A unique example of structural and magnetic diversity in four interconvertible copper(II) - azide complexes with the same schiff base ligand: A monomer, a dimer, a chain, and a layer

Article abstract of DOI:10.1021/ic1005456

Four new Cu(II)-azido complexes of formula [CuL(N₃)] (1), [CuL(N₃)]₂ (2), [Cu₇L₂(N ₃)₁₂]_n (3), and [Cu₂L(dmen)(N ₃)₃]_n (4)

Full text of DOI:10.1021/ic1005456

6616 Inorg. Chem. 2010, 49, 6616–6627
DOI: 10.1021/ic1005456
A Unique Example of Structural and Magnetic Diversity in Four Interconvertible
Copper(II)-Azide Complexes with the Same Schiff Base Ligand: A Monomer,
a Dimer, a Chain, and a Layer
§
Subrata Naiya,† Chaitali Biswas,†,‡ Michael G. B. Drew, Carlos J. Gomez-Garcıa,*,^ Juan M. Clemente-Juan,^,||
ꢀ
´
and Ashutosh Ghosh*,†
†Department of Chemistry, University College of Science, University of Calcutta, 92, APC Road,
Kolkata-700 009, India, ‡Sarojini Naidu College for Women, 30 Jessore Road, Kolkata-700 028, India,
§School of Chemistry, The University of Reading, P.O Box 224, Whiteknights, Reading RG 66AD, U.K.,
^
´
Instituto de Ciencia Molecular (ICMol), Parque Cientıfico, Universidad de Valencia, 46980 Paterna,
||
ꢀ
Valencia, Spain, and Fundacion General de la Universidad de Valencia (FGUV), Valencia, Spain
Received March 23, 2010
Four new Cu(II)-azido complexes of formula [CuL(N3)] (1), [CuL(N3)]2 (2), [Cu7L2(N3)12]n (3), and [Cu2L(dmen)-
(N3)3]n (4) (dmen = N,N-dimethylethylenediamine) have been synthesized using the same tridentate Schiff
base ligand HL (2-[1-(2-dimethylaminoethylimino)ethyl]phenol, the condensation product of dmen and 2-hydroxy-
acetophenone). The four compounds have been characterized by X-ray structural analyses and variable-temperature
magnetic susceptibility measurements. Complex 1 is mononuclear, whereas 2 is a single μ-1,1 azido-bridged
dinuclear compound. The polymeric compound 3 possesses a 2D structure in which the Cu(II) ions are linked by
phenoxo oxygen atoms and two different azide bridges (μ-1,1 and μ-1,1,3). The structure of complex 4 is a double
helix in which two μ-1,3-azido-bridged alternating one-dimensional helical chains of CuL(N3) and Cu(dmen)(N3)2 are
joined together by weak μ-1,1 azido bridges and H-bonds. The complexes interconvert in solution and can be obtained
in pure form by carefully controlling the conditions. The magnetic properties of compounds 1 and 2 show the presence
of very weak antiferromagnetic exchange interactions mediated by a ligand π overlap (J = -1.77) and by an
asymmetric 1,1-N3 bridge (J = -1.97 cm-1), respectively. Compound 3 presents, from the magnetic point of view, a
decorated chain structure with both ferro- and antiferromagnetic interactions. Compound 4 is an alternating helicoidal
chain with two weak antiferromagnetic exchange interactions (J = -1.35 and -2.64 cm-1).
Introduction
the construction of such species, the Cu(II)-azide system is
one of the most popular among synthetic chemists. A variety
of copper-azido complexes with discrete or one-, two-, and
three-dimensional polymeric structures have been reported,
in which the azido ligand exhibits diverse bridging modes
ranging from μ-1,1 (end-on, EO) and μ-1,3 (end-to-end, EE)
to μ-1,1,1, μ-1,1,3, μ-1,1,1,1, μ-1,1,3,3, and μ-1,1,1,3,3,3,
depending upon the steric and electronic demands of the
co-ligands2-4 (Scheme 1). This diversity in the structures of
The rational design and synthesis of polynuclear coordina-
tion complexes is an area of continuing interest in order to
understand the structural and chemical factors that govern
the exchange coupling between paramagnetic centers and to
establish magnetostructural correlations in molecular systems
with the aim of controlling and developing new functional
molecular-based materials.1 Although many different brid-
ging groups and transition metal ions have been employed for
*To whom correspondence should be addressed. E-mail: carlos.gomez@
uv.es; ghosh_59@yahoo.com.
(2) (a) Stamatatos, T. C.; Papaefstathiou, G. S.; MacGillivray, L. R.;
Escuer, A.; Vicente, R.; Ruiz, E.; Perlepes, S. P. Inorg. Chem. 2007, 46, 8843.
(b) Gao, E.-Q.; Bai, S.-Q.; Wang, C.-F.; Yue, Y.-F.; Yan, C.-H. Inorg. Chem. 2003,
42, 8456. (c) Abu-Youssef, M. A. M.; Escuer, A.; Mautnerc, F. A.; Ohrstrom, L.
Dalton Trans. 2008, 3553.
(3) (a) You, Y. S.; Yoon, J. H.; Kimb, H. C.; Hong, C. S. Chem. Commun.
2005, 4116. (b) Nanda, P. K.; Aromíb, G.; Ray, D. Chem. Commun. 2006, 3181.
(c) Lazari, G.; Stamatatos, T. C.; Raptopoulou, C. P.; Psycharis, V.; Pissas, M.; Perlepes,
S. P.; Boudalis, A. K. Dalton Trans. 2009, 3215. (d) Lorosch, J.; Paulus, H.; Haase, W.
Inorg. Chim. Acta 1985, 106, 101. (e) Boudalis, A. K.; Sanakis, Y.; Clemente-Juan,
J. M.; Donnadieu, B.; Nastopoulos, V.; Mari, A.; Coppel, Y.; Tuchagues, J.-P.; Perlepes,
S. P. Chem.;Eur. J. 2008, 14, 2514, and references therein.
(1) (a) Kostakis, G. E.; Powell, A. K. Coord. Chem. Rev. 2009, 253, 2686.
(b) Ambrosi, G.; Formica, M.; Fusi, V.; Giorgi, L.; Micheloni, M. Coord. Chem.
Rev. 2008, 252, 112. (c) Biswas, B.; Pieper, U.; Weyherm€uller, T.; Chaudhuri, P.
Inorg. Chem. 2009, 48, 6781. (d) Stamatatos, T. C.; Christou, G. Inorg. Chem.
2009, 48, 3308. (e) Baca, S. G.; Malaestean, I. L.; Keene, T. D.; Adams, H.; Ward,
M. D.; Hauser, J.; Neels, A.; Decurtins, S. Inorg. Chem. 2008, 47, 11108.
€
ꢀ
ꢀ
(f) Suarez-Varela, J.; Mota, A. J.; Aouryaghal, H.; Cano, J.; Rodríguez-Dieguez,
A.; Luneau, D.; Colacio, E. Inorg. Chem. 2008, 47, 8143. (g) Biswas, C.;
ꢀ
Mukherjee, P.; Drew, M. G. B.; Gomez-García, C. J.; Clemente-Juan, J. M.;
Ghosh, A. Inorg. Chem. 2007, 46, 10771.
r
pubs.acs.org/IC
Published on Web 06/18/2010
2010 American Chemical Society

Article
Inorganic Chemistry, Vol. 49, No. 14, 2010 6617
Scheme 1. Various Azido Bridging Modes
ion remains coordinately unsaturated and thus allows the
azide to bridge the copper centers. Recently it has been shown
that in neutral Cu-azido systems with chelating diamine
ligands the dimensionalities of the polynuclear structures
are determined by the relative molar quantities of copper
and the diamine ligands.6 The products, however, more often
than not, are unexpected and result from uncontrolled
self-assembly. In order to truly design the synthesis of a
desired polynuclear material, one should have a clear knowl-
edge regarding the synthetic conditions that lead to the
formation of a particular product and/or the interconver-
sion between the products. To the best of our knowledge, in
the Cu-azide system, apart from attempts of changing
the molar proportions,6 the effects of other factors (e.g.,
temperature, template effect of the counteranions, solvents,
etc.) that are well known to modulate the structure of
compounds in several other systems7 have not been studied
systematically until the present work. Herein, we report the
synthesis, structural characterization, and variable-tempera-
ture magnetic behavior of four new compounds: a mono-
nuclear [CuL(N3)] (1), a single μ-1,1-azido-bridged dinuclear
[CuL(N3)]2 (2), a 2D coordination polymer with μ-1,1 and
μ-1,1,3 azide bridges [Cu7L2(N3)12]n (3), and a μ-1,3- and
μ-1,1-azido-bridged double-stranded helix, [Cu2L(dmen)-
(N3)3]n (4). All four complexes have been prepared using
the same N,N,O donor Schiff base ligand (2-[1-(2-
dimethylaminoethylimino)ethyl]phenol) (HL), the conden-
sation product of N,N-dimethyl-1,2-diaminoethane (dmen)
and 2-hydroxyacetophenone (Scheme 1) by varying the
reaction conditions. The factors that allow the conversion
of one compound into another have also been explored in
detail. During the course of this work, the crystal structure
and magnetic properties of compound 4 have been reported
by others.8 However, the previous authors reported just one
compound formed with this ligand and as a result did not
study any possible interconversion to related compounds.
Therefore, we retain mention of this compound in our paper,
as it is an integral part of the interconversion of the com-
pounds, but do not describe the structure in detail. Moreover,
we fit the magnetic data of this compound with a different
model and obtain a better fit, which is also reported here.
the Cu(II) systems is a result of its flexibility in coordination
numbers (ranging from 4 to 6) and geometry,2-4 and taken
together with its interesting magnetic properties, Cu(II) has
therefore become the metal ion of choice for such studies. As a
result of the extensive research of the last two decades, the
superexchange mechanisms through various bridging modes
of azide are now well established. For example, symmetric
μ-1,3 Cu(II) azide complexes are strongly antiferromagnetic,
whereas Cu(II) complexes with double symmetric μ-1,1
azide bridges are strongly ferromagnetic, provided that the
Cu-Nazide-Cu angle is less than 108ꢀ.5 Usually asymmetric
μ-1,3 azido bridges lead to weak antiferromagnetic coupl-
ing, whereas asymmetric μ-1,1 azide bridges propagate weak
to moderately strong ferro- or antiferromagnetic interac-
tions.5c-e,i-k Since other structural paramaeters certainly
affect magnetic exchange, a number of exceptions have been
reported, and it has been pointed out that other structural
parameters such as Cu-N bond lengths and the Cu(N)2Cu
dihedral angles5 need also to be considered.
On the contrary, the many possible bridging modes of
azide and the flexible coordination number (and geometry) of
Cu(II) make the synthesis of tailored compounds a real
challenge for coordination chemists and crystal engineers.
There are very few studies focusing on the variables that
dictate the isolation of a particular polynuclear species. A
common strategy for the synthesis of polynuclear complexes
is to choose the blocking ligand in such a way that the copper
Experimental Section
Materials. The reagents and solvents used were of commer-
cially available reagent quality.
(4) (a) Goher, M. A. S.; Mautner, F. A. Polyhedron 1995, 14, 1439. (b) Gu,
Z.-G.; Xu, Y.-F.; Yin, X.-J.; Zhou, X.-H.; Zuo, J.-L.; You, X.-Z. Dalton Trans.
2008, 5593. (c) Zhang, X.-M.; Zhao, Y.-F.; Wu, H.-S.; Battenb, S. R.; Ng, S. W.
Dalton Trans. 2006, 3170. (d) Guo, G. C.; Mak, T. C. W. Angew. Chem., Int. Ed.
1998, 37, 3268.
Synthesis of the Schiff BaseLigand HL [(2-[1-(2-dimethylamino-
ethylimino)ethyl]phenol]. The monocondensed Schiff base ligand
HL (Scheme 2) has been synthesized as reported before.8-10
A
(5) (a) Tandon, S. S.; Thompson, L. K.; Manuel, M. E.; Bridson, J. N.
Inorg. Chem. 1994, 33, 5555. (b) Ribas, J.; Escuer, A.; Monfort, M.; Vicente, R.;
Cortes, R.; Lezama, L.; Rojo, T. Coord. Chem. Rev. 1999, 193-195, 1027.
(c) Koner, S.; Saha, S.; Mallah, T.; Okamoto, K.-I. Inorg. Chem. 2004, 43, 840.
(d) Zbiri, M.; Saha, S.; Adhikary, C.; Chaudhuri, S.; Daul, C.; Koner, S. Inorg.
Chim. Acta 2006, 359, 1193. (e) Mukherjee, P. S.; Dalai, S.; Mostafa, G.; Lu,
T.-H.; Rentschlerc, E.; Chaudhuri, N. R. New J. Chem. 2001, 25, 1203.
(f) Manikandan, P.; Muthukumaran, R.; Justin Thomas, K. R.; Varghese, B.;
Chandramouli, G. V. R.; Manoharan, P. T. Inorg. Chem. 2001, 40, 2378. (g) Dey,
S. K.; Mondal, N.; Fallah, M. S. E.; Vicente, R.; Escuer, A.; Solans, X.Font-Bardía,
M.; Matsushita, T.; Gramlichjj, V.; Mitra, S. Inorg. Chem. 2004, 43, 2427. (h)
Mukherjee, P. S.; Maji, T. K.; Mostafa, G.; Mallah, T.; Chaudhuri, N. R. Inorg.
Chem. 2000, 39, 5147. (i) Comarmond, P.; Plumere, P.; Lehn, J. M.; Augus, Y.;
Louis, R.; Weiss, R.; Kahn, O.; Morgenstern-Badarau, I. J. Am. Chem. Soc. 1982,
104, 6330. (j) Ray, M. S.; Ghosh, A.; Bhattacharya, R.; Mukhopadhyay, G.; Drew,
M. G. B.; Ribas, J. Dalton Trans. 2004, 252. (k) Ray, M. S.; Ghosh, A.; Chaudhuri,
S.; Drew, M. G. B.; Ribas, J. Eur. J. Inorg. Chem. 2004, 3110.
(6) (a) Mondal, K. C.; Mukherjee, P. S. Inorg. Chem. 2008, 47, 4215.
(b) Mukherjee, S.; Gole, B.; Chakrabarty, R.; Mukherjee, P. S. Inorg. Chem.
2009, 48, 11325.
(7) (a) Hu, X.-X.; Xu, J.-Q.; Cheng, P.; Chen, X.-Y.; Cui, X.-B.; Song,
J.-F.; Yang, G.-D.; Wang, T.-G. Inorg. Chem. 2004, 43, 2261. (b) Withersby,
M. A.; Blake, A. J.; Champness, N. R.; Cooke, P. A.; Hubberstey, P.; Li, W.-S.;
€
Schroder, M. Inorg. Chem. 1999, 38, 2259. (c) Forster, P. M.; Burbank, A. R.;
ꢀ
Livage, C.; Ferey, G.; Cheetham, A. K. Chem. Commun. 2004, 368. (d) Tong,
M.-L.; Kitagawa, S.; Changa, H.-C.; Ohbaa, M. Chem. Commun. 2004, 418.
ꢀ
(e) Mukherjee, P.; Drew, M. G. B.; Gomez-García, C. J.; Ghosh, A. Inorg. Chem.
2009, 48, 5848.
(8) Adhikary, C.; Sen, R.; Tuchagues, J.-P.; Chaudhuri, S.; Ianelli, S.;
Solzi, M.; Koner, S. Inorg. Chim. Acta 2009, 362, 5211.
(9) Biswas, C.; Drew, M. G. B.; Asthana, S.; Desplanches, C.; Ghosh, A.
J. Mol. Struct. 2010, 965, 39.
(10) Talukder, P.; Sen, S.; Mitra, S.; Dahlenberg, L.; Desplanches, C.;
Sutter, J.-P. Eur. J. Inorg. Chem. 2006, 329.

6618 Inorganic Chemistry, Vol. 49, No. 14, 2010
Scheme 2. Schiff Base Ligand
Naiya et al.
shaped, blue-colored, X-ray quality single crystals of complex 4
appeared at the bottom of the vessel.
Yield: 0.10 g, 19%. Anal. Calcd for C16H29Cu2N13O: C,
35.16; H, 5.35; N, 33.31. Found: C, 35.09; H, 5.45; N, 33.23.
IR (KBr, cm-1): 3126, 3217, and 348 ν(N-H), 1598 ν(CdN),
2036 ν(N-N), λmax (nm) [εmax (dm3 mol-1 cm-1)] (methanol),
612 (203).
The complex was also prepared in much higher yield (0.43 g,
78%) by adding a methanol solution (20 mL) of HL and dmen
(2 mmol each) to a methanol solution (10 mL) of Cu(ClO4)
3
6H2O (1.462 g, 4 mmol) followed by an aqueous solution (1 mL)
of NaN3 (0.520 g, 8 mmol). The crystalline compound 4 started
to separate within half an hour on keeping the mixture at room
temperature. The product was collected after about 24 h.
Physical Measurements. Elemental analyses (C, H, and N)
were performed using a Perkin-Elmer 240C elemental analyzer.
IR spectra in KBr pellets (4500-500 cm-1) were recorded using
a Perkin-Elmer RXI FT-IR spectrophotometer. The 1H NMR
spectra at 300 MHz were recorded in CDCl3 on a Bruker DRX
300 spectrometer. The numbers in parentheses after the H atoms
in the NMR spectral data are the numbers of carbon atoms to
which the hydrogen atoms are attached, as in Scheme 2. Elec-
tronic spectra in methanol (1000-200 nm) were recorded in
a Hitachi U-3501 spectrophotometer. High-resolution mass
spectra (HRMS) electrospray ionization (ESI) was recorded
on a Qtof Micro YA263 high-resolution mass spectrometer. For
HRMS (ESI), the sample was taken in CH3OH. Thermal
analyses (TG-DTA) were carried out on a Mettler Toledo
TGA/SDTA 851 thermal analyzer in a dynamic atmosphere
of dinitrogen (flow rate 30 cm3 min-1). The samples were heated
in an alumina crucible at a rate of 10 ꢀC min-1. Variable-
temperature magnetic susceptibility measurements were carried
out in the temperature range 2-300 K with an applied magnetic
field of 0.1 T on polycrystalline samples of compounds 1-4
(with masses of 51.70, 25.77, 39.14, and 41.39 mg, respectively)
with a Quantum Design MPMS-XL-5 SQUID susceptometer.
Isothermal magnetization measurements were performed on the
same samples at 2 K with magnetic fields up to 5 T. The
susceptibility data were corrected for the sample holders pre-
viously measured using the same conditions and for the dia-
magnetic contributions of the salt as deduced by using Pascal’s
constant tables (χdia = -239.3 ꢀ 10-6, -184.1 ꢀ 10-6, -803.6 ꢀ
solution of 2-hydroxyacetophenone (20 mmol, 2.41 mL) and
N,N-dimethyl-1,2-diaminoethane (20 mmol, 2.20 mL) in metha-
nol (50 mL) was refluxed for one hour. The resulting dark yellow
solution was then used directly for complex formation. The ligand
was isolated as a yellow solid by removing methanol completely
from the resultant solution under vacuum to result in a viscous
liquid, followed by addition of a mixture of petroleum ether and
chloroform. The ligand was characterized by elemental analyses
and 1H NMR study.
Ligand (HL). Anal. Calcd for C12H18N2O: C, 69.87; H, 8.80;
N, 13.58. Found: C, 69.81; H, 8.89; N, 13.45. 1H NMR (CDCl3,
300 MHz) (Figure S1): 2.11 (s, 3H, CH3); 2.28 (s, 6H, 2 ꢀ
N-CH3); 2.65(t, J = 6.9 Hz, 2H, CH2); 3.61 (t, J = 6.9 Hz, 2H,
CH2); 6.87(dd, 3J = 8.4 Hz and 4J = 0.9 Hz, 1H, C13-H); 7.22
(dt, 3J = 8.7 Hz and 4J = 1.5 Hz, C14-H); 6.70 (dt, 3J = 8.1 and
4J = 1.2 Hz, 1H, C15-H); 7.44 (dd, 3J = 8.1 Hz and 4J = 1.5 Hz,
1H, C16-H); 16.38 (s, 1H, OH).
Synthesis of the Complexes [CuL(N3)] (1) and [CuL(N3)]2 (2).
To a 30 mL methanolic solution of Cu(ClO4)2 6H2O (5.00 mmol,
3
1.828 g) and the HL (5.00 mmol) was added an aqueous solution
(1 mL) of NaN3 (5.00 mmol; 0.325 g). The mixture was stirred for
2 h and filtered. The filtrate was then divided into two equal parts;
one of the parts was kept at room temperature and the other was
kept in a refrigerator (-10 ꢀC). Thin, deep-blue-colored, block-
shaped single crystals of 1 were obtained after 2 days from the
filtrate which was kept at room temperature, whereas dark blue,
needle-shaped, single crystals of 2 were obtained after one week
from the other part of the filtrate which was kept in the refrig-
erator. Crystals of 1 and 2 were suitable for X-ray analysis.
Compound 1. Yield: 0.56 g, 72%. Anal. Calcd for C12H17Cu-
N5O: C, 46.37; H, 5.51; N, 22.53. Found: C, 46.21; H, 5.66; N, 22.42.
IR (KBr, cm-1): 3363 ν(N-H), 1594 ν(CdN), 2044 ν(N-N), λmax
(nm) [εmax (dm3 mol-1 cm-1)] (methanol), 603 nm (66).
10-6, and -394.3 ꢀ 10-6 emu mol-1 for 1-4, respectively).
3
Crystallographic Studies. Intensity data for the four com-
pounds were collected with Mo KR radiation at 150 K using an
Oxford Diffraction X-Calibur CCD system. The crystals were
positioned at 50 mm from the CCD; 321 frames were measured
with a counting time of 10 s. Data analyses were carried out with
the CrysAlis program.11 The structures were solved using direct
methods with the Shelxs97 program.12 The non-hydrogen atoms
were refined with anisotropic thermal parameters. The hydro-
gen atoms bonded to carbon were included in geometric posi-
tions and given thermal parameters equivalent to 1.2 times (or
1.5 times for methyl hydrogen atoms) those of the atom to which
they were attached. Absorption corrections were carried out
using the ABSPACK program.13 The structures were refined on
F2 using Shelxl97.12 Selected dimensions in the four complexes
are given in Table 1.
Compound 2. Yield: 0.52 g, 67%. Anal. Calcd for C24H34Cu2-
N10O2: C, 46.37; H, 5.51; N, 22.53. Found: C, 46.31; H, 5.46; N,
22.68. IR (KBr, cm-1): 3364 ν(N-H), 1594 ν(CdN), 2046 ν(N-N),
λ
max (nm) [εmax (dm3 mol-1 cm-1)] (methanol), 594 nm (114).
Synthesis of the Complex [Cu7L2(N3)12]n (3). To a 30 mL
methanolic solution of [Cu2(OAc)4 (H2O)2] (7.00 mmol, 1.397 g)
3
and HL (2.00 mmol) was added an aqueous solution (1 mL) of
excess NaN3 (12.00 mmol, 0.780 g) with constant stirring. During
stirring, a dark-blue-colored, fine crystalline precipitate of 3 was
separated. The mixture was then filtered. Blue-colored, plate-
shaped, diffraction quality single crystals of 3 were obtained after
a few days on slow evaporation of the filtrate. The IR spectra of the
fine crystalline precipitate and that of single crystals were identical.
Yield: 1.02 g, 75%. Anal. Calcd for C24H34Cu7N40O2: C,
21.20; H, 2.52; N, 41.21. Found: C, 21.33; H, 2.42; N, 41.11. IR
(KBr, cm-1): 3361 ν(N-H), 1587 ν(CdN), 2062-2085 ν(N-N),
Results and Discussion
Synthesis of the Complexes and Their Interconversion.
The monocondensed tridentate Schiff base ligand HL
(2-[1-(2-dimethylaminoethylimino)ethyl]phenol) and its
λ
max (nm) [εmax (dm3 mol-1 cm-1)] (methanol), 599 (189).
Synthesis of the Complex [Cu2L(dmen)(N3)3]n (4). The proce-
dure was similar to that for complex 3. An aqueous solution of
-
-
Cu(II) complexes with ClO4 and BF4 anions are
excess NaN3 (15 mmol, 0.780 g) was added, with constant
stirring, to a methanolic solution of Cu(ClO4) 6H2O (7.00
3
(11) CrysAlis; Oxford Diffraction Ltd.: Abingdon, U.K., 2006.
(12) Sheldrick, G. M. Shelxs97 and Shelxl97, Programs for Crystallo-
graphic Solution and Refinement. Acta Crystallogr. 2008, A64, 112.
(13) ABSPACK; Oxford Diffraction Ltd: Oxford, U.K., 2005.
mmol, 2.559 g) (instead of [Cu2(OAc)4 (H2O)2] for 3) and HL
3
(2 mmol). The mixture was then refluxed for one hour. The
resulting solution was left to stand overnight in air, when plate-

Article
Inorganic Chemistry, Vol. 49, No. 14, 2010 6619
Table 1. Crystal Data and Structure Refinement of Complexes 1-4
1
2
3
4
formula
fw
space group
cryst syst
a/A
C12H17CuN5O
310.85
Cc
monoclinic
19.635(3)
12.194 (2)
12.111(2)
(90)
107.06(2)
(90)
2772.2(7)
8
1.490
C24H34Cu2N10O2
621.70
P21/n
monoclinic
11.783(4)
19.194(5)
12.154(3)
(90)
94.96(2)
(90)
2738.4(13)
4
1.508
C24H34Cu7N40O2
1359.69
P1
C16H29Cu2N13
546.62
C2/c
monoclinic
24.928(5)
9.054(2)
22.412(2)
(90)
119.75(1)
(90)
4391.7(14)
8
1.653
O
triclinic
10.3496(11)
11.3763(10)
11.7155(10)
98.344(8)
115.39 (1)
98.852(8)
1196.4(2)
1
b/A
c/A
R/deg
β/deg
γ/deg
V/A3
Z
calcd density (g cm-3
)
)
1.887
3.127
677
3
absorp coeff (μ, mm-1
F(000)
1.575
1288
1.595
1288
1.976
2256
cryst size (mm)
θ range (deg)
R(int)
0.12 ꢀ 0.14 ꢀ 0.17
2.4, 30.0
0.052
0.03 ꢀ 0.03 ꢀ 0.22
2.5, 30.0
0.326
0.02 ꢀ 0.19 ꢀ 0.19
2.2, 30.0
0.080
0.05 ꢀ 0.17 ꢀ 0.17
2.44, 30.0
0.062
total reflns
7860
5116
2462
0.0848
0.2386
0.90
18 932
7928
2048
0.1043
0.2023
0.81
8037
6384
1907
0.0839
0.1151
0.82
14 808
6365
4019
0.0660
0.1622
0.99
no. of unique data
data with I > 2σ(I)
R1 on I > 2σ(I)
wR2 on I > 2σ(I)
GOF on F2
known,9,10 both of which are aqua-bridged dinuclear
compounds. In the present investigation, we have used
this ligand for the synthesis of complexes with Cu(II)-
azide. This group of mononegative N,N,O donor ligands is
well known to produce mononuclear or dinuclear double
asymmetric EO azide-bridged complexes.5c-e,7e,14-16 There-
fore, it is not surprising that the mononuclear compound 1
could be isolated as deep-blue-colored, block-shaped crys-
held together by azide bridges (μ-1,1 μ-1,1,3), is obtained.
Compound 3 is sparingly soluble in methanol at room
temperature. In boiling methanol its solubility increases,
and interestingly, this solution (obtained after separation
of undissolved solid by filtration) yielded compound 1 on
evaporation at room temperature, indicating that com-
pound 3 converts in solution to compound 1 (as is evident
from the mass spectra of the solution, identical to that of
1). Even more surprising is the fact that compound 1 also
converts into 3 in a rather easy way; when a methanolic
solution of an excess of copper acetate and sodium azide
is added to a methanolic solution of 1, compound 3 starts
to crystallize within an hour. The lower solubility of 3
seems to be the driving force in the equilibrium toward
its formation. By contrast, when an excess of copper(II)
salts other than copper acetate (perchlorate, nitrate, or
chloride) and azide are allowed to react with HL at room
temperature, compound 4 is formed along with 1. How-
ever if the reaction mixture is refluxed and then allowed to
stand at room temperature, compound 4 is the only
product that crystallizes. The structure of compound 4
shows that it is an azide-bridged chain containing alter-
nate Cu(L)(N3) and Cu(dmen)(N3)2 moieties. The dmen
ligand might come from the hydrolysis of the Schiff base
ligand or from a small portion of unreacted dmen in the
solution of HL used for complex formation. However, it
is found that compound 1 can be converted into com-
pound 4, when a methanolic solution of 1 is reacted with
excess copper perchlorate and azide. Therefore, hydro-
lysis of the Schiff base certainly takes place, but this is not
very surprising, as metal ion catalyzed hydrolysis of Schiff
bases is well documented in the literature.14,17,18 Com-
pound 4 transforms completely into 1 when 2-hydroxy-
acetophenone is added to a methanol solution of 4 (3:2
molar ratio) and the solution is refluxed for 1 h (Scheme 3).
tals by reacting a methanolic solution of Cu(ClO4)2 6H2O
3
with a methanolic solution of HL in 1:1 molar ratio
followed by the addition of aqueous solutions of NaN3
at room temperature (25-35 ꢀC) (Scheme 3). To study if
temperature has any effect on the molecular structure,
the same reaction mixture was kept in a refrigerator (ca.
-10 ꢀC) when dark-blue-colored, needle-shaped crystals
of 2appeared, havingdistinctlydifferent morphologyfrom
that of 1. Interestingly, compounds 1 and 2 are easily
interconvertible: a methanolic solution of 1 or 2 after slow
evaporation at room temperature yields compound 1, but
the same solution at low temperature (-10 ꢀC) produces
compound 2 (Scheme 3). The results clearly indicate that
the crystallization temperature plays a crucial role in the
final structure obtained. The HRMS-ESI of 1 and 2
recorded in methanolic solutions are identical (Figure S2,
Supporting Information). The spectra show a less intense
(42%) peak at ca. m/z 268, corresponding to the azide-free
cationic species of the mononuclear complex [Cu(L)]þ.
The most intense (100%) peak observed, at ca. m/z 578,
can be assigned to the association of two mononuclear
species by a single azide ion, [Cu2L2(N3)]þ.
On the other hand, when HL is allowed to react with
excess [Cu2(OAc)4 (H2O)2] and azide ions (3:7) at room
temperature, the 2D polymer [Cu7L2(N3)12] (3), which is
3
(14) Mukherjee, P.; Drew, M. G. B.; Ghosh, A. Eur. J. Inorg. Chem. 2008,
3372.
(15) (a) Jiang, Y.-B.; Kou, H.-Z.; Wang, R.-J.; Cui, A.-L. Eur. J. Inorg.
Chem. 2004, 4608. (b) You, Z.-L.; Jiao, Q.-Z.; Niu, S.-Y.; Chi, J.-Y. Z. Anorg.
Allg. Chem. 2006, 632, 2481.
(16) Talukder, P.; Datta, A.; Mitra, S.; Rosair, G.; Fallah, M. S. E.;
Ribas, J. Dalton Trans. 2004, 4161.
~
(17) Mandal, D.; Bertolasi, V.; Ribas-Arino, J.; Aromı, G.; Ray, D. Inorg.
´
Chem. 2008, 47, 3465.
(18) Sarkar, B.; Ray, M. S.; Drew, M. G. B.; Figuerola, A.; Diaz, C.;
Ghosh, A. Polyhedron 2006, 25, 3084.

6620 Inorganic Chemistry, Vol. 49, No. 14, 2010
Naiya et al.
Scheme 3
The isolation of 3 in the presence of copper acetate but not
of the other salts indicates that acetate ion definitely has an
important role in its formation. One possibility is that the
acetate ion helps to hold two Cu(II) ions together via a
copper acetate dimer, thus facilitating the formation of the
azido-bridged polynuclear structure 3 via a templating
effect.
azide bridge, while two overlapping but distinct peaks at
2062 to 2085 cm-1 for complex 3 corroborate the presence
of two types of azido bridging (μ-1,1, and μ-1,1,3) in the
complex. Complex 4 shows a broad and single peak at
2036 cm-1. The electronic spectra in methanolic solution
of the four complexes display a single absorption band at
603, 594, 599, and 612 nm for complexes 1-4, respec-
tively. These spectra are typical of a square-based environ-
ment for Cu(II).
Structure of [Cu(L)(N3)] (1). In 1, there are two discrete
molecules of CuL(N3), called A and B in the asymmetric
unit, which have equivalent structures. The structure of
1A is shown in Figure 1 together with the numbering
scheme in the metal coordination sphere. Bond distances
and bond angles are given in Table 2. Both molecules, 1A
and 1B, contain a Cu(II) atom in a four-coordinate
square-planar geometry bonded to the three donor atoms
O11, N20, and N23 of the tridentate ligand L together with
an azide nitrogen N1 with bond distances of 1.834(8),
IR and UV-Vis Spectra of Complexes. Spectroscopic
data and their assignments are given in the Experimental
Section. The IR spectra of the complexes are similar and
.
show strong sharp peaks for ν(CdN) at 1594, 1594, 1587,
and 1598 cm-1 for complexes 1-4 respectively, confirm-
ing the presence of the Schiff base. Two sharp peaks at
3217 and 3126 cm-1 for complex 4 are attributed to the
presence of a free amine group, which is produced by the
hydrolysis of the Schiff base ligand, consistent with the
single-crystal X-ray structure. The peaks at 2044 and
2046 cm-1 in the IR spectra of 1 and 2, respectively,
appear to be due to the stretching of the terminal or μ-1,1

Article
Inorganic Chemistry, Vol. 49, No. 14, 2010 6621
Figure 1. Structure of complex 1. (a) Structure of 1A with ellipsoids drawn at 30% probability. The structure of 1B is equivalent. (b) Dimeric structure
formed by a C-H/π interaction between molecules 1A and 1B.
Table 2. Selected Bond Lengths (A) and Bond Angles (deg) for Complexes 1 and 2
tively. Bond distances and bond angles in the coordina-
tion spheres are given in Table 2. However in this case
bond
1A
1B
2A
2B
CuA has a weak axial interaction with atom N1B from an
azide at 2.529(8) A, thus forming a square-pyramidal
coordination. The angles subtended by N1B with N1A,
O11A, N20A, and N23A at the metal are 93.4(3)ꢀ,
94.5(3)ꢀ, 87.9(3)ꢀ, and 101.2(3)o, respectively, confirming
that it is in an axial position. CuB shows no such axial
interaction. Indeed the CuB-N1A distance is 3.585(11) A
and CuB is tetracoordinated with a square-planar envi-
ronment. Deviations of donor atoms O11, N20, N23,
and N1 from the mean plane passing through them are
0.130(6), -0.135(7), 0.130(7), and -0.124(9) A, respec-
tively, for 2A and -0.194(6), 0.200(7), -0.190(7), and
0.184(7) A, respectively, for 2B. In 2A, the deviation of the
copper atom from this plane is rather high (0.117(3) A),
directed toward the coordinated N1B, as is expected for a
square-pyramidal geometry. The Addison parameter (τ)
of Cu is 0.20. However, the copper atom deviates only by
0.022(3) A from the mean plane in 2B with an angle of
9.39ꢀ between the planes N1B-Cu-O11B and N20B-
Cu-N23B, indicating only a small tetrahedral distortion.
Cu1-N1
1.932(14)
1.834(8)
1.976(9)
1.981(11)
93.4(4)
175.0(4)
84.4(4)
92.9(5)
1.937(14)
1.903(9)
1.993(12)
2.007(9)
91.0(5)
177.0(5)
87.5(5)
91.1(5)
1.946(8)
1.917(6)
1.918(8)
2.064(8)
91.0(3)
163.5(3)
84.8(3)
93.1(3)
1.957(8)
1.923(7)
1.919(8)
2.063(9)
92.6(3)
166.9(3)
85.1(3)
91.2(3)
Cu1-O11
Cu1-N20
Cu1-N23
O11-Cu1-N20
O11-Cu1-N23
N20-Cu1-N23
O11-Cu1-N1
N20-Cu1-N1
N23-Cu1-N1
168.8(5)
89.9(5)
176.0(6)
90.2(5)
175.6(3)
90.9(3)
169.9(3)
93.3(3)
1.976(9), 1.981(11), 1.932(14) A and 1.903(9), 1.993(12),
2.007(9), 1.937(14) A for 1A and 1B, respectively. In both
molecules there are no significant interactions in axial
positions, the closest contact being 3.28(1) A for CuB to
O11A (0.5þx, 0.5þy, 1þz). Deviations of donor atoms
O11, N20, N23, and N1 from their mean plane are -0.116,
0.119, -0.119, and 0.116 A, respectively, for 1A, showing a
tetrahedral distortion, whereas the four donor atoms are
ꢀ
nearly coplanar in 1B, with an rms deviation of 0.006 A.
The central Cu(II) atom deviates slightly from the basal
plane by 0.044(6) A in 1A and 0.051(6) A in 1B. The
tetrahedral distortion is apparent in 1A since one pair of
mutually trans donor atoms clearly lie below the plane
while the other pair is above the plane with the metal ion
nearly in the mean plane. The dihedral angles between the
two planes (Nl-Cu-O11 and N20-Cu-N23) are 10.58ꢀ
for 1A and 4.23ꢀ for 1B. Each monomer, 1A and 1B, is
stacked by a weak C-H/π interaction (Figure 1b). The
C-H/π distance from C19-H19E of the methyl group
of 1B to the centroid of the phenyl ring (CG) of 1A at
The Cu Cu distance within the dimer is 3.866 A. In this
3 3 3
compound, there is one weak intra-dimer C-H/π inter-
action between the H24E atom of molecule B and the
phenyl ring of molecule A with dimensions C24B-
H24E CG 2.97 A and γ = 20.4ꢀ. There are also two
3 3 3
other inter-dimer interactions: one between H19A from
molecule A and the phenyl ring of B (0.5þx, 0.5-y,
0.5þz) and the other between H19F from molecule B
and the phenyl ring of A (-0.5þx, 0.5-y, -0.5þz) with
dimensions C19A-H19A CG 2.74 A, γ = 7.6ꢀ and
3 3 3
ꢀ
symmetry operation 0.5þx, 0.5þy, 1þz is 2.91 A with
C19B-H19F CG 2.93 A, γ=9.5ꢀ. These C-H/π inter-
3 3 3
γ = 17.11ꢀ.
actions form a one-dimensional chain along the b direc-
tion, as illustrated in Figure 2b.
Structure of [Cu2L2(N3)2] (2). The structure of 2,
Cu2L2(N3)2, also contains two copper atoms in the asym-
metric unit (Figure 2a). Again the basal plane in both
cases (2A and 2B) consists of the three donor atoms O11,
N20, and N23 of the Schiff base ligand together with an
azide nitrogen atom, N1, at 1.917(6), 1.918(8), 2.064(8),
1.946(8) A and 1.923(7), 1.919(8), 2.063(9), 1.957(8) A
from the Cu(II) atom in molecules 2A and 2B, respec-
It is to be noted that compounds 1 and 2 are polymer-
ized isomers: 1 is a monomer and 2 is a dimer. There are
several examples of polymerized isomers in coordination
chemistry,19 but in azido-bridged systems transformation
(19) Haddad, S. F.; Piekardt, J. Transition Met. Chem. 1993, 18, 377.

6622 Inorganic Chemistry, Vol. 49, No. 14, 2010
Naiya et al.
Figure 2. Structure of complex 2. (a) Cu2 dimer with ellipsoids at 30% probability. (b) Polymeric 1D structure of 2 formed by the interdimer C-H/π
interactions, shown as dotted lines.
Table 3. Selected Bond Lengths (A) and Bond Angles (deg) for Complex 3
between the isomers (2 to 1 and 1 to 2) is very rare. In the
present system, the isomerization takes place only in
bond
distance (A)
bond
angle (deg)
solution. The TG/DSC study of compounds 1 and 2 does
not show any interconversion in the solid state before
decomposition (195 ꢀC) (Figure S3). A unique feature of
compound 2 is that it is a single-azido-bridged dimer. In
fact, although there are several Cu(II)-azide compounds
with this type of NNO Schiff base ligand, in most cases
the compounds are double asymmetric EO azide-bridged
dimers,14,5c,5d and the only example of a single bridged
azido complex is not a dimer but a polymer.15b
Cu1-N1
Cu1-O11
Cu1-N20
Cu1-N23
Cu1-N7
Cu2-N7
Cu2-N4
Cu2-N9a
Cu2-N13
Cu2-N16
Cu3-N4
Cu3-N10
Cu3-O11
Cu3-N10b
Cu3-N1
Cu4-N16
Cu4-N13
1.968(8)
1.925(8)
1.947(10)
1.997(11)
2.357(10)
2.000(9)
2.008(8)
2.336(11)
1.977(9)
2.002(8)
1.946(11)
1.963(9)
1.991(7)
2.006(10)
2.242(10)
1.917(10)
1.977(8)
N13-Cu2-N4
N13-Cu2-N9a
N13-Cu2-N16
N4-Cu2-N9a
N4-Cu2-N16
N9a-Cu2-N16
N7-Cu2-N13
N7-Cu2-N4
96.4(4)
91.9(4)
77.4(4)
96.4(4)
165.1(4)
97.3(4)
166.7(4)
92.6(4)
96.8(4)
91.5(4)
91.1(4)
88.2(3)
97.0(4)
97.6(5)
88.8(4)
174.1(5)
87.9(4)
83.5(4)
172.0(4)
96.9(4)
100.8(4)
88.6(3)
106.4(4)
161.0(4)
170.4(4)
99.9(4)
78.2(4)
75.3(3)
93.5(3)
92.4(4)
79.4(4)
N7-Cu2-N9a
N7-Cu2-N16
O11-Cu1-N20
N7-Cu1-O11
N7-Cu1-N20
N7-Cu1-N23
N7-Cu1-N1
Structure of [Cu7L2(N3)12]n (3). The structure of com-
pound 3, Cu7L2(N3)12, is more complicated than that of 1
or 2 since there are four independent copper atoms in the
asymmetric unit, three in general positions and one,
namely Cu4, on an inversion center. Bond distances and
angles are given in Table 3. There is only one independent
Schiff base ligand, but the structure is completed with as
many as six independent azide anions (Figure 3). Cu1 has
a coordination geometry similar to that observed in 1 and
2 in that it is bonded to the three donor atoms of the
ligand, O11, N20, and N23, and to an azide nitrogen
atom, N1, in a nearly planar geometry. The only differ-
ence is the presence of an additional azide ligand coordi-
nated through N7 in an axial position with a long Cu-N
bond distance (2.357(10) A), giving rise to a square-
pyramidal geometry for Cu1. The basal plane has a rms
O11-Cu1-N23
N20-Cu1-N23
O11-Cu1-N1
N20-Cu1-N1
N23-Cu1-N1
N4-Cu3-N10
N4-Cu3-O11
N4-Cu3-N1
N4-Cu3-N10b
N10-Cu3-O11
N10-Cu3-N1
N10-Cu3-N10b
O11-Cu3-N1
O11-Cu3-N10b
N1-Cu3-N10b
N13-Cu4-N16
ꢀ
deviation of 0.036 A with the copper atom 0.059(7) A
from the plane in the direction of N7, as usually observed
in square-pyramidal Cu(II) complexes. In addition, N7 is
bridged to Cu2, while O11 and N1 are both bridged to
Cu3. The coordination sphere of Cu2 consists of four
azides in a basal plane, all of which bridge to other copper
atoms. Thus, N13 and N16 are bonded to Cu4, N4 is
bonded to Cu3, and, as stated above, N7 is bonded to
Cu1. The square-pyramidal geometry around Cu2 is
completed by N9a (a = -x, 2-y, 2-z), which occupies
an axial position with a long Cu-N bond distance of
2.336(11) A. The rms deviation of donor atoms in the
basal plane is 0.044 A with Cu2 0.194(6) A from this plane
in the direction of the axial N9 atom. The azide
N7-N8-N9 is the only one to bridge in a μ-1,1,3 fashion.
As can be seen from Figures 3 and 4, two such azides
bridge two Cu2 atoms across a center of symmetry. Cu3
a Symmetry elements = -x, 2-y, 2-z. b Symmetry elements = -x,
1-y, 2-z.
shows a square-pyramidal geometry, where the axial
position is occupied by an azide N atom, N1, at
2.242(10) A that bridges Cu3 to Cu1. The basal plane
contains three azide N atoms: N4 connecting Cu3 with
Cu2 and two N10 (from two equivalent azide ligands)
connecting Cu3 with its symmetry-related Cu3 ion. The
fourth basal position is occupied by the O11 atom from the
Schiff base ligand, which connects Cu3 to Cu1 (Figure 3).
Here the basal plane is quite distorted, witha rms deviation
ꢀ
ꢀ
ꢀ
of 0.194 A. As is found for the other Cu(II) ions, Cu3 is
displaced from the basal plane toward the axial position by
0.133(4) A, Cu4 is located on an inversion center and has a
square-planar environment with four N atoms from four

Article
Inorganic Chemistry, Vol. 49, No. 14, 2010 6623
Figure 3. Centrosymmetric fragment of the polymeric structure of 3 with ellipsoids at 30% probability. Atoms Cu4 are positioned on additional centers of
symmetry. Symmetry transformation: b = -x, 1-y, 2-z, c = 1-x, 2-y, 2-z.
Figure 4. 2D polymeric structure of compound 3. Hydrogenatomsare omitted forclarity. Colorcode:Cu =lightblue, N=darkblue, O=red, C=white.
azide ligands, two N13 and two N16, connecting each Cu4
ion with its two Cu2 neighbors.
second bridging ligand (e.g., carboxylate)20 or by employ-
ing chelating diamine ligands, such as ethylenediamine and
its derivatives,6,21 or the use of more azido ligaþnd2s2 by
In summary, five of the six independent azide ligands
act as μ-1,1-N3 bridges, Cu1-N1-Cu3, Cu2-N4-Cu3,
Cu3-N10-Cu3, Cu2-N13-Cu4, and Cu2-N16-Cu4,
giving rise to a one-dimensional zigzag chain along the
crystallographic b axis. The only μ-1,1,3-N3 bridge is
N7-N8-N9, which connects two Cu3 atoms of neigh-
boring chains, forming a two-dimensional polymeric
structure, as shown in Figure 4. Two of the μ-1,1-N3
ligands are disordered at the terminal nitrogen atom, viz.,
N12 and N18, and in both cases two sites were refined,
each with occupancies of x, 1-x, with x refining to 0.66(2)
and 0.51(2), respectively.
adding a countercation, such as Csþ or N(CH3)4
.
As
far as we know, complex 3 is the first example where a
mononuclearcopper(II) Schiff base complex is extended to
a 2D coordination polymer by simply introducing copper-
(II) azide. In this structure the Cu(II) ions are present in
both square-planar and square-pyramidal geometries, de-
monstrating the flexibility of Cu(II) with coordination
numbers four and five.
Structure of [Cu2L(dmen)(N3)3]n (4). As mentioned ear-
lier, the structure of 4 has already been reported.8 So we
The overall structure of compound 3 can be viewed as
formed by five Cu(N3)2 andtwo[Cu(L)(N3)] units(identical
to compound 1), which are connected through phenoxo and
μ-1,1 and μ-1,1,3 azido bridges to form the unprecedented
two-dimensional structure shown in Figure 4. In this con-
text, it should be mentioned that the common strategies for
the extension of metal-azido assemblies are to introduce a
(20) Chen, Z.-L.; Jiang, C.-F.; Yan, W.-H.; Liang, F.-P.; Batten, S. R.
Inorg. Chem. 2009, 48, 4674, and references therein.
(21) Gu, Z.-G.; Zuo, J.-L.; You, X.-Z. Dalton Trans. 2007, 4067.
(22) (a) Goher, M. A. S.; Cano, J.; Journaux, Y.; Abu-Youssef, M. A. M.;
Mautner, F. A.; Escuer, A.; Vicente, R. Chem.;Eur. J. 2000, 6, 778.
(b) Mautner, F. A.; Cortes, R.; Lezama, L.; Rojo, T. Angew. Chem., Int. Ed.
1996, 35, 78. (c) Mautner, F. A.; Hanna, S.; Cortes, R.; Lezama, L.; Barandika,
M. G.; Rojo, T. Inorg. Chem. 1999, 38, 4647.

6624 Inorganic Chemistry, Vol. 49, No. 14, 2010
Naiya et al.
Table 4. Selected Bond Lengths (A) and Bond Angles (deg) for Complex 4
bond
distance (A)
bond
angle (deg)
Cu1-N1
Cu1-O11
Cu1-N20
Cu1-N23
Cu1-N9
Cu2-N31
Cu2-N7
Cu2-N4
Cu2-N34
Cu2-N3a
1.994(4)
1.902(3)
1.963(4)
2.049(4)
2.406(5)
1.985(4)
1.987(4)
1.999(4)
2.089(4)
2.440(5)
N7-Cu2-N34
N4-Cu2-N34
N31-Cu2-N3a
N7-Cu2-N3a
N4-Cu2-N3a
N34-Cu2-N3a
N31-Cu2-N7
N31-Cu2-N4
N7-Cu2-N4
N31-Cu2-N34
O11-Cu1-N20
O11-Cu1-N23
N20-Cu1-N23
O11-Cu1-N1
N20-Cu1-N1
N23-Cu1-N1
89.8(2)
174.1(2)
91.8(2)
95.5(2)
91.9(2)
92.3(2)
171.0(2)
91.1(2)
94.0(2)
84.6(2)
92.5(2)
172.2(2)
84.9(2)
91.4(2)
162.1(2)
88.9(2)
Figure 5. Two 1D chains in 4 connected through long asymmetric
μ-1,1-N3 bridges and H-bonds (shown as dotted lines). All hydrogen
atoms, except the amine ones implicated in the H-bonds, are omitted for
clarity. Color code: Cu = light blue, N = dark blue, O = red, C = black,
H = yellow.
a Symmetry elements = x, -1þy, z.
do not describe the structure in detail here. However our
crystallographic data are better, as the previous data were
collected at room temperature and only to a 2θ max of
50ꢀ. In addition there are significant differences in the cell
parameters, so we include the structural parameters in
Table 1 and bond distances and angles in Table 4. One
point that was missed by the previous authors is that the
μ-1,1-N3 azide bridges lead to a helical chain, as can be
seen in Figures 5 and 6, where Cu1 and Cu2 atoms are
connected through basal-axial μ-1,3-N3 bridges. These
chains are connected through long asymmetric μ-1,1-N3
bridges between two Cu2 ions of different chains and
through two intermolecular H-bonds, forming the double-
helical chain (Figures 5 and 6).
It is interesting to note that although several μ-1,3-N3-
bridged helical chains have been reported with Mn(II/III)
and Ni(II),23-25 there are very few examples of such helical
chains with Cu(II). In fact, a careful search in the CCDC
database shows that there are only three Cu(II) chains
connected through single μ-1,3-N3 bridges,26 and only one
of these26a presents a helical structure (the other two are
zigzag chains), although it has not been magnetically
characterized. Therefore, compound 4 is the first magne-
tically characterized example of a helical Cu(II) chain
connected through a single μ-1,3-N3 bridge.
Figure 6. Helical structure of complex 4: (a) Single 1D helical chain; (b)
two helical chains through long asymmetric μ-1,1-N3 bridges; (c) space-
filling view of a single helical chain.
ture, χmT shows an abrupt decrease, reaching a value of
ca. 0.48 emu K mol-1 at 2 K. This behavior indicates
3
3
that compound 1 presents a weak antiferromagnetic
coupling, responsible for the observed decrease of χmT
at low temperatures.
Although the structure of this compound shows the
presence of Cu(II) monomers, a closer inspection shows
that eachmonomerhasaclose,symmetry-related, neighbor
(1A and 1B, see Figure 1b) that presents a close C-H/π
contact from a methyl group of 1B to a phenyl ring of 1A.
Accordingly, we have fitted the magnetic properties to
the simple Bleaney-Bowers S = 1/2 dimer model.27 This
model reproduces very satisfactorily the magnetic proper-
ties of compound 1 (solid line in Figure 7) with g = 2.085(2)
and J = -1.77(2) cm-1 (Table 5, the Hamiltonian is
written as H = -JS1S2). As expected, the weak antiferro-
magnetic coupling observed in compound1 is confirmed by
the isothermal magnetization that shows, at 2 K and 5 T, a
value of ca. 1.2 μB, well below the expected one for two
noninteracting Cu(II) ions (Figure 8).
Magnetic Properties
[CuL(N3)] (1). The thermal variation of the molar
magnetic susceptibility per two Cu(II) ions times the
temperature (χmT) for compound 1 shows at room tem-
perature a value of ca. 0.80 emu K mol-1, the expected
3
3
value for two isolated Cu(II) S = 1/2 ions with g ≈ 2.07
(Figure 7). On cooling the sample, the χmT product
remains constant down to ca. 15 K. Below this tempera-
(23) (a) Ko, H. H.; Lim, J. H.; Kim, H. C.; Hong, C. S. Inorg. Chem. 2006,
45, 8847. (b) Yuan, M.; Zhao, F.; Zhang, W.; Wang, Z.-M.; Gao, S. Inorg. Chem.
2007, 46, 11235.
(24) Reddy, K. R.; Rajasekharan, M. V.; Tuchagues, J. P. Inorg. Chem.
1998, 37, 5978.
(25) (a) Monfort, M.; Ribas, J.; Solans, X.; Font-Bardıa, M. Inorg. Chem.
´
[CuL(N3)]2 (2). The thermal variation of the molar mag-
netic susceptibility per Cu(II) dimer times the temperature
1996, 35, 7633. (b) Ftibas, J.; Monfort, M.; Diaz, C.; Bastos, C.; Mer, C.; Solans,
X. Inorg. Chem. 1995, 34, 4986.
€
(26) (a) Cabort, A.; Therrien, B.; Bernauer, K.; Suss-Fink, G. Inorg.
Chim. Acta 2003, 349, 78. (b) Adhikary, C.; Mal, D.; Okamoto, K.; Chaudhuri,
S.; Koner, S. Polyhedron 2006, 25, 2191. (c) Yuan, C. L. Acta Crystallogr. Sect.
E 2007, 63, m3148.
(27) (a) Bleaney, B.; Blowers, K. D. Proc. R. Soc. (London) Ser. A 1952,
214, 451. (b) Kahn, O. Molecular Magnetism; VCH Publishers, 1993.

Article
Inorganic Chemistry, Vol. 49, No. 14, 2010 6625
Figure 9. Thermal variation of the χmT product per Cu(II) dimer for
compound 2. The solid line shows the best fit to the S = 1/2 dimer model
(see text).
Figure 7. Thermal variation of the χmT product for compound 1. The
solid line shows the best fit to the S = 1/2 dimer model (see text).
the presence of antiferromagnetic interactions in this
compound.
Table 5. Structural and Magnetic Parameters of Compounds 1-4
compound structure
bridge(s)
g
J (cm-1
)
[Cu7L2(N3)12]n (3). The thermal variation of the molar
magnetic susceptibility per seven Cu(II) ions times the
temperature (χmT) for compound 3 shows at room tem-
perature a value of ca. 2.70 emu K mol-1, close to that
1
2
3
Cu-dimer C-H/π-overlap
2.085
-1.77
-1.97
20.1
-68.1
-4.0
-1.35
-2.64
Cu-dimer single 1,10-N3 (asymmetric) 2.102
Cu-chain double 1,10-N3 (symmetric) 2.110
single 1,10-N3 (symmetric)
3
3
expected for seven isolated Cu(II) S = 1/2 ions with g = 2
(Figure 10). When cooling the sample, the χmT product
shows a continuous decrease to reach a value of ca. 2.5
emu K mol-1 at ca. 50 K and then a smoothness until 20 K
single oxo (symmetric)
Cu-chain single 1,3-N3 asymmetric
single 1,3-N3 asymmetric
4
2.186
3
3
(Figure 10). Below ca. 20 K the χmT product shows a more
abrupt decrease to reach a value of ca. 1.2 emu K mol-1 at
3
3
2 K. This behavior indicates the presence of predominant
antiferromagnetic interactions in this compound.
Since the structure of this compound shows a complex
2D structure with six different azide bridges, we have
necessarily made some approximations in order to fit the
magnetic properties. In a first step we have discarded the
interchain connection (through a long Cu-N bond of
2.336(11) A) and the Cu1-Cu2 asymmetric azide bridge
(through a long Cu-N bond of 2.357(10) A) since they
are the weakest ones. In this way, we can fit the magnetic
data to an alternating decorated chain, as indicated in the
inset of Figure 10. Since even this simplified model implies
the presence of up to four different exchange constants, in
a second approach we have assumed that the coupling
constant through Cu2-Cu4 (J24) is equal to that through
Cu3-Cu3 (J33). This assumption is based on the fact that
both double 1,1-N3 bridges show very similar structural
parameters such as Cu-N bond distances and Cu-
N-Cu bond angles; see below. With these simplifications
it is possible to reproduce the magnetic properties of com-
pound 3 by using a 14 Cu(II) center closed-chain model,28
corresponding to two heptameric units -Cu2-Cu4-
Cu2-Cu3[Cu1]-Cu3[Cu1]- (solid line in Figure 10)
Figure 8. Isothermal magnetization at 2 K for compounds 1 (per Cu2
unit), 2 (per Cu2 unit), 3 (per Cu7 unit), and 4 (per Cu2 unit).
(χmT) for compound 2 shows at room temperature a value
of ca. 0.82 emu K mol-1, expected for two isolated Cu-
3
3
(II) S = 1/2 ions with g = 2.09 (Figure 9). When cooling the
sample, χmT remains constant down to ca. 50 K and shows
a progressive decrease below this temperature to reach a
value of ca. 0.50 emu K mol-1 at ca. 2 K (Figure 9). This
3
3
behavior indicates that compound 2 also presents a weak
antiferromagnetic coupling, responsible for the progressive
decrease observed below ca. 50 K.
Since the structure of this compound shows the pre-
sence of Cu(II) dimers in which the Cu(II) ions are
connected through a long and asymmetric 1,10-N3 bridge
(Figure 2a), we have fitted the magnetic properties of this
compound to the Bleaney-Bowers dimer model for two
S=1/2 ions.27 This simple model reproduces very satis-
factorily the magnetic properties of compound 2 in the
whole temperature range (solid line in Figure 9) with
g = 2.102(2) and J = -1.97(3) cm-1 (Table 5, the Hamilto-
nian is written as H = -JS1S2). Confirmation of the
weak antiferromagnetic coupling found in compound 2
is provided by the isothermal magnetization at 2 K that
shows almost a linear field dependence, without reach-
ing saturation at 5 T (Figure 8). At 5 T the magnetization
value is ca. 1.2 μB, well below the expected one for two
noninteracting S = 1/2 Cu(II) ions (2 μB), confirming
with the following set of parameters: g = 2.110, J24
=
J33 = 20.1 cm-1 (corresponding to the double 1,10-N3
bridges), J23 = -68.1 cm-1 (corresponding to the single
1,10-N3 bridge), and J13 = -4.0 cm-1 (corresponding to
the single oxo bridge, Table 5). This model reproduces
very satisfactorily the magnetic properties of compound 3
in the temperature range 300-30 K. Below this tempera-
ture the experimental points are below the theoretical
ꢀ
(28) (a) Borras-Almenar, J. J.; Clemente-Juan, J. M.; Coronado, E.;
ꢀ
Lloret, F. Chem. Phys. Lett. 1997, 275, 79. (b) Borras-Almenar, J. J.;
Clemente-Juan, J. M.; Coronado, E.; Tsukerblat, B. S. Inorg. Chem. 1999, 38,
ꢀ
6081. (c) Borras-Almenar, J. J.; Clemente-Juan, J. M.; Coronado, E.; Tsukerblat,
B. S. J. Comput. Chem. 2001, 22, 985.

6626 Inorganic Chemistry, Vol. 49, No. 14, 2010
Naiya et al.
Figure 11. Thermal variation of the χmT product for compound 4. Solid
line shows the best fit to the alternating S = 1/2 chain model (see text).
Figure 10. Thermal variation of the χmT product per seven Cu(II) ions
for compound 3. The solid line shows the best fit to the S = 1/2 chain
model (see text). Inset shows the exchange pathways considered in the
model.
1.958(6) (Table 5); the Hamiltonian is written as H =
-JSiSj). As in the other compounds, the isothermal
magnetization measurements at 2 K (Figure 8) confirm
the presence of a weak antiferromagnetic coupling. Thus,
at 2 K, compound 4 shows an almost linear dependence of
the magnetization on the magnetic field and no saturation
for fields of 5 T. The magnetization value at 2 K per
Cu(II) dimer (ca. 1.7 μB) is below expected for two
noninteracting Cu(II) ions (ca. 2.0 μB), confirming the
presence of weak antiferromagnetic interactions.
The antiferromagnetic couplings found in the four
compounds can be readily explained from the magneto-
structural correlations established for 1,1-N3 and 1,3-N3
symmetric and asymmetric bridges.31,5e,5i Thus, the weak
antiferromagnetic coupling found in compound 1 can be
attributed to the close C-H/π contact from a methyl
group of monomer 1B to a phenyl ring of the monomer
1A. This interaction is expected to give rise to a weak
antiferromagnetic coupling, in agreement with the experi-
mental results.
ones, indicating, as expected, the presence of additional
antiferromagnetic interactions (mainly those that were
neglected in the first step). Unfortunately, the presence of
so many different exchange pathways precludes a more
accurate fit of the magnetic properties.
The antiferromagnetic nature of the coupling found in
compound 3 is also observed in the isothermal magneti-
zation at 2 K, which shows no saturation at 5 T with a
magnetization value of ca. 1.4 μB per Cu7 unit (Figure 8),
well below the expected one for seven noninteracting S =
1/2 Cu(II) ions (ca. 7 μB).
[Cu2L(dmen)(N3)3]n (4). The thermal variation of the
molar magnetic susceptibility per two Cu(II) ions times
the temperature (χmT) for compound 4 shows at room
temperature a value of ca. 0.90 emu K mol-1, a value
3
3
close to that expected for two isolated Cu(II) S = 1/2 ions
with g ≈ 2.19 (Figure 11). When cooling the sample, the
χmT product remains constant down to ca. 10 K. Below
this temperature, χmT shows an abrupt decrease, reaching
a value of ca. 0.44 emu K mol-1 at 2 K. This behavior
Compound 2 presents a Cu(II) dimer with an asym-
metric 1,10-N3 bridge (Figure 2a). DFT calculations
performed on this kind of bridge show that for a long
Cu-N bond distance of ca. 2.53 A the expected coupling
should be ca. -5 cm-1 for a double bridge (ca. -2.5 cm-1
for a single one), in agreement with the experimental
value obtained in this compound (Table 5).
3
3
indicates that compound 4 presents a weak antiferro-
magnetic coupling, responsible for the observed decrease
of χmT at low temperatures.
Since the structure of this compound shows the pre-
sence of a Cu(II) chain formed by two different Cu(II)
ions, Cu1 and Cu2, connected through two different
asymmetric alternating 1,3-N3 bridges (-N1-N2-N3-
and -N7-N8-N9-, Figure 5), we can use two dif-
ferent models to reproduce the magnetic properties. First,
if we assume that the two bridges are equivalent, then
we can use the regular S= 1/2 antiferromagnetic chain
model derived by Hatfield et al.29 Unfortunately, this
model is not able to reproduce the magnetic properties
adequately, no doubt because of significant variations in
the bridging angles (see below), and therefore we have
used a second model involving alternating S = 1/2 anti-
ferromagnetic chains, also derived by Hatfield et al.30
This model reproduces very satisfactorily the magnetic
properties of compound 4 (the solid line in Figure 11) with
the following set of parameters: g = 2.1858(7), J1 =
-1.35(1) cm-1 and J2 = -2.64(1) cm-1 (R = |J2|/|J1| =
Compound 3 has a much more complex structure since
there are six different bridges connecting the four inde-
pendent Cu(II) ions. Nevertheless, as already mentioned,
we can, in a first approach, assume that the long asym-
metric azide bridges are negligible as compared with the
short and symmetric ones, and, thus, the coupling scheme
is reduced to only four different coupling constants. A
further simplification can be made if we assume that the
structural parameters of the two double 1,10-N3 bridges
connecting Cu2-Cu4 and Cu3-Cu3 are very similar (see
Figure 12), and in order to reduce the number of adjus-
table parameters, we can assume that they are equivalent.
As already explained above, the two other bridges are a
symmetrical single 1,1-N3 bridge, connecting the Cu2 and
Cu3 atoms (Figure 12), and a single oxo bridge that
connects, together with an asymmetric 1,10-N3 bridge,
the Cu1 and Cu3 atoms (Figure 12).
ꢀ
(31) (a) Triki, S.; Gomez-Garcıa, C. J.; Ruiz, E.; Sala-Pala, J. Inorg.
´
(29) Brown, D. B.; Donner, J. A.; Hall, J. W.; Wilson, S. R.; Wilson, R. B.;
Hodgson, D. J.; Hatfield, W. E. Inorg. Chem. 1979, 18, 2635.
(30) Hall, J. W.; Marsh, W. E.; Weller, R. R.; Hatfield, W. E. Inorg. Chem.
1981, 20, 1033.
Chem. 2005, 44, 5501. (b) Ruiz, E.; Cano, J.; Alvarez, S.; Alemany, P. J. Am.
Chem. Soc. 1998, 120, 11122. (c) Fabrizi de Biani, F.; Ruiz, E.; Cano, J.; Novoa,
J. J.; Alvarez, S. Inorg. Chem. 2000, 39, 3221.

Article
Inorganic Chemistry, Vol. 49, No. 14, 2010 6627
and very weak (of a few cm-1), in agreement with the
experimental results. Furthermore, since this correlation
indicates that the longer the Cu-N bond distance, the
weaker the magnetic coupling, we can attribute the weaker
coupling constant (-1.35 cm-1) to the longer Cu2-N3
bond(2.440(5) A) and the strongerone (-2.64cm-1) tothe
shorter Cu1-N9 bond (2.406(5) A).
Conclusions
The rational synthesis of compounds 1-4 constitutes a
rare example where subtle changes in the synthetic conditions
(such as the crystallization temperature or the Cu(II) pre-
Figure 12. View of the coordination environment of the Cu(II) ions in
compound 3 showing the bridging bond distances (in A) and angles (in
deg). Color code: Cu = green, N = blue, O = pink.
-
cursor) or the Cu(II)/HL/N3 ratio lead to a monomer, a
dimer, a helical chain, and a hexagonal layer with the same
Schiff base ligand. Even more exceptional is the fact that the
four complexes can easily be interconverted in solution. The
structures of compounds 1 and 2 show that they differ only in
the presence of a weak asymmetric 1,1-N3 bridge in the dimer
2. It seemsthat lower temperaturesfacilitate suchassociation.
The composition and structure of 3 demonstrate that a
complex polymeric structure can be obtained by simply
inserting Cu(II) azide into a mononuclear Schiff base com-
plex. The bridging ability of the phenoxo oxygen atom
together with the flexible coordination modes of azide and
the coordination flexibility of Cu(II) play an important role
in the formation of such an atypical structure. The magnetic
couplings in compounds 1 and 2 are weak and antiferro-
magnetic, in agreement with the close C-H/π contacts
observed in 1 and the asymmetric 1,1-N3 bridge present in
2. Compound 3 is a unique example of an alternating and
decorated azide-bridged copper chain presenting ferro- and
antiferromagnetic couplings. The ferromagnetic coupling
found in the two symmetric 1,1-N3 bridges fully agrees with
previous magnetostructural correlations and DFT calcula-
tions. On the other hand, the weak antiferromagnetic cou-
plings found for the single 1,1-N3 and oxo bridges also agree
with the magnetostructural correlations in these kinds of
bridges. Compound 4 presents alternating weak antiferro-
magnetic couplings, as expected for such asymmetric 1,3-N3
bridges.
The weak antiferromagnetic coupling found for the
single oxo bridge (-4.0 cm-1) can be easily rationalized
from the magnetostructural correlations that predict a
weak antiferromagnetic coupling for this kind of
bridge.32 The additional asymmetric 1,10-N3 bridge con-
necting the Cu1 and Cu3 atoms is also expected to give
rise to a weak antiferromagnetic coupling given the long
Cu-N bond distance.31 The single symmetric 1,10-N3
bridge connecting the Cu2 and Cu3 atoms is more
difficult to correlate since this is a very rare 1,10-N3 bridge
with a very large Cu-N-Cu bond angle (116.3ꢀ). This
angle is expected to lead to strong antiferromagnetic
coupling, but the big dihedral angle between the basal
planes of the two Cu(II) ions connected through this
bridge, Cu2 with Cu3, is expected to significantly reduce
the antiferromangetic coupling constant, leading to the
observed moderate coupling (-68.1 cm-1). Finally, for
the double 1,10-N3 bridges connecting Cu2 with Cu4 and
Cu3 with the other symmetry-related Cu3 ion, magneto-
structural correlations and DFT calculations show that the
two main parameters determining the sign and magnitude
of the exchange coupling are the Cu-N bond distances
and the Cu-N-Cu bond angles.31b In the case of the 1,1-
N3 bridges of compound 3, the Cu-N bond distances (in
the range 1.92-2.01 A, see Figure 12) and the Cu-N-Cu
bond angles (in the range 100.6-101.8ꢀ) indicate that
the exchange coupling through these bridges should be
ca. 20 cm-1, in very good agreement with experiment.
The magnetic couplings found in compound 4 are also
easy to explain since the asymmetric 1,3-N3 bridge is well
known to originate weak antiferromagnetic couplings.5e,i
Thus, the correlation established for asymmetric 1,3-N3
bridges based on DFT calculations indicate that the
coupling constant mainly depends on the long Cu-N
bond distance.31a For bond distances in the range 2.4-
2.5 A this coupling is expected to be antiferromagnetic
Acknowledgment. We thank CSIR, Government of India
[Junior Research Fellowship to S.N., Sanction no. 09/028
(0702)/2008-EMR-I], the European Union (MAGMANet
ꢀ
Network of Excellence), the Spanish Ministerio de Educaciony
Ciencia (Projects MAT2007-61584 and Consolider-Ingenio
2010 CSD 2007-00010 in Molecular Nanoscience), and the
Generalitat Valenciana (Project PROMETEO/2009/095) for
financial support. We also thank EPSRC and the University
of Reading for funds for the X Calibur system.
Supporting Information Available: This material is available
free of charge via the Internet at http://pubs.acs.org.
(32) Chiari, B.; Helms, J. H.; Piovesana, O.; Tarantelli, T.; Zanazzi, P. F.
Inorg. Chem. 1986, 25, 2408.