Amorphous Materials

One of the Voyles' group specialities is fluctuation electron microscopy (FEM). FEM uses nanodiffraction in a TEM or STEM to measure nanoscale structural order in amorphous materials. The information we obtain from FEM is extremely difficult, if not impossible to obtain from any other experimental techniques. Paul learned FEM as a graduate student from its inventors, Murray Gibson and Mike Treacy. To learn more about the technique itself, see the introduction to FEM elsewhere on this site.

Chalcogenide Glasses

Building on earlier work using FEM to study nanoscale structural order in tetrahedral amorphous semiconductors like germanium and silicon, we are now studying another form of covalent bond ntwork, chalcogenide glasses. Binary chalcogenides have some four-fold and some two-fold coordinated sites in the network, and the fraction of each type can be adjusted by changing the composition. This should in principle allow greater structural variety, which may be responsible for the fact that chalcogenides are true thermodynamic glasses, unlike the tetrahedral semiconductors. We will test this hypothesis using FEM.

Chalcogenides also exhibit a variety of physical phenomena of interest either from a fundametal perspective or for applications which may be tied to nanoscale structure. We will investigate the structural underpinnings of photodarkening, which is a metastable, light-induced change in bandgap that has been used to pattern waveguides for photonics, and the vibrational intermediate phase, which has been proposed to be a dynamical signature of structural self-organization on the nanoscale.

Amorphous Metals: Crystallization and Plasticity

Amorphous metals offer an entirely different regime in which to study nanoscale structure in glasses without the strong directional bonding of a covalent network. Various structural models for metallic glasses have been proposed, from icosehedral nanoscale order to clusters of atoms constructed around efficient local packing of hard spheres of different sizes; all of them agree with at least a subset of the available structural data from diffraction and occasionally EXAFS or resonant scattering. This suggests that the available data sets are incomplete and do not sufficiently constrain the structure at the nanometer length scale. We seek to add FEM data to the mix, in the hopes of providing more stringent tests of existing theoreis and perhaps developing new ones.

In addition to the basic underlying structure, we are working on two aspects of amorphous metals which are thought to be controlled by their nanoscale structure: crystallization and plasticity. We are studying crystallization in Al-based glasses which require an extremely high cooling rate (>10⁶ K/s) to become amorphous. Perepezko has proposed that these glasses do not in fact escape the formation of crystals, only their growth, and that these very small "frozen-in" crystalline nuclei explain the dramatic nanoscale crystallization of these alloys. Using FEM, we have found residual crystal-like order in the amorphous phase of some of these alloys, and are working to understand its structure, size distribution, and stability.

We are studying plasticity in bulk amorphous alloys based on Zr. Metallic glasses fail during plastic deformation by the formation of shear bands - small (10-1000 nm wide) regions of localized, intense plastic deformation. While the phenomenology of shear bands is well-established, the atomic structure underpinning their formation is not well-understood, nor even is their a completely adequate continuum mechanics description of their effects. We are using FEM to probe the structure of as-cast and deformed glasses, and correlating that structure with advanced nanoindentation measurements of plasticity by Prof. Don Stone's group.

Magnesium Diboride

MgB₂ is a novel, medium-temperature superconductor. It superconducts by the conventional BCS mechanism with phonon coupling, but it is unusual in that it has two superconducting gaps arising to two distinction, weakly coupled electronic bands. This gives it a variety of interesting properties, including the potential for extremely high upper critical magnetic field compared to other low-temperature superconductors, and novel vortex structures and phase textures either by itself or in junctions.

We are part of a large, multi-institution program lead by the Florida State Applied Superconductivity Center to study the fundamental materials science and physics of MgB₂ and to develop it for applications primarily in magnets. We use conventional TEM, advanced STEM imaging, and high resolution microanalysis by EDS and EELS to study the microstructure and microchemistry of various forms of MgB₂, including alloyed and pure thin films and bulk pellets, wires, and tapes.

Three-Dimesional Imaging at Nanometer Resolution

The advent of multipole electron optics that correct the third-order spherical aberration of round electron lenses is in the process of revolutionizing electron microscopy. One part of that revolution is that electron microscopes are finally achieving large enough numerical apertures that the depth of focus becomes smaller than the sample thickness. This opens the door to three dimensional imaging in the EM using optical sectioning: stacking up images acquired at different values of defocus to make a 3D image, like piling slices of bread to make a loaf. This technique has long been used in light microscopy, especially using confocal microscopes, but has only recently been demonstrated with electron optics.^1,2

Optical sectioning in the electron microscope will be limited by residual, higher-order lens aberrations and by the strong dynamical scattering of electrons in matter. We have investigated the optical theory of confocal imaging with corrected electron lenses³ and the effects of dynamical scattering on optical sectioning in confocal STEM and Z-contrast STEM.⁴ Not surprisingly, for zone-axis oriented crystals and current optics, dynamical scattering signficantly complicates the behavior of the intensity as a function of defocus, making the straightforward interpretation of a defocus series of images in terms of pure optical sectioning incorrect.^3,4

This area of research is currently on a hiatus, waiting for better instrumentation and a source of funding.

References

1. "Submicron imaging of buried integrated circuit structures using scanning confocal electron microscopy",
   S. P. Frigo, Z. H. Levine, and N. J. Zaluzec, Appl. Phys. Lett. 81, 2112 (2002).
2. "Three-dimensional imaging of individual hafnium atoms inside a semiconductor device", K. van Benthem, A. R. Lupini, M. Kim, H. S. Baik, S.-J. Doh, J.-H. Lee, M. P. Oxley, S. D. Findlay, L. J. Allen, J. T. Luck and S. J. Pennycook, Appl. Phys. Lett. 87, 034104 (2005).
3. "Prospects for 3D, nanometer-resolution imaging by confocal STEM", J. J. Einspahr and P. M. Voyles, Ultramicroscopy 106, 1041 (2006).
4. "Imaging single atoms with Z-contrast STEM in two and three dimensions", P. M. Voyles, Microchimica Acta 155, 5 (2006).