Mechanisms and applications of cyclometalated Pt(ii) complexes in photoredox catalytic trifluoromethylation

Abstract

The incorporation of a trifluoromethyl group into an existing scaffold can provide an effective strategy for designing new drugs and agrochemicals. Among the numerous approaches to trifluoromethylation, radical trifluoromethylation mediated by visible light-driven photoredox catalysis has gathered significant interest as it offers unique opportunities for circumventing the drawbacks encountered in conventional methods. A limited understanding of the mechanism and molecular parameters that control the catalytic actions has hampered the full utilization of photoredox catalysis reactions. To address this challenge, we evaluated and investigated the photoredox catalytic trifluoromethylation reaction using a series of cyclometalated Pt(ii) complexes with systematically varied ligand structures. The Pt(ii) complexes were capable of catalyzing the trifluoromethylation of non-prefunctionalized alkenes and heteroarenes in the presence of CF₃I under visible light irradiation. The high excited-state redox potentials of the complexes permitted oxidative quenching during the cycle, whereas reductive quenching was forbidden. Spectroscopic measurements, including time-resolved photoluminescence and laser flash photolysis, were performed to identify the catalytic intermediates and directly monitor their conversions. The mechanistic studies provide compelling evidence that the catalytic cycle selects the oxidative quenching pathway. We also found that electron transfer during each step of the cycle strictly adhered to the Marcus normal region behaviors. The results are fully supported by additional experiments, including photoinduced ESR spectroscopy, spectroelectrochemical measurements, and quantum chemical calculations based on time-dependent density functional theory. Finally, quantum yields exceeding 100% strongly suggest that radical propagation significantly contributes to the catalytic trifluoromethylation reaction. These findings establish molecular strategies for designing trifluoromethyl sources and catalysts in an effort to enhance catalysis performance.

Scheme 1. Photoredox catalytic trifluoromethylation of alkenes and heteroarenes.

Fig. 1. (a) UV-vis absorption spectra of the Pt(ii) complexes (10 μM in acetonitrile). (b) Photoluminescence decay traces of the Pt(ii) complexes (50 μM in deaerated acetonitrile) after nanosecond pulsed photoexcitation at 377 nm: λ _obs = 465 nm (Ptdfppy), 483 nm (Ptppy), and 543 nm (PtOMe).

Fig. 2. Photoredox catalytic trifluoromethylation of 1-dodecene (circles) and N-methylpyrrole (triangles) by Ptdfppy (blue), Ptppy (green), and PtOMe (red). Empty and filled symbols correspond to the substrates and the trifluoromethylated products, respectively. Conditions: 2.0 mL of a deaerated acetonitrile solution containing 0.50 mmol substrate, 0.010 mmol Pt catalyst, 1.0 mmol TMEDA (or 1.0 mmol DBU), and 1.5 mmol CF₃I was photoirradiated under blue LEDs (450 nm, 7 W) at room temperature. The progress of the reaction was monitored using gas chromatography, with dodecane as the internal standard.

Scheme 2. Proposed mechanism for the trifluoromethylation of alkenes and heteroarenes.

Fig. 3. Determining the rate constants for photoinduced electron transfer (k _PeT). (a) Phosphorescence decay traces of 50 μM PtOMe (deaerated CH₃CN; λ _obs = 543 nm) over the range of CF₃I concentrations (0–20 mM). (b) Plot of the electron transfer rates (1/τ _obs(CF₃I) – 1/τ _obs, τ _obs(CF₃I) and τ _obs are the phosphorescence lifetimes of the Pt(ii) complexes in the presence and absence of CF₃I, respectively) as a function of the concentration of CF₃I. The k _PeT values were determined based on the pseudo-first order fits of the electron transfer rates. Standard deviations were determined from three independent measurements.

Fig. 4. Determination of the rate constant for back electron transfer (k _BeT) by nanosecond laser flash photolysis (λ _ex = 355 nm) for a deaerated acetonitrile solution containing 100 μM PtOMe (O.D. at 355 nm = 0.52) and 50 mM CF₃I. (a) Transient absorption spectra recorded at 4 μs (red), 8 μs (blue), and 252 μs (black) after photoexcitation. The dotted grey line and grey bars are the simulated (CPCM(CH₃CN)-TD-B3LYP/LANL2DZ:6-311+G(d,p)//B3LYP/LANL2DZ:6-311+G(d,p)) absorption spectrum and calculated absorbance, respectively, of the one-electron oxidized species of PtOMe. (b) Plot of ε/absorbance of the 770 nm band vs. time. The k _BeT value was determined from the second-order linear fit (red). The results obtained from other Pt(ii) complexes are shown in the ESI, Fig. S8.

Fig. 5. Determination of the rate constants for the regeneration (k _regen) of the Pt(ii) complex catalysts. (a) Decay traces of the 770 nm absorption band of the one-electron oxidized PtOMe (100 μM in deaerated CH₃CN, O.D. at 355 nm = 0.52) in the presence (50 μM) or absence of TMEDA after nanosecond pulsed photoexcitation at 355 nm. (b) Plots of the regeneration rate (1/τ(TMEDA) – 1/τ, where τ(TMEDA) and τ are the half-lives of the 770 nm traces in the presence and absence of TMEDA, respectively) vs. the concentration of TMEDA. The k _regen values were determined according to the single exponential curve fitting.

Fig. 6. Plots of the log(rate constant) (log k _eT) vs. the driving force for oxidative photoinduced electron transfer (k _PeT, filled triangles), back electron transfer (k _BeT, empty triangles), and regeneration of the Pt(ii) complex (k _regen, filled circles) at 298 K. Curves show the theoretical plots of eqn (1) at λ = 1.2 eV (grey curve) and 2.7 eV (black curve).