The common goal of this workshop was to bring together computational groups from the two major branches of model approaches and atomistic simulations with experimental groups, in order to discuss and define possible interconnections and regions of common interest. The composition of the scientific program with 15 experimental and 25 theoretical invited talks reflects this ambition. The format of the meeting, with no parallel sessions and with ample time for discussion and opportunities to exchange ideas, was also designed with that purpose in mind.
Several key lectures on contemporary measurements have presented their results in order to uncover remaining and new challenges for the theory of molecular conduction. Participants from both the experimental and computational-theoretical side have attended the workshop with high resonance.
Topics of discussion included
On the experimental side, the field of single-molecule contacts has been successful in contacting individual molecules using various techniques like mechanically controlled break junctions, electromigrated break junctions, nanoparticle dimers, Scanning tunneling microscopes in a much better reproducible way. The parameters in experiments range from ultralow temperatures to room temperature, with experiments in vacuum, in solvent or in electrolyte. Each technique has shown advantages and limitations. However, a full physical picture of molecular-scale charge transport becomes only accessible by careful comparison of the entity of available data and in correlating experimental results with computational predictions.
Simultaneously, the community has experienced that structure property relationships are very interesting on the qualitative level, but have difficulties in reproducing results due to the nearly unavoidable sample-to-sample fluctuations. This is also a major handicap for delivering reliable data for the comparison with theoretical predictions. Techniques to master these problems have matured and now noise, Kondo physics, molecular magnetism, molecular spin crossover and other effects become visible on the single-molecule level, which is a very desirable development for addressing improved quantitative understanding by computational studies.
The meeting was extremely successful for many reasons. One of these had to do with the quality of the discussions: it was particularly striking that a number of questions were asked by postdoctorals and graduate students, from labs in Australia, Denmark, the US, Germany, and France. The closeness of the group, their common interests, and the very open atmosphere, combined with the very welcoming lecture room, to make it easy for questions to be asked, and discussions to be begun.
The group was obviously up on the current literature, and many questions about unpublished work were asked. Finally, there was an equal representation in the discussions of formal theoretical matters (methods, functionals, convergence, self consistency, …) and physical issues (decoherence, activation processes, geometric modifications, thermal dependences, experimental comparison).
The fundamental issues involved in understanding molecular charge transport and its magnetic generalizations were aired at this meeting, and many different approaches were illustrated. These ranged from very useful formal models to computations intended to be quantitative in nature. Bringing all of these groups together, for an intense discussion of five days, was extremely successful.
As the 21st century begins, science in many areas has gotten to the point where we hope to be able to control, as well as measure, what nature does. In the particular areas of molecular charge transport and spintronics, difficulties come about because the systems are by their nature far from equilibrium (the molecule sees two different voltages at the two different electrodes, or at three different electrodes when gating occurs). We then are left with a non-equilibrium problem, for which the variational principle is invalid, and really only quantum kinetics remains as a rigorous approach. This rigorous approach was utilized throughout the workshop, and difficulties were brought out (applicability of and improvements to density functional theory, basis set issues, vibronic issues, etc.). The close comparison with some of the very best experimentalists in the world made this a much more interesting workshop than it would have been without the experimentalists there – the experimental people continually bothered the theorists about the details, because some of the details did not agree with the experiments.
The future direction of this area is clearly going along two pathways. The first is towards generalization to systems that are more than one single molecule, such as adlayer films, photoconduction, and energy capture in solar cells. The second is going to be single molecule and single nanoparticle transport, with or without magnetic field effects. Both of these will require appropriate theoretical methodologies. This workshop certainly put density functional theory, including tight binding density functional theory, into everybody’s consciousness as one of the best ways to go after these problems. Nevertheless, there need to be continuing workshops (perhaps approaching the value of this one!) to work out how to extend the current density functionals towards more accurate representations, and how to mix in things like indirect pathways, quantum coherence, noise measurements, and quantum states rather than quantum levels.