您当前的位置: 首页 > 网页快照
Uncovering the tiny defects that make materials fail – Physics World
29 Nov 2022
When a material fails, it can have devastating consequences – making bridges collapse or pipelines explode. Tomas Martin and Stacy Moore describe how innovative and complementary microscopy and spectroscopy techniques can reveal the underlying atomic-scale mechanisms behind a material’s degradation
Materials make up the world around us. They are everywhere, from the wood, plastics and ceramics in our homes, to the metals and concrete used to construct buildings and bridges. But over time, materials can degrade,?their structure changes, they become less reliable ? and sometimes they even fail?altogether – with catastrophic consequences.
One big challenge with engineering materials such as steel is therefore to ensure they last as long as possible. That means finding ways to counteract “materials degradation” processes such as?fatigue from cyclic stresses; creep (slow deformation) caused by mechanical stress at high temperatures; wear and tear from components rubbing against each other; and corrosion triggered by chemicals in the environment including water, salts and aggressive gases.
Degradation can build subtly until the entire structure suddenly fails
Understanding the way materials change during these processes can be challenging, as the underlying mechanisms often occur at the atomic level. Slight movements or reactions of individual atoms are imperceptible to the human senses, but when multiplied across billions or trillions of atoms, they build up into dramatic changes in the material. These alterations may occur at small levels for years before a noticeable change is observed in a component, and degradation can build subtly until the the entire structure?suddenly fails.
Imperfect crystals .
Many important materials, such as metals, silicon or diamond, are crystals – highly-ordered repeating units of atoms. Their regular lattice formations can produce a myriad of?useful?properties, such as strength, heat conductivity, electrical conductivity and optical transparency. While these properties are hugely important for applications and are optimized by a perfect crystal structure, it is the deviations from perfection that are key in materials degradation.
1 Flaws in perfection .
There are three common types of defects in a crystal lattice: (a) a vacancy or missing atom, (b) a dislocation or misaligned row of atoms, (c) grains, or regions of atoms with differing crystal orientation. (Courtesy: Interface Analysis Centre, University of Bristol, UK)
These “defects” can take many forms, the simplest case being a missing atom – or vacancy – in the repeated lattice ? (figure 1). More complex long-range defects include dislocations, where whole lines or spirals of atoms can be out of place. There are also grain boundaries, where regions of crystal that formed at different angles meet, which can leave a line of atoms with misaligned bonds. The inclusion of additional chemical elements to a material can complicate its structure even more. New phases known as precipitates can form, and as these are likely to have different structures to the bulk crystal, they introduce areas with different properties.
When a defect experiences external influences, such as stress force, a change in temperature or even a chemical attack, many complex and interesting interactions can occur. The atoms at defects don’t have the same bond structure as those in the main crystal, and can be missing bonds entirely. This means the defects are easier to move under stress, and can more readily react with other chemical elements to form new bonds.
Understanding how a simple defect is affected by a single degradation mechanism such as increased stress, temperature or chemical reaction can be relatively straightforward. But a component in a bridge, aircraft, or nuclear reactor might have billions of such interactions occurring in complex environments. Uncovering how these individual microscopic processes combine into complicated macroscopic changes across a component is hugely challenging, but innovative tools and approaches are now enabling materials scientists to study these degradation problems in new ways.
In particular, the rise of high-speed microscopes lets us characterize atomic-scale defects faster and over larger areas than ever before. Meanwhile, techniques such as machine learning, image recognition, and data processing?means we can ? study bigger datasets.?Taken together, we are obtaining new, atomic-level insights into how materials degrade, which in turn is letting us make better predictions of how materials might ultimately fail.
Stress corrosion cracking .
2 Cracks in steel .
Optical (top) and electron microscopy (bottom) of stress corrosion cracks in steel. (Courtesy: Interface Analysis Centre, University of Bristol, UK)
One particularly complex way in which materials degrade is “stress corrosion cracking” (SCC). It occurs in metals when a susceptible material experiences a high stress in a corrosive environment, with the combination of these three factors ultimately leading to sudden and unexpected cracking. SCC can happen at both high temperatures – for example, in aircraft engines, and coolant circuits in nuclear reactors – and low temperatures, such as with offshore wind or oil platforms. It is particularly prevalent where salt is present, putting materials out at sea especially at risk. The end result can be catastrophic failure – boats sink, engines fail, bridges collapse, and gas pipelines explode.
In order to fully understand this unique failure process, we need to work out how it starts. However, this is very difficult to do as the event occurs at random times, and if a crack has already begun, the origins of the process are?probably?hidden by the damage created.
To tackle the problem, our team at the? University of Bristol is using multiple microscopy methods to watch cracks as they expands in real time. One method ? that’s turned out to be?particularly useful for analysing small-scale variations in microstructural surface features is high-speed atomic force microscopy (HS-AFM) (see box).
High-speed atomic force microscopy (HS-AFM) .
As with a conventional atomic force microscope (AFM), HS-AFM produces topographic images of a surface by monitoring the movement of a tiny (10 nm) sharp probe on the end of a cantilever beam as it traces its way over the sample.?When this tip encounters bumps or pits, it is deflected upwards or downwards, respectively – much like the needle in a vinyl record player or a fingertip across braille. The detection system measures this motion and builds up a map of the surface pixel by pixel.
The key difference between HS-AFM and conventional AFM is that it’s much faster. An AFM can typically scan a 5 ?m by 5 ?m area in over the duration of a few minutes, while a HS-AFM can measure the same area in less than a second. This enhanced speed means entirely new experiments can be performed. For example, using HS-AFM you can analyse the spatial distribution of nanoscale features, such as precipitates, over millimetre, or even centimetre scales in a matter of hours – a feat that would take a standard AFM years to do. This type of characterization is key for understanding nanoscale variation because it is these small changes in structure or composition that lead to large-scale changes in material properties, such as strength, hardness, or ductility.
The HS-AFM?can also image ? in liquid or gaseous environments, allowing for in-situ , real-time analysis of the nanoscale changes occurring during processes such as corrosion. This combination of capabilities, alongside the instrument’s high throughput, is unique to HS-AFM,?letting?us carry out novel and exciting experiments into various nanoscale phenomena.
HS-AFM is ideal for studying SCC because experiments can take place in a liquid, and the degradation can be observed in real-time. Our team therefore designed a bending apparatus that can hold a sample under tensile stress within a corrosive liquid environment – and were able to conduct the first experiment of its kind ( npj Materials Degradation ?5 3 ).
The material we tested was a sample of stainless steel that had been heat treated to make the microstructure more susceptible to SCC – the heat changes the size of grains and presence of precipitates, and it also moves chemical elements around and makes the grain boundaries more vulnerable to chemical attack. Tensile stress, i.e. stress that acts to pull the sample apart, was applied to the steel via a three-point bend rig (figure 3). At the same time, the sample was held in a corrosive liquid environment of 395 ppm sodium thiosulfate, which is?often found?in oil and gas pipelines.
These conditions are particularly relevant within nuclear applications, and are known to induce intergranular SCC – where the crack forms along the grain boundaries rather than through the grain. Measurements by HS-AFM were therefore concentrated along the grain boundaries of the material, in order to observe the processes before and during SCC.
With some skill, some luck, and a whole lot of patience, we were able to perform in-situ and real-time observations of SCC
Many attempts are often required to successfully image SCC, as there is little way of predicting which grain boundaries the crack will initiate at and which it will progress along. With some skill, some luck, and a whole lot of patience, we were able to perform in-situ and real-time observations of SCC as the crack progressed along a grain boundary, as shown in figure 3. This measurement gave new insight into cracking behaviour, revealing the way the grains parted. Rather than simply pulling apart in plane, the crack also caused one grain to lift as the crack progressed, producing a shearing motion. This behaviour was found to be the result of sub-surface crack propagation, causing movement of the grains at the sample surface.
3 Stress corrosion cracking in real time .
To observe SCC in real-time, the steel sample was analysed using HS-AFM while being subjected to a corrosive liquid (395 ppm sodium thiosulfate) and tensile stress (via a three-point bend rig 2 cm across) (left). The HS-AFM images show the crack widening along a grain boundary (right). (Courtesy: npj Materials Degradation ? 5 3)
The ability to take high-resolution topographic images of the crack propagation is especially useful, as it helps improve computational models of SCC. This information is powerful – by knowing which part of the material’s structure is attacked by SCC, we can help to design coatings and new materials to protect against attack and make components last longer. However, the picture is incomplete, and often we need complementary techniques to conclude the story.
Complementary analysis .
Corrosion processes, such as SCC, are complex systems consisting of both physical and electrochemical changes. New techniques, like HS-AFM, enable researchers to unlock additional insights into such mechanisms, but to gain full understanding of a material’s behaviour often one technique is not sufficient on its own. Multiple complementary techniques are required, allowing for measurement of surface and sub-surface processes, chemical changes, and electrical signals across different length and timescales.
4 Finding the right combination .
Different analysis techniques can provide varying information about a sample over different time and length scales.
There are many techniques that can be used together to unlock different information about a material (figure 4). For example, electron backscatter diffraction (EBSD) in a scanning electron microscope (SEM) or diffraction in a transmission electron microscope (TEM) can tell us about the relative angles of the crystal lattice within different regions (or grains) of a material?(figure 5). This gives insights into the local stresses at a crack, and why a particular region of a material may be vulnerable to attack first.
5 Grains of many colours .
Electron backscatter diffraction (EBSD) of a crack edge. Each colour is a grain where the crystal lattice is aligned at a difference angle. (Courtesy: Interface Analysis Centre, University of Bristol, UK)
Techniques such as energy-dispersive X-ray spectroscopy (EDX) on both TEM and SEM, as well as atom probe tomography (APT) yield information about the elemental composition of a specimen, providing clues about the chemical changes that occur when corrosive reactions take place. X-ray and ultraviolet photoemission spectroscopy?using?an electron spectroscopy for chemical analysis (NanoESCA) instrument can give incredible information about the local electronic environment at a sample surface. It can tell us, for instance, about how likely different regions of a material are to lose an electron, and therefore why they might be more vulnerable to corrosion.
Each of these advanced microscopy techniques has its own strengths and can give information for different length scales of a material, from the scale of millimetres down to individual atoms. Using the right mix of techniques, scientists can bring together unparalleled insights into the structure, chemistry, local stress and chemical environment so that we can unpick the origins of failure at new levels of detail.
Want to read more? .
Register to unlock all the content on the site
E-mail Address .
Register
Tomas Martin is a senior lecturer in materials physics and Stacy Moore is an EPSRC Doctoral Prize Fellow in the School of Physics at the University of Bristol, UK. e-mail tomas.martin@bristol.ac.uk and stacy.moore@bristol.ac.uk , Twitter @tomas.l.martin and @S_mooreorless .
From:
系统抽取对象
人物     
(1)
(1)
系统抽取主题     
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)