Efficient Electron Hopping Transport through Azurin-Based Junctions.

We conducted a theoretical study of electron transport through junctions of the blue-copper azurin from Pseudomonas aeruginosa. We found that single-site hopping can lead to either higher or lower current values compared to fully coherent transport. This depends on the structural details of the junctions as well as the alignment of the protein orbitals. Moreover, we show how the asymmetry of the IV curves can be affected by the position of the tip in the junction and that, under specific conditions, such a hopping mechanism is consistent with a fairly low temperature dependence of the current. Finally, we show that increasing the number of hopping sites leads to higher hopping currents. Our findings, from fully quantum calculations, provide deep insight to help guide the interpretation of experimental IV curves on highly complex systems.

Molecular Identification from AFM Images Using the IUPAC Nomenclature and Attribute Multimodal Recurrent Neural Networks.

Spectroscopic methods─like nuclear magnetic resonance, mass spectrometry, X-ray diffraction, and UV/visible spectroscopies─applied to molecular ensembles have so far been the workhorse for molecular identification. Here, we propose a radically different chemical characterization approach, based on the ability of noncontact atomic force microscopy with metal tips functionalized with a CO molecule at the tip apex (referred as HR-AFM) to resolve the internal structure of individual molecules. Our work demonstrates that a stack of constant-height HR-AFM images carries enough chemical information for a complete identification (structure and composition) of quasiplanar organic molecules, and that this information can be retrieved using machine learning techniques that are able to disentangle the contribution of chemical composition, bond topology, and internal torsion of the molecule to the HR-AFM contrast. In particular, we exploit multimodal recurrent neural networks (M-RNN) that combine convolutional neural networks for image analysis and recurrent neural networks to deal with language processing, to formulate the molecular identification as an imaging captioning problem. The algorithm is trained using a data set─which contains almost 700,000 molecules and 165 million theoretical AFM images─to produce as final output the IUPAC name of the imaged molecule. Our extensive test with theoretical images and a few experimental ones shows the potential of deep learning algorithms in the automatic identification of molecular compounds by AFM. This achievement supports the development of on-surface synthesis and overcomes some limitations of spectroscopic methods in traditional solution-based synthesis.

Experimental Data Confirm Carrier-Cascade Model for Solid-State Conductance across Proteins.

The finding that electronic conductance across ultrathin protein films between metallic electrodes remains nearly constant from room temperature to just a few degrees Kelvin has posed a challenge. We show that a model based on a generalized Landauer formula explains the nearly constant conductance and predicts an Arrhenius-like dependence for low temperatures. A critical aspect of the model is that the relevant activation energy for conductance is either the difference between the HOMO and HOMO-1 or the LUMO+1 and LUMO energies instead of the HOMO-LUMO gap of the proteins. Analysis of experimental data confirms the Arrhenius-like law and allows us to extract the activation energies. We then calculate the energy differences with advanced DFT methods for proteins used in the experiments. Our main result is that the experimental and theoretical activation energies for these three different proteins and three differently prepared solid-state junctions match nearly perfectly, implying the mechanism's validity.

Revisiting Vibrational Spectroscopy to Tackle the Chemistry of Zr6O8 Metal-Organic Framework Nodes.

The metal-organic framework MOF-808 contains Zr6O8 nodes with a high density of vacancy sites, which can incorporate carboxylate-containing functional groups to tune chemical reactivity. Although the postsynthetic methods to modify the chemistry of the Zr6O8 nodes in MOFs are well known, tackling these alterations from a structural perspective is still a challenge. We have combined infrared spectroscopy experiments and first-principles calculations to identify the presence of node vacancies accessible for chemical modifications within the MOF-808. We demonstrate the potential of our approach to assess the decoration of MOF-808 nodes with different catechol-benzoate ligands. Furthermore, we have applied advanced synchrotron characterization tools, such as pair distribution function analyses and X-ray absorption spectroscopy, to resolve the atomic structure of single metal sites incorporated into the catechol groups postsynthetically. Finally, we demonstrate the catalytic activity of these MOF-808 materials decorated with single copper sites for 1,3-dipolar cycloadditions.

QUAM-AFM: A Free Database for Molecular Identification by Atomic Force Microscopy.

This paper introduces Quasar Science Resources-Autonomous University of Madrid atomic force microscopy image data set (QUAM-AFM), the largest data set of simulated atomic force microscopy (AFM) images generated from a selection of 685,513 molecules that span the most relevant bonding structures and chemical species in organic chemistry. QUAM-AFM contains, for each molecule, 24 3D image stacks, each consisting of constant-height images simulated for 10 tip-sample distances with a different combination of AFM operational parameters, resulting in a total of 165 million images with a resolution of 256 × 256 pixels. The 3D stacks are especially appropriate to tackle the goal of the chemical identification within AFM experiments by using deep learning techniques. The data provided for each molecule include, besides a set of AFM images, ball-and-stick depictions, IUPAC names, chemical formulas, atomic coordinates, and map of atom heights. In order to simplify the use of the collection as a source of information, we have developed a graphical user interface that allows the search for structures by CID number, IUPAC name, or chemical formula.

Hydrogen bonded trimesic acid networks on Cu(111) reveal how basic chemical properties are imprinted in HR-AFM images.

High resolution non-contact atomic force microscopy measurements characterize assemblies of trimesic acid molecules on Cu(111) and the link group interactions, providing the first fingerprints utilizing CO-based probes for this widely studied paradigm for hydrogen bond driven molecular self assembly. The enhanced submolecular resolution offered by this technique uniquely reveals key aspects of the competing interactions. Accurate comparison between full-density-based modeled images and experiment allows to identify key structural elements in the assembly in terms of the electron-withdrawing character of the carboxylic groups, interactions of those groups with Cu atoms in the surface, and the valence electron density in the intermolecular region of the hydrogen bonds. This study of trimesic acid assemblies on Cu(111) combining high resolution atomic force microscopy measurements with theory and simulation forges clear connections between fundamental chemical properties of molecules and key features imprinted in force images with submolecular resolution.

A Deep Learning Approach for Molecular Classification Based on AFM Images.

In spite of the unprecedented resolution provided by non-contact atomic force microscopy (AFM) with CO-functionalized and advances in the interpretation of the observed contrast, the unambiguous identification of molecular systems solely based on AFM images, without any prior information, remains an open problem. This work presents a first step towards the automatic classification of AFM experimental images by a deep learning model trained essentially with a theoretically generated dataset. We analyze the limitations of two standard models for pattern recognition when applied to AFM image classification and develop a model with the optimal depth to provide accurate results and to retain the ability to generalize. We show that a variational autoencoder (VAE) provides a very efficient way to incorporate, from very few experimental images, characteristic features into the training set that assure a high accuracy in the classification of both theoretical and experimental images.

Surface-controlled reversal of the selectivity of halogen bonds.

Intermolecular halogen bonds are ideally suited for designing new molecular assemblies because of their strong directionality and the possibility of tuning the interactions by using different types of halogens or molecular moieties. Due to these unique properties of the halogen bonds, numerous areas of application have recently been identified and are still emerging. Here, we present an approach for controlling the 2D self-assembly process of organic molecules by adsorption to reactive vs. inert metal surfaces. Therewith, the order of halogen bond strengths that is known from gas phase or liquids can be reversed. Our approach relies on adjusting the molecular charge distribution, i.e., the σ-hole, by molecule-substrate interactions. The polarizability of the halogen and the reactiveness of the metal substrate are serving as control parameters. Our results establish the surface as a control knob for tuning molecular assemblies by reversing the selectivity of bonding sites, which is interesting for future applications.

Impact of Small Adsorbates in the Vibrational Spectra of Mg- and Zn-MOF-74 Revealed by First-Principles Calculations.

In this work, we analyze the influence of small adsorbates on the vibrational spectra of Mg- and Zn-metal-organic framework MOF-74 by means of first-principles calculations. In particular, we consider the adsorption of four representative species of different interaction strengths: Ar, CO2, H2O, and NH3. Apart from a comprehensive characterization of the structural and energetic aspects of empty and loaded MOFs, we use a fully quantum ab initio approach to evaluate the Raman and IR activities of the normal modes, leading to the construction of the whole vibrational spectra. Under this approach, not only are we able to proceed with the complete assignment of the spectra in terms of the usual internal coordinates but also we can discern the most relevant vibrational fingerprints of the adsorbates and their impact on the whole MOF spectra. On the one hand, some of the typical vibrational modes of the small molecules are slightly shifted but still visible when adsorbed on the MOFs, especially those appearing at high wavenumbers where the empty MOFs lack IR/Raman signals. On the other hand, some bands arising from the organic ligands are affected by the presence of the absorbates, displaying non-negligible frequency shifts, in agreement with recent experiments. We find a strong correlation between all of these frequency shifts and the interaction strength of the adsorbate with the hosting framework. The findings presented in this work expand the capabilities of vibrational spectroscopy techniques to analyze porous materials and can be useful for the design of sensors and new devices based on MOF technology.

Mechanical Deformation and Electronic Structure of a Blue Copper Azurin in a Solid-State Junction.

Protein-based electronics is an emerging field which has attracted considerable attention over the past decade. Here, we present a theoretical study of the formation and electronic structure of a metal-protein-metal junction based on the blue-copper azurin from pseudomonas aeruginosa. We focus on the case in which the protein is adsorbed on a gold surface and is contacted, at the opposite side, to an STM (Scanning Tunneling Microscopy) tip by spontaneous attachment. This has been simulated through a combination of molecular dynamics and density functional theory. We find that the attachment to the tip induces structural changes in the protein which, however, do not affect the overall electronic properties of the protein. Indeed, only changes in certain residues are observed, whereas the electronic structure of the Cu-centered complex remains unaltered, as does the total density of states of the whole protein.

A comprehensive study of the molecular vibrations in solid-state benzylic amide [2]catenane.

The interpretation of vibrational spectra is often complex but a detailed knowledge of the normal modes responsible for the experimental bands provides valuable information about the molecular structure of the sample. In this work we record and assign in detail the infrared (IR) spectrum of the benzylic amide [2]catenane, a complex molecular solid displaying crimped mechanical bonds like the links of a chain. In spite of the large size of the unit cell, we calculate all the vibrational modes of the catenane crystal using quantum first-principles calculations. The activity of each mode is also evaluated using the Born effective charges approach and a theoretical spectrum is constructed for comparison purposes. We find a remarkable agreement between the calculations and the experimental results without the need to apply any further empirical correction or fitting to the eigenfrequencies. A detailed description in terms of the usual internal coordinates is provided for over 1000 normal modes. This thorough analysis allows us to perform the complete assignment of the spectrum, revealing the nature of the most active modes responsible for the IR features. Finally, we compare the obtained results with those of Raman spectroscopy, studying the effects of the rule of mutual exclusion in vibrational spectroscopy according to the different levels of molecular symmetry embedded in this mechanically interlocked molecular compound.

A Solid-State Protein Junction Serves as a Bias-Induced Current Switch.

A sample-type protein monolayer, that can be a stepping stone to practical devices, can behave as an electrically driven switch. This feat is achieved using a redox protein, cytochrome C (CytC), with its heme shielded from direct contact with the solid-state electrodes. Ab initio DFT calculations, carried out on the CytC-Au structure, show that the coupling of the heme, the origin of the protein frontier orbitals, to the electrodes is sufficiently weak to prevent Fermi level pinning. Thus, external bias can bring these orbitals in and out of resonance with the electrode. Using a cytochrome C mutant for direct S-Au bonding, approximately 80 % of the Au-CytC-Au junctions show at greater than 0.5 V bias a clear conductance peak, consistent with resonant tunneling. The on-off change persists up to room temperature, demonstrating reversible, bias-controlled switching of a protein ensemble, which, with its built-in redundancy, provides a realistic path to protein-based bioelectronics.

High-accuracy large-scale DFT calculations using localized orbitals in complex electronic systems: the case of graphene-metal interfaces.

Over many years, computational simulations based on density functional theory (DFT) have been used extensively to study many different materials at the atomic scale. However, its application is restricted by system size, leaving a number of interesting systems without a high-accuracy quantum description. In this work, we calculate the electronic and structural properties of a graphene-metal system significantly larger than in previous plane-wave calculations with the same accuracy. For this task we use a localised basis set with the Conquest code, both in their primitive, pseudo-atomic orbital form, and using a recent multi-site approach. This multi-site scheme allows us to maintain accuracy while saving computational time and memory requirements, even in our exemplar complex system of graphene grown on Rh(1 1 1) with and without intercalated atomic oxygen. This system offers a rich scenario that will serve as a benchmark, demonstrating that highly accurate simulations in cells with over 3000 atoms are feasible with modest computational resources.

Ab initio electronic structure calculations of entire blue copper azurins.

We present a theoretical study of the blue-copper azurin extracted from Pseudomonas aeruginosa and several of its single amino acid mutants. For the first time, we consider the whole structure of this kind of protein rather than limiting our analysis to the copper complex only. This is accomplished by combining fully ab initio calculations based on density functional theory with atomic-scale molecular dynamics simulations. Beyond the main features arising from the copper complex, our study reveals the role played by the peripheral parts of the proteins. In particular, we find that oxygen atoms belonging to carboxyl groups which are distributed all over the protein contribute to electronic states near the HOMO. The contribution of the outer regions to the electronic structure of azurins had so far been overlooked. Our results highlight the need to investigate them thoroughly; this is especially important in prospect of understanding complex processes such as the electronic transport through metal-metalloprotein-metal junctions.

Substrate-induced enhancement of the chemical reactivity in metal-supported graphene.

Graphene is commonly regarded as an inert material. However, it is well known that the presence of defects or substitutional hetero-atoms confers graphene promising catalytic properties. In this work, we use first-principles calculations to show that it is also possible to enhance the chemical reactivity of a graphene layer by simply growing it on an appropriate substrate. Our comprehensive study demonstrates that, in strongly interacting substrates like Rh(111), graphene adopts highly rippled structures that exhibit areas with distinctive chemical behaviors. According to the local coupling with the substrate, we find areas with markedly different adsorption, dissociation and diffusion pathways for both molecular and atomic oxygen, including a significant change in the nature of the adsorbed molecular and dissociated states, and a dramatic reduction (∼60%) of the O2 dissociation energy barrier with respect to free-standing graphene. Our results show that the graphene-metal interaction represents an additional and powerful handle to tailor the graphene chemical properties with potential applications to nano patterning, graphene functionalization and sensing devices.

Unveiling the atomistic mechanisms for oxygen intercalation in a strongly interacting graphene-metal interface.

The atomistic mechanisms involved in the oxygen (O) intercalation in the strongly interacting graphene (G) on Rh(111) system are characterized in a comprehensive experimental and theoretical study, combining scanning tunneling microscopy and density functional theory (DFT) calculations. Experimental evidence points out that the G areas located just above the metallic steps of the substrate are the active sites for initializing the intercalation process when some micro-etching points appear after molecular oxygen gas exposure. These regions are responsible for both the dissociation of the oxygen molecules and the subsequent penetration to the G-metal interface. Unlike in other species, the DFT calculations exclude single-point defects as additional entry paths to the interface. After penetration, the intercalation proceeds inwards due to the high mobility of atomic oxygen at the interface following mid-height paths connecting the higher areas of the rippled graphene structure. At larger coverages, the accumulation of O atoms under the high areas increases the G-metal distance in the neighboring low areas, paving the way for the O incorporation and the G detachment that leads to the final O-(2 × 1) structure. Furthermore, our results show that these mechanisms are possible only at temperatures slightly lower than those in which graphene etching takes place.

Detecting Mechanochemical Atropisomerization within an STM Break Junction.

We have employed the scanning tunneling microscope break-junction technique to investigate the single-molecule conductance of a family of 5,15-diaryl porphyrins bearing thioacetyl (SAc) or methylsulfide (SMe) binding groups at the ortho position of the phenyl rings (S2 compounds). These ortho substituents lead to two atropisomers, cis and trans, for each compound, which do not interconvert in solution under ambient conditions; even at high temperatures, isomerization takes several hours (half-life 15 h at 140 °C for SAc in C2Cl4D2). All the S2 compounds exhibit two conductance groups, and comparison with a monothiolated (S1) compound shows the higher group arises from a direct Au-porphyrin interaction. The lower conductance group is associated with the S-to-S pathway. When the binding group is SMe, the difference in junction length distribution reflects the difference in S-S distance (0.3 nm) between the two isomers. In the case of SAc, there are no significant differences between the plateau length distributions of the two isomers, and both show maximal stretching distances well exceeding their calculated junction lengths. Contact deformation accounts for part of the extra length, but the results indicate that cis-to-trans conversion takes place in the junction for the cis isomer. The barrier to atropisomerization is lower than the strength of the thiolate Au-S and Au-Au bonds, but higher than that of the Au-SMe bond, which explains why the strain in the junction only induces isomerization in the SAc compound.

Purely substitutional nitrogen on graphene/Pt(111) unveiled by STM and first principles calculations.

Nitrogen doping of graphene can be an efficient way of tuning its pristine electronic properties. Several techniques have been used to introduce nitrogen atoms on graphene layers. The main problem in most of them is the formation of a variety of C-N species that produce different electronic and structural changes on the 2D layer. Here we report on a method to obtain purely substitutional nitrogen on graphene on Pt(111) surfaces. A detailed experimental study performed in situ, under ultra-high vacuum conditions with scanning tunneling microscopy (STM), low energy electron diffraction (LEED) and Auger electron spectroscopy (AES) of the different steps on the preparation of the sample, has allowed us to gain insight into the optimal parameters for this growth method, that combines ion bombardment and annealing. This experimental work is complemented by first-principles calculations and STM simulations that provide the variation of the projected density of states due to both the metallic substrate and the nitrogen atoms. These calculations enlighten the experimental findings and prove that the species found are graphitic nitrogen. This easy and effective technique leads to the possibility of playing with the amount of dopants and the metallic substrate to obtain the desired doping of the graphene layer.