Application Notes

Detailed EBSD Study of a Bivalve Shell

Author: Oxford Instruments

Published: 16 Jan 2019 · Last updated: 16 Jan 2019

Tags: EBSD

Introduction

The structure of natural shell samples has for many years been of significant interest, not only to researchers in the biological sciences, but also to material scientists. This is primarily because the shells often have physical properties far exceeding those of their constituent minerals: in particular the mother of pearl structure, or "nacre", consists of small, interlocking tablets of aragonite (orthorhombic CaCO3) separated by organic matrix membranes (Fig. 1). Nacre has a very high strength when compared to single aragonite crystals and researchers are mimicking this structure in order to produce similarly high-strength materials. The fact that the aragonite tablets have a thickness similar to the wavelength of visible light (e.g. ~ 500 nm) results in nacre's characteristic iridescence, but poses significant challenges for effective microstructural characterisation.

Over the past decade, EBSD has become a routine technique for studying biomineralised structures such as egg shells, mollusc shells (e.g. oysters and snails) and even fossilised structures. However, the majority of these applications involve the study of relatively coarse grained calcite (trigonal CaCO3) structures that are easy to analyse. Aragonite nacre, in contrast, is not only extremely fine grained but is also beam sensitive and as a result there are few published studies showing successful EBSD characterisation of nacre layers. One such study demonstrated that low beam energies were required in order to successfully analyse nacre, showing orientation maps with 100 nm measurement step sizes collected at 8 kV accelerating voltage. The low beam energy significantly improves the spatial resolution of the EBSD technique, but often at the expense of signal and, even with sensitive CCD-based EBSD detectors, long exposure times are required in order to collect each diffraction pattern. The secret to successful measurement of nacre structures is not only low beam energies, but also a low electron dose at each point. This can be achieved by lowering the electron beam current (for example, via the use of smaller apertures or spot sizes) or by lowering the exposure time for collection of each EBSD pattern; in both cases, having a sensitive EBSD detector is essential.

In this application, we demonstrate how the revolutionary Symmetry® detector, utilising custom-developed CMOS technology, combines extreme sensitivity with exceptional speed, making it the ideal detector for analysing challenging, beam-sensitive and fine grained structures such as aragonite nacre. We show results from routine analyses of calcite-aragonite interfaces in the shell of the mollusc Mytilus edulis, better known as the common mussel.

Experimental Details

Sections of the mussel shell were mounted in epoxy, and subsequently ground and polished for EBSD analysis, finishing with 5 minutes of colloidal silica polishing. The sample was then coated with ~ 5 nm of carbon and grounded to a suitable SEM stub using conductive Ag paint. The cross-sections of the shell were then examined in a field emission gun SEM operating at 12 kV accelerating voltage and with a beam current of ~ 7.5 nA, at a working distance of 15 mm. The Symmetry EBSD detector was used to collect EBSD patterns at a resolution of 312 x 256 pixels. Two maps were collected, the first using a step size of 80 nm on a grid of 1508 x 680 measurements with a camera exposure of 3 ms (indexing at 330 patterns per second), and a second with a 60 nm step size, a 1093 x 1413 grid and with a 2.5 ms exposure (395 indexed pps). The indexing hit rates were 97.7% and 96.7% respectively, and minor processing was performed to remove non-indexed and incorrectly-indexed pixels.

Results

Area 1

The results from area 1 are summarised in Fig. 2. The phase map (Fig. 2a) shows the irregular nature of the boundary between calcite and aragonite at this location. The orientation map (Fig. 2b) shows that the calcite has a very strong texture, whereas the aragonite has domains up to 5 μm across that cross numerous nacre layers. The aragonite boundary misorientation distribution shows prominent peaks at approximately 64° and 53° misorientation: these are associated with aragonite triplet twinning, with crystallographic rotations about the <001> direction. These twin boundaries are highlighted in red and yellow in Fig. 2b, and account for 37.7% and 11.3% of high angle boundaries in the aragonite, respectively.

The subtle changes in orientation in both phases are highlighted using the Disorientation colouring scheme (Fig. 2c): this illustrates clearly the tablet structure in the nacre (the "hypostracum"), as well as the scalloped sub-grains within the calcite (the "ostracum" or "prismatic calcite" layer).

The nanoscale microstructural information available within this dataset is shown in more detail in Fig. 3, a small portion of the large dataset at the interface between calcite and aragonite. The data here have had isolated incorrectly indexed pixels removed but otherwise had no subsequent filtering. The detail within and across sub-500 nm wide nacre layers is apparent, as are intricate structures at the interface itself.

Fig. 2a. Area 1 phase map: calcite - blue, aragonite - red

Fig. 2a. Area 1 phase map: calcite - blue, aragonite - red

Fig 2b. IPF orientation map, showing high angle boundaries in black and low angle boundaries in grey. In aragonite, special boundaries are marked in red (63.8° <001>) and yellow (52.4° <001>).

Fig. 2b. IPF orientation map, showing high angle boundaries in black and low angle boundaries in grey. In aragonite, special boundaries are marked in red (63.8° <001>) and yellow (52.4° <001>)

Fig. 2c. Disorientation coloured map, highlighting small orientation colours in both phases.

11.3% of high angle boundaries in the aragonite, respectively.

The subtle changes in orientation in both phases are highlighted using the Disorientation colouring scheme (Fig. 2c): this illustrates clearly the tablet structure in the nacre (the “hypostracum”), as well as the scalloped sub-grains within the calcite (the “ostracum” or “prismatic calcite” layer).

The nanoscale microstructural information available within this dataset is shown in more detail in Fig. 3, a small portion of the large dataset at the interface between calcite and aragonite. The data here have had isolated incorrectly indexed pixels removed but otherwise had no subsequent filtering. The detail within and across sub500nm wide nacre layers is apparent, as are intricate structures at the interface itself.

Fig. 3a. Pattern quality map from a small section of Area 1.

Fig. 3b. IPF orientation map, showing high angle boundaries in black and low angle boundaries in grey. In aragonite, special boundaries are marked in red (63.8° <001>) and yellow (52.4° <001>).

Area 2

In Area 2, the nacre layers were in general narrower (typically 300 – 600nm), necessitating a smaller measurement step size (60nm). The results in figure 4 once again show the intricate detail within both phases: with the exception of a single broad aragonite layer close to the calcite-aragonite interface, the width of the nacre layers decreases as this interface is approached. The pattern quality map (Fig. 4a) shows the stability of the sample during the analysis and the fine resolution of even the most detailed structures. The orientation map (Fig. 4b) highlights the abundance of twin boundaries in the aragonite nacre, and a dominant orientation in the calcite.

Fig. 4a. Pattern quality map from Area 2.

Fig. 4a. Pattern quality map from Area 2

Fig. 4b. IPF orientation map, showing high angle boundaries in black and low angle boundaries in grey. In aragonite, special boundaries are marked in red (63.8° <001>) and yellow (52.4° <001>).

The orientations of the two phases are shown in the pole figures in Fig. 5. The near-single crystal nature of the calcite is clearly shown, as is the very strong alignment of {001} in the aragonite – a well documented texture in nacreous layers.

Fig. 5a. Scattered pole figure showing poles to {0001}, {11-20} and {10-10} in the calcite in Area 2.

Fig. 5b. Scattered pole figure showing poles to {100}, {010} and {001} in the aragonite in Area 2.

Conclusion

The study of shell structures using EBSD is particularly challenging due to the beam-sensitive nature of the carbonate minerals, and the sub-μm scale of layers in nacre. There exist few successful studies of aragonite nacre in the published literature, and those that exist almost certainly required extensive post-analysis data filtering in order to reveal the intricate structures of the nacre layers.

In this study we show how the combination of high speed and extreme sensitivity of the new Symmetry CMOS-based EBSD detector represents a technological breakthrough for the study of biomineralised samples. In the two examples presented here, large areas across the calcite-aragonite interface in a common mussel were analysed with sub-100 nm resolution. Excellent diffraction patterns could be collected and indexed in less than 3 ms, enabling damage-free analyses with exceptional indexing hit rates. The resulting phase and orientation maps highlight the complex structures within these shells, shedding light on the microtexture, twinning and detailed structure in these remarkable materials.

References

  1. Mayer, G. (2005). Science, 310, 44–47.
  2. Schmahl, W.W. et al. (2009). Particle & Particle Systems Characterization, 25, 474–478.
  3. Perez-Huerta, A. et al. (2008). Mineralogical Magazine, 72, 563–568.
  4. Griesshaber, E. et al. (2013). Acta Biomaterialia, 9, 9492–9502.

← Back to Learning