Science Programme

Accessing the nano-structure of condensed matter with high time resolution is one of the primary goals of the MID instrument. This will be possible through conventional X-ray scattering experiments with high time resolution or by taking advantage of the outstanding coherence properties of the SASE radiation. Structural dynamics will become accessible down to 220 ns (in multi-exposure mode) and down into the ps regime by employing split-delay techniques or by special linac filling patterns.

Glass formation and dynamics

When rapidly cooled below the freezing point, most liquids, e.g. organic liquids, metallic alloys and oxides, polymeric materials, water and many others, organize in metastable glassy or amorphous phases. This vitirification process finds numerous applications in material science and industry, but is until today only partly understood. With the significant gain in coherent flux compared to today's most powerful synchrotron sources, experiments can be taken into a new regime, spanning a wide range of timescales, from 10-12 to 103 s, in order to observe the evolution of the dynamics from liquid to glassy behavior. Especially metallic glasses are targeted as ideal candidates to shed light on the glass transition phenomenology, due their simpler matallic bonding based structure. Further new research opportunities span from organic molecular glass formers to various supercooled liquids with various degrees of complexity. Here, the MID instrument will offer exciting new possibilities with both sequential and split-delay XPCS. Both, diffusive dynamics and angular correlations could be tracked at micro- to pico-second timescales, unravelling the molecular dynamics in many different glass formers. Additional stuctural information on local, short range order can be obtained from angular cross-correlation anaylsis, described in the following.

Surface and interface dynamics

X-rays are ideal tools for surface-sensitive measurements thanks to the very limited penetration depth when applied at a grazing incidence angle. So far, major obsticales were the limited scattering power of such small amounts of material and the strong decay of scattering away from the specular reflection. With the European XFEL beam, it will be possible to study surface and interface dynamics at the atomic and molecular scale by X-ray scattering. The self-organization, self assembly, and growth of nano-particles can therefore be followed by coherent X-ray scattering, allowing kinetics and dynamics of such processes to be studied at interfaces with unprecedented time resolution.

The additional option to perform XPCS at photon energies up to 25 keV (or higher) opens up the possibility to probe dynamics at buried liquid–liquid and liquid–solid interfaces or in complex sample environments. For example, the in situ growth and diffusion of nano-particles on reactive solid surfaces plays an important role for many catalytic processes, and the MID setup ensures that user-specific sample environments can be hosted to perform surface chemistry and catalytic experiments under realistic reaction conditions. Further topics to be adressed are capillary wave dynamics, surface glass transitions, biological and soft matter interface dynamics, wetting and de-wetting processes, aggregation and coalescence dynamics as well as phase transitions and critical phenomena.

Non-equilibrium dynamics

Many disordered soft solids, e.g. colloidal gels, clays, polymer gels, and concentrated emulsions show pronounced deviations from simple diffusive dynamics such as compressed, faster-than exponentially decaying correlation functions implying hyper-diffusive, convective-like motion. The wide assortment of soft materials displaying these dynamics suggests a generic underlying mechanism; however, no clear consensus about their microscopic origin has emerged.

Often, a characteristic feature is the out-of-equilibrium behaviour, which can complicate the analysis of XPCS measurements. In this case, the correlation analysis must be performed explicitly as a function of time, or “age”, and the time averaging of the correlation function is replaced by an ensemble averaging leading to a two-times correlation function. Several systems investigated to date display multiple relaxations where the faster dynamics only is indirectly visible. The boost in coherent flux available at MID will allow to record time-resolved correlation functions that are not accessible with today’s X-ray sources. Furthermore, the two-times scheme can also be extended to the
ultrafast XPCS mode with the split-delay technique.

Ultrafast coherent scattering and XPCS

Processes such as ultrafast demagnetization, investigations of frustrated ferromagnetic compounds (“spin ice”), and interplay between spin-orbital and ordering fluctuations, e.g. in complex metal oxides, will in most cases require split-delay XPCS due to the fast timescales of interest. Traditionally, such experiments have been performed by inelastic X-ray or neutron scattering, but often either inelastic techniques have difficulties due to the required energy resolution for the targeted time scale, or direct detection in the time domain is preferred, e.g. to study non-equilibrium dynamic behaviour and non-Gaussian fluctuations.

Access to ultrafast timescales can be achieved by application of split-delay techniques, where a pulse is separated into two parts by a beam splitter. A time delay is introduced between the two pulses by a difference in path length travelled. The two pulses hit the sample and scatter into the detector with a time difference less than 4.5 MHz, therefore the detector will record a sum of two speckle patterns. This double-exposed detector image can be analysed in terms of its contrast, for instance, using speckle visibility techniques, spatial auto-correlations, or intensity histograms, with the potential to obtain a time resolution in XPCS of sub-ns or smaller. This method will allow pushing the fast limits of XPCS, and the split-delay line technique is in principle only limited by the duration of a pulse.

Time-resolved SAXS and WAXS

Due to the high flux and distinguished time structure of the European XFEL, but also due to the multiple experimental schemes possible, the MID station is well suited to perform many "regular" time-resolved scattering experiments. These are already today succesfully performed at third-generation synchrotron sources. However, many studies on phenomena such as folding of proteins, self-assembly of viruses, surfactant or polymer systems or various phase transitions is biological and soft matter systems suffer from the constant need for more flux, in order to study faster processes and weaker signals.

Such studies often make use of specialised sample environments, which are able to homogenously trigger the desired kinetic with high temporal resolution. The MID multi-purpose sample chamber together with the movable long detector arm and X-ray mirrors make many different scattering geometries with various sample environments possible. Therefore, several well-established experimental schemes such as time-resolved SAXS and WAXS and gracing incidence scattering can be perfomed in combination with stopped-flow or magnetic quenching devices, small cryostats, high-precision scanning stages and sample environments for high-speed tomography or imaging.

Biological imaging

The imaging of biological objects and materials is challenging in most of the cases, due to the fragility and radiation sensitivity of the samples, their low-density contrast to X-rays and the need for 3D and atomic resolution. Many different experimental schemes and techniques are proposed, using the outstanding X-ray beam properties of the European XFEL.

Coherent Diffraction Imaging

Coherent Diffraction Imaging (CDI) has appeared as a promising technique to obtain 3D images of living cells. The method of chioce has been Transmission Electron Microscopy (TEM) so far, which however requires the sample to be at cryogenic temperatures as well as sectioned and stained to increase the contrast. Biological CDI at the European XFEL has the potential to increase the resolution of CDI experiments, so far mainly performed at synchrotron sources, matching those of cryo-TEM investigations, together with the opportunity to perform these measurements under the samples' native conditions. Additionally, the high time resolution offered by XFEL experiments is unmatched by other techniques.

Diffraction from 2D lattices

Another important example for new imaging experiments possible at the XFEL is provided by membrane proteins that cannot be crystallized in usual 3D crystalline structures. For the first time, diffraction experiments on single 2D protein lattices could be possible and calculations suggest that radiation damage effects can be avoided in single-shot experiments, even on non-frozen lattices in their natural wet environment. Single protein lattices, typically 0.5 μm in diameter, can be deposited in large arrays on thin SiN membranes. At every tilt angle a new lattice is positioned into the beam so that the diffraction patterns are rotated randomly with respect to each other. The rotation angles can be determined if many different tilt angles are measured and phasing of the diffraction patterns can be achieved by molecular replacement. The advantage of this new methodology is that high-resolution 3D structures of ground and intermediate states of single proteins can be obtained from 2D lattices in the non-frozen state and in a more natural environment. This avoids systematic errors due to 3D crystal packing effects that have appeared as a limiting factor in crystallography of intermediates.


Coherent radiation from X-ray free-electron lasers (FELs) also provides the unique opportunity for imaging biological objects using holographic techniques, for instance studying large, difficult to crystallize membrane proteins complexes or the ribosome from E. coli. A gold sphere can be fixed to the sample by a strand of DNA and the whole assembly sprayed into the X-ray beam using an aerosol injector. In order to obtain the final 3D image of the sample, fourier transform holograms of thousands of these particles need to be analysed and the images classified with respect to the different orientations. In test experiments at synchrotron sources, gold spheres of a few 100 nm had been used to create the reference wave in 3D imaging. In XFEL experiments, the size of the reference object can be downscaled to 2 nm, hence promising a much improved resolution.

Imaging in nanoscience

Visualizing the internal structure of nano- and micron-sized materials can be obtained by using oversampled diffraction patterns in combination with iterative phase retrieval algorithms. This has for example been demonstrated by Coherent X-ray Diffraction Imaging (CDXI) and later technical developments using holographic and ptychographic techniques. CDI on polycrystalline samples, that are typical in materials science, is not trivial, particularly if the goal is to approach atomic resolution. Using a combination of CXDI phase retrieval methods, e.g. the hybrid input-output algorithm, and indexing multiple Bragg spots by 3D diffraction methods could be a possible way forward to address this problem. The possibility to perform these experiments at the MID station with higher photon energies than previously available for coherent scattering techniques is of paramount importance in order to investigate samples of nonvanishing thickness. Together with the exploitation of the time structure of the machine, this will provide new possibilities for in situ CXDI studies in materials science on the ns–ms timescale, or even faster using split-delay techniques.

Angular cross-correlation studies

The calculation of angular correlations from scattering data is a method originally proposed about 30 years ago by Zvi Kam and coworkers, in order to derive 3D molecular structures from solution scattering. Recently, the technique was rediscovered to obtain the single particle form factors from scattering of many particle systems or to uncover local symmetries in glassy colloidal suspensions.

Of high interest is the occurcence of local five fold order symmetries, where the structural arrangement can minimize the energy on the local scale, but it of course prevents the formation of long-range ordering. Hence, these local symmetries may play a decisive role in stabilizing the glassy state. The possibility of revealing bond orientation order in molecular glasses, together with the prospects for uncovering fundamental aspects of the glass transition, is one of the main reasons why this technique will be pursued at the MID station.

Conceptual Design Report

  • Conceptual Design Report: Scientific Instrument MID