Open and closed copper chain coordination polymers with alternating ferromagnetic and antiferromagnetic interactions

Abstract

Two copper(II) coordination polymers formulated {[Cu₃(HL)₂(MeOH)₂(H₂O)₂]·2H₂O}_n (1) and {[Cu₄(L)₂(H₂O)₈]·5H₂O}_n (2) were obtained utilizing the polycarboxylate ligand 2-(bis(carboxymethyl)amino)terephthalic acid (H₄L) and characterized by X-ray diffraction, FT-IR spectra, elemental analysis and EPR spectra. 1 crystallizes in P2₁/c space group and shows a closed 1D double zigzag chain composed of trinuclear building block [Cu₃(HL)₂(H₂O)₄]. 2 crystallizes in a chiral space groupP2₁ and exhibits an open chain structure with tetranuclear building block [Cu₄(L)₂(H₂O)₈]. Fitting the variable-temperature magnetic susceptibilities of two compounds gave g = 2.16, J = 4.25 cm⁻¹, zJ′ = −0.88 cm⁻¹ for 1, and g = 2.13, J₁ = −2.61 cm⁻¹, J₂ = 2.97 cm⁻¹, zJ′ = −0.12 cm⁻¹ for 2, respectively, indicating the alternating ferromagnetic and antiferromagnetic interactions.

---

Introduction

More and more attention has been focused on the development of novel coordination polymers, due to the general and significant values such as for gas storage, catalysis, sensors, magnetic materials, host–guest chemistry, etc.^1–5 To design such compounds, some known factors need to be mentioned. The topologies of diversified coordination polymers based on the molecular building block usually affected by the coordination geometry of the central metal ions, the organic ligands and even some subtle effects from solvent, template, concentration, temperature, counteranion, etc. So far, polycarboxylate compounds especially some with symmetric units serving as precursor ligands (or linkers) are extensively developing, which have been a class of recognized good candidates for assembling such coordination polymers.^6–9 The preferred selection for them may be due to the diversity of coordination fashions of carboxylic group and their abundance, diversity, multifunctionality, and overall stability.¹⁰ Investigations also reveal that copper ions were relatively easy to be embedded in such system,¹¹ owing to their diversity of coordination geometry and the special affinity to oxygen atoms. Up to now, the studies on magneto-structural relationship show that carboxylate bridged systems tend to be observed serving as a vital way for superexchange effects. Especially, various bridging modes such as syn–syn, syn–anti and anti–anti configuration and bridging tridentate to two metal centers, may lead to different magnetic interactions in paramagnetic metal ions.^12,13 In contrast, polycarboxylate ligands with asymmetric structure are less utilized in constructing organic–inorganic coordination polymers.¹⁴

Attracted by the various frameworks and interesting magnetic behavior of polycarboxylate complexes, we introduced 2-(bis(carboxymethyl)amino)terephthalic acid (H₄L) as a precursor ligand with the aim to explore some new magnetic systems. This ligand featuring a terephthalic acid adding an amino acetic acid at phenyl group, is first applied in the synthesis of coordination polymers, as shown in Scheme 1. Previous studies on this ligand lead to the thermodynamic aspect of some complexes without any definite structural information.¹⁵ Unlike most of the complexes containing carboxylate ligands, which were obtained under solvothermal conditions,^16a intense environment is unfavorable for amino acetic compounds to assembly coordination polymers.^16b,c Reacting with a copper salt under mild conditions, H₄L gave rise to the coordination polymers {[Cu₃(HL)₂(MeOH)₂(H₂O)₂]·2H₂O}_n (1) with a closed double zigzag chain and {[Cu₄(L)₂(H₂O)₈]·5H₂O}_n (2) with an open asymmetric chain. To our knowledge, these two chain models were rarely observed in previous Cu(II) complexes bridged by a carboxylate ligand. Both compounds have been characterized by X-ray diffraction analyses, variable temperature magnetic measurements, and X-band EPR analyses. In addition, the non-covalent force plays an important role in assembling the desired supramolecular structure.

Scheme 1 Ligand H₄L and its coordinate modes in complexes 1 (II) and 2 (III) and (IV)

Experiment

Materials and measurements

2-(bis(Carboxymethyl)amino)terephthalic acid (H₄L) was prepared as in a similar method in literature.^16a All other reagents were of A.R. grade obtained commercially and used without purification. Elemental analyses for carbon, hydrogen, and nitrogen were carried out on a Vario EL III elemental analyzer. The infrared spectrum was taken on an Bruker tensor 27 FT-IR spectrophotometer in the 4000–400 cm⁻¹ region, using KBr pellets. The variable temperature magnetic susceptibility data for the crystalline samples of the two complexes were measured on a Quantum Design MPMS-7 SQUID magnetometer over a temperature range f 2–300 K with an applied field of 5 kG. Diamagnetic corrections were made with Pascal's constants for all of the constituent atoms. The EPR spectra were measured on a BRUKER EMX-6/1 EPR spectrometer.

Preparation

Synthesis of {[Cu₃(HL)₂(MeOH)₂(H₂O)₂]·2H₂O}_n1.
Method I. Methanol solution (10 mL) of Cu(NO₃)₂·3H₂O (1.5 mmol, 0.362 g) was mixed with 10 mL water solution containing (0.5 mmol, 0.149 g) H₄L and 0.73 mL N(CH₃)₄OH, after stirring the mixture for about 2 h, then filtering it, and the filtration was allowed to stand in air for two days. A green precipitate was isolated and resolved with methanol and HCl at pH 3.5. Standing again for about two weeks, a green rectangular single crystal was obtained. Yield 15%. Elemental analysis, calc. C₂₆H₂₄N₂O₂₄Cu₃ (%) C 33.22, H 2.56, N 2.98. Found (%), C 33.29, H 2.60, N 2.87. FT-IR spectra (KBr pellets), 3563 (bs), 1584(vs), 1538(m), 1409(vs), 1314(m), 1279(m), 1248(s), 1177(m), 963(w), 933(w), 840(m), 775(s), 705(w), 653(w), 621 (w), 549(w).
Method II. When replacing Cu(NO₃)₂ and N(CH₃)₄OH for CuSO₄ with excess amount of p-chloride amino phenyl, the yield went up to 58%.

Synthesis of {[Cu₄(L)₂(H₂O)₈]·5H₂O}_n2. The first product of 2 was prepared by H₄L (0.25 mmol, 0.074 g) and KOH (1.5 mmol, 0.084 g) in 10 mL MeOH was added to (1.25 mmol, 0.300 g) Cu(NO₃)₂·3H₂O in 10 mL MeOH, following by filtering and allowing the filtrate to stand in air. A dark green block crystal was isolated after one month. The second method is similar to 1. A solution of H₄L (0.3 mmol, 0.089 g) and LiOH (0.50 mmol, 0.222 g) in 10 mL mixture of MeOH/H₂O (1 : 1) was added to a solution of CuCl₂·2H₂O (1 mmol, 0.170 g) and NaClO₄ (5 mmol, 0.700 g), and then stir it for 2 h. Green square crystal was obtained after leaving in air for one month. Yield (41%), elemental analysis, Calc C₂₄H₃₈N₂O₂₈Cu₄ (%) C 27.25, H 3.60, N 2.65. Found (%), C 27.16, H 3.64, N 2.72. FT-IR spectra (KBr pellets), 3394, 1581(vs), 1406(s), 1361(s), 1312(m), 1266(m), 962(w), 935(w), 771(m), 606(w).

X-Ray crystallography . Single crystals of 1 and 2 with suitable dimensions were mounted onto thin glass fibers, all the intensity data were collected at 294 K on a Bruker SMART 1000-CCD diffractometer (Mo Kα radiation, λ = 0.71073 Å) in ω scan modes. The structures were solved by the direct method followed by difference Fourier syntheses, and then refined by full-matrix least squares refinement on F² using SHELXTL.¹⁷ All nonhydrogen atoms were refined with anisotropic parameters while H atoms were placed in calculated positions and refined using a riding model. Crystallographic data and refinement parameters are presented in Table 1.

Table 1 Crystallographic data of complexes 1 and 2

	1	2
Empirical formula	C25H26N2O24Cu3	C24H34N2O28Cu4
M r	913.10	1058.69
T/K	294(2)	294(2)
Wavelength/Å	0.71073	0.71073
Crystal system	Monoclinic	Monoclinic
Space group	P21/c	P21
a/Å	8.903(5)	7.6207(10)
b/Å	23.565(14)	20.686(3)
c/Å	8.164(5)	11.6092(16)
β/°	111.178(9)	97.296
V/Å ^3	1597.1(17)	1815.3(4)
Z	2	2
D/Mg·m^−3	1.899	1.926
Absorption coefficient/mm^−1	2.083	2.420
F(000)	922	1064
Crystal size/mm	0.20 × 0.18 × 0.14	0.16 × 0.24 × 0.36
θ/°	1.73 to 25.02	1.77 to 25.02
Limiting indices	−10 ≤ h ≤ 10, −28 ≤ k ≤ 20, −9 ≤ l ≤ 9	−6 ≤ h ≤ 9 −23 ≤ k ≤ 24 −11 ≤ l ≤ 13
R(int)	0.0427	0.0276
Refinement method	Full-matrix least -squares on F^2	Full-matrix least-squares on F^2
Data/restraints /parameters	2826/5/268	5660/10/568
GOF on F^2	1.090	1.015
R1, wR2 [I > 2σ(I)]	0.0372, 0.1159	0.0280, 0.0644
R1, wR2 (all data)	0.0456, 0.1212	0.0310, 0.0657
Absolute structure parameter		0.001(9)
Largest diff. peak and hole/e Å^−3	1.488, −0.483	0.413, −0.534

---

Result and discussion

Structural description

{[Cu₃(HL)₂(MeOH)₂(H₂O)₂]·2H₂O}_n (1). 1 was synthesized by two methods. The structure of 1 has been characterized by X-ray diffraction analysis. The ORTEP structure and crystallographic data were shown in Fig. 1 and Table 1, respectively. As illustrated in Fig. 1, 1 was assembled by trinuclear [Cu₃(HL)₂(MeOH)₂(H₂O)₂] building blocks with HL³⁻ as skeleton ligands and two H₂O molecules in a crystal lattice. The trinuclear block with a C₂ symmetric unit is composed of three Cu²⁺ ions, two trivalent anionic HL³⁻, MeOH and two H₂O molecules.

Fig. 1 The coordination of complex 1 with a numbering scheme (a), symmetry code, A: −x, 0.5 + y, 0.5 −z. (b) The diagram of complex 1. Carbon atoms are labeled with medium grey circles, red dots represent oxygen atoms, nitrogen atoms are blue and cyan represents the copper atoms.

The three copper centers are located in different coordinate environment. Cu1 and Cu3 ions were five-coordinated in square pyramids of NO₄, and Cu2 was six-coordinated surrounded by an octahedron of O₆. Two HL³⁻ displayed the same tripodal coordination modes to seize the symmetry related Cu1 ions, in which the long arm acted by the phenyl carboxylate group offers one equatorial O atom. The residual chelating positions including two equatorial sites and one apical site were occupied by N1 and its two short arms contributing to two oxygen atoms, respectively. Then, a slightly distorted square pyramidal geometry around the centrosymmetric copper ion is completed along with the coordination of one H₂O molecule. Cu1 had a deviation of about 0.1953 Å from its NO₃ plane. The apical Cu1–O5 distance of 2.173(3) Å is 0.212 Å longer than the mean Cu–O distance of 1.951 Å in the equatorial plane, which is slightly shorter than 2.258 Å of the axial Cu–O distance in the square pyramidal geometry.¹³ Cu1–O and Cu1–N distances in the equatorial plane are in good agreement with those of the reported compounds.^13,18 The residual two O atoms of the amino acetic groups bind to one intrablock and one interblock Cu²⁺ centers along their square plane, respectively. That is, each HL³⁻ afforded one such carboxylic oxygen atom to link the adjacent trinuclear units through binding to the six coordinate copper(II) ions along their equatorial plane, meanwhile the two short arms at the N atom bridge both the intraunit and interunit Cu1 and Cu2viacarboxylate groups in a syn–anti fashion. One direct effect from the coordination mode is the formation of the perfect square plane CuO4, the other valuable result is that trinuclear units were linked into a 1D double zig-zag chain. The two axial positions of the Cu2 ion were occupied by two methanol molecules with a Cu–O distance of 2.568 Å, which is 0.610 Å longer than the mean Cu–O bond length of the equatorial plane. Obviously, the elongation should be owing to the Jahn–Teller effect of Cu2. The distances are 5.266, 5.338, 5.761 and 8.903 Å for Cu1⋯Cu2, Cu1⋯Cu2b, Cu1⋯Cu1b and Cu2⋯Cu2b, respectively. The terminal carboxylic group at the phenyl ring can not coordinate with copper(II) ions. The only role which it played in this compound is to link the neighboring chains with their O2 atoms viaH-bond and then extend to a 2D supermolecular framework as shown in Fig. 2. The O–H⋯O distance is 2.554 Å and the angle is 170.92°. The solvent water molecule was distributed in these tunnels staked by them.

Fig. 2 The 2D supramolecular structure of 1 linked viaH-bonds viewed along the b (left, the water molecules were omitted for clarity) and the a axis (right). Carbon atom, medium grey; oxygen, red circle; copper atoms, green and purple polyhedra.

{[Cu₄(L)₂(H₂O)₈]·5H₂O}_n (2). 2 can also be obtained through two methods at room temperature. The coordination mode and the crystallographic data were exhibited in Fig. 3 and Table 1, respectively. The X-ray diffraction analysis reveals that 2 is also a 1D coordinate polymer. Different from 1, this complex presents an open single chain with a second building unit of [Cu₄(L)₂(H₂O)₈]. As illustrated in Fig. 3, two tetravalent anionic L⁴⁻ acting as hexadentate ligands, four Cu²⁺ ions and eight coordinated H₂O molecules are involved in the unit. The environments around the four copper(II) ions are all square pyramidal. Two are in O₅ and half in NO₄ environment. For convenient description, exhibited in Scheme 1, we entitle the two coordination modes m₁ and m₂ shown by the L⁴⁻, respectively. In m₁, the carboxylic group bridges the Cu1 and Cu2 centers along the basal planes in a syn–anti fashion. The basal positions of Cu1 were occupied by two opposite H₂O molecules and two oxygen atoms from the carboxylic group. A water molecule occupies the apical site of Cu1. Besides being linked by such three-atom bridge with Cu1, the Cu2 center was connected to one amino N atom along the basal plane and the two carboxylic oxygen atoms O5 and O3 along the basal and axial directions. The coordination of O20 (H₂O) ultimately completed the Cu2 geometry. The environment composition of the Cu3 and Cu4 centers is similar to Cu1 and Cu2, respectively, except for the bridging group and bridging sites. L⁴⁻ linking the Cu3 and Cu4 ions was named m₂. In this mode, Cu3 and Cu4 were bridged together by the syn–anti bridged O11C22O22 (from amino acetic group) along the basal and axial directions, respectively. As a result, the square pyramidal Cu3 were rotated by 143.93° with respect to Cu1 geometry, while there is only an angle of 7.30° between Cu2 and Cu4 geometries. Furthermore, as a bridging unit, the delocalization within O11C22O22 made less distortion of Cu4 geometry than Cu2 geometry, which can be concluded from the crystallographic data. The change of bridging groups led to different inter-metallic distances as well, 4.816, 5.293, 9.086 and 8.980 Å for Cu1⋯Cu2, Cu3⋯Cu4, Cu2⋯Cu3 and Cu4⋯Cu1, respectively. Connected by the ending carboxyl groups, the second building blocks were extended to a 1D open chain structure.

Fig. 3 The ORTEP diagram of complex 2 with a numbering scheme. Hydrogen atoms have been omitted for clarity (a), and ball and stick diagram of 2 (b). Carbon atom, medium grey; oxygen, red circle; nitrogen atoms, blue circles; copper atoms, cyan circle.

An interesting feature of 2 is the supramolecular framework. Neighboring chains arranged at parallel directions are linked together via O20⋯O4 H-bonds and form a 2D net framework with rectangular grids of about 23.72 × 7.62 Å. As illustrated in Fig. 4, the grids are so large that they can allow other parallel chains to cross through, and result in an interpenetrating structure. In fact, the 2D interpenetrating nets were also linked by two nonconvalent forces, one is O–H⋯O H-bonds formed between the coordinate water of the basal plane in Cu1 and the uncoordinated carboxyl oxygen atom at the first site of m₂. All hydrogen bond distances and angles were listed in the ESI.† Another interaction is π⋯π stacking interactions. The interpenetration of the net and the existence of H-bonds shorten the distances and increase the overlap between the phenyl planes in m₁ and m₂. The calculation results show that upper overlapped m₁phenyl rings with m₂ have data as centroid–centroid 3.726 Å, inter-planar distance 3.681 Å and 3.531 Å, and dihedral angle 10.4°, respectively. The lower one with m₂ has data as centriod–centriod 3.963 Å, inter-planar distance 3.557 Å and 3.793 Å and the same dihedral angle as the former. All the above data fall in the normal range for π⋯π stacking,¹⁹ suggesting the existence of the π⋯π stacking interactions shown in Fig. 5.

Fig. 4 The interpenetrating framework of 2, (a) and (b): The topology framework. The long lines represent one repeated Cu₄ unit which interacts with the neighbouring parallel chain via an H-bond labelled with a short line.

Fig. 5 The polyhedral framework of 2 with π⋯π stacking interactions labeled with orange dashes.

TG analysis

The thermal weight measurements for the two compounds were carried out in the range 25–700 °C. 1 lost weight gradually. From 25 to 157 °C, the weight loss of 15.85% agrees well with the loss of two lattice waters, two coordinate waters and two methanol molecules (15.97%). The following weight loss in the range of 157–251 °C with a value of 12.96% was attributed to the rest coordinate water molecules and the part of broken carboxylic acid groups. Upon heating, the mass experienced abrupt and slow loss from 251 to 477 °C which corresponds to the decomposition of the organic ligand with the sum value of 40.40% (calc. 40.76%). The thermal analysis for 2 showed that the lattice water and coordinate water molecules were lost successive until 180 °C with a value of 15.56% (less than the calc. 17.05% because of the slow breaking of the short Cu–O bond in the Cu3 basal plane). Higher than 180 °C the tetravalent organic ligand was decomposed rapidly with a max rate at 280 °C, the total weight loss is 46.84% and finished at about 419 °C.

EPR spectra and magnetic properties

To ensure the purity of the crystals, PXRD measurements were performed, and the result agrees well with their structures ESI).† The X-band EPR spectra of 1 and 2 are displayed in Fig. 6.

Fig. 6 EPR spectra of 1 (left) and 2 (right). The three lines were experimental data at rt, simulated and experimental spectra at 120 K from up to down.

1 shows a three-line well resolved feature. The broad resonance lines represent a compromise between two lineshapes, which is typical of exchange-coupled extension.²⁰ Temperature dependence behavior of 1 was shown at 120 and 298 K. The signal at 2974 G became weak and broad as the temperature decreased, meanwhile the signals at 3099 and 3236 G got sharper than room temperature and accompanied the vanishing of a weak signal at 3145 G, indicating the existence of a magnetic coupling interaction. Compared to 1, 2 shows little change between 298 and 120 K with a three resonance line spectra as well. Cooling the samples only made the three peaks sharpen. The spectra were simulated by assuming S = 1/2 for compounds 1 and 2, respectively. The fitting procedure yields g_av = 2.16 and g_av = 2.15 for 1 and 2, respectively.

The temperature dependence of χ_mT (the χ_m was used to express the molar magnetic susceptibilities of the chain system) for 1 is presented in Fig. 7. The χ_mT value is 1.35 cm³K mol⁻¹ at 300 K, which is slightly larger than the expected value of 1.13 cm³K mol⁻¹ for three uncoupled Cu²⁺ ions (S = 1/2, g = 2.0). Upon cooling, the χ_mT values remain about the same value in the range of 300–100 K, then the χ_mT begin to rise up to the maximum at 16 K, indicating ferromagnetic behavior. As the temperature lowers, the plots decrease abruptly until reaching the minimum of 0.79 cm³K mol⁻¹ at 2 K, which may be a feature of antiferromagnetic behavior in this region.

Fig. 7 Temperature dependence of the magnetic susceptibilities χ_mvs. T (empty circle) and χ_mTvs. T (empty quadrangle) plots of 1 under an applied field of 5 kOe with the best-fitting data as red and purple lines, respectively.

From the structure of 1, the 1D double zigzag chain was assembled by centrasymmetric trimeric blocks, Cu2 ion was linked with the surrounding Cu1 centers via carboxylic groups in syn–anti fashions and both along the equatorial direction. However, the Cu₃ unit in each close ring was distinct. Besides the bridging fashion mentioned above, there is a syn–anti bridging mode along the equatorial and axial direction, respectively. As per the description above, such bridges link linear Cu₃ fragment together and complete the double zigzag chain. In view of this case, two different exchange coupling constant J values were considered, the intra-trinuclear can be expressed by J = J₁₂ = J₂₃, the other interactions was considered as an inter-unit coupling constant zJ′, applying the molecular-field approximation.²¹ The coupling mode was illustrated in Scheme 2. The exchange Hamiltonian used the modified form H = −2J(S₁·S₂ + S₂·S₃. The corresponding molar magnetic susceptibilities are expressed by eqn (1) and (2)^20b

χ_m = χ_tri/[1−χ_tri(2zJ^′/Ng²β²)] (2)

Scheme 2 The magnetic coupling modes of 1 (a) and 2 (b).

As a result, the fitting gave good agreement with the experimental plots with g = 2.16, J = 4.25 cm⁻¹, zJ′ = −0.88 cm⁻¹, Nα = 180 × 10⁻⁶ and R = 9.0 × 10⁻⁵(R = ∑│(χ_mT)_exp − (χ_mT)_calc│²/∑(χ_mT)_exp²), in which the value of g fully complies with that obtained from EPR analysis. The different coupling constants can be interpreted according to the Hatfield theory,²² and the plot details under the condition of J/zJ′ > 1 correspond to the rule drawn by Coronado.²³ The results revealed intrachain alternating ferromagnetic and antiferromagnetic interactions.^24a According to the structural information, the magneto-structural relations could be concluded as follows. The carboxylate group bridging pair of copper ions at the equatorial planes forced the two Cu ions out of the related carboxylate plane (0.1992 Å for Cu1 and 0.1195 Å for Cu2), which therefore led to ferromagnetic interactions.^24b However, the bridge from a basal position to an axial position resulted in antiferromagnetic interactions between the two copper ions owing to a nearly planar ⋯Cu⋯O⋯C⋯O⋯Cu⋯ unit (0.0439 Å for Cu1 and 0.0506 Å for Cu2). As a result, an alternating F/AF magnetic behavior was exhibited in this system. Syn–anticarboxylate bridged copper compounds tend to predict weak magnetic coupling. Although less than J = 12 for dominating ferromagnetic interactions in a copper compound containing a monoatomic bridge,²⁵ the main ferromagnetic coupling in 1 is also important.

Plots of χ_mvs. T and χ_mTvs. T of 2 were represented in Fig. 8. Measurement at room temperature gave a χ_mT value of 1.71 cm³K mol⁻¹ which slightly was larger than the four Cu²⁺ ions with a spin only value S = 1/2 (the theoretical value is 1.50 cm³K mol⁻¹). The χ_mTvs. T plot shows that χ_mT is a nearly constant at 1.70 cm³K mol⁻¹ with a tiny increase from 300 to 50 K, and then decreases slowly until 10 K, indicating the simultaneous presence of two independent magnetic interactions and reaching a similar balance in this temperature range. Reducing the temperature successively, the χ_mT value dropped abruptly to reach 1.08 cm³K mol⁻¹ at 2 K, suggesting that antiferromagnetic coupling strived for the absolute advantage at low temperature. The observation is in concordance with the fact that complex 2 contains two different (μ-carboxyl)dicopper(II) fragments, Cu1–COO–Cu2 and Cu3–COO–Cu4. Although both units show five-coordinate tetragonal pyramidal geometry and are bridged by the three atomic carboxylates, respectively, the different positions of the bridging groups, result in distinct angles of the axial z orbitals in the two metal centers of the two fragments. This phenomenon is similar to the cases reported by Gutiérrez-Zorrilla and Lezama²⁶ which also contain two kinds of independent magnetic interactions resulting from different structural conformations. An exact theoretical treatment of the experimental data for 2 is extremely complicated due to the highly asymmetric structure of this complex. Therefore, some approximations and evaluations are needed to obtain an operative expression for the magnetic susceptibilities. For reducing the number of adjustable parameters, all the interfragment magnetic interactions (J₃, J₄, J₅ and J₆) are not included owing to their long approaches. The Hamiltonian for this type of exchange is H = −2J₁S₁S₂ − J₂S₃S₄. The magnetic behavior of 2 can be regarded as the sum of the independent contributions of two dimmer fragments.²⁶ Thus, the χ_m can be expressed as eq.(3)

Fig. 8 Temperature dependence of the magnetic susceptibilities χ_mvs. T (empty triangle) and χ_mTvs. T plots (empty quadrangle) of 2 under an applied field of 5 kOe with the best-fitting data as purple lines and red lines, respectively.

Interchain exchange was accounted for the addition of a mean field correction term to eqn (3), the equation for the susceptibilities then has the form as eqn (4). J′ is the interchain exchange and z is the number of the nearest neighbors. The best fit of eqn (4) to the data in the range 2–300 K yielded g = 2.13, J₁ = −2.61 cm⁻¹, J₂ = 2.97 cm⁻¹, zJ′ = −0.12 cm⁻¹ and R = 1.0 × 10⁻⁵ (R = ∑│(χ_mT)_exp − (χ_mT)_calc│²/∑(χ_mT)_exp²). All magnetic parameters were allowed to vary freely in the fitting calculations. Consequently, the results confirm the assumption of F/AF behavior in 2.

Copper(II) complexes with F/AF alternating interactions are not often documented. To the best of our knowledge, most cases took place in some azide bridged copper(II) coordinate polymers.^21,27,28 There are some common features for these compounds with one particular example excluded:²⁸ (i) coexistence of two kinds of bridge modes EE and EO for the azide anion; (ii) closed chain with a dimeric copper unit; (iii) copper(II) ions alternatively bridged by double EE and EO arranged in cis-mode, being almost perpendicular to each other and therefore present in a μ-1,1,3- or μ₂-1,3-azide. Besides these species, several O bridged or mixed O and bpm bridged copper chains^23,25,29 have also been reported, characterized by an alternating chain with a dinuclear unit. All metal centers were located either in square pyramidal or octahedral geometry, whereas no hints of carboxylic acid (syn–anti) bridged copper(II) complexes with such character were observed previously in literature. The two compounds in this work are the first example of a double zig-zag chain and a rare case of open chains with alternating F/AF magnetic behavior

Conclusions

In conclusion, two copper coordination polymers have been obtained using several methods with a asymmetric polycarboxylate ligand and copper salt, realizing the closed double zig-zag chain which is unusually observed in copper chains reported previously, and an open asymmetric chain. The transformation from a closed to open structure not only changed the bridge group but shortened the intermetallic distances. Fitting of the temperature dependence of the magnetic susceptibility plots reveals that either the closed chain or the open one exhibited two magnetic coupling interactions within the chains, which offers two novel copper chain models with F/AF (ferromagnetic/antiferromagnetic) systems.

Acknowledgements

The authors gratefully acknowledge helpful discussion with Prof. Li-Cun Li. This work was supported by the National Natural Science Foundation of China (Grants 20631030 and 20425103), the NSF of Tianjin (Grant 06YFJZJC009000), the State Key Project of Fundamental Research of MOST (Grant 2007CB815305) and MOE (Grant 20060055039) of China

References

1. (a) S. Takamizawa, E. Nakata, H. Yokoyama, K. Mochizuki and W. Mori, Angew. Chem., Int. Ed., 2003, 42, 4331; (b) G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surble and I. Margiolaki, Science, 2005, 309, 2040; (c) Y. Liu, J. F. Eubank, A. J. Cairns, J. Eckert, V. C. Kravtsov, R. Luebke and M. Eddaoudi, Angew. Chem., Int. Ed., 2007, 46, 3278.
2. C. D. Wu, A. Hu, L. Zhang and W. Lin, J. Am. Chem. Soc., 2005, 127, 8940.
3. (a) H. J. Choi and M. P. Suh, J. Am. Chem. Soc., 2004, 126, 15844; (b) B. D. Chandler, A. P. Côté, D. T. Cramb, J. M. Hill and G. K. H. Shimizu, Chem. Commun., 2002, 1900.
4. (a) R. L. Carlin, Magnetochemistry, Springer-Verlag, Berlin, 1986; (b) O. Kahn, Molecular Magnetism, VCH, New York, 1993; (c) Molecular Magnets: Recent Highlights, ed. W. Linert and M. Verdaguer, Springer-Verlag, Wien, 2003.
5. (a) B. X. Dong, J. Peng, C. J. Gómez-García, S. Benmansour, H. Q. Jia and N. H. Hu, Inorg. Chem., 2007, 15, 5933; (b) S. Hermes, D. Zacher, A. Baunemann, C. Wöll and R. A. Fischer, Chem. Mater., 2007, 19, 2168; (c) C. Janiak, Dalton Trans., 2003, 2781.
6. D. Cheng, M. A. Khan and R. P. Houser, J. Chem. Soc., Dalton Trans., 2002, 4555.
7. (a) S. K. Ghosh and P. K. Bharadwaj, Inorg. Chem., 2004, 43, 5180; (b) N. L. Rosi, J. Eckert, M. Eddaoudi, D. T. Vodak, J. Kim, M. O'Keeffe and O. M. Yaghi, Science, 2003, 300, 1127.
8. M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. O'Keeffe and O. M. Yaghi, Science, 2002, 295, 469.
9. K. Kanamori, K. Ino, H. Maeda, K. Miyazaki, M. Fukagawa, J. Kumada, T. Eguchi and K. Okamotos, Inorg. Chem., 1994, 33, 5547.
10. S. Dominguez, A. Mederos, P. Gili, A. Rancel, A. E. Rivero, F. Birito, F. L. Ioret, X. Solans, C. Ruiz-Perez, M. L. Rodriguez and I. Birito, Inorg. Chim. Acta, 1997, 255, 367.
11. (a) K. E. Holmes, P. F. Kelly, S. H. Dale and M. R. J. Elsegood, CrystEngComm, 2005, 7, 202; (b) B. Paul, B. Zimmermann, K. M. Fromm and C. Janiak, Z. Anorg. Allg. Chem., 2004, 630, 1650; (c) Y.-F. Zhou, F. L. Jiang, D. Q. Yuan, B. L. Wu, R. H. Wang, Z. Z. Lin and M. C. Hong, Angew. Chem., Int. Ed., 2004, 43, 5665; (d) P. Cheng, S. P. Yan, C. Z. Xie, B. Zhao, X. Y. Chen, X. W. Liu, C. H. Li, D. Z. Liao, Z. H. Jiang and G. L. Wang, Eur. J. Inorg. Chem., 2004, 2369; (e) J. Yang, J. F. Ma, Y. Y. Liu, J. C. Ma, H. Q. Jia and N. H. Hu, Eur. J. Inorg. Chem., 2006, 1208; (f) R. Gheorghe, P. Cucos, M. Andruh, J. P. Costes, B. Donnadieu and S. Shova, Chem.–Eur. J., 2006, 12, 187; (g) P. Wang, C. N. Moorefield, M. Panzer and G. R. Newkome, Chem. Commun., 2005, 465.
12. (a) M. Murugesu, R. Clérac, C. E. Anson and A. K. Powell, J. Phys. Chem. Solids, 2004, 65, 667; (b) P. R. Levstein and R. Calvo, Inorg. Chem., 1990, 29, 1581.
13. M. L. Kirk, W. E. Hatfield, M. S. Lah, D. Kessissoglou, V. L. Pecoraro, L. W. Morgan and J. D. Petersen, J. Appl. Phys., 1991, 69, 6013.
14. (a) Y. Y. Zhang, B.-Q. Ma, S. Gao, J. R. Li, Q. D. Liu, G. H. Wen and X. X. Zhang, J. Chem. Soc., Dalton Trans., 2000, 2249; (b) Y. Yan, C. D. Wu, X. He, Y. Q. Sun and C. Z. Lu, Cryst. Growth Des., 2005, 2, 821.
15. (a) R. Krannich and E. Uhlig, Z. Anorg. Allg. Chem., 1973, 402, 285; (b) A. Martin and E. Uhlig, Z. Anorg. Allg. Chem., 1970, 376, 282.
16. (a) Y. Q. Xu, D. Q. Yuan, B. L. Wu, L. Han, M. Y. Wu, F. L. Jiang and M. C. Hong, Cryst. Growth Des., 2006, 5, 1168; (b) E. F. K. McCandlish, T. K. michael, J. A. Neal, E. C. Lingafelter and N. J. Rose, Inorg. Chem., 1978, 17, 1383; (c) J. Sanchiz, S. Dominguez, A. Mederos, F. Brito and J. M. Arrieta, Inorg. Chem., 1997, 26, 4108.
17. (a) G. M. Sheldrick, SHELXS-97, Program for X-ray Crystal Structuture solution, Göttinen University, Germany, 1997; (b) G. M. Sheldrick, SHELXL-97, Progrm for X-ray Crystal Structure Refinement, Göttingen University, Germany, 1997.
18. (a) L. Antolini, L. P Battaglia, G. B. Gavioli, A. B Corradi, G. Grandi, G. Marcotrigiano, L. Menabue and G. C. Pellacani, J. Am. Chem. Soc., 1983, 105, 4327; (b) M. J. Roman-Alpiste, J. D. Martin-Ramos, A. Castineiras-Campos, E. Bugella-Allamirano, A. G. Sicilia-Zafra, J. M. Gonzalez-Perez and J. Niclos-Gutierrez, Polyhedron, 1999, 18, 3341.
19. (a) D. L. Reger, J. R. Gardinier, R. F. Semeniuc and M. D. Smith, Dalton Trans., 2003, 1712; (b) J. C. Liu, G. C. Guo, J. S. Huang and X. Z. You, Inorg. Chem., 2003, 42, 235.
20. (a) A. Bencini, D. Gatteschi. EPR of exchange coupled systems. Springer-Verlag. Berlin, Heildelberg, Germany, 1990, p. 39; (b) A. Tamayo, J. Casabó, L. Escriche, P. González, C. Lodeiro, A. C. Rizzi, C. D. Brondino, M. C. G. Passeggi, R. Kivekäs and R. Sillanpä, Inorg. Chem., 2007, 46, 5665.
21. E. Q. Gao, S. Q. Bai, C. Fe. Wang, Y. F. Yue and C. H. Yan, Inorg. Chem., 2003, 42, 8456.
22. D. K. Towle, S. K. Hoffmann, W. E. Hatfield, P. Singh and P. Chaudhuri, Inorg. Chem., 1988, 27, 394.
23. J. J. Borras-Almenar, E. Coronado, J. Curely, R. Georges and J. C. Gianduzzo, Inorg. Chem., 1994, 33, 5171.
24. (a) A. Rodríguez-Fortea, P. Alemany, S. Alvarez and E. Ruiz, Chem.–Eur. J., 2001, 7, 627; (b) E. Colacio, J. M. Dominguez-Vera, J. P. Costes, R. Kivekäs, J. P. Laurent, J. Ruiz and M. Sundberg, Inorg. Chem., 1992, 31, 774.
25. P, S. Mukherjee, T. K. Maji, G. Mostafa, J. Ribas, M. S. E. Fallah and N. R. Chaudhuri, Inorg. Chem., 2001, 40, 928.
26. S. Reinso, P. Vitoria, J. M. Gutiérrez-Zorrilla, L. Lezama, J. M. Madariaga and L. S. Felices, Amaia Iturrospe, Inorg. Chem., 2007, 46, 4010.
27. (a) G. D. Munno, M. G. Lombardi, M. Julve, F. Lloret and J. Faus, Inorg. Chim. Acta, 1998, 282, 82; (b) T. K. Maji, P. S. Mukherjee, S. Koner, G. Mostafa, J. P. Tuchagues and N. R. Chaudhuri, Inorg. Chim. Acta, 2001, 314, 111; (c) Z. Shen, J. L. Zuo, Z. Yu, Y. Zhang, J. F. Bai, C. M. Che, H. K. Fun, J. J. Vittal and X. Z. You, J. Chem. Soc., Dalton Trans., 1999, 3393; (d) Y. S. You, C. S. Hong and K. M. Kim, Polyhedron, 2005, 2, 249; (e) Y. S. Xie, Q. L. Liu, H. Jiang, C. X. Du, X. L. Xu, M. G. Yu and Y. Zhu, New J. Chem., 2002, 26, 176.
28. H. Y. Shen, W. M. Bu, E. Q. Gao, D. Z. Liao, Z. H. Jiang, S. P. Yan and G. L. Wang, Inorg. Chem., 2000, 39, 396.
29. (a) G. D. Munno, M. Julve, F. Lloret, J. Faus, M. Verdaguer and A. Caneschi, Inorg. Chem., 1995, 34, 157; (b) I. Vasilevesky, N. R. Rose, R. Stenkamp and R. D. Willett, Inorg. Chem., 1991, 30, 4082; (c) S. Dalai, P. S. Mukherjee, G. Rogez, T. Mallah, M. G. B. Drew and N. R. Chaudhuri, Eur. J. Inorg. Chem., 2002, 12, 3292.

---

Footnote

† Electronic supplementary information (ESI) available :  Selected bond length and angles of 1 and 2 (Tables S1 and S2); H-bond data in 2 (Table S3); TGA and DTA analysis of 1 and 2 (Fig. S1); Experimental and simulated PXRD patterns for 1 and 2 (Fig. S2 and S3). CCDC reference numbers 656885 and 656886. For ESI and crystallographic data in CIF or other electronic format see DOI :  10.1039/b816161p

---