SMS_F2013_08

Randomly, uh, but I might try to do it logically. Nothing wrong with doing it logically. Um, but it doesn't matter too much. Um, no questions. If you ask questions I can digress. When you're done with your presentations the class is over, as long as you've done another, you know, another module online. Isn't that great? You don't have to work over Thanksgiving — not that you would have — but you don't have to work in December, right? That's what I think it's good about. To me it's nice to have one of your classes out of the way, right? It's also nice to have a class you don't have to worry about studying for quizzes. It's nice to have a class that is over. Whatever. I guess there's lots of ways just to look at it, okay.

And Christine gets back, is it on? Oh okay, we're all set, you recording? Okay. Um, so other than that, this is just a historical note. I was kind of looking at some other things for other reasons. This is the U— this Department's curriculum from the MIT catalog of 1889 and 1890. And uh, since Dr. Belmar has been giving me a hard time about I'm teaching an economic class to you, um, it was required in your first term sophomore year to take political economy back in 1889. Uh, it was also required to take three semesters of German. Anyone know why you had to take German? They didn't give you a choice, you had to take German.

In all of science — I mean, all the quantum mechanics, weren't all those, you know, uh, [Pauli] and you know, um, [Bohr] and the Bohr atom, and uh Schrödinger and all — the Germany was the place for science. It wasn't the United States. In fact all the way up through World War II it was still — I mean, the United States was coming on, but it was really after we stole all the great German scientists after World War II and brought them back to the United States that we sort of took over the field.

And so now if you go to many uh universities — I remember the second year I was on the faculty, we had a uh young woman who had decided, I think she was of Scandinavian descent, she decided she wanted to go to uh University of Oslo for a semester abroad. This is before we had all these programs. And she came back and she was just starting her junior year and I was her adviser, and I said, well, how did you learn Norwegian? Oh, I just, you know, I just went over there. I said, how did you take the classes? Cuz I found out she'd gone over there five days before classes started at the University. Said, how'd you take the classes? Oh, I just took them. What she didn't tell me, I found out later, is in many countries including Norway, if there is one student in the class who doesn't speak the native language, they will require all the other students to take the class in English. So the Norwegian students had to take the class — and the professors all speak English. You talk to Germans, you know, everybody speaks English.

The only place where people want to speak — want you to speak their native language even if they speak English — is France. And Professor Pelloux [?], who — he's retired now from this department, but he was — studied in France and he came over here and got his PhD in this department and became — went to Boeing and then he came back here as a faculty member. It was sort of funny, he would, um, he would force the French to speak English when they came to visit him here. Okay, so he'd been sort of Americanized.

But anyway, this is some of the courses that people would take. And at that time there was a metallurgy option, but uh course 3 originally at MIT was mining engineering. Metallurgy didn't come along until 1888, so far as that goes. And to show you what — this is where the um Laboratory for Electronic or Advanced Materials is, you know the nice little glass-in thing across from the eating way over here, it's right up here. You're, if you were suspended on the second floor, this is the basement, this is the first floor. Until about 1962 this was a high bay, and they — this was the uh pyrometallurgy laboratory. They used to reduce ores and they did casting and things like that. Uh, and so the laboratories have changed, okay. They don't look the same as they used to. I don't know what that had to do with anything, but just gives you a little historical perspective.

Anybody else that came in have any questions? So we finished up talking about productivity of steel, um, and all I want to point out is, even in my lifetime there's been a change from about six person-hours per ton to about a third of a person-hour per ton. That's a factor of almost — about 20 in productivity. And as a result of that change in productivity, all through the 1980s — about 1980 it was 6 person-hours a ton, by the end of the '80s about 3 person-hours per ton — and because they had already put in continuous casting and basic oxygen furnaces, and the productivity — well, the productivity came from those types of things, and the labor rates went down because of those things. All of a sudden there was a worldwide overcapacity, and the steel companies all started losing money, okay.

There was only one steel company I know of in the United States that never had an unprofitable quarter all through the 1980s. It was Allegheny Steel. Anybody know why they never had an unprofitable quarter? Well I'm not sure there's any uh direct thing, except in my opinion it's because they had a CEO, a guy named Richard Simmons — and not the guy who does the exercises, but uh Dick Simmons of Simmons Hall, okay. Uh, Dick Simmons was an undergraduate at MIT. He had a hard time getting through MIT. He learned to work hard at MIT, he said. And he went off, became CEO of Allegheny, and he understood the industry he was in. He didn't go to some business school to learn how to manage a business. He kind of figured it out on his own. And he knew when they should invest in a new technology, not because it made economic sense to all the bean counters, the financial people, but because he knew this is going to revolutionize the industry. And so it helps to have a CEO who's not just a business-school-trained automaton. Did I say that? Okay, anyway.

Um, but there's been a tremendous — I mean, a 20-fold increase in 30 years is a pretty dramatic change. And as a result of all this increase in productivity, which continued on with the mini-mills, Wall Street became very um disenchanted with the steel industry because they weren't making money, and they said steel was dying. And that's why in the early '90s I wrote that article I gave you on the future of metals, okay, where I basically tried to argue that steel is still an important material today.

I want to tell you about a similar type of productivity increase in the aluminum industry, okay. Now this um started out — this is a book uh about the formation of Alcoa. You can't see it because of the glare but it says From Monopoly to Competition, the subtitle is The Transformation of Alcoa from 1888 to 1986. Well, um, aluminum was discovered in uh about 1825 by — I think his name is Hans Christian Ørsted, okay. Ørsted is for the magnetic field. He was a German [Danish], so far as that goes. And he took aluminum trichloride um and reacted it with potassium and made little beads of aluminum, okay, which he then hammered out and found it was a ductile metal. And that was sort of interesting, and silver and light, um, and so it was interesting but it was still expensive.

And so it turns out that in 1850 the French were interested particularly in aluminum. This is a Napoleonic baby rattle made out of aluminum, because it was more valuable than gold, okay. So Napoleonic babies got to uh rattle their aluminum chains. Anyway, the top, the tip of the Washington Monument, um, which was like 18— should have been 1886 because it was supposed to be the Centennial from 1776, but I think um, or something like that. Anyway, the top of the Washington Monument is this little piece of aluminum at the tip. It was actually not at that time more valuable than gold, but almost half the value of gold. And in fact the price over time — oh there's the guy who, Hall, Charles Martin Hall in 1885, about the time he discovered the way to electrolytically refine aluminum. Here's his sister, who also helped him, in spirit of gender equality, she was sort of also-ran.

Anyway, um, these are prices of aluminum in 1857, 1852 to 1897. And you can see back in 1852 in France someone calculated a price of $545 an ounce. Is that — oh that's per pound, uh, per pound. So you got to divide by 12 or 16, I don't know if it's avoirdupois or Troy, but anyway that's more than $25 an ounce, which is the price of gold back then. We had a fixed price of gold for about almost a century. And by 1854, gee, the price had dropped, it was cheaper than gold. Um, but they were still making it by reducing aluminum chloride with sodium metal. Uh, they switched from potassium to sodium. Uh, and they could only make little beads, and those little beads they'd try to melt them together and they'd have a hard time. And so they used a flux, and they found out something called cryolite, which was sodium aluminum fluoride, was a pretty good flux for remelting these little beads to make small beads into one big bead.

This is a naturally occurring mineral, and they found that if you melted the aluminum under this — put took a torch to it — you could, it — they didn't know what was going on, but the aluminum oxide skin would be reduced by this hot molten fluoride, and you'd be able to get the little beads to melt together. And so you could make baby rattles and sell them to the king of France, okay. Not a very big market, sort of a boutique niche market. But nonetheless, um, here's a picture of Andrew W. Mellon, Carnegie Mellon University, the Mellon Bank, uh, with his aluminum Pierce-Arrow uh automobile in the 1930s. So aluminum automobiles have been around for a long time.

So why this — oh let me go back to the price. The price dropped — where's my price page? Uh, 1886, $8 a pound, down to 36 cents a pound. Or if you want to say 1862 or stuff, from $12. Anyway, about a factor of 40 in 30 years, or a factor of almost 25 or so in 20 years. All because of two people: Charles Martin Hall in Oberlin, Ohio — I think he was in Oberlin, Ohio — um, and Paul Héroult, H-E-R-O-U-L-T, uh, both were experimenting with this cryolite, and a new source of energy that had just become available in the mid-1880s. Anybody know what the new source of energy was?

You ever heard of Westinghouse or Edison? Okay. Thomas Edison had started a company — I don't know if he started the company by then — he started a company called General Electric. With a partner of his, Edison had 400 patents, and his partner is a guy named uh Elihu Thompson, uh, Tom— um, is AC versus DC, that was Westinghouse versus Edison. But Edison and Elihu Thompson — Edison had 400 patents and Thompson had 380, and the two of them got together. Anybody know what Elihu Thompson's day job was? He was MIT professor of electrical engineering. He also for one year was president of MIT, acting president of MIT in like 1907. And they started General Electric at a little plant up here just across the Lynn River in Lynn, Massachusetts. There's a General Electric plant there today, although they're about to close it, they've been about to close it for 50 years. But it was the first plant of General Electric, okay.

And I have some great quotes from uh actually in my thing about leadership and management. MIT — I think I gave a quote from Edison about MIT has the best students in the world or something. Anyway, um, there's a quote in there — there's several good quotes from Edison about the quality of MIT students. Um, and in any case, Westinghouse — George Westinghouse — was interested in AC electricity, and Edison was interested in DC, and so they had this big fight. But in any case, all of a sudden electric energy was available in large scale to people, mostly to replace lights, the Edison light bulb and things like that. But that's when arc welding started, was the 1880s, started in Russia, but nonetheless — um, uh, actually it was Bernardos in France, but he was a Russian or something, anyway.

Um, but Hall in the United States and Paul Héroult in France both tried putting electricity through molten cryolite that had dissolved aluminum oxide, and they were able to make metal. And they had a big patent fight, protecting the whole patent is right here, okay. Um, and uh, it turns out, I think — I can't remember, but it's like days or hours apart in terms of their filing at the patent office. Um, Hall won by hours or days of the filing of this patent to make aluminum. And all of a sudden a new economical way to make aluminum, the price dropped. Uh, Charles Martin Hall created a company called the Pittsburgh Reduction Company. He was from Ohio, he had Andrew Mellon and other people from Pittsburgh as his backers. They started the Pittsburgh Reduction Company, and it later changed its name to the Aluminum Corporation of America, which we now know as Alcoa.

And in fact at one time there were three companies that were merged together: Alcoa, Boeing, and Pratt & Whitney were all one company. And the trust-busters — remember the trust-busters in the early 1900s? — they broke them up, okay. Can you imagine what those people would be doing in the aerospace industry if they kept that monopoly? Um, anyway. Um, tremendous decrease in price, and everyone thought this was a wonderful industry. It was the darling of Wall Street because there was no aluminum industry before, and all of a sudden you had another metal to compete with steel for certain structural applications. And so Wright Brothers — the engine block of the Wright Brothers aircraft was aluminum-copper alloy, okay. Heat-treated aluminum-copper alloy. We'll talk about aluminum, um, not this later this week, but next week when I'm lecturing some more about why you heat-treat it and things like that.

But in fact, it's sort of the difference — what was the difference, can you even think why the difference was? Steel had a 30-fold prod— or 20-fold productivity increase from 1980 to today, 30 years, and Wall Street hates it. Whereas aluminum had a 30-fold price reduction, productivity increase from 1880 to uh um 1900, and everyone loved it. What's the difference? Aluminum shiny — but what does Wall Street care about? Profits, right. Aluminum was a new industry and it was making money. The problem with the steel industry from 1980 to 1990 and all the way actually up to early 2000s was there was an excess capacity. The world didn't need more steel plants. Their productivity had come on top of an already huge industrial base, and therefore everybody's slitting each other's throat to cut price to try to get the market. So aluminum was all new markets, steel, you know, in 1980s was all old markets. And they were losing — they weren't actually losing the amount of steel used, was increasing — but the profitability was going down because there was an over-capacity. There were too many — too much, too many steel plants in the world. And it took about 25 years to demolish the old inefficient steel plants. Plus, at the same time, countries were building steel mills because it was so fundamental to their economy, um, that they were building new facilities.

So, um, I talked — so that's it on kind of the economics of aluminum and this big change. There was also a big change in the use of magnesium about 1850s, as people learned to increase the use of magnesium um, but it never caught on. Uh, and we'll talk about magnesium and its properties later. Um, I showed you the big increase in uh turbine blades and the operating temperatures when they started going to internal cooling. Nice kind of slow rise as they get better metals, and then they started putting coatings on the metals. Like here's a ceramic-coated uh aluminum [nickel] blade with a bunch of cooling holes and stuff. If I ever pass that around, if I haven't — that's got — first they take the nickel-based alloy like they did the one I passed around before, and they um they make it um they spray on a nickel-titanium bond coat — it's also on my cookware in my home, in my kitchen now, a nickel-titanium bond coat — and then they take zirconium oxide, just like my cookware, although they were doing this about 25 years before my cookware, okay.

They had to change from the old Teflon in cookware, cuz people were afraid that you'd get te— uh the fluorine in your food. Well, who cares? You brush your teeth with fluoride toothpaste. But nonetheless, um, there were some rules about Teflon and cookware, so they sort of got rid of a lot of that, and they've gone to ceramic-coated cookware, okay. Still a little pricey, but um uh it's coming down in price. But the aerospace guys were using that 35 years ago. So they put the bond coat on, and then they put — they actually electron-beam evaporate the zirconium oxide in this nice little $20 million evaporator that can do about 10 blades at a time, okay. So it's a little pricey, but there's a lot of value in those blades. Um, and so they put an aluminum oxide, they put a zirconium oxide — the reason they use the zirconium oxide because it has a coefficient of thermal expansion about uh twice that of most other ceramics, which comes closer to matching the metal. One of the problems of trying to make ceramic-metal bonds is the difference in coefficient of thermal expansion. When you heat them up you don't want them spalling off, okay.

Anyway, there are other properties. I've shown you mag— increase in magnetic properties of permanent magnet materials, um, I could — if I had a plot I could show you soft magnetic materials and the improvement, you know. I just found a quarter, I think I found a quarter, anyway, I'm rich. [Tom finds a coin on the floor.] Anybody leave a quarter on the floor? Anyway, uh, we talked about magnetic materials, I talked about steel productivity, I've talked about um the increasing temperature of turbine blades. This is the increasing — or decreasing — optical loss, increasing percent transmission per kilometer of glass. I don't know, the Egyptians didn't have very many kilometer-thick pieces of glass, but you can calculate it, right. And the Phoenicians had better glass. This is a very nonlinear scale here. Then they had optical glass around 1900, where they were starting to grind um good-quality eyeglasses. And over that period of time there was about a four-order-of-magnitude increase in the clarity of the glass. If you go down to Jamestown, Virginia, and see how they make glass, it's sort of a lime-green color glass, okay. That's because of the impurities that are in the glass. As we learned to improve glasses, um, we ended up getting uh clear glasses.

Which brings me to — did I mention the glass flowers? Okay, does anyone know what the glass flowers are? You've lived here all your life and you don't know? Uh, it's the most interesting thing in the Boston area, okay. If you're into art, it's at the Harvard Peabody Museum. And I used to tell freshmen to get out of the classroom and go do something useful, um, and that they should go see some things. And one of — I'd always first ask, what's — were the glass flowers. Only one out of 10 tour books in Boston even mentions the glass flowers, okay. And you ask people who've lived in Boston all their life, have you ever seen the glass flowers, and they say, huh?

Anyway, Harvard in the 1880s decided they wanted a botany museum, and they were considering having different greenhouses with climates from around the world. And they'd have to keep the plants up, and when they die they may have to send someone back to Central Africa or Nepal or something to bring back a plant, right. And that was going to be kind of pricey. So they found in Prussia, uh, they found a glass blower named Blaschka, and he was making little sea urchins in colored glass. And so they said, well, can you try to make some flowers? And so he made some flowers, and he shipped them across, and they got — they were broken when it got across the ocean. They didn't fly them over, they shipped them over. Uh, didn't have a lot of planes in the 1880s. And anyway it was promising, so some wealthy lady on Beacon Hill or somewhere uh gave some money, and for the next 40 years Blaschka and his son — or next 60 years, Blaschka and his son — would copy the flowers that the Harvard botanists would bring back, in colored glass.

And he'd do sections of their seeds to show whether they're monocot or dicot. It's a science exhibit, it's an art exhibit, it's unique in the world. And 75% of them are no longer on display because they've started to degrade. The glass science was not such as — they just make up — they'd mix different minerals, chlorides and other things to get the colors they wanted. They didn't worry about the long-term stability. They needed — they should have had a material scientist working with them. Anyway, but some of them are degrading, but some of them are — they're fairly amazing. It's the only place I've ever seen a cashew plant, for example. Anyway, so go see the optical — the glass flowers, great thing to do on a rainy Saturday. Um.

Anyway, so optical glass got better and better, and now they took off on a different tangent, because they're changing their metric here. They're looking at light going down an optical fiber. And if you walk down the Infinite Corridor, you'll see kind of almost directly across where we are right now, there's some optical fibers, and it shows how they grow the optical fibers, and I'm going to talk about that later. But they went from a — this is um a percent transmission per kilometer of 10 to the minus 100, up to about 96%. And this was in 1983, that was before Professor Yablonovitch [?] did his doctoral thesis here, where he basically kind of used some of the principles of quantum mechanics — what they did to get this better and better optical fibers at Corning, and this basically kept Corning from going out of business, sort of. Um, they perfected making the optical fibers for all the digital communications, okay.

And this is the percent increase in transmissivity. Uh, when they first started doing this in the late '60s and 1970s, they would try to lay an optical fiber across the Atlantic Ocean and they'd have to have repeater stations about every 10 kilometers. And they got better and better until I think you can go all the way across the Atlantic with no repeater station now, okay, because they've improved orders of magnitude their transmissivity. And they essentially make layers, um, so you have a central core of a clear glass, but then you have layers that will reflect the light, because the light might be coming through at a slight angle, you want it to reflect back, and just bounces down this wave guide, this light guide.

But also this technology had a tremendous effect on the copper industry. Um, copper has four or five uses, and when I get to my lecture on copper we'll talk about those. But what makes copper sort of unique — color, okay. There's not very many metals that have that kind of copper or gold color, like copper and gold, about the only two I know of. Uh, all the others are various shades of gray or silver. Um, so it has color, but it also has very good electrical conductivity, has very good thermal conductivity, and it has a couple of other properties. Well, turns out because of this very good electrical conductivity, it was being used for all those Edison and Westinghouse and Alexander Graham Bell applications of carrying electricity. But anybody know what kind of the maximum frequency that you can put down a copper wire?

Yeah. You do? What's the clock speed on a computer? When you go buy a computer it says, tells you how many gigahertz it is for the CPU. Yeah. Okay, so it's gigahertz over very short distances. Over longer distances, like miles, it may be hundreds of megahertz, because you have parasitic capacitive and inductive losses and things like that. So it turns out one of the limiting factors in computers is how far can the electrons travel through the copper.

And there was a big — it was a big deal about 20 years ago when IBM announced that they were able to put copper on silicon. They had always been using aluminum on silicon. And it turns out copper on silicon was a very interesting concept, because ordinarily copper would diffuse into the silicon and destroy all the silicon properties, okay. Aluminum didn't have that problem, okay. So they were using aluminized semiconductors all through the 1980s. And IBM announced around 1990 that they'd come up with a mystery barrier layer that would prevent diffusion of copper into the silicon. And, you know, they wouldn't tell anybody what it was, but most people could figure out it was tungsten, okay. They'd lay down a thin layer of tungsten, then they put the copper down, and the copper wouldn't diffuse through the tungsten. Everybody knows what it is now, and most of your better semiconductors today have copper conductors.

But in fact, um, if I put this up — I do have this with me — this is a — [Tom holds up a chip package.] This was a Pentium 5 [Pentium 4] from the mid-'90s, okay. This is state-of-the-art. This actually doesn't have the chip on it, but it's an Intel chip, and it's got a — what we call a tape-automated-bonding chip. But this little one-square-cm chip — and I don't remember what the Pentium 5 [4] operated at, or how many transistors, it wasn't all that many, maybe 10 million or something. What's — current chip today has got half a billion or something — I mean it's got a lot of, it's got over 100 million, okay. In any case, the Pentium 6 [Pentium Pro / II] was actually two chips, one square cm and one about a half a square centimeter, and it limited the clock speed because you had to go from here to there, and that limited the speed that you could run the clock for this thing. You'll learn more, when — if you take the module on solid-state bonding and soldering and stuff, when I talk about that stuff.

But what happened to the copper industry was — I don't think I have any, I never carry change with me — um, back when I was your age, pennies were made out of copper. And at that time copper cost 60 cents a pound, and I can't remember, 400 pennies made a pound or something. Um, and whenever the price of copper got to $2 a pound it was going to be cheaper to melt your pennies and sell it for the copper than the value of a penny. And so the government was looking at — inflation is going to kill the copper penny. And I don't remember, in the early '80s they came up with the zinc penny, which is the pennies you know of today. They're actually zinc with a copper plate on the top of them. And now that zinc price is going from 40 cents a pound up to whatever it is today, 80 cents or a dollar a pound, the extra processing means that it's getting to the point where it's going to be cheap enough to melt your pennies to get the zinc out of it. Because it currently costs like 1.6 cents to make a penny. Okay, the pennies cost the Treasury more than a penny to make, okay. But what they're really worried about is not their fabrication cost, because pennies last a long time in service. They're worried about the fact that when this is the cheapest source of zinc, okay, or copper, people are just going to melt them down. And at that point they can't afford to make enough pennies to replace all the ones that people can melt illegally. But nonetheless they'll do it.

Um anyway, in the early '70s, uh, people were worried that they could not, because of the limiting speed of however many hundreds of megahertz, that you could send data down uh the telephone lines or anything else, because of the limitations of copper. And just the another— number of telephones, there was not enough room in the manholes or the, you know, the tunnels under New York to carry the amount of copper wire that was necessary to carry all the telephone conversations and the electrical power. So two things came along that were going to save this. Um, and actually one of them did come along.

And one was optical fibers. So all the data communication could go from hundreds of megahertz to — I don't know, I didn't calculate it, but optical fibers are probably operating uh around 100 GHz, okay. So a thousand times. Now you can stream video and everything, cuz it's all on optical fibers. And all of a sudden the price of copper, which was climbing up towards $2 a pound, dropped, because for all the data applications you didn't need copper, okay. And so they could use the copper for the electrical power things. They were still worried about the fact they were running out of room under the skyscrapers in New York City because of the amount of electrical power, okay. And so you could go to higher voltages, but you can't go to higher voltages forever, because ultimately things will arc over and you have problems.

So what did they do? They had the salvation in 1989. What was the salvation? Superconductors. High-temperature superconductors. The problem is this one didn't work, okay. Uh, I always say the only good thing came out of high-temperature superconductors is I got my office on the great court, uh, when Greg Yurek resigned in 1989 to start American Superconductor. Um, but American Superconductor has never made money. Um, but they certainly get hundreds of millions of dollars in government research grants and industrial research grants because there is this promise of superconductivity, okay. As a material that would solve all the problems of all the power in the big cities, because you don't have enough room to carry all the power to all those buildings and things like that. Um, they still have the problem and they're running out of space, and it will limit the ability to build buildings as big as we want.

But um, does anyone know why superconductors — um, well, why the high-temperature superconductors did not um solve all the problems? It's because it's not just high temperature that's important, okay. This is a doctoral thesis from mid-1970s, one of the best ones ever presented at MIT, written by a guy named Tom Eagar, okay. This is my doctoral thesis. I worked on superconductivity. And the first figure in my pieces was a schematic phase diagram — I'm always touchy-feely, never put numbers on anything — of critical temperature for the superconducting phase field. So the superconducting phase is, if you get below a certain critical temperature, below a certain critical magnetic field, and below a certain critical current density. We had known this since let's see — superconductivity was discovered in 1913 — we probably learned this about 1914 or something, okay.

Um, and here, a brief history of superconductive materials: "Although the phenomenon of superconductivity was discovered over 60 years ago" — well yeah, in 1975 — "it has only been in the past decade that its technological potential has begun to be utilized." That's because General Electric started selling niobium-tin [Nb₃Sn] superconducting magnets, and uh other people had made niobium-titanium superconducting magnets around 1960, mostly for high-energy physics experiments where they needed high magnetic fields. "Early in its history, superconductivity was envisioned as a means of lossless electrical power transmission" — solving the problems under New York City — "and of generation of large magnetic fields," such as superconducting — you know, high-field magnets for um, for all kinds of things, but at that time mostly uh physics experiments. "In fact, in terms of energy input, a superconducting magnet would consume a few kilowatts of refrigeration compared to the megawatts of resistive heat loss," blah blah blah.

Very well-written. And not only that, you notice this right- and left-justified — this was the first word-processed thesis in the department. It cost me about $3,000 to do that off my NSF grant. Um, I had to go over to building — want to — building 38 or something, and sit there at a terminal and type it in on a mainframe computer. And the mainframe computer had like 64K of memory. It's just fantastic, you know. Anyway, um, but in any case, superconductivity, um, it turns out there's a relationship between critical temperature, critical field, and current density. But there's also a relationship that says that in order to get that current density, we have to have imperfections in the material, which you call pinning centers, and blah blah blah. Anyway, I don't want to teach you about superconductivity in particular.

But we knew, um, those of us that had worked on superconductivity in 1989, knew that a huge uh critical temperature increase was going to translate to real problems in trying to get any decent critical current. And you can't make a magnet without putting current through it, okay. Um, you would have great uh critical temperatures and critical magnetic fields, but you had to have pinning centers that were smaller than an electron. And it's very hard to do, okay, with all the knowledge we had um at that time.

And so I like to say at that time there were five faculty in this department who had worked on superconductivity, out of about 30 faculty, at some time prior in the previous 30, 40 years. And the four who did not — almost everybody — there's this lemmings heading to the cliff. The National Science Foundation announced in 1989 that the only new proposals they would entertain would be on superconductivity. So if you had any other topic — ceramics, polymers, metals, didn't matter — if your topic wasn't superconductors, high-temperature superconductors, they were not going to fund you for a whole year at National Science Foundation. And so, I mean, everybody was telling Congress this was going to transform our whole energy landscape, and blah blah blah. And so within one year, half of the faculty in this department — it went from one person doing active superconductivity research to about 20. And of the five of us who had worked on superconductivity before, only one of them, one of the five, continued to work on superconductors. That was a guy named Dave Redman [?]. He was an assistant or associate professor and he continued to work on superconductivity because it was an important field and he was in it. And the four of us that already had tenure, we wouldn't touch that field with a 10-foot pole, because we knew enough about it that we said this is not going anywhere, okay.

Um, and I already told you the story about my one time five years later, my department head and I — made someone ask me a question at a conference over in Lille, Cressy [?], and I said, oh no, we won't see high-field magnets in our lifetimes and our children's lifetimes. And you know, one of my colleagues took me to the dean uh over my negative attitude about high-field superconductors.

Um, but in fact, if you really know the limits — what I really want to point out — there are limits to materials, and superconductors are an excellent example. You have to know not just one property. This is fracture toughness and density. If you look at these Ashby plots there's a limit, okay. They're only — things only get so light, okay. Polymer foams or cork, okay, or bolts of wood are very light, but they don't have very much strength, okay. There is a correlation between — this is toughness and strength, but in any case, you'd love to be super light and super strong. You'd love to be over here, super light and super strong. But there's nothing that exists over there, okay.

And why is that? Well, fundamentally it's because everything that we know about materials is related to their bonding. And Dr. Belmar put up a plot like this, which is the energy versus distance between two atoms. If I bring two atoms together in space, just two atoms alone, they will find an equilibrium distance of separation, which we often call a₀, because this is the energy potential well. And it's an electrostatic energy to the minus 1 over R², you learned about electrostatic energy in freshman physics, okay, and it goes as 1 over R². This is a nuclear repulsive term, and it goes as something like R to the 8th, R to the 10th, okay, um, and the two of them compete with each other. And if you try to squeeze them any tighter, they just don't want to squeeze together. And he pointed out that this curve is actually steeper than that curve. He pointed out — if you blow up this region right here, um, it's actually asymmetric, it's not a nice symmetric parabola. But in fact he talked about — you can explain Poisson ratio because of the energy potential. This is sometimes — in the old days, uh, some of the first ones in the 1920s that the quantum-mechanics people generated, were called the Lennard-Jones potential. Uh, they've got all kinds of names nowadays.

But in fact the curvature in here is nothing more than the modulus — the strength — modulus. So if we're talking about modulus of elasticity, it's proportional — well the force plot, let me do a force plot. The force is equal to dE/dR. And the force plot looks something like this, where a₀ is where it crosses the zero force. That's the equilibrium distance of separation. So if I look at this thing coming down here, that's a zero, and so you got a force. And if you differentiate the force you get — so it's now, the modulus is going to be proportional to ∂²E/∂R², okay. So there's a math, okay, um, to prove I'm an MIT professor, okay.

Um, but in fact, most of what we know about strength of materials is contained right here, in the bonding potential between those metals. That's for structural materials. And everything you need to know — magnetic properties, optical properties, um, piezoelectric properties — is contained, if you know everything about where the electrons are and how free they are in the material. So I'm not teaching a course on structural materials or on functional materials — in structural materials you need to know the Lennard-Jones bonding potential, and from this fundamentally you could generate everything else in the world. And that's what computational material science today is all about.

They were able to do this for very simple systems in 1925 out of quantum mechanics, but they didn't have the computing horsepower to do it for more complex systems until somewhere the 1990s, computational material science — the computers got powerful enough that they could start doing this. And now a third of this department, the faculty, are working on computational material science. Why? Well, it's a new tool, the computer, to be able to do this. The other is, it's a lot cheaper and a lot easier than to go to that laboratory and actually do something with your hands, okay. You can sit at a computer terminal, right. And not only that, most people can't check your work yet — it's even better, okay. No one knows whether you're right or wrong. You can just produce junk, and if you convince people you're right, good for you.

Um, but there are limits to materials, whether it's optical transmission, magnetic fields, fracture toughness, and strength. We would love to be way out of here, but in fact nature limits us to certain materials that have certain bonding strength. The bonding strength is this stress, okay, is this force plot — how, oops, well — it's basically how deep is this energy well, and how much force does it take to pull the things apart, okay. And if you take my soldering and brazing uh thing, I'll go through this and talk to you about how the fundamental strength of two atoms when you bring them together is about 2 or 3 million pounds per square inch, but we don't get that strength, we only get about a tenth of that, and that's another story for another module.

But in any case, so we'd like to improve the properties — uh, we have not just one property. Ashby did it in two dimensions in his book. He will sell you a $50,000 program that can do it in 10 dimensions. You tell them all the properties you want to optimize, and it will go through and try to select a material for you. And if you got $50,000, you don't have to take this course, okay. You can just plug in and get a bad answer out of the computer. Whereas you can ask me and I'll give you a bad answer and I won't charge you $50,000, I only charge you — what is it — $48,000 to be a student. Anyway.

Um, so one of the things we do to solve this problem is we make composites. And as much as — you know that I now think steel is a wonderful material, but remember, your homework is to figure out what the Achilles' heel of steel is. Cuz next week when I'm lecturing again I'm going to ask you if you figured out what the Achilles' heel of steel is. It's got fantastic strength, toughness, and cost — those three, just those three multifactors — it's just better than any other material. But um, it doesn't have all the properties we want. No material has all the properties. If it did, that's the material we would be using for everything.

So one thing that gives material scientists a job is they get to try to figure out what the best material is. And composites are wonderful materials. I've shown you my $12,000-a-pound piece of the X-33 space plane. I've shown you this thing of the V-22 Osprey. Could not have built this without composites — graphite-epoxy composites. Uh, if they had made it out of something heavy like aluminum, they never could have gotten it to fly, okay. It's the first large all-composite aircraft. They had made smaller little, you know, single- or two-person jets and stuff out of all composites. But the Osprey was the first one that was all composites.

The 777 — I think I told you it was supposed to be like 80, 90% composite until Boeing started to design it in the early '90s, and they found they couldn't afford the composites for commercial aircraft. This was a military aircraft, and you start figuring out what the V-22 Osprey cost on a dollars-per-pound, and it's probably up in the thousands of dollars a pound range, which you could not afford for a commercial aircraft, okay.

Um, so composites — um, I sometimes in my pejorative way say composites are the materials of the future, they always have been and they always will be the material of the future. They are starting to find their own. The 787 is like 80% composites, okay, by weight or whatever. Uh, however it was also about three years late on a three-year development plan, so it took six years to develop it, in part because of all the problems of developing a manufacturing base for making composites, and making them reliably, okay.

Now in spite of my negative comments about composites, um, composites are some of the most important structural materials we have in the world, and in the largest volume. Anybody know what I'm talking about? Concrete is a composite. Concrete, they put stone in as just an extender, okay. Mortar is just the cement, uh, but if you put stone in it and make it concrete, you also put something else in concrete — rebar. [Tom holds up a piece of rebar.] Oh, just happened to have a piece of rebar, okay, I didn't bring a piece of stone, but you can find one outside. Uh, rebar, you put steel in because one of concrete's problems is it's not very good in tension. It's a ceramic, it's got lousy fracture toughness. So you put rebar in it. You put stone in it to make it lighter — or cheaper, okay. Here's this old plot that we've been putting up, and what's cheaper than cement and used in higher quantities? Stone, okay.

Um, what's another composite that you drive on? Asphalt. It's just tar, but you put stone in it as an extender, okay. It's sort of like putting — I don't know, uh, I'm not thinking of a good example right now of — I usually like to do food analogies, but uh, uh, uh — you put silicon dioxide in your toothpaste as an extender, okay.

So um, what are the problems with high-volume, very-low-cost materials like stone and cement? What are the limiting factors in use of these materials? It has nothing to do with production of the materials. It's weight, and what — could be failure, but we usually use them and just lay them down on the ground, you know, concrete or steel or asphalt. It's transportation cost. So it is weight. At something that's only worth $40 a ton you can't ship it very far, okay. And so there's stone quarries all over the world, in part because there's stone all over the world, but not necessarily the best stone. You will also often use whatever is locally available, even if it's not the best stone, because it costs too much to ship stone. There are not big ocean-going vessels that are just taking crushed stone and taking them from one continent to another, okay. It's transportation cost when you get way down on this end of this plot, okay.

Now cement, it turns out, uh, there's limestone in most parts of the world. No limestone in Hawaii, because limestone came up from the ocean, and all of Hawaii is volcanic, okay. So there's no little sea urchins, or sea animals that died and gave their life to make limestone for your garden. Um, but so you have to ship cement to Hawaii. And it might be cheaper to ship steel, which is 10 times stronger — it might be twice as heavy, but it's 10 times stronger, so on a per-pound basis you might want to build parking garages in Hawaii out of steel rather than concrete.

But we can ship concrete around the world. I mean, Nigeria one time, in their corrupt government of 25 years ago — which is different than the corrupt government of today — um, they had a guy who just ordered, it was like 200 shiploads of cement. He was going to rebuild Nigeria with all this oil money, uh, before he got deposed. And at one time there were, about 20 or 25 years ago, there were 200 ships sitting in the Nigerian harbor full of cement waiting to unload, because they had a strike or something, I don't know, at the loading docks, or they ordered all this stuff and they never figured out how they were going to unload it, I don't know what all the problems were. But the ships were just sitting there waiting to unload, and in the meantime all the humidity was causing the concrete to sort of set up, okay. And all these ships are going to be junk, okay. This is sort of a major crisis in the shipping industry 25 years ago, but a lot of the world's shipping capacity was tied up outside Lagos, Nigeria area. But anyway.

Um, cement and concrete are two composite materials and they're very important. Uh, cement, concrete, and asphalt — um, there are some very important composite materials that are used in very high volume. There are also at the other end of the spectrum some very important composite materials that have fantastic properties, but they tend to be boutique materials. That X-33 space plane — well, they built two liquid hydrogen tanks, and they weighed a total of 8,000 pounds. And I don't care if you get $12,000 a pound for it, it's only a $100 million market, okay. So if you look at — also market size, large markets up in this corner, these are the boutique materials, may have fantastic properties.

Now, I remember uh one of — uh, when we started the manufacturing program at MIT in '89, one faculty member of mechanical engineering, who's no longer in composites but got his tenure in composite materials, was supervising students at Boeing. And this was back in the days of, they were looking at the 777 and trying to — they were going to make it all composite and stuff — and um, the students went off and found that the value of a pound saved was $188, on the confidential Boeing information. Now we all knew it was $200 a pound — Boeing knew it was $188, so they had a 6% advantage on all the rest of us, big deal, okay. But I remember the students came back after doing their thesis, and this faculty member who had spent eight years, 10 years of his life getting tenure working on better composites, came back and said, I've been working on the wrong thing, I should have been working on cheaper composites.

Cuz anybody, just like all-aluminum automobiles in the 1930s, anybody can do it. You can build a spacecraft out of composites, but can you build a Ford Taurus out of composites? The answer is no, okay. And I will predict that we will not be riding around in all-composite Ford Tauruses 20 years from now, okay. 20 years ago — I, 25 years ago — I predicted we would not be riding around in all-aluminum Ford Tauruses, at that point 20 years ago. And about 10 years later I quit predicting that, okay, because we might be riding around in all-aluminum Ford Tauruses in the next 10 or 15 years, depends. I still don't think we will, but I'm not going to predict it. But I can predict — um, and I don't know what I'm going to have time to tell you about — well, I'll tell you anyway, you might have to leave it right there.

Anyway, so you can turn that off, okay. I don't like to gamble.