Multifrequency cw-EPR and DFT Studies of an ... - ACS Publications

Jan 27, 2016 - Geoffrey B. Jameson,. §. Malcolm A. Halcrow,. ∥. Allan G. Blackman,*,⊥ and Carole Duboc*,†. †. Univ. Grenoble Alpes, CNRS UMR ...
1 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/IC

Multifrequency cw-EPR and DFT Studies of an Apparent Compressed Octahedral Cu(II) Complex Nikita Hall,† Maylis Orio,‡ Marcello Gennari,† Christopher Wills,† Florian Molton,† Christian Philouze,† Geoffrey B. Jameson,§ Malcolm A. Halcrow,∥ Allan G. Blackman,*,⊥ and Carole Duboc*,† †

Univ. Grenoble Alpes, CNRS UMR 5250, DCM, F-38000 Grenoble, France Institut des Sciences Moléculaires de Marseille, Aix Marseille Université, CNRS, Centrale Marseille, ISM2 UMR 7313, 13397 Marseille, France § Institute of Fundamental Sciences and the MacDiarmid Institute for Advanced Materials and Nanotechnology, Masssey University, Private Bag 11222, Palmerston North 4442, New Zealand ∥ School of Chemistry, University of Leeds, Woodhouse Lane, Leeds UK LS2 9JT, United Kingdom ⊥ School of Applied Sciences, Auckland University of Technology, Private Bag 92006, Auckland 1142, New Zealand ‡

S Supporting Information *

ABSTRACT: The syntheses and single-crystal X-ray structures of the mononuclear complexes [Cu(bmet)](ClO4)2· H2O, [Cu(bmet)]Br2·2MeCN, and [Zn(bmet)](ClO4)2·H2O (bmet = N,N′-bis(2,2′-bipyridin-6-ylmethyl)ethane-1,2-diamine) are described. All three complexes feature a central metal ion bound to all six N atoms of the bmet ligand, which displays a meridional-facial-facial-meridional (mf f m) configuration. The three complexes show one N−M−N axis to be significantly shorter than the others in agreement with an apparent compressed octahedral geometry. The X-ray structures of a single crystal of [Cu(bmet)](ClO 4 ) 2 · 0.375H2O resolved from data recorded at different temperatures display no remarkable structural modifications. However, they all display both as a powder and, in solution, an axial g1 > g2 ≳ g3 > ge electron paramagnetic resonance (EPR) pattern at low temperature, which is indicative of tetragonally elongated octahedra, while at room temperature the Q-band EPR spectra display a more rhombic g1 ≳ g2 > g3 > ge pattern. The fully density functional theory optimized structure of the CuII complexes displays significant structural modifications only along one Nimine− M−Namine axis resulting in an elongated octahedral structure. Furthermore, the EPR parameters predicted from this structure are comparable to those determined experimentally from the axial EPR signal recorded at low temperature, consistent with the unpaired electron residing mainly in the {3dx2−y2} orbital. The structural and electronic properties of [Cu(bmet)]2+ are different from those in other previously described dynamic Jahn−Teller systems. We propose that these data can be rationalized by a dynamic Jahn−Teller effect perturbed by the strain of the hexadentate bmet ligand.

■

three molecular axes.4 While both distortions are possible, Jahn−Teller elongations are by far preferred, due to 4s−3dz2 orbital mixing, which lowers the energy of the 3dz2 orbital compared to the 3dx2−y2 orbital.5 In principle, the detection of Jahn−Teller distortions in octahedral CuII complexes may simply be achieved by determining the metal−ligand distances by X-ray crystallography; in general, the long and short Culigand bonds usually differ in length by >20%. However, these bond lengths are not necessarily unambiguous, owing to the possibility of masked static and/or dynamic disorder. EPR spectroscopy is an appropriate tool to determine the nature of the Jahn−Teller distortion, because the distortion (elongation or compression) dictates the electronic ground state, with the

INTRODUCTION In 1937, Jahn and Teller used symmetry arguments to demonstrate that a nonlinear system in a degenerate energy state cannot be stable. As a result, it will spontaneously distort itself in some way so that the energy state will split to remove its degeneracy.1 This is particularly common in transition metal complexes, where the ligand field splitting of the d-orbitals can often lead to degenerate electron states.2,3 Complexes with partial occupancy of the eg orbitals can often exhibit significant Jahn−Teller distortions, with the textbook example being octahedral copper(II) complexes; the d9 electron configuration ensures single occupancy of one orbital of the eg set. In these complexes, the Jahn−Teller theorem predicts that in a perfect octahedral environment the degeneracy of the 2Eg ground state will be split, reflecting a concomitant elongation or compression of the Cu-ligand bonds parallel to one of the © XXXX American Chemical Society

Received: October 5, 2015

A

DOI: 10.1021/acs.inorgchem.5b02287 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Chart 1. Structures of Previously Described CuII Complexes Displaying Dynamic Jahn−Teller Distortions (1A−1D and 1F) or a Ground State with a {dz2}1 Electronic Configuration (1 × 10−1 G)

the bmet ligand (Chart 2), as well as their characterization by single-crystal X-ray crystallography and multifrequency con-

unpaired electron occupying either the 3dx2−y2 or 3dz2 orbital, respectively (hereafter these are denoted the {3dorb}1 electronic configurations). These two configurations yield very different EPR spectra, whose line shapes are diagnostic of the identity of the singly occupied d-orbital in the molecule. For a {3dx2−y2}1 complex, a g1 > g2 ≥ g3 > 2.00 pattern is expected, whereas a {3dz2}1 complex yields g1 ≥ g2 > g3 ≈ 2.00.6−8 Although there are several crystallographically characterized CuII complexes displaying axial compression, it has been demonstrated that the majority of these compressions are artifacts, resulting from either static or dynamic disorder.5,9−14 In fact, there are very few genuine Jahn−Teller compressed copper(II) compounds, most of them being inorganic salts. CuF64− doped into BaZnF6 and KAlCuF6 provided the first examples of octahedral CuII complexes in a ligand field of identical ligator atoms to display an actual tetragonal compressed molecular structure,15−17 and these systems are still topics of interest.18 To the best of our knowledge, there are only two “molecular” octahedral CuII complexes that display a genuine compressed octahedral geometry, both having been reported by Halcrow’s group (Chart 1, 1E and 1G).19,20 X-ray crystallographic analysis of [Cu(terpy)2]2+ and related complexes (Chart 1, 1A−1C) has also given evidence for axially compressed octahedral geometries. However, this compression is only “apparent” and results from a dynamic Jahn−Teller distortion, as evidenced by variable-temperature X-ray crystallography and electron paramagnetic resonance (EPR) investigations.5,11,12,21−23 A dynamic disorder of an elongated Jahn−Teller axis over two sets of trans Cu-ligand bonds leads to high-temperature EPR spectra with a particular signature that arises from the fact that two elongated octahedra coexist for which the g-values have been averaged. As the temperature decreases, the Jahn−Teller axis becomes localized in one of the two disordered orientations and gives rise to the typical spectrum of a {3dx2−y2}1 CuII center. The [Cu(bpp)]2+19 and [Cu(pma)]2+20 complex cations, which contain sterically nonbulky ligands (Chart 1, 1D and 1F), also display this dynamic behavior. Other factors can be involved in dynamic Jahn−Teller systems, including random strain effects and reduction factors in EPR spectra recorded at low temperatures.24 Recently, we reported a series of hexadentate nitrogen-based ligands in which two bipyridine units are linked by an aliphatic diamine chain, our objective being to study the effect of altering the length of the aliphatic chain on the nuclearities of the resulting metal complexes.25,26 In the present study, we describe the syntheses of [Cu(bmet)](ClO4)2 and [Cu(bmet)]Br2, two mononuclear octahedral CuII complexes of

Chart 2. Schematic Representation of the bmet Ligand

tinuous-wave (cw)-EPR spectroscopy. The synthesis and characterization of the ZnII complex [Zn(bmet)](ClO4)2, isostructural to [Cu(bmet)](ClO4)2, is also described, to assist in rationalizing the structural properties of the CuII complexes. We show that, while the EPR spectra of [Cu(bmet)]2+ changes with increasing temperature from an axial to a more rhombic EPR signal, consistent with an evolution of the electronic structure of the CuII ion, no obvious structural modification can be discerned from variable-temperature X-ray crystallographic data. Theoretical results based on density functional theory (DFT) calculations are reported, the purpose of which is to aid in the understanding of the behavior of the electronic properties of the CuII complex. On the basis of this combined experimental and theoretical investigation, a proposed rationalization of the data is presented.

■

EXPERIMENTAL SECTION

General. The bmet ligand was prepared according to a previously reported procedure.26 All other reagents and solvents were used as received. Mass spectra were recorded on a Bruker MicrOTOF-Q (Bruker Daltronics, Bremen, Germany) using electrospray ionization (ESI) in positive mode. The spectra were measured over the range 200−2000 Da. Complex Syntheses. The complexes [Cu(bmet)](ClO4)2· 0.5H2O, [Cu(bmet)](Br)2·2MeCN, and [Zn(bmet)](ClO4)2·0.5H2O were prepared by reaction of the appropriate metal salt with the bmet ligand in a 1:1 mole ratio in MeCN under aerobic conditions. Microanalytical data for the bulk perchlorate salts suggested slight desolvation occurred on drying the crystallographic samples. [Cu(bmet)](ClO4)2·0.5H2O. The blue solution that is formed on mixing acetonitrile solutions (1 mL) of bmet (54.0 mg, 1.36 × 10−4 mol) and Cu(ClO4)2·6H2O (49.8 mg, 1.34 × 10−4 mol) gave the product as small blue X-ray quality crystals on standing at room temperature overnight (65.3 mg, 75%). Anal. Calcd for [Cu(C24H24N6)](ClO4)2· 0.5H2O: C, 43.16; H, 3.77; N, 12.59; Cl, 10.62. Found: C, 43.19; H, 3.62; N, 12.40; Cl, 10.84%. ESI-MS: Calcd for C24H24ClCuN6O4+, 558.0838; found 558.0839 [M+]. This material was crystallographically characterized as monoclinic [Cu(bmet)](ClO4)2·H2O. A later synB

DOI: 10.1021/acs.inorgchem.5b02287 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

(Grid4 and GridX4 in ORCA convention) and tight SCF convergence criteria were used in the calculation. The integration grids were increased to an integration accuracy of 11 (ORCA convention) for the metal center. Picture change effects were applied for the calculation of the hyperfine tensors. The cationic component of complexes [Cu(bmet)](ClO4)2·H2O and [Zn(bmet)](ClO4)2·H2O formed the starting point of calculations.

thesis under identical conditions produced tetragonal [Cu(bmet)](ClO4)2·0.375H2O. [Cu(bmet)]Br2·2MeCN. Mixing acetonitrile solutions (1 mL) of bmet (47.2 mg, 1.19 × 10−4 mol) and CuBr2 (26.6 mg, 1.19 × 10−4 mol) gave a green precipitate. Acetonitrile (1 mL) was added to the mixture, and heating resulted in dissolution. A light green precipitate formed upon cooling and standing overnight. The precipitate was filtered and air-dried, and the filtrate was then left to evaporate, giving dark green X-ray quality crystals. IR spectroscopy showed the precipitate and crystals to be identical. Anal. ESI-MS: Calcd for C24H23N6Cu+, 458.1275; found 458.1237 [M+]. Acceptable microanalytical data for this complex could not be obtained. [Zn(bmet)](ClO4)2·0.5H2O. A light yellow solution formed on mixing acetonitrile solutions (1 mL) of bmet (50.3 mg 1.27 × 10−4 mol) and Zn(ClO4)2·6H2O (46.5 mg, 1.25 × 10−4 mol). The solution was left to evaporate over a period of two weeks, giving colorless block X-ray quality crystals (61.9 mg, 71%). Anal. Calcd for [Zn(C24H24N6)](ClO4)2·0.5H2O: C, 43.05; H, 3.76; N, 12.55; Cl, 10.59. Found: C, 42.97; H, 3.76; N, 12.57; Cl, 10.66%. ESI-MS: Calcd for C24H24ClN6O4Zn+, 559.0834; Found 559.0825 [M+]. This material was crystallographically characterized as monoclinic [Zn(bmet)](ClO4)2·H2O. A later synthesis produced tetragonal [Cu(bmet)](ClO4)2·0.375H2O crystals after diffusion of isopropyl ether into a solution of [Cu(bmet)](ClO4)2, prepared as described above, in acetonitrile. X-ray Crystallography. X-ray diffraction data for [Cu(bmet)](ClO4)2·H2O, [Cu(bmet)](Br)2·2MeCN, and [Zn(bmet)](ClO4)2· H2O were collected at 89 K on a Bruker Kappa APEX-II27 system using graphite-monochromated Mo Kα radiation with 0.5° frames. The data were corrected for Lorentz and polarization effects using SAINT28 and scaled using SADABS.29 X-ray diffraction data at 100, 200, and 300 K were collected on a Bruker Kappa CCD system using a multilayer mirrors-monochromated Mo Kα radiation from an Incoatec high brilliance microsource with 1.2° frames. The data were integrated using EvalCCD and corrected for absorption using SADABS. All structures were solved using SIR-9730 or SIR-200431 running within the WinGX package,32 and weighted full-matrix least-squares refinement on F2 was performed using SHELXL-97.33 Hydrogen atoms attached to carbon and nitrogen atoms were included in calculated positions and were refined as riding atoms with individual (or group, if appropriate) isotropic displacement parameters. Hydrogen atoms on oxygen atoms of solvent water molecules were not found. Disorder was found in one of the perchlorate anions of both [Cu(bmet)](ClO4)2·H2O and [Zn(bmet)](ClO4)2·H2O. Crystallographic data for these complexes and for [Cu(bmet)]Br2·2MeCN are summarized in Table S1. The structure of the complex [Cu(bmet)](ClO4)2· 0.375H2O was determined at 100, 200, and 300 K. Crystallographic data are summarized in Table S2. Electron Paramagnetic Resonance Spectroscopy. X-band EPR spectra were recorded with a Bruker EMX equipped with an ER-4192 ST Bruker cavity and an ER-4131 VT for the 100 K experiments. Powder Q-band EPR spectra were recorded with an ER5106 QTW Bruker cavity and an Oxford Instruments ESR-900 continuous-flow helium cryostat for the Q-band for the 4.5 K experiments. Computational Chemistry. All DFT theoretical calculations were performed using the ORCA program package.34 Full geometry optimizations were performed for all complexes using the GGA functional BP8635−37 in combination with the TZV/P38 basis set for all atoms and by taking advantage of the resolution of the identity approximation in the Split-RI-J variant39 with the appropriate Coulomb fitting sets.40 Increased integration grids (Grid4 in the ORCA convention) and tight self-consistent field (SCF) convergence criteria were used. Electronic structures as well as molecular orbitals were obtained using the hybrid functional B3LYP41,42 and the TZV/P basis set. The g-tensors and hyperfine coupling constants were obtained from single-point calculations using the B3LYP functional. Scalar relativistic effects were included with ZORA paired with the SARC def2-TZVP(-f) basis sets43,44 and the decontracted def2-TZVP/ J Coulomb fitting basis sets for all atoms. Increased integration grids

■

RESULTS Structural Properties of the Copper(II) and Zinc(II) Complexes. The structures of both [Cu(bmet)](ClO4)2·H2O and [Cu(bmet)](Br)2·2MeCN were established by singlecrystal X-ray diffraction. A selection of bond lengths and angles is given in Table 1, and the ORTEP diagram of the CuII cation Table 1. Selected Bond Lengths (Å) and Angles (deg) for [Cu(bmet)](ClO4)2·H2O, [Cu(bmet)](Bt)2·2MeCN, and [Zn(bmet)](ClO4)2·H2Oa at 89 K

M−N1 M−N2 M−N3 M−N4(N3′) M−N5(N2′) M−N6(N1′) N1−Cu + Cu− N3 N2−Cu + Cu− N5(N2′) N6−Cu + Cu− N4 N1−Cu−N3 N6−Cu−N4 N2−Cu−N5

[Cu(bmet)] (ClO4)2·H2O

[Cu(bmet)] (Br)2·2MeCN

[Zn(bmet)] (ClO4)2·H2O

2.131(4) 1.970(4) 2.178(5) 2.252(5) 2.015(4) 2.235(4) 4.309

2.171(7) 1.978(6) 2.210(7) 2.210(7) 1.978(6) 2.171(7) 4.381

2.202(3) 2.081(3) 2.206(3) 2.187(3) 2.087(3) 2.209(3) 4.408

3.985

3.956

4.268

4.487

4.381

4.396

154.77(16) 148.45(15) 171.32(16)

152.3(2) 152.3(2) 174.7(4)

150.37(10) 147.69(10) 169.328(10)

a

The Zn and Cu perchlorate monohydrate species are isomorphous in space group C2/c (Table S1); the bromo complex is in space group Pnna with crystallographically imposed twofold symmetry.

in [Cu(bmet)](ClO4)2·H2O is given in Figure 1. Crystallographic and metrical data for [Cu(bmet)](ClO4)2·0.375H2O are in Supporting Information Tables S2 and S3. Bond lengths and angles within the [Cu(bmet)]2+ cations in [Cu(bmet)](ClO4)2·H2O and [Cu(bmet)](Br)2·2MeCN recorded at 89 K are similar, but note that, in the latter complex, the CuII ion lies on a crystallographic twofold axis, thereby rendering pairs of Cu−N distances equivalent. In both cases, the [Cu(bmet)]2+ cation displays a distorted octahedral geometry with the ligand exhibiting a meridional-facial-facial-meridional (mf f m) configuration.45 The CuII ion is bonded to all six nitrogen atoms of the bmet ligand, with bond distances ranging from 1.970(4) to 2.235(4) Å (ClO4− salt) and from 1.978(6) to 2.210(7) Å (Br−− salt). In both complexes, the two Cu−Naliphatic bond distances (Cu1−N3, Cu1−N4; 2.178(5), 2.252(5) Å (ClO4− salt): Cu1−N3, Cu1−N3′; 2.210(7), 2.210 (7) Å (Br− salt)) are longer than the four Cu−Naromatic bond distances. Most notably, the Cu−N distances along the N2−Cu1−N5 (ClO4− salt) or N2−Cu1−N2′ (Br− salt) axis are significantly shorter than the other Cu−N distances, leading to an apparently compressed octahedral geometry. X-ray data were recorded at three temperatures (100, 200, and 300 K) for [Cu(bmet)](ClO4)2·0.375H2O on the same single crystal but for a tetragonal I41/acd polymorph of that in Table 1 (see Tables S2 and S3). Here the ethylenediamine C

DOI: 10.1021/acs.inorgchem.5b02287 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 1. ORTEP diagrams of the cations of the isomorphous [Cu(bmet)](ClO4)2 H2O (left) and [Zn(bmet)](ClO4)2 H2O (right) at 89 K. Perchlorate anions, the water molecule, and hydrogen atoms were omitted for clarity. The atomic displacement parameters for the Cu complex are noticeably higher than those for the Zn complex.

H2O, implying a similar coordination sphere around the metal ion and the same packing environment. The [Zn(bmet)]2+ cation displays a distorted octahedral geometry, with the ligand exhibiting, again, an mf f m configuration. The ZnII ion coordinates to all six nitrogen atoms of the bmet ligand, with bond distances ranging from 2.082(3) to 2.209(3) Å. As also found in the CuII cations, a compressed octahedral geometry is observed, with the two Zn−N bonds along the N2−Zn1−N5 axis being apparently shorter than the other four. Closer inspection of the X-ray structures of [Cu(bmet)](ClO4)2·H2O and [Cu(bmet)](Br)2·2MeCN reveals a possible masked disorder of an elongated Jahn−Teller axis in both cases. TLS analysis of both structures using PLATON46 gave evidence for significant librational effects along the bonds Cu1−N3, Cu1−N4, and Cu1−N6 (with respective ⟨d2⟩/Å2 values of 0.016, 0.021, and 0.011) in the monoclinic perchlorate salt and Cu1−N3 and Cu1−N3′ (⟨d2⟩/Å2 value of 0.013) in the bromide salt at 89 K. These observations are consistent with a disorder of the Jahn−Teller axis that is not immediately apparent from the ORTEP diagrams of the cations. The X-ray structure of the [Zn(bmet)]2+ cation displays no such librational effects, with ⟨d2⟩/Å2 values ranging from 0.0006 to 0.0036 for the Zn−N bonds. In the tetragonal structure [Cu(bmet)](ClO4)2·0.375H2O the ethylenediamine moiety of the bmet ligand is disordered, and a long and a short pair of Namine−Cu−Nimine axes emerges. Solid-State and Solution X- and Q-Band Continuous Wave Electron Paramagnetic Resonance Investigation. The [Cu(bmet)](ClO4)2·H2O, [Cu(bmet)](ClO4)2·0.375H2O, and [Cu(bmet)]Br2·2MeCN complexes exhibit the same powder EPR spectra, suggesting that the different counterions and different crystal packings do not significantly influence the electronic structures of the complexes. This is consistent with the fact that both structures exhibit a similar coordination sphere around the CuII ion. Therefore, the rest of the EPR investigation was performed using [Cu(bmet)](ClO4)2·H2O exclusively. The X-band EPR powder spectra of [Cu(bmet)](ClO4)2·H2O were recorded between 5 and 298 K. They all display similar isotropic signals, whose intensity increases as the temperature decreases, following a Curie law behavior expected for an S = 1/2 system (Figure 3).

moiety is clearly disordered. No noticeable change in the structural properties of this complex is observed with temperature in these structures. Although the structures of the [Cu(bmet)]2+ cations are similar, there are significant differences in the three-dimensional structures of [Cu(bmet)](ClO4)2·H2O and [Cu(bmet)](Br)2· 2MeCN. The structure of [Cu(bmet)](ClO 4 ) 2·H2 O is composed of layers of molecules in the (001) plane, defined by the a and b axes (Figure 2, bottom). The projection down

Figure 2. Packing diagrams of [Cu(bmet)](Br)2·2MeCN (top) and [Cu(bmet)](ClO4)2·H2O (bottom) viewed down the (A and C) b and (B and D) c axes.

the c-axis shows chains of encapsulated perchlorate anions. In contrast, [Cu(bmet)](Br)2·2MeCN (Figure 2, top) comprises pleated sheets, which create channels occupied by the MeCN molecules and Br− ions. The tetragonal [Cu(bmet)](ClO4)2· 0.375H2O shows a highly distinctive structure, where octets of the cationic complex form squares in the a−b plane and stack to surround columns parallel to axis c of highly disordered perchlorate ions. At the corners where the squares meet, a second, less disordered column of perchlorate ions is stacked (see Supporting Information). Selected bond lengths for [Zn(bmet)](ClO4)2·H2O are presented in Table 1, while an ORTEP diagram of the cation is given in Figure 1. The crystal structure of [Zn(bmet)](ClO4)2·H2O is isomorphous with that of [Cu(bmet)](ClO4)2· D

DOI: 10.1021/acs.inorgchem.5b02287 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 3. (left) Powder X-band EPR spectra of [Cu(bmet)](ClO4)2·H2O recorded between 97 and 230 K. (right) Field position of the maximum and minimum of the transition between 5 and 298 K.

and 200 K (Supporting Information). It can be clearly seen that the EPR properties of the complex are temperature-dependent and that the EPR spectra steadily evolve between two types of signatures from 10 to 298 K. At low temperatures, the EPR spectrum of [Cu(bmet)](ClO4)2·H2O displays a quasi-axial signal with g1 > g2 ≳ g3 > ge, indicating tetragonally elongated octahedra. At 298 K, the Q-band EPR spectrum is significantly different; a more rhombic signal with the g1 ≳ g2 > g3 > ge pattern is now observed. Another strategy to obtain well-resolved EPR spectra is to record the data in solution. An MeCN solution of [Cu(bmet)](ClO4)2·H2O exhibits a steady change of the X-band EPR spectra as a function of temperature (Table 2), with a comparable behavior to that observed for the powder sample (Supporting Information). The low-quality resolution of the spectra, which is a result of the very low concentrations required to observe the hyperfine coupling interactions, precludes a precise analysis of the evolution of the intensity of each transition as a function of the temperature. These low concentrations, coupled with the lower sensitivity of the EPR spectrometer at the Q-band frequency, meant that no Q-band EPR signal could be observed in solution. Density Functional Theory Calculations. To obtain further information on the role of the bmet ligand in generating and stabilizing a possible compressed octahedral geometry around the CuII metal center, DFT calculations were performed. One of our objectives was to determine whether or not an elongated distortion could be afforded in a CuII complex of the bmet ligand, especially since the ZnII complex displays a marked compression. First, fully DFT-optimized structures, initiated from the X-ray structures of the cations in [Cu(bmet)](ClO4)2·H2O and [Cu(bmet)](Br)2·2MeCN, were calculated. The two resulting optimized structures are identical and denoted Cu(bmet)*. The EPR parameters were predicted by DFT calculations on Cu(bmet)* (Table 2), and the calculated g- and A1-values nicely reproduce the experimental powder and solution spectra of [Cu(bmet)]2+ at low temperature, thereby validating our approach. The calculated Cu(bmet)* structure displays six different Cu−N bond distances (Table 3), with two long Cu−N distances along the N4−Cu−N6 axis (summing to 4.654 Å) and four short Cu−N distances along the N1−Cu−N3 and N2−Cu−N5 axes (summing to 4.259 and 4.001 Å, respectively), resulting in an elongated octahedral geometry.

Careful examination of the spectra shows that the temperature does not affect the field position of the minimum of the EPR transition. However, this is not the case for its maximum, which is slightly shifted to higher fields when the temperature decreases (Figure 3). To improve the resolution of the g-anisotropy, powder Qband EPR spectra were recorded (Figure 4), allowing the

Figure 4. Experimental and simulated (black line) powder Q-band EPR spectra of [Cu(bmet)](ClO4)2·H2O at 298 K (red) and 20 K (blue). Parameters used for the simulation: g1 = 2.196, g2 = 2.145 g3 = 2.030 at 298 K; g1 = 2.193, g2 = 2.058 g3 = 2.054 at 20 K.

determination of the components of the g-tensor (Table 2). Qband EPR spectra were obtained at temperatures between 15 Table 2. Experimental EPR Parameters Determined for [Cu(bmet)](ClO4)2·H2O under Different Experimental Conditions Together with the Calculated EPR Parameters Predicted for Cu(bmet)*

expt [Cu(bmet)] (ClO4)2·H2O

calcd Cu(bmet)*

phase

T, K

g1

g2

g3

powder

298

2.196

2.145

2.030

powder MeCN MeCN

20 225 26

2.193 2.170 2.195 2.180

2.058 2.135 2.081 2.084

2.054 2.035 2.034 2.039

A1, MHz

480 562 E

DOI: 10.1021/acs.inorgchem.5b02287 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Table 3. Principal Cu−N Bond Distances and N−Cu−N Axis Lengths (Å) in Cu(bmet)* and Zn(bmet)* Cu(bmet)* Zn(bmet)* a

Cu−N1

Cu−N2

Cu−N3

Cu−N4

Cu−N5

Cu−N6

N1−Cu−N3a

N2−Cu−N5a

N4−Cu−N6a

2.095 2.202

1.965 2.113

2.164 2.236

2.371 2.231

2.036 2.112

2.283 2.208

4.259 4.438

4.001 4.225b

4.654b 4.439

Sum of the two bond lengths. bValues in bold show the most elongated or compressed axis in the octahedron.

configuration having a pronounced {dx2−y2}1 character consistent with an elongated octahedron. However, the rhombic EPR signal observed at room temperature arises from a mixture of high-temperature dynamically average conformations of two octahedra elongated along the two Namine−Cu−Nimine axes (z and y, respectively) resulting in an apparent compression along the molecular x-axis, corresponding to the {dx2−y2}1 and {dx2−z2}1 ground states. In addition, the g3 value found in [Cu(bmet)]2+ is notably higher than those values found in previously reported dynamic Jahn−Teller systems. Consistent with the low-temperature EPR spectrum, DFT calculations demonstrated that the steric constraints imposed by the bmet ligand cannot completely prevent the adoption of an elongated octahedral geometry. Indeed, the fully optimized structure Cu(bmet)* displays a somewhat elongated highly distorted octahedral geometry associated with an electronic configuration having a pronounced mixed 40% 3dx2−y2 and 29% 3dx2−z2 character. However, no significant variations in the X-ray determined Cu−N distances are observed for [Cu(bmet)](ClO4)2· 0.375H2O over the 100−300 K temperature range, except that at 100 K, the Namine−Cu−Nimine axis is marginally longer than at 200 and 300 K, consistent with the more axial EPR spectrum observed at low temperatures (Table 2). Nonetheless, taking also into account the significant librational motions (observed for [Cu(bmet)](ClO4)2·H2O and [Cu(bmet)](Br)2· 2MeCN at 89 K) and disorder effects (observed for [Cu(bmet)](ClO4)2·0.375H2O at 300, 200, and 100 K), the Cu−N distances and angles should be interpreted with caution. Indeed, X-ray diffraction is a technique that usually provides time-averaged structural information, whereas EPR probes the local electronic environment of the copper sites on a much shorter time scale (in the nanoseconds). In the case of [Cu(bmet)](ClO4)2·0.375H2O, where the ethylenediamine bridge is disordered, the axial bond lengths at 100 K are very similar to those calculated for [Cu(bmet)]2+ (Cu(bmet)*), starting from the less obviously disordered [Cu(bmet)](ClO4)2·H2O and [Cu(bmet)](Br)2·2MeCN structures. The expected evolution of the structural properties of [Cu(bmet)]2+ as a function of the temperature may thus be masked and, indeed, may not be discernible, given the asymmetry of the complex revealed crystallographically and by DFT calculations, which together did not strongly define a unique Jahn−Teller elongation axis. In summary, the experimental data can be interpreted as a Jahn−Teller effect perturbed by “ligand effects”. The constraint of the rigid bmet ligand leads to short M−N bonds along the tetragonal axis that opposes the Jahn−Teller effect. However, the dissymmetry of the bmet ligand,26 in which the nature of the trans ligands is different (a pyridine trans to an amine) breaks the symmetry of the complex inducing a pseudo Jahn− Teller effect.47 The pseudo Jahn−Teller effect is a variant of the Jahn−Teller effect that occurs when the ground state interacts with excited states in nondegenerate systems. Despite the fact that temperature-dependent structural changes in the X-ray structure of [Cu(bmet)]2+ are quite

Calculation of the singly occupied molecular orbital (SOMO) of Cu(bmet)* showed it to be localized on the CuII ion, with 40% 3dx2−y2 and 29% 3dx2−z2 character suggesting that this orbital does not display the character of a pure single d orbital (x axis along the N2−Cu−N5 axis). A fully DFT-optimized structure, initiated from the X-ray structure of the cation in [Zn(bmet)](ClO4)2·H2O, denoted Zn(bmet)*, was also calculated, to determine if the structural modifications observed in Cu(bmet)* will also occur during the optimization process. The comparison between the experimental and calculated structures (Tables 1 and 3) provides evidence that the compressed octahedral geometry experimentally observed around the ZnII ion is preserved throughout the optimization process. Notably also for Zn(bmet)* there are three pairs of different Zn−N distances compared to six distinctly different distances for Cu(bmet)*, highlighting the complex orbital mixing occurring for Cu(bmet)*.

■

DISCUSSION The [Cu(bmet)]2+ complex cation displays an apparently compressed octahedral geometry in its X-ray structures, at different temperatures, and in the presence of different counteranions. This compression is favored by the presence of the rigid hexadentate ligand bmet, which imposes an mf f m configuration45 around the copper center, similar to that afforded by two meridional tridentate ligands. The constraints associated with bmet contribute to particularly short M−N bonds involving the mutually trans pyridine rings along the tetragonal axis. A compressed tetragonal distortion is found also in the previously reported CoIII, MnII, FeII, and NiII analogues26 and, most interestingly, in the parent d10 [Zn(bmet)]2+ complex. In this latter species, the compression must mainly originate from the steric constraints imposed by the ligand itself (additional packing effects can be also envisaged) because no electronic factors can affect this d10 system. However, the compressed octahedral geometry apparently observed in [Cu(bmet)](ClO4)2·H2O is more pronounced than in the isomorphous [Zn(bmet)](ClO4)2·H2O complex. Indeed, if we calculate Δe‑a = (∑aequatorial)/4) − (∑aaxial)/2) with a being the different Cu−N bond distances, the Δe‑a values are 0.12 and 0.21 Å for the Zn and Cu derivatives, respectively. This difference is indicative of an additional Jahn−Teller effect in the case of [Cu(bmet)](ClO4)2·H2O. As previously described for compressed octahedral CuII complexes, the tetragonal compression can be “real” or “apparent”, the latter resulting from a dynamic Jahn−Teller distortion. While the structural and EPR properties are temperature-independent for “true” compressed systems, these properties vary with temperature in the case of an “apparent” compression. The case of [Cu(bmet)] 2+ is ambiguous. Indeed, while the EPR spectra of [Cu(bmet)]2+ clearly evolve as a function of the temperature, such an evolution is not obvious in the X-ray crystal structures at different temperatures. The axial EPR signal recorded at low temperature can be assigned to a CuII ion with an electronic F

DOI: 10.1021/acs.inorgchem.5b02287 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

(12) Narr, E.; Zimmermann, H.; Godt, A.; Goldfarb, D.; Jeschke, G. Phys. Chem. Chem. Phys. 2003, 5, 3959−3967. (13) Hitchman, M. A.; Maaskant, W.; van der Plas, J.; Simmons, C. J.; Stratemeier, H. J. Am. Chem. Soc. 1999, 121, 1488−1501. (14) Simmons, C. J.; Stratemeier, H.; Hitchman, M. A.; Reinen, D.; Masters, V. M.; Riley, M. J. Inorg. Chem. 2011, 50, 4900−4916. (15) Atanasov, M.; Hitchman, M. A.; Hoppe, R.; Murray, K. S.; Moubaraki, B.; Reinen, D.; Stratemeier, H. Inorg. Chem. 1993, 32, 3397−3401. (16) Masters, V. M.; Riley, M. J.; Hitchman, M. A. J. Synchrotron Radiat. 1999, 6, 242−243. (17) Steffen, G.; Reinen, D.; Stratemeier, H.; Riley, M. J.; Hitchman, M. A.; Matthies, H. E.; Recker, K.; Wallrafen, F.; Niklas, J. R. Inorg. Chem. 1990, 29, 2123−2131. (18) Reinen, D. Inorg. Chem. 2012, 51, 4458−4472. (19) Solanki, N. K.; McInnes, E. J. L.; Mabbs, F. E.; Radojevic, S.; McPartlin, M.; Feeder, N.; Davies, J. E.; Halcrow, M. A. Angew. Chem., Int. Ed. 1998, 37, 2221−2223. (20) Holland, J. M.; Liu, X.; Zhao, J. P.; Mabbs, F. E.; Kilner, C. A.; Thornton-Pett, M.; Halcrow, M. A. J. Chem. Soc., Dalton Trans. 2000, 0, 3316−3324. (21) Henke, W.; Reinen, D. Z. Z. Anorg. Allg. Chem. 1977, 436, 187− 200. (22) Mack, K.; Wunsche von Leupoldt, A.; Forster, C.; Ezhevskaya, M.; Hinderberger, D.; Klinkhammer, K. W.; Heinze, K. Inorg. Chem. 2012, 51, 7851−7858. (23) Meyer, A.; Schnakenburg, G.; Glaum, R.; Schiemann, O. Inorg. Chem. 2015, 54, 8456−8464. (24) Garcia-Fernandez, P.; Trueba, A.; Barriuso, M. T.; Aramburu, J. A.; Moreno, M. Dynamic and Static Jahn-Teller Effect in Impurities: Determination of the Tunneling Splitting. In Vibronic Interactions and the Jahn-Teller Effect; Springer: Netherlands, 2012; Vol. 23, p 6. (25) Baird, L. J.; Black, C. A.; Blackman, A. G. Polyhedron 2007, 26, 378−384. (26) Hall, N. A.; Duboc, C.; Collomb, M. N.; Deronzier, A.; Blackman, A. G. Dalton Trans. 2011, 40, 12075−12082. (27) APEX-II; Bruker AXS Inc: Madison, WI, 2004. (28) SAINT; Bruker AXS Inc: Madison, WI, 2001. (29) Sheldrick, G. M. SADABS, Program for Absorption Correction, Program for Absorption Correction; University of Goettingen: Goettingen, Germany, 1996. (30) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.; Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna, R. J. Appl. Crystallogr. 1999, 32, 115−119. (31) Burla, M. C.; Caliandro, R.; Camalli, M.; Carrozzini, B.; Cascarano, G. L.; De Caro, L.; Giacovazzo, C.; Polidori, G.; Spagna, R. J. Appl. Crystallogr. 2005, 38, 381−388. (32) Farrugia, L. J. Appl. Crystallogr. 1999, 32, 837−838. (33) Sheldrick, G. M.; Schneider, T. R. Methods in Enzymology. In Macromol. Cryst. B 1997; Vol. 277, pp 319−343.10.1016/S00766879(97)77018-6 (34) Neese, F. Wiley Interdiscip. Rev. Comput. Mol. Sci. 2012, 2, 73− 78. (35) Perdew, J. P. Phys. Rev. B: Condens. Matter Mater. Phys. 1986, 33, 8822−8824. (36) Perdew, J. P. Phys. Rev. B: Condens. Matter Mater. Phys. 1986, 34, 7406−7406. (37) Becke, A. D. Phys. Rev. A: At., Mol., Opt. Phys. 1988, 38, 3098− 3100. (38) Schäfer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100, 5829−5835. (39) Neese, F. J. Comput. Chem. 2003, 24, 1740−1747. (40) Weigend, F. Phys. Chem. Chem. Phys. 2006, 8, 1057−1065. (41) Becke, A. D. J. Chem. Phys. 1993, 98, 1372−1377. (42) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37, 785−789. (43) Pantazis, D. A.; Neese, F. J. Chem. Theory Comput. 2009, 5, 2229−2238.

small, the observed temperature dependence of the EPR spectra argues in favor of a dynamic (pseudo) Jahn−Teller effect that leads to subtle changes in 3dx2−y2 and 3dx2−z2 orbital mixing in this low-symmetry system. The investigation of CuII systems with bmet ligands containing bulky groups on the external pyridine should validate this interpretation. Indeed, this greater steric restriction may allow the trapping of long Nimine−Cu bonds along the Nimine−Cu−Namine axis, thus leading to real compressed octahedral geometries.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.5b02287. Illustrated structures, experimental powder Q-band and X-band EPR spectra, DFT-calculated SOMO of Cu(bmet)*, tabulated summary of X-ray crystallographic data, tabulated selected bond lengths and angles. (PDF) X-ray crystallographic information. (CIF) X-ray crystallographic information. (CIF) X-ray crystallographic information. (CIF) X-ray crystallographic information. (CIF) X-ray crystallographic information. (CIF) X-ray crystallographic information. (CIF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. (C.D.) *E-mail: [email protected]. (A.G.B.) Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS N.H. thanks the Département de Chimie Moléculaire for her Ph.D. fellowship. This work has been partially supported by the Labex ARCANE (ANR-11-LABX-0003-01). The authors also thank the TGE Réseau National de RPE interdisciplinaire, FR3443, and the COST Action CM1305 EcostBio (Explicit Control Over Spin-states in Technology and Biochemistry). We are especially grateful for insightful and helpful comments from the referees.

■

REFERENCES

(1) Jahn, H. A.; Teller, E. Proc. R. Soc. London, Ser. A 1937, 161, 220− 235. (2) Bersuker, I. B. Chem. Rev. 2001, 101, 1067−1114. (3) Bersuker, I. B. In Electronic Structure and Properties of Transition Metal Compounds; John Wiley & Sons, Inc: 2010; p 238−323. (4) Cotton, F. A.; Wilkinson, G.; Murillo, C. A.; Bochmann, M. Advanced Inorganic Chemistry; Wiley-Interscience: New York, 1999. (5) Folgado, J. V.; Henke, W.; Allmann, R.; Stratemeier, H.; BeltranPorter, D.; Rojo, T.; Reinen, D. Inorg. Chem. 1990, 29, 2035−2042. (6) Goodman, B. A.; Raynor, J. B. In Advances in Inorganic Chemistry and Radiochemistry; Emelus, H. J., Sharpe, A. G., Eds.; Academic Press: 1970; Vol. 13, pp 135−362. (7) Hathaway, B. J.; Billing, D. E. Coord. Chem. Rev. 1970, 5, 143− 207. (8) Hathaway, B. J. Struct. Bonding (Berlin) 1984, 57, 55. (9) Halcrow, M. A. Dalton Trans. 2003, 4375−4384. (10) Halcrow, M. A. Chem. Soc. Rev. 2013, 42, 1784−1795. (11) Allmann, R.; Henke, W.; Reinen, D. Inorg. Chem. 1978, 17, 378−382. G

DOI: 10.1021/acs.inorgchem.5b02287 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry (44) Pantazis, D. A.; Chen, X. Y.; Landis, C. R.; Neese, F. J. Chem. Theory Comput. 2008, 4, 908−919. (45) Hartshorn, R. M.; House, D. A. J. Chem. Soc., Dalton Trans. 1998, 2577−2588. (46) Spek, A. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2009, 65, 148−155. (47) Bersuker, I. B. Chem. Rev. 2013, 113, 1351−1390.

H

DOI: 10.1021/acs.inorgchem.5b02287 Inorg. Chem. XXXX, XXX, XXX−XXX