Laser-driven soft-x-ray undulator source [Elektronische Ressource] / vorgelegt von Matthias Fuchs

Matthias Fuchs

Sujets

• Physik

Informations

• Publié par: ludwig-maximilians-universitat_munchen
• Publié le: 01 janvier 2010
• Nombre de lectures: 10
• Langue: English
• Poids de l'ouvrage: 18 Mo

Extrait

Laser-Driven Soft-X-Ray
Undulator Source
Matthias Fuchs
Ludwig-Maximilians-Universitat Munchen
August 2010
Matthias FuchsLaser-Driven Soft-X-Ray
Undulator Source
Matthias Fuchs
Dissertation
an der Fakult at fur Physik
der Ludwig{Maximilians{Universit at
Munc hen
vorgelegt von
Matthias Fuchs
aus Ellwangen/Jagst
Munc hen, den 04. August 2010Erstgutachter: Prof. Dr. Florian Gruner
Zweitgutachter: Prof. Dr. Toshiki Tajima
Tag der mundlic hen Prufung: 21. September 2010INTRODUCTION AND ABSTRACT
In order to put the results presented in this thesis into context, a short introduction is
given at this point. A brief summary of the experimental results is given on page (iii).
The discovery of X-ray radiation has shed light on unexplored territories in almost
all disciplines of science, ranging from chemistry, biology, physics, materials science and
10medicine to industrial applications. A wavelength in the Angstr om range (10 m) and
thus on the order of chemical bond lengths enables the radiation to resolve matter on
the atomic scale. In order to simultaneously gain temporal insight into dynamics on the
atomic scale, X-ray pulses with durations on the picosecond to femtosecond scale are
required.
The most powerful sources capable of delivering such pulses are based on synchrotron
radiation. Here, a magnetic eld de ects a pulse of ultra-relativistic electrons in a
direction transverse to its propagation. As a result of this acceleration, the electrons
emit bursts of highly-directed synchrotron radiation.
Synchrotrons are well established facilities with excellent control over the beam pa-
rameters. The discovery of synchrotron radiation goes back into the 1940s [Elder et al.,
1947]. Persistent research and a growing user community led the development from par-
asitic operation at high-energy accelerator facilities to dedicated high-brilliance (third
generation) X-ray sources. Brilliance is a measure for the ux, focusability and trans-
verse coherence of the radiation. The increase in brilliance has been due to both the
improvement of the electron beam in terms of emittance and pulse duration as well as
the evolution of the magnetic structures from bending magnets to more sophisticated in-
sertion devices such as undulators or wigglers. The pulse duration of the X-ray emission
is mainly given by that of the electron beam and is on the order of 100 ps for standard
third generation synchrotron sources based on storage rings. It can reach the sub pico-
second scale only with sophisticated techniques [Khan et al., 2006] and a signi cant loss
in photon ux. The development of sources emitting more brilliant beams has recently
culminated with the impressive demonstration of the world’s rst X-ray free-electron
laser (FEL) [Emma, 2009]. FELs emit pulses of coherent radiation with signi cantly
shorter duration than typical synchrotrons which increases the brilliance by more than
six orders of magnitude. However, both of these sources are based on kilometer-scale
radio-frequency accelerators, which makes them extremely costly and therefore only a
few facilities exist worldwide. This means that they cannot completely serve the large
user community.
iINTRODUCTION AND ABSTRACT
In 1979, Tajima and Dawson laid the theoretical foundation for a new generation of
compact particle accelerators [Tajima and Dawson, 1979]. In this scheme a laser pulse
18 2with an intensity on the order of 10 W=cm and a pulse duration of half the plasma
wavelength (typically a few tens of femtoseconds) is focused into a plasma where it ex-
cites a plasma wave. The wave trails the laser pulse at its group velocity and generates
accelerating electric elds which exceed the strength of those in conventional accelera-
tors by more than three orders of magnitude. Either background plasma electrons or
externally injected electrons can get trapped by these elds and { by \sur ng" them {
get accelerated to GeV-scale energies over distances of only a few centimeters.
However, this scheme relies on laser pulses with ultra-high intensities. Therefore, the
eld of laser-wake eld acceleration is very strongly dependent on advances in laser tech-
nology. Only the invention of the chirped-pulse ampli cation (CPA) 1985 [Strickland
and Mourou, 1985] made it possible to generate laser pulses with a su ciently high
intensity. It took until the mid-1990s for lasers to become mature enough to be used as
drivers allowing for wake eld-acceleration rst for externally injected electrons ([Clay-
ton et al., 1993], [Nakajima et al., 1995]), followed soon after by the acceleration of
self-injected electrons [Modena et al., 1995]. In these early experiments, the thermally
shaped electron spectra showed a high-energy cuto at a few tens of MeV. The experi-
mental developments were supported by advances in the theoretical description such as
analytical models of nonlinear waves, acceleration of electrons in these waves and non-
linear evolution mechanisms for ultra-intense laser pulses in plasmas (see [Esarey et al.,
1996] and references therein).
Developments in laser technology led to a decrease in pulse duration from the pi-
cosecond to the tens of femtoseconds range, while still maintaing the pulse energy and
thus increasing the peak intensity. The application of such laser pulses for wake eld-
acceleration resulted in a further increase in the accelerating gradient to reach 200 MV/m
and extended the high-energy cuto of the (still thermal) electron spectra to 200 MeV
[Malka et al., 2002]. Meanwhile computational power became large enough to perform
3D simulations including the highly non-linear evolution of the plasma and the laser
pulse which led to a deeper understanding of the processes that occur during the ac-
celeration. 3D particle-in-cell (PIC) simuations (see for example [Dawson, 1983]) led to
the discovery of a new acceleration scheme: the \bubble regime" [Pukhov and Meyer-ter
Vehn, 2002], which predicted the production of quasi-monoenergetic electron beams.
In the experiments described above, electrons were accelerated in the self-modulated
regime (for self-modulation, see section 2.6.4), which means that the laser pulse is longer
than a plasma period. Parts of the laser pulse are interacting with the accelerated elec-
tron bunch which leads to thermal energy spectra. The development of laser pulses with
ultra-high intensities and durations shorter than the plasma period resulted in 2004
in the acceleration of quasi-monoenergetic electron beams with energies of a few hun-
dred MeV and 100 pC of charge [Faure et al., 2004], [Geddes et al., 2004], [Mangles
et al., 2004]. Although the laser intensities in these experiments were technically not
high enough to operate in the bubble regime, PIC simulations showed that the laser
undergoes nonlinear processes. While propagating through the plasma these processes
signi cantly increase the laser intensity allowing a pulse with an initially insu cient in-
iitensity to drive an acceleration in the bubble regime. In 2006, electron beams of energies
of 1 GeV were demonstrated using acceleration distances of only 3 cm [Leemans et al.,
2006]. The shot-to-shot reproducibility of LWFA beams has increased in recent years
through careful control over various parameters of the laser pulse [Mangles et al., 2007]
or the gas target. One such control is the steady-state- ow gas cell scheme [Osterho
et al., 2008], which is used as the driver for the experiments described in this thesis.
New injection schemes ([Faure et al., 2006], [Geddes et al., 2008]) have the potential
to further increase the stability as well as to signi cantly improve the electron beam
quality. If su ciently high beam qualities and staging of several acceleration sections
can be shown, laser-wake eld accelerators could be potential candidates to drive future
ultra-high-energy particle colliders [Schroeder et al., 2009], [Tajima, 2010]
The experimental results described in this thesis demonstrate the successful
synergy between the research elds described above: the development of an undulator
source driven by laser-plasma accelerated electron beams. First e orts in this new eld
have led to the production of radiation in the visible to infrared part of the electro-
magnetic spectrum [Schlenvoigt et al., 2008]. In contrast to these early achievements,
the experiment described here shows the successful production of laser-driven undulator
radiation in the soft-X-ray range with a remarkable reproducibility. The source pro-
duced tunable, collimated beams with a wavelength of 17 nm from a compact setup.
Undulator spectra were detected in 70% of consecutive driver-laser shots, which is
a remarkable reproducibility for a rst proof-of-concept demonstration using ultra-high
intensity laser systems. This can be attributed to a stable electron acceleration scheme
as well as to the rst application of miniature magnetic quadrupole lenses with laser-
accelerated beams. The lenses signi cantly reduce the electron beam divergence and its
angular shot-to-shot uctuations
The setup of this experiment is the foundation of potential university-laboratory-sized,
highly-brilliant hard X-ray sources. By increasing the electron energy to about 1 GeV,
X-ray pulses with an expected duration of 10 fs and a photon energy of 1 keV could
be produced in an almost identical arrangement. It can a