- Email: [email protected]
Magnetic and structural properties of a cyano-bridged 1D ruthenium(III)–nickel(II) coordination polymer
Polyhedron 25 (2006) 2705–2709 www.elsevier.com/locate/poly
Magnetic and structural properties of a cyano-bridged 1D ruthenium(III)–nickel(II) coordination polymer Joseph A. Duimstra b
a,1
, Charlotte L. Stern a, Thomas J. Meade
a,b,*
a Department of Chemistry, Northwestern University, Evanston, IL 60201, USA Departments of Biochemistry and Molecular and Cell Biology, Neurobiology and Physiology, and Radiology, Northwestern University, 2145 Sheridan Road, Evanston, IL 60201, USA
Received 14 February 2006; accepted 29 March 2006 Available online 15 April 2006
Abstract Reaction of [Ni(cyclam)](ClO4)2 (cyclam = 1,4,8,11-tetraazacyclotetradecane) with Na[Ru(salen)(CN)2] (2) (salen = ethylenebis(salicylimine)) generated the 1D coordination polymer [Ru(III)(salen)(CN)2][Ni(II)(cyclam)](ClO4) (1). X-ray diffraction studies of 1 revealed a structure consisting of alternating Ni(II) and Ru(III) ions bridged by cyanide ligands. The cyano C-terminus remained bound to the Ru(III) center while the perchlorate anion induced a deviation from linearity along the Ru–CN–Ni axis. Structural characterization of 2 disclosed a 2D framework where the sodium counterions bridge between [Ru(salen)(CN)2] units. Magnetic measurements of 1 indicate ferromagnetic coupling between Ni(II) and Ru(III) with no long-range magnetic order above 3 K. 2006 Elsevier Ltd. All rights reserved. Keywords: Molecular magnetism; Coordination polymers; Ruthenium; Nickel; Cyanides; Crystal structure
1. Introduction Low-dimensional metal-based coordination polymers have attracted attention due to unique magnetic properties of these compounds such as slow relaxation and quantum tunneling of the magnetization [1]. Compounds that display slow relaxation of their magnetization typically consist of paramagnetic metal ions bridged by a small coordinating ligand such as azide, oxalate or cyanide. Examples of azide bridged coordination polymers include [Co(2,2 0 -bithiazoline)(N3)2]n, a 1D polymer that displays single-chain magnetic behavior with a blocking temperature of 5 K [2], while a large class of 2D coordination polymers based on the oxalate (ox) ligand has also been
*
Corresponding author. Tel.: +1 847 491 2481; fax: +1 847 491 3832. E-mail addresses: [email protected], meadesec@chem. northwestern.edu (T.J. Meade). 1 Visiting Predoctoral Fellow from: Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA 91125, USA. 0277-5387/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.poly.2006.03.024
investigated. The oxalate compounds have the general formula [M(II)M 0 (III)(ox)3], where M is a 3d ion such as Mn, Fe, Co or Cu and M 0 = Cr [3,4] or Fe [5]. These compounds display a range of magnetic behavior depending on the different metals and cations involved. The cyanide ligand plays an important design role in the field of low dimensional magnetic coordination polymers as it effectively bridges two metal centers in a predictable linear arrangement [6,7]. A cyanide-bridged 1D single chain magnet comprised of Fe(III) and Co(II) ions arranged in a zigzag fashion displayed slow relaxation of the magnetization below 8 K [8], and superparamagnetic behavior has been observed in a cyanide-bridged Fe(III)2Cu(II) coordination polymer [9]. Incorporation of paramagnetic 4d ions into low dimensional coordination polymers is intriguing due to the more covalent metal-ligand bonds and increased spin–orbit coupling present in the second row transition metals compared to the 3d ions. Several studies have focused on molybdenum(III). Long and coworkers have used Mo(III)(Me3tacn)(CN)3 (Me3tacn is N,N 0 ,N00 -trimethyl-1,4,7-triazacyclononane) to
2706
J.A. Duimstra et al. / Polyhedron 25 (2006) 2705–2709
produce clusters of Mo(III) magnetically coupled to Ni(II) ions [10]. These clusters show higher magnetic anisotropy and better orbital overlap compared with the analogous compounds containing Cr(III) substituted for Mo(III). Other examples such as 2D K2Mn(II)3(H2O)6[Mo(III)(CN)7]2 Æ 6H2O have used MoðIIIÞðCNÞ7 4 as a building block [11]. Examples of low-dimensional ruthenium(III) based polymers are rare however, having been limited until very recently [12] to M(II)Ru(III)oxalato complexes (M is a 3d ion) which typically adopt a 2D network [13,14]. Initially the sign of the exchange coupling in the M(II)Ru(III)oxalato complex (M = Cu(II)) was thought not to obey simple symmetry considerations [13], but an alternative interpretation showed that the interaction does indeed conform to symmetry-based expectations [14]. We report here the first example of a structurally characterized 1D cyano-bridged coordination polymer containing Ru(III). The magnetic coupling in this compound conforms to symmetry-based expectations, showing ferromagnetic coupling between Ru(III) and Ni(II). No long-range ordering or slow relaxation of the magnetization was observed above 3 K. 2. Experimental 2.1. General [Ni(cyclam)](ClO4)2 was prepared according to the literature procedure [15]. Mass spectra were obtained by infusing a methanolic solution of the compound into a Varian 1200 L single quadrupole mass spectrometer. Infrared spectra were recorded of KBr pellets using a Biorad FTS-60 FTIR spectrometer. Elemental analyses were performed by Desert Analytics (Tucson, AZ). Variable temperature magnetic susceptibility measurements were taken of powder samples on a Quantum Design MPMS SQUID magnetometer in the temperature range of 2–300 K. Susceptibilities were corrected for the sample holder and the diamagnetism of the sample calculated from Pascal’s constants. Heat capacity measurements were performed in zero applied field on a Quantum Design PPMS. 2.2. Synthesis Na[Ru(salen)(CN)2], (2) was prepared according to Leung and Che [16]. IR (KBr pellet) mCN = 2099 cm1. MS (ESI) m/z (MNa) 420.1. Anal. Calc. for C18H14N4NaO2Ru Æ 1.5H2O C, 46.06; H, 3.65; N, 11.94. Found: C, 46.04; H, 3.64; N, 11.67%. Blue crystalline plates of 2 were grown from slow evaporation of a methanolic solution. [Ru(III)(salen)(CN)2][Ni(II)(cyclam)](ClO4), (1): [Ni(cyclam)](ClO4)2 (104 mg; 0.226 mmol) in 5 mL of acetonitrile were added dropwise to a solution of 100 mg (0.226 mmol) 2 in 5 mL of water. The resulting blue precipitate was isolated by centrifugation, washed with water and acetronitrile and dried in vacuo yielding 126 mg 1 (72%).
IR (KBr pellet) mCN = 2126 cm1. Anal. Calc. for C28H38ClN8NiO6Ru: C, 43.23; H, 4.92; N, 14.41. Found: C, 43.39; H, 5.11; N, 14.09%. Blue needle-like crystals of 1 suitable for X-ray diffraction resulted from slow diffusion of Ni(II)(cyclam)(ClO4)2 in acetonitrile into an aqueous solution of 2. 2.3. X-ray crystallography Crystallographic data were collected on a Bruker SMART 1000 X-ray diffractometer with CCD detector using graphite monochromated Mo Ka radiation ˚ ). Data sets for 1 and 2 were obtained at (k = 0.71073 A 153(2) K. For 2 the data were collected with a h range of 1.91–29.12. Data were obtained in 0.3 oscillations with 10 s exposures. For 1, the h range for data collection was 1.46–28.65 and data were collected in 0.3 oscillations with 25 s exposures. The crystal-to-detector distance was 50.00 mm with the detector at the 28 swing position for both 1 and 2. Data were processed using SAINT-NT from Bruker and were corrected for Lorentz and polarization effects. The structures were solved by direct methods [17], expanded using Fourier techniques and refined by fullmatrix least-squares on F2 [18]. The non-hydrogen atoms were refined anisotropically. Hydrogen atoms were included in idealized positions but not refined with the exception of the hydroxyl hydrogen in the methanol solvate of 2 and the hydrogen atoms of the water solvent molecule in 1. These H-atoms could not be located in reasonable positions. Crystallographic data for 1 and 2 are tabulated in Table 1 with selected bond lengths and angles given in Tables 2 and 3 for 2 and 1, respectively.
Table 1 Crystallographic data for 1 and 2
Empirical formula Formula weight Crystal system Space group ˚) a (A ˚) b (A ˚) c (A a () b () c () ˚ 3) V (A Z q (calc) (g cm3) l (mm1) Reflections collected/ unique [Rint] Reflections observed [I > 2r(I)] Data/restraints/parameters Goodness-of-fit on F2 Final R indices [I > 2r(I)] R indices (all data)
1 Æ H2O Æ MeCN
2 Æ MeOH
C30H43ClN9NiO7Ru 836.93 triclinic P-1 9.6253(11) 14.2047(17) 14.2607(17) 94.109(2) 99.171(2) 107.940(2) 1816.1(4) 2 1.531 1.064 16 822/8465 [0.0381] 6322 8465/0/446 1.085 R1 = 0.0600, wR2 = 0.1547 R1 = 0.0831, wR2 = 0.1748
C19H18N4NaO3Ru 474.43 monoclinic P21/c 7.3220(9) 12.4263(16) 20.662(3) 90 92.362(2) 90 1878.3(4) 4 1.674 0.886 17 596/4667 [0.0411] 4062 4667/0/254 1.072 R1 = 0.0235, wR2 = 0.0608 R1 = 0.0286, wR2 = 0.0624
J.A. Duimstra et al. / Polyhedron 25 (2006) 2705–2709 Table 2 ˚ ) and bond angles () for compound 2 Æ MeOH Selected bond lengths (A Ru(1)–N(2) Ru(1)–N(1) Ru(1)–O(2) Ru(1)–O(1) Ru(1)–C(17) Ru(1)–C(18) Ru(1)–Na(1) N(2)–Ru(1)–N(1) N(2)–Ru(1)–O(1) N(1)–Ru(1)–O(1) N(2)–Ru(1)–O(2) N(1)–Ru(1)–O(2) O(1)–Ru(1)–O(2) N(2)–Ru(1)–C(17) N(1)–Ru(1)–C(17) O(1)–Ru(1)–C(17) O(2)–Ru(1)–C(17) N(2)–Ru(1)–C(18) N(1)–Ru(1)–C(18) O(1)–Ru(1)–C(18) O(2)–Ru(1)–C(18)
1.9818(14) 1.9844(14) 2.0305(12) 2.0159(12) 2.0745(18) 2.0773(18) 3.2777(8) 83.69(6) 175.63(5) 92.34(5) 93.03(6) 176.25(5) 91.00(5) 84.72(6) 94.12(6) 93.77(6) 87.37(6) 89.01(6) 88.54(6) 92.72(6) 89.59(6)
Na(1)–O(1) Na(1)–O(3) Na(1)–N(3)#1 Na(1)–N(4)#2 Na(1)–O(2) N(3)–C(17) N(4)–C(18) N(3)–C(17)–Ru(1) N(4)–C(18)–Ru(1) C(17)–Ru(1)–C(18) O(1)–Na(1)–O(3) O(1)–Na(1)–N(3)#1 O(3)–Na(1)–N(3)#1 O(1)–Na(1)–N(4)#2 O(3)–Na(1)–N(4)#2 N(3)#1–Na(1)–N(4)#2 O(1)–Na(1)–O(2) O(3)–Na(1)–O(2) N(3)#1–Na(1)–O(2) N(4)#2–Na(1)–O(2) Ru(1)–O(1)–Na(1) Ru(1)–O(2)–Na(1)
2.3104(13) 2.3621(18) 2.4162(19) 2.4916(18) 2.6157(15) 1.153(2) 1.150(2) 170.45(16) 172.18(16) 172.88(7) 143.52(7) 114.84(6) 101.30(7) 92.19(5) 76.21(6) 115.05(6) 71.42(4) 95.90(6) 104.03(6) 140.91(6) 98.28(5) 88.81(5)
2707
and cyanide N4#2 from another nearby Ru(III) complex (at 1 + x, y, z) giving rise to a 2D layered structure (Fig. 2). As a result, the C17–Ru1–C18 angle is 172.88(7), while Ru1–C17–N3 and Ru1–C18–N4 display angles of 170.45(16) and 172.18(16), respectively. The coordination sphere of sodium is completed by salen oxygens O1 and O2 and a methanol solvent molecule. Comparison with trans-NBu4[Ru(III)(salen)(CN)2] shows the significant distortion arising from the sodium counterion. As a tetrabutylammonium salt, there is no interaction between Ru(III) centers resulting in nearly linear angles for C17–Ru1–C18 (178.3(7)), Ru1–C17–N3 (178(1)), and Ru1–C18–N4 (176(1)) [12]. The Ru–C and C„N bond lengths for 2 fall within the range observed by Yeung et al.
Symmetry transformations used to generate equivalent atoms: #1 1 x, 1/2 + y, 1/2 z; #2 1 + x, y, z.
Table 3 ˚ ) and bond angles () for compound Selected bond lengths (A 1 Æ MeCN Æ H2O Ru(1)–N(2) Ru(1)–N(1) Ru(1)–O(2) Ru(1)–O(1) Ru(1)–C(17) Ru(1)–C(18) N(3)–C(17) N(2)–Ru(1)–N(1) N(2)–Ru(1)–O(1) N(1)–Ru(1)–O(1) N(2)–Ru(1)–O(2) N(1)–Ru(1)–O(2) O(1)–Ru(1)–O(2) N(2)–Ru(1)–C(17) N(1)–Ru(1)–C(17) O(1)–Ru(1)–C(17) O(2)–Ru(1)–C(17) N(2)–Ru(1)–C(18) N(1)–Ru(1)–C(18) O(1)–Ru(1)–C(18) O(2)–Ru(1)–C(18)
1.994(4) 2.001(3) 2.016(3) 2.026(3) 2.058(4) 2.072(4) 1.135(6) 82.64(15) 171.97(14) 89.79(13) 90.96(15) 172.45(14) 96.78(13) 92.16(15) 86.98(14) 90.09(14) 89.27(14) 86.91(15) 95.58(16) 91.19(14) 88.04(15)
Ni(1)–N(5) Ni(1)–N(6) Ni(1)–N(3) Ni(2)–N(8) Ni(2)–N(7) Ni(2)–N(4) N(4)–C(18)
2.068(4) 2.075(4) 2.080(4) 2.057(4) 2.072(4) 2.134(4) 1.144(5)
N(3)–C(17)–Ru(1) N(4)–C(18)–Ru(1) C(17)–Ru(1)–C(18)
179.6(4) 173.5(4) 177.14(16)
N(5)–Ni(1)–N(6) N(5)–Ni(1)–N(3) N(6)–Ni(1)–N(3) N(8)–Ni(2)–N(7) N(8)–Ni(2)–N(4) N(7)–Ni(2)–N(4) C(17)–N(3)–Ni(1) C(18)–N(4)–Ni(2)
94.62(16) 89.61(14) 89.73(15) 85.78(17) 87.14(15) 91.67(14) 175.6(4) 159.9(4)
Fig. 1. Thermal ellipsoid (50%) plot of 2 Æ MeOH.
3. Results and discussion 3.1. Synthesis and structural characterization Crystals of Na[Ru(salen)(CN)2] (2) [16] suitable for Xray diffraction were grown from slow evaporation of a methanolic solution (Tables 1 and 2 and Fig. 1). The sodium counterion interacts with cyanide atom N3#1 from one adjacent Ru(III) complex (at 1 x, 1/2 + y, 1/2 z)
Fig. 2. View of 2D plane of 2 Æ MeOH with MeOH solvates removed for clarity. The 2D network lies in the plane defined by the crystallographic a and b axes.
2708
J.A. Duimstra et al. / Polyhedron 25 (2006) 2705–2709
Reaction of equimolar Ni(II)(cyclam) perchlorate [15] in solvents such as water, acetone and acetonitrile with 2 in water resulted in immediate precipitation of a fine steel blue particulate of formulation [Ru(III)(salen)(CN)2][Ni(II)(cyclam)](ClO4) (1). X-ray quality needles of 1 resulted from slow diffusion of Ni(II)(cyclam)(ClO4)2 in acetonitrile into an aqueous solution of 2 (Tables 1 and 3, Fig. 3). The C„N stretching frequency shifts to higher energy from 2099 cm1 in 2 to 2126 cm1 in 1, with concomitant bond ˚ for N3– shortening of the cyano moiety from 1.153(2) A ˚ C17 and 1.150(2) A for N4–C18 in 2 to 1.135(6) and ˚ , respectively, in compound 1. These results are 1.144(5) A comparable to that seen by Yeung et al. upon addition of NiCl2 and cyclam to RuðacacÞ2 ðCNÞ2 and are consistent with N coordination to the Ni(cyclam) subunit. The ˚ while Ru1–C18 is simRu1–C17 distance in 1 is 2.058(4) A ˚ ilar to that in 2 at 2.072(4) A. Around Ni, the Ni1–N3 dis˚ while Ni2–N4 is 2.134(4) A ˚ . With the tance is 2.080(4) A exception of Ni2–N4, these distances are identical to those reported for the trimeric Ni(cyclam)[Ru(acac)2(CN)2]2 [12]. The presence of the perchlorate anion results in a deviation from linearity of the –(Ru–CN–Ni–NC)– fragment giving rise to two different Ni(II) sites. The two Ni atoms lie on independent inversion centers. Along the metal cyanide axis, the Ni1 site is nearly linear with an angle of 175.6(4) between Ni1–N3–C17 and an angle of 179.6(4) between N3–C17–Ru1. The analogous measurements
Fig. 3. (a) Thermal ellipsoid plot (50%) of [Ru(III)(salen)(CN)2][Ni(II)(cyclam)](ClO4) Æ H2O Æ MeCN (1 Æ H2O Æ MeCN). Water and MeCN solvates and hydrogen atoms have been removed for clarity. (b) View of cationic portion of 1 Æ H2O Æ MeCN perpendicular to crystallographic a axis showing several repeat units. The chain axis runs along the vector difference of the crystallographic b and c axes.
around Ni2 display a more bent shape with 159.9(4) for Ni2–N4–C18 and 173.5(4) between N4–C18–Ru1. 3.2. Magnetic properties The results of variable temperature magnetic susceptibility measurements performed on powder samples of compounds 1 and 2 in a field of 1 kOe are shown in Fig. 4. At room temperature vMT for 2 is 0.51 cm3 mol1 K, considerably higher than the calculated spin-only value of 0.38 cm3 mol1 K for S = 1/2, but comparable to other monomeric Ru(III) complexes such as transPh4P[Ru(III)(acac)2(CN)2] [19] and Ru(III)(acac)3 [20]. The shape of the curve closely resembles that observed for trans-Ph4P[Ru(III)(acac)2(CN)2] and polymeric [CoCp2*][ZnRu(ox)3] [13], all of which display a gradual decrease in vMT from 300 K to approximately 15 K with a subsequent steep drop below 15 K. Correction for a temperature independent paramagnetism (TIP) contribution of 5 · 104 cm3 mol1 renders vMT independent of temperature from 300 to 100 K with a value of 0.36 cm3 mol1 K in agreement with the spin only value. As noted by Coronado et al. [13], the low temperature behavior below 15 K arises from spin–orbit coupling and incomplete quenching of the orbital contribution to the magnetic moment and can cause difficulties in the interpretation of susceptibility data containing Ru(III) ions as seen below. For chain 1, vMT is 1.57 cm3 mol1 K at 300 K, higher than the calculated spin only value of 1.38 cm3 mol1 K for uncorrelated S = 1 Ni(II) and S = 1/2 Ru(III) ions, but in good agreement with the calculated result of 1.51 cm3 mol1 K using the data obtained for 2. Applying the same TIP correction used in the analysis of 2 gives a vMT of 1.42 cm3 mol1 K at 300 K but does not remove the minimum at 55 K. The presence of the minimum would seem to indicate an antiferromagnetic interaction between Ru(III) and Ni(II). This conclusion was drawn from simi-
Fig. 4. The product of temperature and magnetic susceptibility as a function of temperature for compounds 1 and 2 in the range of 2–300 K at an applied field of 1 kOe. TIP correction is not included.
J.A. Duimstra et al. / Polyhedron 25 (2006) 2705–2709
2709
Ru(III) can be incorporated into a 1D coordination polymer and provide further evidence that simple symmetry considerations can be used to predict whether the ion will display a ferro- or antiferromagnetic exchange interaction. Acknowledgement We acknowledge Dr. Oleksandr Chernyashevskyy for assistance with the magnetic measurements. This research was supported by NIH R01-A147003. Appendix A. Supplementary material
Fig. 5. Magnetization of 1 as a function of applied field at 2.0 K.
lar data by Kahn and co-workers for the 2D (NBu4)[Cu(II)Ru(III)(ox)3] framework [14]. An alternative interpretation that appears to be more applicable in this case attributes the minimum to the anisotropic Ru(III) ion and indicates that the type of coupling must be determined through examination of the low temperature field-dependent magnetization data [13]. The magnetization of 1 as a function of applied field at 2 K is depicted in Fig. 5. At 50 kOe, the magnetization is nearly saturated with a value of 2.46 lB. The calculated saturation magnetization (Ms) for a spin-only system consisting of ferromagnetically coupled Ni(II) (S = 1) and Ru(III) (S = 1/2) is 3 lB. Antiferromagnetic coupling of the two ions results in a calculated Ms of 1 lB. The experimentally observed value clearly indicates ferromagnetic coupling, a conclusion that agrees with the results of Coronado et al. and with the expected orthogonality of magnetic orbitals between Ni(II) and Ru(III) [12,13]. The lower than expected saturation value can be attributed to the Ru(III) ion [13,20]. Zero-field cooled (zfc) and field-cooled (fc) magnetization experiments performed on 1 in applied fields of 3 and 50 Oe show no inflection point and little difference down to 3 K. These measurements indicate a lack of any type of long-range magnetic order above 3 K. Further evidence for the absence of 3D ordering above 2 K is given by the lack of a k-peak in the heat capacity measurements of 1 indicating that there is no magnetic phase transition. Compound 1 displays negligible hysteresis at 2 K. Although 1 does not display slow relaxation of the magnetization above 3 K, the salen ligand is readily amenable to synthetic modification which could be used to affect the magnetism of the compound through changes in the electronics at the Ru(III) center. It is interesting to note that a similar species, Ni(cyclam)[Ru(acac)2(CN)2]2, is trimeric [12], reflecting the marked effect small structural changes can have on the dimensionality of the resulting mixed metal system. The current results demonstrate that
Crystallographic data for structural analysis (cif files for the two structures) have been deposited with the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK, fax: +44 1223 336 033 or e-mail: [email protected] or http://www.ccdc.cam. ac.uk. Refer to CCDC number 286496 for 1 and number 286497 for 2. Supplementary data associated with this article can be found, in the online version, at doi:10.1016/ j.poly.2006.03.024. References [1] E. Coronado, F. Palacio, J. Veciana, Angew. Chem., Int. Ed. Engl. 42 (2003) 2570. [2] T.-F. Liu, D. Fu, S. Gao, Y.-Z. Zhang, H.-L. Sun, G. Su, Y.-J. Liu, J. Am. Chem. Soc. 125 (2003) 13976. [3] H. Tamaki, Z.J. Zhong, N. Matsumoto, S. Kida, M. Koikawa, N. Achiwa, Y. Hashimoto, H. Okawa, J. Am. Chem. Soc. 114 (1992) 6974. [4] S. Decurtins, H.W. Schmalle, H.R. Oswald, A. Linden, J. Ensling, P. Gu¨tlich, A. Hauser, Inorg. Chim. Acta 216 (1994) 65. [5] S. Decurtins, H.W. Schmalle, P. Schneuwly, R. Pellaux, J. Ensling, Mol. Cryst. Liq. Cryst. Sec. A 273 (1995) 605. [6] L.M.C. Beltran, J.R. Long, Acc. Chem. Res. 38 (2005) 325. [7] K.R. Dunbar, R.A. Heintz, Prog. Inorg. Chem. 45 (1997) 283. [8] L.M. Toma, R. Lescoue¨zec, F. Lloret, M. Julve, J. Vaissermann, M. Verdaguer, Chem. Commun. (2003) 1850. [9] S. Wang, J.-L. Zuo, S. Gao, Y. Song, H.-C. Zhou, Y.-Z. Zhang, X.-Z. You, J. Am. Chem. Soc. 126 (2004) 8900. [10] M.P. Shores, J.J. Sokol, J.R. Long, J. Am. Chem. Soc. 124 (2002) 2279. [11] J. Larionova, O. Kahn, S. Gohlen, L. Ouahab, R. Cle´rac, J. Am. Chem. Soc. 121 (1999) 3349. [12] W.-F. Yeung, P.-H. Lau, T.-C. Lau, H.-Y. Wei, H.-L. Sun, S. Gao, Z.-D. Chen, W.-T. Wong, Inorg. Chem. 44 (2005) 6579. [13] E. Coronado, J.R. Gala´n-Mascaro´s, C.J. Go´mez-Garcı´a, J.M. Martı´nez-Agudo, E. Martı´nez-Ferrero, J.C. Waerenborgh, M. Almeida, J. Solid State Chem. 159 (2001) 391. [14] J. Larionova, B. Mombelli, J. Sanchiz, O. Kahn, Inorg. Chem. 37 (1998) 679. [15] B. Bosnich, M.L. Tobe, G.A. Webb, Inorg. Chem. 4 (1965) 1109. [16] W.H. Leung, C.M. Che, Inorg. Chem. 28 (1989) 4619. [17] G.M. Sheldrick, SHELXS-97, Program for Crystal Structure Determination, University of Go¨ttingen, Germany, 1997. [18] G.M. Sheldrick, SHELXL-97, Program for the Refinement of Crystal Structures, University of Go¨ttingen, Germany, 1997. [19] W.-F. Yeung, W.-L. Man, W.-T. Wong, T.-C. Lau, S. Gao, Angew. Chem., Int. Ed. Engl. 40 (2001) 3031. [20] B.N. Figgis, P.A. Reynolds, K.S. Murray, B. Moubaraki, Aust. J. Chem. 51 (1998) 229.
Magnetic and structural properties of a cyano-bridged 1D ruthenium(III)–nickel(II) coordination polymer Joseph A. Duimstra b
a,1
, Charlotte L. Stern a, Thomas J. Meade
a,b,*
a Department of Chemistry, Northwestern University, Evanston, IL 60201, USA Departments of Biochemistry and Molecular and Cell Biology, Neurobiology and Physiology, and Radiology, Northwestern University, 2145 Sheridan Road, Evanston, IL 60201, USA
Received 14 February 2006; accepted 29 March 2006 Available online 15 April 2006
Abstract Reaction of [Ni(cyclam)](ClO4)2 (cyclam = 1,4,8,11-tetraazacyclotetradecane) with Na[Ru(salen)(CN)2] (2) (salen = ethylenebis(salicylimine)) generated the 1D coordination polymer [Ru(III)(salen)(CN)2][Ni(II)(cyclam)](ClO4) (1). X-ray diffraction studies of 1 revealed a structure consisting of alternating Ni(II) and Ru(III) ions bridged by cyanide ligands. The cyano C-terminus remained bound to the Ru(III) center while the perchlorate anion induced a deviation from linearity along the Ru–CN–Ni axis. Structural characterization of 2 disclosed a 2D framework where the sodium counterions bridge between [Ru(salen)(CN)2] units. Magnetic measurements of 1 indicate ferromagnetic coupling between Ni(II) and Ru(III) with no long-range magnetic order above 3 K. 2006 Elsevier Ltd. All rights reserved. Keywords: Molecular magnetism; Coordination polymers; Ruthenium; Nickel; Cyanides; Crystal structure
1. Introduction Low-dimensional metal-based coordination polymers have attracted attention due to unique magnetic properties of these compounds such as slow relaxation and quantum tunneling of the magnetization [1]. Compounds that display slow relaxation of their magnetization typically consist of paramagnetic metal ions bridged by a small coordinating ligand such as azide, oxalate or cyanide. Examples of azide bridged coordination polymers include [Co(2,2 0 -bithiazoline)(N3)2]n, a 1D polymer that displays single-chain magnetic behavior with a blocking temperature of 5 K [2], while a large class of 2D coordination polymers based on the oxalate (ox) ligand has also been
*
Corresponding author. Tel.: +1 847 491 2481; fax: +1 847 491 3832. E-mail addresses: [email protected], meadesec@chem. northwestern.edu (T.J. Meade). 1 Visiting Predoctoral Fellow from: Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA 91125, USA. 0277-5387/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.poly.2006.03.024
investigated. The oxalate compounds have the general formula [M(II)M 0 (III)(ox)3], where M is a 3d ion such as Mn, Fe, Co or Cu and M 0 = Cr [3,4] or Fe [5]. These compounds display a range of magnetic behavior depending on the different metals and cations involved. The cyanide ligand plays an important design role in the field of low dimensional magnetic coordination polymers as it effectively bridges two metal centers in a predictable linear arrangement [6,7]. A cyanide-bridged 1D single chain magnet comprised of Fe(III) and Co(II) ions arranged in a zigzag fashion displayed slow relaxation of the magnetization below 8 K [8], and superparamagnetic behavior has been observed in a cyanide-bridged Fe(III)2Cu(II) coordination polymer [9]. Incorporation of paramagnetic 4d ions into low dimensional coordination polymers is intriguing due to the more covalent metal-ligand bonds and increased spin–orbit coupling present in the second row transition metals compared to the 3d ions. Several studies have focused on molybdenum(III). Long and coworkers have used Mo(III)(Me3tacn)(CN)3 (Me3tacn is N,N 0 ,N00 -trimethyl-1,4,7-triazacyclononane) to
2706
J.A. Duimstra et al. / Polyhedron 25 (2006) 2705–2709
produce clusters of Mo(III) magnetically coupled to Ni(II) ions [10]. These clusters show higher magnetic anisotropy and better orbital overlap compared with the analogous compounds containing Cr(III) substituted for Mo(III). Other examples such as 2D K2Mn(II)3(H2O)6[Mo(III)(CN)7]2 Æ 6H2O have used MoðIIIÞðCNÞ7 4 as a building block [11]. Examples of low-dimensional ruthenium(III) based polymers are rare however, having been limited until very recently [12] to M(II)Ru(III)oxalato complexes (M is a 3d ion) which typically adopt a 2D network [13,14]. Initially the sign of the exchange coupling in the M(II)Ru(III)oxalato complex (M = Cu(II)) was thought not to obey simple symmetry considerations [13], but an alternative interpretation showed that the interaction does indeed conform to symmetry-based expectations [14]. We report here the first example of a structurally characterized 1D cyano-bridged coordination polymer containing Ru(III). The magnetic coupling in this compound conforms to symmetry-based expectations, showing ferromagnetic coupling between Ru(III) and Ni(II). No long-range ordering or slow relaxation of the magnetization was observed above 3 K. 2. Experimental 2.1. General [Ni(cyclam)](ClO4)2 was prepared according to the literature procedure [15]. Mass spectra were obtained by infusing a methanolic solution of the compound into a Varian 1200 L single quadrupole mass spectrometer. Infrared spectra were recorded of KBr pellets using a Biorad FTS-60 FTIR spectrometer. Elemental analyses were performed by Desert Analytics (Tucson, AZ). Variable temperature magnetic susceptibility measurements were taken of powder samples on a Quantum Design MPMS SQUID magnetometer in the temperature range of 2–300 K. Susceptibilities were corrected for the sample holder and the diamagnetism of the sample calculated from Pascal’s constants. Heat capacity measurements were performed in zero applied field on a Quantum Design PPMS. 2.2. Synthesis Na[Ru(salen)(CN)2], (2) was prepared according to Leung and Che [16]. IR (KBr pellet) mCN = 2099 cm1. MS (ESI) m/z (MNa) 420.1. Anal. Calc. for C18H14N4NaO2Ru Æ 1.5H2O C, 46.06; H, 3.65; N, 11.94. Found: C, 46.04; H, 3.64; N, 11.67%. Blue crystalline plates of 2 were grown from slow evaporation of a methanolic solution. [Ru(III)(salen)(CN)2][Ni(II)(cyclam)](ClO4), (1): [Ni(cyclam)](ClO4)2 (104 mg; 0.226 mmol) in 5 mL of acetonitrile were added dropwise to a solution of 100 mg (0.226 mmol) 2 in 5 mL of water. The resulting blue precipitate was isolated by centrifugation, washed with water and acetronitrile and dried in vacuo yielding 126 mg 1 (72%).
IR (KBr pellet) mCN = 2126 cm1. Anal. Calc. for C28H38ClN8NiO6Ru: C, 43.23; H, 4.92; N, 14.41. Found: C, 43.39; H, 5.11; N, 14.09%. Blue needle-like crystals of 1 suitable for X-ray diffraction resulted from slow diffusion of Ni(II)(cyclam)(ClO4)2 in acetonitrile into an aqueous solution of 2. 2.3. X-ray crystallography Crystallographic data were collected on a Bruker SMART 1000 X-ray diffractometer with CCD detector using graphite monochromated Mo Ka radiation ˚ ). Data sets for 1 and 2 were obtained at (k = 0.71073 A 153(2) K. For 2 the data were collected with a h range of 1.91–29.12. Data were obtained in 0.3 oscillations with 10 s exposures. For 1, the h range for data collection was 1.46–28.65 and data were collected in 0.3 oscillations with 25 s exposures. The crystal-to-detector distance was 50.00 mm with the detector at the 28 swing position for both 1 and 2. Data were processed using SAINT-NT from Bruker and were corrected for Lorentz and polarization effects. The structures were solved by direct methods [17], expanded using Fourier techniques and refined by fullmatrix least-squares on F2 [18]. The non-hydrogen atoms were refined anisotropically. Hydrogen atoms were included in idealized positions but not refined with the exception of the hydroxyl hydrogen in the methanol solvate of 2 and the hydrogen atoms of the water solvent molecule in 1. These H-atoms could not be located in reasonable positions. Crystallographic data for 1 and 2 are tabulated in Table 1 with selected bond lengths and angles given in Tables 2 and 3 for 2 and 1, respectively.
Table 1 Crystallographic data for 1 and 2
Empirical formula Formula weight Crystal system Space group ˚) a (A ˚) b (A ˚) c (A a () b () c () ˚ 3) V (A Z q (calc) (g cm3) l (mm1) Reflections collected/ unique [Rint] Reflections observed [I > 2r(I)] Data/restraints/parameters Goodness-of-fit on F2 Final R indices [I > 2r(I)] R indices (all data)
1 Æ H2O Æ MeCN
2 Æ MeOH
C30H43ClN9NiO7Ru 836.93 triclinic P-1 9.6253(11) 14.2047(17) 14.2607(17) 94.109(2) 99.171(2) 107.940(2) 1816.1(4) 2 1.531 1.064 16 822/8465 [0.0381] 6322 8465/0/446 1.085 R1 = 0.0600, wR2 = 0.1547 R1 = 0.0831, wR2 = 0.1748
C19H18N4NaO3Ru 474.43 monoclinic P21/c 7.3220(9) 12.4263(16) 20.662(3) 90 92.362(2) 90 1878.3(4) 4 1.674 0.886 17 596/4667 [0.0411] 4062 4667/0/254 1.072 R1 = 0.0235, wR2 = 0.0608 R1 = 0.0286, wR2 = 0.0624
J.A. Duimstra et al. / Polyhedron 25 (2006) 2705–2709 Table 2 ˚ ) and bond angles () for compound 2 Æ MeOH Selected bond lengths (A Ru(1)–N(2) Ru(1)–N(1) Ru(1)–O(2) Ru(1)–O(1) Ru(1)–C(17) Ru(1)–C(18) Ru(1)–Na(1) N(2)–Ru(1)–N(1) N(2)–Ru(1)–O(1) N(1)–Ru(1)–O(1) N(2)–Ru(1)–O(2) N(1)–Ru(1)–O(2) O(1)–Ru(1)–O(2) N(2)–Ru(1)–C(17) N(1)–Ru(1)–C(17) O(1)–Ru(1)–C(17) O(2)–Ru(1)–C(17) N(2)–Ru(1)–C(18) N(1)–Ru(1)–C(18) O(1)–Ru(1)–C(18) O(2)–Ru(1)–C(18)
1.9818(14) 1.9844(14) 2.0305(12) 2.0159(12) 2.0745(18) 2.0773(18) 3.2777(8) 83.69(6) 175.63(5) 92.34(5) 93.03(6) 176.25(5) 91.00(5) 84.72(6) 94.12(6) 93.77(6) 87.37(6) 89.01(6) 88.54(6) 92.72(6) 89.59(6)
Na(1)–O(1) Na(1)–O(3) Na(1)–N(3)#1 Na(1)–N(4)#2 Na(1)–O(2) N(3)–C(17) N(4)–C(18) N(3)–C(17)–Ru(1) N(4)–C(18)–Ru(1) C(17)–Ru(1)–C(18) O(1)–Na(1)–O(3) O(1)–Na(1)–N(3)#1 O(3)–Na(1)–N(3)#1 O(1)–Na(1)–N(4)#2 O(3)–Na(1)–N(4)#2 N(3)#1–Na(1)–N(4)#2 O(1)–Na(1)–O(2) O(3)–Na(1)–O(2) N(3)#1–Na(1)–O(2) N(4)#2–Na(1)–O(2) Ru(1)–O(1)–Na(1) Ru(1)–O(2)–Na(1)
2.3104(13) 2.3621(18) 2.4162(19) 2.4916(18) 2.6157(15) 1.153(2) 1.150(2) 170.45(16) 172.18(16) 172.88(7) 143.52(7) 114.84(6) 101.30(7) 92.19(5) 76.21(6) 115.05(6) 71.42(4) 95.90(6) 104.03(6) 140.91(6) 98.28(5) 88.81(5)
2707
and cyanide N4#2 from another nearby Ru(III) complex (at 1 + x, y, z) giving rise to a 2D layered structure (Fig. 2). As a result, the C17–Ru1–C18 angle is 172.88(7), while Ru1–C17–N3 and Ru1–C18–N4 display angles of 170.45(16) and 172.18(16), respectively. The coordination sphere of sodium is completed by salen oxygens O1 and O2 and a methanol solvent molecule. Comparison with trans-NBu4[Ru(III)(salen)(CN)2] shows the significant distortion arising from the sodium counterion. As a tetrabutylammonium salt, there is no interaction between Ru(III) centers resulting in nearly linear angles for C17–Ru1–C18 (178.3(7)), Ru1–C17–N3 (178(1)), and Ru1–C18–N4 (176(1)) [12]. The Ru–C and C„N bond lengths for 2 fall within the range observed by Yeung et al.
Symmetry transformations used to generate equivalent atoms: #1 1 x, 1/2 + y, 1/2 z; #2 1 + x, y, z.
Table 3 ˚ ) and bond angles () for compound Selected bond lengths (A 1 Æ MeCN Æ H2O Ru(1)–N(2) Ru(1)–N(1) Ru(1)–O(2) Ru(1)–O(1) Ru(1)–C(17) Ru(1)–C(18) N(3)–C(17) N(2)–Ru(1)–N(1) N(2)–Ru(1)–O(1) N(1)–Ru(1)–O(1) N(2)–Ru(1)–O(2) N(1)–Ru(1)–O(2) O(1)–Ru(1)–O(2) N(2)–Ru(1)–C(17) N(1)–Ru(1)–C(17) O(1)–Ru(1)–C(17) O(2)–Ru(1)–C(17) N(2)–Ru(1)–C(18) N(1)–Ru(1)–C(18) O(1)–Ru(1)–C(18) O(2)–Ru(1)–C(18)
1.994(4) 2.001(3) 2.016(3) 2.026(3) 2.058(4) 2.072(4) 1.135(6) 82.64(15) 171.97(14) 89.79(13) 90.96(15) 172.45(14) 96.78(13) 92.16(15) 86.98(14) 90.09(14) 89.27(14) 86.91(15) 95.58(16) 91.19(14) 88.04(15)
Ni(1)–N(5) Ni(1)–N(6) Ni(1)–N(3) Ni(2)–N(8) Ni(2)–N(7) Ni(2)–N(4) N(4)–C(18)
2.068(4) 2.075(4) 2.080(4) 2.057(4) 2.072(4) 2.134(4) 1.144(5)
N(3)–C(17)–Ru(1) N(4)–C(18)–Ru(1) C(17)–Ru(1)–C(18)
179.6(4) 173.5(4) 177.14(16)
N(5)–Ni(1)–N(6) N(5)–Ni(1)–N(3) N(6)–Ni(1)–N(3) N(8)–Ni(2)–N(7) N(8)–Ni(2)–N(4) N(7)–Ni(2)–N(4) C(17)–N(3)–Ni(1) C(18)–N(4)–Ni(2)
94.62(16) 89.61(14) 89.73(15) 85.78(17) 87.14(15) 91.67(14) 175.6(4) 159.9(4)
Fig. 1. Thermal ellipsoid (50%) plot of 2 Æ MeOH.
3. Results and discussion 3.1. Synthesis and structural characterization Crystals of Na[Ru(salen)(CN)2] (2) [16] suitable for Xray diffraction were grown from slow evaporation of a methanolic solution (Tables 1 and 2 and Fig. 1). The sodium counterion interacts with cyanide atom N3#1 from one adjacent Ru(III) complex (at 1 x, 1/2 + y, 1/2 z)
Fig. 2. View of 2D plane of 2 Æ MeOH with MeOH solvates removed for clarity. The 2D network lies in the plane defined by the crystallographic a and b axes.
2708
J.A. Duimstra et al. / Polyhedron 25 (2006) 2705–2709
Reaction of equimolar Ni(II)(cyclam) perchlorate [15] in solvents such as water, acetone and acetonitrile with 2 in water resulted in immediate precipitation of a fine steel blue particulate of formulation [Ru(III)(salen)(CN)2][Ni(II)(cyclam)](ClO4) (1). X-ray quality needles of 1 resulted from slow diffusion of Ni(II)(cyclam)(ClO4)2 in acetonitrile into an aqueous solution of 2 (Tables 1 and 3, Fig. 3). The C„N stretching frequency shifts to higher energy from 2099 cm1 in 2 to 2126 cm1 in 1, with concomitant bond ˚ for N3– shortening of the cyano moiety from 1.153(2) A ˚ C17 and 1.150(2) A for N4–C18 in 2 to 1.135(6) and ˚ , respectively, in compound 1. These results are 1.144(5) A comparable to that seen by Yeung et al. upon addition of NiCl2 and cyclam to RuðacacÞ2 ðCNÞ2 and are consistent with N coordination to the Ni(cyclam) subunit. The ˚ while Ru1–C18 is simRu1–C17 distance in 1 is 2.058(4) A ˚ ilar to that in 2 at 2.072(4) A. Around Ni, the Ni1–N3 dis˚ while Ni2–N4 is 2.134(4) A ˚ . With the tance is 2.080(4) A exception of Ni2–N4, these distances are identical to those reported for the trimeric Ni(cyclam)[Ru(acac)2(CN)2]2 [12]. The presence of the perchlorate anion results in a deviation from linearity of the –(Ru–CN–Ni–NC)– fragment giving rise to two different Ni(II) sites. The two Ni atoms lie on independent inversion centers. Along the metal cyanide axis, the Ni1 site is nearly linear with an angle of 175.6(4) between Ni1–N3–C17 and an angle of 179.6(4) between N3–C17–Ru1. The analogous measurements
Fig. 3. (a) Thermal ellipsoid plot (50%) of [Ru(III)(salen)(CN)2][Ni(II)(cyclam)](ClO4) Æ H2O Æ MeCN (1 Æ H2O Æ MeCN). Water and MeCN solvates and hydrogen atoms have been removed for clarity. (b) View of cationic portion of 1 Æ H2O Æ MeCN perpendicular to crystallographic a axis showing several repeat units. The chain axis runs along the vector difference of the crystallographic b and c axes.
around Ni2 display a more bent shape with 159.9(4) for Ni2–N4–C18 and 173.5(4) between N4–C18–Ru1. 3.2. Magnetic properties The results of variable temperature magnetic susceptibility measurements performed on powder samples of compounds 1 and 2 in a field of 1 kOe are shown in Fig. 4. At room temperature vMT for 2 is 0.51 cm3 mol1 K, considerably higher than the calculated spin-only value of 0.38 cm3 mol1 K for S = 1/2, but comparable to other monomeric Ru(III) complexes such as transPh4P[Ru(III)(acac)2(CN)2] [19] and Ru(III)(acac)3 [20]. The shape of the curve closely resembles that observed for trans-Ph4P[Ru(III)(acac)2(CN)2] and polymeric [CoCp2*][ZnRu(ox)3] [13], all of which display a gradual decrease in vMT from 300 K to approximately 15 K with a subsequent steep drop below 15 K. Correction for a temperature independent paramagnetism (TIP) contribution of 5 · 104 cm3 mol1 renders vMT independent of temperature from 300 to 100 K with a value of 0.36 cm3 mol1 K in agreement with the spin only value. As noted by Coronado et al. [13], the low temperature behavior below 15 K arises from spin–orbit coupling and incomplete quenching of the orbital contribution to the magnetic moment and can cause difficulties in the interpretation of susceptibility data containing Ru(III) ions as seen below. For chain 1, vMT is 1.57 cm3 mol1 K at 300 K, higher than the calculated spin only value of 1.38 cm3 mol1 K for uncorrelated S = 1 Ni(II) and S = 1/2 Ru(III) ions, but in good agreement with the calculated result of 1.51 cm3 mol1 K using the data obtained for 2. Applying the same TIP correction used in the analysis of 2 gives a vMT of 1.42 cm3 mol1 K at 300 K but does not remove the minimum at 55 K. The presence of the minimum would seem to indicate an antiferromagnetic interaction between Ru(III) and Ni(II). This conclusion was drawn from simi-
Fig. 4. The product of temperature and magnetic susceptibility as a function of temperature for compounds 1 and 2 in the range of 2–300 K at an applied field of 1 kOe. TIP correction is not included.
J.A. Duimstra et al. / Polyhedron 25 (2006) 2705–2709
2709
Ru(III) can be incorporated into a 1D coordination polymer and provide further evidence that simple symmetry considerations can be used to predict whether the ion will display a ferro- or antiferromagnetic exchange interaction. Acknowledgement We acknowledge Dr. Oleksandr Chernyashevskyy for assistance with the magnetic measurements. This research was supported by NIH R01-A147003. Appendix A. Supplementary material
Fig. 5. Magnetization of 1 as a function of applied field at 2.0 K.
lar data by Kahn and co-workers for the 2D (NBu4)[Cu(II)Ru(III)(ox)3] framework [14]. An alternative interpretation that appears to be more applicable in this case attributes the minimum to the anisotropic Ru(III) ion and indicates that the type of coupling must be determined through examination of the low temperature field-dependent magnetization data [13]. The magnetization of 1 as a function of applied field at 2 K is depicted in Fig. 5. At 50 kOe, the magnetization is nearly saturated with a value of 2.46 lB. The calculated saturation magnetization (Ms) for a spin-only system consisting of ferromagnetically coupled Ni(II) (S = 1) and Ru(III) (S = 1/2) is 3 lB. Antiferromagnetic coupling of the two ions results in a calculated Ms of 1 lB. The experimentally observed value clearly indicates ferromagnetic coupling, a conclusion that agrees with the results of Coronado et al. and with the expected orthogonality of magnetic orbitals between Ni(II) and Ru(III) [12,13]. The lower than expected saturation value can be attributed to the Ru(III) ion [13,20]. Zero-field cooled (zfc) and field-cooled (fc) magnetization experiments performed on 1 in applied fields of 3 and 50 Oe show no inflection point and little difference down to 3 K. These measurements indicate a lack of any type of long-range magnetic order above 3 K. Further evidence for the absence of 3D ordering above 2 K is given by the lack of a k-peak in the heat capacity measurements of 1 indicating that there is no magnetic phase transition. Compound 1 displays negligible hysteresis at 2 K. Although 1 does not display slow relaxation of the magnetization above 3 K, the salen ligand is readily amenable to synthetic modification which could be used to affect the magnetism of the compound through changes in the electronics at the Ru(III) center. It is interesting to note that a similar species, Ni(cyclam)[Ru(acac)2(CN)2]2, is trimeric [12], reflecting the marked effect small structural changes can have on the dimensionality of the resulting mixed metal system. The current results demonstrate that
Crystallographic data for structural analysis (cif files for the two structures) have been deposited with the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK, fax: +44 1223 336 033 or e-mail: [email protected] or http://www.ccdc.cam. ac.uk. Refer to CCDC number 286496 for 1 and number 286497 for 2. Supplementary data associated with this article can be found, in the online version, at doi:10.1016/ j.poly.2006.03.024. References [1] E. Coronado, F. Palacio, J. Veciana, Angew. Chem., Int. Ed. Engl. 42 (2003) 2570. [2] T.-F. Liu, D. Fu, S. Gao, Y.-Z. Zhang, H.-L. Sun, G. Su, Y.-J. Liu, J. Am. Chem. Soc. 125 (2003) 13976. [3] H. Tamaki, Z.J. Zhong, N. Matsumoto, S. Kida, M. Koikawa, N. Achiwa, Y. Hashimoto, H. Okawa, J. Am. Chem. Soc. 114 (1992) 6974. [4] S. Decurtins, H.W. Schmalle, H.R. Oswald, A. Linden, J. Ensling, P. Gu¨tlich, A. Hauser, Inorg. Chim. Acta 216 (1994) 65. [5] S. Decurtins, H.W. Schmalle, P. Schneuwly, R. Pellaux, J. Ensling, Mol. Cryst. Liq. Cryst. Sec. A 273 (1995) 605. [6] L.M.C. Beltran, J.R. Long, Acc. Chem. Res. 38 (2005) 325. [7] K.R. Dunbar, R.A. Heintz, Prog. Inorg. Chem. 45 (1997) 283. [8] L.M. Toma, R. Lescoue¨zec, F. Lloret, M. Julve, J. Vaissermann, M. Verdaguer, Chem. Commun. (2003) 1850. [9] S. Wang, J.-L. Zuo, S. Gao, Y. Song, H.-C. Zhou, Y.-Z. Zhang, X.-Z. You, J. Am. Chem. Soc. 126 (2004) 8900. [10] M.P. Shores, J.J. Sokol, J.R. Long, J. Am. Chem. Soc. 124 (2002) 2279. [11] J. Larionova, O. Kahn, S. Gohlen, L. Ouahab, R. Cle´rac, J. Am. Chem. Soc. 121 (1999) 3349. [12] W.-F. Yeung, P.-H. Lau, T.-C. Lau, H.-Y. Wei, H.-L. Sun, S. Gao, Z.-D. Chen, W.-T. Wong, Inorg. Chem. 44 (2005) 6579. [13] E. Coronado, J.R. Gala´n-Mascaro´s, C.J. Go´mez-Garcı´a, J.M. Martı´nez-Agudo, E. Martı´nez-Ferrero, J.C. Waerenborgh, M. Almeida, J. Solid State Chem. 159 (2001) 391. [14] J. Larionova, B. Mombelli, J. Sanchiz, O. Kahn, Inorg. Chem. 37 (1998) 679. [15] B. Bosnich, M.L. Tobe, G.A. Webb, Inorg. Chem. 4 (1965) 1109. [16] W.H. Leung, C.M. Che, Inorg. Chem. 28 (1989) 4619. [17] G.M. Sheldrick, SHELXS-97, Program for Crystal Structure Determination, University of Go¨ttingen, Germany, 1997. [18] G.M. Sheldrick, SHELXL-97, Program for the Refinement of Crystal Structures, University of Go¨ttingen, Germany, 1997. [19] W.-F. Yeung, W.-L. Man, W.-T. Wong, T.-C. Lau, S. Gao, Angew. Chem., Int. Ed. Engl. 40 (2001) 3031. [20] B.N. Figgis, P.A. Reynolds, K.S. Murray, B. Moubaraki, Aust. J. Chem. 51 (1998) 229.