X-ray pulse compression using strained crystals

https://doi.org/10.1016/S0030-4018(02)01253-1Get rights and content

Abstract

The use of strained crystals to time-compress chirped X-ray free-electron laser (FEL) pulses is proposed. The chirped beam diffracts from a crystal in which the strain varies with depth. Low-Z crystals are favored to reduce absorption over the required path lengths and small Bragg angles are favored to increase the diffraction efficiency of the strained crystal. The best compression ratios found for a 230 fs, 0.5% chirped beam at a wavelength of 1.54 Å are achieved with graphite and diamond crystals. The calculations indicate for graphite that the pulse may be compressed in time by a factor of up to 10 with an efficiency such that this produces 6.8 times as many photons as a comparable time-slicing method.

Introduction

Synchrotrons have had an enormous impact of many areas of science and technology. X-rays of very high brightness are now available from modern third-generation sources. However there appears to be an insatiable demand for X-rays of even greater brightness, or coherence, and work towards fourth generation, free-electron laser (FEL) X-ray sources [1] is now being actively pursued.

The Linac Coherent Light Source (LCLS) [2] is such a proposed fourth-generation source and will produce X-rays with a high-transverse coherence in the wavelength range 1.5–15 Å. This device is proposed for construction at the Stanford Linear Accelerator Center (SLAC). The photon output will consist of high brightness, transversely coherent pulses with duration 230 fs and relative spectral bandwidth (Δλ/λ) of 0.05%. At a wavelength of 1.5 Å the peak power of the pulse will be 9 GW, or approximately 2×1012 photons per pulse. The peak brightness of the source will be about ten orders of magnitude greater than current third-generation synchrotron sources [3].

A major area of application for this instrument will require high fluence X-ray pulses with a pulse length somewhat shorter than the 230 fs pulses currently envisaged. Indeed, the proposed biological imaging experiments will need a pulse length of 20 fs or less [4]. Shorter X-ray pulses may be generated by either compressing the pulse using techniques analogous to those used for optical lasers, or selecting a small time section out of the X-ray pulse (time-slicing). It has been proposed to “chirp” the FEL electron pulse [5] to imprint a time measure on the X-ray pulse and use this as the basis for the generation of shorter X-ray pulses. As an example of a time-slicing approach, it would in principle be possible to extract a short segment of the pulse with a monochromator of appropriate bandwidth [6], [7]. While it is possible to envisage simple time-slicing approaches, they are inevitably only able to use a fraction of the total pulse energy (equal to the fractional reduction in pulse length). Pulse compression, on the other hand, attempts to preserve the total pulse energy. The short wavelength and lack of purely refracting materials means that a simple adaptation of the ideas developed for optical lasers, based on dispersive elements such as gratings [8], [9], lead to impractical geometries and poor efficiencies that do not offer a significant advantage over time slicing. The aim of this paper is propose and analyze a technique for compressing a chirped X-ray pulse by diffraction from a strained crystal.

Section snippets

The strained crystal compressor concept

The basic concept of the strained crystal compressor is that the lattice spacing varies with depth into the crystal and different wavelengths of the chirped beam reflect at different depths according to Bragg diffraction (see Fig. 1). The different wavelengths in the pulse therefore experience different optical paths in such a manner as to compress the pulse. That is, the leading wavelength of the pulse, which diffracts more deeply in the crystal, is delayed by an entire pulse length relative

Conclusions

This paper has presented a simple technique that shows promise of being an effective and efficient method for the compression of X-ray FEL pulses to the length at which they become interesting for biological imaging experiments. The method should be simple to implement using technology that is largely available now. An interesting remaining challenge is the development of the technology to measure the length of X-ray pulses in the femtosecond time domain.

Acknowledgements

The authors acknowledge helpful discussions with Alan Wootton, John Arthur, Carl Schroeder, and Jerry Hastings. This work was performed under the auspices of the US Department of Energy by the University of California, Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48.

References (21)

  • A.K. Freund et al.

    Nucl. Instr. Meth. A

    (2001)
  • J.B. Murphy et al.
  • J. Arthur, SLAC report 521 (1998). Available from...
  • J. Arthur et al.

    Proc. SPIE

    (2001)
  • R. Neutze et al.

    Nature

    (2000)
  • P. Emma, LCLS-TN-00-6...
  • R. Tatchyn et al.

    Proc. SPIE

    (2001)
  • R. Bionta, LCLS-TN-00-7...
  • C. Pellegrini, LCLS-TN-99-5...
  • V. Bharadwaj, A. Chao, M. Cornacchia, P. Emma, T. Kotseroglou, P. Krejcik, I. Lindau, H.-D. Nuhn, G. Stupakov, R....
There are more references available in the full text version of this article.

Cited by (12)

  • Enhanced X-ray free-electron-laser performance from tilted electron beams

    2017, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
    Citation Excerpt :

    First, it is straightforward to generate multi-color XFEL pulses by simply adding a multiple-slotted foil [46] at the undulator entrance, since only the electrons traveling through the slots will not be deteriorated by the foil and therefore will produce XFEL radiation. Second, it may be possible to obtain ultra-short XFEL pulses by compressing the large-bandwidth XFEL radiation [47,48]. Scientific fields such as bioimaging and non-linear optics are demanding XFEL pulses with shorter pulses and higher power than the ones obtained in standard facilities, see for instance Refs. [49–51].

View all citing articles on Scopus
View full text