SMS_F2013_09

You can hear me okay? We were talking about composites and I just wanted to wrap up a little bit on composites. I talked about composites, we use huge numbers of, compare huge amounts of structural composites, concrete for buildings, asphalt for roads. But most of the work done on composites we talked about is done on aerospace composites, because there's certain things that just can't be built without the lightweight and the advantages of aerospace composites. And I put up the V-22 Osprey before — that aircraft could never have flown if it weren't for high-performance composites.

Which is, concrete and asphalt are not exactly high-performance composites, although there are still people who do research, actually there's quite a bit of research now over in civil engineering on improved concrete because there's so much concrete used and so inexpensive, and they'd like to make it better, perfect it, particularly in tensile strength. A lot of work done on asphalt particularly in colder climates, because the asphalt expands and contracts. And you have to, in Alaska you get a much above Fairbanks and they don't even bother to pave the roads, because you'd have to rebuild the road every spring. The winters are so severe it would contract and crack and whatnot.

But they still do a lot of research. I was at University of Alaska in Fairbanks ten or fifteen years ago and they had a big program for the southern half of Alaska to get better asphalts, because they spend a tremendous amount of money repairing those roads, not because of a deficiency in the material, but there just is no material that survives the severe winters. And that's why you can watch Ice Road Truckers, right, go north of a certain level and they just use ice for the road, right? Anyway.

Composites are fundamentally the marriage of several materials to get, what the composites people tell you, the benefits of both. Now the other side of that is to get the detriments of both, okay, which they don't usually talk to you about. So back thirty, forty years ago, thirty-five years ago, someone up in Canada developed a technique to mix silicon carbide in with aluminum and cast it. And so you had a very inexpensive way of making aluminum silicon carbide composites. I can't remember the name of the company, but it's just the late seventies, early eighties, and they were getting press all over the place okay. There were lots of interest in their material because they were good public relations people.

And so I sort of modeled this particular graph on that material. In fact, this graph comes out of probably the mid-eighties. But the problem is this is sort of looking at the metal specifically, in this case I'm thinking of aluminum, and for any material, if you're going to select it there's not just one parameter, there's a number of parameters that might be important and depending on the application those parameters vary, okay. Strength is always an important property in structural materials. We've talked about toughness, it's an important property. Aluminum is just not a very strong material compared to a lot of other things like the ceramics or steels or whatever, so aluminum, it would be great to improve its strength. Aluminum has excellent formability, excellent joinability if it's not too highly alloyed, it's got very good corrosion resistance, it's actually pretty good corrosion resistance in general, and it's reasonably affordable, and it's lightweight of course as far as that goes.

So this is sort of, in the circle here is supposed to represent how much of a property you need, okay. And if you're beyond the circle, if you're beyond the circle you've got enough of the property. But aluminum just doesn't have enough strength for many applications, and so people thought, well, you just marry that with the ceramic. There's my ceramic, here's my ceramic, okay. A ceramic has tremendous strength, and ceramics, you know, bricks and mortar, lots of aluminum oxide is a ceramic, have reasonable affordability, they aren't terribly — and the ceramics like to tell you that they have fantastic corrosion resistance, and in many applications they do, certainly atmospheric or aqueous corrosion ceramics are excellent. They've got lousy formability — you ever try to whack a piece of ceramic with a hammer, you end up with lots of pieces of ceramic, okay. If you look at joinability, it's a problem. Toughness is a problem. But you could marry it with the metal which has deficient strength, and hopefully you'll get an improved product that has the advantages of both.

But in fact if you look at what you get with the composite, these alumina [aluminum] silicon carbide composites for example, yeah you can get the strength you need as you add the hard silicon carbide particles to the aluminum, but your toughness is lousy, your formability is lousy, your joinability is lousy, your corrosion resistance turns out is lousy, and turns out your affordability bites the dust. So in fact if you put everything together on one plot it kind of looks like this for the all of them. You started out with one material with high strength, one material with low strength, you marry the two and you get a composite between, that is beyond what I need for strength, is beyond the circle, but I lose everything in these other areas, okay.

So the point is, composites can be used as a marriage of materials to get the advantages you must have, but you're going to pay a price, and the two prices are going to be cost price, but there's also property price. Virtually every ceramic, every composite has paid a big price in recyclability, formability, sometimes corrosion resistance. Composites have very specialized niche applications. And in fact if you think of concrete and asphalt they have the largest application and huge volumes, but they're still sort of niche. I mean what else do you build out of asphalt rather than roads, okay. And concrete can be used in a lot of things and it's fairly inventive. And actually if you've seen artisans in concrete, I mean there are some people who do buildings that look like fancy stonework and it's fantastic. I've always been surprised they didn't do more of that. But anyway, the point is composites enhance a few of the properties while degrading other properties, and you've got to be aware of that, you can't just buy one thing.

Student: [question about concrete in tension, inaudible]

Essentially you give concrete zero strength in tension. It actually has two or three ksi strength in tension, but in compression it's got like 60 ksi strength, okay. So if you're talking about loading a beam, okay, here's a beam that's flexible — if you're talking about loading a beam, then yeah the bottom's in tension but the top's in compression. Concrete's great there. But basically on the bottom, when they calculate the strength of that concrete beam, they treat it only steel on the bottom. I mean this is a cross-sectional area of that rebar. And the rebar doesn't have to be welded, you can overlap it. And you just use — and remember the rebar, it'll take a — oh I still have it. [Tom produces a piece of rebar.] Here's rebar, passing around, was galvanized rebar, but it's got these — they rolled on these notches to give it mechanical interlock into the concrete. But you can overlap it, and as long as you have enough overlap it's as if it was joined, okay, in terms of the tensile strength, okay. Does that answer your question? Okay. But yes, in the design they give it zero strength in tension, but that's a little unfair okay. Other questions? Okay.

Composites people will talk about composites being inexpensive. Don't believe that. Composites inherently require more processing steps. That was one of the reasons the aluminum and silicon carbide was so exciting, you could make it from with casting technology, so it wasn't going to be much more expensive. But most of the time you have to make material A, you have to make material B, and then you have to join them together. Now there are some intermediate composites that are, have been used widely, the one I'm thinking of, rather than asking you, playing the quiz game, is fiberglass okay. Been using it for sixty, seventy years, ever since plastics really started to become popular in the 1950s. After World War Two plastics started coming into their own, and people found, oh you can take glass which can be extremely strong, okay. People don't realize how strong glass fibers can be, most lay people don't, but glass can be, if freshly made glass in terms of strength not toughness is stronger than steel okay. It's just like piano wire, it's super strong. And if you coat the surface so it doesn't get attacked by the moisture in the air, then the glass will retain much of its strength. If you have lots of fibers you have the redundancy and you can get a very good structure.

When I talk about glass being strong, when it's freshly made, glass has no imperfections on the surface okay, if it's made properly. And it's the imperfections on the surface of a brittle material that weaken it. And so I can't do it, my old thesis advisor used to teach 3.091, and he tells me that he used to live, him and still lives up on the North Shore, but they had a Sylvania lightbulb plant up there in Danvers, and he used to stop in the morning before his lecture and get freshly made lightbulbs okay. There's fresh-blown glass, and he'd bring them in and for his 10 o'clock lecture he could throw them across the room and they would bounce because they were freshly made. The glass had not been attacked by the humidity in the air. After a few days you throw it across the room and it'll shatter just like any lightbulb you know of okay. But freshly made glass, before it starts to get etched by the moisture in the air — it actually does on a microscopic scale, but it's so brittle these submicron imperfections destroy its strength, okay.

But fiberglass, people making fiberglass realized you can have tremendous strength, and you marry a protective adhesive coating, basically the plastic, with the fiberglass, and you end up making boats that don't corrode, which is an advantage. What else? What automobiles, what were the first fiberglass automobiles? The Corvette. Anybody can make a fiberglass $50,000 car, you can't make a $20,000 car out of fiberglass. Although in the 1990s we tried to start making truck beds out of plastics, or plastic resin molds and composites and stuff, and it turns out people love them. Joel Clark was working with, I think it was Ford, and they made some of these truck beds for pickup trucks, and they gave them to a mine site to try to, you know, get a lot of wear on them, and they gave some steel truck beds and some fiberglass truck beds, and they were supposed to put them in service for six months and send them back so they could evaluate them at the research center to see how they did. The companies, the coal companies didn't want to give back the fiber of the resin mold and composite beds of the trucks, because they didn't — they didn't crack, you know, they were resilient. The steel dented and you couldn't close the tailgate and all this other stuff, you know if you're dumping things into a pickup truck and abusing it like they do in a coal mine. And so it turns out the fiber, not fiberglass, but the resin-molded composite truck beds — and now what they do is they give you liners okay, it's still steel on the outside, and the reason for that is the plastic is still just more expensive okay.

This is something I put together years ago, but for high-volume structural materials — sorry about, I have to make some of these again — we look at both cost and fracture resistance. We quickly find that for constant volume, okay, the approximate cost per — for actually this cost per pound in dollars — steels back then was about 20 cents, about 40 cents now. Aluminum is a buck twenty. Simple composites like this truck bed were two to ten dollars a pound. Now you get five times the volume with that, so it's down to 40 cents to $2 on a volume basis, but it's still more expensive. And plastic, just simple plastics not composite plastics, you know, are still more — everything's more expensive, even wood is more expensive than steel. Aluminum has a two-and-a-half-fold advantage in density, and so doing this on a cost-per-pound is not necessarily the fairest way to do it. You can figure it out, and there's all kinds of metrics you can use and stuff to figure out what is best. That's about all I'm going to say about composites right now, I may talk about it a little bit more later. Does anybody have any questions on composites or anything else? Okay.

So I want to talk now, I'm going to get into specific materials and what their advantages and disadvantages are. And what's the first material I'm going to talk about? Steel okay. Actually composites was sort of the first material I talked about, but steel, because it's in such high volumes. It's made everywhere okay. There are hundreds, literally hundreds of very large steel companies around the world, and I told you whole countries will invest in building their steel mill, because the big integrated steel mills are too expensive for a single company to build. In the last forty years, what people don't realize, too — let's take aluminum, the second most widely used metal — and on a weight basis there's forty times less aluminum made than steel. Aluminum is a little over two percent of all metals, steel's over ninety percent. So on a weight basis aluminum's forty times less, on a volume basis it's twenty or fifteen or twenty times less in volume of material that we make. Aluminum's been around technologically almost as long as steel, but in fact these — well it's because steel has this unique combination of strength, toughness, and cost.

Gone over that night up here, but you need to understand that there are hundreds of steel companies. How many aluminum companies can you think of? Well you can think of one, Alcoa right. But there are other aluminum companies, you've heard of Kaiser Aluminum, maybe. No you haven't. If you're from the West Coast you would have heard of Kaiser Aluminum. You've heard of Reynolds Aluminum, right? You might know how Reynolds Aluminum got started. It's the same Reynolds as Reynolds Tobacco, and they needed something to wrap their tobacco in, and so they started making a little foil around 1900 and they kind of grew the aluminum business separate of the tobacco. But it's the same RJ Reynolds Tobacco Company, RJ Reynolds family out of Richmond, and that's the headquarters of Reynolds. There's Alcan, a little company of Canada which is actually a spinoff — trust busters broke up Alcoa. And Boeing, I told you how Alcoa, Boeing, and Pratt & Whitney were all once together, and at some point they broke them up, and then they also broke off the Aluminum Company of Canada, Alcan. Alcoa just had a monopoly, but there was Charles Martin Hall who kind of was the founder of Alcoa, but there was Paul Héroult in France who had also discovered the same thing and just got to the patent office a little bit too late, and Hall beat him. But his company was Pechiney in France, and Pechiney was just bought by somebody else, but anyway I can't remember. But anyway you can't even name very many aluminum companies. I can't name more than a dozen, and I've been in this business for years, longer than you've been alive. But I could name, I could go off and rattle off the names of dozens of steel companies.

Although to a certain extent they've been consolidating into some huge conglomerates, mostly because their profitability was so bad. And why was their profitability bad for the last thirty years? Because their productivity was so good okay. They had improved their productivity so dramatically that they almost drove themselves out of business competing with each other because of their increased productivity, okay. And so Wall Street, those people over at that side of campus, can't figure these things out. That was not a good — that — they were not an ancient material that was no longer useful. They thought they were just an unprofitable backward industry. Well they were, maybe some of their management was backwards, but it's still an important material. And there was this guy [Mittal] in India who just started buying up steel companies for a song, and now ArcelorMittal is probably the second largest steel company in the world, and he's making billions okay. Actually starting to lose some billions right now, but anyway.

So what is the most technologically advanced material you can think of? Most people would think silicon, all right? We know a lot about silicon, right, it's fundamentally semiconductors. What's the size of the silicon industry that makes computer chips, you know the Intels of the world or the Advanced Micro Devices out of Boise, Idaho? Anybody have an idea how big that industry is? Tens of billions, well actually Intel's about 30 billion last time I looked okay. It turns out about 200 billion. Intel is a significant fraction of all the world's semiconductors, and I'm talking about the chips, I'm just talking about the silicon. The foundry business to grow single-crystal silicon, that might be 10 billion dollars or something, but the actual chips when they add all that functionality to the little one-path silicon, single-crystal silicon, it's a couple hundred billion. Whereas the steel industry tends to be about a trillion-dollar industry worldwide, and the aluminum industry tends to be about a hundred to two hundred billion dollar industry. So the aluminum industry and the silicon industry are similar in size, and we often think — if you go and look at the course offerings at MIT, you would think that silicon is the wonderful wonder material. This is the silicon age, and it is a wonder material in many ways, but in fact the most technologically advanced material in terms of how much we know scientifically about it is steel okay.

And I've only said that to a few people a few times. I haven't — well I've had some faculty challenge me on that and then I start pointing out some things. But we know a tremendous amount just because of the size of the industry. There are people, all kinds of people around the world doing research on it because of this economics and importance. Now it turns out it has not seen tremendous growth in properties, whereas when you look at silicon you can say, well the number of transistors on a chip is doubling every two years according to Moore's Law and stuff, and that's true, and the performance of the semiconductor is much greater. And people say, oh but silicon, you know, it may only be two hundred billion dollars to sell the chips but it goes into all kinds of products that are worth a lot more than that. And I said, yeah but did you know steel goes into products like buildings and bridges and automobiles that are worth a lot more? You know about twenty-five percent of the world's GDP relates to structural materials, okay. Bridges, buildings, aircraft, I mean — and steel's a large fraction of that.

Why does steel have some of these advantages? Strength, toughness, cost, but there's other things. The ability to form it, it's actually very easy to form steel and stamp out, you can bend big heavy sections, you can join it easily. One of the advantages that it has — Christine's thing about, she wants two swords — is you can form it when it's soft and you can transform it to being hard. That's also an advantage with aluminum. Both steel and aluminum, their inherent strength: steel's about thirty, thirty-five ksi basic strength, and you can harden it to almost ten times that in certain conditions. Aluminum is about eight ksi, and you can even better than ten times, you can get about twelve times the strength. And so actually your homework problem for Monday is to think about what is the highest-strength application for aluminum alloys, and I'll bet none of you get it. You're going to guess aircraft and missiles or spacecraft, it's not okay. Anyway, I don't know if you can even find it, you can find it on the web, but you'd have to know where to look, okay. But I asked you for a homework problem, and I think Dr. Bell [Ballinger] Martin gave it away — he tells me I gave it away earlier this semester.

What is the Achilles heel of steel? Corrosion, rust, yes okay. Why do they galvanize that rebar? Why are they, when you're driving around New England in the summer, why are they always tearing up the bridges on the highways? Because the rebar in the concrete rusted. The rust expands, causes the concrete to fall, and you end up with potholes on your highways. If someone hits them at 70 miles an hour they'll have a wreck. And so they have to go in about every twenty years and rebuild the bridge deck to replace the rebar. People have tried making stainless steel rebar, plastic-coating the rebar, going to stainless steel rebar, they've even gone to nickel-base rebar, which you couldn't afford to build all the bridges, there's not enough nickel in the world to do that. But nonetheless people have tried all kinds of things to fight that form of corrosion. The corrosion scientists estimate that in the United States we lose about 200 billion dollars a year in corrosion, the cost of corrosion. And guess what, probably 80, 90 percent of that is steel because it's 90 percent of all the metal made is steel okay.

So the Achilles heel is corrosion, and I did have something — talks about the eight forms of corrosion, and I have a handout for you. Comes out of the eight forms of corrosion. There are actually two different eight forms of corrosion. One comes out of a book, the seventh, third edition, Fontana and Greene, which is — Ohio State has a big corrosion research center going all the way back to the 1930s. The professor, Lou Tennyson, who used to be your Professor Ballinger's thesis advisor, came out of Penn State and Ohio State studying corrosion. And Fontana and Greene have, their index of the book talks about the eight forms of corrosion, except their eight forms are different than this eight forms. Good thing I handed out something.

This is mechanics of attack okay, and the eight forms that Fontana and Greene break corrosion into are sort of the chemistry of attack, okay. So they're a little different. But there's something called uniform attack — in the area of steel we call that rust okay, so far as that goes. We have localized attack, which can be pitting or it can be gouges with crevice corrosion. We have erosion, cavitation where you have bubbles collapse and great pressures that break off the protective surface oxide. One of the problems, most steel does not have a protective surface oxide. Now there's a lot of steel that does. [Tom produces a piece of black iron pipe.] Here's a piece of steel, black iron pipe. If you go around the basement of most buildings or a boiler room or something you'll see this stuff, it's unpainted okay. So black iron pipe when it's hot coming off the rolling mill will form at high temperatures Fe3O4, which is called magnetite, which is a magnetic form of iron.

And rust, which forms at lower temperatures in the air with the moisture around, is Fe2O3 dot a bunch of H2Os, and that's hematite. And there's also wüstite at higher temperatures. But turns out magnetite has a volume ratio and coefficient of thermal expansion that is such that when it forms on steel, it forms and has a coefficient of thermal expansion that it will adhere. Rust on the other hand, when it forms, can have a four to ten times volume change depending on how many H2Os are in it and how much chlorine might be around and whatnot, and it will flake off. And so we've all seen rust okay, flaking off. Typical atmospheric corrosion rate of steel is about four thousandths of an inch per year. So forty years you lose an eighth of an inch okay or so, from both surfaces. So if it's a big heavy steel beam, you know, why are they going to rebuild the Longfellow Bridge? It's rusty okay, but it's been there for a hundred years.

But they started out with big heavy sections, and even if it lost an eighth of an inch in forty years, the growth usually goes as the square root of time, so in eighty years it only loses a quarter of an inch. But after eighty years or a hundred years or whatever, the Longfellow Bridge has gotten rusty in enough spots — I mean if you've driven by it or ridden, walked across it or whatever, the handrails where the snow sits, and let's face it, snow in New England is acid rain, it's just frozen acid rain. And when it melts, it's pH 5.5, and it will start eating away that steel. And you see the fence rails, the guardrails to keep you from falling into the Charles, are all rusted away all the way through, perforated, after a hundred years okay. Now some of them, it's cast iron, and cast iron forms — itself, it's got a lot of silicon, and it forms a glassy silicon layer, so it's a lot better, or grows less. Cast iron grows less.

But there are these eight forms of corrosion, and actually I do a short corrosion course for the Navy folks, of about six or seven lectures. Anyway, so there you can talk about eight forms of corrosion. One form, not all forms, occur to steel, but one form that does occur is stress corrosion cracking, okay. Steel will form cracks in certain environments even if it doesn't rust. It can crack in other environments. Stress corrosion cracking is most common with the stainless steels, and we'll talk about that maybe on Monday. But the form of stress corrosion cracking that's most harmful to the steels is hydrogen embrittlement, and Dr. Ballinger talked about that. But hydrogen embrittlement: the steel has been around, exposed to the atmosphere, and hydrogen — it reacts, iron, and the oxygen forms Fe3O — Fe2O3 releasing hydrogen when it reacts with the water. And if the hydrogen diffuses into the steel under certain conditions, or if it's there from the steel manufacturing, and you put it under tension, it will cause what we call delayed cracking.

Basically they call it a static fatigue test. Just take a bar of steel, it's got hydrogen in it, or you can charge it with hydrogen while it's under tension — you put it under tension, if it's got hydrogen in it and you're at more than maybe 40% of the yield strength in tension, the hydrogen will diffuse to little imperfections like inclusions in the steel. And the hydrogen likes to go to the larger crystal lattice, which is the part that's in tension. If you have an imperfection around an inclusion and the inclusion is in tension, then you will essentially — if you think of your volumetric strain field, there's more space here on the sides, and the hydrogen will go there. It will weaken the steel, and you just hold it under constant stress, not fatigue, just hold it under constant stress, and in somewhere between 48 and 72 hours it will crack, it will separate okay.

So they're welding nuclear submarines down in Groton, Connecticut, and if they don't control the hydrogen while they're welding these big egg-crate type of constructions to make a high-strength pressure vessel, guys will be welding during the day and the guys at night who are welding will hear a bang like a rifle shot in the shop. And that's a crack, that's just a brittle crack that just ran through the steel and broke it right in two okay. So the hydrogen embrittlement, also called delayed cracking. Now it doesn't usually occur, well it doesn't occur usually within the first one to four hours. If you look at ASTM specs, if you do something like electroplate steel, that could introduce hydrogen, the ASTM specs say you must put it in an oven at 375 degrees Fahrenheit. So this is like cooking a cake, it's not terribly hot, and that will diffuse the hydrogen out of the steel before it has a chance to accumulate at the imperfections and cause cracks.

People have taken polished sections of steel and put them in a microscope in an aqueous solution, and they actually can look in the microscope and see bubbles of hydrogen forming, okay, right there at those crack tips okay. Over time the hydrogen accumulates at this, the tensile stress region, because there's more space, and if enough hydrogen gets there you get hydrogen embrittlement, and steel will crack. Dr. Ballinger sent around a piece of bolt that had cracked in service, and then he and I are still debating about whether it's hydrogen embrittlement or stress corrosion cracking. The difference between hydrogen embrittlement and stress corrosion cracking — seventy percent of the time metallurgists make no distinction — but in fact one occurs at the anode and one occurs at the cathode. And yesterday you learned about anodes and cathodes in corrosion. Hydrogen embrittlement occurs at the cathode, corrosion, oxidation corrosion, occurs at the anode. And you can very simply change the polarity, and you'll either be cathodically protected and you'll get stress, you can get stress corrosion cracking, you won't get stress corrosion cracking if you're cathodically protected, but you might introduce hydrogen and get cracking for that, from that.

There's a little heat-treating shop over here near the airport. Once a number of years ago, they just put in a big steel tank where they were going to heat-treat — they were making steel toes, I'm surprised they're doing this in the United States, but the steel toes for steel-toe boots okay. And they're very high-strength steel, very ductile steel, because you want lightweight steel toes, and you don't want it to crack if you drop a brick on them or whatever. And so they were putting in a big heat-treating tank, and it was all welded construction, and they were filling it up like a swimming pool, about 50 yards long and about 5 yards wide and 3 yards deep, and all of a sudden it started leaking. There were cracks, and it turns out I looked at the cracks around the welds, weld residual stresses, clearly looked like stress corrosion cracking. And I go to look in the literature, and I find that they were using sodium nitrate salts for the heat-treating bath, and I found that, oh yeah, this is sodium steel and sodium nitrate is susceptible to stress corrosion cracking.

So the guys I was working with, this firm that Steve used to work with, the guy wanted to run a test. So we took it, we did a test, we just made a U-bend out of some of the pieces of steel, put a bolt through it, stuck it in the solution and made it anodic, couldn't crack it. We ran tests for three weeks and we couldn't crack it. And finally I thought, you know the literature says it's stress corrosion cracking, what if it's actually hydrogen embrittlement? And so I switched the polarity on the test, broke within 24 hours, okay. So polarity does make a difference. And it turns out the literature was wrong, they called it stress corrosion cracking, it was actually hydrogen embrittlement, but most of the time metallurgists don't even make a distinction okay. But there can be a distinction, and in that case it was a big distinction, and all of a sudden we had the right mechanism and everybody was happy okay.

So hydrogen embrittlement occurs all the time. The US Navy down here in Groton was going to a higher-strength submarine steel, from HY-80 to HY-100. Didn't seem like a big change. They had promised Congress in the early nineties that they would eventually start making the 21st-century submarine out of HY-130, which they developed in the 1960s, but they never had the nerve to actually build a whole ship out of. So they're starting to build a whole ship out of HY-100, yep, a whole ship — this is the Sea Wolf, the first edition of the 21st-century submarine, around 1992. And so it's higher strength by 20 ksi, going from 80 to 100 ksi yield strength for this high-strength steel. And a grinder — they look for flaws that are an eighth of an inch in size or larger, and without going through fracture mechanics, they were looking, they had all their inspectors had looked for those, found none in the welds. And some guy's grinding the surface of the hull to make a smooth contour — you can't even have a little weld reinforcement, when you're going through the water it creates enough noise that they can pick it up on sonar okay. So they have to grind it smooth. And this grinder's grinding, and he noticed that the grinding swarf, the little iron oxide particles, was sort of lining up in these little less-than-eighth-of-an-inch-long parts in the weld metal. And he was smart enough to tell someone, and they came in, they did more and more thorough non-destructive testing, and all the welds — 18 percent of the hull had been completed — all the welds were full of little microscopic cracks okay. They had never seen before.

And so it ended up they had to scrap the whole 18 percent of the vessel. It slowed down the whole program by about a year. It ended up costing the Navy about two billion dollars okay to fix the problem. And one of the ways you fix the problem is you heat the steel okay, you have to preheat the steel, so that when you're heating it it's like it's already in an oven. The heat from welding raises the temperature above that 375 and it'll diffuse out. After you finish making your weld it's still hot and the hydrogen diffuses out. Well the harder the steel is to weld, the higher the temperature you have to preheat. And so for this particular steel which is highly alloyed, they actually had people going in and welding in what they call blue jelly suits, because they had to preheat the steel to 400 degrees Fahrenheit. And these guys were literally welding inside an oven. They had a mask so they didn't have to breathe 400-degree air okay, and they would wear a suit that — they had a blue liquid and they actually recirculated the liquid. And I mean these guys were sent into these closed containers in the ship to weld the steel together, and it was heated to 400 degrees Fahrenheit. That's how desperate they were to keep welding so Congress wouldn't cancel the whole multi-billion-dollar program.

Anyway, there are lots of different types of steels. If you go look at the metallurgy of steel — again, how many people in here are not Course 3? That's okay, so half of you. I'm going to show you a phase diagram, I apologize, not everybody can appreciate phase diagrams. However, this phase diagram you should learn to appreciate, those of you Course 3. This is the iron-carbon phase diagram. Is anyone who is in Course 3 not ever seen this before? Okay, so you've all seen it. I had a graduate student once who had been in mechanical engineering from another university, and Professor Elliot, who's passed away, put up the iron-carbon phase diagram in a class and just started talking about it. And the student raised his hand, says, "Professor, can you explain what that is?" He says, "You've never seen this iron-carbon phase diagram?" Just tried to humiliate him in front of the class, anyway. This is the iron-carbon phase diagram. Steels are everything on this side below about 1% carbon.

The maker of this diagram — doesn't say it on this one — is a guy named John Chipman okay, our departmental, you know, breakout room is the Chipman Room. John Chipman taught the world how to make steel about 1940 okay, and he's, in the steel industry he's a hero. Anyway, in any case, this is the iron-carbon phase diagram. Steel has a face-centered cubic crystal structure up here, body-centered cubic, and you can go through, you can quench it, you can heat it up in this region, quench it down to this region, and it will form a metastable body-centered tetragonal, similar to the body-centered cubic, which is martensite, which is extremely hard. You can go from this 30 ksi-type material that you started with, you formed into a sword or bolt or whatever, when it was soft, and you can quench it in water or oil. The swords, they used to take a living slave — they thought this gave the swords more power, if you took a slave and you just quenched it by sticking a hot sword through — they'd lose more slaves that way. But anyway, but it's true, they thought this gave the steel vitality, okay. This is sort of making someone sick in the back, right, but this way it was, I mean the world was different. Actually the world is still the same way in certain parts of the world, but anyway.

Student: [comment about cauterizing, inaudible]

It cauterizes the wound, the hot steel, yeah. We used to, in this country, I know — I tell you that my oldest daughter was born in Saint Elizabeth's Hospital in Bethlehem, Pennsylvania, beautiful hospital, because they were always bringing maimed steelworkers down to the hospital. I mean, some guy gets his arm cut off by a piece of hot steel, just sliced through your arm like butter, but there's no bleeding because it cauterizes as it goes. Anyway, so you quench it and you get an increase of five or ten in strength, okay, except now it's so strong that it's somewhat brittle, and you have to temper the steel. So you go to a True Value hardware store and you'll see True Temper tools. They've quenched, they've heated it, they've quenched it in water or oil, they then have to temper it to 500 or 1000 degrees Fahrenheit. This is on, 1000 degrees Fahrenheit right in here. You tend to avoid the 600 to 900 range because the steel can get brittle for other reasons.

And this explains one of the reasons why there are so many steels okay. There are so many steels because there's several things. One's called hardness, one's called hardenability, and then the other is called hardening ability. So this is out of a materials dictionary, and if you look at, hardness is the resistance of any material to indentation. So a hardness test, you take either a diamond or a hard steel ball and you just put a weight on it and you see how deep it goes into the surface of the material. You're probably only going to go in a hundred microns deep, a little thicker than a human hair, to get a little indent. And you measure the depth very precisely in a special machine, and you get a hardness number. And so this is used all the time in industry. Less than, if you go out and get a bunch of them done at some testing lab, they'll charge you five bucks a pop okay, to get a hardness test.

There's hardenability — in ferrous alloys, being steel, the property associated with chemical composition of the base metal which determines the depth and distribution of hardness that can be induced by quenching, what I told you about quenching. And how deep will that go, and how is it distributed? The hardness will be maximum at the surface and it will die off, but well how does it die off? And there are whole books listing all the different steels showing you the depth of stuff. And so a hundred years ago if you'd been a graduate student in this department, you might have been generating these plots for different compositions of steels and trying to understand why different things do different things, different stick compositions behave the way they do.

Hardening ability is, in steels, the property that describes the maximum hardness of base metal can be treated. It turns out the maximum hardness is very strongly a function of the carbon content of this iron-carbon alloy. If you're over here at a tenth of a percent carbon, like automotive steels, the maximum hardness might be Rockwell C 25, whereas if you're over here at six-tenths of a percent carbon, only six times as much, the maximum hardenability might be Rockwell C 60. Well Rockwell C 60 is as hard as a file okay, and if I put it in a vise like this and hit it with a hammer, hit it, even if it's been tempered, I could probably shatter the file if I hit it hard enough. Take that ten steel — it might make an automotive sheet out of, quench it and temper it — I can't get 200 ksi strength, like a nail file or a metal file. But if I put it in there and I quench it in water and I do it well enough, I might be able to get 100 ksi okay. I can't get 200 ksi. So the carbon content is important.

So if I start looking at the actual compositions of steels — these are transformation curves, we don't have to go through that. Actually this is the plot of hardness versus carbon content, and you can see at low carbon contents you're going to be Rockwell C 25, and higher carbon contents you'll get up — actually this one is, oh this is hardness of pearlite, this is not the one I was thinking of. This one would be 0.1, but it's a similar shaped curve, the increasing in carbon and then leveling off, the one I really wanted. But there are lots of steels briefly, and we'll talk some more on Monday, but they have different designations. A 1050 steel has 0.5 carbon, a 1065 has 0.65 carbon, and these are all the plain carbon steels.

And then you get down here to the ones that are 4000 or 5000, and those are what we call alloy steels. They have nickel, chrome, vanadium, niobium alloying elements, and they will have different amounts of carbon, but they have different alloying elements. Some of these things, the plain carbon steels, you probably can't harden more than an eighth of an inch deep. Well if you start putting nickel and chrome and things, you can make something four inches thick or eight inches thick that has a hardening depth, a hardening ability or a hardenability that can go for eight inches deep okay. If you don't put the alloying elements in, you'll get just the top eighth of an inch that's hardened okay. So you need to add alloying elements for heavy steels, you need to adjust the carbon for the maximum hardness you're going to get. And if you're an automotive company you want just the right amount of everything, because it could make a difference of $10 a ton or $20 a ton, and if you're buying a million tons then it makes a difference okay.

And so there are literally hundreds of steel compositions, and you need to be a heat-treater, a steel heat-treater, to be able to figure out which one of these you're going to use okay. But it's because mostly the automotive industry and railroad industries, they want just the right composition for their thickness and their wear resistance or corrosion resistance or whatever. And so they use different alloying elements, different amounts of carbon, and so there are literally hundreds of variations because when people buy a million tons of steel they want the lowest possible price, and tenths of a penny per pound make a big difference okay. So I'll see you Monday. I'm actually lecturing.