Nano-optoelectronics aims to manipulate optical and electrical properties at the nanoscale, leveraging the optoelectronic properties of individual molecules for applications like sensors and LEDs. Single-molecule photoswitching, where molecular geometry or conductance is reversibly switched by light, is crucial. Conventional switches often rely on photochromic molecules, limiting flexibility. Localized surface plasmons (LSPs) offer a route to sub-nanoscale reaction control, enhancing photochemical reactions and enabling reaction pathways inaccessible via far-field excitation. While plasmon-induced photochemistry has been demonstrated in metallic structures, its application to semiconductors remains largely unexplored. This study explores metal-molecule-semiconductor junctions as a platform for LSP-induced single-molecule optoelectronics, addressing the challenge of achieving efficient field enhancement on a non-plasmonic semiconductor substrate like silicon. The authors propose that the significant variety of functional molecule-semiconductor systems could enable the miniaturization of optoelectronic devices and the acquisition of novel functions through effective light-matter interactions.

Previous work on single-molecule photoswitching has focused on using photochromic molecules like azobenzene and spiropyran derivatives in contact with electrodes. Studies on metal-molecule-semiconductor nanojunctions have primarily focused on the photovoltaic effect under far-field irradiation. Research using scanning tunneling microscopy (STM) with laser excitation has characterized plasmon-induced reactions in real space and achieved controlled photochemistry at the submolecular scale. However, plasmon-mediated chemical reactions on semiconductors have not been extensively characterized, particularly at the single-molecule level. The use of LSPs in nanojunctions, particularly those involving semiconductors, offers a novel avenue for achieving highly precise control over single-molecule switching behavior.

The researchers used a low-temperature STM with laser excitation to study single-molecule switching of perylene 3,4,9,10-tetracarboxylic dianhydride (PTCDA) and its derivatives (PMI and PDI) on a Si(111)-7 × 7 surface. Sharp Ag tips, fabricated using focused ion beam (FIB) milling, were used to create the plasmonic nanojunction. The switching was monitored via STM current (I_STM) and tip-enhanced Raman scattering (TERS). Density functional theory (DFT) calculations were used to understand the molecular adsorption and geometry. The tip height was precisely controlled to investigate the influence on the switching rate. Control experiments were performed to elucidate the mechanism, including varying laser power and using Ag tips with different plasmon resonance profiles. Finite element method (FEM) simulations were used to model the plasmonic field enhancement in the junction. The study also involved characterization of the adsorption of PTCDA, PMI, and PDI molecules on the Si surface, with DFT calculations used to determine their adsorption energies.

Illumination of the single-molecule junction at the LSP-resonant wavelength induced dissociation of O-Si bonds between the PTCDA molecule and the Si surface, resulting in reversible switching between two configurations (ON and OFF states). The STM current exhibited telegraph noise, characteristic of a two-state switching system. TERS measurements showed that the ON state is associated with a point contact formation between the Ag tip and the molecule after O-Si bond breakage. The switching rate was controlled with 0.1-Å precision by adjusting the tip height. The chemical substituents on the perylene core (PTCDA vs. PMI vs. PDI) significantly influenced switching behavior. Anhydride groups showed switching activity while imide groups were inert. The switching mechanism is attributed to hot carriers (HCs) generated by LSP excitation in the Ag tip and transferred to the molecule, enabling the O-Si bond dissociation. The laser power dependence supported the HC-mediated mechanism. The non-contribution of bias voltage supports this and suggests that the plasmon-enhanced electric field, rather than the direct bias voltage, drives the reaction. FEM simulations confirmed the limited contribution of the Si substrate to plasmon enhancement, highlighting the role of the Ag tip in controlling the reaction.

The findings demonstrate atomic-scale control over plasmon-induced single-molecule switching in a metal-semiconductor nanojunction. The ability to precisely control the switching rate by adjusting the tip height with sub-angstrom precision highlights the potential for creating highly controllable optoelectronic devices. The chemical sensitivity demonstrated by the different switching behaviors of PTCDA, PMI, and PDI illustrates the importance of molecular design in tailoring plasmon-driven processes. The hot-carrier transfer mechanism proposed provides a plausible explanation for the observed switching behavior. The findings are significant for advancing the field of nano-optoelectronics and open new possibilities for developing highly miniaturized, efficient, and highly controllable optoelectronic devices. The ability to selectively activate specific chemical moieties within the molecule provides pathways for developing sophisticated molecular-scale circuitry.

This work successfully demonstrated atomic-precision control of plasmon-induced single-molecule switching in a metal-semiconductor nanojunction. The precise control over switching rate through tip height manipulation, and the ability to tailor reactivity via chemical substitution, highlight the potential of this approach for developing advanced nano-optoelectronic devices. Future research could focus on exploring other molecule-semiconductor combinations and investigating the potential for integrating these switches into more complex nanoscale circuits.

The short duration of the ON state and the related difficulty in obtaining detailed information about the ON-to-OFF transition hinder a complete understanding of the reaction mechanism. While the study focuses on specific molecules and substrates, further investigation is needed to determine the generalizability of the findings to other systems. The complexity of the experimental setup and the required expertise in STM-TERS techniques may limit widespread adoption of this methodology.