Del Mar Photonics

The Banff Meeting on Structural Dynamics

Ultrafast Dynamics with X-rays and Electrons

Brochures to present at the exhibition:

Product Data Sheets

Pulse strecher/compressor
Avoca SPIDER system
Buccaneer femtosecond fiber lasers with SHG Second Harmonic Generator
Cannon Ultra-Broadband Light Source
Cortes Cr:Forsterite Regenerative Amplifier
Infrared cross-correlator CCIR-800
Cross-correlator Rincon
Femtosecond Autocorrelator IRA-3-10
Kirra Faraday Optical Isolators
Mavericks femtosecond Cr:Forsterite laser
OAFP optical attenuator
Pearls femtosecond fiber laser (Er-doped fiber, 1530-1565 nm)
Pismo pulse picker
Reef-M femtosecond scanning autocorrelator for microscopy
Reef-RTD scanning autocorrelator
Reef-SS single shot autocorrelator
Femtosecond Second Harmonic Generator
Spectrometer ASP-100M
Spectrometer ASP-150C
Spectrometer ASP-IR
Tamarack and Buccaneer femtosecond fiber lasers (Er-doped fiber, 1560+/- 10nm)
Teahupoo femtosecond Ti:Sapphire regenerative amplifier
Femtosecond third harmonic generator
Tourmaline femtosecond fiber laser (1054 nm)
Tourmaline TETA Yb femtosecond amplified laser system
Tourmaline Yb-SS femtosecond solid state laser system
Trestles CW Ti:Sapphire laser
Trestles femtosecond Ti:Sapphire laser
Trestles Finesse femtosecond lasers system integrated with DPSS pump laser
Wedge Ti:Sapphire multipass amplifier

Multi-terawatt lasers overview
Hydrogen Thyratrons - Deuterium Thyratrons - Untriggered Spask Gaps - Triggered Spask Gaps - X-ray tube
Rincon 800 third-order scanning cross-correlator for aligning 20 Terawatt Ti:Sapphire laser
MCP + phosphorous screen for imaging of XUV radiation (14eV- 160-eV) in high harmonics experiments
Femtosecond autocorrelator Reef-RTD 700-1300 nm
New Trestles fs/CW laser system which can be easily switched from femtosecond mode to CW and back.
Femtosecond Two-stage Amplifier System Wedge-XL (table-top terawatt system) - pdf
CORTES-800 40 TERAWATT LASER SYSTEM
New Beacon Femtosecond Fluoresscence Upconversion System
Tamarack C1560 femtosecond fiber laser
Pacifica THz Time Domain Spectrometer
Wedge TiSapphire Multipass Amplifier
New Hatteras femtosecond transient absorption system
Photon Scanning Tunneling Microscope - Power Point presentation (use read-only mode)
Atomic Force Microscope AFM HERON - sample quotes
Near-field Scanning Optical Microscope (NSOM) for nano-characterization and nanomanufacturing
Yb-based high-energy fiber laser system kit, model Tourmaline Yb-ULRepRate-07
Ytterbium-doped Femtosecond Solid-State Laser Tourmaline Yb-SS400
Pismo pulse picker for 1500-1600nm range

Del Mar Photonics Product brochures - Femtosecond products data sheets (zip file, 4.34 Mbytes) - Del Mar Photonics

Program and Notes

Probing molecular dynamics with short X-ray pulses from a
synchrotron.
Laurent Guerin, Marco Cammarata and Michael Wulff
European Synchrotron Radiation Facility, Grenoble, Cedex 38043, France
email: wulff@esrf.fr
Abstract
Fast protein nanocrystallography
We have examined the structure of laser excited molecules in solution by X-ray scattering using
short pulses of X-rays from the European Synchrotron. The experiments are performed on beamline
ID09B, a beamline for pump-probe experiments in physical, chemical and biological systems.
Fast reactions are typically triggered by ultrafast optical pulses and the scattering (or diffraction)
from delayed 100 ps pulses of X-rays are used to probe that structure of the sample at that time. I
will review the latest experiments and beamline techniques, in particular the installation of a fast
FReLoN CCD detector that has increased the efficiency of the beamline by a factor 5-10. It is now
possible to record around 1000 scattering spectra per hour, which highlights the need for on-line
data analyses. Finally we will show our plans for a new advanced pump-probe beamline to be built
in 2012 within the framework of the ESRF upgrade program.
The LCLS X-Ray FEL Facility
John Arthur
SLAC National Accelerator Laboratory, Menlo Park, CA 94025 USA
email: jarthur@slac.stanford.edu
Abstract
The Linac Coherent Light Source (LCLS) at the SLAC National Accelerator Laboratory is a freeelectron
laser based on self-amplified spontaneous emission (SASE) in the wavelength range 1.5
15A° . It includes an FEL undulator about 100m in length, driven by high-brightness electron pulses
with energy in the range 4.3-13.6 GeV prepared by a photoelectron gun and about 1km of linac.
The FEL x-ray pulses can be directed into any one of 6 experimental stations, which are being optimized
for various types of experiments. During 2009, LCLS began commissioning and supported
experiments in the first operational experimental station, optimized for studying the interaction of
soft-x- ray FEL pulses with atoms, molecules, and clusters in the gas phase. Lasing to saturation
was achieved throughout the LCLS design energy range, with every indication that the facility
can ultimately provide a significantly wider FEL photon energy range. By adjusting electron parameters,
FEL pulse widths were adjusted between about 10 and 300fs. Pump-probe experiments
with 50fs resolution were demonstrated using a Ti:sapphire laser pump and LCLS x-ray probe. The
overall stability and reliability of the LCLS x-ray source rivals that of synchrotron sources. In summary,
during early operation LCLS has proven to be a highly flexible and precisely-controllable
x-ray source, and has already exceeded all of its technical design goals. Outfitting of the remaining
experimental stations is well underway, with a three new stations expected to be commissioned in
2010. Plans for a major upgrade to the facility have already begun, promising increased photon
energy range and experimental capacity by about 2017.
A new generation of soft x-ray free electron lasers
R.W. Falcone, K. Baptiste, J. M. Byrd, J. Corlett, P. Denes, L. Doolittle, H. Gang, J.
Kirz, W. McCurdy, H. Padmore, G. Penn, J. Qiang, D. Robin, F. Sannibale, R.
Schoenlein, J. Staples, C. Steier, M. Venturini, W. Wan, R. Wells, R. Wilcox, A.
Zholents
Lawrence Berkeley National Laboratory Berkeley, CA, USA email: rwf@berkeley.edu
Abstract
Recent reports have identified the scientific requirements for a future soft x-ray light source and
a high-repetition-rate free-electron laser (FEL) facility responsive to them is being studied at
Lawrence Berkeley National Laboratory. The facility is based on a continuous-wave superconducting
linear accelerator with beam supplied by a high-brightness, high-repetition-rate photocathode
electron gun operating in CW mode, and on an array of FELs to which the accelerated beam is
distributed, each operating at high repetition rate and with even pulse spacing. Dependent on the
experimental requirements, the individual FELs may be configured for either self-amplified spontaneous
emission, seeded high-gain harmonic generation, echo- enabled harmonic generation, or
oscillator mode of operation, and will produce high peak and average brightness x-rays with a
flexible pulse format ranging from sub-femtoseconds to hundreds of femtoseconds. This new light
source would serve a broad community of scientists in many areas of research, similar to existing
utilization of storage ring based light sources.
We are developing a design concept for a 10-beamline, coherent, soft xray FEL array powered by
a 2.5 GeV superconducting accelerator operating with a 1 MHz bunch repetition rate. Electron
bunches of charge 10 pC to 1 nC are fanned out through a spreader, distributing beams to an array
of 10 independently configurable undulators and FEL beamlines with nominal bunch rates up
to 100 kHz. Additionally, one beamline (the last in the array) could be configured to operate at
higher repetition rate of 10 MHz or greater, in a dedicated operating mode, while simultaneously
operating the other nine FEL beamlines at 100 kHz. The FELs may be seeded by optical lasers to
control the X-ray output characteristics or may use SASE techniques, including generation of lowcharge,
high-brightness bunches with intrinsically short duration. Users specify the wavelength,
pulse duration, and polarization, so that the 10 simultaneously operating beamlines can be individually
optimized for specific experiments, including broad spectral coverage and multiple beam
capability. The spectral range is from 10 eV to 1 keV, with harmonics to approximately 5 keV at
reduced intensity. The beams may also be synchronized with optical lasers or IR and THz sources
for pumpprobe experiments. Three principal modes of operation are proposed: ultrashort pulse
(300 as-10 fs), short pulse (10 fs-100 fs), and high spectral resolution (requiring pulses from 100-
500 fs). The spectral bandwidth in each mode is anticipated to approach fundamental transform
limits. Other features include the capability to achieve high peak power ( 1 GW) for nonlinear
optics, control of peak power to reduce sample damage, and high average power ( 1-10 W) for
low-scattering-rate experiments. With up to 10 FEL beamlines and 20 X-ray beamlines, the facility
will be capable of serving 2000 users per year.
Key Laser Technologies for Future X-ray Sources
Franz X. K¨artner1, William S. Graves2 and David E. Moncton2
1Department of Electrical Engineering and Computer Science, and Research Laboratory of
Electronics, 2Nuclear Reactor Laboratory Massachusetts Institute of Technology, 77
Massachusetts Avenue, Cambridge, Massachusetts 02139, USA email: kaertner@mit.edu.
Abstract
Over the last few years, advances in femtosecond lasers have opened up the possibility to construct
fully coherent soft and hard x-ray sources that range from table-top size to kilometer long seeded
FELs. The later facilities will be combined laser and accelerator laboratories. In this presentation,
we discuss some of the laser technologies and physics central to the development of such sources,
including novel concepts for compact x-ray sources based increasingly, and perhaps entirely, on
lasers. First, we explain the origin of ultralow timing jitter of femtosecond lasers and discuss the
consequences for the control of electron-laser interactions with the precision of a few attoseconds,
or potentially better. As an example, long term stable timing distribution for large scale x-ray FELs
is shown. Systems designed at MIT now operate with sub- 10 fs precision over multiple days and
are currently implemented at the FERMI FEL in Trieste. Timing distribution systems approaching
attosecond level precision appear possible. Second, we discuss the production of laser radiation
in the EUV and XUV via high harmonic generation. We have derived and experimentally verified
closed form analytic expressions for high harmonic conversion efficiencies, verified them experimentally,
and predict the possibility of highly efficiency EUV sources, which may achieve 1%
conversion of optical power into a single harmonic for wavelengths as short as 13.5 nm. Such
sources can stand alone for EUV lithography, for example, or be used for seeding of FELs to reach
hard x-ray wavelengths with high longitudinal coherence. Third, we discuss our progress in the
development of an energy and power scalable single-cycle waveform synthesizer based on a fewcycle
optical parametric chirped pulse amplifier (OPCPA) system delivering synchronized 800 nm
and 2 micron pulses for attosecond pulse generation. The pump laser system developed as the
OPCPA driver is based on cryogenically cooled Yb:YAG. The large average power capabilities of
cryogenically cooled Yb-doped lasers together with advances in superconducting accelerator technology
enables ultrafast, bright and intense x-ray sources based on Inverse Compton Scattering
(ICS). In particular, we consider sources that are based on high repetition rate (100 MHz), high
brilliance electron beams from continuous wave superconducting accelerators operating at 4K. The
pulsed electron beam collides with a 1 MW optical beam of pico- or femtosecond laser pulses in
an enhancement cavity fed by a kW-class cryogenically cooled Yb-laser. Such a source is suitable
for a university or industrial laboratory and can generate quasi monochromatic x-ray beams with
average flux and brightness similar to a second generation synchrotron. Furthermore, the ICSsource
has a spot size of a few microns (much smaller than a synchrotron beam) enabling improved
high resolution phase contrast imaging and protein crystallography using 10-micron sized
crystals. Since the source output consists of ultrafast x-ray pulses, time-resolved x-ray diffraction
experiments in the sub pico-second regime are possible. Low repetition rate sources generating
femtosecond pulses of up to 1010 photons appear to be feasible. In the future, attosecond control
of the electron emission from nanostructured photocathodes and laser acceleration may produce
fully coherent x-rays from sources exploiting the ICS geometry. We discuss the physics and beam
properties of such sources as time allows.
Breaking the attosecond, Angstrom and TV/m field barriers with
ultra-fast electron beams
J.B. Rosenzweig1, G. Andonian1, P. Bucksbaum2, M. Ferrario3, S. Full1, A.
Fukusawa1, E. Hemsing1, M. Hogan2, P. Krejcik2, P. Muggli4, G. Marcus1, A.
Marinelli1, P.Musumeci1, B. OShea1, C. Pellegrini1, D. Schiller1, G. Travish1
1UCLA Dept. of Physics and Astronomy, 405 Hilgard Ave., Los Angeles, CA 90095
2Istituto Nazionale di Fisica Nucleare Laboratori Nazionali di Frascati, via Enrico Fermi 40,
Frascati (RM) Italy
3Stanford Linear Accelerator Center, Menlo Park, CA
4University of Southern California, Dept. of Engineering Physics, Los Angeles, CA
Abstract
Recent initiatives at UCLA concerning ultra-short, GeV electron beam generation have been aimed
at achieving sub-fs pulses capable of driving X-ray free-electron lasers (FELs) in single- spike
mode. This uses of very low charge beams, which may allow existing FEL injectors to produce
few-100 attosecond pulses, with very high brightness. Towards this end, recent experiments at the
Stanford X-ray FEL (LCLS, first of its kind, built with essential UCLA leadership) have produced
2 fs, 20 pC electron pulses. We discuss here extensions of this work, in which we seek to exploit
the beam brightness in FELs, in tandem with new developments at UCLA in cryogenic undulator
technology, to create compact accelerator/undulator systems that can lase below 0.15 Angstroms,
or be used to permit 1.5 Angstrom operation at 4.5 GeV. In addition, we are now developing experiments
which use the present LCLS fs pulses to excite plasma wakefields exceeding 1 TV/m,
permitting a table-top TeV accelerator for frontier high energy physics applications. In this scenario,
one focuses the beam to 100 nm transverse dimensions, where the surface Coulomb fields
are also at the TV/m level. These conditions access a new, novel regime for high field for atomic
physics, allowing frontier atomic physics experiments, including sub-fs plasma formation via barrier
suppression ionization (BSI) for subsequent wake excitation. Plans for experiments at SLAC
based on achieved beam parameters are presented, in which we evaluate the schemes for beam
focusing, BSI ionization, TV/m plasma wakefields excitation and ion collapse.


In-air femtosecond X-ray source
Jiro Matsuo and Masaki Hada
Department of Nuclear Engineering, Kyoto University, Sakyo, Kyoto, Japan Quantum Science
and Engineering Center, Kyoto University, Gokasho, Uji, Kyoto, Japan
email: matsuo@nucleng.kyoto-u.ac.jp
Abstract
The dynamical behavior of a crystalline structure is not only of scientific interest, but also has
technological importance. For instance, the ultra-fast phase transition used in recording materials
and laser-induced recrystallization for electrical devices are studied intensively. However, the transitional
mechanism of these materials in the femtosecond time scale has to be well understood. In
order to develop advanced materials and processing with better performance, fundamental considerations
are quite important. To explore various materials, there is a strong need for a compact and
easyto-use femtosecond X-ray source.
We have demonstrated that high-reputation rate and low peak power laser can deliver an amount
of X-ray similar to that generated with a low reputation rate and high peak power laser. A high
reputation rate and low peak power laser is commercially available nowadays. In addition, the new
X-ray source can be operated in He ambient. This new compact X-ray source is quite useful for
analyzing many different materials.
We will report on the performance of this X-ray source and discuss the possible applications for
ultra fast phenomena.
Synchrotron Radiation from Laser Accelerated Electrons
Heinrich Schwoerer1, Hans-Peter Schlenvoigt2
1Laser Research Institute, Stellenbosch University, Priv. Bag X1, Matieland 7602, South Africa,
email: heso@sun.ac.za
2Laboratoire pour l’Utilisation des Lasers Intenses, ´ Ecole Polytechnique, 91128 Palaiseau, France
email:hans-peter.schlenvoigt@polytechnique.edu
Abstract
Femtosecond laser pulses have revolutionized the knowledge of intramolecular and microscopic
solid state dynamics in the last two decades. This became possible since the duration of the light
pulses is on the order of the characteristic microscopic time scales, and the photon energy is in
the range of relevant electronic excitations. Transition states of photoinduced molecular and condensed
phase dynamics can be observed in real time by applying a pump probe spectroscopy
technique with ultrashort laser pulses or even femtosecond laser-generated electron pulses. However,
the wavelength regime accessible for femtosecond lasers is limited around the visible spectral
range and thereby restricts the interaction with matter to electronic transitions and their coupling
to the atomic motion.
Shorter wavelengths down to a few nanometers can be generated using electron storage rings
or linear accelerators equipped with undulators. This synchrotron radiation opens a more direct
view into intermolecular or solid state dynamics via time-resolved photon diffraction in crystals
and recently also of molecules which is of interest for a wide range of interdisciplinary research. If
an undulator is operated in the free-electron-laser mode (FEL), extremely brilliant, ultrashort, polarized,
and coherent light pulses are produced. FELs promise a wide applicability, spanning from
atomic and cluster physics through temporally resolved structural analysis of complex molecules
to plasma physics. However, they require km long LINACs producing several GeV electron energies
due to the limited energy gain per length of less than 50 MeV/m. Bridging the gaps between
femtosecond laser spectroscopy and synchrotron radiation sources may become possible with relativistic
laser plasma physics. Femtosecond lasers can be used to generate light intensities exceeding
1020 W/cm2, providing fields strong enough for electron particle acceleration up to GeV within a
few mm, with a few percent bandwidth and within a well-collimated beam [1]. The energy gain
per length for a laserplasma accelerator is significantly larger than for radio-frequency accelerators,
because the acceleration is based on a plasma.
In this paper, we discuss the status of generation of synchrotron radiation from laser-accelerated
electrons. A proof of principle experiment was reported by the authors [2,3], but significant improvements
in terms of energy, shot-to-shot reproducibility, pointing stability, and spectral width
of the driving electron beam have been realized since then. We discuss the potential and the limitations
of this novel all-optical synchrotron light source, as it might become an interesting ultrashort
pulsed (fs), tuneable VUV to x-ray coherent source, being smaller and more flexible compared to
accelerator-based sources.
[1] Leemans et al. Nature Physics, 2, 696 (2006), [2] Schlenvoigt et al. Nature Physics, 4, 130 (2008), [3] Schlenvoigt
et al IEEE Trans. Plasma Sci., 36, 1773 (2008).
Single-shot Ultrafast Electron Diffraction
O.J. Luiten, W.J. Engelen, S.B. van der Geer, A.J.C. Klessens, T. van Oudheusden,
P.L.E.M. Pasmans, M.P. Reijnders, E.P. Smakman, G. Taban, E.J.D. Vredenbregt
Department of Applied Physics, Eindhoven University of Technology, P.O. Box 513, 5600 MB
Eindhoven, The Netherlands, e-mail: o.j.luiten@tue.nl
Abstract
The development in recent years of ultrafast electron diffraction (UED) techniques has enabled the
first atomic-level, sub-ps studies of condensed matter phase transition dynamics. UED has also
been applied successfully to determine transient molecular structures with 1 ps resolution during a
chemical reaction of small molecules in the gas phase. Unfortunately, however, the application of
UED up to now has mostly been limited to processes which are sufficiently reproducible, because
recording a full diffraction pattern of sufficient quality requires 106 electrons, corresponding to,
typically, at least 100 shots. The number of electrons in a pulse is limited by space-charge forces,
which cause rapid expansion of the pulse and therefore loss of temporal resolution. A possible way
out is to accelerate the electron bunches to relativistic speeds, which slows down the space-charge
expansion and thus allows single-shot UED with sub-ps resolution.
We have developed a method to produce sub-ps electron bunches suitable for single-shot UED at
non-relativistic energies. The method relies on the use of radio-frequency (RF) techniques to invert
the space-charge expansion. We will report on the first experiments demonstrating RF compression
of 0.1 pC, 100 keV electron bunches. We have used these bunches to produce high- quality, singleshot
diffraction patterns of poly-crystalline gold.
In all UED experiments up to now electron bunches have been generated by femtosecond photoemission
from metal cathodes The transverse coherence length of the ensuing beams is limited to a
few nm for crystal samples of 100 μm size, and therefore does not allow the study of, e.g., protein
samples. As reported elsewhere, we are developing an ultracold electron source which should
enable coherence lengths of a few tens of nm for crystal samples with a size of 100 μm. We will
show that 0.1 pC, sub-ps, 100 keV electron bunches can be extracted from such a source, while retaining
the transverse beam coherence, by applying RF acceleration and phase-space manipulation
techniques. This should enable single-shot studies of macromolecular crystals.
Experimental realization of an ultracold electron source
E.J.D. Vredenbregt, G. Taban, M.P. Reijnders, E.P. Smakman, W.J. Engelen, S.B.
van der Geer, and O.J. Luiten
Department of Applied Physics, Eindhoven University of Technology, P.O. Box 513, 5600 MB
Eindhoven, The Netherlands, e-mail: e.j.d.vredenbregt@tue.nl
Abstract
We report on the development of an ultracold electron source, which is based on near-threshold
photo- or eld-ionization of a cloud of laser-cooled atoms. Such a source offers the unique combination
of low emittance and extended size that may be essential for achieving single-shot, ultrafast
electron diffraction of macromolecules. As reported elsewhere, a photo-emission electron source
that could provide 106 electrons in a 200 μm rms spot size with a 3 nm coherence length is
currently un- der development in our labs. However, sources that provide even larger coherence
lengths for a similar amount of electrons in a similar spot size are required to study the dynamics
of larger objects such as proteins. An appropriate source must have a small enough emittance in
order for all electrons to contribute to the diffraction pattern at the required coherence length. In
addition, the local electric accelerating eld at the source must be substantially larger than the eld
due to image charges in order for the pulse not to be lengthened and transversely distorted. The rst
criterion can be met by a variety of sources as it represents a trade-off between the source size and
the effective source temperature. The second criterion, however, favors an extended source, such
as the ultra-cold electron source presented in this contribution. Here we present measurements
of the effective temperature of such a pulsed electron source employing rubidium atoms that are
magneto-optically trapped at the center of an accelerator structure. Transverse source temperatures
ranging from 200 K down to 10 K are demonstrated, controllable with the wavelength of the ionization
laser. Together with the 50 μm source size, the achievable temperature enables a transverse
coherence length of 20 nm for a 100 μm sample size. On the order of 105 electrons are contained
in a (calculated) 50 ps long pulse when the trapped atoms are rst converted to a “frozen”Rydberg
gas from which electrons are extracted by a fastelectric eld pulse.
RF photoinjector based ultrafast relativistic electron diffraction
P. Musumeci, J. T. Moody, C. M. Scoby
UCLA Department of Physics and Astronomy, Los Angeles, CA 90095-1547
Abstract
Electron diffraction holds the promise to yield real time resolution of atomic motion in a easily accessible
environment like a university laboratory at a fraction of the cost than 4th generation x- ray
sources. Currently the limit in time-resolution for conventional electron diffraction is set by how
short an electron pulse can be made. A possible solution to maintain the highest possible beam
intensity without excessive pulse broadening from space charge effects is to increase the electron
energy to the MeV level where relativistic effects significantly reduce the space charge forces.
Rf photoinjectors can in principle deliver up to 107 -108 electrons packed in bunches of 100 fs
length allowing an unprecedented time resolution and enabling the study of irreversible phenomena
by single shot diffraction patterns. The UCLA Pegasus laboratory has recently demonstrated
time resolved single shot electron diffraction using a 200 fs long relativistic beam from an rf
photoinjector. We use this novel technique to study the evolution of the laser induced solid-liquid
transition in metal foils of different thicknesses. The preliminary results of this experiment and the
future directions of ultrafast electron diffraction with relativistic electrons will be discussed.







Hard X-Ray Emission Spectroscopy
Pieter Glatzel1, Gyorgy Vanko2, Christian Bressler3, Marcin Sikora4, Amelie
Juhin5, Frank de Groot5, Simo Huotari1, Grigory Smolentsev6
1European Synchrotron Radiation Facility, BP 220, F-38043 Grenoble Cedex, France
email:glatzel@esrf.fr
2KFKI Research Institute for Particle and Nuclear Physics, H-1525 Budapest, Hungary
3European XFEL, c/o DESY, Notkestrasse 85, D-22 607 Hamburg, Germany
4Faculty of Physics and Applied Computer Science, AGH University of Science and Technology,
30-059 Krakow, Poland
5Department of Inorganic Chemistry and Catalysis, Utrecht University, 3584 CA Utrecht, The
Netherlands
6Faculty of Physics and Research center for Nanoscale Structure of Matter, Southern Federal
University, 344090 Rostov-on-Don, Russia
Abstract
Inner-shell spectroscopies using hard X-rays provide an element-selective and truly bulk- sensitive
probe with great flexibility regarding the sample environment. Analysis of the emitted X-rays
(XES) [1, 2] as opposed to scanning the energy of the incident X-ray beam to measure the absorption
(XAS) appears to be an attractive option for upcoming 4th generation sources. The instrumentation
and theoretical understanding of X-ray emission spectroscopy has made important progress
and the technique is nowadays routinely used for the characterization of the local coordination and
electronic structure.
XES includes a number of techniques. The least challenging in terms of instrumentation is nonresonant
excitation of the sample with an incident beam of large energy bandwidth (tens of eV).
Non-resonant XES may provide information on the oxidation and spin-state as well as the ligand
orbitals. Resonant XES or resonant inelastic X-ray scattering (RIXS) requires a monochromatic
beam that is tunable within a few eV around the Fermi energy. RXES is used to study electronelectron
interactions and crystal field splittings in detail. The technique also enables to measure
charge-neutral (i.e. non-ionizing) and element-selective excitations within the valence band. The
spectral range is thus similar to UV-Vis spectroscopy but with different selection rules for electron
transitions and the energy range can be extended well beyond 6 eV that limits standard optical
spectroscopies.
The presentation will provide in introduction to the various techniques and discuss their potential
for applications at hard X-ray free electron lasers. The instrumentation for single shot experiments
and the feasibility will be addressed.
[1] F.M.F. de Groot and A. Kotani, Core Level Spectroscopy of Solids. Advances in Condensed Matter Science, ed.
D.D. Sarma, G. Kotliar, and Y. Tokura. Vol. 6. 2008, New York: Taylor and Francis.
[2] P. Glatzel and U. Bergmann, High resolution 1s core hole X-ray spectroscopy in 3d transition metal complexes -
electronic and structural information, Coord. Chem. Rev. 249 65-95 (2005).
Phase sensitive x-ray imaging and ultrafast chemical dynamics
C. Rose-Petruck1, B. Ahr1, V. Ortiz1, Y. Liu1, G. Diebold1, Z. Derdak2, J. Wands2,
B. Adams3, M. Chollet3
1Department of Chemistry, Box H, Brown University, Providence, RI 02912, USA email:
crosepet@brown.edu
2The Liver Research Center, Rhode Island Hospital and Warren Alpert Medical School of Brown
University, Providence, RI 02912, USA
3Advanced Photon Source, Argonne National Laboratory 9700 S. Cass Ave, Argonne, IL 60439,
USA
Abstract
Recent progress in the area of phase-sensitive x-ray imaging of bio-medical tissues as well as
density waves in materials is discussed. Furthermore, recent 2-ps resolution x-ray absorption data
from our experiment at the Advanced Photon Source (APS), ID7-C will be presented.
The high transverse coherence of the x-rays produced form laser-driven x-ray sources has been
used for in-line holographic hard x-ray imaging of murine livers as well as clathrate hydrate slurries.
The employed phase-sensitive x-ray imaging method is fundamentally different from conventional
x-ray shadowgraphy because the mechanism of image formation does not rely on differential
absorption by matter. Instead, x-ray beams undergo differential phase shifts and subsequently
interfere constructively or destructively at the x-ray detector. Hence, material densities are distinguished
by the differences between the real parts of their refractive indices rather than their
absorptive properties. Example images of cancer bearing livers are presented. The chemical application
of x-ray phase contrast imaging aims to observe the melting dynamics of clathrate hydrates
in water solutions. These compounds are examples of chemical guest-host systems and are of
interest for the capture of CO2 and contaminant gases from power plant flue gases.
Recently, the first x-ray absorption spectroscopic measurements of the ligand substitution of
Fe(CO)5 have been carried out at the APS with 2-ps temporal resolution. This resolution is
achieved in a 400-nm pump x-ray probe arrangement by detecting the x-ray pulses transmitted
through the sample solution with a steak camera after photo excitation. An ultrafast Fe K-edge
shift with subsequent recovery has been observed, which is consistent with impulsive Fe-CO bond
elongation and recovery.
Theory and Simulation of Time-Resolved X-ray Diffraction
Klaus Braagaard Møller
Department of Chemistry, Technical University of Denmark, 2800 Kgs. Lyngby, Denmark email:
klaus.moller@kemi.dtu.dk
Abstract
Pulsed x-ray sources can be used for real-time observation of chemical dynamics [1]. Recently,
we derived the basic theoretical formulation for x-ray diffraction with pulsed fields using a fully
quantized description of light and matter [2], which is in contrast to previous accounts on timeresolved
x-ray diffraction on dynamic non-equilibrium structures, where the pulsed radiation field
was treated classically [3-5]. We present some of the key features of our derivation and apply the
theory to the laser-induced bond dynamics of simple molecules in particular, in the context of
the upcoming free-electron x-ray lasers producing high-intensity x-ray pulses with duration of 100
femtoseconds or less [6,7]. The talk will highlight the differences between the expression for the
time-dependent scattering signal we obtain from a first principles treatment and what one gets from
just “adding”time to the expression for signal from time-independent scattering theory, and we will
argue why the latter may work well for difference scattering images. The talk will also touch upon
two issues that become important when moving from the 100 ps time resolution available at current
synchrotron sources to 100 fs time resolution, namely laser-induced anisotropy and the importance
of taking both vibrational population and vibrational hole dynamics into account.
[1] Ihee, H. et al. Science 2005, 309, 1223.
[2] Henriksen, N. E.; Mller, K. B. J. Phys. Chem. B 2008, 112, 558.
[3] Cao, J.; Wilson, K. R. J. Phys. Chem. A 1998, 102, 9523.
[4] Bratos, S. et al. J. Chem. Phys. 2002, 116, 10615.
[5] Tanaka, S.; Chernyak, V.; Mukamel, S. Phys. Rev. A 2001, 63, 063405.
[6] Tschentscher, T. Chem. Phys. 2004, 299, 271.
[7] Gaffney, K. J.; Chapman, H. N. Science 2007, 316, 144.
First Experiments with the AMO Instrument at LCLS
John D. Bozek, Christoph Bostedt and Jean Charles Castagna
Linac Coherent Light Source, SLAC National Accelerator Laboratory,
2575 Sand Hill Road, Menlo Park, CA 94025, USA
email: jdbozek@slac.stanford.edu, bostedt@slac.stanford.edu, castagna@slac.stanford.edu
Abstract
An instrument has been designed, built and commissioned to take advantage of the unique ultrafast
duration and ultra-intense x-ray beam of the Linac Coherent Light Source (LCLS) for atomic,
molecular and optical (AMO) physics experiments. The instrument was commissioned over the
summer of 2009 and used for the first set of peer-reviewed and facility approved user experiments at
the LCLS in the subsequent months through the end of the year. Without exception the experiments
were successful and numerous exciting new results were obtained, some of which are reported
separately here at this meeting. The design and performance of the AMO instrument along with
the performance of the LCLS will be presented here.
The LCLS is the first of three x-ray free electron lasers (FELs) being built in the U.S., Japan
and Germany to begin operations. From the first time electrons were accelerated in the linac
and injected into the undulators to the most recent experiments, the LCLS has been fantastically
successful. In spite of early reservations among numerous reviewers evaluating the x-ray free
electron laser concept, the LCLS lases robustly. The source has been very dependable in its first
five months of operation with very few (and short) unscheduled down times. The LCLS x-ray FEL
source has also proved to be very versatile, producing pulses ranging in duration from a few fsec to
300 fsec over a photon energy range of 800-2000eV with pulse energies up to 3.5mJ. Currently in a
scheduled maintenance period, the LCLS will begin delivering its design goal 0.15nm radiation to
the first hard x-ray experiments, when it is started up again in May 2010, satisfying another design
goal of the facility.
The AMO instrument was designed to capitalize on the unique properties of the short intense pulses
of x-rays generated by the LCLS to study some of the simplest forms of matter; atoms, molecules
and clusters. It consists of focusing optics that produce a 1um focus in the interaction region of
the first experimental chamber and 5um in the second chamber. Two experimental chambers are
located about 1m and 3m downstream of the optics. The first chamber utilizes a skimmed, pulsed
supersonic jet to introduce sample into the middle of an ion time-of-flight (TOF) spectrometer and
five electron TOF spectrometers. Downstream, in the second chamber, a capillary is used to inject
a steady stream of gas into the interaction region of a magnetic bottle spectrometer. IR or higher
harmonics from a synchronized optical laser have been used for pump-probe experiments in both
chambers. Special attention was paid to the data acquisition system to be able to handle the large
amounts of data resulting from measurement of complete spectra from all instruments for each
shot of the LCLS.
First results on nonlinear dynamics in diatomic molecules using the
LCLS free electron laser
M. Hoener1
,
2, L. Fang 1, M. Guehr 3,C. Blaga 4, C. Bostedt 5, J.D. Bozek 5, P.
Bucksbaum 3, C. Buth 3
,
6, R. Coffee 3, J. Cryan 3, L. DiMauro 4, O. Gessner 2, J.
Glownia 3, E. Hosler 2, E. Kanter 6 , O. Kornilov 2, E. Kukk 8 , S. Leone 2, B.K.
McFarlan 3, B. Murphy 1, S.T. Pratt 6 , D. Rolles 9 , and N. Berrah 1
1Western Michigan University, Physics Department, Kalamazoo, MI, 49008, USA
email: mhoener@lbl.gov
2Lawrence Berkeley National Laboratory, Berkeley, CA, 94720, USA
3Ohio State University, Department of Physics, Columbus, OH, 43210, USA
4LCLS, Menlo Park, CA, 94025, USA
5PULSE Institute, SLAC, Menlo Park, CA 94025, USA
6Louisiana State University, Baton Rouge, LA, 70803, USA
7Argonne National Laboratory, Argonne, IL 60439, USA
8Dept. of Physics and Astronomy, University of Turku, FI-20014 Turku, Finland
9Max Planck ASG, CFEL, 22761 Hamburg, Germany
Abstract
The unprecedented peak power at x-ray wavelengths of the Linac Coherent Light Source (LCLS)
at the SLAC National Accelerator Laboratory, was used to study ultra fast, nonlinear and x-ray
multiphoton physics in molecules. We report on fundamental questions concerning the creation
and decay of multiple core-holes and, in particular, double core-holes in N2 . We investigated both
the Auger and secondary electron relaxation pathways subsequent to multiple core vacancies in
molecules, and the fragmentation patterns and charge-state distributions of the resulting ions as
function of wavelength, pulse duration and intensity. The new light source allows the characterization
of complex molecular ionization and dissociation dynamics and provides new insight into
the correlated motion of the electrons remaining in the targets and into fundamental aspects of
ultrafast molecular physics and chemistry. In addition our work contributes to the foundation for
future imaging experiments on molecules. The LCLS photon beam was focused to about 1μm
diameter spot producing an intense x-ray laser beam of up to 1018 W/cm2 , sufficient to investigate
multiphoton, multiple core-holes, and multiple-ionization processes.
The experiment was performed at the AMO beamline, which is equipped with an ion time-offlight
spectrometer to determine the charge state and kinetic energy distribution of the ions as
well five angle and energy resolving electron time-of-flight spectrometers to detect the emitted
photoelectrons and Auger electrons.
This work was supported by the DOE-SC-BES, Chemical Sciences, Geosciences and Biosciences
Division.
In-situ observation of irreversible reactions in liquids and gases by
Dynamic Transmission Electron Microscopy (DTEM)
N. D. Browning1
,
2
,
3, G. H. Campbell1,1 J. E. Evans1
,
3, K. L. Jungjohann2, W. E.
King1, T. B. LaGrange1, B. W. Reed1, M. Santala1
1Condensed Matter and Materials Division, Physical and Life Sciences Directorate, Lawrence
Livermore National Laboratory, 7000 East Avenue, Livermore, Ca 94550. USA email:
nbrowning@ucdavis.edu
2Department of Chemical Engineering and Materials Science, University of California-Davis,
One Shields Ave, Davis, Ca 95616. USA
3Department of Molecular and Cellular Biology, University of California-Davis, One Shields
Ave, Davis, Ca 95616. USA
Abstract
In response to a need to be able to observe dynamic phenomena in materials systems with both
high spatial ( 1nm or better) and high temporal ( 1μs or faster) resolution, a dynamic transmission
electron microscope (DTEM) has been developed at Lawrence Livermore National Laboratory
(LLNL). The high temporal resolution is achieved in the DTEM by using a short pulse laser
to create the pulse of electrons through photo-emission (here the duration of the electron pulse is
approximately the same as the duration of the laser pulse). This pulse of electrons is propagated
down the microscope column in the same way as in a conventional high- resolution TEM. The
only difference is that the spatial resolution is limited by the electron- electron interactions in the
pulse (a typical 10ns pulse contains 108 electrons). To synchronize this pulse of electrons with
a particular dynamic event, a second laser is used to “drive”the sample a defined time interval
prior to the arrival of the laser pulse. The important aspect of this dynamic DTEM modification is
that one pulse of electrons is used to form the whole image, allowing irreversible transitions and
cumulative phenomena such as nucleation and growth, to be studied directly in the microscope.
The use of the drive laser for fast heating of the specimen presents differences and several advantages
over conventional resistive heating in-situ TEM such as the ability to drive the sample into
non-equilibrium states. So far, the drive laser has been used for in-situ processing of nanoscale
materials, rapid and high temperature phase transformations, and controlled thermal activation of
materials. In this presentation, a summary of the development of in-situ stages for both the existing
DTEM at LLNL and a new DTEM being installed at UC-Davis will be described. Particular
attention will be paid to the potential for gas stages to study catalytic processes and liquid stages
to study biological specimens in their live hydrated state. The potential improvements in spatial
and temporal resolution that can be expected through the implementation of upgrades to the lasers,
electron optics and detectors used in the new DTEM will also be discussed along with the correlation
of dynamic results with conventional high resolution imaging and spectroscopic methods in
TEM.
Aspects of this work are performed under the auspices of the U.S. Department of Energy by Lawrence Livermore
National Laboratory and supported by the Office of Science, Office of Basic Energy Sciences, Division of Materials
Sciences and Engineering, of the U.S. Department of Energy under Contract DE-AC52-07NA27344. Aspects of this
work at UC-Davis were supported by DOE NNSA-SSAA grant number DE-FG52-06NA26213 and NIH grant number
RR025032-01.
Transient Electric Fields Induced By Ultrafast Pulsed Laser
Irradiation and Implications for Time-Resolved Reflection Electron
Diffraction
Hyuk Park1 and Jian-Min Zuo2
Department of Materials Science and Engineering, and Frederick Seitz Materials Research
Laboratory University of Illinois, Urbana-Champaign Urbana, IL 61801, USA
email:1hyukpark@illinois.edu, 2jianzuo@illinois.edu
Abstract
Studies of ultrafast processes using time resolved reflection high energy electron diffraction have
revealed unusual lattice contraction and expansion and phase transitions in a number of materials.
The dynamic processes are initiated by ultrafast laser irradiation. Understanding the interaction
of ultrafast pulsed laser with matter is thus critical for understanding these phenomena. It is also
important for understanding the physics of laser ablation and the laser induced non- equilibrium
carrier dynamics in metals and semiconductors, including plasmonics. When an intense laser
pulse of femtoseconds (fs) in duration hits the surface of a targeted matter, it excites a hot electron
gas. Part of the hot electrons is emitted from the surface in a way similar to thermionic emission.
Electrons can also be emitted through multiphoton photoemission (MPPE) or thermally assisted
MPPE. The emitted electrons travel at speeds that create transient electric fields (TEFs). To detect
TEFs and study the dynamics of emitted electrons, we have developed a time resolved an electron
beam imaging technique that allows us to measure TEFs above a sample surface at picoseconds
time resolution. We have also developed a model of the TEFs based on the propagation of emitted
electrons and the percentage of electrons escaping from the surface. The results will be reported for
silicon and graphite. The measured field strength and direction change with time; at the pump laser
fluence of 67.7mJ/cm2, the maximum field reaches 34 kV/m at 0.29 mm away from the silicon
surface. We show that the TEF can induce large deflection of the reflected electron beams and
changes in their intensity. The implications of our results for previous reported ultrafast structural
studies will be discussed in the talk.




Dynamics of cooperative lattice-charge (spin) coupled phenomena
induced by fs laser light irradiation studied by time-resolved X-ray
diffraction
S.Koshihara1
,
2, H.Ichikawa2, S.Nozawa2
,
6, T.Sato2, A.Tomita2, K.Ichiyanagi2,
M.Chollet1, L.Guerin2, N.Dean3, T.Arima4, H.Sawa5
,
6, S.Adachi2
,
6 and K.Miyano7
1JST, CREST & Department of Materials Science & FRC, Tokyo Institute of Technology,
Meguro-ku, Tokyo 152-8551, Japan,email: skoshi@cms.titech.ac.jp
2Non-equilibrium Dynamics Project, ERATO, JST, Tsukuba 305-0801, Japan,
3Department of Physics, University of Oxford, Clarendon Laboratory, Parks Rd. Oxford, OX1
3PU, UK
4Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai
980-8577, Japan,
5Department of Applied Physics, Nagoya University, Nagoya 464-8603, Japan.
6Photon Factory, Institute of Materials Structure Science, High Energy Accelerator Research
Organization, Tsukuba 305-0801, Japan,
7Research Center for Advanced Science and Technology, University of Tokyo, Tokyo 153-8904,
Japan,
Abstract
We make a report on the pico-second dynamics of normal and super lattice structures triggered by
fs laser irradiation utilizing ps time-resolved x-ray diffraction technique in the thin film of manganite
with charge and orbital ordering; (Nd0.5Sr0.5)MnO3. The Jahn-Teller distortion becomes
weak, i.e. orbital ordering melts, in the photo-induced state leading to the large changes in optical
and magnetic properties, though structural coherence is kept even well after excitation. The
obtained results shows that the origin of the gigantic photo-response is due to appearance of new
state characteristic only for light-induced far-equilibrium condition but not the simple mixture of
orbital ordered and disordered states in nanometer scale as like thermally induced phase change
Optical Control in Complex Solids
A. Cavalleri
Max Planck Group for Structural Dynamics, University of Hamburg in CFEL
email: andrea.cavalleri@mpsd.cfel.de
Abstract
In this talk I will cover some of our recent work in studying photo-induced dynamics in complex
solids. I will focus on the time dependent response of Peierls insulators and on measurements
in Mott Insulators on the timescale of hopping and correlations. I will also discuss the case of
half doped manganites, where we have combined a variety of time-resolved measurements, spanning
THz to soft x-ray wavelengths, to understand how light pulses perturb electronic and lattice
structure, as well as magnetic and orbital arrangements on the Ultrafast timescale.
Structural dynamics of the nearly commensurate phase in the
Charge Density Wave compound 1T-TaS2 probed by ultrafast
electron diffraction
Maximilian Eichberger1, Hanjo Schfer1, Jure Demsar1, Helmuth Berger2, Gustavo
Moriena3, Germn Sciaini3, and R.J. Dwayne Miller3
1Department of Physics, University of Konstanz, D-78457, Germany
email: maximilian.eichberger@uni-konstanz.de
2Physics Department, EPFL, CH-1015 Lausanne, Switzerland, email: helmuth.berger@epfl.ch
3Institute for Optical Sciences and Departments of Chemistry and Physics, University of Toronto,
Toronto, ON, M5S 3H6, Canada email: dmiller@lphys.chem.utoronto.ca
Abstract
Femtosecond spectroscopy is becoming an important tool for investigation of the so called strongly
correlated systems due to its intrinsic ability to determine the interaction strengths between various
degrees of freedom which lead to the fascinating phenomena like superconductivity or colossal
magnetoresistance. Low dimensional charge density wave (CDW) systems, with their inherently
multi-component order parameter (modulation of carrier density is accompanied by the modulation
of the underlying lattice) present no exception. In the past decade various one and two dimensional
CDWs have been studied by time-resolved optical1−5 as well as photoemission6,7 techniques focusing
on the dynamics of photoexcited electrons and collective modes. Recently, first systematic
studies on the photoinduced melting of the CDW order has been reported, where the results suggest
that on the sub-picosecond time scale when melting and subsequent initial recovery of the
electronic order takes place the lattice remains unperturbed in its modulated state8.
Here we report on the first studies of photoinduced CDW transition where the dynamics of the
CDW modulation following photoexcitation with an intense optical pulse was probed directly by
means of ultrafast electron diffraction. The results demonstrate an extremely fast suppression of
the CDW modulation (within 200 fs) and the sub-picosecond recovery dynamics. The possible
mechanisms of such rapid recovery of the CDW order are going to be discussed.
1. J. Demsar, K. Biljakovic, D. Mihailovic, Phys. Rev. Lett. 83, 800 (1999).
2. J. Demsar, et al., Phys. Rev. B 66, 041101 (2002).
3. K. Shimatake, Y. Toda, and S. Tanda, Phys. Rev. B 75, 115120 (2007).
4. D.M. Sagar et al., J.Phys. Cond. Mat. 19, 436208 (2007).
5. R.V. Yusupov, et al., Phys. Rev. Lett. 101, 246402 (2008).
6. L. Perfetti, et al., Phys. Rev. Lett. 97, 067402 (2006).
7. F. Schmitt, et al., Science 321, 1649 (2008).
8. A. Tomeljak, et al., Phys. Rev. Lett. 102, 066404 (2009).
Coherent Phonon in Iron Pnictide Superconductor Ba(Fe1−x Cox )2
As2 (x=0.06 and x=0.08)
D. Boschettoa,
, B. Mansartb , A. Savoiaa , F. Rullier-Albenquec , A. Forgetc , D.
Colsonc , A. Roussea , and M. Marsib
aLaboratoire dOptique Appliqu´ee, ENSTA, CNRS, Ecole Polytechnique, 91761
Palaiseau, France,email: davide.boschetto@ensta.fr, annunziata.savoia@ensta.fr
bLaboratoire de Physique des Solides, CNRS-UMR 8502, Universit Paris-Sud,
F-91405 Orsay, France, email: mansart@lps.u-psud.fr, marsi@lps.u-psud.fr
cService de Physique de l’Etat Condens´e , Orme des Merisiers, CEA Saclay (CNRS URA 2464),
91195 Gif-Sur-Yvette cedex, France
Abstract
What’s the role of phonon in high temperature superconductivity? The opinions of the scientists
on this theme diverge, giving rise to interesting debate and fancy exper- imental essays. The recent
discovery of high critical temperature superconductivity in iron pnictide compounds has driven the
attention of a large and multidisciplinary community. In these complex materials, the interplay between
all the degrees of freedom of the crystal such as spin, charge and lattice, entails the existence
of an interesting phase diagram. Here, understanding the role of phonon in the supercon- ducting
phase transition is a key point to better understand the occurrence of the superconductivity in these
compounds. We will report on the rst study of coherent A1g optical phonon mode in the superconductor
iron pnictide Ba(Fe1−x Cox )2 As2 (x=0.06 and x=0.08) [1], excited and detected in time
domain in a pump and probe scheme by a 40 fs laser pulse. The transient reectivity was measured
for different crystal temperatures and doping. The optical phonon parameters such as amplitude,
frequency and damping time, are measured across the superconductivity phase transition. Our results
suggest that the A1g optical phonon mode do not participate to the superconductivity phase
transition in these compounds.
[1] B. Mansart el al., Physical Review B 80, 172504 (2009).




Coherent and incoherent femtosecond structural dynamics in solids
studied by x-ray diffraction
S. L. Johnson1
,
, P. Beaud E1, E. Vorobeva1, C. J. Milne1, R. De Souza1, U. Staub1,
´ E. D. Murray3, S. Fahy4, Q. X. Jia5, G. Ingold1
1Swiss Light Source, Paul Scherrer Institut, Villigen, Switzerland
2Laboratoire de Spectroscopie Ultrarapide, Ecole Polytechnique F´ed´erale de Lausanne, Lausanne,
Switzerland
3Department of Physics and Astronomy, Rutgers University, Piscataway, New Jersey, USA
4Tyndall National Institute and Department of Physics, University College, Cork, Ireland
5Los Alamos National Laboratory, Los Alamos, NM, USA
email: *steve.johnson@psi.ch
Abstract
The fundamental time scales for structural dynamics in a crystalline solid is set by the periods of
the normal mode lattice vibrations, typically on the order of 100 fs for the fastest modes. Perturbation
of the crystal on time scales comparable to or even shorter than these periods can lead to
novel non-equilibrium structural phenomena. X-ray diffraction applied on the femtosecond time
scale offers a way to directly study these non-equilibrium structural dynamics. The femtosecond
slicing facility at the Swiss Light Source has in this way been able to apply x-ray diffraction to
observe several different types of ultrafast structural phenomena in solids. In this talk we discuss
examples including coherent phonons, phonon squeezing and the photo-induced melting of charge
and orbital order in a manganite.
Real-time structural dynamics in materials on femtosecond and
picosecond time-scales
A.M. Lindenberg
Department of Materials Science and Engineering / Photon Science Stanford University / SLAC
National Accelerator Laboratory 476 Lomita Mall, Stanford CA 94305
email: aaronl@stanford.edu
Abstract
The use of femtosecond x-ray pulses to probe materials opens up new windows into atomic- scale
structural and electronic dynamics and the functional properties of materials through both x-ray
scattering and x-ray absorption techniques. With the advent of new sources of femtosecond x-rays
at synchrotrons and free electron lasers in recent years, the range of accessible time-scales and
length-scales that can be probed has been dramatically increased, and provides new methods for
elucidating how atoms move in materials in real time. In this talk, I will present recent work probing
ultrafast dynamics in the solid and liquid phase, at atomic-scale resolution. We will present
hard x-ray diffraction measurements of the first steps in the solid- liquid phase transition and the
dynamics of the resulting disordered/liquid phase in both bulk and nanocrystalline systems, including
the dynamics of a unique intermediate phase associated with superionicity at the nanoscale. We
will show how ultrafast x-ray studies can be used to capture the polarization dynamics associated
with perovskite ferroelectrics, leading towards all-optical control of the ferroelectric polarization.
Finally we will present recent soft x-ray transmission measurements of ultrafast bond-breaking
dynamics in the liquid phase of water.
Femtosecond x-ray powder diffraction
M. Woerner, F. Zamponi, Z. Ansari, J. Dreyer, T. Elsaesser
Max-Born-Institut, Max Born Strasse 2A, 12489 Berlin, Germany
email: woerner@mbi-berlin.de, zamponi@mbi-berlin.de, ansari@mbi-berlin.de
dreyer@mbi-berlin.de, elsasser@mbi-berlin.de
Abstract
Fast protein nanocrystallography
We report on the rst femtosecond x-ray powder diffraction experiment in which we directly map
the transient electronic charge density in the unit cell of a crys talline solid with 30 picometer
spatial and 100 femtosecond temporal resolution. X-ray diffraction from polycrystalline powder
samples, the Debye Scherrer diffrac tion technique, is a standard method for determining equilibrium
structures. The intensity of the Debye Scherrer rings is determined by the respective x-ray
structure factor which represents the Fourier transform of the spatial electron density.
In our experiments, the transient intensity and angular positions of up to 20 Debye Scherrer
reections from a polycrystalline powder are measured and unravel for the rst time a concerted
electron and proton transfer in hydrogen-bonded (NH4)2 SO4 crystals. Photoexcitation of ammonium
sulfate induces a sub-100 fs electron transfer from the sulfate groups into a highly conned
electron channel along the z-axis of the unit cell. The latter geometry is stabilized by transferring
protons from the adjacent ammonium groups into the channel. Time-dependent charge density
maps derived from the diffraction data display a periodic modulation of the channels charge den
sity by low-frequency lattice motions with a concerted electron and proton motion between the
channel and the initial proton binding site. A deeper insight into the un derlying microscopic
mechanisms is gained by quantum chemical calculations with the result that the photo-excited
electron from the sulfate groups triggers up 15 proton transfer events along the reaction trajectory
NH
+
+ SO
2−
$ NH3 + HSO
−.
Our results set the stage for femtosecond structure studies in a wide class of (bio)molecular
materials.
Tracking Consecutive Steps of Photoinduced Switching Dynamics of
Spin-Crossover Materials by X-ray Diffraction & Optical
Pump-Probe Experiments.
Eric Colleta,b, Ch´erif Bald´ea, Maciej Lorenca, Marina Servola, Marylise Burona,
Herve Cailleauaa,b, Marie-Laure Boillotc, Shin-ya Koshiharad, Laurent Gu´erinec,
and Michael Wulffe.
aInstitut de Physique de Rennes, University of Rennes 1, France email:
eric.collet@univ-rennes1.fr, cherif.balde@univ-rennes1.fr, maciej.lorenc@univ-rennes1.fr,
marina.servol@univ-rennes1.fr, marylise.buron@univ-rennes1.fr, herve.cailleau@univrennes1.
fr
bInstitut Universitaire de France, Paris, France.
cInstitut de Chimie Molculaire et Matriaux d’Orsay, University of Paris-Sud, France email:
mboillot@icmo.u-psud.fr
dTokyo institute of Technology, Tokyo, Japan. email: skoshi@cms.titech.ac.jp
eEuropean Synchrotron Radiation facility, Grenoble, France. email: laurent.guerin@esrf.fr ,
wulff@esrf.fr
Abstract
Light may direct the functionality of a material through spectacular collective and/or cooperative
photoinduced phenomena in the solid state. This can trigger the transformation of the material
towards another macroscopic state of different electronic and/or structural order, for instance from
non magnetic to magnetic or from insulator to conductor. This addresses photosteady instabilities
as well as light pulse driven transformations. The increase of sophisticated instrumentation, including
ultra-fast time-resolved diffraction [1], gives fascinating capabilities not only to observe
and understand the elementary dynamic processes in materials but also to watch how matter works
and can be directed to a desired outcome. We present here detailed investigation of the out-ofequilibrium
spin-state switching dynamics of a molecular Fe(III) spin-crossover solid triggered
by a femtosecond laser flash. The time-resolved x-ray diffraction and optical results [2-4] show
that the dynamics span from sub- picosecond local photo-switching followed by volume expansion
on nanosecond time scale and thermal switching on microsecond) time-scale. We discuss a
physical picture of the consecutive steps in the out-of-equilibrium dynamics associated with the
photo-switching of such molecular materials.
[1] E. Collet Ed. ”Time-resolved structural science”, special issue of Act. Cryst. A 66(2) (2010).
[2] N. Moisan et al., C.R. Chimie 11 (2008) 1235.
[3] M. Lorenc et al., Phys. Rev. Lett. 103 (2009), 028301.
[4] E. Collet et al., Z. Krystallogr. 223 (2008) 272.
Short-pulse laser induced transient structure formation and ablation
studied with time-resolved coherent XUV-scattering
K. Sokolowski-Tinten1, A. Barty2, S. Boutet3, U. Shymanovich1, M. Bogan2, S.
Marchesini4, S. Hau-Riege5, N. Stojanovic6, J. Bonse7, Y. Rosandi8, H. Urbassek8,
R. Tobey9, H. Ehrke9, A. Cavalleri2
,
9, S. Dsterer6, H. Redlin8, M. Frank5, S. Bajt2,
J. Schulz2, M. Seibert10, J. Hajdu10, R. Treusch6, H. Chapman2
1University of Duisburg-Essen, Lotharstr. 1, 47048 Duisburg, Germany, e-mail:
Klaus.Sokolowski@uni-due.de
2Centre for Free-Electron Laser Science, Hamburg, Germany.
3Stanford Linear Accelerator Laboratory, Menlo Park, CA, USA.
4Lawrence Berkeley National Laboratory, Berkeley, CA, USA.
5Lawrence Livermore National Laboratory, Livermore, CA, USA.
6HASYLAB, DESY, Hamburg, Germany.
7Bundesanstalt f¨ur Materialforschung und-pr¨ufung (BAM), Berlin, Germany.
8Technische Universit¨at Kaiserslautern, Kaiserlautern, Germany.
9University of Oxford, Oxford, United Kingdom.
10Uppsala University, Uppsala, Sweden.
Abstract
XUV- and X-ray free-electron-lasers (FEL) combine short wavelength, ultrashort pulse duration,
spatial coherence and high intensity. This unique combination of properties opens up new possibilities
to study the dynamics of non-reversible phenomena with ultrafast temporal and nano- to
atomic-scale spatial resolution. This contribution discusses results of time-resolved experiments
performed at the XUV-FEL FLASH (HASYLAB/Hamburg) aimed to investigate the nano-scale
structural dynamics of laser-irradiated materials. Thin films and fabricated nano- structures, deposited
on Si3N4-membranes, have been excited with ultrashort optical laser pulses. The dynamics
of the non-reversible structural evolution of the irradiated samples during laser- induced melting
and ablation has been studied in an optical pump XUV-probe configuration by means of singleshot
coherent scattering techniques.
In a first set of experiments we investigated the formation of laser induced periodic surface structures
(LIPSS) on the surface of thin Si-films. Time-resolved scattering using femtosecond XUVpulses
at 13.5 nm and 7 nm allowed us to directly follow the LIPSS evolution on an ultrafast
time-scale and with better than 40 nm spatial resolution. The observed scattering patterns show
almost quantitative agreement with theoretical predictions and reveal that the LIPSS start to form
already during the 12 ps pump pulse.
In the second set of measurements we studied picosecond and femtosecond laser induced ablation
and disintegration of fabricated nano-structures. Time-dependent auto-correlation functions were
obtained from the coherent diffraction patterns measured at various pump-probe time delays and
reveal the expansion dynamics of the irraditated samples. Under certain circumstances (e.g. adequate
sampling) it became also possible to reconstruct real-space images of the object as it evolves
over time [1].
[1] Barty et al., Nat. Phot. 2, 415 (2008).



Femtosecond Electron Diffraction: “Making the Molecular Movie”
Germ´an Sciaini
Institute for Optical Sciences and Departments of Chemistry and Physics, University of Toronto,
80 St George Street, Toronto, Ontario M5S 3H6, Canada.
email: gsciaini@lphys.chem.utoronto.ca
Abstract
Imagine one being able to follow chemical reactions and phase transformations with atomic spatial
and temporal resolution. This dreamed experiment has been entitled “Making the Molecular
Movie”(1). Recent advances in ultrafast time-resolved X-ray (2) and electron diffraction (1, 2)
techniques have shown that such a dream became real. Femtosecond Electron Diffraction (FED) is
very promising table-top technique that holds a great potential for the study of ultrafast structural
phenomena of matter. In FED a femtosecond laser pulse excites the sample and the photoinduced
structural changes are probed by an ultrashort electron pulse that scatters off the irradiated area
to generate a diffraction pattern downstream. By varying the time delay between the excitation
and the electron pulses atomic-level movies can be reconstructed after Fourier analysis. We have
fully characterized our electrons pulses employing transient optical gratings in order to scatter
electrons off via ponderomotive forces (3). With the development of our 4th generation electron
gun, we were able to reduce the electron pulse duration to 200 fs to provide enough time resolution
and brightness to study structural changes of matter occurring under strongly driven nonreversible
conditions. Nonthermal melting in Si caused by the promotion of 10% of its valence electrons to
the conduction band (4), bond hardening in warm dense Au (5) and strongly accelerated atomic
motions in Bi (6, 7) are some examples of the very different phenomena that were observed by
FED. During my talk I will present an overview of the ongoing efforts put forward the development
of ultrafast X-ray and electron diffraction techniques for the study of structural dynamics of
matter and show some recent result obtained by FED at University of Toronto.
(1) Dwyer J. R et al. Phil. Trans. R. Soc. A 364, 741 (2006).
(2) Chergui M. and Zewail A. H. ChemPhysChem. 10, 28 (2009).
(3) Hebeisen et al. Opt. Express 16, 3334 (2008).
(4) Harb M. et al. Phys. Rev Lett. 100, 155504 (2008).
(5) Ernstorfer R. et al. Science 323, 1033 (2009).
(6) Sciaini G. et al. Nature 458, 56 (2009).
(7) Cavalleri A. Nature 458 (news & views), 42 (2009).
The economical support provided by Canada Foundation for Innovation is acknowledged.
Studying Nanoscale Material Processes in the Dynamic Transmission
Electron Microscope (DTEM)
Thomas LaGrange1, Geoffrey H. Campbell1, Bryan W. Reed1, Nigel D.
Browning1
,
2
,
3, and Wayne E. King1
Condensed Matter and Materials Division, Physical and Life Science Directorate, Lawrence
Livermore National Laboratory, P.O. Box 808, Livermore, CA USA, email: lagrange2@llnl.gov
Department of Chemical Engineering and Materials Science, University of California-Davis,
Davis, CA
Department of Molecular and Cellular Biology, University of California-Davis, One Shields Ave,
Davis, Ca 95616. USA
Abstract
Often materials macroscopic properties and behavior under external stimuli can be described
through observation of its microstructural features and dynamical behavior. Materials models and
computer simulations that are used to predict material behavior in different environments, e.g.,
phase transformation kinetics under high pressure loading, typically require experimental data for
validation or interpretation of simulated quantities. However, most materials dynamics are extremely
rapid, making it difficult to capture transient, fine-scale features of the material process,
especially on the length and time scale relevant for most simulations. In effort to meet the need for
studying fast dynamics in material processes, we have constructed a nanosecond dynamic transmission
electron microscope (DTEM) at Lawrence Livermore National Laboratory to improve the
temporal resolution of in-situ TEM observations.
The DTEM consists of a modified JEOL 2000FX transmission electron microscope that provides
access for two pulsed laser beams. One laser drives the photocathode (which replaces the standard
thermionic cathode) to produce the brief electron pulse. The other strikes the sample, initiating
the process to be studied. A series of pump-probe experiments with varying time delays enable,
for example, the reconstruction of the typical sequence of events occurring during rapid phase
transformations. This presentation will discuss the core aspects of the DTEM instrument citing
specific examples for which the DTEM has been used to elucidate the kinetics of rapid martensitic
phase transformations, the morphologies rapid solidification and chemical reaction fronts and high
temperature crystallization processes in amorphous metallic films.
Work was performed under the auspices of the U.S. Department of Energy by the Lawrence Livermore
National Laboratory and supported by the Office of Science, Office of Basic Energy Sciences,
Division of Materials Sciences and Engineering, of the U.S. Department of Energy under contract
No. DE-AC52-07NA27344.



Ultrafast Electron Diffraction at Surfaces: From Non-Thermal Heat
Transport to Strongly Driven Phase Transitions
Michael Horn von Hoegen
Department of Physics, University of Duisburg-Essen, 47057 Duisburg, Germany
email: horn-von-hoegen@uni-due.de
Abstract
The multitude of possible processes that can occur at surfaces cover many orders of magnitude
in the time domain. While large scale growth and structure formation, for instance, happens on
a timescale of minutes and seconds, diffusion processes are already much faster. Energy transfer
processes take place on the femto- and picosecond timescale and are important for electron excitation
and relaxation, chemical reactions, phonon dynamics, nanoscale heat transport, or even phase
transitions.
In order to study such ultrafast processes at surfaces we have combined modern surface science
techniques with fs laser pulses in a pump probe scheme. We use a reflection high energy electron
diffraction (RHEED) setup with grazing incident electrons of 7 - 30 keV to ensure surface sensitivity
[1,2]. Utilizing the Debye Waller effect the cooling of vibrational excitations in monolayer
adsorbate systems or the nanoscale heat transport through a heterofilm interface is studied on the
lower ps-time scale [3-5]: the heat transport of ultrathin Bi(111) films on Si(001) is dominated by a
pronounced non-equilibrium distribution in the phonon system resulting in a much slower cooling
rate.
In order to demonstrate the huge potential of this technique I will shortly present examples for
the dynamics of strongly driven structural phase transitions at surfaces upon excitation with a fslaser
pulse: the famous order-disorder phase transition from c(4x2) to (2x1) on Si(001) at 200 K
and the Indium induced Peierls-like transition from c(8x2) to (4x1) on Si(111) at 80 K which is
additionally accompanied by the formation of a charge density wave [6].
[1] A. Janzen, B. Krenzer, P. Zhou, D. von der Linde, and M. Horn-von Hoegen, Surf. Sci. 600, 4094 (2006)
[2] A. Janzen, B. Krenzer, O. Heinz, P. Zhou, D. Thien, A. Hanisch, F.-J. Meyer zu Heringdorf, D. von der Linde, and
M. Horn-von Hoegen, Rev. Sci. Inst. 78, 013906 (2007)
[3] B. Krenzer, A. Janzen, P. Zhou, D. von der Linde, and M. Horn-von Hoegen, New J. Phys. 8, 190 (2006)
[4] A. Hanisch, B. Krenzer, T. Pelka, S. Mllenbeck, and M. Horn-von Hoegen, Phys. Rev. B 77, 125410 (2008)
[5] B. Krenzer, A. Hanisch-Blicharski, P. Schneider, Th. Payer, S. Mllenbeck, O. Osmani, M. Kammler, R. Meyer and
M. Horn-von Hoegen, Phys. Rev. B 80, 024307 (2009)
[6] S. Mllenbeck, A. Hanisch-Blicharski, P. Schneider, M. Ligges, P. Zhou, M. Kammler, B. Krenzer, and M. Hornvon
Hoegen, MRS-Proceedings (submitted)
Femtosecond Molecular Photocrystallography
Hubert Jean-Ruel, Cheng Lu, Ryan Cooney, Meng Gao, Germn Sciaini, Gustavo
Moriena, R. J. Dwayne Miller
Department of Physics and Chemistry, University of Toronto, Canada,
email: hubert@lphys.chem.utoronto.ca, clu@lphys.chem.utoronto.ca
Abstract
Diarylethenes are a class of photochromic compounds which undergo well documented conformational
changes in both the solution and crystal phase [1]. The photoreversible isomerization
involves ring-closing and -opening of the molecular system, which leads to distinct absorptive
features in the visible and UV spectral regions respectively. Of particular interest is the recent
development of diarylethene derivatives that exhibit not only pronounced thermal stability of the
open and closed-ring isomers, but also a high degree of fatigue resistance in the crystal phase suggesting
the potential for optical switching and memory applications. Here we present preliminary
results of a femtosecond electron diffraction (FED) study on such a derivative.
FED will provide a direct observation of the structural dynamics involved in the conformational
changes of diarylethene with femtosecond time resolution and atomic level details [2]. Among
other studies, FED has now been successfully used to study ultrafast structural dynamics in the
order-to-disorder phase transition of strongly driven melting in gold [3], and the electronically
driven melting of silicon [4]. In FED, an ultrashort laser pulse initiates the reaction in the sample
under study and an electron bunch probes its structure via diffraction; by varying the time delay
between the laser and electron pulses, the recorded diffraction patterns temporally resolves changes
in the molecular structure. In the case of diarylethene, a third beam is required to bring the sample
back to its initial state before the next pump-probe event. To complement the electron diffraction
study, an optical pump-probe absorption measurement will first be performed to characterize the
required experimental parameters and insure complete reversion to the initial conditions.
[1] M. Irie, Diarylethenes for Memories and Switches, Chem. Rev. 100, 1685 (2000).
[2] J. R. Dwyer, C. T. Hebeisen, R. Ernstorfer, M. Harb, V. B. Deyirmenjian, R. E. Jordan, and R. J. D. Miller,
Femtosecond electron diffraction: ‘making the molecular movie’. Phil. Trans. Roy. Soc. A 364, 741778 (2006).
[3] Ralph Ernstorfer, Maher Harb, Christoph T. Hebeisen, Germn Sciaini, Thibault Dartigalongue, R. J. Dwayne
Miller, The formation of warm dense matter: experimental evidence for electronic bond hardening in gold, Science
323, 1033-1037 (2009).
[4] M. Harb, R. Ernstorfer, C.T. Hebeisen, G. Sciaini, W. Peng, T. Dartigalongue, M.A. Eriksson, M.G. Lagally,
S.G. Kruglik, and R.J.D. Miller, “Electronically Driven Structure Changes of Si Captured by Femtosecond Electron
Diffraction”, Phys. Rev. Lett, 100, 155504/1-4 (2008).






First Results of Coherent Diffraction Experiments at LCLS
Henry N. Chapman
Center for Free-Electron Laser Science, DESY, Notkestrasse 85, Hamburg, Germany
University of Hamburg, Hamburg, Germany email: henry.chapman@desy.de
Abstract
The ultrafast pulses from X-ray free-electron lasers may enable the determination of structures of
proteins that cannot be crystallized. The specimen would be completely destroyed by the pulse, but
that destruction will ideally only happen after the termination of the pulse. In order to address the
many challenges that we face in attempting molecular diffraction, we have carried out experiments
in coherent diffraction from protein nanocrystals at the Linac Coherent Light Source (LCLS) at
SLAC. The periodicity of these objects gives us much higher scattering signals in order to determine
the effects of pulse duration and fluence on the high-resolution structure of single objects.
The crystals are filtered to sizes less than 2 micron, and are delivered to the pulsed X-ray beam in
a liquid jet. Diffraction patterns are recorded at the LCLS repetition rate with pnCCD detectors.
Preliminary results will be presented on our first LCLS experiments. This work was carried out as
part of a collaboration, for which Henry Chapman is the spokesperson. The collaboration consists
of CFEL DESY, Arizona State University, SLAC, Uppsala University, LLNL, The University of
Melbourne, LBNL, the Max Planck Institute for Medical Research, and the Max Planck Advanced
Study Group (ASG) at the CFEL. The names and addresses of all do not fit on one page. The
experiments were carried out using the CAMP apparatus, which was designed and built by the
Max Planck ASG at CFEL. The LCLS is operated by Stanford University on behalf of the U.S.
Department of Energy, Office of Basic Energy Sciences.
Fast protein nanocrystallography
J.C.H.Spence, P. Fromme, B. Doak, K. Schmidt, U. Weierstall, M. Hunter, R.
Kirian, M. Hunter, X. Wang, H. Chapman*, T. White*, J. Holton**
Dept. of Physics, Arizona State University, Tempe, Az. USA 85287, spence@asu.edu
CFELS, DESY/U.Hamburg, Notkestrasse 85, 22607 Hamburg, Germany
henry.chapman@desy.de
*ALS, Lawrence Berekley Laboratory, Berkeley , Ca. USA, 94720 JMHolton@lbl.gov
Abstract
The invention of the hard X-ray laser has opened the way for a new form of protein microcrystallography
under an entirely new regime of radiation-damage conditions (1, 2). When combined with
pump-probe methods, this “diffract-and-destroy”mode, in which an X-ray pulse terminates before
damage begins, promises dramatic advances in the study of protein dynamics, and of structures
which have never been seen at high resolution because of their radiation sensitivity. In this talk our
recent diffraction data obtained at LCLS from individual sub-micron crystallites of Photosystem
1 membrane protein will be discussed, where femtosecond pulses (with repetition rate of 30 Hz)
were used at 2 kV with a 3 micron X-ray beam diameter to obtain tens of thousands of patterns
from individual crystallites fired in single-file across the beam by a protein-beam injector. Previous
work at Flash (2), and simulations (3), have indicated the difficulties in phasing and orientation
determination for single non-periodic bioparticles (such as viruses or single macromolecules) due
to the very low counts at high angle (much less than unity). The Bragg amplification of coherent
scattering in “stills”(snap-shot diffraction) from nanocrystals increases counts greatly, providing
high resolution information, but requiring a new form of data analysis. Additionally, since Miller
indices are coordinates in reciprocal space, the ability to index these stills solves the molecular
orientation problem.
This talk will focus mainly on data analysis methods (4), in which, following indexing, we achieve
a Monte-Carlo integration over particle size and orientation by adding together all “spots”(partials)
with the same index from different crystals. Our crystals are roughly sorted by size, but are not
identical particles. This makes whole-particle phasing a challenging excercise (5) since it requires
sorting by both size and orientation - if that can be done, by selecting phases only on lattice points,
a new method of phasing would be possible for protein crystallography. The method of aperture
photometry (as used in Astronomy) is used to integrate over the crystal shape-transform on each
pattern. Simulations showing the convergence of these orientation and size summations to yield
wanted structure factors will be discussed. These address the question of how many patterns are
needed for a required accuracy, with a given photon count per pulse. Details of the indexing
method, of the protein-beam injector (6,7), of hit rates, and membrane protein hydration will also
be discussed.
(1) Howells, M. et al J. Elec. Spectr. Rel Phenom. 170, 4 (2009). (2) Chapman, H. Nature Materials 8, 299 (2009) (3)
Starodub, D. et al. J. Synch. Res. 15, 62 (2008) (4) Kirian et al. Optics Express. Submitted (2010). (5) Fung et al
Nature Physics 5, 64 (2009). (6) DePonte et al Micron 40, 507 (2009). (7) Shapiro et al J. Synch Res. 15, 593. (2009).
Work supported by DOE award DE-SC0002141.
Linear and nonlinear imaging with XFEL: results from ab-initio
computations
Andrea Fratalocchi and Giancarlo Ruocco
Dept. of Physics, Sapienza University, P.le A. Moro 2, 00185 Rome, Italy.
email: andrea.fratalocchi@uniroma1.it, giancarlo.ruocco@uniroma1.it
Abstract
The ultimate frontier of single molecule imaging with XFEL sources is currently hampered by
several challenging questions concerning sample damage, time-gating imaging and the role of
nonlinearity. By employing an original ab-initio approach, as well as exceptional resources of
parallel computing, we provide a decisive answer to them. Our model, directly stemming from
the quantum-mechanical equations governing the dynamics of atoms subjected to electromagnetic
elds, try to denitively settle down the theoretical grounds for present and future ab-initio researches
on XFEL science. More specically, our approach combines classical molecular dynamics, nonlinear
Scrh¨odinger and Maxwell’s equations into a single efcient parallel environment, which features
original second order propagators designed with state-of-the-art methods and algorithms. By analyzing
a selection of atoms and molecules, we address the problem of sample radiation damage,
thus predicting a large sample photoionization in a few of femtoseconds (with external electron
emission in hundreds of attoseconds). We then deeply analyze the the scattered far eld, highlighting
the role of nonlinearity and anticipating the possibility to spread out the XFEL application
domain to nonlinear coherent imaging. We nally investigate the coherent imaging capabilities of
XFEL sources, collecting snapshots of integrated far field (as retrieved by a standard camera),
reporting ab-initio molecular images and discussing image blurring versus XFEL pulse length.
Single shot soft X-ray holography using extended references
David Gauthier, Xunyou Ge, Willem Boutu, Xiaochi Liu, Bertrand Carr, Hamed
Merdji
SPAM, CEA Saclay, 91191 Gif sur Yvette, France, email: david.gauthier@cea.fr
hamed.merdji@cea.fr
Manuel Guizar-Sicairos and James R. Fienup
The Institute of Optics, University of Rochester, Rochester, N.Y. 14627, US
Abstract
X-ray lensless imaging is demonstrating a very high potential in performing images of isolated
nanoscale objects with unprecedented space and time resolution. Active research is actually pursued
to push the capability of this technique using coherent X-ray sources recently available. In
this context, we present a generalization of Fourier transform holography. A major advance shown
here is the use of extended holographic reference to perform soft X-ray nanoscale imaging. The
direct reconstruction process of the object is simple and robust. Moreover, the design of the holographic
reference is easy to implement. We demonstrate here single shot imaging with table top
soft X-ray source based on the high harmonics generation process. A spatial resolution of 110 nm
is obtain with an integration time resolution of 20 fs. Using harder X-rays available at femtosecond
X-ray free electron lasers, extended holographic references can be used to capture dynamical
processes at a sub-nanometer scale and in real time.
Ultra-fast, A¨ ngstro¨m Scale Structure Determination of Molecules via
Photoelectron Holography
Faton Krasniqi1
,
, Bennaceur Najjari2, Alexander Voitkiv2, Sascha Epp1, Daniel
Rolles1, Artem Rudenko1, Lutz Foucar1, Yin-peng Zhong1, Benedikt Rudek1,
Benjamin Erk1, Robert Hartmann1
,
3, Robert Moshammer2, Klaus-Dieter Schr¨oter2,
Simone Techert1
,
4, Lothar Str¨uder1
,
5, Ilme Schlichting1
,
5, Joachim Ullrich1
,
2
1Max Planck Advanced Study Group at CFEL, 22761 Hamburg, Germany
2Max-Planck-Institut f¨ur Kernphysik, 69117 Heidelberg, Germany
3Max-Planck Halbleiterlabor, 81739 M¨unchen, Germany
4Max Planck-Institut for Biophysical Chemistry, 37077 G¨ottingen, Germany
5Max-Planck-Institut f¨ur medizinische Forschung, 69120 Heidelberg, Germany
email: Faton.Krasniqi@asg.mpg.de
Abstract
We examine a new scheme that enables us to realize a molecular movie with femtosecond time
and A° ngstro¨m spatial resolution for small and medium sized molecules based on the (i) upcoming
brilliant X-ray Free Electron Laser (FEL) sources, (ii) novel energy and angular dispersive,
large-area electron imagers and (iii) the photoelectron holography. Here, photoelectrons produced
via core-level excitation and launched at specific and well-defined atomic sites will scatter on
“their way out”on the multi-atomic potential of the parent molecule generating a hologram on the
detector that encodes the molecular structure at the instant of photoionization. Due to the large
photo-absorption and electron elastic scattering cross sections the method extends X-ray diffraction
based, time-dependent structure investigations envisioned at FELs to new classes of samples
that are not accessible by any other method. Among them are dilute samples in the gas phase such
as aligned, oriented or conformer selected molecules, ultra- cold ensembles and/or molecular or
cluster objects containing mainly light atoms that do not scatter X-rays efficiently.
Explosions of Xe-Clusters in Intense Soft-X-Ray and X-Ray Pulses
H. Thomas1
,
, K. Hoffmann1, N. Kandadai1, A. Helal1, J. Keto1, T. Ditmire1, C.
Bostedt2 , T. M¨oller3, U. Saalmann4, C. Gnotke4, J.M. Rost4, B. Iwan5, N.
Timneanu5, J. Andreasson5, S. Schorb3, T. Gorkhover3, D. Rupp3, M. Adolph3, G.
Doumy6, L.F. DiMauro6, J. Bozek2
1Fusion Research Center, University of Texas, Austin, TX 78712 USA
2LCLS, Stanford Linear Accelerator Center, Menlo Park, CA 94025, USA
3Institut fr Optik und Atomare Physik, Technische Universitt Berlin, 10623 Berlin, Germany
4Max-Planck-Institut fr Physik komplexer Systeme, 01187 Dresden, Germany
5Uppsala University, Uppsala, Sweden and Stanford University, Menlo Park, CA 94025, USA
6Department of Physics, The Ohio State University, Columbus, Ohio 43210, USA
corresponding author email: drmetal@web.de
Abstract
Intense femtosecond x-ray pulses from free electron lasers open the door for novel experiments
in a wide spectrum of sciences ranging from atomic, molecular and plasma physics over chemical
and surface dynamics to diffraction imaging of non-periodic objects and biological samples. The
interaction of intense x-ray pulses with matter is so far only scarcely investigated, even though its
understanding is a prerequisite for virtually all future experiments in this field. Clusters, bridging
the gap between the atom and bulk solid, are ideal to investigate the light matter interaction. They
exhibit the density of bulk solids but due to their finite size hidden energy dissipation into the
surrounding media is virtually absent.
The presentation will show results of the interaction of Xe-Clusters consisting of up to <N>
10,000 atoms with the FLASH radiation at a photon energy of 90 eV at a pulse length of 10 fs
resulting in a maximum intensity of 8x1014 W/cm2 in the focus. At this photon energy one photon
can ionize the 4d-innershell electrons of xenon. The absorption of 90 eV-photons is rather complex
for xenon including multi-photon processes and auger effects.
Simulating the ion kinetic energies in an electrostatic model suggests that highly charged ions
explode off the surface due to Coulomb repulsion while the inner core expands in a hydro- dynamic
expansion [1]. The current results yield evidence for efficient ionization of the clusters in addition
to direct multistep photoemission [2,3]. Further a model for the induced multi- electron dynamics
can be shown which reveals that fast electrons originate from an equilibrated electron plasma of
supra-atomic density [3]. The plasma has sufficiently high temperature to support fast electrons
without traditional laser plasma heating, which is not operative at 90 eV. This results will be compared
to results of the very recent experiments at LCLS on xe-clusters at a photon energy up to 2
keV and similar pulse lengths. In the experiments at FLASH and LCLS ion- and electon-spectra
were recorded using the time-of-flight-technique.
[1]Shell explosion and core expansion of xenon clusters irradiated with intense femtosecond soft x-ray pulses, H.
Thomas et al, J. Phys. B: At. Mol. Opt. Phys. 42, 134018 (2009)
[2]Fast electrons from multi-electron dynamics in xenon clusters induced by 90 eV FLASH pulses, H.Thomas et al,
submitted to PRL
[3]Multistep Ionization of Argon Clusters in Intense Femtosecond Extreme Ultraviolet Pulses, Bostedt et al, PRL 100,
133401 (2008)



Watching proteins function in real time via 150-ps time-resolved
X-ray diffraction and solution scattering
Philip Annrud, Friedrich Schotte, Hyun Sun Cho, Naranbaatar Dashdorj, and
William Royer*
National Institutes of Health, Laboratory of Chemical Physics/NIDDK, 5 Memorial Dr.,
Bethesda, MD 20892-0520 USA email: annrud@nih.gov, schotte@nih.gov,
HyunSunC@intra.niddk.nih.gov, ndn@nih.gov
*Dept. of Biochemistry and Molecular Pharmacology, University of Massachusetts Medical
School, Worcester, MA 01605 USA email: William.Royer@umassmed.edu
Abstract
To generate a deeper understanding into the relations between protein structure, dynamics, and
function, we have developed X-ray methods capable of probing changes in protein structure on
time scales as short as 100 ps. This infrastructure was rst developed on the ID09B time- resolved
X-ray beamline at the European Synchrotron and Radiation Facility, and more recently on the
ID14B BioCARS beamline at the Advanced Photon Source. In these studies, a picosecond laser
pulse rst photoexcites a protein, then a suitably delayed picosecond X-ray pulse passes through
the laser-illuminated volume of the sample and the scattered X-rays are imaged on a 2D detector.
When the sample is a protein crystal, this pump-probe approach recovers time- resolved diffraction
snapshots whose corresponding electron density maps can be stitched together into movies
that unveil correlated protein motions at near atomic resolution. When the sample is a protein
solution, we recover time-resolved small- and wide-angle X-ray scattering patterns that are sensitive
to changes in the size, shape, and structure of the protein. Scattering studies of proteins in
solution unveil structural dynamics without the constraints imposed by crystal contacts; thus, these
scattering “fingerprints”at low spatial resolution complement results obtained from high-resolution
diffraction studies. Studies of structural dynamics in wild-type and mutant scapharca hemoglobin,
a homodimer that exhibits cooperative ligand binding, unveil non-exponential tertiary relaxation
followed by a quaternary R to T structural change that alters the binding afnity of its two ligand
binding sites. The structural dynamics characterized by X- ray scattering are highly correlated with
spectral changes observed via time-resolved optical spectroscopy, thereby allowing us to make a
structural assignment for the spectroscopic states. These studies are leading to a comprehensive
characterization of the structural dynamics that contribute to the cooperative binding of ligands in
this allosteric protein. This research was supported in part by the Intramural Research Program of
the NIH, NIDDK.
But my crystals aren’t light-sensitive....help!
Keith Moffat
Institute for Biophysical Dynamics, Center for Advanced Radiation Sources, University of
Chicago
email: moffat@cars.uchicago.edu
Abstract
Studying the structure-based, ultrafast dynamics of biological systems by e.g. Laue crystallography
requires a means of initiating the reaction in the crystal rapidly, smoothly and with high
efficiency. In practice, this has meant using a brief laser pulse in the fs to ns range, and restricting
the crystals under study to those of naturally light-sensitive systems such as photoreceptors
or the CO-complexes of heme proteins. This substantially restricts the applicability of ultrafast
time-resolved crystallography, a fact not lost on peer reviewers.
There are two possible rejoinders: find other means of ultrafast rapid initiation e.g. temperature
jump or dielectric relaxation, particularly challenging if time scales less than s are to be probed; or,
confer light sensitivity on otherwise light-inert systems. In tackling the latter, we base our approach
on key features of natural signaling photoreceptors: they are modular in architecture, containing
several compactly-folded domains; and different functions are located in different domains. For
example, they contain one or more sensor or input domains that respond to a physical signal e.g.
absorption of light, or a chemical signal e.g. binding of a small molecule, and an effector or output
domain whose activity e.g. catalytic, DNA binding is influenced by the signal. Thus information
is transferred from the sensor domain to the effector domain. Further, the sensor domain(s) is
usually located near the N-terminus of the effector domain, and is covalently joined to it by a
linker that may be -helical or a coiled coil. One class of sensor domain e.g. a blue-light-sensing
LOV domain is found joined to many different types of effector domains. The last argues against
structure-specific interaction between the sensor and effector domains.
We exploit these natural principles to confer sensitivity to light on the DNA-binding trp repressor
(Strickland et al., PNAS 105, 10709-14 (2008)), a histidine kinase (Moeglich et al., J.Mol.Biol.
385, 1433-44 (2009)) and kinases with more than one sensor domain (unpublished). The last raises
the additional complexity of interaction between signals: such molecules can act as logic elements
whose output depends on more than one input.
This new area has been labelled “optogenetics”(see Miesenbock, Science 326, 395-9 (2009)),
the genetic encoding of natural e.g. channelrhodopsin and artificial, designed light-sensitive systems.
Molecular Structural Dynamics Visualized by Pump-Probe X-ray
Liquidography and Crystallography
J. Kim and H. Ihee
Center for Time-Resolved Diffraction, Department of Chemistry, Graduate School of
Nanoscience & Technology (WCU), Daejeon 305-701, South Korea email: ihee57@gmail.com
Abstract
The principle, experimental technique, data analysis, and applications of time-resolved X-ray
diffraction and scattering to study spatiotemporal reaction dynamics of proteins in single crystals
and solutions will be presented. X-ray crystallography, the major structural tool to determine
3D structures of proteins, can be extended to time-resolved X-ray crystallography with a laserexcitation
and X-ray-probe scheme, and all the atomic positions in a protein can be tracked during
their biological function. However time-resolved Crystallography has been limited to a few model
systems with reversible photocycles due to the stringent prerequisites such as highly- ordered and
radiation-resistant single crystals and crystal packing constraints might hinder biologically relevant
motions. These problems can be overcome by applying time-resolved X-ray diffraction directly
to protein solutions rather than protein single crystals. To emphasize that structural information
can be obtained from the liquid phase, this time-resolved X-ray solution scattering technique is
named time-resolved X-ray liquidography (TRXL) in analogy to time- resolved X-ray crystallography
where the structural information of reaction intermediates is obtained from the crystalline
phase. Using ultrashort optical pulses to trigger a reaction in solution and detecting time-resolved
X-ray diffraction signals to interrogate the molecular structural changes, TRXL can provide direct
structural information generally difficult to extract from ultrafast optical spectroscopy such
as the temporal progression of bond lengths and angles of all molecular species including shortlived
intermediates over a wide range of times, from picoseconds to milliseconds. TRXL elegantly
complements ultrafast optical spectroscopy because diffraction signals are sensitive to all chemical
species simultaneously and the diffraction signal from each chemical species can be quantitatively
calculated from its three- dimensional atomic coordinates and compared with experimental TRXL
data. Application examples on spatiotemporal kinetics and structural dynamics of a halomethane,
a triatomic molecule, haloethanes, and an organometallic catalyst are presented. In addition, we
demonstrate tracking of proteins structural changes in solution using TRXL. TRXL permitted us
to investigate the tertiary/quaternary conformational change of human hemoglobin in nearly physiological
conditions triggered by laser induced ligand photolysis. Data on optically induced tertiary
relaxations of myoglobin and refolding of cytochrome c are also reported to illustrate the
wide applicability of the technique. By providing insights into the structural dynamics of proteins
functioning in their natural environment, TRXL complements and extends results obtained with
time-resolved spectroscopy and X-ray crystallography.
Molecular Snapshot in Solar Energy Conversion Processes Taken by
Ultrafast X-rays
Lin X. Chen, Jenny Lockard, Andrew B. Stickrath, Xiaoyi Zhang, Klaus
Attenkofer, Guy Jennings
Chemical Sciences and Engineering Division and X-ray Science Division, Argonne National
Laboratory, Argonne, IL 60439
Department of Chemistry, Northwestern University Evanston, IL 60208
email: lchen@anl.gov
Abstract
A decade of studies on excited state structures of transition metal complexes for solar energy conversion
using laser and x-ray transient absorption spectroscopy will be briefly reviewed including
the details for the excited state dynamics, structural diversity in solution and hot vibrational states.
We will discuss three examples on 1) metalloporphyrins excited state structure and photoinduced
ligation/deligation, 2) interplays of structure and dynamics of MLCT excited state transition metal
complexes for photoinduced charge separation and electron transfer, and 3) excited state transition
metal complexes at interfaces of hybrid material for solar electricity generation/catalysis. The current
advances and limitations in resolving excited state structures during photochemical reactions
will be presented. New needs in theoretical computation and modeling will be addressed for these
studies to exert the full potentials of resolving otherwise elusive excited state structures. The potential
and prospective in excited state structural dynamics studies using new light sources, such as
XFEL will be discussed.
Towards Femtosecond X-Ray Spectroscopies
Christian Bressler, Andreas Galler, Wojciech Gawelda, Majed Chergui†, Chris
Milne†, Van-Thai Pham†, Renske van der Veen†, Steven Johnson*, Rafael Abela*
European XFEL GmbH, Albert-Einstein Ring 19, D-22607 Hamburg, Germany
email: christian.bressler@xfel.eu, andreas.galler@xfel.eu, wojciech.gawelda@xfel.eu
† EPF Lausanne, ISIC Bt. CH, CH-1015 Lausanne, Switzerland email: majed.chergui@epfl.ch,
chris.milne@psi.ch, vanthai.pham@epfl.ch, renske.vanderveen@psi.ch
*Paul-Scherrer Institut, CH-5232 Villigen-PSI, Switzerland email: steven.johnson@psi.ch,
rafael.abela@psi.ch
Abstract
Femtosecond X-Ray Science is an emerging field aiming to deliver a detailed understanding of
the ultrafast elementary steps in complex processes involving changes in nuclear, electronic and
spin states. Such processes are vital ingredients in chemistry and biology, but also in technological
applications, including efficient charge transport in solar energy converters and ultrafast switchable
molecular magnets.
This talk will present results obtained on a prototype spin transition phenomenon in aqueous
Fe(bpy)3. Optical techniques explore the ultrafast changes in the valence states, but ultrafast xray
spectroscopies reveal the underlying nuclear and electronic changes during this spin transition
process. While picosecond resolved XANES and EXAFS are exploited to understand the
altered geometrical structure of the molecule after the spin transition is complete, Femtosecond
XANES is able to monitor the evolution of this process in real-time. Finally, a recent experiment
exploiting time-resolved XES of the K emission with picosecond resolution established a direct
measurement of the short-lived (0.7 ns) high-spin state. Combing these spectroscopic tools with
the intense intensity and femtosecond time resolution at x-ray free electron lasers will allow us
to deliver a motion picture of the interplay between the nuclear, electronic and spin degrees of
freedom during complex chemical reactions, and an outlook towards exploiting XFEL machines
currently in operation or under construction will be given.
Photo-Induced Spin-State Conversion in Transition Metal
Complexes Probed via Ultrafast Soft X-ray Spectroscopy
Nils Huse, Hana Cho, Tae Kyu Kim, Lindsey Jamula, James K. McCusker, Frank
M. F. de Groot, and Robert W. Schoenlein
Chemical Sciences Division, Lawrence Berkeley National Laboratory, Berkeley,
CA 94720, USA, nhuse@lbl.gov & rwschoenlein@lbl.gov
Department of Chemistry, Pusan National University, Geumjeong-gu, Busan
609-735, Korea, hcho@lbl.gov & tkkim@pusan.ac.kr
Department of Chemistry, Michigan State University, East Lansing, MI 48824,
USA, jamulali@msu.edu & jkm@chemistry.msu.edu
Department of Chemistry, Utrecht University, 3584 CA Utrecht, Netherlands,
F.M.F.deGroot@uu.nl
Abstract
A precise understanding of the transient valence charge distribution in solvated tran- sition metal
complexes is of great scientic interest due to their important role in chemical reaction and biological
processes. By exploiting the capability of time- resolved L-edge spectroscopy to deliver
unique information about transient valence electronic states in transition metal compounds, we
investigated the photo-induced spin crossover reaction in a solvated iron (II) model complex via
femtosecond soft x- ray spectroscopy. Our recent experimental results in combination with charge
trans- fer multiplet calculations relate to important aspects of general chemistry and reveal a wealth
of information on the changes of the electronic valence charge distributions and the role of ligand
-back-bonding in different molecular structures. Upon photo-excitation to the singlet metal-toligand
charge transfer state, the in- tricate coupling of nuclear and electronic degrees of freedom
results in an ultra- fast singlet-to-quintet spin state conversion within 200fs mediated by large
structural and electronic changes. The transient valence electronic structure of the metastable highspin
state features strongly altered orbital hybridization and delocalization, de- creased ligand-eld
splitting, and strongly suppressed -back-bonding, increasing the ionic character of the central
transition metal atom in the dilated ligand cage.
Observation of multiphoton processes in the x-ray regime: First
experiments at LCLS
Linda Young1, Elliot Kanter1, Bertold Krssig1, Yuelin Li1, Anne Marie March1,
Stephen Pratt1, Robin Santra1, Stephen Southworth1, John Bozek2, Christoph
Bostedt2, Marc Messerschmidt2, Lou DiMauro3, Gilles Doumy3, Chris Roedig3,
Nora Berrah4, Matthias Hoener4, Li Fang4, Phil Bucksbaum5, David Reis5, James
Cryan5, Mike Glownin5
1Argonne National Laboratory, Argonne, IL 60439 USA, email: young@anl.gov
2LCLS, SLAC National Accelerator Laboratory, Menlo Park, CA 94025 USA
3Ohio State University, Columbus, OH 43210 USA
4Western Michigan University, Kalamazoo, MI 49008 USA
5PULSE, SLAC National Accelerator Laboratory, Menlo Park, CA 94025 USA
Abstract
The worlds first x-ray free electron laser, the LCLS at SLAC National Accelerator Laboratory, provides
access to ultraintense x-ray radiation for the first time. Understanding the atomic response
to such radiation is of fundamental importance for planning any future work where matter will be
exposed to ultraintense x-ray beams. Therefore, we investigated the most basic aspects of intense
x-ray/matter interactions by observing photoionization of the prototypical neon atom, an atom that
exhibits rich physics over the initial energy range of 800-2000 eV. At high photon energy, one
expects sequential single photon absorption to dominate the XFEL-atom interaction with, e.g., sixphoton
absorption leading to fully-stripped neon. Such processes depend only on the fluence of
the radiation. However, with ultraintense x-ray radiation (focused intensities of 101
8 W/cm2)
one can photoinduce sequential K-shell absorption prior to the intraatomic Auger decay (2.4 fs) to
create exotic hollow atom states with high probability. By contrast, at low x-ray intensity hollow
atoms are only formed indirectly via rare one-photon, two- electron processes that require electron
correlation. The versatility of the LCLS allowed us to investigate the nature of photoelectric x-ray
absorption processes over a wide range of photon energy, pulse energy and pulse duration. We
tracked the evolution of the neon atom using one ion and five electron spectrometers to view the
interaction region. The electron energy and angular distributions reveal details of the photoabsorption
mechanism. We observed an intensity-induced transparency, photoproduction of hollow neon,
and considerable valence ionization. The observations are qualitatively explained by sequential
multiphoton processes. A comparison with a simple rate equation model demonstrates the need to
include shake and double Auger processes for quantitative agreement. The simplicity of the neon
target provides useful diagnostics of the XFEL beam.
Results of the CAMP Instrument Commissioning at LCLS
Daniel Rolles1
,
, Artem Rudenko1, Sascha Epp1, Lutz Foucar1, Benedikt Rudek1,
Benjamin Erk1, Carlo Schmidt1, Andr´e H¨omke1, Faton Krasniqi1, Robert
Hartmann1
,
2, Nils Kimmel2, Christian Reich2, G¨unther Hauser2, Daniel
Pietschner2, Peter Holl2, Lothar Str¨uder1
,
2, Hubert Gorke3, Helmut Hirsemann4,
Guillaume Potdevin4, Tim Erke4, Jan-Henrik Mayer4, Michel Matysek4, Sebastian
Schorb5, Daniela Rupp5, Marcus Adolph5, Tais Gorkhover5, Marc Simon6, Loic
Journel6, Kioyshi Ueda7, Kiyonobo Nagaya8, Nora Berrah9, Christoph Bostedt10,
John Bozek10, Marc Messerschmidt10, Joachim Schulz11, Lars Gumprecht11,
Andrew Aquila11, Nicola Coppola11, Frank Filsinger12, Nina Rohringer13, Kai-Uwe
Khnel14, Christian Kaiser41, Ilme Schlichting1
,
15, Joachim Ullrich1
,
14
1Max Planck Advanced Study Group at CFEL, 22761 Hamburg, Germany
2Max Planck Halbleiterlabor, 81739 Mnchen, Germany
3FZ J¨ulich, 52428 J¨ulich, Germany
4Deutsches Elektronen Synchrotron, 22607 Hamburg, Germany
5Technische Universit¨at Berlin, 10623 Berlin, Germany
6Laboratoire de Chimie Physique-Mati`ere et Rayonnement, 75231 Paris, France
7Tohoku University, Sendai 980-8577, Japan
8Kyoto University, Kyoto 606-8501, Japan
9Western Michigan University, Kalamazoo, MI 49008, USA
10LCLS, Menlo Park, CA 94015, USA
11CFEL, Deutsches Elektronen Synchrotron, 22607 Hamburg, Germany
12Fritz-Haber-Institut der MPG, 14195 Berlin, Germany
13Lawrence Livermore National Laboratory, 94551 Livermore, USA
14Max-Planck-Institut f¨ur Kernphysik, 69117 Heidelberg, Germany
15Max-Planck-Institut f¨ur medizinische Forschung, 69120 Heidelberg, Germany
email: *Daniel.Rolles@asg.mpg.de
Abstract
The CFEL-ASG MultiPurpose (CAMP) instrument designed and constructed by the Max Planck
Advanced Study Group at CFEL has recently been commissioned during the first user run at LCLS
in November/December 2009. The general layout and capabilities of the CAMP instrument will
be reviewed and first results of the successful instrument commissioning will be reported.
Structural Dynamics with Bound Electrons: Isomeric and
Conformeric Motions of Hot Molecules
Peter M. Weber, Michael P. Minitti, Sanghamitra Deb, Joseph Bush
Department of Chemistry, Brown University, Providence, R.I. 02912, USA,
email: peter weber@brown.edu
Abstract
The binding energy of a Rydberg electron that orbits a positively charged ion core is a uniquely
sensitive probe of the structure of the underlying molecular ion core. The structure sensitivity can
be traced to the very same phase shifts that give rise to electron diffraction patterns. When the
electron binding energy is measured in an ionization transition, the resulting spectrum is free of
vibrational progressions: the spectrum is purely electronic in character. Showing only the usual
orbital and magnetic angular momentum states of the Rydberg electrons, the complexity of the
spectra does not scale with the size of the molecular system. Moreover, since the Rydberg orbits
are large compared to the dimensions of usual molecules, the structure sensitivity extents to the
entire molecule. The global structure sensitivity coupled with the insensitivity towards vibrations
makes Rydberg electron binding energy spectra ideally suited to observe structural dynamics, including
transformations between isomeric and conformeric forms of highly excited molecules. The
drawback of the technique is that unlike a diffraction pattern, the data cannot easily be inverted to
obtain molecular structures. This talk outlines the essential features of the technique and illustrates
it with examples from a series of investigations on tertiary amines.
All tertiary amines exhibit a very rapid structural change that can be traced to the initial planarization
of the amine bond upon electronic excitation. In tripropylamine and trimethylamine, little
further signature of structural dynamics is found. Triethylamine, however, shows a rich timedependent
spectrum. The ethyl groups of triethylamine, rotating about the C-N single bond, create
a complex energy landscape that serves as a model system for conformational dynamics with
highly coupled degrees of freedom. Electronic excitation to a 3p or 3s Rydberg level leads to a
high-energy Rydberg state conformer that rapidly relaxes to other, more stable conformeric forms
with a 232 fs time constant. A new equilibrium is established on a sub- picosecond time scale.
Even so, the molecules retain a large dispersion of molecular structures about the equilibrium position.
For the close-lying minima in the energy landscape, the variation of the Rydberg electron
binding energy is the determining parameter of the landscape.
N,N-dimethylphenethylamine (PENNA), a molecule with two functional groups, is able to form
an intramolecular cation-pi bond between a positive ion core at the amine site and the phenyl ring.
Excitation of the initially stretched molecule to a 3p Rydberg state triggers the formation of the
cation-pi interaction, which is seen in the binding energy spectrum as a sizable time- dependent
shift. Structural dispersion in this system is again large, leading to a broad line width.
The Rydberg electron binding energy also depends strongly on the presence of neighboring molecules,
opening an experimental avenue to study the kinetics of transitions between isomeric forms of
molecular clusters. In tetramethylethyldiamine and dimethylpropylamine clusters, we observe that
the binding energies of small molecular clusters (n<10) are shifted by about 0.5 eV from their
monomer energies. The time dependence of the spectrum reveals the reorganization of the solvent
surrounding the newly formed molecular ion core.


Ultrafast Electron Diffraction from Selectively Aligned Molecules
Martin Centurion1, Peter Reckenthaeler2, Werner Fuß
2, Sergei A. Trushin2, Ferenc
Krausz2
,
3, and Ernst E. Fill2
1University of Nebraska, Lincoln, NE 68588-0111, USA, Email: mcenturion2@unl.edu
2Max-Planck-Institut fuer Quantenoptik, Hans-Kopfermann-Straße 1, D-85748 Garching,
Germany
3Ludwig-Maximilians-Universitaet Muenchen, Am Coulombwall 1, D-85748 Garching, Germany
Abstract
Electron diffraction has been very successful for determining the structure of molecules in the gas
phase, and also for investigating ultrafast conformational changes. However, due to the random
orientation of the molecules in the gas phase only 1D information (the interatomic distances) can
be extracted from the diffraction patterns, which limits the size of molecular structures that can be
studied. Having a sample of aligned molecules would greatly increase the information encoded in
the diffraction pattern and potentially allow for reconstructing the full 3D molecular structure.
Here we show electron diffraction patterns recorded from a sample of transiently aligned molecules.
In our experiments molecules are aligned selectively using photodissociation of C2F4I2 (1,2- diiodotetrafluoroethane).
The diffraction pattern is captured by probing the sample with picosecond
electron pulses shortly after dissociationbefore molecular rotation causes the alignment to vanish.
The transition dipole moment of C2F4I2 is parallel to the C-I bond, along which the dissociation
takes place. Therefore, the C2F4I radicals emerge preferentially with the dissociated C-I direction
aligned along the laser polarization vector. Our results clearly show that the angular distribution
of the molecules becomes anisotropic after dissociation. The alignment was found to decay with a
time constant of 2.6±1.2 ps.
Radio-frequency compression of electron bunches applied to
Ultrafast Electron Diffraction at kV energies
Robert P. Chatelain, Chris Godbout, Vance R. Morrison, Bradley J. Siwick
Departments of Physics and Chemistry, Center for the Physics of Materials,
801 Sherbrooke St. W., Montreal, QC, H3A 2K6 Canada.
email: robert.chatelain@mail.mcgill.ca, christopher.godbout@mcgill.ca,
vance.morrison@mail.mcgill.ca, bradley.siwick@mcgill.ca
Abstract
Ultrafast Electron Diffraction (UED) has evolved into a versatile tool for studies of structural dynamics
in molecules and materials at sub-Angstrom spatial resolution. The time resolution obtainable
with this approach has steadily improved since the “picosecond barrier”was broken in 2003.
In fact, electron pulse durations of several hundred femtoseconds are available from state-of-theart
kV electron sources as long as the bunch charge is kept below approximately 2 fC. These are
impressive advances, however it is important to note that time resolution below 100 fs is required
for many experiments, and that an electron beam dose in the range of 1 - 1000 pC is needed for
diffraction patterns of sufficient quality for most studies. This is a combination of requirements that
cannot be currently realized due to the space-charge temporal broadening inherent to high charge
density electron bunches. Thus, improvements in electron source performance are desirable for
the further development of UED. In this work we will show how the introduction of a specially
designed Radio- Frequency (RF) cavity into the UED beamline removes many of the technical
limitations on the current generation of electron sources. For example, state-of-the-art particle
tracking simulations show that it is possible to produce electron pulses below 100 fs that contain
less than 1 pC of charge at the kV energies preferred for electron crystallography experiments.
In addition, this approach allows for much greater control over the electron beam illumination
conditions (at the specimen) than is possible with the current generation of sources. Finally, the
fundamental limit to the performance of a UED diffractometer will be discussed. It will be shown
that the space-charge temporal broadening of electron bunches is but a hurdle to overcome; that is,
the true limit to performance results from the required transverse coherence length of the electron
beam for a given experiment, and the initial brightness of the photoemission itself.
Ultra Fast Electron Sources A New Conclusion
Ben Cook and Pieter Kruit
Faculty of Applied Science, Delft University of Technology,
Lorentz weg 1, 2628CJ Delft, The Netherlands
email: b.j.cook@tudelft.nl
Abstract
According to our research most ultra fast electron sources waste much of the current they so
painstakingly create, obtaining a brightness that does not match that of a continuous source.The
reduced/normalised brightness (which scales as current over normalised emittance) Br is a key
source parameter, because apart from statistical interactions it is a conserved quantity. Also Br
denes the current I in an illuminated area A,
I = A 2 V Br (1)
where is the half opening angle and V the potential.We examined existing and proposed sources,
making a table of Br , pulse length and energy spread (where pos- sible at source and sample).
We concluded: (1) Accurate information about source design and performance is limited;(2) Surprisingly,
despite modern mode-locked lasers, pulsed, experimentally proven, Br is much below
continuous eld emitters and Schottky(thermal eld) emitters. We nd photoeld emission very promising,
both [1] and [2] have claimed Br > 10
10
A/(m
2
srV ) but no proper, experimental evidence
is given. For a Schottky emitter Van Veen showed that statistical coulomb forces decrease Br as
early as 10
8
A/(m
2
srV ) [3]. The photoeld emitter may do even worse [4].
We suggest chopping a high Br continuous source as an alternative for stroboscopic imaging.
This could also be used for ultra fast ion microscopy, unleashing a whole new area of research.
[1] C. A. Brau. NUCL INSTRUM METH A, 407(1):1, 1998.
[2] P. Hommelhoff, C. Kealhofer, and M. A. Kasevich. PHYS REV LETT, 97(24):4, 2006.
[3] AHV van Veen, CW Hagen, JE Barth, and P Kruit. J VAC SCI TECHNOL B, 19(6):2038, 2001.
[4] M. S Cook, B Bronsgeest and P Kruit. In 7th International Vacuum Electron Sources Conference- awaiting publication,
2008.
Building a Modular Compact/Radio-Frequency Ultrafast Electron
Diffractometer: First Experiments in Compact Geometry
Chris Godbout, Vance R. Morrison, Robert P. Chatelain, and Bradley J. Siwick
Departments of Physics and Chemistry, Center for the Physics of Materials, McGill University,
801 Sherbrooke St. W., Montr´eal, Quebec, Canada, H3A 2K6 email:
christopher.godbout@mcgill.ca, vance.morrison@mail.mcgill.ca,
robert.chatelain@mail.mcgill.ca, bradley.siwick@mcgill.ca
Abstract
We will present our progress towards the development and implementation of a flexible new ultrafast
electron diffractometer at 100-150kV energies. This diffractometer can be congured in both a
compact geometry and expanded into a geometry that allows for the temporal compression of electron
pulses using a RF cavity. In the compact geometry the electrons are allowed to freely expand
via space-charge interactions so it is important to have the ability to place the electron source as
close as possible to the sample. This conguration provides temporal resolution of approximately
800fs with 104 electrons per pulse. The RF conguration uses a synchronized RF cavity to temporally
compress the electron pulses to below 100fs while allowing up to 6x106 electrons per pulse;
this is an improvement of several orders of magnitude compared to the current state of the art.
We will report on initial experiments to characterize the diffractometer in compact geometry.
These experiments include studies of the electron relaxation dynamics and lattice heating in thin
film gold. The films are excited using approximately 50 femtosecond 400nm optical pump pulses
below the damage threshold. The relatively slow heating dynamics of the gold thin film leads
it to be an excellent initial experiment to characterize our system by comparing it to previously
published results. Progress towards implementing RF pulse compression in this instrument will
also be described.
A picosecond time-resolved X-ray scattering facility at BioCARS
T. Graber*, R. W. Henning, I. Kosheleva, Z. Ren, V. Srajer, and K. Moffat
Center for Advanced Radiation Sources, The University of Chicago, Chicago, IL 60637
H-S. Cho, N. Dashdorj, F. Schotte, and P. Anfinrud
NIDDK, National Institutes of Health, Bethesda, MD 20892
Abstract
BioCARS, a national user facility for time-resolved X-ray scattering studies at the Advanced Photon
Source (APS), has recently completed commissioning of a focused pink-beam beamline for
single-shot laser-pump/X-ray-probe measurements with a time resolution of 100 ps. Each x-ray
pulse can contain up to 3 x 1010 photons, giving a time- averaged flux similar to that of fourthgeneration
free electron laser sources. A broadly tunable laser system provides a pulse width of 1
to 150 ps depending on its configuration and has an energy density of 5 mJ/mm2 at the sample.
Two in-line undulators with periods of 23 and 27 mm give continuous 6.8-20 keV first-harmonic
coverage and can be combined for maximum flux at 12 keV. In combination with a high-heat-load
shutter that reduces the average power load, a Kirkpatrick-Baez mirror system focuses the x-ray
beam to a spot size of 90 μm (horizontal) by 20 μm (vertical). A high-speed J¨ulich shutter isolates
radiation from individual 100-ps storage-ring bunches at a 1-kHz rate and is compatible with the
most common storage ring fill patterns. This strategy allows almost full utilization of the entire
run period at the APS. The facility will be described, along with some recent scientific results
that highlight the unique features of the beamline. Additionally, a proposed experiment to use
energy-chirped X-ray pulses at the Linac Coherent Light Source will be discussed.
To apply for beamtime or for more information about the BioCARS facility, visit http://biocars.org.
* Corresponding author: graber@cars.uchicago.edu
Evaluation of lattice motion with vacuum-free compact designed
time- resolved X-ray diffraction
Masaki Hada and Jiro Matsuo
Department of Nuclear Engineering, Kyoto University, Sakyo, Kyoto, Japan
email: hadamasaki@nucleng.kyoto-u.ac.jp
Quantum Science and Engineering Center, Kyoto University, Gokasho, Uji, Kyoto, Japan
email: matsuo@nucleng.kyoto-u.ac.jp
Abstract
Hard X-ray from femtosecond laser-produced plasma has gained much interest, as unique time
resolved X-ray diffraction (TRXRD) experiments demonstrated and reveal ultrafast atomic dynamics
of chemical reactions, phase transitions and coherent phonon vibrations.[1-3] Elucidating
such ultrafast phenomena will lead to the fundamental understanding of energetic beam science
and also further understanding of physical phenomena in the uncharted nanoscale extreme conditions.
Recently, compact tabletop millijoule femtosecond lasers have been reported to be available
for generating hard X-ray in vacuum with an intensity of about 1081010 cps/sr with the K X-ray
conversion efficiency of 105106.[4-5] The experimental scale of a femtosecond laser could be reduced
with a tabletop laser; however difficulties remain when using a huge and complex vacuum
chamber system. We have constructed a compact designed and high intensity ultrafast pulsed Cu
X-ray source in helium atmospheric pressure. A vacuum- free TRXRD system has also been constructed
with this X-ray source. It is possible to reduce the overall size of X-ray source system
without the complexity of a vacuum system.[6,7] It is also feasible to place the samples which are
measured with TRXRD close to the X-ray source without vacuum system, enabling the use of the
generated X-ray more efficiently.
We performed TRXRD on the 3 mJ/cm2 infrared femtosecond laser irradiated bulk sample of a
CdTe single crystal with this vacuum-free compact designed TRXRD system. The CdTe is one
of the suitable samples for TRXRD because the penetration depths of infrared light and Cu K
X-ray into CdTe are almost the same degrees about 0.5 mum.[3] The integrated intensities of K
X-ray diffraction lines from CdTe (111) were decrease by 5.6% in the time scale of 100 ps. The
irradiation of infrared light at the intensity of 3 mJ/cm2 raises the temperature of CdTe by 50
K, and the thermal lattice vibration and expansion could occur. They would reduce the intensity
of X-ray diffraction line by 56% due to the change of Debye-Waller factor. It takes 100 ps
for the thermalized lattice in CdTe with acoustic velocity to expand 0.5 μm depth. Thus, the
changes of the integrated intensity of X-ray diffraction line would be induced by thermal vibration
and expansion of CdTe lattice. This vacuum-free compact designed TRXRD system would be a
desirable tool for time-resolved atomic dynamics measurements.
[1] C. Rose-Petruck, et. al., Nature 398, 310 (1999).
[2] K. Sokolowski-Tinten, et. al., Nature 422, 287 (2003).
[3] K.G. Nakamura, Appl. Phys. Lett. 93, 061905 (2008).
[4] C.L. Retting, et. al. Appl. Phys. B 93, 365 (2008).
[5] C.G. Serbanescu, et. al., Rev. Sci. Instruments 78, 103502 (2007).
[6] B. Hou, et. al., Appl. Phys. Lett. 92, 161501 (2008).
[7] M. Hada, et. al., Appl. Phys. B submitted.
Ultrafast Time Resolved Electron Diffraction of Dynamics of
Adsorbates on Silicon Surfaces
M. Kammler, S. M¨ollenbeck, A. Hanisch-Blicharski, A. Kalus, P. Schneider, B.
Krenzer, and M. Horn-von Hoegen
Department of Physics and Center for Nanointegration Duisburg-Essen (CeNIDE) University of
Duisburg-Essen, 47057 Duisburg, Germany, email: martin.kammler@uni-due.de
Abstract
Dynamic processes of surfaces like electron excitation and relaxation, electron-phonon coupling,
phase transition and phonon-phonon coupling take place on the femto- and picosecond timescale.
Ultrafast time resolved electron diffraction is an excellent technique to study such processes on
surfaces after excitation by a fs laser pulse. The laser energy will excite the electron system and
heat the topmost atomic layers by electron-phonon coupling. In our experiment surface sensitivity
is obtained by a RHEED (reflection high energy electron diffraction)-geometry [1]. In order to
study the energy dissipation of an adsorbate systems after vibrational excitation we have performed
time resolved measurements on the (
p
3x
p
3) Pb reconstruction on a Si(111) surface. (
p
3x
p
3) Pb
reconstruction has a coverage of 4/3 monolayer and was prepared by deposition of Pb on Si(111)
- (7x7) at 300 K followed by an annealing step to 500 K. After excitation of the Pb layer the
heat transport into the silicon substrate is determined by studying the cooling process using the
Debye Waller effect on the diffraction patterns taken at different delays between pumping laser
pulse and probing electron pulse. The measured time constant of 150 ps can be explained by the
huge difference in mass of Si and Pb atoms which prevents effective coupling of the Pb vibrational
modes to the phonon bath in Si substrate. In order to study the dynamics of strongly driven phase
transitions at surfaces far away from thermal equilibrium we performed time resolved experiments
on the Peierls like phase transition from a (8x”2”) to a (4x1) reconstruction of a Indium terminated
Si(111) surface upon laser excitation at a sample temperature of 40 K [2]. The In-chains
form 1-dimensional system currently being discussed whether the formation of a charge density
wave (CDW) or the rearrangement of atoms in the In-chains is responsible for the formation of
reconstruction. After excitation the (8x”2”)-diffraction spots instantaneously disappears, while
the intensity of the (4x1)-spots increases. This increase of the (4x1) spot intensity excludes an
explanation by the Debye-Waller-Effect and is evidence for a true structural phase transition at a
surface.
[1] A. Janzen, B. Krenzer, O. Heinz, P. Zhou, D. Thien, A. Hanisch, F.-J. Meyer zu Heringdorf, D. von der Linde, and
M. Horn-von Hoegen, Rev. Sci. Inst. 78,013906 (2007)
[2] S. M¨ollenbeck, A. Hanisch-Blicharski, P. Schneider, M. Ligges, P. Zhou, M. Kammler, B. Krenzer, and M. Hornvon
Hoegen, MRS-Proceedings (submitted)
Miniaturized RF Technology Towards a Novel Technique for
Sub-Picosecond Electron Bunch Generation
A. Lassise, P.H.A. Mutsaers, O.J. Luiten
Eindhoven University of Technology, Dept. of Applied Physics,
P.O. Box 513, 5600 MB Eindhoven, The Netherlands
email: A.C.Lassise@tue.nl, P.H.A.Mutsaers@tue.nl, O.J.Luiten@tue.nl
Abstract
Due to the extremely fast nature of interactions that occur on the microscopic scale, appropriate
spatial and temporal resolution of these processes is desired. One way this is being pursued is
through the use of electron bunches with a temporal length below picoseconds (<10−12 s). To
date, sub-picosecond electron bunches have been realized through the use of femtosecond lasers
interacting with photocathodes.
We present a novel technique utilizing RF technology towards the creation of sub-picosecond electron
bunches without the compulsory use of femtosecond laser systems. Utilizing RF technology
and tricks developed as far back as the 1930s, we show through simulations and calculations that
sub-picosecond electron bunches can be created with extremely low emittance growth to the electrons.
The design implements a 30 keV electron source from an SEM and highly underdamped electromagnetic
standing wave cavities designed for high field strengths with low power consumption.
The experimental setup is currently in the construction phase. Initial measurements are planned to
progress shortly hereafter.
Coherent acoustic phonons in ultrathin monocrystalline Bismuth
Gustavo Moriena1 , Masaki Hada2 , Jiro Matsuo2 , Cheng Lu2 , Hubert Jean-Ruel1 ,
Meng Gao1 , Ryan Cooney1 , Angelo Karantza1 , Germ´an Sciaini1 and R.J. Dwayne
Miller1
1Institute for Optical Sciences and Departments of Physics and Chemistry, University of Toronto,
80 St. George Street, Toronto, ON, M5S 3H6, Canada
2Quantum Science and Engineering Centre, Kyoto University, Gokasho, Uji, Kyoto 611-0011,
Japan
Abstract
Femtosecond electron diffraction (FED) is a very important technique to study structural dynamics
of matter in ultrathin lms. When a femtosecond laser pulse photoexcites a thin lm it generates
electronic and thermoelastic stresses which are nally released as acoustic waves. The propagation
of those waves, being constrained by the lm thickness, is responsible for the launching of coherent
acoustic modes. The corresponding vibrational periods of those modes are in good agreement with
that predicted by standing waves established by the boundary conditions[1]. FED is very sensitive,
due to their very small de Broglie wavelength, to lattice displacements in transverse direction.
When an ultrashort electron pulse probe the sample, reveals information about elastic properties
of those lms, including shear modes which are usually within the noise in all-optical studies[2].
Taking into account the speed of sound in solids ( 5 km/s) and the thickness of the lms (tens
of nanometers), the period of those oscillations is typically in the order of few picoseconds. In
this work we report on the generation and detection of coherent acoustic phonons in free-standing
single crystalline Bismuth lms.
A Compact Ultrafast Electron Diffractometer with MeV Electron
Pulses Generated by RF photocathode
Y. Murooka, N. Naruse, J. Yang, and K. Tanimura
The Institute of Scientific and Industrial Research (ISIR), Osaka University, Mihogaoka 8-1,
Ibaraki, Osaka 567-0047, JAPAN
email: ymuro@sanken.osaka-u.ac.jp, naruse@sanken.osaka-u.ac.jp, yang@sanken.osaka-u.ac.jp,
tanimura@sanken.osaka-u.ac.jp
Abstract
For determining transient structures in dynamical phenomena further and for developing electron
microscope with fs-temporal resolution, we have constructed an ultrafast electron diffraction system
of a transmission mode for a pulse including 107 electrons. A compact gigahertz rf (S band)
photocathode, with an extremely small energy spread ( E/E < 10−4) and emittance (<0.1mm
mrad) was specially designed to make the entire diffraction system a laboratory-sized equipment
[1]. Photoelectrons from Cu target were generated by the third harmonics of Ti:Sapphire laser, and
accelerated by rf with a repetition rate of 10Hz. For a pulse with 106 electrons generated by 70-fs
laser pulse and accelerated to 2MeV, the temporal width is estimated to be as short as 80 fs.
The system is designed to be especially rich in the electron beam configuration equipped with a
condenser lens, an objective lens, and a projector lens, similar to a conventional transmission electron
microscope. Therefore, both electron diffraction and imaging are possible. The illuminations
with parallel/focused electron beam are easily switched, and the camera length is also adjustable.
The sample chamber is at an ultra-high-vacuum ( 10−9Pa) with several manipulation capabilities.
Diffraction patterns can be recorded in two ways: one is real-time imaging with a sensitive CCD
camera combined with an efficient scintillator for pump-probe experiments of reversible phenomena,
and the other is for single shot experiments of non- reversible phenomena with extremely
sensitive emulsion films used for high-energy physics experiments.
The photocathode was stable over hours, and the current density could be tuned precisely for
various types of experiments. The current is in the range of 0.1 2pA, corresponding to 106
107
electron/pulse that is sufficient for single shot experiments. Using the CCD based detection, high
quality diffraction patterns were recorded from a thin film (70nm) of polycrystalline aluminium.
Diffraction rings were clearly resolved up to 1.4°A−1 that is sufficient for further processing to
obtain, for example, the radial distribution function. It seems that a sample with the thickness
close to the penetration depth of the laser can be investigated. Diffraction patterns were recorded
also from single crystal mica without obvious degradation in the pattern due to possible charge
buildup. The capability of single-shot imaging is reported, and the challenges to the goal of fstime
resolved electron microscope are discussed.
In situ observations of amorphous Silicon and Germanium
nanocrystallisation by Ultrafast Transmission Electron Microscopy
Liliya Nikolova1, Shona McGowan2, James Evans3
,
4, Thomas LaGrange3, Bryan
W. Reed3, Mitra L. Taheri5, Nigel D. Browning3
,
4, Jean-Claude Kieffer1, Bradley J.
Siwick2 and Federico Rosei1
1Institut National de la Recherche Scientifique Center Energy Materials Telecommunications
1650, boul. Lionel-Boulet, Varennes, Qu´ebec, J3X 1S2, Canada
2Departments of Physics and Chemistry, Center for the Physics of Materials, McGill University
801 Sherbrooke St. W., Montreal, Quebec, H3A 2K6 Canada
3Lawrence Livermore National Laboratory 7000 East Ave., Livermore, CA 94550-9234,
Livermore, California, USA
4University of California Davis, Department of Chemical Engineering & Materials Science One
Shields Ave., Davis, California, USA, 95616
5Department of Materials Science and Engineering, Drexel University 3141 Chestnut Street,
Philadelphia, PA 19104 U.S.A.
Abstract
High quality structural information on the equilibrium states of most materials can be routinely
obtained through several standard approaches. Detailed structural characterization of short-lived
nonequilibrium states of materials, however, has proved very challenging since revealing the dynamics
of structural transformation requires direct observations on the nanosecond to femtosecond
timescale with spatial resolution of few nanometers.
The transmission electron microscope (TEM) is a powerful and versatile tool for the characterisation
of materials, offering high spatial resolution (as low as 0.5A° ); however, due to the poor
temporal resolution of conventional TEMs it is rarely used for in situ direct imaging of structural
transitions. In this work we will discuss recent developments in enhancing the temporal resolution
of TEMs to produce a new class of Dynamic Transmission electron microscope (DTEM) at
Lawrence Livermore National Lab. By improving TEM temporal resolution to the nanosecond
timescale while preserving high spatial resolution studies of even irreversible structural transformations
can be made.
We have used this new capability to study the crystallization dynamics of Amorphous Silicon (a-
Si) and Germanium (a-Ge) specimens at a temporal resolution of 20 ns. Crystallization of these
amorphous films has been induced by 532nm nanosecond laser pulses of variable fluence. Timeresolved
TEM images have shown that the crystallisation process for a-Si begins at approximately
20ns and its duration is strongly influenced by the incident fluence of the laser beam. At low
fluences the a-Si undergoes solid-state nanocrystallisation. At intermediate fluences a melt pool
is generated and large radially oriented crystals eventually form. At high fluences the film was
entirely melted and dewetting of the surface occurs with eventual crystallisation in large droplets
onto the supporting SiO2 membrane. Numerical modeling of heat conduction in the laser excited
film was also performed and is in good agreement with the observed TEM images.
This work performed in part under the auspices of the U.S. Department of Energy by Lawrence Livermore National
Laboratory under Contract DE-AC52-07NA27344 and supported in part by the US Department of Energy, Office of
Basic Energy Sciences.
Four-Dimensional Visualization of Electron Dynamics by Attosecond
Diffraction
P. Baum
Max-Planck-Institute of Quantum Optics, and Ludwig-Maximilians-Universit¨at M¨unchen,
Am Coulombwall 1, 85748 Garching, Germany.
email: peter.baum@lmu.de
Abstract
We report here on the extension of ultrafast electron diffraction to the attosecond regime of charge
densities in motion. Four-dimensional imaging of electronic structures and their changes by
diffraction requires electron pulses with attosecond duration, in free space and at keV-range energies.
We present two of our concepts, using synchronized microwave cavities or counter- propagating
optical fields for electron pulse compression towards durations approaching 15 attoseconds
[1-2]. Results on the roles of space charge and phase matching are presented. In contrast
to attosecond photon pulses at around 100 eV [3], these attosecond electron pulses have by factors
of 1000 shorter wavelengths and allow for diffraction with atomic-scale resolution [4]. Two
potential applications are discussed for the example of molecular iodine: One involves measuring
changes in bond order and the associated reshaping of the molecular charge density; the other
regards attosecond charge oscillations in dielectrics and the buildup of the refractive index at optical
frequencies [4]. We also present the results of quantum model simulations of the electron
scattering process on an attosecond time scale and investigate the magnitude of radiation damage,
the role of electron exchange interaction, and the influence of the molecular orbitals to diffraction
[5]. These calculations support the possibility of using electron diffraction for imaging the structural
motion of charge density in four dimensions, and also point out ways for exciting attosecond
electron dynamics with keV-range electron pulses.
[1] P. Baum, A. H. Zewail, PNAS 104, 18409 (2007).
[2] F. Kirchner, F. Krausz, P. Baum, in preparation (2010).
[3] F. Krausz and M. Ivanov, Rev. Mod. Phys. 81, 163 (2009).
[4] P. Baum, A. H. Zewail, Chem. Phys. 366, 28 (2009).
[5] P. Baum, J. Manz, A. Schild, Sci. China G, submitted (2009).
Dynamic Transmission Electron Microscopic Investigation of
Telluride Phase Change Materials
B. W. Reed1, S. Meister2, G. H. Gilmer1, D. J. Masiel3, M. K. Santala1, T.
LaGrange1, G. H. Campbell1, and N. D. Browning1
,
3
1Physical and Life Sciences Directorate, Lawrence Livermore National Laboratory, 7000 East
Avenue, Livermore, CA 94551, USA; email: reed12@llnl.gov, gilmer1@llnl.gov,
santala1@llnl.gov, lagrange2@llnl.gov, campbell7@llnl.gov
2Materials Science and Engineering, Stanford University, 476 Lomita Mall, Stanford, CA 94305,
USA; smeister@stanford.edu
3Department of Chemical Engineering and Materials Science, University of California, Davis,
Davis, CA 95616, USA; email: djmasiel@ucdavis.edu, nbrowning@ucdavis.edu
Abstract
Using the technologically important phase change material Ge2Sb2Te5 as an example, we show
how a combination of single-shot real-space nanosecond transmission electron microscope imaging,
time-resolved electron diffraction, and computation can reveal details of the interactions
among geometry, optical absorption, and nucleation and growth kinetics in amorphous-crystalline
transformations. We find the crystal nucleation density in this material to be exceedingly high
(with many nuclei appearing per cubic m even after nanosecond-scale incubation times), such that
large-scale molecular dynamics simulations are directly relevant for interpretation of the results.
Grain growth and ensuing morphological changes happen much more slowly, on the scale of microseconds.
We also show how principal component analysis of time-resolved diffraction data can
provide a multi-dimensional picture of the evolution of various aspects of the transformation while
suppressing noise and irrelevant information. Finally, we explore the interaction between geometry
and laser absorption through the in situ study of nanostructured phase change materials coupled
with multiphysics finite element simulations.
This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National
Laboratory under Contract DE-AC52-07NA27344.
How to Extract Time-resolved Signal from Laue Diffraction by an
Energy-Chirped Hard X-ray Pulse: a Proposal
Zhong Ren, Tim Graber, and Keith Moffat
Center for Advanced Radiation Sources, The University of Chicago 9700 South Cass Avenue,
Building 434B, Argonne, IL 60439, USA email: renz@uchicago.edu
Abstract
A method to generate an energy-chirped hard X-ray pulse using a scheme based on overcompression of electron
bunches is currently being developed at the LCLS. These energy- chirped pulses are expected to reach 1-1.5% bandwidth
at 8 keV and a subpicosecond temporal width. We propose to use these chirped pulses to study light-initiated
reactions in biological macromolecules like myoglobin and photoactive yellow protein at ultra-fast time-resolution in
the time domain from 100 ps to 10 fs. One of the research and development areas required by this study is an effective
numerical algorithm to extract time-resolved signal from Laue diffraction images produced by these chirped pulses.
We will present a proposal of such algorithm and some preliminary data.
Single crystal Laue diffraction by a polychromatic X-ray beam is recorded as a pattern of spots on an area detector
when the sample is stationary during exposure. Most spots arise from satisfying the Bragg condition at specific
wavelengths within the bandwidth of the polychromatic source. A small fraction of the spots, known as multiples, are
caused by satisfying the Bragg conditions simultaneously at two or more harmonic wavelengths, all represented within
the source bandwidth. If the bandwidth is small enough, as it would be for the proposed energy chirp, there would
be virtually no multiple spots. In an oversimplified statement, single crystal diffraction by such a narrow bandwidth,
polychromatic source produces a pattern of spots, each of which can be traced back to a specific wavelength present
in the source. If the source features an energy-chirped pulse, i.e. the arrival time at the crystal of each X-ray photon
is highly correlated with its energy, each spot in a Laue pattern can be further mapped to its time of diffraction. This
spot-to-time mapping suggests that Laue patterns produced by chirped pulses, although they do not appear to differ
from those produced by conventional, unchirped pulses, are capable of recording time-dependent information with an
intrinsic time-resolution substantially less than their pulse duration. A sufficient number of these Laue patterns may
yield time- resolved data that is complete in diffraction space and span the entire desired time range.
The energy-angle correlation inherent in Braggs Law suggests an even more detailed mapping between each detector
pixel and time. Each pixel associated with a Bragg spot has a known mean energy and spans a small energy range
proportional to its linear dimension. In all previous analyses of Laue diffraction images, each spot spanned many
(often 25) pixels, integration of diffraction intensities was carried out across all pixels and each spot was associated
with a single X-ray energy. Here, our basic strategy is to analyze all Laue spots pixel by pixel without spot integration
in order to take full advantage of the pixel-to-time mapping. This strategy requires joint modeling of the crystal mosaic
structure and the spectral distribution of photon energy in each chirped pulse. The spectral distribution is anticipated
to vary markedly from pulse to pulse, but both it and the mosaic structure are constant across the few hundred spots
on each image. When these functions are jointly modeled, the remaining variation in pixel intensity across a spot
arises from a combination of time-resolved signal, that is, the desired quantity synchronized in time from spot to spot,
and experimental noise. This gives us the opportunity to apply singular value decomposition to extract the signal
synchronized in the time domain.
Progress of mega-electron volt ultrafast electron diffraction at
Tsinghua University
Renkai Li, Wenhui Huang, Yingchao Du, Huaibi Chen, Taibin Du, Qiang Du,
Jianfei Hua, Jiaru Shi, Lixin Yan and Chuanxiang Tang
Department of Engineering Physics, Tsinghua University, Beijing 100084 China
email: *Tang.xuh@tsinghua.edu.cn
Abstract
Time-resolved ultrafast electron diffraction (UED) is a promising tool to probe structural changes
on the fundamental temporal and spatial scales of atomic motions. There have been recent efforts
to employ mega-electron volt (MeV) electron beam from photocathode radio-frequency (RF) gun
for UED application, mainly to achieve a better temporal resolution and eventually single-shot
patterns with good signal-to- noise ratio. While, when using RF technology and MeV electron
beam, several issues are worth careful consideration before applied for scientific experiments, e.g.
the RF amplitude jitter, the RF-to-laser synchronization jitter, and how to detect MeV electrons
with high enough efficiency. We optimized the configuration and parameters of a MeV UED
system by start-to-end numerical simulation, and built and optimized such a prototype system at
the Tsinghua Thomson scaterring X-ray source (TTX) facility. We obtained high-quality singleshot
diffraction patterns of a 200 nm polycrystalline aluminum foil in which the first few rings are
clearly distinguishable. We will also present considerations on improving several key components
and discuss the futural plan.
Ultracold plasma electron source for imaging biological molecules
Mark Junker, Simon Bell, David Sheludko, Sebastian Saliba, Andrew McCulloch
and Robert Scholten
Centre of Excellence for Coherent X-ray Science,
The University of Melbourne
VIC 3010, Australia
email: scholten@unimelb.edu.au
Abstract
The molecular structure of biological molecules such as bacteriorhodopsin can be determined by
electron diffraction, but general application of the technique has been limited by the brightness of
conventional electron sources. Brightness is proportional to current and inversely proportional to
temperature. Recent advances in atomic physics have made the prospect of high brightness electron
beams from cold atomic clouds a promising alternative to conventional high temperature (104 K)
sources [1,2]. Cold atoms in a magneto-optic trap (MOT) can be photoionized with a laser tuned
just above threshold, releasing electron bunches with temperatures as low as 10 K. Although the
number of electrons that can be extracted from a MOT is relatively small, the dramatic reduction
in temperature may enable brightness that is competitive with conventional alternatives.
We created a MOT of 108 85Rb atoms, which were then ionized by two-step photoexcitation using
the 780 nm MOT trapping beams and a 5 ns pulsed dye laser tuned near the ionization threshold
(480 nm). The electrons were accelerated by an electrostatic field up to 200 V/cm between parallel
accelerator plates, and electrostatically focussed using a third electrode. The electron bunches were
detected using a microchannel plate, phosphor screen, and standard scientific CCD camera.
We are investigating the coherence and brightness of the extracted electron bunches, and in particular
the effect of controlling the initial spatial distribution of the atoms to generate a uniform
density elliptical charge distribution. Such elliptical bunches intrinsically preserve their brightness,
and can for instance be refocused with conventional accelerator techniques [3].
[1] T.C. Killian, T. Pattard, T. Pohl and J.M. Rost, J.M. (2007). Ultracold Neutral Plasmas. Physics Reports 449, 77
[2] B.J. Claessens, M.P. Reijnders, G. Taban, O.J. Luiten and E.D.J. Vredenbregt (2007). Cold electron and ion beams
generated from trapped atoms. Physics of Plasmas 14, 093001
[3] B.J. Claessens, S.B. van der Geer, G. Taban, E.J.D. Vredenbregt and O.J. Luiten (2005). Ultracold electron source.
Phys. Rev. Lett. 95, 1649801
Ultra-fast dynamics of dimeric rhodium in a rigid ligand framework
Mirko Scholz1 , Faton S. Krasniqi2 , Ren´e Mor´e 1 , J¨org Hallmann1 , Simone
Techert1
,
2
1Max-Planck-Institute for Biophysical Chemistry, Am Fassberg 10, 37077 G¨ottingen, Germany
2Advanced Study Group at Centre for Free Electron Laser Science, Notkestraße 85, 22607
Hamburg, Germany
email: mscholz5@gwdg.de (M. S.), faton.krasniqi@asg.mpg.de (F. S. K.), rmore@gwdg.de (R.
M.), jhallma1@gwdg.de (J. H.), stecher@gwdg.de (S. T.)
Abstract
[Rh2 dimen4 ](PF6 )2 is a system with unusual bond shortening upon photo-excitation. In order to
understand the switching behavior of this compound in more detail ultra- fast X-ray diffraction and
transient absorption spectroscopy have been performed. The transient optical spectroscopy in the
NIR regime suggests a coherent behavior on the femtosecond time scale, where as time resolved
X-ray diffraction reveals mod- ulations of the integrated intensities of the observed Bragg reections
on slower time scales (about 10 ps). However, no dynamics of the crystal lattice was induced with
the excitation power used. The experimental data will be compared to theoretical calculations of
the cation at TD-DFT level of theory.
The time resolved XANES and X-ray fluorescence high average
power beam- lines at the Advanced Laser Light Source (ALLS)
facility
C. Serbanescu, S. Fourmaux, J.C. Kieffer
INRS-EMT, blvd Lionel Boulet, Varennes, Qu´ebec, Canada
Abstract
We are investigating performances of ultrafast laser-based x-ray sources for dynamic imaging of
various materials using time resolved X-ray spectroscopy [1,2]. We will present our effort in developing
time resolved XANES and X-ray fluorescence beam lines at the ALLS facility at INRS
with femtosecond and picosecond resolutions. A prototype beam line has been developed and coupled
to the 100Hz laser system at ALLS [3]. This Ti:Sapphire CPA system is delivering 100mJ at
800nm with 100Hz repetition rate (giving 10W of average power) and 25 fs pulses (giving 4TW
of peak power). Our most recent improvements include the control of the thermal loading of the
beam line components at the 10W average power level in order to achieve high brightness and high
stability x-ray source, and very high signal to noise ratio data collection. The source performances
will be discussed and our preliminary experiments to follow the dynamics of photoexcited myoglobin
will be presented. The ongoing effort to achieve sub-hundred femtosecond x-ray pulses
with the 200TW/50W ALLS system (5J, 10Hz, 25fs) will be briefly sketched.
The ALLS facility has been funded by the Canadian Foundation for Innovation (CFI). This work is supported by
NSERC, the Canada Research Chair Program and by Ministre de lducation du Qubec.
[1] F. Raksi et al, J. Chem. Phys. 104, 6066 (1996)
[2] A. Cavalleri et al, Phys. Rev. Lett. 95, 067405 (2005)
[3] S. Fourmaux et al, Rev. Sci. Instrum. 78, 113104 (2007)
Extreme phonon softening in laser-excited Bismuth towards an
inverse Peierls-transition
K. Sokolowski-Tinten1, W. Lu1, M. Nicoul2
,
1, U. Shymanovich1, A. Tarasevitch1,
M. Kammler1, M. Horn von Hoegen1, D. von der Linde1
1University of Duisburg-Essen, Lotharstr. 1, 47048 Duisburg, Germany, e-mail:
Klaus.Sokolowski@uni-due.de
2University of Cologne, Z¨ulpicher Straße 77, 50937 K¨oln, Germany e-mail:
matthieu.nicoul@uni-koeln.de
Abstract
Irradiation of a solid material with intense ultrashort laser pulses can lead to significant changes of
the interatomic forces. Upon photoexcitation electrons are usually promoted from bonding states
to less bonding or even anti-bonding states, thereby setting off atomic motion in the system. A
prominent example is the so-called displacive excitation of coherent phonons (DECP) [1]. It has
been found that DECP occurs only in materials with phonon modes of A1-symmetry which do not
lower the symmetry of the material, and that only A1-modes are excited. The equilibrium structure
of these materials can be derived by a Peierls-type transition from a state of higher symmetry.
Bismuth is a prominent example in which this type of coherent vibrational excitation has been
studied in great detail. The majority of published results are based on time-resolved all-optical
studies which cannot provide direct structural information. More recently time-resolved X-ray
diffraction has also been used to directly follow the atomic motion associated with the laser- excited
coherent phonon [2-4]. In particular the work performed at the Sub-Picosecond Pulse Source [3]
has allowed, for the first time, to quantitatively measure the transient changes of the potential
energy surface which underlie DECP and the softening of the phonon modes. In the present work
we have used time-resolved X-ray diffraction to extend our studies of coherent optical phonons in
laser-excited Bismuth to a higher fluence range that has not been studied previously. Femtosecond
X-ray pulses at 8 keV (Cu K ) from a laser-produced plasma served as probe pulses in an optical
pump X-ray probe experiment. The transient changes of the (111)- and the (222)-diffraction peaks
of a crystalline, 50 nm thick Bismuth film have been measured in a symmetric Bragg-configuration.
For absorbed laser fluences above 2 mJ/cm2 our experimental data reveal an extreme softening of
the A1g-mode down to frequencies of about 1 THz, only 1/3 of the unperturbed A1g-frequency.
The observed softening follows qualitatively the predictions of density functional calculations [5].
For even higher fluences (above 3 mJ/cm2) the measured diffraction signals no longer exhibit an
oscillatory behaviour. Our experimental observations present strong indication that upon intense
laser-excitation the Peierls-transition which determines the equilibrium structure of Bismuth can
be reversed and that the material is transformed into a transient ordered state of higher symmetry.
[1] H. J. Zeiger et al., Phys. Rev. B. 45, 768 (1992).
[2] K. Sokolowski-Tinten et al., Nature 422, 287 (2003).
[3] D. M. Fritz et al., Science 315, 633 (2007).
[4] S. L. Johnson et al., Phys. Rev. Lett. 100, 155501 (2008).
[5] E. D. Murray et al., Phys. Rev. B 72, 060301 (2005).
Time-resolved crystallographic studies of heme proteins
Vukica Srajer1, Marius Schmidt2, James Knapp3 and William E. Royer4
Center for Advanced Radiation Sources, The University of Chicago, Chicago IL, USA, email:
v-srajer@uchigago.edu
University of Wisconsin-Milwaukee, Milwaukee, WI, USA, email: m-schmidt@uwm.edu
Department of Biomedical Science, Mercer University School of Medicine, Savannah, GA, USA,
email: KNAPP JE@Mercer.edu
University of Massachusetts Medical School, Worcester, MA, USA,
email: william.royer@umassmed.edu
Abstract
The ultimate goal of time-resolved crystallographic studies of biological macromolecules is to
visualize intermediate states along a reaction pathway at atomic resolution and at physiological
temperatures, without trapping of the intermediates by chemical or physical methods. This is accomplished
by taking X-ray snapshots of the molecule in the crystal as a reaction proceeds following
the reaction initiation. The technique has reached a mature phase with demonstrated ability to
detect small structural changes on ns and sub-ns time scale (1-6) and with important advances in the
analysis of time-resolved crystallographic data, such as the use of Singular Value Decomposition
(SVD) method to determine the structures of intermediates and elucidate the reaction mechanism
(3-5). We present results of time-resolved crystallographic studies of heme proteins: structural relaxation
processes in myoglobin and allosteric action in real time in a more complex, cooperative
dimeric hemoglobin, as well as ligand migration pathways in both molecules. Myoglobin studies
reveal sub-ns protein relaxation following ligand photo- dissociation and provide first direct experimental
evidence of the main ligand exit pathway (7). Dimeric hemoglobin studies capture an early
photoproduct intermediate and identify a possible trigger for a transition from the initial R-state to
a tertiary T-like state that occurs on a s time- scale (8,9). These time-resolved experiments were
conducted at the BioCARS beamline 14-ID at the Advanced Photon Source (USA).
1) Srajer et al. Biochemistry 40, 13802 (2001)
2) Schotte et al. Science 300, 1944 (2003)
3) Schmidt et al. PNAS 101, 4799 (2004)
4) Ihee et al. PNAS 102, 7145 (2005)
5) Rajagopal et al. Structure 13, 55 (2005)
6) Bourgeois et al. PNAS 103, 4924 (2006)
7) Schmidt et al. PNAS 102, 11704 (2005)
8) Knapp et al. PNAS 103, 7649 (2006)
9) Knapp et al. Structure 17, 1494 (2009)
Direct Observation of Domain Behavior in a Multiferroic Structure
Under Applied DC Bias
Christopher Winkler1, Lane W. Martin2, Craig Johnson1, Mitra L. Taheri1
,

1Department of Materials Science & Engineering, Drexel University, 3141 Chestnut Street,
Philadelphia, PA 19104, USA; email: mtaheri@coe.drexel.edu
2Department of Materials Science & Engineering, University of Illinois at Urbana-Champaign,
1304 W. Green St., Urbana, IL 6180, USA; email: lwmartin@illinois.edu
contact author
Abstract
Select multiferroic materials exhibit a coupling between ferroelectric and magnetic order parameters,
mediated by a quantum-mechanical exchange interaction. One of the most widely studied
magneto-electric multiferroics is the perovskite BiFeO3 (BFO). The magneto-electric coupling in
BFO allows for control of the ferroelectric domain structures via applied electric fields. Recent
advances in synthesis techniques have enabled the growth of high quality, epitaxial thin films.
Because of these unique properties, BFO and other magneto-electric multiferroics constitute a
promising class of materials for incorporation into devices such as high density ferroelectric and
magnetoresistive memories, spin valves, and magnetic field sensors. Before BFO can be integrated
into devices, an understanding of its ferroelectric and antiferromagnetic domain behavior across a
range of time and length scales needs to be developed. Improved control of ferroelectric domain
structures is critical for increasing the performance of ferroelectric and magnetoresistive memories,
because memory switching speed and capacity are limited by domain wall mobility and domain
size, respectively. We investigated the ferroelectric domains in BFO using transmission electron
microscopy (TEM). Diffraction contrast was used to distinguish adjacent domains with different
polarization directions, and high resolution images were analyzed to determine the atomic structure
of domain walls. We present in situ TEM experiments designed to probe the response of
BFO thin films to an applied DC bias, thereby enabling control of ferroelectric switching in the
BFO thin film. Domain wall movement will be captured using digital streaming video at 30Hz, at
both low and high magnifications. In our experiments, domain motion was studied at millisecond
timescales; however, as industry aims to reduce device sizes, we look to ultrafast TEM to examine
domain kinetics at timescales otherwise unattainable.
Ultrafast coherent imaging using UV-X harmonic beamline
Willem Boutu, David Gauthier, Xunyou Ge, Xiaochi Liu, Bertrand Carr´e, Hamed
Merdji
SPAM, CEA Saclay, 91191 Gif sur Yvette, France, email: willem.boutu@cea.fr
hamed.merdji@cea.fr
Manuel Guizar-Sicairos and James R. Fienup
The Institute of Optics, University of Rochester, Rochester, N.Y. 14627, US
Filipe Maia and Janos Hajdu
Laboratory of Molecular Biophysics, Uppsala University, SE-751 24 Uppsala, Sweden
Abstract
X-ray lensless imaging extends standard X-ray diffraction towards imaging of individual nanosystems
with unrivalled space and time resolutions. Up to now, this ability was limited to intense
femtosecond coherent pulses from a free electron laser. High harmonics generation (HHG) sources
would represent an excellent alternative since they are widely available and show the required properties.
Moreover, HHG pulses are synchronized on sub-femtosecond time scale with the driving
infrared femtosecond laser, allowing a vast flexibility in time resolved experiments. However, their
brightness has so far restricted their application to static phenomena. In Saclay we developed a new
harmonic source, based on a significant improvement of UV-X yield from HHG in gases, driven in
enhanced phase-matching conditions. Using a long gas cell and a long focal length lens (5.5 m),
together with a high quality UV-X optical line, allows reaching up to 2x109 photons per shot for
the 25th harmonic (=32nm) on the sample.
This high energy level allowed performing coherent imaging under several different configurations.
We first realized a coherent diffracting imaging (CDI) experiment. The UV-X light is diffracted
by a micrometer size sample. The diffraction pattern of the object exit wave is collected on a
CCD camera in the far field regime. Since only the intensity of the diffracted coherent wave is
measured, the phase information is missing and must be recovered. Image reconstruction with
60nm resolution was carried out using iterative phase retrieval techniques. To demonstrate the
potential of our CDI beamline we then decreased the exposure times down to 20 femtoseconds.
In the single shot acquisition regime, we achieved a 120nm resolution (Ravasio et al., PRL 103
028104 (2009)), which we recently lowered down to 80nm after optimization of the harmonic
wavefront.
We then implemented a recently proposed holographic technique using extended references. This
technique, easy to implement, allows a direct non iterative image reconstruction. In the single shot
regime, we demonstrated a spatial resolution of 110nm.
This opens fascinating perspectives in imaging dynamical phenomena to be spread over a large
scientific community. Investigation of ultrafast phase transitions in mesoscopic systems, ultrafast
spin-reversals of magnetic nano-domains or large molecule rearrangements in biological environments
are some examples.

(pdf)