Research Activities
Introduction
page 004 - 009
Chapter 1 page 010 - 043 Chapter 2 page 044 - 067 Chapter 3 page 068 - 101 Chapter 4 page 102 - 121 Chapter 5 page 122 - 143 Chapter 6 page 144 - 167
Facts and Figures
page 168 - 176
Introduction Lasers X-ray sources driven by ultrashort laser pulses Particle acceleration by laser Applications in molecular, biomedical, and material science Plasma and high energy density physics Exotic physics and theory Facts and Figures
Introduction
Introduction Extreme Light Infrastructure (ELI) is one of the 35 large-scale European projects identified on the ESFRI Roadmap published in 2006 and updated in 2008 (Roadmap of 44 large-scale research infrastructure projects of high priority identified by the European Strategic Forum on Research Infrastructures). Officially launched in Paris on February 21st and 22nd 2008, the Preparatory Phase of ELI involves nearly 40 research and academic institutions from 13 EU Member States. With an end scheduled in December 2010, this 36-month phase aims at bringing the project to the level of legal, organisational, financial and scientific maturity. The Preparatory Phase of ELI is organized around nine work packages coordinated by 15 participating institutions. The Institute of Physics of the Academy of Sciences of the
Introduction
Czech Republic is involved in this process as the coordinator for the strategic work package on lasers. The main objectives of the ELI Project include the construction of a modern, cutting-edge laser facility and realization of many research and application projects involving interaction of light with matter at intensities that are 100 to 1000 times greater than the values achieved at present. ELI will be delivering ultrashort laser pulses lasting typically a few femtoseconds (10-15 fs) and its peak output will approximate 200 PW. In September 2008, the Czech Republic submitted its bid along with France, Hungary, Romania and the United Kingdom for hosting ELI. The decision about location of ELI was made in October 2009 and the Czech Republic, Hungary and Romania jointly received from the ELI-PP Consortium the mandate to proceed towards the construction of ELI. Fully supported by the European Commission, the decision to implement ELI as a distributed infrastructure will lead
to the construction by end 2015 of three facilities dedicated to three of the scientific pillars of the project (attosecond science, beamline generation of secondary sources and laser-driven nuclear physics). The location of the fourth one – the ultra-high intensity pillar – will be decided in 2012. The Extreme Light Infrastructure might be managed in accordance with the new governance model designed for the European Research Infrastructure Consortium (ERIC). The ELI-ERIC will involve the founder members, the Czech Republic, Hungary and Romania, as well as the major partners of the ELI-Preparatory Phase Project, Germany, United Kingdom, France and potentially others.
Introduction ELI in the Czech Republic The primary mission of the ELI Beamlines Facility will consist of producing an entirely new generation of secondary sources driven by ultra-intense lasers. These secondary sources will produce pulses of radiation and particles such as flashes of X-rays and gamma-rays, bunches of accelerated electrons, protons and ions, etc., exploitable as unprecedented research tools in many research disciplines and in the development of new technologies. The research agenda using the ultrashort and ultra intense pulses delivered by the ELI laser will focus on the following main activities which are described in detail in next chapters: • X-ray sources driven by ultrashort laser pulses • Particle acceleration by lasers
• Applications in molecular, biomedical, and material sciences • Physics of plasmas, physics of high energy densities and of warm dense matter • Exotic and frontier physics The ultrashort and ultra-intense pulses of light and particles will find many applications in fundamental research and in chemistry, biology, medical technologies, and development of new materials. In fundamental research for the first time it will be possible to study the phenomena connected to e.g. quantum electrodynamics, space-time dependent radiation fields, structure of vacuum, and many others, in the lab. Furthermore, it will help understanding the astrophysical phenomena involved in mechanisms of radiation emitted by pulsars, brown dwarfs, and giant planets for example. In the field of practical applications, the new laser-driven sources will enable significant improvements in screening techniques in medical diagnos-
tics, and the capability of ultrashort pulses to provide high-resolution snapshots of molecules will contribute to better understanding of complex diseases such as cancer, and will enable development of personalized medicaments and medical treatments. The capability of ultrashort light pulses to obtain snapshots that have been thus far inaccessible and “frozen in time” of physical materials and chemical molecules will help the development of new materials for electronics, optoelectronics, nanotechnologies, and many others. ELI will bring short-, medium- and long-term economic opportunities associated with the construction and particularly with the development of specific technologies required for R&D facilities such as high-repetition lasers, control systems, etc. This in turn will present a unique opportunity for laser, optical and vacuum industry. By participating in building of such major facility, these industries will be able to extend their current portfolio of products that will find use in
ELI: Extreme Light Infrastructure www.eli-beams.eu
7
industry, in small- and medium-scale research laboratories, and will participate on the development of qualitatively new technologies.
Management and location
The most positive impact will be seen on a longterm time horizon. Amongst the technologies and derivative applications that are expected to bring most long term economic impacts are the new techniques used in medical imaging and diagnostics, radiation therapy, tools for development and testing of new materials, new techniques of disposal of radioactive waste, new X-ray optics, etc. It is also expected that ELI will contribute to tackling selected problems related to thermonuclear fusion.
The project is managed in the Czech Republic by a dedicated team of the Institute of Physics of the Academy of Sciences of the Czech Republic and supported by the Ministry of Education, Youth and Sports, the Academy of Sciences and the Central Bohemia Region. The Consortium ELI-CZ, which already includes 14 Czech universities and research institutions, is another key player that demonstrates the strong support of the Czech scientific community to the project.
The ELI facility will also create an attractive platform for bringing up a new generation of PhD students, scientists and engineers. This will increase significantly the visibility of the host country of such cutting edge research facility and the Czech Republic will attract further investments in advanced technologies.
The site that will host the ELI Beamlines Facility is a 6-hectare lot located in the town of Dolní Břežany, in the Central Bohemia Region. This location is accessible from downtown Prague by public transportation within less than half an hour. It is in close proximity to the nearly completed Prague motorway ring, which directly connects to the European motorway network
Introduction
and provides direct communication with the Prague International Airport. Dolní Břežany is a developing and ambitious community showing a clear and coherent concept of local and regional development. Employees and visitors will enjoy a pleasant working environment with sufficient services, newly built accommodation capacities and opportunities for relaxation, leisure and other outside activities. Situated within an area reserved by the town for the development of a technological park, the ELI site itself already offers enough room for potential upgrades of the facility, but also for spin-off companies or future industrial activities related to laser and optical science. This cluster approach will certainly foster the scientific and economic impact of the future facility.
Introduction
Architectonic design was made by Hamiltons Architects.
ELI: Extreme Light Infrastructure www.eli-beams.eu
9
Chapter 1 Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Author: Ing. Bedřich Rus, Ph.D.
Introduction The ELI laser system will be a complex facility that will be providing laser pulses for all research activities. The baseline design of the ELI laser system features multipurpose mission of this facility in fundamental and applied research, as well as versatility of its operation and of user (access) research. The laser pulses with various levels of energy will be delivered into the individual experimental areas by a versatile switching system. The ELI laser, as designed by the ELI-Preparatory Phase consortium, is a complex instrument consisting of three main sections (pillars), as shown in Figure 2: a)
Attosecond facility
b)
Beamlines facility
c)
High-intensity facility
Figure 1: Relationship between the individual Research Activities of this project. The laser systems that are subject of the Research Activity 1 will, as a backbone instrument of the ELI facility, deliver high-intensity pulses for the Research Activities 2 to 6 and will enable their realization.
These facilities will be optimized to provide specific combination of laser pulse energy, repetition rate, and peak and/or average power. Each of these facilities represents a major advancement compared to the currently available laser facilities in the world; implementation of each of these facilities has also its own technology challenge. Following recommendation of the ELI-Preparatory Phase consortium, implemen-
tation of the attosecond and beamline facilities will be a priority of the project with delivery date 2015; simultaneously, vigorous technology development shall be carried out in international collaboration with the goal to prototype and test critical components for the high-intensity facility before 2015. The project of ELI facility in the Czech Republic reflects all these recommendations of the international ELI-Preparatory
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1 100 mJ but will use other amplification techniques (especially cryogenic multislab amplifiers) to provide ultrashort petawatt-class pulses with up to 50 J of energy, at a repetition rate of up to 10 Hz. As emphasized by the ELI-PP consortium, ideally all the beamlines should run at the 10 Hz repetition rate, in order to enable ELI to become, amongst other, a highly competitive source of accelerated electrons or protons for applications.
Figure 2: Baseline scheme of the laser ELI laser facility, as resulted from works and analysis carried out within the ELI-Preparatory Phase project, and as approved in the ELI-Preparatory Phase Mid-Term Report (MTR) submitted to EC in July 2009.
Phase consortium. The attosecond facility will be largely based on the Petawatt Field Synthesizer (PFS, description see below) technique optimized for delivering ultrashort pulses with duration of ~5 fs and operating at a repetition rate of typically kHz. This facility will be composed from a single (or
several) petawatt-class systems, and will have a specific mission in generation of coherent XUV and X-ray pulses in the attosecond time domain for fundamental research and applications. The ELI Beamlines Facility will equally exploit the PFS technology in the front end up to a few
The high-intensity facility is designed to deliver a sum peak power of about 200 PW (0.2 EW), with nominal repetition rate 0.1 Hz or less. The primary goal of this facility will be to provide pulses for fundamental research in frontier (exotic) physics, using focused intensities of ≼1024 Wcm-2 which i s approximately 100x larger value than those available at current cutting-edge laser facilities and entering the ultrarelativistic regime. The high-intensity facility will consist of laser blocks providing 10 to 30 PW. The design
ELI: Extreme Light Infrastructure www.eli-beams.eu
13
options for laser systems, as well as techniques of coherent combination of their pulses, will be prototyped and tested within this project. Another part of ELI considered since recently and not represented yet in the Mid-Term Report scheme as shown in Figure 2, is the photonuclear facility using a 10-PW class laser system to explore nuclei with high energy photons and for studies and development of new techniques of gamma-ray generation. This facility will likely use a flashlamp-pumped laser with repetition rate about 0.1 Hz, and will use some technologies common to those that will be implemented in the high-intensity facility.
The schematic layout of the laser system of Beamlines Facility is in Figure 3. The laser technologies of the individual sections of this system are described below.
Figure 3: Schematic layout of the ELI laser of the Czech Republic’s facility. Its primary mission will be to operate as the ELI Beamlines Facility. The two 10-PW laser blocks will serve to development of the high-intensity section and to testing of selected technologies. Pulse compressors are not represented in this scheme.
The laser front end will consist of the oscillator section (three optically synchronized identical
oscillators) delivering ~5 fs pulses with >300 nm equivalent bandwidth, pre-amplified by the PFS technique to approximately 10 mJ level. The front end will supply pulses into booster
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1 repetition-rate amplifiers based on the PFS technique and pumped by thin disk lasers currently prototyped at the Max-Planck Institute for Quantum Optics (MPQ) in Garching and at the Max-Born Institute (MBI) in Berlin. The system will involve two types of the booster amplifiers: the first type running at 1 kHz and providing equivalent energy of 200 mJ in the compressed pulse, the second type running at 100 Hz and delivering equivalent energy of about 1 J in the compressed pulse. The first type will deliver pulses directly to the experimental areas, whereas the second type will feed the beamline power amplifiers and the 10-PW test laser blocks. The beamline power amplifiers will be based on the OPCPA technique, driven by repetition rate diode pumped solid-state lasers (DPSSL) at a frequency of 10 Hz. The OPCPA technique is preferred as nominal solution due to its large bandwidth and a possibility to provide high out-
put energy at a significant repetition rate. The DPSSL pump systems for the beamline power amplifiers will be a critical component for successful delivery of this research activity. These pump systems will be based on a multi-slab cryogenic amplifiers and will represent today’s cutting-edge technology. Development and optimization of these lasers and associated technologies is to be accomplished, partially within this project. Given that 2015 is the commissioning date of the facility.
J and one 50 J beamline will be dedicated to the target area for electron and acceleration, and will be compressed to provide flexible pulse lengths between about 50 and 200 fs.
The designed laser system will involve two types of repetition-rate beamline power amplifiers, both operating at 10 Hz. Two identical beamlines of each type will be installed in the facility. The first type will deliver approximately 10 J of compressed energy, implying requirement for ~100 J (at the fundamental wavelength 1 Âľm) pump pulses. The second type will provide about 50 J of compressed energy and will thus require ~500 J pump pulses. One 10
The designed ELI laser involves two blocks of the future high-intensity end. Each of these nominally 10-PW (potentially 20-30 PW) blocks will provide at their output compressed pulses with energy of 200-300 J and pulse duration of 20-30 fs, with repetition rate of approximately 0.1 Hz. Two amplification technology options, currently examined by ELI-PP partners, exist for these blocks: OPCPA architecture that will
The fallback solution for the beamline power amplifiers will be Ti:Sapphire technology using as pump systems conventional Nd:YAG flashlamp lasers (or Nd:GGG heat capacity flashlamp pumped lasers). This technology matured and is commercially available. If used here, the beamlines will provide repetition rate of 0.1 Hz.
ELI: Extreme Light Infrastructure www.eli-beams.eu
15
be tested at the Rutherford Appleton Laboratory (10 PW Vulcan project) and Ti:Sapphire architecture that will be tested at the Institut de Lumiere Extreme (Apollon 10 PW project). The pumping system in both cases will be flashlamp pumped lasers (with the active material of Nd:glass, Nd:GGG, or Nd:YAG ceramics). The mission of the 2x10 PW laser section will be twofold: to develop and test technologies of multi-10-PW lasers such as coherent beam combination, and to deliver focused intensity of 1023 Wcm-2 or higher to perform experiments in the field of exotic physics with adaptable beam geometries as required for specific high-field experiments.
will be possible operatively anytime both during implementation of the designed facility before 2015, or later.
The architecture of the designed ELI facility allows straightforward upgrade of the high-intensity section by providing space for implementation of up to 6 high-intensity laser units and compressors (see also Figure 4). This upgrade of specifications of the high-intensity section
d) High intensity 2x10 PW section, including pulse compressors.
The layout of the laser shown in Figure 3 is reflected in the structure of the budget of this project see Chapter 3 of ESOP. The budget of the laser consists of four parts: a) Front end including the oscillator section and the three booster amplifiers; b) Beamlines with output energy 10 J, including pulse compressors; c) Beamlines with output energy 50 J, including pulse compressors;
Scheme of implementation of the laser in the designed building in the district of Dolnà BřeŞany
is shown in Figure 4. The individual laser systems are located in the ground floor of a monolithic structure ensuring high level of mechanical and thermal stability. The laser area is divided into four subsections, hosting the front end a booster amplifiers, 10-J repetition rate beamlines, 50-J repetition rate beamlines, and highintensity amplifiers. The laser front end (block of oscillators and preamplifiers) will be located in an enclosed compartment in the first subsection. The layout features easy maintenance and economy of operation, and possibility for future upgrades. Overhead the laser systems in the first elevated floor are pump and supporting systems (drivers of diode lasers, flashlamp pump lasers, capacitor banks, cryogenic systems, etc.) The final compressors of the 10-PW laser blocks are located below the amplifiers in the underground floor which provides flexibility of delivery of the large-aperture high-intensity beams into the Exotic physics and Particle acceleration experimental areas.
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1
Figure 4: Designed implementation of the ELI Beamlines Facility in the Czech Republic, with indicated locations of the research activities exploiting the laser. The laser systems, including oscillator & front end, repetition rate beamlines and a test high intensity system are located in the ground floor, the driving and pumping systems (DPSSL drivers, pump lasers, capacitor banks, etc.) are placed in the 1st floor.
ELI: Extreme Light Infrastructure www.eli-beams.eu
17
Key research topics The general goal here is to implement the designed ELI laser facility with primary mission consisting in beamline applications and in testing key elements of the future high-intensity facility (with 200 PW projected peak power). The particular key research topics will consist in development, optimization and implementation of those individual components of the ELI laser which cannot be obtained commercially off the shelf or as a custom solution. Delivery of the designed ELI facility represents a significant research and engineering task, with proportion of development and engineering works varying according to a particular part of the laser system. From the point of view of laser development and the scheduled delivery date 2015, the most challenging part of the laser are
the diode-pumped solid-state lasers which will be the backbone of all the projected repetitionrate beamlines. International cooperation in development of these systems will be of crucial importance for success of this project. Key research topics which are necessary for delivery of the designed ELI laser include: 1. Implementation and optimization of the femtosecond high-contrast front end This part of the laser chain includes broadband oscillators (5 fs) and preamplifiers (about 10 mJ). The oscillators will largely employ commercially available technology; the preamplifiers will use the Petawatt Field Synthesizer amplification technology currently prototyped at the Max Planck Institute in Garching (Germany).
2. Booster amplifiers: femtosecond OPCPA high rep-rate systems pumped by thin disk lasers The booster amplifiers will provide peak power of typically 200 TW and will amplify the front end signal at a repetition rate of 100 Hz (two units providing input signal for the beamlines and the high-intensity block) and 1 kHz (one unit providing pulses directly for applications in molecular, biomedical, and material sciences). The booster amplifiers will be based on the PFS amplification technique and will be pumped by thin disk lasers providing 1-2 ps long pulses. These thin disk pump lasers for both repetition rates will be based on the architecture currently developed at the Max Planck Institute in Garching and at the Max Born Institute in Berlin, constituting today’s cutting-edge technology; collaboration on prototyping and optimization of these lasers will be the subject of this ELI project.
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1 3. Repetition-rate multi-10-J beamlines pumped by slab DPSSL systems
4. High-intensity beamlines (10 PW and multi-10-PW)
5. Control systems of the laser chains and of pulse distribution
The laser design involves two types of repetition beamlines, providing respectively 10 J and 50 J in the compressed pulse, both at repetition rate 10 Hz. The amplification technique will be OPCPA. The most critical element of the beamlines will be DPSSL pump systems providing respectively 100 J and 500 J at the fundamental harmonics (1.05 microns for Yb:YAG ceramic). These pump lasers have been conceptually designed at the Rutherford Appleton Laboratory (UK), based on the technology demonstrated by the Mercury laser at the Lawrence Livermore National Laboratory (60 J with 10 Hz rep-rate). The architecture uses a stack of variably doped cryo-cooled Yb:YAG ceramic slabs, face-pumped by laser diode arrays. Extensive international collaboration, especially with the Rutherford Appleton Laboratory, on prototyping and optimization of these pump lasers will be one of the key points.
Two options for the ELI high-intensity pillar are suggested: OPCPA amplification and Ti:Sapphire amplification techniques, providing peak power of at least 10 PW per beam, i.e. 200-300 J in 15-20 fs pulses. Both options will use as the driver Nd:glass flashlamp-pumped lasers (or, alternatively. Nd:GGG heat capacity flashlamp pumped lasers). As the primary option, OPCPA technology will be used to build two nominally 10-PW units within this project, in collaboration with the Rutherford Appleton Laboratory which is currently establishing a 10 PW OPCPA-based laser facility. The secondary option for these 10PW units are Ti:Sapphire amplifiers. The option that will ultimately be selected for the designed laser will be made by end 2012. Amongst the research topics that will be carried out in connection to these lasers will be beam diagnostics and diagnostics of PW and multi-10-PW pulses, which currently do not exist.
Control systems of the lasers and of delivery of the laser pulses to the required experimental areas will be a major element of efficient operation of the facility and of its flexibility. The laser operations will be controlled and supervised from control rooms that will feature efficient interfacing with control rooms of the individual experimental areas. Works on this research topic will involve mostly engineering development activities, involving system engineering due to complexity of the facility. The control systems will be based on state of the art industrial standards in automation engineering, ensuring low susceptibility to faults and high operation reliability. Modular PLC (programmable logic controller) systems and distributed input-output peripheral systems located in the vicinity of associated sensors and actuators will be used. Data acquisition and graphical user interface will be based e.g. on the LabVIEW platform.
ELI: Extreme Light Infrastructure www.eli-beams.eu
19
6. Compression of multi-10-PW ultrashort pulses A specific research and development task that has to be addressed in implementation of the ELI laser is compression of 10-PW and multi10-PW laser pulses delivered by beams with apertures as large as 400 to 600 mm (or even larger). In order to compress broadband few hundred picosecond to nanosecond pulses down to 10-15 fs, close to their bandwidth limitation with low pedestal, the required bandwidth of the compressor has to be at least 150 nm. Compressors for such large beams and operating with such broadband constitute a unique technological and engineering task. Due to the beam size, large compression gratings (~1 m diagonal) are required, while due to the bandwidth the gratings must have relatively low line density (~900 lines/mm). Compressor for 10PW large aperture beams has been designed at the Rutherford Appleton Laboratory (UK), and a supplier of the required grating (with gold layer)
was identified. Another type 10-PW compressor will be developed at the Institut Lumiere Extreme (France), which is furthermore involved in development of new high damage threshold (~J/cm²) large bandwidth gratings using metaldielectric layers. An international cooperation with the two above-mentioned institutions and other partners in Germany (University of Jena) will be an important element for successful delivery of the compression units at the facility. 7. Techniques of coherent superposition of multi-100-J femtosecond pulses Implementation of the full scale high-intensity ELI pillar (6 to 12 multi-10-PW beams) will require to coherently superimpose the individual ultrashort pulses in the time domain in order to generate through their constructive interference the 200 PW design power. Another technique that has to be developed for full scale highintensity ELI pillar is the accurate spatial super-
position of the individual beam foci. Although coherent temporal superposition of low-energy femtosecond pulses has been demonstrated, superposition of large beams with multi-10-PW power will require developing new techniques. The systems for phase control and temporal superposition of the compressed beams will exploit the existing well-proven technology of adaptive optics in astronomy, especially from technology of large push-pull/tip-tilt mirrors and actuators. The large adaptive optics will operate in closed loop with sensors of coherent combination of the pulses. A leading candidate for measurement of the temporal and spatial superposition of femtosecond beams is interference pattern detection.
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1
Directions of implementation at ELI Implementation and optimization of the femtosecond highcontrast front end As mentioned above, the oscillator section is designed to involve three optically synchronized oscillators providing effectively 5 fs equivalent bandwidth. The oscillator signal will be preamplified at 1 kHz first in a Ti:Sapphire preamplifier (regenerative or multipass type) and subsequently in PFS preamplifiers. Oscillators providing ~300 nm bandwidth are nowadays available commercially (e.g. Octavius-85M 6 fs Ti:Sapphire oscillator, Menlo Systems Inc.); another alternative of a 5-fs oscillator is the approach developed at MPQ and described below.
Figure 5: Nominal scheme of the oscillator unit for the designed ELI facility. The output produced by a commercially available broadband oscillator (âˆ†Îť>300 nm) is pre-amplified to 10-mJ level and is used to produce mutually synchronized seed and pump input pulses of the booster amplifiers, as well as 1 kHz optical synchronization signal. The oscillator section will consist of three identical units optically synchronized.
ELI: Extreme Light Infrastructure www.eli-beams.eu
21
The oscillator / front end solution currently developed at MPQ where the required ~300 nm bandwidth is generated by frequency broadening the output of commercial Ti:Sapphire oscillator in a photonic crystal fiber (PCF), is shown in Figures 6 and 7. The wavelength shifting, accomplished by soliton self-frequency shift in the fiber, serves to adjust the wavelength of the pulse that feeds the thin-disk DPSSL pumping laser chain (see Figure 7). The current system operates at frequency of 70 MHz, with energy between 10-20 nJ per pulse. The Petawatt Field Synthesizer technique (PFS) is OPCPA-based system using high-repetition rate DPSSL pumping. The pump laser, which is based on the thin-disk architecture, produces pulses of the order of ~1ps, with typically kilohertz repetition rate. These pulses are used to pump KDP or DKDP crystals acting as OPCPA amplifiers. Due to the thin OPCPA amplifiers, the PFS technique possesses inherently large
Figure 6: Scheme of the oscillator unit applicable for ELI, developed at MPQ Garching. The pulses produced by Ti:Sapphire oscillator are spectrally broadened and wavelength shifted in a photonic-crystal fiber (PCF), upon which the signal is amplified in a commercial Yb:glass fiber amplifier (14 nJ pulses, 1 W at 70 MHz).
bandwidth while the ps pumping allows achieving high pump intensities with low sensitivity to damage issues, and reduces also contribution of the fluorescence in the output signal, thereby ensuring a high contrast. PFS represents an advanced approach of the OPCPA technology. It benefits from availability of large diameter KDP or DKDP crystals and en-
sures high-fidelity re-compression of the amplified pulse, as the higher order dispersion terms introduced by the stretcher and the compressor can be better matched for low-compression ratio (ps to fs, rather than ns to fs as for the “conventional� OPCPA). The PFS beamlines have to propagate in vacuum in order to avoid deterioration of the B-integral by picosecond pulses in the air).
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1
Figure 7: Scheme of the ELI oscillator and front end section as being currently prototyped at MPQ Garching within the ELIPreparatory Phase project. The current frequency of 10 Hz will be boosted to 1 kHz by using thin-disk systems that are currently developed (courtesy of Stefan Karsch, MPQ).
ELI: Extreme Light Infrastructure www.eli-beams.eu
23
Booster amplifiers: femtosecond OPCPA high rep-rate systems pumped by thin disk lasers The PFS technique is due to its bandwidth a prime candidate both for the ELI booster amplifiers. The key issue for achieving the designed output energies is the pumping system operating at 100-200 Hz or at 1 kHz, based on the thin-disk technology. Collaboration on development of these systems with MPQ Garching and MBI Berlin will constitute a major activity of this key research topic. Figure 8: Generic scheme of the PFS technique using OPCPA with the amplified pulses stretched to typically 1-2 ps.
Principle of the thin-disk DPSSL systems is in Figure 9. One of the currently demonstrated
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1 Figure 9: Principle of the think disk repetition rate amplifier (courtesy of P.Nickles, MBI Berlin). The pumping diodes delivering ~1 ms pulses operate either at the required frequency (for 100-200 Hz systems) or in CW regime (for operation at 1 kHz or higher frequencies).
systems (MPQ) provides 25 mJ in compressed pulses, at repetition rate 3 kHz (see Figure 10). The system operates in the regime of a regenerative amplifier. This system exhibits very high stability both pulse-to-pulse (0.7% rms) and in long term operation. An up-scaled system that will operate as multipass amplifier (20 passes) is being presently implemented at MPQ. It will provide about 300
mJ in the compressed pulses at repetition rate 10 kHz and will be pumped by 3 kW of CW pump power. The pump region will be increased to 8.2 mm. This thin-disk system, after its completion and testing in 2010, will be directly relevant to the considered booster amplifiers of the designed ELI facility. By straightforward, low-risk scaling of this system, thin-disk delivering 2 J at repetition rate 1 kHz will be obtained (see Figure 11). Participation on development of this
Figure 10: Thin-disk based laser system prototype developed in MPQ Garching for pumping PFS amplification chains (courtesy of T. Metzger and G. Korn, MPQ Garching). The system provides 25 mJ in compressed pulses (1.6 ps) at repetition rate 3 kHz and is CW pumped by laser diodes delivering 280 W (5.2 kW/cm²). The size of the output beam is 2.5 mm. The system exhibits excellent stability of operation and diffraction limited beam quality (M2=1.1).
system, in collaboration with MPQ Garching and MBI Berlin, will be undertaken as a major part of this key research topic.
ELI: Extreme Light Infrastructure www.eli-beams.eu
25
Repetition-rate multi-10-J beamlines pumped by slab DPSSL systems
Figure 11: Scheme of thin-disk head providing 2 J at the frequency 1 kHz (left), with optical pump homogenizer for 15 kW pump power (right). The pumped region of the crystal is 18.6 mm. Courtesy of G. Korn and S. Karsch, MPQ Garching.
DPSSL technology is a prime choice for the realization of the designed ELI repetition-rate beamlines. The beamline mission of ELI requires a laser source delivering energies of 10 J and 50 J on target in a sub-20 fs pulse at a repetition rate of up to 10 Hz. The final amplifier of this source, be it Ti:Sapphire or OPCPA, will require a pump laser delivering multi-100J pulses of few-ns duration at around 500 nm. These demands can be best met by a diode pumped solid state laser delivering around 500 J at the fundamental wavelength of ~1Âľm. Different options for various aspects of the final
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1 amplifier, the most challenging part of such a laser, have been evaluated by Rutherford Appleton Laboratory and form a part of the ELI-Preparatory Phase Mid-Term Report, submitted in July 2009. Following these considerations a concept for an amplifier is described that is thought to present the best compromise between the different constraints and requirements. a)
Gain Material
The requirements for the gain material are: • Availability in large size to handle high energy. • Good thermo-optical and thermo mechanical properties: to handle high average power. • Sufficiently low saturation fluence: to facilitate energy extraction.
• Long fluorescence lifetime: to minimize number of pump diodes required. Materials that have been suggested are Nd:Glass, Yb:CaF2 and ceramic Yb:YAG. Any glass-based material is not suitable due to the very poor thermal properties. In the case of Nd:Glass, the short fluorescence lifetime is another disadvantage. Yb:CaF2 suffers from a very high saturation fluence which makes both pumping and efficient energy extraction while avoiding optical damage very difficult. Also, the growth of sufficiently large crystals with good optical quality remains yet to be demonstrated. This leaves ceramic Yb:YAG as the most promising candidate. The required pump and extraction fluence levels for efficient operation are still quite high at room temperature, but this can be overcome by cooling the medium to cryogenic temperatures around 150 K.
b)
Amplifier geometry and cooling
The two following considerations apply with regards to amplifier geometry and cooling: • For effective cooling, the amplifier medium needs to have a high surface-to-volume ratio • To minimize wave front distortion caused be temperature gradients, heat flow must not be perpendicular to the beam propagation direction; ideally it should be in the same direction. This leaves us with a face-cooled thin slab geometry, of which there are currently two variants: the active mirror concept and the gascooled slab-stack architecture as demonstrated in the Mercury laser system (see Figure 12).
ELI: Extreme Light Infrastructure www.eli-beams.eu
27
The active mirror (or thin disk) concept has been very successful for CW and high repetition rate (repetition period << fluorescence lifetime) applications where tremendous average power and efficiency levels have been achieved. In these applications both the amount of energy stored in the gain medium (and hence the gain) and the peak intensity are not very high. This is not the case for high energy amplifiers and this leads to two severe problems: â&#x20AC;˘ Because of the high aspect ratio of the gain medium, the (parasitic) transverse gain is very much higher than the (useful) longitudinal gain. To avoid high amplified spontaneous emission (ASE) losses, such an amplifier can only be operated at a very low gain, which in turn means that many amplifier heads are required to reach required output pulse energy. This leads to a technically very complex system and to increased optical losses. Due to the quasi three level
nature of Yb:YAG, efficient operation with low gain is also only possible at very low temperatures. â&#x20AC;˘ Since the input and output pulse overlap with each other in the gain medium and at the surface, the optical intensity is increased up to a factor of four, with the according consequences for optical damage and nonlinear effects.
This leaves the Mercury-type architecture as the most promising one, since the aspect ratio of the gain medium can be chosen freely without compromising cooling. Also, the concept should be applicable to any temperature and one does not need to find a medium that is liquid in the chosen temperature range, as would be the case for the active mirror concept.
Figure 12: Mercury DPSSL repetition rate demonstrator built at the Lawrence Livermore National Laboratory (USA). The system provides 62 J at the repetition rate 10 Hz, in ~10 ns pulses, and provides beam quality of 4x of the diffraction limit.
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1 c)
Amplifier Concept
The basic structure of the amplifier is outlined in Figure 13. The amplifier medium consists of a stack of quadratic ceramic Yb:YAG slabs. Cold helium gas is forced through the gaps between the slabs for cooling. The amplifier is end or face-pumped from both sides. Employing slabs with increasing doping level towards the centre of the amplifier reduces the required overall thickness for a given maximum gain coefficient and also equalizes the heat load for all slabs.
Figure 13: Schematic of the design of DPSSL high-energy amplifiers studied at the Rutherford Appleton Laboratory (courtesy of J. Collier and K. Ertel, RAL).
ELI: Extreme Light Infrastructure www.eli-beams.eu
29
d)
Design parameters
Size-independent parameters Pumping of the amplifier has been simulated with a spectrally and temporally resolved 1-D code. The parameters used are listed in Table 1. Pump intensity (each side)
5 kW/cm2
λpump
~ 940 nm
∆λpump
5 nm FWHM
Pump duration
1 ms
Temperature
175K
Table 1: Pump modelling parameters for 1-kJ amplifier DPSSL multislab head.
The pump intensity was chosen as it seems to represent a value that can realistically be achieved and maintained over a certain depth of field with the brightness of currently avail-
able high power diodes. The same applies to the pump spectral width. The duration was chosen equal to the fluorescence lifetime which offers a reasonable compromise between fluorescence losses and required diode peak power. The operating temperature was chosen because spectrally resolved absorption data is available for this value. The main quantity calculated by the numerical model is the pump efficiency which is extractable fluence divided by pump fluence. The losses considered are fluorescence, quantum defect, un-absorbed pump and minimum population needed (due to quasi 3-level nature). For the given parameters quantum defect and fluorescence limit the efficiency to 58%. The other two loss mechanisms need to be balanced off against each other and an optimum is found for a certain optical depth (doping concentration times thickness) of the amplifier medium.
Using absorption data published in [2] and assuming a lower laser level population of 0.64%, the results listed in Table 2 are obtained. Maximum pump efficiency
50.2%
Optimum pump wavelength
939 nm
Optimum optical depth (doping 3.15 % concentration times) cm Table 2: Modelling results for 1-kJ amplifier DPSSL multislab head.
Despite the non-ideal pump spectrum and modest pump intensity, an efficiency quite close to the theoretical maximum is achieved. Assuming a stimulated emission cross section of 5.2x10-20 cm2, the resulting small-signal gain is 3.8.
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1 Aperture scaling Nothing is said about transverse size and aspect ratio of the gain medium in the previous section. The aspect ratio is governed by ASE management requirements and is independent of aperture size: if the aperture is increased, thickness needs to be increased by the same factor and the doping concentration lowered accordingly in order to maintain the optimum optical depth. This way the transverse gainlength product (small signal gain coefficient times aperture size, called transverse gain in the following) remains constant. A rule of thumb says the transverse gain should be kept below 3 for all possible paths. For an amplifier with a square cross section, the highest transverse gain is found along the diagonal across the outer face of the amplifier. For a constant doping level and a maximum
transverse gain, the required aspect ratio is 0.74, meaning the gain medium (excluding cooling channels) has to be thicker than the beam width. The thickness can be reduced by using different doping levels for the individual slabs in the amplifier. Using ten slabs of equal thickness with an optimised doping profile, the aspect ratio can be increased to 1.4. If a higher maximum transverse gain turned out to be acceptable, the aspect ratio would increase proportionally.
Refinement of model The parameters presented in the previous section present a point design that may not be accurate (due to incorrect spectroscopic data) and is probably not optimum either. It is also possible that with progress in diode laser technology, the parameters listed in Table 1 can be improved upon. Further numerical and especially experimental investigations are therefore required.
The required aperture size for a given output energy is mainly governed by the stored energy. The required aperture size for 1 kJ output is 14x14 cm2 and the required total thickness 10 cm. It is unlikely, however, that the total stored energy can be extracted without losses, hence it is either necessary to increase the amount of stored fluence by stronger pumping or to employ more than one amplifier head in series.
ELI: Extreme Light Infrastructure www.eli-beams.eu
31
Risks and disadvantages Currently the apparent risks and drawbacks of the presented concept are: â&#x20AC;˘ Cooling system and concept has unknown
Technical complexity
Cost
Effect on transmitted wave front
â&#x20AC;˘ YAG has a high nonlinear refractive index, resulting in high B-integral in large aperture amplifiers (as amplifier thickness scales linearly with aperture size). For a 1 kJ system however, a tentative calculation of the amplification process yields a B-integral value of 0.7 for a 4 ns pulse, which is quite acceptable.
High-intensity beamlines (10 PW and multi-10-PW) As mentioned above, OPCPA architecture option is considered as nominal for the two designed 10-PW blocks; the second solution is the architecture option based on Ti:Sapphire. The final decision on implementation of particular technology will be made by end 2012 following prototyping works carried out at the Rutherford Appleton Laboratory (Vulcan 10 PW OPCPA project) and at the Institute Lumiere Extreme (Apollon 10 PW Ti:Sapphire project). The repetition rate will be at least 1 shot per min, ideally 0.1 Hz. The design based on the 10 PW Vulcan OPCPA project proposes to use 6 beam lines for fullscale ELI to deliver a total of 200 PW to target. The architecture proposed by Rutherford Apple-
ton Laboratory is based on optical parametric chirped amplification. OPCPA has already been demonstrated at significant power levels. A number of different combinations of pulse duration and energy that could be envisioned to achieve this ranges from 10 fs and 300 J to 30 fs and 900 J. Shorter pulse duration has implications for the seed source and the longer pulse durations conversely have implications for the pump laser. The design to achieve these peak powers is based on the parameters in the Table 3 below. The proposed seed generation scheme is shown in Figure 14. As can be seen for the seed generation scheme, OPA is used to generate the pulses at 900 nm, and is seeded and pumped by the output of a single oscillator ensuring good optical synchronicity. The architecture for the power amplification stages is shown in Figure 15.
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1 Parameter
Value
Assumption
N [mm-1]
900
900 line / mm gratings available: avoid -1st order over BW
Size [mm, DxND]
~
Pump Energy [J]
1200 green
Ein [J]
480
Assumes 40% OPCPA efficiency
O/P Fluence [mJcm-1]
170
Long pulse LIDT > Short pulse LIDT
θin [deg]
16.9
∆θ of 14° is ok for efficiency
Separation [mm]
3670
To separate i/p and o/p beams
∆λ out [nm]
123
145 nm FWHM 3rd order super Gaussian input. Grating bandwidth supports this
O/P Fluence [mJcm-1]
102
No difference between gold and di-electric in short LIDT
ηnet [%]
60
Assumes 4 Gold Gratings @ 90 % efficient per grating (65%)
Eout [J]
290
t compressed [fs]
15
Power [PW]
~
530 x ~530
Grating availability
Fourier Transform Limit
19.2
Table 3: Key parameters of the design of 10-PW block for the ELI high-intensity (from ELI-PP Mid Term Report).
ELI: Extreme Light Infrastructure www.eli-beams.eu
33
Figure 14: Schematic for the architecture to generate the seed pulses at the wavelength 900 nm, with >150 nm bandwidth.
Figure 15: Architecture of the 10-PW OPCPA chain designed at the Rutherford Appleton Laboratory, showing the power amplification stages.
The choice of DKDP for the power amplification stages makes the requirement on the seed that it must be cantered at 900 nm. There are a number of schemes that can be used: direct seeding at 900 nm, although a suitable seed
source has not been found, or using nonlinear process. Two alternative schemes have been demonstrated of generating ultra-short pulses at 910 nm for amplification in DKDP. One scheme uses a Cr:Fosterite laser operating at
1250 nm to seed an OPA pumped at 527 nm to generate 910 nm. This scheme has demonstrated sufficient bandwidth for 45 fs pulses. The alternative is to use a chirp compensated OPA scheme seeded at 710 nm and pumped at 400 nm. This scheme has the advantage that the seed and pump pulses can be derived from the same source a broad bandwidth Ti:Sapphire laser; consequently, both pulses are highly optically synchronized. This scheme has been demonstrated to have sufficient bandwidth to generate a <15 fs pulse. The chirp on the pump beam enables the broadband amplification by relaxing the phase matching requirement that the pump and signal beam must be phase matched for all wavelengths simultaneously, because it is chirped and the instantaneous phase matching at the pump wavelength is relatively narrow, the fluorescence in the idler beam will also be chirped. Since the chirp of the idler and the chirp of the fluorescence is in the same direction when the pulse is compressed, the fluorescence will be as well.
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1 We see this as an essentially solved problem with a suitable seed source now available. If greater bandwidth is required then white light generation in a PCF might be employed to generate a large bandwidth. a)
Energy Requirement
To achieve the required energy onto target requires relatively large beams in the OPA stages; consequently, this limits the choice of nonlinear crystal. There are three potential candidates that can be grown in sufficient size: KDP, its deuterated isomorph DKDP and YCOB. Of the existing high peak power designs DKDP has been the most studied and pursued. Calculations have demonstrated that an OPA system based on a LBO preamplifier system and DKDP power amplifier stages will have sufficient bandwidth to support 15 fs pulses. As the size of the gain crystal is increased,
care has to be taken that the transverse gain is insufficient to lead to parasitic lasing or significant gain depletion due to ASE. This problem is one that Ti:Sapphire is prone to. On the contrary, due to the phase matching conditions of the OPA process there is no transverse gain and consequently no depletion of the gain due to ASE. Parasitic OPO does become a problem for gain stages with large small signal gains, but this can be mitigated by using crystals that have wedged faces preventing reflections experiencing multiple passes during the gain duration of the pump pulse. b)
Pump Beam Requirement
The principal requirements of the pump beam are to have an approximately Top-Hat profile in the temporal and spatial domains within the nonlinear crystal. Optical Parametric Amplification is relatively intolerant to the pump wavelength. This means that conventional Nd:glass lasers
(位=1053 nm/ 527 nm) can be used to pump large aperture OPAs or with sufficient development Yb:X (位=1030 nm/ 515 nm). High energy Nd:glass lasers with sufficient energy to achieve the required pump power have been developed and demonstrated, though with the current and projected technology they have repetition rates on the minute timescales. Nevertheless, recent development of diode pumped Yb:Host and even Nd:Host laser has made significant advances in laser pulse energy generation. At present they have been limited to the ~100J region, but it is expected with the added interest from other laser development programmes that systems will be designed to be capable of delivering 1kJ of energy at the fundamental on a repetition rate measured in seconds or a fraction thereof. The calculated phase matching angle for DKDP for these two pump wavelengths is shown in Figure 16. As can be seen, both pumping wavelengths enable a broad bandwidth to be achieved, with the phase matching decreasing more rapidly for the longer wavelengths. The gain centre for
ELI: Extreme Light Infrastructure www.eli-beams.eu
35
the 515 nm pump also appears to be shifted slightly to shorter wavelength with respect to the 527 nm.
c)
Contrast Limitations
On the nanosecond timescale the contrast is limited by the amount of parasitic parametric fluorescence (PPF) that is generated in the amplifier stages. This will be generated for the duration of the pump pulse within the OPA stages. PPF can be reduced by ensuring that as much gain is achieved for shorter stretched pulse durations and injecting a clean seed pulse into the power amplification stages. The amount of PPF that might impinge onto target can be estimated by: The impact of temporal modulations on the pump pulse has recently come to the attention
Figure 16: The calculated phase matching angle for non-collinear OPA pumping at 515 nm (solid) and 527 nm (dashed). From ELI-PP Mid-Term Report.
of a number of authors. Temporal modulations of the pump beam impinge onto the spectral amplitude of the seed pulse being amplified. The cause of the temporal modulations has been attributed to ASE within the pump laser
leading to a coherent noise modulation of the laser output. On compression this spectral modulation will lead to a plateau on the picosecond timescale that has a contrast ratio of the order of 104 to 106 depending on the pump
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1 laser properties. The temporal modulations of the pump beam can be appropriately reduced if the OPA amplifiers are run in saturation, which is likely at the expected energies of the system. Operating the OPA stages in saturation will also reduce the impact of any modulations on the spectral amplitude modulations.
Control systems of the laser chains and of pulse distribution Appropriately designed control systems are one of the key factors in reliable operation of the whole facility and in providing effective environment for its mission in research and development. Therefore the control systems will be one of the first systems that will be engineered since the beginning of this project, followed by design and engineering of particular components of the ELI laser.
The systems will be based on robust industrial controllers using modular PLC systems and distributed input-output peripheral systems (most likely based on PLC Siemens industrial standard); similar systems are now used in cuttingedge automotive and aerospace applications. The baseline scheme of the control system is displayed in Figure 17. The core of the system consists of an array of PLCs (programmable logic controllers). The master PLC and the slave PLCs are linked by a robust, interference-resistant, system bus (e.g. Profibus). Each PLC is located near a specific technological unit/node (e.g. amplifier head, pulse compressor, cryogenic unit, vacuum chamber, etc.) that has to be individually controlled. The PLC units provide hardware warning signals, which ensures maximum robustness and operation reliability. The PLC units will be equipped with HMI (human-machine interface, e.g. touchscreens, keyboards) for local manual control, and will
also make use of e.g. Labview for data visualization. The local manual control (local peripheries) will be exploited especially for the start-up period, system testing, and maintenance purposes. The local peripheries are linked to the local PLC by means of standard I/O links or standard interfaces. The sensors and actuators (e.g. beam sensors, valve coils, position sensors, limit switches) of the monitored and/or controlled components are linked to the local PLC via distributed I/O units and local bus. This concept ensures robustness of the system and resistance to interferences, minimizes the need of cabling, and allows for expansion of the system. All important data (configuration, working stages, faults, accesses, etc.) are stored in the main data storage unit. The whole control system will be entirely separated from other local control subsystems, LANs, personal computers, etc., and will have a dedicated UPS power backup. This will ensure
ELI: Extreme Light Infrastructure www.eli-beams.eu
37
maximum operational reliability of the control system. In order to facilitate interaction with the control of particular experiments, privileged users will be able to enter the system through a sole external access point. This exclusive access point/gate using dedicated HW and SW tools will protect the control system against external environment. Figure 18 shows the layout of the beam delivery architecture. Two beam/pulse switchyards will be located along the ground floor, while one switchyard system to distribute additionally large-energy pulses will be located in the underground experimental floor. This pulse delivery is one the facility systems that will be remotely controlled and fully automated, and will integrate fundamental beam diagnostic systems. Figure 17: Baseline concept of a system for the control of the laser chains and of pulse distribution into the experimental areas. It is based on robust, state-of-the-art, modular PLC units and distributed input-output peripheral systems ensuring transmission of individual signals to the PLC via interference-resistant serial bus. The system is expandable; selected components can be commanded remotely by e.g. TCP/IP communication.
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1 Compression of multi-10-PW ultrashort pulses
Figure 18: Designed beam delivery implementation of the ELI facility (preliminary version). Appearing in red are two beam switchyards along the laser areas in the ground floor, and one L-shape beam switchyard in the underground floor.
The limitation on the minimum stretched pulse duration is governed by the pump laser. The limit is set by the B-integral of the pump pulse as it is amplified; consequently for pump pulse energies of the low kilojoules the stretched pulse will need to be ~3 ns long. The size of the bandwidth requires that the stretcher is all reflective to prevent chromatic aberration. For the proposed bandwidth the line density of the gratings and the beam polarization also play an important role. Figure 19 shows the calculated diffraction efficiency for different grating possibilities. As can be seen the 900 lines/mm gratings demonstrate a broader bandwidth than the 1100 lines/mm option and the Littrow outof-plane design has the broadest and flattest diffraction efficiency predicting a bandwidth of
ELI: Extreme Light Infrastructure www.eli-beams.eu
39
160 nm with diffraction efficiency greater than 95%. Different designs for achromatic stretcher systems have been demonstrated that they have the possibility of being extended to incorporate the full bandwidth of the proposed system. It is expected that a cavity design will be needed to
generate the required stretch. The energy that it will be possible to deliver to target is limited by the gratings in the compressor. The current proposal is based upon gold gratings; however, it is envisioned that the development of the MLD gratings might have achieved sufficient maturity to provide a better
solution. Although in the short pulse duration laser induced damage threshold gold and MLD gratings demonstrate similar damage threshold levels. The design uses gratings that are 530 mm x 530 mm. A manufacturer capable of producing gratings of this dimension and at the required line density has been identified.
Figure 19: Calculated diffraction efficiency for three different grating options: 900 lines/mm at out-of plane Littrow angle (left), 1100 lines/mm in plane non-Littrow (centre), and 900 lines/mm in plane non-Littrow angle (right).
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1 to 92%. Were these two improvements in grating technology to be made then this would increase the potential output power to 50 PW.
Figure 20: Visualization of the 10-PW compressor design that will be built at as a part of the 10 PW Vulcan project at the Rutherford Appleton Laboratory. The compressor is designed for 900 lines/mm gratings and for out-of diffraction plane Littrow configuration (courtesy of J.Collier, RAL).
However we envisage that larger gratings will become available in the near future. With a size of 920 x 530 mm this should allow an approximately 70% increase in the amount of energy that the compressor is capable of withstanding.
If a MLD or hybrid MLD/gold grating solution can be found then the efficiency of the throughput of the compressor could be increased as well. If we assume 98% diffraction efficiency for the MLD gratings then the throughput will increase
ELI: Extreme Light Infrastructure www.eli-beams.eu
41
Techniques of coherent superposition of multi-100-J femtosecond pulses To attain focused intensities exceeding significantly 1023 Wcm-2 it will be typically required to superimpose several laser beams. Since the duration of the laser pulses is several femtoseconds, temporal superposition of the pulses delivered to the focus will have to be better than one micron in order to achieve their constructive interference. The goal of this key research topic will be to develop and demonstrate techniques for accurate on-line measurement of the mutual phase between several focused femtosecond laser pulses, and to demonstrate techniques of active on-line phase control of 10-PW and multi-10-PW beams. The developed diagnostics will make possible accurate spatial and
temporal characterization of the focus of produced by several multi-10-PW laser beams.
controlling the temporal and spatial field distribution in the focus.
The leading candidate for the detection of superposition of femtosecond pulses in both temporal and spatial domain are interferometric techniques. A specific challenge of development in the context of the proposed ELI facility is the need of a technique that will allow accurate phase detection of non-collinear beam wavefronts delivered at low-repetition rate (0.1 Hz) at a precisely defined target plane. The superposition detector will likely use low-energy high repetition rate (100 Hz to 1 kHz) track lasers propagating along the optical path of the 10-PW beams. The output from the superposition detector will provide signal to the adaptive optics (push-pull/tip-tilt mirrors) that will form the closed loop and will provide the required phase corrections. The developed techniques of coherent superposition of multi-100-J femtosecond pulses will also be potentially used for
Lasers generating repetition-rate ultrashort pulses and multi-petawatt peak powers
Chapter 1
ELI: Extreme Light Infrastructure www.eli-beams.eu
43
Chapter 2 X-ray sources driven by ultrashort laser pulses
Authors: Ing. Tomáš Mocek, Ph.D. / Ing. Michaela Kozlová
Introduction One of the main goals within the ELI scientific community is to produce ultrashort X-ray beamlines, both coherent as well and non-coherent, which can open an entirely new field of research in multidisciplinary fields. In fact, the discovery of X-rays by W.C. Roentgen in 1895 [1] has been one of the most important driving forces to push forward understanding and knowledge in numerous scientific areas. Applications range from structure analysis in solid-state, atomic physics and molecular chemistry via imaging applications in medicine and the life sciences to the discovery of the basic building blocks of life, in particular the DNA, and more generally the structure of proteins and other macromo足 lecules. Having access to the spatial resolution of molecular structure and electron orbitals is only one side of the coin to be explored by X-rays. It took about one century to flip the coin to the other side showing the temporal resolution of the atomic and molecular motion, making it possible to monitor the dynamics of molecules and electrons on their natural time scale,
which is now in the attosecond range. The early X-ray generation devices and techniques such as X-ray tubes, electrical discharges or the first synchrotron sources have not been able to deliver X-ray pulses that had a duration of less than several nanoseconds and thus could not be used to gather both types of information, spatial and temporal, simultaneously. An insight into the temporal dynamics of quantum systems was first gained by using probes of much larger wavelength than that of X-rays: namely infrared (IR), visible (VIS) and ultraviolet (UV) laser pulses. Today, the fundamental limit of ultrashort laser pulses is one single optical cycle, lasting about 1 fs in the UV to several femtoseconds in the IR spectral region [2-3]. Having at hand the possibility of monitoring molecular and electronic motion, unfortunately it is not possible to use these lasers to directly image molecular structure at atomic resolution owing to the large wavelength of the laser photons. The principal solution to this problem is to combine X-rays and lasers to take advantage of both the short wavelength
and the temporal coherence properties to create ultrashort pulses of X-ray radiation. Ultrashort, high-intensity lasers can produce very bright bursts of X-rays through the production of very short lived, high temperature laser plasma. Such X-ray bursts may be used to study ultrafast structural dynamics in solids and complex molecules like proteins. In fact, intense attosecond X-ray pulses will make it possible to the movement of electrons in an atom or molecule as it undergoes a quantum or chemical transition. In particular, coherent Xrays present a very important secondary source perfectly suitable for the nano-scale metrology because of its very short wavelength combined with cohe足rence. X-ray interferometry with X-ray lasers can be an excellent method to measure the surface properties with nm-scale resolution, thanks to the fact that X-ray lasers are extremelly monochromatic sources. Possible applications thus include X-ray holography, co-
X-ray sources driven by ultrashort laser pulses
Chapter 2 herent imaging or measurement of fine structures on any solid surface. It may be also used for inspection of defects in lithography masks having thus great impact on the industry. The high-energy/high-peak power lasers developed at the ELI Beamlines Facility will be able to generate complementary X-ray sources such as X-ray lasers, high-order harmonics, X-ray freeelectron-laser (XFEL), K-alpha source, betatron radiation, with extreme parameters far beyond current limits. From the user´s point of view, high repetition rate and high average power will be particularly important.
XFEL, advanced K-alpha sources, betatron radiation, and ultra-high-order harmonic generation in keV region. The output parameters of X-ray sources at ELI Beamlines Facility should be comparable and sometimes even better than the parameters of large-scale XFEL facilities planned world-wide, but on a much smaller scale. The particular advantages of these advanced X-ray sources include: ultrashort pulse duration, beam collimation, spatial and temporal coherence, full synchronization with an ultrafast IR/VIS laser pulses for pump-probe investigations, and extremelly high brightness.
The main goal of this research activity is to provide intense, brilliant, ultra-short X-ray beams for multidisciplinary applications. Since there is no â&#x20AC;&#x153;idealâ&#x20AC;? photon source for all currently imaginable applications, complementary X-ray sources will be developed and optimized. These include new injection-seeded, plasma-based X-ray lasers operating in the water window,
ELI: Extreme Light Infrastructure www.eli-beams.eu
47
Current Status of X-ray Source Development and Applications X-ray lasers Since the first demonstration of lasing action at optical frequencies in 1960 [4], there has been considerable theoretical and experimental effort aimed at the extension of lasing into the X-ray spectral region. To date, lasing at wavelengths between 3 nm and 50 nm has been observed in numerous ionic species [5] and X-ray lasers have the potential to extend current optical laser applications to much shorter space and timescales, while probing much deeper into the core of matter. X-ray time-resolved imaging of biological samples, nanolithography, or laboratory astrophysics are typical examples of applications that are recently starting to be explored as X-ray lasers become closer to reality. The use of X-ray lasers in these novel applications is obviously limited by the parameters of the driver creating population inversion. For the first X-ray laser demonstration nearly 20 years ago, kilo-
joules of laser pump energy were required to generate a laser at ~20 nm [6]. However, year after year, the required pump energy has been dramatically reduced (nevertheless at the expense of output pulse energy) thanks to the improvement of plasma and X-ray laser dynaÂmics. To date, the most energetic collisional X-ray laser [7] has been demonstrated at the kJ-class laser facility PALS [8] in the Czech Republic. It operates in a quasi steady state regime where the driving sequence consists of a weak prepulse followed by a strong main pulse, both being linearly focused on solid Zn target (Fig. 1). The double-pass zinc X-ray laser at 21.2 nm (58.5 eV) delivers up to 10 mJ (1015 photons), 150 ps pulses in a narrowly collimated beam with divergence of about 4Ă&#x2014;6 mrad2. To reduce the pumping energy and increase the repetition rate, transient collisional excitation scheme was proposed and demonstrated [9]. In this scheme, a longer (ns) pulse, which
Figure 1: Typical geometry of quasi steady state X-ray laser using half cavity.
geÂnerates plasma and the required closed-shell ionization balance, is followed by a much shorter (ps) pulse, which rapidly heats the performed plasma and generates a transient population inversion. In 2005, an extension of the transient pumping scheme was demonstrated using a grazing-incidence angle for pumping (GRIP), achieving lasing at 18.9 nm close to saturation
X-ray sources driven by ultrashort laser pulses
Chapter 2 [10]. Similarly to the original prepulse pumping mentioned above, a long pulse (200-600 ps) producing 1012 to 1013 Wcm-2 is first applied to the target, generating plasma with the required (Ni-like) ionization balance. Subsequently, a short pulse (1-5 ps) is injected into the plasma under a small angle, sampling a specific electron density region where it instantaneously raises the electron temperature and efficiently creates population inversion (Fig. 2). Using this approach, strong laser emission at wavelengths down to 10.9 nm was demonstrated [11]. In 2004, an optical field ionization X-ray laser amplifier produced by femtosecond laser excitation of a krypton gas cell was seeded with the 25th harmonic of a Ti:sapphire laser to generate saturated amplification in the 32.8 nm laser line of Ni-like krypton [12]. This seeded X-ray laser provides about 1011 (0.8 µJ) photons per pulse in a narrowly collimated beam with divergence of as low as 1 mrad. Also the amplification of
Figure 2: Generic scheme of the grazing incidence pumping.
19th harmonic of a Ti:Sapphire laser in Pd-like xenon laser at 41.8 nm was demonstrated [13]. By further elaboration of this approach, the injection-seeded X-ray amplifier using a solid target (Ne-like Ti) was demonstrated at 32.6 nm in 2006 [14]. Using the same experimental setup, seeded X-ray lasers at wavelengths down to 13.2 nm [15] have been achieved so far, delivering coherent X-ray beams with excellent parameters (aberration-free, Fourier-limited, polarized) but still with low output pulse energy on a sub-microjoule level only.
The use of extremely powerful lasers developed at the ELI Beamlines Facility as a driver for injection-seeded X-ray lasers presents an excellent opportunity to boost significantly the X-ray laser output up to the mJ level and even more, using the concept of “slab amplifier” consisting in increasing the active surface of the amplifier. This scheme is purely pump dependent, and so there is no principal limit for the output X-ray laser pulse energy. At ELI Beamlines Facility, seeding technique will be extended down to the “water window” (2.2 – 4.4 nm) through pump-
ELI: Extreme Light Infrastructure www.eli-beams.eu
49
of CNTs were retrieved with a spatial resolution of 46 +/- 2 nm, comparable to the illumination wavelength. Finally, the diffractive “lensless” imaging with the spatial resolution of 70 nm was achieved by storing and processing a single diffraction pattern from the object using a phase-retrieval algorithm.
ing energy scaling laws. In particular, energetic Ni-like Au X-ray laser at 3.6 nm operating at high repetition rate can finally become a reality, being an ideal source for X-ray imaging and holography of living cells (Fig. 3) as suggested long time ago [16] but never accomplished. As the seeding technique matures, the X-ray laser beamlines may actually overcome the performances of current and even projected XFEL facilities, i.e. ultra-high X-ray intensities above 1020 Wcm-2 can be available. Regarding X-ray applications, coherent radiation is necessary for interferometry, holography, diffractive imaging or nanopatterning. For such applications free electron lasers or plasma Xray lasers are the best sources. For example, using a 46.9 nm wavelength EUV radiation from a capillary discharge laser a series of experiments was performed [17]. An interferometric lithographic technique was implemented using a Lloyd’s mirror interferometer to pattern a cone
Figure 3: The use of X-ray lasers for 3-D imaging of living cells may become a reality at ELI Beamlines Facility [16].
shaped, 58-nm FWHM diameter nano-dots over the area of 500 x 500 µm2 [18]. The scheme allows also to pattern lines, holes and oval shaped features in PMMA and HSQ photoresists. Holographic imaging in Gabor’s in-line scheme was performed using 50-80 nm diameter carbon nanotubes CNTs as an object. After a numerical reconstruction of the hologram the images
Apart from the above-mentioned applications, sufficiently strong X-ray lasers are the only sources allowing for dense plasma probing and interferometry. X-ray laser interferometry can be used for measurements in material science, metrology and especially dense plasma diagnostic. Two types of X-ray laser interferometry have been applied in practice: amplitude division interferometry (Mach-Zehnder interferometer [19]) and wavefront division interferometry (Fresnel double-mirror [20] or Lloyd’s mirror [21]). Amplitude division interferometry requires high temporal coherence, while wavefront division interferometry requires high transverse coherence. Injection-seeded X-ray lasers elab-
X-ray sources driven by ultrashort laser pulses
Chapter 2 orated at ELI Beamlines Facility will possess both of these features. X-ray laser holography offers the potential to obtain high-resolution 3-D images of biological and other specimens. High-resolution images can be obtained thanks to the high coherence combined with the high brightness of X-ray lasers that enables a single shot record. Exposure times can be reduced to < 1 ns, which eliminates motion blur in biological samples as the dynamic processes in biology have time scales of typically 1 msec. Note that synchrotron X-ray holography and undulator X-ray holography typically needs about 1000 second exposure times from many shots. The first demonstration of X-ray laser holography was carried out using a Gabor in-line geometry and a hologram of a 3-D structure of a gold test pattern was recorded by using the Ne-like Se X-ray laser at 20.6 nm [22]. Since then, unfortunately, no big progress has been done due to the lack of sufficiently strong X-ray laser. One of our challenging missions will be to revive this topic.
X-ray laser microscopy offers the opportunity to observe biological specimens in an aqueous environment [23]. The main goal here is to have lasing at wavelengths near or inside the water window (2.3 - 4.5 nm) where the high contrast in absorption cross section between carbon in biological specimens and oxygen in water can be obtained. However, as current X-ray lasers cannot be routinely produced at wavelengths much shorter than ~10 nm, other wavelength ranges have been employed for X-ray microscopy to date. The high brightness and short duration of X-ray lasers again allow eliminating the of problems associated with motion blur and radiation-induced chemical decomposition of any biological specimen. Researchers at LLNL demonstrated X-ray laser microscopy using a Fresnel zone plate lens and imaged a test pattern with a resolution of 75 nm with a Ni-like Ta X-ray laser at the wavelength of 4.5 nm in 1992 [24], and subsequently rat sperm nuclei were also observed [25]. In this proof-of-principle experiment the X-ray laser had an energy
of 10 ÎźJ in 400 ps and structures with size of about 50 nm were resolved. Since then, few demonstrations of X-ray laser microscopy have been reported until now. One reason is that no saturated X-ray laser at a wavelength inside the water window has been achieved. Moreover, in this particular case the coherence characteristics of X-ray lasers are not so beneficial for microscopy because of the problem of fringes and/or speckle. To stimulate further progress in this field, X-ray laser saturation at a wavelength in the water window is required. X-ray laser radiography is useful for measuring the plasma parameters. The short pulse duration of X-ray laser is of the same scale as the hydrodynamic timescale of laser-produced plasma. The high brightness of X-ray lasers allows them to be used as a backlighter for the imaging bright sources such as high temperature plasmas. The very short wavelength enables the X-ray laser to probe large and near-solid den-
ELI: Extreme Light Infrastructure www.eli-beams.eu
51
sity plasmas. Measurement of the early laser imprint and subsequent Rayleigh-Taylor instability is of great interest in studying direct drive inertial confinement fusion [26]. An important method to measure densities and temperature of plasmas is the Thomson scattering. X-ray laser Thomson scattering has a great potential for probing plasmas at much higher density than optical probing, even above the solid density [27]. A laser entering a plasma is scattered by electrons. The plasma temperature can be measured by observing the spectrum of scattered light and the density of plasma can be measured by observing the amount of scattered light. As the Thomson scattering cross-section is very small, it is difficult to gain a signal above the noise level caused by plasma emission. Because of this fact, X-ray laser Thomson scattering experiments have not succeeded in observing the scattered spectrum so far. To improve the situation, X-ray lasers of higher brightness
and at a shorter wavelength are absolutely necessary.
X-ray free electron lasers Recent advances in laser wakefield accelerators have allowed the production of electron beams with energies ranging from tens of MeV to more than 1 GeV within a centimeter scale, and with pulse durations of several femtoseconds. The enormous progress in improving the beam quality and stability makes them serious candidates for driving the next generation of compact light sources. The versatile laserbased radiation sources range from infrared to X-ray energies. They attract a large user community because they are extremely reliable sources of short radiation pulses tunable up to keV photon energies, making them useful for structural analysis of matter. The development of Free-Electron Lasers has led to an enormous increase of brilliance and coherence. However, accelerator technology currently limits the pulse duration of synchrotron sources just under a
X-ray sources driven by ultrashort laser pulses
Chapter 2 picosecond and demands large and expensive facilities due to the accelerating field gradients of conventional accelerators, which are restricted to 20–100 MV/m by radiofrequency-cavity electrical breakdown. In contrast, a plasma, which is already fully broken down, can sustain electric fields that are 3–4 orders of magnitude higher, exceeding 100 GV/m. The first combination of a laser-plasma wakefield accelerator (producing 55–75 MeV electron bunches) with an undulator in order to generate visible synchrotron radiation, has been successfully demonstrated [28]. This experiment, aimed to demonstrate the production of undulator synchrotron radiation by laser-accelerated electrons, has been carried out using the high-intensity Ti:sapphire laser JETI at Jena. The set-up (see Fig. 4) consists of a laser wakefield accelerator as a source of relativistic electrons, driven by an 80 fs laser pulse with an intensity of 5×1018 W/cm2. The electrons traverse a 50
Figure 4: The principle of laser-plasma based XFEL [28].
periods, 1-m-long undulator with a deflection parameter K = 0.6, and are then analyzed by a calibrated electron energy spectrometer. The measured electron spectrum showed a distinct maximum at 64 MeV with a width of 3.4 MeV (FWHM), and a total charge of 28 pC. The corresponding undulator radiation spectrum was centered at 740 nm and contained 2.84 x 105 photons. Moreover the wavelength scaling with energy and the narrow-bandwidth spectra had been demonstrated.
XFEL generated by using wakefield electrons could provide orders of magnitude brighter Xray radiation (1012 photons/pulse/0.1% BW) than conventional sources. In this scheme, the laser-accelerated electron beam is injected into permanent magnet undulators, shaped into microbunches separated by the resonance wavelength of the magnetic structure; then, as it propagates into the undulator, a burst of bright X-rays is produced from the coherent emission of all microbunches. Owing to the high electron
ELI: Extreme Light Infrastructure www.eli-beams.eu
53
beam peak current generated from laser-wakefield accelerators, saturation lengths for X-ray emission of a few meters can be obtained, offering the opportunity to develop very compact XFEL devices. This is in contrast to the existing huge scale facilities world-wide, which are based on radiofrequency technology acceleration requiring hundred-meter-long undulators and kilometer-long accelerating lines.
K-alpha Laser driven K-shell thermal and Kα sources are currently the most widely applied laserplasma X-ray (defined as photons with energies that exceed 1 keV) sources. These sources are bright, short-lived, quasi-monochromatic and may be well synchronised to a laser. This fulfils the basic requirements for radiography and X-ray scattering, thus the area that has seen widest application for these sources is dense plasma physics. These sources are often used in point projection and area backlighter configurations. Kα sources are monochromatic (∆λ/ λ<10-4) and available with photon energies between 1.48 keV to 68.8 keV from aluminium and gold targets respectively. The emission is fast electron driven. These fast electrons are created during the laser-target interaction, and propagate into the target material and slow down rapidly through bremsstrahlung losses and
collisions. Collisions result in atomic inner-shell ionisation and subsequent de-excitation can result in the emission of a Kα photon of specific and material dependent, wavelength. Bremsstrahlung losses results in continuum emission and ultimately degrade the contrast between the monochromatic source and background. The resulting X-ray pulse is short, lasting a few picoseconds longer than the laser pulse [28]. Sub-20 ps temporal resolution is readily achieved. The conversion efficiency from laser energy to Kα energy is good at around 10−4 albeit into a solid angle of 4π. In general, higher laser intensity yields brighter K-alpha sources as more laser energy couples into these nonthermal electrons. However, inner-shell ionization probability increases and then falls off as the energy of the fast electrons increases. Source optimisation is target material dependent and is possible by controlling laser intensity and prepulse characteristics.
X-ray sources driven by ultrashort laser pulses
Chapter 2 Other X-ray sources Non-coherent laser-plasma X-ray sources equipped with collecting optics offer relatively high fluence reaching 1 J/cm2 in a wide or narrow spectral range depending on the optics. Radiation of such sources has been used for many applications related to X-ray imaging, material investigations, material processing and micromachining. Non-coherent X-ray imaging can be performed using imaging mirrors, zone plates or combination of both types. The most interesting wavelength range is again the â&#x20AC;&#x153;water windowâ&#x20AC;? and few pilot experiments were performed with laser-plasma sources emitting radiation in this range. Concerning material investigations there were some attempts to perform L-shell fluorescence and Resonant Inelastic X-ray Scattering (RIXS) investigations using laser-plasma EUV source. The average intensity with existing sources is until now too low for such investiga-
tions to give comparable results with synchrotrons; however, this will completely change with the extremelly powerful and highly energetic ELI lasers. There have been measurements of gas density distribution in gas targets using X-ray backlighting method. Laser-plasma sources were also used for micro- and nanomachining by photo-etching. The experiments were performed employing contact masks or by direct writing with a proper demagnifying optical system. Relatively high aspect ratio (about 10) of the structures created in PTFE and FEP was obtained. Higher values are in principle possible but they would require a special optical system giving a parallel beam with sufficient fluence. Laser-plasma EUV source was also used for modification of polymer surface and creation of self-organized micro- and nanostructures. Irradiation of polymer surface with EUV radiation offers new possibilities and modification of a very thin near surface layer.
ELI: Extreme Light Infrastructure www.eli-beams.eu
55
Directions of implementation at ELI Beamline Facility X-ray lasers Current X-ray lasers are mainly based on electron collisional pumping scheme. At the proposed ELI Beamlines Facility, in order to make the X-ray laser sources fully competitive with current and projected state-of-the-art XFEL, major effort will be devoted to improve the beam quality, shorten the pulse duration, and boost the output pulse energy. At this moment, the most promising way is to elaborate the concept of injection-seeded X-ray laser which ensures excellent beam quality in terms of spatial homogenity [30], aberration-free wavefront [31], high degree of coherence, polarization, and short pulse duration limited only by the laser transition bandwidth. We will start by investigating the physical phenomena constituting a base is for understand-
Figure 5: Basic geometries for seeded X-ray lasers: (a) normal incidence, and (b) grazing incidence.
ing and development of novel X-ray lasers with a special emphasis on the medium excitation and the consecutive generation, amplification and propagation of the X-ray radiation. There are two possibilities of the amplifier architecture: normal incidence (Fig. 5(a)), and grazing incidence pumping geometry (GRIP, Fig. 5(b)). In the normal incidence scheme, a weak long prepulse (ns, 100 mJ level) followed by a much shorter stronger pump pulse (ps, multi-J level) are sent perpendicularly onto a bulk solid tar-
get. In the GRIP scheme, only prepulse is sent at normal incidence to the target whilst the energetic pump beam is injected under an angle. The latter case has an advantage of reduced pump beam energy. In both cases we will seed the X-ray laser plasma amplifier with a highorder harmonic beam to improve the output beam quality (divergence, coherence, photon number).
X-ray sources driven by ultrashort laser pulses
Chapter 2 With a pulse of sufficiently high energy available at ELI Beamlines Facility, it will be possible to create multi-stage X-ray laser chains (Fig. 6) that will deliver beams with excellent quality of wavefront and energy on the multi-mJ level in sub-ps pulses. Considering the nominal pump laser parameters (50 J / 15 fs operating at 10 Hz), several X-ray laser beamlines at various wavelengths may be constructed, all of them being synchronized to each other. With appropriate focusing, X-ray intensities on the order of 1020 Wcm-2 could be reached [12]. The highenergy, high-power ELI laser driver will allow to realize novel X-ray laser schemes in the “water window” region that are desired for investigation of biological species in their natural environment with a hundreds time higher resolution than with any existing optical microscope. The alternative schemes such as “recombination scheme” and “inner-shell ionization scheme” will be also explored as they have in-
Figure 6: Principal scheme of the XRL beamline at ELI Beamline Facility.
herent potential to overcome the wavelengthlimits of collisionally pumped X-ray lasers. Achieving high gain in a fast recombining plasma is a very desirable in the pursuit of X-ray lasers. Compared to collisional X-ray laser schemes, where a very high degree of ionization is needed, recombination schemes require relatively low pumping power. This, combined with the high quantum efficiency achieved by using the transition to the ground state, makes the creation of a compact X-ray laser in “water window” feasible. However, rather stringent experimental conditions are required in order
to achieve recombination gain. X-ray lasers based on inner shell transitions in atoms and ions are also very attractive for the possibility of generating gain at wavelengths below 1 nm. Although inner shell transitions in ions may lead to shorter wavelengths, the accompanying plasma electrons can accelerate undesired process of quenching population inversion and gain due to fast Auger process. While the physics of recombination X-ray lasers is based on a process of plasma relaxation which follows very rapid heating of cold matter, lasing in the inner shell transitions of ions should occur near the plasma
ELI: Extreme Light Infrastructure www.eli-beams.eu
57
peak temperature. The very short heating phase generates free electrons associated with a distribution of inner-shell vacancies formed in ions. So, laser transitions can occur between highly populated upper levels and inner shell vacancies before equilibrium restoration. The idea of inner shell transition for lasing in sodium at 37 nm was first proposed about 40 years ago [32] but never realized. It is based on the fact that cross section for photo-ionization of atom or ion in deeper states can be larger than for more shallow states. More challenging experiments are expected for X-ray laser gain generations using several very innovative approaches to obtain lasing in inner shell transitions, which may be very attractive for wavelengths significantly below 1 nm. However, it will require pumping pulses of very high intensities and very short durations in the order of 100 as. With the advent of the ELI project, such pulses are already on lasers “horizon”.
K-alpha Regarding the implementation of this X-ray source at the ELI Beamlines Facility, a good starting place is to consider the provision from future XFEL sources which are supposed to produce exceptionally high quality X-ray beams. As a guide we can assume that these sources will deliver photons with 10 keV energy, 1012 photons per pulse with 100 fs duration at 100 Hz, with excellent transverse coherence, poor temporal coherence, and be highly collimated in sub-mm diameter beam at the target station. Though these are quite challenging source characteristics, it may be possible to compete, but excitement is likely found by providing a simultaneous mix of synchronized sources (photon Terahertz to gamma, electron, proton, and neutron) to an experiment. The features that distinguish Kα sources from XFEL and potentially make them unique are high diver-
gence and high photon energy (68.8 keV). To attract the widest science base it is important to develop Kα sources with high-average and high-peak brightness sources at approximately 10 keV and above. Ultimately, this demands a source repetition rate of 100 Hz. The next step in Kα source development should be to optimize the yield, enhance the source to background contrast, and reduce the source size from ~10 µm down to ~2 µm. Optimisation and contrast improvement are closely related. This part of work will require comprehensive experimental and computation studies to assess optimum laser wavelength, and acceptable laser contrast, optimal laser pulse duration and shape. A predictive capability is needed to determine optimal laser characteristics for materials of different atomic number. Progress demands precise measurements using highenergy and high-performance laser systems.
X-ray sources driven by ultrashort laser pulses
Chapter 2 Contrast and photon flux may be improved by using polycapillary micro-lenses, although these devices temporally stretch the X-ray pulse. The need for high throughput target manufacture, target and debris handling in high repetition sources is well known and needs to be addressed. As a part of this we should attempt to exploit the coherent properties of the KÎą source. Spatial resolution is a major incentive to reduce the effective source size to and below the few micron scale.
ELI: Extreme Light Infrastructure www.eli-beams.eu
59
Source Development and Advanced Source Use As an international large-scale research infrastructure, ELI Beamlines Facility will offer secondary X-ray sources with excellent parameters and complementary characteristics allowing experimental setup that is not feasible in any other existing laser laboratory. In practice, it means enough flexibility in manipulation and timing of several X-ray pulses. Taking the advantage of inherent synchronization of all ELI lasers which originate from one master oscillator, it will be possible to perform pump-probe experiments with multiple, ultra-intense X-ray beams such as X-ray lasers, high-order harmonics, and/ or non-coherent X-rays (e.g. K-alpha). This is extremelly difficult with current XFEL facilities which provide intense but merely one single X-ray beam. Second, despite the typical picosecond time duration, laser-plasma based X-ray lasers are still unbeatable in terms of photon flux and thus are capable of probing near-solid density plasmas. Considering 50 J / 15 fs laser operating at 10 Hz, a typical user at ELI may require 2-3 mutually synchronized energetic (e.g.
X-ray laser) and/or ultrashort (e.g. high-order harmonic) X-ray beams along with several high power, auxiliary NIR laser beamlines. This will make possible e.g. complex investigation of fusion relevant plasmas via X-ray laser Thomson scattering and X-ray interferometry at the same time, as well as various pump-probe configurations with multiple directional quantum beams. The X-ray sources developed within this Research Programme should deliver both coherent and non-coherent radiation at much higher intensities than any other existing plasma source of that type. Concerning coherent sources the most promising are X-ray lasers seeded with high-order harmonics. There are however also some ideas for X-ray lasers based on inner-shell excitation. Two possibilities are considered: direct excitation of inner shells by ultra-high power laser pulse or indirect excitation with a high power non-coherent laser-plasma X-ray source. Coherent radiation, however, is generally not
necessary for excitation. Instead of wavelength matching a sufficient power density is needed. Efficient focusing of non-coherent radiation should deliver the proper conditions. A flexible advanced target arrangement should be designed and constructed for the investigations of X-ray laser itself. It should allow to use targets of different parameters and experimental conditions for the source optimization. As a result of these investigations, coherent X-ray beams at different wavelengths and reliable parameters will be generated. These X-ray beams will be then manipulated (wavefront correction, spatial filtering, focusing) and delivered to the users.
X-ray sources driven by ultrashort laser pulses
Chapter 2 X-ray lasers The possibility of using X-ray lasers in combination with microscopy and holography for the study of cell biology is very attractive even at 10 nm. Of course, X-ray lasers will have a much broader application potential in biology when they become operational in the â&#x20AC;&#x153;water windowâ&#x20AC;? where cells in their natural environment (water, atmospheric pressure) can be observed with a high resolution X-ray microscope. Structures of biomolecules such as protein with atomic resolution can be also determined by X-ray diffraction. The high brightness of X-ray lasers should produce enough scattering to record diffraction pattern of very small samples such as single particles of large macromolecular assemblies without the need for samples to be crystalline where large signals are obtained because of coherent addition of the electric field of the Xrays by Bragg reflection. Femtosecond pulses
are short enough to avoid image blurring due to the vibrational motion of nuclei, thus atomic structure can be obtained without averaging over any periodic atomic motions. X-ray lasers will be suitable for observing biological specimens in their natural wet environment. Though radiation damage may become a problem, the very short period (femtosecond) can be expected to allow the collection of diffraction patterns before radiation damage that distorts or destroys the biomolecule structure occurs. A large degree of spatial coherence of an X-ray laser is just perfect to record a 3-D image by holographic technique. Imaging at the atomic scale can be obtained by advanced X-ray microscopy making it possible to observe amorphous and disordered materials, including polymers, crystals with strains and defects, inorganic structures such as nanotubes, and bio-molecules that are difficult to crystallize. Femtochemistry will be one of the key applica-
tions of X-ray lasers at ELI Beamlines Facility. Femtochemistry by pump-probe spectroscopy aims at real-time observation of transient molecular behaviours including the atomic positions, bond lengths and angles during chemical reactions. As chemical reactions occur at timescales of order of femtoseconds or picoseconds and structural changes occur on an atomic scale, high spatial and temporal resolution corresponding to these scales are required to observe molecular motion in real time. The advanced X-ray lasers should have sufficient spatial and temporal resolution for this purpose. Even for such a short pulse, the photon flux per pulse can be enough to allow spectroscopic observations with a single pulse. Short pulse X-ray lasers can provide an opportunity to observe various nanoscale dynamics spectroscopically, such as viscoelastic flow, protein folding, crystalline phase transitions, and magnetization dynamics of nanostructures over a wide range of time scales.
ELI: Extreme Light Infrastructure www.eli-beams.eu
61
Finally, some fundamental scientific experiments that have not been able to be carried out so far with other light sources such as synchrotron radiation can be realized with X-ray lasers at ELI Beamlines Facility. They include studying of plasma, warm dense matter and atomic physics, e.g. determination of energies, ionization cross sections, lifetimes and transition rates. The high X-ray intensity will make it possible to study nonlinear effects that lead to multiphoton processes in the keV region. Thanks to the expected high repetition rate of the source, X-ray lasers may be utilized to create controlled structures in the nanometre range which is a big challenge for the semiconductor industry (nanolithography).
X-ray free electron-lasers With future multi-petawatt lasers like ELI, ultrahigh beam currents of about 100 kA in 10 fs can be expected, allowing a drastic reduction in the undulator length of free-electron lasers. However, the very extreme electron beam properties required to produce an XFEL, or at least X-ray synchrotron light, are not available yet. A crucial parameter will be the charge contained in the 0.1% spectral bandwidth: 1 nC at 1 GeV energy will be required to produce an efficient XFEL. FELs operating in the ultraviolet or Xray region rely on self-amplified spontaneous emission, which requires electron beams with extremely small emittance and long and highly engineered undulators. However, the selfamplified spontaneous emission process also depends on the current density of the electron
beam, which could be very high for wakefield accelerators. Therefore, with future reduction in energy spread, laser-plasma accelerators may be optimal compact drivers for XFEL. Very recent experiments by Leemans et al. (LBL, U.S.A.), utilizing only 40-TW laser pulse of 38 fs duration and a gas density of 4.3Ă&#x2014;1018 cm-3, clearly showed that 1 GeV electron beams can be produced by means of capillary discharge. However, the measured charge of 30 pC is significantly below the goal of 1 nC. In order to further improve the resulting current, an alternative scheme has been proposed [34]. The number of bubble electrons can be increased if high density plasmas are used, because the feeding process is more efficient if a high amount of plasma electrons is available. An increased gas density requires shorter laser-pulses (sub10-fs), such that the laser pulse length fits into one plasma period. In this case, the entire laser pulse energy can be used, while longer pulses
X-ray sources driven by ultrashort laser pulses
Chapter 2 lose energy during self-shortening for entering the bubble regime. Therefore, in contrast to standard laser-acceleration experiments, a different driver laser, namely a 5 fs-Petawatt laser (like the Petawatt-field-synthesizer at MPQ, Germany), has to be used. PIC simulations show that such ultrashort lasers can capture more than 1 nC charge inside the bubble. The smaller plasma period leads to smaller bubbles, hence shorter bubble stems and thus higher currents (1 nC in 10 fs). Recent analytical estimates and simulations have shown that laser-plasma accelerator-based FELs can be operated with meter-scale undulators. The key parameter is the ultra-high electron peak current, which significantly reduces the gain length and increases the tolerance with respect to the energy spread and emittance.
Other X-ray sources Other non-coherent laser-plasma X-ray sources at present time utilize mainly nanosecond lasers. In this case plasma is thermalized what means that its spectrum is defined by its temperature and atomic number of the excited material. Ultra-high intensity femtosecond pulses will generate X-rays with essentially different spectrum because of the non-thermal excitation. It will allow obtaining intense X-ray radiation with controllable spectrum. The other important parameter of X-ray radiation is time duration of the pulse. Because of non-thermal excitation the resulting X-ray pulses can be very short comparing to current EUV pulses created with nanosecond laser pulses. Assuming relatively high conversion efficiency, high power X-ray pulses can be obtained this way. Apart from that short duration of the pulse would allow to perform measurements of high speed
processes requiring X-rays. The non-coherent X-ray source will be equipped with separate chambers with different target systems (solid, liquid, gas targets) to assure experiments with different spectral distributions. The chambers for solid and liquid targets must be equipped with a debris mitigation system. Effective utilization of the intense, non-coherent X-ray radiation will require novel and highly durable X-ray collecting optics. Every chamber should be nominally equipped with such optics which can be, depending on application, either broad-band or wavelength-selective optics. Because the noncoherent X-rays will be emitted in a wide solid angle, each chamber should employ at least two collectors: one grazing incidence mirror for wide band collection and one multilayer mirror for narrow band collection. The collecting X-ray mirrors can be paraboloids to give parallel beams propagating in specified directions. In this way, the X-ray sources at ELI Beamlines Facility will have beamlines for different purposes similarly to synchrotrons.
ELI: Extreme Light Infrastructure www.eli-beams.eu
63
Potential for Applications, Business and Technology Transfer The development of the short wavelengths sources for the scientific and technological use in the next 30 years is absolutely crucial. In particular, the 2007 report of US Department of Energy explicitly states that: “During the halfcentury since the laser was conceived, the use of coherent light has become an indispensable part of our world, with applications such as telecommunications and optical data storage playing a trillion-dollar role in the global economy. In today’s world, the development of tools and techniques that use coherent light at evershorter wavelengths – in the extreme ultraviolet and soft X-ray regions of the spectrum – is assuming increasing importance”. Considering the size and future needs of X-ray sources for research, academia and industry, it is not sur-
prising, that huge effort is being devoted to develop these sources of coherent X-ray radiation as well as methods for their application. The main advantages of ultrashort and intense X-ray sources are threefold: first, the spatial resolution information to be extracted from a system is proportional to the wavelength of the probing electromagnetic field. Sufficiently strong X-ray sources will greatly help science and technology to move from the micrometer to the nanometer scale range and beyond. Coherence and polarization of such sources can further improve the quality of the obtained informations. Second, the intensity of X-ray sources to be elaborated at the ELI Beamlines Facility, will reach the highest intensities ever at these wavelengths. As a consequence, transformations in the matter can be initiated on a nanometer scale range in a controlled way, with temporal resolution on the order of femtoseconds. Last but not least, the spectral quality of these X-ray sources can reach a very high spectral
purity, below 10-5 relative bandwidth, making them suitable for very precise investigation of the dynamics of sample structure and spectral properties. The above aspects can specifically impact the following R&D areas with certain commercial potential: 1. Material science where nanotechnologies are quickly developing: magnetism, phase transitions, smart materials, liquids and disordered systems, cluster vibrations and reactions, phase transitions and cluster melting, warm dense matter generation, and imaging plasmas are waiting for the ultra-high spatial resolution to be achieved with the advanced X-ray sources at ELI. 2. Time-resolved studies aimed at better understanding of ionization dynamics of atoms, molecules, clusters and highly charged ions, multiple core hole formation, nonlinear effects in the X-ray domain,
X-ray sources driven by ultrashort laser pulses
Chapter 2 Coulomb explosion of clusters, generation of plasmas at solid density, hydrodynamic response to X-ray pulses, non-equilibrium plasmas generation and the study of materials under extreme conditions. 3. High-resolution atomic physics: excitation of highly charged ions, study of X-ray nonlinear effects, resonant elastic scattering, plasma spectroscopic time-resolved diagnosis, application of X-ray lasers in nuclear spectroscopy. As mentioned earlier, the high peak current of laserâ&#x20AC;&#x201C;plasma electron beams could lead to compact XFEL facilities and underpin a wide range of applications. High dissemination towards multidisciplinary users is then foreseen in fundamental science (time-resolved X-ray diffraction, photoelectron spectroscopy, ultrafast studies of material damage, plasma physics, radiography of dense materials, X-ray microscopy
and X-ray imaging) but also in more societal fields; e.g. the implementation in existing hospitals of phase-contrast imaging techniques, developed at synchrotrons to provide high-resolution images with micrometer resolution, could enable significant advancement in clinical diagnostics (the X-ray energy of a tabletop XFEL can be as large as required for medical diagnostics, i.e. above 20 keV). Ultrashort light pulses are powerful tools for time-resolved studies of molecular and atomic dynamics. They arise in the visible and infrared range from femtosecond lasers, and at shorter wavelengths, in the ultraviolet and X-ray range, from synchrotron sources and free-electron lasers. The ultrashort duration of these radiation beams will provide unprecedented timeresolved measurements down to the motion of electrons on atomic scales, and a zooming onto the two fundamental molecular building blocks, the electron and the atom. It will enable
exposure of atoms and molecules to relativistic intensities before their disintegration. Coherent diffraction on single molecules will then become accessible, opening an entire new field of research. Time-resolved absorption spectroscopy and Thomson scattering of high-density plasmas require penetrating radiation such as X-rays and an ultrafast time resolution to reveal the properties of the warm dense matter produced in a laserâ&#x20AC;&#x201C;plasma experiment. Time-dependent measurements of plasma temperature and density will provide a valuable contribution to the understanding of degeneracy and coupling, as well as long and short-range interactions between charged particles in dense plasmas. Finally, the simultaneous use of particles and radiation as probe or pump beams offers unique opportunities. As an example with societal is-
ELI: Extreme Light Infrastructure www.eli-beams.eu
65
sues, the study of the ultrafast kinetics following matter excitation by high-energy particles is a major subject in radiation physics and in nuclear technology, with implications on nuclear reactor lifetime. At present, the physical effects of intense particle energy deposition can only be accessed through modelling. There is a crucial need to look at vacancy dynamics that occur in the few-hundred-femtosecond timescale, by ultrafast X-ray or visible probing. Novel laserbased sources will provide the necessary tools. Laser-based accelerators imply a large reduction in infrastructure, particularly for shielding, and thus in size and cost. A purely laser-driven light source would inherently produce ultrashort-duration pulses and be perfect temporally synchronized with the driving laser. Time-resolved experiments require at least two pulses, one to initiate the fast process to be observed and another to probe after a well-defined delay. The pump and probe would automatically be
synchronized with the electron bunch and thus the undulator radiation. However, there are no restrictions to laser pulses, and these may be converted into other forms, for example, electron or proton pulses. Thus, a laser-based accelerator facility would be very flexible.
timize the existing sources and even to develop completely new concepts.
The advanced X-ray sources at ELI Beamlines Facility will be unique and complementary to the existing sources, delivering significantly higher photon flux and narrower spectral bandwidth for the experiments, helping to draw the details of the whole picture in a given scientific investigation. Moreover, the development of laser based radiation sources, such as attosecond pulses, electrons, protons, neutrons and further accelerated particles will greatly benefit from the development of X-ray sources at ELI while it will be possible to have an in situ X-ray diagnostic of the laser irradiated sample and its subsequent dynamics. The goal will be to understand the mechanisms of the radiation production, to op-
X-ray sources driven by ultrashort laser pulses
Chapter 2 References
[21]
J.J. Rocca et al., Opt. Lett. 24, 420 (1999)
[1]
W.C. Roentgen, Nature 53, 274 (1896)
[22]
J.E. Trebes et al., Science 238, 517 (1987)
[2]
B. Schenkel et al., Opt. Lett. 28, 1987 (2003)
[23]
S. Suckewer et al., Science 247, 1553 (1990)
[3]
K. Yamane et al., Opt. Lett. 28, 2258 (2003)
[24]
L.B. DaSilva et al., Opt. Lett. 17, 754 (1992)
[4]
T.H. Maiman, Nature 187, 493 (1960)
[25]
L.B. DaSilva et al., Science 258, 269 (1992)
[5]
H. Daido, Rep. Prog. Phys. 65, 1513 (2002)
[26]
D.H. Kalantar et al., Phys. Rev. Lett. 76, 3574 (1996)
[6]
D.L. Matthews et al., Phys. Rev. Lett. 54, 110 (1985)
[27]
H.A. Baldis et al., Rev. Sci. Instrum. 73, 4223 (2002)
[7]
B. Rus et al., Phys. Rev. A 66, 063806 (2002)
[28]
H.P. Schlenvoigt et al., Nature Phys. 4, 130 (2008) and Refs. therein
[8]
K. Jungwirth et al., Phys. Plasmas 8, 2495 (2001)
[29]
H.S. Park et al., Phys. Plasmas 13, 056309 (2006)
[9]
J. Dunn et al., Phys. Rev. Lett. 84, 4834 (2000)
[30]
F. Brunel et al., Phys. Rev. Lett. 59, 52 (1987)
[10]
R. Keenan et al., Phys. Rev. Lett. 94,103901(2005)
[31]
J.Ph. Goddet et al., Opt. Lett. 32, 1498 (2007)
[11]
Y. Wang et al., Phys. Rev. A 72, 053807 (2005)
[32]
J.Ph. Goddet et al., Opt. Lett. 34, in press (2009)
[12]
Ph. Zeitoun et al., Nature 431, 426 (2004)
[33]
M.A. Duguay et al., Appl. Phys. Lett. 10, 350 (1967)
[13]
T. Mocek et al., Phys. Rev. Lett. 95, 173902 (2005)
[34]
F. Gruner et al., Appl. Phys. B 86, 431 (2007) and Refs. Therein
[14]
Y.Wang et al., Phys. Rev. Lett. 97, 123901 (2006)
[15]
Y. Wang et al., Nature Photonics 2, 94 (2008)
[16]
P. JaeglĂŠ, La Recherche 184, 16 (1987)
[17]
S. Heinbuch et al., Opt. Express 13, 4050 (2005)
[18]
P.W. Wachulak et al., Opt. Express 15, 3465 (2007)
[19]
L.B. DaSilva et al., Phys. Rev. Lett. 74, 3991 (1995)
[20]
F. Albert et al., Opt. Commun. 142, 184 (1997)
ELI: Extreme Light Infrastructure www.eli-beams.eu
67
Chapter 3 Particle acceleration by lasers
Authors: RNDr. Josef Krรกsa, CSc. / Daniele Margarone, Ph.D.
Introduction The ELI Beamlines Facility is supposed to be built as central laser facility with different experimental stations that can be used and accessed independently by various communities to best develop and use new particle and photon beams for scientific progress, but also to transfer the corresponding knowledge to industrial and societal applications. ELI will probably be the first facility providing beam time for pumpprobe experiments using moderately relativistic ultra-short ion beams and synchronized highintensity lasers, electron beams, X-ray sources or attosecond sources. What makes ELI additionally unique is the ability to vary all these different secondary sources independently as allowed by having independent individual driving laser beams and adapting them optimal to the requirement of the specific experiment. The envisioned laser driven electron and proton/ ion sources with huge energies (up to the GeV level), in combination with other high-energy
particle and radiation sources, require a specific environment close to that of conventional accelerators. In contrast to the latter, which are planned for a dedicated energy range, the laser driven source is, according to the laser parameters that will be available on ELI and the projected parameters of the produced source, a â&#x20AC;&#x153;broadbandâ&#x20AC;? facility, which has to fulfil at least two aspects: first, to explore the optimum mechanisms to create efficiently the highest energy and flux accompanied by shaping the energy spectra with the available laser parameters, and second, to use them for the envisioned scientific program and different other applications. In recent years the dramatic rise in attainable laser intensity has generated an even more dramatic emergence and new evolution of the fields of research associated with non-linear laser-matter interaction. Production and acceleration of electrons up to 1 GeV over accelerating distances around 1 mm (100 meters for
Particle acceleration by lasers
conventional accelerators) and hadron acceleration to 100 MeV, are the clearly visible results of this evolution. The spectacular increase in brightness and decrease in pulse duration of particle beams will revolutionize the way of investigating matter. Fundamental events in biology, chemistry and solid-state physics can be recorded with angstrom space resolution to capture electronic, atomic or molecular transient dynamics. Source compactness, broad spectral range and perfect synchronization of particle and radiation bursts are unique properties that could extend the breadth of applications. The high peak current of laserâ&#x20AC;&#x201C;plasma electron beams could lead to compact XFEL facilities, on a size affordable by small-scale laboratories. High dissemination towards multidisciplinary users is then foreseen in fundamental science, but also in other fields. Finally, time-resolved experiments, would significantly extend the field of investigation in the
Chapter 3 dynamics of matter, compared with currently available techniques using a visible pump and X-ray or visible probes. It seems possible, at the first stage, to pump intermediate laser amplifiers (50 J level per beam at target) with laser diode technology instead of flash lamp based lasers. This change opens the possibility to strongly increase the laser quality, shot-to-shot stability and repetition rate, which follows the path to provide particle and radiation sources with sufficient reproducibility to be interesting to users and for further source development. The last five years witness a very significant progress on laser-plasma accelerators. Moreover, significant improvements are expected on the state-of-the-art of particle accelerators. The sources parameters we are proposing for users are not yet demonstrated but they seem rea-
sonable for a 6-year horizon. It is also expected that a significant improvement on the source parameters related to the quality of the beam would be improved in future with multi-stage and all-optical injection techniques. The end stage of ELI envisions a powerful laser, delivering a few kJ in 15 fs (~200 PW) with low repetition rate (minute based). Since all this stages will be constructed sequentially and the laser technology will evolve quickly, it is likely that the laser in 2015 will be further upgraded. Thus, it will be possible to get the 100 GeV electron laser acceleration, which is one of the goals ELI intends to achieve. The emergence of many PW-class systems, in the next few years, will be used for demonstrating the 10 GeV electron laser acceleration; thus ELI and the rest of the world ultra intense laser labs can nicely cooperate to sharpen their overall skills and knowhows in a collaborative network in this development. On the other hand, this energy level could trig-
ger significant interest from the physics community and could be a vehicle for getting their involvement. In summary, the ELI Beamlines Facility will offer a versatile electron and proton/ion source emitting in an unprecedented energy range and, moreover, it is worth noting that the facility with its unique variability in the parameter range given by the optional different laser beam parameters would not be interesting exclusively for basic science studies. The particle â&#x20AC;&#x153;beamlinesâ&#x20AC;? with the concomitant environment (diagnostics, radiation protection, etc.) will also allow accomplishing multidisciplinary societal applications, which are already in the focus of present day activities.
ELI: Extreme Light Infrastructure www.eli-beams.eu
71
Current Status in Particle Acceleration Techniques Electrons Current CPA lasers range from the Terawatt to the Petawatt power and can produce laser pulses with durations below 20 fs. In this way a focused laser beam can produce intensities above 1019 W/cm2 with short durations. In laser plasma-electron accelerators, a longitudinal accelerating electric field is generated by the ponderomotive force of the ultrashort and intense laser. This force, proportional to the gradient of the laser intensity, pushes the plasma electrons out of the laser beam path, separating them from the ions. This creates a travelling longitudinal electric field, in the wake of the laser beam, with a phase velocity close to the speed of light, most suitable for accelerating particles to relativistic energies. This electric field can reach amplitudes of several hundred gigavolts per meter. Electrons need to be injected into the wakefield with a sufficient initial energy so that
they can be trapped and accelerated. Experimentally, two injection mechanisms have recently demonstrated the generation of highquality quasi-monoenergetic electron beams. In the first mechanism, a single laser pulse is used to drive the wakefield to large enough amplitude such that electrons are injected into the rear of the first wake oscillation through transverse breaking of the plasma wave. The electrons then surf the wake and after outrunning the wave they form a monoenergetic electron bunch. This is known as the ‘bubble’ regime [1]. With current laser technology, electron beams in the 100 MeV range have been produced over millimeter distances [2,3], with relative energy spreads of the order of 5–10% and charge of hundreds of picocoulombs. A 1 GeV electron beam has been reported in a recent experiment, where the laser pulse was guided and evolved over a few centimeters in a capillary plasma discharge [4].
Particle acceleration by lasers
The second mechanism is based on the use of several laser pulses. In its simplest form, the scheme uses two counter-propagating ultrashort pulses with the same wavelength and polarization. The first laser pulse, the ‘pump’ pulse, creates a wakefield, whereas the second laser, the ‘injection’ pulse, is only used for injecting electrons into this wakefield. The laser pulses collide in the plasma and their interference creates an electromagnetic beatwave pattern that pre-accelerates some electrons. A fraction of these have enough energy to be trapped in the wakefield driven by the pump pulse and further accelerated to relativistic energies. Although this scheme is more complicated experimentally, it also offers more flexibility: experiments have shown that the electron beam energy can be tuned continuously from 10 to 250 MeV [5]. The electron beam has a quasi-monoenergetic distribution with energy spread in the 5–10% range, charges in the 10–100 pC range and its parameters are stable within 5–10%. This approach is promising for the control of the elec-
Chapter 3 tron beam parameters and might enable tuning of both the charge and the energy spread. For instance, increasing the beam energy to the gigaelectronvolt range should decrease the relative energy spread to the 1% level. The current state-of-the-art in laser-plasma electron accelerators is the demonstration of mono-energetic electron bunches in the sub300 MeV range [2,6,7], the demonstration of 1 GeV mono-energetic acceleration using external guiding [4] and the first demonstration of all-optical injection resulting in an electron beam with low emittance and energy dispersion [5].
Ions Using present-day common laser facilities in the 100 TW range, low-emittance (Cowan 2004) protons are accelerated up to 20-30 MeV energies from thin (~20 Âľm) solid metallic target. The present record in terms of kinetic energy has been set by the Nova 1 PW laser facility at LLNL which has produced up to 56 MeV protons (Snavely 2000) and the Trident laser facility which has produced up to 59 MeV (Filippo 2008). These beams can be refocused and energy selected (Schwoerer 2006, Toncian 2006, Schollmeier 2008). In standard conditions, protons contained in a surface layer (~20 Ă&#x2026; thick) of hydrocarbon contaminants are preferentially accelerated due to their highest q/m ratio. Heating of the target prior to the experiment eliminates these hydrogen
contaminants and allows acceleration of heavier ions (Hegelich 2006). By simply changing the substrate, laser-acceleration can therefore produce a versatile, easy to modify source of ions. Conversion efficiency up to a few % has been measured from laser energy to proton energy integrated in the spectrum from 10% to 100 % of the cut-off energy. Several mechanisms have been demonstrated to induce proton and ion acceleration, namely target sheath normal acceleration (TNSA) (Hatchett 2000), front-surface acceleration (Nemoto 2001), and shock acceleration (Silva 2004), the TNSA mechanism being the one, up to the presently accessible highest laser intensities that produces highest energy ions. Acceleration takes then place at the target-vacuum interfaces where laser-accelerated relativistic electrons form a dense electron plasma sheath (field ~TV/m) that ionizes surface atoms and accelerates ions along the target normal direction.
ELI: Extreme Light Infrastructure www.eli-beams.eu
73
High laminarity of the ion beam requires acceleration from an initially cold surface (i.e. that has a sharp vacuum interface). Usually, this is ensured only at the target non-irradiated rear surface. Using very high temporal contrast laser pulses (i.e. contrast >1010), laminar acceleration is also possible from the target front that is then also unperturbed by the low-level laser pedestal (Ceccotti 2007). The increase of the laser temporal contrast also allows to use extremely thin targets (in the nm range) which have been shown to yield higher energy ions with increased efficiency (Ceccotti 2007, Antici 2007, Neely 2006), due to a combination of increased laser absorption and electron recirculation. Refocusing of the naturally divergent beam (due to the curvature of the electron sheath) is possible by using curved targets (Patel 2003) or optics, either conventional (Schollmeier 2008, Ter-Avetisyan 2008) or plasma-based (Toncian 2006).
Monochromaticity is made possible either by accelerating only a monolayer of elements (all ions then experience the same electric field and are accelerated to the same energies) deposited on a higher-Z material foil (Hegelich 2006), by using the specific electron distribution on the surface of isolated microdroplets (Ter-Avetisyan 2006) or by selecting differentially focused spectral components (Fourkal 2002, Toncian 2006). Acceleration has been tested using low-density plasmas (Willingale 2006, Antici 2009) and has been shown to be also efficient. Although it produces beams of lower quality than from solids, it offers the advantage of being compatible with a high-repetition rate for the laser source.
Particle acceleration by lasers
Chapter 3
Scaling Laws for ELI laser intensity regime Electrons In order to estimate the electron beam parameters that will be available for users of the ELI Beamlines Facility it is mandatory to specify the laser beam characteristics. The main laser parameters are:
Beam Number up to 4 (electron sources can use up to 3 beams) Energy
up to 50 J @ 10 Hz
Wavelength
800 nm
pedestal cannot disturb the plasma density; therefore we can set two contrast levels: on ns time-scale we need to avoid the gas ionization (intensity bellow 1012 W/cm2) while on ps-time scale we need at least to avoid bulk charge separation in the plasma (intensity bellow 1016 W/cm2). We present three pulse energy levels possible at ELI at a repetition rate compatible with a source for users: 100 mJ â&#x20AC;&#x201C; 1kHz, 1.5 J â&#x20AC;&#x201C; 100 Hz, and 45 J â&#x20AC;&#x201C; 10 Hz. We have reported the laser and electron source main parameters for the three simple regimes (see Table 1): blowout self guided, blowout external guided and bubble.
sensitive to experimental parameters and can only be anticipated by fully explicit three-dimensional heavy particle-in-cell simulations. This source specification should be tuned before the construction phase since the state-ofthe-art is in fast progress and new particle injection techniques may be included to improve the resulting electron beam parameters. Additionally, scaling laws for laser-plasma electron accelerators resulted from these analyses can be used to design future electron accelerators [8].
Pulse Duration 15 to 150 fs Contrast
ns contrast (prepulse) and ps contrast (pulse pedestal)
The maximum intensity on pre-pulses or pulse
At this point, there is no straightforward method to calculate electron beam properties, besides maximum energy and charge. Other important electron beam quantities, such as bunch energy dispersion, emittance and bunch length are very
ELI: Extreme Light Infrastructure www.eli-beams.eu
75
Blowout
Self guiding
Laser energy 100 mJ on target 1kHz τ [fs] 9.8
1.5 J 100 Hz 24.2
65.8
W0 [µm]
4.4
10.9
29.6
ne[10 cm ]
23.2
3.83
0.63
L [mm]
0.2
3.3
54.6
∆ E[GeV]
0.1
0.62
4.59
Q [nC]
0.13
0.31
1.04
τ [fs]
15.6
38.4
119.2
W0 [µm]
18
External guiding
Bubble
-3
45 J 10 Hz
7.0
17.3
53.7
ne[1018 cm-3]
4.62
0.76
0.08
L [mm]
1.8
26.4
793.1
∆ E[GeV]
0.26
1.56
15.06
Q [nC]
0.1
0.25
0.78
τ [fs]
6.6
12.2
26.8
2.0
3.7
8.1
ne[10 cm ]
57.3
24.92
8.74
L [mm]
0.03
0.2
1.8
∆ E[GeV]
0.03
0.18
1.5
0.55
1.54
5.71
W0 [µm] 18
Q [nC] Table 1: Electron sources for users – experimental parameters
-3
Particle acceleration by lasers
Chapter 3 Ions The â&#x20AC;&#x153;initial sourceâ&#x20AC;? parameters available for ion beam acceleration will obviously be the same as already defined for electron acceleration (50 J per beam laser pulses, with a minimum laser pulse duration of 15 fs and a repetition rate of ~10 Hz) with an important constraint for ion generation, which is the prepulse laser contrast (about 1010). A review of the maximum energy and proton number obtained in various experiments is shown in Figure 1. It is clearly shown that the maximum proton energy increases with the laser pulse duration. It has been demonstrated that the efficiency of the acceleration process also increases with the laser pulse duration and the laser irradiance as well (Fuchs 2006).
Figure 1: Review of cut-off maximum proton beam energy, as a function of laser intensity, as reported from published data. Experimentally measured data are small dots, boxes and crosses corresponding to three pulse duration ranges are shown. Simulations performed at higher laser intensities planned for ELI are reported as big purple dots. Note that, as experimentally proven up to 1020 W.cm-2, the maximum proton energy for the extreme short pulses is ~ I, whereas for the longer pulses ~ I1/2.
ELI: Extreme Light Infrastructure www.eli-beams.eu
77
ELI, in its various stages, will allow exploring new, efficient, ion acceleration regimes that have been observed in numerical simulations. As a result, and as shown in Fig. 1, several hundred MeV up to a few GeV protons could be achievable with the planned on-target intensities at ELI in its final stage (i.e. up to 1025 W/cm²). Supposing that the proton number would be similar to that obtained currently on existing facilities, and considering a high repetition rate (e.g. 10 Hz), then we could obtain currents up to a few tens of nA, comparable with a conventional accelerator. Based on the present knowledge and known acceleration mechanisms, the source specifications would be: •
200 MeV for protons
•
50 MeV/nucleon for heavy ions
•
>1012 ions/bunch
•
< ps duration at the source
•
% ∆E/E
•
µm virtual source size
•
few degrees angular distribution
•
kA current at the source
According to simulations, the predominant ion acceleration regime should be, when increasing the laser intensity, first (i) shock accelerated at the target front surface or in its interior for I>1021 W/cm2 (Silva 2004), followed by (ii) radiation pressure acceleration, so-called RPA, (Esirkepov 2004, Macchi 2005), when I≥1023 W/cm2, i.e. when the electromagnetic wave is directly converted into ion energy via the spacecharge force related to the displacement of all electrons in a thin (nm scale) foil, allowing to reach GeV-scale energies. Multiple stage acceleration using stacked foils would offer the
Particle acceleration by lasers
additional prospect of further increase of the ion maximum energy (Kim 2009). At a similar intensity regime, other simulations have also shown that ions could be accelerated to energies higher than 10 GeV in the “bubble” regime of wakefield acceleration using near-critical density plasmas and mixed ions (Shen 2007). Such underdense plasma targets would have the benefit of allowing high repetition rate operation. ELI Beamlines Facility will thus offer the prospect of producing and study a versatile ions source, at high repetition rate, while enhancing simultaneously the high-energy end of the spectrum, the beam monochromaticity and the laser-to-ions conversion efficiency, all of which being crucial points for the development of applications in various areas. The possibility to vary the pulse duration of the ELI laser beamlines offers also a way to create proton/ion bunches of different durations, offering a way to adapt the source to the various envisioned applications.
Chapter 3
Directions of Implementation at ELI Electrons A precise work plan is required to develop a laser-facility of accelerated electron beams in order to: study the different acceleration technologies and choose between them the most adapted to ELI goals and performances; set the full metrology of the electron beams; detail the building constraints especially if using a multistage accelerator; consider the evolution of the accelerated particle beam considering ELI laser full performances; compare the results from numerical modelling with experiments to validate the codes and use them for improving the highenergy electron sources; etc. The evolution of laser plasma accelerators until 2015 (possible date for ELI activity) will develop in very different directions in order to improve or extend the parameters of the electron bunches.
Practical plasma densities for laser plasma acceleration with current laser technology range from 1017 cm-3 to 1019 cm-3. These densities correspond to plasmas obtained from low pressure gases (10â&#x20AC;&#x201C;500 mbar). The traditional way to obtain an adequate gas target consists in the use of a supersonic nozzle. Another possible target for laser-plasma accelerators are preformed plasma channels [9,10]. In this case the plasma is completely ionized and presents a radial parabolic density profile that can act as a graded index optical fiber. This external guiding can keep the laser beam focused for long distances increasing the energy gain of the accelerator. It will be necessary to develop an adequate plasma target for high repetition rate (>10 Hz), short length (<1mm), high density (>1019 cm-3) to take advantage of the short pulse capability of ELI. Such a plasma target would allow to explore the â&#x20AC;&#x153;bubble regimeâ&#x20AC;? with the laser fo-
cused to intensities > 1020 W/cm2 and producing high charge electron bunches (> 1 nC) with energies of about 1 GeV. The work plan will be also aimed to improve the existing plasma sources for external guiding. The more important improvements are the reproducibility and uniformity of the plasma, the possibility of guiding laser beams with spot sizes focused to less than 50 microns (this requires a strong radial density gradient in the plasma density), the length of the plasma channel, which needs to be increased to the 20 cm range in order to match the dephasing lengths available for ELI parameters. It is mandatory to develop a new class of plasma sources of very low axial plasma density, strong radial density gradient and at least 1-meter length. This plasma source could be extended to 2-meter length in order to accommodate a
ELI: Extreme Light Infrastructure www.eli-beams.eu
79
demonstration of a 15 GeV laser plasma collider. The development of meter long low-density plasma sources will be a crucial step to break the few (~4) GeV barrier possible for current plasma sources.
loading techniques tested so far with success used longitudinal density steps or the collision of laser beam to promote localized injection of charge and therefore to reduce the energy spread.
It will be also possible to develop a multi-stage acceleration setup. The acceleration in multiple stages is the key to high-energy acceleration techniques. The use of multi-stage acceleration combined with 50 J laser (as planned for ELI) allows four meter-long 15 GeV stages (excluding laser and electron beam optics).
Significant technical developments are also required for the source development. Real-time laser beam metrology will be required to assure electron beam reproducibility and reduce the activation of the accelerator setup. Some of the electron beam diagnostics can be adapted from the conventional accelerators, others need further development and will require the use of short pulse lasers (for time resolved measurements of the electron bunches with 10 fs duration). A hypothetical source design is reported below.
It will be important to develop efficient beam loading techniques. The use of external beam loading is probably the most effective direction. Two approaches can be followed: i) to maximize the bunch charge, or ii) to reduce the bunch energy spread. The theoretical and experimental work on beam loading is just starting. The beam
â&#x20AC;˘ Laser Target Laser targets for electron accelerators are low
Particle acceleration by lasers
atomic number gases (H2 or He) or pre-formed highly ionized plasmas made of these gases. These laser targets imply the introduction of significant amount of gas in the vacuum system. At ELI, these electron sources will operate for the first time at high repetition rate (10 Hz to 100 Hz). The amount of gas introduced in the vacuum system will increase significantly requiring: i) reduction of the gas release per shot, ii) increase of the pumping speed and optimization of the geometry of the vacuum chambers including gas traps (taking advantage of the directionality of gas jets), iii) use of differential pumping between sections of the vacuum system. The gas is introduced in the interaction point by means of: i) supersonic gas jets, ii) differential pumped gas cells, iii) other plasma sources. Supersonic gas jets introduce a much higher amount of gas in the system than gas cells or discharge based plasma sources.
Chapter 3 â&#x20AC;˘ Laser beam dumping / electron charge collection In a laser-plasma electron accelerator, the great part of the laser beam energy will be absorbed by the plasma and transferred to the electron beam. However, for a practical high repetition rate accelerator, special care needs to be taken with the laser light transmitted by the plasma, especially if sensitive applications are ahead in the beam axis. One possibility to mitigate this problem consists in the inclusion of a short section of a high-Z gas (example Ar) and uses it as a divergent lens for the laser light (due to ionization induced defocusing). It is important to reduce the amount of electromagnetic noise resulting from the ultra-fast rise-time pulsed currents in the accelerator. This noise strongly affects the electronic systems in the laboratory. This problem can be well mitigated if a collimator/beam dump consisting in a cylinder with a hole on the axis collects all the higher divergence (lower energy) electron ejected from the
plasma. This collimator/beam dump will also concentrate most of the radiation in the target chamber reducing the activation of other elements. These two elements need to be carefully designed in order to maximize their efficiency in such a way that the electron beam properties would not be affected.
netic field used to deflect the beams out of the original axis should have the freedom of being deflected to at least two different axes: the experiment target beamline and the diagnostic beamline. This property is very important since the beam can be optimized and then sent to the experiment by changing only the magnetic field.
â&#x20AC;˘ Electron beam optics / beam deflection
â&#x20AC;˘ Electron beam diagnostics
Current experiments are producing a significant amount of hard radiation due to betatron oscillations of the electron beam in the ion channel. This radiation may affect the electron beam experimental setup. Therefore the electron beam should have the possibility to be magnetically deflected from its original axis as well as to be imaged from the laser ion channel exit to the target. Since the electron beams will be considered as a source for users, we have to introduce some of the beamline concepts of conventional accelerators as well as diagnostics. The mag-
Different electron beam diagnostics are necessary to measure and control the beam properties. These include both beamlines (FCT/ICT current transformers, beam positioning, pepper-pot emittance diagnostic, beam shape optical transmission radiation, OTR, diagnostic) and diagnostic beam line (high-resolution magnetic electron spectrometer, temporal bunch characterization).
ELI: Extreme Light Infrastructure www.eli-beams.eu
81
• Electron beam target station The target stations for electron beams should be flexible in order to accommodate multipurpose experiments. These can range from single electron beam experiments to complex pumpprobe experiments involving the electron beam and multiple laser beams (synchronized with the electron beam). • Electron beam dump The electron beam must be absorbed in a material medium. Depending on the energy of the electron beam and on the atomic number of the material, the beam dump may require several meters to stop the electrons generated with ELI beams (up to 15 GeV / 1 nC / 10 Hz) on a single stage accelerator. • Vacuum system Typical vacuum systems for laser-plasma inter-
action are high-vacuum systems (~10-4 mbar), while conventional accelerators are ultra-highvacuum systems (< 10-8 mbar). This is because the first ones release considerable quantities of gas and the latter ones require ultra-high-vacuum to avoid damage by electrical disruption. At ELI we see no reason to improve the vacuum level. However, the particle sources may release considerably higher gas quantities for the vacuum system due to the high repetition rate (10 Hz and higher). This requires a clever design of the vacuum system in order to keep the high-vacuum necessary for high-intensity laser operation. • Materials High repetition rate laser-plasma electron accelerators will activate the materials in the in the experimental areas. We can anticipate, based on previous experience with smaller scale lasers, that operation of the rooms will be strongly affected by the activation of the room compo-
Particle acceleration by lasers
nents if the materials in the room are not properly chosen. The materials in contact with the primary as well as secondary radiation should have low atomic number and short decay lifetimes such as aluminum, plastics and carbon fiber. Materials as stainless steel, iron or copper should be avoided since they have intermediate decay lifetimes when activated, increasing the radiation levels in the room. • Target station space organization We can foresee two types of target stations: i) basic experiments without electron beamlines and ii) beamline experiments. A possible generic setup for these two possible target stations is presented, as a simplified schematic, in the figures below (2a and 2b). The second type will require a larger area since the electron beam optics and diagnostics will require a considerable length. In Figure 2 we can see a simple schematic of the second type of target station. • Compact laser-focusing in plasma ac-
Chapter 3 celerators
Figure 2: Target station setup in experiments without electron beamlines (a) and with electron beamline (b) experiments.
In current laser-plasma accelerator experiments the laser beam is approximately Gaussian and the focal distances are in the range 0.2-3 meters, depending mostly of the available laser pulse energy. ELI will allow scaling the electron bunch energy by more than an order of magnitude. As a consequence, the required focal distance should scale also one order of magnitude. Such a long focal distance requires large vacuum vessels and laboratory space, introduces instability in the laser pointing and considerably constrains the design of the electron beam optics between acceleration stages. Therefore, the source design would be strongly simplified if the focal distance could be reduced to the meter range. Two possible directions to achieve this reduction were proposed in the previous section: reduction of the beam diameter for nonfully compressed pulses, and the use of plasma lenses. Other possibilities are the use of a divergent mirror with high-damage threshold coating
ELI: Extreme Light Infrastructure www.eli-beams.eu
83
(technique already used at Rutherford Appleton Laboratory, U.K.) or random phase plates.
Ions Based on the presently available projections for the production of proton and ions sources (see Figure 1), the maximum energy at the first phase of ELI project could reach ~200 MeV for protons and ~50 MeV/nucleon for carbon ions. The source design follows from the fact that the dominant mechanism (TNSA) is currently well-known and optimal designs exist to exploit it. They however require a stringent temporal contrast for the laser pulse and the associated sophisticated targetry (ultra-thin target). In order to use the high repetition rate of the laser, one needs also to consider all problems related to such “high-repetition” targetry, i.e. a flexible way of replacing the target once it has been shot by the laser. One possible option could be to use a “film-target” such as a strip of metal
Particle acceleration by lasers
or a very huge metallic surface. However, these solutions might not be adapted for very thin targets (nm scale) where a different type of targetry needs to be implemented (e.g. a series of independently mounted targets or gas-jets). A very general set-up delivering an ion beam is shown in Figure 3. The probe beam can be used to diagnose the electron population at the front side (Antici 2008), the auxiliary beam at the right side can generate protons that can be used for proton radiography of the rear surface field (Romagnani 2005), whereas the film behind the target can be used as detector for protons. The spectrometer (spectro) and Thomson parabola behind the target can be used as complementary diagnostics for detecting proton/ion energies.
Chapter 3 Figure 3: Schematic of an experimental setup for generating laser-accelerated ions. OAP stands for off-axis parabola, T.P. for Thomson parabola.
The diagnostics should comprise: • Spectrometer • Emittance measurement • Source size • Divergence measurements • Source duration • Flux/charge
much higher proton numbers, still providing excellent spatial and spectral resolution and absolutely calibrated. Compound detectors also will measure electron propagation inside solid density material using (γ,n) reactions with real and virtual photons. Those equipments will be used for the initial source as a pool and the expertise shared by several research teams. The diagnostics for higher repetition rate with high spatial resolution will need further improvements.
With the maximum energy planned for the initial ion source, most of these diagnostics, working adequately in the projected energy range, have already been designed, calibrated and used extensively in several laboratories (nuclear activation diagnostics, films, spectrometers for protons and electrons, etc.). For energy resolved beam profiling, new detectors are required to deal with the higher particle energies and fluxes. New detectors based on nuclear activation are currently developed; they could detect the
ELI: Extreme Light Infrastructure www.eli-beams.eu
85
Based on the above source specifications, the following points will be required from ELI in order to be able to deliver the source: • minimum two full power beams + lower power probe (non-harmonic preferably, possibility to be used chirped), all with their own compressors to vary the pulse duration • feedback control of synchronization better than 10 fs • temporally stacked pulses on the same beam to perform plasma conditioning • polarization control (Linear and Circular) • knowledge of pulse energy on target • laser parameters stability (better than 1%) • tunable temporal contrast with the possibility of having high temporal contrast (i.e. a laser pedestal with laser intensity <1010 W/cm2)
• frequency-doubling or plasma mirrors setup will be required if laser temporal contrast does not meet requirements • variable duration for the laser pulses and single shot control of the pulse shape • tightest focusing (with pointing better than 0.1 mrad) • focal spot shaping, wave front metrology and control • repetition rate (single pulse may be needed for specific applications) The interface with other sources represents a strong need to take advantage of unique synergies. For example, coupling with high-energy electron beam production would allow using the ion beams to probe the fields in the electron acceleration and thus to obtain a better understanding of the physical mechanisms. Coupling with X-ray sources would allow using them as
Particle acceleration by lasers
a probing tool for warm dense matter produced by ion beams, thus measuring the plasma parameters and probing the atomic structure of the heated matter. This would allow for micron resolution radiography with highest temporal resolution of ELI experiments. Enough space needs also to be reserved to build an ion optical system for proton-irradiation experiments, which will benefit also from linking one target area to the other. Besides a source for ultimate high particle energy (with reduced repetition rate) it should be possible to operate also a proton source with high repetition rate and a broad variability in the energy and flux, allowing also for more technical relevant applications as well as for studies on the coupling with common accelerator setups (seeding of cavities, etc.). In order to obtain a highly performing ion source,
Chapter 3 we will need to address and optimize the following general issues: maximum charge, energy spread, divergence, emittance, maximum energy. This will likely require some laser pulse shaping (uniform/flattop), the development of high repetition-rate capability (for the detectors, targets, etc.), new detectors (potentially very large for GeV ions detection), and some possible coupling to conventional technology for post-acceleration/focusing. A crucial point to address will be the source reproducibility (laser stability). Since we expect temporary activation during experiments at high repetition rate, the experimental stations should be isolated by thick walls in order to allow the access to other experimental and laboratory common areas. In this way, it would be possible to use permanently the most expensive elements in the laboratory (pulse compressor, beam delivery).
ELI: Extreme Light Infrastructure www.eli-beams.eu
87
Source Development Electrons As already explained above, we are proposing two experimental programs/rooms for laserplasma based electron sources: one dedicated to supply sources for users (previous section) and the other dedicated to source development and advanced source use (this section). The experimental program for electron source development should focus on the most important elements of the roadmap for future laser-plasma accelerators requiring one or more laser beams with the ELI size as well as a large laboratory. When using external guiding (a plasma channel with a length equal to the dephasing length) the maximum energy gain of the accelerator can be optimized. Therefore, external injection can result in a great enhancement of the accelerator efficiency. Moreover, the external injection
may also result in the improvement of the energy spread and emittance of the final electron beam. This improvement is related with the fact that the injector stage can be designed for a higher plasma density than the second acceleration stage and therefore the electron bunch will be accelerated by a more uniform electric field than in the injection stage. At this point we can foresee three techniques for external injection: i) the use of two separated stages with laser and electron beam optics between them, ii) the use of a higher plasma density zone in the beginning of the accelerator, and iii) the use of an all-optical injection scheme. So far, only the last method was demonstrated experimentally with success in a configuration of colliding pulses. The methods i) and iii) require at least an additional synchronized high power laser beam. The method ii) just requires that the plasma source acts as a lens over the laser pulse, changing its intensity and size ac-
Particle acceleration by lasers
cording to the plasma density. Such a experimental setup, with one or two ELI 50 J beams, would be able to produce electron bunches in excess of 1 nC and 10 GeV depending on the configuration. Acceleration in multiple stages will allow to multiply the final energy of the bunch by the number of stages keeping a high level of compactness of the entire system. By means of staging it is possible to achieve higher energy electron bunches, by using a set of synchronized laser beams, than in a single stage accelerator, by using a laser beam with the same total energy. However, the final bunch charge in the staged accelerator is smaller. As an example, the use of four 50 J ELI beams in a four-stage accelerator would produce an electron beam with a charge of 0.78 nC and an energy beam higher than 60 GeV, according to the electron acceleration scaling laws. Using the same scaling laws in the external guiding blowout regime, a laser beam
Chapter 3 with the total energy of 200 J would produce an electron bunch of about 1.25 nC with an energy close to 40 GeV. The long focal length interaction station of ELI will enable to extend scaling of particle acceleration in nearly one-dimensional theoretical expectation that in fact has never been conclusively tested in latest lesser energy laser experiments. For the latter purpose, it is necessary to avoid self-trapping and coherent wakefield without break and with weak transverse focus to prevent transverse emittance blowup. Unlike the contemporary experiments at relatively high plasma density, this high energy capability of ELI allows experiments with lower density and thus higher energy gain with near one-dimensional laser focus. The acceleration of more than TeV electrons with ELI seems to be possible in one stage at lower density than contemporary experiments. This should pave a way to investigate the scaling of the future path in terms of
Ions
the energy frontier. The benefits of staging need to be balanced with the cost of synchronization of different laser beams with the electron bunch being accelerated in addition to the electron beam optics required between each stage. However, staging will be a high demanding technological challenge since a perfect synchronization of the laser and electron beams is required in a distance of several tens of meters.
This section is intended to outline the program that is envisioned to push the development of the ion sources within ELI above the limit that is set by the first stage of ELI (i.e. up to 50 J laser pulses), and that has been addressed in the previous section. Based on the projection of Figure shown above, several hundred MeV up to a few GeV protons could be obtained with the ultimate ELI parameters, i.e. using laser intensities toward 1024 W/cm2. This would represent extremely significant improvements, compared to present state-of-the-art, of ion accelerators. The acceleration length can be as low as a few tens of microns, however with a downside associated with a non-guided process that is a higher beam emittance and divergence. These parameters have not been demonstrated because the required laser parameters are not available. However, based on simulations, they seem rea-
ELI: Extreme Light Infrastructure www.eli-beams.eu
89
sonable. It needs also to be demonstrated, but is equally expected from simulations, that the beam quality can be further improved (mostly in terms of divergence and monochromaticity, as the present ion sources are already extremely bright). This would also be a major achievement and would serve users with great benefits. One should add that at these intensities protons should reach the ultra-relativistic regime and in that regime different acceleration schemes than what mentioned above might occur. The ELI ultrarelativistic feature introduces the dynamics of protons to behave similarly to electrons in the intense laser fields. Because of this feature, it is foreseeable that intense laser pulse from ELI beyond intensity 1024 W/cm2 can drive protons to high energies just like electrons are driven by laser wakefield beyond 1018 W/cm2. The energy gain scaling in this regime is proportional to the laser intensity and since accelerated protons move together
with the laser pulse (i.e just like a piston), the energy gain is adiabatic and thus the energy spectrum is monoenergetic. We will have to take into consideration not only the maximum ion energy that will be reached through the various mechanisms under study, but also the following parameters, in order to decide for the most suited ones for downstream applications: • laser to ions conversion efficiency • ability for beam handling & selection (either through target engineering or conventional solutions, e.g. quadrupoles) • ability for energy selection (with particular target/cylinder/capillary/others) • ability for increased beam stability (energy distribution, particle numbers, emittance)
Particle acceleration by lasers
• increased efficiency, even at modest energies The design strategy for the source development will therefore require to study various paths which can be pursued for ion energy increase, including: • simple laser intensity/energy increase • use of ultrathin targets (requires very high contrast, circular polarization), production of protons and heavy ions • use of small (mass-limited) targets (solids, μ droplets, spray) • exploit/explore new mechanisms/regimes, other than TNSA, at higher intensities (e.g. shock acceleration, transparency regime, ponderomotive acceleration, etc). For this, the preplasma will have to be tuned at the target front, which will require an adjustable
Chapter 3 prepulse • test the use of underdense targets (foam, gas, cluster, spray). Different set-ups will be used to study these various aspects. Once a route is chosen, based on the results obtained throughout the study, a final set-up will be chosen and implemented. For all set-ups used through the course of the study, the issue of target development will be addressed with consideration for repetition rate, and the study of using either structured solid targets (layered, pre-ionized, embedded dots, isolated targets, μ droplets of different size) or of less than solid density where optimum prepulse will need to be adjusted. We will need, along this source development program, to clarify the paths that are today envisioned from theoretical studies and that could lead to energy increase up to the range of
200 MeV - 1 GeV. Such energy levels are extremely attractive to reach as they will open many new applications with high impact. As mentioned above (in the state of the art section), a possible physical mechanism for ion acceleration that could be used to reach high ion energies would be the radiation pressure in solids. In this regime the plasma electrons are pushed steadily by this force, and ions are accelerated in the strong electrostatic field forming a shock-like structure. The use of circularly polarized laser light might further improve the efficiency of such ponderomotive ion acceleration, avoiding strong electron overheating. It has been shown in simulations that it allows one to obtain a quasi-monoenergetic ion bunch in a homogeneous medium by adjusting the laser pulse and plasma parameters. Compared to the target normal sheath ion acceleration, which is the dominant mechanism at the currently (2009) explored laser intensities, the pondero-
motive mechanism allows a better control of the ion energy distribution and acceleration of a much larger number of ions from a smaller target area. It will be also necessary to investigate in parallel the influence of radiation losses increasing with the intensity. A series of PIC simulations, to complete the scaling of Figure and to check the parameters that could be expected for the ion source at the 100 PW range, have been performed (see Figure 4). This analysis, made for ELI PP-WP6, focuses on specifying the source term for the most extreme conditions provided – 133 PW – assuming a beam energy of 2 kJ in a 15 fs duration. The energy spectra from these simulations allow the conclusion that particles with up to a few GeV can be generated with this setup (Table 2 and 3).
ELI: Extreme Light Infrastructure www.eli-beams.eu
91
In the studied configurations the particle beams are not collimated. The divergence can be actually quite large, with denser side peaks of less energetic protons. It was also found that a significant fraction of energetic protons is travelling backwards. Thicker targets make a relatively uniform electron distribution in all the directions with some random spikes. Concerning the conversion efficiency, the 10 and 20 µm thick targets reach lower peak energies, but the proton yield is higher for the protons down to half of the peak energy:
General
Target
Simulation PIC 2D fully relativistic code
Type
Thin solid target
Material
H
Density [g.cm-3]
0.088
Thickness [mm]
1 ; 10 ; 20
Energy [J]
2000
Pulse length [fs]
15
Wavelength [nm]
800
Intensity [W/cm ]
1.6 1023
a0
276
Focusing system (f/#)
f/10
Beam waist [mm]
5.1 (1/e2 intensity radius)
Polarization
Circular
Angle of incidence [˚]
0
2
Laser
• 1 µm, 2-4 GeV range: ~5 x105 protons/ (MeV.mrad. µm) • 10 and 20 µm, 0.5-1 GeV range: ~5x106 protons/(MeV.mrad. µm)
Type
Layout
Table 2: Summary of the input conditions for the Osiris simulations
Particle acceleration by lasers
Chapter 3 Target thickness (mm)
Peak energy (GeV)
Proton FWHM divergence (rad)
Proton
Electron
1
4.6
2.9
0.5
10
1.3
0.9
1.6
20
1.3
0.5
1.3
Table 3: The maximum energy and beam divergence for the different target thicknesses
These numerical simulations foresee an order of magnitude increase of the peak energy relative to the most energetic particles accelerated by todayâ&#x20AC;&#x2122;s operating lasers. They also show that for radio safety concerns, it will be important to optimise the configuration, namely by laser and/or target shaping, in order to improve the beam collimation and, eventually, create or reinforce its mono-energetic features. Further 3D simulations will be performed along the course of the ELI planning phase to accurately confirm the expected parameters of the future source.
Figure 4: Maximum proton energy scaling with laser intensity. Purple dots are experimental values, blue are from simulations. The F/# rectangles show the intensity ranges achievable with the highest intensity ELI beam line using three different focal distances. The two darker blue squares at 1.0x1023 W.cm-2.mm2 show the results from the Osiris simulations on the 1 and 10 Âľm targets.
ELI: Extreme Light Infrastructure www.eli-beams.eu
93
Another potentially promising way to accelerate ions to high energies is to use underdense or near-critical density targets (Esirkepov 1999, Yamagiwa 1999, Sentoku 2000). That approach has recently received much attention (Willingale 2006, Yogo 2008, Antici 2009). Compared to solid targets where laser absorption is limited to the target surface, the laser pulse inside low density plasmas heats electrons on a large volume leading to higher laser absorption. This acceleration regime is also advantageous for applications since less debris are produced. It is more adapted for utilization with high repetition rate lasers. The laser contrast, which can be problematic with thin solid targets, is less detrimental as the gas jets are transparent for laser intensities below 1013 W/cm2. The main acceleration processes in this low density regime are still being debated. It appears from simulations that the acceleration processes depend strongly on the characteristics of the density gradient. We should also mention that, at extreme laser intensities reachable with the ultimate stage of
ELI, the use of underdense targets also offers the potential for bubble regime acceleration at extreme > GeV energies (proton acceleration using mixed ions). The experimental setup designs will benefit from progress that will be accomplished a route to ELI, in partner laboratories, and also in the initial phase of ELI when the initial source will be delivered to users. The amount of work that will have been done at that time will allow to decide on the best experimental set-ups that will be most adequately suited to conduct the above mentioned design strategy for optimal source development. The diagnostic requirements are the same as for the initial source but specific issues will arise from the strong energy increase of the particles produced at extreme laser intensities. There will be an additional requirement for: pion, muon
Particle acceleration by lasers
and neutrino beam detection; neutron beam detection and transport. As mentioned above, ELI, in its final phase, will likely be able to produce various kind of radiations (photons, particles) whose energy will range from MeV to GeV. Besides the number of particles per shot will be huge (>1012 protons in the 5-6 GeV range). The characterization in energy and beam spreading of these particles will be a challenge since no unique detector can be used. This constitutes the specification for the diagnostics to be developed for the source development program. Beyond 100 MeV and in the GeV range one must compose between the brightness of the radiation source and the variety of particles emitted together. Specific measures are required to plan for a beam stop of the bright proton/ion bunches in order to reduce radioactivations within a manageable space of ELI infrastructure. Nuclear high energy techniques could be considered, but these are huge equipments requiring very large experimental area.
Chapter 3
Directions of Implementation at ELI Electrons Laser-plasma accelerators produce short pulse electron beams (typical duration 10 fs) the bunch charge depends on laser energy and can exceed 1 nC with ELI. The emittance is naturally low due to the acceleration in the ion channel. The energy spread of the bunch can as low as a few percent. Except the energy spread, which can be improved with new beam loading techniques, these characteristics make these accelerators ideal for driving free-electron-lasers. A new generation of short period undulators, specially designed for free-electron-lasers with laser-plasma accelerators, was started to be developed. Another application exploring the unique properties of laser-plasma accelerators is the generation of collimated beams of hard X-rays by
betatron oscillation of the electron bunch in the ion channel [11]. Most of the applications of conventional accelerators are also possible with laser-plasma accelerators and sometimes at a reduced cost. Up to now, X-rays with energies of a few megaelectronvolts represent the vast majority of ionizing radiations used for cancer radiotherapy of several million patients throughout the world. Higher quality, more energetic electron beams, such as those produced by laser–plasma accelerators, could be used for radiotherapy and provide better clinical results. It was shown that such beams are well suited for delivering a high dose peaked on the propagation axis, a sharp and narrow transverse penumbra, combined with a deep penetration [12]. The real-time investigation of relativistic particle interactions with biomolecular targets opens ex-
citing opportunities for the sensitization of confined environments (aqueous groove of DNA, protein pockets) to ionizing radiation. However, compared with classical dose rate delivery in radiotherapy (1Gy/min) the very high dose rate delivered with laser–plasma accelerators (1013 Gy/s) may challenge our understanding of biomolecular repair, as ultrafast radiation perturbations may be triggered on the timescale of molecular motions, angstrom or sub-angstrom displacements [12]. Electron beams produced in laser–plasma accelerators can be used to generate secondary radiation sources. The electron beam energy is efficiently converted into multi-megaelectronvolt Bremsstrahlung photons when it interacts with a solid target of high atomic number, providing a submillimetre pulsed-ray source that is significantly smaller (450 μm) and of shorter duration (in the picosecond range) than other sources available today. Ultrashort-ray sources
ELI: Extreme Light Infrastructure www.eli-beams.eu
95
are interesting for several applications, including imaging material compression to high density. A train of short laser pulses may enable recording of movies of dense objects under fragmentation, or of the damage evolution of structures with a spatial resolution of 100 Îźm. Light and flexible devices for non-destructive material inspection would also be interesting, with potential applications in motor engineering, aircraft inspection and security. The ultrashort duration of these particle and radiation beams will provide unprecedented time-resolved measurements down to the motion of electrons on atomic scales, and a zooming onto the two fundamental molecular building blocks, the electron and the atom [12]. A hypothetical but credible scenario which could be achieved is the realization of a TeV collider that is based on laser acceleration and potentially cheaper than the conventional technology. However, this deliberation is based on
future anticipated extension of technology of lasers and others. Another interesting physics objectives is the investigation of the space-time structure using high energy photons and, in particular the Landau-Pomeranchuk-Migdal effect, which is the modification of the interaction of photons and electrons within high density matter. An additional application of laser acceleration is the electron-positron pair creation by laser-accelerated electrons colliding with an intense laser beam. The most attractive and fundamental characteristics of ELI facility is that its suite of beams comes in perfect synchronism with its optical pulses themselves. Thus all the high energy beams (electrons, ions, gamma-rays) are synchronous with one another and with the optical beam, whose structure is typically constituted of ultrafast bunches. This provides a basis for arranging a marriage between different beams to collide or influence the others. An example may
Particle acceleration by lasers
be a set of intense optical beams counterpropagating against a laser accelerated high energy electron bunch. Such a setup may be useful for exploring the highly nonlinear QED effect.
Chapter 3 Ions Laser-accelerated ion beams are currently used mainly in two applicative areas: (i) warm dense plasma generation and (ii) probing of matter and/or electric and magnetic fields, but have potentially very close parameters to other significant applications. At the boundary between condensed matter and plasma physics, the study of warm dense matter (WDM), i.e. of dense matter (1-10 g/cm3) at high temperature (1-100 eV) (Ichimaru 1982), is essential since it is relevant to a wide range of disciplines. Indeed, the matter in this state is found not only in astrophysical objects but also in the transition from solid density matter to the plasma state. Using high currents of laser-accelerated ions presents a significant advantage over other techniques to produce WDM (Patel
2003). Indeed, ions have intrinsically in-depth energy deposition capability. Moreover, being produced in a very short time (few ps), the ions can also deposit energy potentially in a shorter time scale than the expansion time of matter heated to a few eV. This technique allows one to produce sufficiently large volumes of WDM to permit quantitative observation, thus opening the way for measurements of WDM parameters, such as stopping power, equation of state and others. ELI Beamlines Facility by producing a high number of ions at high energy in a compact manner and in a short time-scale, will offer a decisive advantage when producing such WDM by allowing to increase the volume of heated matter and to push the achieved temperatures up to 10 keV range, as required e.g. for laboratory astrophysics.
Protons are also relevant as a diagnostic tool for fusion science, they are e.g. employed in experiments that study laser-plasma or hydrodynamic instabilities or as a tool to diagnose target compression. Short, low-emittance ion sources produced by laser-acceleration are already used to perform pump/probe ps-resolved and Âľm-resolved experiments to probe electric and magnetic fields in plasmas (Borghesi 2002). This technique has proved to be an essential tool, yielding the discovery of entirely new phenomena by diagnosing previously inaccessible observables (Rygg 2008). Besides the possibility to take 2D snapshots of electric and magnetic field structures in plasmas the usually broader energy distribution of protons in combinations with their time of flight effects allow also to track the whole dynamics of field evolution in plasmas (Romagnani 2005, Sokollik 2008). Due to the relatively modest ion energies achieved up to now, this
ELI: Extreme Light Infrastructure www.eli-beams.eu
97
technique is however limited to probing fields in low-density media. For the same reason, dense matter (e.g. shocked matter) probing is also possible using this technique in order to retrieve density (and not field) information, but is presently limited to very thin samples. At relatively modest energies (~100-200 MeV) ELI will allow extending the range of such field/ matter probing application. It will then allow to study electron transport in dense matter (as required for the optimization of secondary X-rays or of nuclear reactions within dense materials to produce positrons or Îł-rays) or shocks in compressed material for laboratory astrophysics and geophysics. Higher ion energies (up to 800 MeV), will allow to radiograph very dense and thick objects (e.g. matter at 50 g/cm2). Here, compared to conventional accelerators (e.g. the LANSCE 800 MeV proton accelerator found at LANL for dense matter probing), the excellent emittance of ELI ion
sources, in combination with its short pulse duration, will be a significant advantage, allowing unprecedented spatial and temporal resolution. The first potential area of application for ion sources produced by ELI lies within the area of bio-chemistry helped by ion beam irradiation for conditioning materials (using radicals, e.g. hydroxyl radical yields in the tracks of high energy C6+ and Ar18+ ions in liquid water) to steer reactions. For this, ELI will offer interesting complementary prospects with cyclotron facility like GANIL as, again, the ion beams will be produced with the possibility of offering simultaneously other sources of companion radiation (e.g. X-rays, electrons, etc). High yield medical applications can also be investigated by using ELI-based ion sources. For medicine, two main applications can be considered: first, proton therapy assisted by lasers
Particle acceleration by lasers
(Malka 2003, Bulanov 2002). Here, the proton energies ought to be within the therapeutic window, i.e. between 70 and 250 MeV with a ~ nA current (1010 part/s) with spectral and spatial beam shaping. Not only protons, but also other type of ion beams, such as carbon or lithium beams, will also be produced in this prospect, as they appear nowadays to be of major interest for hadron therapy. Another potential advantage of such a source is that the energy and quantity of the ions could be adapted from shot to shot (i.e. during the treatment time), which allows a more modular form of treatment compared to conventional accelerators. By fostering development of such technique, ELI would help promoting proton source for hospitals to treat cancer tumors. ELI will allow estimation of the cost and optimization of the source through the guidance of radiotherapists. Moreover, electron and proton beams with a pulse length in the ps or fs time scale will provide a unique opportunity to study new and important biological radiation effects.
Chapter 3 ELI could also be used for time-resolved biology studies. There is a fundamental difference between the impact of a heavy ion and a proton beam. The difference in the delta electron density caused by the impacting ion results in biological effects, which are not known below a microsecond timescale. It manifests itself in certain tumors which are resistant to proton therapy, but not to e.g. carbon therapy. Increasing the dose rate, but keeping the total dose constant could lead to higher biological effectiveness and would allow proton therapy to new types of cancer not applicable today. Here basic research is required to investigate dose dependent effects, like water radiolysis close to cell DNA or multiple impacts within the chemical recombination timescales. Short proton and ion bunches from ELI would increase the accessible range of particle energy and flux by orders of magnitude from the present status. Higher particle energies could also stretch the range for radiobiological research from solar cosmic radiation to the realm of interstellar radiation
background experiments. Reaching extremely high ion energies, in the GeV range, would allow to use ELI in another extremely attractive area, namely as a source for spallation. This opens all the downstream study of neutron beams, or as a source for the transmutation of radio-nucleides. Currently there is only one very costly facility (SNS, recently completed in the USA) that is able to perform transmutation of nuclear waste, moreover, conventional neutron sources are currently not very flexible and are of low brillance. Here ELI would again offer a complementary tool to conventional sources like the SNS to probe matter but in a time-resolved manner that is not possible at the SNS. Indeed, the neutron, produced through D-D monochromatic reactions, should be bunched and keep the short duration of the initial ion source.
Finally, proton beams with energies > GeV could be used for laser-driven heavy-ion collider experiments (e.g. ion beams driven by opposing two ultrarelativistic laser pulses) with about 1 million events per laser shot (Esirkepov 2004). For this, beams containing about 1012 protons in the 5-6 GeV range, as predicted by (Pukhov 2003), could be suitable. Pair creation through laser-proton collisions (Keitel 2004, Di Piazza 2008) would also be a unique application allowed by ELI when using both the highfield laser beams and the extremely high energy proton beams, both produced by the facility. The potential for drastic cost reduction of particle accelerators for cancer treatment is potentially extremely important for medical business. The interest is that, by utilizing laser accelerated protons, the generating laser beams can be transported close to the Gantry-like set-up where the protons are produced close to the patients. Therefore no expensive measures for
ELI: Extreme Light Infrastructure www.eli-beams.eu
99
radiation protection of the proton beam transport as in conventional facilities are required. This arrangement makes also easier to vary the proton dose in terms of energy spectra and pulse duration. If realized, this scheme could stimulate a large market of photo-medical industry with huge investment leverage. In fact, current limitation of particle beam therapy with the conventional accelerators is its high cost. Thus, small laser accelerators could lead to a revolutionary change in the cancer therapy scenery and thus open a vast photo-medical industrial market. Currently there are about 30 conventional particle therapy facilities in operation worldwide: e.g. Loma Linda University in California, USA (http:// www.protons.com), PSI (Paul Scherrer Institut) in Villigen, Switzerland (http://www.psi.ch), HIMAC (Heavy Ion Medical Accelerator) in Chiba, Japan (http://www.nirs.go.jp), Institut Curie in Orsay, France (http://protontherapie.curie.info),
NPTC-MGH (Northeast Proton Therapy Center - Massachussets General Hospital) in Boston, USA (http://www.massgeneral.org). Thousands of patients per year are treated in the existing facilities. The average cost of a conventional proton therapy facility is 70-100 M USD consisting of 40 M for building and radiation shielding, 15-30 M for accelerator itself, and 15-30 M for gantries and other related costs. The cost for a carbon therapy facility is about 2-3 times higher. On the other hand, the best estimate for a laser-based proton facility is currently 4.5 - 7.5 M USD consisting of 0.5 M for building/ shielding, 3-5 M for accelerator (laser + target), and 1-2 M for gantries/others. The main advantage is evidently the drastic decrease in the cost of building and shielding (about 100 times), which could permit rapid spread of commercial proton facilities around the world. The ELI facility can provide a major contribution for the development of future high-quality
Particle acceleration by lasers
and low-cost proton sources for cancer therapy. In particular, ELI can first optimize the ion source, and then in collaboration with industrial partners design and test the prototypes. In particular, shielding and beam transport design will concern magnet systems for electron stopping and ion energy selection, beam stoppers, collimators. The sources to be considered will be primary protons/electrons, secondary protons/ electrons, X-rays and neutrons, implying different shielding materials (lead, tungsten, copper, steel, polyethylene, heavy concrete). Thus the radiation protection, extremely important for practical implementation in therapeutic facilities, can be verified. Co-operation with companies working in the field of conventional accelerators, such as HITACHI (http://www.hitachi-hitec.com/global), IBA (http://www.iba-worldwide.com), VARIAN (http://www.varian.com/us/oncology), SIEMENS (http://www.medical.siemens.com) or ACC-
Chapter 3 SYS (http://www.accsys.com/) is anticipated. In fact these companies can also transfer their know-how and suggest the way to fit conventional techniques to the small laser accelerator machines. Thus, an efficient technology transfer towards commercial sector and back is expected, e.g. in terms of accelerator and gantry coupling, beam transport, particle selection, collimation, monochromaticity, scanning beam systems, etc. It is worth to mention the importance of the closely collaboration between ELI laser accelerator researchers and the conventional accelerator community. In fact, the level of sophistication and the long years of experiences that the accelerator physics community has pioneered and accumulated are among the most spectacular technologies that the last 100 years of human history have witnessed.
Besides the medial applications, protons are used in a variety of industrial applications, ranging from radiography of paintings and art objects to colouring of precious stones. It is therefore likely that a facility, offering high-precision, high luminescence protons beam, better and cheaper than conventional proton facilities, will soon attract industries for their potential applications. However, since the characteristics of the achieved proton beams are still to be refined, a precise definition of the technology transfer between industry and ELI is still premature.
References [1] A. Pukhov & J. Meyer-ter-Vehn, Appl. Phys. B 74 (2002) 355– 361. [2]
S. Mangles et al., Nature 431 (2004) 535–538.
[3]
J. Faure et al., Nature 431 (2004) 541–544.
[4]
W.P. Leemans et al. Nature Phys. 2 (2006) 696–699.
[5]
J. Faure et al. Nature 444 (2006) 737–740.
[6]
C.G.R. Geddes et al., Nature 431 (2004) 538
[7]
J. Faure et al., Nature 431 (2004) 541
[8]
W. Lu, et al., Phys. Rev. ST Accel. Beams 10 (2007) 061301
[9]
A. Butler, et al., Phys. Rev. Lett. 89, 185003 (2002)
[10] R. Bendoyro., et al., IEEE Trans. Plasma Sci. 36, 1728 (2008) [11] M. Fuchs et al., accepted by Nature Physics (2009) [12] V. Malka et al., Nature Phys. 4 (2008) 447
ELI: Extreme Light Infrastructure www.eli-beams.eu
101
Chapter 4 Applications in molecular, biomedical, and material sciences
Author: Ing. Libor Juha, CSc.
Introduction Funding and building such a large-scale facility like ELI Beamlines require providing a conclusive evidence that the infrastructure is able to take up the great scientific challenges of the future. Great challenges in molecular, biomedical, and material sciences (MBM sciences) of the 21st century can be formulated as follows: 1. Measuring the mechanisms of physical and chemical processes at the atomic scale. We need to develop â&#x20AC;&#x17E;camerasâ&#x20AC;&#x153; for making molecular movies and learn how to use them effectively to probe matter in quick motion. 2. Controlling electronic processes in matter. In addition to that, nuclear dynamics following the electronic events should represent a subject of control.
they are present in nature. 4. Nanometre scale imaging of arbitrary objects in their native state, e.g., capturing a living cell at nanometre resolution. Nanometrically resolved dynamics of their responses to various stimuli. Below, we are going to describe how ELI will provide numerous powerful tools allowing efficient research in the above-mentioned areas.
Why MBM sciences need the ultra-short pulses ? Temporal scales of particular processes are as follows: looking at electron dynamics requires resolution better than 1 fs; nuclear dynamics (molecular vibrations, phonon dynamics) have characteristic times in ~ 10 fs; intramolecular dynamics in large molecules take times > 1 ps and only molecular fragmentation, real chemical change, radiative transitions in molecules (fluorescence) occur at longer time scales. Using a single sub-picosecond pulse also avoids the influence of radiation damage to probed system on the measurement.
3. Understanding the complexity - efficient methods should be developed to control and investigate various processes in real, i.e., highly complex, systems in the state as
Applications in molecular, biomedical, and material sciences
Chapter 4 Why MBM sciences need energetic photons and charged particles ? a) because of probing and/or pumping very diluted systems – if we investigate a system containing the target species (e.g. ions, clusters, molecules) at very low concentration, we need very high „concentration“ of photons. Typical examples of such a system represent a cloud of highly charged ions, species doping a certain solid material and/ or atomic, molecular or cluster beams.
tion of the pulse energy is utilized to initiate and/or visualize the particular species and/ or process investigated. We should take into account that most of systems whose investigation is motivated by an application are very complex, e.g., cells, tissues, organisms, artificial nanostructures, geological matter, etc.
Why MBM sciences need ELI ? Since ELI provides a unique opportunity of perfect spatial overlap and temporal synchronization of an fast optical laser beam with beams of ionizing radiation at highest parameters achievable using the technology of the near future. Such a combination allows investigating very early stages of photochemical and/or radiation chemical processes.
b) because of probing and/or pumping a particular process in very complex system – if the system is composed of numerous compounds in several phases, the deposited radiation energy is distributed into many different channels. Thus only a minor por-
ELI: Extreme Light Infrastructure www.eli-beams.eu
105
Directions of implementation at ELI At the MBM station intended to be built and operated at ELI, following main layouts will be constructed, commissioned, and utilized in molecular, biomedical and materials research: a) the pulse radiolysis device with sub-ps resolution should be based on a combination of ELI-driven particle (i.e. electron, proton, highly charged ions) and/or energetic photon beam and properly delayed portion of the primary ELI beam for analysis of particle-generated transients, b) ELI-driven electron beam will serve in the time-resolved electron diffraction apparatus; timing with a portion of the ELI optical beam allow to investigate structural changes in photo-transforming systems,
c) coherent XUV/X-ray sources operated at ELI will be used for diffractive imaging of various objects, from single molecules to living cells; timing the short-wavelength beam with the long-wavelength one offers a possibility to investigate a fast response of the investigated system to high fluxes of low-energy photons, and d) ELI-drive short-wavelength sources allow looking also at spatio-temporal momentum patterns of photo-electrons and secondary electrons; this layout may provide important information on fast electronic and structural dynamics of highly energized molecular and supra-molecular systems. Coupling the short- and long-wavelength laser fields allows to clock the system on the very fine time scale (optical streak camera with ultrahigh temporal resolution will be developed).
Applications in molecular, biomedical, and material sciences
Key research topics X-ray coherent imaging with atomic resolution Measuring atomic-resolution images of materials with X-ray photons during chemical reactions or physical transformations resides at the technological forefront of X-ray science. New Xrayâ&#x20AC;&#x201C;based experimental capabilities have been closely linked with advances in X-ray sources, a trend that will continue with the impending arrival of X-rayâ&#x20AC;&#x201C;free electron lasers driven by electron accelerators. X-ray diffraction has proven to be among the most powerful tools for determining the atomic structures widely used in scientific disciplines ranging from biology to solid state physics. Although all objects scatter X-rays, crystal formation has been an essential step in the measure-
Chapter 4 ment of atomic resolution structures for the majority of materials. Because of the regular arrangement of atoms in a crystal as opposed to an amorphous material, the X-ray scattering from the repeating structural unit adds coherently at periodically arranged Bragg peaks. This amplification greatly reduces the X-ray fluence required to measure a high-resolution diffraction pattern, but away from the Bragg peaks the scattered intensity remains too weak to be measured, losing much of the information contained in the continuous molecular transform of the underlying structural unit. Even so, crystallography has remained the only method to measure diffraction at the large scattering angles required for the determination of high-resolution structures. The inability to crystallize important samples, however, has impeded progress in materials science and structural biology [1]. Measuring the continuous X-ray scattering pattern directly from a nonperiodic object provides
an alternative method for structure determination [2]. A real-space image with atomic resolution can be obtained from the measured X-ray scattering pattern computationally. The numerical determination of the image requires that the phase of the diffracted light has to be determined. However the scattering pattern contains the information on the intensity of the scattering radiation rather than the amplitude and thus no phase information can be directly measured. A variety of methods have been developed for alleviating the information deficit in crystallography, such as examining the wavelength dependence of the diffraction pattern near an atomic absorption edge or by knowing part of the structure or a similar structure. With coherent diffractive imaging, an alternative route to reconstructing the scattered X-rays into an image can be used. Sayre has noted that the continuous diffraction pattern of a coherently illuminated unit cell contains twice the information obtained from the diffraction pattern of a crystalline arrangement of identical copies of that unit cell [2,3]. If
ELI: Extreme Light Infrastructure www.eli-beams.eu
107
adequately sampled, this pattern provides the exact amount of information needed to solve the phase problem and deterministically invert the X-ray scattering pattern into an image of the object. This imaging technique requires the illuminating X-ray beam to maintain phase coherence across the entire width of the object. For coherent imaging, as with crystallography, X-ray exposure determines the achievable resolution, and radiation damage sets the maximum dose. The necessity of limiting the dose without the benefit of Bragg amplification inhibits coherent imaging from achieving atomic resolution at synchrotrons designed to produce high average, but low peak, flux, and has limited the technique to relatively large objects [4,5]. On the other hand, free electron lasers operating in the extreme ultraviolet have demonstrated that â&#x20AC;&#x153;instantaneous imagingâ&#x20AC;? can be achieved [6] and X-rayâ&#x20AC;&#x201C;free electron lasers (LCLS, XFEL)
Figure 1: (A) Schematic representation of photogenerated softening of the interatomic potential in InSb. (B) Time-dependent distribution of atomic positions following bond softening. The initial Gaussian distribution broadens linearly with time and with a velocity determined by the root mean square atomic velocities before laser excitation. (C) Schematic representation of a photogenerated shift in the equilibrium bond length in a bismuth crystal. (D) Time-dependent distribution of atomic positions after displacive vibrational excitation. The frequency of the coherently excited vibration determines the period of the oscillation in average atomic position, whereas the magnitude of the shift in equilibrium position determines the amplitude of the oscillation. Reproduced from ref. [8]; for the original data see references cited therein.
Applications in molecular, biomedical, and material sciences
Chapter 4 [7] have the promise of making atomic resolution imaging possible [8]. Delivering the radiation dose to the sample in an extremely intense single X-ray pulse before the onset of radiation degradation provides a strategy for highresolution X-ray imaging without crystallization [9,10]. Moreover, ELI would make possible to study non-equilibrium structural dynamics of transient phenomena initiated by ultrashort laser pulse in femtosecond and attosecond time scale.
Figure 2: Schematic depiction of single-particle coherent diffractive imaging with an XFEL pulse. The intensity pattern formed from the intense X-ray pulse (incident from left) scattering off the object is recorded on a pixellated detector. The pulse also photoionizes the sample. This leads to plasma formation and Coulomb explosion of the highly ionized particle, so only one diffraction pattern (a single two-dimensional (2D) view) can be recorded from the particle. Many individual diffraction patterns are recorded from single particles in a jet (travelling from top to bottom). The particles travel fast enough to clear the beam by the time the next pulse (and particle) arrives. The data must be read out from the detector just as quickly. Reproduced from ref. [8].
ELI: Extreme Light Infrastructure www.eli-beams.eu
109
X-ray holography with atomic resolution Another method for imaging small atomic clusters is X-ray holography with atomic resolution [11,12]. In contrast to X-ray coherent scattering, X-ray holography records three-dimensional image on a two-dimensional recording medium and, in principle, all the information needed for the image reconstruction can be obtained in a short single shot. The holograms contain information not only on the intensity but on the phase of the scattered radiation as well. Namely, in the case of X-ray fluorescence holography, the fluorescence emitted by the atoms of a selected element (emitters) is used as a holographic reference wave. The portion scattered by neighbouring atoms constitutes an object waves. The interference of the reference wave and the object wave constitutes the ho-
Figure 3: (a)-(d) Holograms of CoO measured at Ephot 足6925, 13 861, 17 444, and 18 915 eV, respectively. (e) Atomic images obtained from the combination of these measurements. Reproduced from ref. [14].
logram. Several concepts of X-ray holography with atomic resolution using X-ray fluorescence [11-14] or anomalous X-ray diffuse scattering has been developed [15-17]. Nowadays, a large number of emitter atoms have to be illuminated in order to get a detectable signal even when using most powerful
synchrotron sources of x rays. Then a numerical reconstruction of measured holograms provides an average neighbourhood of all emitters illuminated by an incident beam. Consequently, useful structure information can be obtained only in the case where all the emitters have identical environments. This fact strongly limits the samples suitable for holography experiments. Obviously, all single crystals fulfil this criterion.
Applications in molecular, biomedical, and material sciences
Chapter 4 However, whereas orientational order is necessary, translational periodicity is not required and, in principle, the holography can be applied to small clusters of atoms like protein molecules, viruses, nanostructures etc. when using a much more powerful source of X-rays. Again, radiation damage by deposition of the energy of X-ray photons into the sample limits the structural studies on non-repetitive and/or nonreproducible structures. However, as shown by computer simulations [9], useful structural information can be obtained before radiation damage destroys the sample when using high X-ray dose rates and ultra-short exposures, i.e. femtosecond X-ray pulses generated at ELI.
Time-resolved X-ray diffraction The extension of time-resolved X-ray diffraction, widely used in nanosecond and picosecond time domains [18,19], to the femtosecond domain is an important challenge, as the nature of chemical reactions and phase transitions is determined by atomic motions on these timescales. An ultimate goal is to study the structure of transient states with a time resolution shorter than the typical period of vibration along a reaction coordinate. Recent developments in time-resolved X-ray diffraction, using both synchrotron and laser-plasma based sources, have led to the capability of directly observing structural dynamics on picosecond and femtosecond time scales. This has resulted in a number of novel experiments, including the investigation of structural changes induced by short-pulselaser irradiation of organic materials [20,21] as
well as ultra-fast laser-induced phase transitions [22-24]. The use of standard X-ray diffraction methods (monochromatic and Laue single-crystal diffraction, powder diffraction) in a pump-probe mode using still shorter pulses (ELI is supposed to provide femtosecond or attosecond flashes) may be very beneficial to the study of ultrafast changes in the electron density in materials induced by high-intensity ultrashort laser pulses available at the ELI Beamlines Facility.
ELI: Extreme Light Infrastructure www.eli-beams.eu
111
Sub-picosecond pulse radiolysis The pulse radiolysis method represents a general technique for the study of fast chemical reactions and related processes. In particular, this technique contributed significantly to understanding the chemical and biological action of ionizing radiation. The technique of pulse radiolysis is based on the external perturbation of a chosen system by a nanosecond (recently picosecond) to microsecond pulse of ionizing radiation (typically 10-MeV electrons) of sufficiently high intensity to produce an instantaneous concentration of transient species which is high enough to be directly observable (mostly due to time-resolved optical absorption, luminescence, EPR, electrical conductivity, etc.). The pulse radiolysis celebrates fifty-year anniversary right now, because its development became feasible in 1959 upon the availability of pulsed electron
accelerators exhibiting appropriate pulse durations and energy contents. Actually, it is one of the earliest pump-probe techniques. More details on earlier stages of the development of this technique can be found for example in [25,26]. The advent of technologies providing ultra-short laser pulses in the eighties triggered the new period in the pulse radiolysis development allowing the study of primary radiation chemical processes with a picosecond temporal resolution. In the currently used techniques, the ultrashort laser pulse is split into two parts. One of them represents and/or drives a pulse beam while another one serves as a probe. There are two particular ways how to utilize these pulses in the pulse radiolysis. a) The first portion of the light pulse triggers the photo-injector in a suitable electron accelerator providing an electron bunch
which can be well synchronized with the second part of the light pulse. The second light flash may be delayed using an adjustable path length to be used as a probe of transient events initiated by the electron bunch. More details on such devices and their use can be found in [27-29]. b) The ultra-short laser pulses have sufficient intensity to ionize irradiated system due to multiple photon ionization. Therefore, the optical laser can provide a pump beam directly. The second part of the beam helps to follow the event, usually to be converted into synchronized white light flash. For more details see [30-33]. The temporal resolution achieved with the first class of ultra-fast pulse radiolysis devices remain at a picosecond level [29]. Following the latter way, we may achieve a sub-picosecond resolution [30]. However, this arrangement is
Applications in molecular, biomedical, and material sciences
Chapter 4 associated with several difficulties (e.g., only a minor part of absorbed energy is ionizing the matter; a significant portion of the deposited energy is dissipated resulting in a temperature jump [32] which makes the interpretation of results difficult) and it should always be kept in mind that the pulsed laser beam ionizes irradiated matter in a different manner than the real, high-energy ionizing radiation. The picosecond resolution limit should be broken because products of radiation induced chemical reactions are determined by rapid primary processes [34] such as energy transfer, thermalization and solvation. Some experiments carried out using the above mentioned methods demonstrate that these initial events occur in liquid water on time scales < 5 ps and involve a complicated interplay between electronic relaxation and vibrational energy redistribution. These experiments also show that the chemical processes originating from ionizing radia-
Figure 4: Fig. 4. Primary events initiated by high-energy ionizing radiation of liquid water [33]. The wide impact of ionizing radiations inducing electronic or vibrational excitations and ionizations concerns inelastic interactions of energetic charged particles and X- or γ-rays with water molecules. Most of these physical and chemical steps take place in less than one picosecond (< 10-12 s) and involve confined spaces commonly called spurs. The interaction of ionizing radiations (relativistic particles, γ- or X-rays) with water molecules induces energy deposits and secondary electrons in tracks. These physical events trigger the formation of nonequilibrium electronic configurations (quasi-free or dry electron eqf−), p-like prehydrated electron ep−, electron–radical pairs and fully hydrated electron ground state es−. Simultaneously, the primary water molecular cation (H2O+) reacts with surrounding water molecules via an ultrafast ion–molecule reaction. The latter event occurs in less than 100 fs, yielding a strong oxidant (hydroxyl radical OH) and hydronium ion H3O+ (hydrated proton). This reaction is likely one of the fastest that occurs in polar molecular solvents and represents an ideal case to learn more about ultrafast proton transfer in water. A favorable structured environment induced by an aqueous hydronium ion (H3O+)aq and hydroxyl radical (OH)aq can be created before an excess electron gets to an equilibrium state.
ELI: Extreme Light Infrastructure www.eli-beams.eu
113
tion are unique and cannot be reproduced by laser photons alone. Due to the lack of a suitable femtosecond source of ionizing radiation, knowledge of the primary processes in radiation chemistry and radiobiology remains poor. The large variety of femtosecond sources of ionizing radiation, which could be in the pulse radiolysis technique utilized as a pump, synchronized with the UV-Vis-NIR laser beam, which would be used as a probe, will be available at the ELI Beamlines Facility. Initial pulse radiolysis experiments [33,35,36] have already been carried out using the ultraintense, ultra-fast laser beams both generating and accelerating electrons to MeV energies. Their results clearly indicate that an application of more advanced sources of this kind could shed light on the course of elementary ionizing events in nascent spurs formed in irradiated liquid water, schematically shown in Fig. 4 [33]. Experiments with femtosecond pulses of
ionizing radiation promise providing the direct information about the magnitude of primary ionization events, depending up to now just on indirect approaches such as the inverse Laplace transformation of a concentration dependence of the solvated electron yield using scavengers or stochastic modelling of non-homogeneous radiolytic events with Monte Carlo methods.
Influencing and probing diluted systems As we mentioned in the Introduction, very high â&#x20AC;&#x17E;concentrationâ&#x20AC;&#x153; of photons and charged particles provided by various ELI-driven secondary sources allows to investigate very diluted systems, i.e., systems containing the target species (e.g. ions, clusters, molecules, nanocrystals) at very low concentration. Typical examples of such systems represent streams of weakly bound (van-der-Waals) clusters and/or cluster ions, which we describe below in details. Photochemical processes of molecules in bulk environments are dictated by the interaction of molecules with the solvating species. By the way of example, the photoexcitation in aqueous systems can lead to the production of the H3O
Applications in molecular, biomedical, and material sciences
Chapter 4 molecular radical (or solvated electron) which in turn can interact with embedded species such as various biomolecules and transfer the excitation to these molecules leading to their dissociation. On the other hand, the solvent molecules can change the photochemistry of embedded molecules by closing some dissociation channels. Such processes are relevant e.g. to DNA radiation damage, etc. However, the understanding to these complex processes at the molecular level can hardly be provided by the condensed phase studies due to the complexity of the systems involved. The gas phase studies of the isolated molecules, on the other hand, neglect the important action of the solvent and therefore the insight is also very limited. Here molecular clusters were established as a tool to reveal the details of photodissociation and photochemistry of solvated molecules. The relevance of cluster studies can range from biological species (e.g. heteroaromatic ring molecules in water clusters) to chemistry of atmosphere (hydrogen halides, HNO3, NOx, etc. in
water clusters and nanoparticles). We propose to build, at a specialized MBM station operated at ELI Beamlines Facility, a molecular beam apparatus to produce free molecules and clusters in vacuum. Additionally, an electron gun will be used to excite the molecules and clusters and eventually to produce negative clusters by an electron attachment or positive ones by electron impact ionization. The neutral or charged species will be subsequently exposed to ultra-intense flash of VUV/XUV/X-ray radiation. Several tools will be used to analyze the interaction outcome: Ion imaging technique will provide 3-D information about the velocity of a dissociation product, from which the details of the photodissociation dynamics can be revealed. Quadrupole and time-of-flight mass spectrometry will analyze the cluster composition and intracluster reactions.
Thus a versatile apparatus will be obtained for the study of a variety of important processes in molecules and clusters: e.g. photodissociation at low energies (< 10 eV), and photoionization at higher energies (>10 eV). These processes differ in the bare molecules and clusters pointing to the important effects of solvation. The tunability of ELI-driven sources will provide a unique insight into these processes. For example by scanning the radiation energy the experiment can be walked through the landscape of the system potential energy hypersurfaces, where various reaction channels can be opened at various photon energies. Enormous number of photons in a single pulse provided by numerous ELI-powered sources would allow studying processes with extremely small photo-ionization and/or photo-dissociation cross-sections. Another interesting option is represented by studying not so rare processes but occurring in low-abundant species.
ELI: Extreme Light Infrastructure www.eli-beams.eu
115
out also with molecules, molecular ions, covalent clusters, nanocrystallites, nanodroplets, and so on. Such research has strong sources of motivation in areas of plasma physics (especially in the controlled fusion research and development and plasma astrophysics), plasma chemistry, cosmochemistry and astrobiology, chemistry of lower and upper atmosphere, chemical catalysis, bioorganic chemistry, nanotechnology, tailored syntheses of molecules (e.g., pharmaceuticals) and advanced materials, etc.
Figure 5: Proposed experimental arrangement for a VUV/XUV/X-ray beam-induced photoionization and photodissociation of vander-Waals clusters, to be implemented at MBM station of ELI.
The simple layout proposed in Fig. 5, which is based on a single ELI-provided beam, would be in the second stage of the project turned into a pump-probe scheme combining two pulses in time and space to investigate ultra-fast dynam-
ics of various elementary events in clusters with a femtosecond temporal resolution. Experiments described above for van-der-Waals clusters can be in the proposed scheme carried
Surfaces and interfaces of condensed phases can be also considered as diluted systems because of their restricted volume, i.e., the strongly limited number of their constituents interacting with the incident radiation. Some experiments (see for example [37]) have already been done with high-order harmonics [38] in this area, but these investigations suffer from relatively low number of photons in the probe. The ELI driven sources of XUV radiation can be used for real-
Applications in molecular, biomedical, and material sciences
Chapter 4 time observation of electron transfer processes at interfaces [39]. Electron transfer processes of valence electrons constitute the fundamental step in photochemistry, electrochemistry, and electron transport across boundaries. Understanding the underlying mechanisms has important implications for a range of fields from materials to life sciences. The time scale of electron transfer is determined by the interaction strength of the donor and acceptor states. For weakly interacting states, e.g. image potential states at metal surfaces, the electron transfer dynamics has been studied in the last decade by femtosecond laser pulses. For stronger covalent interactions, electron transfer occurs on an attosecond time scale and has been inaccessible in the time domain so far. Frequency-domain experiments using high-resolution resonant X-ray spectroscopy indicate transfer times well into the attosecond regime, e.g. for sulphur covalently bound on a transition metal substrate. However, this method does not provide real-time resolution as in pump-probe ex-
periments and is based on assumptions about the relaxation dynamics. A question highly relevant to fundamental physics is the dynamics of electron correlations and screening of charge carrier excitations in solidstate materials with high electron density. Theoretical estimations show that the screening is built up within the inverse plasma frequency, i.e., well below 1 fs. Various ELI-driven XUV sources with its attosecond XUV-pump/XUV-probe capability offer a unique opportunity to address these problems experimentally. Fig. 6 [39] illustrates possible generic schemes for probing electron transfer and screening processes at interfaces in a wide range of systems in real time. In a more general context, these studies aim at probing the evolution of non-stationary electron wave functions in condensed matter. In scheme (a) a XUV ultra-short pulse hν1 resonantly populates an occupied state in an adlayer. A subsequent pulse hν2 monitors the ultra-fast electron
transfer to the substrate, which provides a sink for electrons. The electron transfer between the solid (acceptor) and the adlayer (donor) is thereby analyzed. Various model interfaces such as Bi/Si, CO/Pt, and H2O/metal allow addressing key questions for material science, catalysis, and life science. On the other hand, scheme (b) shows that charge screening can be studied by probing the change in binding energy upon electron injection into an unoccupied state of the adlayer. Due to electron-electron interaction the binding energy of a shallow core level will be renormalized upon electron injection. This change in binding energy is monitored in the photoelectron kinetic energy by a time-delayed second XUV pulse.
ELI: Extreme Light Infrastructure www.eli-beams.eu
117
Potential for applications, business and technology transfer Numerous particular problems, both technical and scientific, could be of great interest for industry: A. Better knowledge on early stages of radiation processes occurring on solid surfaces of various materials can be used for an optimization of industrial radiation processing.
Figure 6: Proposed schemes for attosecond real time probing electron transfer and screening processes at interfaces [37].
B. Ultra-high emittance beams of charged particles driven by ELI and above mentioned probe techniques will serve for both development and optimized use of compact particle sources for radiation therapy of cancer. C. The ultra-short pulses carrying an enor-
Applications in molecular, biomedical, and material sciences
Chapter 4 mous number of energetic particles make it possible to investigate behaviour of complex systems composed of many chemical constituents in several coexisting phases. Such a complex system, e.g., living tissue, polymer composite and/or industrially polluted water represent a typical subject of interest in numerous industrial and biomedical applications.
mercial sector absorbing and stimulating results obtained from layouts looking at elementary molecular processes with femtosecond and sub-nanometer temporal and spatial resolution, respectively.
D. Ultra-fast signal transmission, detection and timing techniques developed for MBM experiments could represent germs for innovative activities, especially in the communications and aerospace industry. E. Procedures and techniques developed for time-resolved imaging would very likely represent an impulse for further development and major upgrades of conventional techniques of biomedical imaging. F. Last but not least, pharmaceutical industry should represent the most motivated com-
ELI: Extreme Light Infrastructure www.eli-beams.eu
119
References [1]
Z. Dauter, Acta Crystallogr. D62, 1 (2006).
[2]
D. Sayre, Struct. Chem. 13, 81 (2002).
[3]
D. Sayre, Acta Crystallogr. 5, 843 (1952).
[4]
J. W. Miao, H. N. Chapman, J. Kirz, D. Sayre, K. O. Hodgson, Annu. Rev. Biophys. Biomol. Struct. 33, 157 (2004).
[5]
H. N. Chapman et al., J. Opt. Soc. Am. A 23, 1179 (2006).
[6]
H. N. Chapman et al., Nature Phys. 2, 839 (2006).
[7] (a) XFEL – The European X-Ray Free-Electron Laser. Technical Design Report (Eds. M. Altarelli et al.). Hamburg DESY 2006-097. (b) J. D. Bozek, Eur. J. Phys. Special Topics 169, 129 (2009) and references cited therein [8]
K. J. Gaffney and H. N. Chapman, Science 316, 1444 (2007).
[9]
R. Neutze, R. Wouts, D. van der Spoel, E. Weckert, J. Hajdu, Nature 406, 752 (2000).
[10] J. C. Solem, G. C. Baldwin, Science 218, 229 (1982).
[17] M. Kopecký et al., Phys. Rev. Lett. 100, 195504 (2008).
[35] V. Malka et al., Nature Physics 4, 447 (2008).
[18] J. Wark, Contemp. Phys. 37, 205 (1996).
[36] D. A. Oulianov et al., J. Appl. Phys. 101, 053102 (2007).
[19] Time-resolved Diffraction, Eds. J. R. Helliwell, P. M. Rentzepis (Oxford: Clarendon Press 1997).
[37] C. Lei et al., Phys. Rev. B 66, 245420 (2002).
[20] C. Rischel et al., Nature 390, 490 (1997). [21] F. Schotte et al., Science 300, 1944 (2003). [22] B. J. Siwick et al., Science 302, 1382 (2003). [23] A. Rousse et al., Nature 410, 65 (2001).
[26] The Study of Fast Processes and Transient Species by Electron Pulse Radiolysis, Eds. J. H. Baxendale and F. Busi (Reidel 1982). [27] T. Kobayashi et al., J. Nucl. Sci. Technol. 39, 6 (2002). [28] J. Belloni et al., Nucl. Instrum. Meth. Phys. Res. A 539, 527 (2005).
[12] M. Tegze, G. Faigel, Nature 380, 49 (1996).
[31] Y. Gauduel, in Lasers in Chemistry, Vol. 2, Ed. M. Lacker (Weinheim: Wiley-VCH), p. 861 (2008).
[15] M. Kopecký, J. Appl. Cryst. 37, 711 (2004). [16] M. Kopecký et al., Appl. Phys. Lett. 87, 231914 (2005).
p. 18; http://www.extreme-light-infrastructure.eu/pictures/ELI-scientific-case-id17.pdf
[25] M. S. Matheson, L. M. Dorfman: Pulse Radiolysis, (Cambridge(MA)London: The MIT Press, 1969).
[29] Y. Muroya et al., Rad. Phys. Chem. 77, 1176 (2008).
[14] M. Tegze et al., Phys. Rev. Lett. 82, 4847 (1999).
[39] F. Amiranoff et al., Proposal for a European Extreme Light Infrastructure (ELI),
[24] A. M. Lindenberg et al., Science 308, 392 (2005).
[11] A. Szöke, in Short Wavelength Coherent Radiation: Generation and Applications (AIP Conf. Proc. No. 147), Eds. D. T. Attwood and J. Boker (New York: AIP), p. 361 (1986).
[13] T. Gog et al., Phys. Rev. Lett. 76, 3132 (1996).
[38] D. Riedel et al., Rev. Sci. Instrum. 72, 1977 (2001).
[30] A. Migus et al., Phys. Rev. Lett. 58, 1559 (1987).
[32] R. A. Crowell et al., J. Phys. Chem. 108, 9105 (2004). [33] B. Brozek-Pluska et al., Rad. Phys. Chem. 72, 149 (2005). [34] R. A. Crowell et al., Nucl. Instrum. Meth. Phys. Res. B 241, 9 (2005).
Applications in molecular, biomedical, and material sciences
Chapter 4
ELI: Extreme Light Infrastructure www.eli-beams.eu
121
Chapter 5 Plasma and high energy density physic
prof. Ing. Jiřří Limpouch, Csc. / Ing. Jaroslav Nejdl
Introduction The ELI Beamlines Facility will open chances to study previously inaccessible states of matter and interaction regimes of laser, X-ray and charged particle beams with various targets. The focused intensity of ELI laser will be up to three orders of magnitude higher than on any existing laser, and consequently, the parameters of the secondary sources (X-rays, charged particle beams) will be massively superior to the existing ones. The intensity will be so high that electrons will oscillate ultra-relativistically with Îł up to 1000 in the laser field, and ion oscillations will become relativistic. Plasma with relativistic bulk electron temperature Te â&#x2030;Ľ 500 keV will be formed and thermal bremsstrahlung will generate Îł-radiation capable to produce electronpositron pairs massively. The laser-accelerated ion beams and X-ray laser beams will be able to heat macroscopic mount of solid density matter isochorically to keV temperatures. Thus, fasci-
nating, fundamentally new plasma physics will be investigated on ELI infrastructure. The additional essential advantage of the ELI infrastructure will be the synergy of laser and secondary sources. Thus, one experiment can use synchronized laser and X-ray and/or particle beams for plasma production and interaction and/or for diagnostics. For example, electric fields in laser-produced plasma will be detected by beam of accelerated protons. First, the development of the secondary sources (X-ray and charged particle beams) of ELI will benefit from the research of laser-target interactions. There is still vast space for improvement of the existing schemes for X-ray emission and charged particle acceleration, and innovative schemes will be proposed based on laserplasma interaction research. Second, plasma is a suitable medium for parametric amplifi-
cation and compression of laser pulse. It can sustain much higher intensities than any other medium and application of stimulated Raman (SRS) and Brillouin (SBS) scattering is currently being explored experimentally. Third, the envisioned research will have impact on many other physics and science field. For example, novel fusion schemes will be proposed and explored. High energy density systems will be formed with parameters near to or scalable to those important for astrophysics and thus the basic data for modelling certain astrophysical systems will be obtained. Laser-produced plasmas may be also utilized in the technology of the classical accelerators as ultraintense electric or magnetic lens focusing intense particle beams (e.g. Large Hadron Collider) to a significantly narrower focal spot than via classical technology. ELI will be a unique user facility and the experimental studies of plasma and interaction physics, and of high energy density physics will
Plasma and high energy density physics
Chapter 5 be proposed and guided by user teams. Consequently, the research topics proposed here cannot be complete by any means, and they are not intended for setting any limitations on the future research. They can only serve as guidelines setting an insight that is required for the design conceptions of the interaction and diagnostic complex. The infrastructure ELI must provide very flexible environment that will enable very broad range of experimental set-ups.
this new interaction regime was published soon together with an overview of its prospective applications [3].
Current status Interactions of high-intensity lasers with targets have been studied since the second half of 1960s, immediately after the invention of laser. The driving idea behind the largest lasers was inertial confinement fusion (ICF). Nanosecond laser pulses were used and the interaction was non-relativistic, the maximum laser intensities were up to 1016 W/cm2. Warm dense matter was produced near to the ablation surface and also by intense shock waves inside solid targets. Solid matter was compressed several times and temperature behind the shock wave reached up to 10 eV. The interaction physics was summarized in Ref. [1], the physics of laser-produced warm dense matter is described in Ref. [2]. With the invention of chirped pulse amplification (CPA), the studies of interactions of much shorter, femtosecond laser pulses with targets started in 1989 and the physical description of
The relativistic intensities were firstly reached at Los Alamos CO2 laser Helios in 1978 [4]. Due to the long wavelength 10.6 µm of the CO2 laser, intensity of 1016 W/cm2 was sufficient for electron to oscillate relativistically. However, the programme of fusion CO2 lasers was abandoned soon, because laser energy was predominantly transformed into fast electrons that preheat the fuel making fuel compression unfeasible. Relativistic intensities were reached again in the end of 1990s by femtosecond solid state lasers, relativistic γ ≈ 25 was achieved in 1999 in LLNL where one of the NOVA laser beams was converted to CPA femtosecond laser [5]. High-power femtosecond lasers are either high-energy long-pulse (~500 fs) low-repetition (10-3–10-4 Hz) Nd-glass-based lasers (Vulcan-PW, RAL, UK, FIREX-I, Osaka, Japan and Omega-EP, Rochester, USA), or lower-energy, short-pulse (~30
ELI: Extreme Light Infrastructure www.eli-beams.eu
125
fs), higher-repetition Ti:Sapphire lasers (Astra Gemini, UK, Texas Petawatt, USA). Extremely sharp focusing lead to the record laser intensity of 2Ă&#x2014;1022 W/cm2 at 300 TW-Ti:Sapphire laser Hercules at Univ. Michigan, USA. ELI will fully exploit the existing laser technologies and at the same time explore the emerging novel approaches such as DPSSL to generate extreme laser power and intensity.
increased due to ion energy enhancement via Coulomb explosion. Strong shock and radiative blast wave in cluster media can also be scaled to astrophysical conditions [10]. Isochoric heating of thin foils by laser-accelerated protons was studied experimentally [11]. We have mentioned only a few important examples, for compilation and presentation of a complete list is beyond the scope of this text.
Research of relativistic laser interactions with various targets is driven mainly by prospective applications. Laser propagation in long underdense plasmas is examined in connection with laser wakefield acceleration of electrons [6]. Laser interaction with thin foils is investigated in order to improve ion acceleration process [7]. Mass-limited targets are used to diminish undesirable lateral energy spread out of the focal spot [8], and thus energy density is enhanced. Cluster targets are used for initialization of various nuclear reactions [9] as the reaction rate is
We would like to stress that all pre-requisites for future successful plasma and high energy density physics already exist. There is a strong European community of researchers with vast experience in experimental, theoretical and numerical studies of relativistic laser interactions with various targets. This community is interconnected as interpretation of experiments is not realistic without numerical simulations as full characterization of the system is not feasible with the required temporal and spatial resolutions. Many experiments are proposed
by theorists and guided by numerical simulations. There is long-time experience with running large user facilities, including lasers, targets chambers and diagnostics. Broad range of methods for diagnostics of laser-target interaction and of laser-produced plasmas, including active and passive optical, X-ray and particle methods, have been already developed.
Plasma and high energy density physics
Chapter 5
Directions of plasma, HEDP and interaction studies at ELI In the following, we shall discuss several important directions of plasma and high energy physics research at the ELI Beamlines Facility. This list is in no aspect complete and we cannot guarantee that the listed topics will be the most important. Firstly, no one can fully foresee the evolution of this booming field in the next five years; secondly the research topics will be selected from the user proposals. This list is intended as a guideline for the design of flexible target area facilities devoted to plasma and HEDP studies, and for the choice of basic onsite diagnostic equipment. The research topics are organized into the following particular topics.
Nonlinear optics of plasmas and laser interactions with underdense plasmas The propagation of high-intensity, coherent, electromagnetic radiation in plasmas, and the resulting modifications of the plasma state, is the subject of the nonlinear optics of plasmas. Coherent radiation can scatter from and decay
into collective plasma modes and can create radiation at new frequencies. The plasma oscillations may become self-organized, nonlinear, kinetic states. Simple modes that have linear and fluid limits can be strongly influenced by nonlinear kinetic modes, such KEEN and KEIN waves. This subject is very exciting from theoretical, computational, and experimental points of view because of the myriad states made possible by the interaction of multimode fields, together with wave-particle resonances, the nonlocality of the self-consistent, plasma-particle dynamics [12].
Figure 1: Proton projection image of wake taken 20 ps after interaction of 35 fs Ti:Sapphire laser pulse of intensity 5Ă&#x2014;1019 W/cm2 with tenuous plasma formed in front of 0.9-Âľm-thick mylar foil by laser prepulse [13]. Small scale structures in the wake are clearly depicted.
ELI: Extreme Light Infrastructure www.eli-beams.eu
127
This topic includes wake formation behind the laser pulse and studies of the wake properties (Fig.1) that are important for electron acceleration. Presently, the interest is concentrated on the bubble regime [14] that leads to formation of monoenergetic beam of electrons accelerated up to GeV energies. However, the limits of electron, positron and ion acceleration in the wakefields are not clear. Wakefields are multidimensional and the electron density in the wake may reach values orders of magnitude higher than the background density. The intensity of ELI laser will be so high that ions could be trapped and accelerated in the wake. The optimum configuration of laser beams for stable propagation over long distances in tenuous plasma is not known to date. The optimum regime for electron acceleration at ultra-relativistic intensities may differ substantially from the present situation. Parametric scattering like SRS and SBS can influence propagation of laser beams and at
Figure 2: Electromagnetic soliton detected in numerical simulations of SBS [16]; blue â&#x20AC;&#x201C; density, red â&#x20AC;&#x201C; transverse electric field squared, green â&#x20AC;&#x201C; magnetic field squared.
Figure 3: The phase space of relativistic KEEN wave plotted in Vlasov simulations of SRS [17].
Plasma and high energy density physics
Chapter 5 high intensities they could create the conditions for trapping laser light and formation of relatively long-lived electromagnetic solitons [15]. These solitons have been observed in three-dimensional PIC simulations and more recently in the nonlinear stage of SBS evolution [16] (Fig. 2). They can stay in the same place for a sufficiently long time (of about tens of ps) or slowly drift if the ambient plasma is inhomogeneous. Fig. 3 shows SRS-induced non-linear KEEN wave. ELI will allow such structures to be manipulated, and used for studies of long-term effects of extremely high electric and magnetic fields on isolated nuclei. Backward Raman and Brillouin scattering can also lead to laser pulse amplification and compression [18] that could be used in interaction experiments or in laser technology. ELI will allow studying this process at much higher intensities.
Relativistic HED plasma ELI will create plasmas with temperatures exceeding 1 MeV. In these plasmas bremsstrahlung photons will have energy sufficient for creation of electron-positron pairs (Fig. 4). This already observed process [19] is particularly efficient with high-Z target materials like Figure 5: Electron and positron spectrum, when 1 ps pulse of intensity 1020 W/cm2 was incident on thick gold target [19].
Figure 4: Diagram of pair creation. Electron colliding with high-Z ion emits bremsstrahlung (Îł photon) that collides with another high Z ion and electron-positron pair is created.
gold (Fig. 5). Creation of plasma with positron density comparable to ion density will be possible only with the next generation lasers like ELI. Understanding how these plasmas behave when either their directed flow or their average energy is relativistic is a fascinating area for research. This includes understanding how to create useful electron-positron plasmas in the
ELI: Extreme Light Infrastructure www.eli-beams.eu
129
laboratory. This topic is important in high energy astrophysics, in the development of compact photon and energetic-particle sources, and in extending the boundaries within which high energy density systems have been explored.
Laser interaction with solid targets and dense matter When an intense laser like ELI is incident on a solid-density target, a complicated suite of phenomena occurs. The electric field in such lasers is orders of magnitude larger than the fields that bind the electrons to the atoms, single free electrons can oscillate with energies in excess of 1 GeV, and the radiation pressure can exceed 10 Tbar. The absorption of an intense laser at a dense plasma boundary is not completely understood because of the complexity of the electron-distribution function and fields (both electric and magnetic) at the interface. Investigation of radiation pressure effects on density steepening and consequently on electron spectrum and temperature is important. Research of laser interaction with solid foils is needed for
optimization of proton and ion acceleration for various applications, i.e. for optimization of fast ion source at ELI. As probe beams one can use either portions of the few-cycle laser pulse itself or secondary electron, ion, or X-ray pulses of similar time duration or even much shorter, attosecond XUV pulses that can be generated by means of laser harmonics. Attosecond time scales govern electron motion not only in atoms and molecules, but also in metals and soliddensity plasmas. When ELI laser will interact with the corona of solid target, laser will penetrate to dense plasma via relativistic transparency and channel boring by ponderomotive force. Recent theoretical paper [20] demonstrates that pulses with intensities exceeding 1022 W/cm2 may penetrate deeply into the plasma as a result of efficient ponderomotive acceleration of ions in the forward direction. The penetration depth as large as hundreds of microns depends on the
Plasma and high energy density physics
Chapter 5 Laser interactions clusters and masslimited targets
Figure 6: Simultaneous subpicosecond time framed interferogram (a) and Schlieren image (b) of a radiative blast wave 24 ns after being launched into a medium of 6 nm Ar clusters by a 700 mJ, 750 fs laser pulse. As material is heated and ionized, the resulting free electron refractive index change causes fringes to bend in the image (a). Image (b) highlights refractive index gradients. Here, a thin shelled blast wave has formed, with a strong radiative precursor leading the compression wave [10].
laser fluence, which has to exceed a few tens of GJ/cm2. The fast ions, accelerated at the bottom of the channel with an efficiency of more than 20%, show a high directionality and may heat the precompressed target core to fusion conditions.
Atomic clusters have been shown to be very efficient absorbers of intense laser radiation. During Coulomb explosion of laser-heated clusters, high energy ions and hard X-rays are produced. The possibility of applying clusters for various applications, such as tabletop nuclear fusion from exploding deuterium clusters and synchrotron radiation, is very challenging. Clusters were recently used to create high energy density plasmas that drive strong shocks (Mach > 50) and radiative blast waves [10] (Fig. 6). The plasma states generated can be described by magneto-hydro-dynamic equations that have a number of similarities. Careful application of these equations and similarities allow experiments to be scaled to astrophysi-
ELI: Extreme Light Infrastructure www.eli-beams.eu
131
cal phenomena that have spatial and temporal scales that are greater by as much as 15–20 orders of magnitude. In this way, the radiative blast waves in the laboratory have been scaled those experienced in supernova remnants and the physics governing their dynamics investigated under controlled conditions. Mass-limited targets are used in order to limit undesirable energy spread out of the laser focal spot. Thus, high energy density deposition is achieved. For example, deposition of 1 J in a droplet 10 μm in diameter would correspond to deposition of about 10 keV for each particle. Since the expansion time of such a droplet is greater than 1 ps, during this time period one can study matter in a quite unusual state of very high density and temperature in the same time. The limited-mass targets made from transparent materials allow one to avoid negative effects associated with pre-pulses and they could be used in many applications, in particular for ef-
Figure 7: (Left) XUV image of proton heating of a 60 µm Al foil at D/ r=1.5, converted to temperature. (Right) X-ray pinhole camera image of a laser irradiated Al hemisphere (where protons are accelerated) with a 100 µm thick Al foil at D/ r=1.8 [22].
ficient ion acceleration to high energies [8].
Plasma and high energy density physics
Chapter 5 Warm dense matter studies At ELI it will be possible to heat macroscopic amount of solid density matter up to keV temperatures either directly by laser or by the secondary sources. While direct heating by laser uses heating by hot electrons confined by self-generated magnetic fields and shock waves launched at material interfaces [21], using ion or proton beam is very straightforward. Isochoric heating of 15 Âľm-thick Al foils by laser-accelerated protons up to 80 eV has already been demonstrated at Gekko petawatt laser [22] (Fig. 7). Equation of state, opacities and transport properties of warm dense matter are key data for astrophysics that will be well measurable with ELI. The proton beam generated by ELI will unite
properties unreached so far by any laser generated proton beam: high energy combined with a high intensity and focusability. This makes it an ideal instrument for a plasma generation. Since the ELI performance covers a broad interval of energies and intensities of the generated proton beam, it will be possible to generate plasmas ranging from a deeply non-ideal state (warm dense matter) to plasmas with high energy content, close to that formed in the tempers meant for a proton-driven inertial fusion. The warm dense matter is an object of physical interest in its own right, not to speak about the astrophysical relevance. Any relevant quantity measured on this exotic state of matter will be of interest. For the astrophysical purposes especially the optical properties like the opacity coefficient in the optical and X-ray region, emission spectra, etc. On the high energy end we are likely to obtain proton-beam-generated plasma in a high power density regime. If such a regime is reached with the laser-generated proton beam of sufficiently high power density, the hot dense
matter beam plasma would become available to the physical research and, moreover, a very interesting transition from a highly non-ideal to the ideal state could be traced by simply varying the properties of ELI generated proton beam. With ELI, it will be possible to create high energy density plasmas and probe it with the secondary hard X-ray source. Thus, extremely important plasma opacities will be measured in this regime.
ELI: Extreme Light Infrastructure www.eli-beams.eu
133
Stopping of proton beam in a pre-generated plasma Plasma heating by proton beams is of interest for both the fundamental research as well as for the particle beam driven fusion experiments. Experiments of this kind have been going on for a long time using combined equipment of lasers to provide the stopping plasma and of accelerators to supply the proton beam. Unfortunately, just a very few laboratories can host experiments of this kind and they are mostly located on the accelerator sites. ELI will be the first purely laser laboratory, where the necessary proton beam could be generated by one of the ELI main laser beams. The stopping of charged particles in plasmas is a complex physical process [23], which can hardly be reduced to binary Coulomb collisions governed simply by the Coulomb logarithm. Even then the Coulomb
logarithm is somewhat vaguely defined quantity and its determination is often provided by a hand-waving argument. Its measurement would thus shed more light on hitherto insufficiently clarified questions of plasma kinetics, especially in non-ideal plasmas. But even in rarefied plasmas the stopping power is determined by other processes then binary collisions. The onset of collective phenomena is to be expected under the effect of the impinging proton beam, whose spontaneously created electric fields will add to the plasma stopping power. This phenomenon might show a resonant behaviour, anomalous dependences on the beam energy and plasma parameters. The expected flexibility of the ELI performance might help map out broad regions of both the stopping plasma and the proton beam parameters not yet explored in the classical way. For the purpose of pre-forming the stopping plasma a part of the ELI pumping beams will have to be diverted on the ns or even ps level.
ELI Beamlines Facility is not designed for complex inertial confinement fusion (ICF) experiments. However, many physical issues of advanced ICF schemes, such as fast ignition or shock ignition can be addressed at ELI. First, transport of high-current electron beams in dense plasmas is crucial for fast ignition via fuel heating by electrons (both for schemes using or avoiding a cone). The ELI laser pulses will drive relativistic electron currents of the order of gigaamperes in overdense plasma targets [24]. In vacuum, electron current transport is limited by IA = 17 βY kA due to magnetic self-interaction. But, return current in a plasma tends to suppress B-fields. The system of counter propagating beam and plasma currents is, however, unstable and leads to beam filamentation and a host of secondary filament dynamics and merging events (Fig. 8). At present, it is unclear to what extent the beam energy can be transported to the compressed cores of fusion targets.
Plasma and high energy density physics
Chapter 5 The filamentation instability may be reduced by transverse beam heating and other effects. Again, the present predictions are based only on 3-D simulations [26] and require experimental verification. The filamentary structures evolve and change on the microscopic plasma time scale, which is in the attosecond regime at these densities. Second, it will be possible to study shock-ignition-relevant strong shock propagation in dense plasmas and shock collision with a counter propagating shock wave at ELI. The temporal visualization of this process is obviously important for shock ignition feasibility and optimization. The vital temporal visualization of the colliding shocks inside the compressed fuel can be done with a hard X-ray or even Îł-ray imaging coming from the ELI main beam.
Figure 8: 3-D-hybrid-PIC simulation showing a continuous 1 GA beam of 1 MeV electrons with 120 keV transverse temperature injected into plasma with density rising exponentially from 2 to 100 g/cm3 after 1.2 ps [25].
ELI: Extreme Light Infrastructure www.eli-beams.eu
135
Directions of implementation at ELI ELI must provide versatile, user-friendly environment for a broad range of plasma and high energy density physics experiments. Experiments will profit from two basic distinctions: a) superior focused laser intensity, and b) synergy of laser and secondary sources. It is not possible to foresee all important experiment proposals, but the brief list above indicates that the possibility to exploit at least two energetic femtosecond laser beams together with at least one beam with longer pulse (noncompressed nanosecond pulse or pulse compressed to ps range) would be vital for many experiments. The experiments may generate X-rays and charged particles for the interaction and/or for plasma diagnostics directly inside the target chamber. However, it may be of advantage for many experiments to exploit the advanced secondary sources available at ELI, i.e. coherent
or incoherent X-rays, electron and ion beams generated in an extra target chamber. Consequently, special ports have to be available for the secondary sources and special equipment for pointing and focusing of secondary sources into experimental target chamber. Very precise synchronization of the secondary sources with the interaction laser beam has to be provided.
fielded. Fast X-ray, Îł-ray and neutron detectors will be provided. Streaked and time integrated X-ray spectra are important part of diagnostics. Charge collectors, Thomson parabolas and electron and ion spectrometer have to be provided. Active optical and X-ray diagnostics using shadowgraphy, interferometry and Thomson scattering must be possible.
The interaction chamber has to be prepared for various types of targets, such as gas jets of various forms, cryogenic gas jets producing clusters, micro-droplets produced by pulse nozzles, capillary targets for laser guiding, thin foils etc. High-precision automatic target manipulation and pointing must be provided.
The interaction chamber has to be shielded properly. The shielding requirements against radiation (Îł-rays, neutrons and activation) have to rely on scaling of experimental electron distribution and on scaling of radiation data from present petawatt experiments. Scaling will rely also on numerical simulations of at experiments at laser intensities expected at ELI. For the first stage the shielding should be comparable with shielding of Vulcan petawatt target chamber [27]. For medium and high-Z targets, an interaction chamber shield blanket of 15 cm lead and 10 cm of high density polyethylene, and a
The interaction chamber must be equipped with the basic optical, X-ray and particle diagnostics. Optical and X-ray streak camera with the best available time resolution have to be
Plasma and high energy density physics
Chapter 5 materials will be used for the interaction chamber construction.
Figure 9: Installation of shielding at Vulcan petawatt target chamber at RAL, U.K.
shielded fire escape door were required. This shielding is combined with the 60 cm thick concrete walls surrounding the interaction area gives an attenuation of Îł-rays with the direct attenuation of 50,000 at 6 MeV. Low activation
The complex for plasma and high energy density physics should include at least two target areas (not counting the target area(s) for the secondary sources like X-rays and laser-accelerated particles), so that two experiments may be carried out and/or prepared in parallel. The target areas may be developed gradually, so that the first target chamber will be ready for the first phase of ELI. This target area will stay in operation in the ELI second and third stage for the experiments that do not demand full ELI power. The main target area has to comply with the requirements for experiments at laser intensities up to 1024 W/cm2. The design can be finalized using experience from novel high-intensity facilities including first stage of ELI.
Complex development for plasma and HEDP studies and its use As an international large-scale user facility, ELI Beamlines Facility must offer interaction areas with excellent parameters and complementary characteristics allowing experimental setups which are not feasible in any other existing laser laboratory. In practice, it means enough flexibility in manipulation and timing of several laser pulses and/or pulses of secondary sources (X-rays and charged particle beams). Taking the advantage of inherent synchronization of all ELI lasers which originate from one master oscillator, it will be possible to perform various interaction experiments with preformed plasmas and pump-probe experiments. The interaction areas must offer possibility of easy and accurate po-
ELI: Extreme Light Infrastructure www.eli-beams.eu
137
sitioning of various types of targets. ELI has to offer a broad range of standard on-site optical, X-ray and particle diagnostics. Active diagnostic methods must be developed, tested and ready for fielding in selected experiments. The fundamental part of the complex for plasma and high energy density studies will be the interaction areas. In the first stage of ELI, the project of interaction area(s) can start from the existing interaction areas at petawatt lasers like Vulcan petawatt interaction area (Fig. 9) mentioned above. The entry ports for incident laser beams should be based on the parameters and geometry of ELI laser beams. Entry ports for secondary sources must be designed in accordance with their projects. The radiation shielding (Îł-rays, neutrons and activation) must be designed according to the data from present experiments extrapolated to ELI parameters. Materials used for the construction of the interaction chamber should have low atomic numbers and short de-
cay lifetimes such as aluminum, plastics and carbon fiber. Materials as stainless steel, iron or copper should be avoided. When the volume of interaction chamber is specified, the vacuum pumping system will be designed. High-vacuum systems (~10-4 mbar) are typical for laser-plasma interaction, ultrahigh vacuum is not needed. The basic requirement is a high pump power enabling pumping of large interaction chamber in a reasonably short time. The stands for positioning of various types of primary and secondary targets should be specified and designed. The basic set of on-site diagnostics should be specified based on a preliminary set of typical experiments like those specified above. The diagnostic ports should be specified. The typical setups for active diagnostic methods should be developed and tested at existing installations.
The plasma and high energy physics complex will be exploited as a user facility with experimental time granted to the most important and challenging project proposals by an international body. Only projects that cannot be carried out on other smaller facilities will be accepted. Often, the investigated plasma object will be prepared by a long or short laser prepulse, or by X-ray or particle source and then the interaction of the main laser beam with the object will be examined. The typical interaction temporal and spatial scales do not allow obtaining sufficiently detailed information using simple diagnostic methods. Thus, auxiliary laser beams, X-rays and/or particle beams will be used for active diagnostics. For example, underdense plasmas will be diagnosed by Thomson scattering of an auxiliary laser beam and dense plasmas will be diagnosed by Thomson scattering of beam of secondary X-ray laser. The electric fields in the target will be imaged by beam of accelerated protons.
Plasma and high energy density physics
Chapter 5 Consequently, the typical experimental set-ups will be complex and thus experiments should be prepared for a relatively long time. The key elements of the setup and of diagnostic methods should be tested at smaller installations, if possible. Strong technical support for the experimental groups must be provided by ELI infrastructure. Experimental time must be awarded only to fully prepared experiments in order to avoid inefficient use of the infrastructure.
Potential for Applications and Technology Transfer Experimental research in plasma and high energy density physics at ELI infrastructure will be mainly oriented to fundamental science. This will accumulate scientific knowledge of new regimes of laser and secondary source interactions with target and investigate unique states of matter that can be prepared only with ELI infrastructure.
mental breakthrough was achieved in 2004 for electrons and in 2006 for ions. New acceleration mechanisms are still being proposed and the acceleration process is not yet optimized even for present day intensities. Numerical simulations and theory predict possibility of significant improvements of the acceleration process at intensities achievable at ELI but an extensive search for suitable experimental set-ups will be inevitable. The possible improvements in the parameters of the secondary sources would significantly enhance their application potential that is discussed in detail in the descriptions of the particular activities.
However, there is also a large application potential of the plasma and interaction physics investigated at the ELI Beamlines Facility. First, the basic physics of laser-target interaction is essential for optimization of the secondary sources. Generation of quasi-monoenergetic electron and ion beams is particularly novel and funda-
Interaction experiments will be directly used in material science, for example for understanding the aging process in construction materials of nuclear power plants. One can think about experiments when ultrafast processes induced in material by laser-accelerated ion beam will be detected by synchronized ultrafast laser or
ELI: Extreme Light Infrastructure www.eli-beams.eu
139
X-ray probe. Indeed, the mechanisms leading to defect creation or phase transformation in materials subjected to low or high-energy ions involve as initial steps ultrafast processes whose elucidation is still a challenge: in the case of elastic scattering, collision cascades – at present observed only through atomistic simulations – or intense and short-lived electronic excitation in the case of high-energy ions – for which essentially two models were proposed long ago; and there is no clear answer yet concerning their respective validity. It is essential in both cases to obtain observations of the state of the target in the first few picoseconds after passage of the particles, and only laser-based sources can make such observations possible. Additionally, ELI laser-target interactions will be intense ultrashort source of neutrons and γ-rays, and ultrafast response of materials on radiation damage could be probed by synchronized laser and X-ray pulses. Such experiments could bring essential progress to radiation physics that is vital for many major technologies.
Intense and short positron pulses produced at ELI may be also interesting from application point of view. Positrons are used in several techniques of material analysis: The lifetime of the positrons in solids, this being in the range of picoseconds, is inversely proportional to the electron density. Positrons are easily trapped at vacancy-type defects, where the positron lifetime is prolonged. The lifetime spectrum, therefore, tells about the type and density of the defect distribution in solids. On this basis, it has been possible, for instance, to build a positron microscope able to observe inside a material the appearance of clusters of vacancies located at the head of a scratch. In addition, the chemical surroundings can be analyzed from the Doppler shift of the annihilation radiation. The Doppler effect can also be used to analyze the momentum distributions of the electrons in the solid in great detail. By focusing ELI laser on high-Z material we shall
be able to produce intense pulse of γ-rays that can be used for nuclear transmutation of longlived radioactive isotopes into less radioactive or short-lived products. This concept is being developed in the world for nuclear waste management. The primary risk isotope is long-lived iodine-129 with high radiotoxicity and mobility, and this may be transformed to iodine-128 that decays with a half-life of 25 minutes to stable inert xenon-128. The experiments may demonstrate the feasibility of laser-induced transmutation process. Laser-produced plasmas can be used in the classical accelerator technology. Laser-ion accelerator can serve as an intense ion source for classical accelerators. Additionally, laserproduced plasmas may serve as ultrastrong lenses for focusing of classical accelerator beam as they can sustain quasistatic electric and magnetic fields higher than any other system. Recently, a micro-lens for focusing and en-
Plasma and high energy density physics
Chapter 5 ergy selection of laser-accelerated MeV protons has been demonstrated [28] (Fig. 10, 11). It is anticipated that with laser of ELI scale it could be possible to develop ultrastrong lenses that could narrow the focus of large accelerator like LHC (Large Hadron Collider) significantly, thus beam intensity and collision probabilities will be enhanced. Figure 10: Schematic of the micro-lens device. A proton beam accelerated from a planar foil by the CPA1 laser pulse propagates through a hollow cylinder side-irradiated by the CPA2 pulse.
The above discussed technologies could be potentially transferred to industry in the future and could bring an important (incl. commercial) profit for the whole society.
Figure 11: RCF layers show that 7.5-MeV protons were selected and focused [28].
ELI: Extreme Light Infrastructure www.eli-beams.eu
141
References [1] M.N. Rosenbluth, R.Z. Sagdeev, A. Rubenchik, S. Witkowski (eds.), Handbook of Plasma Physics vol 3 Physics of Laser Plasma, North-Holland, Amsterdam 1991 [2]
S. Eliezer, A. Ghatak, H. Hora, E. Teller, Fundamentals of Equations of State, World Scientific Singapore 2002
[3] P. Gibbon, E. Förster, Short-pulse laser–plasma interactions, Plasma Phys. Control. Fusion 38 (1996), 769-793 [4]
R.L. Carlson et al., Helios - a 15-TW carbon-dioxide laser-fusion facility, IEEE J. Quantum Electronics 17 (1981), 1662-1678
[5] D.M. Pennington et al., Petawatt laser system and experiments, IEEE J. ST in Quantum Electronics 6 (2000), 676-688 [6] B. Walton et. al., Measurements of forward scattered spectra from intense laser interactions in the forced laser wake-field regime, Plasma Phys. Control. Fusion 48 (2006), 29-37 [7] M. Borghesi et al., Propagation issues and energetic particle production in laser-plasma interactions at intensities exceeding 1019 W/cm2, Laser Part. Beams 20 (2002), 31-38 [8]
M. Schnürer et al., Ion acceleration with ultrafast laser driven water droplets, Laser Part. Beams 23 (2005), 337–343
[9] T. Ditmire et al. Nuclear fusion from explosions of femtosecond laser-heated deuterium clusters, Nature 398 (1999), 489-492 [10] P.A. Norreys et al., Intense laser-plasma interactions: New frontiers in high energy density physics, Phys. Plasmas 16 (2009), 041002 [11] P. Antici et al., Isochoric heating of matter by laser-accelerated high-energy protons, J. Physique IV 133 (2006), 1077-1079
[12] Advancing the Science of High Energy Density Laboratory Plasmas, report of the Panel on High Energy Density Laboratory Plasmas, United States Department of Energy, January 2009
[23] R. Singleton, Charged particle stopping power effects on ignition: Some results from an exact calculation, Phys. Plasmas 15 (2008), 056302
[13] M. Borghesi et al., Plasma Ion Evolution in the Wake of a High-Intensity Ultrashort Laser Pulse, Phys. Rev.Lett. 94 (2005), 195003
[24] G. Mourou et al., Proposal for a European Extreme Light Infrastructure (ELI), Paris 2008
[14] J. Faure et al., A laser-plasma accelerator producing monoenergetic electron beams, Nature 431 (2004), 541-544
[25] J. Meyer-ter-Vehn et al., On electron transport in fast ignition research and the use of few-cycle PW-range laser pulses, Plasma Phys. Control. Fusion 47 (2005), B807–B813
[15] G.A. Mourou, T. Tajima, S.V. Bulanov, Optics in the relativistic regime, Rev. Modern Phys. 78 (2006), 309-371 [16] S. Weber et al., Electromagnetic solitons produced by stimulated Brillouin pulsations in plasmas, Phys. Plasmas 12 (2005), 112107 [17] A. Ghizzo et al., Stimulated-Raman-scatter behavior in a relativistically hot plasma slab and an electromagnetic low-order pseudocavity, Phys. Rev. E 74 (2006), 046407
[26] J.J. Honrubia, J. Meyer-ter-Vehn, Fast ignition of fusion targets by laser-driven electrons, Plasma Phys. Control. Fusion 51 (2009), 014008 [27] C.N. Danson et al., Vulcan petawatt: Design, operation and interactions at 5x1020 W/cm2, Laser Part. Beams 23 (2005), 87–93 [28] O. Willi et al., Laser triggered micro-lens for focusing and energy selection of MeV protons, Laser Part. Beams 25 (2007), 71–77.
[18] A.A. Andreev et al., Short light pulse amplification and compression by stimulated Brillouin scattering in plasmas in the strong coupling regime, Phys. Plasmas 13 (2006), 053110 [19] H. Chen et al., Relativistic Positron Creation Using Ultraintense Short Pulse Lasers, Phys. Rev. Let. 102 (2009), 105001 [20] N. Naumova et al., Hole Boring in a DT Pellet and Fast-Ion Ignition with Ultraintense Laser Pulses, Phys. Rev. Lett. 102 (2009), 025002 [21] Y. Sentoku et al., Isochoric heating in heterogeneous solid targets with ultrashort laser pulses, Phys. Plasmas 14 (2007), 122701 [22] R.A. Snavely et al., Laser generated proton beam focusing and high temperature isochoric heating of solid matter, Phys. Plasmas 14 (2007), 092703
Plasma and high energy density physics
Chapter 5
ELI: Extreme Light Infrastructure www.eli-beams.eu
143
Chapter 6 Exotic physics and theory
Authors: RNDr. Karel Rohlena, CSc. / Bc. Jan Prokůůůpek
Introduction Ultra-high fields of high-power short-pulse lasers pose important possibilities for fundamental physics. Albert Einstein's equivalence of matter and energy, elegantly expressed as E=mc2, has been confirmed countless times, most dramatically whenever a nuclear weapon detonates. But the process also occurs naturally in stars shining because atoms in its core fuse, transforming a sliver of matter into light. However, any equation works in both directions, at least theoretically, consequently it should be possible to convert energy into matter. The recent progress in laser technology calls for a reassessment of intensity effects in Quantum Electrodynamics (QED) and the new prospects of measuring them. QED describes electromagnetic fields, such as those of light, and their interactions with matter. Physicists are eager to study it in order to convert photon energy directly into creation of electrons and positrons. There is a plethora of strong-field QED
processes, which may be roughly categorized into two classes: “loop” (strong-field vacuum polarization, spontaneous pair production) and “tree-level” (perturbative pair production, pair annihilation, Compton scattering) processes. One-loop processes are of order h and thus of a genuine quantum nature, while tree-level processes generically do have a classical limit. As a result, one can introduce two distinct parameters which characterize the different physics involved. The first parameter is the QED electrical field (see below), EC, which depends on the Planck’s constant and the speed of light, showing that it originates from a relativistic quantumfield theory. In an electric field of strength Ec an electron acquires an electromagnetic energy equal to its rest mass mc2 upon traversing a distance of a Compton wavelength:
Exotic physics and theory
Hence, Ec may be viewed as the critical field strength above which vacuum pair production becomes abundant. This is also borne out by Schwinger pair-creation probability given by the tunneling factor p~exp(-πEC/E), where E denotes the “ambient” electric field one succeeds in achieving. Currently, this is E ≅ 1014 V/m implies a huge exponential suppression.
Chapter 6 Current status As already mentioned, turning matter into light, heat, and other forms of energy is nothing new. In 1997 a team of physicists at the Stanford Linear Accelerator Center (SLAC) in the framework of the E-144 experiment have demonstrated the inverse process: "the first creation of matter out of light". They collided large crowds of photons together so violently that the interactions spawned electron-positron pairs [1]. In order to generate a field as close as possible to Ec, they used a 0.5 TW short-pulse glass laser (λ=527 nm, EL=500 mJ) into a beam measuring 6 micrometers across at its narrowest point. Moreover, in order to increase the energy of the photons, the team collided the pulses with SLAC's 30-micrometer-wide pulsed beam of high-energy electrons (46.6 GeV).
When laser photons collided head-on with the electrons, they got a huge energy boost, much like ping-pong balls hitting a speeding Mack truck, changing them from visible light to very high energy gamma rays. Because of the laser's intensity, these backscattered gamma photons sometimes encountered several incoming laser photons simultaneously; a collision with four of them concentrated enough energy in one place to produce electron-positron pairs, as depicted in Figure 1. The highly energetic electrons, undergoing nonlinear Compton scattering with n laser photons, produce backscattered photons of 27–30 GeV which then collide with the laser photons producing electron-positron pairs via the multiphoton Breit–Wheeler mechanism. The number of positrons measured in 21.962 laser pulses was 175±13 [2]. The experiment was a proof of principle for a technology, based on intense laser beams boosted to enormous energies, for exploring the
Figure 1: An electron beam intersects a laser pulse, boosting photons to gamma energies and triggering an interaction that spawns particles [2].
quantum electrodynamics theory. The result can be viewed as the first direct demonstration of the “vacuum sparking" phenomenon, in which the energy of a very strong electromagnetic field promotes some of the fleeting "virtual" particles that inhabit the vacuum, according to QED, to become pairs of real particles.
ELI: Extreme Light Infrastructure www.eli-beams.eu
147
Electron-positron pairs are often spawned in accelerator experiments that collide other particles at high energies, and photons produced in the collision are what actually generate the pairs. But at least one of the photons involved is virtual, i.e. produced only for a brief moment in the strong electric field near a charged particle. The SLAC experiment marked the first time that matter has been created entirely from ordinary photons. In 2000 at ATLAS laser facility of Max-PlanckInstitut, a team of scientists succeeded in converting low-energy photons into positrons via electron acceleration in a plasma channel, by using a femtosecond table-top laser system [3]. The average intensity of this source of positrons was estimated to be equivalent to 23.108 Bq and it exhibits a very favourable scaling for higher laser intensities. In the paper are described measurements of positrons produced by MeV electrons from a relativistically self-focused la-
Figure 2: Experimental setup on positron production using laser at MPQ [3].
ser channel in underdense plasma. The scheme employed is analogous to that used for the generation of positrons from high-Z moderators in linear electron accelerators. It consists of two steps: first, using laser pulses from the 790 nm, 220 mJ, 130 fs, 10 Hz ATLAS laser system, a beam of electrons was generated in a gas-jet target; second, the electrons were converted in
Exotic physics and theory
a 2-mm-thick Pb slab to positrons. They were able to produce 106 positrons per laser pulse with a mean energy of about 2 MeV. Although for radiation safety reasons the measurements were performed using single laser pulses, there are no technical constraints hindering the operation at the laser repetition rate, i.e., 10 Hz. Thus, in repetition mode configuration, this ex-
Chapter 6 periment represents a positron source with an activity of 107 Bq. Very interesting results have been also obtained using the Petawatt Laser at Lawrence Livermore National Laboratory [4]. This system uses one arm of the NOVA laser to amplify a frequency-chirped pulse to 1 MJ energy before temporal compression to 450 fs. The peak power is well in excess of 1 PW, and the beam can be focused to a 10 µm spot, with a typical intensity of 3x1020 W/cm2. At these intensities the electric field at the laser focus is higher than 1013 V/m. On the face of the solid targets used in these experiments a plasma was generated by a pre-pulse at 2 or 10 ns before the main pulse. The pre-plasma served as a medium to selffocus the laser, thereby increasing the ponderomotive energy, and to produce electrons of even higher energy by a variety of laser acceleration mechanisms. The interaction of the electrons with the solid target produces positron-electron
pairs. The yield of positron-electron pairs in this experiments was of order 104 of the electron yield in the energy range 5-10 MeV, as reported in Figure 3 showing the positron distribution for a thin, 125 µm Au target. The number of recorded positrons clearly exceeds the calculated yield expected for external pair conversion of Bremsstrahlung photons generated in the target. The additional mechanism of “trident production”, i.e., direct positron-electron pair production by electron-ion collisions, is expected to have a similar total yield as the Bremsstrahlung component for these experimental parameters. These data suggest that in future generations of ultra-high-intensity laser experiments, it might be possible to observe pair creation directly in the collisions of the quivering electrons with the ions of the target plasma, and perhaps even progress towards a terrestrial relativistic pair plasma, which is relevant for astrophysical models of gamma-ray bursts or accretion disks.
Figure 3: Spectra of electrons and positrons measured in 600 J/0.5 ps shot on 125 µm Au target. Histograms show PIC simulation of electrons, and corresponding prediction for positron creation by Bremsstrahlung in the Au target [4].
The above reported experiments have led to a recent interest of the subject especially as to whether modern laser technology can produce the strong electric field required for experimental verification. In the recent paper [5], the au-
ELI: Extreme Light Infrastructure www.eli-beams.eu
149
thors treated the problem of electron-positron production in a standing wave of oppositely directed laser beams of plane transverse linearly polarized electromagnetic waves of fixed frequency, using a two level multiphoton resonant approximation. This approach, if experimentally implemented, will result in much higher electron-positron production rate for the case of conventional femtosecond laser systems.
Exotic physics and theory
Chapter 6
Directions of implementation at ELI The technological breakthrough of laser chirped-pulse amplification has led to unprecedented laser powers and intensities, the current records being about 1 PW and 1022 W/cm2, respectively. Up to 3 orders of magnitude may be gained at the planned ELI facility. At a laser intensity of 1026 W/cm2, an electron will undergo an acceleration of 1027 g, comparable to the gravitational acceleration at the event horizon of a black hole. This high acceleration could be used to study Unruh radiation generation. At sufficiently high intensities, even vacuum can be broken down. The field necessary to achieve pair creation (â&#x20AC;&#x153;boil the vacuumâ&#x20AC;?) is the above mentioned Schwinger field, EC. Although such fields are beyond the horizon, other nonlinear quantum electrodynamics effects could be accessed at more modest fields. For petawatt, kilojoule-class lasers, a nontrivial electron-positron pair can be created. Nowadays positron emission from direct interaction of
petawatt laser pulses with solid targets has the main disadvantage of no operation at high repetition rate, limiting the yield of positron emission. However this limit will be easily overcome at the future ELI facility. In this point of view ELI will explore new regimes and try to map out the basic phenomena concerning the QED theory. Finally, it is noted that an alternate path to the Schwinger field could be an X-ray free-electron laser. In the future proposed XFELs field strengths of 1017 V/m are expected and fundamental physics with these fields constitutes an important research area. Various QED phenomena that can be study at ELI facility are reported and discussed below.
Electron-positron plasmas Electron-positron plasma is a plasmatic state of matter along with antimatter where heavy ions are replaced by the electron antimatter equivalent and where relativistic effects play an important role. The early pre-stellar period of the universe evolution is characterized by sufficiently high temperatures at which the creation of particles-photons, electron-positron pairs is a natural process. Other sources of quite dense electron-positron plasma are pulsars. Most theories state that plasmas near a black hole are so hot that electrons and positrons exist in equilibrium with each other. Such matter-antimatter is created due to the Hawking radiation near the event horizon of massive gravitational body. On Earth, there is no such extreme gravitation
ELI: Extreme Light Infrastructure www.eli-beams.eu
151
field, but according to the theory of quantum electrodynamics the photon can be split up to electron-positron pair in presence of an electric field with a threshold value of 1.3x1018 V/m, also known as the Schwinger limit (analytically defined above). The Schwinger field corresponds to a laser intensity of 5x1029 W/cm2. With weaker fields the sub-threshold production can be studied, where in the oscillating fields enhanced production of electron-positron pairs and their annihilation radiation are predicted. For a strongly focused laser pulse pair production is predicted already at 1027 W/cm2 [6], while for two counter-propagating pulses the threshold should occur at 1026 W/cm2 [7]. In this sub-threshold field strength region theoretical predictions differ by many orders of magnitude [8]. Experimental clarification of these conflicting predictions is required and from the above mentioned facts the investigation of propagation of strong e.m. waves (pulses) in relativistic hot electron-positron plasma becomes very topical.
These matter-antimatter plasmas can be created in laboratory using an ultra-high intensity lasers (PW class and more). Laser intensities exceeding 1018 W/cm2 couple most of their energy to superthermal electrons with temperature exceeding the rest energy of the electron (relativistic plasma). Positrons are created when the relativistic electrons interact with high-Z target ions. High magnetic fields generated by relativistic electrons help to confine them. PW class lasers with sufficient pulse length can generate positron densities (approximately 1022 cm-3 for solid gold targets) which exceed any laboratory source of positrons [9].
Figure 4: Double illumination of a target with two petawatt laser beams could lead to the production of the electronpositron pair plasma [9].
ELI laser facility might demonstrate the acceleration of electron beams in the 100 GeV energy range, which can result in an increase of the electric field strength up to 1018 V/cm in the electron beam frame. Using the imaginary time method this field predicts a production of up to 105 electron-positron pairs per shot in a small
volume of the time-space. The research programme at ELI can start with electron-positron generation from thin high-Z targets and then proceed to the lasers generating high electron beams which would collide and produce electron-positron pairs in vacuum without a target. This could be carried out at laser intensities of the order of 1023 W/cm2.
Exotic physics and theory
Chapter 6 Vacuum processes Vacuum four-wave mixing Direct observation of elastic photon-photon scattering among real photons would be an important benchmark test of laser-based Quantum Electrodynamic (QED) experiments. Deviations from the expected scattering rate would indicate new physics in the low-energy regime. Throughout the last decades, several suggestions on how to detect elastic photon-photon scattering, using laser assisted schemes, have been put forward. Crossing electromagnetic waves can interact and yield new modes of different frequencies. One of the most prominent modes in such a mechanism is given by the four-wave interaction mediated mode satisfying the resonance condition between the frequencies and wavevectors (i.e. photon energy and momentum conservation). It is therefore not a
surprise, given the evolution of laser powers and frequencies, that the search for photonphoton scattering using resonant four-wave interactions has caught the attention of researchers in this area. This approach has progressed furthest in the experimental attempts to detect elastic scattering among photons. Using four-wave mixing to stimulate photonphoton scattering has the advantage of not being limited by the low scattering cross section, in the optical range. The resonance conditions should be ω4=ω1+ω2-ω3 and k4=k1+k2-k3, between the vacuum generated photons and the laser pump sources, respectively (see Figure 5). When the scattering of a laser beam by another laser beam is stimulated by a third laser beam, the scattering probability becomes to be measurable [10]. Since a perfect vacuum in the interaction chamber is in practice impossible to achieve, there will be competing scattering processes present. The effect of these
Figure 5: Configuration of the incoming laser beams (represented by the wave vectors k1, k2 and k3) and the direction of the scattered wave (with wave vector k4).
processes will rather effectively be suppressed due to plasma cavitations caused by the strong laser pulses. Progress of low-energy QED experiments could also prove to be useful for dark matter searches. Four-wave mixing technique can be also used
ELI: Extreme Light Infrastructure www.eli-beams.eu
153
for an ultrashort pulse source in the UV and VUV spectral range pulse generation in hollow waveguides with unprecedented short-pulse durations. The generation of a 2.5 fs pulse at 160 nm with energy of about 1 nJ has been predicted by pumping all the energy into the fundamental mode, well above its phase-matching pressure [11]. The energy can be increased by a factor of 20 using higher-order transverse modes. By changing the input idler frequency a detuning of the VUV signal can be realized. Vacuum polarisation In quantum field theory, and specifically quantum electrodynamics, vacuum polarization describes a process in which a background electromagnetic field produces virtual electron-positron pairs changing the distribution of charges and currents which have generated the original electromagnetic field. It is also sometimes referred to as the self energy of the gauge
boson (photon). According to quantum field theory, the ground state of a model with interacting particles is not simply empty space. Rather, it contains short-lived virtual particle-antiparticle pairs which are created out of the vacuum and then annihilate each other. Some of these particle-antiparticle pairs are charged, e.g., virtual electron-positron pairs. Such charged pairs act as an electric dipole. In the presence of an electric field, e.g. the electromagnetic field around an electron, these particle-antiparticle pairs reposition themselves, thus partially counteracting the field (partial screening effect, i.e. dielectric effect). The field therefore will be weaker than it would be expected if the vacuum were completely empty. This reorientation of the short-lived particle-antiparticle pairs is known as vacuum polarization. Electromagnetic waves interact in vacuum owing to the presence of virtual electron-positron pairs. In a typical experimental setup one strong
Exotic physics and theory
electromagnetic field polarizes the vacuum and another one, usually a weak field, probes the induced vacuum polarization. The polarized vacuum behaves like a dielectric medium and the light spot generated by a strong focused laser field acts as a microscopic piece of matter that diffuses the probe field. This diffusion or diffraction alters the polarization state of the probe and if the probe initially is linearly polarized, then after the interaction it results elliptically polarized, with the main axis of the ellipse rotated with respect to the initial polarization direction. The observation of vacuum polarization effects is feasible in laser fields significantly weaker than the Schwinger field, thus it can be easily investigated in the ELI intensity regime. Moreover the detection of vacuum birefringence is already possible by the available laser technique with 1022 W/cm2 intensities if X-rays are employed as a probe.
Chapter 6 Vacuum birefringence High quality vacuum according with a strong magnetic field (more than 5 T) serves as a birefringent medium. Due to the vacuum birefringence linear polarization of the laser light turns into elliptical polarization. The basic experimental setup should consist of a polarizer to reach as highest polarization as possible of the incoming laser beam, a strong dipole magnet (or a superconductive magnet) with the magnetic field B0 exceeding 5 T, and a sensitive ellipsometer as depicted in Figure 6. It is necessary to have large interaction region in order to make stronger elliptical polarization for its better measuring. For the plane wave in the magnetic field the electric intensity and magnetic induction will be E(r,t)=Ewei(k⋅r-ωt) and B(r,t)=Bwei(k⋅r-ωt) + B0, respectively. It should be also possible to make vacuum birefringence experiment without permanent mag-
Figure 6: The experimental scheme for the vacuum birefringence observation [12].
Figure 7: The vacuum birefringence turns the linear polarization of the laser beam into the elliptical one in presence of the strong magnetic field [12].
netic field because the polarized vacuum acts as a medium with preferred directions dictated by the external fields which we assume to be generated by a high-power laser of frequency ω0. Accordingly, there are two different refractive indices for electromagnetic probe beams of different polarization states. Distorting the vacuum with lasers has been suggested long time ago [13] but was not considered experimentally for lack of sufficient laser power. ELI laser intensity regime and recent progress in X-ray detection will open the way towards novel experimental capabilities. It is therefore due time to specifically address the feasibility of a strong-
ELI: Extreme Light Infrastructure www.eli-beams.eu
155
field laser experiment to measure vacuum birefringence. Such experiments are also necessary in order to test whether strong electromagnetic fields provide windows into new physics. In the experimental setup proposed in [13] a high-intensity (petawatt class) laser pulse is focused by an f/2.5 off-axis parabolic mirror. A hole is drilled into the parabolic mirror in alignment with the z-axis in such a way that an X-ray pulse can propagate along the z-axis through the focal region of the high-intensity laser pulse. Using a polarizer-analyzer pair the ellipticity of the X-ray pulse may be detected. In Figure 8: a linearly polarized laser-generated ultra-short X-ray pulse is aligned collinearly with the focused optical laser pulse; after passing through the focus the laser induced vacuum birefringence should lead to a small ellipticity of the X-ray pulse which can be detected by a high contrast X-ray polarimeter [14]; extension of the setup for the generation of counter propagating
Figure 8: Proposed experimental setup for the demonstration of vacuum birefringence [14].
laser pulses and a high-intensity standing wave which may be used for pair creation (in gray). The whole setup should be located in an ultrahigh vacuum chamber and entirely computer controlled. Accurate control over spatial and temporal overlap was convincingly demonstrated carrying out an autocorrelation of the laser pulses at full in-
Exotic physics and theory
tensity [15] and generating Thomson backscattered X-rays from laser-accelerated electrons [16]. This counter propagating scheme, a tabletop “photon collider”, may also be employed for pair creation from the vacuum. For the X-ray probe pulse it was chosen an X-ray source of photon energy ω ≅ 1 keV, since the birefringence signal is proportional to ω2. X-ray polarimetry is currently sensitive to el-
Chapter 6 lipticities of just about 10−4. Thus it should require intensities in the upper range of ELI specifications (1026 W/cm2). However, the situation changes if polarized photon beams of MeV energies could be produced. Then the signal should increase significantly. Photon sources of this type can be obtained via Thomson or Compton backscattering off an electron beam generated, i.e., via laser wake field acceleration. Backscattering a probe laser (with a0 < 1) off 3 GeV electrons, for instance, would yield 50 MeV γ -rays.
Unruh radiation Unruh radiation is a quantum correction to the classical radiation rate that grows large only in situations where quantum fluctuations in the radiation rate become very significant. An observer or detector undergoing a uniform acceleration a experiences the vacuum as a thermal bath with the Unruh temperature T=ha/2πkBc, c is the speed of light, h is the reduced Planck's constant and kB is the Boltzmann's constant. The Unruh effect is close to the Hawking radiation at the horizon of a black hole where the observer staying near the horizon sees the thermal bath of the virtual particles whereas observer falling to the black hole does not see any. The linearly accelerated electron will execute a Larmor radiation and, as a response, the electron reacts to the vacuum fluctuations with a quivering motion in its proper frame and in turn triggers an additional radiation. At the classical level, the same
linear acceleration induces a Larmor radiation. The Unruh radiation is induced by the reaction to the Larmor radiation and is a minute perturbation of it. The space where we are interested in detecting the Unruh radiation is along the acceleration direction where the Larmor radiation is the weakest. Therefore the two radiations can be treated as independent processes without interference. The Unruh radiation can be discriminated from classical Larmor radiation via the different angular and spectral distributions and the distinct photon statistics. In the quantum case, the photons are always emitted in pairs with maximally entangled polarizations whereas the classical radiation corresponds to a coherent state with Poisson statistics. The Larmor radiation has a blind spot in forward (and backward) direction where the Unruh radiation is maximal leading to a cone of “quantum domination” with a small angle. The Larmor and the Unruh radiation are
ELI: Extreme Light Infrastructure www.eli-beams.eu
157
comparable when acceleration exceeds 1030 g.
Figure 9: Two-photon amplitudes for Unruh radiation on the left side and Larmor radiation on the right side [17].
The signatures of the Unruh effect might be detectable in future laser facilities generating electric fields not too far from the Schwinger limit (corresponding to an intensity of order 1029 W/cm2). However, when a 50 GeV electron collides head-on with a laser beam of field strength E then the field strength in the rest frame of the electron is Eâ&#x20AC;&#x2122;=2ÎłE. Hence if the laboratory field strength of the focused laser beam is E=6.5x1010 V/cm, the field appears to be critical from the point of view of the electron. The corresponding laser intensity is I=1.1x1019 W/cm2 [18]. An experiment was proposed for the indirect signature in the Unruh effect [19]. The uniformly
accelerated detector acts as if it is immersed in a thermal bath, there is a finite probability that it absorbs a virtual particle from this bath and passes to an excited state. This process corresponds to the emission of a real particle for a non-accelerating observer [20]. The opposite process, when the detector re-emits the virtual particle back into the bath in the accelerated frame and goes back to its ground state, also corresponds to the emission of a real particle for the non-accelerating observer. In the limit case in which the time between absorption and reemission becomes arbitrarily small, the detector transforms into a scatterer which scatters virtual particles from one mode into another mode of the thermal bath in the accelerated frame. This process corresponds to the emission of two real particles by the accelerated scatterer for the non-accelerating observer. This effect is analogous to moving-mirror radiation [21] and can be interpreted as a signature of the Unruh effect.
Exotic physics and theory
QED cascades: Inverse Compton Scattering Electromagnetic cascade initiated by very high energy electrons and photons is very important in a number of astrophysical situations. In the early Universe the high energy photons might be produced by some exotic sources such as primordial black holes and unstable heavy particles. In so-called Inverse Compton Scattering (ICS) a laser pulse collides with an electron bunch (see Figure 10) and forces the electrons to wiggle. This wiggle motion makes the electrons emit highly collimated synchrotron radiation in the forward direction, similarly to an electromagnetic undulator in a conventional synchrotron. The Doppler energy upshift allows one to reach
Chapter 6 high photon energies, e.g. 100 MeV γ -rays with a 10-GeV electron beam. While in linear Compton scattering only the fundamental wavelength is generated, non-linear ICS allows one to produce higher harmonics where the critical harmonic number scales with the third power of the dimensionless laser amplitude a0: nc~a03. At high intensities one expects to see a substantial red shift of the usual kinematic Compton edge of the photon spectrum caused by the large, intensity-dependent effective mass of the electrons within the laser beam. In addition, it was demonstrated that the motion of the center-of-mass frame for a given harmonic becomes intensity dependent. Tuning the intensity then effectively amounts to changing the frame of reference, going continuously from inverse to ordinary Compton scattering with the center-of-mass kinematics defining the transition point between the two [22].
Figure 10: Inverse Compton scattering scheme.
As above reported an important parameter comes into play, the “dimensionless laser amplitude”, given as the ratio of the electromagnetic energy gained by an electron across a laser wavelength to its rest mass:
This is a purely classical ratio which exceeds unity once the electron’s quiver motion in the laser beam has become relativistic. It is sufficient to adopt a useful rule-of-thumb formula expressing a0 in terms of laser power:
So that a0 is of order 102 for a laser in the petawatt class. If the electrical field amplitude associated with the laser beam stays far below the Schwinger pair-creation threshold, the natural process that comes to mind is the strong-field Compton scattering where a high-intensity beam of laser photons γL collides with an electron beam emitting a photon γ. In this case one has to sum over all n-photon processes of the type:
ELI: Extreme Light Infrastructure www.eli-beams.eu
159
Then a pair cascades can be initiated by the Compton scattered photons as depicted in Figure 11. The γ photon, in fact, can absorb a further energy from the laser beam and generate an electron-positron pair. Finally, each of them can in turn absorb energy and generate a γ photon, giving rise to a cascade process.
tron beam of sufficiently high energy γ ≥ 102 produced in a conventional accelerator or by a suitable laser plasma acceleration mechanism. The cross sections or photon emission rates are enhanced with increasing dimensionless laser amplitude. The possible effects are mostly of classical nature, being fundamentally due to the mass shift:
caused by the relativistic quiver motion of an electron in a laser field. Ranked in order of their relevance the main intensity effects are: Figure 11: Pair cascade induced by inverse Compton scattering.
The physical scenario assumed is the collision of a high-intensity laser beam with an elec-
1. red shift of the kinematic Compton edge for the fundamental harmonic
Exotic physics and theory
2. appearance of higher harmonic peaks n > 1 in the photon spectra 3. possible transition from inverse Compton scattering (ωI > ω) to Compton scattering (ωI < ω) upon tuning a0. The redshift may be explained in terms of the larger effective electron mass, m* > m, the generation of which costs energy that is missing when it comes for “boosting” the photons to higher frequencies. To avoid significant energy losses, reducing the photon frequency, the amplitude a0 should not exceed a limit value. However, the generation of higher harmonics might partially overcome this problem. The transition from inverse to ordinary Compton scattering, once a0 increases beyond 2γ illustrates the energy “loss” just mentioned. When a0 = 2 the laboratory frame can be interpreted as an intensity-dependent center-of-mass frame for which ωIn = nωn, at least for low harmonics. Thus,
Chapter 6 there is no longer energy gain of the emitted photons: the laser beam becomes so “stiff” that, in this frame, electrons begin to bounce back from it gaining energy rather than vice versa. Nonlinear Compton scattering provides a unique testing ground for strong-field QED. ELI pulse power and duration will allow to perform the experiments required for measuring the effects listed above. In fact, ICS radiation produced with the 10-fs ELI laser system will have a duration comparable to the electron bunch duration, which in turn is expected to be on the same time scale as the laser pulse one. The laser amplitude provided by ELI can be as large as 280 (in the case of 150 PW) and can provide intense beams of ultrafast radiation in the MeV energy region.
Quark-gluon plasmas
case of a vanishing baryon chemical potential.
A quark-gluon plasma (QGP) is a state of matter in which quarks and gluons are no longer confined to volumes of hadronic sizes. The fundamental theory for the strong interaction, Quantum Chromodynamics, estimates for the critical temperature at which the transition be-
Figure 12: At a certain density (temperature) hadrons lose their identity and a system of free quarks and gluons is created (quark gluon plasma) [23]
tween ordinary nuclear matter and quark gluon plasma appears. Calculations indicate that the critical temperature should be 175 MeV in the
The study of the QGP is also a testing ground for finite temperature field theory, a branch of theoretical physics which seeks to understand particle physics under conditions of high temperature. Such studies are important to understand the early evolution of our Universe: if the quark-gluon plasma is possible, the conditions of temperature and energy density in the very early stages (the first 25 μs) of the Big Bang would have certainly resulted in such a state existing before the temperature fell as the expansion proceeded and the quark-gluon “soup” froze out into hadrons. This should be crucial to the physics goals of a new generation of universe observations. In laboratory, in order to produce a quark-gluon plasma, a high density state has to be created by means of symmetric compression by ultra-
ELI: Extreme Light Infrastructure www.eli-beams.eu
161
high intensity lasers or nuclear collisions based on laser accelerating techniques. The evidence for bulk properties consistent with quark gluon plasma formation are accompanied by large energy density, entropy growth, plateau behaviour of the thermodynamic variables, unusual expansion and lifetime properties of the system, presence of thermodynamic equilibration, fluctuations of particle number or charge balance. The evidence for modifications of specific properties of particles thought to arise from their interactions with quark gluon plasma can be the modification of widths and masses of resonances, modification of particle production probabilities due to colour screening and modification of parton properties due to interaction with other partons in a dense medium.
friction force and Quantum Electrodynamics effects to nonlinear vacuum polarization and electron positron pair creation [24]. In this limit novel mechanisms of the ion acceleration come into play. Since the energy of the resulting ion bunch can be over 100 GeV per nucleon, this ion acceleration regime is suitable for quarkgluon plasma studies [25] and in connection with an application to the investigation of neutrino oscillations [26].
In the extremely-high intensity limit the laser pulse-plasma interaction acquires new properties, ranging from the effects of the radiation
Exotic physics and theory
Chapter 6
Potential for applications, business and technology transfer First of all, it should be noted that nowadays the exotic physics is far away from any imaginable application and a huge effort in this fundamental research field has to be done. The ultra-high, ultra-intense laser pulses themselves will be the main tool for experimental studies relevant to QED. Ultra-intense lasers have a unique potential as an effective source of electron-positron plasmas due to laser matter interaction and/ or vacuum nonlinear effects. The positrons can be separated from the electron-positron pair plasmas and might be further utilised e.g. for positron emission tomography. The nature of the vacuum under extreme conditions can be also one of the most important subjects to be studied. At the certain intensity threshold the vacuum acts in a nonlinear regime as a dielectric or birefringence material due to the effects of the virtual particles. The interaction of high intensity laser pulses with underdense and overdense plasmas pres-
ents a manifestation of one of the most basic nonlinear processes in physics: the high-order optical harmonic generation. High order optical harmonics have been observed in laser plasma interactions with radiation intensities ranging from moderate up to relativistic intensities. Nonlinear orders as high as 300 have been reported recently [24]. In addition to its fundamental interest in the theory of nonlinear waves, this radiation presents unique properties of coherence and short pulse duration that makes high order harmonics a useful XUV source of short coherent radiation such as EUV for lithography, holography, etc. The photon-photon collider is considered as the best tool for addressing and discovering new physics: Higgs physics, extra dimensions, supersymmetry, and top quark. In a photon collision any charged particles can be produced [24]. The cross sections for pairs are significantly higher than in the electron-positron col-
lisions. The γ-γ collider relies on the scattering of photons from a high-intensity laser by an ultrarelativistic electron beam. After scattering, photons have energy close to the electron energy [24] and the process efficiency is excellent with one electron scattering one γ photon. The photon beams after focusing correspond approximately to the electron beam size. These are additional meeting points of laser and high energy charged particles. In some of these applications it is possible to explore nonlinear QED and also to produce large amounts of highenergy γ-photons through the inverse Compton scattering process useful for high-energy and nuclear physics [27]. Tajima proposed the process for realizing a possible nuclear transmutation in combination with efficient lasers such as the free-electron laser [28]. A gamma ray initiating an electron-positron pair production has a necessary condition of presence of electric field which has to reach the
ELI: Extreme Light Infrastructure www.eli-beams.eu
163
Schwinger limit. With the presence of the trident process (a process in which a nucleus plays the role of an additional photon in the interaction among electrons and photons), the electric field condition is relaxed [29-31] at an intensity level directly accessible (1022 W/cm2). γ-rays may be also used in photonuclear physics. For example, such a photon interacting with nuclear matter may lead to a new field of investigation which relies on the coupling between weak and strong interactions [32]. Moreover, polarized γ-photons may be used for creation a large flux of polarized positrons, which may have an important role in future collider-beam sources to enhance the signal-to-noise ratio of desired events [33].
energetic electrons as well as acceleration of other particles can be provided for the creation of substantial numbers of isotopes; this technique has already been demonstrated [35-38]. Transmutation in the minor actinides may be carried out via a new fission decay mechanism in which the vibrational levels created by the hyperdeformation of nuclei (as in the formation of isomers) are populated [39]. Such a process may be initiated by gamma rays generated by inverse Compton scattering of the laser pulse with high energy electron beam. The produced gamma rays can induce various (γ,n) nuclear processes, as opposed to the more common (n, γ) processes which can be seen in nature.
In experiments conducted on solid targets highenergy electrons have been generated, leading to the creation of high-energy gamma photons in the solid by the bremsstrahlung process. These gamma rays might be possible to induce a nuclear transmutation [34]. The production of
A further advantage of high intensity laser field is the application in gravitational field effects that could be measurable in laboratory according to the main postulate of general relativity, the Einstein principle of equivalence. The effect of a homogenous gravitational field is equiva-
Exotic physics and theory
lent to that of a uniform accelerated reference frame. In the past various experiments have been performed in order to test the equivalence principle in its weak limit in the laboratory using neutron beams with a spinning mirror [40]. By using ultra-high ultra-intense lasers it should be possible to test the equivalence principle in its strong limit. The theory of quantum gravity has been recently advanced [41-46], stating that gravitational effects, having extra dimensions, could be observed over macrodistances. Ultra-intense lasers could provide a new way to test extradimensional effects. Probably for a sufficiently intense laser field the distance of the electron to its horizon might become on the order (or even smaller) of the proper distance over which the effects of extra dimensions could be observed. The entire universe began with an explosion of energy, the Big Bang, and accelerator physicists have witnessed the conversion of energy into
Chapter 6 matter (virtual photons) by smashing atoms together. But such virtual photons are not under the direct control of physicists; these photons arise as part of a complex chain of events starting with a collision of two particles of matter. The conversion of light into matter could give particle physicists a new source of positrons that are exceptionally uniform in energy and momentum. Moreover, an electron-positron pair beam can be accelerated and focused by a properly shaped ponderomotive potential. The high-field experiments at ELI facility could shed light on phenomena at the surface of neutron stars, where magnetic fields are very strong, as well as in other exotic astrophysical settings.
ELI: Extreme Light Infrastructure www.eli-beams.eu
165
References [1]
D.L. Burke, et al., Phys. Rev. Lett. 79 (1997) 1626
[2] Volume 277, Number 5330 Issue of 29 August 1997, p. 1202, The American Association for the Advancement of Science. [3]
Gahn et al., Appl. Phys. Lett. 77 (2000) 17
[4]
T.E. Cowan et al., Nucl. Instr. Meth. Phys. Res. A 455 (200) 130139
[5]
H.K. Avetissian et al., Phys. Rev. E 66 (2002) 016502
[6] Popov, A. K. et al. Large Enhancement of Fully Resonant SumFrequency Generation Through Quantum Control Via Continuum States. Physical Review 69 (2004) [7]
Narozhny, N. B. Et al. JETP Letters, Vol. 80 (2004) 382–385
[8]
Popov, A. K. et al. Physical Review 69 (2004)
[9]
The Science and Applications of Ultrafast, Ultraintense Lasers. A report on the SAUUL workshop held (2002)
[17] R. Schützhold, G. Schaller and D. Habs, Phys.Rev.Lett. 97 (2006) [18] McDonald, K. T. Positron Production by laser light, http://www. hep.princeton.edu/~mcdonald/e144/ssitalk.pdf [19] R. Schützhold, Gernot Schaller and Dietrich Habs, Phys.Rev.Lett. 97 (2006)
[32] Fujiwara, M., et al., 2005, in Exotic Nuclear Systems, edited by Z. Gasci, Zs. Dombradi, and A. Krasznahorkay, AIP Conf. Proc. No. 802 (AIP, Melville, NY), p. 246 [33] Omori, T. et al., Nucl. Instrum. Methods Phys. Res. A 500 (2003) 232-252
[20] W. G. Unruh and R. M. Wald, Phys. Rev. D 29, 1047 (1984)
[34] Cowan, T. E., et al., 2000a, in High Field Science, edited by T. Tajima, K. Mima, and H. Baldis (Kluwer, New York), p. 145
[21] N. D. Birrell and P. C. W. Davies, Quantum Fields in Curved Space, (Cambridge University Press, Cambridge, England 1982)
[35] Yamagiwa, M., Koga, J., J. Phys. D 32, 2526 (1999)
[22] Harvey et al., Phys. Rev. Lett. 79 (2009) 063407 [23] V. Wagner – lectures on Elements of Nuclear Physics [24] Mourou, G. A. et al., Reviews on modern physics 78 (2006) 309371 [25] Esirkepov, T. et al., Regime. Phys.Rev.Lett. 92 (2004)
[36] Ledingham, K. W. D. et al., Science 300, 1107 (2003) [37] Leemans, W. P. et al., Phys. Plasmas 8, 2510 (2001) [38] [38] Nemoto, K. et al., Appl. Phys. Lett. 78, 595 (2001) [39] Shizuma, T., T. Hayakawa, and K. Nishio, in Proceedings of the Workshop on Applications of IRFEL and Nuclear Isomers 2002, edited by T. Yamauchi. JAERI, Tokai (2002)
[26] Bulanov, S. V. et al. Neutrino Oscillation Studies With Laser-Driven Beam Dump Facilities. INFN Laboratori Nazionali di Frascati (2004)
[40] Bonse, U., Wroblewski, T., Phys. Rev. Lett. 51, (1983) 1401 1404
[11] I. V. Babushkin et al. Optic letters 2 (2008) 938-940
[27] Fujiwara, M., et al., 2005, in Exotic Nuclear Systems, edited by Z. Gasci, Zs. Dombradi, and A. Krasznahorkay, AIP Conf. Proc. No. 802 (AIP, Melville, NY), p. 246 [29]
[41] Akama, K., 1983, in Gauge Theory and Gravitation: Proceedings of the International Symposium, Nara, Japan, Lecture Notes in Physics No. 176, edited by K. Kikkawa, N. Nakanishi, and H. Nariai (Springer-Verlag, Berlin), p. 267
[12] S. Y. Buhmann. On the search for nothing: Vacuum birefringence and Casimir-Polder forces, Institutsseminar TPI 2005 FSU Jena
[28] Tajima, T., Proceedings of the Japan-Hungary Seminar. IIAS, Kyoto (2002)
[42] Roessl, E., Shaposhnikov, M. Localizing, Phys. Lett. 125B, (1983) 136
13] E. Brezin and C. Itzykson, Phys. Rev. D3, 618 (1970)
[29] Shrearer, J. W. et al., Phys. Rev. A 8, (1973) 1582 – 1588
[14] M. Hart et al., Rev. Sci. Instrum. 62, 2540 (1991)
[30] Mima, K., et al. Power-Law Energy Spectrum and Orbital Stochasticity. AIP, New York (1991) p. 27
[43] Amelino-Camelia, G., et al. Planck-scale deformation of Lorentz symmetry as a solution to the ultrahigh energy cosmic ray and the TeV-photon paradoxes. Nature (London) 393, (1998) 763
[31] Liang, E. P., et al., Phys. Rev. Lett. 81 (1998) 4887 – 4890
[44] Arkani-Hamed, N., et al., Phys. Rev. D 59, 086004 (1999)
[10] Moulin, F., Bernard, D. Optics Communications 164 (1999) 137144
[15] B. Liesfeld et al., Appl. Phys. Lett., 86, 161107 (2005) [16] H. Schwoerer et al., Phys. Rev. Lett. (2005)
Exotic physics and theory
Chapter 6 [45] Giddings, S. B., Thomas, Phys. Rev. D 65, 056010 (2002) [46] Rubakov, V. A., Usp. Fiz. Nauk 173 (2003) 219 [43]
ELI: Extreme Light Infrastructure www.eli-beams.eu
167
Facts and Figures
1. Members of Consortium ELI-CZ
Existing and projected facilities as of September 2009
2. Strategic location of the ELI facility in Europe
Facts and Figures
Facts and Figures 3. Overview of the high-power laser facilities worldwide
Worldwide High Power Laser facility overview. Existing and projected facilities as of September 2009
ELI: Extreme Light Infrastructure www.eli-beams.eu
171
4. Overview of the high-power laser facilities in Europe
Existing and projected facilities as of September 2009
Facts and Figures
Facts and Figures 5. Positioning of high power laser facilities with respect to pulse duration and peak power
ELI: Extreme Light Infrastructure www.eli-beams.eu
173
6. Positioning of high power laser facilities with respect to pulse energy and peak power
Facts and Figures
Facts and Figures 7. World Optics and Photonics Clusters
World Optics and Photonics Clusters as of September 2009
ELI: Extreme Light Infrastructure www.eli-beams.eu
175
8. European Optics and Photonics Clusters
European Optics and Photonics Clusters as of September 2009
Facts and Figures
List of Abbreviations PFS OPCPA DPSSL PLC PCF KDP DKDP CW YAG CPA ASE OPA YCOB LBO OPO PPF I/O LAN UPS HW tools SW tools MLD XFEL FEL UV IR VIS DNA GRIP PMMA HSQ CNTs LLNL FWHM
Petawatt Field Synthesizer Optical Parametric Chirped Pulse Amplification Diode Pumped Solid-State Lasers Programmable Logic Controller Photonic Crystal Fiber Monopotassium phosphate Deuterated potassium dihydrogen phosphate Continuous Wave Yttrium Aluminium Garnet Chirped Pulse Amplification Amplified Spontaneous Emission Optical Parametric Amplification Yttrium Calcium Oxyborate Lithium triborate Optical Parametric Oscillator Pulse-to-Pulse Fluctuation Input/Output Local Area Network Uninterruptible Power Supply Hardware tools Software tools Multi-Layer Dielectric X-ray Free Electron Laser Free Electron Laser Ultraviolet Infrared Visible Deoxyribonucleic Acid Grazing Incidence Pumping Poly(methyl methacrylate) Hydrogen Silsesquioxane Carbon NanoTubes Lawrence Livermore National Laboratory Full Width at Half Maximum
EUV RIXS PTFE FEP NIR R&D LCLS EPR VUV XUV ICF CPA HEDP KEEN waves KEIN waves SRS SBS PIC simulation RPA FCT ICT OTR TNSA QED WDM LANL LANSCE GANIL SNS SLAC ICS QGP
Extreme Ultra-Violet Resonant Inelastic X-ray Scattering Polytetrafluoroethylene (teflon) Fluorinated Ethylene Propylene Near Infrared Research and Development Linac Coherent Light Source Electron Paramagnetic Resonance Vacuum Ultraviolet Extreme Ultra-Violet Inertial Confinement Fusion Chirped Pulse Amplification High Energy Density Physics Kinetic Electrostatic Electron Nonlinear waves Kinetic Electrostatic Ion Nonlinear waves Stimulated Raman Scattering Stimulated Brillouin scattering Particle-in-Cell simulation Radiation Pressure Acceleration Fast Current Transformer Integrating Current Transformer Optical Transition Radiation Target Normal Sheath Acceleration Quantum Electrodynamics Warme Dense Matter Los Alamos National Laboratory Los Alamos Neutron Science Center Grand Accélérateur National d’Ions Lourds Spallation Neutron Source Stanford Linear Accelerator Center Inverse Compton Scattering Quark-Gluon Plasma
Institute of Physics of the Academy of Sciences CR, v.v.i. Na Slovance 2 Prague 8 - 182 21 CZECH REPUBLIC www.eli-beams.eu eli-cz@fzu.cz +420 266 052 207