Invited article on cover of Review of Scientific Instruments
A review paper focusing on the technology of the complete attosecond beamline at Imperial College
was published in the July issue of
Review of Scientific Instruments.
We describe a complete technological system at Imperial College London for Attosecond Science
studies. The system comprises a few-cycle, carrier envelope phase stabilized laser source which delivers
sub 4 fs pulses to a vibration-isolated attosecond vacuum beamline. The beamline is used for the generation
of isolated attosecond pulses in the extreme ultraviolet (XUV) at kilohertz repetition rates through
laser-driven high harmonic generation in gas targets. The beamline incorporates: interferometers for
producing pulse sequences for pump-probe studies; the facility to spectrally and spatially filter the
harmonic radiation; an in-line spatially resolving XUV spectrometer; and a photoelectron spectroscopy
chamber in which attosecond streaking is used to characterize the attosecond pulses. We discuss the
technology and techniques behind the development of our complete system and summarize its performance.
This versatile apparatus has enabled a number of new experimental
investigations which we briefly describe.
Can we freeze time? Laser adventures in the realm of the nano-nanosecond. - John Tisch's Inaugural Lecture
Inaugural Lecture: Video of the inaugural lecture
John Tisch (project PI) has his professorial Inaugural Lecture at Imperial College, June 22, 2011. In this lecture
John Tisch explains some of the exciting possibilities in ultrafast phenomena and measurement techniques to broad audience.
This is the perfect starting point non-specialists to find out more about attosecond science. Take a tour of high-speed measurement
technology that provides startling insight into the world around us, from galloping horses to electrons.
Numerical simulation of attosecond nanoplasmonic streaking
Video abstract: Numerical simulation of attosecond nanoplasmonic streaking
The characterization of the temporal profile of plasmonic fields is important both from the fundamental point
of view and for potential applications in ultrafast nanoplasmonics. It has been proposed by Stockman
(2007 Nat. Photonics 1 539) that the plasmonic electric field can be directly measured by the attosecond
streaking technique; however, streaking from nanoplasmonic fields differs from streaking in the gas phase because
of the field localization on the nanoscale. To understand streaking in this new regime, we have performed
numerical simulations of attosecond streaking from fields localized in nanoantennas.
In this paper, we present simulated streaked spectra for realistic experimental conditions and
discuss the plasmonic field reconstruction from these spectra. We show that under certain circumstances
when spatial averaging is included, a robust electric field reconstruction is possible.
LSI: Farfield intensity (left) and phase (right) of the 13th, 19th
and 25th harmonics. Experimental results (blue, shaded represents standard deviation)and simulation (red).
Lateral Shearing Interferometry of High-Harmonic Wavefronts
In work led by Dane Austin from the group of Prof Ian Walmsley at Oxford, Lateral Shearing Interferometry (LSI), was demonstrated
for the first time in the Attosecond Laboratory at Imperial college to characterise the wavefront of individual high harmonics.
Knowing the space-time structure of high harmonics will shine light on the physical processes involved in the atomic as well
as the macroscopic origin of high harmonic generation (HHG).
LSI for HHG uses two tilted replicas of an IR pulse as the driving field that are generated in a Mach-Zehnder interferometer.
The two resulting foci lead to two HHG sources that are spectrally resolved on a two-dimensional flat-field XUV-spectrometer.
Wavelength is dispersed vertically, and the harmonics propagate freely in the horizontal direction. As a result a two-source
spatial interference pattern is recorded at the detector from which the spatial phase can be extracted.
We measured the 13-25th harmonic generated by a 14fs Ti:Sa laser pulse in krypton. The experimental results
were compared against a simulation. The single atom response is modelled using quantum orbits and the propagation
effects are described with a simplified model, neglecting ionization and dispersion effects on the driving field.
The experimental and theoretical results are shown in the figure on the right.
Sub 4 fs High Energy Pulse
SPIDER reconstruction of a sub-4 fs pulse. Click on image to see spectrum and phase.
We demonstrate the generation of a 3.8 fs pulse with energies of up to 250 μJ. An octave spanning spectrum was
produced by coupling 30fs, ~700 μJ Ti:Sa laser with a 1kHz repetition rate into a 1 m long hollow fibre.
Compression was achieved with ultrabroadband chirped mirrors.
The pulse was measured with a SEA-SPIDER setup and the reconstructed pulse duration was 3.8 fs, only slightly
than the transform limited pulse (3.5 fs) given by the spectrum. Spectrum and Phase
Amplification of Impulsively Excited Molecular Rotational Coherence
Molecular phase modulation (MPM) uses the rapid variation of refractive index
in an ensemble of coherently vibrating or rotating molecules to spectrally
modify radiation, allowing broadband radiation to be generated. Typically,
coherent molecular motion is prepared using a rapidly changing pump field, or
fields, which drive the dynamics. A key challenge is to control the phase of the
molecular dynamics with respect to additional ultrafast optical sources. In a
recent Physical Review Letter, researchers at the University of Oxford proposed
and demonstrated a solution to this problem. The scheme involves preparation of
high-coherence molecular dynamics which are phase-stable with respect to
FROG-CRAB trace measured at Imperial College
attosecond pulses measured in the UK
One of the fastest processes in nature is the motion of an
electron in an atom. This occurs on the attosecond timescale (1
attosecond is 10-18s). In order to "track" electron
motion on this timescale, one must controllably produce pulses
of light with similar or even shorter duration, and take
"snapshots" of the process as it evolves over time - much like
a conventional camera but with a shutter speed more than a
million billion times faster. The production and measurement of
such pulses is notoriously difficult and has only recently
become feasible in a few laboratories worldwide. Our recent
measurements at Imperial College London are the first in the
United Kingdom, adding another powerful attosecond technique to
our toolbox of attosecond technology and bringing the UK to the
forefront of attosecond metrology.
The scheme used is similar to the pioneering development by
Ferenc Krausz and colleagues at the Max Planck Institute,
Garching, Germany (work started at the Technical University in
Vienna). Laser pulses (central wavelength 800nm) from a
commercial femtosecond kHz source are spectrally broadened to
cover the visible-infrared spectrum using a
differentially-pumped hollow-core fibre developed here at
Physics B 85,4 (2006): 525-529]. These pulses are
compressed to sub-10 femtoseconds (1 femtosecond is 10-15s)
duration, near the absolute limit possible at this wavelength.
These "few-cycle" pulses are waveform-stabilised through a
process known as carrier-envelope phase stabilisation, ensuring
that the electric field waveform of each pulse is nearly
identical -- a vital requirement for the characterisation and
stable generation of isolated attsecond pulses. In order to
realize the even shorter (attosecond) bursts of light desired
we exploit a process known as high harmonic generation in neon
gas, which frequency-multiplies the light into the xuv
spectrum. Alone, this technique produces a regular train of
pulses, one pulse for each half-cycle of the generating laser
waveform. However, through careful spectral filtering of the
pulses, using the combination of a zirconium foil and a
molybdenum-silicon multilayered mirror we isolate a single
attosecond pulse with ~92eV photon energy (wavelength about
The production of attosecond pulses is only the first step
-- any application necessitates a full characterisation of
these pulses. For this we have custom built an attosecond
streak camera, used to obtain a FROG-CRAB trace. The isolated
attosecond pulse is inherently synchronised with the infrared
laser pulse used to generate it. Both pulses are focussed onto
another gas jet of neon atoms. A neon atom absorbs the
energetic attosecond pulse, knocking free a photo-electron,
which carries away the excess energy. Simultaneously, the
electric field of the infrared laser pulse pushes the
photoelectron towards or away from a time-of-flight detector,
thereby streaking the momentum of the photoelectron. This
momentum shift depends on the vector potential of the laser
pulse at the instant in time it overlaps with the attosecond
pulse. By precisely scanning the delay between both pulses the
waveform of the laser pulse is mapped out, and more
importantly, the full characteristics (phase and amplitude) of
the attosecond pulse can be retrieved. Initial analysis
indicates an attosecond pulse duration of approximately 280
attoseconds, close to the limit anticipated for the arrangement
With a source of isolated attosecond pulses available the
next step is to use them in time-resolved studies of atoms,
molecules and surfaces. For example, we have just completed the
construction of a new surface-science target chamber with early
experiments set to look at nano-plasmonic
We made the first observation of “half-cycle cut-offs” (or
HCOs) during high harmonic generation (HHG). HCOs are the
highest frequency bursts of radiation emitted each half-cycle
of the laser pulse during HHG. They have been theoretically
predicted but had not previously been measured. We showed that
HCOs are extremely sensitive to the sub-cycle (i.e. attosecond)
details of the electric field evolution of the few-cycle
driving laser pulse. In particular, the centre frequency of the
HCOs was found to be a function of the CEP of the laser pulse,
an essential parameter in the characterisation of few-cycle
pulses, but notoriously difficult to measure. We used this to
develop a completely new measurement technique of the CEP that
offers many advantages over the handful of existing measurement
techniques. Most notably it can work for much longer pulses (up
to 5 cycles) and on a single-shot basis, This result was
published in Nature Physics in January 20072, making the front
cover of that issue.
The motion of hydrogen and deuterium is captured in the
first few femtoseconds after ionisation by an intense laser
field, by observation of the high harmonic spectrum emitted on
recollision of the ionised electron. This technique allows
these first few moments of motion of the molecule to be
captured with a resolution of <100 attoseconds.