Low-valent Ge2 and Ge4 species trapped by N-Heterocyclic gallylene

Article abstract of DOI:10.1002/anie.201204440

Much π and no ω: Quantum chemical calculations showed that the Ge atoms of the Ga₂Ge₂ core in Ge₂[Ga(DPP)] ₂ are not bonded by ω interactions, but rather by a transannular π interaction (see picture). The compound is formed by reduction of (PCy₃)·GeCl₂ with Ga(DDP)/KC₈ which also yielded a further product Ge₄[Ga-(DPP)]₂ with a Ge₂ tetrahedron (DDP=HC(CMeNC₆H₃-2,6-iPr ₂)₂).

Full text of DOI:10.1002/anie.201204440

.
Angewandte
Communications
DOI: 10.1002/anie.201204440
Metalloid Clusters
Low-Valent Ge2 and Ge4 Species Trapped by N-Heterocyclic
Gallylene**
Adinarayana Doddi, Christian Gemel, Manuela Winter, Roland A Fischer,*
Catharina Goedecke, Henry S. Rzepa, and Gernot Frenking*
Much progress has been made in the field of low-valent main
(NHC = :C{N(2,6-iPr2C6H3)CH}2) with Ga(DDP) in fluoro-
=
group 13–15 compounds. The disilicon molecule LSi SiL (L =
benzene (258C, 8d), respectively (see Scheme S1 in the
Supporting Information). In case of 2, the anticipated by-
product Cl2Ga(DDP) was observed by 1H NMR spectroscopy
as well as single-crystal X-ray crystallography. Liquid injec-
tion field desorption ionization mass spectrometry (LIFDI-
MS) studies of the reaction solution reveal that the formation
of 1 (m/z 1120.3, [M]+C) and 2 (m/z 1264.08, [M]+C) proceeds
very rapidly with both clusters being present after short
reaction times (30 min). The ratio of the products changes in
favor of 2 with longer reaction times. No other (larger)
clusters are found by LIFDI-MS (see Figure S1a and S1b for
details). Our attempts to optimize the conditions to achieve
more selective reactions and increased isolated yields were
unsuccessful. Nevertheless, 1 and 2 were characterized by
solution NMR spectroscopy, matching isotope patterns
(LIFDI-MS) (Figure S6a and S6b, Figure S7a and S7b),
UV/VIS spectroscopy, single-crystal X-ray analysis
:C{N(2,6-iPr2C6H3)CH}2)[1] and the closely related species
E2L2 (i.e. “Ge2”,[2] “P2”,[3] “As2”)[4] were major discoveries in
recent years.[5] Even the dicarbon congeners C2L2 should be
accessible, as was deduced from quantum chemical studies.[6]
N-heterocyclic carbenes (NHCs) turned out to be the key for
opening the door to this new chemistry, and their heavier
main group 13–14 analogs such as gallylenes (NHGa),
silylenes (NHSi), germylenes (NHGe), expand this library
of “multitalented” Lewis base ligands for stabilizing reactive
species and unusual bonding situations.[7] In particular the
NHGa compound Ga(DDP) (DDP = HC(CMeNC6H3-2,6-
iPr2)2)[8] behaves as a potent trapping ligand and a selective
reducing agent as well. This unique combination of properties
has been used for the synthesis of the metalloid tin clusters
[9]
Sn7L’2, Sn17L’4
and the bismuthenes Bi2L’2 (L’ =
(DDP)XGa; X = Cl, CF3SO3).[10]
ꢀ
Herein, we report about the reduction of GeCl2 with
Ga(DDP)/KC8 and the properties of two interesting products
(Figure 1), namely Ge2[Ga(DPP)]2 (1) and Ge4[Ga(DPP)]2
(2). Compound 1 reveals a planar, four-membered Ge2Ga2
(Figure 1), and elemental analysis. The absence of Ge H
moieties is confirmed by H NMR as well as Raman and IR
1
spectroscopy.[12] The experimental and complete analytical
details are given in the Supporting Information.
ring with the two NHGa ligands in a bridging position much in
Compound 1 crystallizes in the monoclinic space group
C2/c and compound 2 crystallizes in the trigonal space group
P3121 with half of the molecule as well as one disordered
solvent molecule in the asymmetric unit.[13] Both molecules 1
and 2 reveal centers of symmetry in their solid-state
structures. The planar, four-membered Ge2Ga2 rhombic ring
[3]
=
contrast to the chain-structured congener LGe GeL with
terminal NHC ligands L. The structure of 2 is derived from
a Ge4 tetrahedron by insertion of Ga(DDP) into two opposite
ꢀ
Ge Ge edges. This structural motif is unknown for ligand-
stabilized metalloid GenRm clusters (n > m; R = bulky aryl,
silylamide, etc.).[11]
of 1 exhibits two sets of equivalent Ga Ge bonds of 2.3899(8)
ꢀ
The new compounds Ge2[Ga(DPP)]2 (1) and Ge4[Ga-
(DPP)]2 (2) were reproducibly isolated in low yields (1, about
7%; 2, about 11%) upon treatment of (PCy3)·GeCl2 with
Ga(DDP)/KC8 in THF (258C, 2 h) and (NHC)·GeCl2
and 2.4113(8) ꢀ which are shorter than in 2 (2.5059(6) and
ꢀ
2.4755(6) ꢀ). The distance Ge(1) Ge(1’) of 2.8714(11) ꢀ of
ꢀ
1 is much longer than the four Ge Ge bonds of 2, which only
slightly vary around the average value of 2.459 ꢀ and match
the value of elemental germanium (2.45 ꢀ).[14] Larger metal-
loid GenRm clusters (n > 4; m < n) exhibit longer Ge–Ge
contacts, that is, Ge6R2 (2.546(1)–2.886(2) ꢀ; R = C6H3-2,6-
Dipp2; Dipp = C6H3-2,6-iPr2)[15a] and Ge8R6 (2.50 to 2.67 ꢀ;
R = N(SiMe3)2).[15b] The structure of 2 can be viewed as
derived from an ideal Ge4 tetrahedron with two opposite
edges being opened by insertion of two carbenoid Ga(DDP)
ligands in trans fashion which results in two long Ge–Ge
distances of 2.952 ꢀ. Compound 2 may be compared with
P4[Al(DDP)]2.[16] The latter was described by an ionic
[*] A. Doddi, Dr. C. Gemel, M. Winter, Dr. R. A. Fischer
Lehrstuhl fꢀr Anorganische Chemie II
Organometallics and Materials, Fakultꢁt fꢀr Chemie und Biochemie
Ruhr-Universitꢁt Bochum, 44780 Bochum (Germany)
E-mail: roland.fischer@rub.de
C. Goedecke, Dr. G. Frenking
Fachbereich Chemie, Philipps-Universitꢁt Marburg
Hans-Meerwein-Straße, 35032 Marburg (Deutschland)
E-mail: frenking@chemie.uni-marburg.de
Dr. H. S. Rzepa
Department of Chemistry, Imperial College, London (UK)
4ꢀ
bonding model as P4 butterfly-shaped Zintl anion coordi-
nated by two cationic [Al(DDP)]2+ units. Thus, 2 may be
regarded as a similar contact ion pair of a Ge44ꢀ anion and two
[Ga(DDP)]2+ units. However, such a description suggests to
[**] Support by the German Chemical Industry Fund is gratefully
acknowledged.
Supporting information for this article (experimental, analytical, and
computational details for compounds 1 and 2) is available on the
WWW under http://dx.doi.org/10.1002/anie.201204440.
ꢀ
view the very long Ge···Ge distances of 2.952 ꢀ as Ge Ge
450
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 450 –454

Angewandte
Chemie
Figure 1. Molecular structures of Ge2[Ga(DDP)]2 (1) and Ge4[Ga(DDP)]2 (2). Anisotropic displacement parameters are depicted at 50% probability
level. Hydrogen atoms have been removed for clarity. Interatomic distances (ꢂ) and angles (8) of the Ge2Ga2 core of 1: Ge(1)-Ge(1’), 2.8714(11)
[2.911]; Ga(1)-Ge(1)=2.4113(8) [2.467]; Ga(1)-Ge(1’)=2.3899(8) [2.443] and Ga(1)-Ga(1’)=3.848 [3.954]; Ge(1)-Ga(1)-Ge(1’)=73.46(3) [72.7];
Ga(1)-Ge(1)-Ga(1’)=106.54(3) [107.3] (The respective computational values for 1 at the BP86/SVP level of DFT theory are given in brackets).
Selected interatomic distances (ꢂ) of the Ge4Ga2 core of 2: Ge(1)-Ge(1’)=2.4545(9), Ge(1)-Ge(2)=2.4637(6), Ge(2)-Ge(2’)=2.4609(9), Ga(1)-
Ge(1)=2.5062(6), Ga(1’)-Ge(2)=2.4750(6) ꢂ. Further details on the molecular structure determination and bonding parameters of 1 and 2 are
given in the Supporting Information (Table S1).
4ꢀ
bonding (note: the electron precise Ge4 Zintl ion is
tetrahedral).
the four-membered ring of the former compound are bonded
to nitrogen atoms which have lone-pair p orbitals that can
donate electronic charge into the formally empty p(p) AOs of
Si. The gallium atoms in 1M do not possess lone-pair
p orbitals which leads to a bonding situation at the dicoordi-
nated Ge atoms that is quite different from that of the
We carried out quantum chemical calculations[17a] of
compounds 1 and 2 to analyze the electronic structure and
the bonding situation in both molecules. Herein we will focus
on compound 1, which reveals an unusual bonding situation.
A detailed account of the bonding aspects of both molecules
will be given in a forthcoming full paper.[17b] We optimized the
model structure 1M at the BP86/SVP level of theory (1M
contains CH3 groups at the N atoms instead of the bulky aryl
substituents in 1). A comparison of the most important
calculated bond lengths and bond angles (see the Figure 1 and
Table S2) of 1M shows a good agreement with the exper-
imental results of 1. The calculated Ga–Ge and Ga–N
distances are slightly longer than the experimental values.
Intermolecular forces usually lead to shorter bond lengths,
which holds particularly for long and weak bonds.[18] The
calculated transannular Ge–Ge distance (2.911 ꢀ) is also a bit
longer than the measured data (2.871 ꢀ).
2
!
Figure 2 shows the Laplacian distribution 1(r) of the
electron density in the plane of the Ge2Ga2 ring of 1M. The
ꢀ
contour line diagram shows four Ge Ga bond paths and one
ꢀ
ring critical point but there is no Ge Ge bond path. There are
2
!
four small areas of charge concentration ( 1(r) < 0, solid
ꢀ
lines) along the Ga Ge bond paths which are on the Ge side
of the bond critical points. This is in agreement with the
higher electronegativity of Ge (2.0) compared to Ga (1.8). A
2
!
striking aspect of the Laplacian distribution
1(r) is the
2
!
Figure 2. Contour line diagram of the Laplacian distribution 1(r) of
absence of electron density at the Ge atoms which would
1M in the plane of the four-membered ring. Solid lines indicate areas
come from s lone-pair orbitals at Ge. The latter were found in
2
!
of charge concentration ( 1(r)<0) while dotted lines show areas of
2
!
the Laplacian distribution 1(r) of the electron density at the
2
!
charge depletion ( 1(r)>0). The thick solid lines connecting the
Si atoms in the four-membered ring of Si2(NAr)2 which was
recently synthesized where two N atoms instead of Ga are
bonded to Si.[19] We want to point out that the silicon atoms in
atomic nuclei are the bond paths. The thick solid lines separating the
atomic basins indicate the zero-flux surfaces crossing the molecular
plane.
Angew. Chem. Int. Ed. 2013, 52, 450 –454
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
451

.
Angewandte
Communications
the Ge–Ga s bonds of the
Ge2Ga2 ring. The most impor-
tant result is the finding of two
disynaptic basins above and
below the four-membered ring
which feature the Ge-Ge p
bond. In contrast, there is no
disynaptic basin in the plane
which would indicate a Ge–Ge
s bond. Thus, the ELF calcu-
lations also suggest that 1M
exhibits a Ge–Ge p bond but
no genuine Ge–Ge s bond!
This unusual bonding situation
of 1M which is suggested by the
AIM investigation, the shape
analysis of the MOs, and the
ELF results is further sup-
ported by the results of NBO
calculations. Table 1 gives the
Figure 3. Shapes and eigenvalues e of important vacant and occupied orbitals of 1M which are relevant
for the bonding situation.
dicoordinated Si atoms in Si2(NAr)2. This difference comes
clearly to the fore when the relevant occupied and vacant
valence orbitals in the two compounds are compared with
each other.
Table 1: Natural orbital occupations for the bonds in the four-membered
ring of 1M and Wiberg bond indices WBI.
Type
Atom(s)
Occ.
WBI
Figure 3 shows the shape of two occupied and two vacant
orbitals of 1M which are crucial for the discussion of the
bonding situation. The complete set of occupied valence
orbitals is shown in the Supporting Information (Figure S20).
The HOMOꢀ1 and the LUMO of 1M are the plus and minus
s bond
s bond
s bond
s bond
s bond
s bond
s bond
s bond
p bond
lone pair
lone pair
Ga1-Ge1
Ga1-Ge2
Ga2-Ge1
Ga2-Ge2
Ga1-N1
Ga2-N2
Ga2-N3
Ga2-N4
Ge1-Ge2
Ge1
1.836
1.843
1.843
1.836
1.943
1.943
1.943
1.943
1.761
0.774
0.774
0.951
0.983
0.983
0.951
0.405
0.405
0.405
0.405
1.203
–
combination of the out-of-plane p(p ) AOs of Ge where the
?
former orbital has a small contribution from Ga. The shape of
the HOMOꢀ1 clearly identifies it as transannular Ge–Ge p-
bonding MO. The HOMOꢀ2 and the LUMO + 4 of 1M are
the plus and minus combination of the s lone-pair orbitals at
the Ge atoms.
Ge2
–
There is a striking difference to the bonding situation in
Si2(NAr)2, where the orbital analogs to the LUMO + 4 of 1 is
occupied while the HOMOꢀ1 is vacant.[19] The Ge2Ga2 ring in
1M has only one occupied s lone-pair MO which is the
HOMOꢀ2 while there are two occupied s lone-pair MOs at
Si in Si2(NAr)2. This explains the emergence of charge
concentration at Si in the Laplacian distribution which
comes from the s lone-pair orbitals. There is no such charge
concentration at the Ge atoms which have only one-half lone-
pair MO each. Instead of a second occupied s lone-pair MO
bonding orbitals and their occupation as well as the Wiberg
bond indices (WBI). The NBO results give four Ga–Ge
s bonds and four Ga-N s bonds. The occupation of the natural
orbitals is close to a value of two, which is a normal value for
a two-electron bond. Furthermore, there is a Ge–Ge p bond
which has also a rather high occupation of 1.795. Note that the
ꢀ
WBI value for the nonpolar Ge Ge bond is very high while
ꢀ
the WBI value for the very polar Ga N bonds is much
there is the occupied Ge–Ge out-of-plane p HOMO-
smaller. The calculated value of 1.203 for the bond order
indicates that the Ge–Ge p bond is supported by some weak
Ge–Ge s bonding which comes from the small backside lobes
of the orbitals in the HOMOꢀ2. The NBO analysis also gives
one lone-pair orbital for each germanium atom which,
however, is occupied by only about 0.8 electrons. There is
no genuine Ge–Ge s orbital in the NBO calculation. The best
description of this bonding situation in terms of the Lewis
formula concept features equivalent mesomeric structures
(Figure 4). Alternatively, the situation may be described by
two unpaired electrons at the germanium atoms possessing
opposite spins which are weakly coupled.
?
1 (Figure 3). Note that 1M may be regarded as hypoelectronic
(12 valence electrons for the Ge2Ga2 ring) with respect to the
Si2(NAr)2 reference compound (16 valence electrons for the
Si2N2 ring). In spite of the difference between the occupation
of the valence orbitals, both compounds Si2(NAr)2 and 1M
have electronic singlet ground states. Calculation of 1M in the
triplet state gave a structure which is 19.1 kcalmolꢀ1 higher in
energy than the singlet.[20]
The unusual bonding situation in 1M in terms of Ge–Ge
p-bonding but no s-bonding is supported by the NBO analysis
and by the ELF (electron localization function) calcula-
tions.[21] Two trisynaptic areas are found which connect the
germanium atoms with gallium (Figure S21). They represent
In summary, the bonding analysis of 1M clearly rules out
an ionic description of Ge2[Ga(DPP)]2 (1) and provides
452
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 450 –454

Angewandte
Chemie
[11] a) A. Schnepf, Eur. J. Inorg. Chem. 2008, 1007 – 1018; b) A.
Schnepf, Chem. Soc. Rev. 2007, 36, 745 – 758; c) A. Schnepf,
Coord. Chem. Rev. 2006, 250, 2758 – 2770.
[12] All signals observed in the 1H NMR spectrum of 1 are assigned
and there are no signals in the typical region of Ge-H groups
(about 3.92 and 8.0 ppm) also the IR and Raman spectra does
not provide any indications for the presence of terminal Ge-H
moieties: a) K. C. Thimer, S. M. I. Al-Rafia, M. J. Ferguson, R.
McDonald, E. Rivard, Chem. Commun. 2009, 7119 – 7121;
b) S. L. Choong, W. D. Woodul, C. Schenk, A. Stasch, A. F.
Richards, C. Jones, Organometallics 2011, 30, 5543 – 5550.
[13] Crystal Structure analyses (Oxford Excalibur 2 diffractometer):
The structural solution and refinement was performed using the
programs SHELXS-86 and SHELXL-97: G. M. Sheldrick,
SHELXS-97, Program for the Solution of Crystal Structures,
Universitꢁt Gçttingen, 1997; G. M. Sheldrick, SHELXL-97,
Program for Crystal Structure Refinement, Universitꢁt Gçttin-
gen, 1997; a) X-ray crystal structure analysis of 1:
C58H82Ga2Ge2N4, M = 1119.90, monoclinic, space group C2/c,
a = 24.2323(10) ꢀ, b = 15.2254(8) ꢀ, c = 15.0722(8) ꢀ, b =
Figure 4. Schematic description of the bonding situation in the four-
membered ring of 1 using two equivalent Lewis structures (above) or
spin-coupled lone-electrons (below). The formal negative charge at Ga
which cancels with a positive charge at N has been omitted because
this is not relevant for the discussion.
ꢀ
evidence for covalent Ge Ga bonding. The striking feature is
94.492(4)8, V= 5543.8(5) ꢀ3, Z = 4, 1calcd = 1.342 gcmꢀ3
, T=
ꢀ
the important transannular p(Ge Ge) single bond interac-
110(2) K, F(000) = 2336, m(MoKa) = 2.075 mmꢀ1, prismatic crys-
tal, 30770 reflections measured, 4869 unique (Rint = 0.1312), 303
parameters, R1 = 0.0564 (I > 2s(I)), wR2 = 0.1024 (I > 2s(I)),
GOF = 1.026. The overall data quality is limited, as revealed
by the R(int)value of 0.1312, because the crystal was small and
only weakly diffracting. No time dependency of the quality
within the measurement has been observed. Attempts to obtain
better crystals have not been successful; b) X-ray crystal
structure analysis of 2: C64H87FGa2Ge4N4, M = 1361.18, Trigonal,
space group P3121, a = 19.5740(3) ꢀ, c = 15.2771(3) ꢀ, V=
tion. The main difference in the bonding situation of the four-
membered cyclic moieties between 1M and Si2(NAr)2[19] is the
absence of p lone-pair donation into the group 14 atoms of
the former molecule.[20] Thus, compound 1 represents an
unusual example of a bonding situation where two (heavier)
main group atoms are bonded in p fashion without an
additional s bond between them. We are not aware of
a previous example where a molecule exhibits a p bond
between two atoms without a s bond.[22–24]
5069.10(15) ꢀ3, Z = 3, 1calcd = 1.338 gcmꢀ3
, T= 110(2) K, F-
(000) = 2094, m(MoKa) = 2.583 mmꢀ1, prismatic crystal, 10993
reflections measured, 6540 unique (Rint = 0.0203), 307 parame-
ters, R1 = 0.0295 (I > 2s(I)), wR2 = 0.0812 (I > 2s(I)), GOF =
1.034. The disordered solvent molecule in 2 was squeezed out
with the program Platon 1.13: A. L. Spek, Acta Crystallogr. Sect.
A 1990, 46, C34. The fluorobenzene molecule is included in the
empirical formula. CCDC 833302 (1) and CCDC 833301 (2)
contain the supplementary crystallographic data for this paper.
These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
Received: June 7, 2012
Revised: August 15, 2012
Published online: November 21, 2012
Keywords: bonding analysis · density functional calculations ·
.
gallium · germanium · metalloid clusters
[14] J. S. Kasper, S. M. Richards, Acta Crystallogr. 1964, 17, 752 – 755.
[15] a) A. Schnepf, R. Kçppe, Angew. Chem. 2003, 115, 940 – 942;
Angew. Chem. Int. Ed. 2003, 42, 911 – 913; b) A. F. Richards, H.
Hope, P. P. Power, Angew. Chem. 2003, 115, 4205 – 4208; Angew.
Chem. Int. Ed. 2003, 42, 4071 – 4074.
[16] Y. Peng, H. Fan, H. Zhu, H. W. Roesky, J. Magull, C. E. Hughes,
Angew. Chem. 2004, 116, 3525 – 3527; Angew. Chem. Int. Ed.
2004, 43, 3443 – 3445.
[1] Y. Wang, Y. Xie, P. Wei, R. B. King, H. F. Schaefer, P. von R.
Schleyer, G. H. Robinson, Science 2008, 321, 1069 – 1071.
[2] A. Sidiropoulos, C. Jones, A. Stasch, S. Klein, G. Frenking,
Angew. Chem. 2009, 121, 9881 – 9884; Angew. Chem. Int. Ed.
2009, 48, 9701 – 9704.
[3] Y. Wang, Y. Xie, P. Wei, R. B. King, H. F. Schaefer, P. von R.
Schleyer, G. H. Robinson, J. Am. Chem. Soc. 2008, 130, 14970 –
14971.
[17] a) The quantum chemical methods are described in the Support-
ing Information. b) The cyclic structure of 1 is remarkably
different from its NHC stabilized, chain like congeners LE = EL
with a distinct multiple bond situation between the main group
elements (E = Si, Ge; L = NHC).[1,2] The NHGa prefers the
bridging position in contrast to the terminal bonded NHC. The
NHC and NHGa ligand properties are characteristically differ-
[4] M. Y. Abraham, Y. Wang, Y. Xie, P. Wei, H. F. Schaefer,
P. von R. Schleyer, G. H. Robinson, Chem. Eur. J. 2010, 16,
432 – 435.
[5] a) Y. Wang, G. H. Robinson, Dalton Trans. 2012, 41, 337 – 345;
b) Y. Wang, G. H. Robinson, Chem. Commun. 2009, 5201 – 5213.
[6] J. L. Dutton, D. J. D. Wilson, Angew. Chem. 2012, 124, 1506 –
1509; Angew. Chem. Int. Ed. 2012, 51, 1477 – 1480.
[7] M. Asay, C. Jones, M. Driess, Chem. Rev. 2011, 111, 354 – 396.
[8] N. J. Hardman, B. E. Eichler, P. P. Power, Chem. Commun. 2000,
1991 – 1992.
ꢀ
ent which has consequences for the Ge Ge bond situation.
[18] V. Jonas, G. Frenking, M. T. Reetz, J. Am. Chem. Soc. 1994, 116,
8741 – 8753.
[19] R. S. Ghadwal, H. W. Roesky, K. Prçpper, B. Dittrich, S. Klein,
G. Frenking, Angew. Chem. 2011, 123, 5486 – 5490; Angew.
Chem. Int. Ed. 2011, 50, 5374 – 5378.
[9] G. Prabusankar, A. Kempter, C. Gemel, M. K. Schrçter, R. A.
Fischer, Angew. Chem. 2008, 120, 7344 – 7347; Angew. Chem. Int.
Ed. 2008, 47, 7234 – 7237.
[20] One referee suggested that the reason for the Ge-Ge p bonding
could be the slightly larger electronegativity of Ge compared
with Ga. We checked this by calculating the silicon and tin
[10] G. Prabusankar, C. Gemel, P. Parameswaran, C. Flener, G.
Frenking, R. A. Fischer, Angew. Chem. 2009, 121, 5634 – 5637;
Angew. Chem. Int. Ed. 2009, 48, 5526 – 5529.
Angew. Chem. Int. Ed. 2013, 52, 450 –454
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
453

.
Angewandte
Communications
homologues of 1M. The electronegativities of Si (1.7) and Sn
(1.7) are smaller than that of Ga (1.8) but the electronic structure
of the group-14 homologues showed also Si-Si and Sn-Sn
p bonding, respectively.
which are suggested for the bonding in S2I42+[23b] refer to orbitals
which have p symmetry with respect to the S2 and I2 fragments
ꢀ
but the S I bonding orbitals have s symmetry. A similar
situation is presented in the work of Bertrand[24a] and Driess.[24b]
Figure 4 in the former article shows nicely how two P-P p*
orbitals of three-membered cyclic fragments yield P-P s bonds.
The same holds true for the related systems which are described
by Driess[24b] where two S-S p* orbitals of three-membered cyclic
fragments yield S-S s bonds.
[21] A. D. Becke, K. E. Edgecombe, J. Chem. Phys. 1990, 92, 5397 –
5403.
[22] One referee suggested that related cases of p bonding between
main-group elements have been discussed in early works by
Passmore and co-workers[23] and in recent studies by Bertrand[24a]
and Driess.[24b] The article of Passmore present sketches of Lewis
structures for S82+[23a] and S2I42+[23b] which can not be considered
[23] a) C. G. Davies, R. J. Gillespie, J. J. Park, J. Passmore, Inorg.
Chem. 1971, 10, 2781 – 2785; b) J. Passmore, G. Sutherland, T.
Whidden, P. S. White, J. Chem. Soc. Chem. Commun. 1980, 289 –
290.
as proof for p bonding without s bonding. Moreover, the authors
2+[23b]
suggest that in S2I4
p* orbitals of the S2 and I2 fragments
interact such that the valence p orbitals of the atoms are aligned
along the bond which yields s bonding. We want to point out that
s and p define the symmetry of orbitals of a bond A-B with
respect to a plane that contains the bond. The p*-p* interactions
[24] a) Y. Canac, D. Bourissou, A. Baceiredo, H. Gornitzka, W. W.
Schoeller, G. Bertrand, Science 1998, 279, 2080 – 2082; b) S. Yao,
C. Milsmann, E. Bill, K. Wieghardt, M. Driess, J. Am. Chem. Soc.
2008, 130, 13536 – 13537.
454
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 450 –454