ERC FUNDED PROJECTS

Computer models based on known physical laws are our primary tool for predicting climate change. Yet the state-of-the-art models exhibit a disturbingly wide range of predictions of future climate change, especially when examined at the regional scale, which has not decreased as the models have become more comprehensive. The reasons for this are not understood. This represents a basic challenge to our fundamental understanding of climate. The divergence of model projections is presumably related to systematic model errors in the large-scale fluxes of heat, moisture and momentum that control regional aspects of climate. That these errors stubbornly persist in spite of increases in the spatial resolution of the models suggests that they are associated with errors in the representation of unresolved processes, whose effects must be parameterised. Most attention in climate science has hitherto focused on the thermodynamic aspects of climate. Dynamical aspects, which involve the atmospheric circulation, have received much less attention. However regional climate, including persistent climate regimes and extremes, is strongly controlled by atmospheric circulation patterns, which exhibit chaotic variability and whose representation in climate models depends sensitively on parameterised processes. Moreover the dynamical aspects of model projections are much less robust than the thermodynamic ones. There are good reasons to believe that model bias, the divergence of model projections, and chaotic variability are somehow related, although the relationships are not well understood. This calls for studying them together. My proposed research will focus on this problem, addressing these three aspects of the atmospheric circulation response to climate change in parallel: (i) diagnosing the sources of model error; (ii) elucidating the relationship between model error and the spread in model projections; (iii) understanding the physical mechanisms of atmospheric variability.

2 489 151 €

Start date: 2014-03-01, End date: 2020-02-29

Active Matter systems, such as self-propelled colloids, violate time-reversal symmetry by producing entropy locally, typically converting fuel into mechanical motion at the particle scale. Other driven systems instead produce entropy because of global forcing by external fields, or boundary conditions that impose macroscopic fluxes (such as the momentum flux across a fluid sheared between moving parallel walls). Nonequilibrium statistical physics (NeSP) is the basic toolbox for both classes of system. In recent years, much progress in NeSP has stemmed from bottom-up work on driven systems. This has provided a number of exactly solved benchmark models, and extended approximation techniques to address driven non-ergodic systems, such as sheared glasses. Meanwhile, work on fluctuation theorems and stochastic thermodynamics have created profound, model-independent insights into dynamics far from equilibrium. More recently, the field of Active Matter has moved forward rapidly, leaving in its wake a series of generic and profound NeSP questions that now need answers: When is time-reversal symmetry, broken at the microscale, restored by coarse-graining? If it is restored, is an effective thermodynamic description is possible? How different is an active system's behaviour from a globally forced one? ADSNeSP aims to distil from recent Active Matter research such fundamental questions; answer them first in the context of specific models and second in more general terms; and then, using the tools and insights gained, shed new light on longstanding problems in the wider class of driven systems. I believe these new tools and insights will be substantial, because local activity takes systems far from equilibrium in a conceptually distinct direction from most types of global driving. By focusing on general principles and on simple models of activity, I seek to create a new vantage point that can inform, and potentially transform, wider areas of statistical physics.

2 043 630 €

Start date: 2017-10-01, End date: 2022-09-30

Mathematical proofs have always been prone to error. Today, proofs can be hundreds of pages long and combine results from many specialisms, making them almost impossible to check. One solution is to deploy modern verification technology. Interactive theorem provers have demonstrated their potential as vehicles for formalising mathematics through achievements such as the verification of the Kepler Conjecture. Proofs done using such tools reach a high standard of correctness. However, existing theorem provers are unsuitable for mathematics. Their formal proofs are unreadable. They struggle to do simple tasks, such as evaluating limits. They lack much basic mathematics, and the material they do have is difficult to locate and apply. ALEXANDRIA will create a proof development environment attractive to working mathematicians, utilising the best technology available across computer science. Its focus will be the management and use of large-scale mathematical knowledge, both theorems and algorithms. The project will employ mathematicians to investigate the formalisation of mathematics in practice. Our already substantial formalised libraries will serve as the starting point. They will be extended and annotated to support sophisticated searches. Techniques will be borrowed from machine learning, information retrieval and natural language processing. Algorithms will be treated similarly: ALEXANDRIA will help users find and invoke the proof methods and algorithms appropriate for the task. ALEXANDRIA will provide (1) comprehensive formal mathematical libraries; (2) search within libraries, and the mining of libraries for proof patterns; (3) automated support for the construction of large formal proofs; (4) sound and practical computer algebra tools. ALEXANDRIA will be based on legible structured proofs. Formal proofs should be not mere code, but a machine-checkable form of communication between mathematicians.

2 430 140 €

Start date: 2017-09-01, End date: 2022-08-31

This proposal details the mechanism-based discovery of ground-breaking new catalyst systems for a broad range of arylation processes that will be of immediate and long-lasting utility to the pharmaceutical, agrochemical, and materials chemistry industries. These industries have become highly dependent on coupling technologies employing homogeneous late transition metal catalysis and this reliance will grow further, particularly if the substrate scope can be broadened, the economics, in terms of reagents and catalyst, made more favourable, the reliability at scale-up improved, and the generation of side-products, of particular importance for optical and electronic properties of materials, minimized or eliminated. This proposal addresses these issues by conducting a detailed and comprehensive mechanistic investigation of direct arylation, so that a substantial expansion of the reaction scope can be achieved. At present, the regioselectivity can be very high, however catalyst turnover rates are moderate, and the arene is required to be in a fairly narrow window of activity. Specific aspects to be addressed in terms of mechanistic study are: catalyst speciation and pathways for deactivation; pathways for homocoupling; influence of anions and dummy ligands; protodemetalloidation pathways. Areas proposed for mechanism-informed development are: expansion of metalloid tolerance; expansion of arene scope; use of traceless activators and directors, new couplings via ligand exchange, the evolution of simpler / cheaper and more selective / active catalysts; expansion to oxidative double arylations (Ar-H + Ar’-H) with control, and without resort to super-stoichiometric bias. The long-term legacy of these studies will be detailed insight for current and emerging systems, as well as readily extrapolated information for the design of new, more efficient catalyst systems in academia, and their scaleable application in industry

2 114 223 €

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