Nirmal K. Sinha

Engineering Physics of High-Temperature Materials


Скачать книгу

Spherical (sp2 and Sp3) Between sp2 and sp3 Appearance Dark black/gray, nonmetallic luster Dark gray, metallic luster Colorless (pure) Black Black Hardness Low Medium Very high High Very high Tensile strength Low Medium — Medium Very high Thermal conductivity Low Low High High High Electrical conductivity Varies High Insulator Low Semiconducting – metallic

      Trinity of Glass Structure Models

      RANDOM NETWORK

      RANDOM COIL

      RANDOM CLOSE‐PACKING

      The atomic structure of glasses can be very complex. Three overarching structural models have arisen to describe most ideal amorphous materials.

      1 The continuous random‐network model: This model describes a three‐dimensional network where an irregular network is formed through bridging atoms, but the bonding pattern at network sites is repeated. It is most applicable to covalently bonded glasses, such as oxide‐based glasses (Zachariasen 1932). In oxide‐based glasses, the oxygen atoms form bridges and network formers, such as silicon, phosphorus, or germanium, form strong bonds with them in a randomly arranged network structure. In addition, network modifiers, such as sodium and calcium, generally sit in ionic form within interstitial holes and can drastically alter the properties of the bulk material. The continuous random‐network model describes an ideal – fully amorphous – glass. In Section 2.4, we explore the structure of real glasses in more detail.

      2 The random‐coil model: This model describes the disorder created by entangled chains through the use of three‐dimensional random walks (Flory 1949). In fact, the model presents a statistical distribution of conformations taken by all the chains in a population of macromolecules and is further complicated by chain length distributions, branching, and constrained rotation in real material. It is most applicable to glassy polymers, such as polystyrene and many plastics.

      3 The random close‐packing model: This model describes an irregular structure of molecules that does not contain any short‐ or long‐range order or holes large enough to admit another molecule. It is based upon crystallographic models for liquid structure (Bernal 1959) and is most applicable to metallic glasses.

      An exhaustive review of the knowledge gained on the structure of inorganic oxide and nonoxide glasses and the liquids they are derived from was carried out by Greaves and Sen (2007). They looked at the atomic structure, ranging from the local environments of individual atoms to the long‐range order, which can cover many interatomic distances. In this book, we will concentrate only on inorganic amorphous materials and, in particular, on silicate glass that is used extensively in domestic and industrial buildings, automobiles, and for household purposes. Even though silica‐based glasses have satisfied demands for a remarkable range of consumer goods for a long time throughout human history, the mechanism of glass cracking has not been well understood and is keeping glass scientists active (Michalske and Bunker 1987).

      2.3.1 Solidification of Materials

      Most inorganic solids, like minerals, rocks, and ice, on Earth's surface are solidified or frozen from a liquid phase. The temperature at which a material transitions between the solid and liquid phases is denoted as the melting point (T m) of the material. This transformation temperature represents the point where the free energy of the solid phase and that of the liquid phase are equal. Unless otherwise noted, this T m is assumed to be at standard pressure. However, there are several environmental aspects involved in the solidification process from liquid to solid. These are all contributing factors that determine the microstructure of the solid.

      At a high level, during the solidification process, atoms/molecules from the fluid begin to bond together to form a nucleus. Homogeneous nucleation involves the clustering of liquid molecules/atoms to form a critical‐sized nucleus without an external interface. If a seed crystal of a desired structure is exposed to the melt, it can act as a nucleus and support in controlling the final structure. The nucleation process can also be triggered by the surface of an impurity (or structural material) that lowers the critical free energy required to form a stable nucleus. This process is called heterogeneous nucleation and, in practice, can control the solidification process.

      Generally, multiple nuclei simultaneously develop in the fluid. The nuclei define the crystallographic phase of the grain. The crystals grow by the progressive addition of molecules from the fluid until they impinge upon adjacent crystals. The border between crystals is called the crystal boundary or grain boundary (GB). In metals, crystal growth from the liquid often follows a dendritic pattern where there is a main branch with many side branches caused by the growth of defined crystal planes in the lattice. As the crystal grows and impinges on its neighbors, the interdendritic spaces fill in and the original dendrites may not be apparent in the final microstructure. The bonding structures within the grains will depend not only on the temperature, composition, atmosphere, and pressure, but also on the speed of temperature changes and time at various temperatures. Certain crystals or grains may grow at the expense of others such that with slow cooling the individual grain sizes tend to be larger. Impurities, differing elemental compositions (e.g. due to overall concentration and equilibrium phases), and pores may be segregated into different grains and even into grain and subgrain boundaries. The size/shape/composition of the containment vessel and flow patterns for heat exchange can also play important roles. Microstructural properties, such as grain/crystal size, shape, crystallographic orientation, and intragranular/intergranular impurities, are just a few of the parameters that influence the mechanical response and physical properties of a material.

      The physical properties of a material can also depend on the orientation of the material. The “texture” of a material is used to describe the distribution of crystallographic orientations and can have a strong influence on directionally dependent properties. A perfectly random polycrystalline material will have isotropic properties when the size of the sample is sufficiently larger than the grain size. This is due to the averaging effect across the different grains in different orientations. Some common textures include:

       Wire/fiber: Orientational alignment in the axial direction with nearly random radial orientation.

       Sheets: Compression and rolling in sheet preparation can orient grains in both axes through grain flow; however, annealing can change the texture.

       Thin films: “Fiber textures,” where certain lattice planes preferentially align with the substrate, and “biaxial textures,” where alignment with the substrate occurs.

       Rocks: