§1. The steel industry and the future-of-metals argument [00:03]
Where we are already in steel — at least in North America — we don't use a lot of virgin ore. They still do in other parts of the world, like China. Australia and Brazil provide ore, but that's partly because the use of steel is still growing.
I wrote an article on the future of metals in 1991, because I was sick of hearing that the ceramists at the time were the big thing, and the high-temperature superconductor people. It came out in the Welding Journal — probably the only people who had the guts to publish it.
My point was: I'd hear them talking about how fine ceramics was going to become a $30 billion a year industry. At the time, the steel industry was a $500 billion industry. Fine ceramics was $5 billion, with an estimated annual growth rate of twenty percent. Composites — which were big because of Star Wars in the mid-'80s, everybody was talking about composites — were $15 billion, growing at ten percent a year. Semiconductors were $100 billion, growing at five percent a year. Everybody was telling me how wonderful silicon was, and fine ceramics and composites, and people were starting to say metals were dead. And I said, no, they're not.
I was sort of a voice crying in the wilderness. Yes, steel only grows at two percent a year in volume, but when you've got a $500 billion base, the growth in steel is more than the growth of all of those others combined, in dollar volume. And it turns out there were some people who understood that. There's an Indian guy named [Mittal], and he went around in the late '80s and early '90s buying up every steel company that all the boys on Wall Street wanted to get rid of. He could buy them for two cents on the dollar. They had environmental problems and legacy obligations, but he worked some of those things out in the sales. All the investors wanted to get rid of the steel stocks back then, and this guy [Mittal] in India went around buying them up. Today some of the world's largest steel companies are ArcelorMittal, which is a French company that merged, and British Steel is now part of [Mittal].
Part of the best part of Bethlehem Steel, the Burns Harbor plant, was the last large-scale steel plant built by a company in the world. Bethlehem decided to build a brand-new greenfield steel plant in 1965, and it almost bankrupted them. At the same time they decided to build the Homer Research Labs for $600 million, which is where I was hired in 1974. A $600 million research lab in 1966 — it was a Taj Mahal of research labs.
Bethlehem was the second-largest steel company in the world at the time, after U.S. Steel. And how did U.S. Steel and Bethlehem get to be the first and second largest, with so much market clout that in 1962, when U.S. Steel wanted to raise the price of steel, President Kennedy stood them down and made them roll back the prices? The steel guys say that's when we lost our profitability. But the reason: steel was like oil is today. The world's economy depended on the price of steel. Before the 1973 oil embargo, steel was the commodity that controlled the world's economy. U.S. Steel and Bethlehem controlled 75 percent of the world steel production in 1945. Why? Because we had bombed out all the rest of the capacity. What a great thing — you have a war, you bomb out your competition.
By 1975, when I was working at Bethlehem Steel, U.S. Steel and Bethlehem had between them over forty percent of the world's capacity. By '75 — only thirty years later — we had 25 percent of the world's steelmaking capacity.
I remember going to one of what they called the loop course — all the 500 new college grads that had been hired by Bethlehem that year. The vice president of finance got up and told us that it cost Bethlehem nothing to make steel, because we had 1911 coke ovens and 1912 blast furnaces and they were fully depreciated. And I was stupid enough in the question-and-answer to raise my hand among these 500 new hires and ask: how come it doesn't cost anything to make steel because you've got 1911 and 1912 equipment? Oh, you don't understand accounting. Well, a year and a half later I took an accounting course at Lehigh, and I found — he's right, I didn't understand accounting, but I also learned that he didn't understand accounting. Using old equipment doesn't make you more efficient. Productivity is what's important, not whether your equipment's depreciated. But he thought depreciation — that's why Bethlehem Steel no longer exists. That's why after 13 months I said, where do I want to be five years from now, and the clear answer was, not here. And that's when I came back to MIT. Now I say where do I want to be five years from now, and actually I probably still want to be here, but I don't want to deal with the administration. That's another story.
There's this article on the future of metals, and it shocked people when I told them the steel industry used five times what the silicon industry used in 1991. It's still five times the silicon industry. It's bigger. But so is the silicon industry. We know more about the science of iron and steel than we do about any other material in the world. We know a lot more about steel than we do about silicon. This is sort of like my little thing on corrosion: you're not going to be a corrosion expert after all this, and you're not really even going to be a steel expert after all of this. Hopefully you have an idea of some of the things that go into trying to weld some materials, and in particular steel is the one you're going to deal with the most.
§2. Productivity and the lesson of necessity [07:29]
That 21st century research thing — the Defense Department wanted to know what type of material they were going to use. The Army did a big study in the late '90s, because you had managers in government and in industry and the Wall Street Journal and everything saying the steel companies are dying. Another part of my future-of-metals paper: did you know the productivity of manufacturing in the United States went up like four percent a year in the 1980s, when the Japanese were killing us in the marketplace? We still had the highest productivity in the world in the United States, per man-hour. The steel industry was double the manufacturing average. And the industry that was three times the average of manufacturing was the mining industry. Everybody said these industries were terrible because they were losing jobs. Well, when you have a doubling of productivity in ten years, and not a super increase in consumption, guess what happens to jobs — they decrease. You don't have to be a rocket scientist economist to know that if your consumption is constant and your productivity doubles, your employment goes down by a factor of two.
When I was in the steel industry in 1975, six person-hours per ton. By the mid-'90s it was half an hour per ton — six times better in twenty years. Today it's less than twenty minutes a ton to make steel. If you go up to Saugus Ironworks, it was about 4,000 hours per ton — about one or two person-years of effort, cutting down the trees, making the charcoal, smelting it, all to get a lousy ton of cast iron, it wouldn't even be steel. So productivity in these industries has been tremendous over this period of time. Why? Necessity is the mother of invention. When you see that you're going to be extinct, you're going to lose your company, you start being willing to innovate.
There's a lesson here for the U.S. Navy. When was the last time you really innovated — sorry, that's another story, I won't go into that. There's a little conservatism, particularly in the submarine business, about innovation. But actually we had a very good presentation about innovating in titanium yesterday, right. So you have to be careful about innovation — it can be a problem.
§3. The iron-carbon phase diagram and the lost beta iron [09:43]
Anyway, here is the phase diagram for steel. Steel, in the most basic form, is just an alloy of iron and carbon. So here's iron, 100 percent, going up in temperature, and here you're adding carbon to it, up to — this thing goes out to about twelve percent carbon. This is molten liquid up here. This is face-centered cubic, non-magnetic steel up here. This is body-centered cubic down here. It turns out, above about six hundred something degrees, it's actually non-magnetic body-centered cubic, because you're above the Curie temperature.
They talked about this as being alpha iron or ferrite. You see the alpha iron right there. This is gamma iron. What happened to beta iron? I used to teach this to my sophomores. There was a beta iron — beta iron was the non-magnetic body-centered cubic form — until the 1920s and X-ray diffraction came along. And then Marie Curie [possibly Pierre Curie / Pierre Weiss — Tom said Marie Curie] came along and explained that there is a phase transformation, but it's what we call a second-order phase transition, when it goes from ferromagnetic to non-ferromagnetic. But you don't have to worry about that. There was a beta iron once.
If I look at that same diagram, I can plot on it where different types of materials are. So for thousands of years we used cast iron, until Henry Bessemer. That's because we'd put charcoal and limestone and iron ore in this big furnace, and we actually could melt it, because carbon drops the melting point dramatically, to below 1200 degrees centigrade. I can melt that with a charcoal fire. I can't melt steel — above 1500 degrees centigrade — until Henry Bessemer taught me how to preheat the air coming in. But I could melt cast iron. So at Saugus Ironworks we could melt cast iron, and we could make little frying pans and kettles and things like that. They used to hand-forge nails back in the old days. Four hundred years ago they mostly used wooden pegs to build a house, because most people couldn't afford nails. In fact, when the house got old and rotten, they sometimes would burn it down and then rake through the ashes to recover the nails. At one person-year per ton, there was significant labor content to be gained by recovering the nails in an old house.
So these are cast irons, at above three or three and a half percent carbon, and they melt at low temperatures. Then at very high carbon we have something called tool steels. We have high-carbon steels. We also have medium-carbon steels in here, but I didn't want to crowd it. And then in yellow, with the yellow highlighter, we have low-carbon steel. This is what we weld, low-carbon steels, because carbon — although it strengthens steels — lowers the melting point. It can do really bad things when you can't control the solidification.
§4. Heat treatment, hardness, and the kinds of steel [13:26]
Steel is a very important material, just like clay was a very important material thousands of years ago, because you could take clay and shape it, make complex shapes out of it. They learned to make potter's wheels, get your hands all muddy, play in the mud as an adult. And then you could fire it, burn off all that water, and turn it into pottery, which is a brittle ceramic. You can do the same thing to steel. I can take a piece of steel and harden it, or I can leave it as — this is actually a medium-carbon steel, 1045. This is probably about eight tenths of a percent carbon, typically, right at the lowest transformation temperature. And I can make one of them six or seven times harder than the other. I have a file, and I can machine the soft stuff. I can heat it up and deform it. I can heat this up and deform it, but it's no longer going to be hard and strong.
One of the great things about steel — I wrote my humanities paper as a junior on the hardening of steel, taught by one of the world's great metallurgists, who came here from the University of Chicago when he retired at 65 and became an institute professor in humanities, because he had also studied the archaeology of materials — Cyril Smith. So I took one of my humanities courses to learn about the history of materials under Cyril Smith.
We have different types of steels. We have higher-carbon tool steels, we can hot-work steel, we can cold-work steel, we can make very hard tooling that allows us to form steel very rapidly and make sheet metal for automobiles. But the sheet metal for automobiles is always going to be a low-carbon steel, because we can adjust the carbon. The high-carbon steels are going to be in things like files. We don't do a lot of high-carbon steel. Medium-carbon steels — when we need strength and we need it inexpensively, maybe in a pressure vessel — we might use 0.35 carbon steel. But HY-80 probably has a limit on carbon of 0.2. Point two is going to be right down in the yellow. We really would prefer, in the HSLAs, to be down at point one. And the Japanese have developed some steels that are less than 0.02. But they're really expensive, because I have to use a different form of manganese as an alloying element that doesn't have carbon in it. I can make steels that are very low carbon and they have some interesting properties — but because they're expensive, you have to make them the same way we make stainless steels, and so they're more expensive.
Just to tell another story — I'd rather tell stories than go through my lecture notes, as you can tell. The technology to make very low-carbon steels, which is very important in the stainless steel industry — because too much carbon in some cases can destroy the corrosion resistance of stainless steel, and we'll talk about that later — was invented in the basement of Building 8 here, in the 1950s. Stainless steels were much more expensive than they are today.
It was basic science. The graduate student who did his doctoral thesis on the carbon reaction went to work for a company and applied it, and changed the whole stainless steel industry. If you look at the bottom of this little diagram, the bottom has the name John Chipman. John Chipman was the head of the materials department here from 1945 to 1962, when he turned 65 and retired. John Chipman taught the world how to make reproducible steel. In the 1930s and 1940s, they might make five heats of steel and get four good ones with the right chemistry. In the 1920s they might make two heats of steel to get one good one. They'd have to either reuse the steel for something else, but they couldn't control the chemistry of the steel very well. John Chipman spent his career teaching people how to control the chemistry of steel.
§5. Steel terminology, rolling, and the cost of a steel mill [18:31]
If we look at different types of steels, there are lots of terms. There's cast iron versus cast steel. You can have low-carbon steel, which I've defined for you as less than about 0.2 percent carbon. Medium-carbon — these are not hard-and-fast definitions, but that's what people have called things over the last hundred years. High-carbon steels are above 0.6, so somewhere between 0.3 and 0.6 is medium carbon. Mild steel is ill-defined; it's certainly all of the low-carbon steels, but it's also some of the medium-carbon steel, so mild steel might be up to 0.4 or 0.5 carbon. There's hot-rolled steel and cold-rolled steel.
Hot-rolled steel: if I want to make plate, I've got to heat it up. I can't take a ten-inch slab of steel and cold-roll it. I don't have mills that are big enough. In theory you could, but your mill stands would have to be twenty feet in diameter for the rolls. So you heat it up to 1100 or 1200 degrees centigrade, and you can hot-roll it. At that point, compared to what it was before, it is soft as butter. You can squeeze it, extrude it — not that we do a lot of extrusion of steel, but just like squirting a toothpaste tube, if you get to 1100 or 1200 and you have a big enough hydraulic ram, you can squeeze it out. They can hot-roll steel from ten inches thick in a plate mill. A new modern plate mill may cost you one or two billion dollars. The casting facility might cost you two or three billion. The place where you make your cast iron might cost you two or three billion. A cold-rolling mill will cost you one or one and a half billion. You put all this together to build a new steel mill — probably 15 to 25 billion dollars. The last company-built steel mill in the world that I know of was Bethlehem Steel Burns Harbor — almost bankrupted Bethlehem Steel in the 1960s. But by 1974, once Burns Harbor started, high productivity, turned out steel, it was the best year Bethlehem Steel ever had. They hired me and everything's been downhill since.
Cold-rolled — you can hot-roll in a plate mill, and then you go to a strip mill, or a hot-roll mill, and you can go from about half an inch down to 3/16 of an inch while it's hot. It's coming out of there at 60 miles an hour. Cold-rolling mill: you go from 3/16, and now it's thin enough, you can have big steel rolls made of higher-carbon steel, and you can get that stuff at 60 miles an hour. You can go all the way down to foil — not in those same great big rolls, that's another module you won't take. But you can cold-roll the steel. This is where we make the automotive sheet. The price of steel varies: plate is going to be made in a hot-roll mill, and the sheet will be made in a cold-roll mill, and the break point is basically that plate is hot-rolled and sheet is cold-rolled. Bar can be either one. So this little half-inch bar could be cold-rolled, could be hot-rolled, it depends.
§6. Phases, hardness testing, and quenching practice [22:25]
In steel we have different structures. The alpha phase, which is body-centered cubic, is called ferrite, after ferrous. The face-centered cubic, which is non-magnetic — and most of your stainless steels are face-centered cubic, not all of them — is called austenite, after Roberts-Austen, the British metallurgist of the 1890s. Then there's martensite, which is the hard stuff. You can transform the steel and make it extremely hard — 200 ksi strength for this file. That's named after a French metallurgist named Martens, with an S. And bainite is named after a guy who did his research at U.S. Steel in the 1930s to understand how martensite formed, and in doing the research he came up with another phase called bainite. Why are they all named -ites? Because the ability to measure the structure of the steel was invented by Henry Clifton Sorby in the 1880s in Britain. He was a geologist. Geologists like to polish stones, and they get these beautiful pictures in color using polarized light. Sorby, being a geologist, when he would see these different phases, named them, because he thought of them as minerals. That's how they got their names.
You can take a piece of this medium-carbon steel, heat it up into the FCC region, quench it in water or in oil or sometimes in molten lead, and you can transform it to martensite. It won't transform to ferrite — it'll transform to martensite, which is not a cubic structure, it's body-centered tetragonal [Tom said "to trigonal"]. It traps the carbon in there, becomes extremely hard, just like the nail file. And it also becomes very brittle. One of the reasons we like to use lower-carbon steels for welding and structures is because we don't want to get too much of this martensite, which can be very hard. If you keep the carbon content low, you can control the hardness.
This is hardness — called Vickers hardness. You just have a controlled indenter. There are probably 50 different types of hardness tests. Vickers has been around for 100 years, Rockwell's been around for 100 years — those are two of the most common. You have a little steel ball or a little diamond, and you press it into the steel or the aluminum or the copper or whatever your alloy is — or plastic, we do hardness tests on plastics, we do them on rubbers. You press it into the surface and you measure how deep it goes. The deeper it goes, the softer the material, for a given force. It's not a very sophisticated test, but it's used all the time in industry because it's cheap and quick.
If you look at the hardness of steel versus the carbon, if you quench and temper it, you can get strengths like 300 ksi. You can just divide this by three and get ksi, not exactly, but close. Hardness Rockwell C — there's equivalence. So that file is probably about Rockwell C 55 or 60. It's over 200 ksi. Whereas if I cool it slow, I can get a different structure if it's air-cooled. If I cool it very slow in a furnace, I can get things even softer, very malleable. So I can change the structure of the steel by the way I heat-treat it, and by the carbon content I have.
So there's quenching and tempering. In the old days, some of the first steel that people ever had was basically meteoric iron. They'd find a meteor somewhere, and someone found that if you hammered it, it was malleable. Then someone found if you heated it up in a fire and quenched it, it would become brittle, but it became extremely hard. And then they found if you tempered it, you could get some of the ductility back, and you might lose a little bit of the hardness. The best swords thousands of years ago might be made of meteoric iron. Well, there wasn't a lot of meteoric iron, so only the kings had the meteoric iron things. But they did have wrought iron, and they learned how to make swords. And of course, during the Crusades, the best quenching — they used to take a living slave, and they would quench the hot steel by running it through the slave. Gave it the quality to kill, I guess. A lot of little things that people did in the old days that they don't do anymore.
§7. Annealing, normalizing, and bicycle tubing [27:35]
The steel can be annealed like the spheroidized iron, so it's very soft. You can normalize it, where you heat it up and air-cool it back down, and get a very fine grain size, and that improves toughness. Fine grain size is one of the only things that improves both toughness and strength at the same time. Usually you add alloying elements and things get worse in terms of toughness. You might get strength with alloying elements, but you lose toughness.
A commodity-type material like bicycle tubing might be a steel from the 4000 series, or the 1000 series. You might have heard of a 1010 steel or 1020 steel. That's just low-carbon steel. An alloy steel would be something like 4130. HY-80 would be a 4000-series-type steel — if it had an AISI number, it doesn't, it has the U.S. Navy number, HY-80. But bicycle tubing — you can buy it quenched and tempered. And 4130 bicycle tubing, this is thin-wall tubing, is also sometimes called aircraft tubing. If you build your own little airplane frame, you make it out of the same 4130 stuff, and you can TIG-weld it. You can get it quenched and tempered up to 120 or 130 ksi, you can buy it annealed at about 80 ksi, and you can buy it normalized at about 100 ksi, with the alloying elements. So the alloying elements add some strength, but you can get a whole range of strengths depending on the heat treatment.
§8. Manganese — from Spiegeleisen to the welding lawsuits [29:27]
You have your carbon steels, which traditionally — when Andrew Carnegie was selling to the railroads — were the cheapest; alloy steels would cost a little bit more. We don't have that kind of distinction anymore, we still call them carbon steels and alloy steels for historical reasons. Iron-carbon alloys — people found that just straight iron and carbon wasn't good enough, in the 1870s and 1880s, particularly in Europe, where they had a lot of phosphorus in their steel, and they had a fair amount of sulfur, which came from the quality of coal. They didn't have the really low-sulfur coal that you could get out of Pennsylvania. So the United States could make pretty good steel, but the Europeans were having problems. And someone found that if they added some Spiegeleisen — a German name for an iron-manganese ore, which had a lot of manganese in it — they would no longer have this problem with the sulfur.
Sulfur in steel can be a real problem, because it forms a liquid iron sulfide. You end up with not a solid piece of steel, but something sort of like a Slurpee. It's got a liquid at the grain boundaries — an iron sulfide liquid. Think of a snow cone or a Slurpee, and try to roll it: it sort of cracks up and falls apart. That wasn't very good. But people learned that you put manganese in steel.
Now, a few years ago, some attorneys in the United States were trying to go after the welding industry — because if you were to breathe 100 percent manganese — we've known this since the 1790s — people who worked in manganese ore mines would get something that looked sort of like Parkinson's disease, a nervous disorder. We've since learned, actually some people at Mass General did some of this work, that it attacks a different part of the brain. If you get too much manganese, you can get a neurological deficit, which is permanent, and it's sort of like Parkinson's disease.
Most Parkinson's disease is called idiopathic, which means they don't really know what caused it. They did prove that some of the drug addicts in California ended up with Parkinson's disease, from overdoses on certain drugs. So all of a sudden people started saying chemicals are causing Parkinson's disease — that was in the '60s and '70s. So by the '80s, attorneys decided, ah, the manganese in steel is causing people to get Parkinson's disease. You have to understand that about 20 to 30 percent of the populace by the time they turn 80 will have some sort of essential tremor. Thirty percent of us are going to start shaking by age 80. It's just part of aging.
They finally proved that manganism — a neurological deficit caused by too much manganese — attacks a different part of the brain than Parkinson's disease. But the welding industry was spending over a hundred million dollars in legal fees a few years ago to defend themselves against what would be a hundred thousand cases of all these welders out there, who, you know, 20 or 30 percent of them when they get older are going to develop essential tremor. And the attorneys would say, oh, this is caused by welding. It's not caused by welding. They finally proved it, they settled all of them, and all the plaintiffs' attorneys went to find something else to do with their life. Many of them had started out in the asbestos business. Just one of the wonderful things about our system.
I actually helped defend the welding industry on those back in the '90s. I was in Danville, Illinois. Manganese actually is essential for health too. You can go to a pharmacy and they will sell you manganese pills. If you get a big overdose, like 50 or 100 times as much, you can start getting manganese poisoning — just like you can get ice cream poisoning if you eat too much. If you ate 50 pints of ice cream every day, you would get ice cream poisoning.
I'm on the witness stand, and the plaintiff's attorney says, well, Dr. Eagar, do you know how much manganese Mr. [Sedans] would have breathed in a day? And I said, oh, it's about equivalent to eating a handful of raisins. He says, excuse me? I said, you know those little red boxes of raisins you put in your lunch? It's about equivalent to eating a box of raisins. He says, excuse me? At which point I just unloaded on them about how manganese is essential for health, and we don't have a minimum daily requirement for manganese because it's in most of the foods we eat naturally. You don't really need to take supplements, unless you have a really weird diet — like one kid I was in the infirmary with once, a student, he lived on M&M's and Coca-Cola. If that's your diet, you probably need some manganese. Green leafy vegetables, fruits, nuts, dried fruits — I rattled off four or five health foods that have manganese. And at the end he says, objection your honor. The judge says, you're objecting to your own question? What he meant to say was "move to strike" — he could have asked to strike my testimony as being non-responsive — but he said "objection." Everybody in the courtroom started laughing at him. It wasn't my joke, it was the judge's joke. But anyway, he lost that case.
§9. Hydrogen still in the bottle — closing demonstration [35:34]
Now, let's finish up here with this. You still see a few bubbles, but if I shake it, you see more bubbles, fine bubbles. So there's still CO2 in here. If this were a weld, there would still be hydrogen in there after an hour. There would be less hydrogen — I can't get as much as I had in the very beginning — but it's still there. I can nucleate those bubbles by shaking it, because I create pressure waves, and the pressure waves help with the nucleation. But we don't have to go through nucleation and growth. Remember that, because we will start talking a little bit about hydrogen embrittlement tomorrow. And after we talk a little bit about hydrogen embrittlement, we'll have a couple of presentations, and that's it for today. Okay, thanks.