Importance of Materials Science in Understanding Failures
Over the years, I’ve had many opportunities to increase my knowledge of my technical field. I have also had the chance to teach at multiple colleges and universities, as well as customize materials-engineering course material.
Experience has helped me to appreciate the value of the concepts that were “beat into me” by my college professors. It has also resulted in the creation of my “crash materials degree,” which I shared with my Industrial Heating blog readers. Here are the key concepts condensed from my original blogs.
1 The essence of materials science and engineering is that every material must be made into a part by a process, which creates a multilevel structure. This results in a constellation of properties, characteristics or behaviors in a given environment. Traditionally, we show a triangle with “process” on the bottom left, “structure” on the bottom right and “properties” on top (Fig. 1).
2 If the characteristics or behavior differ in two materials that are nominally “the same,” then they are not the same. Either the composition is different or the processing was different, which (either way) resulted in a different structure and thus different properties. This is a direct consequence of #1. Sometimes differences are trivial, but sometimes differences in composition or processing would seem to be trivial but result in significant property differences. Sometimes 5-10°F difference in a heat-treat operation can cause a problem.
3 There are two main forces that create structure in a given material. The first is thermodynamic tendency. This means that given a group of atoms in a specific environment, the atoms will want to group themselves and position themselves in a specific way. If there is more than one type of atom present, there will be a preferred way in which the atoms either avoid or seek out those of the other “species.” The second “force” that influences the structure and arrangement of the atoms within a material is kinetics. This has to do with how fast the atoms can move in their environment.
A classic example in the ferrous-metallurgy field is that carbon in iron would really rather be present as graphite. Think about it.
Carbon is totally different in about every way (e.g., diamonds and coal) from iron (steel is mostly iron). But steel generally has the carbon present as carbides or as individual dissolved carbon atoms between the iron atoms.
Why is this? It is because of kinetic considerations. When processing ferrous materials, we don’t generally allow time for the carbon atoms to find each other and push all the iron atoms out of the way. There are a lot more iron atoms in even the highest-carbon steel than there are carbon atoms. It’s “easier” for the carbon to compromise and form carbides with the iron neighbors.
So we have thermodynamic “desire” fighting kinetic “resistance,” and we get what we get in a given alloy composition subjected to a given process history.
Mechanical action can strongly influence kinetics. The blacksmith does more than change the shape. The blacksmith changes the arrangements of the atoms within the alloy.
Traditionally, we have “equilibrium diagrams” that show us how the atoms prefer to group themselves at a given temperature. Time-temperature diagrams help us figure out the effects of the kinetic influences.
After integrating the concept of the process-structure-property materials triangle that we discussed, the next most important concept in materials science for failure analysis of metal and polymeric components is related to structure and its multiple overlapping levels.
Let’s get to the nitty-gritty of structure. We require concepts that allow us to understand the multiple levels of structure inherent in each component (Fig. 1). If we are talking about a solid material as a materials scientist, the finest level of structure that concerns us on a day-to-day basis is the atomic structure. We need to have a basic idea of what an atom is.
The ancient Greek traveler and natural philosopher (they didn’t have scientists back then) Democritus (c. 460-370 BC) was the first to formulate the idea of the atom as the smallest piece of matter that could exist. Check out Wikipedia’s entry on Democritus for more insight.
We now understand that atoms are not spherical, that the electron clouds that surround the nucleus are not spherical, and the three-dimensional geometry of the electron cloud (Fig. 2) surrounding the nucleus of the atom is extremely important in determining which other types of atoms will be attracted to a given type of atom.
For our practical purposes, metallurgists working at the level of my daily activities don’t need to be concerned with anything more complicated than this. The details of the shape and electrical charge of the electron clouds have a strong effect on the thermodynamic driving force we discussed previously.
The important thing to understand is that the outer electrons in metallic atoms, when present mainly with other metallic atoms, are held very loosely by the nuclei of the atoms. This creates the possibility for electrical conductivity and also malleability – two of the key properties of metals.
Now that we have a sound foundation to understand the structure of atoms and atomic bonds, it’s time to move on to the concept of microstructure.
Microstructure is associated with relatively large groupings of atoms – large enough to see in an optical microscope, when the specimens are properly prepared. Check out my June 2014 blog for useful tips on how to prepare metallographic specimens for microstructural evaluation (use the website “Blog” tab).
How does the microstructure develop in a given component? Think of our earlier discussion of the twin concepts of thermodynamic driving force and the kinetic reality.
Thermodynamic driving force is what makes a given type of atom want to seek out its fellows or, alternatively, avoid them in preference to other types of atoms. Thermodynamic tendency has one other very important effect. In any system that has areas of more than one composition, there will be boundaries between the two compositions. There may also be boundaries between two areas of similar composition. We call these grain boundaries.
Boundaries in solid materials take energy to create and maintain. In most steels, for example, we have an iron-rich matrix, which would be of one composition and particles of carbides, oxides, sulfides, etc. Each of these “-ides” is of a different composition. There is a boundary between the matrix that contains these particles and the particles themselves. Different types of boundaries take different amounts of energy to maintain.
If there is enough energy (temperature) available in the environment, it may be used to minimize the surface area of the boundaries. How is this done? One way is by diffusion of the atoms that are in the smaller particles toward the larger particles. Another way is by changing the shape of flat or pointy particles into rounded ones. The sphere has the lowest ratio of surface-to-volume area of any shape.
Heat, or thermal energy, allows faster movement toward the low-energy (lazy) state that thermodynamics desires. Figure 3 shows a piece of annealed steel. The small protruding particles visible in the flat matrix are carbides. Most of the smaller ones have become quite rounded due to exposure to a suitable spheroidizing-anneal temperature.
Some of the larger particles, seen at the white arrows, still have shapes that deviate strongly from spherical or spheroidal. Note that the upper white arrow shows a smaller particle than the lower. Or does it? We always need to remember that we are viewing a cross section. What we are seeing might be a slice through a narrow area of a larger particle.
Longer time at the suitable spheroidizing annealing temperature would allow the larger particles to become more spheroidal as well.
Note also that increasing grain size is another way for metals to reduce their energy, even if they need temporary infusions of energy to accomplish the change. The small grains get absorbed into the larger grains, leaving less grain-boundary area per unit of volume.
So far, we have talked about different levels of structure. We started with atomic structure and went on to microstructure. We’ve talked about how process influences microstructure. We discussed the important concept of boundary energy.
Microstructure is intentionally attained in many different materials by the choice of processing parameters. This includes hot rolling, forging and extruding, cold rolling, cold forming and heat treating. All of these fall into the category of “thermomechanical” processing. Cold forming that disrupts the crystal structure is also used to modify the characteristics of materials.
Figure 4 shows a metallographic mount of a piece of low-carbon steel that has been etched to reveal the grain boundaries. This is a stamped part. We can clearly see that the grains in area A are blockier than the ones in area B. This is due to the deformation that happened – to a greater extent in area B – as a result of the forming process.
We would expect area B to be harder than area A. Room-temperature deformation usually makes the material harder and often makes it stronger because it locks energy into the volume of the material.
As we see here, in this small area the microstructure varies a lot. If we assume this is a bracket and the dark area at the top is the plastic mounting compound, and that the service stresses tend to open the angle, we might possibly expect that if the stresses are too high, a crack could start at the middle (vertical) of the three arrows and grow into the core of the steel. Because there is a gradient of deformation, the crack may wander to the left (Fig. 4) because that material is softer and probably weaker. Cracks start and propagate when and where the stress exceeds the strength. We see in this image that the microstructure and resultant strength can vary quite a bit.
We need to understand the microstructure at the crack-initiation area in order to understand fracture. That’s why we need to be quite confident that we find the initiation area. Figure 5 shows some aluminum that was overheated. The material got so hot that it actually melted in some areas. Where does aluminum melt when that happens? This typically happens at “triple points,” where three grain boundaries join at a single point. It requires less heat to melt because of the extra grain-boundary energy. The dark circles have all experienced “incipient melting,” with A and B actually showing evidence of “dendrites,” a feature that forms during solidification. Someone went to a lot of trouble to work the original cast billet, and now it has areas of lower-strength “micro-castings” scattered throughout.
Figure 6 shows a cold-headed bolt that was friction welded onto a piece of bar stock of the same diameter as the head. The bolt cracked when it was dropped on the floor. Why did it crack in an area that was probably around three times as large as the adjacent threads? Clearly, there was something related to the cold deformation, the weld heat and the composition that made that area’s microstructure much weaker than the threaded area.
Understanding the basic concepts of materials science and engineering includes the foundational facts about the relationship between structure and behavior and how structure is developed by selection of specific processing details. The optimization of material performance requires a knowledge of microstructures and how they are developed. The particular microstructure that develops in a given location in a specific part is dependent on many factors.