MSE_F2017_05

Materials Selection and Economics Fall 2017 Session · 9 sections 16 cases · Watch on YouTube ↗ all files
Layer 3 — readable edition

§1. Course logistics and working habits [00:00]

§1.p1

We are encouraging you — Dr. Belmar and I both have to be away most of the week of October 2nd, so we're giving you that week off from coming to class at 9 o'clock. If I were you, I would watch some of these other videos, whether at 9 o'clock in the morning or some other time. Somewhere around the end of the second week in October is when we want the one-page summary on one of your modules. We're trying to get you to finish everything up by Thanksgiving — paper and all that other stuff. Front-load this class. Learn to get things done early rather than late. There are certain advantages to that, however most people don't necessarily believe that.

§1.p2

I've always gotten in early, and one of my graduate students had a hard time getting in to see me when I was department head. Students had to come see me at 5:30 in the morning, because from nine o'clock on I just did meetings all day. Terrible life. He told me a couple years ago that the way he solved the problem, he would just stay up all night, come to see me at 5:30, and then go to bed. Works for him.


§2. Sprinkler systems and plastic pipe failures [01:43]

§2.p1

We talked about conveying gas and different materials that have been used through the ages, and how we now tend to use a lot of plastic or corrugated stainless steel, and these materials tend to develop different types of problems. Let me go to another material that has developed problems that were sort of surprising. We used to use carbon steel pipe for sprinkler systems. This is a fire sprinkler system, and this one is actually now spreading a bunch of water. These things operate around 200 psi, so this gets a lot of water. But no one has ever died in a home that's had a sprinkler system in over a hundred years. So it works. You might end up with a home that, just like Houston, is flooded out, but it works. It keeps people from dying in fires.

§2.p2

Nonetheless, they decided they wanted to save some money. Instead of using copper or steel, they decided to go with plastic. It turns out it's not a problem with the plastic, but with the water that comes in on some of the old steel pipes. They put a biocide on the inside, they sprayed the inside of the steel pipe with something to keep the bugs and the corrosion down. It turns out that polymer they put on the inside of the steel pipe gets in the water, and it's stagnant water — it's not flowing through like in a potable water system, it just sits there — and it cracks the plastic. So there are lots of buildings in this country right now that have sprinkler systems leaking all the time. There are a lot of secondary effects you don't always think about that you run into.


§3. Gas turbine materials and the temperature ceiling [03:40]

§3.p1

I want to shift gears and take a couple of industries and talk about what's happened in some of them. This is the gas turbine industry. If you're interested in aerospace and engines — Whittle was the British guy who developed a gas turbine; there's a nice quote from him about how he invented the gas turbine around 1936. Then during World War Two it actually came in. The Germans had V2 rockets and things that had turbines — some of the first gas turbines. Those are the original gas turbines. If you go back in time, they didn't operate at very high temperatures. In fact, most of them worked at stainless steel temperatures, 500 or 700 degrees centigrade, which is not terribly hot.

§3.p2

People started working on better and better materials. These were uncooled solid materials, and they started going to nickel-based superalloys in the 50s and 60s and 70s. As they went to the nickel-based superalloys, they learned you didn't like finer grain sizes — at high temperature that meant creep; the metal would deform. So they went to directionally solidified turbine blades, and they went to single crystal blades by the early 1990s. The guy who really led that at Pratt & Whitney was Gell [unclear — captioner garble; likely a Pratt & Whitney metallurgist with MIT faculty ties], who had been a faculty member here at MIT in both mechanical and materials. He went to Pratt & Whitney and led all that effort in the 60s.

§3.p3

Then they went to cooling. [Tom holds up a turbine blade with internal cooling passages.] It's got internal cooling passages. If you take some of the other courses, I'll talk more about how they make that. They went to internal cooling, and the interesting thing here is you see this jump in temperature — about 1,300 degrees centigrade. That's above the melting point of the material. If you didn't have that cooling, your engine would melt. The cooling is actually thousand-degree Fahrenheit air that comes from the compressor. As long as the turbine is running, you always have cooling air — unless something gets plugged up.

§3.p4

Then they came to thermal barrier coatings. [Tom holds up a stator with a ceramic coating.] This is a stator with a ceramic coating on the outside. There's a lot of technology in these thermal barrier coatings, TBCs. That allowed them to get to even higher temperatures. On a growth-rate curve, you can see when you just had solid metal, there's only so much you can do with the metal. When the metal melts at 1250, these things are getting pretty close to the melting temperature of the metal, and you can't exceed that unless you do some active cooling.

§3.p5

All the thermal barrier coatings are zirconium oxide. Anybody know why? Because it's the only ceramic that matches the coefficient of expansion of the nickel. Most ceramics have about one-fifth the coefficient of thermal expansion. You put a ceramic like straight aluminum oxide on the nickel alloy, it's just going to crack off as things heat up and cool down. They actually have to put a fancy metal inner layer, typically intermetallic nickel-aluminum or something that's nice and porous, and then they spray the ceramic on. It actually vaporizes — an electron beam vaporizer, a $2,000,000 machine, and each blade may have to be in that machine for an hour as you vapor-deposit the zirconium oxide. That's why these things can cost $5,000 a blade. They're not cheap.

§3.p6

There's tremendous cost savings as you go to higher temperatures, because this is just what you learned in thermodynamics about heat engines. Delta T over T is the theoretical heat efficiency of the engine. Now they're predicting we're going to go to silicon-based ceramic matrix composites — different materials — and different types of coating, then heat pipe cooling in the future, and eutectic oxide ceramic-metal composites. Don't plan on seeing these things in your lifetime, because this is where someone's predicting they're going to have all these fancy things. But why are they doing it? Because every 50-degree-centigrade increase in operating temperature, that engine is worth two billion dollars a year to the commercial airlines in fuel savings. It doesn't take much to end up with a lot of money. We can talk about the value of a pound saved in a structure, but when you start looking at the engine driving it, you can have multiples, and that's why people spend so much money on engine technology.

§3.p7

Student: [inaudible question about ceramic matrix composites]

§3.p8

This figure was probably developed in 2004 or '05, and people were looking at niobium silicides and things like that, because they're intermetallics and have higher melting points. They're not 1250 centigrade as a melting point — it might be 2,000 degrees centigrade — but they're brittle as can be. No one's solved that brittleness problem. They say, well, we're going to eventually solve the brittleness problem in our ceramics. Yeah, good. I'm not invested in that one right now. But people are working on it.


§4. Military vs. civilian engines, and Mach 17 [10:01]

§4.p1

I remember serving on a committee back in about this time frame that was looking at how the Air Force should spend its three hundred million dollars a year in research money for engine development. It was a very interesting committee — there were about 40 of us on it. Two of us were materials people. A number of people were from General Electric, the chief engineer on such-and-such engine. One guy sitting next to me was the guy who developed the F100 engine for Pratt & Whitney, I think that's right. There was a problem starting in the 1990s: military technology for engines diverged from civilian technology, and the civilians went to high-bypass engines. You know what high-bypass engines are? I'm seeing a lot of nods.

§4.p2

It's simple freshman physics — if you have mass and you have momentum and energy, one's m times v and the other is one-half mv squared. A high-bypass engine, you can blow a lot of air through there, so you have a big M but not as much V, and you get a more efficient engine. But it doesn't have the performance of one of these things where you can just dump fuel in and basically — well, there are turbojets, but this is basically when you just throw fuel into your afterburner.

§4.p3

I grew up in Virginia Beach, and we had two naval air stations right there in the Norfolk area. One guy was out flying over the Atlantic and he had a flameout, and he just turned on his afterburner and he made it back to Oceana Naval Air Station within a hundred yards. He kind of crashed within a hundred yards of the field because he just ran out of fuel. He had probably half a tank of fuel, but to go 10 miles from out in the ocean to Oceana, he burned a whole half a tank of fuel — but you can go really fast. In his case, he didn't have a choice because the main engine had failed.

§4.p4

So the military technology was looking for performance and not efficiency, and they started diverging, and that's been a real problem for the whole engine development, because the commercial engines before that all built off the military technology. Now they can't, because they're heading in different directions in their designs. So I was sitting next to this guy, and at the beginning of the committee they asked everyone to go around: what are the key points we ought to develop? The blade designers — the guys who design these cooling passages — would say, well, we need better materials, we can go to higher temperatures. And those of us materials people would say, well, we've kind of gone to the limit of our materials.

§4.p5

I can go back to an Ashby plot — maximum service temperature — and there's nothing that goes to 3,000 centigrade. It just doesn't exist. We don't have things that have strong enough bonds. Diamond might do it if it didn't sublime before then or burn in the air. So we would say we need better designs. Well, this is like a 1975 design. This is a 1980 design. This is not the latest stuff. They're getting these things down to less than a millimeter thick — a half a millimeter thick. Do you know how hard it is to cast that with any accuracy?

§4.p6

The reason we have CAT scanners for metals today, primarily, is because every turbine blade gets a full CAT scan to measure the wall thickness. You have a core that you cast this thing around, and you grow a single crystal around it. If that core is off by two or three thousandths of an inch from its perfect position — and its perfect position is not just a center axis, it's the center axis in a twist — you're off by two or three thousandths over a couple of inches on something that's going to heat up to 1,300 degrees centigrade. It'll melt, and it's a piece of junk. All of a sudden, your manufacturability goes like that.

§4.p7

They continue to make progress because they spend billions of dollars. The problem the Air Force had is they only had 300 million a year, and they wanted to know how to spend it. The other problem they had was that they didn't just want regular old engines. They would have been delighted with people developing engines that could go Mach 17 in the air.

§4.p8

I remember going to this meeting and saying, wait a second, who needs Mach 17? Even the Air Force — I said, why do you need Mach 17 in the air? This is unmanned, isn't it? They said yes. So at least I got that right — it wasn't manned.

§4.p9

They gave me an answer which I thought was interesting. Remember this is like 2005, so 9/11 had already occurred, and Osama bin Laden had not yet been captured. They said that one time, a year or two before, they knew where Osama bin Laden was — this is all classified, and they didn't give us the details — but the closest ordnance they could get to him, because he was off in some mountain in Pakistan or Afghanistan, was on an aircraft carrier about eight hundred miles away. They only had 10 or 15 minutes to get the ordnance from 800 miles away to where he was, to wipe him out. They needed Mach 17 to do that. Well, I guess if you could just shoot a laser anywhere in the world. But Mach 17 is what they go in space with, no atmosphere.

§4.p10

At Mach 17, your skin temperature is going to be — from heat flow and friction heating in air — 10,000 degrees. I said, there is no material that can do that. They said, oh, we'll just use active cooling. Let me tell you what these guys think of for designs. They were going to do this for the National Aerospace Plane. You know what the National Aerospace Plane was? It's like Elon Musk now. We're gonna go from Los Angeles to New Delhi in two hours, because you're gonna go up into space, and you're gonna go Mach 17 up there, and you're gonna re-enter the atmosphere, and you're gonna do this for a $180 airfare ticket.

§4.p11

They were going to have copper as the skin, and they were going to have liquid hydrogen as the fuel, because it's light. Weight is important if you're going fast — you've learned that here. They were going to have cooling from the hydrogen. On this side of the thing, it would be four or five thousand degrees centigrade out here in the air — that's the frictional heating across the surface. And it would be cooled with liquid hydrogen. This would be a copper alloy. Copper alloys are not exactly known as high-temperature alloys, but it's going to be cooled. They calculated this should be no more than two millimeters thick, 80 thousandths of an inch. If you ever got a hole, a leak, it's the Space Shuttle Challenger all over again — kaboom. You've got hydrogen and air at 4,000 degrees; it burns.

§4.p12

The risk-benefit here — you know what risk is defined as? Risk is defined nowadays by psychologists as the hazard times the consequence. If I'm manufacturing bricks for a building, I can probably tolerate one bad brick in a thousand, because if a brick is bad, the guy will probably throw it out, or even if he does mold it into the building, the building's not gonna collapse because of one bad brick with a crack in it. But if I'm building Hubble telescopes, my quality should be more like one chance in a billion of failure, because the consequence is you can't fix it. Actually, they did do some fixes, but they're very difficult and very expensive.

§4.p13

The hazard for the National Aerospace Plane was pretty severe. You'll blow up the whole aircraft if you have one small leak. And what are the thermal stresses across here? From minus 270 centigrade and 4,000 centigrade. I thought these were the Air Force's best designers, but they're actually people designing things to get money funding from Congress, so it has nothing to do with anything practical. You look at these things and say, is this the best investment of our money? Couldn't we buy potato chips instead?


§5. The real challenge: cost, not properties [21:56]

§5.p1

So aside from that — we did this review a couple of days ago. Here we had good old Ashby basically pointing out a lot of our traditional metals where development is slow. He said mostly quality control and processing. We've hit the limits on the turbine blades, we've hit the limits on steel. I often point out that steel is the most important metal in the world. It's not the most important material — that's stone. But steel — we spend more money on steel research than we do on silicon, a lot more. Why? Because the industry is ten times larger. Even if we only had 20% of the research on steel, it would be twice the research on silicon, even though the percentage research on silicon might be higher.

§5.p2

I actually wrote an article back in the 90s about the real challenge in materials engineering, and I said it was basically to make things less expensive, not to go for tremendous property improvements, because frankly over the 20th century we did a lot to improve properties, and we are getting to the limits of some of them, like the melting temperature. Ashby and other people pooh-poohed that. I got the Dean of Northwestern University, who's an MIT grad — he's passed away now — he wrote a vitriolic letter about my having said that the real challenge was in quality control and processing of these metals. Now, there are all kinds of interesting developments that can go on in polymers and composites and ceramics and glasses, and that's what you read about in the Wall Street Journal.


§6. Dwellings, concrete homes, and Maslow [24:20]

§6.p1

Let's talk about something — there's Maslow's hierarchy of needs. People haven't talked about that in here yet. You know what Maslow's hierarchy of needs is? It's a little pyramid. At the bottom is food, shelter, and clothing, and then you have other needs, and up at the top is self-actualization. People first are interested in basic needs of life — food, shelter, clothing, housing. Then they can worry about getting an education or other things. Maslow came up with this back in the 50s, and now a lot of people criticize it, but nonetheless it's fairly well known.

§6.p2

There have been some other things added below Maslow's bottom level. The first one is Wi-Fi — more important than food and shelter — and the other is battery life, more important than Wi-Fi. So we've added to Maslow.

§6.p3

Dwellings — first of all, we made dwellings with the materials we had, which were things like grass, and then people went to stone, and now we do wood and metal and concrete. People thought about concrete homes many years ago. Here's a patent assigned to T.A. Edison for a way to cast a concrete home. Here's your concrete mixer and a little conveyor that dumps the concrete in the top of this mold, and you mold a house. Here's a picture of Edison with his concept. Here's a picture of the house. I think this was in Ohio, but he was going to mass-produce these. Talk about a housing development where everyone's home looks the same.

§6.p4

I was thinking this morning — what are the advantages and disadvantages of a concrete home? Can you think of an advantage or a disadvantage?

Student: [Earthquakes.]

You can't have an earthquake. Actually, if it's a small home, a small block, it's not going to — but you're right, a big building, earthquake, you've got problems. Part of the problem is, I don't think Edison was doing reinforced steel. Nowadays we wouldn't build a structure like this — codes wouldn't allow it unless we reinforced it with steel, because of its brittle nature, its lousy tensile properties.

§6.p5

Advantages?

Student: Fire.

Fire — thermal mass could be a very good one, although it turns out insulation is better. Stone and concrete actually have low thermal conductivity, much lower than metals, but nowhere near what air gives you when you have fiberglass insulation, which just traps a bunch of air. Or polyurethane is so good that it burns down homes. Have you ever heard about polyurethane burning down homes? It's an exothermic reaction — to form the polymer. We got it up to a couple hundred degrees. We just made blocks and cast the polyurethane and measured it. The codes say you're not allowed to put it down in layers more than two inches thick at a time. People go in spraying it in homes, and they would spray it four inches, six inches thick, and it's such a good insulator that the exothermic nature would basically ignite it. So you burn down your home by putting in insulation. Now they have codes and standards to correct for it.

§6.p6

Other thoughts? It's going to be very fire-resistant.

Student: You can't remodel it.

You can't remodel it. Not only does yours look like everyone else's in the neighborhood, but you can't do much. Can you imagine what people are going to be trying to do? You can't put nails in the wall as easily as wood. No creativity in refurbishing your home. So there are lots of things to start thinking about.

§6.p7

This was in the 1930s, but they probably didn't cast it until the 1910s or something. I doubt it's still around, because concrete — actually, Roman concrete lasts for thousands of years, and that's an interesting study. I've had students do papers on it. There's a lot of science behind Roman concrete and why it lasts for 2,000 years, when modern concrete typically doesn't last for more than 100 or 200 years. It has to do with a volcanic ash they used to make the concrete, and you get a certain chemistry. I don't know that much about it. So concrete will last.

Student: [Location of Edison house?]

North Mountain — in some state. You could go ask the neighbors what it's like to live in a concrete house.

§6.p8

People have been living in concrete block houses — that was a project a student did once. She was from Mexico, and she did adobe. Adobe — I didn't know this, but they tend to use manure for the adobe in many cases because it keeps the flies away. So it's just straw and manure, and you build your house out of that. One of the problems is it does smell for a few years.


§7. Melting metals and the early history of iron [30:52]

§7.p1

I want to talk about metals now rather than concrete. I brought up the Edison concrete home as an example of people looking for different things to use. They look at concrete because it's cheap. You can't exactly move that home — it would be cheaper to blow it up than to move it. It's kind of heavy. But metals were not used in earlier times very much, because they were very expensive. They used to burn down all the wooden buildings that had nails in them to recover the nails from ashes. That's how valuable the nails were, because it took about a person-year of effort to make a ton of cast iron or steel 400 years ago, and now it takes 20 minutes or less of labor content.

§7.p2

The first metals were not steel and cast iron. The adiabatic flame temperature in air of burning carbon or hydrogen in oxygen is about a thousand degrees centigrade. Copper melts right near there, so you can build a little furnace that will hold the heat or preheat the air coming in, and you can melt copper, or you can make bronze with some tin that lowers the melting point of the copper. For years we could not melt steel — I'll talk about that. Cast iron melts at 1150, so you can actually build a blast furnace. This is a picture of an old blast furnace. You have some guy at the top, precariously, dumping coal and limestone and iron ore into the blast furnace. There are bellows here.

§7.p3

If you go to Saugus Ironworks — the Saugus National Historical Site, the first cast iron foundry in the United States — there's a waterwheel, that's the sluice for the waterwheel, and this is the platform they drop the stuff in the top. This is the stone blast chamber. The sluice would run a bellows that blows the blast of air in to get up to the 1,150 degrees you need to melt cast iron. Cast iron was a strategic material militarily, because that's what your cannonballs and your guns and your cannons were made out of. The first cannons were made out of brass, but they weren't as strong, and they were expensive. Eventually people developed cast iron.

§7.p4

Steel has unique things — you can form it in soft condition and then harden it. I did a paper in my junior humanities class on the hardenability of steel, learning how to harden steel, and it goes back thousands of years. People made swords, quenched and tempered steel. The Achilles' heels of steel are corrosion, hydrogen cracking, and low-temperature toughness, which we've talked a little bit about — that was part of the Liberty ships, and that's why you need fracture energy.

§7.p5

If you go to the history of steelmaking, we had two types of steel — well, cast iron and wrought iron. You know the difference between the two of them? What's the difference in composition? Cast iron is like three or four percent carbon and is very brittle, unless we now learn — from World War Two — how to make it ductile. We call it ductile iron, but the traditional cast iron for thousands of years is very brittle. Wrought iron — you take cast iron and you blow air through it, and they called it puddling of iron. People would have bellows and blow, and they'd burn the carbon off, and they would end up with something that's mostly iron with very low carbon. Now it can be forged and beaten by blacksmiths into useful things.

§7.p6

Student: [question about smelting]

I'm going to talk about that. It is a diffusion process to smelt it. You don't melt it — you take the iron ore, put it in with a bunch of charcoal in an oxygen-depleted environment, and you get a reducing carbon monoxide atmosphere which reduces the ore, and you get sponge iron. You get something that looks like a sponge, it's all porous, and you have to forge that back into a solid chunk, and you lose 50% of it by oxidation in the forge as you're pounding it to get all the pores out. But it is a solid-state process at 700 C. They later learned not to do this at 700 C, but to take the cast iron and blow oxygen through it and make puddled iron. That was at higher temperatures, and that's what they were doing in the 1850s when Bessemer came along. He was trying to improve puddling of iron, and he learned a way to make steel without puddling it.

§7.p7

But this is low carbon. It would be forged, whereas high carbon has to be cast and is brittle and cannot be forged — it will break when you hit it. If you look at the iron-carbon phase diagram, here's pure iron at 1500-something centigrade melting temperature. If you add three or four percent — cast iron — you get a eutectic just under 1148 C. You have different phases in here. They didn't really know — they didn't even know what a phase diagram was until J. Willard Gibbs came along in the 1870s and told them what a phase was. Then in the 1880s they developed metallography, and that was the beginning of metallurgy.

§7.p8

MIT didn't start out with the Department of Metallurgy. It started out as the Department of Geology, and it became the Department of Metallurgy in 1888, because this guy in England learned how to polish things and look at them in a microscope and see the structure of the material. So this guy Roberts-Austen — and Franz Martens — and austenite — they all have geological names, like -ite on the end. Ferrite just means "iron mineral," right? That's because metallurgy came out of geology.


§8. Bessemer, Carnegie, and integrated steel mills [38:00]

§8.p1

You had this guy who was one of the greatest British scientists of all time. In 1856 he developed — this is his original Bessemer converter. He would take the cast iron and blow oxygen in here, and he essentially came up with a way to make puddled iron without puddling it — you blow air into cast iron and turn it into steel. That's the beginning of the Bessemer Age, of Andrew Carnegie becoming the richest man in the world. It's an interesting story that could be a good paper — how Andrew Carnegie did some sort of shady things in Pennsylvania and ended up with part of a steel company. He became the richest man in the world, so far as anyone we know, in modern times, in constant dollars. He actually was a very kind Scotsman in many ways.

§8.p2

When you're that rich, you can be kind. "Take away my people, leave my factories, and soon grass will grow on our factory floors. Take away my factories, leave my people, and soon they will have new and better factories." That's not the philosophy of most businessmen today. Andrew Carnegie was — they sort of were competing to be well thought of after they raped and pillaged the economy. Rockefeller [misspoken as "Roosevelt"], who started Standard Oil, used to go around giving away dimes. A dime was sort of like a $10 bill nowadays. He'd just go down the street and hand out dimes to people. Actually, I have a good story — Andrew Carnegie and my grandfather both — I'll do that some other time. My grandfather helped start the University of Chattanooga, and he got the last money from Andrew Carnegie.

§8.p3

Andrew Carnegie's real thing, just like Bill Gates wants to improve health in the third world, was education. So there are Carnegie libraries, Carnegie Mellon University. He gave a tremendous amount of his fortune away to universities. Integrated steel mills — most world steel is still made in integrated steel mills. The last commercially built integrated steel mill in the world was Burns Harbor, Indiana, built by Bethlehem Steel starting in 1964, finished about 10 years later, right on Lake Michigan. This is only part of it. It stretches for miles. It's as big as the MIT campus, and a lot more industrial.

§8.p4

I've estimated it would cost twenty billion dollars today to build an integrated steel mill. It is the most productive steel mill in the United States, by far, because it's the only modern one. When I worked for Bethlehem Steel, they were very proud of the fact their blast furnaces were built in 1912 and their coke ovens were built in 1911. The guy said, "and it doesn't cost us anything to make steel because all our equipment's fully depreciated." I asked, well, why's that? He said, "well, that's because you don't understand finance." I later learned that this vice-president of finance didn't understand finance, and that's why Bethlehem Steel is bankrupt now. The Japanese had to build new steel mills after World War Two, and they built them, and they were beating our pants off in the 1970s with better quality steel at lower cost.

§8.p5

Typically these mills will be five million tons per year, several blast furnaces, and these blast furnaces can run for two or three years before they shut them down to realign the ceramic. That's based on $180-a-ton hot metal, which was the price in the 1970s, and it's still the price today — not constant dollars; actually the price has gone down by about a factor of three because inflation has been about a factor of three since then. The blast furnace has stoves. There was just something on the MIT website about someone with a new idea to use an old technology for solar energy — they're going to have hot bricks. People have been using hot bricks to hold the heat — just a massive bunch of bricks. Every blast furnace had two sets of brick stoves, and you would have your exhaust gases go through one heating up the bricks, and your incoming air going through the other one, which had hot bricks. Every 12 hours, you would shift the flow to reheat the cold bricks and use the heat from them. It's a heat recovery. They do the exact same thing in glass furnaces, because the energy cost is such a big fraction, you have to do something to conserve your energy.


§9. Open hearth, BOF, and the productivity revolution [43:18]

§9.p1

If you look at this — this is an open hearth process. This is what Andrew Carnegie had. You have these huge — they don't show them as big as they were — they're about the size of a football field and about two stories tall. Just brickwork, latticework of bricks. The actual furnace is not much bigger — your room may be twice the size of this room, but the furnace is not the size of these two things. The brickwork is a couple of football fields.

§9.p2

In the 1950s, there was a guy at Linz in Austria who decided, all you're doing in the open hearth and the other furnace is basically the Bessemer process of passing hot air over cast iron to burn the carbon off, to turn it into a lower-carbon version, which is basically steel. He decided, why don't we just use pure oxygen? The Germans had developed the Linde process for liquefying air, and you could get pure oxygen. So they built a basic oxygen furnace. In the old open hearth process, it would be 400 tons of steel in the furnace, and it used to take anywhere from 24 to 48 hours to process those 400 tons of steel. The BOF furnace, which they started building around 1960, by the 1970s had taken over the entire industry. The last basic open hearth was built by U.S. Steel in the early 70s, and they were stupid to build that one. The BOF process is 300 tons of steel in 20 minutes.

§9.p3

Student: [question about forging and cooling]

If you have low-carbon forged steel, you can air-cool it. With higher carbon stuff, it cools down too fast, you'll get martensite and might crack. In a very high-quality forge shop, you would have furnaces and you do controlled cooling on big parts. But small parts — if it's the right size part, air cooling is great. It's called normalization. It's a heat treatment step. For the majority of forged parts, they do something to control it. If you're low-carbon steel, if it's just some hinge and it's not critical, it's not a big deal. But most forging is medium-carbon and high-carbon going into turbines and things, and if you did that, General Electric would fire you right away.

§9.p4

Student: [question about fuel in the BOF]

The fuel in this one is oxygen burning the carbon out of the cast iron. I've got four percent carbon in that cast iron, so I've got 12 tons — four percent times 300 tons of cast iron — and I blow pure oxygen in there. There's plenty of heat; I burn up 12 tons of carbon. This whole thing stands about four or five stories tall, and the bottom is maybe only four or five feet tall. When they blow the oxygen in there at supersonic velocities, the steel becomes a froth and fills up the whole three- or four-story-tall container. It's a foam — it is fireworks galore. It's real exciting. You're not allowed to stand anywhere except in the doghouse — the control room, which is insulated — or outside the building. There are literally sparks flying. It's a very violent reaction. But it's very efficient. It takes a 20-minute blow to get all the carbon out, whereas in the old days when you just had a boundary layer and you just blew air across the top and you didn't get a froth, it could take a day or two.

§9.p5

That's one of the reasons — we'll go through the productivity next time — but productivity has gone from about 3,000 hours per ton of cast iron back in the days of Saugus Ironworks 400 years ago, to about 20 minutes today per person-hour to produce it. That has made a world of difference, and that's one of the reasons steel is so widely used — it's cheap, because they have spent so much money. I'll show you the curve I came up with, and it's really dramatic. It's a log plot of costs going down, and it's getting steeper in recent years. Okay, we ran over. Sorry.

Cases referenced

  • Fire sprinkler system plastic pipe failures §2.p1

    Plastic sprinkler pipes cracking due to biocide polymer leaching from steel pipes into stagnant water. Used to extend the previous lecture's theme on material substitution failures in fluid-conveyance systems.

  • Single-crystal turbine blades (Pratt & Whitney) §3.p2

    Development arc from uncooled solid blades → nickel superalloys → directionally solidified → single crystal by early 1990s, led at Pratt & Whitney by an MIT-affiliated metallurgist (name garbled in captioning).

  • Turbine blade cooling and thermal barrier coatings §3.p3

    Internal air cooling above the melting point of the alloy; zirconium oxide TBCs as the only ceramic matching the CTE of nickel; $5,000-per-blade cost driven by EB vapor deposition.

  • Air Force engine research committee and Mach 17 / National Aerospace Plane §4.p1

    Tom's service on an Air Force engine-research advisory committee (~2005) where the Mach 17 air-breathing strike requirement was justified by the bin Laden ordnance-reach problem. Used to teach the divergence of military and civilian engine technology, and the implausibility of the National Aerospace Plane copper-skin / liquid-hydrogen design.

  • Hubble Space Telescope quality requirement §4.p12

    Risk = hazard × consequence framing. One-in-a-billion quality required because the system cannot be repaired (with a parenthetical acknowledgment that fixes were ultimately done).

  • Niobium silicide intermetallics (Air Force research) §3.p8

    Higher melting point (~2000°C) but unsolved brittleness problem; Tom's dismissive close ("I'm not invested in that one right now").

  • Edison concrete homes §6.p3

    Edison's patent for casting concrete houses. Used to teach the advantages (fire resistance, thermal mass) and disadvantages (earthquake brittleness, no remodel, uniformity) of structural concrete in residential construction.

  • Polyurethane insulation home fires §6.p5

    Exothermic polymerization of polyurethane sprayed too thickly (more than 2 inches) self-ignites. Demonstrated experimentally by Tom's group with cast blocks. Codes now restrict layer thickness.

  • Roman concrete longevity §6.p7

    2,000-year durability of Roman concrete vs. 100–200-year modern concrete, attributed to volcanic ash chemistry. Source for student papers.

  • Adobe construction (Mexican student project) §6.p8

    Past student project on traditional adobe in Mexico — straw plus manure (the manure repels flies, but the house smells for a few years).

  • Saugus Ironworks §7.p3

    First cast-iron foundry in the United States, with waterwheel-driven bellows reaching the 1150°C needed to melt cast iron. Anchors the productivity-revolution arc that closes the lecture (3,000 person-hours per ton at Saugus vs. 20 minutes today).

  • Bessemer process invention §8.p1

    Henry Bessemer in 1856 develops air-blown cast-iron conversion to steel, enabling the Carnegie steel era.

  • Andrew Carnegie and integrated steel §8.p1

    Carnegie's rise via Bessemer steel, his quote ("Take away my people, leave my factories..."), his philanthropy (Carnegie libraries, Carnegie Mellon), juxtaposed with Rockefeller [misspoken as "Roosevelt"] handing out dimes. Tom's grandfather's role at the University of Chattanooga briefly invoked.

  • Bethlehem Steel Burns Harbor §8.p3

    Last commercially built integrated steel mill in the world (1964, completed ~10 years later), still the most productive mill in the U.S. because it's the only modern one. Tom's anecdote about the Bethlehem finance VP who didn't understand depreciation — used to teach why Bethlehem went bankrupt and why Japanese postwar mills outcompeted U.S. integrated mills in the 1970s.

  • Linz BOF development (Austria) §9.p2

    1950s Linz, Austria — replaced air-blown Bessemer with pure-oxygen blow using Linde liquefaction. 300 tons of steel in 20 minutes vs. 24–48 hours in open hearth. By the 1970s had taken over the global industry. U.S. Steel built the last open hearth in the early 70s — "they were stupid to build that one."

  • U.S. Steel's last open hearth §9.p2

    Built in the early 1970s when the BOF had already won; Tom's verdict: "they were stupid to build that one."

Layer 2 — cleanup edit
p1 00:00

Um, we are encouraging you — I, Dr. Belmar and I both have to be away most of the week of October 2nd, so we're giving you that. I think Brian or I sent something around that we're giving you the week of October 2nd off. Well, giving it to you off from coming to class at 9 o'clock. If I were you, I would watch some of these other videos, okay, whether it's at 9 o'clock in the morning or some other time. Because I think Brian sent something around, and somewhere around the end of the second week in October is when we want the one-page summary on one of your modules, okay. Because we're trying to get you to finish everything up by Thanksgiving, okay, paper and all that other stuff. Front-load this class. Learn to get things done early rather than late. There are certain advantages to that, however most people don't necessarily believe that.

p2 01:02

You know, I've always — you can go ahead and start if you want — but I've always gotten in early, and one of my students, graduate students, had a hard time getting in to see me when I was department head. Students have to come see me at 5:30 in the morning, okay, because I just, from nine o'clock on, I just did meetings all day. Terrible life. But he told me a couple years ago that the way he solved the problem, he would just stay up all night, come to see me at 5:30, and then go to bed, okay. Works for him.

p3 01:43

Okay, we talked about conveying gas and different materials that have been used through the ages, and how we now tend to use a lot of plastic or corrugated stainless steel, and these materials tend to develop different types of problems, okay. Let me go to another one, another material that has developed problems that were sort of surprising. We used to use carbon steel pipe for sprinkler systems. So this is a fire sprinkler system, and this is one that's actually now spreading a bunch of water. These things actually operate around 200 psi, and so this gets a lot of water. But no one has ever died in a home that's had a sprinkler system in over a hundred years, okay. So it works. You might end up with a home that, just like Houston, that's flooded out, but it works. It keeps people from dying in fires.

p4 02:42

Nonetheless, they decided they wanted to go and save some money. Instead of using copper or steel, they decided to go with plastic. And it turns out it's not the problem with the plastic, but it's the problem that the water comes in on some of the old steel pipes — and the steel pipes, they put a biocide on the inside, they sprayed the inside of the steel pipe with something to keep the bugs and the corrosion down. It turns out that polymer they put on the inside of the steel pipe gets in the water, and it's stagnant water, it's not flowing through in a potable water system, it just sits there, and it cracks the plastic. So there are lots of buildings in this country right now that have sprinkler systems that work all the time, okay, they're leaking. And so there's a lot of secondary effects that you don't always think about, okay, that you run into.

p5 03:40

I didn't — I want to kind of shift gears and take a couple of industries and talk about what's happened in some of these industries. This is the gas turbine industry. So if you're interested in aerospace and engines — it turns out Wittle [Whittle] was the British guy who developed a gas turbine, and his nice quote from him how he invented the gas turbine like 1936. And then it was during the World War Two that it actually kind of came in. Two things in though — the Germans actually had V2 rockets and things that had turbines and stuff, some of the first gas turbines. But those are the original gas turbines. If you go back in time, they didn't operate at very high temperatures. In fact, most of them worked at stainless steel temperatures, so you were talking 500 or 700 degrees centigrade, which is not terribly hot.

p6 04:33

And in any case, people started working on better and better materials, and these were uncooled solid materials, and they started going to nickel-based superalloys, okay, in the 50s and 60s and 70s. And then they basically came up with — well, as they went to the nickel-based superalloys, they learned you didn't like grain sizes — finer grain size at high temperature meant creep, the metal would deform. And so they went to directionally solidified turbine blades, they went to single crystal blades by the early 1990s. And in fact, the guy who really led that at Pratt Whitney was Gotbutshank [Gell-Mann? — likely "Goldschmidt" or similar; captioner garble, see log] who had been a faculty member here at MIT in both mechanical and materials, and he went to Pratt and Whitney and kind of led all that effort in the 60s, okay.

p7 05:25

But then they went to cooling. I've sent around — sitting around again if you want this thing several times, so you can look at — I won't pass around, you probably have a chance to unless you wanted to see it, but it's got the internal cooling passages. And if you take some of the other courses, I'll talk more about how they make that and whatnot. But they went to internal cooling, and the interesting thing here is you see this jump in temperature — 230 — about almost 1300, it eventually got to 1,300 degrees centigrade. That's above the melting point of the material, okay. If you didn't have that cooling, your engine would melt. Now the cooling is actually thousand-degree Fahrenheit air that comes from the compressor. So as long as the turbine is running, you always have cooling air, okay, unless something gets plugged up. So it's okay.

p8 06:18

And then they came to thermal barrier coatings. I don't think I passed this one around. This is a stator, but it's got a ceramic coating on the outside. It actually is a lot of technology in these thermal barrier coatings, TBCs, okay. And that allowed them to get to even higher temperatures. And on a new growth rate curve, you can see when you just had solid metal, there's only so much you can do with the metal. When the metal melts at 1250, these things are getting pretty close to the melting temperature of the metal, and you can't exceed that unless you do some active cooling. And so they have active cooling with thermal barrier coatings.

p9 06:58

And it turns out all the thermal barrier coatings are zirconium oxide. Anybody know why? Because it's the only ceramic that matches the coefficient of expansion of the nickel. Most ceramics have about one-fifth the coefficient of thermal expansion. You put a ceramic like straight aluminum oxide on the nickel alloy, it's just going to crack off as things heat up and cool down, okay. So they actually have to put a fancy metal inner layer, typically intermetallic nickel aluminum or something that's nice and porous, and then they spray the ceramic on. It actually vaporizes — an electron beam vaporizer, $2,000,000 machine, and each blade may have to be in that machine for an hour as you vapor-deposit the zirconium oxide and stuff. That's why these things can cost $5,000 a blade, okay. They're not cheap.

p10 07:52

But there's tremendous cost savings as you go to higher temperatures, because this is just what you learned in thermodynamics about heat engines, okay. Delta T over T is the heat efficiency of the engine, right, theoretical efficiency. And now they're predicting here that, oh, we're going to go to other types of silicon-based ceramic matrix composites — these are different materials — and we have different types of coating, and then there you have heat pipe cooling in the future, and eutectic oxide ceramic metal composites. Well, don't plan on seeing these things in your lifetime, okay, because this is where someone's predicting they're gonna have all these fancy things. But why are they doing it? Because every 50 degree centigrade increase in operating temperature, that engine is worth two billion dollars a year to the commercial airlines, okay, in fuel savings. It doesn't take much to end up with a lot of money. So we can talk about the value of a pound saved in a structure, but when you start looking at some engine as driving it, you can have multiples, and that's why people spend so much money on the engine technology. Yep.

p11 09:12

Student: [inaudible question]

p12 09:23

Yeah, they're not here. Yes, this figure was probably developed in 2004 or five, and people were looking at niobium silicides and things like that, because they're actually sort of — well, they're intermetallics, okay, and they have higher melting points. They're not 1250 centigrade as a melting point, it might be 2000 degrees centigrade, but they're brittle as can be, okay. And so no one's solved that brittleness problem, and no one — they say, well, we're going to eventually solve the brittleness problem in our ceramics. Well, yeah, good, okay. I'm not invested in that one right now, okay. But people are working on it, okay.

p13 10:01

And I remember serving on a committee back in about this time frame that was looking at how the Air Force should spend its three hundred million dollars a year in research money for engine development, okay. It was a very interesting committee because there were about 40 of us on it. Two of us were materials people. A number of people were people from General Electric who was the chief engineer on such-and-such engine, okay. And one guy sitting next to me one time was the guy who developed the F100 engine for Pratt & Whitney, I think that's right, Whitney — anyway, I don't remember which engines they were. But there was a problem starting around in here in the 1990s, in that military technology for engines diverged from civilian technology, and the civilians went to high bypass engines. You know what high bypass engines are? I'm seeing a lot of nods, okay.

p14 10:57

So it's just simple little freshman physics of, if you have mass and you have momentum and energy — one's m times v and the other one's one-half mv squared, right? I mean, you learn that in freshman physics. And it turns out a high bypass engine, you can blow a lot of air through there, so you have a big M but not as much V, and you get a more efficient engine. But it doesn't have the performance of one of these things that you can just dump fuel in and basically, whether you call them ram— well, not a ramjet, they're a scramjet, but basically — okay well, there's turbojets, but this is basically when you just kind of throw fuel into your afterburner. Or maybe it's an afterburner, yeah I guess it's just the afterburner.

p15 11:45

I remember, I grew up in Virginia Beach, and we had two naval air stations right there in the Norfolk area, and there was one — I used to pass it every day going on the bus to school in high school — but one guy was out flying over the Atlantic and he had a flameout, and he just turned on his afterburner and he made it back to Oceana Naval Air Station within a hundred yards. He kind of crashed within a hundred yards of the — but 'cause he just ran out of fuel. I mean, he had probably half a tank of fuel, but to go 10 miles from out in the ocean to Oceana, you burned a whole half a tank of fuel, but you can go really fast, okay. In his case, he didn't have a choice because the main engine had failed.

p16 12:31

So the military technology was looking for performance and not efficiency, and they started diverging, and that's been a real problem for the whole engine development, because the commercial engines before that all built off the military technology, but now they can't, because they're heading in different directions in their designs. So anyway, I was sitting next to this guy, and they asked, at the beginning of this committee that lasted for a number of meetings, you know, to go around and what are the key points that we ought to develop, okay. And these guys, the blade designers, okay — they design these cooling passages and things — they would say, well, we need better materials, they can go to higher temperatures. And those materials people, we say, well, we've kind of gone to the limit of our materials.

p17 13:32

I mean, I can go back to an Ashby plot here somewhere. I think it's this plot that says maximum service temperature, and there's nothing that goes to ten— 3,000 centigrade, okay. It just doesn't exist, okay. We don't have things that have strong enough bonds. Diamond might do it if it didn't sublime before then or burn in the air. So we would say we need better designs. Well, this is like a 1975 design, folks. This is a 1980 design. This is not the latest stuff. They're getting these things down less than a millimeter thick, okay, like a half a millimeter thick. And they have designs — 210 — so a millimeter thick for the walls, excuse me. I mean, do you know how hard it is to cast that with any accuracy?

p18 14:30

They do — the reason we have CAT scanners for metals today, primarily, is because every turbine blade gets a CAT scan, a full CAT scan, to measure the wall thickness, because you got a core that you cast this thing around, and grow a single crystal around this thing. And if that core is off by two or three thousandths of an inch from its perfect position — and its perfect position, it's not just a center axis, it's the center axis in a twist, right — you're off by two or three thousandths over a couple of inches on something that's going to heat up to 1,300 degrees centigrade and melt it and everything else, and it's a piece of junk. Well, all of a sudden, you find your manufacturability goes like that.

p19 15:16

And we are — they continue to make progress because they spend billions of dollars. The problem the Air Force had — am I going the wrong direction? The problem the Air Force had is they only had 300 million a year, and they wanted to know how to spend it. The other problem that they had — I've been going the wrong way, whether wrong way — the other problem they had was that they didn't just want regular old engines. They would be delighted — well, I don't really need this one now anymore — they would have been delighted with people developing engines that could go Mach 17 in the air.

p20 16:02

I remember going to this meeting and saying, wait a second, who needs Mach 17, okay? I mean, even the Air Force — I said, why do you need Mach 17 in the air? I said, this is unmanned, isn't it? They said yes, okay. So at least I got that right, okay. It wasn't in the air. What happened to my — what happened to my cursor? Anyone see my cursor? Oh, there it is, okay. Thanks. So let's cancel that one.

p21 16:41

Yeah, well, they gave me an answer which I thought was interesting. Now you gotta remember this is like 2005, so 9/11 had already occurred, and Osama bin Laden had not yet been captured. But they said that they knew where Osama bin Laden was one time, a year before or two years before — this is all classified, and they didn't give us the details — but they knew where Osama bin Laden was, but the closest ordinance that they could get to him, because he was off in some mountain in Pakistan or Afghanistan or somewhere, was on an aircraft carrier about eight hundred miles away. And they only had 15, 10 or 15 minutes to get the ordinance from 800 miles away to where he was, to wipe him out, and they didn't — they needed Mach 17 to do. Okay, well, I guess if you just could shoot a laser anywhere in the world. But Mach 17 is what they go in space with, no atmosphere, right?

p22 17:42

And at Mach 17, your skin temperature is going to be — if you — they had plots for us just from heat flow and friction heating in air, your skin temperature is gonna be 10,000 degrees. And I said, there is no material that can do that. They said, oh, we'll just use active cooling. And let me tell you what these guys think of for designs. I mean, your engineers — they were going to do this for the National Aerospace Plane. I'll just tell you about the National Aerospace Plane in the 1980s. You know what the National Aerospace Plane was? Well, is it now — this is like Elon Musk. We're gonna go from Los Angeles to New Delhi in two hours, right, because you're gonna go up into space, and you're gonna go Mach 17 up there, and you're gonna re-enter the atmosphere, and you're gonna do this with all — you know, for a $180 airfare ticket, right?

p23 18:44

But they were gonna have copper as the skin, and they were gonna have liquid hydrogen as the fuel, because it's light. You know, weight is important if you're going fast, right? You've learned that here. And they were going to basically bleed the hydrogen — oh no, they're going to have the cooling of the hydrogen. And you're going to have on this side of the thing — they would have, back in the National Aerospace Plane, it would be four thousand degrees centigrade out here, four or five thousand degrees centigrade out here in the air, okay, that's the frictional heating across the surface. And it would be cooled with liquid hydrogen. This would be a copper alloy. Copper alloys are not exactly known as high-temperature alloys, okay, but it's going to be cooled. And this would be — they calculated this should be about no more than two millimeters thick, 80 thousandths of an inch. Now, if you ever got a hole, a leak, it's, you know, the space shuttle Challenger all over again — kaboom. Cause you got hydrogen and air at 4,000 degrees, it burns, okay.

p24 19:59

The risk-benefit here, okay — the, or — you know what risk is? People know what risk is defined as? Risk is defined nowadays as the hazard times — now this is done by psychologists — the opportunity, or a consequence, I'm sorry, consequence. And so I have — actually I was just talking to somebody yesterday — I used to talk about, if I'm manufacturing bricks for a building, I can probably tolerate one bad brick in a thousand, because, you know, if a brick is bad, the guy will probably throw it out, or even if he does mold it into the building, the building's not gonna collapse because of one bad brick with a crack in it, right. But if I'm building Hubble telescopes, my quality should be more like 1/2 billion, one chance in a billion of failure, because the consequence is you can't fix it, okay. If it's the Hubble telescope — or actually, they did do some fixes, but they're very difficult and very expensive.

p25 21:07

The hazard here for the National Aerospace Plane was pretty severe. You'll blow up the whole aircraft if you have one small leak across something. And what are the thermal stresses across here, okay? I mean, there are some thermal stresses, right? 4— 4— you know, from minus 270 centigrade and 4,000 centigrade, okay. And I thought these are the Air Force's best designers, but they're actually people designing things to get money funding from Congress, so it has nothing to do with anything practical. You look at these things and you say, is this the best investment of our money? Couldn't we buy potato chips instead? You know, I mean — anyway, okay.

p26 21:56

So aside from that — now I gotta find my cursor again, or get this thing working. What happened to my cursor? You can't see my cursor? I gotta — there's my cursor, okay. So we did this review yesterday, or a couple of days ago. We did this review. We did this, and here we had good old Ashby basically pointing out a lot of our traditional metals where development is slow. He said mostly quality control and processing. We've hit the limits on the turbine blades, we've hit the limits on steel. I often point out, since I point out, steel is the most important metal in the world. It is most important metal. It's not the most important material — it's stone. But steel — we spend more money on steel and research on steel than we do on silicon, a lot more. Why? Because the industry is ten times larger. Even if we only had 20% of the research on steel, it would be twice the research on silicon, even though the percentage research on silicon might be higher.

p27 23:13

And so I actually wrote an article back in the 90s about the real challenge in materials engineering, and I would say it was to basically make things less expensive, not to go for tremendous property improvements, because frankly over the 20th century we did a lot to improve properties, and we actually are getting to the limits of some of the properties, like the melting temperature, okay. But anyway, so Ashby and other people who pooh-poohed that — I got the Dean of Northwestern University who's an MIT grad, he's passed away now, but he wrote a vitriolic letter about my having said, you know, that the real challenge was in quality control and processing of materials, as these metals. Now there's all kinds of interesting developments that can go on in polymers and composites and ceramics and glasses, and that's what you read about in the Wall Street Journal and stuff, okay.

p28 24:20

Let's talk about something — there's Maslow's hierarchy of needs. People actually haven't talked about that in here yet. You know what Maslow's hierarchy of needs are, where are they? It's a little pyramid scheme, right? Not exactly a pyramid scheme, but — and at the bottom is food, shelter, and clothing, okay, and then you have other needs, and up here it's self-actualization, okay. But people first are interested in basic needs of life — food, shelter, clothing, housing. Then they can worry about getting an education or other things. And so you can look up Maslow's hierarchy of needs. He came up with this back in the 50s, and now a lot of people criticize it, but nonetheless it's fairly well known.

p29 25:11

I actually have a good slide to find it if you wanted, but there's been some other things added down here to Maslow. There's two other things below this. The first one is Wi-Fi — more important than food and shelter — and the other is battery life, more important than Wi-Fi. Anyway, so we've added to Maslow.

p30 25:41

In any case, dwelling — so I mean, first of all, we made dwellings with the materials we had, which were things like grass, okay, and then people went to stone, and now we do wood and metal and concrete. People thought about concrete homes many years ago. Here's a patent assigned to Thomas Edison, E.A. Edison [T.A. Edison], for a way to cast a concrete home. Here's your concrete mixer and a little conveyor that dumps the concrete in the top of this mold, and you mold a house. And here's a picture of Edison with his concept, okay. And here's a picture of the house, okay. I think this was in Ohio, but he was gonna mass-produce these. So talk about a housing development where everyone's home looks the same.

p31 26:35

I was thinking about this morning, and I was thinking, what are the advantages and disadvantages of a concrete home? Can you think of an advantage or a disadvantage? Student: [inaudible — mentions earthquake] You can't have an earthquake. Actually, if it's a small home, you can — I mean, a small block, it's not gonna — but you're right, a big building, earthquake, no, you got problems. And part of the problem is, I don't think he was doing reinforced steel, you know, reinforced structures and stuff. But yeah, earthquakes would be a problem for big ones. And nowadays we wouldn't build a structure like this — codes wouldn't allow it unless we reinforced it with steel, okay, because of its lousy — or its brittle nature, its lousy tensile properties.

p32 27:16

Advantages? Student: Fire. Fire, thermal mass — could be a very good one, although it turns out insulation is better. I mean, stone and concrete actually have low thermal conductivity, lower than — much lower than metals, but nowhere near what air gives you when you have fiberglass, you know, insulation, okay, which just traps a bunch of air. Or polyurethane is so good that it burns down homes. Have you ever heard about polyurethane burning down homes? What about polyurethane burning down homes? Yeah, it's an exothermic reaction. To form the polymer, right? And it goes up to — well, we got it up to a couple hundred degrees, okay. We just made blocks and cast the polyethylene and just measured, you know, a polyurethane, and measured it. But basically, the codes say you're not allowed to put it down in layers more than two inches thick at a time. And so if people go in there spraying it in homes, and they would spray it four inches, six inches thick, and it's such a good insulator that the exothermic nature, it would basically ignite, okay. And so you burn that — you burn down your home by putting insulation. But now they have codes and standards to correct for it.

p33 28:46

Other thoughts? Fire — is it — it's gonna be very fire-resistant. Student: You can't remodel it. You can't remodel it. Not only does yours look like everyone else's in the neighborhood, okay, but you can't do much. And can you imagine what people are gonna be trying to do? You can't put nails in the wall very as easily as wood. I mean, it's — I G — I mean, no creativity in refurbishing your home. So there's lots of things to start thinking about when you start thinking of these things.

p34 29:16

I don't know — it was in the 1930s, okay, but they probably didn't cast it until the 1910s or something. But I doubt it's still around, because concrete — actually the Roman concrete lasts for thousands of years, and that's an interesting study. And I've had students do papers on that. There's a lot of science behind Roman concrete and why it lasts for 2,000 years, and modern concrete typically doesn't last for more than a hundred or 200 years. And it has to do with a volcanic ash that they used to make the concrete, and you get a certain chemistry. I don't know that much about it. But so concrete will last. Are there — yeah, okay, where are they? Do they say? Student: North Mountain. North Mountain, what? Okay, well, in some state, okay. So you go and you could go ask the neighbors what it's like to live in a concrete house.

p35 30:18

Actually, people have been living in concrete block houses — actually, that was a project that a student did once. She was from Mexico, and she did adobe, okay. And adobe — I didn't know, but they tend to use manure for the adobe in many cases because it keeps the flies away. So it's just straw and manure, okay, and you build your house out of that. One of the problems is it does smell for a few years.

p36 30:52

Anyway, melting of metals, okay. I want to talk about metals now rather than concrete. But people will look — I guess I just brought up the Edison concrete home as an example of people looking for different things to use, things in concrete. They look at it because it's cheap, okay. You can't exactly move that home — be cheaper to blow it up than to move it. It's kind of heavy. But metals, on the other hand, were not used in earlier times, in fact, very much, because they were very expensive. And it turns out they used to burn down all the wooden buildings that had nails in them to recover the nails from ashes, okay. That's how valuable the nails are, and that's because it took about a person-year of effort to make a ton of cast iron or steel 400 years ago, and now it takes 20 minutes or less, okay, of labor content. And then I talked about that.

p37 31:58

But some of the problems were, the first metals were not steel and cast iron. The adiabatic flame temperature in air of burning carbon or hydrogen in oxygen in the air is about a thousand degrees centigrade. Copper melts right near here, so you can build a kind of little furnace that will hold the heat or preheat the air coming in, and you can melt copper, or you certainly can make brown — bronze — with some tin that lowers the melting point of the copper. For years we could not melt steel, and I'll talk about that. Cast iron melts at 1150, so you can actually build a blast furnace. And this is a picture of an old blast furnace. You have some guy at the top, precariously here, he's dumping coal and limestone and iron ore into this blast furnace. There's bellows here.

p38 32:48

If you go to Saugus Ironworks — this is the Saugus National Historical Site, the first cast iron foundry in the United States — there's actually a waterwheel, that's the sluice for the waterwheel, and this is the platform they drop the stuff in the top. This is actually the chamber, the stone blast chamber. And the sluice would run a bellows that blows the blast of air in here to get up to the 1,150 degrees that you need to melt cast iron. And cast iron was a strategic material militarily, because that's what your cannonballs and your guns and your cannons were made out of. The first cannons were made out of brass, but they weren't as strong, and they were expensive. And eventually people developed cast iron.

p39 33:43

So if we go to steel — you've already heard about it, it's got unique things, and you can form it in soft condition and then harden it. That's a — I did a paper in my junior humanities class on the hardenability of steel, learning how to harden steel, and it goes back thousands of years. People made swords, and quenched and tempered steel, okay. And the Achilles heels of steel are corrosion, hydrogen cracking, and low-temperature toughness, which we've talked a little bit about, and that was part of the Liberty ships, and that's why you need fracture energy.

p40 34:20

Now, if you go to the history of steelmaking, we had two types of steel. We had cast iron, wrought iron anyway. I know the difference between the two of them, all right. Yep. But why — what's the difference in the composition? Cast iron is like three or four percent carbon and is very brittle, unless we now learn — for World War Two — how to make it ductile. We call it ductile iron, but the traditional cast iron for thousands of years is very brittle. The wrought iron — you basically take cast iron and you blow air through it, and they called it puddling of iron, and people would just, you know, have bellows and blow, and they burn the carbon off, and then they would end up with something that's mostly iron with very low carbon, and now it can be forged and beat into, by blacksmiths, into useful things. Yes.

p41 35:18

Student: [question about smelting]

Oh, I'm gonna talk about that, okay. Well, it is a diffusion process to smelt it, okay. You don't melt it — you actually take the iron ore, put it in with a bunch of charcoal, put it in an oxygen-depleted environment, and you get a reducing carbon monoxide atmosphere which reduces, and you get sponge iron. You basically get something that looks like a sponge, it's all porous, and you have to forge that back into a solid chunk, and you lose 50% of it by oxidation in the forge, okay, as you're pounding it to get all the pores out of it. But it is a solid-state process at 700 C. They later learned to basically not do this at 700 C, but to take the cast iron and blow oxygen through it and make puddle iron, okay. That was at higher temperatures, and that's what they were doing in the 1850s when Bessemer came along. He was trying to improve puddling of iron, and he learned a way to make steel without puddling it, okay.

p42 36:26

But this is low carbon. It would be forged, and high carbon has to be cast and is brittle and cannot be forged — it will break when you hit it. If you look at the iron-carbon phase diagram, which we won't go through, but here's pure iron at 1500-something centigrade melting temperature. If you add three or four percent — cast iron — you get a eutectic that's just under 12— 1148 C. You have different phases in here that we don't have to go through. But they didn't really know — well, they didn't even know what a phase diagram was until J. Willard Gibbs came along in the 1870s and told them what a phase was. And then in the 1880s, they developed metallography, and that was the beginning of metallurgy.

p43 37:23

MIT didn't start out with the Department of Metallurgy. It started out as the Department of Geology, okay, and it became the Department of Metallurgy in 1888, because this guy in England learned how to polish things and look at them in a microscope and see the structure of the material. And so this guy Roberts-Austen, and I love Martens — Franz Martens, site — and austenite — and they all have geological names, like -ite on the end. Ferrite just means "iron mineral," right? But that's because it came out of geology, so far as that goes.

p44 38:00

You had this guy who was one of the greatest British scientists of all time, people say, because in 1856 he developed — this is his original Bessemer converter. He would take the cast iron and he would blow oxygen in here, and he essentially came up with a way to make puddled iron, where you blow air into cast iron and turn it into steel. And that's the beginning of the Bessemer Age, of Andrew Carnegie becoming the richest man in the world, when he basically — it's an interesting story that could be a good paper, is the paper of how Andrew Carnegie did some sort of shady things in Pennsylvania and ended up with part of a steel company. But he became the richest man in the world, so far as anyone we know, in modern times, in constant dollars. And there, you know, you can see the environment he produced for everything. But he actually was a very kind Scotsman in many ways.

p45 38:58

When you're that rich, you can be kind. "Take away my people, leave my factories, and soon grass will grow on our factory floors. Take away my factories, leave my people, and soon they will have new and better factories." That's not the philosophy of most businessmen today. But Andrew Carnegie was — they sort of were competing to be well thought of after they raped and pillaged the economy. Roosevelt [Rockefeller], who started Standard Oil, used to go around giving away dimes. But a dime was sort of like a $10 bill nowadays, okay. He'd just go down the street and hand out dimes to people. That was Rockefeller, Rocke— sorry. Actually, I have a good story — Andrew Carnegie and my grandfather both — do that some other time. My grandfather helped start the University of Chattanooga, and he got the last money from Andrew Carnegie.

p46 39:56

But anyway, Andrew Carnegie's real thing, just like Bill Gates wants to improve health in the third world, Andrew Carnegie's thing was education. So there's Carnegie libraries, Carnegie Mellon University. He gave a tremendous amount of his fortune away to universities. Integrated steel mills, okay. It turns out most of the steel — world steel is still made in the integrated steel mills. The last commercially built integrated steel mill in the world was Burns Harbor, Indiana, built by Bethlehem Steel starting in 1964, finished about 10 years later, and right here on Lake Michigan in Indiana. And this is only part of it, okay. It stretches for miles. It's as big as the MIT campus, and a lot more industrial.

p47 40:47

I've estimated in the past that it would cost twenty billion dollars today to build an integrated steel mill. It turns out it is the most productive steel mill in the United States, by far, because it's the only modern one. When I worked for Bethlehem Steel, they were very proud of the fact their blast furnaces were built in 1912 and their coke ovens were built in 1911, and the guy said, "and it doesn't cost us anything to make steel because all our equipment's fully depreciated." And I asked him, well, why's that? And he said, "well, that's 'cause you don't understand finance." Well, I later learned that this vice-president of finance didn't understand finance, and that's why Bethlehem Steel is bankrupt now. But nonetheless, the Japanese had to build new steel mills after World War Two, and they built them, and they were beating our pants off in the 1970s with better quality steel at lower cost.

p48 41:46

Typically these will be five million tons per year, several blast furnaces, and these blast furnaces can run for two or three years before they shut them down now, okay, to realign the ceramic and stuff. And that's based on $180-a-ton hot metal, which was the price in the 1970s, and it's still the price today, okay — not constant dollars, actually the price has gone down by about a factor of three because inflation has been about a factor of three since then. But the blast furnace takes all these things, and it basically has stoves. And I just — there was just something on the MIT website about someone had this new idea to use an old technology to worry about solar energy and things like that. They're going to have hot bricks. Well, people have been using hot bricks to hold the heat — just a massive bunch of bricks. Every blast furnace had two sets of brick stoves, and you would have your exhaust gases go through one heating up the bricks, and your incoming air going through the other one, which had hot bricks. And every 12 hours, you would shift the flow to reheat the cold bricks and use the heat from them. So it's a heat recovery, okay. They do the exact same thing in glass furnaces, okay, because the energy cost is such a big fraction, you have to do something to conserve your energy.

p49 43:18

So if you look at this — this is an open hearth process. This is what Andrew Carnegie had, and you have these huge — they don't show them as big as they were — they're about the size of a football field and about two stories tall. Just brickwork, latticework of bricks, okay. The actual furnace is not much bigger — well, a little bigger than this — your room may be twice the size of this room, but it's not the size of these two things. The brickwork is a couple of football fields.

p50 43:45

But then in the 1950s — now finished with this — there was a guy at Linz in Austria who decided, all you're doing in the open hearth and the other furnace is basically the Bessemer process of passing hot air over cast iron to burn the carbon off, to turn it into a lower carbon version, which is basically steel. He decided, well, why don't we just use pure oxygen? The Germans had developed the Lindy [Linde] process for liquefying air, and you could get pure oxygen. So they built a Bessemer converter — not a Bessemer, a basic oxygen furnace. And it used to take, in the old open hearth process, which would be 400 tons of steel in here, it used to take anywhere from 24 to 48 hours to process that 400 tons of steel. And a BOF furnace, which they started building in about 1960s or so, by the 1970s that had taken over the entire industry. The last basic open hearth, the last one was built by U.S. Steel in the early 70s, and they were stupid to build that one, okay. This process is 300 tons of steel in 20 minutes.

p51 45:00

Student: [question about forging / cooling]

If you have low-carbon forged steel, you can do that. You have higher carbon stuff and it cools down too fast, you'll get martensite and might crack, okay. So in a very high-quality forge shop, okay, you would have furnaces and you do controlled cooling, okay, on big parts. But small parts — I mean, if you're talking parts this size, a lot of times you actually, if it's the right size part, air cooling is great. It's called normalization, okay. It's actually a heat treatment step. But for the majority of forged parts, they do something to control it. But if you're low carbon steel, if it's just you're forging some hinge or something and it's not critical, it's not a big deal. But most of your forging is medium-carbon and high-carbon going into turbines and things, and you did that, General Electric would fire you right away. Yes.

p52 46:09

Student: [question about fuel in BOF]

The fuel in this one is oxygen burning the coal — the carbon — out of the cast iron. I got four percent — cast iron, carbon in that cast iron. So I got — I got 12 tons, four percent times 300 tons of cast iron, and I blow through pure oxygen in there. There's plenty of heat, I burn up 12 tons of — right? — just— and that 300 tons — this whole thing stands about four or five stories tall, okay, and this is maybe only four or five feet tall to the bottom. And when they blow the oxygen in there at supersonic — pure oxygen at supersonic velocities — this becomes a froth and fills up this whole three or four-story-tall container, okay. And it's just a foam, it is fireworks galore, okay. It's real exciting. You're not allowed to stand anywhere — well, you got to be in the doghouse, which is where the control room, which is, you know, insulated in there, everything, or you got to be outside the building. I mean, there's literally sparks just flying. It's a very violent reaction. But it's very efficient. Like I said, it takes a 20-minute blow to get all the carbon out, whereas the old days when you just had a boundary layer and you just blew air across the top and you didn't get a froth, okay, it could take you know, a day or two.

p53 47:39

And that's one of the reasons — we'll go through the productivity next time — but the productivity has gone from about 3,000 hours per ton of cast iron back in the days of Saugus Ironworks 400 years ago, to about 20 minutes today per person-hour to produce it. And that has made a world of difference, and that's one of the reasons why steel is so widely used — it's cheap, because they have spent so much money. And I'll show you the curve that I came up with, and it's really dramatic. It's a log plot of costs going down, okay, and it's getting steeper, okay, in recent years. Okay, we ran over. Sorry.