Search

Atomistic molecular dynamics simulations were carried out to obtain information on the rheological, aggregation and disintegration properties of carboxylated (TEMPO-oxidized) cellulose nanofibrils with different functionalization levels. The magnitude of the inter-fibril interaction was quantified for parallel nanofibrils using the umbrella sampling method. The obtained potential of mean force was found highly sensitive to the charge configuration for intermediate functionalization levels. This feature was further studied with an electrostatic model for similar charge configurations and system periodicity as in the case of the molecular dynamics simulations. The electrostatic contribution of the charged surfaces varied from repulsive to attractive depending on the distribution of the carboxylate groups and nearby counter-ions, as well as the distance between the fibrils. The simulated deviations from average behavior for single fibrils in both models suggest heterogeneity in their aggregation and disintegration behavior. This was seen in disintegration experiments, where the differences in disintegration energy and in the structural variation qualitatively agreed with the model predictions. As to aggregation behavior, the studied case with parallel fibrils reflects the upper boundary of the repulsive interaction.

We uncover the electronic structure of molecular graphene produced by adsorbed CO molecules on a copper (111) surface by means of first-principles calculations. Our results show that the band structure is fundamentally different from that of conventional graphene, and the unique features of the electronic states arise from coexisting honeycomb and Kagome symmetries. Furthermore, the Dirac cone does not appear at the K-point but at the Γ-point in the reciprocal space and is accompanied by a third, almost flat band. Calculations of the surface structure with Kekule distortion show a gap opening at the Dirac point in agreement with experiments. Simple tight-binding models are used to support the first-principles results and to explain the physical characteristics behind the electronic band structures.

We use atomistic molecular dynamics simulations to study the crystal structure of cellulose nanofibrils, whose sizes are comparable with the crystalline parts in commercial nanocellulose. The simulations show twisting, whose rate of relaxation is strongly temperature dependent. Meanwhile, no significant bending or stretching of nanocellulose is discovered. Considerations of atomic-scale interaction patterns bring about that the twisting arises from hydrogen bonding within and between the chains in a fibril.

The formation of extended electron states in one-dimensional nanostructures is of key importance for the function of molecular electronics devices. Here we study the effects of strong electron-phonon interaction on the formation of extended electronic states in intentionally created Cl vacancy pairs and chains in a NaCl bilayer on Cu(111). The interaction between the vacancies was tailored by fabricating vacancy pairs and chains of different orientation and separation with atomic precision using vertical manipulation. Small separation of divacancies led to the formation of symmetric and antisymmetric vacancy states and localized interface-states. By scanning tunneling spectroscopy (STS) we measured their energy splitting and broadening as a function of the inter-vacancy separation. Unexpectedly, the energy splitting between the vacancy states is enlarged by level repulsion resulting from phonon dressing of the electronic states, as evidenced by theory. Already for a few coupled vacancies we observe an emerging band structure of the defect band., 9 pages, 4 figures

Transformation of cellulose into monosaccharides can be achieved in a chemical process performed by a special group of enzymes known as cellulases. We have used atomistic molecular dynamics simulations to study endoglucanase II (Cel5A) that is one of the proteins in this group. Based on the atomistic simulation results, we discuss how the Cel5A enzyme interacts with cellulose fibrils comprised of both crystalline and amorphous regions. We show that the enzyme’s carbohydrate-binding domain prefers to interact with crystalline regions of cellulose, while the catalytic domain has a high affinity to the amorphous regions of fibrils. In particular, through electrostatic interactions the catalytic domain attracts loose glucose chains to its catalytic cleft. The atomistic details of the enzyme–cellulose interaction are presented and the implications for practical applications are briefly discussed.

We study tip-adsorbate–substrate interactions in scanning tunneling microscopy (STM) manipulation and imaging, and the influence of impurities on the images. Thence, we perform molecular dynamics simulations and calculate qualitative STM images for oxygen on Pt(1 1 1) surface. The adsorption site of the oxygen molecule is found to be in accordance with ab initio calculations. The calculated STM image has a good resemblance to the experimental ones. The contamination of the tip by oxygen or water alters the STM image strongly. Molecular dynamics simulations on manipulations of oxygen on the surface reveal several mechanisms of how molecular oxygen can be either produced or decomposed with STM tip. Finally, we find out that transfer of oxygen from the surface to an STM tip is not very probable.

Man-made artificial graphene has attracted significant attention in the past few years due to the possibilities to construct designer Dirac fermions with unexpected topological properties and applications in nanoelectronics. Here we use a first-principles approach within density-functional theory to study molecular graphene similar to the experiment by Gomes~{\em et al.}, Nature {\bf 483}, 306 (2012). The system comprises carbon monoxide molecules arranged on a copper (111) surface in such a way that a hexagonal lattice is obtained with the characteristic electronic properties of graphene. Our results show in detail how carbon monoxide molecules modify the copper surface (and regions beneath) and create a hexagonal lattice of accumulated electrons between the adsorbate molecules. We also demonstrate how the properties of the formed Dirac fermions change as the CO density is tuned, and provide a direct comparison with experimental scanning tunneling microscope images.

Molecular dynamics simulations are performed to study adsorption of methane on Pd(1 1 0). The direct dissociation is found to take place due to transfer of incident translational energy into vibrational modes of the molecule. The deformation modes of the molecule appear to be the most susceptive to this energy transfer. At high incident translational energies the molecules tend to dissociate directly, while at intermediate energies the molecules are bounced off the surface. At low energies, the molecules are trapped into a physisorption state, which may be a gateway to thermally assisted dissociation within a longer time.

Dissociative adsorption of methane has been investigated on Pd {1 1 0} by using molecular beam surface scattering. The initial sticking probability has been determined in the translational energy range of 7–95 kJ/mol and at selected vibrational energies from 300 to 700 K. The measured initial sticking probability is found to increase strongly with both translational and vibrational energy of CH4 molecules. The activation of the dissociative chemisorption of CH4 induced by the vibrational energy is shown to depend on the translational energy and is attributed to the excitation of the bending modes of the incident molecule. We have also performed molecular dynamics simulations to investigate the dissociation mechanism theoretically. The simulations clearly demonstrate that an efficient energy transfer occurs upon adsorption between the translational and vibrational energies of the incident CH4 molecule, which thereby facilitates the deformation of the molecular structure of CH4 resulting in dissociation.

A method to calculate local electronic structure and interatomic forces for molecular dynamics simulation of a materials system with two size scales, is described. In the method, the interactions for the larger scale subsystem are modeled with semiempirical potentials, and the interactions for the smaller scale subsystem are calculated quantum mechanically. Interatomic Hellman-Feynman forces can be calculated using the off-diagonal matrix elements of density matrix derived from Green's function and the gradients of the matrix elements of the tight-binding Hamiltonian matrix. The elements of Green's function for semiempirical particles are modeled in some simple predefined form, while Green's function for the rest of the particles is calculated in tight-binding basis using Dyson's equation. As an example, dissociative adsorption of ${\mathrm{O}}_{2}$ on a stepped $\mathrm{Pd}{110}$ surface is simulated.

We present scanning tunneling microscopy (STM)-based single-molecule synthesis of linear metal−ligand complexes starting from individual metal atoms (iron or nickel) and organic molecules (9,10-dicyanoanthracene) deposited on an ultrathin insulating film. We directly visualize the frontier molecular orbitals by STM orbital imaging, from which, in conjunction with detailed density functional theory calculations, the electronic structure of the complexes is inferred. Our studies show how the order of the molecular orbitals and the spin-state of the complex can be engineered through the choice of the metal atom. The high-spin iron complex has a singly occupied delocalized orbital with a large spin-splitting that points to the use of these engineered complexes as modular building blocks in molecular spintronics.

We present a fast and efficient tight-binding (TB) method for simulating scanning tunneling microscopy (STM) imaging of adsorbate molecules on ultrathin insulating films. Due to the electronic decoupling of the molecule from the metal surface caused by the presence of the insulating overlayer, the STM images of the frontier molecular orbitals can be simulated using a very efficient scheme, which also enables the analysis of phase shifts in the STM current. Au-pentacene complex adsorbed on a NaCl bilayer on Cu substrate provides an intricate model system, which has been previously studied both experimentally and theoretically. Our calculations indicate that the complicated shape of the molecular orbitals may cause multivalued constant current surfaces -- leading to ambiguity of the STM image. The results obtained using the TB method are found to be consistent with both DFT calculations and experimental data., 19 pages, 5 figures

A combined study using density functional calculations and scanning tunneling microscopy experiments shows that individual silver adatoms on ultrathin sodium chloride films on copper surfaces are stable in three different charge states—neutral, negatively, and positively charged adatoms. The charge states of the individual adatoms were manipulated by voltage pulses. The key parameters determining the stability of various charge states are identified and discussed within a simple model.

A covalent bond between an individual pentacene molecule and a gold atom was formed by means of single-molecule chemistry inside a scanning tunneling microscope junction. The bond formation is reversible, and different structural isomers can be produced. The single-molecule synthesis was done on ultrathin insulating films that electronically isolated the reactants and products from their environment. Direct imaging of the orbital hybridization upon bond formation provides insight into the energetic shifts and occupation of the molecular resonances.

Broad Gaussian line shapes are observed in scanning tunneling spectroscopy of single, localized electronic states induced by Cl vacancies in ultrathin NaCl films on Cu surfaces. Using a simple inelastic resonance tunneling model, we show that the observed broad line shapes are caused by a strong coupling between the localized state and the optical phonons in the film. The parameters for the model are obtained from density functional calculations, in which the occupation of the vacancy state temporarily taking place in the experiment has also been accounted for.

Dissociation of methane on $\mathrm{Pd}(110)$ and $\mathrm{Pd}(320)$ is investigated using molecular dynamics simulations and statistical analysis of excitations of vibrational modes of the molecule. Simulations are made by varying translational energy and excitations of vibrational stretching $({\ensuremath{\nu}}_{1})$ and bending $({\ensuremath{\nu}}_{2})$ modes of methane. The role of these factors in dissociation is found to depend on the adsorption path in a rather complicated way. Especially, different vibrational modes seem to facilitate dissociation at different adsorption site and for different orientation of the molecule. Statistical analysis of excitations of vibrational modes indicate that dissociation is governed at low vibrational temperatures by bending modes and on higher temperatures by stretching mode. Translational energy is still found to be more efficient activator of dissociation than vibrational energy.

We compare computer simulations to experimental scanning tunneling microscopy (STM) images of chloronitrobenzene molecules on a Cu(111) surface. The experiments show that adsorption induced isomerization of the molecules takes place on the surface. Furthermore, not only the submolecular features can be seen in the STM images, but different isomers can also be recognized. The Todorov-Pendry approach to tunneling produces simulated STM images which are in good accordance with the experiments. Alongside with STM simulations in a tight-binding basis, ab initio calculations are performed in order to analyze the symmetry of relevant molecular orbitals and to consider the nature of tunneling channels. Our calculations show that while the orbitals delocalized to the phenyl ring create a relatively transparent tunneling channel, they also almost isolate the orbitals of the substitute groups at energies which are relevant in STM experiments. These features of the electronic structure are the key ingredients of the accurate submolecular observations.

A change in bias voltage can reverse the contrast of a scanning-tunneling microscopy (STM) image of an adsorbate molecule on a solid surface. This contrast change may take place not only in a case of tunneling in the neighborhood of a resonance level of the adsorbate, but it can also be seen in the case of a wide energy gap between highest occupied molecular-orbital and lowest unoccupied molecular-orbital levels. In order to find a unified description of contrast changes in STM images, we make comparisons between different approaches to tunneling. We especially consider the role of phase differences between possible tunneling channels. These phase differences can be related to the imaginary and real parts of the Green's function of the system which reveal whether the tunneling takes place through a resonant state or a forbidden gap. Finally, we compare experimental and calculated images of oxygen molecules on a silver surface and analyze the mechanisms of the contrast change in terms of phase shifts in different tunneling channels.

We perform a molecular dynamics (MD) study of non-dissociative adsorption of O 2 molecule on Pt{111} surface at low temperatures, and analyse the STM image of the obtained geometry. The interatomic forces for the simulation are calculated using tight-binding method in density matrix formalism with the constraint of local charge neutrality. The STM image is constructed by calculating tunneling current between a sharp metal tip and the sample. The MD simulation gives an adsorption site with a very plausible bonding behaviour. Furthermore, the corresponding STM image is similar to recent experimental images. The present simulation suggests formation of a larger complex of the oxygen molecule and two Pt atoms. It turns out that the STM image does not directly show the orientation of the O 2 molecule, and has such a complicated origin in interference of atomic orbitals of the formed complex, that an optimization of the surface structure using MD simulation, instead of intuitive considerations, is necessary.