The electronic structure of [Cr(tren)(3,6-DTBSQ)]2+, where tren is tris(2-aminoethyl)amine and 3,6-DTBSQ is 3,6-di-tert-butylorthosemiquinone, has been studied by self-consistent-field non-local gradient-corrected density functional theory. The results are consistent with a Heisenberg exchange formulation where a Cr(S = 3/2) ion is antiferromagnetically coupled to the semiquinone(S = 1/2) giving rise to a S = 1 ground state. Population analyses were carried out which show net α and β spin densities at the chromium ion and semiquinone, respectively. Some orbital interactions have been identified that allow partial delocalization from the semiquinone toward the chromium ion giving rise to an antiparallel alignment of their electron spins. The isotropic exchange constant J of the Heisenberg Hamiltonian ℋex = J S 1·S 2 has been determined from the self-consistent-field energies at the U-BLYP/6-311** and U-B3LYP/6-311** levels and is consistent with previously reported magnetic susceptibility data. The triplet state wave function shows some spin contamination from the higher-lying quintet state. Accordingly, approximate spin and energy projections were performed to account for the quintet admixture. Some magneto-structural correlations between the Cr−OSQ and OSQ−CSQ bond lengths and the magnitude of J have also been investigated. It was found that J decays in a nearly exponential fashion with increasing bond distances. The electronic structure of the free semiquinone ligand has also been studied and correlated to its bonding with chromium. Finally, a single-crystal X-ray structure of the related complex [Cr(tren)(3,6-DTBCat)]+ was obtained and used to carry out similar self-consistent-field calculations. Analysis of the quartet ground state wave function of the catecholate complex produced spin densities consistent with a Cr(S = 3/2)−catechol(S = 0) formulation.

The synthesis, structural and spectroscopic characterization of monosemiquinone and monocatechol complexes of chromium(III) are described. Compounds of the general form [Cr(N(4))Q](n+), where N(4) represents a tetradentate or bis-bidentate nitrogenous ligand or ligands and Q represents a reduced form of an orthoquinone, have been prepared by two different routes from Cr(III) and Cr(II) starting materials. The complex [Cr(tren)(3,6-DTBSQ)](PF(6))(2), where tren is tris(2-aminoethyl)amine and 3,6-DTBSQ is 3,6-di-tert-butylorthosemiquinone, crystallizes in the monoclinic space group P2(1)/c with a = 11.9560(2) Å, b = 17.0715(4) Å, c = 17.1805(4) Å, beta = 90.167(1) degrees, V = 3506.6(1) Å(3), Z = 4, with R = 0.056 and R(w) = 0.070. Alternating C-C bond distances within the quinoidal ligand confirm its semiquinone character. Variable temperature magnetic susceptibility data collected on solid samples of both [Cr(tren)(3,6-DTBSQ)](PF(6))(2) and [Cr(tren)(3,6-DTBCat)](PF(6)) in the range 5-350 K exhibit temperature-independent values of 2.85 +/- 0.1 &mgr;(B) and 3.85 +/- 0.1 &mgr;(B), respectively. These data are consistent with a simple Cr(III)-catechol formulation (S = (3)/(2)) in the case of [Cr(tren)(3,6-DTBCat)](PF(6)) and strong antiferromagnetic coupling (|J| > 350 cm(-)(1)) between the Cr(III) and the semiquinone radical in [Cr(tren)(3,6-DTBSQ)](PF(6))(2). The absorption spectrum of the semiquinone complex exhibits a number of sharp, relatively intense transitions in fluid solution. Group theoretical arguments coupled with a qualitative ligand-field analysis including the effects of Heisenberg spin exchange suggest that several of the observed transitions are a consequence of exchange interactions in both the ground- and excited-state manifolds of the compound. The effect of electron exchange on excited-state dynamics has also been probed through static emission as well as time-resolved emission and absorption spectroscopies. It is suggested that the introduction of exchange coupling and subsequent change in the molecule's electronic structure may contribute to an increase of nearly 4 orders of magnitude in the rate of radiative decay (k(r)), and a factor of ca. 10(7) in the rate of nonradiative decay (k(nr)).

The electronic structures of the three oxidation states of the "noninnocent" ligand 3,6-di-tert-butylorthoquinone (3,6-DTBQ) have been studied by nonlocal gradient-corrected density functional theory. Optimized structures obtained at the B3LYP/6-31G* and BLYP/6-31G* levels show that neutral 3,6-DTBQ has two equivalent CO double bonds and a nonaromatic six-membered carbon ring. Upon one-and two-electron reduction to its semiquinone (3,6-DTBSQ) and catechol (3,6-DTBCat) oxidation states, respectively, the single bonds of the ligand become shorter whereas its double bonds elongate. The carbon ring of catechol acquires nearly aromatic character perturbed by a long C1-C2 bond. The calculations confirm that 3,6-DTBQ and 3,6-DTBCat have closed-shell configurations and singlet ground states whereas the 3,6-DTBSQ has an openshell configuration and a doublet ground state. Analogous calculations have also been carried out on the 3,5-di-tert-butylsemiquinone (3,5-DTBSQ) isomer. Single point calculations at the U-B3LYP/6-311G** level show that both semiquinone isomers have smaller negative charge densities at the carbons bonded to their tert-butyl groups relative to other carbons of their six-membered rings. The spin densities of both semiquinone isomers are mainly localized at their oxygens with somewhat different delocalization patterns throughout the six-membered ring. Detailed descriptions of the composition of frontier molecular orbitals are given that reveal subtle differences between charge distributions and molecular orbital energies across the orthoquinone/ semiquinone/catechol redox series. Finally, optimized geometric parameters for the closely related molecule 1,2-benzoquinone have been obtained and compared with its X-ray structure to assess possible discrepancies between experimental and theoretical methods.

Bimolecular quenching between photosensitizers and exchange-coupled transition metal complexes has been studied in an effort to experimentally establish a link between Heisenberg spin exchange and chemical reactivity. The acceptors are members of the oxo/hydroxo-biscarboxylato class of dinuclear Fe(III) compounds, where protonation of the oxo bridge provides a means for modulating the magnitude of spin exchange within the cluster. Photoexcitation of solutions containing Ru(II) polypyridyl sensitizers and the Fe(III) complexes results in quenching of emission from the (3)MLCT excited state of the Ru(II) chromophores; nanosecond time-resolved absorption measurements demonstrate that quenching occurs, in part, by electron transfer. Decoupling electron transfer driving force (DeltaG(0)(ET)) from changes in the magnitude of spin exchange was achieved by varying the bridging carboxylate to afford a series of complexes of the form [Fe(2)O(H)(O(2)CR)(2)(Tp)(2)](n)(+) (n = 0, 1, 2). Electrochemical measurements reveal a greater than 500 mV shift in cluster reduction potential across the series (i.e., R = CH(3) to CF(3)), whereas variable-temperature magnetic susceptibility measurements demonstrate a corresponding invariance in spin exchange between the metal centers (J(oxo) = -119 +/- 4 cm(-1) and J(hydroxo) = -18 +/- 2 cm(-1) for H = -2JS(1).S(2)). Structural analyses suggest that reorganization energies (lambda) associated with electron transfer should be identical for all molecules within a given series (i.e., oxo or hydroxo bridged); likewise Deltalambda between the series is expected to be small. A comparison of quenching rates for the two extended series firmly establishes that neither reorganization energy nor electron transfer driving force considerations can account for differences in reactivity between oxo-bridged (large spin exchange) and hydroxo-bridged (small spin exchange) quenchers. Upon consideration of energy transfer contributions, it is determined that reactivity differences between the oxo- and hydroxo-bridged quenchers must lie in the relative rates of Dexter energy transfer and/or electron transfer, with the origin of the latter linked to something other than DeltaG(0)(ET) or lambda. Finally, the extent to which spin exchange within the dinuclear Fe(III) quenchers can be identified as the key variable influencing these reactivity patterns is discussed.