.
Angewandte
Communications
cated by a color change from yellow to dark green
(Scheme 3). After filtration and concentration, the solution
was stored at À258C, resulting in the deposition of violet
crystals (black appearance, 60% yield) which were unequiv-
ocally characterized (X-ray, elemental analysis, NMR, IR,
Raman spectroscopy) as [ClSb(m-NTer)2Sb][GaCl4] (3a;
Figure 1, left). Interestingly, upon addition of GaCl3, neither
monomerization nor any transformation to trimeric or
oligomeric species was observed.[17] The same procedure
was also applied to the analogous Bi species 2b. However,
after initial formation of a dark-brown solution, indicating,
Scheme 5. Synthesis of cyclo-dibisma(III)diazenium cation 6.
over 30 min. After filtration, the black residue was dissolved
in CH2Cl2 and the dark violet solution (black in appearance)
was again filtered, concentrated, and stored at À258C for
several hours. This process resulted in the deposition of black
crystals (yield 46%), which were identified by single-crystal
X-ray studies (Figure 1, right) as the cyclo-dibismadiazenium
tetrakis(pentafluorophenyl)borate dichloromethane solvate
[IBi(m-NTer)2Bi][B(C6F5)4]·3CH2Cl2 (6·3CH2Cl2).
1
according to H and 13C NMR studies, the presence of the
[ClBi(m-NTer)2Bi]+ ion (Scheme 3), the isolation as [GaCl4]À
salt (3b) from this mixture was impossible owing to rapid
decomposition upon crystallization conditions (high concen-
tration). Utilization of other chloride-abstracting reagents,
such as SbCl5 and Ag[SbF6], gave the same result. Obviously,
the anions [GaCl4]À, [SbCl6]À, and [SbF6]À were involved in
the decomposition process and were not innocent. Therefore,
we decided to use the chemically robust weakly coordinating
anion [B(C6F5)4]À as counterion for the [ClBi(m-NTer)2Bi]+
ion to avoid decomposition triggered by cation–anion inter-
actions. As the reaction of [ClBi(m-NR)]2 (2b) with Ag[B-
(C6F5)4] was surprisingly slow and immediate decomposition
was observed, a chlorine/iodine exchange was attempted to
yield [IBi(m-NTer)]2. It is known that element triflate
compounds are good starting materials for a triflate–iodine
substitution, which is easily achieved by addition of Me3SiI,
[5]
À
inducing the elimination of Me3Si OTf. Thus, at first
[ClBi(m-NTer)]2 (2b) was transformed to the hitherto
unknown compound [(TfO)Bi(m-NTer)]2 (4b) in the reaction
of 2b with two equivalents of AgOTf (Scheme 4).[16] It should
be noted that this route also worked well for the Sb species,
Figure 1. ORTEP representation of the molecular structure of the
cations in 3a (left) and 6·3CH2Cl2 (right). Ellipsoids are set at 50%
probability (at 173 K); hydrogen atoms, anions, and solvent molecules
are omitted for clarity. Selected bond lengths [ꢂ] and angles [8]: 3a:
Sb1–N1 1.996(2), Sb1–N2 2.024(2), Sb2–N2 2.071(2), Sb2–N1
2.115(2), Sb2–Cl 2.3646(8), Sb1···Sb2 3.2100(3), N1–C1 1.405(3), N2–
C25 1.399(3); N1-Sb1-N2 78.84(9), N2-Sb2-N1 75.16(9), N2-Sb2-Cl
99.30(7), N1-Sb2-Cl 87.87(6), C1-N1-Sb1 131.4(2), C1-N1-Sb2
123.9(2), Sb1-N1-Sb2 102.7(1), C25-N2-Sb1 127.6(2), C25-N2-Sb2
124.9(2), Sb1-N2-Sb2 103.2(1); 6·3CH2Cl2: Bi1–N2 2.114(3), Bi1–N1
2.155(3), Bi2–N1 2.196(3), Bi2–N2 2.241(3), Bi2–I 2.8580(4), Bi1···Bi2
3.4155(3), N1–C1 1.402(4), N2–C25 1.402(4); N1-Bi2-N2 74.0(2), N1-
Bi2-I1 100.11(9), N2-Bi2-I 89.34(9), N2-Bi1-N1 77.4(2), C1-N1-Bi1
127.9(2), C1-N1-Bi2 126.9(2), Bi1-N1-Bi2 103.4(2), C25-N2-Bi1
130.0(3), C25-N2-Bi2 125.3(2), C1-N1-Bi1 127.9(2), Bi1-N2-Bi2
103.3(2).
Scheme 4. Synthesis of iodine species 5 by Cl/OTf/I exchange.
yielding the desired triflate species [(TfO)Sb(m-NTer)]2 (4a)
in good yields (Sb: 65%, Bi: 88%) as pure crystalline solids.
In the next step, 4b was reacted with an excess of Me3SiI,
which resulted in the formation of [IBi(m-NTer)]2 (5).[16]
Compound 5 could also be obtained when 2b was treated
with two equivalents of NaI, but a complete conversion,
separation from side products, and NaCl by extraction was
very difficult to achieve and was time-consuming. Thus an
overall yield of only 20% was obtained, while with the triflate
substitution route pure crystalline 5 was isolated in an overall
yield of more than 60%.Compound 5 (Scheme 5) is stable as
Both salts 3a and 6·3CH2Cl2 bearing the cyclo-dipnicta-
diazenium ions are air- and moisture-sensitive but stable in
argon over a long period as solids.[16] The black appearance of
3a and 6·3CH2Cl2 vanishes rapidly when traces of H2O or
oxygen are present. Compounds 3a and 6·3CH2Cl2 are easily
prepared in bulk and are stable for long periods when stored
in a sealed tube and kept at À308C in the dark. Compound 3a
is thermally stable up to 2958C, whereas 6·3CH2Cl2 can be
heated up to 1918C. Decomposition starts at these temper-
atures. X-ray studies of crystals from the reaction sequences
illustrated in Scheme 3 (Sb) and Scheme 5 (Bi) reveal that
salts 3a and 6·3CH2Cl2 with a cyclo-1,3-dipnicta-2,4-diaze-
nium cation that is kinetically protected in the pocket formed
by the terphenyl groups (Figure 1). Compound 3a crystallizes
1
a dimer in toluene at ambient temperature, as shown by H
and 13C NMR studies. However, after adding a solution of
[Ag(toluene)3][B(C6F5)4] in toluene at À788C, the initially
deep-red solution turned violet and a black precipitate
formed. The black suspension was stirred for 30 min at
À788C and then allowed to warm up to ambient temperature
2
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
These are not the final page numbers!