🔬 materials science

Rationalizing charge transfer: a single-molecule case study across three distinct surfaces

This study combines Kelvin probe force microscopy, scanning tunneling spectroscopy, density functional theory, and model Hamiltonian calculations to demonstrate that isolated macrocyclic emeraldine molecules donate one electron to Au(111) and FeCl $_2$ substrates but two electrons to a superconducting NbSe $_2$ surface, thereby establishing a robust methodology for quantifying charge transfer across diverse material interfaces.

Original authors: Benjamin Lowe, Jan Berger, Martin Ondráček, Oleksandr Stetsovych, Shayan Louie, Marco Lozano, Diego Soler-Polo, Shiladitya Karmarkar, Andrés Pinar Solé, Iva Plutnarová, Pingo Mutombo, Manish Kumar, Mi

Published 2026-10-26

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Benjamin Lowe, Jan Berger, Martin Ondráček, Oleksandr Stetsovych, Shayan Louie, Marco Lozano, Diego Soler-Polo, Shiladitya Karmarkar, Andrés Pinar Solé, Iva Plutnarová, Pingo Mutombo, Manish Kumar, Michael Píro, Ali Cem Demirok, Michael L. Steigerwald, Zdeněk Sofer, Colin Nuckolls, Pavel Jelínek

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine building a complex machine not by assembling pre-made gears, but by arranging individual atoms on a surface to form new materials with custom properties. This is the promise of "on-surface synthesis," a field where scientists place organic molecules onto solid surfaces to create tiny, one-dimensional wires or two-dimensional networks. For years, this work has been almost exclusively done on shiny, noble metals like gold. These metals are excellent at conducting electricity and helping molecules stick together, but they are also very reactive. To unlock new technologies, such as building magnetic switches or superconducting circuits, researchers need to move these delicate molecular structures onto other types of surfaces: insulators that don't conduct electricity, magnets that have their own internal order, or superconductors that carry current without resistance. However, moving a molecule from a metal surface to a different material changes how it behaves. The most critical change involves the flow of electric charge. When a molecule lands on a surface, electrons may jump from the molecule to the surface, or vice versa, depending on how tightly the surface holds onto its own electrons. Understanding exactly how many electrons move, and why, is essential for designing these future materials, yet measuring this tiny exchange has proven difficult.

In a recent study, researchers tackled this challenge by placing a single, ring-shaped organic molecule called emeraldine onto three very different surfaces: a standard gold crystal, a thin layer of magnetic iron chloride, and a superconducting crystal made of niobium and selenium. The team wanted to see if the molecule would give up one electron, two electrons, or none at all, and how the nature of the surface dictated this behavior. They used a suite of advanced tools, including microscopes that can see individual atoms and measure their electrical properties, alongside computer simulations to interpret what they saw. The results revealed a clear pattern: the molecule acted as a donor, giving up electrons to the surface, but the number of electrons it gave away depended entirely on the surface's ability to hold onto them. On the gold surface and the magnetic iron chloride layer, the molecule donated exactly one electron. On the superconducting surface, it donated two. This difference, though seemingly small, fundamentally alters the molecule's magnetic and electronic character, turning it from a neutral object into a charged one with specific magnetic properties.

To reach these conclusions, the scientists first prepared their samples with extreme care. They created a surface that was half gold and half covered with a single layer of iron chloride, then gently deposited the ring-shaped molecules onto it. They also prepared a separate sample of the superconducting crystal. Using a scanning tunneling microscope, which can map the surface with atomic precision, they confirmed that the molecules landed alone, isolated from one another, rather than clumping together. This isolation was crucial, as it allowed them to study the behavior of a single molecule without interference from its neighbors. They then used a technique called Kelvin probe force microscopy to measure the electrical potential of the surfaces. This measurement helped them determine the "work function" of each material, a property that describes how much energy is required to remove an electron from the surface. They found that the gold surface had a work function of about 5.25 electron volts, while the superconducting surface was higher at 5.9 electron volts. The magnetic iron chloride layer fell in between, at roughly 5.35 electron volts.

The researchers then looked at how the molecules responded electrically. By scanning the molecules with a voltage, they could see which energy levels were occupied by electrons. On the gold surface, the data showed a distinct signature indicating that the molecule had lost one electron. This loss created a state where the molecule held an unpaired electron, giving it a magnetic spin. On the metallic gold, this magnetic spin interacted with the sea of electrons in the metal, creating a specific signal known as a Kondo effect, which appeared as a sharp peak in their measurements at zero voltage. This confirmed that the molecule had indeed donated a single electron. On the superconducting surface, the picture was different. The electrical signals showed that the molecule had lost two electrons, leaving its highest energy orbital completely empty. This meant the molecule had donated a double charge to the surface. The magnetic iron chloride surface presented a more complex puzzle. The electrical signals suggested the molecule had lost an electron, but the presence of the insulating iron chloride layer made the measurement noisy and difficult to interpret directly.

To solve the ambiguity on the magnetic surface, the team combined their experimental data with a theoretical model that accounted for how the surface screens, or shields, the electric charge of the molecule. They realized that the insulating layer of iron chloride reduced the screening effect compared to the bare metal, which would normally make it harder for the molecule to give up a second electron. When they applied this model to their measurements, it strongly supported the conclusion that the molecule on the magnetic surface had also donated just one electron, matching the behavior on the gold. The study demonstrated that a relatively small difference in the surface's work function—about 0.65 electron volts between the gold and the superconductor—was enough to trigger a jump from donating one electron to donating two. This finding is significant because it shows that scientists can control the charge and magnetic state of a molecule simply by choosing the right surface to place it on.

The work also highlighted the limitations of relying on a single method to understand these systems. The researchers found that standard computer simulations alone were not enough to accurately predict the exact number of electrons transferred, as they often struggle to account for the subtle ways surfaces screen electric charges. Similarly, looking only at the electrical spectra without the context of the surface's work function could lead to incorrect conclusions. By combining direct measurements of the surface properties, high-resolution imaging, and a refined theoretical model, the team was able to build a complete picture of the charge transfer process. Their approach provides a new toolkit for future researchers who want to engineer molecular materials on diverse surfaces, from insulators to magnets. It proves that by carefully selecting the substrate, one can tune the fundamental properties of a molecule, opening the door to creating new types of quantum materials with tailored magnetic and electronic behaviors. The study does not claim to have solved all the mysteries of molecule-surface interactions, but it offers a clear, quantified method for understanding how charge moves at the atomic scale, a necessary step for the next generation of molecular electronics.

Technical Summary: Rationalizing Charge Transfer in a Single-Molecule Case Study

Problem Statement
Organic molecules serve as versatile building blocks for bottom-up fabrication of low-dimensional quantum materials. While on-surface synthesis has successfully produced complex magnetic and conductive architectures on noble metal surfaces, translating these protocols to insulating, magnetic, or superconducting surfaces remains a challenge. A critical factor governing the electronic properties of adsorbates on these alternative surfaces is charge transfer. The extent of electron transfer between a molecule and a substrate depends on the alignment of the molecule's ionization potential (IP) and electron affinity (EA) with the substrate's work function ( $\Phi$ ). However, quantifying this charge transfer is difficult; standard Scanning Tunneling Microscopy/Spectroscopy (STM/STS) and Density Functional Theory (DFT) are often insufficient to unambiguously determine integer charge states, particularly when screening effects and hybridization complicate the interpretation of spectral features.

Methodology
The authors conducted a systematic study of isolated macrocyclic emeraldine (Me6) molecules adsorbed on three distinct surfaces:

Metallic Au(111): A standard noble metal surface.
Ferromagnetic Semiconducting Monolayer (FeCl2/Au(111)): An insulating magnetic layer grown on Au(111).
Superconducting 2H-NbSe2: A bulk superconductor cleaved under vacuum.

The experimental approach combined multiple scanning probe microscopy techniques:

STM and STS: Used to map the spatial distribution of electronic states ( $dI/dV$ ) and identify resonances near the Fermi level.
Non-contact Atomic Force Microscopy (ncAFM): Performed with CO-terminated tips to resolve molecular geometry and adsorption height.
Kelvin Probe Force Microscopy (KPFM): Utilized to measure contact potential differences ( $V_{CPD}$ ) and analyze the electrostatic interaction between the tip and the charged molecule at varying tip-sample distances.

Theoretical support included:

DFT Calculations: To optimize molecular geometries on each surface and calculate projected density of states (DOS).
Model Hamiltonian Calculations: A many-body model incorporating static image-charge polarization ( $\Delta \hat{H}_{im}$ ) and dynamic electron exchange ( $\hat{H}_{hyb}$ ) between the molecule and a simplified 1D metallic chain representing the substrate. This model was used to simulate charge transfer as a function of work function changes.

Key Results

Structural Properties: ncAFM and DFT confirmed that Me6 molecules physisorb on all three surfaces in a non-planar, three-fold symmetric configuration. Adsorption heights were consistent with weak coupling (physisorption), with values of $\sim4.2$ Å on Au(111) and slightly larger on FeCl2 and NbSe2.
Electronic States and Charge Transfer:
- Au(111): STS revealed a zero-bias resonance (Kondo peak) and symmetric features at $\pm 0.2$ V. Spatial maps matched the simulated gas-phase Positive Ion Resonance (PIR), indicating a molecular charge of +1 (one electron transferred to the substrate). The half-filled HOMO splits into SOMO and SUMO orbitals separated by Coulomb repulsion ( $U$ ).
- NbSe2: STS showed features exclusively at positive bias ( $\sim 0.45$ V) and negative bias ( $\sim -0.77$ V), with spatial distributions corresponding to the PIR and PIR-1/-2 orbitals, respectively. This indicates a molecular charge of +2 (two electrons transferred), resulting in a completely depopulated gas-phase HOMO.
- FeCl2: STS showed a charging feature at $\sim -1.2$ V (attributed to a double-barrier tunnel junction effect) and a molecular state at $+0.35$ V resembling the PIR. While the positive bias feature suggested charge transfer, the negative bias region was obscured by the charging peak, leaving the charge state ambiguous via STS alone.
Quantification via KPFM and Modeling:
- KPFM measurements on FeCl2 revealed a work function ( $\Phi$ ) of $\sim 5.35$ eV, intermediate between Au(111) ( $\sim 5.25$ eV) and NbSe2 ( $\sim 5.9$ eV).
- Analysis of $V_{CPD}$ trends with tip-sample distance, combined with a model accounting for screening lengths, quantitatively supported a charge state of +1 for Me6 on FeCl2. The model fit for $Q_M = +2$ failed to converge with realistic parameters.
- The Model Hamiltonian demonstrated that image-charge renormalization significantly lowers the energy barrier for electron transfer. It showed that a work function increase of only $\sim 0.5$ eV is sufficient to transition the molecule from a +1 to a +2 state, consistent with the experimental difference between Au(111) and NbSe2 ( $\sim 0.65$ eV).

Key Contributions

Methodological Advancement: The study demonstrates that STS and DFT alone are insufficient for quantifying charge transfer in these systems. The authors establish a robust methodology combining KPFM measurements with model Hamiltonian calculations to unambiguously determine integer charge states across different substrates.
Charge State Control: The work provides experimental evidence that a relatively small change in substrate work function ( $\sim 0.65$ eV) can induce a transition between distinct integer molecular charge states (+1 to +2).
Role of Screening: The research highlights the crucial role of image-charge screening in governing charge transfer processes. Theoretical modeling confirms that electrostatic renormalization, rather than just energy level alignment, dictates the stability of specific charge states.

Significance
The paper claims to provide essential experimental and theoretical tools for rationalizing and quantifying charge transfer on diverse surfaces. By establishing a link between substrate work function, image-charge screening, and molecular charge state, the work offers a pathway for engineering the electronic and magnetic properties of surface-supported molecular architectures. This is particularly relevant for the future development of low-dimensional quantum materials on insulating, magnetic, and superconducting surfaces, where controlling the spin and charge of adsorbates is a prerequisite for realizing correlated phases and functional devices. The authors modestly note that while they identified a local magnetic moment on Au(111) via the Kondo effect, the interaction of this moment with the ferromagnetic FeCl2 surface remains an open question for future investigation.

Technical Summary: Rationalizing Charge Transfer in a Single-Molecule Case Study

More like this