 Let's welcome to the stage Heather Jackson, thanks everyone for coming tonight. My name is Heather Jackson, and I'm a metal basher. But seriously, I'm interested in not necessarily why things break, but more how and when, and what we can do to prevent it. So everyone knows that some things, when you put them under stress, they crack or they shatter and we call that brittle. Other things, they dent or deform and we call that ductile. They're tough. They can take a licking and keep on ticking. So I think this car is a kind of a good example of these differences. So this, the crumple zone of your car is designed to deform and crush to keep the passenger compartment up top rigid and then the glass, as we all know, it just shatters. But if you ever wondered how these different materials got this way, tonight we're going to look inside the materials that our world is made of, focusing on this whole brittle ductile thing. Where does it come from and where do we have the power to do something about it? Not in a van. All right, so brittle or ductile have everything to do with which atoms are in a material and how they're put together. So these are the structures of the most common materials that we deal with metals and polymers. We know that they're ductile. We know that they deform. They're malleable on the atomic scale. This happens because in a metal, the atoms are packed closely together, kind of like this stack of cannonballs here. That's the structure of a metal like aluminum or gold or stainless steel. And they all kind of share atoms so that they, the atoms are easily able to move around. The bonds can easily break and reform and basically planes of atoms slide around to form new shapes. So thank you. So if you've ever seen these, these bucky balls, this, this is one of the funnest toys ever if you can still get a hold of it because I think there's some problems if you swallow them. But it, I mean it kind of gives you the sense of what's going on. They're magnetic, which is not quite the same thing as what's going on inside of a metal, but it gives you a sense of how metals can be so pliable. Polymers, on the other hand, they're ductile for different reasons. So think about, it's a big ball of string on the molecular scale. And if you're pulling on, thank you, if you're pulling on a polymer, you're basically just pulling on the strands in the ball. So everybody, everybody knows what happens with your silly putty or your modeling clay. A little bit of effort gets you a lot of deformation. Other materials, like glasses and ceramics, very different. So these, these are just some pictures to represent silicon dioxide in two forms, a crystal and a glass. And the atoms are all bonded together in kind of the same way, just one of them has a long range regular structure and the other one doesn't. But very different. The atoms are not closely packed together. The bonds can be very strong. So it's really hard for the atoms to rearrange themselves. It's really hard to get much deformation. And the bonds would really just rather break. Ductile materials still break, of course, but they deform. They can deform quite a lot before they break. And that's a key difference between brittle and ductile behavior. So that's a simple picture, right? Metals are brittle. I'm sorry, metals are ductile, glasses are brittle. Except sometimes metals can be brittle as well. So we did a little experiment. We picked up some, thank you very much, we picked up some high carbon steel nails from the hardware store. They're called masonry nails, and they're very strong and very tough, made for nailing things into concrete. So fingers crossed because we got a little video here. Oh, and we even got sound too. Okay, so off the shelf, this nail is engineered to be tough. You know, we thought, what could we do if we... It's possible. I'll turn this down a little tiny bit. Okay, so the nails, the steel and nails is treated specially to make it very tough. And we thought, what can we do to mess with that treatment? So we took a blow torch to a second nail, and we've got it here on a fire brick, let it slow cool. Then we put it back in the vise. The first nail, as we saw, you know, it was pretty darn tough. We beat the hell out of it, and it bent eventually, but it never broke. I'm sorry for the blatant ad for Coca-Cola, and that also probably shouldn't be in the lab either. So that, you know, kind of similar to the first nail, but a lot easier to bend it. The third nail, we heated it red hot with a blow torch again, but this time with a twist. Instead of letting it slow cool, we quenched it in a bucket of water, which we'll hear in about a second. So really quick, quick cool as opposed to slow cool. And then we're going to do the same thing again. We're going to put it back in that vise and whack it with a hammer. A little less in focus this time, but you'll still get the point. Oh, did I, did it disappear? Yeah, all right. So we're going to, I can make it, all right, all right. Pick up where we left off. I'm not going to move. All right, so, you know, it's bending a bit, bam. And I promise you it didn't just, you know, slip out. It actually broke. So, and up top, that's our slow cooled nail. In the middle, that's our quenched nail. What happened? We took a strong and tough nail, and in one case we made it softer and weaker, and in another case we made it brittle. So who cares? Well, the trouble with embrittlement is that when something breaks, it doesn't give you any warning, it's about to fail, and that can be a problem. So how do metals get brittle? Well, the point is defects. No real material is perfect or pure. All materials contain some imperfections in their structure. Just starting with at the level of the crystal structure, you get point defects. You can have an interstitial, which are extra atoms. So steel is iron with carbon wedged into the gaps. You can get substitutions. Stainless steel becomes stainless when you swap some of the iron for chromium. You can get vacancies, which are missing atoms. Dislocations. So if you pile up enough metal atoms, most of the place, they're going to form a perfectly regular ordered structure. But unless you're at absolute zero, it's not practical for it to be perfect. So things are going to be out of place. You're going to get extra planes of atoms, missing planes of atoms, and that line right at the edge of that plane, that's called a dislocation. Dislocations make it a lot easier for metals to deform. Kind of like this guy here. It's easier for him to move that rug across the room if he makes a little kink in the rug and pushes that kink across the room instead of trying to move the whole rug. And then the flip side is if you get dislocations intersecting, then they get tangled up and it makes it a bit more difficult to deform the metal. And as I said, most of the place, you're getting perfect crystal, but what real materials usually end up being is small regions that are their own little perfect crystal and they're all slightly differently oriented. And those are called grains, and grains are surrounded by grain boundaries. And the size and the shapes of the grains, and what's going on with grain boundaries, and how many dislocations you've gotten a material, all these things define a material's microstructure. All right, so microstructure actually tells you a lot about how a material was made and how it's going to behave when you put it under stress. So grains come in all different kinds of shapes and sizes. These are just some cross sections through different kinds of metals. Subtop on the left, we've got aluminum. Aluminum alloys tend to be, they're often made in sheet metal or extrusions, think of the skin of an airplane. And so what happens is all the grains get smushed out into these long, flat, pancake-shaped grains. So the properties are going to be different in different directions, but they're going to be strongest in the directions the grains flow, which is what you want in the skin of your airplane. On the other hand, this stainless steel in the upper right, this is an annealed structure, and the grains are all kind of the same shape in different directions, and its properties are going to also be similar in different directions. Meteorites are pretty cool. You don't see them too often. They can have really huge grains, because this is an iron nickel alloy that's forming in the core of a planet. And these grain, this pattern is called a Widman statin pattern. You'll see it if you go to a museum, and they have some polished cross sections of meteorites there. But the grains can get really huge, because these things are cooling over millions of years, and things grow bigger when they grow slower. On the other hand, a casting, this is molten metal, got poured into a mold, and what we can learn from this is that the metal on the outside that was in contact with the mold, it froze first. And then the freezing traveled inward to the center, and so the grains solidified in that direction. So these are just a few examples of different types of microstructures that you can get in materials. That's great. What does that have to do with fracture? So metals typically deform to some extent before they fracture. And if you look at the microstructure, you'll see this is a cross section on the right of a fractured specimen, like up top, or like our slow-cooled nail, if it had broken. And you can see the grains are all uniformly deformed, and then they tear. But if we look at the fracture surface, there it is. If you look at the fracture surface closely in a microscope, so that's scanning electron microscope. And the scale bar here is 10 microns. So we're looking at pretty high magnification. This is about 10,000 times magnification, not something you're going to get with a light microscope. So what you'll see are these dimples on the surface. You see this in all kinds of ductile fractures. And what dimples tell you, that gives you a clue to how it actually failed. Let's try it with this guy now. Okay, so this is just a little cartoon, but what's going on is you get voids in the microstructure, these guys here. And as the materials put under stress, these voids stretch and stretch, and the ligament between them stretches and stretches, and then the voids link up, and that's how a ductile crack grows. I mean, it's still a crack, it's still breaking, but this process absorbs a lot of energy, and it's slow, and that energy would otherwise go into making new surfaces, which are cracks. All right, back we go. All right, on the other hand, when deformation becomes difficult, then cracks grow. Cracks can either grow across the grains, that's called transgranular fracture or cleavage, which always leads to, it puts nerves in my heart whenever I'm doing some kind of search for data on the internet at work. But what cleavage is, so a crack like this wants to grow in a particular direction, or plane through a crystal, and all these crystal lights are oriented slightly differently, so the crack is going to change directions when it moves to a new grain, and this is actually the fracture surface of our quenched masonry nail, and these little facets here, or these cleavage facets are just where those different planes link up. On the other hand, you could get intragranular fracture, which is when the cracks grow at the grain boundaries, and the surface looks like rock candy, that's kind of the hallmark of an intragranular fracture. But the point is, in either one of these cases, you're not going to, on the macro scale, it's going to look kind of like this bolt up here, you're not going to see really any deformation before it fractures. So now we can start to piece together what happened with our masonry nails. It turns out that the steel microstructure also comes in lots of different shapes and sizes, and heat treatment changes the very structure of the steel. So at high temperature, this is our cannonball pile, steel likes to be in this structure called austenite, which is called face-centered cubic. And then if you cool it down, there's a transformation and steel would rather be in this structure called ferrite, which is called body-centered cubic structure. This is actually a giant tribute to iron in Brussels called the atomium, left over from the 1958 World's Fair, but it's very special for us metallurgists. So ferrite, it doesn't like to have as much carbon as austenite, so some of it's got to come out of solution, forms carbides. And then when we look in cross-section at the microstructure, these regions here, it's a layered structure of ferrite and carbide called perlite, very soft, not very strong. These nails are really high carbon steel, so there's a lot of extra carbide, which is the white stuff. When we quench the nail, we cheated. We cheated equilibrium, and we formed a different phase called martensite. So with martensite, there's no time to form carbides. It's super saturated with carbon. It's got a lot more carbon than it wants to have, and so this structure is under high stress. It's very hard, very brittle. You get these needle-shaped grains. The nails, as they came out of the box, were neither of these. They were called quenched and tempered, and tempering just means you heat it up again. You first you quench it, you heat it up again for a while to allow some of that martensite to go to equilibrium, get softer, form ferrite and carbide. But that's the trade-off that we need to get nails that are both hard and strong and tough. We talked about defects on a micro scale, on the scale of the crystal structure. Defects are also important on a macro scale, on the scale of a part or a component. I'm not here to talk about Bay Bridge bolts, but it's a good visual example of a stress concentration. A stress concentration could be anything from a crack or a notch, or even just a sharp edge in a part that you've made, or a part that needs to have bolt threads like this. This notch here is a really great place to form cracks, and what a stress concentration is, is it amplifies whatever stress is on the components as a whole, which is an issue because rather than deforming throughout the whole entire part, it's going to deform or it's going to crack first, right there. This is a representation of the magnitude of the stress at that concentration. These are also things we have to worry about as engineers. Then there's corrosion. Corrosion, it's natural, it's inevitable. It's metals returning back to the ore they came from, if you really think about it. Even the National Association of Corrosion Engineers can't argue with that. Corrosion is an issue of course when it's eating away slowly at all your metal parts, but it's an even more insidious problem when you get localized attack like at grain boundaries. All right, so when you have a situation where you have a material that's susceptible to corrosion, you've got a corrosive environment and some kind of tensile stress, you end up with a situation called stress corrosion cracking. In this case, this is a stainless steel weld, the environment is salt water, and right around the weld, we've got high residual stresses. So what happened? To make this crack grow in this way. So when a weld is basically liquid metal and heats up the area around it as well, so when it solidifies, everything's kind of cooling down slowly, and what happens in stainless steel is some of the chromium and some of the carbon in the stainless steel. It mixes together and it forms carbides right at the grain boundaries, chromium carbides, which isn't really the issue, but now you've got less chromium in the stainless steel to make it stainless, and then the corrosion and the cracking follows the grain boundaries. So what can we do about this? Well, it's like a tripod. You take away one of the legs, no more stress corrosion cracking. We could remove our corrosive environment by absolutely limiting the amount of chloride dissolved in the water. We could reduce the tensile stress by heat treating our weld after we weld it, or we could make our material less susceptible by keeping the chromium in the steel and out of the carbides. So for welding, usually really low carbon grades are specified. That's it. That's all we've got to do. All right, before I wrap up, I just want to put this question to you. Is failure always a bad thing? Can structures ever be designed to fail? How about this? You know, if this perforation weren't here just so to make this part fracture in a controlled manner right there, you would need a can opener to drink your soda every day. I mean, I guess I'm a bit young, but I imagine that's how it was done in the past. Have you ever wondered how the space shuttle was standing on the launch pad one minute and taking off the next? Well, there was a series of nuts called frangible nuts, which means that they fracture, that were holding the shuttle down on the launch pad. And at the moment of liftoff, explosive charges went off so that these things would fracture in a controlled manner. Boom, let the space shuttle take off. So I do think that there is a role for failure and fracture and for understanding how to kind of harness that power in engineering. Why do we care? Well, I think engineering allows us humans to do pretty daring things. And we want those things to be reliable and not fail and not cost us our lives and our society's money. So what can we do? As engineers, especially in research, which is near to my heart, in my former life, we engineers can investigate the mechanisms that make metals brittle at other materials as well. Composites, glasses, ceramics, they all have a role in engineering our world. And we can understand what is the role of chemistry and microstructure that determines how different materials break in different ways. And once we have a better understanding of those things, we can optimize the behavior and the properties of material at the small scale so that we can impact their performance on a large scale. Thank you. Yeah, go ahead. So the question was about the difference in the structure and the properties of your crumple zone of your auto frame versus the passenger compartment. And yeah, there is definitely engineering that goes into the chemical compositions of those different alloys, how their heat treated, it affects the structure to control the way that they deform. And I wouldn't so much say that your rigid compartment has got to be brittle, but you can definitely make it stronger, so a lot stronger than whatever loads you're going to expect it to take on. But it's not going to ever reach the ultimate strength of the metal, you know what I'm saying? So you're not going to see it as brittle behavior, whereas it's desirable for the front party or car, the bumper, and all those expendable parts, where you would prefer the energy to be absorbed, right? Did you say damping? That's a very good point because steel and aluminum, I mean I guess if it's carbon steel and aluminum, they've got different crystal structures, they've got different what's called the elastic modulus, so the stiffness. So rather than reflecting how material deforms permanently, it's more like an elastic band, you know, how stretchy it is. And yeah, aluminum and steel have different elastic properties. So that's definitely going to affect that behavior you experience. Yeah? Okay, cool. Right there in the back. It depends what the application is. I'm sorry, the gentleman in the back asks if you're on a desert island, what one material would you take with you? Or I'm putting words in his mouth. My answer is it depends on the application, of course, because different properties are good for different things, and I'm sure that you can think of many examples from your everyday life, but I mean I think if it were an object, I'd have to take a nail clipper. Two more questions. Another gentleman in the back. I'm curious to learn more about that myself, so I organize technical talks for the American Society materials in the South Bay and Mountain View, and our September speaker is, he's a young Chung from Berkeley, but he's going to come talk about the metallurgy of the failure, but also I think some of the organizational failures and the failure analysis, so I'm really interested to learn more about that. I mean on a technical level, you can work out what went wrong in the processing of a steel. What's most interesting to me is what went wrong in the processes to get that part from being designed, and the raw metals fabricated, and the quality assurance, and how it was actually used, and maintenance, all of these things, so those are, that's some of the books that I recommended for the reading list delved more into the human factors, and I find that so interesting. Sorry, that's not an answer, but I'm interested to learn more myself, and then you had a question right here. So the question was what makes steel and steel and iron alloys a good choice for structural application? If cost were not an issue, what would you be choosing? So cost, I mean in industry, cost can be one of the biggest considerations. If I'm a researcher and I'm developing the coolest new alloy, and it's got all of these cool features, if it's not going to be cheap enough to displace what's already the solution to that problem, nobody's ever going to use it. So there are definitely a lot of components to material selection. Iron is really available. Aluminum is really cheap, but aluminum is not nearly as strong as steel, but aluminum is a lot lighter and stiffer than steel, so that's why you want to make things that you got to launch up into the air out of aluminum, but something that sits on the ground, it's fine if it's steel. Something that needs to operate in a really corrosive environment, like you're processing chemicals, you're making acids, you need to have super corrosion resistant alloys, and that jacks up the price a lot, but it depends on really what do you need it for and how important is it to do what it's got to do. All right, thank you.