[Paper Review] Proposal for a scheme to generate 10 TW-level femtosecond x-ray pulses for imaging single protein molecules at the European XFEL

Svitozar Serkez, Vitali Kocharyan+4 more|arXiv (Cornell University)|Jun 4, 2013

Advanced X-ray Imaging Techniques12 references3 citations

TL;DR

This paper proposes a cost-effective upgrade to the European XFEL using self-seeding, undulator tapering, and an emittance spoiler foil to generate 10 fs, 50 mJ x-ray pulses at 3.5–4 keV with peak power up to 10 TW—enabling single protein molecule imaging at near-atomic resolution via 'diffraction before destruction.' The scheme achieves a 100-fold increase in peak power over baseline SASE sources, validated through start-to-end simulations using a 1 nC electron bunch.

ABSTRACT

Single biomolecular imaging using XFEL radiation is an emerging method for protein structure determination using the "diffraction before destruction" method at near atomic resolution. Crucial parameters for such bio-imaging experiments are photon energy range, peak power, pulse duration, and transverse coherence. The largest diffraction signals are achieved at the longest wavelength that supports a given resolution, which should be better than 0.3 nm. We propose a configuration which combines self-seeding and undulator tapering techniques with the emittance-spoiler method in order to increase the XFEL output peak power and to shorten the pulse duration up to a level sufficient for performing bio-imaging of single protein molecules at the optimal photon energy range, i.e. around 4 keV. Experiments at the LCLS confirmed the feasibility of these three new techniques. Based on start-to-end simulations we demonstrate that self-seeding, combined with undulator tapering, allows one to achieve up to a 100-fold increase in peak-power. A slotted foil in the last bunch compressor is added for x-ray pulse duration control. Simulations indicate that one can achieve diffraction to the desired resolution with 50 mJ (corresponding to 1e14 photons) per 10 fs pulse at 3.5 keV photon energy in a 100 nm focus. This result is exemplified using the photosystem I membrane protein as a case study.

Motivation & Objective

• To enable high-resolution single biomolecule imaging at the European XFEL using x-ray pulses with sufficient peak power and ultrashort duration.
• To address the limitation of baseline SASE undulators, which saturate at ~50 GW, far below the 10 TW required for optimal imaging of single protein molecules.
• To develop a cost-effective upgrade path using established techniques—self-seeding, undulator tapering, and emittance spoiler foil—without major infrastructure changes.
• To achieve a 100-fold increase in peak power and pulse duration control down to 10 fs, enabling 10^14 photons per 10 fs pulse at 3.5 keV.
• To demonstrate feasibility for 0.3 nm resolution imaging of 10 nm-sized proteins like photosystem I using realistic simulation models.

Proposed method

• Self-seeding via a crystal monochromator in a single-chicane setup to narrow the bandwidth and enhance spectral brightness.
• Undulator tapering to increase the FEL gain and boost output peak power by coherently enhancing the radiation emission along the undulator.
• Emittance spoiler foil placed at the center of the final bunch compressor (BC3) to selectively increase transverse emittance in most of the electron bunch while preserving a narrow, low-emittance temporal slice.
• The slotted foil induces Coulomb scattering in the bulk of the beam, suppressing FEL gain there, while the unspoiled central slice generates a short, high-gain x-ray pulse.
• Chromatic dispersion in the chicane transforms the energy chirp into a transverse bunch tilt, enabling precise localization of the unspoiled slice at the foil center.
• Start-to-end simulations using a 1 nC electron bunch model the full beamline from electron gun to final x-ray pulse, incorporating beam dynamics, FEL gain, and radiation propagation.

Experimental results

Research questions

• RQ1 Can the peak power of SASE-based XFEL sources be increased by up to 100-fold to reach 10 TW levels using existing techniques?
• RQ2 Can femtosecond x-ray pulses as short as 10 fs be generated at the European XFEL using a minimal hardware upgrade, specifically an emittance spoiler foil?
• RQ3 Is it feasible to achieve 10^14 photons per 10 fs pulse at 3.5 keV with a 100 nm focus, sufficient for single protein molecule imaging at 0.3 nm resolution?
• RQ4 Can self-seeding and undulator tapering jointly enhance spectral brightness and peak power without compromising transverse coherence?
• RQ5 What is the required photon fluence and pulse duration to achieve sufficient signal-to-noise for 3D reconstruction of single biomolecules like photosystem I?

Key findings

• The combination of self-seeding, undulator tapering, and emittance spoiler foil enables a 100-fold increase in peak power, reaching up to 10 TW for 10 fs x-ray pulses.
• A 50 mJ x-ray pulse at 3.5 keV with 10 fs duration and 100 nm focus delivers 10^14 photons, satisfying the fluence requirement of ~10^22 photons/mm² for 0.3 nm resolution imaging.
• Simulations confirm that a 10 fs pulse with 10^14 photons at 3.5 keV produces a diffraction signal sufficient for 3D structure determination of 10 nm-sized proteins like photosystem I.
• The emittance spoiler foil reduces the effective electron bunch length in the FEL interaction region, enabling pulse compression to ~10 fs without additional hardware in the undulator line.
• The radially averaged scattering intensity profile shows detector pixel values well above one photon per pixel up to the detector edge, indicating strong signal-to-noise for single-shot 3D reconstruction.
• The scheme is compatible with existing European XFEL infrastructure and requires only minimal modifications—specifically, a single-chicane self-seeding setup and a slotted foil in BC3—making it a cost-effective upgrade path.