Applications of Electron Channelling Contrast Imaging

Electron channelling contrast imaging (ECCI) on modern FEG-SEM instruments allows for the routine imaging of lattice defects close to the surface of bulk specimens.  The development of the methodology since its inception in the 1960s is outlined in another editorial ‘Historical Development of Electron Channelling Contrast Imaging’.  This early work established that the relatively low contrast levels coupled with a need for both good spatial resolution and limited convergence of the incident beam meant that a Schottky FEG SEM was required.  Widespread availability of such instruments, their ease of use, and the advantages of using bulk samples makes the method very attractive and competitive with TEM diffraction contrast in many cases. 

The notes below attempt pick-out some of the contributions that have attracted my attention – this by no means an exhaustive list (there are other fine examples in the ECCI collection at ScienceOpen).

Applications to Semiconductors Materials:

Trager-Cowan et al (2006, 2007), – ECCI and EBSD of epitaxial GaN layers grown on sapphire substrates employing the highly tilted geometry of EBSD with forescatter detectors positioned at low take off angles.  Threading dislocations and atomic steps imaged in GaN layers grown to different thicknesses beyond coalescence.  Further observations made on GaN layers grown on substrates patterned with silica stripes forcing epitaxial lateral overgrowth (ELOG) to reduce defect density.  EBSD shows lattice tilting and strains across seed, wing and coalescence regions with ECCI showing threading dislocations swept out of wing regions and into coalescence boundaries.

Picard et al (2007, 2008), Twigg et al (2010) – using a highly tilted sample and forescatter detector geometry compatible with EBSD studies they imaged atomic step terraces and threading dislocation structures on CVD grown 4H-SiC.  Diffraction conditions were inferred from changes in image intensities and contrast as the sample was tilted with orientation measured by EBSD.  Particularly striking images were presented of atomic steps in spiral morphologies expanding out from central screw dislocations which are thought to act as sources for steps required for two-dimension step-flow homoepitaxial growth.

Naresh-Kumar et al (2012) – the Strathclyde group implemented beam rocking so as to acquire ECP from small regions on an FEI Sirion FEG-SEM using non-standard configuration of the electron optics,  The sample was tilted from 30° to 70° and images collected using a purpose built forescatter detector system.  Comparing images of the same dislocations imaged under differing diffraction conditions allowed screw, mixed and edge dislocations to be distinguished.

Kamaladasa et al (2011, 2013) - ECCI studies of dislocations in SrTiO₃ in FEI Quanta 600 using annular BSE detector at the pole piece and near normal incidence sample geometry.  Low (x20) magnification imaging used to generate ECP and align imaging conditions.  Initial paper examines dislocations and establishes imaging methods on large single crystal.  Second paper examines Pt-SrTiO₃-Pt lateral devices establishing similar resistive switching performance for high and low dislocation density SrTiO₃ regions.  Furthermore, ECCI was used to establish that dislocation were only generated in higher power dissipation ‘electroforming’ events.

Carnevale et al (2014) - ECCI used to image lattice defects including threading dislocations, misfit dislocations, and stacking faults, in heteroepitaxially grown samples of GaP on Si.  Low magnification imaging in FEI Sirion FEG-SEM used to generate crude ECP allowing diffraction condition to be set.  A near normal incidence geometry was used with a conventional annular BSE detector at the pole piece.  The utility of ECCI in covering large areas was exploited to show the significant variation in misfit dislocation distribution across the width of the entire sample grown on a 4” Si substrate. 

Yan et al (2015) – use of ECCI to image threading dislocations and ‘meandering loop’ structures of antiphase boundaries in La_0.7Sr_0.3MnO₃ thin films on (110)-oriented SrTiO₃.  Striking contrast is seen at the APBs which is consistent with a displacement vector R=½[001]_LSM.

Yaung et al (2016) – use of ECCI to quantify threading dislocation densities in various GaAs_yP_1−y/Si_1−xGe_x structures for solar cell applications.  ECCI observations made at near normal incidence using an annular BSE detector at the pole piece with ECP generated by low magnification imaging used to control diffraction conditions.  Direct comparison of ECCI with EBIC imaging and defect sensitive etching is presented.

Vilalta-Clemente et al (2017) – ECCI and HR-EBSD used to investigate threading dislocations in In_XAl_1-XN/AlN/GaN high electron mobility transistor (HEMT) structures grown by MOVPE with X nominally 20%.  Effects of different substrates SiC and sapphire on defect density were assessed.  ECCI images of the same region under different (but unknown, as surface roughness prohibited ECP collection) allowed pure edge dislocations to be distinguished from screw and mixed types and their densities to be determined.  Assessment of dislocation densities from lattice curvature measured by HR-EBSD gave further constraint and allowed densities to be established for pure edge, pure scree and mixed threading dislocations.  Densities were markedly lower for the SiC substrate.

Applications to Metals and Alloys:

Ahmed et al (1997), Ahmed et al (1999), and Ahmed et al (2001) – a series of papers using ECCI in a JEOL SEM with LaB₆ source to image dislocation substructure in Cu single crystals after cyclic deformation.  Near normal incidence geometry and an annular BSE detector at the pole piece were used.  Striking images of self-organised patterning in the dislocation structures including vein and PSB ladder-like structures.  Compared to TEM diffraction contrast the spatial resolution is very limited but the advantages of using bulk specimens included intermittent observations of the same sample at different points through its cyclic life to show evolution of the dislocation structure, observations over large areas, and examination of structures close to extrusions, intrusions and cracks.

Gutierrez-Urrutia et al (2012, 2013) – Gutierrez-Urrutia and co-workers have been significant users of the ECCI technique in exploring deformation mechanisms in steels of varying levels of complexity – some of this is reviewed in the 2013 paper.  The 2012 paper is an interesting development in making quantitative dislocation density measurements in large grained Fe-3Si electrical steel using ECCI.  Quantification follows methods familiar to TEM practitioners but with the need to estimate the depth to which ECCI can reveal dislocation lines so that the probed volume can be defined.
electrical steel Fe-3Si: Gutierrez-Urrutia et al (2012), Eisenlohr et al (2012)
TWIP steel: Gutierrez-Urrutia et al (2009, 2011, 2012)
austenitic 316L: Yan et al (2014)

Yamasaki et al (2015) – amazing work by a group at Kyushu University combining FIB serial sectioning with ECCI to capture three dimensional details of dislocation structures.  The slice thickness of 10 nm was chosen to be smaller than the depth to which dislocations were visible so as to aid registration of the stack of 33 images.  The method was used to reconstruct the 3D dislocation substructure in a Ni-based alloy after creep deformation to ~1.5% strain.

Monsour et al (2015), Ben Saada et al (2017) – present some detailed analysis of (very) low angle boundaries combining EBSD mapping of crystal misorientations and ECCI imaging of the discrete dislocations making up the sub-grain boundary.  The 2015 work examines an 0.13° boundary in an Fe-Si steel using ECCI to determine the nature and spacing of individual dislocations in the boundary array using g·b=0 and g·bxu=0 criteria and compare this to quantification of the boundary misorientation.  The 2017 work gives some beautiful ECCI images of more complex dislocation arrays at grain boundaries in UO₂ after high temperature creep deformation.  EBSD gave overall statistics on grain and sub-grain sizes while ECCI was used for more detailed of selected examples.

Zhang et al (2015) – ECCI used to study dislocation structures generated around nano-indents in differently oriented grains in a Fe–22Mn–0.65C (wt%) TWIP steel.  Dislocation density was quantified as a function of distance from indent centre.  Structures observed were used as comparison to discrete dislocation plasticity simulations.

Ram et al (2016) – using ECCI (along with EBSD) to study effects of pre-existing low angle boundaries on the creep deformation of single crystal Ni-based superalloy.  ECCI was instrumental in demonstrating that these low angle boundaries were not the sources of dislocations dominating plastic creep even at relatively small strains (~0.1%).  Instead rapid avalanches and dislocation multiplication away from the boundaries gives a drastic and rapid increase in dislocation density.

Wang et al (2018) – Examination of dislocation structures in Nb single crystals, for use as superconducting radio frequency (SRF) cavities, after a selection of heat treatments and deformation.  ECCI at near normal incidence using an annular BSE detector at the pole piece, and employing SA-ECP through beam rocking to establish well-controlled diffraction conditions. 

Dunlap et al (2018) – Study of dislocations near spherical-conical nanoindents in Ta which makes careful quantitative comparison of direct observations via ECCI and GND density estimates from HR-EBSD mapping.  ECCI at near normal incidence using an annular BSE detector at the pole piece, and employing SA-ECP through beam rocking to establish well-controlled diffraction conditions. 

Angus J Wilkinson
December 2018