Electron transfer rate modulation with mid-IR in butadiyne-bridged donor-bridge-acceptor compounds.

Controlling electron transfer (ET) processes in donor-bridge-acceptor (DBA) compounds by mid-IR excitation can enhance our understanding of the ET dynamics and may find practical applications in molecular sensing and molecular-scale electronics. Alkyne moieties are attractive to serve as ET bridges, as they offer the possibility of fast ET and present convenient vibrational modes to perturb the ET dynamics. Yet, these bridges introduce complexity because of the strong torsion angle dependence of the ET rates and transition dipoles among electronic states and a shallow torsion barrier. In this study, we implemented ultrafast 3-pulse laser spectroscopy to investigate how the ET from the dimethyl aniline (D) electron donor to the N -isopropyl-1,8-napthalimide (NAP) electron acceptor can be altered by exciting the CC stretching mode ( ν CC ) of the butadiyne bridge linking the donor and acceptor. The electron transfer was initiated by electronically exciting the acceptor moiety at 400 nm, followed by vibrational excitation of the alkyne, ν CC , and detecting the changes in the absorption spectrum in the visible spectral region. The experiments were performed at different delay times t 1 and t 2 , which are the delays between UV-mid-IR and mid-IR-Vis pulses, respectively. Two sets of torsion-angle conformers were identified, one featuring a very fast mean ET time of 0.63 ps (group A) and another featuring a slower mean ET time of 4.3 ps (group B), in the absence of the mid-IR excitation. TD-DFT calculations were performed to determine key torsion angle dependent molecular parameters, including the electronic and vibrational transition dipoles, transition frequencies, and electronic couplings. To describe the 3-pulse data, we developed a kinetic model that includes a locally excited, acceptor-based S2 state, a charge separated S1 state, and their vibrationally excited counterparts, with either excited ν CC (denoted as S1A tr , S1B tr , S2A tr , and S2B tr , where tr stands for the excited triplet bond, ν CC ) or excited daughter modes of the ν CC relaxation (S1A h , S1B h , S2A h , and S2B h , where h stands for vibrationally hot species). The kinetic model was solved analytically, and the species-associated spectra (SAS) were determined numerically using a matrix approach, treating first the experiments with longer t 1 delays and then using the already determined SAS for modeling the experiments with shorter t 1 delays. Strong vibronic coupling of ν CC and of vibrationally hot states makes the analysis complicated. Nevertheless, the SAS were identified and the ET rates of the vibrationally excited species, S2A tr , S2B tr and S2B h , were determined. The results show that the ET rate for the S2A species is ca. 1.2-fold slower when the ν CC mode is excited. The ET rate for species S2B is slower by ca. 1.3-fold if the compound is vibrationally hot and is essentially unchanged when the ν CC mode is excited. The SAS determined for the tr and h species resemble the SAS for their respective precursor species in the 2-pulse transient absorption experiments, which validates the procedure used and the results.