[Paper Review] Ultrafast X-ray pulse measurement method

Gianluca Geloni, Vitali Kocharyan+1 more|arXiv (Cornell University)|Jan 20, 2010

Particle Accelerators and Free-Electron Lasers5 references11 citations

TL;DR

This paper proposes a novel, low-cost method to measure femtosecond X-ray pulses at X-ray free-electron lasers (XFELs) by using a 'fresh bunch' technique with a magnetic chicane to create a variable delay between the electron bunch and a seed X-ray pulse. The method measures the intensity autocorrelation function of the X-ray pulse via shot-to-shot averaged energy detection, enabling pulse width estimation with a deconvolution factor of 1.5, achieving sub-1.5 μm resolution for 1.4 μm FWHM pulses.

ABSTRACT

In this paper we describe a measurement technique capable of resolving femtosecond X-ray pulses from XFEL facilities. Since these ultrashort pulses are themselves the shortest event available, our measurement strategy is to let the X-ray pulse sample itself. Our method relies on the application of a "fresh" bunch technique, which allows for the production of a seeded X-ray pulse with a variable delay between seed and electron bunch. The shot-to-shot averaged energy per pulse is recorded. It turns out that one actually measures the autocorrelation function of the X-ray pulse, which is related in a simple way to the actual pulse width. For implementation of the proposed technique, it is sufficient to substitute a single undulator segment with a short magnetic chicane. The focusing system of the undulator remains untouched, and the installation does not perturb the baseline mode of operation. We present a feasibility study and we make exemplifications with typical parameters of an X-ray FEL.

Motivation & Objective

• To address the lack of reliable methods for measuring sub-10 fs X-ray pulses at XFEL facilities.
• To develop a technique that enables direct measurement of ultrashort X-ray pulse durations using only minimal hardware modifications.
• To provide a non-invasive, baseline-compatible solution that does not disrupt the standard operation of XFELs.
• To demonstrate feasibility of measuring the intensity autocorrelation function of X-ray pulses via electron bunch seeding and variable delay.
• To enable accurate pulse width estimation using only integrated energy measurements and simulation-based deconvolution.

Proposed method

• The method uses a magnetic chicane installed in place of a single undulator segment to delay the electron bunch relative to the X-ray pulse, washing out microbunching and enabling a variable delay.
• The X-ray pulse from the first undulator segment acts as a seed for the second undulator segment, which is resonant at the same wavelength and amplifies the signal via seeded FEL gain.
• The shot-to-shot averaged power in the second undulator is measured as a function of delay τ, yielding the intensity autocorrelation function A(τ) = ∫<P(t−τ)><P(t)> dt.
• The autocorrelation trace is obtained experimentally by scanning the delay using an X-ray optical delay line within the chicane, enabling full trace acquisition.
• For the simplified version, only a magnetic chicane is used without an optical delay line, requiring simulation-based reconstruction of the missing autocorrelation data.
• The method relies on 1D FEL theory, with power growth modeled as <P(t)> = P₀ exp[2L_w Re(Λ(t))], and the final autocorrelation is derived from the product of two such power envelopes.

Experimental results

Research questions

• RQ1 Can the intensity autocorrelation of a femtosecond X-ray pulse be measured using only a magnetic chicane and a seeded FEL configuration?
• RQ2 What is the achievable temporal resolution of the proposed method for measuring X-ray pulse widths?
• RQ3 How does the method perform when only a magnetic chicane is used, without an active optical delay line?
• RQ4 To what extent can the pulse width be accurately estimated from the measured autocorrelation trace using deconvolution?
• RQ5 Can the method be implemented without perturbing the baseline operation of an XFEL facility?

Key findings

• The method successfully measures the intensity autocorrelation function of a femtosecond X-ray pulse using only a magnetic chicane and a seeded FEL configuration, enabling pulse width estimation.
• The full autocorrelation trace is obtained by scanning the delay between the electron bunch and the seed X-ray pulse, with the measured energy per pulse directly yielding A(τ).
• For a 1.4 μm FWHM X-ray pulse, the method estimates the width as 1.3 μm when applying a deconvolution factor of 1.5, showing good agreement with simulation.
• The simplified version using only a magnetic chicane provides a partial autocorrelation trace, which can be reconstructed using simulations and prior knowledge of electron bunch properties.
• The technique is robust and non-invasive, requiring only a single undulator segment to be replaced with a chicane, preserving the baseline focusing structure and operational mode.
• The feasibility study confirms that the method is suitable for typical XFEL parameters, with a measurement accuracy sufficient for practical pulse characterization.