Effect of the Substitution Pattern (Peripheral vs Non-Peripheral) on the

Institute of Applied Physics of the Academy of Sciences of Moldova, Academiei strasse 5, Chisinau , Moldavia. Inorg. Chem. , Article ASAP. DOI: 10.102...
24 downloads 3 Views 2MB Size
Article Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX

pubs.acs.org/IC

Effect of the Substitution Pattern (Peripheral vs Non-Peripheral) on the Spectroscopic, Electrochemical, and Magnetic Properties of Octahexylsulfanyl Copper Phthalocyanines Tulin Ateş Turkmen,† Lihan Zeng,‡ Yan Cui,‡ Iṡ mail Fidan,§ Fabienne Dumoulin,*,§ Catherine Hirel,§ Yunus Zorlu,§,△ Vefa Ahsen,§ Alexander A. Chernonosov,∥ Yurii Chumakov,†,⊥ Karl M. Kadish,*,‡ Ayşe Gül Gürek,*,§ and Sibel Tokdemir Ö ztürk*,† †

Department of Physics, §Department of Chemistry, and △Institute of Nanotechnology, Gebze Technical University, 41400 Gebze, Kocaeli, Turkey ‡ Department of Chemistry, University of Houston, Houston, Texas 77204-5003, United States ∥ Institute of Chemical Biology and Fundamental Medicine, 8 Lavrentiev Avenue, Novosibirsk 630090, Russia ⊥ Institute of Applied Physics of the Academy of Sciences of Moldova, Academiei strasse 5, Chisinau, Moldavia S Supporting Information *

ABSTRACT: In order to investigate the substitution position effect on the spectroscopic, electrochemical, and magnetic properties of copper phthalocyanines, a detailed structure− property analysis has been performed by examining two copper phthalocyanines that are octasubstituted by hexylsulfanyl chains respectively in the peripheral (Cu-P) and non-peripheral (CuNP) positions. Cu-NP showed a marked near-IR maximum absorption compared to Cu-P and, accordingly, a smaller HOMO−LUMO energy gap, calculated via the electrochemical results and simulations in the gas phase, as well as for Cu-NP from its crystallographic data. An electron-spin resonance (ESR) technique is used to extract the g values from the powder spectra that are taken at room temperature. The g values were determined to be g∥ = 2.160 and g⊥ = 2.045 for Cu-P and g∥ = 2.150 and g⊥ = 2.050 for Cu-NP. These values indicate that the paramagnetic copper center in both phthalocyanines has axial symmetry with a planar anisotropy (g∥ > g⊥). The ESR spectra in solution could be obtained only for Cu-P. Curie law is used to fit the experimental data of the magnetic susceptibility versus temperature graphs, and the Curie constant (C) and diamagnetic/temperature-independent paramagnetic (α) contributions are deduced as 0.37598 (0.39576) cm3·K/mol and −23 × 10−5 (25 × 10−5) cm3/mol respectively for Cu-P and Cu-NP. The room temperature magnetic moment value (1.70 μB) is close to the spin-only value (1.73 μB) for the peripheral complex, showing that there is no orbital contribution to μeff. In contrast, at room temperature, the value of the magnetic moment (1.77 μB) is above the spin-only value, showing an orbital contribution to the magnetic moment. Cu-NP’s room temperature magnetic moment value is larger than the value for Cu-P, demonstrating that the orbital contribution to the magnetic moment depends upon the substituent position. The magnitudes of the effective magnetic moment values also support that both Cu-P and Cu-NP complexes have square-planar coordination. This result is consistent with the determined g values. The spin densities were determined experimentally, and the results suggest that the positions of the substituents affect these values (0.469 for Cu-P and 0.490 for Cu-NP). known to affect these properties.2 Most Pcs have a squareplanar structure,3 promoting intermolecular interactions, such as aggregation, liquid crystallinity, or specific packing. Numerous applications benefit from the Pcs’ maximum absorption commonly centered at ∼700 nm, and efforts have been made to further red-shift this band. In addition to providing an extension of the electronic delocalization in the

1. INTRODUCTION Phthalocyanines (Pcs) are molecular materials exhibiting properties exploited in various applications, such as pigments in paints and printing inks, IR security devices, information storage and computer disk writing, or photodynamic therapy of cancer.1 Many of these applications are based on the spectroscopic, electronic, and/or photochemical properties of the compounds. The nature, position, number, and even bulkiness of the substituents on the Pcs, as well as the nature of the metal and possible axial substituents, are structural factors © XXXX American Chemical Society

Received: February 28, 2018

A

DOI: 10.1021/acs.inorgchem.8b00528 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 1. Structure of the Pcs investigated.

ature-dependent magnetization and have been fitted to a suitable function to obtain the Curie constant (C) and the diamagnetic/temperature-independent paramagnetic (TIP) contribution (α) and Weiss constant (θ). The Curie constant was used to determine the spin magnetic moment. The sign of the Weiss constant reflects the nature of the interactions. Positive values are representative of ferromagnetic coupling, while negative values are representative of antiferromagnetic coupling. The magnitude of θ, in either the positive or negative direction, is indicative of the strength of the intermolecular interactions between spins. The results and interpretations have been supported by calculations.

case of naphthalocyanines,4 fused derivatives,5 or metalation by group 15 elements,6 non-peripheral octasulfanyl substitution has proven to be an easy way to shift the Pcs’ maximum absorption up to 800 nm. On the other hand, and except for the specific case of lanthanide double-decker complexes,7 studies of the magnetic properties exhibited by appropriately metalated derivatives have been limited mainly to unsubstituted Pcs: the magnetic properties of unsubstituted copper phthalocyanine have been investigated in detail by many laboratories.8 Nonetheless, the effect of the substitution pattern, and, in particular, of the position of the substituents, has remained unexplored. In order to address this point, we have performed a structure−property analysis of this effect by examining two copper phthalocyanines that are octasubstituted by hexylsulfanyl chains respectively in the peripheral (Cu-P) and non-peripheral (Cu-NP) positions of the four benzene rings on the macrocycle (Figure 1). These substituents have been selected because they confer an appropriate solubility to the Pc core and an important variation of the maximum absorption in relation to their position. Octasubstitution, unlike tetrasubstitution, avoids obtainment of the Pcs as regioisomeric mixtures, hence allowing them to work on well-defined compounds. After comparative analyses of the electronic and electrochemical properties of the two compounds, the investigation of their magnetic properties was carried out with data from magnetic susceptibility measurements and electron-spin resonance (ESR) spectroscopy. The use of vibrating-sample magnetometry (VSM) at constant magnetic field intensity allowed for determination of the compounds’ magnetic moment and magnetic susceptibility, which, in turn, gave information on intermolecular interactions between the metal centers of neighboring molecules and on the orbital contribution to the spin. ESR measurements at room temperature were used to determine the g values of each molecule, giving information not only about the paramagnetic atom itself but also about its surroundings, in particular about the symmetry of the paramagnetic center. Additional lowtemperature solution ESR measurements were used to determine hyperfine and superhyperfine interactions besides the g values respectively for copper and nitrogen atoms. Magnetic susceptibility data were deduced from the temper-

2. RESULTS AND DISCUSSION Absorption Spectroscopy. Ground-state electronic absorption spectra of Cu-P and Cu-NP were recorded in tetrahydrofuran (THF) and chloroform (Figure 2) and provide

Figure 2. UV−vis spectra of Cu-P (green) and Cu-NP (brown) in THF (solid line) and chloroform (dashed line).

evidence for the significant effect of the substituent position on the maximum absorption wavelengths and on the degree of aggregation in both solvents (see Table S1 for spectra at different concentrations). Compared to THF, chloroform induces slight hypochromic and bathochromic shifts (6 nm for the peripheral Pcs and 14 nm for the non-peripheral Pcs) for each compound. The aggregation of Pcs depends on the B

DOI: 10.1021/acs.inorgchem.8b00528 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 3. Left: ORTEP drawing with atomic labeling for the compound studied. Thermal ellipsoids are shown with a 50% probability level. Middle: Fragment of molecular packing in the crystal of Cu-NP. Right: Crystal packing of Cu-NP representing the stacking column along the [100] direction. Hydrogen atoms were omitted from the packing diagram for clarity.

indicates a different crystal packing of these compounds. In the solid phase, unsubstituted CuPc forms neighboring columns, resulting in a herringbone-type structure.12 Metalated Pcs are macrocyclic planar aromatic molecules in which a central metal atom is bound to the organic structure through four inwardly projecting nitrogen atoms. A striking feature of the studied Pcs is their stacking mode within a molecular column; the molecular stacking direction projected on a molecular plane is different. The metal is in a distorted octahedral environment, where the two nitrogen atoms from the adjacent molecules are coordinated at the axial positions. The apical distances of Cu− N are equal to 3.283 Å, and the N−Cu−N angles are equal to 180°.13 Electrochemistry. The electrochemical behavior of copper phthalocyanines was investigated by cyclic voltammetry. The effect of the substitution pattern on the redox potentials was analyzed comparatively in both pyridine and CH2Cl2. The eight hexylthio groups greatly enhance the solubility, and both Pcs are very soluble in pyridine and CH2Cl2, with no significant aggregation in either solvent.14 It is well-known that pyridine, when used as the solvent, may also bind with the central metal ion of copper phthalocyanines and thus prevent aggregation. Additionally, the time scale of the measurement is different in pyridine, with spectral changes on the slower time scale being better defined. Redox processes of copper phthalocyanines occur only on the macrocyclic ring because the central copper metal ion is redox-inactive. Cu-P exhibits three well-defined one-electron reductions in both pyridine and CH2Cl2 when the potential sweep is reversed at −2.0 V (Figure 4). A fourth reduction is also possible at more negative potentials.15 The first reduction of Cu-P in CH2Cl2 is not fully reversible, maybe because of the unstability16 of the trianionic CuPc forms [Cu2+Pc3−]−. The reduction potentials of Cu-P in pyridine and CH2Cl2 are quite similar to each other, although the first reduction in CH2Cl2 is 0.10 V lower compared to that in pyridine. Cu-NP displays similar redox behavior, with the first and second reductions being at almost the same potentials as the first two reductions of Cu-P. On the other hand, the third reduction of Cu-NP is located at −1.39 V in pyridine, while no reduction process is seen in CH2Cl2 in this potential range. Each complex is also characterized by two successive quasireversible or irreversible oxidation waves in either solvent

structural factors (number, position, and nature of the substituents and nature of the metal) and also on external factors such as the temperature, concentration, or solvent itself. As for similar compounds with a central zinc atom,2 THF appeared to be a solvent favoring the monomeric state of Pcs, whereas halogenated solvents (dichloromethane or chloroform) induce aggregation to a greater extent, probably because of the possible coordination of THF on the central metal. As was previously observed for zinc phthalocyanines, the nonperipheral substitution pattern induces a bathochromic shift of nearly 80 nm in THF and 90 nm in chloroform. Solid-State Molecular Arrangement. The magnetic properties (to be discussed in the next section) depend on both molecular (nature of the metal and of the ligand) and intermolecular parameters such as the distance between paramagnetic centers. To be able to interpret future measurements, taking all of these parameters into account, data on the relative molecular arrangement of the molecules in the solid state were collected. In line with the previously reported facile obtainment of suitable crystals of the non-peripheral octasulfanyl phthalocyanines,9 we obtained suitable crystals of Cu-NP and appropriate XRD data, which are reported for the first time for this compound. The fact that Cu-P is waxy at room temperature was already reported.10 The crystal packing of Cu-NP has a columnar structure (Figure 3), as can be expected for Pcs. When the angles between the pair of the planes generated by the isoindole units on either side of the Pc are compared, the ring planarity of Cu-NP is nearly similar to that of previously reported metal-free (CSD ref. Code JUBPON)2c and lead (CSD ref code AJUVIK-1B9a) analogues but is different from other structures of lead analogues (CSD ref codes AJUVIK-019b and AJUVIK-1A;9a see Table S4 and Figure S11). These experimental X-ray parameters provide evidence that the central copper atom has a four-coordinate, square-planar geometry, without contribution of the substituents to its coordination sphere. The apical distances of Cu−S7, 3.272 (1) Å (1 − x, 1 − y, 1 − z), and Cu−S5, 3.499 (1) Å (−x, 1 − y, 1 − z), are greater than the sum of the van der Waals radii [rvdw(Cu) + rvdw(S) = 3.20 Å],11 and the S7−Cu−S5 angle is equal to 162.72(3)°. To obtain insight into the packing of the other derivatives, their powder XRD (PXRD) spectra were recorded (Figure S1). The different behavior of the PXRD profiles of each derivative C

DOI: 10.1021/acs.inorgchem.8b00528 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

In summary, Cu-P and Cu-NP display three successive reductions and two oxidations. Besides, according to previous reports, one cannot exclude an oxidation of the substituents, which anyway does not affect the characterization of the neutral species.17 Calculations of the HOMO−LUMO Orbitals and Energy Gaps. The energy gaps between the highest occupied molecular orbitals (HOMOs) and the lowest unoccupied molecular orbitals (LUMOs) have been determined for both compounds using three methods: (i) in solution from the electrochemical data, (ii) in the gas phase after optimization of their geometry (Figure S2), and (iii) in the solid state for CuNP, using its X-ray structure for electronic structure calculation (Figure S3). The three related values are presented in Table 2. Although the relaxation and solvation effects are not taken into account by this method, the average HOMO−LUMO energy gap can be extracted from the electrochemistry data because it corresponds to the potential difference between the first oxidation and the first reduction of the conjugated macrocycles. It is equal to the energy difference between the eg and a1u orbitals in which the π−π* transitions of the Q band are involved. The values of the gap range from 1.33 to 1.57 V (Table 2) and are similar to values in the literature.18,19 The nonperipheral derivative Cu-NP has a smaller gap in comparison to that of the peripheral derivative Cu-P, which is in accordance with the longer-wavelength absorption of Cu-NP. The electron density distributions at HOMO and LUMO (spin-up and spin-down) determined from the X-ray structure for Cu-NP are displayed in Figure S3; those determined by density functional theory (DFT) calculations after geometry optimization are presented in Figures S4 and S5. Figure 5 shows the HOMO and LUMO calculated from the X-ray structure of Cu-NP and the model of Cu-P spin-up electrons. The plot of the total spin densities is given in Figure S12. These DFT calculations indicate that the HOMOs of Cu-NP and CuP are located on the isoindole moieties of the macrocycles coupled with the contribution from the sulfur atoms. The LUMOs of Cu-NP (from X-ray) and Cu-P are localized on two opposite isoindole units and nitrogen bridges (Figure 7), while the LUMO of model Cu-NP is distributed mainly on the isoindole moieties (Figure S5). The calculated values of the energy gaps varied depending upon the method used to determine them.20 This is because the calculations were carried out without taking into account the solvents used in the electrochemical measurements. However, whatever the method employed, the same trend is observed, with the energy gaps for Cu-P being higher than those for Cu-NP. Determination of the g Values by ESR Measurements. Powder ESR spectra (simulated and experimental) of peripheral and non-peripheral copper phthalocyanines are

Figure 4. Cyclic voltammograms of Cu-P and Cu-NP in pyridine and CH2Cl2 containing 0.1 M TBAP. Concentrations for the measurements in pyridine: 5.98 × 10−4 M for Cu-P; 5.07 × 10−4 M for Cu-NP. Concentrations for the measurements in CH2Cl2: 6.99 × 10−4 M for Cu-P; 5.02 × 10−4 M for Cu-NP.

(Figure 4). In pyridine, Cu-P is oxidized at 0.74 V, while the first oxidation of Cu-NP in pyridine is located at 0.48 V. The easier oxidation of Cu-NP is in harmony with its longer wavelength absorption. All of the measured potentials for CuNP and Cu-P in pyridine and CH2Cl2 are summarized in Table 1. Table 1. Summary of Redox Potentials versus SCE for Cu-P and Cu-NP in Pyridine and CH2Cl2 Containing 0.1 M TBAP Cu-P Cu-NP

pyridine CH2Cl2 pyridine CH2Cl2

Ox2

Ox1

Red1

Red2

Red3

0.94 0.85 1.00 1.21

0.74 0.57 0.48 0.50

−0.83 −0.93 −0.85 −0.85

−1.14 −1.14 −1.18 −1.24

−1.91 −1.90 −1.39

Table 2. Determination of the HOMO and LUMO Energies and the HOMO−LUMO Energy Gaps DFT after geometry optimization Cu-NP Cu-P

a

DFT from the X-ray structure

electrochemistry

MO

EHOMO (eV)

ELUMO (eV)

Egap (eV)

EHOMO (eV)

ELUMO (eV)

Egap (eV)

Egap (average) (eV)

spin-up spin-down spin-up spin-down

−4.68 −4.69 −5.33 −5.34

−2.76 −2.73 −3.21 −3.19

1.92 1.96 2.12 2.15

−4.30 −4.31

−2.57 −2.55

1.73 1.76

1.33a 1.35b 1.57a 1.50b

In pyridine. bIn dichloromethane. D

DOI: 10.1021/acs.inorgchem.8b00528 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

spectrum matches the experimental spectrum, after optimization of the input values. The extracted values of g are summarized in Table 3. For Cu-P, the g values were determined to be g∥ = 2.160 and g⊥ = 2.045. Both values are in agreement with the data from an earlier study.21 The average g value (gav) is calculated as 2.083. For Cu-NP, the g values were determined to be very close, g∥ = 2.150 and g⊥ = 2.050, and gave the same gav value (2.083) as that of Cu-P. The values between both complexes are in the range of other experiments conducted on copper phthalocyanines with different substituents.8b The measured components of the g factors indicate that the paramagnetic copper center in both Pcs is axially symmetric, with gz = g∥ and gx = gy = g⊥.22 The g values of copper for both complexes are on the order of g∥ > g⊥ > ge (free electron g value, ge = 2.0023), which implies that the dx2−y2 orbital is occupied by an unpaired electron.23 No hyperfine splittings could be observed for either Pcs in the powder samples. For both cases, g∥ > g⊥, which indicates a planar anisotropy where the copper is within the plane of the four inner nitrogen atoms of the Pc. The unpaired electron remains localized in dx2−y2, which spreads into the molecular plane with a σ characteristic, which prevents an overlap of the wave function with the neighboring Pc atoms. Nitrogen atoms of the neighboring Pcs are too far to establish extra axial coordination. The critical distance for magnetic exchange interactions is actually possible when the distance between the metal and nitrogen atoms is around 3.4 Å.24 This was previously observed for a MnPc complex.25 The ligation of the manganese atom by pyridine changed its coordination from 4 to 6. This induced a change in its magnetic properties because the distance between the axially located nitrogen atoms and the manganese atom was 3.4 Å, providing a suitable pathway for magnetic exchange interactions. The X-ray structure of Cu-NP confirms that the copper atom does not coordinate with the atoms of the nearest-neighboring molecules. Several attempts have been done to record the ESR spectra in solution because it would give better insight into their electronic structure, thanks to a better resolution and the possibility of distinguishing hyperfine and superhyperfine splittings. Three different solutions (chloroform, THF, and pyridine) of Cu-P and Cu-NP (2 × 10−3 M) have been prepared first for the ESR experiments. At room temperature, the spectra were not much different from the powder spectra. At lower temperature (110 K), only the ESR spectrum of Cu-P in THF showed resolved parallel hyperfine features due to copper (I = 3/2). In the low-field region, the further split due to superhyperfine interaction with four equivalent nitrogen nuclei (I = 1) could not be resolved for the first two of them because of intermolecular dipolar interactions, but the last one at the high field clearly shows this interaction. In the perpendicular region, partial resolution is, however, observed, with the superhyperfine features due to nitrogen atoms. Figure 7 shows the experimental and simulated spectra of Cu-P concentrated in THF. Unfortunately, the respective intensities

Figure 5. 3D plots of the HOMO and LUMO (spin-up electrons) for Cu-NP (from the X-ray data) and Cu-P (after geometry optimization).

given in Figure 6. The g values were extracted from the powder spectra by using the simulation program WinEPR Simfonia

Figure 6. Experimental (solid line) and simulated (dotted line) powder ESR spectra at room temperature.

v1.25 (Bruker), and the components of the g value, g∥ and g⊥, are the inputs for this program. The g average (gav) value is g

2g

calculated according to the gav = 3 + 3⊥ formula. The bestfitted g∥ and g⊥ values were obtained when the simulated

Table 3. ESR Data for Cu-P and Cu-NP in Powder (Room Temperature) and in Solution (at 110 K in THF)

Cu-P Cu-NP

powder solution powder solution

g∥

g⊥

gav

A∥(Cu) (G)

A⊥(Cu) (G)

A∥(14N) (G)

A⊥(14N) (G)

2.1600 2.1784 2.1500

2.0450 2.0525 2.0500

2.0830 2.0945 2.0830

211

16.5

16

16

E

DOI: 10.1021/acs.inorgchem.8b00528 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 7. Experimental (black) and simulated (red) ESR spectra at 110 K of Cu-P in a THF solution.

of the peaks in the simulated and experimental spectra do not perfectly match, but all of the simulated splittings are observed in the experimental spectrum. The g values are corrected according to the g value of the 2,2-diphenyl-1-picrylhydrazyl reference sample (g = 2.0036). The g value obtained from the simulation showed that Cu-P in solution has the same axial symmetry as that in powder. The axial g values are determined to be g∥ = 2.1784 and g⊥ = 2.0525. The hyperfine interaction due to copper also has axial symmetry, and the values are determined to be A∥(Cu) = 211 G and A⊥(Cu) = 16.5 G. On the other hand, the superhyperfine interaction due to the four nitrogen atoms is determined to be isotropic, and the value is A(Niso) = 16.0 G. The obtained values are listed in Table 3. Unfortunately, no resolved spectrum for Cu-NP could be obtained in the three solvents tested. Cu-NP was too aggregated at 2 × 10−3 M. Attempts to work at more dilute concentrations prevented the obtainment of a good signal: we tried as dilute as 7 μM but were not able to observe any signal. No resolved signals could be obtained even at a higher concentration (up to 20 mM). Therefore, the spin Hamiltonian parameters of the g value, copper hyperfine and nitrogen superhyperfine splittings for both Cu-P and Cu-NP complexes, could not be compared in solution. The g values are different in solution and in powder; this is due to the different respective arrangements of the molecules at the solid state and in solution. Molar Magnetic Susceptibility and Molar Inverse Susceptibility. The magnetic susceptibility of all complexes was obtained in the temperature range of 10−300 K. The temperature dependences of the molar magnetic susceptibility χm (the solid lines in the graph represent the best fits to the experimental data with these parameters) and of the molar inverse susceptibility 1/χm of all molecules for both the uncorrected and corrected diamagnetic or TIP contribution data are shown in Figure 8. The experimental values of the magnetic parameters are summarized in Table 4. The α, C, and θ values were determined by applying the C Curie−Weiss law ( χm (T ) = α + T − θ ), which takes into account intermolecular interactions, where C is the Curie constant and θ is the Weiss constant. Negative values of α correspond to a diamagnetic contribution, whereas positive values of α correspond to a TIP contribution of metal. TIP can originate from coupling of the ground and excited states and the effect of the orbital moments of the d electrons.26 The

Figure 8. Temperature dependences of (a) the molar magnetic susceptibility (the solid lines represent a fit by the Curie law), (b) the inverse susceptibility, and (c) the α-corrected inverse susceptibility: Cu-P (green); Cu-NP (brown).

Table 4. Experimental Values of the Magnetic Parameters α (cm3/mol) Cu-P Cu-NP

Curie−Weiss law Curie law Curie−Weiss law Curie law

−25 −23 8 25

× × × ×

−5

10 10−5 10−5 10−5

C (cm3·K/mol)

θ (K)

0.37598 0.37359 0.42000 0.39576

−0.071 −0.720

intercept with the T axis gives both the sign and value of θ. In this model, a positive θ value indicates ferromagnetic intermolecular interactions, whereas a negative θ indicates antiferromagnetic intermolecular interactions between neighboring molecules. The magnitude of the Weiss constant, in either a positive or a negative direction, is indicative of the strength of the intermolecular interactions between spins. By using the Curie−Weiss law to fit the experimental data of the susceptibility versus temperature graphs, θ was determined to be −0.071 K for Cu-P and −0.720 for Cu-NP. The small negative values of θ indicate that there are very weak antiferromagnetic intermolecular interactions for both complexes. However, the ESR and XRD data do not support these interactions and show that the use of the Curie−Weiss law is not adequate in our case and does not allow interpretation of the variations in the θ values. For this reason, the experimental data were fitted again using C the Curie law and applying the relationship χm (T ) = α + T . C and α were determined to be 0.37359 cm3·K/mol and −23 × 10−5 cm3/mol for Cu-P and 0.39576 cm3·K/mol and 25 × 10−5 cm3/mol for Cu-NP. The signs of the α values are opposite, indicating diamagnetic and TIP contributions of the copper atom to the magnetic susceptibility for both complexes. Cu-P has a negative α value, which means that the magnetic susceptibility has a diamagnetic contribution. All compounds show that some diamagnetism arises from the interaction of the magnetic field with the motion of the electrons in their orbitals. On the other hand, Cu-NP has a positive α value, which means F

DOI: 10.1021/acs.inorgchem.8b00528 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

quantum number. All data are summarized in Table 5. For each molecule, the spins are close to that for an S = 1/2 system. After the diamagnetic correction, the magnetic moment μeff of Cu-P remains constant with only a small scatter in the temperature range of 10−300 K, with a value at room temperature of 1.70 μB, with the average value (1.734 μB) in this studied temperature range being close to the spin-only value (1.73 μB) for the peripheral complex, showing that there is no orbital contribution to μeff. In contrast, the magnetic moment μeff of Cu-NP slightly increases from 1.77 to 1.83 μB in the range 300−270 K, decreases from 1.83 μB at 270 K to 1.75 μB at 62 K, then starts to increase, and reaches a maximum value of 1.85 μB at 44 K. Upon further cooling, μeff decreases again up to 1.77 μB at 10 K. This behavior suggests that there is an orbital contribution to the magnetic moment in the 62−10 K range. At room temperature, the value of the magnetic moment (1.77 μB) is above the spin-only value too, showing an orbital contribution to the magnetic moment in the 300−260 K range due to a mixing of some spin−orbital angular momentum from excited states via spin−orbit coupling.20 Different magnetic moment values support that the difference in the spacing and inclination of the molecules in both complex structures is the reason for these different amounts of angular momentum contribution to the magnetic moment. The Cu-NP complex’s room temperature magnetic moment value is larger than the value for Cu-P, demonstrating that the orbital contribution to the magnetic moment depends on the substituent position. According to Ray and Sen,29 four-coordinated copper complexes (such as in copper phthalocyanine) can be divided into two categories, one with magnetic moments in the range of 1.72−1.82 μB and the other one with moments in the range of 1.90−2.20 μB. In the first group, the copper has square-planar coordination, whereas it is either tetrahedral or planar in the second group. Our measured effective magnetic moment values are in the range of the first group for both Cu-P and Cu-NP. This indicates that both complexes have a square-planar coordination. This result is consistent with the determined g values. Spin densities calculated in the gas phase show that the substituent position has no significant effect and, hence, that the coordination is not significantly modified when the molecules are independent of each other (in the gas phase). In contrast, the spin densities (0.469 for Cu-P and 0.490 for Cu-NP) determined experimentally from the magnetic moment by applying Curie’s law gave evidence for a strong effect of the molecular packing because the values are different from those in the gas phase but closer to each other for each derivative. For both copper complexes, the spin density is not entirely localized on the copper atom, the non-peripheral derivative having a larger value. There is an important effect of the substituent position: Cu-NP occupies more space in the unit cell compared to Cu-P and is a larger ligand than Cu-P. One can suggest that this larger ligand induces a larger contribution to the total moment of the copper atom.30

that the magnetic susceptibility has a TIP contribution. A coupling between the ground and low-lying excited states through the Zeeman perturbation27 creates a positive TIP contribution to the magnetic susceptibility for Cu-NP. Thus, the position of the substituent has an important effect on these parameters. For both copper complexes, the temperature dependence of the inverse susceptibility has very different values above 100 K and, hence, slightly different slopes for the peripheral and nonperipheral structures. The plot of χm−1 = f(T) for both systems shows a clear nonlinearity, which is due to diamagnetic and TIP contributions to the magnetic susceptibility. The plot of the corrected χm−1 = f(T) for a system obeying the Curie law gives a straight line having a slope of 1/C.28 The intercept with the T axis gives both the sign and value of θ, which is negligible in our case. Magnetic Moments and Spin Densities. The temperature-dependent effective magnetic moments μeff for both uncorrected and corrected diamagnetic (or TIP) contribution data are shown in Figure 9 for all molecules. It has been

Figure 9. Temperature dependences of (a) the effective magnetic moment and (b) the α-corrected effective magnetic moment: Cu-P (green); Cu-NP (brown).

calculated according to the relationship μeff = 2.83(χT)1/2 and is expressed in Bohr magneton (μB). The experimental data include both the α-uncorrected and -corrected contributions. Assuming that the magnetic moment is due to the spin moment only, the spin of the copper atom for all Pcs was determined using eq 1: χ=

Nge 2μβ 2 3kBT

S(S + 1) =

C T

(1)

where N is the total number of paramagnetic molecules, ge is the electron’s g factor, μB is the Bohr magneton, kB is the Boltzmann constant, and S is the spin angular momentum Table 5. Magnetic Moments and Spin Densities

Cu-P Cu-NP

coordination number of the central atom

no. of unpaired electrons

calculated (gas phase) μeff (μβ)

exptl μeff (μβ)

calcd spin densities (gas phase)

exptl spin densities

4 4

1 1

1.32 1.31

1.70 1.77

0.659 0.652

0.469 0.490

G

DOI: 10.1021/acs.inorgchem.8b00528 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

3. CONCLUSIONS Ground-state electronic absorption spectra of the two studied Pcs were recorded in THF and chloroform and evidence the dramatic effect of the substituent position in both solvents, on the maximum absorption wavelengths, and on the aggregation. The electrochemical behaviors of copper phthalocyanines were investigated by cyclic voltammetry. Cu-P and Cu-NP display three successive reductions and two oxidations. The effect of the substitution patterns on the redox potentials has been analyzed comparatively in both pyridine and CH2Cl2. The reduction potentials of Cu-P in pyridine and CH2Cl2 are almost identical, with the exception that the first reduction in CH2Cl2 is 0.10 V lower compared with that in pyridine. Cu-NP displays similar redox behaviors, with the first and second reductions being at almost that same potentials as the first two reductions of Cu-P, respectively. However, the third reduction of Cu-NP was observed at −1.39 V in pyridine, while no redox wave in CH2Cl2 was within the potential range. It was found that the HOMO−LUMO energy gap varies in the range from 1.33 to 1.57 V and the non-peripheral derivative has a smaller gap value in comparison to that of the peripheral derivative, which is in accordance with the longer-wavelength absorption of Cu-NP. The DFT calculations indicate that the HOMOs of Cu-NP and Cu-P are distributed on the isoindole moieties of the macrocycles coupled with the contribution from the sulfur atoms and the LUMOs of both studied molecules are localized on two opposite isoindole units and nitrogen bridges. The calculated energy gap of Cu-NP is smaller than that of Cu-P, which is in accordance with the electrochemical measurements. The XRD measurements have been performed to analyze the solid-state molecular arrangement of the compounds studied. It was found that the crystal packing of Cu-NP has a columnar structure and the central copper atom has a four-coordinate, square-planar geometry, without any contribution of the substituents to its coordination sphere. However, the PXRD spectra revealed different crystal packings of these compounds. The g values are extracted from the powder ESR spectra (simulated and experimental) of Cu-NP and Cu-P. For both molecules, the g∥ > g⊥ order indicates a planar anisotropy, where copper is planar with the four inner nitrogen atoms of the Pc. The unpaired electron remains localized in dx2−y2, which spreads into the molecular plane with a σ characteristic, which prevents overlap of the wave function with neighboring Pc atoms. The nitrogen atoms of neighboring Pcs are too far to establish extra axial coordination. The magnetic susceptibility of the complexes studied was obtained in the temperature range of 10−300 K. For Cu-NP and Cu-P, the temperature dependence of the inverse susceptibility has very different values above 100 K and, hence, slightly different slope values for peripheral and non-peripheral structures. The measured effective magnetic moment values for the Cu-P and Cu-NP complexes indicate that both complexes have square-planar coordination. This result is consistent with the determined g values. Spin densities calculated in the gas phase showed that the substituent position has no significant effect and, hence, that the coordination is not significantly modified when the molecules are independent from each other (in the gas phase). On the contrary, experimental spin densities (0.469 for Cu-P and 0.490 for Cu-NP) determined from the magnetic moment by applying Curie’s law evidenced a strong effect of the molecular packing because the values are different from those in the gas phase but closer to each other for each derivative. For both copper

complexes, the spin density is not entirely localized on the copper atom, with the non-peripheral derivative having a larger value. There is an important effect of the substituent position: Cu-NP occupies more space in the unit cell compared to Cu-P and is a larger ligand than Cu-P. One can suggest that this larger ligand induces a larger contribution to the total moment of the copper atom. Future studies should deal with similar studies on Pcs with different metalation or different substituents.

4. EXPERIMENTAL SECTION Synthesis. All Pcs were prepared following methods reported in the literature.9,16,27 Their structures were confirmed by their Fourier transform infrared (FT-IR) spectra (Figure S6) recorded on a PerkinElmer Spectrum 100 FT-IR spectrophotometer, by their matrix-assisted laser desorption ionization (MALDI) spectra recorded on a MALDI Bruker Microflex LT (Figure S7), by their high resolution mass spectrometry (HRMS) spectra (Figures S8 and S9; experimental details and the mass accuracy are also described in the Supporting Information), and by their elemental analyses (CHNS; see the Supporting Information). Computational Methods. The electronic structure calculations for the compounds studied were performed at the unrestricted B3LYP/6-311G level of theory by DFT of the Gaussian09 suite of quantum-chemical codes.31 The models were constructed for both non-peripheral and peripheral octasubstituted copper phthalocyanines, and their geometry optimizations were performed (Figure S2). The electronic structures for the spin-up and spin-down electrons were calculated for the models and X-ray structure of Cu-NP. ESR and VSM. The 10−300 K magnetization measurements were obtained from powder samples in a Quantum Design Physical Properties Measurement System with VSM. χ−T graphs were recorded under a constant magnetic field of 1 kOe. ESR data were collected from the powder samples at room temperature. All spectra were obtained from a Bruker EMX spectrometer operating at the X band (ca. 9.8 GHz) with suitable microwave power (15−20 mW) and 100 kHz magnetic field modulation. ESR spectra for solutions at low temperature (110 K) were obtained from a JEOL spectrometer operating at the X band (ca. 9.06 GHz) with suitable microwave power of around 9 mW. Electrochemistry. Pyridine and CH2Cl2 used for the electrochemical measurements were purchased from Sigma-Aldrich Chemical Co. Tetra-n-butylammonium perchlorate (TBAP), used as the supporting electrolyte, was purchased from Sigma-Aldrich, recrystallized from ethyl alcohol, and dried under vacuum at 40 °C for at least 1 week prior to use. Cyclic voltammetry was carried out at 298 K by using an EG&G Princeton Applied Research (PAR) 173 potentiostat/galvanostat. A homemade three-electrode electrochemistry cell was used for cyclic voltammetric measurements and consisted of a glassy carbon working electrode, a platinum wire counter electrode, and a saturated calomel reference electrode (SCE). The SCE was separated from the bulk of the solution by a fritted-glass bridge of low porosity, which contained the solvent/supporting electrolyte mixture. All potentials are referenced to the SCE. X-ray Crystallography. The solid-state structure of Cu-NP was confirmed by X-ray crystallography. The data were obtained on a Bruker APEX II QUAZAR three-circle diffractometer using monochromatized Mo Kα radiation (λ = 0.71073 Å). Indexing was performed using APEX2.32 Data integration and reduction were carried out with SAINT.33 Absorption correction was performed by a multiscan method implemented in SADABS.34 The structures were solved and refined using the Bruker SHELXTL software package.35 All non-hydrogen atoms were refined anisotropically using all reflections with I > 2σ(I). The carbon-bound hydrogen atoms were positioned geometrically and refined using a riding mode. The crystallographic data and refinement details of the data collection for Cu-NP are given in Table S2. The selected bond lengths and angles are given in Table H

DOI: 10.1021/acs.inorgchem.8b00528 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry S3. Crystallographic data for the structure described in this study have been deposited at the Cambridge Crystallographic Data Centre as CCDC 1470853.



5134. (c) Zorlu, Y.; Kumru, U.; Isci, U.; Divrik, B.; Jeanneau, E.; Albrieux, F.; Dede, Y.; Ahsen, V.; Dumoulin, F. 1,4,8,11,15,18,22,25Alkylsulfanyl phthalocyanines: effect of macrocycle distortion on spectroscopic and packing properties. Chem. Commun. 2015, 51, 6580−6583. (d) Cidlina, A.; Svec, J.; Ludvová, L.; Kuneš, J.; Zimcik, P.; Novakova, V. Predominant effect of connecting atom and position of substituents on azomethine nitrogens’ basicity in phthalocyanines. J. Porphyrins Phthalocyanines 2016, 20, 1122−1133. (e) Lin, M.-J.; Wang, J.-D. Effect of non-peripheral alkyloxy substituents on the structure and spectroscopic properties of metal-free phthalocyanines. J. Mol. Struct. 2007, 837, 284−289. (3) (a) Robertson, J. M. An X-ray study of the phthalocyanines. Part II. Quantitative structure determination of the metal-free compound. J. Chem. Soc. 1936, 1195. (b) Linstead, R. P.; Robertson, J. M. The stereochemistry of metallic phthalocyanines. J. Chem. Soc. 1936, 1736− 1738. (c) Ingram, D. J. E.; Bennett, J. E. Paramagnetic resonance in phthalocyanine, haemoglobin, and other organic derivatives. Discuss. Faraday Soc. 1955, 19, 140−146. (4) Makarov, S. G.; Suvorova, O. N.; Schnurpfeil, G.; Wöhrle, D. Synthesis of a Dinuclear Naphthalocyanine Analogue with Absorption in the NIR. Eur. J. Inorg. Chem. 2010, 4617−4621. (5) Quartarolo, A. D.; Pérusse, D.; Dumoulin, F.; Russo, N.; Sicilia, E. Hydrophilic annulated dinuclear zinc(II) phthalocyanineas Type II photosensitizers for PDT: a combined experimental and (TD)-DFT investigation. J. Porphyrins Phthalocyanines 2013, 17, 980−988. (6) Furuyama, T.; Satoh, K.; Kushiya, T.; Kobayashi, N. Design, Synthesis, and Properties of Phthalocyanine Complexes with MainGroup Elements Showing Main Absorption and Fluorescence beyond 1000 nm. J. Am. Chem. Soc. 2014, 136, 765−776. (7) Gürek, A. G.; Ahsen, V.; Luneau, D.; Pecaut, J. Synthesis, Structure, Spectroscopic Properties, and Magnetic Properties of an Octakis(Alkylthio)-Substituted Lutetium(III) Bisphthalocyanine. Inorg. Chem. 2001, 40, 4793−4797. (8) (a) Achar, B. N.; Lokesh, K. S. Studies on tetra-amine phthalocyanines. J. Organomet. Chem. 2004, 689, 3357−3361. (b) Seelan, S.; Agashe, M. S.; Srinivas, D.; Sivasanker, S. Effect of peripheral substitution on spectral and catalytic properties of copper phthalocyanine complexes. J. Mol. Catal. A: Chem. 2001, 168, 61−68. (c) Wu, W.; Fisher, A. J.; Harrison, N. M.; Wang, H.; Wu, Z.; Gardener, J.; Heutz, S.; Jones, T.; Aeppli, G. Theoretical modeling of the electronic structure and exchange interactions in Cu(II)Pc. J. Phys.: Conf. Ser. 2012, 391, 012119−012127. (d) Kroll, T.; Aristov, V. Yu.; Molodtsova, O. V.; Ossipyan, Yu. A.; Vyalikh, D. V.; Büchner, B.; Knupfer, M. Spin and Orbital Ground State of Co in Cobalt Phthalocyanine. J. Phys. Chem. A 2009, 113, 8917−8922. (9) (a) Burnham, P. M.; Cook, M. J.; Gerrard, L. A.; Heeney, M. J.; Hughes, D. L. Structural Characterisation of a Red Phthalocyanine. Chem. Commun. 2003, 16, 2064−2065. (b) Burnham, M.; Chambrier, M. I.; Hughes, D. L.; Isare, B.; Poynter, R. J.; Powell, A. K.; Cook, M. J. Synthesis, lead lability and ring deformation of three 1,4,8,11,15,18,22,25-octasubstituted phthalocyaninato lead derivatives. J. Porphyrins Phthalocyanines 2006, 10, 1202−1211. (10) Basova, T.; Kol’tsov, E.; Gürek, A. G.; Atilla, D.; Ahsen, V.; Hassan, A. K. Investigation of liquid-crystalline behaviour of copper octakisalkylthiophthalocyanine and its film properties. Mater. Sci. Eng., C 2008, 28, 303−308. (11) Bondi, A. van der Waals Volumes and Radii. J. Phys. Chem. 1964, 68, 441−451. (12) Brown, C. Crystal Structure of β-Copper Phthalocyanine. J. Chem. Soc. A 1968, 0, 2488−2493. (13) Figgis, B. N.; Kucharski, E. S.; Reynolds, P. A. Electronic Structure of Cobalt Phthalocyanine: A Charge Density Study. J. Am. Chem. Soc. 1989, 111, 1683−1692. (14) Gürek, A. G.; Bekâroğlu, Ö . Octakis(alkylthio)-substituted Phthalocyanines and Their Interactions with Silver(I) and Palladium(II) ions. J. Chem. Soc., Dalton Trans. 1994, 1419−1423. (15) Clack, H. W.; Hush, N. S.; Woolsey, I. S. Reduction Potentials of Some Metal Phthalocyanines. Inorg. Chim. Acta 1976, 19, 129−132.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.8b00528. UV−vis, FT-IR, MALDI-MS, and MALDI-TOF HRMS spectra, PXRD profiles, optimized geometry, 3D plots, HRMS experimental details and mass accuracy, experimental details and results of the elemental analyses, scan of the report of the elemental analysis results, crystal data and refinement parameters, selected bond lengths and angles, angles between the normals, crystal structures, and total spin density plots (PDF) Accession Codes

CCDC 1470853 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing data_ [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.



AUTHOR INFORMATION

Corresponding Authors

*E-mail: *E-mail: *E-mail: *E-mail:

[email protected] (F.D.). [email protected] (K.M.K.). [email protected] (A.G.G.). [email protected] (S.T.Ö .).

ORCID

Iṡ mail Fidan: 0000-0002-3649-0764 Catherine Hirel: 0000-0002-4500-7594 Karl M. Kadish: 0000-0003-4586-6732 Ayşe Gül Gürek: 0000-0002-8565-2424 Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS Support of the Robert A. Welch Foundation (Grant E-680 to K.M.K.) is gratefully acknowledged. REFERENCES

(1) (a) Bouvet, M. In The Porphyrin Handbook; Academic Press: New York, 2003. (b) Leznoff, C. C.; Lever, A. B. P. Phthalocyanines: properties and applications; VCH: Weinheim, Germany, 1989−1996; Vols. 1−4. (c) Bilgin, A.; Ertem, B.; Gök, Y. Highly Organosoluble Metal-Free Phthalocyanines and Metallophthalocyanines: Synthesis and Characterization. Eur. J. Inorg. Chem. 2007, 12, 1703−1712. (d) Bottari, G.; de la Escosura, A.; González-Rodríguez, D.; de la Torre, G. Tomás Torres’ research in a nutshell. J. Porphyrins Phthalocyanines 2016, 20, 966−986. (e) Photosensitizers in Medicine, Environment, and Security; Nyokong, T., Ahsen, V., Eds.; Springer, 2012. (2) (a) Topal, S. Z.; Isci, U.; Kumru, U.; Atilla, D.; Gürek, A. G.; Hirel, C.; Durmus, M.; Tommasino, J.-B.; Luneau, D.; Berber, S.; Dumoulin, F.; Ahsen, V. Modulation of the electronic and spectroscopic properties of Zn(II) phthalocyanines by their substitution pattern. Dalton Trans. 2014, 43, 6897−6908. (b) Kobayashi, N.; Ogata, H.; Nonaka, N.; Luk’yanets, E. A. Effect of Peripheral Substitution on the Electronic Absorption and Fluorescence Spectra of Metal-Free and Zinc Phthalocyanines. Chem. - Eur. J. 2003, 9, 5123− I

DOI: 10.1021/acs.inorgchem.8b00528 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry (16) Ö zkaya, A. R.; Gürek, A. G.; Gül, A.; Bekâroğlu, Ö . Electrochemical and Spectral Properties of octakis(hexylthio)-substituted phthalocyanines. Polyhedron 1997, 16, 1877−1883. (17) (a) Mbambisa, G.; Tau, P.; Antunes, E.; Nyokong, N. Synthesis and electrochemical properties of purple manganese(III) and red titanium(IV) phthalocyanine complexes octa-substituted at nonperipheral positions with pentylthio groups. Polyhedron 2007, 26, 5355−5364. (18) Ou, Z.; Jiang, Z.; Chen, N.; Huang, J.; Shen, J.; Kadish, K. M. Electrochemistry and spectroelectrochemistry of tetra-α-substituted metallophthalocyanines. J. Porphyrins Phthalocyanines 2008, 12, 1123− 1133. (19) L’Her, M.; Pondaven, A. Electrochemistry of Phthalocyanines: Phthalocyanines: Spectroscopic and Electrochemical Characterization. In The Porphyrin Handbook; Kadish, K. M., Smith, K. M., Guilard, R., Eds.; Elsevier Science, New York, 2003; Vol. 16, Chapter 104. (20) Nemykin, V. N.; Hadt, R. G.; Belosludov, R. V.; Mizuseki, H.; Kawazoe, Y. Influence of Molecular Geometry, Exchange-Correlation Functional, and Solvent Effects in the Modeling of Vertical Excitation Energies in Phthalocyanines Using Time-Dependent Density Functional Theory (TDDFT) and Polarized Continuum Model TDDFT Methods: Can Modern Computational Chemistry Methods Explain Experimental Controversies? J. Phys. Chem. A 2007, 111, 12901− 12913. (21) Bennett, J. E.; Ingram, D. J. E. Paramagnetic Resonance in Copper Phthalocyanine. Nature 1955, 175, 130−131. (22) Abragam, A.; Bleaney, B. EPR of Transition Ions; Dover Publications: New York, 1986; p 157. (23) Di Mauro, E.; Mara Domiciano, S. Covalent bonding in NH4Br:Cu2+. J. Phys. Chem. Solids 1999, 60, 1849. (24) Mitra, S.; Gregson, A. K.; Hatfield, W. E.; Weller, R. R. Singlecrystal magnetic study on ferromagnetic manganese(II) phthalocyaninate. Inorg. Chem. 1983, 22, 1729−1732. (25) Janczak, J.; Kubiak, R.; Sledz, M.; Borrmann, H.; Grin, Y. Synthesis, structural investigations and magnetic properties of dipyridinated manganese phthalocyanine, MnPc(py)2. Polyhedron 2003, 22, 2689−2697. (26) Earney, J. J.; Finn, C. P. B.; Najafabadi, B. M. Van Vleck temperature-independent paramagnetism in some rare earth double nitrates. J. Phys. C: Solid State Phys. 1971, 4, 1013−1021. (27) Sakamoto, K.; Ohno-Okumura, E.; Kato, T.; Soga, H. Synthesis of near-infrared absorbed metal phthalocyanine with S-aryl groups at non-peripheral positions. J. Porphyrins Phthalocyanines 2010, 14, 47− 54. (28) Kahn, O. Molecular Magnetism, 1st ed.; Wiley-VCH, 1993. (29) Ray, P.; Sen, D. N. Magneto-Chemical Studies in Valency. 6. Bond-Type and Stereochemistry of 4-Coordinated Copper Complexes. J. Indian Chem. Soc. 1948, 25, 473. (30) Wisbey, D.; Feng, D.; Bremer, M. T.; Borca, C. N.; Caruso, A. N.; Silvernail, C. M.; Belot, J.; Vescovo, E.; Ranno, L.; Dowben, P. A. Electronic Structure of a Metal−Organic Copper Spin-1/2 Molecule: Bis(4-cyano-2,2,6,6-tetramethyl-3,5-heptanedionato)copper(II). J. Am. Chem. Soc. 2007, 129, 6249−6254. (31) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson, G. A.; Nakatsuji, H.; Li, X.; Caricato, M.; Marenich, A.; Bloino, J.; Janesko, B. G.; Gomperts, R.; Mennucci, B.; Hratchian, H. P.; Ortiz, J. V.; Izmaylov, A. F.; Sonnenberg, J. L.; Williams-Young, D.; Ding, F.; Lipparini, F.; Egidi, F.; Goings, J.; Peng, B.; Petrone, A.; Henderson, T.; Ranasinghe, D.; Zakrzewski, V. G.; Gao, J.; Rega, N.; Zheng, G.; Liang, W.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Throssell, K.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Millam, J. M.; Klene, M.; Adamo, C.; Cammi, R.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Farkas, O.; Foresman, J. B.; Fox, D. J. Gaussian 09, revision A.02; Gaussian, Inc.: Wallingford, CT, 2016.

(32) (33) (34) (35)

J

APEX2; Bruker AXS Inc.: Madison, WI, 2007. SAINT; Bruker AXS Inc.: Madison, WI, 2007. SADABS; Bruker AXS Inc.: Madison, WI, 2001. SHELXTL; Bruker AXS Inc.: Madison, WI, 2003.

DOI: 10.1021/acs.inorgchem.8b00528 Inorg. Chem. XXXX, XXX, XXX−XXX