§1. Class business and the Hertz fellowship interview [00:00]
Where are we — anybody have any questions? I'm going to pass around the sign-up sheet for presentations. If your name is on here, it's because when I sent around the sign-up sheets you said you were available those days. We have forty people in the class. If your name is not anywhere on the sheet, please let me know, because then we don't know you're part of the course.
If I don't have to go out of town we'll finish before Halloween. If I have to go out of town we'll finish before Veterans Day, November 11th. I think the Armistice was at eleven o'clock — eleven, eleven, eighteen.
Does anybody have any questions? The schedules will come around. You should be signed up for one of the days you said. We'll start presentations next Wednesday. The presentation should be brief, no more than ten overheads, a focused subject. By focused I don't mean the automotive industry or the steel industry. Someone suggested they want to do Venetian glass — that sounds great to me. You can tell everybody what Venetian glass is and why it's different and what's interesting about it. The topic can be anything that's of interest to you.
I could tell you the story of when I interviewed for a Hertz fellowship. Hertz is the Hertz rental car fortune, and he was sort of a right-winger. Edward Teller was on his board of directors. The guy who interviewed me — if I tell someone at Lawrence Livermore National Lab that the guy who interviewed me was Lowell Wood, they say, "So you interviewed with Lowell Wood." Lowell Wood was Edward Teller's right-hand man. Edward Teller was the father of the hydrogen bomb — Hungarian, super right wing, advisor to Eisenhower and Kennedy.
I made the first round, and the second round was when they made the final decision. I was in the dean's conference room one evening interviewing with this guy, and frankly he was sort of a jerk. Everyone I've talked to since says, "Oh, you interviewed with him? Yeah, he was a jerk" — because that's how he ran Lawrence Livermore. He was the number two guy there at the time. He first asked me, who is Count Rumford? Anybody know who Count Rumford was? Neither did I. Count Rumford was a philanderer from Medford, Massachusetts who disproved the phlogiston theory of matter in the 1790s. He was reaming out gun barrels and he showed he could get an infinite amount of heat from friction, and therefore heat was not a constituent of the material — it was generated by the work. That was the early days of thermodynamics. Count Rumford was not actually a count. He went over to France so he could be a philanderer with some of the royal people and call himself Count Rumford. One of the deans here at MIT got really interested and wrote a book about him. I didn't have a clue.
So he asked another question: who unified the Prussian states in the 1870s? I said Metternich. At that time Henry Kissinger was an advisor to Nixon — early '70s. And I said Henry Kissinger did his doctoral thesis on Metternich. He was really impressed that I knew that. Big deal — I read it in some magazine. We got onto those questions because he said, "I've been asking you questions, why don't you pick the topic." I was sort of fed up with this jerk, and I said, well, how about cooking, because I like to cook. He said, "Well, I was thinking something more technical." I said okay, what about history. So we got onto Count Rumford and Metternich. The only thing that impressed him was that I knew Henry Kissinger's doctoral thesis topic. I didn't get the fellowship. I got an NSF fellowship instead.
So this sheet — please sign up if you haven't. We're scheduled through October 23rd. We've got three empty dates on these days and four empty slots on those days. Hopefully you can fit into some of this and we'll finish up early. Do something that's fun and interesting.
§2. Tempered and laminated glass [06:44]
We were talking about glass, and I mentioned that glass manufacturing productivity had improved markedly over the ages. I also mentioned glass is a brittle material, but it is a structural material. It has one very interesting property — it's transparent — and because it's transparent we want to use it in structures. The problem is, it actually corrodes. If you look at annealed glass, which comes out of the float glass process — float glass is produced virtually stress-free — and it breaks, it shatters, and it ends up with long sharp shards that have killed people when they walk through a glass door without realizing it. So now there are codes that you have to either use tempered glass or laminated glass. More often than not they use plastic in screen doors. Since you'd like to have something optically clear, we use glass in a number of different forms.
You have to anneal glass, because glass has low thermal conductivity, is brittle, and has small flaws that can cause it to crack and shatter with hazardous, potentially lethal shards. We have to do something to make it safer. What do they do on windshields, or side windows?
Student: Laminate it on the windshield, temper it on the side.
They laminate it, and on the side windows they temper it. They put a piece of plastic in between, they make a composite. If you've seen an accident where someone impacted their windshield, you'll see it breaks, but the plastic acts as a catcher's mitt and holds the stuff so it doesn't go shattering all over everywhere and cutting people.
Tempering is a process of heating the glass up to a range where it's in a viscous plastic state, and then cooling it rapidly — they're just blowing air on it. As you cool it rapidly, the top surface goes into tension because it's shrinking. With poor thermal conductivity the center is still warm and viscous, but because it's viscous it actually contracts while the outside is also contracting. The outside is going below what we call the strain point — the temperature at which glass will relax within about an hour. If you cool it rapidly in tens of seconds, you basically get the outside to harden up and the core to shrink. When you cool all the way down, the core has shrunk and everything gets back to uniform temperature, and the outside surface is in compression.
One of the rules of residual stresses due to thermal processes is that whatever stress you have at high temperatures will be reversed at low temperatures. At high temperatures you have tension on the surface, but at low temperatures you have compression on the surface, and you have to overcome those compressive stresses. If you look at the tempering plots, this is temperature through the thickness of the glass. You start cooling it down — quenching time in seconds, so this is over tens of seconds. The center cools down more slowly than the surface, and you set up these temperature gradients. When you get down to room temperature, you end up with a residual stress pattern that looks something like this — stress in megapascals. You can end up with compressive residual stresses on the order of 75 megapascals, which is not a whole lot. It's about 10 ksi. But 10 ksi compressive stress on the surface — as long as you don't break through that, you can bend the glass and put stresses of up to 10 ksi on there, whereas untempered glass would have a strength of about 1 ksi due to the notches from the float glass process.
Another thing we can do, since glass is an important material, is ion exchange. You take a glass — in this case they took a sodium aluminosilicate glass — and you do an ion exchange with potassium. You put it in a salt bath with potassium, and just because of entropy effects the sodium exchanges with the potassium. Potassium is a larger atom. You're doing this at relatively low temperatures, below any viscous diffusion, for hours. After four hours you can increase the strength of the glass from 140 megapascals — about 20 ksi — up to 400 megapascals. This is chemically tempered glass, where you get compressive residual stresses by exchanging large ions for small ions.
§3. The John Hancock Building and the depth of ion exchange [13:15]
The John Hancock building over here — when I was a student they were building it, and after they first built it, the winds came along, creating a suction force, and the windows were popping out of the building and falling to the ground.
Student: The Plywood Palace?
Yeah, probably the Plywood Palace. Someone who lived in the same house I was in, two years younger — the day after they put the top I-beam on, he snuck in, climbed up to the top and signed the top I-beam. He was sort of a risk taker. A couple of years later he died skydiving in Colorado when his parachute didn't open. But in any case, it's chemically tempered glass, because they needed a stronger glass — the building was flexing. They now have the building under a slight negative pressure, with big vents. And they took a whole floor and put a big weight in there. The swaying frequency of the building is several seconds, and it's a damper.
Student: Tuned mass damper?
Right, it's a seismic balance — not for earthquakes, but to keep the thing from bowing too much. They're reducing the vibration of the building. Most buildings just sway. About the same time I went to Chicago, went to the top of the Sears building. They have an observation deck, and you go in the restroom and you see the water sloshing in the toilet, because the building is sloshing. The John Hancock, they didn't want it to do that — they wanted to damp it seismically. They have a whole floor with a weight rotating in opposition to the wind motion. They spent a lot of money — hundreds of millions of dollars — to fix that building because they used glass as a structural material and ran into problems. Yes?
Student: [question about ion exchange depth]
No, it doesn't come down the whole profile. It's only the first two or three thousandths of an inch. This is typically a plate glass, a quarter of an inch thick, no thinner than an eighth of an inch. Float glass might be a sixteenth, an eighth, or 3/16. Plate glass is where you pour it on a steel plate and grind it and polish it — quarter inch to half inch or thicker. The ion exchange is relatively thin. You could scratch it with a glass cutter and not be beneath the chemically tempered compressive layer. But if you really got a good gouge in it — took a diamond and gouged — you'd get below that, and this'll shatter on you as you get into the tensile residual stresses on the inside. So they have to make it 50 microns deep, and that's why it takes hours to do. You put it in there, you'll get a very thin layer, but then that slab gets scratched, and now you've created a situation where you've stored energy in it and you release the spring and kapow. It's got to be a thick enough layer. But on a hard material like glass, it's hard to scratch it three thousandths deep. You can, but if you do you're in trouble.
I wish I had it, but Professor Uhlmann, who used to teach glass here when I was a student, had a piece of chemically tempered glass that was probably about eighteen inches long and bowed about three inches high, and he could put it on the table and just flatten it with his hand. He wasn't exceeding the compressive stresses in bending. This only works in bending, doesn't help in tension. But big flat sheets fail in bending. So it works. People have done a lot of things to make glass a useful material in spite of its brittleness. So all those things I said about how wonderful steel is because it has toughness — well, that's true, but people are clever enough to figure out how to make glass into something useful even though it's brittle.
§4. Transparent armor and the RPG problem [17:40]
Now, other stories. This is from protective armor systems. [Tom displays a transparent armor sample.] Three layers of glass with one layer of polycarbonate. Polycarbonate is a very tough plastic in its own right — about 10 ksi strength if you pull it slowly. They used to use it as a bullet stopper. Now there are copolymers — I just bought some Vitamixes for my children last night, and they use Eastman Chemical Tritan, a polyester copolymer. They used to make those out of polycarbonate, which is very tough and transparent. Now they make armor for bullets. This was an inch and a quarter, four layers adhesively bonded together. You can't have any bubbles in the adhesive — this has to go through an autoclave in a very controlled roller process. This will stop a 50 millimeter round. It does it by cracking the glass, absorbing the energy, hitting the next layer, cracking and shattering and absorbing the energy. Eventually it gets to the polycarbonate, and hopefully you've slowed it down enough.
That will not stop a rocket-propelled grenade — an RPG. However, they make things that are just under six inches thick that will stop an RPG. The president's limousine has five and three-quarter inch thick material made by these folks. It looks like a nice big car, but if you get in it, it's pretty scrunched inside. It is a big car, but the doors are about ten inches thick. It looks like a regular old Cadillac or Lincoln, but it's been modified. You can buy one yourself. A lot of sheikhs and other people in the Middle East buy these, because they have some enemies who would just as soon hit them with an RPG. Humvees might have something like that. The big things in Iraq —
Student: MRAPs.
MRAPs. The MRAPs were these sixty-ton personnel carriers that replaced the Humvees, because of the V-shape on the bottom. And it turns out six layers of glass as the armor.
What happened — I can't tell you everything because some of it's classified — but I used to sit on the committee that reviewed the materials and mechanics group at Aberdeen Proving Ground. This used to be the materials group out at Watertown — Watertown Arsenal Mall now. Since the 1860s that was the U.S. Army's materials lab. They used to make cannonballs there in the 1860s, in the mall area. Some of those buildings were almost original. They moved to Aberdeen Proving Ground, where the Army does all their ballistic studies, and built them a beautiful new building.
What happened was, the Secretary of the Army — we were losing a lot of soldiers in Iraq — came and said, look, you're our people who are supposed to be developing armor for us. We want you to solve this problem. Within eighteen months they solved it. Soldiers had been being blown up in their Humvees, but within eighteen months we quit losing soldiers being blown up in their Humvees. One guy on our committee was a captain who'd done two tours in the Middle East, and he said the most important thing was the eyes of the soldier. They go down a road every day, and if they saw anything change on the road, or within fifty yards of the road, they would stop and start looking for some buried whatever it was. These people weren't very sophisticated — they didn't have remote control. They'd be hiding twenty-five yards off the road ready to trigger the device when someone came by. The best thing was the eyes of the soldier. But in fact they developed things — they didn't know what armor was going to be best, so they tried steel, titanium, aluminum, glass. They had reasons to believe glass might be good. They just blew them up down there at Aberdeen, all within a month or two, and they found out glass was the best.
I can't tell you why, because that's classified, but glass worked better than anything else at repelling the rocket. These are shaped charges that send off a stream of molten copper that will just cut through steel and destroy aluminum. Sonic velocities of this massive dense copper that will cut through steel. A shaped charge that shoots out a thirty-six inch piece of copper will penetrate thirty-six inches of steel. I've seen the steel that was penetrated by one of these things. So you can't put thirty-six inches of steel underneath the V-shaped MRAP. You put six sheets of half-inch, three-quarter-inch glass. I can tell you that because any soldier out there can look and see that he's being protected by glass.
So hardness is an important property in armor protection. Glass is harder than steel. It's not stronger, because flaws will make it break. But if you cause it to break, you're absorbing energy. Maybe only one time, but you only need to do it one time. If you get hit by an RPG, you're probably not going to use that MRAP again anyway. Yes?
§5. Shaped charges, IEDs, and the armor-penetration race [24:36]
Student: Same kind of material used for IED protection and the underbelly of the Cougars?
Yeah. Improvised explosive devices are nothing more than shaped charges. Let me give you a little of the history. When they first went into Iraq — President George Bush's time — they would use a four-inch black iron pipe, just a steel pipe four inches in diameter. They'd put a bottom on it — machined or welded. Then you put the explosive in, and you have to have a piece of copper that's sort of dish-shaped, like a saucer. When you set off the charge, the copper folds outward — remember little clickers you had as a kid that you could click and they'd buckle? The copper bows outward, and as it does, the stress causes enough frictional heating that the copper melts. You end up getting what's called a shaped charge.
The U.S. Navy discovered this in the 1880s at the Brooklyn Navy Yard, by mistake, while they were doing some explosives work, and they found it would cut through steel. I wish I had one of Doc Edgerton's pictures of a milk drop splash — you have a drop of water come down in a little dish and it goes out in a wave and comes back as a wave, and when the wave comes back you get this big spike of milk. Basically all the copper melts from the energy of the explosion and the mechanical shape — it's called a shaped charge — and you form a jet of molten copper. That will eat through steel of equal thickness to the length of the jet, approximately.
RPGs have fins on the back so they'll fly. But the reason they have the shape is because they have a V-shaped piece of copper and the explosives down below, and it implodes and makes this big jet that will go right through a Humvee. Four-inch diameter — I think this was 200 kilojoules of energy. They got a little better and went to eight inches. The energy goes as the cube, so they went to 1.6 megajoules. That blew a Humvee into smithereens. People were dying in Humvees, so the Secretary of the Army came and said, you've got to do something.
Within eighteen months they designed and developed armor for the MRAP vehicles. The seats in the MRAP are sort of interesting — I got to sit in one. They're big shock absorbers, because if you get a 1.6 megajoule IED under you, even though it's shaped like a V for the armor to deflect the blast, it will still lift that sixty-thousand-pound vehicle four or five feet in the air, and you're taking off like a rocket too. So you have to be strapped in, with these big springs, so that you don't get a serious spine injury or hit your head on the ceiling and end up dying from a concussion. You don't want to be hit while you're in one of these, but it won't blow the thing to smithereens.
And of course, now that we've moved out of Iraq, what are they doing with the MRAPs? We read in the papers, they're giving them to the local National Guard. Where was it — Missouri — they come in after the riots, after the policeman killed the teenager, with their MRAP. So now the rioters in Missouri better have their eight-inch-diameter improvised explosive devices, and the police will be safe. Isn't it great.
To machine the copper disc you rely on symmetry — you can't just use any old machine shop, and there is no machine shop in Iraq that can make those. I can't tell you what country was giving these to the insurgents, because that's classified, but we know where it came from and who's doing the machining. All you had to do was take a little disc of copper, and pipe — they've got plenty of pipe in Iraq, that's what they use for the oil business. They just went from four inch to eight inch and they could wipe out Humvees.
The technology for weapons is sort of amazing. In 1985, they were first developing some of these really good RPGs. One of my students who was in the Army at the time, over at Watertown Arsenal doing his doctoral thesis, told me a story. They lined up three tanks on a battlefield — probably down at Aberdeen — and they shot an RPG, and it went through six layers of armor, through both sides of all three tanks. Apparently the general tossed his cookies right there on the battlefield. He just got sick.
Within a year they had developed new types of armor that could not be punctured. In 1984 there was no armor that could not be penetrated; by 1986 there was no propellant that could penetrate the modern armor. The modern armor was a composite — layered with metal and ceramic. It's actually a metallurgical reaction that allows the copper to go through steel — it's an embrittlement reaction of copper and steel. You put ceramic in between, you don't embrittle ceramic with copper, so you can stop it.
The other thing you can do is — these penetrators, I talked about depleted uranium. They might be a sabot, thirty-six inches long, half or three-quarter inch diameter depleted uranium rod shot out of a shoulder-mounted weapon or a tank weapon. When it hits, it gets stressed to close to 400,000 pounds per square inch. If you have on the surface of your armor an explosive that ignites and sends a shock wave through that rod, you'll send a stress wave through it that reflects off the other end, and the whole rod will shatter. So they have what they call explosive armor — armor with an explosive on the top surface.
That works great for the first shot. So what did I see in 1985? They were doing supercomputer modeling, and they had sabots with a twelve-inch piece that would set off the shock wave, and another twenty-four-inch piece right behind it that would go through the hole, because they'd already set off the explosive. Modeling this on supercomputers, finding ways to develop armor that couldn't be penetrated and penetrators that couldn't be defeated by armor. It's sort of gory, but it's interesting to see the technology that goes into this stuff. None of what I just told you is classified — you can find it in books. But there's plenty of stuff that is classified that I can't tell you about.
§6. Touchscreens, fusion-drawn glass, and humidity attack [33:42]
I want to tell you about your computer screens. They're glass. They're actually a composite. I showed you LED screens — they might be six or seven layers. You've got a layer of plastic on the very outside — nowadays we actually have glass on the outside because you have indium tin oxide, this transparent semiconductor that allows you to use a touch screen. It's very interesting to have a transparent material that conducts electricity. Ordinarily if it's transparent that means it doesn't absorb photons in the visible range, which is several electron volts. But if it conducts electrons, it's got to have some band gaps — it doesn't have very big band gaps. So how can you be transparent in one case, which means you have to have a band gap in the visible range, and then conduct electrons at the same time? Indium tin oxide is about the only material that can do this. I've never looked into the band gap structure, but it can, and that's why we can use automatic teller machines and touch screens on iPads.
But what about the glass? The glass is 0.3 millimeters thick — twelve thousandths of an inch — and you can get it in sheets 1.8 meters by 1.6 meters wide. You can do big-screen TVs, laminate it with all this other stuff, and that gives it some strength. But glass can become brittle within hours by being attacked by humidity. In the float glass process, the top air surface — where it just uses gravity and air to give you a flat surface — gives you an extremely pure flat surface, and if they use the right composition glass it won't be attacked by moisture. But the side on the tin bath — although they keep the tin bath from oxidizing, you shouldn't have big defects — if you go in there with an atomic force microscope and measure the flatness, the tin side is nowhere near as flat and as perfect as the air side.
So what do they do for these very thin things? They want both sides to be float glass. I've seen the Corning plant that does this, but I had to sign a confidentiality agreement. They basically — I won't tell you how they do it, but they take two sheets of float glass and they weld them together. The outside surface on both sides is float glass, and the bad tin surfaces are buried as they weld the glass together. I can't tell you the details. Interesting technology — about a quarter-million-dollar plant. Yes?
Student: [question about glass structure]
If you look at the structure of glass, you've got silica atoms with tetrahedral bonds in lots of different orientations. They're not perfect tetrahedra because glass is amorphous, not crystalline. You've got silicons with oxygens in between — a silicon-oxygen-silicon bond, which is one of the strongest bonds. It's potentially stronger energetically than even a carbon-carbon bond in diamond. But pure quartz is not as strong, because of fracture properties, as diamond. To break the network up, you have some sodium — you put in some Na₂O and some alumina. Some of those sodium atoms will hit the surface, and when sodium sees moisture it might form sodium hydroxide, might form a little pit. It won't keep forming the pit forever, and depending on the composition and network structure, the silica is fairly immune to moisture corrosion, but the sodium and potassium are not. You form hydroxides.
You might say, that's only a few angstroms deep. Yeah, well, go look at the fracture toughness of glass and the Ashby plot — critical flaw size, 10⁻⁴ millimeters. All you have to have is a few sodiums together and you get a pit a little deeper. Not very deep — you can't see it in a regular microscope. If you go in polished glass, like plate glass, you'll never get the strength you get from float glass in bending, because the whole polishing process leaves little scratches all over the surface, microscopic. You can't see them — they're smaller than the wavelength of light, hundreds of nanometers. But if you go in there with a fine enough microscope, you'll see notches, and moisture can do that. It doesn't take long. Yes?
Student: So if you use ion exchange you don't really change the sensitivity to humidity?
No, you don't. Well, you can, but now you've ended up with a compressive layer that's hundreds of microns deep, not angstroms — maybe 500 or a thousand microns deep. So what do I care about little small flaws? They're still in a compressive state.
§7. Glass strength, the brittleness limit, and Corian countertops [40:00]
[Tom locates a reference handout.] Annealed glass, not a safety glazing material, has very low internal stress, is brittle, and can be easily broken. You get large sharp dangerous shards. Unlaminated glass cannot be used in motor vehicles. Tempered glass has very high internal stresses — over distances of microns or tens of microns, even a hundred microns — that increase impact resistance. Upon breakage, small blunt particles result.
Glass is a brittle material — it fails instantaneously and completely. The ultimate strength of glass is not so much a function of its chemistry, although the potassium and sodium matter. With no residual stresses, like in float glass, I get critical flaw sizes. If I can introduce compressive stresses, or I can protect it by lamination — pristine glass fibers, those without flaws, have exhibited tensile strength of 14 gigapascals, 2000 ksi. That's when they're brand new and haven't been attacked. However, ordinary everyday glass is never in pristine condition. The surfaces of flat glass are cumulatively damaged on a microscopic scale during the initial float glass manufacturing process as well as subsequent operations — handling, cutting, grinding, tempering. You only need flaws on the order of tens of nanometers to significantly weaken that from its 2000 ksi strength level.
That's four times the best steel wire we can come up with. If we get iron single-crystal whiskers with a screw dislocation down the middle, we'll get steel with 2000 ksi strength. The inherent strength of the bonds is fantastic, but we never achieve it in metals because dislocations move before we ever get to those strength levels. In brittle materials like glass, they shatter by fracture mechanics before we ever get there, unless we take nice new fibers and encase them in protective epoxy so they don't get attacked. You can make composites out of glass, and we do have transparent glass structural materials.
Here's a lesson on — eventually I'm going to get to how to pick your kitchen countertops. Transparent structural materials: polycarbonate — that stuff between glass layers in transparent armor — acrylic (stretched acrylic is basically polymethyl methacrylate, Plexiglas), or chemically tempered glass. Tensile strengths are very low for the plastics, very high for the glass. Modulus of elasticity — glass is very stiff. That plastic window, you can flex it, you can lean against it and pop the window out. Design stress for the plastics — they give you three or four ksi, that's nothing compared to a metal. They won't even give you a design stress for glass, because it's brittle — you'd better not put it in tension. It's fine in compression, but don't let it see tension. Just like Portland cement — about 1 ksi tensile strength, but the codes give it zero in tension. You have to put steel in there as reinforcement.
Glass-ceramic compared to marble and granite as building materials. How many of you have a home where your parents have marble or granite countertops? All right. Now, you want to know what's wrong with that? Glass-ceramic versus marble and granite — you can get a bending strength three times as great in the glass-ceramic. Water absorption — glass-ceramic doesn't absorb water; marble and granite do. In fact, you can change the color of granite all the way through by ion exchange, half an inch through. You can take white granite and turn it into red granite. Red granite sells for a lot more. Marble and granite are beautiful. But the water absorption — do you really want to be fixing food on something that's absorbing water?
But that's not the real problem. Acid resistance and alkaline resistance. Marble — acid resistance, one percent sulfuric acid, gets a 10.3 rating, which means it's lousy. Glass-ceramic is hardly attacked. Granite is actually pretty good, because it's a mixture of feldspar and silica, but it's got pores. What's wrong with marble? Why is marble attacked by acid? Anybody know the chemical composition of marble? Calcium carbonate. What happens to calcium carbonate in an acid? It decomposes — to carbon dioxide and calcium hydroxide. It sort of seals the surface, but it dulls it. So don't spill lemon juice or vinegar on your marble countertops. They might be beautiful now — they won't be beautiful once you get a little acid on them. They'll lose their polish. Are you going to fix food on this? Well, no — you shouldn't. When they bought it, I'll bet they said, oh, don't get this or that on your marble countertops. So nowadays, rather than marble and granite, they're selling glass-ceramic countertops — much better material selection for food production. Yes?
Student: Don't they seal them?
They do seal them — they put a plastic coating on it, and they're beautiful. One of my favorite sets of elevators in the country is at the Williams Tower, just west of Houston. It's the tallest building on the Beltway at the nine o'clock position, a seventy-story building, and each elevator has a different type of marble from Europe. It's got twelve elevators with twelve different marbles in different colors. Yes, Johnny, you're right, they do seal them. But if you scratch through the sealant, you run into problems.
Student: [follow-up]
I still have Corian countertops, which is a polymer DuPont was selling in the mid-'80s, and it's about time to redo the kitchen. I cooked or had a hot frying pan — one of the griddles — and put it on there without a reflective surface underneath, and heated it up. By thermal expansion it was in compression, but when it cooled back down we got cracks. It turns out DuPont came out with an improved Corian about three years later. We were an early adopter. You learn all kinds of things.
§8. Aluminum: from precious metal to commodity [49:17]
Any questions about any of that? I took a little bit longer as I usually do. I was going to talk about aluminum, and maybe I can just quickly do some of the history of aluminum. Aluminum was once more valuable than gold. In 1850, a baby rattle from the royal family in France was made out of aluminum. It's light — babies aren't that strong. What's on top of the Washington Monument, built in the 1880s? A piece of aluminum. It's a lightning rod, intended to attract lightning, and it's aluminum.
Aluminum in 1852 was $545 a pound. If you divide by sixteen, it was more expensive than gold, which was about $25 an ounce. The price came down. By 1854, Napoleon thought this would be great armor for his soldiers, so he funded the development of aluminum in France. At the time they were using either sodium or potassium to reduce aluminum chloride — still a very intensive process. It wasn't until the 1880s in the United States, the Hall process dropped the price — 1886 I think it was. Hall cut it in half. Remember what happens if you cut the price of a structural material in half? You go up four times in the quantity used. They kept dropping the price. By 1897 it was thirty-six cents a pound.
Aluminum had tremendous growth. In 1889 they sold 7,000 pounds. By 1909 they were selling fifteen thousand tons. The real growth of aluminum came in World War II. This is the world's total for aluminum production — Germany was making more than we were in 1939. By the end of the war we'd outstripped them. The Germans hadn't been able to increase much, but we went from 148 to 695. All of a sudden Alcoa, which owned most of the business in the United States, was making 2.6 million tons of aluminum. Here's Andrew Mellon's aluminum Pierce-Arrow automobile. They were making horse carriages out of aluminum in New Kensington, Pennsylvania in 1909.
[Tom advances slides.] This is what Alcoa said the Pittsburgh Reduction Company looked like — this idyllic picture. And here's the actual picture. It's not quite as idyllic. There's a lot of pollution in the smelting and refining of aluminum.
There's one plot in here about Alcoa's command of the market. Alcoa had a monopoly. They had 100% of the market for bauxite in the world. They were a true monopoly in the aluminum business. George Kenney, who was Joel Clark's first doctoral student, did his doctoral thesis on aluminum and magnesium and showed pretty convincingly that Alcoa and Dow had fixed prices after World War II. Alcoa said, we won't go into the magnesium business if you stay out of the aluminum business. So Dow took over magnesium and had a monopoly there. You could match their prices over time from 1945 up until 1975 when George did his thesis.
When we hear these terrible things about other countries doing terrible things, just remember — we do the same thing, we just don't advertise it the same way. Any questions? You can have Monday off. Today is Tuesday. We've got class Wednesday and Thursday. I'll be here on Thursday — we'll finish up the rest of the periodic table for structural materials. Dr. Belmar will be here next Tuesday for the last of our two sets of lectures. If you haven't signed up, please do. If anybody has questions about the presentations, come see me.