STML: Scanning Tunneling Microscopy Induced Luminescence

• Scanning tunneling microscopy induced luminescence is a phenomenon where inelastic electron tunneling excites photon emission from nanoscale junctions via molecular and plasmonic pathways.
• It employs advanced spectroscopic and imaging techniques, including time-resolved measurements and plasmon–molecule hybridization, to reveal ultrafast energy and charge dynamics.
• Experimental setups integrate high-efficiency optical collection and quantum photonics methods to probe single-photon emissions and nanoscale exciton behavior.

Scanning tunneling microscopy induced luminescence (STML) refers to the phenomenon in which the tunneling current or voltage generated by a scanning tunneling microscope (STM) excites optical emission from a nanoscale junction. STML exploits the local chemical, electronic, and structural environment to probe and manipulate light emission mechanisms—including molecular fluorescence, vibronic transitions, plasmonic modes, excitonic recombination, and even single-photon emission—down to atomic and single-molecule spatial scales. The field encompasses foundational models of electronic tunneling, many-body quantum electrodynamics, plasmon–molecule hybridization, ultrafast charge and energy transfer, and advanced spectroscopic and imaging methodologies.

1. Theoretical Foundations and Mechanisms

STML originates from inelastic tunneling events in which electrons traversing the potential barrier between STM tip and sample transfer energy to the local environment, producing photon emission via two principal pathways:

• Electron-Induced Molecular Excitation: An electron tunnels and excites a molecular electronic transition (often coupled to vibrational degrees of freedom). The excited molecule relaxes radiatively, emitting a photon. Theoretical models typically use effective two-level or multilevel Hamiltonians, where the rate of excitation is governed by the inelastic tunneling matrix element—often approximated by a dipolar interaction Hamiltonian $$H_{\mathrm{el-m}} \sim -e\,\vec{\mu}\cdot\vec{r}/|\vec{r}|^3$$—and Franck–Condon physics for vibronic coupling (Dong et al., 2020, Wen et al., 2023).
• Plasmonic Excitation and Enhancement: The STM junction (metal tip and sample surface) forms a plasmonic nanocavity supporting localized surface plasmon resonances. Inelastic tunneling excites these plasmons, which can, in turn, transfer energy to a molecule (plasmon-assisted molecular emission) or emit light directly (plasmonic luminescence) (Miwa et al., 2012, Tian et al., 2010). The Purcell effect enhances spontaneous emission rates in such nanocavities.

Complexity arises when these mechanisms coexist and interact. Interference, hybridization, and energy exchange between electronic, vibrational, and plasmonic modes are rigorously described within frameworks such as density matrix formalism (Tian et al., 2010), nonequilibrium Green’s functions (NEGF) (Miwa et al., 2012, Miwa et al., 2013), and canonical transformations (e.g. Lang–Firsov) to handle exciton–vibron coupling.

The emission spectrum $I_\mathrm{em}(\omega)$ derived from these models often takes the form:

$$I_\mathrm{em}(\omega) = \frac{1}{\pi} \sum_{j,i} \frac{\sigma_{ji} \tau}{(\omega_{ji} - \omega)^2 + \tau^2}$$

where $\sigma_{ji}$ is the emission cross-section integrated over relaxation and energy-exchange contributions, and $\tau$ is the emission linewidth (Tian et al., 2010).

2. Exciton–Plasmon Coupling and Vibronic Structure

In molecular junctions, coupling between a discrete molecular exciton and a surface plasmon continuum leads to hybridized quasiparticles with properties of both exciton and plasmon (Miwa et al., 2012, Miwa et al., 2013). This exciton–plasmon coupling ($V$, typically appearing in Hamiltonians as $V(a d^\dagger + d a^\dagger)$) modifies the absorption and emission spectra:

• Enhancement and Suppression: Coupling introduces new excitation channels, shifts peak energies, and alters transition rates. The emission spectra exhibit both enhanced peaks (especially from vibronically excited states) and suppression (dips or dents) due to reabsorption or destructive interference.
• Vibronic Progression: When electron–vibration (vibronic) coupling is significant, as characterized by the Huang–Rhys factor $S$, the emission spectrum displays a progression of peaks corresponding to different vibrational quanta. The distribution of intensities is governed by the Franck–Condon envelope:

$$| \langle \nu | \nu' \rangle |^2 = e^{-S} \frac{S^{\nu'}}{\nu'!}$$

and can be extracted from second-derivative analysis of tunneling current versus bias voltage (Wen et al., 2023).

• Upconversion Luminescence: Vibrational excitations facilitate emission of photons with energy exceeding the applied bias, a haLLMark of upconversion enabled by energy stored in vibrational modes and transferred via hybrid channels (Miwa et al., 2013).

Interference between direct molecular absorption and plasmon reabsorption can be constructive or destructive, affecting the overall emission profile (Miwa et al., 2013). These effects are quantified through Green’s function formalisms and explicit calculation of spectral functions, e.g. $J_L(\omega) = -\Im[L^<(\omega)]/\pi$, for the molecule and $J_P(\omega)$ for the plasmon (Miwa et al., 2012).

3. Experimental Methodologies and Optical Collection

Advances in experimental STML hinge on the integration of highly efficient light collection systems:

• Optical Design: Plug-in off-axis parabolic mirrors and clip-on lens systems are implemented to maximize photon collection efficiency from the nanocavity. An off-axis parabolic mirror can, when optimally positioned, collect ~$72\%$ of the emission hemisphere, with overall system efficiency (including the spectrometer) near $50\%$ (Román et al., 2022). A clip-on lens, positioned within millimeters from the tunneling junction, can exploit nearly the entire numerical aperture; for NA = 0.3, this can approach $9\%$ collection efficiency for highly directional plasmonic emission (Cahlík et al., 5 May 2024).
• Spectroscopic Modalities: In addition to STML, systems may perform cathodoluminescence (STM-CL), in situ photoluminescence (PL), and Raman spectroscopy—often within the same cryogenic UHV platform (Román et al., 2022).
• Time-Resolved Techniques: Nanosecond or sub-nanosecond voltage pulses, synchronized with single-photon avalanche photodiodes, enable time-resolved STML (TR-STML), accessing single charge and exciton dynamics on nanometer and nanosecond scales (Rosławska et al., 2018, Rosławska et al., 2020). Gigahertz frame rate imaging further enables direct visualization of charge injection and exciton formation (Rosławska et al., 2020).

A summary table of common collection configurations:

Optics Type	Collection Efficiency (%)	Key Features
Off-axis parabolic mirror	~72 (hemisphere), ~50 sys	High etendue, high spectral resolution (Román et al., 2022)
Clip-on lens (NA=0.3)	~9 (directional)	Maximal use of NA, in-situ mounting (Cahlík et al., 5 May 2024)

4. STML in Functional Nanostructures: Spectral and Spatial Resolution

STML is applied to a broad spectrum of sample types, revealing localized and composition-sensitive emission phenomena:

• Single Molecule and Molecular Aggregate Emission: Atomically or sub-molecularly resolved photon maps correlate spectral features (e.g. vibronic peaks) with molecular symmetry modes, as demonstrated on phthalocyanines and polycyclic aromatic hydrocarbons (Doppagne et al., 2016, Kröger et al., 2018). Distance-dependent Fano line shapes in emission encode information about the interplay between molecular excitonic transitions and plasmon continua.
• Quantum Wells and Heterostructures: STML in InGaN/GaN quantum wells reveals Anderson-type carrier localization driven by alloy disorder, with nanometer-resolved emission lineshapes elucidating the local confining potential as predicted by landscape theory ($H u(\vec{r})=1$) and its reciprocal acting as an effective potential (Hahn et al., 2018, Sauty et al., 2022). Carrier diffusion lengths (~40 nm) and defect-induced spectral features emerge directly from spatially resolved STM-EL maps.
• 2D Materials and Van der Waals Heterostructures: In monolayer TMD systems (e.g. MoSe₂ on graphene/Au or p-doped WSe₂), STM tunneling current directly injects carriers, spawning excitonic (neutral and charged) emission with emission energies and linewidths modulated by local environment—such as moiré superlattices, strain, doping, and dielectric decoupling layers (López et al., 2022, Román et al., 2022). The quantum yield, emission ratios, and exciton binding energies are inferred from emission maps and correlated with nanoscale features.
• Activation and Modulation of Luminescence: With deft control over the STM tip and local environment, new light emission modes can be activated. For example, the fluorescence of Ni(II) complexes—normally “dark” due to ultrafast ISC—is switched on by STM-induced resonant energy transfer from a neighboring molecule, selectively crossing the ISC barrier only in desired local configurations (Hung et al., 2023).

5. Quantum Optics, Single-Photon Emission, and Dynamics

STML platforms offer unique access to quantum optical phenomena at the nanoscale:

• Single-Photon Emission: In C₆₀ thin films, the formation of tip-induced "split-off states" below the LUMO band, and consequent Coulomb blockade, result in the emission of single photons (antibunching with sub-200 ps 1/e lifetime), even in the absence of excitonic origin. Tight-binding models and HBT interferometry confirm the quantum nature and mechanism of these emissions (Leon et al., 2019).
• Time-Resolved Charge and Exciton Dynamics: The combination of pulsed STM and fast photodetectors allows direct measurement of single charge injection, exciton formation, and carrier/exciton diffusion dynamics via rate equations and four-dimensional (3D + time) mapping (Rosławska et al., 2018, Rosławska et al., 2020). The measured luminescence transients provide quantitative parameters such as $\langle T_\mathrm{h} \rangle$ (hole injection time), $T_\mathrm{ex}$ (exciton lifetime), and spatially-varying electron injection barriers $\Delta E(x)$.
• Fine Structure Resolution: Modulation techniques using alternating current (AC) bias exploit the Fourier decomposition of photon counting/current signals, enabling highly precise detection of molecular energy levels (and vibrational structure) at the "knee" points of the zero-frequency component's first derivative (Wen et al., 2021).

6. Advanced Analysis: Vibronic Coupling and Quantitative Extraction

STML provides a platform for the quantitative extraction of molecular parameters:

• Huang–Rhys Factor ($S$) Determination: Analysis of bias-dependent steps in dI/dV and peak intensities in d²I/dV² enables extraction of the Huang–Rhys factor directly from experimental data, quantifying electron–vibration coupling strength with the relation $[(d^2I/dV^2)_1/(d^2I/dV^2)_0] \approx S$ (Wen et al., 2023).
• Interference Analysis and Fano Line Shape Parameters: The Fano asymmetry parameter $q$ is tuned by tip–molecule distance in polycyclic aromatic hydrocarbon fluorescence, offering insight into direct versus indirect excitation strengths, and thus into microscopic coupling mechanisms (Kröger et al., 2018).
• Localization Landscape Analysis: The mapping of the effective confining potential via the landscape function $u(r)$ and estimation of localized state energies provides direct spectroscopic evidence for Anderson localization and the statistical dispersion of emission energies in disordered quantum wells (Hahn et al., 2018).

7. Implications and Future Prospects

STML stands as a uniquely powerful tool for nano-optoelectronic characterization and control:

• Device Design: Comprehensive theoretical models enable predictive design of molecular optoelectronic and photonic devices, with fine-tuning of emission via plasmonic coupling, vibrational structure, and local environment (Tian et al., 2010, Miwa et al., 2012).
• Nanoscale Sensing and Quantum Control: The exquisite spatial, spectral, and temporal resolution of STML facilitates chemical fingerprinting, quantum light sources, and the probing of energy transfer processes—from vibrational relaxation to resonant energy transfer and upconversion.
• Instrumentation: Ongoing developments in light collection optics (mirrors, clip-on lenses) and experimental geometries are key to maximizing photon yields and expanding the scope of STML applications (Román et al., 2022, Cahlík et al., 5 May 2024).

Current challenges include thermal drift in optical setups, alignment precision, environmental control (especially for 2D semiconductors), and the faithful theoretical modeling of hybrid many-body interactions. Prospects for future research involve the integration of advanced photon correlation measurements, active field-of-view scanning, and multiprobe spectroscopy modes (Román et al., 2022).

In summary, scanning tunneling microscopy induced luminescence provides an atomically precise window into the quantum electrodynamics of solids, molecules, and nanostructures, delivering deep insight into electronic, vibronic, and plasmonic phenomena and heralding applications in nanoscale photonics, spectroscopy, and quantum technologies.