Electronic Coupling in [Mo2]–Bridge–[Mo2] Systems with Twisted

Nov 18, 2015 - Electronic Coupling in [Mo2]–Bridge–[Mo2] Systems with Twisted ..... Discover the Most-Read Physical Chemistry Articles of February...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/IC

Electronic Coupling in [Mo2]−Bridge−[Mo2] Systems with Twisted Bridges Hong Li Zhang, Guang Yuan Zhu, Gangyi Wang, Miao Meng, Chun Y. Liu,* and Hao Lei* Department of Chemistry, Jinan University, 601 Huang-Pu Avenue West, Guangzhou 510632, China S Supporting Information *

ABSTRACT: In order to evaluate the impact of bridge conformation on electronic coupling in donor−bridge−acceptor triad systems, two Mo2 dimers, [Mo2(DAniF)3]2[μ-1,4-{C(O)NH}2-Naph] (1, DAniF = N,N′-di(p-anisyl)formamidinate and Naph = naphthalenyl) and [Mo2(DAniF)3]2[μ-1,4-(CS2)2-2,5Me2C6H2] (2), have been synthesized and structurally characterized. These two compounds feature a large dihedral angle (>60°) between the central aromatic ring and the plane defined by the Mo−Mo bond vectors, which is distinct from the previously reported phenylene bridged analogues [Mo2(DAniF)3]2[μ-1,4{C(O)NH}2-ph] (I) and [Mo2(DAniF)3]2[μ-1,4-(CS2)2-C6H4] (II), respectively. Unusual optical behaviors are observed for the mixed-valence (MV) species (1+ and 2+), generated by single-electron oxidation. While 2+ exhibits a weak intervalence charge transfer (IVCT) absorption band in the near-IR region, the IVCT band is absent in the spectrum of 1+, which is quite different from what observed for I+ and II+. Optical analyses, based on superexchange formalism and Hush model, indicate that, in terms of Robin−Day classification, mixed-valence species 1+ belongs to the electronically uncoupled class I and complex 2+, with Hab = 220 cm−1, is assigned to the weakly coupled class II. Together with I+ and II+, the four MV complexes complete a transition from class I to class II−III borderline as a result of manipulating the geometric topology of the bridge. Given the structural and electronic features for the molecular systems, the impacts of electrostatic interaction (through-space) and electron resonance (through-bond) on electronic coupling are discussed.

■

INTRODUCTION Electronic coupling (EC) and electron transfer (ET) are among the most fundamental phenomena in chemical and biological reactions, and are involved in many naturally occurring or artificial processes.1,2 Understanding of such phenomena has highly relied on the rational design and construction of smallmolecule model systems, which contain three basic components, i.e., electron donor (D), electron acceptor (A), and bridge (B), namely, D−B−A triad. Starting with the Creutz− Taube complex,3 [(NH3)5Ru(pyrazine)Ru(NH3)5]5+, numerous D−B−A model compounds have been developed by synthetic chemists, for which a diversity of D, B, and A components have been unitized.4,5 The three units in a D−B− A assembly function in different ways to affect the donor− acceptor coupling and electron transfer. The D (and A) site controls the ET reaction by their electron donating (or accepting) ability. These two units, either metal complexes4,6,7 or purely organic groups,8 are redox active in general; thus, the electron donating ability for D and electron accepting ability for A can be measured using electrochemical methods. The central unit or bridge mediates the donor−acceptor interaction through its electron (or charge) transporting ability, which is determined by the geometric and electronic natures of this moiety. For a D−B−A system with given D and A sites, the bridge (B) plays a critical role in tuning the extent of electronic © XXXX American Chemical Society

coupling effect and controlling the electron transfer rate. Therefore, much work has focused on the impact of various factors involving the bridge, such as length9,10 conformation,11,12 and conjugation.13,14 Recently, a series of D−B−A complex models consisting of quadruply bonded dimolybdenum complex units (Scheme 1) have been developed and studied in terms of electronic Scheme 1. Molecular Skeleton of [Mo2]−Bridge−[Mo2] Model Compounds

Received: August 20, 2015

A

DOI: 10.1021/acs.inorgchem.5b01923 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry coupling and electron transfer.15,16 In these systems, owing to the unique electronic configuration (σ2π4δ2) for the Mo24+ units, the transferring electron(s) is identified to be the δ electron(s), which greatly facilitates the spectroscopic and electrochemical analyses for the mixed-valent species. Previous works from different groups have demonstrated that sulfurcontaining head groups of the bridging ligand are capable of promoting stronger electronic communication compared to those with nitrogen and oxygen atoms.17−19 In addition, it is found that extension of the bridge by adding another phenylene group results in significant weakening of the electronic interaction between the two [Mo2] sites.20 In our studies, the geometric length of the spacer in these models, rather than the distance between the two Mo2 centers, is proposed to be the effective electron transfer distance (r′ab), which is critical for accurate determination of the electronic coupling matrix element Hab.16,17 These results reveal that the d(δ)−p(π) orbital interactions between the dimetal units and bridging ligand play a key role in promoting strong metal−metal electronic coupling and accelerating electron migration via the charge transfer platform.16 It is generally recognized that increasing the conjugated π system of the bridge would enhance the electronic coupling. Polycyclic aromatic hydrocarbons (PAHs) are classical conjugated systems and have been widely applied as part of the bridging ligand in various organometallic and organic D− B−A systems. For instance, bimetallic complexes with naphthalene- or anthracene-based bridging ligand have been reported and their ET properties investigated.21,22 One of the compounds that contain both quadruply bonded dimetal moieties and PAH bridge backbone is [Mo2(DAniF)3]2(C14H8N4) (DAniF = N,N′-di(p-anisyl)formamidinate), in which the bridging ligand tetraazatetracene dianion (C14H8N42−) is a large aromatic π system with four fused six-membered rings coplanar with the Mo−Mo bond vectors.23 In this mixed-valence complex, the odd electron is fully delocalized over the two Mo2 units as shown by the intense intervalence absorption band. However, the overall coupling effect resulting from the interplay of various factors needs detailed elucidation through structurally and electronically well-defined molecular systems. In light of this, we decided to investigate the EC and ET properties of [Mo2]−bridge−[Mo2] complexes where the bridging ligands vary in conjugation, conformation, and chelating atoms. Herein, we report the study of two newly synthesized and structurally characterized molybdenum dimers of dimers [Mo2(DAniF)3]2[μ-1,4-{C(O)NH}2-Naph] (1, Naph = naphthalenyl) and [Mo2(DAniF)3]2[μ-1,4-(CS2)2-2,5-Me2C6H2] (2) (Scheme 2). These two compounds have similar D−

A distances but differ in bridging donor atoms (N, O, and S) and the aromatic spacer that connects the two [Mo2] units. More importantly, both of them show a very large dihedral angle (>60°) between the central aromatic ring and the plane defined by the Mo−Mo bond vectors. Furthermore, compounds 1 and 2 are related to two previously studied molecules [Mo 2 (DAniF) 3 ] 2 [μ-1,4-{C(O)NH} 2 -ph] (I) 1 9 and [Mo2(DAniF)3]2[μ-1,4-(CS2)2-C6H4] (II),16 respectively, by having nearly equal bridge length and identical [Mo2] units as the electron donor and acceptor. Therefore, comparison of the two species in the pairs (1 and I, 2 and II) allows direct evaluation of the interplaying impacts of these factors on electronic coupling. It is found that increasing the torsion angle of the bridge interrupts the d(δ)−p(π) conjugation, consequently, lowering the donor−acceptor interaction while the charge transfer distance remains unchanged.

■

RESULTS AND DISCUSSION Syntheses and Molecular Structures. The synthetic routes for compounds 1 and 2 are outlined in Scheme 3. For preparation of 1, naphthalenedicarboxylic acid was selected as the starting material to react with thionyl chloride and aqueous ammonia successively, affording the expected naphthalenedicarboxamide in high yields. The bis(amide) was then reacted with two equivalents of dimolybdenum precursor Mo2(DAniF)3(O2CCH3) in the presence of NaOC2H5. By following this known synthetic protocol,11,15 crystalline products of 1 were obtained in 73% yields after standard workup and crystallization procedures. The composition of 1 is established by its 1H NMR spectrum in deuterated chloroform. In the spectrum, three sets of resonances for the aromatic protons on the central naphthalene ring are observed at 7.29, 7.82, and 8.55 ppm; singlet resonance for the amide protons is detected in the downfield region (8.80 ppm) and integrated into two protons, which is comparable to that of the terephthalamidate analogue [Mo2(DAniF)3]2[μ-1,4-{C(O)NH}2-C6H4] (I, 9.01 ppm) reported earlier.19 The dimethyl substituted bridging ligand for compound 2 was synthesized starting with p-xylene, to which two bromomethyl groups were introduced by treating the reaction with paraformaldehyde and 31% HBr solution in acetic acid. Subsequently the benzyl bromide was converted to tetrathioterephthalate using elemental sulfur and sodium ethoxide. The product was isolated in the form of piperidinium salt and reacted with Mo2(DAniF)3(O2CCH3) to afford complex 2 (Scheme 3). After standard workup, the greenish blue compound was isolated as needle crystals in 60% yield. The composition and structure in solution are determined by 1H NMR spectra in deuterated chloroform. The two methyl groups on the bridging ring give a singlet resonance at 2.60 ppm, and all the other resonances are observed in the expected region and integrated into the corresponding number of protons. Single crystals of 1 and 2 suitable for X-ray crystallography analysis were obtained by slow diffusion of ethanol into a dichloromethane solution of the corresponding compound. Both complexes crystallized in the triclinic space group P1̅ with Z = 1. The crystal structures of 1 and 2 are depicted in Figure 1, and the crystallographic data and collection parameters are presented in Tables S1 and S2. Generally speaking, both compounds adopt the same molecular skeleton and show similar bond lengths in comparison to those for I and II (Table S2).16,19 When I and II are taken as the reference compounds

Scheme 2. Two New [Mo2]−Bridge−[Mo2] Complexes (Left, 1 and 2) and Related Phenylene Spaced Analogues (Right, I and II), [Mo2] = [Mo2(DAniF)3]+

B

DOI: 10.1021/acs.inorgchem.5b01923 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Scheme 3. Synthetic Routes for Complexes 1 and 2a

(i) SOCl2/DMF, reflux; (ii) NH3 (aq)/THF, 0 °C; (iii) Mo2(DAniF)3(O2CCH3), NaOC2H5, ethanol/THF; (iv) (CH2O)n, 31% HBr in CH3COOH, reflux; (v) (1) S8, NaOC2H5/C2H5OH, reflux; (2) HCl, piperidine.

a

between the amide hydrogen atom and the hydrogen atoms on naphthalene ring. Steric hindrance between the coordinated amidate group and the spacer on the bridging ligand was also seen in other N-substituted terephthaloyldiamidate bridged analogues,24,25 which showed a torsion angle of 34−50°. By introducing two methyl groups at the 2- and 5-positions of the central phenylene ring, a very large torsion angle (62°) between the central phenyl ring and the Mo2 chelating rings is observed in the structure of 2. The structural parameters are compared with those for II as presented in Table S2. Both complexes show very similar bonding distances and angles, except for the torsion angle values. For example, the Mo2···Mo2 distance in 2 (12.19 Å) is essentially equal to that in II (12.24 Å), in spite of the presence of the two methyl groups. The geometric relevancies between 2 and II are the same as those between 1 and I. Therefore, the molecular structures set up that the two species in each pair have similar distance-dependent electrostatic interaction. Expectedly, in 1 and 2, the d(δ)−p(π) conjugation along the charge transfer platform is interrupted, to some extent, by the torsional orientation of the aromatic spacer, which would significantly weaken the metal−metal interaction. Electrochemical Studies. Electrochemical cyclic voltammograms (CVs) and differential pulse voltammograms (DPVs) for 1 and 2 were measured in CH2Cl2 for general evaluation of the electronic coupling effects between the two Mo2 redox sites. As shown in Figure 2, the two redox events corresponding to the successive removal of one δ electron from each [Mo2] unit are not resolved for compound 1, indicating a very weak electronic interaction between the two dimetal centers. In comparison, two separated redox waves are detected in the voltammograms for 2, implying appreciable electronic

Figure 1. X-ray crystal structures for 1 (top) and 2 (bottom) with thermal ellipsoids shown at the 30% probability level. Hydrogen atoms are omitted for clarity, except for those on the N atoms of bridging ligand in 1.

for 1 and 2, respectively, a topologic difference in molecular geometry between the two species of each pair is that, in 1 and 2, the central bridging aromatic rings largely deviate from the plane defined by the two Mo−Mo bond vectors. In the solid-state structure of 1, the two nitrogen and two oxygen coordinating atoms are arranged in a trans-fashion across the naphthalene spacer. The Mo−Mo bond distances in 1 are 2.0916(8) Å, similar to those of I (2.0892(5) Å). The Mo2···Mo2 separations in 1 (11.35 Å) and I (11.36 Å) are essentially the same, implying similar electrostatic (through space) interactions between the two Mo2 redox sites for them. The Mo(1)−N(7) and Mo(2)−O(7) bond distances in 1 are 2.126(5) and 2.161(5) Å, respectively, which are comparable to the corresponding values in I19 and the N-alkyl24 or N-aryl25 substituted analogues. As mentioned above, the striking feature for the crystal structure of 1 is the great deviation of the bridging naphthalene group from the plane defined by the two Mo2 vectors. A torsion angle of 65° is observed in the structure of 1, in sharp contrast to the “near co-planarity” reported for I (Table S2).19 Such structural difference is likely due to the bulky naphthalene ring, more specifically, the steric repulsion

Figure 2. Differential pulse voltammograms (DPVs, top) and cyclic voltammograms (CVs, bottom) for complexes 1 (red) and 2 (blue). C

DOI: 10.1021/acs.inorgchem.5b01923 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

and II, about half of the contribution, as measured by ΔE1/2, is made by electronic resonant effect. However, this portion of interaction is mostly eliminated in 1 and 2 by the enlarged torsion angles. This comparison also indicates that the ligating atom effects (S > O > N)16,19 on electronic coupling in [Mo2]− bridge−[Mo2] are based upon efficient d(δ)−p(π) orbital interaction. Significantly, these results show that, by manipulating the geometric topology of the molecules, one can evaluate the two major effects of electronic coupling separately. Electronic Structures and Spectroscopic Properties for 1 and 2. Both compounds 1 and 2 display an intense absorption band in the electronic spectra (Figure 3).

communication. The potential separations (ΔE1/2) between the two redox couples can be obtained from the DPV data by Richardson and Taube’s methods.26 Given the ΔE1/2 value, the constant (Kc) and free energy change (ΔGc) for the comproportionation equilibrium in solution can be derived from Kc = exp(ΔE1/2/25.69) (at 298 K) and ΔGc = −RT ln(Kc), which measures the thermodynamic stability of the mixed-valent species in solution. Compared to the related complexes (I and II), both complexes show much smaller ΔE1/2 values as listed in Table 1. More specifically, the potential Table 1. Electrochemical Measurements and Parameters for the Comproportionation Equilibriuma compd

E1/2(1) (mV)

1 2 Ib IIc

/ 633 327 502

E1/2(2) (mV) ΔE1/2 (mV) / 722 411 697

68 101 96 195

Kc

ΔGc (cm−1)

14 51 42 1980

−549 −817 −774 −1572

a

Parameters for 1 and 2 were measured on the CH2Cl2 solution with supporting electrolyte nBu4NPF6 (0.10 M), Pt working and auxiliary electrodes, and a Ag/AgCl reference electrode. bData cited from ref 19. cData cited from ref 16.

separation for compound 1 (68 mV) is smaller than those of complexes I (96 mV) and the N,N′-diarylterephthalamidate (100 mV)25 and N,N′-diethylterephthalamidate analogues (85 mV).24 These complexes share a common Mo2 coordination shell and Mo2···Mo2 distance, but they differ in the twisting degree of the spacer, ca. 50° and 34° for the N-diethyl24 and Ndiaryl derivatives,25 respectively. Clearly, the ΔE1/2 values vary as a function of the torsion angle. Therefore, the large torsion angle (65°) is likely responsible for the reduced electronic communication. It is interesting to note that ΔE1/2 of 68 mV for 1 is essentially the same as that for the Mo2 dimer bridged by 1,4-cyclohexylene dicarboxylate (69 mV),27 in which the δ−δ interaction is completely disrupted by the saturated sixmembered ring. This means that electron resonance, which accounts for the electron delocalization, makes no contribution to ΔE1/2. Similarly, the ΔE1/2 value for 2 (101 mV) is about half of the corresponding value for the structurally related complex II (195 mV),16 even though sulfur atoms and electron donating methyl groups, which are capable of enhancing electronic coupling, are introduced to the charge transfer platform. In a symmetrical D−B−A system, there are two major kinds of interactions, i.e., electrostatic (through space) and electron resonant (through bonding), that are involved in the donor− acceptor coupling.28 It is interesting and, meanwhile, challenging to evaluate the two contributors in a more quantitative manner.29 The former is correlated to donor− acceptor distance, while the latter mostly relies on the orbital interaction and accounts for electron delocalization. Thus, a conjugated coplanar structure is a prerequisite for the throughbonding interaction to be in effect. In the case of 1 and 2, the d(δ)−p(π) conjugation across the bridge should be destroyed to a large extent by the large torsion angles. The redox potential separations for these two complexes result mainly from the through-space Coulomb interaction. In contrast, in the related compounds I and II, whose structures feature better coplanarity, the through-space and through-bond interactions take effects mutually, yielding large potential separations (ΔE1/2). By comparison of the ΔE1/2 values for the two species in each pair (Table 1), it is roughly estimated that, in I

Figure 3. Electronic absorption spectra of 1 (red) and 2 (blue) in CH2Cl2 solutions.

Theoretical work on similar complexes has confirmed that this band is attributed to the electronic transition from δ orbital of the Mo2 unit (HOMO) to the empty π* orbital of the bridging ligand (LUMO). Thus, it is assigned as metal-to-ligand charge transfer (MLCT).16,30 As shown in Figure 3, complexes 1 and 2 exhibit the absorption maxima at 464 and 556 nm, respectively. For I, an absorption peak appears at 490 nm, significantly lower in energy than that for 1.19 In comparison, compound 1 has a relatively large HOMO−LUMO gap, which is likely due to its higher π* orbital energy, as seen in organic amine systems with the bridge changing from benzene to anthracene via naphthalene.31 For 2, the band maximum is largely blue-shifted relative to that for II (715 nm),16 although the methyl groups on the benzene spacer are expected to lower the π* orbital energy. As is known, for conjugated D−B−A systems, enlarging the π bridge and/or introducing sulfur atoms reduces the HOMO−LUMO gap. The opposite variation trend of MLCT energies observed here is indicative of an interruption of the metal−ligand conjugation in 1 and 2, as a result of increasing the torsion angles. Mixed-Valence Properties for the Singly Oxidized Complexes 1+ and 2+. To better understand the donor− acceptor electronic coupling in the conjugation interrupted D− B−A systems, the properties of the mixed-valence species 1+ and 2+ were investigated. Compound 1 or 2 was oxidized using one equivalent of ferrocenium hexafluorophosphate (FcPF6), yielding the corresponding radical cation 1+ or 2+, respectively. We prefer to use this oxidizing reagent because the reduced species Fc is soluble in hexane and, thus, the product can be readily purified. The electron paramagnetic resonance (EPR) spectra of the resultant products were measured in situ in CH2Cl2 solution. For both, there is a single, symmetric signal in the EPR spectra (Figure 4). It is noted that, similar to those for D

DOI: 10.1021/acs.inorgchem.5b01923 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 5. Visible and near-infrared spectra for the mixed-valence complexes 1+ (top) and 2+ (bottom) in CH2Cl2 solutions, along with the spectra of the neutral compounds 1 and 2 for comparison. The intervalence absorption band for 2+, as shown in the inset, is simulated with a Gaussian-shaped curve (red dashed line).

Figure 4. X-band EPR spectra of the radical cations (1+ and 2+) generated by single electron oxidation of 1 and 2. Samples were measured in CH2Cl2 solution at 173 K.

D−B−A assemblies, ideally, electron hopping from the donor to the bridge gives a MLCT absorption, whereas hole hopping from the acceptor to the bridge (or electron hopping from the bridge to the acceptor) produces a LMCT band in the spectrum. Based upon the McConnell theory, Creutz, Newton, and Sutin proposed the so-called “CNS” model that correlates the optical behaviors to the electronic coupling parameter (HMM′). In application of the CNS model (eq 1), besides the metal−ligand coupling constants HML and HLM, the effective energy gaps for charge transfer, i.e., ΔEML and ΔELM, are the key parameters dominating the magnitude of HMM′.35 H H H H HMM ′ = ML M ′ L + LM LM ′ 2ΔEML 2ΔE LM (1)

other Mo2 dimers in this category,15,16,20 the EPR spectra lack characteristic hyperfine structures, which is different from those with carboxylate supporting ligands, as reported by Chisholm and co-workers.32 Similar g values, 1.941 for 1+ and 1.940 for 2+, are measured from the spectra, which are significantly smaller than 2.0023 for an organic radical. Thus, the EPR spectra provide solid confirmation for the presence of an unpaired electron in each of the oxidized complexes, which resides in the higher energy metal orbital, the δ orbital. It is worthwhile to note that the g values for 1+ and 2+ are appreciably smaller than that reported for I+ (1.946)19 and II+ (1.947).16 As a matter of fact, among the MV complexes {[Mo2]−bridge−[Mo2]}+, radical cations 1+ and 2+ show the smallest g values.16,19 In previous work, we have observed that the g values vary depending on the extent of charge delocalization, that is, a larger g value for a stronger delocalized system.16 Here, the small g values reflect the nature of electron localization for 1+ and 2+. Mixed-valence species 1+ and 2+ exhibit an intense absorption band in the visible region in the vis−near-IR spectra, with the band maximum slightly blue-shifted with respect to the MLCT band for the corresponding neutral precursor, as shown in Figure 5. Therefore, this band is attributed to the electronic transition from the metal-based orbital (δ) to the bridging ligand-based orbital (π*). These results are consistent with the observation in other analogous systems, showing similar metal to ligand transition energies for the closed and open shell electronic configurations for a given molecular system.16,18−20 In mixed-valence chemistry for metal complex systems, the MLCT and LMCT (ligand to metal charge transfer, if applicable) absorption bands provide important and detailed information that allows EC mechanistic investigation and ET kinetic assessment. According to the McConnell superexchange mechanism for donor−acceptor electron transfer, electron hopping and hole hopping are the two efficient pathways.33,34 In the situation involving complex

Our study has shown that, in the Mo2 systems,18,20 ΔEML and ΔELM, determined from spectroscopic data for the MV species,35 are numerically close to the MLCT and LMCT band energies, respectively. Thus, through eq 1, understanding the influence of metal−ligand charge transfer (MLCT and LMCT) on metal−metal electronic coupling is quite straightforward. In the present case, the LMCT absorption bands, which appear as a shoulder (1+) or a small bump (2+) on the low energy side of the MLCT band, are very weak (Figure 5); the high MLCT energies (large ΔEML for eq 1), compared to those for I+ and II+, reduce the electronic coupling between the two [Mo2] sites and impede the electron delocalization crossing the bridge. The characteristic optical behavior for MV compounds is the low energy intervalence charge transfer (IVCT) or metal to metal charge transfer (MMCT) absorption in the vibronic region. For the Mo2 systems having similar bridge backbones, a single, very broad absorption band across the near- and mid-IR areas is attributed, in a localized sense, to the δ electron transfer from the neutral [Mo2]0 unit (donor) to the oxidized one [Mo2]+ (acceptor).16,18−20 Interestingly, such an IVCT band is absent in the spectrum of 1+. As shown in Figure 5, there is no E

DOI: 10.1021/acs.inorgchem.5b01923 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Electronic Coupling Matrix Elements and MixedValency. Quantitative and quantum mechanical evaluation of electronic coupling relies upon the magnitude of electronic coupling matrix element (Hab). The Mulliken−Hush expression (eq 2) has been widely used for determination of Hab based on spectroscopic data from the IVCT absorption band, i.e., energy (EIT), intensity (εmax) and bandwidth at half-height (Δν1/2), and charge transfer distance (rab).37

absorption in the near- to mid-IR area in the spectrum. The spectrum in the low energy area is almost identical to that for the neutral precursor (1) and overlapped with the baseline. This optical property is in good agreement with the chemical results, indicating insignificant electronic coupling. In contrast, for I+, a typical IVCT band appears in the near-IR spectrum with a band maximum at 4650 cm−1 and a molar extinction coefficient (ε) of 1171 M−1 cm−1.19 For cation 2+, a faint (ε = 380 M−1 cm−1) and broad absorption band with a maximum at 4100 cm−1 is observed in the spectrum, which should be assigned to the IVCT band. Simulated with Gaussian-shaped profile, the half-height bandwidth is estimated to be 2470 cm−1. This IVCT band is blue-shifted in acetonitrile solution from 4650 to 8000 cm−1 (Figure S5). All these band characters point to extremely weak electronic coupling for 2+, which is abnormal for systems with coordinating S atoms and electron donating methyl groups on the charge transfer platform. In comparison with the spectrum for II+, the IVCT band for 2+ is largely shifted toward high energy, the intensity greatly lowered and the bandwidth tremendously broadened.16 The only explanation for this unusual optical behavior is the large dihedral angle between central phenyl group and the connected chelating groups, which blocks the electron delocalization across the bridge. Figure 6 schematically illustrates the influence of the

Hab = (2.06 × 10−2)

(Δv1/2εmax E IT)1/2 rab

(2)

Determination of effective electron transfer distance is critical for accurate calculation of Hab from eq 2 in that application of the geometric donor−acceptor separation would underestimate the Hab value.38 In [Mo2]−bridge−[Mo2] systems, since the coordinatively saturated complex unit [Mo2] serves as the electron donor (acceptor), the actual bridge that connects the two [Mo2] units is the central moiety, the phenylene group for I and II, for example. This aromatic entity (spacer) undergoes a small dipole moment change during the intramolecular charge transfer; thus, the size of it is taken as the effective ET distance. For example, using the length of the “−CC6H4C−” group for I+ and II+ (ca. 5.8 Å), calculation from the Mulliken−Hush expression yielded electronic coupling matrix elements Hab that are in excellent agreement with the results (HMM′) from CNS formalism.16 Apparently, the four complexes 1+, 2+, I+, and II+ share this length as the effective electron transfer distance. In application of eq 2, Hab = 0 is determined for 1+ because of its zero absorption for intervalence transition, and for 2+, Hab = 220 cm−1. As is known, a large conjugated naphthalene bridge and the electron donating methyl groups on the bridging phenyl group are able to enhance the electronic coupling. However, for 1+ and 2+, opposite effects, shown by the electrochemical and optical data measured in solution, are observed in comparison with the respective analogues I+ (Hab = 600 cm−1) and II+ (Hab = 864 cm−1). We believe that the unusual phenomena should be attributed to the increased torsion angles. Therefore, these results indicate that the bridge conformation remains in solution, although the torsion angles may vary within a certain range due to the molecular dynamics in solution. Significantly, for the series, the overall effects resulting from the conformational modification are to invoke a transition of the mixedvalence systems in different regimes in Robin−Day classification. Previous work has shown that complex I+ belongs to the weakly coupling class II, while the thiolated species II+ is considered, from its unusual optical behaviors, on the class II− III borderline.16 Contrarily, based on the optical analyses, complexes 1+ and 2+ are assigned unambiguously to the noninteracting class I and weakly coupled class II, respectively. As is noted, system transition in different Robin−Day classes has been a research focus in mixed-valency chemistry.16,39,40 The current study exemplifies that the system transitions are realized by conformational change of the bridge, while the composition of the electron donor and acceptor and the charge transfer distance remain the same. While much attention has been drawn to the class II to III transition, class I to II transition is scarcely seen. From their distinct optical behaviors, the two oxamidate bridged Mo2 isomers, {α-[Mo2(DAniF)3]2(μoxamidate)}+ (no IVCT band) and {β-[Mo2(DAniF)3]2(μoxamidate)}+ (high energy and narrow IVCT band), might be considered to belong to class I and class III, respectively;

Figure 6. Schematic illustration of d(δ)−p(π) orbital interactions along the charge transfer platform of [Mo 2]−bridge−[Mo2 ] molecules, showing the impact of bridge conformation on electronic coupling. (A) Delocalization of δ electron allowed in complexes with a coplanar conjugated bridge as seen in I and II. (B) Delocalization of δ electron blocked in complexes with a largely twisted bridge, as seen in 1.

bridge conformation in charge transfer or electron transporting. Electronically, for [Mo2]−bridge−[Mo2] systems, a charge transfer platform is established for δ electron transfer via d(δ)− p(π) conjugation when a conjugated bridge is employed; however, it is disrupted by the large torsion angle in the cases of 1+ and 2+. Undoubtedly, this is an important issue with respect to the mechanistic aspects of electronic coupling and electron transfer, which has been addressed recently in different D−B− A systems by experimental and theoretical approaches.14,31,36 Here, cation radicals 1+ and 2+, in comparison with I+ and II+, render this issue a straightforward and pictorial rationalization due to the well-defined molecular structures and electronic configurations. F

DOI: 10.1021/acs.inorgchem.5b01923 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry however, the oxamidate bridges in the α and β forms are in different coordination modes which are not interchangeable.11,41 Furthermore, it is generally accepted that both electrostatic coupling (through space) and electron resonance (through bond) contribute to invoke electronic coupling, but there is no direct evidence supporting that through-space coupling induces long distance (>5 Å) charge transfer, although through-space electron transfer is observed in face to face (3.2 Å) π−π systems.42 In this study, Hab = 0 for 1+ means that adiabatic electron transfer is not an efficient pathway, but electrochemical measurements show that there exists appreciable electronic interaction (ΔE1/2 ≫ 0) between the two redox centers, as observed for [Mo2]−bridge−[Mo2] compounds having saturated bridges.9 Nevertheless, this work illustrates that thermal donor−acceptor electron transfer can be efficiently impeded by manipulating the geometric topology of the bridge as predicted from the superexchange formalism, in the case of which the electrostatic interaction plays an insignificant role. It is hoped that further elucidation on this issue can be achieved in the followed study.

published methods. Bridging ligands 1,4-naphthalenedicarboxamides44 and piperidinium 2,5-dimethyltetrathioterephthlate45 were synthesized by modification of literature procedures. Physical Measurements. Electronic spectra were measured on a Shimadzu UV-3600 UV−vis-NIR spectrophotometer in CH2Cl2 solution. 1H NMR spectra were recorded on a Bruker Avance 300 spectrometer. Cyclic voltammograms (CVs) and differential pulse voltammograms (DPVs) were obtained using a CH Instruments model CHI660D electrochemical analyzer in 0.10 M CH2Cl2 solution of nBu4NPF6, with Pt working and auxiliary electrodes, a Ag/AgCl reference electrode, and a scan rate of 100 mV s−1. EPR spectra were measured using a Bruker A300-10-12 electron paramagnetic resonance spectrometer. Measurements for the mixed-valence complexes were carried out in situ after single electron oxidation of the corresponding neutral compounds by FcPF6. X-ray Crystal Structure Determinations. Single-crystal data for 1·2CH2Cl2 were collected on an Agilent Xcalibur Nova diffractometer with Cu Kα radiation (λ = 1.54178 Å) at 100(2) K, and single-crystal data for 2·2CH3CH2OH·2H2O were collected on an Agilent Gemini S Ultra Xcalibur Nova diffractometer with Cu Kα radiation (λ = 1.54178 Å) at 173(2) K. For both, the empirical absorption corrections were applied using spherical harmonics, implemented in the SCALE3 ABSPACK scaling algorithm.46 All the structures were solved using direct methods, which yielded the positions of all non-hydrogen atoms. Hydrogen atoms were placed in calculated positions in the final structure refinement. Structure determination and refinement were carried out using SHELXS-97 and SHELXL-97 programs, respectively.47 For the two measured crystal structures, the solvent molecules were disordered in multiple orientations, which were refined isotropically. All non-hydrogen atoms were refined with anisotropic displacement parameters. 1,4-Naphthalenedicarboxamide. A mixture having 1,4-naphthalenedicarboxlic acid (4.0 g, 0.019 mol) and N,N-dimethylformamide (two drops) in thionyl chloride (20 mL) was allowed to reflux for 12 h. After the reaction mixture was cooled to ambient temperature, all the volatiles were removed under reduced pressure. The residue was dissolved in THF (20 mL). To the resultant solution was dropwise added 5 mL of ammonia aqueous solution (30%) at 0 °C, yielding yellow precipitates. After filtration, the solid product was washed with distilled water and then ethanol. The target compound (3.24 g) was obtained in a yield of 82%. 1H NMR δ (ppm in DMSOd6): 8.31 (dd, 2H, aromatic H), 8.05 (br, 2H, −NH2), 7.68 (br, 2H, −NH2), 7.62 (s, 2H, aromatic H), 7.60 (d, 2H, aromatic H). Piperidinium 2,5-Dimethyltetrathioterephthlate. p-Xylene (3.1 mL, 24.0 mmol), paraformaldehyde (1.53 g, 20 mmol), and HBr/acetic acid solution (31 wt %, 27.18 g, 28 mmol) were mixed and refluxed under nitrogen for 12 h. After the reaction solution was poured into 100 mL of water, the solid product 1,4-bis(bromomethyl)2,5-dimethylbenzene was formed, collected by filtration, and dried under vacuum. Yield: 8.50 g (86%). To a 100 mL flask containing 1,4-bis(bromomethyl)-2,5-dimethylbenzene (1.46 g, 5.5 mmol), sulfur (0.64 g, 20 mmol), and sodium ethoxide (1.30 g, 20 mmol) was added 30 mL of ethanol. The mixture refluxed overnight, generating a brown solution suspended with some precipitates. After filtration, the diluted HCl (25.0 mL) was added slowly to the solution, and then 30 mL of CH2Cl2 was used to extract the desired organic acid. With stirring, piperidine (6.0 mL, 10 mmol) was added to the organic component, yielding orange piperidinium 2,5-dimethyltetrathioterephthlate. The product was washed with Et2O three times and then dried in vacuo overnight. Yield: 4.04 g (53%). 1H NMR δ (ppm in DMSO-d6): 8.51 (br, 4H, −NH2), 6.53 (s, 2H, aromatic H), 3.01 (t, 8H, −CH2NH2CH2−), 2.14 (s, 6H, −CH3), 1.66 (m, 8H, −CH2CH2NH2CH2CH2−), 1.57 (m, 4H, −NH2CH2CH2CH2−). [Mo2(DAniF)3]2[μ-1,4-{C(O)NH}2-C10H6] (1). To a mixture of Mo2(DAniF)3(O2CCH3) (0.203 g, 0.20 mmol) and NaOC2H5 (0.0136 g, 0.20 mmol) were added 30 mL of THF and 10 mL of ethanol. After being stirred for 30 min, the mixture was transferred to a THF (10 mL) solution dissolving 1,4-naphthalenedicarboxamide (0.0257 g, 0.12 mmol). After addition, the mixture was stirred for 4

■

CONCLUDING REMARKS Employing naphthalene bis(amidate) and 2,5-dimethyltetrathioterephthalate as the bridging ligands, two dimolybdenum dimers 1 and 2, respectively, have been synthesized as D−B−A model compounds for evaluating the impacts of torsional direction of bridge on metal−metal electronic coupling. These two compounds feature commonly a large dihedral angle (>60°) between the central aromatic moiety and the associated Mo2 chelating rings, and thus, the d(δ)−p(π) conjugation along the charge transfer platform is interrupted. For each of the two compounds, electrochemical measurements showed appreciable electrostatic interaction (through space), but the potential separations (ΔE1/2) for the two dimetal redox centers are largely lowered in comparison with that for the previously reported phenylene spaced complex I corresponding to 1, and II corresponding to 2. Notably, mixed-valence species 1+, generated by one-electron oxidation using ferrocenium hexafluorophosphate, lacks an IVCT absorption band in the vis−near-IR spectrum. For 2+, the IVCT band is largely shifted toward high energy, the band intensity greatly lowered, and the bandwidth tremendously broadened, relative to the IV absorption band for II+. Spectroscopic results indicate that the electron resonance effect (through bond) is either eliminated (as in 1+) or significantly weakened (as in 2+), as a result of increasing the torsion angle. The values of electronic coupling parameter (Hab), Hab = 0 cm−1 for 1+ and Hab = 220 cm−1 for 2+, are determined from the Mulliken−Hush expression, which are much smaller than those for the phenylene bridged analogues, ca. 600 cm−1 for I+ and 864 cm−1 for II+. Accordingly, mixed-valence cations 1+ and 2+ are considered in the class I and class II regimes, respectively, in terms of Robin−Day classification. Therefore, together with I+ and II+, a transition from class I to class II−III borderline is observed in the present [Mo2]−bridge−[Mo2] series, which is realized by modifying the geometric topology.

■

EXPERIMENTAL SECTION

Materials and Methods. All the manipulations were performed in a nitrogen-filled glovebox or by using standard Schlenk-line techniques. All solvents were freshly distilled and dried over appr opriate drying a gent s u nder N 2 . H DA niF 4 3 a n d Mo2(DAniF)3(O2CCH3 )11,15 were prepared by following the G

DOI: 10.1021/acs.inorgchem.5b01923 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

(5) (a) Nelsen, S. F.; Ismagilov, R. F.; Trieber, D. A., II Science 1997, 278, 846−849. (b) Heckmann, A.; Lambert, C. Angew. Chem., Int. Ed. 2012, 51, 326−392. (6) (a) Ito, T.; Hamaguchi, T.; Nagino, H.; Yamaguchi, T.; Kido, H.; Zavarine, I. S.; Richmond, T.; Washington, J.; Kubiak, C. P. J. Am. Chem. Soc. 1999, 121, 4625−4632. (b) Ito, T.; Imai, N.; Yamaguchi, T.; Hamaguchi, T.; Londergan, C. H.; Kubiak, C. P. Angew. Chem., Int. Ed. 2004, 43, 1376−1381. (c) Salsman, J. C.; Kubiak, C. P. J. Am. Chem. Soc. 2005, 127, 2382−2383. (7) (a) Kaim, W.; Lahiri, G. K. Angew. Chem., Int. Ed. 2007, 46, 1778−1796. (b) Cayton, R. H.; Chisholm, M. H.; Huffman, J. C.; Lobkovsky, E. B. J. Am. Chem. Soc. 1991, 113, 8709−8724. (c) Xu, G.L.; Crutchley, R. J.; DeRosa, M. C.; Pan, Q.-J.; Zhang, H.-X.; Wang, X.; Ren, T. J. Am. Chem. Soc. 2005, 127, 13354−13363. (d) Yao, C.-J.; Zhong, Y.-W.; Yao, J. J. Am. Chem. Soc. 2011, 133, 15697−15706. (8) (a) Lambert, C.; Nöll, G. J. Am. Chem. Soc. 1999, 121, 8434− 8442. (b) Heckmann, A.; Lambert, C. J. Am. Chem. Soc. 2007, 129, 5515−5527. (9) Cotton, F. A.; Donahue, J. P.; Murillo, C. A. J. Am. Chem. Soc. 2003, 125, 5436−5450. (10) (a) Xu, G.-L.; Zou, G.; Ni, Y.-H.; DeRosa, M. C.; Crutchley, R. J.; Ren, T. J. Am. Chem. Soc. 2003, 125, 10057−10065. (b) Launay, J.P. Chem. Soc. Rev. 2001, 30, 386−397. (11) Cotton, F. A.; Liu, C. Y.; Murillo, C. A.; Villagrán, D.; Wang, X. J. Am. Chem. Soc. 2003, 125, 13564−13575. (12) Benniston, A. C.; Harriman, A. Chem. Soc. Rev. 2006, 35, 169− 179. (13) (a) Barybin, M. V.; Chisholm, M. H.; Dalal, N. S.; Holovics, T. H.; Patmore, N. J.; Robinson, R. E.; Zipse, D. J. J. Am. Chem. Soc. 2005, 127, 15182−15190. (b) Chisholm, M. H.; Chou, P.-T.; Chou, Y.-H.; Ghosh, Y.; Gustafson, T. L.; Ho, M.-L. Inorg. Chem. 2008, 47, 3415−3425. (14) Albinsson, B.; Eng, M. P.; Pettersson, K.; Winters, M. U. Phys. Chem. Chem. Phys. 2007, 9, 5847−5864. (15) Lei, H.; Xiao, X.; Meng, M.; Cheng, T.; Shu, Y.; Tan, Y. N.; Liu, C. Y. Inorg. Chim. Acta 2015, 424, 63−74. (16) Xiao, X.; Liu, C. Y.; He, Q.; Han, M. J.; Meng, M.; Lei, H.; Lu, X. Inorg. Chem. 2013, 52, 12624−12633. (17) (a) Cotton, F. A.; Li, Z.; Liu, C. Y.; Murillo, C. A. Inorg. Chem. 2007, 46, 7840−7847. (b) Chisholm, M. H.; Patmore, N. J. Dalton Trans. 2006, 3164−3169. (c) Han, M. J.; Liu, C. Y.; Tian, P. F. Inorg. Chem. 2009, 48, 6347−6349. (18) Liu, C. Y.; Xiao, X.; Meng, M.; Zhang, Y.; Han, M. J. J. Phys. Chem. C 2013, 117, 19859−19865. (19) Shu, Y.; Lei, H.; Tan, Y. N.; Meng, M.; Zhang, X. C.; Liu, C. Y. Dalton Trans. 2014, 43, 14756−14765. (20) Xiao, X.; Meng, M.; Lei, H.; Liu, C. Y. J. Phys. Chem. C 2014, 118, 8308−8315. (21) (a) Carano, M.; Careri, M.; Cicogna, F.; D'Ambra, I.; Houben, J. L.; Ingrosso, G.; Marcaccio, M.; Paolucci, F.; Pinzino, C.; Roffia, S. Organometallics 2001, 20, 3478−3490. (b) Gao, L.-B.; Zhang, L.-Y.; Shi, L.-X.; Chen, Z.-N. Organometallics 2005, 24, 1678−1684. (c) de Montigny, F.; Argouarch, G.; Costuas, K.; Halet, J.-F.; Roisnel, T.; Toupet, L.; Lapinte, C. Organometallics 2005, 24, 4558−4572. (22) (a) Vilà, N.; Zhong, Y.-W.; Henderson, J. C.; Abruña, H. D. Inorg. Chem. 2010, 49, 796−804. (b) Fabre, M.; Bonvoisin, J. J. Am. Chem. Soc. 2007, 129, 1434−1444. (c) Ou, Y.-P.; Jiang, C.; Wu, D.; Xia, J.; Yin, J.; Jin, S.; Yu, G.-A.; Liu, S. H. Organometallics 2011, 30, 5763−5770. (23) Cotton, F. A.; Li, Z.; Liu, C. Y.; Murillo, C. A.; Villagrán, D. Inorg. Chem. 2006, 45, 767−778. (24) Cotton, F. A.; Li, Z.; Liu, C. Y.; Murillo, C. A. Inorg. Chem. 2006, 45, 9765−9770. (25) Cotton, F. A.; Daniels, L. M.; Donahue, J. P.; Liu, C. Y.; Murillo, C. A. Inorg. Chem. 2002, 41, 1354−1356. (26) Richardson, D. E.; Taube, H. Inorg. Chem. 1981, 20, 1278− 1285. (27) Cotton, F. A.; Donahue, J. P.; Lin, C.; Murillo, C. A. Inorg. Chem. 2001, 40, 1234−1244.

h, during which the color of the solution gradually changed from yellow to red. All the volatiles were removed under reduced pressure, and the residue was extracted with dichloromethane and subsequently filtered through a Celite-packed funnel to remove the insoluble impurity. The filtrate was evaporated under vacuum, and the solid product was washed with ethanol (20 mL × 3) and collected by filtration. Yield: 0.155 g (73%). Diffusion of ethanol into a dichloromethane solution of the compound yielded orange red needle crystals. 1H NMR δ (ppm in CDCl3): 8.80 (s, 2H, −C(O)NH), 8.55 (dd, 2H, aromatic H), 8.49 (s, 2H, −NCHN−), 8.47 (s, 4H, −NCHN−), 7.82 (s, 2H, aromatic H), 7.28 (d, 2H, aromatic H), 6.62 (d, 24H, aromatic H), 6.50 (d, 4H, aromatic H), 6.43 (d, 12H, aromatic H), 6.28 (d, 8H, aromatic H), 3.70 (s, 30H, −OCH3), 3.65 (s, 6H, −OCH3). UV−vis, λmax nm (ε, M−1 cm−1): 464 (1.18 × 104). Anal. Calcd for C102H98Mo4N14O14 (1): C, 57.58; H, 4.64; N, 9.22. Found: C, 56.08; H, 4.80; N, 9.12 [Mo2(DAniF)3]2[μ-1,4-(CS2)2-2,5-Me2C6H2] (2). A procedure similar to that described above was employed to prepare this compound. The reaction produced 0.132 g of greenish blue solid compound in a yield of 60%. Single crystals suitable for X-ray diffraction were obtained by diffusion of ethanol into a dichloromethane solution of the compound. 1H NMR δ (ppm in CDCl3): 8.51 (s, 2H, −NCHN−), 8.32 (s, 4H, −NCHN−), 7.61 (s, 2H, aromatic H), 6.65 (d, 16H, aromatic H), 6.60 (d, 16H, aromatic H), 6.44 (d, 8H, aromatic H), 6.14 (d, 8H, aromatic H), 3.74 (s, 24H, −OCH3), 3.68 (s, 12H, −OCH3), 2.60 (s, 6H, −CH3). UV−vis, λmax nm (ε, M−1 cm−1): 556 (2.08 × 104). Anal. Calcd for C100H98Mo4N12O12S4 (2): C, 55.30; H, 4.55; N, 7.74. Found: C, 57.08; H, 4.68; N, 7.86.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.5b01923. Crystallographic information (CIF) Crystallographic data and spectral details (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We are grateful for the financial support from National Natural Science Foundation of China (21371074, 21301070 and 90922010), Jinan University.

■

REFERENCES

(1) (a) Osyczka, A.; Moser, C. C.; Daldal, F.; Dutton, P. L. Nature 2004, 427, 607−612. (b) Blondin, G.; Girerd, J. Chem. Rev. 1990, 90, 1359−1376. (c) Wang, H.; Lin, S.; Allen, J. P.; Williams, J. C.; Blankert, S.; Laser, C.; Woodbury, N. W. Science 2007, 316, 747−750. (2) (a) Huang, Y. H.; Rettner, C. T.; Auerbach, D. J.; Wodtke, A. M. Science 2000, 290, 111−114. (b) Kumar, A.; Sevilla, M. D. Chem. Rev. 2010, 110, 7002−7023. (3) (a) Creutz, C.; Taube, H. J. Am. Chem. Soc. 1969, 91, 3988− 3989. (b) Creutz, C.; Taube, H. J. Am. Chem. Soc. 1973, 95, 1086− 1094. (4) (a) Creutz, C. Prog. Inorg. Chem. 1983, 30, 1−73. (b) Crutchley, R. J. Adv. Inorg. Chem. 1994, 41, 273−325. (c) Ren, T. Chem. Rev. 2008, 108, 4185−4207. (d) Ito, T.; Hamaguchi, T.; Nagino, H.; Yamaguchi, T.; Washington, J.; Kubiak, C. P. Science 1997, 277, 660− 663. H

DOI: 10.1021/acs.inorgchem.5b01923 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry (28) (a) Gong, Z.-L.; Zhong, Y.-W.; Yao, J. Chem. - Eur. J. 2015, 21, 1554−1566. (b) Quardokus, R. C.; Lu, Y. H.; Wasio, N. A.; Lent, C. S.; Justaud, F.; Lapinte, C.; Kandel, S. A. J. Am. Chem. Soc. 2012, 134, 1710−1714. (29) (a) Richardson, D. E.; Taube, H. Coord. Chem. Rev. 1984, 60, 107−129. (b) Evans, C. E. B.; Naklicki, M. L.; Rezvani, A. R.; White, C. A.; Kondratiev, V. V.; Crutchley, R. J. J. Am. Chem. Soc. 1998, 120, 13096−13103. (30) Cotton, F. A.; Donahue, J. P.; Murillo, C. A.; Pérez, L. M. J. Am. Chem. Soc. 2003, 125, 5486−5492. (31) Lambert, C.; Risko, C.; Coropceanu, V.; Schelter, J.; Amthor, S.; Gruhn, N. E.; Durivage, J. C.; Brédas, J.-L. J. Am. Chem. Soc. 2005, 127, 8508−8516. (32) (a) Chisholm, M. H.; Patmore, N. J. Acc. Chem. Res. 2007, 40, 19−27. (b) Brown-Xu, S. E.; Chisholm, M. H.; Durr, C. B.; Lewis, S. A.; Spilker, T. F.; Young, P. J. Inorg. Chem. 2014, 53, 637−644. (33) McConnell, H. M. J. Chem. Phys. 1961, 35, 508−515. (34) Newton, M. D. Chem. Rev. 1991, 91, 767−792. (35) Creutz, C.; Newton, M. D.; Sutin, N. J. Photochem. Photobiol., A 1994, 82, 47−59. (36) (a) Karafiloglou, P.; Launay, J.-P. Chem. Phys. 2003, 289, 231− 242. (b) Fraind, A. M.; Sini, G.; Risko, C.; Ryzhkov, L. R.; Brédas, J.L.; Tovar, J. D. J. Phys. Chem. B 2013, 117, 6304−6317. (37) (a) Hush, N. S. Prog. Inorg. Chem. 1967, 8, 391−444. (b) Hush, N. S. Electrochim. Acta 1968, 13, 1005−1023. (38) (a) Nelsen, S. F.; Newton, M. D. J. Phys. Chem. A 2000, 104, 10023−10031. (b) D’Alessandro, D. M.; Dinolfo, P. H.; Hupp, J. T.; Junk, P. C.; Keene, F. R. Eur. J. Inorg. Chem. 2006, 2006, 772−783. (39) (a) Nelsen, S. F. Chem. - Eur. J. 2000, 6, 581−588. (b) Brunschwig, B. S.; Creutz, C.; Sutin, N. Chem. Soc. Rev. 2002, 31, 168−184. (c) Demadis, K. D.; Hartshorn, C. M.; Meyer, T. J. Chem. Rev. 2001, 101, 2655−2685. (40) (a) Lambert, C.; Amthor, S.; Schelter, J. J. Phys. Chem. A 2004, 108, 6474−6486. (b) DeRosa, M. C.; White, C. A.; Evans, C. E. B.; Crutchley, R. J. J. Am. Chem. Soc. 2001, 123, 1396−1402. (41) Cotton, F. A.; Liu, C. Y.; Murillo, C. A.; Zhao, Q. Inorg. Chem. 2007, 46, 2604−2611. (42) Sun, D.-L.; Rosokha, S. V.; Lindeman, S. V.; Kochi, J. K. J. Am. Chem. Soc. 2003, 125, 15950−5963. (43) Lin, C.; Protasiewicz, J. D.; Smith, E. T.; Ren, T. Inorg. Chem. 1996, 35, 6422−6428. (44) Gao, M. Z.; Reibenspies, J. H.; Zingaro, R. A.; Wang, B.; Xu, Z. L. J. Heterocycl. Chem. 2004, 41, 899−908. (45) Van der Made, A. W.; Van der Made, R. H. J. Org. Chem. 1993, 58, 1262−1263. (46) CrysAlis RED, Version 1.171.31.7; Oxford Diffraction Ltd: Abington, U.K., 2006. (47) Sheldrick, G. M. SHELXTL, Version 6.12; Bruker Analytical Xray Systems, Inc.: Madison, WI, 2000.

■

NOTE ADDED AFTER ASAP PUBLICATION This paper was published on the Web on November 18, 2015, with minor text errors in the Supporting Information file. The SI file was corrected and reposted on November 19, 2015.

I

DOI: 10.1021/acs.inorgchem.5b01923 Inorg. Chem. XXXX, XXX, XXX−XXX