Porphyrins bearing a polyaromatic or a heterocyclic group are prepared to study their fundamental and photovoltaic properties. Solar cells sensitized with a fluorene-modified porphyrin outperform other dyes in the series, reaching ~90% efficiency of N719 dye.

Three new hydridosilazido ligands, -N(SiHMe2)Aryl (Aryl = Ph, 2,6-C6Me2H3 (dmp), 2,6-C6iPr2H3 (dipp)) and their rare earth complexes Ln{N(SiHMe2)Aryl}3(THF)n (Ln = Sc, Y, Lu; Aryl = Ph, n = 2; Aryl = dmp, n = 1; Aryl = dipp, n = 0) were synthesized to study the relationships between ligand steric properties, secondary Ln↼H-Si bonding, and the reactivity of amide and SiH groups. In these compounds, the steric encumbrance of the aryl group was systematically increased from phenyl to 2,6-dimethylphenyl to 2,6-diisopropylphenyl. NMR, IR and X-ray diffraction studies of the complexes characterize the number of secondary interactions and additional THF ligands coordinated to the rare earth centers. The complexes with the smallest phenylsilazido ligands, Ln{N(SiHMe2)Ph}3(THF)2, contain features associated with three non-bridging 2c-2e Si-H. Characterization of intermediate-sized Ln{N(SiHMe2)dmp}3THF reveals three and two Ln↼H-Si interactions for yttrium and lutetium analogues, respectively, with both metals having one coordinated THF per complex. Ln{N(SiHMe2)dipp}3 is formed solvent-free, and all three ligands adopt Ln↼H-Si bonding modes. The reaction between Ln{N(SiHMe2)dipp}3 and ketones provides the hydrosilylated product via addition of C=O and Si-H bonds, which occurs rapidly even at low temperature. This reaction is proposed to occur through an associative mechanism on the basis of negative activation entropy measured for substitution of pyridine in Ln{N(SiHMe2)dipp}3•NC5H5.

Porphyrins bearing a polyaromatic or a heterocyclic group are prepared to study their fundamental and photovoltaic properties.

Six new solvent-free, homoleptic paramagnetic tris­(alkyl)­lanthanides Ln­{C­(SiHMe2)3}3 (1Ln) and Ln­{C­(SiHMe2)2Ph}3 (2Ln) (Ln = Gd, Dy, and Er) were synthesized to investigate the magnetic properties of 4f organometallic compounds stabilized by secondary Ln↼H–Si and benzylic interactions. The unit cell of 1Gd contains one independent molecule (Z = 2), while 1Dy and 1Er crystallize with four independent isostructural molecules per unit cell (Z = 16). In all molecules, as in other 1Ln compounds, the three tris­(dimethylsilyl)­methyl ligands form a trigonal planar LnC3 core, and six secondary interactions involving Ln↼H–Si bonding in Ln­{C­(SiHMe2)3}3 form above and below the equatorial plane. Two and five crystallographically independent molecules of each 2Ln (2Gd, Z = 8; 2Dy, Z = 20) form with three π-coordinated phenyl groups in addition to either one or two secondary Ln↼H–Si interactions per molecule. The packing of these midseries organolanthanide compounds contrasts the single crystallographically unique molecules in previously reported La­{C­(SiHMe2)3}3 (1La, Z = 2, Z′ = 1) and La­{C­(SiHMe2)2Ph}3 (2La, Z = 2, Z′ = 1/3). 2La doped with 2Dy can adopt the crystallographic structure of 2La, which promotes magnetic properties, namely a higher χm T value at low temperatures as well as stronger magnetic anisotropy. The ac susceptibility data for 10% 2Dy doped into 2La suggests slow relaxation at low temperatures with a relaxation barrier of ∼45 K. The computed saturated magnetization of 1Er (M ≈ 4.5 μB) and 1Dy (M ≈ 6 μB) matches the experimental values, while the computed value for 2Dy better matches the value measured for 2Dy diluted in 2La (M ≈ 5 μB). Gas-phase calculations predict that the ground-state and first excited-state multiplet separations are larger for 1Er than 2Er, while the ordering for dysprosium is 1Dy > 2Dy.

The proligands C5H5CMe2CHPhOxR (R = Me2, CHMe2, and CMe3) react with M­(NMe2)4 (M = Ti, Zr, and Hf) to give monodeprotonated (OxRCHPhCMe2C5H4)­M­(NMe2)3, doubly deprotonated (C5H4CMe2CPhOxR)­M­(NMe2)2, or a mixture of both. The observed products depend on reaction conditions, the oxazoline substituent, and the metal center, with 4-t-butyl-oxazoline or Ti giving only the monodeprotonated (OxRCHPhCMe2C5H4)­M­(NMe2)3. Amine elimination from (OxRCHPhCMe2C5H4)­M­(NMe2)3 to give (C5H4CMe2CPhOxR)­M­(NMe2)2 is reversible for 4,4-dimethyl- and 4-isopropyl-oxazoline-based ligands in Zr or Hf complexes. Temperature-dependent kinetic studies of the equilibration of (C5H4CMe2CPhOxMe2)­Zr­(NMe2)2, HNMe2 and (OxMe2CHPhCMe2C5H4)­Zr­(NMe2)3 provide the experimental thermodynamic parameters ΔS° = −17.4 ± 2.6 cal·mol–1K–1 and ΔH° = −6.8 ± 0.8 kcal·mol–1. An Eyring plot of the rate constants, determined from the system as it approaches equilibrium, gives the activation entropy and activation enthalpy for the addition of (C5H4CMe2CPhOxMe2)­Zr­(NMe2)2 and HNMe2 as −36 ± 4 cal·mol–1 K–1 and 8.9 ± 1.3 kcal·mol–1, respectively; the elimination of HNMe2 from (OxMe2CHPhCMe2C5H4)­Zr­(NMe2)3 is characterized by ΔS ⧧ = −19 ± 5 cal·mol–1 K–1 and ΔH ⧧ = 15.7 ± 1.5 kcal·mol–1. DFT computational models indicate a single-step, nonlinear transfer of the H between the benzylic position of the noninnocent, oxazoline-coordinated ligand and NMe2. Computations also confirm the negative activation entropy and the trends in the barriers support the experimental results. Together, these studies indicate the importance of steric effects from the oxazoline ligand, metal center, ancillary ligands, and leaving group on the shuttling of the proton between HNMe2 and the noninnocent ligand. Overall, these effects suggest that coordination of oxazoline to the metal center is a key part of the benzylic deprotonation and noninnocent behavior of the cyclopentadienyl-oxazoline ligand.