Antenna-coupled infrared nanospectroscopy of intramolecular vibrational interaction

Abstract

Many photonic and electronic molecular properties, as well as chemical and biochemical reactivities are controlled by fast intramolecular vibrational energy redistribution (IVR). This fundamental ultrafast process limits coherence time in applications from photochemistry to single quantum level control. While time-resolved multidimensional IR-spectroscopy can resolve the underlying vibrational interaction dynamics, as a nonlinear optical technique it has been challenging to extend its sensitivity to probe small molecular ensembles, achieve nanoscale spatial resolution, and control intramolecular dynamics. Here, we demonstrate a concept how mode-selective coupling of vibrational resonances to IR nanoantennas can reveal intramolecular vibrational energy transfer. In time-resolved infrared vibrational nanospectroscopy, we measure the Purcell-enhanced decrease of vibrational lifetimes of molecular vibrations while tuning the IR nanoantenna across coupled vibrations. At the example of a Re-carbonyl complex monolayer, we derive an IVR rate of (25±8) cm^-1 corresponding to (450±150) fs, as is typical for the fast initial equilibration between symmetric and antisymmetric carbonyl vibrations. We model the enhancement of the cross-vibrational relaxation based on intrinsic intramolecular coupling and extrinsic antenna-enhanced vibrational energy relaxation. The model further suggests an anti-Purcell effect based on antenna and laser-field-driven vibrational mode interference which can counteract IVR-induced relaxation. Nanooptical spectroscopy of antenna-coupled vibrational dynamics thus provides for an approach to probe intramolecular vibrational dynamics with a perspective for vibrational coherent control of small molecular ensembles.

Keywords: intramolecular vibrational redistribution; light–matter interaction; nanospectroscopy.

Fig. 1.

Purcell-enhanced IR antenna–coupled vibrational spectroscopy. (A) Experimental design of femtosecond IR nanospectroscopy of monolayer Re–carbonyl complex covalently linked to IR gold nanoantenna resonators. (B) Symmetric A′(1), and antisymmetric stretch vibrations A′(2) and A′′, of the Re complex (41). (C) Grazing incidence far-field FTIR absorption spectrum with corresponding mode assignment. (D) Near-field amplitude A_NF spectra (fitted) of selected infrared nanowire antennas of variable length, with an example laser spectrum.

Fig. 2.

IR nanoimaging and vibrational free-induction decay of coupled molecular vibrations. (A) AFM topography of nanoantenna (Top, l = 1,400 nm), imaginary part of two-phase heterodyne nano-IR imaging (Middle), and FDTD calculated E-field amplitude of the antenna–monolayer–tip system (Bottom). (B) Corresponding interferogram measured at the right antenna terminal (Top, black) with model fit (magenta), and decomposed (Bottom) into the driving laser field (gray), the antenna response (blue), and vibrational response (red) of A′(1) mode in this case, and with characteristic distinct lifetime and different relative phases. (C) Corresponding Fourier-transform of time-domain signals in (B). The spectral minimum near the vibrational resonance position is the result of antenna and vibrational interference (Fano resonance).

Fig. 3.

Antenna-coupled vibrational free-induction decay and IVR. Interferometric nano-IR time traces at a fixed laser excitation frequency of ${\overline{\nu }}_{\mathrm{Laser}}=$ 2,020 cm⁻¹, for antennas of variable lengths, with (A) 1,400 nm (${\overline{\nu }}_{\mathrm{ant}}=$ 2,022 cm⁻¹), (B) 1,440 nm antenna (${\overline{\nu }}_{\mathrm{ant}}=$ 1,978 cm⁻¹), and (C) 1,520 nm antenna (${\overline{\nu }}_{\mathrm{ant}}=$ 1,937 cm⁻¹). (D) Extracted T_FID decay times of the A′(1) vibrational mode when tuning the antenna frequency. (E) Amplitude of the A′(1) vibrational signal normalized on the antenna signal. (F) Phase difference between the antenna and the A′(1) signal.

Fig. 4.

(A) Vibrational dephasing time T_{FID, A′(1)} of the symmetric A′(1) mode subject to laser driving at 2,020 cm⁻¹, as a function of antenna resonance frequency ω_ant. Experiment (circles) and analytical results with (solid blue line) and without (solid gray line) anti-Purcell effect. For fixed antenna and molecular parameter of γ_S = 14 cm⁻¹, ${\overline{\nu }}_{\mathrm{A}\prime (1)}$ = 2,028 cm⁻¹, γ_A = 30 cm⁻¹, ${\overline{\nu }}_{\mathrm{A}\prime (\mathrm{E})}$ = 1,919 cm⁻¹, κ_ant = 80 cm⁻¹, g_S = 20 cm⁻¹, g_A = 2.7g_S, and η = 0.5, a vibration–vibration coupling rate ζ_SA = 25 ± 8 cm⁻¹ is derived. (B) Evolution of T_FID time of the effective A′(E) vibrational mode for laser driving at 1,919 cm⁻¹.

Fig. 5.

Model T_{FID, A′(1)} as a function of antenna frequency ${\overline{\nu }}_{\mathrm{Ant}}$, obtained from 1/T_FID⁰ + 1/2T_{1, P} + 1/2T_{1, IVR − P}. (A) For varying ratios of the oscillator strengths 0 ≤ g_A/g_S ≤ 4.0, for fixed vibration–vibration coupling ζ_SA = 25 cm⁻¹, g_S = 20 cm⁻¹ and weighting anti-Purcell factor η = 0. (B) For varying 0 ≤ ζ_SA < 40 cm⁻¹, with fixed ratio g_A/g_S = 3.3 and η = 0. (C) For varying 0 ≤ η < 0.6, g_A/g_S = 2.7 and ζ_SA = 25 cm⁻¹. The red marked lines correspond to the parameters with the best fit to the experimental data in Fig. 4A.