Femtosecond Electron Imaging and Spectroscopy

Proceedings of the Conference on Femtosecond Electron Imaging and Spectroscopy, FEIS 2013, December 9–12, 2013, Key West, FL, USA

Martin Berz; Philip M. Duxbury; Kyoko Makino1; Chong-Yu Ruan    Michigan State University, East Lansing MI 48824, USA
1 Corresponding author: email address: makino@msu.edu

1 Introduction

We are witnessing tremendous opportunities in ultrafast sciences with the development of extremely bright radiation sources to investigate the structure and spectroscopy of matter with atomistic space and femtosecond time resolution. While generally a strong focus has been on X-ray sources—notably free electron laser (FEL) sources—the use of femtosecond electron pulses has also shown enormous promise in the last decade, especially in the investigation of materials from the sub-micrometer down to the angstrom scale, facilitated by the high sensitivity of electron scattering and the relative ease in designing electron optics for imaging and diffraction from nanomaterials. Moreover, important innovations have been achieved by incorporating ultrafast photoemission sources into various electron microscope setups. Most recently, a new trend of integrating the FEL high-brightness electron beam concept into the ultrafast electron diffraction and microscope system design is likely to open up new prospects and applications of femtosecond diffraction, imaging, and spectroscopy with high throughput.

The conference on Femtosecond Electron Imaging and Spectroscopy (FEIS 2013) was held on December 9–12, 2013 in Key West, Florida. FEIS 2013 built on the potential synergy between related technology developments and various emerging scientific opportunities and brought together leaders engaged in cutting-edge development of high-brightness electron and X-ray beam systems and their applications to frontier science problems. FEIS 2013, the first in this series, was organized with the goal of initiating conversation between different communities with the following objectives in mind: (1) to review the current state-of-the-art development and open issues of ultrafast electron imaging technologies; (2) to discuss emerging scientific opportunities enabled by ultrafast imaging and spectroscopy; (3) to identify the key technical challenges in the design and applications of ultrafast electron imaging systems; and (4) to forge cross-fertilization between the electron microscopy, accelerator and beam physics, and ultrafast communities, and to have experimentalists and theorists address common challenges and promote synergistic developments.

1.1 Synopsis of FEIS 2013

1.1.1 Current Status of Ultrafast Imaging and Spectroscopy

Functional imaging and spectroscopy at the local level with atomic, electronic, and magnetic sensitivity are highly desirable for understanding structure-property relationships at the nanometer-length scale and in complex materials. Y. Zhu (page 26) presented an overview of the broad scientific opportunities accessible by utilizing high-energy electrons, including atomic imaging, quantitative electron diffraction, energy-loss spectroscopy, and Lorentz and in situ microscopy, with an emphasis on understanding the materials’ functionality through correlative studies. A community that incorporates electronic, magnetic, thermal, and optical excitations into conventional high-resolution electron microscopes for in situ imaging and spectroscopy studies is rapidly developing. In particular, optical excitations can now routinely be employed on the femtosecond timescale, presenting an opportunity for unique photonic control and potentially imaging at high temporal resolution. Ultrafast electron imaging and spectroscopy represents a natural next step of modern electron microscope development.

To form a diffraction pattern or image, typically 105 to 107 electrons are required. In time-resolved electron microscopy, diffraction, and spectroscopy systems, the electron sources are triggered by pulsed lasers, so the electron beams are delivered in discrete bunches, rather than a steady, diluted stream. So-called space charge effects emerge due to the strong electron-electron interaction within a single photoelectron bunch, which may manifest itself in different forms (i.e., virtual cathode, defocusing, and stochastic blur, as discussed later). Several active technologies cleverly circumvent space charge effects and have achieved significant improvements in temporal resolution using electron microscopy, diffraction, and spectroscopy. G. H. Campbell (page 15) presented the dynamic transmission electron microscope (DTEM) project at Lawrence Livermore National Laboratory using the single-shot approach. By initiating intense photoelectron pulses using a 10-ns laser, the average distance between electrons, even at the 108 electron per pulse level, is more than 100 μm apart, suffering nearly no space charge effect except at the acceleration stage and near the focal plane. Single-shot imaging of microstructure formation, including the kinetics of nucleation and phase transitions in semiconductors, phase change materials, and intermetallic compounds at combined ~ 10 ns–10 nm spatiotemporal resolution, have been achieved using the DTEM.

In contrast, by operating at a high repetition rate (~ 100 MHz), as presented by S. T. Park (page 21), near-single-electron-pulse ultrafast electron microscopes (UEMs) developed at California Institute of Technology are used to study highly reproducible site-specific events, such as dynamical modes of nanomechanical systems and surface plasmons. The fs single-electron pulses, initiated on a LaB6 filament, are fully compatible with the existing electron optical system in a TEM, largely preserving its high spatial resolution and achieving in practical implementations an impressive sub-ps-nm resolution in a stroboscopic setup, where hundreds of thousands or more diffraction data sets are collected at each delay time. The concurrence of ultrashort electron probing and fs laser excitation also enables a new modality of imaging, termed photon-induced near field electron microscopy (PINEM), that has been used to map the optically driven charge density distribution of nanoparticle plasmons. The mechanism and implications of such studies were discussed by S. T. Park's and in the talk by B. Barwick's (page 14). Both approaches are operated by modifying a conventional 100–200 keV TEM, maintaining the capability to retrieve local information.

So far, the most widely employed fs imaging protocol is the diffraction mode. This ultrafast electron diffraction (UED) method initiated the field of electron-based ultrafast imaging; it was introduced in the 1990s first in gas-phase studies of chemical reactions and nonequilibrium molecular dynamics, not long after the development of the largely optical spectroscopy–based fs-chemistry. The timely development of single-electron-sensitive CCDs equipped with pixilated electron amplification and Ti-Sapphire amplified laser...