Sub-phonon-period compression of electron pulses for atomic diffraction

Abstract

Visualizing the rearrangement of atoms in a wide range of molecular and condensed-matter systems requires resolving picometre displacements on a 10-fs timescale, which is achievable using pump–probe diffraction, given short enough pulses. Here we demonstrate the compression of single-electron pulses with a de Broglie wavelength of 0.08 ångström to a full-width at half-maximum duration of 28 fs or equivalently 12-fs root-mean square, substantially shorter than most phonon periods and molecular normal modes. Atomic resolution diffraction from a complex organic molecule is obtained with good signal-to-noise ratio within a data acquisition period of minutes. The electron-laser timing is found to be stable within 5 fs (s.d.) over several hours, allowing pump–probe diffraction at repetitive excitation. These measurements show the feasibility of laser-pump/electron-probe scans that can resolve the fastest atomic motions relevant in reversible condensed-matter transformations and organic chemistry.

Diffraction and microscopy with attosecond electron pulse trains

“…2f shows the electron pulse duration as a function of the compression laser's peak field strength, showing a characteristic shortening (compression) and subsequently an increase (over-compression) of the pulse width 13,21 . The best-compressed electron pulses in the train are about 35 times shorter than previously reported for atomic-resolution diffraction 21 . Beam quality is not substantially altered by the compression.…”

“…If a single isolated attosecond electron pulse is generated by single-opticalcycle filtering (see main text), we still can expect 10 5 -10 7 electrons per second. Terahertz-driven or microwave-driven pre-compression 13,21 can improve this value down to the limits given by the space-charge regime 37 , to be clarified by simulations, or alternatively reduce the demands on the electron source.…”

Attosecond electron pulse trains and quantum state reconstruction in ultrafast transmission electron microscopy

“…The underlying technology heavily rests on laser science for the 2 generation and characterization of ever-shorter femtosecond electron 10,14 and xray [15][16][17] probe pulses, with examples in optical pulse compression 18 and streaking spectroscopy [19][20][21] . The temporal structuring of electron probe beams is facilitated by time-dependent fields in the radio-frequency [22][23][24] , terahertz 18,25 or optical domains. Promising a further leap in temporal resolution, recent findings suggest that ultrafast electron diffraction and microscopy with optically phasecontrolled and sub-cycle, attosecond-structured wave functions may be feasible 8,[26][27][28][29][30] .…”

“…Promising a further leap in temporal resolution, recent findings suggest that ultrafast electron diffraction and microscopy with optically phasecontrolled and sub-cycle, attosecond-structured wave functions may be feasible 8,[26][27][28][29][30] . Specifically, light-field control may translate the temporal resolution of ultrafast transmission electron microscopy (UTEM) 31,32 and electron diffraction (UED) 10,33 , currently at about 200 fs 34 and 20 fs 14,23 , respectively, to the range of attoseconds 26,27,35 . However, such future technologies call for means to both prepare and fully analyze the corresponding quantum states of free electrons.…”

Imaging population transfer in atoms with ultrafast electron pulses

“…Nowadays, electron pulses with femtosecond (fs) duration have been reported [10][11][12][13][14]. Recently, single-electron pulses with a full-width at half-maximum (fwhm) duration of 28 fs have been demonstrated [15]. Various schemes for further compression of these pulses to attosecond durations [16][17][18][19][20] and for reaching attosecond resolution by optical gating [21] have been proposed.…”