To date, the study of ultrafast processes has largely relied on femtosecond optical pulses. Since x rays interact with core electronic levels and hence are effective structural probes, the availability of femtosecond x-ray pulses from laser sources and the intrinsic synchronization between laser and x-ray pulses would make it possible to directly probe changes in atomic structure on ultrafast time scales. Such sources could be complementary to recent light source facilities such as that at the Jefferson Laboratory infrared free electron laser. The vast potential of these short-pulse, multicolor operating scenarios for materials and other research and development is only just beginning to be explored.

The Linac Coherent Light Source (LCLS), which is proposed to be constructed in the 2004–2007 time frame, will utilize the last third of the existing SLAC linac. This linac produces a high-current 5- to 15-GeV electron beam that is bunched into 230-fs slices with a 120-Hz repetition rate. When traveling through an ~100-m-long undulator, the electron bunches will lead to self-amplification of the emitted x-ray intensity in the 800- to 8,000-eV energy range and will function as the first x-ray free-electron laser. The emitted coherent x rays will have unprecedented brightness and hence offer a new window on beam-matter interactions.

Two general classes of experiments are proposed for the LCLS. The first class consists of experiments in which the x-ray beam is used to probe the sample without modifying it, as in most synchrotron source experiments today. In the second class, the LCLS beam is used to induce nonlinear photo-processes or to study matter in extreme conditions. The latter experiments will include pump-probe studies of so-called warm dense matter.

When simulation models were first applied to the study of laser-plasma interactions in the 1960s, lasers were nanoseconds long and computing power was measured in megaflops. The parallel development of femtosecond lasers and teraflop computing has led to a convergence of these factors by more than 10 orders of magnitude! As a result, it has recently become possible for the first time to perform detailed simulations of high energy density beam-plasma experiments using algorithms with very low level approximations in three dimensions with unscaled parameters. It is even possible to begin to use the actual number of particles in simulations that are used in some experiments (e.g., 10¹⁰ electrons in a beam experiment). This is leading to a significant change in the way that simulations are used to study high energy density beam-plasma interactions. While scaled simulations were previously used to test theoretical concepts that could then be applied to experi-

Front Matter (R1-R16)

Executive Summary (1-8)

1. Exordium and Principal Findings and Recommendations (9-33)

2. High Energy Density Astrophysics (34-70)

3. High Energy Density Laboratory Plasmas (71-119)

4. Laser-Plasma and Beam-Plasma Interactions (120-148)

Glossary (149-160)