New Experimental Insight into the Nature of Metal?Metal Bonds in Digallium Compounds: J Coupling between Quadrupolar Nuclei

Article abstract of DOI:10.1002/chem.201600999

Multiple bonding between atoms is of ongoing fundamental and applied interest. Here, we report a multinuclear (¹H,¹³C, and⁷¹Ga) solid-state magnetic resonance spectroscopic study of digallium compounds which have been proposed, albeit somewhat controversially, to contain single, double, and triple Ga?Ga bonds. Of particular relevance to the nature of these bonds, we have carried out two-dimensional⁷¹Ga J/D-resolved NMR experiments which provide a direct measurement of J(⁷¹Ga,⁷¹Ga) spin–spin coupling constants across the gallium?gallium bonds. When placed in the context of clear-cut experimental data for analogous singly, doubly, and triply bonded carbon spin pairs or boron spin pairs, the⁷¹Ga NMR data clearly support the notion of a different bonding paradigm in the gallium systems. Our findings are consistent with an increasing role across the purported gallane–gallene–gallyne series for classical and/or slipped π-type bonding orbitals.

Full text of DOI:10.1002/chem.201600999

DOI: 10.1002/chem.201600999
Full Paper
&
NMR Spectroscopy |Hot Paper|
New Experimental Insight into the Nature of MetalꢀMetal Bonds
in Digallium Compounds: J Coupling between Quadrupolar Nuclei
Libor Kobera, Scott A. Southern, Gyandshwar Kumar Rao, Darrin S. Richeson, and
David L. Bryce*[a]
Abstract: Multiple bonding between atoms is of ongoing
fundamental and applied interest. Here, we report a multinu-
clear (1H, 13C, and 71Ga) solid-state magnetic resonance spec-
troscopic study of digallium compounds which have been
proposed, albeit somewhat controversially, to contain single,
double, and triple GaꢀGa bonds. Of particular relevance to
the nature of these bonds, we have carried out two-dimen-
sional 71Ga J/D-resolved NMR experiments which provide
a direct measurement of J(71Ga,71Ga) spin–spin coupling con-
stants across the galliumꢀgallium bonds. When placed in
the context of clear-cut experimental data for analogous
singly, doubly, and triply bonded carbon spin pairs or boron
spin pairs, the 71Ga NMR data clearly support the notion of
a different bonding paradigm in the gallium systems. Our
findings are consistent with an increasing role across the
purported gallane–gallene–gallyne series for classical and/or
slipped p-type bonding orbitals.
exhibiting a double bond.[12] Robinson and co-workers subse-
quently carried out calculations on the more realistic complex
Na2[(C6H5)2C6H3GaGaC6H3(C6H5)2)][13,14] and the claim was made
that the nonbonding orbital reported by Cotton et al.[12] is ac-
tually a bonding orbital, and so the originally reported gallyne
Introduction
The nature of multiple bonding between elements is a topic of
great fundamental and practical interest.[1–6] For example,
a series of recent papers has provided experimental evidence
and a theoretical understanding of the nature of the boronꢁ should indeed be described as containing a weak triple
boron triple bond.[7,8] The excitement around this work is
somewhat reminiscent of that generated by the first report by
Robinson in 1997 of a gallyne,[9] purported to contain a galli-
umꢁgallium triple bond. The description of multiple bonds be-
tween gallium atoms has attracted much attention and gener-
ated some controversy in the scientific community, with vary-
ing conclusions on the nature of GaꢀGa bonding, both in “gal-
lenes” and in “gallynes”. The original gallyne system was stabi-
lized by two sodium ions, and the short GaꢀGa distance of
2.319 ꢀ was a primary argument for the triple bond.[9] Subse-
quently, Klinkhammer proposed, on the basis of a natural
bond order analysis, that the galliumꢀgallium bond in the
model system HGaGaH2ꢀ consists of a s-bond, a p-bond, and
a non-classical “slipped” p-bond,[10] and similar conclusions
were reached by topographical electron localization function
analysis.[11] However, Cotton et al. concluded on the basis of
density functional theory (DFT) that the gallynes involve only
one s-bond orbital and one p-bonding orbital, along with
a nonbonding p orbital, and thus can only be considered as
bond.[14] Robinson has further described theoretical models de-
fending this claim, showing that simple compounds such as
(Me)GaGa(Me) and HGaGaH, along with their reduced species,
include a weak triple bond comprised of two coordinate
bonds and a p-bond.[15,16] An experimental study and analysis
reported by Power and co-workers in 2002 on IArGaGaArI (1),
ArGaGaAr (2), and NaArGaGaArNa (3) [Ar=2,6-Dipp2C6H3,
Dipp=2,6-iPr2C6H3; Scheme 1], however, called into question
the bonding order in the purported gallene (2) and gallyne
(3).[17] A more recent analysis of the electron localization func-
tion concluded that the GaꢁGa bond is a non-classical triple
bond consisting of a regular p-bond, a “slipped” p-bond, and
a distorted s-bond.[18] The sodium cations were found to con-
tribute greatly to the short GaꢀGa distance and to participate
in the p-bond, and it was argued that these compounds
should be viewed as comprising Ga2Na2 clusters.[18] An analysis
based on domain-averaged Fermi holes of (PhꢀGaGaꢀPh)Na2
and PhꢀGaGaꢀPh2ꢀ suggested that of the three electron pairs
contributing to the GaꢁGa bond, only the p-bond can be con-
sidered a classical shared electron pair, while the two other
bonds are complex and have partial lone pair character, but
cannot be considered fully bonding.[19] The previous discussion
of galliumꢀgallium bonding has thus largely focused on X-ray
diffraction analyses coupled with computational chemistry and
theoretical considerations.
[a] Dr. L. Kobera, S. A. Southern, Dr. G. K. Rao, Prof. Dr. D. S. Richeson,
Prof. Dr. D. L. Bryce
Department of Chemistry and Biomolecular Sciences and
Centre for Catalysis Research and Innovation, University of Ottawa
10 Marie Curie Pvt. D’Iorio Hall, Ottawa, Ontario K1N 6N5 (Canada)
E-mail: dbryce@uottawa.ca
Our laboratory has recently reported the development of
a series of J-resolved solid-state NMR spectroscopy tech-
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201600999.
Chem. Eur. J. 2016, 22, 1 – 10
1
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Scheme 1. Compounds studied in this work. Such compounds have been described as featuring single (1), double (2), and triple (3) galliumꢀgallium bonds;
however, the disputed nature of these bonds is acknowledged by the grey shading.
niques.[20] These methods are valuable in that they enable the
Results and Discussion
measurement of indirect nuclear spin-spin (J) coupling con-
1
Initial characterization by H and 13C solid-state NMR spec-
stants for pairs of quadrupolar nuclei (spin quantum number
troscopy
I>1= ) in the solid state. Such measurements are often impossi-
2
ble in solution due to rapid quadrupolar relaxation of many
nuclides of interest. In the solid state, quadrupolar broadening
of the line shapes for nuclei such as 11B, 71Ga, (both I=3/2)
etc., is often orders of magnitude larger than the J coupling
constant, thereby obscuring the effects of the latter in a stan-
dard one-dimensional solid-state NMR (SSNMR) spectrum. Our
two-dimensional J-resolved experiments alleviate these prob-
lems and have enabled the measurement of J(11B,11B) couplings
in solid diboranes, diborenes, and diborynes,[8] as well as
J(71Ga,71Ga) couplings across single GaꢀGa bonds in
Ga2Cl4(dioxane)2, Ga2Cl4(THP)2, and Ga2Cl4(THF)2.[21] The signifi-
cance of the measurements of these J couplings arises due to
their intimate relationship with the electronic structure of the
bond. In the case of multiply bonded boron compounds, we
demonstrated a strong correlation between the reduced
carbonꢀcarbon coupling constants (K=4p2 J/gNgN‘h)[22] for alka-
nes, alkenes, and alkynes with those of diboranes, diborenes,
and diborynes, thereby providing key insight into the nature
of the boronꢁboron triple bond.
Solid-state NMR spectroscopy was used to characterize the di-
gallium compounds. High-resolution 1H DUMBO magic-angle
spinning (MAS) NMR and 13C cross-polarization (CP)/MAS NMR
experiments were used for primary characterization and the re-
sulting spectra are shown in Figure 1. The assignments of the
spectra are based on the literature[17] (Table 1).
Table 1. 1H DUMBO/MAS NMR and 13C CP/MAS NMR chemical shift as-
signments for digallium compounds 1, 2, and 3.
Assignment
1H NMR (ppm)
1
2
3
CH3ꢀ
1.10
2.58
6.92
7.71
1.25
2.72
~7.46
~7.46
0.95
2.65
~7.23
~7.23
>CHꢀ
=CHꢀ (m-)
=CHꢀ (p-)
Assignment
13C NMR (ppm)
1
2
3
CH3ꢀ
23.1; 24.3; 25,8; 27.5
30.2
123.5
24.9
30.4
122.7
24.7
30.3
122.8
The persistent interest and outstanding ambiguity in the de-
scription of GaꢀGa bonding prompted us to study this prob-
lem via NMR spectroscopy. Specifically, how do the reduced
galliumꢀgallium coupling constants for a gallane–gallene–gal-
lyne series compare to the values for the well-understood
carbonꢀcarbon and boronꢀboron analogues, and what insight
does this provide into the nature of galliumꢀgallium bonding?
Here, we have focused our attention on the preparation and
solid-state NMR characterization of a prototypical series of di-
gallium systems which have been proposed to contain single
(1), double (2) and triple (3) bonds (Scheme 1). The gallium
bonding environments in these compounds have been charac-
>CHꢀ
m-Dipp
p-C6H3
i-Dipp
m-C6H3
o-C6H3
p-Dipp
o-Dipp
i-C6H3
130.3
128.1
140.0
146.7
127.8
140.4
146.1
139.3
144.9
147.9
149.1
1
The good resolution of the H DUMBO/MAS NMR spectrum
for 1 allows for clear identification not only of methyl (ꢀCH3)
and methine (>CHꢀ) groups in the aliphatic region of the
spectrum, but also the meta and para protons (Figure 1, left).
The 13C CP/MAS NMR spectrum distinguishes more than
twenty individual peaks, which suggests a high degree of crys-
tallinity of compound 1. Both spectra show some impurity
peaks which may be attributed to unreacted intermediate
product (see Figure 1 and Figure S3 in the Supporting Informa-
1
terized by H, 13C, and 71Ga SSNMR spectroscopic methods. In
particular, advanced 71Ga J-resolved NMR techniques[23,24] have
been applied to overcome issues associated with dipolar and
quadrupolar
spectral
broadening,
and
homonuclear
J(71Ga,71Ga) couplings have been extracted from the NMR spec-
tra to provide new insights into the nature of galliumꢀgallium
bonds.
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
2
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
1
Figure 1. Experimental H DUMBO/MAS NMR spectra collected at 10 kHz MAS (left panel) and 13C CP/MAS NMR spectra acquired at 8 kHz MAS (right panel) of
digallium compounds 1, 2, and 3. B0 =9.4 T. Asterisks denote impurities. A regular 4 mm ZrO2 rotor was used. Spinning sidebands are indicated by “ssb”.
tion). Some splittings on the order of 100 Hz observed in the
13C CP/MAS NMR spectra may also be attributed to through-
space residual dipolar coupling (RDC) between carbon atoms
and the quadrupolar 127I nuclei (I=5/2).[25] (Supporting Informa-
tion, Figure S3).
plied magnetic field must be used, because both isotopes typi-
cally give rise to relatively broad spectral lines, often covering
hundreds of ppm.[21,27] For nuclei with large quadrupolar inter-
actions, the static quadrupolar Carr–Purcell–Meiboom–Gill
(QCPMG) experiment, or a modification thereof (wideband uni-
form-rate smooth truncation (WURST)-QCPMG) are often ad-
vantageous.[28–30] For compounds 1 to 3, it was found that fast
T2(71Ga) relaxation rendered these experiments impractical. In-
stead, regular 71Ga spin-echo NMR experiments were carried
out at 21.1 T (Figure 2). The line shapes are largely due to the
71Ga nuclear electric quadrupolar interaction between the galli-
um quadrupole moment and the electric field gradient (EFG).
In principle, the spectra are also influenced by gallium chemi-
cal shift anisotropy (CSA); however, the impact of CSA is ex-
pected to be small relative to the dominant quadrupolar cou-
pling and, as such, the spectra were fit only on the basis of the
The similarity between the 1H DUMBO/MAS NMR spectra,
and between the 13C CP/MAS NMR spectra, of the digallium
compounds 2 and 3 indicates a comparable degree of overall
crystallinity. Additionally, a decreased presence of impurities in
2 and 3 is noted. In the 13C CP/MAS NMR spectrum of 2,
a small signal at 69 ppm is attributed to residual diethyl ether
1
solvent.[26] An impurity at 0.1 ppm is clearly visible in the H
DUMBO/MAS NMR spectrum of system 3. This impurity can be
attributed to the silicone grease used to join reaction flasks
during sample preparation.[26]
While these spectra provide useful characterization informa-
tion, they do not provide direct information on the galliumꢀ quadrupolar interaction and an isotropic chemical shift. The re-
gallium bond. Therefore, the systems were further investigated
by 71Ga NMR spectroscopy.
sulting NMR parameters are listed in Table 2. Errors were deter-
mined on an iterative heuristic basis, but then increased to ac-
count for our neglect of small CSA effects.
Typical gallium isotropic chemical shifts range from ꢀ700 to
700 ppm with respect to aqueous Ga2(NO3)3, reflective of the
oxidation state and coordination geometry.[27] Chemical shifts
for GaI range from ꢀ700 to 0 ppm, those for GaII from 200 to
250 ppm, and from 0 to 700 ppm for GaIII.[21,27] Additionally,
a strong correlation between 27Al NMR and 71Ga NMR chemical
shifts of gallium and aluminum oxides, silicates or phosphates
Gallium-71 NMR spectroscopy
Gallium has two NMR-active isotopes, 69Ga and 71Ga (I=3/2),
with natural abundances of 60.1 and 39.9%, respectively. The
less-abundant 71Ga isotope is observed presently due to its
smaller quadrupole moment, which contributes to its higher
spectral receptivity relative to 69Ga.[27] Regardless, a strong ap-
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
3
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 2. 71Ga spin-echo spectra of static powdered digallium compounds 1, 2, and 3. Experimental data, solid lines; simulated data, dashed lines. The asterisk
denotes an impurity resonance in 1.
gest a planar, trans-bent (CꢀGaꢀGaꢀC) core geometry with an
Table 2. 71Ga NMR parameters and structural information for the GaꢀGa
oxidation state of +1. The chemical shift values and large
spin pairs in digallium compounds 1, 2, and 3.[a]
values of CQ and h can be explained by structural asymmetry
along the GaꢀGa bond.
Cmpd[b] Site diso
[ppm]
CQ
[MHz]
h
rGaGa Spin
[ꢀ] system group number
Space CCDC
The aforementioned 71Ga NMR spectra provide useful infor-
mation about the oxidation states and the coordination envi-
ronments of each of the gallium sites, but they do not provide
direct proof of GaꢀGa bonding nor the bond order. Direct ex-
amination of the GaꢀGa bond was carried out using shifted-
echo two-dimensional J/D-resolved NMR experiments, which
allow for deep insight into the nature of GaꢀGa bonds
(Figure 3). These experiments are quite challenging on these
systems due to the dominant quadrupolar broadening of the
spectra. In addition, fast T2 relaxation led to a complete loss of
the Ga signal when using a double-quantum filtered version of
the experiment. Unfortunately, the unfiltered spectra are affect-
ed by a dominant peak at zero frequency. This peak is due in
part to spin states which are not affected by J-modulation
during the evolution of the pulse sequence (see references
[21] and [23]). Despite this, a simple two-pulse experiment pro-
vided sufficient sensitivity and resolution (Figure 3).
1
Ga(1) 200ꢂ50 21.2ꢂ1 0.60ꢂ0.1 2.493 AX
Ga(2) 180ꢂ50 20.5ꢂ1 0.52ꢂ0.1 2.493 AX
P21/c 186154
¯
P1
2
3
Ga
Ga
66ꢂ50 19.0ꢂ1 0.74ꢂ0.1 2.627 A2
95ꢂ50 19.3ꢂ1 0.72ꢂ0.1 2.347 A2
186153
P21/n 186155
[a] CQ =eV33Q/h and h=(V11ꢀV22)/V33, where Q is the nuclear electric quad-
rupole moment, and V11, V22, and V33 are the principal components of the
electric field gradient tensor. [b] NMR parameters of the impurity (unreact-
ed intermediate product) in 1: diso =35(ꢂ20) ppm; CQ =3.6(ꢂ0.5) MHz; h<
0.1.
can be seen.[31,32] Generally, 27Al chemical shifts correspond
with AlꢀO environments: for 4-coordinate Al the chemical shift
ranges from 50 to 80 ppm, the chemical shift range of 5-coor-
dinated Al is about 30 to 40 ppm, and 6-coordinated Al shifts
range from about ꢀ10 to 15 ppm.[33] Analogously, the 71Ga
chemical shifts of 4-coordinate Ga sites fall within the range of
90 to 120 ppm, 5-coordinate Ga sites resonate between 0 ppm
and 30 ppm, and 6-coordinate Ga sites fall between ꢀ60 and
10 ppm.[27,31,32] The Ga coordination also has a direct influence
on the quadrupolar interaction and this trend is observed with
71Ga in a wide range of inorganic compounds. The 71Ga quad-
rupolar coupling constants (CQ) of various gallium oxides in
tetra-, penta-, and hexagonal coordination environments range
from a few MHz up to 17 MHz.[34] However, in our case, as well
as in previous work,[21] CQ reaches values exceeding 19 MHz,
which is indicative of the increasing magnitude of the EFG,
consistent with asymmetric gallium environments in the digal-
lium systems. These compounds also exhibit unexpected
chemical shift values which fall outside of the traditional re-
gions for these coordination environments.[27,31–33]
In such experiments, we have shown that the J-splitting
magnitude and shape of each doublet component depends in-
timately on whether or not the quadrupolar spins are related
by an inversion center.[21] The X-ray crystal structures show that
the directly bonded gallium spin pairs are related by an inver-
sion center in compounds 2 and 3 (resulting in A2 spin sys-
tems), but that they are crystallographically and magnetically
non-equivalent (AX spin system) in gallane 1 (Table 2).[17]
Armed with this knowledge, the doublets of the 71Ga J/
D-resolved spectra may be fitted to extract both the isotrop-
ic J coupling constant and the effective dipolar coupling con-
stant (Reff =RDDꢀDJ/3). The experimental anisotropic J coupling
(DJ) can be determined, if the direct dipolar coupling constant,
RDD (equal to (m0g1g2ꢀh<r12ꢀ3 >/8p2), is known.[21] The effect of
Reff on the spectra shown in Figure 3 is manifested in the
breadth and sense of the powder patterns associated with
each component of the doublet.[21] The shape of each doublet
The similar chemical shifts and the quadrupolar parameters
for both crystallographically distinct gallium sites of compound
1 (Table 2) suggest the existence of an asymmetric planar (Cꢀ component is an anisotropic powder pattern with a breadth of
GaꢀGaꢀC) core with a formal GaII oxidation state.[21,35] The NMR
3Reff/2 in the case of the AX spin system (compound 1), and
9Reff/2 in the case of A2 spin systems (compounds 2 and 3).[21]
Additionally, the sense of these powder patterns is opposite
for the A2 case compared to the AX case. The value of
J(71Ga,71Ga) is determined by measuring the difference be-
parameters of the impurity (CQ =3.6 MHz, hQ =0 and diso
=
35 ppm) are indicative of the presence of a mononuclear GaIII
site in a pseudo-octahedral coordination environment.[36] The
chemical shift and relatively large CQ of systems 2 and 3 sug-
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
4
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Figure 3. Experimental static 71Ga J/D-resolved NMR (upper) and simulated spectra (red dashed line) of compounds 1, 2, and 3 measured at 21.1 T, with deter-
mined J or 3J coupling values (left). On the right are experimental 71Ga J/D-resolved NMR (upper) and simulated (bottom) spectra after subtracting simulation
of the dominant zero-frequency signal.
tween the centers of gravity of each of the two doublet com-
ponents. In the case of an A2 spin system (where I=3/2), this
splitting is equal to three times J(71Ga,71Ga); in the AX case, the
splitting is directly equal to J(71Ga,71Ga).[20,21,23]
ing only monovalent (non-coupled) gallium atoms (Supporting
Information, Figure S4); this spectrum shows only a broad
zero-frequency peak, with no evidence of doublet structure, as
anticipated.
Spectra were fit using two independent methods. First, the
raw data sets containing both the zero-frequency peak and
the lower-intensity doublets of interest were fit. Second, the
zero-frequency peak was removed from the frequency domain
through a simple subtraction of a Gaussian peak of appropri-
ate breadth. The resulting doublet spectra were of very good
quality for compounds 1 and 2 (Figure 3) and these doublets
were also fit directly. The breadth of the zero-frequency peak
and low intensity of the doublet peaks for compound 3 pre-
cluded the application of the subtraction approach. Neverthe-
less, the full spectrum for 3 could be fit directly, and special
care was taken to ensure the correctness of the fit (Supporting
Information, Figure S5).
The most well-defined powder patterns are obtained for the
doublet components of 2 (perhaps in part due to the absence
of nearby 127I or 23Na spins which are present for 1 and 3, re-
spectively), and their sense is consistent with an A2 spin
system (see right-hand panel of Figure 3). The spectrum of
1 was fit using powder patterns of opposite sense to those for
2 and 3; however, their shape is not well-resolved due to spec-
tral broadening.
The experimental J(71Ga,71Ga) coupling constants are listed
in Table 3 and do not show a clear trend with respect to the
proposed bond orders for these compounds. This behavior
suggests non-classical bonding character (non-traditional mul-
tiple bonds) in these systems. The influence of the bond order
on the J coupling constants can be assessed using direct com-
parisons with experimental data for different spin pairs such as
Several precautions were taken in the fitting of the spectra
in Figure 3 to account for the impact of direct dipolar cou-
pling; these have been described previously.[21] Briefly, we
made the assumption that the effective dipolar coupling con-
stant is relatively insensitive to spectral offset. This is a valid as-
sumption given the moderate overall breadth of the 71Ga
powder patterns and the high values of the quadrupolar asym-
metry parameter (h; far from axial symmetry), as explained in
reference [21]. To alleviate concerns about the weakness of the
observed doublet peaks as well as their partial overlap with
the zero-frequency peak, we acquired the corresponding shift-
ed-echo 2D J/D-resolved NMR spectrum for a system contain-
13
11
13Cꢀ C or 11Bꢀ B in systems where the nature of the bonding
is well-established and accepted.[8] We have converted the ex-
perimental J coupling constants to reduced coupling constants
(K), which removes the dependence of the values on nuclide-
specific magnetogyric ratios, thereby providing a more suitable
parameter for the comparison of differences or similarities in
electronic structure between different spin pairs. Several points
are worthy of discussion. First, the K(71Ga,71Ga) constants are
overall roughly an order of magnitude larger than the K(13C,
13C) or K(11B, 11B) values, which is in agreement with established
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
5
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
trends for heavier elements.[22] Shown in Figure 4b is a plot of
the experimental K(71Ga,71Ga) values for 1, 2, and 3 versus the
experimental K(13C, 13C) values for ethane, ethene, and ethyne
and also versus the K(11B, 11B) values for tetramethyldiborane,
a diborene stabilized by two N-heterocyclic carbenes (NHC),
and an NHC-stabilized diboryne.[8] While there is an excellent
linear correlation between the K(13C, 13C) and K(11B, 11B) values
(Figure 4a),[8] no such clear correlation exists between the
K(71Ga,71Ga) and the K(13C, 13C) or K(11B, 11B) values, providing
a strong indication of a different bonding mechanism in the
series of gallium compounds.
Table 3. Experimental 71Ga,71Ga spin–spin coupling values for digallium
systems.
[a]
[b]
[b]
Cmpd
RDD
Reff
jJiso
j
DJ[c]
[kHz]
[Hz]
[Hz]
[Hz]
1
2
3
728
618
866
465ꢂ50
168ꢂ50
68ꢂ50
1415ꢂ50
849ꢂ50
1268ꢂ50
0.79ꢂ0.15
1.35ꢂ0.15
2.40ꢂ0.15
[a] Direct dipolar coupling constant calculated from the experimental gal-
lium–gallium distances determined by X-ray diffraction.[17] [b] Obtained
from spectral fitting. [c] Determination of experimental DJ values was car-
ried out according to Reff =RDD
DJ/3, where RDD was calculated from the X-
ꢀ
ray crystal structures. DJ=J33ꢀ(J11 +J22)/2.
Hardman’s crystallographic measurements[17] indicate that
the bond distances increase in the order 3<1<2 (Table 2);
the measured J(71Ga,71Ga) values do not show a clear positive
or negative correlation with these distances. This is in contrast
to the established dependence of J(11B,11B) on the bond length
within a series of s-bonded diboranes,[23] thereby discrediting
the possibility that compounds 2 and 3 are held together only
by longer or shorter s-bonds. All of the measured J(71Ga,71Ga)
coupling constants are an order of magnitude smaller than
those observed previously for the single GaꢀGa bonds in gal-
lium(II) halides stabilized by organoligands.[21] This diversity can
be induced by mixed oxidation states of gallium (I, III) chlor-
ides combined with different steric shielding[15] in comparison
with the presently investigated systems.
Figure 4. a) Correlation between reduced indirect nuclear spin–spin coupling
constants (K) for B,B spin pairs in B2Me4, a NHC-stabilized diborene, and a di-
boryne, with C,C spin pairs in ethane, ethane, and ethyne, respectively (see
reference [8] for further details). b) Correlation between K for Ga,Ga versus
~
C,C spin pairs (* and solid line) and between Ga,Ga and B,B spin pairs (
and dotted line) for 1 (XꢀX), 2 (X=X), and 3 (XꢁX) compounds. The lack of
a positive correlation between the data sets in (b) indicates a different elec-
tronic structure for the digallium compounds in comparison with the classi-
cal s/p model.
gands.[21] Nevertheless, the experimentally observed trend is
indeed reproduced, that is, jJ(71Ga,71Ga)j increases in the order
2<3<1. Furthermore, the calculations successfully reproduce
the experimentally observed order of magnitude difference in
the values of J(71Ga,71Ga) for the presently studied compounds
(ꢃ1 kHz) compared to those reported for the single GaꢀGa
bonds in gallium(II) chlorides stabilized by organoligands
Given the extensive previous theoretical and computational
studies on the nature of multiple galliumꢀgallium bonds, we
prefer not to rely overly on computation in the present work.
Nevertheless, in order to understand the nature of galliumꢀ (ꢃ10 kHz).[21]
gallium bonds in these digallium systems in more detail, and
to corroborate the experimentally measured J(71Ga,71Ga)
values, a series of quantum chemical calculations using the
B3LYP functional was performed (Supporting Information,
Table S1). This method was chosen based on the quality of
previously obtained results.[21] For compound 1, the electronic
structure of which is the least controversial, excellent agree-
ment is obtained between experimental and computed values
(1415(ꢂ50) Hz vs. 1472 Hz). For systems 2 and 3, discrepancies
between the experimental and calculated values of a few hun-
dred Hz are noted; this level of agreement is consistent with
a previous report on gallium(II) chlorides stabilized by organoli-
The experimental trend in DJ(71Ga,71Ga), that is, 1<2<3, is
not in concordance with the known experimental trend for
DJ(13C,13C) for the ethane, ethene, and ethyne series, where the
smallest value is noted for ethene and the largest for
ethyne.[37] This provides further clear evidence for a different
electronic structure in the purported gallane–gallene–gallyne
systems relative to the prototypical carbon systems. Generally,
the anisotropies of the spin–spin coupling tensors report on
local symmetry and on the electronic structures of the bonds
between the coupled nuclei.[37–39] The J coupling involves the
Fermi-contact (FC), spin-orbital (SO), and spin-dipolar (SD) con-
tributions, as well as a FCꢂSD cross-term which only contrib-
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
6
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
utes to the anisotropy. The FC interaction is often generated
due to covalent bonding between nuclei, mediated by over-
lapping orbitals with s character. The SO contribution reflects
the interaction between the magnetic moments of the nuclei
and the orbital angular momentum of the surrounding elec-
trons, and may be subdivided into diamagnetic (DSO) and par-
amagnetic (PSO) parts. A coupling between the nuclear spins
and the electronic spins is defined as the SD mechanism.[37,38]
In the case of the galliumꢀgallium bonds, all of our NMR
data are consistent with a difference in the electronic structure
across the series of compounds 1, 2, and 3, but are clearly in-
consistent with a classical s/p-bonding model across this
series. The large values of jJ(71Ga,71Ga)j for 1 and 3, which
have the shortest galliumꢀgallium distances, are consistent
with a dominant contribution from the FC mechanism and a s-
bond; however, as mentioned, there is no clear correlation be-
tween bond length and jJ(71Ga,71Ga)j across the full series of
three compounds, suggesting that some non-FC mechanisms
also play a nontrivial role. This is supported by our DFT calcula-
tions, and is in contrast to the case of carbonꢀcarbon bonding
where the FC mechanism is overwhelmingly dominant.[37]
Indeed, our calculations show that the contribution from the
PSO term increases substantially in the order 1<2<3 (see the
Supporting Information).
order 2<3<1 and this was reproduced by hybrid DFT calcula-
tions. The experimental values of DJ increase in the order 1<
2<3. Both of these trends are in disagreement with the
known trends for coupling constants in singly, doubly, and
triply bonded carbon and boron systems. The jJ(71Ga,71Ga)j
values also do not correlate simply with the galliumꢀgallium
distances. Taken together, these results uphold the somewhat
enigmatic reputation that multiple galliumꢀgallium bonds
have garnered in the literature. While the values of coupling
constants in isolation cannot be translated directly into bond
orders, the following conclusions are reached: 1) there is signif-
icant covalent bonding between pairs of gallium atoms in all
three samples; 2) the measured values of jJ(71Ga,71Ga)j and DJ
show that the electronic structure across the 1, 2, 3 series is
clearly different from that for related carbonꢀcarbon and
boronꢀboron series; 3) the increase in DJ in the order 1<2<
3, and lack of a straightforward correlation between galliumꢀ
gallium bond length and J(71Ga,71Ga), is consistent with an in-
creased role across the series for classical and/or slipped p-
type bonding orbitals, perhaps of the type described by Klink-
hammer.[10]
Experimental Section
The experimental trend observed for DJ clearly indicates
a significant contribution from non-FC terms since the FC term
is purely isotropic. It is well known that couplings between
heavy elements often feature non-FC contributions from the
SO and SD mechanisms.[22,40] As the paramagnetic SO term is
known to be important in bonding situations which do not in-
volve s-type orbitals,[41] the experimentally observed increase
in DJ is consistent with an increased contribution from p-
bonding orbitals in the order 1<2<3.
The participation of sodium anions in the gallium-gallium
triple bond[18] evokes questions about their arrangement and
mobility, because this information may provide an improved
understanding of the nature of this multiple bond. The experi-
mental 23Na MAS NMR spectrum of 3 is depicted in Figure S6
in the Supporting Information. Interestingly, the narrow and
relatively symmetric resonance centered at ꢀ3.25 ppm may
suggest that the Na2Ga2 cluster comprises relatively mobile
Na+ ions.
Sample preparation
GaI was prepared according to a literature procedure.[36] GaI
(0.784 g, 4 mmol) in toluene (10 mL) was cooled to ꢀ788C using
a dry-ice acetone bath. (LiAr)2 (Ar=2,6-Dipp2C6H3, Dipp=2,6-
iPr2C6H3) was prepared according to a literature procedure.[42] A so-
lution of (LiAr)2 (1.616 g, 2 mmol) in toluene (20 mL) was added
dropwise over a period of 1 h. The mixture was allowed to warm
to room temperature over a period of 6 h. The dark green solution
was filtered and concentrated to about 10 mL to give a light
yellow precipitate. This solution was kept at room temperature for
an additional hour and then the precipitate was isolated by de-
cantation. This was then washed with 2 mL hexane and dried to
give compound 1 (IArGaGaArI). The dark green liquid was dried to
give a red crystalline solid. This was recrystallized from 2 mL of
warm hexane to give 2 (ArGaGaAr). A solution of 2 (0.268 g,
0.5 mmol) in diethyl ether (10 mL) was added to sodium metal
(0.161 g, 7 mmol) and stirred for 4 h at room temperature. The
dark purple solution was filtered to separate the excess sodium.
This was concentrated, and cooled in a freezer at ꢀ208C for one
week to afford a dark red solid (3; NaArGaGaArNa). These com-
pounds have been synthesized according to literature proce-
dures.[17] The acquired H NMR spectra correspond with the litera-
1
Conclusions
ture data for 1, 2, and 3[17] (Supporting information, Figure S1). The
prepared samples were stored under inert atmosphere in a glove-
box at 277 K.
In this contribution, we have demonstrated that two-dimen-
sional J/D-resolved NMR experiments are powerful tools to
measure spin–spin coupling interactions between quadrupolar
metal nuclei which give rise to broad spectral lines, including
those with fast relaxation times. These experiments provide
a new experimental handle on the nature of galliumꢀgallium
bonds. In a prototypical series of digallium compounds, iso-
tropic and anisotropic parts of the J(71Ga, 71Ga) coupling tensor
were measured and interpreted in the context of analogous
13C and 11B NMR data for singly and multiply bonded species
featuring carbonꢀcarbon or boronꢀboron bonds. The experi-
mental values of jJ(71Ga,71Ga)j were found to increase in the
Solid-state NMR spectroscopy
Dry powdered samples of 1, 2, and 3 were packed into individual
ZrO2 rotors. Solid-state NMR spectra were recorded at 9.4 and 21.1
T using Bruker AVANCE III and Bruker AVANCE II spectrometers, re-
spectively. A 4 mm cross-polarization (CP)/magic angle spinning
(MAS) probe was used for H and 13C experiments at Larmor fre-
1
quencies of n(1H)=400.130 MHz and n(13C)=100.613 MHz, respec-
tively. The 13C CP/MAS NMR spectra were collected at 8 kHz and
the 1H DUMBO/MAS NMR spectra were collected at 10 kHz spin-
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
7
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
1
ning speed. The H MAS NMR and 13C MAS NMR isotropic chemical
[1] U. Ruschewitz, Angew. Chem. Int. Ed. 2016, 55, 3264–3266; Angew.
Chem. 2016, 128, 3320–3322.
shifts were calibrated with adamantane (1H: 1.85 ppm; central
signal) and glycine (13C: 176.03 ppm; carbonyl signal) as external
standards. A H DUMBO/MAS NMR set-up experiment on glycine is
described and depicted in the Supporting Information, Figure S2.
[2] Y. Wang, G. H. Robinson, Chem. Commun. 2009, 5201–5213.
[3] P. P. Power, J. Chem. Soc. Dalton Trans. 1998, 2939–2951.
[4] a) S. Shaik, D. Danovich, W. Wu, P. Su, H. S. Rzepa, P. C. Hiberty, Nat.
Chem. 2012, 4, 195–200; b) S. Shaik, H. S. Rzepa, R. Hoffmann, Angew.
Chem. Int. Ed. 2013, 52, 3020–3033; Angew. Chem. 2013, 125, 3094–
3109; c) G. Frenking, M. Hermann, Angew. Chem. Int. Ed. 2013, 52,
5922–5925; Angew. Chem. 2013, 125, 6036–6039; d) D. Danovich, S.
Shaik, H. S. Rzepa, R. Hoffmann, Angew. Chem. Int. Ed. 2013, 52, 5926–
5928; Angew. Chem. 2013, 125, 6040–6042.
1
For 23Na, a 4 mm CP/MAS probe was used at a Larmor frequency
of n(23Na)=105.842 MHz. A MAS rate of 11 kHz was employed. The
23Na isotropic chemical shift was calibrated with NaCl(s) as an ex-
ternal standard (23Na: 7.2 ppm). The /2 pulse length was 2.1 ms and
p
the recycle delay was 5 s. 71Ga NMR spectra were collected on sta-
[5] D. Danovich, P. C. Hiberty, W. Wu, H. S. Rzepa, S. Shaik, Chem. Eur. J.
2014, 20, 6220–6232.
[6] Multiple Bonds Between Metal Atoms (Eds.: F. A. Cotton, C. A. Murillo,
R. A. Walton), Springer, 2005.
tionary samples using
a 4 mm CP/MAS probe operating at
a Larmor frequency of n(71Ga)=274.510 MHz (B0 =21.1 T). The ex-
p
ternal standard Ga2(SO4)3(aq.) was used to determine the /2 central
transition (CT)-selective pulse as well as for chemical shift calibra-
tion (71Ga: 0.0 ppm).[43] The 2D shifted echo two-pulse 71Ga J-re-
solved NMR techniques were developed in our laboratory and eval-
uated in previous work.[21] For the current work, this two-pulse ex-
periment (908CT–(t1/2)–1808CT–(t1/2)–acq.) was carried out under
static conditions at B0 =21.1 T. The 908CT selective pulse length was
1.2 ms and the recycle delay was 0.5 s. The t1 evolution period con-
sisted of 32 to 48 increments each consisting of 2048 or 3072
scans, depending on the sample. All NMR experiments were per-
formed at room temperature.
71Ga NMR spectra of stationary samples were simulated using the
“QUAD central” model in the line shape analysis feature of Bruker’s
Topspin 3.5. CSA was not included in the fits. The simulated 71Ga J/
D resolved spectra were prepared using a superposition of line
shapes generated using the “CSA” model in the same software. A
manual analysis of the doublet breadths and positions was also
used to confirm the parameters obtained through the simulations.
[7] H. Braunschweig, R. D. Dewhurst, K. Hammond, J. Mies, K. Radacki, A.
Vargas, Science 2012, 336, 1420–1422.
[8] F. A. Perras, W. C. Ewing, T. Dellermann, J. Bçhnke, S. Ullrich, T. Schꢃfer,
H. Braunschweig, D. L. Bryce, Chem. Sci. 2015, 6, 3378–3382.
[9] J. Su, X.-W. Li, R. C. Crittendon, G. H. Robinson, J. Am. Chem. Soc. 1997,
119, 5471–5472.
[10] K. W. Klinkhammer, Angew. Chem. Int. Ed. Engl. 1997, 36, 2320–2322;
Angew. Chem. 1997, 109, 2414–2416.
[11] H. Grꢄtzmacher, T. F. Fꢃssler, Chem. Eur. J. 2000, 6, 2317–2325.
[12] F. A. Cotton, A. H. Cowley, X. Feng, J. Am. Chem. Soc. 1998, 120, 1795–
1799.
[13] Y. Xie, R. S. Grev, J. Gu, H. F. Schaefer III, P. v. R. Schleyer, J. Su, X.-W. Li,
G. H. Robinson, J. Am. Chem. Soc. 1998, 120, 3773–3780.
[14] Y. M. Xie, H. F. Schaefer, G. H. Robinson, Chem. Phys. Lett. 2000, 317,
174–180.
[15] G. H. Robinson, Acc. Chem. Res. 1999, 32, 773–782.
[16] G. H. Robinson, Chem. Commun. 2000, 2175–2181.
[17] N. J. Hardman, R. J. Wright, A. D. Phillips, P. P. Power, Angew. Chem. Int.
Ed. 2002, 41, 2842–2844; Angew. Chem. 2002, 114, 2966–2968.
[18] J. Sun, L. Meng, S. Zheng, Z. Sun, X. Li, Science China Chemistry 2012,
55, 1370–1376.
Quantum chemical calculations
[19] R. Ponec, G. Yuzhakov, X. Gironꢅs, G. Frenking, Organometallics 2004,
23, 1790–1796.
[20] F. A. Perras, D. L. Bryce, eMagRes 2015, 4, 561–574.
[21] F. A. Perras, D. L. Bryce, J. Phys. Chem. Lett. 2014, 5, 4049–4054.
[22] D. L. Bryce, R. E. Wasylishen, J. Autschbach, T. Ziegler, J. Am. Chem. Soc.
2002, 124, 4894–4900.
[23] F. A. Perras, D. L. Bryce, J. Am. Chem. Soc. 2013, 135, 12596–12599.
[24] F. A. Perras, D. L. Bryce, Chem. Sci. 2014, 5, 2428–2437.
[25] V. V. Terskikh, S. J. Lang, P. G. Gordon, G. D. Enright, J. A. Ripmeester,
Magn. Reson. Chem. 2009, 47, 398–406.
[26] H. E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 1997, 62, 7512–
7515.
DFT calculations were conducted on models of the three gallium
systems. The Cartesian coordinates of the atoms in the models
were obtained from the published crystal structures without fur-
ther optimization (CCDC numbers 186154 (1), 186153 (2), and
186155 (3)).[17] To reduce computational requirements, the isopro-
pyl groups were removed and the resulting terminal CꢀC bonds
were replaced with CꢀH bonds. NMR parameters for each structure
were calculated using Gaussian version 2009, rev. D.01.[44] The
B3LYP functional was used with custom TZP basis sets on all
atoms.[45–48]
[27] R. E. Wasylishen, S. E. Ashbrook, S. Wimperis, NMR of Quadrupolar Nuclei
in Solid Materials, Wiley, 2012.
[28] J. A. Tang, L. A. O’Dell, P. M. Aguiar, B. E. G. Lucier, D. Sakellariou, R. W.
Schurko, Chem. Phys. Lett. 2008, 466, 227–234.
Acknowledgements
Acknowledgement is made to the donors of The American
Chemical Society Petroleum Research Fund for partial support
of this research. D.L.B. thanks the Natural Sciences and Engi-
neering Research Council of Canada for funding. We are grate-
ful to Dr. F. A. Perras for valuable discussions. Access to the
21.1 T NMR spectrometer was provided by the National Ultra-
high-Field NMR Facility for Solids (Ottawa, Canada), a national
research facility funded by a consortium of Canadian Universi-
ties, supported by the National Research Council Canada and
Bruker BioSpin, and managed by the University of Ottawa
(http://nmr900.ca).
[29] L. A. O’Dell, A. J. Rossini, R. W. Schurko, Chem. Phys. Lett. 2009, 468,
330–335.
[30] A. W. MacGregor, L. A. O’Dell, R. W. Schurko, J. Magn. Reson. 2011, 208,
103–113.
[31] S. M. Bradley, R. F. Howe, R. A. Kydd, Magn. Reson. Chem. 1993, 31, 883–
886.
[32] J. T. Ash, P. J. Grandinetti, Magn. Reson. Chem. 2006, 44, 823–831.
[33] K. J. D. MacKenzie, M. E. Smith, Multinuclear Solid-State Nuclear Magnetic
Resonance of Inorganic Materials, Pergamon, Amsterdam, 2002.
[34] D. Massiot, T. Vosegaard, N. Magneron, D. Trumeau, V. Montouillout, P.
Berthet, T. Loiseau, B. Bujoli, Solid State Nucl. Magn. Reson. 1999, 15,
159–169.
ˇ
ˇ
ˇ
ˇ ˇ
[35] Z. Cerny, J. Machꢆcek, J. Fusek, O. Kriz, B. Cꢆsensky, D. G. Tuck, J. Orga-
nomet. Chem. 1993, 456, 25–30.
[36] C. M. Widdifield, T. Jurca, D. S. Richeson, D. L. Bryce, Polyhedron 2012,
35, 96–100.
Keywords: gallium · gallyne · J coupling · multiple bonding ·
[37] J. Kaski, P. Lantto, J. Vaara, J. Jokisaari, J. Am. Chem. Soc. 1998, 120,
3993–4005.
NMR spectroscopy
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
8
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[38] W. P. Power, M. D. Lumsden, R. E. Wasylishen, J. Am. Chem. Soc. 1991,
113, 8257–8262.
[39] M. Kaupp, O. L. Malkina, V. G. Malkin, P. Pyykkç, Chem. Eur. J. 1998, 4,
118–126.
[40] D. L. Bryce, R. E. Wasylishen, Inorg. Chem. 2002, 41, 3091–3101.
[41] J. Vaara, J. Jokisaari, R. E. Wasylishen, D. L. Bryce, Prog. Nucl. Magn.
Reson. Spectrosc. 2002, 41, 233–304.
bayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyen-
gar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B.
Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg,
S. Dapprich, A. D. Daniels, ꢇ. Farkas, J. B. Foresman, J. V. Ortiz, J. Cio-
slowski, D. J. Fox, Gaussian Inc., Wallingford, CT, 2009.
[42] B. Schiemenz, P. P. Power, Angew. Chem. Int. Ed. Engl. 1996, 35, 2150–
2152; Angew. Chem. 1996, 108, 2288–2290.
[43] S. M. Bradley, R. A. Kydd, R. Yamdagni, J. Chem. Soc. Dalton Trans. 1990,
413–417.
[45] P. L. Barbieri, P. A. Fantin, F. E. Jorge, Mol. Phys. 2006, 104, 2945–2954.
[46] S. F. Machado, G. G. Camiletti, A. Canal Neto, F. E. Jorge, R. S. Jorge, Mol.
Phys. 2009, 107, 1713–1727.
[47] C. T. Campos, F. E. Jorge, Mol. Phys. 2013, 111, 167–173.
[44] Gaussian 09, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Ko-
[48] a) D. Feller, J. Comput. Chem. 1996, 17, 1571–1586; b) K. L. Schuchardt,
B. T. Didier, T. Elsethagen, L. Sun, V. Gurumoorthi, J. Chase, J. Li, T. L.
Windus, J. Chem. Inf. Model. 2007, 47, 1045–1052.
Received: March 2, 2016
Published online on && &&, 0000
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
9
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
Must be going GaꢀGa: The nature of
galliumꢀgallium bonding is assessed by
measuring J(71Ga,71Ga) coupling con-
stants by solid-state NMR spectroscopy.
These experimental data demonstrate
that while Ga–Ga systems may feature
multiple bonds, their electronic struc-
ture clearly differs from that established
for analogous carbonꢀcarbon or
&
NMR Spectroscopy
L. Kobera, S. A. Southern, G. K. Rao,
D. S. Richeson, D. L. Bryce*
&& – &&
New Experimental Insight into the
Nature of MetalꢀMetal Bonds in
Digallium Compounds: J Coupling
between Quadrupolar Nuclei
boronꢀboron bonds.
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
10
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!