Scanning Tunneling Microscopy Data Research Articles

Chirality Detection in Scanning Tunneling Microscopy Data Using Artificial Intelligence.

Enantiospecific effects play an uprising role in chemistry and technical applications. Chiral molecular networks formed by self-assembly processes at surfaces can be imaged by scanning probe microscopy (SPM). Low contrast and high noise in the topography map often interfere with the automatic image analysis using classical methods. The long SPM image acquisition times restrain Artificial Intelligence-based methods requiring large training sets, leaving only tedious manual work, inducing human-dependent errors and biased labeling. By generating realistic looking synthetic images, the acquisition of real datasets is avoided. Two state-of-the-art object detection architectures are trained to localize and classify chiral unit-cells in a regular molecular chiral network formed by self-assembly of linear molecular bricks. The comparison of different architectures and datasets demonstrates that the training on purely synthetic data outperforms models trained using augmented datasets. A Faster R-CNN model trained solely on synthetic data achieved an excellent mean average precision of 99% on real data. Hence this approach and the transfer to real data show high success, also highlighting the high robustness against experimental noise and different zoom levels across the full experimentally reasonable parameter range. The generalizability of this idea is demonstrated by achieving equally high performance on a different structure,too.

Solution of the structure of the high-coverage CO layer on the Ru(0001) surface—A combined study by density functional theory and scanning tunneling microscopy

Structures formed by dense CO adsorption layers can provide information about the balance between molecule–surface and molecule–molecule interactions. However, in many cases, the structure models are not clear. Using density functional theory (DFT) and scanning tunneling microscopy (STM), we have investigated the high-coverage CO layer on the Ru(0001) surface. Previous investigations by low-energy electron diffraction (LEED) and vibrational spectroscopy led to conflicting results about the structure. In the present study, 88 models with coverages between 0.58 and 0.77 monolayers have been analyzed by DFT. The most stable structures consist of small, compact CO clusters with an internal pseudo 1×1 structure. The CO molecules in the cluster centers occupy on-top sites in an upright position, whereas the molecules farther outside are slightly shifted from these sites and tilted outward. STM data of the CO-saturated surface at low temperatures, corresponding to a coverage of 0.66 monolayers, show a quasi-hexagonal pattern of features with an internal hexagonal fine structure. Simulated images based on the cluster model agree with the experimental data. It is concluded that the high-coverage CO layer consists of the close-packed clusters predicted by DFT as the most stable structure elements. In the experiment, the sizes and shapes of the clusters vary. However, the arrangement is not random but follows defined tiling rules. The structure remains ordered, almost up to room temperature. The LEED data are re-interpreted on the basis of the Fourier transforms of the STM data, solving the long-standing conflict about the structure.

Optical manipulation of the charge-density-wave state in RbV3Sb5.

Broken time-reversal symmetry in the absence of spin order indicates the presence of unusual phases such as orbital magnetism and loop currents1-4. The recently discovered kagome superconductors AV3Sb5 (where A is K, Rb or Cs)5,6 display an exotic charge-density-wave (CDW) state andhave emerged as astrong candidate for materialshosting a loop currentphase. The idea that the CDW breaks time-reversal symmetry7-14 is, however, being intensely debated due to conflicting experimental data15-17. Here we use laser-coupled scanning tunnelling microscopy to study RbV3Sb5. By applying linearly polarized light along high-symmetry directions, we show that the relative intensities of the CDW peaks can be reversibly switched, implying a substantial electro-striction response, indicative of strong nonlinear electron-phonon coupling. A similar CDW intensity switching is observed with perpendicular magnetic fields, which implies an unusual piezo-magnetic response that, in turn, requires time-reversal symmetry breaking. We show that the simplest CDW that satisfies these constraints is an out-of-phase combination of bond charge order and loop currents that we dub a congruent CDW flux phase. Our laser scanning tunnelling microscopy data open the door to the possibility of dynamic optical control of complex quantum phenomenon in correlated materials.

2D Vanadium Sulfides: Synthesis, Atomic Structure Engineering, and Charge Density Waves.

Two ultimately thin vanadium-rich 2D materials based on VS2 are created via molecular beam epitaxy and investigated using scanning tunneling microscopy, X-ray photoemission spectroscopy, and density functional theory (DFT) calculations. The controlled synthesis of stoichiometric single-layer VS2 or either of the two vanadium-rich materials is achieved by varying the sample coverage and sulfur pressure during annealing. Through annealing of small stoichiometric single-layer VS2 islands without S pressure, S-vacancies spontaneously order in 1D arrays, giving rise to patterned adsorption. Via the comparison of DFT calculations with scanning tunneling microscopy data, the atomic structure of the S-depleted phase, with a stoichiometry of V4S7, is determined. By depositing larger amounts of vanadium and sulfur, which are subsequently annealed in a S-rich atmosphere, self-intercalated ultimately thin V5S8-derived layers are obtained, which host 2 × 2 V-layers between sheets of VS2. We provide atomic models for the thinnest V5S8-derived structures. Finally, we use scanning tunneling spectroscopy to investigate the charge density wave observed in the 2D V5S8-derived islands.

Machine-Learning driven STM images prediction of doped/defective graphene: Towards optimized tools for 2D nanomaterials characterization

Scanning tunneling microscopy (STM) is a key characterization technique that allows for visualization of specific features at nanostructures, given its ability to reveal changes in local charge densities in doping or defective sites. Besides the experimental acquisition of STM data from real samples, there is the theoretical route, which allows for STM images simulation from ab-initio calculations on defined nanostructures. This work presents an alternative route for theoretical STM image prediction by means of Machine Learning (ML) methods, based on the training of deep convolutional neural networks using simulated STM data. The ML-based STM predictions for defective, B- and N-doped graphene quantum dots are compared to ab-initio STM simulations, in terms of accuracy of structural rearrangements, charge density localizations, and computational resources requirements. The results demonstrate that STM images predicted by ML-methods are accurate at doping sites, whereas they show good similarity to simulations of large structural vacancies. This work represents a novel alternative for the use of ML methods in image-based characterization routines of doped/defective graphene, and potentially, for other 2-dimensional nanomaterials.

Automated Generation and Analysis of Molecular Images Using Generative Artificial Intelligence Models.

The development of scanning probe microscopy (SPM) has enabled unprecedented scientific discoveries through high-resolution imaging. Simulations and theoretical analysis of SPM images are equally important as obtaining experimental images since their comparisons provide fruitful understandings of the structures and physical properties of the investigated systems. So far, SPM image simulations are conventionally based on quantum mechanical theories, which can take several days in tasks of large-scale systems. Here, we have developed a scanning tunneling microscopy (STM) molecular image simulation and analysis framework based on a generative adversarial model, CycleGAN. It allows efficient translations between STM data and molecular models. Our CycleGAN-based framework introduces an approach for high-fidelity STM image simulation, outperforming traditional quantum mechanical methods in efficiency and accuracy. We envision that the integration of generative networks and high-resolution molecular imaging opens avenues in materials discovery relying on SPM technologies.

Highly Ordered Co-Assembly of Bisurea Functionalized Molecular Switches at the Solid-Liquid Interface.

Immobilization of stimulus-responsive systems on solid surfaces is beneficial for controlled signal transmission and adaptive behavior while allowing the characterization of the functional interface with high sensitivity and high spatial resolution. Positioning of the stimuli-responsive units with nanometer-scale precision across the adaptive surface remains one of the bottlenecks in the extraction of cooperative function. Nanoscale organization, cooperativity, and amplification remain key challenges in bridging the molecular and the macroscopic worlds. Here we report on the design, synthesis, and scanning tunneling microscopy (STM) characterization of overcrowded alkene photoswitches merged in self-assembled networks physisorbed at the solid-liquid interface. A detailed anchoring strategy that ensures appropriate orientation of the switches with respect to the solid surface through the use of bis-urea groups is presented. We implement a co-assembly strategy that enables the merging of the photoswitches within physisorbed monolayers of structurally similar 'spacer' molecules. The self-assembly of the individual components and the co-assemblies was examined in detail using (sub)molecular resolution STM which confirms the robust immobilization and controlled orientation of the photoswitches within the spacer monolayers. The experimental STM data is supported by detailed molecular mechanics (MM) simulations. Different designs of the switches and the spacers were investigated which allowed us to formulate guidelines that enable the precise organization of the photoswitches in crystalline physisorbed self-assembled molecular networks.

Self-Assembly of N-Rich Triimidazoles on Ag(111): Mixing the Pleasures and Pains of Epitaxy and Strain.

In the present report, homochiral hydrogen-bonded assemblies of heavily N-doped (C9H6N6) heterocyclic triimidazole (TT) molecules on an Ag(111) substrate were investigated using scanning tunneling microscopy (STM) and low energy electron diffraction (LEED) techniques. The planar and prochiral TT molecules, which exhibit a threefold rotation symmetry and lack mirror symmetry when assembled on the substrate, carry multiple hydrogen-bonding donor and acceptor functionalities, inevitably leading to the formation of hexameric two-dimensionally extended assemblies that can be either homo- (RR/SS) or heterochiral (RS). Experimental STM data showing well-ordered homochiral domains and experimental LEED data are consistent with simulations assuming the R19.1° overlayer on the Ag(111) lattice. Importantly, we report the unexpected coincidence of spontaneous resolution with the condensation of neighboring islands in adjacent "Janus pairs". The islands are connected by a characteristic fault zone, an observation that we discuss in the context of the fairly symmetric molecule and its propensity to compromise and benefit from interisland bonding at the expense of lattice mismatches and strain in the defect zone. We relate this to the close to triangular shape and the substantial but weak bonding scheme beyond van der Waals (vdW) of the TT molecules, which is due to the three N-containing five-membered imidazole rings. Density functional theory (DFT) calculations show clear energetic differences between homochiral and heterochiral pairwise interactions, clearly supporting the experimental results.

Formation and stability of Fe-rich terminations of the Fe3O4(001) surface

Understanding how the structure of iron oxide surfaces varies with their environment is essential for rationalizing their role in (geo-)chemistry and optimizing their application in modern technologies. In this paper, we create Fe-rich terminations of Fe3O4(001) by depositing iron directly onto the ‘subsurface cation vacancy’-reconstructed surface, which is the most stable surface under ultrahigh vacuum conditions. Scanning tunneling microscopy and x-ray photoelectron spectroscopy data reveal that the excess iron is initially accommodated as two-fold coordinated adatoms and later incorporates into the subsurface cation vacancies. As the coverage increases, small patches of the octahedral pair termination (also known as the ‘Fe dimer’ termination) nucleate, eventually covering the entire surface after the deposition of 2 iron atoms per (√2×√2)R45° unit cell. This conclusion effectively rules out some existing models for the termination and provides support for the model proposed by Rustad et al (Surface Science 432, L583-L588, 1999), highlighting the need for further theoretical work to complete the Fe3O4(001) surface phase diagram. The octahedral pair termination is found to be unstable above 523 K and upon exposure to molecular O2 because the excess iron atoms agglomerate to form small FeOx clusters.

Machine learning the microscopic form of nematic order in twisted double-bilayer graphene

Modern scanning probe techniques, such as scanning tunneling microscopy, provide access to a large amount of data encoding the underlying physics of quantum matter. In this work, we show how convolutional neural networks can be used to learn effective theoretical models from scanning tunneling microscopy data on correlated moiré superlattices. Moiré systems are particularly well suited for this task as their increased lattice constant provides access to intra-unit-cell physics, while their tunability allows for the collection of high-dimensional data sets from a single sample. Using electronic nematic order in twisted double-bilayer graphene as an example, we show that incorporating correlations between the local density of states at different energies allows convolutional neural networks not only to learn the microscopic nematic order parameter, but also to distinguish it from heterostrain. These results demonstrate that neural networks are a powerful method for investigating the microscopic details of correlated phenomena in moiré systems and beyond.

Deciphering Alloy Composition in Superconducting Single-Layer FeSe1-xSx on SrTiO3(001) Substrates by Machine Learning of STM/S Data.

Scanning tunneling microscopy (STM) is a powerful technique for imaging atomic structure and inferring information on local elemental composition, chemical bonding, and electronic excitations. However, a plain visual analysis of STM images can be challenging for such determination in multicomponent alloys, particularly beyond the diluted limit due to chemical disorder and electronic inhomogeneity. One viable solution is to use machine learning to analyze STM data and identify hidden patterns and correlations. Here, we apply this approach to determine the Se/S concentration in superconducting single-layer FeSe1-xSx alloys epitaxially grown on SrTiO3(001) substrates via molecular beam epitaxy. First, the K-means clustering method is applied to identify defect-related dI/dV tunneling spectra taken by current imaging tunneling spectroscopy. Then, the Se/S ratio is calculated by analyzing the remaining spectra based on the singular value decomposition method. Such analysis provides an efficient and reliable determination of alloy composition and further reveals the correlations of nanoscale chemical inhomogeneity to superconductivity in single-layer iron chalcogenide films.

Defect pairing in Fe-doped SnS van der Waals crystals: a photoemission and scanning tunneling microscopy study.

We investigate the effect of low concentrations of iron on the physical properties of SnS van der Waals crystals grown from the melt. By means of scanning tunneling microscopy (STM) and photoemission spectroscopy we study Fe-induced defects and observe an electron doping effect in the band structure of the native p-type SnS semiconductor. Atomically resolved and bias dependent STM data of characteristic defects are compared to ab initio density functional theory simulations of vacancy (VS and VSn), Fe substitutional (FeSn), and Fe interstitial (Feint) defects. While native SnS is dominated by acceptor-like VSn vacancies, our results show that Fe preferentially occupies donor-like interstitial Feint sites in close proximity to VSn defects along the high-symmetry c-axis of SnS. The formation of such well-defined coupled (VSn, Feint) defect pairs leads to local compensation of the acceptor-like character of VSn, which is in line with a reduction of p-type carrier concentrations observed in our Hall transport measurements.

Over- and underpotential deposition of copper from a deep eutectic solvent: Pt(1 1 1) single crystal versus polycrystalline Pt substrates

• Deposition/dissolution of Cu on Pt from ethaline occur in one-electron reactions. • A full (1x1) monolayer of Cu adatoms is formed on Pt(111) in ethaline. • No voltammetric signals of Cu UPD were observed for Pt(poly) in ethaline. • Surface structure determines the mechanism and rate of Cu OPD at the initial stages. • Cu OPD on Pt(111) and Pt(poly) in ethaline occurs via 2D and 3D growth, respectively. Deep eutectic solvents (DESs) attract ever greater interest as nonaqueous media for the electrodeposition of metals and alloys. This work investigates the initial stages of Cu electrodeposition from DES representing a mixture of choline chloride and ethylene glycol (ChCl:EG) with a (1:2) molar ratio that contains CuCl or CuCl 2 salts. These studies were performed by means of cyclic voltammetry and scanning probe microscopy on single-crystalline Pt(111) and polycrystalline Pt(poly) electrodes. Additional experiments on Cu deposition were carried out using an aqueous Cl - -rich solution (3 M NaCl + 0.01 M CuCl). The process of Cu underpotential deposition (UPD) on the surface of Pt(111) in DES is observed for the first time. It is proposed that a complete (1x1) monolayer of Cu adatoms is formed at potentials more positive than the onset of overpotential deposition (OPD). The comparison of results for Pt(111) and Pt(poly) in ChCl:EG (1:2) and in the aqueous 3 M NaCl solution confirms that the surface morphology and (co)adsorption of solution components play an important role in the Cu UPD process. In DES, as well as in the aqueous solutions containing CuCl, Cu OPD occurs via the reduction of Cu(I) ions. The Cu(I)/Cu(II) redox transition was observed at significantly more positive potentials. We conclude on the basis of voltammetry and in situ scanning tunneling microscopy (STM) data that the mechanism of Cu electrodeposition in DES is determined by the surface structure. STM data demonstrate the layer-by-layer growth of a two-dimensional Cu deposit on Pt(111), while the resulting Cu deposit on Pt(poly) is characterized by a grain-like morphology. We also demonstrate that the mechanism of Cu OPD affects the initial deposition rate under conditions of diffusion limitations.

Identifying the alloy structures of germanene grown on Al(111)

While the growth of germanene has been extensively claimed on many substrates, their exact crystal structures remain controversial. Herein, we systematically explore the surface structures of Ge deposition onto the Al(111) surface by theoretical calculations combined with available scanning tunneling microscopy (STM) and low-energy electron diffraction data. We show that the formation of germanene on Al(111) is energetically unfavorable by $ab\phantom{\rule{0.16em}{0ex}}initio$ evolutionary simulations and high-level random-phase approximation calculations, and the two experimental phases are identified as honeycomb alloys ${\mathrm{Al}}_{3}{\mathrm{Ge}}_{3}/\mathrm{Al}(\sqrt{7}\ifmmode\times\else\texttimes\fi{}\sqrt{7})$ and ${\mathrm{Al}}_{3}{\mathrm{Ge}}_{4}/\mathrm{Al}(3\ifmmode\times\else\texttimes\fi{}3)$. It is interesting to find a vacancy at the interface of ${\mathrm{Al}}_{3}{\mathrm{Ge}}_{4}/\mathrm{Al}(3\ifmmode\times\else\texttimes\fi{}3)$, which is responsible for the dark clover pattern in STM experiments. Our results clarify the structural controversy of the Ge/Al(111) system and indicate that the fabrication of germanene on Al(111) remains challenging.

A multiscale wavelet algorithm for atom tracking in STM movies

High-speed scanning tunneling microscopy (STM) data have become available that provide movies of time-dependent surface processes. To track adsorbed atoms and molecules in such data automatic routines are required. We introduce a multiresolution wavelet particle detection algorithm for this purpose. To identify the particles, the images are decomposed by means of a discrete wavelet transform into wavelet planes of different resolutions. An ‘à trous’ low-pass filter is applied. The coefficients from the wavelet planes are filtered to remove noise. Wavelet planes with significant coefficients from the particles are multiplied, and the product is transformed into a binary particle mask. The precision of the method is tested with data sets of adsorbed CO molecules and O atoms on a Ru(0001) surface. The algorithm can safely detect and localize these particles with high precision, even in the presence of the enhanced noise characteristic for high-speed, constant-height STM data. By linking the particle positions, we obtain extended trajectories with a resolution of ∼0.5 Å or better allowing us to investigate the detailed motion of single atoms on a surface.

Unraveling the Highly Complex Nature of Antimony on Pt(111)

AbstractUnderstanding the growth behavior of group‐V elements on metal surfaces provides valuable information that can shed light on the feasibility of tailoring atomically thin monoelemental 2D polymorphs composed of pnictogens on these metallic substrates. Here, by combining scanning tunneling microscopy (STM), low energy electron diffraction and Auger electron spectroscopy measurements under ultra‐high vacuum conditions, a wide variety of Sb reconstructions on single‐crystal Pt(111) are identified and characterized. At Sb coverage of ≈0.2 ML, STM data are compatible with a scenario in which Sb atoms are randomly embedded into the topmost layer of Pt, in a substitutional configuration and without establishing a periodic structure. This disorder is robust against thermal annealing and quenching. Increasing the surface coverage and whether or not the sample is annealed, different well‐ordered Sb phases are formed. The Sb structures synthesized at room temperature without any heating process are best interpreted as surface alloys that involve only the first atomic layer. In contrast, experimental evidence points towards the development of multilayer alloy phases for the annealed samples.

Automatic indexing of two-dimensional patterns in reciprocal space

An indispensable part of the structure determination of crystalline two-dimensional (2D) materials and epitaxial thin films is the correct indexing of the acquired diffraction patterns. In our previous work, we described an effective algorithm to determine the 3D unit-cell parameters of complex systems comprising different orientations and polymorphs. In this work, we adapt the indexing method to 2D lattices in reciprocal space. Analyzing low-energy electron diffraction and Fourier-transformed scanning tunneling microscopy measurements, the method is exemplarily applied to thin films of conjugated molecules like 3,4:9,10-perylenetetracarboxylic dianhydride (PTCDA), 6,13-pentacenequinone (P2O), and vanadyl phthalocyanine (VOPc) grown by physical vapor deposition on Ag(111). In all cases unit cells (rhomboids) along with their sixfold rotationally or mirror symmetric counterparts are determined. The already known commensurate epitaxial relationship is reproduced for PTCDA on Ag(111), demonstrating the validity of our method. In the case of P2O/Ag(111) a point-on-line epitaxial condition is found. Our algorithm can be equally well applied to all kinds of 2D patterns in reciprocal space where a crystallographic indexing is required, e.g., electron diffraction data [such as transmission electron diffraction, selected area electron diffraction (SAED)] and fast Fourier transforms (FFTs) of scanning probe images. To demonstrate this aspect, we evaluate FFTs of scanning tunneling microscopy data for stacked VOPc/PTCDA heteroepitaxial layers on Ag(111) as well as SAED data of an epitaxial ${\mathrm{TiO}}_{2}/{\mathrm{LaAlO}}_{3}(100)$ heterostructure in cross section.

Tunneling and break junction spectroscopy of the ambient-pressure semiconducting and superconducting gap structures in the ladder compound (Sr, Ca)14Cu24O41

The ladder-chain compound ${(\mathrm{Sr}, \mathrm{Ca})}_{14}{\mathrm{Cu}}_{24}{\mathrm{O}}_{41}$ is a semiconductor at ambient pressure, but becomes a bulk superconductor above the pressure of about 3 GPa. Since the compound is at the verge of the metal-insulator transition, it is reasonable to make tunnel measurements to probe the electronic density of states in its subtleties. We present the results of such measurements carried out at ambient pressure. The break junction (BJ) tunneling gives evidence for the apparent typical gap $2\mathrm{\ensuremath{\Sigma}}$ of about 140 meV at temperature $T$, equal to 4 K. The gap is smeared out at ${T}^{*}\ensuremath{\approx}90--100$ K. We interpret the gap as that induced by the charge density wave (CDW) formation, although its $T$-dependence differs from the usually observed CDW-like mean-field behavior. Quite unexpectedly, BJ spectra also exhibit distinct zero-bias peak accompanied by the low-energy gaps of $2\mathrm{\ensuremath{\Delta}}\ensuremath{\approx}4--8$ meV immediately after BJ are formed at 4 K. The thermally driven disappearance of this apparently superconducting structure at ${T}_{c}\ensuremath{\approx}7--13$ K is consistent with the conventional properties of superconducting tunnel junctions. The resultant ratio $\mathrm{\ensuremath{\Delta}}/{T}_{c}$ is consistent with similar observed values for a high-${T}_{c}$ cuprate superconductor. Therefore, we attribute this feature as well as the Josephson-like zero-bias conductance peak to the superconductivity induced at the freshly created BJ surface of the ${(\mathrm{Sr}, \mathrm{Ca})}_{14}{\mathrm{Cu}}_{24}{\mathrm{O}}_{41}$, which is semiconducting in the sample bulk. Our additional scanning tunneling microscopy data testify that the cracked atomic surface of this compound is substantially modified, which might be the reason for the superconductivity appearance.

Specific Features of Statistical Analysis of Relative Grain Boundary Energies Obtained by Scanning Tunneling Microscopy

An approach based on statistical analysis is proposed for the processing of data obtained using scanning tunneling microscopy of grain boundaries, which allows a numerical estimation of the relative energy of grain boundaries. The proposed statistical model also gives a possibility to separate groups of grain boundaries depending on their average relative energy and fraction in general distribution. Scanning tunneling microscopy data analyses have been carried out on data obtained by investigating copper and nanostructured copper were analyzed coarse-grain commercially pure copper and on copper nanostructured by the equal-channel angular pressing (ECAP) method. Obtained results were compared with available in literature experimental data for these types of materials, received by other methods. It is established that the grain boundaries in coarse-grain copper have significantly lower relative energy in contrast to the grain boundaries of ECAP-treated copper. Besides, there is, except for boundaries with high relative energy, a fraction of boundaries in the deformed sample with energy corresponding to those in coarse-grain copper.

Determining amplitude and tilt of a lateral force microscopy sensor.

In lateral force microscopy (LFM), implemented as frequency-modulation atomic force microscopy, the tip oscillates parallel to the surface. Existing amplitude calibration methods are not applicable for mechanically excited LFM sensors at low temperature. Moreover, a slight angular offset of the oscillation direction (tilt) has a significant influence on the acquired data. To determine the amplitude and tilt we make use of the scanning tunneling microscopy (STM) channel and acquire data without and with oscillation of the tip above a local surface feature. We use a full two-dimensional current map of the STM data without oscillation to simulate data for a given amplitude and tilt. Finally, the amplitude and tilt are determined by fitting the simulation output to the data with oscillation.