A self-assembled, magnetically coupled square Cu16 4 x [2 x 2] grid

Article abstract of DOI:10.1002/anie.200701840

(Chemical Equation Presented) Grid ion: A Cu₁₆ grid prepared with a tetratopic bishydrazone ligand has metal ions in a [4 x 4] array linked by a combination of μ-O_hydrazone and μ-NN_pyridazine bridges. The six-coordinate copper ions display Jahn-Teller distortion, leading to some orbitally orthogonal bridges. Multiple intragrid magnetic-exchange pathways are present and are modeled simply by taking the molecular symmetry and the orthogonal connections into account.

Full text of DOI:10.1002/anie.200701840

Communications
DOI: 10.1002/anie.200701840
Magnetic Grid Complexes
A Self-Assembled, Magnetically Coupled Square Cu16
” [2 ” 2]Grid**
4
Louise N. Dawe and Laurence K. Thompson*
Self-assembly strategies in coordination chemistry have
reached a level of maturity such that predictable structures
can be created by the simple expedient of designing and
synthesizing a ligand with suitably encoded coordination
information. With an appropriate balance between the
coordinating features of the ligand and the coordination
requirements of a metal, square [n ” n]grid structures have
been achieved routinely for n = 2 and 3 (M , M ) with six-
ordered “flat” arrays has created a broad interest in such
systems within the chemistry and physics communities;
[
4–7]
this interest has focused on their novel quantum-based
magnetic properties, and in particular, in the case of
II
III
[Mn Mn ]mixed-spin-state [3 ” 3]grids, their potential to
5
4
[
7]
act as spin “qubits”.
The coordination capacity of such ligands can be
expanded to “tetratopic” by starting with a central dinucleat-
ing pyridazine fragment, rather than pyridine (Scheme 1), and
using similar extension strategies. The tetratopic pyridazine
bishydrazone ligand L1 was shown to produce a linear
4
9
II
II
II
coordinate transition-metal ions (M : M = Mn , Co , Ni ,
4
II
II
II
III
II
II
Cu , Zn ; M : M = Mn , Fe , Cu , Zn ), using ditopic and
9
tritopic (Scheme 1) picolylhydrazone ligands, respectively.
II
[8]
trinuclear Ni complex, with an empty potential coordina-
tion site (occupied by water), and a open, square-based,
[
9]
incomplete Cu12 [n ” n]grid. Other pyridazine-based poly-
topic ligands include tritopic and pentatopic examples, which
[
10]
produce a [3 ” 3]Ag 9 grid complex and a 2 ” [2 ” 5]Ag
20
[
11]
incomplete gridlike complex, respectively,
in which the
pyridazine groups act as bridges, and the congruence of the
bidentate ligand pockets creates four-coordinate (N ) metal
4
sites. A recent review article highlights a number of [n ” n]
[
12]
grid-based systems in this general class.
The Schiff base ligand L2 was recently shown to produce
II [13]
the expected [4 ” 4]M 16 grid with Mn . A mixture of m-O
(hydrazone and OH) and pyridazine bridges in this compact
grid arrangement of S = 5/2 spin centers led to intramolecular
antiferromagnetic exchange. This ligand type has now been
modified further to introduce pyrimidine end groups (e.g. L3).
Pentatopic 2,6-bis(picolyl)hydrazone ligands have also been
prepared by first creating an extended ester, converting it into
the corresponding extended bishydrazone, and then capping
with a pyridine function (e.g. L4). The anticipated [5 ” 5]grid
II[14]
has been produced with Mn
and displays unusual mag-
netic properties, which are dominated by intramolecular
antiferromagnetic exchange with a large residual ground-
state spin.
Scheme 1. Tritopic, tetratopic, and pentatopic ligands.
Reaction of L3 with Cu(CF SO ) in MeOH/CH CN led
3
3
2
3
Spin-exchange interactions occur between the paramagnetic
to the formation of a brown solution, from which brown
crystals of [(L3) Cu (O) (OH) (H O) ](CF SO ) (H O) -
metal ions through direct bridging (e.g. m-O, m-NN) connec-
8
16
2
4
2
2
3
3
6
2
66
[
1–3]
tions.
The ability to concentrate spin-bearing sites in
(CH OH) (1) were obtained. The crystal structure revealed
3 10
II
a square grid arrangement of 16 six-coordinate Cu ions
embedded in an assembly of eight ligands, arranged in two
roughly parallel groups of four, above and below the metal
[
*] L. N. Dawe, Prof. Dr. L. K. Thompson
Department of Chemistry
Memorial University
pseudoplane (Figure 1). The overall molecular dimensions for
2
1
involve a square cation approximately 19 ” 19 Š with a
St. John’s NL A1B 3X7 (Canada)
Fax: (+1)709-737-3702
2
metallic core 10.9 ” 11.2 Š .
A simplified core structure is shown in Figure 2. The
overall structure is best considered as a combination of four
E-mail: lthomp@mun.ca
[
**] This study was supported by the Natural Sciences and Engineering
Research Council of Canada (NSERC).
square Cu -(m-Ohydrazone)4 subunits, with each subunit bridged
4
to its neighbors by a combined m-NNpyridazine/m-O double
bridge, in which the oxygen is exogenous. The overall
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
7440
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7440 –7444

Angewandte
Chemie
and two protons were found on O7, indicating a bridging
water molecule at this site and its symmetry-related counter-
part. No protons were found on O8, and combined with the
very short Cu4ꢀO8 bond length (1.925 Š), this strongly
suggests that O8 and its symmetry-related counterpart are
oxide. Other oxygen bridging sites (O1–O4) within the Cu4
corner subunits are associated with deprotonated hydrazone
oxygen atoms. This combination would lead to a total internal
charge of ꢀ8 on the basis of the molecular symmetry. The two
additional charges could reasonably be associated with lattice
components, for example, hydroxide, but close examination
of the ligand NH protons revealed that H23 bound to N21 is
2
at significantly lower occupancy than the others. Conse-
quently, it was set to an occupancy of 0.5 in the final
refinement to correctly reflect the overall charge difference,
thus indicating further ligand deprotonation under basic
conditions.
Each ligand (see Scheme 1) would of necessity create a
bend in the arrangement of its four bound metal ions, which
would be repeated throughout the grid to lead to a
pronounced overall bending of the core {Cu (m-O) } struc-
Figure 1. Povray structure of the cation in 1 (Cu magenta, Nblue,
O red, C black).
1
6
24
ture itself (Figure 3). The packing of the ligands in close
Figure 3. Core structure of 1 showing the bending (Cu large spheres,
O small spheres).
proximity in roughly parallel and eclipsed pairs must add
significant stability to the structure through p–p interactions
and influence the choice of the preferred syn ligand con-
formation. Closest contacts between atoms in the pyridazine
rings fall in the range 3.39–3.62 Š, whereas within the Cu4
corner subunits similar pyrimidine contacts fall in the range
3.40–3.91 Š. No significant intergrid contacts were found in
the extended lattice structure.
The complexity of the intragrid bridging connections
would suggest that many, and perhaps all, of the bridges
would contribute to the overall magnetic spin-exchange
situation between the copper centers. Previous studies on
Figure 2. Povray core structure of 1.
molecular symmetry (space group Pmmn) leads to four
identical Cu4 subunits related by internal mirror-plane
symmetry. Cu···Cu distances within each m-O-bridged square
fall in the range 3.92–4.12 Š, and Cu-O-Cu angles are in the
range 131–1418. Cu···Cu distances between the pyridazine-
bridged Cu4 subunits are smaller (3.35–3.73 Š), and as a
consequence the Cu-O-Cu bridge angles are smaller (96.8–
1
23.68). The presence of six well-defined lattice CF SO3
3
anions suggests nominally that 26 negative charges exist
within the grid. Each ligand can reasonably be assigned a ꢀ2
charge on the basis of the normal loss of protons from the
hydrazone oxygen atoms. The 10 additional negative charges
are therefore reasonably associated with the exogenous
oxygen bridging sites, and perhaps the ligands. The formation
of hydroxide as a bridge in reactions of pyridazine-based
ligands with copper(II) salts is common, even without the
simpler, and directly related, Cu [2 ” 2]square grids with
4
ditopic picolinic hydrazone ligands show that Jahn–Teller
distortions occur at the copper(II) sites, leading to canted
(908) orientations of the copper magnetic planes and resulting
[
1,17]
in magnetic orbital orthogonality within the grid.
The
mirror-plane symmetry in 1 allows a relatively simple assess-
ment of the magnetic-orbital connectivity by effectively
considering just one quarter of the grid (Figure 4). Each
[
15,16]
II
addition of base.
However, base was added during the
corner Cu subunit comprises four six-coordinate Cu ions
4
synthesis of 1, which would reasonably lead to the formation
of hydroxide and even oxide. Single protons were revealed in
difference maps on O5 and O6, indicating that these sites and
their symmetry-related counterparts are bridging hydroxides,
with m-Ohydrazone bridges between adjacent metal centers.
Examination of CuꢀL bond lengths reveals the expected
Jahn–Teller elongation for Cu2, Cu3, and Cu4 (arrows in
Figure 4; Cu2ꢀO4 2.323 Š, Cu2ꢀN19 2.468 Š, Cu3ꢀO7
Angew. Chem. Int. Ed. 2007, 46, 7440 –7444
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7441

Communications
Figure 4. Jahn–Teller axes within the Cu -(m-O) corner subunits.
4
4
Figure 5. Plots of susceptibility and moment per mole as a function of
temperature for 1 (see text for fitted parameters).
2
2
.251 Š, Cu3ꢀO3 2.134 Š, Cu4ꢀO6 2.453 Š, Cu4ꢀO2
.193 Š), indicating d
x2ꢀy2
magnetic ground states, and orthog-
onal connections through the oxygen bridges. This geometry
would therefore suggest no antiferromagnetic exchange
between Cu4 and both Cu2 and Cu3. Similar connections
connections, with the possibility of very weak antiferromag-
netic coupling between Cu1 and Cu2. However, on the
assumption that the spin-exchange connections from Cu2,
Cu3, and Cu4 to other adjacent subunits are dominated by
much stronger antiferromagnetic exchange (see above), the
corner Cu1 sites can be considered to be essentially isolated,
thus accounting for the four effective residual spins at low
temperature. Magnetization data were obtained at 2 K in the
range 0–5 T (Figure S1 in the Supporting Information) and
show a rise in M to a value of 4.4 Nb at 5 T (1Nb =
5585 cm Gmol ; M = 1.05 Nb for an integer spin at 5 T
with g = 2.21), with no evidence for saturation. The profile is
not consistent with a ground-state system with S = 4/2, but
corresponds very closely to the sum of four independent S =
1/2 centers (solid line in Figure S1 calculated for g = 2.21 at
2 K; see below). There is no structural evidence to suggest
longer range intermolecular exchange interactions.
throughout the Cu squares lead to intramolecular ferromag-
4
[
1,17]
netic exchange.
pressed dz
ground state (Cu1ꢀN4 1.932 Š, Cu1ꢀN16
.920 Š), which, despite the long Cu1ꢀO1
bond
Cu1 is different; it has an axially com-
2
1
hydrazone
(
2.251 Š), could lead to very weak antiferromagnetic
exchange between Cu1 and Cu2. The short Cu2ꢀO5 and
Cu2ꢀN6
contacts (1.947 and 2.003 Š, respectively), and
pyridazine
3
ꢀ1
their symmetry-related counterparts, lead to a combination of
nonorthogonal pyridazine and oxygen atom bridges. As a
result, antiferromagnetic coupling would be expected for both
bridging connections. Cu3 is bridged to its symmetry-related
counterparts by an orthogonal oxygen bridge (O7), and a
nonorthogonal pyridazine bridge, thus leading to antiferro-
magnetic exchange only through pyridazine. For the Cu4
centers and their symmetry-related counterparts, two bridg-
ing connections occur, through O6 and O8, leading to
different bridging combinations. The oxygen O6 bridge is
orthogonal, whereas oxygen O8 is nonorthogonal, but in both
cases the pyridazine bridges are nonorthogonal. In those cases
in which both bridges involve nonorthogonal connections it is
reasonable to assume that any exchange would be consid-
erably larger than in those cases in which just the pyridazine
bridge is magnetically active.
The magnetic consequences of the bridging connections
between the Cu subunits would be the sum of the relevant m-
4
O and m-NN contributions, taking into account the orthogon-
ality situation and the molecular symmetry. Figure 6 repre-
sents a spin-exchange model that subdivides the grid into the
magnetic-exchange components that will be the principal
contributors. As the four corner sites can effectively be
isolated magnetically, and as the connections 2–13, 3–14, 5–14,
6–15, 8–15, 9–16, 11–16, and 12–13 are orthogonal (see
above), the exchange can be modeled as the sum of four
dinuclear terms and one tetranuclear term, as defined by the
Variable-temperature magnetic data were collected on 1
(
2–300 K in a 0.1-T field). Plots of molar magnetic moment
and susceptibility are shown in Figure 5. No discernible
maximum was observed in the c(T) profile, but the drop in
moment on lowering the temperature clearly indicates
dominant intramolecular antiferromagnetic exchange. The
apparent change in slope of the m(T) plot suggests that j J j
values of quite different magnitude are involved, in keeping
with complex and different bridging connectivity. The room-
exchange Hamiltonian [Eq. (1)]. J refers to connections with
1
Hex ¼ ꢀJ fS Á S þ S Á S gꢀJ fS Á S þ S Á S g
1
2
3
8
9
2
5
6
11
12
ð1Þ
ꢀ½J
1
fS13 Á S16 þ S14 Á S15g þ J
2
fS13 Á S14 þ S15 Á S16gŠ
temperature moment (6.6 m ) is consistent with a system
B
II
comprising 16 spin-coupled Cu ions (1.65 m per metal), and
B
the moment appears to approach a slight minimum of 3.7 mB
around 10 K, before dropping slightly below this temperature.
The nonzero low-temperature moment indicates the presence
of significant residual spin, which approximates the presence
II
of four isolated Cu sites (mcalcd = 3.8 mB for g = 2.2). The
corner square Cu subunits have three orthogonal internal
Figure 6. Proposed magnetic exchange model for 1.
4
7442 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7440 –7444

Angewandte
Chemie
[
16,17]
orthogonal oxygen bridges, and J2 refers to those with
nonorthogonal oxygen bridges. The simplifying assumption
is also made that exchange through O5 and O8 is the same,
because the Cu-O-Cu angles are comparable (123.88 and
zine-3,6-dicarboxylate
with hydrazine hydrate in methanol) was
added to the above solution and the pH was adjusted to 7 with acetic
acid. The mixture was refluxed for 18 h. The resulting yellow slurry
was filtered and washed with diethyl ether (3 ” 15 mL) to yield a
yellow powder, which was used without further purification. Yield
1
21.08, respectively).
(
1.24 g, 88%), m.p. > 3308C. Mass spectrum (major mass peaks, m/z):
Appropriate exchange equations were written for the
dinuclear and tetranuclear components, and these were
factored and combined to include corrections for TIP
3
91.4 (MꢀNH ), 302.1, 284.1, 149.1. IR (nujol): n˜ = 3309, 3197 (nNH);
2
ꢀ
1
1646 (nC = O); 1600 cm (nC=N). Elemental analysis calcd (%) for
C H N O ·0.5H O: C 46.27, H 3.64, N 40.47; found: C 46.44, H 3.41,
N 40.72.
Compound 1 was synthesized from the reaction of Cu(CF SO )
3 2
16 14 12 2 2
(
“
temperature-independent paramagnetism) and the four
isolated” corner copper(II) sites. In view of the complexity
of the exchange model, a remarkably good fit was obtained
3
(
0.608 mmol) in 1:1 MeOH/CH CN (15 mL) with L3 (0.295 mmol).
3
This reaction produced a clear, dark green solution to which aqueous
NaOH (4 mL; 0.65m) was added to achieve a neutral pH. The
resulting dark brown solution was stirred with gentle heating for 24 h
and then filtered. The filtrate was preserved for crystallization. Brown
needle-like crystals, suitable for X-ray diffraction, formed upon
standing after 12 days (35% yield). Elemental analysis calcd (%) for
(C16H N O ) Cu16(O) (OH) (H O) ](CF SO ) -
11 12 2 2 2 4 2 2 3 3 6
H O) (CH OH) : C 25.47, H 4.07, N 19.81; found: C 25.24, H 2.13,
N 19.82.
Crystal data for 1: Orthorhombic, Pmmn (no. 59), a = 25.797(2),
b = 29.199(2), c = 16.5920(13) Š, V= 12497.5(18) Š , Z = 2, 1
ꢀ1
(
ꢀ
solid lines in Figure 5) for g = 2.21, J = ꢀ45 cm , J =
1
2
ꢀ
1
ꢀ6
3
ꢀ1
2
350 cm , TIP = 800 ” 10 cm mol , 10 R = 2.3; (R =
2
2 1/2
[
ꢀ(cobsdꢀc ) /ꢀc
] ). J1 is associated only with the
calcd
obsd
pyridazine bridges, whereas J is associated with the pyrida-
2
zine/nonorthogonal oxygen bridging connections. As would
[
(C16
H N O
12 12 2
)
6
be expected, j J j is much larger than j J j , and it is satisfying
2
1
(
2
66
3
10
to compare these values with those associated with compa-
rable simple dinuclear copper compounds involving pyrida-
zine and pyridazine/OH bridges. Weak coupling (ꢀJ =
3
=
calcd
ꢀ1
ꢀ3
ꢀ1
6
5 cm ) was found for the compound [Cu (PPD)Cl ], which
1.505 gcm , m = 1.482 mm , final R = 0.0639 for I > 2s(I), wR
1
2
=
2
4
0
.1825 for all data. The intensity data were recorded on a Rigaku
AFC8-Saturn 70 system with MoKa radiation (l = 0.71070 Š) at
13(2) K. The structure was solved by direct methods and refined
involves only a magnetically active pyridazine bridge in a
[
18]
similar bonding situation. For compounds with magneti-
cally active combinations of pyridazine and m-OH, and large
Cu-OH-Cu angles (116–1268), ꢀJ values in the range 375–
1
2
by full-matrix least-squares analysis on F by using SHELXL. NH2
protons (H4, 5, 10, 11, 16, 17, 22, and 23) were located in difference
map positions. However, H23 appeared much lower in the peak list
relative to the others, and its occupancy was consequently set to 0.5.
ꢀ1
[15,16]
4
50 cm were found.
This comparison strongly supports
the assignment of the bridging connectivity in 1 and the
prediction of magnetic properties on the basis of the orbital
connectivity.
[
18]
The Platon Squeeze
procedure was applied to recover 427.1
3
electrons per unit cell in two voids (total volume 4161.7 Š ).
Disordered solvent water and acetonitrile molecules appeared to be
present prior to the application of Squeeze, though a good point atom
model could not be achieved for them. The application of Squeeze
greatly improved the data statistics and allowed a full anisotropic
refinement of the framework structure and counterions.
CCDC 644634 contains the supplementary crystallographic data for
this paper. These can be obtained free of charge from the Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/
cif.
In summary, ligand design and synthesis has led to a viable
approach to the specific self-assembly of a supramolecular
polymetallic square-grid-shaped Cu16 aggregate. The forma-
tion of a complete [4 ” 4]grid in this case, compared with an
[
9]
incomplete Cu12 grid with the closely related ligand L1, is
reasonably associated with the addition of sufficient base to
fully deprotonate the ligand hydrazone oxygen atoms. This
arrangement leads to a complex network of spin-exchange
pathways through pyridazine, hydrazone, and exogenous
oxygen bridges throughout the grid, which in some cases are
effectively switched off because of the preferred orthogonal
magnetic connections between copper ions in certain grid
positions. This behavior is typical of copper(II) in simpler [2 ”
Variable-temperature magnetic measurements were performed
on a Quantum Design MPMS5S magnetometer in DC mode (0.1-T
field) with appropriate corrections for the sample holder and
diamagnetic complex components.
Received: April 25, 2007
Published online: August 14, 2007
2
]and [3 ” 3]grids, in which ferromagnetic exchange domi-
nates because of a prevalence of orthogonal connections.
However, in the present case, a specific mixture of orthogonal
and nonorthogonal connections allows the antiferromagnetic
exchange terms to dominate. Fitting of the variable-temper-
ature magnetic data to an appropriate model based on this
connectivity gives exchange coupling through m-pyridazine
and m-pyridazine/O combinations comparable to those of
simpler related dinuclear complexes.
Keywords: copper · grid complexes · magnetic properties ·
N,O ligands · self-assembly
.
[
1]C. J. Matthews, K. Avery, Z. Xu, L.K. Thompson, L. Zhao, D. O.
Miller, K. Biradha, K. Poirier, M. J. Zaworotko, C. Wilson, A. E.
Goeta, J.A.K. Howard, Inorg. Chem. 1999, 38, 5266 – 5276.
2]L. K. Thompson, L. Zhao, Z. Xu, D. O. Miller, W. M. Reiff,
Inorg. Chem. 2003, 42, 128 – 139.
[
[
3]V. A. Milway, S. M. T. Abedin, V. Niel, T. L. Kelly, L. N. Dawe,
S. K. Dey, D. W. Thompson, D. O. Miller, M. S. Alam, P. Müller,
L. K. Thompson, Dalton Trans. 2006, 2835 – 2851.
Experimental Section
L3: Methyl pyrimidine-2-carboximidate was generated in situ by
reaction of 2-cyanopyrimidine (0.73 g, 6.96 mmol) with a solution of
sodium methoxide, produced by dissolving sodium metal (0.10 g,
[4]O. Waldmann, R. Koch, S. Schrom, P. Müller, L. Zhao, L. K.
Thompson, Chem. Phys. Lett. 2000, 332, 73 – 78.
[5]O. Waldmann, L. Zhao, L. K. Thompson, Phys. Rev. Lett. 2002,
88, 066401.
4
.35 mmol) in methanol (100 mL). Pyridazine-3,6-dicarbohydrazide
(0.68 g, 3.45 mmol; prepared from the reaction of dimethyl pyrida-
Angew. Chem. Int. Ed. 2007, 46, 7440 –7444
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7443

Communications
[
[
[
6]L. K. Thompson, O. Waldmann, Z. Xu, Coord. Chem. Rev. 2005,
[14]S. K. Dey, T. S. M. Abedin, L. N. Dawe, S. S. Tandon, J. L.
Collins, L. K. Thompson, A. V. Postnikov, M. S. Alam, P. Müller.
Inorg. Chem., in press.
[15]L. K. Thompson, T. C. Woon, D. B. Murphy, E. J. Gabe, F. L.
Lee, Y. Le Page, Inorg. Chem. 1985, 24, 4719 – 4725.
[16]L. K. Thompson, S. K. Mandal, E. J. Gabe, F. L. Lee, Inorg.
Chem. 1987, 26, 657 – 664.
249, 2677 – 2690.
7]O. Waldmann, H. U. Güdel, T. L. Kelly, L. K. Thompson, Inorg.
Chem. 2006, 45, 3295 – 3300.
8]L. K. Thompson, C. J. Matthews, L. Zhao, C. Wilson, M. A.
Leech, J. A. K. Howard, J. Chem. Soc. Dalton Trans. 2001, 2258 –
2262.
[
9]C. J. Matthews, S. T. Onions, G. Morata, M. Bosch Salvia,
[17]L. N. Dawe, T. S. M. Abedin, T. L. Kelly, L. K. Thompson, D. O.
Miller, L. Zhao, C. Wilson, M. A. Leech, J. A. K. Howard, J.
Mater. Chem. 2006, 16, 2645 – 2659.
[18]S. K. Mandal, L. K. Thompson, E. J. Gabe, F. L. Lee, Inorg.
Chem. 1987, 26, 2384 – 2389.
M. R. J. Elsegood, D. J. Price, Chem. Commun. 2003, 320 – 321.
[
10]P. N. W. Baxter, J.-M. Lehn, J. Fischer, M.-T. Youinou, Angew.
Chem. 1994, 106, 2432 – 2434; Angew. Chem. Int. Ed. Engl. 1994,
3, 2284 – 2287.
11]P. N. W. Baxter, J.-M. Lehn, G. Baum, D. Fenske, Chem. Eur. J.
000, 6, 4510 – 4517.
3
[
[
[19]G. H. Spencer, Jr., P. C. Cross, K. B. Wiberg, J. Chem. Phys. 1961,
35, 1939 – 1945.
2
12]M. Ruben, J. Rojo, F. J. Romero-Salguero, L. H. Uppadine, J.-M.
Lehn, Angew. Chem. 2004, 116, 3728 – 3747; Angew. Chem. Int.
Ed. 2004, 43, 3644 – 3662.
[20]S. Sueur, M. Lagrenee, F. Abraham, C. Bremard, J. Heterocycl.
Chem. 1987, 24, 1285 – 1289.
[21]A. L. Spek, J. Appl. Crystallogr. 2003, 36, 7 – 13.
[
13]S. K. Dey, L. K. Thompson, L. N. Dawe, Chem. Commun. 2006,
4967 – 4969.
7444 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7440 –7444