ERC FUNDED PROJECTS

"Large bodies usually follow the classical equations of motion. Deviations from this can be called macroscopic quantum behavior. These phenomena have been experimentally verified with cavity Quantum Electro Dynamics (QED), trapped ions, and superconducting Josephson junction systems. Recently, evidence was obtained that also moving objects can display such behavior. These objects are micromechanical resonators (MR), which can measure tens of microns in size and are hence quite macroscopic. The degree of freedom is their vibrations: phonons. I propose experimental research in order to push quantum mechanics closer to the classical world than ever before. I will try find quantum behavior in the most classical objects, that is, slowly moving bodies. I will use MR's, accessed via electrical resonators. Part of it will be in analogy to the previously studied macroscopic systems, but with photons replaced by phonons. The experiments are done in a cryogenic temperature mostly in dilution refrigerator. The work will open up new perspectives on how nature works, and can have technological implications. The first basic setup is the coupling of MR to microwave cavity resonators. This is a direct analogy to optomechanics, and can be called circuit optomechanics. The goals will be phonon state transfer via a cavity bus, construction of squeezed states and of phonon-cavity entanglement. The second setup is to boost the optomechanical coupling with a Josephson junction system, and reach the single-phonon strong-coupling for the first time. The third setup is the coupling of MR to a Josephson junction artificial atom. Here we will access the MR same way as the motion of a trapped ions is coupled to their internal transitions. In this setup, I am proposing to construct exotic quantum states of motion, and finally entangle and transfer phonons over mm-distance via cavity-coupled qubits. I believe within the project it is possible to perform rudimentary Bell measurement with phonons."

2 004 283 €

Start date: 2015-01-01, End date: 2019-12-31

Self-assembly is the key construction principle that nature uses so successfully to fabricate its molecular machinery and highly elaborate structures. In this project we will follow nature’s strategies and make a concerted experimental and theoretical effort to study, understand and control self-assembly for a new generation of colloidal building blocks. Starting point will be recent advances in colloid synthesis strategies that have led to a spectacular array of colloids of different shapes, compositions, patterns and functionalities. These allow us to investigate the influence of anisotropy in shape and interactions on aggregation and self-assembly in colloidal suspensions and mixtures. Using responsive particles we will implement colloidal lock-and-key mechanisms and then assemble a library of “colloidal molecules” with well-defined and externally tunable binding sites using microfluidics-based and externally controlled fabrication and sorting principles. We will use them to explore the equilibrium phase behavior of particle systems interacting through a finite number of binding sites. In parallel, we will exploit them and investigate colloid self-assembly into well-defined nanostructures. Here we aim at achieving much more refined control than currently possible by implementing a protein-inspired approach to controlled self-assembly. We combine molecule-like colloidal building blocks that possess directional interactions and externally triggerable specific recognition sites with directed self-assembly where external fields not only facilitate assembly, but also allow fabricating novel structures. We will use the tunable combination of different contributions to the interaction potential between the colloidal building blocks and the ability to create chirality in the assembly to establish the requirements for the controlled formation of tubular shells and thus create a colloid-based minimal model of synthetic virus capsid proteins.

2 498 040 €

Start date: 2014-02-01, End date: 2019-01-31

This proposal aims at developing new inference algorithms for graphical models with discrete variables, with a focus on the MAP estimation task. MAP estimation algorithms such as graph cuts have transformed computer vision in the last decade; they are now routinely used and are also utilized in commercial systems. Topics of this project fall into 3 categories. Theoretically-oriented: Graph cut techniques come from combinatorial optimization. They can minimize a certain class of functions, namely submodular functions with unary and pairwise terms. Larger classes of functions can be minimized in polynomial time. A complete characterization of such classes has been established. They include k-submodular functions for an integer k _ 1. I investigate whether such tools from discrete optimization can lead to more efficient inference algorithms for practical problems. I have already found an important application of k-submodular functions for minimizing Potts energy functions that are frequently used in computer vision. The concept of submodularity also recently appeared in the context of the task of computing marginals in graphical models, here discrete optimization tools could be used. Practically-oriented: Modern techniques such as graph cuts and tree-reweighted message passing give excellent results for some graphical models such as with the Potts energies. However, they fail for more complicated models. I aim to develop new tools for tackling such hard energies. This will include exploring tighter convex relaxations of the problem. Applications, sequence tagging problems: Recently, we developed new algorithms for inference in pattern-based Conditional Random Fields (CRFs) on a chain. This model can naturally be applied to sequence tagging problems; it generalizes the popular CRF model by giving it more flexibility. I will investigate (i) applications to specific tasks, such as the protein secondary structure prediction, and (ii) ways to extend the model.

1 641 585 €

Start date: 2014-06-01, End date: 2019-05-31

In spite of their small areal extent, inland waters play a vital role in the carbon cycle of the continents, as they emit significant amounts of the greenhouse gases (GHG) carbon dioxide (CO2) and methane (CH4) to the atmosphere, and simultaneously bury more organic carbon (OC) in their sediments than the entire ocean. Particularly in tropical hydropower reservoirs, GHG emissions can be large, mainly owing to high CH4 emission. Moreover, the number of tropical hydropower reservoirs will continue to increase dramatically, due to an urgent need for economic growth and a vast unused hydropower potential in many tropical countries. However, the current understanding of the magnitude of GHG emission, and of the processes regulating it, is insufficient. Here I propose a research program on tropical reservoirs in Brazil that takes advantage of recent developments in both concepts and methodologies to provide unique evaluations of GHG emission and OC burial in tropical reservoirs. In particular, I will test the following hypotheses: 1) Current estimates of reservoir CH4 emission are at least one order of magnitude too low, since they have completely missed the recently discovered existence of gas bubble emission hot spots; 2) The burial of land-derived OC in reservoir sediments offsets a significant share of the GHG emissions; and 3) The sustained, long-term CH4 emission from reservoirs is to a large degree fuelled by primary production of new OC within the reservoir, and may therefore be reduced by management of nutrient supply. The new understanding and the cross-disciplinary methodological approach will constitute a major advance to aquatic science in general, and have strong impacts on the understanding of other aquatic systems at other latitudes as well. In addition, the results will be merged into an existing reservoir GHG risk assessment tool to improve planning, design, management and judgment of hydropower reservoirs.

1 798 227 €

Start date: 2013-09-01, End date: 2019-08-31