Assembling Di- and Polynuclear Cu(I) Complexes with Rigid Thioxanthone-Based Ligands: Structures, Reactivity, and Photoluminescence

Abstract

Thioxanthone (TX) molecules and their derivatives are well-known photoactive compounds. Yet, there exist only a handful of luminescent systems combining TX with transition metals. Recently, we reported a TX-based PSP pincer ligand (L1) that appears as a promising platform for filling this niche. Herein, we demonstrate that with Cu(I) this ligand exclusively assembles into dimeric structures with either di- or polynuclear Cu(I) cores. With cationic Cu(I) precursors, complexes featuring solvent-bridged bis-cationic cores were obtained. These coordinatively unsaturated bimetallic systems showed surprisingly facile activation of the chloroform C-Cl bonds, suggesting a possible metal-metal cooperation. The reaction of L1 with binary Cu(I) halides afforded dimeric complexes with polynuclear [CuX]_n (n = 3 or 4) cores. With X = Br or I, emissive complexes containing stairstep [CuX]₄ clusters were obtained. Emission lifetimes in the microsecond range measured for these complexes were indicative of a triplet emission (phosphorescence), which according to our time-dependent density functional theory study originates from a halide-metal-to-ligand charge transfer between the [CuX]₄ cluster and the TX backbone of L1. Finally, the distinctive polynucleating behavior of L1 toward Cu(I) was also showcased by a comparison to another PSP ligand with a diaryl thioether backbone (L2), which formed only mononuclear pincer-type complexes, lacking any unusual reactivity or photoluminescence.

Figure 1

Rigid (a) and flexible (b) structural motifs capable of binding two metals in close proximity to each other.

Figure 2

Butterfly-like motion of TX (a) and coordination modes of a TX-based PSP ligand with various metals (b).

Scheme 1. Synthesis of Cu(I) Halide Complexes 1–3 with Tri- and Tetranuclear Cores

Figure 3

Possible coordination modes of a ligand L1 in its dimeric complexes: κ²-P,P (a), μ₂-P,P (b), and μ₂-(κ²-P,S-κ¹-P) (c).

Figure 4

Molecular structure of complexes 1 (a), 2 (b), and 3 (c). All H atoms and solvent molecules have been omitted for the sake of clarity.

Scheme 2. Synthesis of the Cationic Binuclear Copper Complexes 4–6

Figure 5

Molecular structure of complexes 5 (a) and 6 (b). H atoms, as well as BF₄^– counteranion and solvent molecules, are omitted for clarity.

Figure 6

Possible coordination modes of weakly coordinated solvent molecules to a bis-cationic dicopper(I) core.

Figure 7

Optimized structures of model dicopper complexes A, C, and D.

Figure 8

NBOs of the coordination interactions between a dicopper center and acetone in model complex D (a and b) and between dicopper and MeCN complex A (c) and contour-line maps of the electron density Laplacian in the Cu–O–Cu plane in complex D (d) and in the Cu–N–Cu plane in complex A (e).

Scheme 3. Previously Observed Coordination Modes of Ligands L2 and L2′ (a) and Reactions of L2 with Neutral and Cationic Cu(I) Precursors (b)

Figure 9

Molecular structures of complexes 7 (a) and 8 (b). H atoms, counteranions, and solvent molecules are omitted for clarity.

Figure 10

2D DOSY NMR spectra of complexes of 4a (a) and 8 (b) in MeCN at 25 °C.

Figure 11

Photographs under UV light of ligand L1 (a) and its complexes 2 and 3 (b and c, respectively), a comparison of their emission spectra at room temperature (d), and variable-temperature emission spectra of complexes 2 (e) and 3 (f) in the solid state (powder).

Figure 12

Optimized geometries of model complex 3* in the excited T₁ and ground S₀ electronic states overlaid (T₁ geometry is shown at the forefront) (a), their corresponding frontier molecular orbitals (b), and ESP plots (c). H atoms and phenyl rings are omitted for clarity.