Bernhard Braunecker, SPS; René Dändliker, SATW; Thomas Feurer, University of Bern
This year we celebrate the 50th anniversary of the invention of the laser with numerous symposia and commemorative events around the world. It was in the August edition 1960 of Nature [Nature 187, 493 (1960)] where a short article appeared with the title 'Stimulated Optical Radiation in Ruby'. The author, Theodore H. Maiman, wrote '… to demonstrate the above effect a ruby crystal of 1 cm dimensions coated on two parallel faces with silver was irradiated by a high power flash lamp …' and a few lines hereafter '… These results can be explained on the basis that negative temperatures were produced and regenerative amplification ensued …'. This publication marks the starting point of a remarkable development spanning the past 50 years and still ongoing. Besides being an indispensable tool in science, the laser has also become an integral part of our daily lives. The 1960 publication by Maiman was followed by a variety of proposals outlining what one could possible do with such a device.
Only two years later, in 1962, Koichi Shimoda published a paper in Applied Optics [Applied Optics 1, 33 (1962)] were he argues that among the many applications of the extremely high brightness of such a light source could be a high energy electron accelerator. It is instructive to read his conclusions were he states '… It has been shown above that an acceleration of 109 eV/m could be obtained with a maser output of 10 kW/cm2 …'. In the following we briefly discuss what has become of this originally rather exotic laser application; especially in view of the tremendous evolution in focused laser intensities which have soared to record highs of more than 1020 W/cm2. The past 50 years have witnessed a huge amount of publications suggesting different ways and methods to transfer the laser’s electromagnetic field energy to charged particles. One may divide them in two groups, namely those operating in vacuum and those which use plasma as an intermediate energy storage medium. Plasma based schemes were triggered by T. Tajima and J. M. Dawson when they proposed, in 1979, that a pulsed laser can create a wake of plasma oscillations and electrons trapped in such wakes can be accelerated to high energies [PRL 43, 267 (1979)].
The biggest obstacle in vacuum based acceleration schemes is a fundamental theorem which states that the net acceleration of a relativistic charged particle in a plane, transversely polarized wave must be zero when integrated from minus to plus infinity. To first order this is also true for electromagnetic fields composed of plane waves, for example Gauss beams, because (to first order) the net acceleration from the sum equals the sum of accelerations resulting from the individual waves. Therefore, many proposals suggest ways to circumvent this problem by introducing special boundary conditions, applying additional electric or magnetic fields et cetera. One may argue that plasma based schemes have been more successful because these systems offer a larger flexibility. The energy of the electromagnetic field is first transferred to the plasma and only then to the charged particles. This two step process relaxes many constraints which vacuum based schemes face. Thus, we henceforth concentrate our attention on plasma-based schemes.
In plasma based schemes short and intense laser pulses excite charge density oscillations in plasmas, so-called plasma waves. Their wavelength is on the order of tens of micrometers and their amplitude scales roughly with the intensity of the laser. When an ultrashort laser pulse propagates through a plasma the corresponding light pressure, or the ponderomotive force, pushes electrons to the side much alike a ship in the ocean pushing aside the water in front of it. Ions, however, remain because of their much larger mass and a region depleted of electrons follows behind the laser pulse. These regions or pockets of positive ions are sometimes called bubbles. Because of the remaining positive charge a huge field gradient develops and electrons which are injected into this region (in the wake) can be accelerated. Electrons surf on those plasma waves and the longer they remain in phase with the plasma wave, the larger will be their final kinetic energy. For typical experimental parameters the accelerating electric field can be as high as several tens of Gigavolts per meter. Such typical experiments include an intense laser tightly focused into a diluted gas; tight focusing, however, goes hand in hand with a small depth of focus which limits the length of the acceleration volume. Moreover, the laser pulse propagating through the gaseous medium constantly transfers energy to the plasma and, thus, is attenuated along its path of propagation. Thus, efficient laser-based acceleration is feasible only if, first, dephasing, second, energy depletion, and, lastly, defocusing can be avoided or compensated.
Dephasing occurs whenever energy is transferred between two co-propagating systems which have different velocities. For example, in nonlinear optics, where two co-propagating light fields exchange energy, dephasing will limit the conversion efficiency. Here, electrons may have a different velocity than the plasma wakes which can eventually stop or even reverse the energy transfer to the electrons. From many other dephasing problems we know that phase matching can be achieved by limiting the interaction between the two systems to less than the dephasing length, then rearranging the phases before sending both systems in a further interaction volume. That is to say, dephasing can be solved by dividing the whole acceleration structure into several stages where each stage is as long as the electrons remain in phase with the plasma wake. It turns out that this measure can also solve the problem of energy depletion as for every acceleration stage a separate laser may be used.
The problem of defocusing can also be tackled with ideas from classical optics; we know that transverse confinement of light over long distances can be achieved if we employ suitable wave-guiding structures. Over the past years two promising ways to ‘inscribe’ a waveguide into the plasma have emerged. While the first is based on separately generated plasma waveguides, for example through an additional electric discharge, the second relies on a relativistic effect related to the laser pulse itself. Electrons experience a relativistic mass increase in regions of high intensity and they are expelled from this region by the ponderomotive force. Both effects lead to a focusing of light which becomes stronger with decreasing beam diameter and increasing intensity. This relativistic self-focusing in proper balance with diffraction and ionization-induced refraction can lead to a wave-guiding effect and produce light channels over centimeter length scales.
Today, the highest electron energies reached through laser-based acceleration schemes are on the order of several GeV, accelerated in plasmas which are only a few centimeters long, thus, reaching electric field strength on the order of tens of GeV per meter. These electron beams contain approximately 109 electrons, are directional, and have an energy spread of only a few percent. Thus, the operational principle of laser-based electron acceleration has been demonstrated but before such laser based accelerators can replace existing technologies there is still a long way to go. Following similar strategies, also laser-based acceleration of protons and carbon ions was shown with ion energies reaching several tens of MeV. One of the major issues of laser-based acceleration schemes today is the low average power of the driver laser systems and, thus, the resulting low brightness of the charged particle beams. Laser systems suitable for particle acceleration schemes can deliver at most 100 W of average power (with a wall plug efficiency of approximately 1 per mille). That is, given the current laser technology laser-driven particle accelerators may turn out to be useful in niches where brightness is not the decisive factor. One such field could be proton therapy of cancer and a number of groups worldwide explore those possibilities.
Symposium: Laser-Driven Particle Acceleration
In spring 2009 the Swiss Academy of Sciences SCNAT, the Swiss Academy of Engineering Sciences SATW, and the Swiss Physical Society SPS organized on behalf of the Swiss Academies of Arts and Sciences in close cooperation with the Swiss State Secretariat for Education and Research SER a workshop in Engelberg. The intention was to obtain an overview on the status quo of laser-driven particle acceleration and applications thereof. International experts in the field have reported on the state of the art of high power lasers, current concepts of accelerating charged particles by laser light, and applications such as the generation of coherent X-ray radiation or the use of laser-accelerated protons in medical sciences.
All presentations have shown that the scientific progress over the past decades has been tremendous, however, many challenges remain to be solved, especially in view of potential commercial applications. As Prof. T. W. Hänsch (Max-PIanck-Institute for Quantum Optics / Garching) stated at the symposium "… there is obviously a long way from physical experiments to technological maturity, but it is worth to go considering the industrial importance...". His statement complies with the strong technology efforts for a new generation of medical instruments, as was reported by Prof. W. Knüpfer of Siemens AG / Erlangen at the symposium.
[Released: October 2010]