Energy Security and Climate Change: A New Approach for Global Sustainability in the 21st Century.

Transcription

1 Energy Security and Climate Change: A New Approach for Global Sustainability in the 21st Century. Tomas Diaz de la Rubia (LLNL) Projections by the Department of Energy's Energy Information Administration and most other international studies show that worldwide electric power demand will increase from the current level of about 2 Terawatts (TW) to 5 TW by 2050 and likely to as much as 10 TW by A recent IEA 2008 Energy Technologies Perspectives report shows that for the next 30 to 50 years burning fossil fuels will continue to provide most of the world's electricity. In fact, in these baseline scenarios CO2 emissions will be almost two and a half times the current level by In addition, the most recent report from the Intergovernmental Panel on Climate Change has placed a 90% likelihood that human sources of carbon dioxide emissions are significantly affecting the global climate. Clearly, this increasing demand is placing enormous pressure on natural resources, the global ecosystem, and international political stability. Alternative sources of energy are required in order to meet increased energy demand, stabilize the increase of atmospheric carbon dioxide, and mitigate the concomitant climate change. In response, governments are urgently trying to develop new economical, sustainable, and environmentally friendly energy technologies. In this talk, I will discuss an approach to generating carbon-free, economically competitive power from nuclear energy that greatly mitigates proliferation concerns, minimizes nuclear waste and eliminates concerns related to reactor core meltdown accident scenarios. The approach, Laser Inertial Fusion-based Energy (LIFE), combines a modest, neutron-rich fusion source with a subcritical fission blanket into an engine capable of generating several thousand MegaWatts. A LIFE engine can utilize a variety of fertile and fissile fuels, eliminates the need for uranium enrichment and for Spent Nuclear Fuel reprocessing, and minimizes the production of long-lived actinides in nuclear waste to below DOE attractiveness level E (the lowest in the safeguards tables). LIFE thus represents a oncethrough, closed fuel cycle that virtually eliminates actinides from the spent nuclear fuel (SNF) form and reduces fission product volumes per unit of energy generated by factors of 20 to 100. Moreover, LIFE engines can burn the existing inventories of SNF and excess plutonium thereby drastically shrinking the nation s and the world s stockpiles of these special nuclear materials. Because LIFE is safe and minimizes proliferation concerns associated with the nuclear fuel cycle, we envision this technology as capable of providing a global solution to carbon-free energy generation in the 21st century. I will describe progress at LLNL s National Ignition Facility towards achieving fusion ignition and burn the sine qua non condition for LIFE and will discuss the specifics of the LIFE engine design and the basic and applied research challenges associated with making this vision a reality. I will close by discussing how success with LIFE could help meet the carbon-free energy demand gap for the planet and help mitigate potential climate change in the second half of the 21st century.

2 Inelastic transport in nanostructures: the CEID formalism and pdinamo Daniel Dundas (Queens University Belfast) Ehrenfest dynamics is a widely used form of non-adiabatic molecular dynamics in which a quantum description of electrons is coupled to a classical treatment of ions. While Ehrenfest dynamics does allow energy transfer between the two subsystems, it does so in an incomplete way [1]. Ehrenfest dynamics captures correctly the excitation of electrons by energetic ions, but the reverse process - the heating of ionic vibrations by excited electrons - is suppressed. The reason for this failure of the Ehrenfest approximation is the neglect of electron-ion force-momentum correlations, which results in a mean-field description of electron-ion interactions. This problem has recently been addressed in a new method called Correlated Electron-Ion Dynamics (CEID). CEID is an extension of Ehrenfest dynamics that reinstates approximately electron-ion correlations, and thus enables a consistent treatment of the energy exchange between electrons and ions away from equilibrium and of the dynamical response of the two subsystems to inelastic scattering [2]. At QUB the CEID formalism has been extended and applied to the study of electrical conduction in atomic-scale wires. CEID has been combined with electronic open boundaries, to enable dynamical simulations of inelastic quantum transport over long timescales. A rigorous connection between CEID and diagrammatic perturbation theory has been established, which allows the key approximations in CEID to be quantified in diagrammatic terms and, in principle, extended [3]. In this talk we will detail the CEID method and describe its implementation, together with electronic open boundaries, in a parallel code, called pdinamo [4], that has been developed in-house at QUB. Several applications of the method including inelastic current-voltage spectroscopy, the heating and equilibration of quantum ions with current-carrying electrons and the non-conservative nature of current-induced forces will be given [5-8]. Some directions for future research will also be discussed. 1. Horsfield A P, Bowler D R, Fisher A J, Todorov T N and Sanchez C G, J. Phys.: Condens. Matter (2004) 2. Horsfield A P, Bowler D R, Fisher A J, Todorov T N and Sanchez C G, J. Phys.: Condens. Matter (2005) 3. Wang Y, in preparation (2009) 4. Dundas D, McEniry E J, Mason D and Stella L, Capability Computing: The Newsletter of HPCx (2007) 5. McEniry E J, Bowler D R, Dundas D, Horsfield A P, Sanchez C G, and Todorov T N, J. Phys.: Condens. Matter (2007) 6. McEniry E J, Frederiksen T, Todorov T N, Dundas D and Horsfield A P, Phys. Rev B (2008) 7. McEniry E J, Todorov T N, and Dundas D, submitted to J. Phys.: Condens. Matter (2009) 8. Dundas D, McEniry E J and Todorov T N, Nature Nanotech 4 99 (2009)

3 Shot noise and vibration excitations in electron transport through single molecules Jan van Ruitenbeek (Leiden University) STM and break junction techniques at low temperatures allow manipulation of individual atoms and molecules. By such methods one can produce wires connecting the two electrodes, formed by a single (organic) molecule. There are two important tools that help us characterize the molecular junctions. First, the measurement of the intrinsic noise in the current, shot noise, allows determining the number of conductance channels. This number can be compared with DFT calculations and it also helps to demonstrate that the current is actually carried by a single molecule. As a second tool we exploit the fact that the current transport properties of these nanowires show deviations from the well know ohmic current-voltage relation at specific characteristic energies, corresponding to the vibration modes of the molecular junctions. These vibration modes, on the one hand, are exploited for characterization of the molecular wire configurations. On the other hand, the physical mechanisms involved in electron-phonon interaction at the single-molecule level can be investigated in detail. In the near future we plan to explore the effect of vibration mode excitations on shot noise, which will allow us to obtain statistical information on phonon-mode occupation. I will briefly review the predicted effects. Tight Binding Simulations in the Ehrenfest approximation: results from radiation damage using spiced Daniel Mason (Imperial College London) In the limit that spontaneous phonon emission may be neglected, i.e. where the ionic temperature is much greater than the electronic, the Ehrenfest approximation result for the rate of heat transfer between ions and electrons is good. We have focussed on using the Ehrenfest approximation in a region where it is applicable- to look at high energy ions in metals. Evolving the electrons in a model metal using Ehrenfest dynamics is sufficiently computationally inexpensive that simulations of ten thousand atoms and more are now routine. This opens up the possibility of using simulation cells of a sufficient size to contain the trajectories of high energy atoms; and with this the possibility of exploring radiation damage and other phenomena with proper quantum mechanical electrons. We have been able to study the electronic stopping of ions with a kinetic energy of ev up to MeV. In this talk I will report on our recent results illuminating the process of heat transfer between ions and electrons in radiation damage events, and the code we have developed to find them. Spanning time-scales in simulations of irradiated materials Vasily Bulatov (LLNL) Production of crystal defects in materials subjected to irradiation by high-energy particles is affected by energy exchange between electrons and phonons on very short time-scales. This initial damage is the primary energy source driving subsequent long-time evolution of material microstructure. This

4 talk will begin with a brief overview of processes taking place after the initial collision cascades cool down. The resulting evolution entails multiple competing mechanisms - defect diffusion, annihilation, clustering, absorption, etc. - taking place on time scales ranging from picoseconds to years. The rest of the talk will discuss several existing and emerging approaches to simulations of irradiated materials and their potential for accurate predictions of material damage accumulation in nuclear reactors. PolyCEID: towards a better description of non-adiabatic molecular processes by Correlated Electron-Ion Dynamics Lorenzo Stella (University College London) Non-adiabatic processes in molecules often lead to ionic wave-packet splitting in proximity of a Landau-Zener crossing. On the other hand, it is also reasonable to expect a coherent superposition of well separated ionic wave-packets to decay quickly into a mixed state. Indeed, classical molecular dynamics, which ignores any quantum coherence effect, is known to provide a good approximation for long time dynamics of large atomic systems. The original version of CEID, being based on a selfconsistent second order closure of the equations of motion hierarchy, is not suitable to describe ionic wave-packet splitting. On the other hand, a direct extension of the original self-consistent derivation of the CEID equations to higher order closure schemes poses formidable mathematical problems and it might be not viable. For these reasons, the group at UCL started the development of a complementary version of CEID based on a different density matrix expansion in order to apply the method to molecules, in particular conjugated polymers. The theoretical results of these investigations converged into an original computer code, PolyCEID, which is able to provide an accurate description of both wave-packet splitting and electronic decoherence. In this contribution this alternative formulation of the CEID is explained, along with the main features of the code. Two test-cases are also presented to illustrate the code capabilities. Finally, a brief account of the code status and future developments is provided. Charge separation in molecular donor acceptor heterojunctions Jenny Nelson (Imperial College London) Photocurrent generation in organic photodetectors and solar cells relies upon the photoinduced generation of electron-hole polaron pairs at a heterojunction. The dissociation of photogenerated excitons into separate charges is believed to consist of two stages, the formation of a bound (geminate) polaron pair followed by the dissociation of the pair into separate charges. The charge generation process competes both with the transfer of energy to neutral excited states of the system and with the recombination of the geminate charges before separation. The yield and dynamics of the charge separation (CS) process can be monitored experimentally using transient optical spectroscopy. We will report on experimental studies of charge pair generation in a range of polymer:fullerene blend films. We find that in some families of polymers, the charge generation yield correlates to the free energy difference between the excited singlet state and the charge separated state, as expected from Marcus theory. However, the charge generation yield is also influenced by the specific chemical structure of the donor polymer, by the degree of phase

5 segregation and molecular order in the separate phases, and by the formation of emissive charge transfer states. Moreover, the high charge separation yields observed are surprising given the large Coulombic barrier involved. In this talk, we will report on these observations, on several outstanding unsolved problems, and on current theoretical approaches to describe the charge separation process.