A Spin-Frustrated Trinuclear Copper Complex Based on

Article pubs.acs.org/IC

A Spin-Frustrated Trinuclear Copper Complex Based on Triaminoguanidine with an Energetically Well-Separated Degenerate Ground State Eike T. Spielberg,† Aksana Gilb,† Daniel Plaul,† Daniel Geibig,† David Hornig,† Dirk Schuch,† Axel Buchholz,† Arzhang Ardavan,‡ and Winfried Plass*,† †

Institut für Anorganische und Analytische Chemie, Friedrich-Schiller-Universität Jena, Humboldtstraße 8, 07743 Jena, Germany Centre for Advanced Spin Resonance, Clarendon Laboratory, University of Oxford, OX1 3PU Oxford, United Kingdom

‡

S Supporting Information *

ABSTRACT: We present the synthesis and crystal structure of the trinuclear copper complex [Cu 3 (saltag)(bpy) 3 ]ClO 4 ·3DMF [H 5 saltag = tris(2hydroxybenzylidene)triaminoguanidine; bpy = 2,2′-bipyridine]. The complex crystallizes in the trigonal space group R3̅, with all copper ions being crystallographically equivalent. Analysis of the temperature dependence of the magnetic susceptibility shows that the triaminoguanidine ligand mediates very strong antiferromagnetic interactions (JCuCu = −324 cm−1). Detailed analysis of the magnetic susceptibility and magnetization data as well as X-band electron spin resonance spectra, all recorded on both powdered samples and single crystals, show indications of neither antisymmetric exchange nor symmetry lowering, thus indicating only a very small splitting of the degenerate S = 1/2 ground state. These findings are corroborated by density functional theory calculations, which explain both the strong isotropic and negligible antisymmetric exchange interactions.

■

INTRODUCTION Quantum information processing promises enormous innovational progress in a number of key applications. In particular, encryption and database technologies would feel a drastic gain in efficiency by the proposed algorithms.1 In the search for adequate storage units, called quantum bits (qubits), a number of promising candidates have been proposed. As example, systems have been tested based on 31P-doped silicon single crystals2 and Rydberg atoms3 but also molecular systems such as N@C604 or ring-shaped Cr7Ni5 complexes. In recent years, nanomagnets have been proposed as suitable building blocks for quantum computing devices.6 Compared to atomic spin carriers, these molecular systems offer several advantages.7 Because of their extended size, molecular systems are easier to address, they can be modified by chemical means, and their intrinsic energetic (spin) structures can be used as quantum gates.8 A particular promising class of molecules are spin-frustrated systems.9 Spin frustration occurs in a system of competing magnetic interactions, which cannot be satisfied simultaneously.10 For molecular systems, this is observed if an odd number of noninteger spins is coupled in a ring-shaped system with identical antiferromagnetic interactions between the spins. From an energetic point of view, this leads to a doubly degenerate ground state (see Scheme 1). This degeneracy can be lifted by lowering of the molecular symmetry within the coupling scheme or by antisymmetric exchange. The former is caused by a distortion of the ideal C3 © XXXX American Chemical Society

Scheme 1. Coupling Scheme and Energy Splitting of a SpinFrustrated System

symmetry, while the latter is generally explained by the admixture of excited states via spin−orbit interactions.11 For such spin-frustrated systems, Loss and co-workers have proposed an interesting way to manipulate the spin structure via spin-electric coupling.12 Magnetic molecules suited for this kind of application are desired to combine large antiferromagnetic isotropic coupling interactions, a strict C3 symmetry, and small antisymmetric exchange interactions. The former leads to a good thermal isolation of the ground state, while the latter two conditions prevent a splitting of the degenerate ground-state doublet (Δ in Scheme 1). However, these criteria are frequently violated, as has been observed for a large number of compounds.13 Received: December 27, 2014

A

DOI: 10.1021/ic503095t Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

X-ray Crystallography. Suitable single crystals of the complex [Cu3(saltag)(bpy)3]ClO4·3DMF were selected under a polarizing microscope and fixed with epoxy cement on respective fine glass fibers. The diffraction data were collected on a Nonius Kappa CCD diffractometer, using graphite-monochromated Mo Kα (λ = 71.073 pm) radiation. The crystallographic data are summarized in Table S1 in the Supporting Information (SI). The data were corrected for Lorentz and polarization effects but not for absorption effects. The structure was solved by direct methods (SHELXS17) and refined by full-matrix least-squares techniques against F2 (SHELXL201318). All non-hydrogen atoms were refined with anisotropic displacement parameters. CCDC-885145 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre (http://www.ccdc. cam.ac.uk/data_request/cif). Computational Details. Density functional theory (DFT) according to the Kohn−Sham scheme was employed to study the magnetic properties of the molecular cation of [Cu3(saltag)(bpy)3]ClO4. All-electron DFT calculations and geometry optimizations were performed with the quantum-chemical program package TURBOMOLE.19 As implemented in this program, the gradient-corrected (generalized gradient approximation) functional BP86,20 in combination with the resolution-of-the-identity density-fitting technique,21 and the hybrid functional B3LYP22 were used together with the def2TZVP basis set, which is of split-valence triple-ζ quality, with additional polarization functions on all atoms.23 The calculations have been performed on the structural model taken from the crystallographic data for the heavy atoms. The geometry of the hydrogen atoms was optimized at the BP86/def2-TZVP level, whereas the final single-point calculations were performed at the B3LYP/def2-TZVP level of theory. For structure optimization, a geometry gradient norm of 3 × 10−4 hartree bohr−1 was used as a convergence criterion. The magnetic properties were analyzed according to the phenomenological Heisenberg−Dirac−van Vleck (HDVV) Hamiltonian H = −JS1S2, utilizing the broken-symmetry (BS) approach to evaluate the energy of the antiferromagnetically coupled configurations. It has been shown that the BS solution (EBS) can be directly used to approximate the energy splitting24 because the fully selfconsistent BS single determinant can be considered the correct solution of the Kohn−Sham equations as far as the energy of the lowest spin state is concerned.25 Accordingly, it has been shown that, in combination with the inclusion of Hartree−Fock-type exchange contributions by the hybrid functional B3LPY, this leads to a good agreement with corresponding experimental values.26 For polynuclear systems, the BS approach is generally utilized in terms of an Ising-like mapping of the calculated energies of the BS magnetic states onto the HDVV Hamiltonian.27 On the basis of this approximation, the difference in energy between spin configurations can be expressed as the sum of pairwise interactions, which leads to the desired mapping of the calculated BS energies onto the exchange-coupling constants.

In this work, we describe the trinuclear copper complex with the tritopic ligand tris(2-hydroxybenzylidene)triaminoguanidine (H5saltag), which represents a spin-frustrated system that meets the described preconditions in an almost ideal manner.

■

EXPERIMENTAL DETAILS

Materials. All commercially available chemicals were of reagent grade and were used without further purification. Triaminoguanidine hydrochloride14 and H5saltag·HCl15 were prepared as reported. Caution! Perchlorate derivatives may detonate upon scraping or heating. Instrumentation. Elemental analyses were performed on Leco CHNS-932 and El Vario III elemental analyzers by the Microanalytical Laboratory of the Friedrich-Schiller-Universität Jena. IR spectra were recorded on a Bruker IFS55/Equinox spectrometer equipped with a diamond ATR unit. Mass spectra were measured on a Bruker MAT SSQ 710 spectrometer. Electron spin resonance (ESR) spectra on powdered samples were recorded at X-band frequencies on a Bruker ESP 300 E spectrometer. Single-crystal ESR spectra were recorded on a Bruker EMXmicro spectrometer equipped with an ER 4102ST resonator and an Oxford Instruments ESR 4112HV cryostat. Magnetic data were measured with a Quantum-Design MPMS-5 SQUID magnetometer equipped with a 5 T magnet in the range from 300 to 2 K. Fitting of the magnetic data was performed either by applying the appropriate analytical expression utilizing the program OriginPro 8.5 or by using the full-matrix diagonalization routines, as implemented in the MagProp analysis program in the DAVE package.16 The diamagnetic corrections were estimated from Pascal’s constants. Thermogravimetric measurements were carried out from room temperature to 573 K on powdered samples in a nitrogen stream using a NETZSCH STA409PC Luxx apparatus with a heating rate of 1 °C min−1. UV/vis spectra were recorded utilizing a Varian Cary 5000 UV/vis/NIR spectrophotometer equipped with a Praying Mantis accessory. Synthesis of [Cu3(saltag)(bpy)3]ClO4·3DMF (bpy = 2,2′Bipyridine). A solution of the ligand hydrochloride H5saltag·HCl (453 mg, 1 mmol) and triethylamine (606 mg, 866 μL, 6 mmol) in a mixture of N,N-dimethylformamide (DMF) and methanol (MeOH) (25 mL, ratio 2:3) is added dropwise to a solution of Cu(ClO4)2· 6H2O (1.086 g, 3 mmol) and bpy (469 mg, 3 mmol) in DMF (40 mL). The dark-green solution is stirred for ∼5 min at room temperature and subsequently filtered. From the filtrate, a microcrystalline precipitate is formed overnight, which is collected by filtration and dried in air. Total yield: 554 mg (40%). MS (μ-ESI+, MeOH): m/z 914 (100%, [Cu3(saltag)(bpy)2] +), 790 (89%, [Cu3(saltag)(bpy) + MeOH]+). IR (selected bands, cm−1): 1673 (m), 1592 (s), 1537 (w), 1470 (vs), 1460 (vs), 1439 (s), 1356 (s), 1335 (s), 1313 (m), 1197 (m), 1081 (vs), 760 (vs), 622 (m). UV/vis (powder, BaSO4): 428, 648 nm. In order to obtain (large) single crystals, the following procedure can be applied: A solution of Cu(ClO4)2·6H2O (1.086 g, 3 mmol) in DMF (15 mL) is added to a solution of H5saltag·HCl (453 mg, 1 mmol) in DMF (15 mL) followed by the addition of triethylamine (606 mg, 866 μL, 6 mmol). The resulting dark-green-brown mixture is stirred for about 10 min at room temperature and then filtered. The filtrate is carefully layered in a glass tube with DMF (2 mL) and subsequently with a methanolic solution of bpy (469 mg, 3 mmol). The tube was sealed and left to stand at room temperature. Slow diffusion between the two solutions afforded dark-brown crystals of [Cu3(saltag)(bpy)3]ClO4·3DMF within 4 weeks. The crystals were filtered from solution, washed with MeOH, and dried on filter paper. Yield: 95 mg (7%). Elem anal. for a freshly prepared sample of [Cu3(saltag)(bpy)3]ClO4·3DMF. Calcd for C61H60ClCu3N15O10 (1389.31): C, 52.73; H, 4.35; N, 15.12. Found: C, 53.15; H, 4.27; N, 15.28. Elem anal. for a sample of the complex that was stored in air for about 1 month and used for physical measurements. Found: C, 53.13; H, 4.10; N, 15.01. This is consistent with the weight loss of about 12% observed for thermogravimetric analysis in the temperature range between 313 and 460 K.

■

RESULTS AND DISCUSSION Synthesis and Structure. In the presence of triethylamine, the deprotonated ligand saltag5− reacts with copper(II) perchlorate in DMF to form the cationic complex [Cu3(saltag)]+. To avoid the formation of coordination polymers,28 bpy is added as the capping ligand (Scheme 2).29 Via slow diffusion of a methanolic solution of bpy into the DMF solution of the complex fragment, the compound [Cu3(saltag)(bpy)3]ClO4·3DMF is obtained as a crystalline material. The compound crystallizes in the trigonal space group R3̅. Additional crystallographic data and selected bond lengths and angles can be found in Tables S1 and S2 in the SI. The molecular structure of the cationic complex is depicted in Figure 1. The copper ions are coordinated in a Jahn−Teller distorted square-pyramidal environment (τ = 0.32), with the three donor atoms of the triaminoguanidine scaffold located at B


Article

Inorganic Chemistry Scheme 2. Synthesis of the Trinuclear Copper Complex

Figure 2. Dimer formed by π−π interaction between two adjacent complex cations as seen along the b axis (left) and along the c axis (right). Hydrogen atoms and, on the right-hand side, also the bpy ligands have been omitted for clarity.

different trinuclear molecules is with 511 pm somewhat larger than the intramolecular Cu−Cu distance at 481 pm. The closest distance between a copper ion and a nitrogen atom of the neighboring triaminoguanidine backbone is 404 pm, leading to a corresponding trans angle of 166° at the copper ion, with this nitrogen atom and the bpy nitrogen atom located in the axial position. The other motif is based on π−π interactions between bpy coligands of neighboring molecules at a distance of 335 pm, as depicted in Figure 3. This gives rise to an overall hexagonal-

Figure 1. Molecular structure and numbering scheme of the cationic complex fragment [Cu3(saltag)(bpy)3]+. The ellipsoids are drawn at the 30% probability level. Hydrogen atoms have been omitted for clarity.

Figure 3. Interactions between the bpy ligands of adjacent complex molecules within the hexagonal layers. The π−π interactions are indicated by dashed lines. Hydrogen atoms are omitted for clarity.

equatorial positions (N1, N2, and O1). The nitrogen atom N3 of the bpy ligand occupies the fourth equatorial position at the copper ion, whereas the donor N4 is located in the axial position. For the bridging Cu−N−N−Cu diazine units, a dihedral angle of 152° is observed. The complex cation exhibits a hemisphere shape, with the equatorial coordination planes of the copper ions nearly coplanar to the ligand backbone (given by the central carbon atom and the six nitrogen atoms) at a dihedral angle of 14° between the planes. Moreover, the small angle between the vectors defined by the Cu−N4 bonds at the individual copper centers of 4° indicates that the elongation axes of the Jahn− Teller distorted copper(II) ions within the trinuclear complex cation are nearly collinear. The complex fragment is placed on the crystallographic C3 axis, with the central carbon atom directly located on the symmetry element. The given unique combination of a rigid ligand backbone and the crystallographic symmetry relating the copper ions restricts the options of the system to lift the degeneracy of the ground-state doublet by lowering the symmetry. The supramolecular structure in the solid state is essentially governed by two distinct motifs of π−π interactions. One motif is characterized by the association of two complex cations via interactions between the ligand backbones of adjacent molecular hemispheres, as depicted in Figure 2. The two molecular planes (defined by the central carbon and the six nitrogen atoms of each ligand) are separated by 356 pm. The distance between the two nearest copper ions from

layered structure (see Figure S1 in the SI). Within the resulting layers, the shortest distance between the central carbon atoms of two neighboring molecules is 1260 pm. Magnetic Properties. The molar susceptibility χM of the trinuclear complex was determined on powder samples in the temperature range from 2 to 300 K. The temperature dependence of χMT is depicted in Figure 4.

Figure 4. Representation of the temperature dependence of χMT for [Cu3(saltag)(bpy)3]ClO4·3DMF. The solid line represents the simulated behavior with the parameters given in the text. C


Article

Inorganic Chemistry At room temperature, a value of 0.92 cm3 K mol−1 is observed, which is considerably smaller than the value expected for three independent copper(II) ions (χMT = 1.30 cm3 K mol−1 for g = 2.15). With decreasing temperature, the χMT value decreases and reaches a plateau below 100 K. This behavior is typical for an antiferromagnetically coupled trinuclear system (see Scheme 1). The temperature dependence of χMT can be described using the HDVV Hamiltonian given in eq 1. H = −JCuCu (S1S2 + S1S3 + S2S3)

temperature data analyzed (T < 100 K) in terms of the Bleaney−Bowers equation (eq S2 in the SI), leading to a coupling constant Jdimer = −0.82(2) cm−1 with g = 2.150(2) and χTIP = 1.122(5) × 10−3 cm3 K mol−1. This shows that the decrease of χMT below 10 K can be nicely reproduced by a model of two interacting antiferromagnetically coupled S = 1/2 systems (Figure S2 in the SI). To further address possible anisotropy effects, we performed magnetic susceptibility measurements on single crystals of [Cu3(saltag)(bpy)3]ClO4·3DMF. Two orientations have been examined, one along the crystallographic c axis and the other along one of the indistinguishable in-plane axes (a and b). Because of the weak paramagnetic signal, reliable data could only be obtained for temperatures below 200 K (Figure S3 in the SI). The data have been analyzed according to the Hamiltonian given in eq S1 in the SI, again showing no sign for the presence of antisymmetric exchange within the trinuclear system (see Table S4 in the SI). Hence, we also attempted to model the low-temperature single-crystal data in terms of two exchange-coupled S = 1/2 systems given by the stacked trinuclear cations (Figure 2). To this end, we again applied a dimer model to analyze the singlecrystal χMT data below 100 K (Table S5 and Figure S4 in the SI). The obtained coupling constants for the two orientations are Jdimer(x,y) = −0.71(3) cm−1 and Jdimer(z) = −0.76(3) cm−1, indicative of a slight anisotropy. These two values can be decomposed into an isotropic part, Jdimeriso, and a dipole−dipole term, Jdimerdd, according to eq 3.

(1)

Including intermolecular interactions via the mean-field approach (θ) and a correction term for temperatureindependent magnetic contributions (χTIP) leads to the expression given in eq 2. χM T =

Nβ 2g 2 T 1 + 5e3JCuCu /2kT + χTIP T 4k T − θ 1 + e3JCuCu /2kT

(2)

The best agreement with the experimental data is obtained with an isotropic coupling constant JCuCu = −324(2) cm−1 and g = 2.160(1), assuming a temperature-independent magnetic correction of χTIP = 1.1170(7) × 10−3 cm3 K mol−1. Interestingly, the observed coupling constant is in fairly good agreement with an earlier reported magnetostructural correlation found with the dihedral angle of N−N diazine-bridged copper ions,30 which would predict a value of −311 cm−1 for the observed dihedral angle of 152° in [Cu3(saltag)(bpy)3]ClO4·3DMF. The coupling constant represents an energy gap of 486 cm−1 (i.e., 5.8 kJ mol−1 or 700 K) between the ground-state doublets and excited-state quartet. This clearly shows excellent thermal isolation of the degenerate ground state within the trinuclear complex. The decrease of χMT at very low temperatures can be explained very well by molecular field theory, leading to a Weiss temperature θ = −0.405(7) K. This indicates that intermolecular interactions might be responsible for the observed lowtemperature behavior. However, to shed more light on this issue, we further examined the low-temperature behavior by including additional effects. Of particular interest here is the antisymmetric exchange because this could also account for the observed behavior. Toward this end, we performed additional data analyses with a full-matrix diagonalization approach including both possible effects (eq S1 in the SI), i.e., the antisymmetric exchange and intermolecular interactions via the mean-field approximation (for details, see the SI). It is clearly shown that the antisymmetric exchange does not give any appreciable contribution. In fact, the mean-field approximation alone gives a better description than including antisymmetric exchange (Table S3 in the SI). This leads to an estimated upper limit for Gz of −0.9 cm−1 for possible antisymmetric exchange contributions. This is consistent with the structural feature of nearly collinear elongation axes of the Jahn−Teller distorted copper(II) ions because for collinear spin momenta the antisymmetric exchange is strictly vanishing.31 Together with the observation of closely stacked dimers in the crystal structure of [Cu3(saltag)(bpy)3]ClO4·3DMF (Figure 2), this led us to the assumption that the lowtemperature behavior might be mainly governed by intermolecular antiferromagnetic exchange interactions between these two trinuclear fragments. To simulate this, the complex molecules were treated as S = 1/2 systems and the low-

Jdimer = Jdimer

iso

+ Jdimer

⎛1 ⎞ ⎟ ⎜1 ⎟ ⎝− 2 ⎠

dd ⎜

(3)

This leads to rough estimates for the isotropic interaction, Jdimeriso = −0.73 cm−1, and a dipolar part, Jdimerdd = 0.02 cm−1. Interestingly, the latter is in good agreement with the dipolar coupling constant (Jcalcdd = 0.016 cm−1) calculated from the intermolecular Cu−Cu distances within the stacked dimers (511 pm; see Figure 2). Magnetization measurements as a function of the magnetic field were performed at 2 K for both powder samples and single crystals (Figure S5 in the SI). The data are fully consistent with the previous analysis and again did not show any evidence for antisymmetric exchange. In conclusion, only small exchange interactions within the π−π-stacked dimer (Figure 2) are observed, which, however, might obscure even weaker interactions. However, even if the decay at low temperatures would be solely attributed to antisymmetric exchange, the corresponding exchange constant would be smaller than 1 cm−1. However, it should be noted here that it is, in general, difficult to extract very small exchange interactions accurately from magnetic measurements. ESR Spectroscopy. A possible splitting of the ground-state doublet by symmetry lowering and/or antisymmetric exchange could easily be detected by ESR spectroscopy. Such interactions would lead to changes of the effective gyromagnetic ratios within the xy plane (gx,y). Therefore, X-band ESR spectra for powdered samples and single crystals of the complex [Cu3(saltag)(bpy)3]ClO4·3DMF were measured at 5 K. The angular dependence of the ESR observed for singlecrystal measurements can be described by eq 4, where geff denotes the experimentally observed g factor, gx,y and gz are the principal components of the g tensor, θ is the goniometer D


Article

Inorganic Chemistry angle, and θ0 is the angle formed between the crystal and goniometer. geff =

2 gz2cos 2(θ − θ0) + g x,y sin2(θ − θ0)

orientation-dependent susceptibility and magnetization measurements on single crystals. The slight anisotropy observed can solely be described by an axial g tensor without the use of any additional exchange interactions. DFT Calculations. To support analysis of the experimental magnetochemical data and to elucidate the possible mechanism for the exchange interactions within the trinuclear system, DFT calculations on the complex cation [Cu3(saltag)(bpy)3]+ were performed. The magnetic exchange interactions according to the HDVV Hamiltonian given in eq 1 were determined by use of the BS approach. For the trinuclear C3-symmetric cationic complex [Cu3(saltag)(bpy)3]+, two spin configurations need to be constructed, a high-spin (HS) Cu1(↑)−Cu2(↑)−Cu3(↑) and a BS configuration Cu1(↓)−Cu2(↑)−Cu3(↑). Because of the trigonal symmetry, it is irrelevant for which copper ion the spin is inverted for defining the BS state. The HS configuration represents the quartet state (EHS = EQ), with all copper ions ferromagnetically coupled, whereas the BS configuration corresponds formally to the doublet states, which simultaneously have one ferromagnetic and two antiferromagnetic couplings between the copper ions within the trinuclear arrangement. The energy difference between the quartet (EQ) and BS (EBS) states is calculated as ΔE = EQ − EBS = 470 cm−1. Within the framework of the usually applied mapping (EBS − EQ = 2JCuCu),32 this leads to a coupling constant of −235 cm−1. However, it should be noted that the calculated energy difference ΔE very closely resembles the experimentally observed energy splitting between the quartet (EQ) and doublet (ED) states, which is found to be EQ − ED = 486 cm−1. It is therefore tempting to assume that not only is the HS quartet state correctly represented by a single determinate but also the doublet state can be expressed by a single determinant that corresponds to the BS state. This, however, would suggest that the BS single determinant is the correct solution for the doublet state of the trinuclear copper system. According to this interpretation (ED − EQ = EBS − EHS = 3/2JCuCu), the exchangecoupling constant would be calculated as JCuCu = −313 cm−1, which is in excellent agreement with the experimental value of −324 cm−1. In any case, the calculations confirm the very strong antiferromagnetic interactions between the copper ions within the spin-frustrated trimer. Analysis of the magnetic orbitals shows the dominant interaction to be an exchange path through the σ bonds of the N−N diazine bridge within the triaminoguanidine ligand backbone (see Scheme 3). The extended π system of the ligand does not contribute to the exchange. Moreover, the observed exchange pathway also explains the small antisymmetric exchange present in the trinuclear system. For an effective antisymmetric exchange, a large overlap between the magnetic orbital of one copper center (dx2−y2) with empty orbitals excited by spin−orbit coupling (dxy, dxz, and dyz) at another copper center is necessary. The exchange pathway observed for [Cu3(saltag)(bpy)3]+ allows, in fact, for a very efficient overlap of the magnetic orbitals but very poor overlap with the corresponding excited-state orbitals of the neighboring copper ions. Even if there is no strict orthogonality of the orbitals, this leads to very small contributions by antisymmetric exchange, which is in full agreement with the experimental data.

(4)

The least-squares fitting of the single-crystal data to eq 4 gives the values gx,y = 2.0539(5) and gz = 2.2159(5) (see Figure 5), which nicely agree with the observed square-pyramidal coordination environment at the copper(II) ions.

Figure 5. Angular dependence of the effective g factor (geff) as obtained from single-crystal ESR measurements of [Cu3(saltag)(bpy)3]ClO4·3DMF at 5 K for rotation around an axis perpendicular to the crystallographic C3 axis. The dotted line represents the best fit (see the text).

The ESR spectrum of a microcrystalline sample of [Cu3(saltag)(bpy)3]ClO4·3DMF is depicted in Figure 6. The

Figure 6. X-band ESR spectrum for a microcrystalline sample of [Cu3(saltag)(bpy)3]ClO4·3DMF measured at 5 K (solid line) and the simulated spectrum (dashed line). The weak signal at 3350 G, marked with an asterisk, is an artifact of the resonator.

data can be simulated as an axial spectrum, utilizing strong line broadening of the axial transition as determined from the single-crystal measurement (see Figure S6 in the SI). The principle components of the g tensor used for simulation are gx,y = 2.075 and gz = 2.208, which is in good agreement with the values obtained from single-crystal ESR spectra. Moreover, the average g factor deduced from the powder measurement, gav = 2.119, is in good agreement with the g factor obtained from the magnetic susceptibility data. In summary, the X-band ESR measurements both on powdered samples and on single crystals show no indications for a splitting of the ground state, which is in accordance to the

■

CONCLUSIONS The trinuclear copper complex presented is characterized by a thermally well-isolated ground state with a virtually undisturbed spin frustration. The bridging ligand provides two remarkable E


Article

Inorganic Chemistry

(2) (a) Kane, B. E. Nature 1998, 393, 133−137. (b) Morton, J. J. L.; Tyryshkin, A. M.; Brown, R. M.; Shankar, S.; Lovett, B. W.; Ardavan, A.; Schenkel, T.; Haller, E. E.; Ager, J. W.; Lyon, S. A. Nature 2008, 455, 1085−1088. (3) (a) Ahn, J.; Weinacht, T. C.; Bucksbaum, P. H. Science 2000, 287, 463−465. (b) Saffman, M.; Walker, T. G.; Mølmer, K. Rev. Mod. Phys. 2010, 82, 2313−2363. (4) Morton, J. J. L.; Tyryshkin, A. M.; Ardavan, A.; Benjamin, S. C.; Porfyrakis, K.; Lyon, S. A.; Briggs, G. A. D. Nat. Phys. 2006, 2, 40−43. (5) (a) Timco, G. A.; Carretta, S.; Troiani, F.; Tuna, F.; Pritchard, R. J.; Muryn, C. A.; McInnes, E. J. L.; Ghirri, A.; Candini, A.; Santini, P.; Amoretti, G.; Affronte, M.; Winpenny, R. E. P. Nat. Nanotechnol. 2009, 4, 173−178. (b) Wedge, C. J.; Timco, G. A.; Spielberg, E. T.; George, R. E.; Tuna, F.; Rigby, S.; McInnes, E. J. L.; Winpenny, R. E. P.; Blundell, S. J.; Ardavan, A. Phys. Rev. Lett. 2012, 108, 107204. (6) (a) Leuenberger, M. N.; Loss, D. Nature 2001, 410, 789−793. (b) Tejada, J.; Chudnovsky, E. M.; del Barco, E.; Hernandez, J. M.; Spiller, T. P. Nanotechnology 2001, 12, 181−186. (c) Stepanenko, D.; Trif, M.; Loss, D. Inorg. Chim. Acta 2008, 361, 3740−3745. (d) Affronte, M.; Troiani, F.; Ghirri, A.; Carrette, S.; Santini, P.; Schuecker, R.; Timco, G.; Winpenny, R. E. R. J. Magn. Magn. Mater. 2007, 310, e501−e502. (7) Gatteschi, D.; Sessoli, R.; Villain, J. Molecular Nanomagnets; Oxford University Press: Oxford, U.K., 2006. (8) (a) Troiani, F.; Affronte, M. Chem. Soc. Rev. 2011, 40, 3119− 3129. (b) Lehmann, J.; Gaita-Ariño, A.; Coronado, E.; Loss, D. J. Mater. Chem. 2009, 19, 1672−1677. (9) (a) Gatteschi, D.; Pardi, L.; Barra, A. L.; Müller, A.; Döring, J. Nature 1991, 354, 463−465. (b) Gatteschi, D.; Sessoli, R.; Plass, W.; Müller, A.; Krickemeyer, E.; Meyer, J.; Sölter, D.; Adler, P. Inorg. Chem. 1996, 35, 1926−1934. (c) Yamase, T.; Ishikawa, E.; Fukaya, K.; Nojiri, H.; Taniguchi, T.; Atake, T. Inorg. Chem. 2004, 43, 8150−8157. (d) Chaudhuri, P.; Karpenstein, I.; Winter, M.; Butzlaff, C.; Bill, E.; Trautwein, A. X.; Flörke, U.; Haupt, H. J. J. Chem. Soc., Chem. Commun. 1992, 321−322. (e) Ferrer, S.; Lloret, F.; Bertomeu, I.; Alzuet, G.; Borrás, J.; García-Granda, S.; Liu-González, M.; Haasnoot, J. G. Inorg. Chem. 2002, 41, 5821−5830. (f) Angaridis, P. A.; Baran, P.; Boča, R.; Cervantes-Lee, F.; Haase, W.; Mezei, G.; Raptis, R. G.; Werner, R. Inorg. Chem. 2002, 41, 2219−2228. (g) Cage, B.; Cotton, F. A.; Dalal, N. S.; Hillard, E. A.; Rakvin, B.; Ramsey, C. M. J. Am. Chem. Soc. 2003, 125, 5270−5271. (h) Ray, M. S.; Chattopadhyay, S.; Drew, M. G. B.; Figuerola, A.; Ribas, J.; Diaz, C.; Ghosh, A. Eur. J. Inorg. Chem. 2005, 4562−4571. (i) Kortz, U.; Nellutla, S.; Stowe, A. C.; Dalal, N. S.; van Tol, J.; Bassil, B. S. Inorg. Chem. 2004, 43, 144− 154. (j) Mirica, L. M.; Stack, T. D. P. Inorg. Chem. 2005, 44, 2131− 2133. (10) (a) Kahn, O. Chem. Phys. Lett. 1997, 265, 109−114. (b) Schnack, J. Dalton Trans. 2010, 39, 4677−4686. (11) (a) Yoon, J.; Mirica, L. M.; Stack, T. D. P.; Solomon, E. I. J. Am. Chem. Soc. 2004, 126, 12586−12595. (b) Yoon, J.; Solomon, E. I. Coord. Chem. Rev. 2007, 251, 379−400. (c) Belinsky, M. I. Inorg. Chem. 2008, 47, 3521−3531. (d) Belinsky, M. I. Inorg. Chem. 2008, 47, 3532−3539. (12) (a) Trif, M.; Troiani, F.; Stepanenko, D.; Loss, D. Phys. Rev. Lett. 2008, 101, 217201. (b) Trif, M.; Troiani, F.; Stepanenko, D.; Loss, D. Phys. Rev. B 2010, 82, 045429. (13) Ferrer, S.; Lloret, F.; Pardo, E.; Clemente-Juan, J. M.; LiuGonzález, M.; García-Grana, S. Inorg. Chem. 2012, 51, 985−1001. (14) Matyushin, Y. N.; Kon’kova, T. S.; Titova, K. V.; Rosolovskii, V. Y.; Lebedev, Y. A. Izv. Akad. Nauk SSSR, Ser. Khim. 1982, 446−449. (15) (a) Winkelmann, E.; Raether, W. Chem. Abstr. 1972, 77, 126264. (b) Müller, I. M.; Robson, R. Angew. Chem., Int. Ed. 2000, 39, 4357− 4359. (16) DAVE: A comprehensive sof tware suite for the reduction, visualization, and analysis of low energy neutron spectroscopic data: Azuah, R. T.; Kneller, L. R.; Qiu, Y.; Tregenna-Piggott, P. L. W.; Brown, C. M.; Copley, J. R. D.; Dimeo, R. M. J. Res. Natl. Inst. Stand. Technol. 2009, 114, 341.

Scheme 3. Exchange Mechanism via the Diazine Bridge within the Triaminoguanidine Scaffolda

a

The strong overlap between the magnetic orbitals of two adjacent copper ions (left) leads to very strong antiferromagnetic interactions, while the very small overlap between one magnetic orbital and an excited state at an adjacent copper ion yields only small antisymmetric exchange.

features that allow for the observed properties. At first, the ligand mediates strong isotropic and very weak antisymmetric exchange interactions through the σ bonds of the diazine bridging units. Furthermore, the rigid ligand backbone hinders any distortions and thus prevents the system from lifting the degeneracy by lowering the symmetry. This makes trinuclear complexes based on the tritopic triaminoguanidine scaffold a class of promising candidates in the field of molecular magnets as potential qubits. Particularly intriguing is the option of utilizing such molecular systems in terms of spin-electric coupling, which might enable electric control of molecular spin states. To this end, the high synthetic flexibility of the modular ligand system, which allows for a wide range of chemical modifications, is an enormous advantage that can open a broad range of future applications.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional structural details including data in CIF format, further data related to magnetic characterization, and ESR spectroscopy. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work has been supported by the Deutsche Forschungsgemeinschaft (DFG) within the priority program “Molecular Magnetism” (SPP 1136). E.T.S. acknowledges financial support by the Carl-Zeiss Foundation, “Graduiertenakademie Jena”, and DFG. D.P. thanks Freistaat Thüringen for a graduation scholarship. A.A. acknowledges financial support by the Royal Society. The authors thank Dr. H. Görls for collecting the crystallographic data.

■

REFERENCES

(1) (a) Shor, P. W. SIAM J. Comput. 1997, 26, 1484−1509. (b) Grover, L. K. Phys. Rev. Lett. 1997, 79, 325−328. (c) Grover, L. K. Phys. Rev. Lett. 1997, 79, 4709−4712. F


Article

Inorganic Chemistry (17) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Adv. 2008, 64, 112−122. (18) Sheldrick, G. M. SHELXL-2013, Program for the refinement of Crystal Structures form X-ray Data; Universität Göttingen: Göttingen, Germany, 2013. (19) Ahlrichs, R.; Bär, M.; Häser, M.; Horn, H.; Kölmel, C. Chem. Phys. Lett. 1989, 162, 548−549. (20) (a) Becke, A. D. Phys. Rev. A 1988, 38, 3098−3100. (b) Perdew, J. P. Phys. Rev. B 1986, 33, 8822−8824. (21) (a) Baerends, E. J.; Ellis, D. E.; Ros, P. Chem. Phys. 1973, 2, 41− 51. (b) Dunlap, B. I.; Connolly, J. W. D.; Sabin, J. R. J. Chem. Phys. 1979, 71, 3396−3402. (22) (a) Becke, A. D. Chem. Phys. 1993, 98, 5648−5652. (b) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789. (23) (a) Weigend, F. Phys. Chem. Chem. Phys. 2006, 8, 1057−1065. (b) ftp://ftp.chemie.uni-karlsruhe.de/pub/jbasen. (24) Ruiz, E.; Alvarez, S.; Cano, J.; Polo, V. J. Chem. Phys. 2005, 123, 164110. (25) (a) Perdew, J. P.; Savin, A.; Burke, K. Phys. Rev. A 1995, 51, 4531−4541. (b) Perdew, J. P.; Ernzerhof, M.; Burke, K.; Savin, A. Int. J. Quantum Chem. 1997, 61, 197−205. (26) (a) Ruiz, E.; Cano, J.; Alvarez, S.; Polo, V. J. Chem. Phys. 2006, 123, 107102. (b) Plass, W.; Pohlmann, A.; Rautengarten, J. Angew. Chem., Int. Ed. 2001, 40, 4207−4210. (27) (a) Ruiz, E. Struct. Bonding (Berlin) 2004, 113, 71−102. (b) Kortus, J.; Hellberg, C. S.; Pederson, M. R. Phys. Rev. Lett. 2001, 86, 3400−3403. (28) (a) Zharkouskaya, A.; Buchholz, A.; Plass, W. Eur. J. Inorg. Chem. 2005, 4875−4879. (b) Zharkouskaya, A.; Görls, H.; Vaughan, G.; Plass, W. Inorg. Chem. Commun. 2005, 8, 1145−1148. (29) Ion, A. E.; Spielberg, E. T.; Görls, H.; Plass, W. Inorg. Chim. Acta 2007, 360, 3925−3931. (30) Thompson, L. K.; Xu, Z.; Goeta, A. E.; Howard, J. A. K.; Clase, H. J.; Miller, D. O. Inorg. Chem. 1998, 37, 3217−3229. (31) Moriya, T. Phys. Rev. 1960, 120, 91−98. (32) Spielberg, E. T.; Fittipaldi, M.; Geibig, D.; Gatteschi, D.; Plass, W. Inorg. Chim. Acta 2010, 363, 4269−4276.

G


A Spin-Frustrated Trinuclear Copper Complex Based on

Recommend Documents