On the reduction of NC chelated organoantimony(III) chlorides

Article abstract of DOI:10.1002/ejic.201100010

The conversion of organolithium compound LLi, for which L = [o-C ₆H₄(CH=NC₆H₃iPr₂-2,6)] ^-, with antimony chloride gave the molecular chlorides LSbCl ₂ (1) and L₂SbCl (2) depending on the molar ratio used (either 1:1 or 2:1). Bothcompounds were characterized by using ¹H and ¹³C NMR spectroscopy, elemental analysis and single-crystal X-ray diffraction. The reaction of 1 with two molar equivalents of K[B(sBu) ₃H] led to smooth formation of compound L₄Sb₄ (3) as a result of hydrogen elimination from the unstable hydrido compound. On the contrary, a similar reaction between 2 and K[B(sBu)₃H] (1:1) did not result in hydrogen elimination and the formation of expected distibine L₂SbSbL₂, but an addition of an in situ generated L ₂Sb-H bond across the C=N functionality in the pendant arm of one of the ligands was observed. The reduction of compound 2 with an excess amount of magnesium was strongly dependent on the reaction time. The distibine L ₂SbSbL₂ (5) could be isolated after 4-5 h from this reaction mixture. Elongation of the reaction time to 1 d (and more) gave a more complicated reaction mixture, in which three products, distibine 5, compound 3 and organomagnesium compound L₂Mg(THF) (6), were characterized by single-crystal X-ray diffraction in addition to other unidentified products. These results can be rationalized by a migration of the ligand L from the antimony atom to the magnesium, thus explaining formation of both L ₄Sb₄ (3) compound and L₂Mg(THF) (6).

Full text of DOI:10.1002/ejic.201100010

FULL PAPER
DOI: 10.1002/ejic.201100010
On the Reduction of NC[‡] Chelated Organoantimony(III) Chlorides[‡‡]
[a]
[a]
[a]
[a,b]
ˇ
ˇ
˚ˇ ˇ
Libor Dostál,* Roman Jambor, Ales Ruzicka, and Petr Simon
Keywords: Chelates / Antimony / Reduction / Ligand effects
The conversion of organolithium compound LLi, for which L
[o-C6H4(CH=NC6H3iPr2-2,6)]–, with antimony chloride
functionality in the pendant arm of one of the ligands was
observed. The reduction of compound 2 with an excess
=
gave the molecular chlorides LSbCl2 (1) and L2SbCl (2) de- amount of magnesium was strongly dependent on the reac-
pending on the molar ratio used (either 1:1 or 2:1). Both
compounds were characterized by using 1H and 13C NMR
spectroscopy, elemental analysis and single-crystal X-ray dif-
fraction. The reaction of 1 with two molar equivalents of
K[B(sBu)3H] led to smooth formation of compound L4Sb4 (3)
as a result of hydrogen elimination from the unstable hydrido
compound. On the contrary, a similar reaction between 2 and
K[B(sBu)3H] (1:1) did not result in hydrogen elimination and
the formation of expected distibine L2SbSbL2, but an ad-
dition of an in situ generated L2Sb–H bond across the C=N
tion time. The distibine L2SbSbL2 (5) could be isolated after
4–5 h from this reaction mixture. Elongation of the reaction
time to 1 d (and more) gave a more complicated reaction mix-
ture, in which three products, distibine 5, compound 3 and
organomagnesium compound L2Mg(THF) (6), were charac-
terized by single-crystal X-ray diffraction in addition to other
unidentified products. These results can be rationalized by
a migration of the ligand L from the antimony atom to the
magnesium, thus explaining formation of both L4Sb4 (3) com-
pound and L2Mg(THF) (6).
on dibismuthenes and distibenes, stabilized by terphenyl li-
gands, that were prepared by the reduction of chlorido pre-
cursors by alkali metals as well.[8] The common feature of
the majority of reported compounds is the shielding of the
metallic core by ligands that are sterically very demanding,
which gives sufficient kinetic stability to the compounds.
Another approach toward the stabilization of such com-
pounds is the coordination of the central antimony (or bis-
muth) atom to a transition-metal fragment,[9] which elec-
tronically influences and stabilizes the low-valent antimony
or bismuth atom. Quite recently, Breunig et al. proved,[10]
as a reminder of the electronic stabilization of low-valent
antimony and bismuth atoms by transition metals, the abil-
ity of an NC chelating ligand, [o-C6H4(CH2NMe2)]– (Fig-
ure 1, A, denoted as L#) to stabilize both low-valent anti-
Introduction
In 1965, Issleib et al. reported on the synthesis of
tBu4Sb4 as the first example of well-defined molecular orga-
nometallic compounds with central antimony atoms in, up
to that moment, an unusually low oxidation state.[1] This
pioneering work was followed by many other works that
dealt with the preparation of analogous compounds with
low-valent antimony or bismuth centres.[2] This led to the
description of monocyclic derivatives of Sb and Bi RnMn (n
= 3–6),[3] and several exciting polycyclic systems[2a,3h,4] as
well as a significant number of distibines and dibismuth-
ines[4a,5] R2MMR2 were isolated (M = Sb or Bi). Another
landmark in the research in this field was the preparation
of the dibismuthene RBi=BiR as an analogue of alkenes
with a double bond between bismuth atoms by Okazaki
and Tokitoh,[6] which was shortly followed by the isolation
of antimony analogue RSb=SbR.[7] Power et al. reported
[‡] NC designates [o-C6H4(CH=NC6H3iPr2-2,6)]–
[‡‡] The same ligand system for the stabilization of antimony
bromides and sulfides was recently reported by M. Mehring
et al. at the Third EuCheMS Chemistry Congress, Nürnberg,
Germany, poster no. VIIb.097.
[a] Department of General and Inorganic Chemistry, Faculty of
Chemical Technology, University of Pardubice,
Studentská 573, 53210 Pardubice, Czech Republic
Fax: +420-466037068
E-mail: libor.dostal@upce.cz
[b] Institute of Microbiology, Centre of Biocatalysis and
Biotransformation, Academy of Sciences of the Czech
Republic,
Vídenˇská 1083, 14220 Prague, Czech Republic
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201100010.
Figure 1. Structures of selected NC and NCN chelating ligands.
2380
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 2380–2386

Reduction of NC-Chelated Organoantimony(III) Chlorides
mony and bismuth atoms. The same group also proved that
a related NCN pincer-type ligand, [o,o-C6H4(CH2NMe2)2]–
(Figure 1, B, denoted as LЈ) can stabilize organobismuthine
LЈ2BiBiLЈ2.[11,12]
We have used the same ligand (LЈ) for the preparation of
monocyclic stibine LЈ4Sb4 by the reduction of the parent
chlorido compound. More importantly, a unique byproduct
was isolated from the reaction mixture, namely, LЈ3Sb5 clus-
ter compound with a propellane-like structure.[13] Recently
we demonstrated that the use of the more sterically de-
manding NCN ligand (Figure 1, C), which contains two
ketimino donor arms, allowed us to obtain unprecedented
monomeric stibinidene and bismuthinidene.[14]
To shed more light on the possibility of stabilization of
low-valent heavier main-group centres by similar chelating
ligands, we decided to start systematic research into a sim-
ilar NCN pincer as well as analogous sterically demanding
NC ligands. As a part of our effort, we report here on the
use of ligand L, [o-C6H4(CH=NC6H3iPr2-2,6)]– (Figure 1,
D), which contains only one ketimino pendant arm but is
armed with a bulky aryl substituent. Synthesis of starting
chlorido compounds LSbCl2 and L2SbCl and attempts to-
ward their reduction either by means of a dehydro–coupling
synthetic protocol[15] through unstable hydrido precursors
or by conventional reduction with magnesium are dis-
cussed.
Scheme 1. Preparation of compounds 1 and 2.
Analogously to the synthetic protocol used by us recently
for preparation of low-valent antimony compounds, we
treated compounds 1 and 2 with one or two equivalents of
K[B(sBu)3H] with the aim of preparing unstable hydrido
complexes that should eliminate hydrogen gas and form
low-valent metal centres.[13,14] In the case of precursor 1,
the reaction proceeded smoothly in an expected manner
through visible hydrogen-gas evolution. Organoantimony(I)
compound L4Sb4 (3) could be isolated after extraction (tol-
uene) as an air-sensitive reddish solid in 45% yield
(Scheme 2). This compound was characterized by elemental
analysis and 1H and 13C NMR spectroscopy, which re-
vealed only one set of relatively sharp signals, thereby ruling
out any ring–ring equilibrium that would give rise to any-
thing other than the four-membered ring and also proving
the equivalency of all ligands in the structure. Similar find-
ings were reported for four-membered ring systems
[10b]
L#4Sb4
or LЈ4Sb4.[13] Compound 3 is stable for a long
time both in solution and in the solid state under an argon
atmosphere.
Results and Discussion
Treatment of in situ prepared organolithium compound
LLi in diethyl ether, recently isolated by Mu et al. from
hexane,[16] with antimony chloride gave either LSbCl2 (1) or
L2SbCl (2) according to the molar ratio used (Scheme 1).
Both compounds were isolated as air stable white (1) or
yellow (2) solids in reasonable yield (46% and 67%). The
identity of 1 and 2 was established by elemental analysis,
1H and 13C NMR spectroscopy and single-crystal X-ray
Scheme 2. Reaction of 1 with K[B(sBu)3H].
On the contrary, the analogous treatment of 2 with one
diffraction techniques. The 1H NMR spectrum of 1 re- equivalent of K[B(sBu)3H] most probably led to the forma-
vealed one set of sharp signals, in which two doublets are tion of a hydrido compound, which may be postulated as
present for the methyl groups of iPr groups, thus suggesting L2SbH, but instead of the expected hydrogen elimination
the presence of rigid NǞSb interactions in solution. Two and the formation of the target compound L2SbSbL2, an
1
sets of signals were obtained in the H NMR spectrum of addition of an in situ generated Sb–H bond across the C=N
2, thus pointing to the fact that both ligands are nonequiva- double bond in one of the ligand arms was observed. This
lent in the structure of 2 (probably due to different strengths reaction pathway led to isolation of compound 4 in 48%
of the NǞSb interactions), which is consistent with the yield (Scheme 3); it contained a new Sb–N covalent bond
solid state of 2 (vide infra). Interestingly, an analogous com- within an aza-stiba ring. All our attempts to identify the
1
pound that contained two NC ligands, L#2SbCl (Figure 1, intermediate L2SbH failed. The H NMR spectroscopic ex-
A), displayed one set of broad signals at ambient tempera- periment of addition of K[B(sBu)3H] to 2 in C6D6 smoothly
ture as a result of the fluxional behaviour of this com- gave compound 4 after less than 1 min. (see the Supporting
pound, but at –60 °C two sets of signals were resolved, Information). Similarly, mixing K[B(sBu)3H] and 2 at
which was ascribed only to the nonequivalent degrees of –80 °C followed by the addition of an excess amount of
NǞSb coordination.[17] This comparison points to the tBuNC or tBuCN (10 equiv.) did not lead to addition ac-
conclusion that the NǞSb interactions in the case of 2, ross the Sb–H bond, but compound 4 was isolated as the
which contains ligand L, are more rigid than in the ana- major product. The identity of compound 4 and the unam-
logue with classical L# ligand (Figure 1, A).
biguous evidence for addition across the C=N double bond
Eur. J. Inorg. Chem. 2011, 2380–2386
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2381

ˇ
L. Dostál, R. Jambor, A. Ru˚zicˇka, P. Simon
ˇ
FULL PAPER
1
stems mainly from H and 13C NMR spectroscopic data tallized in moderate yield as an orange-red crystalline mate-
1
(Figure S1). The H NMR spectrum of 4 revealed an AX rial. However, in most cases, the crude mixture contained a
pattern at δ = 5.18 ppm, which is indicative of the formation significant amount of starting material 2 and unfortunately
of a methylene group CH2N in the ligand arm. This reso- this compound was also soluble in hexane. Consequently,
nance pattern is typical for classical CN- or NCN-chelated isolation of pure 5 requires several recrystallization steps,
organoantimony compounds with CH2NR2 pendant arms. which cause a significant decrease in the reaction yield
The observation of only one singlet resonance for CH=N (39%). Compound 5 was characterized by elemental analy-
(at δ = 8.05 ppm) in the 1H NMR spectrum of 4, which has sis and 1H and 13C NMR spectroscopy, which revealed only
an integral intensity of 1:2 compared to the AX pattern one set of signals for all four ligands in the structure of
for the new CH2N group, is further proof for the proposed 5, thereby suggesting the highly fluxional structure of 5 in
structure of 4. Finally, the signal at δ = 70.0 ppm in the 13
NMR spectrum of 4 also established the formation of a
C
solution at ambient temperature.
Prolonging the reaction time to 12 or 24 h[10] as a logical
CH2N linkage, and only one signal with a typical chemical attempt to increase the reaction yield of 5 resulted in an
shift at δ = 167.8 ppm for CH=N was detected, thereby fur- even more complicated mixture of products, which we were
ther supporting the proposed structure of 4. Unfortunately, not able to describe and understand completely. Besides the
all of our numerous attempts to obtain suitable single crys- expected and desired compound 5, the compound L4Sb4 3
tals of 4 for X-ray diffraction techniques failed.
was identified, the presence of which was proven based on
1
the analysis of the H and 13C NMR spectra of these mix-
tures. By chance, we were able to obtain from this mixture
single crystals of an organomagnesium derivative
L2Mg(THF) (6) for X-ray diffraction measurements (see the
Supporting Information). This result may be rationalized
by migration of the ligand L from the antimony atom to
magnesium, which is also supported by the identification of
compound L4Sb4 (3), which may be formed in this reaction
mixture only by a loss (migration) of one ligand. Unfortu-
nately, all of our numerous attempts to isolate compound
6 in larger quantities by a controlled procedure from the
treatment of 2 with Mg or, for example, by reduction of 1
with Mg failed. Similarly, the possibility of the formation
of 6 by treating 1 or 2 with MgCl2 was ruled out and only
starting materials were isolated after reaction even when an
excess amount of MgCl2 and long reaction times (more
than 1 week) were applied.
Scheme 3. Reaction of 2 with K[B(sBu)3H].
These findings reveal interesting differences in the reac-
tion between either 1 or 2 and K[B(sBu)3H]. The reduction
of the central atom and formation of compound 3 is ob-
tained when 1 is used as starting material, whereas addition
of the Sb–H bond across the C=N double bond of the li-
gand backbone is preferred in the second case, thereby giv-
ing compound 4.
To overcome the problems with the preparation of com-
pound L2SbSbL2, the reduction of precursor 2 with magne-
sium (Scheme 4), which was proven to be very convenient
for the preparation of similar NC-chelated (L#) organoanti-
mony compounds,[10] was applied. Although this method
allowed us to isolate the desired compound L2SbSbL2 (5),
the reaction was not as straightforward as expected. The
reaction mixture started to turn orange-red immediately,
and if the reaction mixture was evaporated after, typically,
3–5 h and extracted by hexane, compound 5 could by crys-
Molecular structures of 1–3, 5 and 6 were determined by
X-ray diffraction techniques and are depicted in Figures 2,
3, 4, 5 and Figure S3 (for the latter, see the Supporting In-
Figure 2. ORTEP diagram of compound 1 with thermal displace-
ment parameters at 30% probability. Hydrogen atoms were omitted
for clarity. Selected distances [Å] and angles [°]: Sb1–C1 2.149(3),
Sb1–N1 2.416(2), Sb1–Cl1 2.3704(8), Sb1–Cl2 2.4856(7); C1–Sb1–
N1 74.15(8), C1–Sb1–Cl1 95.93(7), C1–Sb1–Cl2 92.62(9), N1–Sb1–
Cl2 163.96(5).
Scheme 4. Reduction of 2 by magnesium.
2382
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 2380–2386

Reduction of NC-Chelated Organoantimony(III) Chlorides
formation). Relevant structural parameters are given in the
figure captions.
Figure 5. ORTEP diagram of compound 5 with thermal displace-
ment parameters at 30% probability. Hydrogen atoms were omitted
for clarity. Selected distances [Å] and angles [°]: Sb1–Sb2 2.9194(6),
Sb1–C1 2.169(5), Sb1–C20 2.198(5), Sb2–C39 2.168(5), Sb2–C58
2.166(4), Sb1–N1 3.018(4), Sb1–N2 2.979(4), Sb2–N3 2.877(4),
Sb2–N4 4.559(5); C1–Sb1–C20 93.16(18), C39–Sb2–C58 98.59(19),
Sb1–Sb2–C39 96.27(14), Sb1–Sb2–C58 94.59(12), Sb2–Sb1–C1
96.09(12), Sb2–Sb1–C20 93.75(13).
Figure 3. ORTEP diagram of compound 2 with thermal displace-
ment parameters at 30% probability. Hydrogen atoms were omitted
for clarity. Selected distances [Å] and angles [°]: Sb1–C1 2.168(3),
Sb1–C20 2.161(2), Sb1–N1 2.416(2), Sb1–N2 2.952(3), Sb1–Cl1
2.5149(7); C1–Sb1–N1 73.81(9), C1–Sb1–N2 163.81(8), C1–Sb1–
C20 95.51(10), C1–Sb1–Cl1 91.86(7), N1–Sb1–Cl1 163.76(6), N2–
Sb1–C1 163.81(8), C20–Sb1–N1 85.43(8), C20–Sb1–N2 68.38(9),
C20–Sb1–Cl1 88.49(7).
equatorial position. There is no intermolecular contact of
significance in the crystal structure of 1; this is in contrast
to dichloro derivative L#SbCl2 (Figure 1, A), in which the
intermolecular Sb···Cl contacts lead to formation of an infi-
nite chain.[17] This fact reflects the higher steric hindrance
of the ligand L.
Both nitrogen atoms N1 and N2 are coordinated to the
central atom Sb1 in 2, but the bond lengths point to a dif-
ferent strength of these contacts: compare Sb1–N1 2.416(2)
to Sb1–N2 2.952(3) Å, as suggested already for the struc-
ture in solution (vide supra). Nevertheless, if the weaker
interaction with the N2 atom is also taken into account, the
resulting coordination polyhedron of the central atom may
be described as a strongly pseudo-tetragonal pyramid. The
apical position is occupied by the C20 atom. The equatorial
plane is formed by the second ipso-carbon atom C1 and
nitrogen donor atoms and the chlorine atom Cl1. The N1,
Cl1 and C1, N2 are placed in trans positions [N1–Sb1–Cl1
163.76(6) and N2–Sb1–C1 163.81(8)].
Figure 4. ORTEP diagram of the compound 3 with thermal dis-
placement parameters at 30% probability. Hydrogen atoms were
3
omitted for clarity. Symmetry operators: a = /2 – x, ½ – y, z; b =
The molecular structure of compound 3 proved the tetra-
meric nature of the compound with the central Sb4 ring.
This ring is strongly puckered and the ligands L are placed
in all-trans fashion around the Sb4 girdle, which coincides
½ + y, 1 – x, 1 – z; c = 1 – y, –½ + x, 1 – z. Selected distances
[Å] and angles [°]: Sb1–C1 2.168(4), Sb1–N1 2.761(4), Sb1–Sb1b
2.8869(4); Sb1b–Sb1–Sb1c 88.12(1).
The nitrogen atom N1 is strongly coordinated to the cen- with the structure described in solution. The Sb–Sb bond
tral atom Sb1 in the structure of compound 1 [Sb1–N1 lengths within the ring are 2.8869(4) Å and the bonding Sb–
2.416(2) Å, Figure 2]. The coordination polyhedron of the Sb–Sb angles amount to 88.12(1)°. These bond lengths are
central atom is, as a result of this NǞSb dative connection, a bit longer in comparison to analogous tetrameric L#4Sb4
best described as a pseudo-trigonal bipyramid in which the (Figure 1, A), and the bonding angles are also significantly
apical positions are occupied by the atoms N1 and Cl2 [the wider [cf. the range of Sb–Sb contacts 2.8474(5)–
bonding angle of N1–Sb1–Cl2 is 163.96(5)°]; the chlorine 2.8605(4) Å
and
Sb–Sb–Sb
angles
78.462(11)–
atom Cl1 and the ipso-carbon atom C1 remain in the equa- 80.825(12)°].[10b] All nitrogen atoms are coordinated to the
torial positions. The bonding angle of C1–Sb1–Cl1 central antimony atoms with bond lengths of 2.761(4) Å, a
[95.93(7)°] is more acute than the ideal value 120° as a con- value that is longer than that in parent chlorido precursor
sequence of repulsion from the lone pair of the antimony LSbCl2 (1), which reflects the lower Lewis acidity of the
atom, which is most probably directed toward the third central antimony in 3 compared to 1. More importantly,
Eur. J. Inorg. Chem. 2011, 2380–2386
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2383

ˇ
L. Dostál, R. Jambor, A. Ru˚zicˇka, P. Simon
ˇ
FULL PAPER
these NǞSb interactions are considerably weaker than leads to a complicated reaction mixture, in which com-
those detected in the monomeric stibinidene stabilized by pounds 3 and 5 were characterized along with organomag-
NCN pincer ligand (Figure 1, C), in which the Sb–N dis- nesium compound 6. This finding may be explained by mi-
tances are 2.352(3) and 2.346(3) Å and indicate very strong gration of the ligand L from the antimony atom to magne-
interactions.[14] These strong NǞSb interactions are crucial sium. We are currently studying the reduction of compound
for the stabilization of monomeric stibinidene. But in the L2SbCl (2) with alkali metals.
case of compound 3, support of the central antimony atom
by only one pendant ketimino donor arm, although more
sterically demanding, is not sufficient for stabilization of the
Experimental Section
monomeric structure and so the backbone of 3 is domi-
General Procedures: All air- and moisture-sensitive manipulations
nated by an internal Sb4 ring, thus rendering the NǞSb
were carried out under an argon atmosphere using standard
interaction less important (weaker) for stabilization of the
Schlenk tube techniques. All solvents were dried by standard pro-
central antimony atom.
cedures and distilled prior to use. 1H and 13C spectra were recorded
The Sb1–Sb2 bond length of 2.9194(6) Å in the molecu-
with a Bruker 400 spectrometer using a 5 mm tuneable broadband
lar structure of 5 is considerably longer than corresponding
values in Ph4Sb2 (2.834 Å), Me4Sb2 (2.86 Å) or [(Me3Si)2-
CH]2(H)2Sb2 (2.8304 Å).[4a,5c,18] The comparable distibine
probe. Appropriate chemical shifts in 1H and 13C NMR spectra
were related to the residual signals of the solvent [CDCl3: δ(1H) =
7.27 ppm and δ(13C) = 77.23 ppm; C6D6: δ(1H) = 7.16 ppm, δ(13C)
= 128.39 ppm]. The starting ligand o-C6H4(CH=NC6H3iPr2-2,6)Br
was prepared according to the literature procedure.[16]
L#2SbSbL# has not, to the best of our knowledge, been
2
reported. The elongation of the Sb–Sb bond may be in part
ascribed to a steric repulsion of four ligands L as well as to
NǞSb interactions. Both nitrogen atoms N1 and N2 are
coordinated to the antimony atom Sb1 [Sb1–N1 3.018(4),
[o-C6H4(CH=NC6H3iPr2-2,6)]SbCl2 (1): nBuLi (7.9 mL, 13 mmol,
1.6 m solution in hexane) was added to
C6H4(CH=NC6H3iPr2-2,6)Br (4.35 g, 13 mmol) in diethyl ether
a solution of o-
Sb1–N2 2.979(4) Å], but in the case of the antimony atom (100 mL) at –70 °C and stirred for 1 h. The resulting yellow-orange
suspension of lithium compound was added to a solution of SbCl3
(2.88 g, 13 mmol) in diethyl ether (100 mL) precooled to –40 °C.
The resulting mixture was allowed to reach room temperature and
stirred for an additional 3 h. The volume of the reaction mixture
was reduced to around 30 mL. The insoluble material was filtered
off and washed with hexane (15 mL). The remaining yellowish solid
was extracted by dichloromethane (20 mL). Evaporation of the ex-
tract gave 1 as yellowish powder (this powder may be crystallized
from dichloromethane/hexane to obtain single crystals suitable for
X-ray studies). Yield 2.66 g, 46%; m.p. 217–219 °C. 1H NMR
(400 MHz, CDCl3, 25 °C): δ = 1.13 [d, 6 H, CH(CH3)2], 1.25 [d, 6
H, CH(CH3)2], 2.91 [sept, 2 H, CH(CH3)2], 7.26 (m, 3 H, C6H3iPr2-
2,6), 7.69 (dd, 1 H, C6H4), 7.83 (m, 2 H, C6H4), 8.49 (s, 1 H,
CH=N), 8.79 (d, 1 H, C6H4) ppm. 13C NMR (100.61 MHz, CDCl3,
25 °C): δ = 25.1 [s, CH(CH3)2], 25.7 [s, CH(CH3)2], 29.0 [s,
CH(CH3)2], 124.4, 127.6, 131.5, 133.0, 134.5, 136.3, 138.9, 140.7,
141.7, 153.3 (s, Ar–C), 169.2 (s, CH=N) ppm. C19H22Cl2NSb
(457.05): calcd. C 49.9, H 4.9; found C 49.7, H 4.6.
Sb2 only the N3 atom remains orientated towards the cen-
tral metal [Sb2–N3 2.877(4) Å], with the second nitrogen
atom N4 being pendant [Sb2–N4 4.559(5) Å]. This situation
is rather similar to the dibismuthine L#2BiBiL# (Figure 1,
2
A), in which only three of four ligand arms are likewise
coordinated to bismuth atoms.[10a] Nevertheless, all NǞSb
interactions in 5 are very weak, if not negligible, which coin-
cides with fluxional behaviour of all four ligands in the
structure of 5 in solution (vide supra).
Conclusion
We have demonstrated that the use of L, with only one
ketimino ligand arm, is not sufficient for stabilization of the
monomeric stibinidene. Only cyclic tetrameric compound
3 could be isolated from the reaction of parent chloride
compound 1 with K[B(sBu)3H]. This fact may be mainly
ascribed, in our opinion, to the absence of steric protection
of the second ortho position in the ligand structure. There-
fore future research will be devoted to the preparation of
NC ligands substituted in the second ortho position with a
bulky group. Interestingly, an analogous reaction with dior-
ganoantimony precursor L2SbCl (2) resulted in the addition
{[o-C6H4(CH=NC6H3iPr2-2,6)]}2SbCl
(2):
nBuLi
(5.9 mL,
9.4 mmol, 1.6 m solution in hexane) was added to a solution of o-
C6H4(CH=NC6H3iPr2-2,6)Br (3.25 mg, 9.4 mmol) in diethyl ether
(60 mL) at –70 °C and stirred for 1 h. The resulting yellow-orange
suspension of lithium compound was added to a solution of SbCl3
(1.1 g, 9.4 mmol) in diethyl ether (50 mL) precooled to –20 °C. The
resulting mixture was allowed to reach room temperature and
stirred for an additional 12 h. The volume of the reaction mixture
of the in situ generated Sb–H bond across the C=N double was reduced to around 10 mL. The insoluble material was filtered
off and washed with cold (–30 °C) hexane (5 mL). The remaining
yellow solid was extracted by dichloromethane (20 mL). Evapora-
tion of the extract gave 2 as yellow powder (this powder may be
crystallized from hot hexane to obtain single crystals suitable for
X-ray studies). Yield 2.2 g, 68%; m.p. 185–187 °C. 1H NMR
(400 MHz, CDCl3, 25 °C): δ = 0.46 [d, 3 H, CH(CH3)2], 0.77 [d, 3
H, CH(CH3)2], 0.93 [d, 6 H, CH(CH3)2], 1.02 [d, 6 H, CH(CH3)2],
1.07 [d, 3 H, CH(CH3)2], 1.10 [d, 3 H, CH(CH3)2], 1.84 [sept, 1 H,
CH(CH3)2], 2.83 [m, 3H-overlap of two signals, CH(CH3)2], 7.01
(d, 1 H, C6H4), 7.10–7.26 (m, 7 H, C6H3iPr2-2,6 and C6H4), 7.45
bond of one of the ligand arms, thereby yielding compound
4 with an aza-stiba ring. The reactivity of heterocyclic com-
pound 4 is currently being extensively studied in our labs,
as are the trapping reactions of the present Sb–H bond with
various multiple-bond-containing substrates. Similar sys-
tems are now being studied with other central atoms as
well.
Finally, distibine 5 can be prepared in low yield by re-
duction of the precursor L2SbCl with magnesium provided
that the reaction time is 3–5 h. Prolonging the reaction time (dd, 1 H, C6H4), 7.60 (d, 1 H, C6H4), 7.70 (dd, 1 H, C6H4), 7.80
2384 Eur. J. Inorg. Chem. 2011, 2380–2386
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reduction of NC-Chelated Organoantimony(III) Chlorides
(dd, 1 H, C6H4), 7.88 (d, 1 H, C6H4), 8.36 (s, 1 H, CH=N), 8.46
recrystallization from hexane was necessary, which lowered the
(s, 1 H, CH=N), 8.84 (d, 1 H, C6H4) ppm. 13C NMR (100.61 MHz, yield]. Yield 0.51 g, 39%; m.p. 224 °C, melt starts to decompose.
CDCl3, 25 °C): δ = 22.3, 24.1, 24.3, 24.6, 25.1, 25.2, 28.0, 28.2, 28.6
[s, region of CH(CH3)2 and CH(CH3)2], 123.1, 123.4, 124.1, 124.5,
126.9, 129.3, 129.7, 132.1, 132.9, 133.4, 133.5, 135.7, 138.0, 138.7
1H NMR (400 MHz, C6D6, 25 °C): δ = 0.96 [d, 6 H, CH(CH3)2],
1.02 [d, 6 H, CH(CH3)2], 2.89 [sept, 2 H, CH(CH3)2], 6.72 (dd, 1
H, C6H4), 6.96 (dd, 1 H, C6H4), 7.06 (m, 3 H, C6H3iPr2-2,6), 7.57
(overlap of two signals), 140.0, 140.3, 140.6, 144.0, 147.1, 148.4, (d, 1 H, C6H4), 7.91 (d, 1 H, C6H4), 8.39 (s, 1 H, CH=N), ppm.
155.9 (s, Ar–C), 164.9 (s, CH=N), 169.7 (s, CH=N) ppm.
C38H44ClN2Sb (685.99): calcd. C 66.5, H 6.5; found C 66.8, H 6.7.
13C NMR (100.61 MHz, C6D6, 25 °C): δ = 24.3 [s, CH(CH3)2], 24.7
[s, CH(CH3)2], 28.6 [s, CH(CH3)2], 123.6, 125.1, 128.4, 131.1, 132.4,
138.7, 141.2, 141.6, 142.3, 149.7 (s, Ar–C), 165.6 (s, CH=N) ppm.
C76H88N4Sb2 (1301.08): calcd. C 70.2, H 6.8; found C 70.5, H 5.7.
{[o-C6H4(CH=NC6H3iPr2-2,6)]Sb}4 (3): A solution of K[B(sBu)3H]
(3.7 mL, 1 m solution, 3.7 mmol) in THF was added to a stirred
solution of 1 (0.84 g, 1.84 mmol) in THF (30 mL) at room tempera-
ture. Immediately after addition, elimination of hydrogen was ob-
served and the reaction mixture turned red-brown. The resulting
mixture was stirred for an additional 1 h at room temperature and
evaporated in vacuo. The residue was washed with hexane (10 mL).
The solid residue was extracted with toluene (30 mL) to give a red-
orange solution. Evaporation of this solution gave 3 as a reddish
powder (this powder may be crystallized from toluene/hexane mix-
ture to obtain single crystals suitable for X-ray studies). Yield
320 mg, 45%; m.p. 162–165 °C. 1H NMR (400 MHz, C6D6, 25 °C):
δ = 1.01 [d, 12 H, CH(CH3)2], 3.31 [br., 2 H, CH(CH3)2], 6.32 (dd,
1 H, C6H4), 6.76 (dd, 1 H, C6H4), 6.99 (m, 3 H, C6H3iPr2-2,6),
7.04 (d, 1 H, C6H4), 8.36 (s, 1 H, CH=N), 8.48 (d, 1 H, C6H4)
ppm. 13C NMR (100.61 MHz, C6D6, 25 °C): δ = 24.8 [s, CH-
(CH3)2], 28.6 [s, CH(CH3)2], 123.5, 125.1, 127.2, 130.4, 135.1,
135.3, 139.0, 139.4, 145.5, 149.4 (s, Ar–C), 166.2 (s, CH=N) ppm.
C76H88N4Sb4 (1544.58): calcd. C 59.1, H 5.7; found C 59.2, H 6.0.
X-ray Crystallography: Suitable single crystals of the compounds
were mounted on a glass fibre with oil and measured with a Kap-
paCCD four-circle diffractometer with a CCD area detector by
monochromatized Mo-Kα radiation (λ = 0.71073 Å) at 150(1) K.
The numerical[19] absorption corrections from the crystal shape
were applied for all crystals. The structures were solved by direct
methods (SIR92)[20] and refined by a full-matrix least-squares pro-
cedure based on F2 (SHELXL97).[21] Hydrogen atoms were fixed
into idealized positions (riding model) and assigned temperature
factors Hiso(H) = 1.2Ueq (pivot atom) or of 1.5Ueq for the methyl
moiety with C–H 0.96, 0.97 and 0.93 Å for methyl, methylene and
hydrogen atoms, respectively, in the aromatic ring. There is a disor-
dered isopropyl group in the structure of 5; this disorder was solved
by splitting the methyl groups into two positions. The final differ-
ence maps displayed no peaks of chemical significance as the high-
est peaks and holes are in close vicinity (ca. 1 Å) of heavy atoms.
CCDC-804248 (for 1), -804249 (for 2), -804250 (for 3), -804251 (for
5) and -804252 (for 6·THF) contain the supplementary crystallo-
graphic 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.
[o-C6H4(CH=NC6H3iPr2-2,6)]Sb[o-C6H4CH2N(C6H3iPr2-2,6)] (4):
A solution of K[B(sBu)3H] (1.8 mL, 1 m solution, 1.8 mmol) in
THF was added to a stirred solution of 2 (1.22 g, 1.8 mmol) in
THF (40 mL) at room temperature. The resulting mixture was
stirred for an additional 30 min at room temperature and evapo-
rated in vacuo. The residue was extracted with hexane (30 mL).
The yellow extract was concentrated to around 10 mL (until the
first yellow precipitate started to emerge) and crystallization for
12 h at room temperature gave a bright yellow powder of 4. Yield
Crystallographic Data for 1: C19H22Cl2NSb, Mr = 457.03, mono-
clinic, P21/c, a = 9.6010(4) Å, b = 13.8080(8) Å, c = 16.8582(11) Å,
β = 120.854(6)°, V = 1918.6(2) Å3, Z = 4, T = 150(1) K, 19556
total reflections, 4373 independent [Rint = 0.031, R1 (obsd. data) =
0.027, wR2 (all data) 0.048].
1
0.57 g, 48%; m.p. 192–194 °C. H NMR (400 MHz, C6D6, 25 °C):
Crystallographic Data for 2: C37H44ClN2Sb·0.75(CH2Cl2), Mr
=
δ = 1.06 [d, 3 H, CH(CH3)2], 1.10 [d, 6 H, CH(CH3)2], 1.13 [d, 6
H, CH(CH3)2], 1.21 [d, 3 H, CH(CH3)2], 1.27 [d, 3 H, CH(CH3)2],
1.55 [d, 3 H, CH(CH3)2], 3.01 [sept, 1 H, CH(CH3)2], 3.23 [br., 2
H, CH(CH3)2], 4.44 [sept, 1 H, CH(CH3)2], 5.18 (AX pattern, 2 H,
NCH2), 6.94 (dd, 1 H, C6H4), 7.02–7.19 (m, 10 H, C6H4 and
C6H3iPr2-2,6), 7.28 (dd, 1 H, C6H4), 7.63 (d, 1 H, C6H4), 8.05 (s,
1 H, CH=N), 8.73 (d, 1 H, C6H4) ppm. 13C NMR (100.61 MHz,
C6D6, 25 °C): δ = 24.3, 25.3 (overlap of two signals), 25.5, 26.0,
28.9, 29.0, 29.6 [s, region of CH(CH3)2 and CH(CH3)2], 70.0 (s,
NCH2), 123.6, 123.9, 124.6, 125.5, 125.7, 126.4, 126.7, 128.7, 128.8,
131.9, 133.5, 134.7, 135.5, 139.9, 140.2, 146.9, 148.1, 149.0, 149.3,
151.8, 151.9, 153.6 (s, Ar–C), 167.8 (s, CH=N) ppm. C38H45N2Sb
(651.55): calcd. C 70.1, H 7.0; found C 70.3, H 7.3.
749.65, monoclinic, C2/c, a = 33.2431(12) Å, b = 12.9942(9) Å, c =
20.0118(7) Å, β = 118.771(9)°, V = 7581.4(3) Å3, Z = 8, T =
150(1) K, 37617 total reflections, 8555 independent [Rint = 0.036,
R1 (obsd. data) = 0.034, wR2 (all data) 0.070].
Crystallographic Data for 3: C76H88N4Sb4, Mr = 1544.50, tetrago-
nal, P4/n, a = 21.0521(12) Å, b = 21.0520(15) Å, c = 8.7760(5) Å,
V = 3889.4(4) Å3, Z = 2, T = 150(2) K, 20526 total reflections,
4453 independent [Rint = 0.040, R1 (obsd. data) = 0.035, wR2 (all
data) 0.078].
Crystallographic Data for 5: C76H88N4Sb2, Mr = 1301.00, ortho-
rhombic, Pna21, a = 15.0210(12) Å, b = 17.6420(19) Å, c =
25.424(12) Å, V = 6737.4(13) Å3, Z = 4, T = 293(2) K, 43363 total
reflections, 12370 independent [Rint = 0.059, R1 (obsd. data) =
0.041, wR2 (all data) 0.068], Flack parameter 0.028(17).
{[o-C6H4(CH=NC6H3iPr2-2,6)]2Sb}2 (5): A solution (20 mL) of 2
(1.37 g, 2 mmol) in THF was added to Mg (73 mg, 3 mmol) in
THF (10 mL) activated by BrCH2CH2Br (150 μL) at room tem-
perature. The resulting mixture was stirred until the yellow colour
changed to red-orange, typically after 3–5 h (the reaction was
stopped if the reaction mixture started to turn back to yellow-
orange and an opalescence formed) and then evaporated in vacuo.
The residue was extracted with hexane (25 mL). The red-orange
extract was concentrated to around 10 mL and crystallization for
5 h at room temperature gave orange-red crystals of 5 [in many
cases the product was contaminated (around 5–10% according to
NMR spectroscopy) by starting compound 2 and additional
Crystallographic Data for 6·THF: C42H52MgN2O, Mr = 625.17,
monoclinic, Cc, a = 14.3911(12) Å, b = 12.3382(9) Å, c =
20.4189(11) Å, β = 93.642(7)°, V = 3618.2(4) Å3, Z = 4, T =
150(1) K, 11439 total reflections, 5750 independent [Rint = 0.041,
R1 (obsd. data) = 0.058, wR2 (all data) 0.133]; Flack parameter for
6 is not reliable.[22]
Supporting Information (see footnote on the first page of this arti-
cle): 1H NMR spectra of compound 4, that is, from treatment of 2
with K[B(sBu)3H]. Molecular structure of compound 6 with rel-
evant structural parameters.
Eur. J. Inorg. Chem. 2011, 2380–2386
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2385

ˇ
L. Dostál, R. Jambor, A. Ru˚zicˇka, P. Simon
ˇ
FULL PAPER
[8] B. Twamley, D. C. Sofield, M. M. Olmstead, P. P. Power, J. Am.
Chem. Soc. 1999, 121, 3357.
Acknowledgments
[9] For example, see: a) A. H. Cowley, N. C. Norman, M. Pakul-
ski, D. L. Bricker, D. H. Russell, J. Am. Chem. Soc. 1985, 107,
8211; b) A. M. Arif, A. H. Cowley, N. C. Norman, M. Pakul-
ski, Inorg. Chem. 1986, 25, 4836; c) G. Huttner, U. Weber, O.
Scheidsteger, L. Zsolnai, Angew. Chem. Int. Ed. Engl. 1982, 21,
215; d) K. Plossl, G. Huttner, L. Zsolnai, Angew. Chem. Int.
Ed. Engl. 1989, 28, 446; e) G. Balázs, H. J. Breunig, E. Lork,
Z. Anorg. Allg. Chem. 2001, 627, 1855.
[10] a) L. Balázs, H. J. Breunig, E. Lork, C. Silvestru, Eur. J. Inorg.
Chem. 2003, 1361; b) L. M. Opris, A. Silvestru, C. Silvestru,
H. J. Breunig, E. Lork, Dalton Trans. 2004, 3575.
[11] L. Balázs, H. J. Breunig, E. Lork, A. Soran, C. Silvestru, Inorg.
Chem. 2006, 45, 2341.
The authors would like to thank the Grant Agency of the Czech
Republic (P207/10/0130) and the Ministry of Education of the
Czech Republic (MSM 0021627501) for financial support.
[1]
[2]
K. Issleib, B. Hamann, L. Schmidt, Z. Anorg. Allg. Chem.
1965, 339, 289.
For review articles, see: a) L. Balázs, H. J. Breunig, Coord.
Chem. Rev. 2004, 248, 603; b) H. J. Breunig, R. Rössler, Chem.
Soc. Rev. 2000, 29, 403; c) N. Tokitoh, J. Organomet. Chem.
2000, 611, 217; d) P. P. Power, Chem. Rev. 1999, 99, 3463; e)
P. P. Powe r, Chem. Rev. 2010, 110, 3877; f) B. D. Ellis, C. L. B.
Macdonald, Coord. Chem. Rev. 2007, 251, 936.
[12] For an example of CNC-chelated low-valent bismuth com-
pounds, see: S. Shimada, J. Maruyama, Y. K. Choe, T. Yamash-
ita, Chem. Commun. 2009, 41, 6168.
[3]
For example, see: a) H. J. Breunig, M. Denker, K. H. Ebert, J.
Organomet. Chem. 1994, 470, 87; b) H. J. Breunig, M. Denker,
K. Häberle, M. Dräger, T. Severngiz, Angew. Chem. Int. Ed.
Engl. 1985, 24, 72; c) O. Mundt, G. Becker, H. J. Wessely, H. J.
Breunig, H. Kischkel, Z. Anorg. Allg. Chem. 1982, 486, 70; d)
G. Balázs, H. J. Breunig, E. Lork, Organometallics 2003, 22,
2919; e) M. Ates, H. J. Breunig, S. Gülec, W. Offermann, K.
Häberle, M. Dräger, Chem. Ber. 1989, 122, 473; f) H. J. Breu-
nig, R. Rössler, E. Lork, Angew. Chem. Int. Ed. 1998, 37, 3175;
g) L. Balázs, H. J. Breunig, E. Lork, Angew. Chem. Int. Ed.
2002, 41, 2309; h) G. Linti, W. Köstler, Z. Anorg. Allg. Chem.
2002, 628, 65; i) G. Linti, W. Köstler, H. Pritzkow, Eur. J. Inorg.
Chem. 2002, 2643.
[13] L. Dostál, R. Jambor, A. Ru˚zicˇka, J. Holecˇek, Organometallics
ˇ
2008, 27, 2169.
ˇ
[14] P. Simon, F. de Proft, R. Jambor, A. Ru˚zicˇka, L. Dostál, An-
ˇ
gew. Chem. Int. Ed. 2010, 49, 5468.
[15] For examples, see: a) R. Waterman, T. D. Tilley, Angew. Chem.
Int. Ed. 2006, 45, 2926; b) E. Rivard, J. Steiner, J. C. Fettinger,
J. R. Guiliani, M. P. Augustine, P. P. Power, Chem. Commun.
2007, 4919.
[16] D. Zhao, W. Gao, Y. Mu, L. Ye, Chem. Eur. J. 2010, 16, 4394.
[17] L. M. Opris, A. Silvestru, C. Silvestru, H. J. Breunig, E. Lork,
Dalton Trans. 2004, 4367.
[18] K. V. Deuten, D. Rehder, Cryst. Struct. Commun. 1980, 9, 167.
[19] P. Coppens, in Crystallographic Computing (Eds.: F. R Ahmed,
S. R. Hall, C. P. Huber), Copenhagen, Munksgaard, 1970, pp.
255–270.
[4] a) G. Balázs, H. J. Breunig, E. Lork, S. A. Mason, Organome-
tallics 2003, 22, 576; b) G. Balázs, H. J. Breunig, E. Lork, Z.
Anorg. Allg. Chem. 2003, 629, 637.
[5] a) G. Balázs, H. J. Breunig, E. Lork, W. Offermann, Organome-
tallics 2001, 20, 2666; b) O. Mundt, H. Riffel, G. Becker, A.
Simon, Z. Naturforsch., Teil B 1984, 39, 317; c) A. J. Ashe III,
E. G. Ludwig, J. Oleksyszyn, J. C. Hoffman, Organometallics
1984, 3, 337; d) G. Balázs, H. J. Breunig, E. Lork, Organome-
tallics 2001, 20, 2584.
[6] N. Tokitoh, Y. Arai, R. Okazaki, S. Nagase, Science 1997, 277,
78.
[7] N. Tokitoh, Y. Arai, T. Sasamori, R. Okazaki, S. Nagase, H.
Uekusa, Y. Ohashi, J. Am. Chem. Soc. 1998, 120, 433.
[20] A. Altomare, G. Cascarone, C. Giacovazzo, A. Guagliardi,
M. C. Burla, G. Polidori, M. J. Camalli, Appl. Crystallogr.
1994, 27, 1045.
[21] G. M. Sheldrick, SHELXL-97, A Program for Crystal Struc-
ture Refinement, University of Göttingen, Germany, 1997.
[22] a) H. D. Flack, G. Bernardinelli, Acta Crystallogr., Sect. A
1999, 55, 908–915; b) H. D. Flack, G. Bernardinelli, J. Appl.
Crystallogr. 2000, 33, 1143–1148.
Received: January 5, 2011
Published Online: April 14, 2011
2386
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 2380–2386