Skip to main content
News   |   Events   |   Safety   |   CHESS-U   |   InSitμ   |   MacCHESS   |   CLASSE

X-RAY RUNS: Apply for Beamtime

2017  Nov 1 - Dec 21

2018  Feb 7 - Apr 3
2018  Proposal/BTR deadline: 12/1/17

2018  Apr 11 - Jun 4
2018  Proposal/BTR deadline: 2/1/18

The F2 station at CHESS currently supports several high-energy (40+ keV) X-ray diffraction techniques for probing the crystalline microstructure of structural engineering materials and the loads (stresses) applied to these crystals comprising the microstructure during in-situ thermomechanical testing. The failure of these materials begins at the microscale, often deep within the bulk of the deforming material, with the formation of voids and cracks. Typically, failure of this type must be studied forensically using destructive optical or electron microscopy post-mortem, making the causes of the failure difficult to determine. With new in-situ high-energy X-ray techniques, long standing questions regarding the initiation of material failure can be answered, improving our ability to produce lightweight, long-lasting engineering components.

These X-ray techniques work by inferring the spacing and orientation of planes of atoms from diffracted X-ray intensity. From the changes in atomic spacing (lattice strain), loads acting upon individual crystals are determined, while changes to sample microstructure are derived from reorientation of the crystal lattice. Depending on the initial microstructure of the specimen and the desired loading rate, the lattice strain and orientation of crystals can be found on a grain-by-grain basis or averaged over sets of grains which share a common crystallographic orientation (orientation averaged). In addition, X-ray computed microtomography can also be performed at the F2 station in order to further characterize specimens before and during loading. Below are brief descriptions of experiments that can be performed at F2 station, along with recommendations to help prospective users decide what type of experiment will be optimal for the material system and loading conditions of interest.

    General considerations for all experiments:
  • Measurements can be performed during in-situ thermomechanical loading or ex-situ on pre-deformed (cold-worked) specimens.
  • Specimens can be loaded incrementally and scanned or diffraction measurements can be performed continuously at strain rates up to 10-1 s-1 (orientation averaged measurements only).
  • Diffraction experiments are typically performed in transmission, but orientation averaged measurements can be performed in reflection if arrangements are made with the science staff.
  • In-situ testing on load frames at CHESS requires specimen designs which fit into existing specimen grips.
  • Users are encouraged to work with the science staff to develop their own custom loading solutions if the load frames available or recommended specimen geometries do not meet their needs.
  • New users can arrange to be trained in methods to process raw X-ray diffraction images / radiographs into relevant mechanical quantities and microstructural descriptions.

Single Grain Lattice State Measurements

This technique uses the positions of multiple diffraction peaks measured on a large-panel area detector placed ~1 m from the sample to reconstruct information about individual diffracting grains including: elastic strain tensor, lattice orientation, and position in the sample. The diffraction peaks from grains in the volume are collected as a sample rotates at a fixed load. With knowledge of the single crystal elastic moduli of the material to be studied, the complete grain average stress tensors can also be determined. In addition, advanced analysis techniques may also be applied by users to estimate intragranular quantities such as lattice misorientation and dislocation content in the grains.

    When planning to perform this experiment on a material of interest:
  • Maximum recommended cross section dimension is 1.5 mm (previously employed specimen design drawings can be obtained by contacting the science staff).
  • Lattice state reconstructions are most successful when between 50-500 grains are in the diffraction volume.
  • Recommended average grain diameter is between 50-250 μm, with few grains smaller than 20 μm.
  • Uncertainty of single grain quantities is related to the condition of diffraction peaks (spots), so ideally the samples should have minimal dislocation content prior to testing.
  • To obtain information about every crystal in the diffraction volume, the sample must be rotated relative to the beam. Each rotation scan of a diffraction volume takes approximately 10 minutes.
  • Typical uncertainties for lattice strain measurements are on the order of 10-4, 0.01° for lattice orientation, and 10μm for crystal positions.

Grain Orientation Spatial Mapping Measurements

This technique reconstructs the spatial variation of crystallographic lattice orientation from the positions and shapes of diffraction peaks on a coupled scintillator optical imaging system placed ~10 mm from the sample. The reconstruction technique employed uses grain orientations measured independently (with the previously described technique) as trial solutions to build a 3-D map of the most-likely crystallographic orientation for a position within a diffraction volume. To build the map, at each position in a discretized diffraction volume, diffracted X-rays are simulated using the trial orientations. The trial orientation at a point which best matches the experimental diffraction data measured is then chosen as the most likely lattice orientation for that position in space. This technique is best suited for non-destructively estimating the size and shapes of grains prior to loading and in-situ during the early stages of plastic deformation.

    When planning to perform this experiment on a material of interest:
  • Maximum recommended cross section dimension is 1.5 mm.
  • Each rotation scan of a diffraction volume takes approximately 2-3 hours.
  • Reconstructions are most successful with a smaller beam height (minimum height is 25 μm, maximum height 400 μm), however, multiple scans will be necessary to reconstruct a large volume. Users must decide on a trade-off between quality of reconstruction and the total size of volume probed.
  • The samples should have minimal dislocation content prior to testing. Therefore, samples should not be cold-worked prior to testing and ideally sample material is annealed after manufacture.
  • It is recommended that the distribution of grain volumes be within one order of magnitude.
  • Minimum reconstruction resolution is equivalent to the effective pixel size of 1.48μm at maximum focus.

 Figure 1: 2-D cross section of an orientation map of grains in a copper alloy. Different colors correspond to different crystal orientations (grains).

Orientation Averaged Measurements

In the single grain experiments, the diffraction volume typically contains hundreds of crystals and the diffracted intensity appears as distinct peaks (spots) on the detector. With a smaller grain size / beam size ratio, the diffraction spots overlap, and complete Debye-Scherrer diffraction rings are measured on a large panel area detector ~1 m from the specimen. These techniques, which are often called ‘powder’ experiments even though the samples are solid, are therefore well suited for small grain sizes or cold-worked specimens.

Each point along the Debye-Scherrer ring has contributions from many grains in the diffraction volume which share a common crystallographic direction. From the distribution of maximum X-ray intensity along the Debye-Scherrer rings measured at different sample orientations, the probability distribution of crystallographic orientation in the diffraction volume can be determined (orientation distribution function or ODF). In addition, from the distribution of X-ray position along the Debye-Scherrer rings measured at different sample orientations, the distribution of lattice strain as a function of crystal orientation can be determined (lattice strain distribution function or LSDF). To make these measurements in-situ, sample loading must be halted and the sample must be rotated to probe a large-subset of lattice plane orientations in the diffraction volume. However, complete distributions may not be necessary for all experiments and the diffraction geometry can be optimized to monitor lattice planes of interest during continuous thermo-mechanical loading. In addition, orientation averaged lattice strain measurements can be used to quantify residual stress in ex-situ specimens.

    When planning to perform this experiment on a material of interest:
  • Measurements can be performed on numerous positions along large specimens, specimen geometry is only limited by X-ray absorption.
  • Measurements are best suited for materials that have small grain sizes (<10 μm diameter) or have been previously cold worked: complete Debye rings should diffract from the specimens, not peaks.
  • If only a few lattice plane orientations need to be probed (such as lattice planes with normal along and perpendicular to the loading direction), the sample does not need to be rotated and continuous framing can be performed on the specimen at up to 10 hz, enabling real-time ‘movies’ of deformation.
  • For complete ODF or LSDF measurements, the sample is illuminated at numerous orientations as the sample rotates, often about multiple axes. Typical scan time for a diffraction volume is 5-30 minutes.
  • Lattice strain measurement resolution is on the order of 10-4.

 Figure 2: A 2-D map of orientation averaged lattice strain around a crack grown by cyclic loading in an aluminum alloy.

Computed Microtomography

Variation in X-ray absorption is used to produce contrast on X-ray radiographs. From series of radiographs taken as a sample rotates, 3-D maps of the absorption coefficient in the specimen can be reconstructed. From these absorption coefficient maps, the presence of secondary phases or voids can be inferred.

    When planning to perform this experiment on a material of interest:
  • High aspect ratio defects (such as a thin crack) require large amounts of projections to reconstruct, so scan time will increase.
  • Filtered-back-projection reconstructions can be performed at the station. Advanced users must supply their own iterative reconstruction solvers if necessary.
  • Minimum reconstruction resolution is equivalent to the effective pixel size of 1.48μm at maximum focus.
  • Large variation in absorption coefficient (for example a high Z and low Z material together) within a specimen will often produce artifacts in the reconstruction that are difficult to remove.



Submitted by: Darren Pagan, CHESS, Cornell University