Metal and Ligand K-edge XAS of Titanium−TEMPO Complexes:  Determination of Oxidation States and Insights into Ti−O Bond Homolysis

Abstract

Ti-TEMPO (TEMPO = 2,2,6,6-tetramethylpiperidine-N-oxyl) provides a means for generating Ti(III) complexes by homolysis of the Ti-O bond. It has been determined that bis-Cp-Ti-TEMPO complexes readily undergo homolytic cleavage while the mono-Cp-Ti-TEMPO complexes do not. Here Ti K- and Cl K-edge XAS are applied to directly determine the oxidation state of TiCl3TEMPO, TiCpCl2TEMPO, and TiCp2ClTEMPO, with reference to Ti(III) and Ti(IV) complexes of known oxidation state. The Ti K-edge data show that Ti(III) complexes exhibit a pre-edge feature approximately 1 eV lower than any of the Ti(IV) complexes; while the Cl K-edges show that Ti(III) complexes have a Cl K- pre-edge feature to approximately 1 eV higher energy than any of the Ti(IV) complexes. Taken together, the Ti and Cl K-edge data indicate that the Ti-TEMPO complexes are best described as Ti(IV)-TEMPO anions (rather than Ti(III)-nitroxyl radicals). In addition, the Cl K-edges indicate that replacement of Cl by Cp weakens the bonding with the remaining ligands, with the Cl 3p covalency decreasing from 25% to 21% to 17% on going from TiCl3TEMPO to TiCpCl2TEMPO to TiCp2ClTEMPO. DFT calculations also show that the electronic structures of the Ti-TEMPO complexes are modulated by the replacement of chloride by Cp. The effect of the Cp on the ancillary ligation is one factor that contributes to facile Ti-O bond homolysis in TiCp2ClTEMPO. However, the results indicate the primary contribution to the energetics of Ti-O bond homolysis in TiCp2ClTEMPO is stabilization of the three-coordinate product by Cp.

Quantifying the Interdependence of Metal–Ligand Covalency and Bond Distance Using Ligand K‐edge XAS

“…Perhaps not surprising given the inherent uncertainties when using ionic radii that are highly dependent on both oxidation state and coordination number, no correlation between %Cl 3p character and relative M−Cl bond distance was observed (Figure S6). However, the plot revealed that a correlation does exist for group IV complexes, especially for the numerous Ti 4+ and Ti 3+ complexes measured by different research groups with diphosphine, cyclopentadienyl, and (2,2,6,6‐tetramethylpiperidin‐1‐yl)oxyl (TEMPO) ligands . Indeed, analysis of the Ti complexes revealed a strong correlation between %Cl 3p character and Ti−Cl bond distance ( R 2 =0.90; Figure )…”

Oxidizable Ketones: Persistent Radical Cations from the Single‐Electron Oxidation of 2,3‐Diaminocyclopropenones.

“…These experiments strongly support the proposal that the Ce IV complex 7 is in equilibrium with aC e III species and the free DACO radical 4a.A lthough metal-oxygen homolytic ligand dissociation has also been seen in aminoxyl complexes, [19] we believe that this is the first example of the reversible homolytic dissociation of aneutral oxygen ligand to afree radical cation. This unusual behavior can be attributed to the unique structure of the cyclopropenone,w hich is ap roficient l-type donor but with as table one-electron oxidation state that can be accessed by strongly oxidizing metals such as Ce IV .…”

“…These experiments strongly support the proposal that the Ce IV complex 7 is in equilibrium with aC e III species and the free DACO radical 4a.A lthough metal-oxygen homolytic ligand dissociation has also been seen in aminoxyl complexes, [19] we believe that this is the first example of the reversible homolytic dissociation of aneutral oxygen ligand to showing the emergenceo fDACO 5a.b)UV/Vis spectrum of 5a (red), 5a plus 2equiv Ce(NO 3 ) 3 (4a) 3 9 (yellow), and 5a plus 4equiv 9 (blue). These measurements were taken in CH 2 Cl 2 solvent.…”

Quantifying the Interdependence of Metal–Ligand Covalency and Bond Distance Using Ligand K‐edge XAS

“…As imilarly poor regression was obtained when the sum of the covalent radii for NiÀCl (2.26 )a nd PdÀCl (2.41 )w ere used instead of ionic radii ( Figure S5). [15] Unlike PK -edge studies of M À Pb onds,w hich are more commonly used to assess electronic structure as it relates to MO energies and oxidation state, [17] there is asignificant trove of reported Cl K-edge XAS data with resolved and quantified M À Cl pre-edge peak areas.T his begs the question as to how our Cl K-edge XAS data compares and if XAS data reported previously correlates to M À Cl bond distance.T ob etter facilitate comparisons across data sets,w ec onverted our integrated Cl K-edge pre-edge peak areas into %Cl 3p character per MÀCl bond using the D 2d -CuCl 4 2À intensity standard described by Solomon et al [8b] Thed ata were then plotted against relative bond distances to account for differences in metal radii in complexes such as MCl 4 nÀ (where M = Fe 2+ ,F e 3+ ,C o 2+ ,N i 2+ ,C u 2+ ,P d 2+ ), [8a, 18] MCl 6 2À (where M = Ti 4+ and Pd 4+ ), [10a,18c] group IV cyclopentadienyl complexes, [19] Ti 4+ diphosphine complexes, [20] and others (see Supporting Information for details). Perhaps not surprising given the inherent uncertainties when using ionic radii that are highly dependent on both oxidation state and coordination number, no correlation between %Cl 3p character and relative M À Cl bond distance was observed ( Figure S6).…”

“…Perhaps not surprising given the inherent uncertainties when using ionic radii that are highly dependent on both oxidation state and coordination number, no correlation between %Cl 3p character and relative M À Cl bond distance was observed ( Figure S6). However,t he plot revealed that acorrelation does exist for group IV complexes, especially for the numerous Ti 4+ and Ti 3+ complexes measured by different research groups with diphosphine, [20] cyclopentadienyl, [19] and (2,2,6,6-tetramethylpiperidin-1yl)oxyl (TEMPO) ligands. [19b] Indeed, analysis of the Ti complexes revealed as trong correlation between %Cl 3p character and Ti À Cl bond distance (R 2 = 0.90;F igure 4).…”