• Home
• Research
• Publications
• CV
• Group
• News
• Contact

News

• Oct 2017. High-resolution submolecular imaging at 5K using the IETS amplitude. This paper combining measurements and simulations is a collaboration with our colleagues from Pavel Jelinek’s group here at the Institute and from the DIPC. In a seminal paper, W. Ho’s group showed that submolecular structure could be imaged from the spatial variation of the IETS frequencies. However, the high resolution needed to track the changes in mode energies at different tip positions was only possible at sub-Kelvin temperatures, limiting the wider application of this technique. Here we show how submolecular resolution is possible at 5K from the spatial variation of the IETS intensity. We approached a FePc molecule on Au with CO-functionalized tip: as the lateral position of the CO+tip was changed, the amplitude of the FR modes of CO remained almost constant, while that of the FT modes changed substantially. The contribution of our group was to explain the sensitivity of these modes to tip position. We used a technique we had developed previously to map the spatial origin of the IETS peaks and found clear differences between FT and FR modes. FR modes are generated mostly on the CO molecule and tip, and the contribution from other atoms is small. To correctly describe the intensity of FT modes, on the other hand, it is necessary to also include the interaction of the CO molecule with the molecular substrate. FT modes will therefore be more sensitive to the lateral position of the CO molecule relative to the molecule. Our colleagues also extended their probe-particle model to describe the common imaging mechanism for all three STM, AFM and IETS channels. Altogether, this paper demonstrates that it is possible to achieve submolecular resolution at 5K comparable to AFM from the spatial variation of the IETS amplitude. Publications

• Oct 2017. Paper on the role of molecular adsorbates in current-induced cooling. In the past we had studied how, when applying a voltage across a molecular junction, the electrical current could heat (but also cool!) the junction. In this new paper, we again consider a biscarbene molecule but this time with an electron-withdrawing NH2 species adsorbed on only one of the tip terminations. We compare the changes in current-induced energy exchange of this system with the ‘clean’ junction (which has no co-adsorbed species). Despite the absence of chemical bonds between the bridging molecule and the adsorbate, the NH2 group in the vicinity results in the cooling of biscarbene modes at all applied voltages. The population and energy stored in biscarbene modes are smaller with co-adsorbed NH2 than for the clean junction, and NH2 modes are heated while biscarbene modes cooled with applied voltage. We find that this is a rather indirect effect, caused by the shift in the biscarbene DOS induced by the presence of the NH2 species, while the effect of the adsorbate DOS itself is small. By changing the nature of the adsorbate (from electron-donating to withdrawing), it should be possible to tune the heating and cooling of molecular junctions. Publications

• July 2017. Conductance of an antiaromatic molecule measured for the first time! Together with our Japanese collaborators the Kiguchi group from Tokyo Institute of Technology and the Shinokubo group from Nagoya University, we study the conductance of a genuinely antiaromatic molecular circuit for the first time. Cyclic and planar molecules are said to be aromatic if they have 4n+2 pi electrons, and antiaromatic if they have 4n pi electrons. There are many aromatic species in molecular nanoscience (oligophenyls, thiophenes, porphyrins…) but antiaromatic molecules are very unstable and hard to synthesize. Antiaromatic molecules had been predicted by Breslow in the 70s to be highly conducting based on electrochemical measurements of oxidation/reduction potentials. In single molecule circuits it was recently shown that less aromatic molecules are more conducting but no genuinely antiaromatic molecule had been measured up to now. In our work we compare porphyrin (aromatic) and norcorrole (antiaromatic) units. The norcorrole core has two fewer Carbon atoms than the porphyrin (thus going from 4n+2 to 4n) but is structurally similar. The antiaromatic molecule is ~25× more conducting than its aromatic counterpart due to a more favourable alignment of frontier orbitals at the interface. Publications

• May 2017. Our paper on Au-C bonds in fullerenes is out! We form fullerene-based structures on Au(111) which are stable at room temperature against diffusion on the surface. We do this by low-energy Ar⁺ sputtering of fullerene films. Sputtering at these low energies (120 eV) was shown in Carbon materials to result in predominantly single vacancy defects. After sputtering we see bright spots on the surface that do not diffuse at room temperature. DFT calculations show that molecules with one missing atom are much more stable at the surface than fullerenes or than molecules with double vacancies. This increased binding comes from the saturation of the bonds around the defect by the Au surface. We interpret the observed features as adsorbed fullerene-based molecules with C vacancies. This is a collaboration with the Nanosurf group here at the Institute, who did the measurements. Publications

• Jan. 2017. Welcome Enrique! Enrique joins the group as a postdoctoral researcher. He will work on single molecule transport theory. Group

• Dec. 2016. Our paper on current-induced cooling is out! We calculate the current-induced excitation and damping of vibrational modes in a carbene-based molecular circuit. Previously, we had seen that for this system, the width and position of the LUMO resonance was determined by the atomistic details of the tip (adatom, pyramid,...). Here we calculate the energy exchange between electronic and vibrational degrees of freedom. Energy transferred to vibrational modes will heat the junction, while in the opposite case it is released in the electrodes far from the junction, which is effectively cooled. We find that for chain-like tip structures (which can result from the stretching of strong Au-C bonds), current can cool the junction at high applied voltage. Current-induced heating and cooling processes are important as they strongly influence the stability of the molecular junction under an applied voltage. Publications

• July 2016. Postdoc position available! A postdoc position is available in the group. Candidates should have experience with first-principles simulations. Preference will be given to candidates with expertise in (Tran)SIESTA codes and FORTRAN programming. Informal inquiries are welcome.

• July 2016. Our paper on the local contributions to IETS is out! Here we completely characterize the different contributions to the inelastic spectrum of a representative molecular junction. These contributions encode information on both the vibrational modes and the electronic structure and thus go beyond the usual figures of the modes. Since we use a local orbital basis, we can determine the spatial origin of all contributions to the vibrational modes, mapping the origin of the inelastic signal. We thereby provide, across all vibrational modes, a quantitative relation between the degree of symmetry of each mode, its inelastic signal, and the locality of selection rules. Publications

• July 2016. Giuseppe starts his Marie Skłodowska-Curie fellowship! In this project we will study current-induced vibrational heating and cooling of molecular junctions. The interaction of vibrational degrees of freedom with the tunneling electrons can result in the heating and, under certain circumstances, even cooling of the junction, and the endless number of chemical structures opens many possibilities to tune and control these processes. To study the electronic structure and the inelastic processes associated to the emission and absorption of molecular vibrations, we will carry out first-principles simulations based on DFT. We look forward to a fruitful fellowship!

• March 2016. Welcome Martin! In addition to his experimental work on nanoprobes in the group of Antonín Fejfar, we are happy Martin is interested in carrying out ab-initio simulations. He will calculate the interface properties of carboranes at metal surfaces. Group

• Feb. 2016. Our paper on tip-induced gating of carbene-based junctions is out! We calculate the transport properties of N-heterocyclic carbene-based (NHC) molecules on gold and see a strong very dependence of the LUMO position on the atomistic details of the tip. Depending on whether the tips have tetrameric, pyramidal, chain-like or adatom terminations, the position of the LUMO shifts by almost ~0.8 eV. We explain these changes through an analysis of the electron density difference. Since transport is LUMO-derived, the calculated transmission at the Fermi level changes by a factor 8× and the tip is effectively gating the carbene-based circuit. Publications

• Feb. 2016. Welcome Narendra! Narendra joins the group. He will work on using and developing DFT-based methods to calculate conductance single in molecular junctions. Group