Work supported in part by US Department of Energy contract DE-AC02-76SF00515 Energy doubling of 42 GeV electrons in a metre-scale plasma wakefield accelerator Ian Blumenfeld 1 , Christopher E. Clayton 2 , Franz-Josef Decker 1 , Mark J. Hogan 1 , Chengkun Huang 2 , Rasmus Ischebeck 1 , Richard Iverson 1 , Chandrashekhar Joshi 2 , Thomas Katsouleas 3 , Neil Kirby 1 , Wei Lu 2 , Kenneth A. Marsh 2 , Warren B. Mori 2 , Patric Muggli 3 , Erdem Oz 3 , Robert H. Siemann 1 , Dieter Walz 1 & Miaomiao Zhou 2 The energy frontier of particle physics is several trillion electron volts, but colliders capable of reaching this regime (such as the Large Hadron Collider and the International Linear Collider) are costly and time-consuming to build; it is therefore important to explore new methods of accelerating particles to high energies. Plasma-based accelerators are particularly attractive because they are capable of producing accelerating fields that are orders of magnitude larger than those used in conventional colliders 1–3 . In these accelerators, a drive beam (either laser or particle) produces a plasma wave (wakefield) that accelerates charged particles 4–11 . The ultimate utility of plasma accelerators will depend on sustain- ing ultrahigh accelerating fields over a substantial length to achieve a significant energy gain. Here we show that an energy gain of more than 42 GeV is achieved in a plasma wakefield accel- erator of 85 cm length, driven by a 42 GeV electron beam at the Stanford Linear Accelerator Center (SLAC). The results are in excellent agreement with the predictions of three-dimensional particle-in-cell simulations. Most of the beam electrons lose energy to the plasma wave, but some electrons in the back of the same beam pulse are accelerated with a field of 52 GV m 21 . This effectively doubles their energy, producing the energy gain of the 3-km-long SLAC accelerator in less than a metre for a small frac- tion of the electrons in the injected bunch. This is an important step towards demonstrating the viability of plasma accelerators for high-energy physics applications. In a plasma wakefield accelerator large-amplitude electric fields result from space-charge waves excited by the passage of an ultra- relativistic electron beam through a plasma 12 . A fully ionized plasma can be formed in a neutral vapour when the radial electric field of the electron beam exceeds the field ionization threshold 13 . The ionization occurs in a narrow region in the front of the beam. This ionization front produces a plasma that has a radius much larger than the beam itself. If the beam density exceeds the plasma density, the plasma electrons are expelled from the volume of the electron pulse, leaving a column of more massive ions behind 14 . Subsequently, the expelled plasma electrons are pulled back (by the ions) to the beam axis behind the pulse, overshoot, and set up a space-charge oscillation or wake. The longitudinal field of this wake varies continuously along the pulse, decelerating its core but accelerating the particles in the back. The ion column also provides a focusing force 15 that guides the beam over many diffraction lengths, allowing an efficient transfer of the beam energy to the wake. This force also causes the transverse size of the beam to oscillate as it propagates through the plasma—the so- called betatron oscillations (see Supplementary Movie 1). Recent plasma wakefield accelerator experiments have shown high-gradient acceleration of electrons using a 10-cm-long plasma 11 . To obtain energy gains of interest to high-energy physics, these high gradients must be extended over metre-scale plasmas. Such an exten- sion transitions the plasma wakefield accelerator from a regime in which the drive beam has no time to distort, deplete or go unstable, to a regime in which it is significantly depleted in energy, deformed owing to combined effects of diffraction and multiple transverse oscillations, and possibly goes unstable because of the electron-hose instability 16 . This work is in this latter regime. A schematic of the experimental set-up is shown in Fig. 1. In the present work carried out at the Final Focus Test Beam facility at SLAC, the nominally 50-femtosecond-long electron beam contain- ing 1.8 3 10 10 particles is focused to a spot size of ,10 mm at the entrance of an 85-cm-long column of lithium vapour with a density n e of 2.7 3 10 17 cm 23 . The nominally 42 GeV beam has a correlated energy spread of approximately 1.5 GeV, with electrons in the front of the beam at higher energies than those at the back. The beam exiting the plasma traverses a metre-long dipole magnet, which disperses the beam electrons according to their energy. The transverse distribution of the dispersed electrons is measured at two distances (planes 1 and 2 in Fig. 1) downstream of the dipole magnet to distinguish the energy changes of the electrons from their possible transverse deflection due to the plasma. Images of the dispersed electrons are recorded along with the relevant beam parameters on a shot-to-shot basis. The energy gain achieved for each shot is determined as described in the Methods section. Figure 2 shows one example of the electron energy 1 Stanford Linear Accelerator Center, 2575 Sand Hill Road, Menlo Park, California 94025, USA. 2 University of California Los Angeles, 405 Hilgard Avenue, Los Angeles, California 90095, USA. 3 University of Southern California, University Park, Los Angeles, California 90089, USA. θ Plasma Spectrometer magnet Plane 1 Plane 2 85 cm Electron Pulse 218 cm 86 cm 100 cm 0 θ 1 Figure 1 | Schematic of the experimental set-up. Two cameras record the energy-dispersed images at planes 1 and 2. A combination of low dispersion at plane 1 and a lower lens magnification on the camera allows a broad energy spectrum of the beam, including energy gain and loss, to be recorded. A higher dispersion at plane 2 coupled with a larger lens magnification is used to record images showing greater detail of the energy gain. The comparison of these two images allows for an independent measurement of vertical deflection and energy gain, as discussed in the Methods section. February 2007 SLAC-PUB-12363 Submitted to Nature