VOLUME 80, NUMBER 2 PHYSICAL REVIEW LETTERS 12 JANUARY 1998 Measurements of High Gain and Intensity Fluctuations in a Self-Amplified, Spontaneous-Emission Free-Electron Laser M. Hogan, C. Pellegrini, J. Rosenzweig, G. Travish, A. Varfolomeev,* S. Anderson, K. Bishofberger, P. Frigola, A. Murokh, N. Osmanov,* S. Reiche, ² and A. Tremaine Department of Physics and Astronomy, University of California, Los Angeles, California 90024 (Received 1 July 1997) We report measurements of large gain for a single pass free-electron laser operating in self-amplified spontaneous emission (SASE) at 16 mm starting from noise. We also report the first observation and analysis of intensity fluctuations of the SASE radiation intensity in the high gain regime. The results are compared with theoretical predictions and simulations. [S0031-9007(97)04953-3] PACS numbers: 41.60.Cr, 41.60.Ap An x-ray laser would offer a unique way to explore the structure of matter at the atomic and molecular scale. Among the various schemes proposed to reach this wave- length region, the free-electron laser (FEL), operating without mirrors in a self-amplified spontaneous emission (SASE) mode, as proposed in [1], and independently in [2], offers a favorable scaling law [3]. It has also been shown [4] that utilizing state of the art linear accelerators and electron sources it is possible to build an x-ray SASE FEL, and this has led to two major proposals to build a SASE x-ray FEL, one at SLAC [5], the other at DESY [6]. The theory on which the SASE x-ray FEL is based [7 – 9] has been developed over many years, but the experimental data to support it are few and incomplete. Very large gain in SASE has so far been observed in the centimeter to millimeter waves [10–12] and in the medium infrared (IR) at Los Alamos [13]; recently, gain in the near IR has been observed at Orsay [14] and at Brookhaven [15]. The intensity distribution function has been previously measured only for spontaneous undulator radiation [16], with no amplification, and long bunches. In this paper we report the results of measurements, at 16 mm, of large gain and of the intensity distribution function for amplified radiation, and for a short bunch length. When a beam traverses an undulator it emits electro- magnetic (EM) radiation at the wavelength l  l u 1 1 K 2 22g 2 (where l u is the undulator period gmc 2 the beam energy, and K the undulator vector potential nor- malized to mc 2 ). If, as it is the case in SASE, there is no input EM field, radiation is emitted when the beam cur- rent is not uniform, and has a Fourier component i v at v  2p cl. The EM field is then proportional to i v and the intensity to ji vj 2 . If the bunch length L is much larger than l, and the beam is generated from a thermionic cathode or photocathode i v, and thus the EM field and intensity are stochastic quantities characterized by a distri- bution function and determined by the random initial elec- tron longitudinal distribution. The dependence of ji vj 2 on charge is ji vj 2  Q1 1 FvQ, where Fv is the bunch form factor. The intensity term quadratic in Q is what is called the coherent spontaneous emission (CSE). For L ¿ l, as in our case, and a smooth charge distribution we have Fv ø 1, as we will again discuss later in the paper, and we will neglect this term. For a long undulator the EM intensity can grow expo- nentially along the undulator axis z as I  aji vj 2 e z L g . (1) The power gain length L g is given, in the simple 1D theory, neglecting diffraction and slippage [7,8] by L g  l u 4 p 3pr, where the FEL parameter r is propor- tional to the beam plasma frequency to the power 23, or QsL 13 , Q being the electron bunch charge, s the beam cross section, and L the bunch length. Saturation occurs after about 20 gain lengths, and the radiation intensity at saturation is about r times beam energy. Diffraction, energy spread, and slippage S  lN u , can increase the gain length over the 1D value if the conditions ´#l4p , s E ,r, S , L are not satisfied, where ´ is the beam emittance and N u the number of undulator periods. In this experiment we measure the gain length and the intensity distribution function for a SASE FEL at 16 mm. The measurements have been done using the Saturnus linac [17], consisting of a 1 1 2 cell Brookhaven National Laboratory photocathode RF gun, and a PWT accelerating structure [18]. The linac is followed by a beam transport line and an undulator built at the Kurchatov Institute [19], providing focusing in both transverse planes with a beta function of approximately 0.1 and 0.4 m, and with field errors of about 0.25%. The characteristics of the electron beam, the undulator, and of the undulator radiation are given in Table I. The linac operates at 5 Hz, with 2.5 ms long macropulses, and one electron bunch per macropulse. The beam transport line from linac to the undulator has steering magnets to control the beam trajectory, and beam instrumentation including slits to measure the emittance [20], an integrating current transformer (ICT), and Faraday cups to measure the beam charge, phosphor screens to measure the beam transverse profiles, and a dipole spectrometer to measure the energy and energy spread. 0031-9007 98 80(2) 289(4)$15.00 © 1998 The American Physical Society 289