SMS_F2014_04

Questions? I'll ask a question, since some... how many of you are Course 3 undergraduates? Okay. So I'll ask a question. For whatever reason, for twenty or twenty-five years, Course 3 undergraduates have had a significantly larger fraction of women students than male students, okay. It's like sixty percent for years and years. Um, so yesterday at the faculty meeting Professor Shu put up a graph: the sophomore class this year is eighty-four percent female in Course 3. And so the faculty was all speculating — this type of speculation is just idle speculation. I said you need to ask the students, okay, cuz there're all kinds of, when I was department head, actually it started back when I was department head, and there were all kinds of theories, and some of them were stupid theories. Um, but uh, so why?

Well actually you don't have to tell me in front of everybody else if you don't want to, but the question is, I mean, eighty-four percent is so far beyond — I mean, it never bothered me that it was sixty percent, didn't bother me at all. In fact it's something we used to kind of say, see we're doing a great job on diversity. Uh, but we were doing nothing on diversity in terms of, you know, how many sophomores come in and who comes in and stuff, other than just trying to attract students. And someone's got to be above average and someone's got to be below average right? In the Institute, average is forty-five percent, and so being at sixty percent, okay, it's not that big a deal. Um, but eighty-four percent is so far beyond the average that you have to say, h— why? So uh, that's an assignment for you. You can email Professor Shu why you think. Now this probably applies more to the women in Course 3 than the men in Course 3, okay, cuz you don't have much to do with this. But why, okay? Um, so anybody have any theories you want to tell me in public?

Student: [inaudible — suggests breakdown by department for biology, chemistry, etc.]

Oh okay, yeah. Well I'm — like, what the breakdown is for like biology, chemistry, that — and the Institute has all that. And I used to see that when I was department head, because we talk about these things in Engineering Council and stuff. Um, of course the data is — I mean, this is not suspect data. I mean, they know the gender of the incoming sophomore class. Uh, and that actually means that now the undergraduate population in Course 3 is about seventy percent female, because you have eighty-four percent, then the current classes were about sixty, you know, that's eighty-four percent sort of skewed things, okay, to above seventy.

Um, but uh, they also — he also put up some data about uh underrepresented minorities. And it turns out, comparing to other schools around the country, uh it turns out MIT has thirty underrepresented minorities in Materials Science, and the next school — I can't remember which one it was, University of Illinois, anyway, whatever — had twenty-seven. And then it goes down very quickly, you know, the fifth or sixth school has like five students who are underrepresented minorities. And on the top ten there was not a single school from Texas. And I raised my hand. I said, wait a second, you're trying to tell me there's no Materials Science departments in Texas? Now why did I say that? Because a substantial fraction of the students in Texas are Hispanic. And you're trying to tell me not a single department in Texas has more than three Hispanic students? Okay, so it didn't make sense.

The point is, you have to question data, okay, whether it's reliable. And someone says, oh in Texas all the Hispanic students go into business. Oh come on, give me a break. I mean, I start hearing answers like that and it just galls me, okay. You're MIT students, you can figure out logic, okay. So what I finally learned is, well someone pointed out that Texas A&M just started a Materials department. Materials used to be part of mechanical engineering. So whoever compiled these statistics probably looked at materials departments, and they didn't look and see if you had a big materials department in some state school that was part of a mechanical engineering or a chemical engineering department, okay. So we got bogus data, okay. I mean, the — or incomplete data. But it didn't matter, Professor Shu was very happy because MIT is at the top in that category, okay, and so it doesn't get beat up by the dean, okay. Dean used to beat me up — actually used to try to beat me up, but I never let him.

Um, in any case, so we do track these things. But I, I'm actually very curious now about why the sophomore class is, you know, double, you know, it's twice the percentage of, uh, women in the class. Why? That's got to be — that's more than just statistical uh, um, variation, okay. So if you've got any good theories, let me know, cuz we actually pay attention to these things, okay.

Okay, now, just as uh — it's time to start anyway. Someone asked me something yesterday at the end of class about diamond, and I was talking about structural materials last time. And this is a product. Um, this is a — well, the product goes together like this and there's another half I have in my office, but — anybody know what this tool is?

[Student response: cutting tool.] It's a cutting tool for reclaiming asphalt, okay. You've seen these — well, you have to be driving at night on the highway, and they have the lights out and they're chewing up the asphalt and reclaiming it and then they repave the road. And this is the type of tool, that cutting tool, that just gouges the road surface, okay. Uh, they have similar types of cutting tools that are um on uh mining tools, okay, you're boring a tunnel in the mountain or something, and they have similar types of cutting tools. And typically the material of choice, structural material of choice, is silicon carbide.

Turns out in this one — we don't need this, I'll pass this other stuff around — and this particular one, and they're not cheap tools. Silicon carbide cost about $10,000 a pound, okay, as a powder, cuz it's not easy to make, has high melting points and you have to sinter it. So this one — this is silicon carbide tip, and it's brazed to a steel shank. This is the steel shank. And then this one actually has a sintered diamond, man-made diamond tip, okay. If you look at it I'll pass it around, you can see the tip, it's brazed right here. You can actually see some copper color where some of the braze alloy came down. And then it's got another braze down here. So you've got very hard material, and hardest material known right there. [Tom passes the tools around.]

And what happened is, a company came to me — or a consulting company came to me who is consulting for a company who was looking at buying the company who developed this technology to put sintered tips on there. And it turns out, um, it's a wonderful technology, these things never wear out. What's the problem with that? Student: No replacement. You — no replacement, I mean, you got to have something to — got to have, you got to have something to sell, you know, six months from now.

I had a similar type of problem once, twenty-five, thirty years ago. Someone asked me to evaluate this little company up here north of Boston. This was a Canadian timber baron, and these guys had developed cutter knives. So if you buy a ream of paper, you know it's 8½ by 11, okay, you can tell it's all been cut all at once. These are the knives they use to cut the paper to 8½ by 11 before they package it, you know, in a ream. And these knives are sharp, let me tell you. They are so sharp that — you know, you take a sharp knife and you run your finger over it to see if it's sharp — you do that with one of these knives and you slash, sl— slice your finger off. You have to run, rub your fingernail against it to see how sharp it is, okay. They — I'm not kidding — uh, these were super-sharp knives.

And these guys — it was a steel, a tool steel — and these guys had developed a new heat-treating process and brazing process similar to what these type — I don't know if it's similar brazing process, but these guys had developed a brazing process for that tool I'm handing around. And apparently making these slitter knives was a $25 million-a-year business back in 1984 or whatever this was. And uh, the guy wanted me to evaluate the brazing process and determine whether he should buy this little company. Well, I go up to wherever it was, Danvers or whatever. I meet with the president of this little company and he says, uh, yes, they're getting six times the life, okay, uh, on these new blades compared to the old blades that everybody else in the world was using, and these guys supplied a fair fraction of the world's uh knives.

And uh, so I, my response to the guy in Canada who was thinking about buying him was, uh, I don't think I'd buy this company right now, unless you can find another application of these knives. Because you can take a $25 million business and turn it into a $3 million business or a $4 million business, because it has four times the life. I mean, if you don't have something to grow the market — it's kind of like what you were talking about before, if you improve a material, okay. Um, the general term, long-term growth, says you can double your market because, as you drop your cost by a factor of two, or your factor of two, your volume should increase by a factor of four on average. But you better know where that new market's coming from in the short run.

So the reason they didn't end up buying this company that had that sintered diamond, because it turns out the process and the sintered diamond was so expensive that even though they could get, you know, two or three or four fold advantage in lifetime, you couldn't justify it for the cost of using sintered diamond in that application, because silicon carbide was good enough as far as that goes. Any questions?

Student: How is that cut? Okay, cut — pardon me — that was cut by an electrical discharge wire, EDM, electrical discharge machine. I used to have one right over here kind of across my office. It's now over at the Plasma Fusion Center because when they redid the LMP [Laboratory for Manufacturing and Productivity], Mike Tantillo didn't think we needed it. It was actually half the size of this room. So, but it's a — you just have a little wire, actually it's about a 7,000th [of an inch], it's copper wire, and it goes through a water bath and it just electrically arcs its way through, okay. Very slow process but very precise. Um, and you can cut through diamond with no problem as long as it's conductive, okay. Diamond is not really conductive, but the sintered diamond is, okay. They have to use other things to sinter the hard diamond together, and that — you have to have a little bit electrical conductivity. You can't cut ceramics, but you can cut anything that's electrically conductive, because you're arcing through. I mean, you're using a 10,000-degree micro-flame, if you will, okay. But it makes a very nice cut.

Student: How are the beads bonding with the silicon carbide? Yeah, it's brazed. It's what we call step braze. They first would braze the um diamond to the silicon carbide at one temperature, let's say 1300°F, and then they would braise the silicon carbide to the steel at maybe 1100°, because you don't want to melt the first braze, obviously. So we call that step brazing, okay. When you have to braze two different things at two different times, you can't do it continuously, okay. Uh, I mean, obviously you'd like to do it in one brazing operation, but the braze alloys that work for one set of materials, when you're doing dissimilar materials, is not necessarily the same as it works for another set of dissimilar materials, okay.

Other questions? Student: For very tough materials like carbide, what's like the failure mechanism that you would get on a cutting tool like that? Well, silicon carbide — we can call it tough, I mean, that's — yes, it's tougher than ceramics, most ceramics, but I just happen to have a plot that I was going to hand out, which I guess I might as well hand out, that's — we're going to talk about fracture toughness and strength today. And so let's see where silicon carbide lies on this plot. [Tom distributes the Ashby plot.]

Okay, so this is one of the Ashby plots. Ashby — I told you about Ashby. And you know, if I had thought about it, if I'd decided what I was going to teach this term before third week of August, I might have chosen this as the textbook. Not that I usually have a textbook, because who wants to go spend 150 bucks for a book, right? Um, so I can hand out pieces of it under the copyright fair-use laws. But in any case, the Ashby plots are these log plots. This is strength versus fracture toughness. And if I look in here, here's ceramics — silicon nitride is here, here's silicon carbide.

So how tough is silicon carbide? Well, it's better than glasses, it's better than pottery, brick, in terms of toughness. In fact it's almost an order of magnitude. But it's not as good as zirconia necessarily, it's not as good as sapphire, Al₂O₃. And we have some sialons, which is silicon aluminum oxide, which is silicon dioxide and aluminum oxide mixed together. Anyway, so silicon carbide is tough, but it's not tough like metals, okay. So it's — yes, it is tough, it's a lot tougher than ceramics. Uh, the failure mechanism — it's extremely hard, okay. In fact, well, look at it on the hardness. So the strength level, which is basically hardness, it's right over here. It's one of the hardest things in the world. In fact, diamond is just slightly better, okay. That goes over that way a little bit. Diamond actually — that circle would be continued over here a little further than silicon carbide. So they're extremely hard.

And so it turns out, they're not pure silicon carbide, they're not pure diamond, they're actually sintered powders. And the sintered matrix is something like cobalt, typically. Where's cobalt? Cobalt's going to be up here somewhere. They don't even have cobalt on here. Uh, but cobalt would be actually up in here, very tough, relatively strong. And so the failure mechanism actually turns out to be just abrasive wear. But they're so hard that it takes forever to wear it out — well, not forever. Um, that thing out of silicon carbide might get one night's use. They tend to resurface the roads in the middle of the night, right? Um, but they might get one night's use. With diamond you might get a week's use, okay, on that tip. But you know, with a factor of five, you know, you got all day, you can replace the tip if you were only getting one hour use, which they were back twenty years ago before they improved the silicon carbide sintered, silicon carbides and stuff.

That was a big problem, because now look at the productivity. You're losing all those workmen standing around while some mechanic's changing the tools out. And if you actually look at — I could pass this around if you want, but this is actually fairly — a quick, a sort of a quick-change tool, it's got a little pin in here, it's got slots to keep it, you know, keep it from rotating and stuff. Anyway, um, but as long as they get to one shift, they're happy. As long as it can last for one shift. There's sort of a barrier to introducing a better material as long as you can make one shift, because you have time for maintenance. When you don't have that — in fact, uh, when I talk about spot welding of automobile sheet, when they went to galvanized steel and they had zinc, the zinc would react with the copper electrode tips, and the problem was, they could use copper tips that — they could use copper tips that would last a whole shift on the automotive assembly line, okay, eight hours. When they went to galvanized steel they couldn't even get an hour, which means they had to stop the line and um have some guy had to go in there and change the tip.

What they wanted was 2,000 welds, cuz that would be the break after two hours. So Ford has a spec of 2,000 spot welds. You had to — the steel company had to supply a steel with galvanizing on it that would last for 2,000 spot welds. That was the Ford spec back in the early 1980s. And the reason was, while everyone else was having their fifteen-minute coffee breaks, a mechanic could go in there and change the tips. So you have to actually look at the whole system.

Student: Tungsten carbide versus silicon carbide? Uh, tungsten carbide — if they had it on here would be a little better in most cases than silicon carbide. But — I'm sorry, I told you silicon carbide was $10,000 a pound, it's tungsten carbide that's $10,000 a pound. Silicon carbide is only about 2,000 a pound. Yeah, but it's actually a sintered tungsten carbide. But tungsten carbide only gives you about ten or twenty percent more hardness. So for machining of, in a tool where you're cutting metal and stuff, tungsten carbide is many times better than silicon carbide depending on the steel — not the steel but the material, the metal being cut.

And there it depends. For example, in cutting tools, um, diamond is not so good for the world's most abundant metal, which is steel, right? Why? Because iron dissolves carbon. Diamond is nothing but carbon. There are temperatures of 1000°C at that interface where you're doing the cutting. People have measured it, okay. Um, you basically use the thermocouple technique where your tool is part of one of the electrodes of your thermocouple, and you can measure the temperature. So you get a 1000°C — well, at 1000°C carbon — the diamond starts diffusing into the steel.

So what they use is cubic boron nitride, which is actually between diamond and silicon carbide. Cubic boron nitride has a diamond crystal structure, but boron doesn't dissolve in steel, and nitrogen doesn't dissolve in steel. So when you're machining steels, regular old carbon steels, you use cubic boron nitride. If you're doing titanium or something, you might use diamond tools, okay, as your tools for machining. But it turns out, diamond, even though it's the hardest material, it's not that much harder, and if the wear mechanism is a chemical wear mechanism, which is basically dissolving the carbon into the steel, diamond's pretty lousy actually, okay.

So, um, and by the way, so oh — that was one of the things, someone yesterday at the end came up and corrected me. I told you that little pad conditioner — the little diamond pad conditioner was what they ground the semi-c— [conductors]. The student was right, she actually worked on it down in, out in Taiwan or somewhere, Intel or somewhere. Um, it's actually the p— it's a pad conditioner. They actually have a slurry that electrochemically dissolves away the surface of the semiconductor, and it is machining the surface of the — they call it planarization, to make it flat. But that slurry clogs these little soft spongy pads, and to clean all that gunk out of the pads they use the pad conditioner, which is this little diamond tool I was talking about. They basically rub off the surface of their polishing pad. So anyway, just correction there. I was taking a shortcut. I actually did know how it worked, but I just told you a pad conditioner's planarized silicon. They clean up the pads that planarized silicon. So it's a two-step process, actually it's a multi-step process.

Other questions? Okay, I'd rather have questions, then — I've heard my lecture before, it's sort of boring to me anyway, okay. So no other questions. Um, this is what we're going to talk about, some later — uh, a little bit later this morning. So the theme from, uh, well I can use it, leave it up there, from Thursday is: structural materials are used in very large volumes. I told you about the six elements in the periodic table that are used in more than a billion tons a year. A billion tons a year, even if you're charging a penny a pound, figure it out, it's big money, okay. We're talking trillion-dollar industries. Um, and in fact it turns out some of these industries are larger than the semiconductor industry. And you can multiply it out — my future of metals paper talked about some of that.

Because of that, cost is crucial in these structural materials industries. When you get down to steel or aluminum, um, automotive companies, when they're using huge amounts of materials to make billions, or, well, you know, tens of millions of cars, um, they will kill for a tenth of a cent per pound, okay, on your structural material. So whereas — you're working on space shuttles and things like that, uh, yeah, they use beryllium and all kinds of things like that, okay, but it's a whole different uh cost business. Um, but you get down to just reclaiming asphalt, sometimes you run into some of these things, okay.

I think it was one of you over here asked the question of the abundance, and I said aluminum was like fifth most abundant element on Earth. And it is in the crust — it's number three, it's eight percent of the crust. But you don't find it in the oceans and you don't find it in the atmosphere. If it were in the ocean it would react to form aluminum oxide, it would precipitate out. And in fact the reason we find it in the Earth's crust, we find it as the mineral — they got, one day it may have been in the ocean, but it gets reacted out and forms all kinds of different minerals, okay.

Iron is abundant. Um, there's something interesting about iron, at least I think it's interesting, it's trivia. Um, where are the heavy elements made? I mean, we know where hydrogen and hel— hydrogen forms into helium. I mean, you know, let's, what — what's hydrogen? It's proton and electron. Somewhere out there in space, electron gets together with a proton and says okay let's make a hydrogen atom. And then inside a sun the hydrogen atoms form lithium and helium and all the light elements — not all the light elements, but — they don't make very heavy elements. I mean, I don't think you're making aluminum in the sun. Where do the, you know, in the grand scheme of, after the Big Bang, where the heavy elements come from? Student: Supernovas. Supernovas, yes, okay. Supernovas. And in fact, well, I probably can't find it, but quickly — supernova explosions are where you have enough energy to start having elements combined into other elements and things like that.

And, oh, maybe I can find it. Um, there are some elements of the ninety-three — they call ninety-three naturally occurring elements. How many occur on Earth? [Tom looks for his periodic table.] I was looking for my periodic table. Anybody know how many elements occur on Earth? Nope, not ninety-three. Ninety-one. The two that don't occur on Earth are francium — the very bottom of the first column, okay, all they've ever seen is — well, actually, yeah, francium occurs on Earth, but all they've ever seen is a spectral line, okay. It's such — and it has a very short radioactive half-life. The other one that doesn't occur on Earth — it didn't, until the last fifty or sixty years — is technetium. And it's right underneath vanadium, if I remember, on the periodic table. Technetium — the longest half-life is ten and a half hours.

But we actually use technetium. I've had people inject technetium in me, okay. We use technetium for under— if you ever have to, later in your life, you'll have to go undergo a stress test. You ever heard of a stress test? You go to the hospital and they make you run on a treadmill, and they put you in a room, they first they monitor everything. And after they made you uh run on the treadmill and your heart's pumping and everything, um, they will take live X-rays, and then they will inject you with technetium, and they will now take new live X-rays, and they image different things, because technetium is decomposing so fast that they can use the technetium itself for the radiograph, okay. And so they can actually sort of see where this fluid is going in your heart and stuff.

So it turns out, when I went there, turns out the radiologist at uh Mount Auburn Hospital happens to be an MIT physics undergrad. Uh, and so he got — when he found out I was an MIT prof, he, in fact, I had an ear infection — this was, I had my stress test six or eight years ago, and then I had um uh prostate cancer, and he injected me with radioactive titanium [technetium]. Uh, and that was five years ago. And then, um, then I had an ear infection back in April. And he saw me on the list of patients, you know, I was all swollen with a sinus infection and stuff. And so he came up — Dr. Abner came up to see me cuz he just likes to chat about MIT, what's going on at MIT. No, MIT alums like to know what's happening. Anyway, uh, when I visited them, they actually have, um, two reactors in North America that are producing technetium, and they have Learjets and things that fly this stuff around to different hospitals. And they might get a weekly dose, and their dose is decreasing by half over the week, right? It may be more frequent than that. There's a concern because the Canadians are about to shut down for economic reasons one of the reactors. And where are they going to get the technetium? But anyway. So not all ninety-three naturally occurring elements occur on Earth, but you can see technetium spectra in the spectral lines.

Um, which I guess leads me to another story. I'll just tell stories. I was looking at the schedule and realized I've already started talking about some of the things ahead of time. I had a case down in Kentucky of a, uh, a school that burned down. I think it was an elementary school, but who cares. The roofers were putting nails in the roof, and the question is, did they hit an electrical cable and cause an arc, and was that what started the fire that burned out the high school, or the school, whatever the school was. And uh, so we had some analysis come back from some laboratory in Kentucky. They did an SEM EDS analysis, and it found thirteen percent technetium. And I said, oh, we can solve this case — it's only worth $8 million, we'll just sell the nails! Because that much technetium is valuable. It's just — this is where again you got to question the data, right? I was telling you earlier today, this was garbage data. Case settled before I got to get up and tell the jury how stupid these other people were. Anyway.

Um, anyway, so aluminum is — this is the abundance on the Earth's crust. Well, you start looking at it, well, what I was going to tell you is iron is the most stable of all the elements. If you're a nuclear scientist and you start looking at the stability of the nuclei, iron 56 just happens to be the most stable of the naturally occurring elements, in terms of half-lives and everything else. Aluminum is much more abundant than iron, but the problem with aluminum is, it's expensive. And the reason it's expensive, I told you yesterday, is because it takes a lot of energy to produce aluminum. And this is all in this price and availability thing from Ashby that I handed out, these are actually his plots. Uh, aluminum is 300 [megajoules] per ton. Um, you could convert that to kilocalories per mole or kilojoules per mole or something. Uh, fairly close. Plastics, I think I told you yesterday, was about 100. Steel is only 50. So there's six times the cost of energy of aluminum [vs.] steel.

And it turns out it's about two and a half to five times the cost per pound for aluminum versus steel. And you could plot energy cost of refining it from its oxide or its sulfide or whatever the ore is, versus the selling price, and you'd find a very good correlation.

Why is copper r— [rising]? Uh, because copper ores — we don't have any good copper ores. The best copper ores in the old days were five percent copper. Uh, right now — anyone from Salt Lake City or Utah? Yes, okay. And what's the Bingham mine? It's a copper mine, right? You sometimes can see it flying into the Salt Lake airport, if they, depending on what route they take. I think it's just west of Salt Lake. Anyway, it's the world's largest open-pit mine. It's like a mile deep and it's between these mountains, and they just kind of have been just driving in circles and digging it deeper and deeper. The Bingham mine is about a half percent copper, which means to get a pound of metal you have to move 4,000 pounds of dirt, okay. And that cost money, and that's why it's rising to 500, okay.

That's, simple answer is, we've lost all our good — we've already mined out our best ores. Whereas steel ores, I mean if you go to what they have in India or what we had in the Mesabi Range, is like forty percent iron, okay. I mean, you can't get more than that — if it's pure iron oxide it's forty percent iron, okay, if you don't have any other impurities in there. And there were some steel deposits. Aluminum, you can get aluminum, um, uh, it's not cryolite, what is it — is Sam here? Sam's not here today. Sam's from the aluminum, he works for Novelis aluminum. What do — Student: Bauxite. Bauxite, yes, thank you. Bauxite is the ore they usually use. They also use laterite and some other things, but bauxite is a fairly pure aluminum hydroxide-oxide. And there are places like Jamaica, okay, Jamaica has two incomes: tourism and bauxite, okay, not much else. Um, most people go there for the uh tourism, not for the bauxite. Uh, so aluminum we have some good ores, steel we have some good ores, zinc, it turns out we have good ores, but copper, there's not really good ores. Copper is a semi— in some ways almost a semi-precious metal.

Student: So the energy rise for copper is attributed to the fact that you just have to move that much material, it's not the actual extraction process that's very intensive? Yeah, I mean, copper actually — the energy cost of copper, it binds oxygen and sulfur less strongly than iron does. Not much less, okay, it's about the same as iron. But the bigger thing is just moving the rock, and grinding it and crushing it and extracting it. And so they have things like — they use leaching just like we do fracking to get the oil and stuff out. They actually send acids down the hole. They don't mine rock all the time, because of this problem of mining rock all the time, um, and having, making a big hole. They actually, you know, build a loop and send acids and other things down and bring up the liquor with the copper dissolved — dissolve it in situ. Now, a lot of people in Arizona don't particularly like that, okay, because environmentally it's not necessarily the greatest. I mean, you want to talk about — I'd rather have fracking in my backyard than a copper leaching mine, okay. Um, but in any case, yes, the way we extract copper at the mine is a lot different than we do steel or aluminum because of the concentration of the ores. It's an entropy problem if you want to think of it that way, okay. Copper is more disperse, right?

So you, let, you just call today a recitation and forget the lecture, that's — but I'd rather do this, okay. Other questions? Yep.

Student: So if you were going to invest, you wouldn't invest in gold, you'd invest in copper? Well, gold — a lot of the gold comes from copper, okay. Um, gold — so there's an interesting correlation, which I probably will cover later in this course. But if copper costs $6 a pound or something like that right now, and we call that a unit of 1x, okay, and we look at silver — we usually talk about — and copper, silver, and gold — if I had my periodic table you'll see that's it, right, they're right above one another, or below — silver, we usually talk about ounces, we sell it by the ounce. And I don't know what it's going for now, was $15 an ounce. I mean it fluctuates between 10 and 20, except when the — what's their name — Brothers down in Texas tried to corner the — pardon me, the C— no, not the Kochs, there were some guys down in Texas who tried to corner the world silver market and they were putting billions of dollars into buying up all the silver, but anyway. Uh, this was fifteen years ago, twenty years ago. 100x. And gold — guess what gold is — $1,000x. Order of magnitude. There's a factor of 100 in each one of those.

Silver and gold are byproducts of copper mining. Now, we also mine just for gold, and we also have a few just for silver mines, but I would bet — I don't know exactly — but I would bet that half of the silver and gold come from copper mining as a byproduct. In fact, sometimes if you have a rich copper ore, you might make as much money off the silver and gold as you do off the copper, okay. But again, they're precious because they're not in very high concentrations in the Earth's crust, okay, they're very disperse.

And I don't, I mean, Ashby wrote that in 1980. Um, I don't think he was right. I think it hasn't risen to 500. Because, hey, who wants to start paying 500, five times as much for their copper? People have been innovative and have come up with different ways of getting things out, they found new ore deposits. I mean, the oil we had in 1980 is not the oil we have, you know, thirty years later.

Does anybody know why — um, the deposits for oil and things — we've been predicting we're going to run out of oil for the last fifty, in twenty years for the last fifty years. Does anyone know why there's a twenty-year time constant? Student: Have we been able to — yeah, but it turns out that's part of it. But the real reason is, once you have enough proven reserves to be forty years ahead, it makes no economic sense to invest a dollar in finding reserves that you're not going to dig out of the ground until forty years later. If you look at the time value of money — this is another economics lesson, okay. You look at the time value of money, we only explore for minerals and oil for twenty years ahead. That's all a financial analyst would tell you. You know, you'll get hammered by Wall Street if you have a hundred-year reserve. Now, if you by chance happen to find a hundred-year reserve — and for some minerals and things we do have hundred-year reserves — but that's just because we found this wonderful thing all at once and we don't use very much of whatever that thing is, and so we have a hundred-year reserve, that's why no one's going out to look for more.

But in oil you can look back for sixty or seventy years, we've had a twenty-year supply. And the reason we have a twenty-year supply is because they quit drilling when they get much above twenty years. And the reason is, you can do the present value calculation, um, if you do the present value — I wrote a paper on this once — if you invest a dollar today you have to recoup — I don't remember what the numbers were — at five percent rate of return — and that's not a very good internal rate of return for a business, right now they're looking at ten or fifteen percent — if you only had a five percent return, you'd have to generate $20 twenty years from now in order to recoup your $1 investment today, okay. It's just time value of money. And that's why we only have a twenty-year supply of oil, and we always have a twenty-year supply of oil, but because someone's going to go out there and search for something else.

And so Ashby was wrong, okay. It's not really rising. It may have doubled over that period of time, because there is a problem there and he's pointing about the problem, but he's assuming — he's assuming everything else in the world would stay the same. Then it would rise to 500 in ten or fifteen years. But not everything stays the same, okay. In fact, um, there's a very, well, fairly popular quote, um, by Harry Truman. Did anyone ever hear the quote from Harry Truman: "What I need is a one-armed economist"? And why did he say it? Does anyone know? Because the economists are always saying "on the one hand," and then "on the other hand." And so he wanted a one-armed economist because he was tired of hearing about the other hand. I often say what we need in the world is a one-armed metallurgist, okay. Because they often have different ways of looking at things.

Anyway, so energy cost is fundamental to the cost of the material, because that's the big thing about getting the material out of the ground. It wasn't always that way. In fact, I'll — since I know the industry, I'll talk about the steel industry, right? Surprise! I'd have Tom Eagar talk about the steel industry. Um, in fact, you know, I've already talked about some of the stuff, but I found ahead this morning, when I was looking ahead at my old lecture notes, I found a plot I put together last year, if I can find it quickly. Um, of my estimate of — we've talked about the hours per ton to make steel. [Tom searches for the plot.]

Here it is, the advantage of anything I write on yellow paper, so it's easier to find, okay. So I told you about — I estimated that in the old days it took, back in the days of Saugus Iron Works, it took um uh 3,000 hours per ton or something, in 1600, okay. Uh, so this is hours per ton on a log scale. Well, it's 3,000 or 4,000, anyway, there's a person-year per ton. And that might at sixty hours a week, that's 3,000 hours, okay. That's Saugus Iron Works, 1600s. Everything was hand-forged.

And then along — nothing really happened much, and there was some decrease to some point. I'm sure there's some increase in productivity as we got better water wheels and things like that and made bigger plants and economies of scale, but that probably didn't improve things by more than about a factor of two. Uh, then Henry Bessemer came along, and he showed us how to melt steel, as opposed to just make cast iron or do a solid-state reduction to make wrought iron — and that we then ran on sponge, which was a solid, which you then had to forge, okay. And then Andrew Carnegie came along with economies of scale. And I estimated that back in the 1850s or 1860s it was 100 hours per ton. So you had 20 person-hours [actually 100] to make a ton of steel. Andrew Carnegie probably dropped that some.

But then someone — what's your name? Yeah, Allison, okay. You might have been the one who asked this question the other day. Um, the be— what was it that made steel so much more productive? And I think I said it was the basic oxygen furnace and continuous casting, mini-mills. There's a lot of process technology. Basic oxygen furnace was first tried out in Linz, Austria in like 1958, went commercial by the mid-60s. Continuous casting — the Japanese were doing this in the late 60s. By 1975, if you didn't have a continuous caster, you were a dinosaur.

And I worked for a dinosaur company, because when I was at Bethlehem Steel, when you do ingot casting, you get about two-thirds yield, okay. Um, and the reason is, if I have an ingot — and we make ingots tapered, the mold looks like this and we pour the metal in, and when it solidifies it shrinks and we get a pipe. And you have to cut off that top part and send it back to remelt it, okay. So this is the only good part. Why do you make it tapered? Student: So it'll fall out. So it'll fall out, exactly, cuz thermal contraction, everything — cast iron doesn't contract as much as steel does, and so the steel, when it cools off, you know, tends to shrink from the mold, and they turn it over, now it'll fall out, okay, because it's got tapered sides.

But anyway, and maybe I didn't do this — maybe the pipe is only down to here, but anyway, you only get — we — there was actually an MIT grad from this department who was a big shot downtown at Bethlehem Steel when I was a young engineer, Mike Kimchuk, who was also the head of the MIT Educational Council. If you were from eastern Pennsylvania and you wanted to go to MIT, you'd have to interview with him. And his whole job — he was just under a vice president — was to try to improve yield from sixty-three percent. If he could get a one percent yield, Bethlehem Steel's profits were probably increased by thirty percent, okay. Just to get a one percent increase in what we call yield. And what's yield? Pounds melted to pounds shipped — okay, actually pounds shipped to pounds melted, okay.

Um, in fact there's another interesting story back in the uh 1930s. There was a guy named Eugene Grace who was chairman of Bethlehem Steel during the Depression. And all the steel companies were losing money back then — everybody was losing money back then — but Bethlehem Steel had six of the highest-paid executives in the United States. And that's because Eugene Grace and a number of his cronies were getting paid a bonus on the number of tons of steel poured, not the number of tons of steel sold, but the number poured. So boy, they kept those blast furnaces going and stuff, because they were getting a profit off of it, okay. Company's losing money, but, you know, they got paid more, Bethlehem produced. Um, so that was sort of the mindset in the steel industry uh back then.

But um, what's the yield? Well today we're, because of continuous casting, we're in the ninety-seven percent. Cuz continuous casting, you don't have to cut anything off, you just keep on pouring into a mold, which has — it's a copper, water-cooled copper mold, and it's slightly tapered, actually sort of curved like this, and you cast in here, you have your pipe, but you just keep on doing — as it comes out you cut it off, okay.

Actually have some pictures of that. This is jumping way ahead, but hey, it actually was going to come up at some point. I don't know if I have — well, I don't have a picture of the actual continuous casting, I'll show it to you another day. Um, cuz I don't think I have — oh well, I do have, I have this picture, if it works. Um, does that work without the glare? Let's see — with — if I put light on it, does it get rid of the glare, and make it worse? Probably going to make it worse. Oh, now I did something, okay.

So this is a — this is coming out of a continuous caster. So you got a building about eight stories tall, and you're pouring steel in at the top floor, and you have this big hot band of steel. In this case they're pouring I-beams, okay, uh, dog-bone shaped I-beams. And that, as the steel comes out, it's just like in fracking, drawing a horizontal well. In fracking, you can bend steel if you have a big enough radius. Uh, and it's hot like this. Um, in fracking, they just bend it while it's cold, they have, you know, a half-mile radius. Uh, when they're drilling in the ground, but here they have a couple of hundred-foot radius. And this big band of hot steel coming out, and it gets horizontal. And then they come along and they don't even use a fuel gas to do the oxygen cutting, all they do is just hit it with pure oxygen. And that's what those torches are, basically — they're just hitting with pure oxygen, and they're burning the iron. And so you can cut it off to length, send it to the rolling mill. Tremendous increases in productivity.

And what we saw starting in around 1980 or so, um, we started seeing, or actually starting in the 1960s we saw this tremendous drop, uh, drop in hours per ton, uh, to make steel. Uh, and I — that was part of my message of the last couple of days or day before. Productivity may not be everything, but as Krugman says, it's nearly everything, okay.

Um, so any other questions? Yeah, Student: Who actually originally developed the continuous casting? Well, continuous casting of steel wasn't the first continuous casting, okay. Um, we start out with smaller things. Copper was continuously cast on industrial scale probably back in the '40s and stuff. I actually have a series of lectures on casting technology, okay, and I talk about the continuous casting and stuff. Uh, but gold was one — because if you're casting gold, do you want sixty percent yield and throw away the top half? Turns out, in making gold, everything is the time value of money.

I mean, I started on the faculty in like 1976, right, 1978 I bought a house. I still live in Belmont. At that time it was three bedrooms and one and a half baths, it's now eight bedrooms and four and a half baths. So it's grown a little bit. I mean, it's still the same six-tenths-acre plot, but I had seven children. My wife thought all the children should have their own bedroom anyway, so we added on. Anyway, but to buy that house I had gotten a consulting job with a company down here in Attleboro, Massachusetts called Leach and Garner. And Leach and Garner, uh, they put me on a $5,000-a-year retainer. Now, you have to remember my salary as a faculty member at that time was probably about $20,000 a year. So this was, okay, a big deal. I could not purchase that house — about two years ago I found the letter in my old files, um, on, about my house. The bank made me get a letter from Mr. Leach and Mr. Garner saying that I had a retainer of $5,000 a year. Cuz I couldn't pay 85 — I didn't, couldn't, if I'm making $20,000 a year, I couldn't afford an $85,000 house. Well, now that house is worth over a million dollars in today's market. And people say, oh, you bought a house for $85,000 in Belmont. Well, yeah, but you know, the year before it sold for 70, you know. Uh, that was the price of things back then.

Um, but in any case, Leach and Garner — the Attleboro is sometimes called the gold capital of the world, okay. And that's where all the jewelry and stuff in the United States comes out of there. And the reason is because in 1899, Phil Leach's father and uh Steve Garner's father started Leach and Garner, which was a gold mill. They would take gold bullion, they would alloy it, they would cast it, and originally they cast it in ingots. But when I went down there in 1978, when a lot of steel companies didn't have continuous casters, they had two continuous casters for gold. And they would take two-inch round gold — or two-inch diameter round-carat gold.

I remember one time I was helping him with a casting problem. He said, come down, we're having some problems, and we're going to make a cast. And so this is like 1980 or so, um, I go down there and I find out this cast was just for me. And I found out there's $10 million in the pot, okay. It's just a little pot, like this, you know, whatever, big pot. This was $10 million in the pot. Oh, okay. Um, it turns out, everything in processing of gold is the time value of money. The interest charges on the gold — if you can get it through that plant in three days, from bullion to ship product, you're going to make a lot more money than if you do it in four days. And so it was great, I mean, they had everything from starting with the bullion, alloying it, casting it, rolling it, shaping it, forming it, making little, you know what they call findings, are the little earring posts and clasps and, you know, for watchbands or whatever. They made all those little components, they would sell them to the jewelry people, okay, all the people, Schwank and, you know, you name some jewelry supplier. But the mill that was providing all the raw materials was Leach and Garner, okay.

And that was great consulting job for me until the early 1990s, and it turns out they inherited — because they used to loan money to their suppliers, and one of their suppliers in Canada kind of came up short, and so they ended up owning this little gold foundry in Canada. And then those guys ended up uh coming up short of $10 million in cash. This — one of the problems in the gold business: people tend to steal, okay. I tell you, I mean, you can talk about security at the airport — I was seeing s— kind of security every time I went into the Leach and Garner plant, or out. Actually they didn't care about going in, but when you came out, you had to take your belt off, okay, you had to take your shoes off, you had to, you know, uh, back in the 1980s, okay, because people do almost anything to get gold out.

Um, my favorite story was not from Leach and Garner, it was from South Africa. One guy was working in one of the gold mines, and uh, he was in the chemistry lab. And it turns out he could bring his lunch in every morning, and he would have a hard-boiled egg every day, and he had a little container of salt in his lunch pail. And of course, he put it on his egg, and everybody knew that's why he had it. But he decided he would start taking, instead of taking salt out, he would take sodium chloride, he would take out gold chloride, which looks just like table salt. And he could produce it in the lab, and it wouldn't show up on the metal scanners because it wasn't metallic, okay. So they finally caught up with him, okay.

But there was another one, um, that was in Attleboro. It wasn't at Leach and Garner, it was another company that made things like the MIT brass rat. And they were uh, they were missing rings, okay. They keep pretty good inventory, they weigh the gold every time it changes hands from one station to another. I mean, if somebody's using it here, and someone else that far away is using it, there's going to be a piece of paperwork and a weighing goes on to make sure this person didn't skim some off the top. And there's tolerances on all this. Anyway, they were missing whole rings. And they found out that there's a river — a little stream runs right through the middle of Attleboro. People were throwing rings out the windows, and then on weekends they'd go dive for gold in the river. So the company bought the mineral rights from the town of Attleboro so they could police the river. So if you were doing a little diving in the river over the weekend, they could come up and check you, because they own the mineral rights.

But then they fired one of these guys, they caught him, and they were still missing some rings, and they could not figure out what it was. Turns out, he had a girlfriend, and she worked in the same plant. And just despite [to spite] the company, she was flushing them down the toilet, okay. But another story of Attleboro is, one guy, Warren Morris, was the head of engineering when I first went down there — it's time to quit, I know — um, but Warren was like great-grandson of Samuel F.B. Morse in the telegraph and stuff. He was from that family. But Warren first had started there about eight or ten years before. And Leach and Garner had been there since 1899. And they had a little drain in the floor of the casting shop. It was just regular little drain, like you see anywhere else, you know, little holes in it and stuff. He decided to go in and put a filter in the drain. He took $110,000 of gold out in the first week. It was just going down the drain. And so then they had to decide — and they did basically excavate the drain all the way — cuz it was a gold mine. It had been getting gold down there for decades. Anyway, so I'll see you on Monday. I'll be lecturing Monday and Tuesday. Uh, if you want to have another recitation, we can. I did cover a couple of slides, okay.