The interplay between fluorescence and phosphorescence with luminescent gold(i) and gold(iii) complexes bearing heterocyclic arylacetylide ligands

Abstract

The photophysical properties of a series of gold(i) [LAu(C[triple bond, length as m-dash]CR)] (L = PCy₃ (1a-4a), RNC (5a), NHC (6a)) and gold(iii) complexes [Au(C^N^C)(C[triple bond, length as m-dash]CR)] (1b-4b) bearing heterocyclic arylacetylide ligands with narrow band-gap are compared. The luminescence of both series are derived from an intraligand transition localized on the arylacetylide ligand (ππ*(C[triple bond, length as m-dash]CR)) but 1a-3a displayed prompt fluorescence (τ _PF = 2.7-12.0 ns) while 1b-3b showed mainly phosphorescence (τ _Ph = 104-205 μs). The experimentally determined intersystem crossing (ISC) rate constants (k _ISC) are on the order of 10⁶ to 10⁸ s^-1 for the gold(i) series (1a-3a) but 10¹⁰ to 10¹¹ s^-1 for the gold(iii) analogues (1b-3b). DFT/TDDFT calculations have been performed to help understand the difference in the k _ISC between the two series of complexes. Owing to the different oxidation states of the gold ion, the Au(i) complexes have linear coordination geometry while the Au(iii) complexes are square planar. It was found from DFT/TDDFT calculations that due to this difference in coordination geometries, the energy gap between the singlet and triplet excited states (ΔE _ST) with effective spin-orbit coupling (SOC) for Au(i) systems is much larger than that for the Au(iii) counterparts, thus resulting in the poor ISC efficiency for the former. Time-resolved spectroscopies revealed a minor contribution (<2.9%) of a long-lived delayed fluorescence (DF) (τ _DF = 4.6-12.5 μs) to the total fluorescence in 1a-3a. Attempts have been made to elucidate the mechanism for the origins of the DF: the dependence of the DF intensity with the power of excitation light reveals that triplet-triplet annihilation (TTA) is the most probable mechanism for the DF of 1a while germinate electron-hole pair (GP) recombination accounts for the DF of 2a in 77 K glassy solution (MeOH/EtOH = 4 : 1). Both 4a and 4b contain a BODIPY moiety at the acetylide ligand and display only ¹IL(ππ*) fluorescence with negligible phosphorescence being observed. Computational analyses attributed this observation to the lack of low-lying triplet excited states that could have effective SOC with the S₁ excited state.

Fig. 1. Selected examples of transition-metal complexes that display dominant fluorescence instead of phosphorescence. Auxiliary ligands coordinated to the metal ions are omitted.

Chart 1. Au(i) and Au(iii) acetylide complexes studied in this work.

Fig. 2. Perspective drawings of the crystal structures of 1a (top left), 4a (top right) and 2b (bottom) with the thermal ellipsoids shown at 30% probability level. Hydrogen atoms have been omitted for clarity.

Fig. 3. UV-vis absorption spectra (left) and emission spectra (right) of 1a–4a (top) and 1b–4b (bottom) in CH₂Cl₂ at 298 K (2 × 10^–5 M).

Fig. 4. ns-TRE spectra of 1a recorded from (a) 0–42 ns and (b) after a time delay of 1 μs in degassed CH₂Cl₂ (5 × 10^–5 M) at 298 K. Inset shows the emission kinetic decay trace. Decay time constants were fitted as mono-exponential decay (λ _exc = 355 nm).

Fig. 5. ns-TRE spectra of (a) 1a and (b) 3a in 77 K glassy solution (EtOH/MeOH = 4 : 1) recorded at different time intervals. λ _exc = 355 nm; integration time: 80 and 200 μs for 1a and 3a, respectively. Insets of (a) and (b) show the kinetic decay traces at the specified wavelengths with the estimated phosphorescence lifetime (τ _phos).

Fig. 6. ns-TRE spectra of 2a in 77 K glassy solution (EtOH/MeOH = 4 : 1) recorded at different time intervals. λ _exc = 355 nm; integration time: 200 μs. Inset shows the log–log plot of emission intensity of DF (λ _DF = 500 nm) and phosphorescence (λ _phos = 596 nm) of 2a in 77 K glassy solution against time; both decay according to a power law: I ∝ t ^–1.

Fig. 7. Nanosecond transient absorption (ns-TA) difference spectra of 1a–3a recorded at selected decay times in degassed CH₂Cl₂ (5 × 10^–5 M) at 298 K. Insets show the ESA kinetic decay trace at the specified wavelengths; decay lifetimes were fitted as mono-exponential decay for 1a and bi-exponential decays for 2a and 3a. (λ _exc = 355 nm; integration time: 200 ns).

Fig. 8. (Top) fs-TRF spectra and (middle) fs-TA difference spectra of 1b–3b in CH₂Cl₂ (5 × 10^–5 M) at 298 K (λ _exc = 400 nm; 120 fs fwhm). Arrows indicate the spectral evolution. (Bottom) ns-TA difference spectra of 1b–3b in degassed CH₂Cl₂ (laser λ _exc: 355 nm). Insets show the kinetic time profiles and the decay time constants at the specified wavelengths.

Fig. 9. Frontier MOs of 1a and 1b at the optimized S₀ geometries. Orbital energies are also given in eV.

Fig. 10. Frontier MOs of 4a and 4b at their optimized S₀ geometries. Orbital energies are also given in eV.

Fig. 11. Illustration of the low-lying singlet and triplet excited states of Au(i) (left) and Au(iii) (right) complexes that accounts for the different photophysical behaviour of the Au(i) and Au(iii) complexes investigated in this work. S₁ and T₁ for both complexes are derived from HOMO → LUMO transitions; S₂ and T₄ excited states of 1a are derived from ^1,3[H–1 → LUMO] transitions while S₃ and T₅ excited states of 1b are derived from ^1,3[H–2 → LUMO] transitions. The d-orbitals involved in the T₂ of 1a and T₄ of 1b have the same orientations as their respective S₁ excited state (see Tables S5, S6 and S9 in ESI†). The wavy blue arrows indicate internal conversion (IC) from the T₅ to T₁ excited state. F = fluorescence and P = phosphorescence.

Fig. 12. Dependence of delayed fluorescence of 1a–3a in 5 × 10^–5 M degassed CH₂Cl₂ with excitation power intensity. Emission intensity measured after a time delay of 1 μs. (Laser λ _exc = 355 nm; 0.2–11 mJ per pulse; diameter = 8 mm, integration time: 800 μs.)