Gating upconversion electroluminescence in a single molecule via adsorption-induced interaction of unpaired spin

Molecules with unpaired spins (radicals) offer promising alternatives to closed-shell molecules as they are less limited regarding the spin statistics in their electroluminescence. Here, we combine scanning tunneling microscopy induced luminescence and density functional theory to study single vanadyl phthalocyanine molecules, which are stable neutral radicals. Two distinct adsorption geometries of the molecule on NaCl/Au(111) lead to a difference in the interaction of the unpaired electron with the substrate, which in turn allows us to investigate its effects on the light emission process. Remarkably, we observe that up-conversion electroluminescence is gated by the adsorption geometry of the molecule, an effect we attribute to a reordering of excited states and enhanced excited state transition probabilities. The profound influence of the unpaired electron via state reordering opens new possibilities for tuning not only molecular electroluminescence but also many other spin dependent phenomena.

💡 Research Summary

The authors investigate how the interaction between an unpaired electron in a radical molecule and its substrate can gate up‑conversion electroluminescence (UCEL) at the single‑molecule level. They combine scanning tunneling microscope‑induced luminescence (STML) with density functional theory (DFT) and time‑dependent DFT (TD‑DFT) to study vanadyl phthalocyanine (VOPc), a stable neutral radical, adsorbed on three monolayers of NaCl on Au(111). Because the central oxygen atom of VOPc can point either toward the substrate (O‑down) or away from it (O‑up), two distinct adsorption geometries are realized, providing a natural “spin‑gate” that modifies the coupling of the molecule’s unpaired electron to the surface.

Both geometries display two emission bands in STML spectra: a higher‑energy Q band at ~1.80 eV, assigned to the neutral molecule’s D₃,₄ → D₀ transition, and a lower‑energy X⁺ band at ~1.45 eV, assigned to the positively charged molecule’s triplet transition. However, only the O‑down configuration exhibits strong UCEL: photons with energy exceeding the applied bias appear already at the onset of the positive‑ion resonance (PIR) around –1.35 V, whereas the O‑up configuration shows no such up‑conversion. The Q band in O‑down is markedly sharper (≈5.8 meV linewidth) than in O‑up (≈40 meV), indicating a longer excited‑state lifetime. Photon‑count versus current measurements reveal a near‑quadratic dependence (exponent α≈1.9) for O‑down, consistent with a two‑electron process, while O‑up shows a sub‑quadratic exponent, typical of single‑electron emission.

Differential conductance (dI/dV) spectra confirm that the PIR, which enables transient positive charging of the molecule, occurs at the same voltage for both geometries, but an additional resonance near –2.0 V appears only for O‑down, suggesting a geometry‑dependent electronic coupling that facilitates electron capture into higher excited states.

DFT calculations show that the frontier orbitals of the macrocycle are essentially unchanged between O‑up and O‑down, but the spin density of the unpaired electron is dramatically different: in O‑down the spin is closer to the substrate, leading to stronger hybridization. TD‑PBE0 and TD‑CHYF excitation spectra identify the Q band with doublet D₃,₄ → D₀ transitions and the X⁺ band with triplet transitions of the cation. Crucially, for O‑down a set of low‑energy “dark” triplet states (T₁–₁₂) with very low oscillator strength appears, which are absent in O‑up. Quadratic‑response TD‑DFT calculations reveal that these dark states have sizable transition dipoles for second‑order excitations from the D₁,₂ doublet states, providing an efficient pathway for population of long‑lived intermediate states.

The authors propose a many‑body mechanism: (1) at –1.35 V an electron tunnels from the molecule to the tip, creating the cationic triplet T⁺₀ (positive‑ion resonance); (2) an electron from the substrate is captured, populating the neutral doublet D₁,₂; (3) in O‑down only, the reordered triplet manifold allows D₁,₂ → T₁–₁₂ via charge‑injection, creating a long‑lived dark state; (4) a second electron capture promotes the system to D₃,₄; (5) radiative decay D₃,₄ → D₀ emits a photon with energy higher than the bias, completing the UCEL cycle. In O‑up the low‑energy triplet states are too high in energy, blocking step 3, so UCEL does not occur.

Thus, the adsorption geometry controls the spin‑dependent electronic structure, effectively gating up‑conversion electroluminescence. This work demonstrates that beyond molecular design, the molecule‑substrate interaction can be harnessed as a “spin gate” to manipulate multi‑electron excitation pathways, opening new routes for efficient single‑molecule light sources, spin‑controlled charge transport, and advanced nano‑optoelectronic devices.