ph2phen)12]Cl,18 and, moreover, similar icosa- remarkable and unprecedented metalloid copper clusters. That
hedral structures are discussed for the pivotal gold cluster complements our results on the sterically little encumbered
[Au55(PPh3)12Cl6].19 It is also worth noting that copper mono- NHC and phosphine copper boryl complexes, where also –
layers with bound NHC ligands have been reported recently.20 smaller – copper clusters have been observed as decomposition
[(IDipp)12Cu179] crystallises in the monoclinic spacegroup products.6 Finally, it must be emphasised, that those copper
type P21/n and is situated on an inversion centre, however, the clusters are subject of ongoing studies and that they have so far
cluster itself does not appear to possess inversion symmetry, been only characterised by X-ray crystallography. Hence, it
resulting in extensive disorder of the copper atoms within cannot be fully excluded that additional hydrido ligands are
[(IDipp)12Cu179]. Similar as [(IDipp)6Cu55] the bigger cluster may present, undetected due to their low scattering power. Moreover,
be dissected into different shells. The inner core consists of 2 for both clusters co-crystallised solvent was found that could not
equivalent copper atoms (C54 and C540) one of them centring an be adequately refined and was removed mathematically.13 How-
icosahedron, the other comprising an apical position of this ever, none of the solvent molecules was close to the cluster Cu
icosahedron. This latter copper atom, however, centres again a atoms, excluding an additional solvent coordination. The single
two-fold base-capped pentagonal prism (Fig. 5, inset). Due to the crystal X-ray structure determinations were for both clusters, but in
particular for [(IDipp)12Cu179], not straightforward and must be
considered with the necessary care.13
Conclusion
IDipp copper boryl complexes have gained considerable attention as
prototypic copper boryl model complexes since their introduction by
Sadighi and co-workers.3 However, the first complexes of the Bneop
ligand, [(IDipp)Cu–Bneop] (2), has only recently been reported –
though not comprehensively characterised – and exhibits a
comparable low stability in solution.8 The Bneop complexes
[(IDipp)Cu–Bneop] (2) and the sterically more encumbered
[(IDipp*)Cu–Bneop] (3) are isolated and thoroughly char-
acterised by multinuclear NMR spectroscopy in solution as well
as structurally in the solid state by single crystal X-ray diffraction.
Both the spectroscopic data as well as the structural data do not
exhibit significant differences to those of other IDipp copper
boryl complexes. The, indeed, rapid decomposition of 2 was found
to lead to predominantly the free NHC ligand and B2neop2 (1) and,
Fig. 5 Left: Views of Ball and Stick model of [(IDipp)12Cu179]; NHC only
represented by the carbene carbon atoms.13 inset: detail of the inner core
of [(IDipp)12Cu179].
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
presumably, elemental copper. Whereas 3 does not show Nova A instrument, using mirror-focused CuKa radiation. The
appreciable decomposition under comparable conditions. The reflections were indexed, integrated and appropriate absorption
destabilisation of 2 compared to other IDipp copper boryl corrections were applied as implemented in the CrysAlisPro
complexes appears to be an intrinsic effect of the Bneop ligand software package.23b The structures were solved employing the
but can be compensated by the use of the sterically more program SHELXT and refined anisotropically for all non-
demanding IDipp* ligand. Besides the mentioned decomposition hydrogen atoms by full-matrix least squares on all F2 using
products of 2 the two remarkable copper clusters [(IDipp)6Cu55] SHELXL software.23c–e Unless noted otherwise hydrogen atoms
and [(IDipp)12Cu179] were obtained occasionally in miniscule were refined employing a riding model; methyl groups were
amounts as single crystals. This shows, together with our earlier treated as rigid bodies and were allowed to rotate about the
reports,6 that the formation of low-valent copper clusters is a E–CH3 bond. During refinement and analysis of the crystallo-
relevant decomposition pathway of copper boryl complexes. This graphic data the programs WinGX, OLEX,2 PLATON/SQUEEZE,
and the possible role and mechanistic implications of those Mercury and Diamond were used.23f–j Unless noted otherwise the
clusters have to be considered in the further development of the shown ellipsoids represent the 50% probability level and hydro-
flourishing field of copper mediated borylation reactions.
gen atoms are omitted for clarity. Adapted numbering schemes
may be used to improve the readability.
Synthetic procedures
Experimental part
[(IDipp)Cu–Bneop] (2). Complex 2 was prepared following
Sadighis procedure.8 Alternatively the procedure given below was
employed. Though both procedures yielded repeatedly pure 2,
both procedures are in our hands not very reliable with respect to
the obtained yields and the purity of the compounds. Moreover,
we were unable to perform the reaction successfully at a larger
scale. [(IDipp)Cu–OtBu] (50.0 mg, 95 mmol, 1.0 eq.) was dissolved
in PhMe (2 mL) in a screw-cap vial and a suspension of 1
(21.5 mg, 95 mmol, 1.0 eq.) in PhMe (0.5 mL) was added at
ambient temperature in the dark. After 5 min of shaking PhMe
was added until a clear solution is obtained (ca. 2.5 mL). The by
now dark solution was left to crystallise at ꢀ40 1C for 36 h to
obtain the solvate 2(PhMe) in crystalline form suitable for X-ray
diffraction analysis. The supernatant mother liquor was decanted
and the residues washed with n-pentane (3 ꢂ 2 mL) and dried
briefly in vacuo to obtain 2(PhMe)40.8, in agreement with the
1H NMR data (10.5 mg, 16 mmol, 17% as 2(PhMe)). 2(PhMe)
obtained may be recrystallised from THF/n-pentane at ꢀ40 1C to
obtain single-crystals of 2(THF) suitable for X-ray diffraction
analysis.
General considerations
[(IDipp)Cu–OtBu]10a,b and [(IDipp*)Cu–OtBu]10c–e were prepared
according to literature procedures, however, KOtBu was sub-
stituted for NaOtBu throughout the synthesis. All other com-
pounds were commercially available and were used as received;
their purity and identity was checked by appropriate methods.
All solvents were dried using MBraun solvent purification
systems, deoxygenated using the freeze–pump–thaw method
and stored under purified nitrogen. All manipulations were
performed using standard Schlenk techniques under an atmo-
sphere of purified nitrogen or in a nitrogen-filled glove box
(MBraun). NMR spectra were recorded on Bruker Avance II 300,
Avance III HD 300, Avance III 400, Avance III 500 or Avance 600
spectrometers. For air sensitive samples NMR tubes equipped
with screw caps (WILMAD) were used and the solvents were
dried over potassium/benzophenone and degassed. Chemical
shifts (d) are given in ppm, using the (residual) resonance signal
of the solvents for calibration (C6D6 : 1H NMR: 7.16 ppm, 13C
NMR: 128.06 ppm, THF-d8 : 1H NMR: 1.72 ppm, 13C NMR: 25.31
ppm).22 11B chemical shifts are reported relative to BF3ꢁEt2O. 13
C
and 11B NMR spectra were recorded employing composite pulse
1H decoupling. 2D NMR techniques were employed if necessary
to assign the individual signals (1H–1H NOESY (1 s mixing time),
M.p. Decomposition to black material above 75 1C. Found:
C, 70.51; H, 8.27; N, 4.31. Found: C, 68.11; H, 7.95; N, 4.71
(prolonged drying). Found: C, 67.66; H, 8.36; N, 4.82 (very long
drying). Calc. for C12H16B2O4 (2): C, 68.02; H, 8.21; N, 4.96. Calc.
for C32H46BCuN2O2 (2(PhMe)): C, 71.27; H, 8.28; N, 4.26. Calc.
1
1
1H–1H COSY, H–13C HSQC and H–13C HMBC). Unless noted
otherwise, Melting points were determined in flame-sealed
capillaries under nitrogen using a Bu¨chi 535 apparatus and are
not corrected. Elemental analyses were performed at the Institut
fu¨r Anorganische und Analytische Chemie of the Technische
2
for C36.6H51.3BCuN2O2 (2(PhMe) ): C, 70.29; H, 8.26; N, 4.47.
3
1
Calc. for C35.5H50BCuN4O4 (2(PhMe) ): C, 69.77; H, 8.25; N, 4.58.
2
dH(500.3 MHz, THF-d8, ꢀ40 1C) 0.64 (6 H, s, C(CH3)2), 1.20
¨
Universitat Braunschweig using an Elementar vario MICRO cube
(12 H, d, J = 6.9 Hz, CH(CH3)2), 1.33 (12 H, d, J = 6.8 Hz,
CH(CH3)2), 2.61 (4 H, sept., J = 6.8 Hz, CH(CH3)2), 3.02 (4 H, s,
CH2), 7.30 (4 H, d, J = 7.7 Hz, 3-CHAr), 7.42 (2 H, s, NCH), 7.43
(2 H, t, J = 7.7 Hz, 4-CHAr). dC(125.8 MHz, THF-d8, ꢀ40 1C) 23.1
instrument.
X-ray diffraction studies
The single crystals were transferred into inert perfluoroether oil (C(CH3)2), 23.8 (CH(CH3)2), 25.6 (CH(CH3)2), 29.5 (CH(CH3)2),
inside a nitrogen-filled glovebox and, outside the glovebox, 31.9 (C(CH3)2), 70.1 (CH2), 123.6 (NCH), 124.3 (3-CHAr),
rapidly mounted on top of a human hair, Mitegen loop or a 130.4 (4-CHAr), 136.2 (1-CAr), 146.5 (2-CAr), 186.5 (CCarbene).
Hampton loop and placed in the cold nitrogen gas stream on dB(160.5 MHz, THF-d8, ꢀ40 1C) No 11B NMR signal was
the diffractometer.23a The data were either collected on a detected. Sample composition according to 1H NMR integration
Rigaku Oxford Diffraction Synergy-S or an Oxford Diffraction 2(PhMe)0.85. dH(600.1 MHz, C6D6, r.t.) 0.61 (6 H, s, C(CH3)2),
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
1.10 (12 H, d, J = 6.9 Hz, CH(CH3)2), 1.54 (12 H, d, J = 6.8 Hz, Notes and references
CH(CH3)2), 2.67 (4 H, sept., J = 6.9 Hz, CH(CH3)2), 3.27 (4 H, s,
1 For reviews see: (a) L. Dang, Z. Lin and T. B. Marder, Chem.
CH2), 6.22 (2 H, s, NCH), 7.07 (4 H, d, J = 7.8 Hz, 3-CHAr), 7.14
(2 H, t, J = 7.9 Hz, 4-CHAr). dC(150.9 MHz, C6D6, r.t.) 23.0
(C(CH3)2), 23.7 (CH(CH3)2), 25.4 (CH(CH3)2), 29.1 (CH(CH3)2),
31.6 (C(CH3)2), 70.0 (CH2), 121.9 (NCH), 124.0 (3-CHAr), 129.03
´
´
Commun., 2009, 3987; (b) J. Cid, H. Gulyas, J. J. Carbo and
´
E. Fernandez, Chem. Soc. Rev., 2012, 41, 3558; (c) E. C. Neeve,
S. J. Geier, I. A. I. Mkhalid, S. A. Westcott and T. B. Marder,
´
Chem. Rev., 2016, 116, 9091; (d) J. Cid, J. J. Carbo and
(4-CHAr), 135.3 (1-CAr), 145.9 (2-CAr), CCarbene was not detected.
11
´
E. Fernandez, Chem. – Eur. J., 2012, 18, 12794.
1
dB(96.3 MHz, C6D6, r.t.) 40 (br. s, Dw = 3500 Hz). Sample
2
1
2 K. Takahashi, T. Ishiyama and N. Miyaura, J. Organomet.
Chem., 2001, 625, 47.
3 D. S. Laitar, P. Mu¨ller and J. P. Sadighi, J. Am. Chem. Soc.,
2005, 127, 17196.
composition according to H NMR integration 2(PhMe)0.75
.
[(IDipp*)Cu–Bneop] (3). [(IDipp*)Cu–OtBu] (26.7 mg, 25.4 mmol,
1.0 eq.) and 1 (5.7 mg, 25.4 mmol, 1.0 eq.) were combined and dis-
solved in THF (2 mL). After approximately five minutes at room tem-
perature, all volatiles were removed under reduced pressure. The oily
residue was washed with n-pentane (3 ꢂ 2 mL) at room temperature
and dried in vacuo to give 3 (19.5 mg, 17.9 mmol, 70%) as a pale
yellow solid. Single crystals were obtained by evaporation of an
ethereal solution of 3 at room temperature under inert conditions.
4 Besides copper boryl complexes derived from diborane(4)
derivatives, and hence of direct relevance for copper cata-
lysed borylation reactions, copper boryl complexes have also
been obtained from boryl lithium species: (a) T. Kajiwara,
T. Terabayashi, M. Yamashita and K. Nozaki, Angew. Chem., Int.
Ed., 2008, 47, 6606; (b) Y. Okuno, M. Yamashita and K. Nozaki,
Eur. J. Org. Chem., 2011, 3951; (c) Y. Segawa, M. Yamashita and
K. Nozaki, Angew. Chem., Int. Ed., 2007, 46, 6710.
5 (a) C. Borner and C. Kleeberg, Eur. J. Inorg. Chem., 2014, 2486;
(b) W. Drescher and C. Kleeberg, Inorg. Chem., 2019, 58, 8215.
6 (a) C. Borner, L. Anders, K. Brandhorst and C. Kleeberg, Organo-
metallics, 2017, 36, 4687; (b) C. Kleeberg and C. Borner, Organo-
metallics, 2018, 37, 4136.
M.p. Decomposition to black material above 189 1C. Found:
C, 80.78; H, 6.30; N, 2.82. Calc. for C74H66BCuN2O2 (3): C, 81.56;
H, 6.10; N, 2.57. Repeated attempts to obtain a more satisfactory
elemental analysis failed. dH(300.1 MHz, C6D6, r.t.) 0.78 (6 H, s,
C(CH3)2), 1.60 (6 H, s, 4-CAr–CH3), 3.48 (4 H, s, CH2), 5.53 (2 H, s,
NCH), 5.77 (4 H, s, CH(CPh)2(CAr)), 6.99 (4 H, s, 3-CHAr, over-
0
lapping), 6.99 (20 H, s, CHPh , overlapping), 7.13 (4 H, app. t, J =
7.6 Hz, 4-CHPh), 7.36 (8 H, app. t, J = 7.7 Hz, 3-CHPh), 7.73 (8 H, d,
J = 7.7 Hz, 2-CHPh). dC(100.7 MHz, C6D6, r.t.) 21.2 (4-CAr–CH3), 23.2
(C(CH3)2), 31.7 (C(CH3)2), 51.7 (s, CH(CPh)2(CAr)), 70.3 (CH2), 122.8
7 (a) K. Semba, M. Shinomiya, T. Fujihara, J. Terao and
Y. Tsuji, Chem. – Eur. J., 2013, 19, 7125; (b) C. M. Wyss,
J. Bitting, J. Bacsa, T. G. Gray and J. P. Sadighi, Organome-
tallics, 2016, 35, 71.
0
0
(NCH), 126.6 (s, 4-CHPh ), 126.8 (s, 4-CHPh), 128.5 (s, 3-CHPh ), 129.0
0
(s, 3-CHPh), 130.0 (s, 2-CHPh ), 130.6 (s, 3-CHAr), 130.8 (s, 2-CHPh),
8 A. J. Jordan, P. K. Thompson and J. P. Sadighi, Org. Lett.,
2018, 20, 5242.
0
135.2 (s, 1-CAr), 140.1 (s, 4-CAr), 141.7 (s, 1-CPh ), 143.5 (s, 2-CAr),
144.1 (s, 1-CPh), 187.0 (CCarbene, 1H–13C HMBC only). dB(160.5 MHz,
C6D6, r.t.) No 11B NMR signal was detected.
9 (a) P. Nguyen, G. Lesley, N. J. Taylor, T. B. Marder, N. L.
Pickett, W. Clegg, M. R. J. Elsegood and N. C. Norman,
Inorg. Chem., 1994, 33, 4623; (b) F. J. Lawlor, N. C. Norman,
N. L. Pickett, E. G. Robins, P. Nguyen, G. Lesley, T. B.
Marder, J. A. Ashmore and J. C. Green, Inorg. Chem., 1998,
37, 5282; (c) W. Clegg, M. R. J. Elsegood, F. J. Lawlor,
N. C. Norman, N. L. Pickett, E. G. Robins, A. J. Scott,
P. Nguyen, N. J. Taylor and T. B. Marder, Inorg. Chem.,
[(IDipp)6Cu55]. The synthesis of 2 was conducted as described
above, but instead of crystallisation at ꢀ40 1C the dark reaction
mixture was left at room temperature under inert condition for
several days. In about 30% of the attempts miniscule amounts of
small (longest axis o0.1 mm) dark black prisms had deposited that
were analysed by single crystal X-ray structure determination.
[(IDipp)12Cu179]. A saturated solution of 2 in Et2O (0.5 mL)
was layered with n-pentane (2 mL) at room temperature. After a
few days at room temperature a dark precipitate had formed. In
about 30% of the attempts, additionally miniscule amounts of
small (longest axis o0.1 mm) dark prisms had deposited that
were analysed by single crystal X-ray structure determination.
1998, 37, 5289; (d) M. Eck, S. Wurtemberger-Pietsch,
A. Eichhorn, J. H. J. Berthel, R. Bertermann, U. S. D. Paul,
H. Schneider, A. Friedrich, C. Kleeberg, U. Radius and
T. B. Marder, Dalton Trans., 2017, 46, 3661.
¨
10 (a) L. Hintermann, Beilstein J. Org. Chem., 2007, 3, 22;
(b) V. Jarkauskas, J. P. Sadighi and S. L. Buchwald, Org.
Lett., 2003, 5, 2417; (c) M. Hans, J. Lorkowski, A. Demonceau
and L. Delaude, Beilstein J. Org. Chem., 2015, 11, 2318;
(d) W.-J. Yoo, T. V. Q. Nguyen and S. Kobayashi, Angew.
Chem., Int. Ed., 2014, 53, 10213; (e) J. Zhai, A. S. Filatov,
G. L. Hillhouse and M. D. Hopkins, Chem. Sci., 2016, 7, 589.
11 To unambiguously identify the very broad 11B NMR signal of
2 the spectrum was processed by subtraction of a suitable
background spectrum. A Lorentz type window function (LB =
30 Hz) was used. C. Kleeberg, A. G. Crawford, A. S. Batsanov,
P. Hodgkinson, D. C. Apperley, M. S. Cheung, Z. Lin and
T. B. Marder, J. Org. Chem., 2012, 77, 785.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
C. K., C. B. and W. D. gratefully acknowledges support by a Research
Grant (KL 2243/5-1) of the Deutsche Forschungsgemeinschaft
(DFG). The authors thank AllyChem Co. Ltd. for a generous gift
of B2neop2.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
12 (a) M. A. Spackman and J. J. McKinnon, CrystEngComm, 21 It was considered that this electron density might represent
2002, 4, 378; (b) M. A. Spackman and D. Jayatilaka, Cryst-
EngComm, 2009, 11, 19.
13 See ESI‡ for details.
hydrido ligand(s). The observed distances of the electron
density maxima to the cluster Cu atoms in a range of 2.25–2.55 Å
are more suggestive for Cu atoms than for hydrogen atoms,
14 M. Mantina, A. C. Chamberlin, R. Valero, C. J. Cramer and
D. G. Truhlar, J. Phys. Chem. A, 2009, 113, 5806.
15 It is noted that the solvate 3(C6D6)2 crystallises from
benzene solutions. However, due to extensive twining and
as crstallographically determined Cu–H distances for mZ2-
hydrogen atoms are typically below 2 Å. CSD database Version
5.41, May 2020C. R. Groom, I. J. Bruno, M. P. Lightfoot and
S. C. Ward, Acta Crystallogr., 2016, B72, 171.
the low over-all quality of the crystals a satisfactory structure 22 G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb,
refinement was not obtained. The cell parameters are:
A. Nudelman, B. M. Stoltz, J. E. Bercaw and K. I. Goldberg,
Organometallics, 2010, 29, 2176.
a 11.0712(6) Å, b 26.527(2) Å, c 26.606(2) Å, a 67.203(7)1,
0
%
b 85.754(6)1, g 89.877(3)1, P1, Z = 4, Z = 2.
23 (a) D. Stalke, Chem. Soc. Rev., 1998, 27, 171; (b) CrysAlisPro,
Rigaku Oxford Diffraction, Versions 1.171.37.35, 1.171.40.21a,
2014–2018; (c) G. M. Sheldrick, Acta Crystallogr., 2015, A71, 3;
(d) G. M. Sheldrick, Acta Crystallogr., 2008, A64, 112; (e) G. M.
Sheldrick, Acta Crystallogr., 2015, C71, 3; ( f ) L. J. Farrugia,
J. Appl. Crystallogr., 2012, 45, 849; (g) O. V. Dolomanov,
L. J. Bourhis, R. J. Gildea, J. A. K. Howard and H. Puschmann,
J. Appl. Crystallogr., 2009, 42, 339; (h) A. L. Spek, Acta Crystallogr.,
2015, C71, 9; (i) C. F. Macrae, I. Sovago, S. J. Cottrell,
P. T. A. Galek, P. McCabe, E. Pidcock, M. Platings, G. P.
Shields, J. S. Stevens, M. Towler and P. A. Wood, J. Appl.
Crystallogr., 2020, 53, 226; ( j) Diamond – Crystal and Molecular
Structure Visualization, Crystal Impact - H. Putz and
K. Brandenburg GbR, Bonn, Germany, 2020.
16 W. Clegg, M. R. J. Elsegood, A. J. Scott, T. B. Marder, C. Dai,
N. C. Norman, N. L. Pickett and E. G. Robins, Acta Crystal-
logr., 1999, C55, 733.
17 R. W. G. Wyckoff, Cryst. Struct., 1963, 1, 7.
18 (a) T.-A. D. Nguyen, Z. R. Jones, B. R. Goldsmith, W. R. Buratto,
G. Wu, Z. L. Scott and T. W. Hayton, J. Am. Chem. Soc., 2015,
137, 13319; (b) T.-A. D. Nguyen, Z. R. Jones, D. F. Leto, G. Wu,
S. L. Scott and T. W. Hayton, Chem. Mater., 2016, 28, 8385.
19 (a) Y. Pei, N. Shao, Y. Gao and X. C. Zeng, ACS Nano, 2010,
¨
4, 2009; (b) N. Jian, C. Stapelfeldt, K.-J. Hu, M. Froba and
R. E. Palmer, Nanoscale, 2015, 7, 885.
20 C. R. Larrea, C. J. Baddeley, M. R. Narouz, N. J. Mosey, J. H.
Horton and C. M. Crudden, ChemPhysChem, 2017, 18, 3536.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020