Bis(1°-amino)cyclodistib(III)azanes: The first structural characterization of cis and trans isomers of a single cyclodipnict(III)azane

Article abstract of DOI:10.1021/ic061660t

The dichlorocyclodistib(III)azane [CISb(μ-N^tBu)]₂ (1) has been shown to exist as the cis isomer in the solid state. A series of bis(1°-amino)cyclodistib(III)azanes [R′NHSb(μ-N^tBu)] ₂ (2, R′ = ^tBu; 3, R′ = Dipp; 4, R′ = Dmp) has been prepared by the reaction of 1 with 2 equiv. of LiNHR′. On the basis of NMR solution spectra, all three derivatives are formed as a mixture of cis and trans isomers. In the case of 3, the structures of both the cis and trans isomers have been determined by X-ray crystallography; cis-3 adopts an endo, endo arrangement for the amido protons of the DippNH groups. Isomerization of trans-3 into cis-3 occurs slowly in solution. Deprotonation of 2 with 2 equiv. of ⁿBuNa or trans-3 with ⁿBuLi produces [Na ₂Sb₂(μ-N^tBu)₄] (5) and [Li ₂Sb₂(μ-N^tBu)₂(μ-NDipp) ₂] (6), whose solvated cubane structures were established by X-ray crystallography. In contrast, the reaction of cis-3 with 2 equiv. of ⁿBuLi produces the tricyclic compound [Li₂Sb(μ-N ^tBu)₂(μ-NDipp)(μ-NHDipp)] (7).

Full text of DOI:10.1021/ic061660t

Inorg. Chem. 2006, 45, 10734−10742
Bis(1°-amino)cyclodistib(III)azanes: the First Structural Characterization
of cis and trans Isomers of a Single Cyclodipnict(III)azane
Dana J. Eisler and Tristram Chivers*
Department of Chemistry, UniVersity of Calgary, Calgary, Alberta, Canada T2N 1N4
Received September 1, 2006
The dichlorocyclodistib(III)azane [ClSb(
A series of bis(1 -amino)cyclodistib(III)azanes [R
been prepared by the reaction of 1 with 2 equiv. of LiNHR
µ
-NtBu)]2 (1) has been shown to exist as the cis isomer in the solid state.
NHSb(
-NtBu)]2 (2, R′ ) tBu; 3, R′ ) Dipp; 4, R′ ) Dmp) has
. On the basis of NMR solution spectra, all three
°
′
µ
′
derivatives are formed as a mixture of cis and trans isomers. In the case of 3, the structures of both the cis and
trans isomers have been determined by X-ray crystallography; cis-3 adopts an endo, endo arrangement for the
amido protons of the DippNH groups. Isomerization of trans-3 into cis-3 occurs slowly in solution. Deprotonation
of 2 with 2 equiv. of nBuNa or trans-3 with nBuLi produces [Na2Sb2(
(6), whose solvated cubane structures were established by X-ray crystallography. In contrast, the reaction of cis-3
with 2 equiv. of nBuLi produces the tricyclic compound [Li2Sb( -NtBu)2(
-NDipp)( -NHDipp)] (7).
µ µ µ-NDipp)2]
-NtBu)4] (5) and [Li2Sb2( -NtBu)2(
µ
µ
µ
Introduction
reported,5 and it has been shown that group 4 metal
complexes of this type can function as catalysts in the
polymerization of ethene.6 One intriguing new development
in the chemistry of cyclodipnict(III)azanes has evolved from
the work of Wright et al., who have used cyclodiphosph-
(III)azanes to prepare the first examples of a new class of
inorganic macrocycles.2,7
Whereas cyclodiphosph(III)azanes have been extensively
studied, the chemistry of the heavier pnictogen (As, Sb, Bi)
analogues has been slower to develop.1,3 This is in large part
due to the difficulty in the systematic preparation of the
Cyclodipnict(III)azanes of the type [XE(µ-NR)]2 (E ) P,
As, Sb, Bi) are well-known inorganic heterocycles. The
chemistry of these four-membered rings has attracted sus-
tained interest for more than 40 years, and highlights of this
research area have been well-summarized in several recent
review articles.1-3 Much of the current research on these
systems has been directed toward the synthesis of anionic
-2
complexes of the type [RNE(µ-NR)]2 and the bis(1°-
amino)cyclodipnict(III)azanes, from which the anions are
easily derived.1-3 Expectedly, neutral cyclodiphosph(III)-
azanes, which have two available phosphorus donors, have
been used to prepare coordination complexes with transition
metals.4 In addition, coordination complexes utilizing the
nitrogen centers of the [RNP(µ-NR)]2-2 dianions have been
(5) (a) Grocholl, L.; Stahl, L.; Staples, R. J. Chem. Commun. 1997, 1465.
(b) Moser, D. F.; Carrow, C. J.; Stahl, L.; Staples, R. J. J. Chem.
Soc., Dalton Trans. 2001, 1246.
(6) (a) Axenov, K. V.; Kotoy, V.; Klinga, M.; Leskela¨, M.; Repo, T. Eur.
J. Inorg. Chem. 2004, 695. (b) Axenov, K. V.; Klinga, M.; Leskela¨,
M.; Kotoy V.; Repo, T. Eur. J. Inorg. Chem. 2004, 4702. (c) Axenov,
K. V.; Klinga, M.; Leskela¨, M.; Repo, T. Organometallics 2005, 24,
1336. (d) Axenov, K. V.; Kilpelainen, I.; Klinga, M.; Leskela¨, M.;
Repo, T. Organometallics 2006, 25, 463. (e) Lief, G. R.; Carrow, C.
J.; Stahl, L.; Staples, R. J. Chem. Commun. 2001, 1562.
* To whom correspondence should be addressed. Telephone: (403) 220-
5741. Fax: (403) 289-9488. E-mail: chivers@ucalgary.ca.
(1) Stahl, L. Coord. Chem. ReV. 2000, 210, 203.
(7) (a) Garcia, F.; Goodman, J. M.; Kowenicki, R. A.; McPartlin, M.;
Riera, L.; Silva, M. A.; Wirsing, A.; Wright, D. S. Dalton Trans. 2005,
1764. (b) Dodds, F.; Garcia, F.; Kowenicki, R. A.; McPartlin, M.;
Steiner, A.; Wright, D. S. Chem. Commun. 2005, 3733. (c) Dodds,
F.; Garcia, F.; Kowenicki, R. A.; McPartlin, M.; Riera, L.; Steiner,
A.; Wright, D. S. Chem. Commun. 2005, 5041. (d) Bashall, A.; Doyle,
E. L.; Tubb, C.; Kidd, S. J.; McPartlin, M.; Woods, A. D.; Wright, D.
S. Chem. Commun. 2001, 2542. (e) Bashall, A.; Bond, A. D.; Doyle,
E. L.; Garcia, F.; Kidd, S.; Lawson, G. T.; Parry, M. C.; McPartlin,
M.; Woods, A. D.; Wright, D. S. Chem.sEur. J. 2002, 8, 3377. (f)
Garcia, F.; Goodman, J. M.; Kowenicki, R. A.; Kuzu, I.; McPartlin,
M.; Silva, M. A.; Riera, L.; Woods, A. D.; Wright, D. S. Chem.s
Eur. J. 2004, 10, 6066.
(2) Doyle, E. L.; Riera, L.; Wright, D. S. Eur. J. Inorg. Chem. 2003,
3279.
(3) Balakrishna, M. S.; Chivers, T.; Eisler, D. J. Chem. Soc. ReV. 2007,
DOI: 10.1039/b514861h.
(4) For recent examples see: (a) Chandrasekaran, P.; Mague, J. T.;
Balakrishna, M. S. Inorg. Chem. 2006, 45, 6678. (b) Chandrasekaran,
P.; Mague, J. T.; Balakrishna, M. S. Inorg. Chem. 2005, 44, 7925. (c)
Schranz, I.; Lief, G. R.; Carrow, C. J.; Haagenson, D. C.; Grocholl,
L.; Stahl, L.; Staples, R. J.; Boomishankar, R.; Steiner, A. Dalton
Trans. 2005, 3307.
10734 Inorganic Chemistry, Vol. 45, No. 26, 2006
10.1021/ic061660t CCC: $33.50
© 2006 American Chemical Society
Published on Web 11/15/2006

Bis(1°-amino)cyclodistib(III)azanes
heavier cyclodipnict(III)azanes, because the synthetic meth-
ods available for producing cyclodiphosph(III)azanes tend
to give more disparate results when applied to the heavier
elements.1,3 Thus, the reaction of SbCl3 with LiNHDmp
(Dmp ) 2,6-dimethylphenyl) produces the bis(1°-amino)-
cyclodistib(III)azane [DmpHNE(µ-NDmp)]2, whereas the
analogous reaction using LiNPh generates the macrocycle
Sb12(NPh)18.8 The difficulty in preparing these heterocycles
is evidenced by the rarity of known examples; there are only
two reported bis(1°-amino)cyclodistib(III)azanes, and [Dip-
pHNBi(µ-NDipp)]2 (Dipp ) 2,6 diisopropylphenyl) is the
sole bismuth analogue.9,10 The related antimony dianion
-2
Figure 1. View of one of the molecules of cis-1. Hydrogen atoms have
been removed for clarity.
[CyNSb(µ-NCy)]2 (Cy ) cyclohexyl) is not synthesized
directly from [CyNHSb(µ-NCy)]2 but is prepared in situ by
the reaction of [Me2NSb(µ-NCy)]2 with LiNHCy, for
-2
example.11
Scheme 1. Synthesis of Bis(1°-amino)cyclodistib(III)azanes
In an effort to devise a cogent synthetic method for the
generation of bis(1°-amino)cyclodistib(III)azanes, we sur-
mised that the reaction of the known reagent [ClSb(µ-Nt-
Bu)]2 with lithium amides would provide a suitable starting
point.12 In this contribution, we report a series of bis(1°-
amino)cyclodistib(III)azanes generated by the successful
application of this synthetic method. The compounds are
shown to exist as a mixture of cis and trans isomers in
solution. In the case of one derivative, the structures of both
isomers in the solid state were established. Investigations of
provide clarity on this subject, we have undertaken a
crystallographic study on 1 that revealed that the compound
adopts a cis configuration in the solid state (Figure 1).
Compound 1 crystallizes with three independent but chemi-
cally equivalent molecules present in the asymmetric unit.
The molecules are nearly identical with the exception of
the puckering of the central Sb2N2 rings. Two of the
molecules present in the asymmetric unit of 1 are noticeably
puckered, with the angles between N-Sb-N planes being
168 and 170°. However, the third molecule is essentially
planar, with an analogous angle of 177°. It has been
previously shown that cis-cyclodipnict(III)azanes have
puckered E2N2 rings, whereas the rings in the trans isomers
are planar.1 Regardless of the differences in the ring
puckering, the molecules all have similar geometric
parameters (Table 1). The Sb-N distances are equal within
error, ranging from 2.009(7) to 2.022(7) Å. Not surprisingly,
the endocyclic N-Sb-N angles are more acute (77.9(3)-
78.6(3)°) than the exoxyclic N-Sb-Cl angles (97.5(2)-
100.4(2)°).
Synthesis and Structures of Bis(1°-amino)cyclodistib-
(III)azanes. The reaction of 2 equiv. of lithium amide with
1 readily produces the bis(1°-amino)cyclodistib(III)azanes
2-4 as outlined in Scheme 1. In each case, the products are
isolated as a mixture of cis and trans isomers, with the cis
isomer being favored in solution. The cis:trans isomer ratio
that is initially isolated varies for each compound (2, 80:20;
3, 60:40; 4, 85:15). Interestingly, in the case of 3, it was
possible to separate the cis and trans isomers because of the
large difference in their solubilities; trans-3, which is isolated
as a yellow solid, is only sparingly soluble in n-hexane,
whereas cis-3 is freely soluble. Furthermore, cis-3 is initially
obtained as a viscous yellow oil that slowly crystallizes upon
standing at -20 °C. Both compounds 2 and 4 are obtained
as thick oils, presumably as a result of the high percentage
n
the metalation of these cis and trans isomers with BuLi
revealed an unanticipated difference in the type of product
that is obtained.
Results and Discussion
X-ray Structure of [ClSb(µ-NtBu)]2 (1). The cyclodistib-
(III)azane [ClSb(µ-NtBu)]2 (1) was first prepared in 1979
via reaction of SbCl3 with LiN(SiMe3)tBu.12 More recently,
Stahl has reported a convenient one-pot synthesis, albeit with
slightly lower yields.13 Although the structure of 1 has not
been previously determined, it was originally assigned to be
cis on the basis of the known structure of the phosphorus
analogue.12 Subsequent work demonstrated that the arsenic
analogue also adopts a cis configuration in the solid state.14
More recently, several derivatives of 1 have been shown to
exist as the trans isomers in the solid state, which led to the
suggestion that, in contrast to the phosphorus and arsenic
compounds, 1 might also have a trans configuration.13 To
(8) Bryant, R.; James, S. C.; Jeffery, J. C.; Norman, N. C.; Orpen, A. G.;
Weckenmann, U. J. Chem. Soc., Dalton Trans. 2000, 4007.
(9) Burford, N.; Cameron, T. S.; Lam, K. C.; LeBlanc, D. J.; McDonald,
C. L. B.; Philips, A. D.; Rheingold, A. L.; Stark, L.; Walsh, D. Can.
J. Chem. 2001, 79, 342.
(10) Wirringa, U.; Roesky, H. W.; Noltemeyer, M.; Schmidt, H. G. Inorg.
Chem. 1994, 33, 4607.
(11) (a) Barr, D.; Edwards, A. J.; Pullen, S.; Paver, M. A.; Raithby, P. R.;
Rennie, M. A.; Russell, C. A.; Wright, D. S. Angew. Chem., Int. Ed.
1994, 33, 1875. (b) Beswick, M. A.; Cromhout, N. L.; Harmer, C.
N.; Paver, M. A.; Raithby, P. R.; Rennie, M. A.; Steiner, A.; Wright,
D. S. Inorg. Chem. 1997, 36, 1740. (c) Bashall, A.; Beswick, M. A.;
Ehlenberg, H.; Kidd, S. J.; McPartlin, M.; Palmer, J. S.; Raithby, P.
R.; Rawson, J. M.; Wright, D. S. Chem. Commun. 2000, 749.
(12) Kuhn, N.; Scherer, O. J. Z. Naturforsch. 1979, 34b, 888.
(13) Haagenson, D. C.; Stahl, L. Inorg. Chem. 2001, 40, 4491.
(14) Bohra, R.; Roesky, H. W.; Noltemeyer, M.; Sheldrick, G. M. Acta
Crystallogr., Sect. C 1984, 40, 1150.
Inorganic Chemistry, Vol. 45, No. 26, 2006 10735

Eisler and Chivers
Table 1. Selected Bond Distances (Å) and Angles (deg) for cis-1
molecule 1
molecule 2
molecule 3
Sb1-N1
2.013(7)
2.017(7)
2.022(6)
2.009(7)
2.445(3)
2.450(3)
77.9(3)
99.9(2)
99.7(2)
77.9(3)
99.8(2)
Sb3-N3
2.014(7)
2.017(6)
2.022(7)
2.016(6)
2.456(7)
2.440(3)
78.6(3)
99.4(3)
97.7(3)
78.4(3)
100.4(2)
98.1(2)
Sb5-N5
2.014(5)
2.014(5)
2.018(6)
2.020(6)
2.449(3)
2.434(3)
78.1(3)
98.9(2)
97.5(2)
77.9(3)
97.7(2)
Sb1-N2
Sb3-N4
Sb5-N6
Sb2-N1
Sb4-N3
Sb6-N5
Sb2-N2
Sb4-N4
Sb6-N6
Sb1-Cl1
Sb3-Cl3
Sb5-Cl5
Sb2-Cl2
Sb4-Cl4
Sb6-Cl6
N1-Sb1-N2
N1-Sb1-Cl1
N2-Sb1-Cl1
N1-Sb2-N2
N1-Sb2-Cl2
N2-Sb2-Cl2
N3-Sb3-N4
N3-Sb3-Cl3
N4-Sb3-Cl3
N3-Sb4-N4
N3-Sb4-Cl4
N4-Sb4-Cl4
N5-Sb5-N6
N5-Sb5-Cl5
N6-Sb5-Cl5
N5-Sb6-N6
N5-Sb6-Cl6
N6-Sb6-Cl6
100.0(2)
98.0(2)
Figure 3. View of the structure of cis-4. Only the amido hydrogen atoms
are shown.
of 161 and 163°, respectively. Presumably, the increased
puckering observed in cis-3 and cis-4 in comparison to that
in cis-1 is due to the large increase in steric bulk arising
from the introduction of the Dipp and Dmp substituents.
There is little difference in the bond distances and angles
observed for cis-3, trans-3, and cis-4 (Table 2). Even the
imido Sb-N and amido Sb-N distances show little variation,
falling in the narrow range of 2.022(4)-2.067(7) Å. Similar
to cis-1, the endocyclic N-Sb-N angles (range 77.5(3)-
78.1(1)°) are in each case more acute than the exocyclic
N-Sb-N angles (range 91.5(3)-99.6(1)°). For both of the
cis isomers, the amido protons of the (H)NR substituents
adopt an endo, endo configuration.
Solution NMR Studies of cis and trans Isomerism.
Solution-state cis and trans isomerism has been well-
established for cyclodiphosph(III)azanes, and it is possible
to distinguish between the isomers from their markedly
different 31P NMR chemical shifts.15 Unfortunately, there are
no isotopes with I ) 1/2 for the heavier pnictogens, and
consequently, NMR studies on heavier cyclodipnict(III)-
azanes are limited. A solution-state equilibrium between cis
and trans isomers has been observed for a series of
cyclodistib(III)azanes [RSb(µ-NtBu)]2 (R ) alkyl, alkoxy,
aryloxy, and silyl).16 Furthermore, the possibility of solution-
state cis and trans isomerism has been considered for bis-
(amino)cyclodistib(III)azanes, although definitive NMR data
Figure 2. (top) View of the structure of trans-3. (bottom) View of
the structure of cis-3. In each case, only the amido hydrogen atoms are
shown.
of the cis isomer present in these samples. All attempts to
separate the cis and trans isomers of 2 and 4 were unsuc-
cessful. The identities of the two isomers cis-3 and trans-3
were unambiguously established by an X-ray structural
determination of each isomer, as depicted in Figure 2. The
structure of cis-4 was also determined crystallographically
and is shown in Figure 3.
The central Sb2N2 ring of trans-3, which has a crystallo-
graphically imposed center of symmetry, is completely
planar, as found for other structurally characterized cyclo-
dipnict(III)azanes that have a trans configuration.1,3 In
contrast, there is considerable puckering of the Sb2N2 ring
in cis-3 and cis-4, with angles between the N-Sb-N planes
(15) Kumaravel, S. S.; Krishnamurthy, S. S.; Cameron, T. S.; Linden, A.
Inorg. Chem. 1988, 27, 4546.
(16) Ross, B.; Belz, J.; Nieger, M. Chem. Ber. 1990, 123, 975.
10736 Inorganic Chemistry, Vol. 45, No. 26, 2006

Bis(1°-amino)cyclodistib(III)azanes
Table 2. Selected Bond Distances (Å) and Angles (deg) for trans-3, cis-3, and cis-4
trans-3
cis-3
cis-4
Sb1-N1
Sb1-N1A
Sb1-N2
N1-Sb1-N1A
N1-Sb1-N2
N1A-Sb1-N2
2.022(4)
2.039(3)
2.058(3)
77.9(2)
95.7(1)
99.6(1)
Sb1-N1
2.027(3)
2.050(3)
2.058(3)
78.1(1)
96.1(1)
98.7(1)
Sb2-N2
2.031(3)
2.045(3)
2.052(3)
78.1(1)
98.5(1)
93.3(1)
Sb1-N1
Sb1-N1A
Sb1-N2
N1-Sb1-N1A
N1-Sb1-N2
N1A-Sb1-N2
2.027(7)
2.047(7)
2.067(7)
77.5(3)
91.5(3)
98.1(3)
Sb1-N2
Sb2-N1
Sb1-N3
Sb2-N4
N1-Sb1-N2
N1-Sb1-N3
N2-Sb1-N3
N1-Sb2-N2
N1-Sb2-N4
N2-Sb2-N4
are lacking.8,17 Solution NMR studies on the series of
cyclodistib(III)azanes 2-4 clearly demonstrated the presence
of two compounds, which had very similar NMR spectra in
each case. 1H-13C HMBC-correlated NMR spectra obtained
for 3 and 4 indicated that, in each case, the amido protons
were associated with the aromatic substituents, thus ruling
out the possibility of the rearrangement isomers [Sb(µ-NAr)-
(NHtBu)]2. The presence of cis and trans isomers was
unambiguously established in the case of 3 by crystal-
lographic studies, as discussed previously.
The ability to separate the two isomers of 3 by physical
methods allowed for the collection of NMR data for each
1
isomer separately. Although very similar H NMR spectra
are obtained for the two isomers, there is a significant
difference in the chemical shifts observed for the amido
protons of cis-3 (δ ) 5.01) and trans-3 (δ ) 5.23). The
difference in shifts can be attributed to the endo, endo
configuration of the amido protons in cis-3 (Figure 2), which
places them in a more shielded environment than that in
Figure 4. 1H NMR spectra showing the transformation of trans-3 to cis-3
in solution. Only the regions containing the amido N-H and methine protons
of the isopropyl groups are shown for clarity.
observed. Although it has not been possible to unambigu-
ously establish the presence of cis and trans isomers for 2
and 4, given the similarities in their physical properties and
spectroscopic data, it seems the most probable explanation.
Unfortunately, all attempts to grow crystals of compound 2
suitable for X-ray diffraction studies were unsuccessful.
Several different mechanisms have been proposed for the
interconversion of cis and trans isomers of cyclodiphosp-
hazanes.1,18 In particular, pyramidal inversion has been
considered as the most likely mechanism for the phosphorus
case.1 For the heavier elements, however, pyramidal inversion
is a considerably higher energy process,18a and so it is highly
unlikely that this mechanism is operating at room temperature
for the compounds 2-4. The possibility of complete dis-
sociation into iminopnictane monomers has also been
considered.1,18b However, whereas this has been demonstrated
to occur for both phosphorus and arsenic, such a dissociation
is unknown for antimony.18c To probe the possibility that
this mechanism was acting in the cis and trans interconver-
sion for the compounds 2-4, we monitored equal mixtures
1
trans-3; similar differences are observed in the H NMR
spectra of cis- and trans-4. For compound 2, the difference
in the chemical shifts is considerably smaller (cis, δ ) 2.65;
trans, δ ) 2.69), although this is not surprising, as the tert-
butyl amido substituents would not be as shielding as the
aromatic rings of the Dipp and Dmp groups.
Initial NMR investigations on the isomer trans-3 indicated
the presence of small amounts of cis-3, and all attempts to
obtain pure samples of trans-3 were unsuccessful. Further
NMR studies revealed that, in fact, trans-3 isomerizes to cis-3
in solution, over the course of ca. 10 days. This transforma-
1
tion can be easily followed by H NMR spectroscopy, as
shown in Figure 4. It should be noted that during this long
time period, some decomposition of 3 occurs in solution.
The solution NMR data for compound 3 clearly indicate that
the trans isomer is the kinetic product, whereas the cis isomer
is the thermodynamic product.
Similar long-term NMR solution studies were performed
on 2 and 4, and it was found that the initially observed cis:
trans ratios do not change. During this work, it was noted
that thick oily samples of 2 stored at -20 °C slowly
transform to a mixture of an oil and a yellow solid. NMR
spectra obtained on these semisolid samples revealed an
increase in the quantity of the trans isomer (up to ca. 50%).
However, within 48 h, the original cis:trans ratio of 80:20 is
1
of 2 and 3 by H NMR, and no evidence of a new mixed
species was observed; only intact 2 and 3 were present even
after one week in solution. A further conceivable possibility
is a partial opening of the central Sb2N2 ring via cleavage
of an Sb-N bond that allows for rotation around the geminal
Sb-N bond and results in the formation of the other isomer
upon ring closure.1 This latter mechanism would appear to
be the most likely possibility in the current systems.
(17) (a) Beswick, M. A.; Wright, D. S. Coord. Chem. ReV. 1998, 176, 373.
(b) Beswick, M. A.; Harmer, C. N.; Hopkins, A. D.; Paver, M. A.;
Raithby, P. R.; Wright, D. S. Polyhedron 1998, 17, 745. (c) Edwards,
A. J.; Leadbeater, N. E.; Paver, M. A.; Raithby, P. R.; Russell, C. A.;
Wright, D. S. J. Chem. Soc., Dalton Trans. 1994, 1479. (d) Edwards,
A. J.; Paver, M. A.; Rennie, M. A.; Raithby, P. R.; Russell, C. A.;
Wright, D. S. J. Chem. Soc., Dalton Trans. 1994, 2963.
(18) (a) Senkler, G. H., Jr.; Mislow, K. J. Am. Chem. Soc. 1972, 94, 291.
(b) Schranz, I.; Moser, D. F.; Stahl, L.; Staples, R. J. Inorg. Chem.
1999, 38, 5814. (c) Burford, N.; Cameron, T. S.; Macdonald, C. L.
B.; Robertson, K. N.; Schurko, R.; Walsh, D. Inorg. Chem. 2005, 44,
8058. (d) Silaghi-Dumitrescy, I.; Haiduc, I. Phosphorus, Sulfur Silicon
Relat. Elem. 1994, 91, 21.
Inorganic Chemistry, Vol. 45, No. 26, 2006 10737

Eisler and Chivers
Scheme 2. Metalation of trans-3 and cis-3 with Butyl-lithium
Figure 5. View of the structure of complex 5. Hydrogen atoms have been
removed for clarity.
[P2(NtBu)4Li2(THF)2] and [{P2(NtBu)4}2Li4].21,22 The solid-
structure structure of complex 5 contains two monosolvated
sodium ions, whereas complex 6 contains only a single
molecule of THF coordinated to one of the lithium centers
(Li1). The second lithium center is too sterically shielded
by the bulky Dipp substituents to allow for coordination of
a second THF molecule. Furthermore, there are several
agostic interactions between Li2 and the surrounding hy-
drogen atoms (range 2.17(1)-2.23(1) Å) as a result of this
steric congestion. In both 5 and 6 it can be assumed that
solvation prevents further aggregation of the cubane struc-
tures.23
Both of the complexes 5 and 6 adopt a distorted cubane
structure, and there is considerable variation of the Sb-N
bond distances (Table 3). In particular, the Sb-N distances
of the Sb2N2 rings are considerably longer (2.088(4)-2.096-
(3) Å) than those in the SbMN2 (M ) Li, Na) rings [1.978-
(5)-2.015(4) Å]. In addition, direct comparison can be made
in the case of 6 with cis-3, in which the analogous distances
ranged from 2.027(3)-2.050(3) Å. Presumably, the differ-
ences in the distances observed in the cluster complexes 5
and 6 are in part a reflection of the higher ionic character of
the bonding resulting from the generation of the [RNSb(µ-
NtBu)]22- dianion. Furthermore, there is clearly an increase
in steric crowding in 6 that may also contribute to the
observed bond lengthening. As was observed in the structures
of 3 and 4, the endocyclic N-Sb-N angles of the Sb2N2
rings are more acute than the exocyclic N-Sb-N angles in
5 and 6.
Figure 6. View of the structure of complex 6. Hydrogen atoms have been
removed for clarity.
Deprotonation of bis(1°-amino)cyclodistib(III)azanes.
Bis(1°-amino)cyclodistib(III)azanes are obvious precursors
2-
for the formation of dianions of the type [RNSb(µ-NR)]2
.
Indeed, the reaction of 2 or 3 with butyl-sodium or butyl-
lithium, respectively, readily produces the heterometallic
cubane complexes [Na2Sb2(µ-NtBu)4] (5) and [Li2Sb2(µ-Nt-
Bu)2(µ-NDipp)2] (6), which both contain dianions of the type
[RNSb(µ-NtBu)]2 (Figures 5 and 6). The cyclohexyl
2-
derivative [CyNSb(µ-NCy)]22- is the only other example of
an antimony-containing dianion of this type that has been
reported.11,19 The observed solid-state structures of complexes
5 and 6 are analogous to that of the solvated bismuth complex
[Bi2(NCy)4Li2(THF)2].20 In contrast, the unsolvated antimony
complex [{Sb2(NCy)4}2Na4] is tetrameric as a result of the
face-to-face dimerization of two [Sb(NCy)4Na2] cubanes.19
A similar effect of alkali-metal solvation is evident in the
structures of the analogous phosphorus-containing complexes
Comparison of the Reactions of cis- and trans-3 with
nBuLi. The ability to separate the isomers cis-3 and trans-3
provided the intriguing opportunity to investigate the chem-
istry of the two isomers independently. The reaction of two
equivalents of n-butyl-lithium with the two isomers produced
some surprising results. We had initially surmised that the
(19) (a) Alton, R. A.; Barr, D.; Edwards, A. J.; Paver, M. A.; Raithby, P.
R.; Rennie, M. A.; Russell, C. A.; Wright, D. S. J. Chem. Soc., Chem.
Commun. 1994, 1481. (b) Bashall, A.; Beswick, M. A.; Harmer, C.
N.; Hopkins, A. D.; McPartlin, M.; Paver, M. A.; Raithby, P. R.;
Wright, D. S. J. Chem. Soc., Dalton Trans. 1998, 1389.
(21) Schranz, I.; Stahl, L.; Staples, R. Inorg. Chem. 1998, 37, 1493.
(22) Brask, J. K.; Chivers, T.; Krahn, M. L.; Parvez, M. Inorg. Chem. 1999,
38, 290.
(20) Edwards, A. J.; Beswick, M. A.; Galsworthy, J. R.; Paver, M. A.;
Raithby, P. R.; Rennie, M. A.; Russell, C. A.; Verhorevoort, K. L.;
Wright, D. S. Inorg. Chim. Acta 1996, 248, 9.
(23) For a discussion of the influence of solvation on the structures of alkali-
metal derivatives of polyimido anions of p-block elements, see: Brask,
J. K.; Chivers, T. Angew. Chem., Int. Ed. 2001, 40, 3960.
10738 Inorganic Chemistry, Vol. 45, No. 26, 2006

Bis(1°-amino)cyclodistib(III)azanes
Table 3. Selected Bond Distances (Å) and Angles (deg) for Complexes 5 and 6
complex 5
complex 6
Sb1-N1
2.095(3)
2.096(3)
1.978(5)
80.2(2)
94.0(1)
94.6(1)
2.445(3)
2.421(3)
2.419(3)
Sb1-N1
2.119(4)
2.089(4)
2.015(4)
80.4(2)
91.7(2)
88.9(2)
2.154(9)
2.101(10)
2.082(9)
Sb2-N1
2.088(4)
2.095(4)
2.002(4)
81.0(2)
90.6(2)
90.2(2)
2.084(10)
2.008(10)
2.063(10)
Sb2-N1
Sb1-N2
Sb2-N2
Sb1-N2a
Sb1-N3
Sb2-N4
N1-Sb1-N1Ab
N1-Sb1-N2
N1-Sb2-N3
Na1-N1
N1-Sb1-N2
N1-Sb1-N3
N2-Sb1-N3
Li1-N1
N1-Sb2-N2
N1-Sb2-N4
N2-Sb2-N4
Li2-N2
Na1-N2
Na1-N3
Li1-N3
Li1-N4
Li2-N3
Li2-N4
a Sb2-N3 ) 1.988(5). b N1-Sb2-N1A ) 80.1(2).
Table 4. Selected Bond Distances (Å) and Angles (deg) for Complex 7
molecule 1 molecule 2
Sb1-N1
Sb1-N2
Sb1-N3
Li1-N1
Li1-N2
Li1-N4
Li2-N3
Li2-N4
2.093(4)
2.097(4)
1.970(4)
2.14(1)
2.13(1)
2.03(1)
1.92(1)
2.00(1)
88.2(2)
91.1(2)
92.5(2)
Sb2-N5
Sb2-N6
Sb2-N7
Li3-N5
Li3-N6
Li3-N8
Li4-N7
Li4-N8
2.105(4)
2.098(5)
1.966(4)
2.13(1)
2.13(1)
2.00(1)
1.94(1)
1.99(1)
87.0(2)
92.7(2)
91.8(2)
N1-Sb1-N2
N1-Sb1-N3
N2-Sb1-N3
N5-Sb2-N6
N5-Sb2-N7
N6-Sb2-N7
Scheme 3. Possible Mechanism for the Formation of Complex 7
Figure 7. View of the structure of Complex 7. Only the amido hydrogen
atoms are shown.
reaction with cis-3 would readily form the cubane complex
6, whereas trans-3 might produce a new structural motif.
However, only the reaction of 2 equiv. of n-butyl-lithium
with trans-3 resulted in the formation of the complex 6; the
analogous reaction with cis-3 consistently resulted in de-
struction of the cyclodistib(III)azane to form the tricyclic
complex [Li2Sb(µ-NtBu)(µ-NDipp)(µ-NH(Dipp)] (7), as
depicted in Scheme 2. The ready formation of the cis-
arranged cubane 6 from trans-3 clearly requires a rapid
isomerization to occur in solution. The NMR studies of 3
described earlier demonstrated that trans-3 isomerizes to cis-3
in solution, albeit slowly (days). Presumably, the formation
of six Li-N bonds provides a thermodynamic driving force
that increases the rate of isomerization in solution upon
deprotonation of trans-3. It is also of interest to note that in
the preparation of complex 5, the highest yields are obtained
when the reaction is carried out on samples enriched with
trans-2; when freshly prepared samples of 2 (ratio of cis:
trans isomers, 80:20) are used, yields of 5 are ca. 20%.
Although these observations are consistent with the reactivity
of compound 3, they do not unambiguously prove that only
trans-2 leads to the formation of complex 5.
(4)-1.970(4) Å), which was also observed in complex 6.
There is little difference in the SbV-N distances observed
in 7 in comparison to the SbIII-N distances found in 5 and
6. Although it could be anticipated that the SbV-N distances
would be shorter than the SbIII-N distances, it has been
previously demonstrated that there is little correlation
between the oxidation state of antimony and Sb-N bond
lengths.24 Unsurprisingly, the N-Sb-N angles of the LiSbN2
ring are similar to those observed in the cluster complex 6
(Table 4).
The formation of complex 7 can be considered to result
from the monolithiation of cis-3 followed by nucleophilic
attack at the second antimony center, which results in
replacement of one Sb in the Sb2N2 ring by Li (Scheme 3).
This difference in reactivity in relation to trans-3 can be
attributed to the endo, endo configuration of the amido
protons in cis-3. Presumably, the two bulky Dipp substituents
Complex 7 crystallizes with two crystallographically
independent but chemically equivalent molecules in the
asymmetric unit; one molecule is depicted in Figure 7. The
Sb-NtBu imido distances (2.089(4)-2.099(4) Å) are con-
siderably longer than the Sb-NDipp imido distances (1.967-
(24) Copsey, M. C.; Gallon, S. B.; Grocott, S. K.; Jeffery, J. C.; Russell,
C. A.; Slattery, J. M. Inorg. Chem. 2005, 44, 5495.
Inorganic Chemistry, Vol. 45, No. 26, 2006 10739

Eisler and Chivers
sterically shield the amido protons, making them less
accessible to approach by n-butyl-lithium. This would allow
nucleophilic attack by n-butyl-lithium at one of the antimony
centers to become a viable alternative reaction that initiates
the opening of the Sb2N2 ring in cis-3. Competition between
deprotonation and nucleophilic substitution has been ob-
served previously for amido-arsenic and amido-boron sys-
tems.25,26 In the current case, the byproduct is presumably
the antimony oligomer (nBuSb)n.27
of a general synthetic route to dichloro reagents of the type
[ClE(µ-NR)]2 (E ) Sb, Bi) remains a significant challenge.
Experimental Section
All reactions and the manipulations of products were performed
under an argon atmosphere by using standard Schlenk techniques
or an inert atmosphere glovebox. Solvents were freshly distilled,
dried, and degassed prior to use. NMR spectra were obtained on
C6D6 solutions at 25 °C using a Bruker AMX 300 spectrometer,
unless otherwise noted. All lithium amides were prepared by
reaction of n-butyl-lithium with the respective amine, isolated as
solids, dried under vacuum, and subsequently stored in an inert
atmosphere glovebox. The reagent [ClSb(µ-NtBu)]2 (1) was pre-
pared by the literature methods.13 X-ray quality crystals of 1 were
grown by slow diffusion of n-hexane into a concentrated solution
of the compound in THF at -20 °C.
2: A mixture of LiNHtBu (0.693 g, 8.76 mmol) and 1 (2.000 g,
4.38 mmol) was placed in a 100 mL Schlenk flask and cooled to
-78 °C. To this was added 50 mL of cold toluene (-78 °C), and
the reaction mixture was stirred for 0.5 h at -78 °C. The solution
was warmed to room temperature and stirred for an additional 1.5
h. The mixture was then filtered through Celite and the solvent
was removed under a vacuum, to give primarily cis-2 as a thick
yellow oil (2.05 g, 88%) that solidifies after storage at -20 °C for
12 h. Anal. Calcd for C16H38N4Sb2: C, 36.26; H, 7.23; N, 10.57.
Found: C, 35.34; H, 7.16; N, 10.08. (cis-2) 1H NMR: δ 1.27, 1.34
(s, 36H, C(CH3)3) ,2.65 (s, 2H, NHtBu). 13C NMR: δ 35.86, 36.44
(C(CH3)3), 53.31, 54.81 (C(CH3)3). (trans-2) 1H NMR: δ 1.28, 1.29
(s, 36H, C(CH3)3), 2.69 (s, 2H, NHtBu). 13C NMR: δ 35.63, 35.69
(C(CH3)3), 53.09, 55.85 (C(CH3)3).
Solution NMR Studies of Complexes 5-7. The NMR
data obtained for complexes 5-7 were in keeping with the
solid-state structures of the complexes. In the case of 5, two
1
separate resonances are observed in the H NMR spectrum
for the methyl protons of the two different tert-butyl imido
groups. For complex 6, one resonance is observed for the
tert-butyl groups, and one doublet and one septet are present
1
for the iso-propyl groups of the Dipp substituents. The H
NMR spectrum of 7 is, as anticipated, more complex because
of the presence of the two different Dipp substituents.
However, the appropriate resonances are clearly distinguish-
able, such as the two separate doublets for the methyl protons
of the isopropyl groups. The amido N-H proton resonance
is also clearly visible and is markedly shielded (δ ) 2.63)
relative to that observed in cis-3 (δ ) 5.01) presumably as
a result of an increase in negative charge at the amido
nitrogen center in the complex 7.
Conclusions
3: A solution of 1 (4.000 g, 8.76 mmol) in 70 mL of cold toluene
(-78 °C) was added to a suspension of LiNHDipp (3.210 g, 17.52
mmol) in 20 mL of cold toluene (-78 °C), and the mixture was
stirred for 30 min at -78 °C. The solution was allowed to warm to
room temperature and stirred for an additional 2h. The mixture was
then filtered through Celite and the solvent was removed under a
vacuum. The residue was washed with n-hexane (15 mL), leaving
trans-3 as a yellow solid (2.75 g, 43%). The n-hexane washings
were collected and the solvent removed under a vacuum to give
cis-3 as a thick yellow oil (2.65 g, 41%). Storage of the oil at -20
°C for 24 h resulted in the formation of X-ray quality crystals of
cis-3. A small amount of trans-3 could be dissolved into hot
n-hexane, and storage of this solution at -20 °C for several hours
produced X-ray quality crystals of trans-3. Anal. Calcd for
C32H54N4Sb2: C, 52.06; H, 7.37; N, 7.59. Found: C, 51.90; H,
7.67; N, 7.32. (cis-3) 1H NMR: δ 1.17 (s, 18H, C(CH3)3), 1.35 (d,
The metathetical reaction of lithium amides with [ClSb-
(µ-NtBu)]2 was used successfully to prepare the first series
of bis(1°-amino)cyclodistib(III)azanes. In all cases, the
compounds were found to exist in solution as a mixture of
cis and trans isomers. In the case of [NHDippSb(µ-NtBu)]2,
it was possible to separate the two isomers and compare their
reactivity with nBuLi. Surprisingly, the trans-isomer gave rise
to the cis-arranged heterometallic cubane [LiNDippSb(µ-Nt-
Bu)]2 (6). In contrast, the cis isomer was found to be
susceptible to nucleophilic attack at antimony, resulting in
destruction of the cyclodistib(III)azane ring and the formation
of the tricyclic complex [Li2Sb(µ-NtBu)(µ-NDipp)(µ-NH-
Dipp)] (7). These unexpected results have significant
implications for the study of cyclodistib(III)azanes, and could
help to shed light on the multifarious results that have been
previously observed in the chemistry of these compounds.1,3
3
3
24H, JHH ) 8 Hz, CH(CH3)2), 3.72 (sept, 4 H, JHH ) 8 Hz,
3
CH(CH3)2), 5.01 (s, 2 H, NHDipp), 6.97, (t, 2 H, JHH ) 8 Hz,
Ar-H), 7.18 (d, 4 H, 3JHH ) 8 Hz, Ar-H). 13C NMR: δ 24.71 (CH-
(CH3)2), 29.49 (CH(CH3)2), 36.11 (C(CH3)3), 54.08 (C(CH3)3),
121.09, 124.11, 138.22, 143.42 (Ar-C). (trans-3) 1H NMR: δ 1.06
Although we have demonstrated the applicability of this
approach to the synthesis of bis(1°-amino)cyclodistib(III)-
azanes, further application of this methodology suffers from
the lack of availability of other [ClSb(µ-NR)]2 dimers, 1
being the only known isolated example. The development
3
(s, 18H, C(CH3)3), 1.34 (d, 24H, JHH ) 8 Hz, CH(CH3)2), 3.62
(sept, 4 H, 3JHH ) 8 Hz, CH(CH3)2), 5.23 (s, 2 H, NHDipp), 6.95,
(t, 2 H, 3JHH ) 8 Hz, Ar-H), 7.19 (d, 4 H, 3JHH ) 8 Hz, Ar-H). 13
C
NMR: δ 24.28 (CH(CH3)2), 29.89 (CH(CH3)2), 35.18 (C(CH3)3),
53.65 (C(CH3)3), 120.33, 124.15, 136.83, 144.51 (Ar-C).
(25) Copsey, M. C.; Jeffery, J. C.; Leedham, A. P.; Russell, C. A.; Slattery,
J. M. Dalton. Trans. 2003, 2103.
4: In a manner similar to that described for 2, this was prepared
from 1 (3.000 g, 6.57 mmol) and LiNHDmp (1.670 g 13.14 mmol)
in 50 mL of cold toluene (-78 °C) to give primarily cis-4 as a
thick, deep yellow oil (3.58 g, 87%). Storage of the oil for several
hours at room temperature resulted in the formation of X-ray quality
crystals of cis-4. Anal. Calcd for C24H38N4Sb2: C, 46.04; H, 6.12;
(26) Brask, J. K.; Chivers, T.; Schatte, G. Chem. Commun. 2000, 1805.
(27) The reaction of BuSbBr2 with magnesium has been reported to produce
yellow solutions that contain a mixture of cyclic oligomers (BuSb)n
(primarily n ) 5) on the basis of solution NMR data. Evaporation of
these solutions gives black solids. Ates, M.; Breunig, H. J.; Gulec,
S.; Offermann, W.; Hanerle, K.; Drager, M. Chem. Ber. 1989, 122,
473.
10740 Inorganic Chemistry, Vol. 45, No. 26, 2006

Bis(1°-amino)cyclodistib(III)azanes
Table 5. Crystallographic Data for Complexes cis-1, cis-3, trans-3, cis-4, 5, 6, and 7a
cis-1
cis-3
trans-3
cis-4
5
6
7
formula
fw
space group
a (Å)
b (Å)
c (Å)
C8H18Cl2N2Sb2
456.64
P21/n
5.965(1)
14.331(3)
51.65(1)
90
91.77(3)
90
4413(2)
12
C32H54N4Sb2
738.29
P21/c
9.955(2)
11.433(2)
31.723(6)
90
95.71(3)
90
3593(1)
4
C32H54N4Sb2
738.29
P21/n
12.111(2)
9.339(2)
15.749(3)
90
103.10(3)
90
1735.0(6)
2
C24H38N4Sb2
626.08
Pbcn
11.254(2)
12.092(2)
19.741(4)
90
90
90
2684.4(8)
4
1.548
C24H52N4Na2O2Sb2
718.18
Pnma
17.305(4)
16.439(3)
11.772(2)
90
90
90
3349(1)
4
1.424
C36H60Li2N4OSb2
822.26
P21/c
18.353(4)
10.586(2)
21.552(4)
90
112.16(3)
90
3878(2)
4
C40H69Li2N4O2Sb
773.62
P1h
10.705(2)
20.816(4)
20.970(4)
107.63(3)
92.50(3)
91.08(3)
4447(2)
4
1.156
0.654
0.525
0.1376
R (deg)
â (deg)
γ (deg)
V (Å3)
Z
d
calcd (g cm-3
)
2.062
4.008
0.0268
0.0499
1.365
1.528
0.0338
0.0742
1.413
1.582
0.0362
0.0949
1.408
1.424
0.0429
0.0880
µ (mm-1
)
2.028
0.0535
0.1612
1.663
0.0391
0.1020
Rb
Rw
c
a T ) 173(2) K; wavelength ) 0.71073 Å. b R ) ∑|Fo| - |Fc||/∑|Fo| (I > 2.00σ(I)). c Rw ) {[∑w(Fo - Fc )2]/[∑w(Fo )2]}1/2 (all data).
2
2
2
1
N, 8.95. Found: C, 45.72; H, 6.05; N, 8.72. (cis-4) H NMR: δ
1.09 (s, 18H, C(CH3)3), 2.42 (s, 12H, Ar(CH3)), 4.74 (s, 2 H,
25.32 (CH(CH3)2), 25.48 (THF), 29.76 (CH(CH3)2), 34.39 (C(CH3)3),
54.74 (C(CH3)3), 68.32 (THF), 117.37, 123.65, 139.81, 157.29 (Ar-
C).
NHDmp), 6.76, (t, 2 H, 3JHH ) 9 Hz, Ar-H), 7.05 (d, 4 H, 3JHH
)
9 Hz, Ar-H). 13C NMR: δ 21.51 (Ar(CH3)), 36.22 (C(CH3)3), 53.92
7: A solution of n-butyl-lithium (1.21 mL, 2.5 M, 3.03 mmol)
in cold THF (-78 °C, 20 mL) was added using a cannula to a
solution of cis-3 (1.130 g, 1.53 mmol) in cold THF (-78 °C, 20
mL). The solution was held at -78 °C for 30 min and then allowed
to warm to room temperature and stirred for an additional 1 h. The
solvent was removed under a vacuum and the residue was dissolved
in 5 mL of n-hexane. Storage of the solution at -20 °C for 12 h
resulted in the formation of X-ray quality yellow crystals of 7 (0.580
g, 50%). Anal. Calcd for C40H71Li2N4O2Sb: C, 61.94; H, 9.23; N,
1
(C(CH3)3), 119.41, 125.64, 129.74, 147.21 (Ar-C). (trans-4) H
NMR: δ 0.99 (s, 18H, C(CH3)3), 2.32 (s, 12H, Ar(CH3)), 4.94 (s,
3
2 H, NHDmp), 6.76 (t, 2 H, JHH ) 9 Hz, Ar-H), 7.02 (d, 4 H,
3JHH ) 9 Hz, Ar-H). 13C NMR: δ 21.34 (Ar(CH3)), 36.13
(C(CH3)3), 53.71 (C(CH3)3), 118.90, 124.76, 129.70, 147.95 (Ar-
C).
5: A suspension of butyl-sodium (0.725 g, 9.06 mmol) in
n-hexane (15 mL) was cooled to -78 °C, and cold THF (-78 °C,
20 mL) was added to dissolve the butyl-sodium. This solution was
added dropwise by using a cannula to a cold solution of 2 (2.400
g, 4.53 mmol) in THF (-78 °C, 25 mL), and the mixture was stirred
for 0.5 h at -78 °C. The solution was allowed to warm to room
temperature and stirred for an additional 1h. The solution was
filtered through Celite; the solvent was then removed under a
vacuum, and the residue was dissolved in n-hexane (ca. 5 mL).
Storage of the solution at -20 °C for 12 h resulted in the formation
of X-ray quality yellow crystals of 5 (1.78 g, 55%). Anal. Calcd
for C24H52N4Na2O2Sb2: C, 40.14; H, 7.30; N, 7.80. Found: C,
1
7.22. Found: C, 61.10; H, 8.96; N, 7.39. H NMR: δ 1.08 (m,
3
8H, THF), 1.32 (s, 18H, C(CH3)3), 1.34, 1.38 (d, 24H, JHH ) 7
Hz, CH(CH3)2), 2.63 (s, 1H, NHDipp), 3.06 (m, 8H, THF), 3.42
3
(m, br, 2H, CH(CH3)2), 4.04 (sept, 2H, JHH ) 7 Hz, CH(CH3)2),
3
3
6.62, 6.86 (t, 2H, JHH ) 7 Hz, Ar-H), 7.10, 7.17 (d, 4H, JHH
)
7 Hz, Ar-H). 13C NMR: δ 24.21 (CH(CH3)2), 24.96 (THF), 25.53
(CH(CH3)2), 28.48, 28.80 (CH(CH3)2), 35.05 (C(CH3)3), 52.37
(C(CH3)3), 68.72 (THF), 111.60, 118.21, 122.94, 123.25, 132.05,
143.35, 158.11, 158.35 (Ar-C).
X-ray Analyses. Individual crystals of cis-1 (pale yellow rod),
cis-3 (yellow needle), trans-3 (yellow plate), cis-4 (yellow plate),
5 (yellow block), 6 (yellow rod), and 7 (yellow plate) were coated
with Paratone 8277 oil (Exxon) and mounted onto thin glass fibers.
Data were collected on a Nonius Kappa CCD diffractometer via ω
and φ scans. Cell parameters, data reduction, and absorption
corrections were performed with the Nonius software (DENZO and
SCALEPACK, B. V. Nonius, 1998). The structures were solved
by direct methods as implemented in SHELXS-97, and refinement
was carried out on F2 against all independent reflections by the
full-matrix least-squares method using the SHELXL-97 program
(G. M. Sheldrick). The hydrogen atoms were calculated geo-
metrically and refined using a riding model unless otherwise
specified. Except as mentioned, all non-hydrogen atoms were
refined with anisotropic thermal parameters. Relevant parameters
for the data collections and crystallographic data are summarized
in Table 5. Thermal ellipsoid plots are shown at the 30% probability
level.
1
40.36; H, 7.61; N, 7.64. H NMR: δ 1.33 (m, 8H THF), 1.44,
1.53 (s, 36H, C(CH3)3), 3.52 (s, 8H, THF). 13C NMR: δ 25.92
(THF), 35.15, 42.15 (C(CH3)3), 55.37, 56.23 (C(CH3)3), 68.40
(THF).
6: A solution of n-butyl-lithium (2.20 mL, 2.5 M, 5.50 mmol)
in cold THF (-78 °C, 5 mL) was added using a cannula to a
solution of trans-3 (2.000 g, 2.71 mmol) in cold THF (-78 °C, 60
mL). The solution was held at -78 °C for ca. 2 min and then
allowed to warm to room temperature and stirred for an additional
40 min. The solvent was removed under a vacuum and the residue
was dissolved in 5 mL of n-hexane, resulting in the immediate
precipitation of yellow crystals of 6. Deposition of the crystals was
allowed to continue for 30 min; the solvent was then decanted and
the crystals dried under a vacuum (1.78 g, 80%). Bulk samples of
6 were contaminated with small amounts of 7 because of the cis-
trans equilibration of 3 (see Results and Discussion). However,
analytically pure 6 can be obtained by sequential recrystallizations.
Anal. Calcd for C36H60Li2N4OSb2: C, 52.58; H, 7.36; N, 6.81.
cis-1: The complex crystallizes with three independent but chem-
ically equivalent molecules in the asymmetric unit. The refinement
was complicated by twinning of the data, although all non-hydrogen
atoms were refined with anisotropic thermal parameters. There were
many indications of twinning, including the fact that the refinement
stalled at 21.18%. Application of the program ROTAX28 determined
1
Found: C, 52.06; H, 7.66; N, 6.99. H NMR: δ 1.32 (s, 18H,
3
C(CH3)3), 1.35 (d, 24H, JHH ) 8 Hz, CH(CH3)2), 3.16 (m, 4H,
THF), 3.44 (sept, 4 H, 3JHH ) 8 Hz, CH(CH3)2), 6.91, (t, 2 H, 3JHH
) 8 Hz, Ar-H), 7.20 (d, 4 H, JHH ) 8 Hz, Ar-H). 13C NMR: δ
3
Inorganic Chemistry, Vol. 45, No. 26, 2006 10741

Eisler and Chivers
that twinning occurred around the [100] direct lattice axis, and the
program WinGX29 was used to prepare an HKLF5 file for further
refinement. The R values, K values, esds, and background noise
were all significantly improved, indicating the correct twin assign-
ment.
cis-3: The molecule was well-ordered with the exception of one
of the isopropyl groups, which was modeled as a 50:50 isotropic
mixture. The amido hydrogen atoms were located in the difference
map.
trans-3: The molecule was well-ordered and no special consid-
erations for the refinement were necessary. The amido hydrogen
atoms were located in the difference map. Only one-half of the
molecule was located in the difference Fourier map as the molecule
is situated on a crystallographic center of symmetry.
cis-4: The molecule was well-ordered, and no special consid-
erations for the refinement were necessary. The amido hydrogen
atoms were located in the difference map. Only one-half of the
molecule was located in the difference Fourier map, as the molecule
is situated on a crystallographic 2-fold rotation axis.
5: The molecule was well-ordered with the exception of the
coordinated THF molecule, which was modeled as a 60:40 mixture;
two of the disordered carbon atoms were refined with isotropic
thermal parameters.
6: The molecule was well-ordered, with the exception of one
carbon position of the coordinated THF molecule, which was
modeled as a 65:35 isotropic mixture. The lithium atoms were
refined with isotropic thermal parameters.
7: There are two chemically equivalent but crystallographically
independent molecules in the asymmetric unit. The molecules were
well-ordered with the exception of the coordinated THF molecules
and one isopropyl group, the latter of which was modeled as a 60:
40 isotropic mixture. Three of the THF molecules were modeled
as 50:50 isotropic mixtures, whereas the fourth THF molecule was
severely disordered, and the best model that could be refined was
a 25:25:20:15:15 isotropic mixture. For all the disordered portions,
light geometric restraints were applied. The amido hydrogen atoms
were located in the difference map.
Acknowledgment. The authors gratefully acknowledge
financial support from the Natural Sciences and Engineering
Research Council (Canada) and the Alberta Ingenuity Fund.
Supporting Information Available: Crystallographic data in
CIF format. This material is available free of charge via the Internet
at http://pubs.acs.org.
IC061660T
(28) Cooper, R. I.; Gould, R. O.; Parsons, S.; Watkin, D. J. J. Appl.
Crystallogr. 2002, 35, 168.
(29) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837.
10742 Inorganic Chemistry, Vol. 45, No. 26, 2006