SMS_F2013_10

Structural Materials Selection Fall 2013 Session · 12 sections 11 cases · Watch on YouTube ↗ all files
Layer 3 — readable edition

§1. Project framing: choosing a material for an application [00:02]

§1.p1

The purpose of the project is for you to present something you're interested in. Fiberglass is fine, but it's a little broad to just say "fiberglass" everywhere because it's been around for so long. You might pick an application and say, okay, people use fiberglass in small craft, in boat hulls — why don't they make them out of wood? Because they do make some out of wood. Why don't they make them out of another type of plastic? Hopefully you've learned some things about properties of plastics, why you want to use a composite in that application — but composites are more expensive than steel or aluminum. People make boats out of aluminum. Why would you make one out of steel? People make boats out of steel. There are lots of materials, and you could say, here's an application, someone has chosen fiberglass, why in that application?

§1.p2

Take small boats, or take tennis rackets — focus it a little bit more. "Fiberglass is a material, glass fiber in a plastic resin" — okay, I want a little more depth than that, a little more thought into why someone would choose fiberglass in this application.

§1.p3

On Thursday we've got a guest lecturer, Adam Powell, who's going to talk about Infinium, a new process that started out here at MIT. The professor went over to Boston University, and now they've started this company that's going through about ten or twenty million dollars of funding for a new environmentally clean way to make all kinds of metals — almost everything on the periodic table. On Friday, if I have time, I've been thinking about doing something like comparing soda bottles, or beverage containers. What are the materials of choice in beverage containers? There's aluminum. Plastic. Paper — actually most paper is a composite, even if it's just the plastic-coated paper of milk cartons. If you go to those drink boxes, that's a seven-layer composite, pretty complex. Glass. And one other: steel.

§1.p4

In Japan — anybody ever lived in Japan for a while, visited Japan? If you go to the vending machine and you get a soda, is that an aluminum can? Nowadays it depends. When I lived there thirty years ago it was always steel, and it was a political thing. Japan doesn't have an indigenous aluminum industry, but their government wanted to protect their steel industry and give them a bigger market. So the Japanese used steel — not because it was cheaper, not because it was better, but because the government was trying to give a market to the steel companies. Not that the steel companies really needed it in 1985, but they still had it. I haven't been back to Japan for ten years now, but I understand they are using more aluminum cans now, because they don't need to protect their steel industry anymore. It gets down to economics.

§1.p5

What do they make cars out of in Europe — steel or aluminum? In general, steel. But at the price of fuel in Europe, aluminum should be cheaper. So why do they make steel cars? Because everybody wants to sell into the United States, and at the price of fuel in the United States, steel is cheaper. If the Europeans were only selling in their own market, all their automobiles would be aluminum. But then they couldn't sell in the United States — they'd be priced out of the market.

§1.p6

One day this week I might ask you to start writing down what your topics are, so think about it. You've only got about ten minutes for the presentation — that means ten overheads max. Practice your presentation in front of a mirror to make sure you can do it in ten minutes. If you can do it in ten minutes with no audience, you'll probably do it in twelve minutes with an audience. Making eye contact, or if someone raises a question in the middle of it.

§1.p7

As a young assistant professor giving my first few talks at conferences, I would go over that talk fifteen times and literally stand in the hotel room in front of the mirror and give the talk over and over. Now I don't have to do that anymore — but maybe I should. If you haven't had a lot of experience giving presentations, you need to practice. You don't just walk in and do it.


§2. Carbon and alloy steels: hardness, hardenability, hardening ability [06:27]

§2.p1

We started talking about steels last time. They're made everywhere; it's one of the most technologically advanced materials, if not the most. We know more about steel than anything else, just because of the size of the market. There are many types of steels, and I gave a handout last time — if you didn't get it, see Jerry.

§2.p2

One of you asked, why are there so many steels? I pointed out last time that the carbon content of the steel determines its strength to a first approximation. A 1050 would be a half-percent carbon steel; here's a 1010, for a carburized part, that's a tenth of a percent. The last two digits are hundredths of a percent of carbon, and typically most carbon steels are less than one percent carbon. So here's a stabilizer bar which is a 1090 steel. Lawnmower blade of 1060. A spring at 1080. High strength, high carbon; low strength, low carbon.

§2.p3

Then we also have alloy steel parts that have typically different amounts of chrome, nickel, molybdenum, and vanadium, and there's a whole wide range of these. There's a difference between hardness, hardenability, and hardening ability. Hardness is just the resistance to indentation. It's a direct measure of the strength. The hardness measurement, if you measure it in kilograms per square millimeter or KSI, is about three times the tensile strength of the material, whether it's steel or copper or aluminum, for a ductile material.

§2.p4

Then there's hardenability, which is how deep you can harden it. I was asked once by a professor — would be now mechanical engineering, but at the time he was naval architecture — working on the America's Cup. He had a four-inch-thick keel beam. The keel beam is not just on a big ship; the keel is the thing you build everything up from. There's a keel beam on a 747, a big aluminum beam they build everything up from. But on a sailboat — that particular America's Cup — they had a big steel piece that created the resistance to being blown sideways. This was supposed to be four inches thick.

§2.p5

HY-130 was a very weldable steel — more weldable than most steels, still got problems, but the Navy had developed it in the 1950s and 60s for submarine hulls, and it needed to be weldable. It's got about two-tenths-percent carbon. Well, this was four inches thick. It had to have enough hardenability to transform to martensite four inches deep, so you get uniform strength through the thickness. With carbon steel, you might be able to get a good eighth of an inch or quarter of an inch thick before you lose your ability to quench it fast enough. But when you start putting these alloying elements in, it slows down the transformation, and you can heat something and quench it such that you get hardness four inches, eight inches, even deeper than that in some steels. For this keel beam, supposed to be four inches thick, they were actually bending it in some of the trials — it was plastically deforming. Better than snapping, but still. They also wanted weight down in the bottom of the ship — they want light weight, but if you're going to have weight you want it in the bottom. They used to put lead in these ships in the bottom — you're paying a huge penalty, but they needed the weight toward the bottom.

§2.p6

Hardenability is the depth and distribution of hardness that can be induced by quenching. Hardening ability is what the maximum hardness can be, and at about 0.6 carbon you get to the maximum hardness in most steels. HY-130 only had about two-tenths-percent carbon, and it only had 130 KSI yield. If you went to 4340, which is readily available in big heavy sections — we use it to make all kinds of high-strength components in thicknesses up to ten inches thick — it's got 0.4 carbon. The 40 is four-tenths-percent carbon, and you can get much higher hardness, much higher strength: 180 KSI rather than 130. So they wanted to use 4340 but didn't know how to weld it in that thickness. That's why he called me up one morning. I showed them how to weld 4340. It's not easy, not cheap, but who cares if you're building America's Cup — it's only one of a kind. You may only be sailing one ship, but they actually have multiple hulls on these things. Very interesting business.


§3. Learning to weld: oxyacetylene and arc [13:06]

§3.p1

Student: Can you weld really well?

§3.p2

Not particularly — I don't do it every day. I can weld. I was taught when I was an engineer at Bethlehem Steel by the technicians. I said I want to learn how to weld, and they took me out and I did a lousy job.

§3.p3

I always remember one time though. I was walking through my lab — not like it looks like now, but they used to have a bunch of boxy welding setups. Just like they have the blacksmith shop in the basement, they used to have a guy — it started when I was a student. My thesis advisor became the head of this little thing. It was to help a technician — the guy who had the welding lab before me left in 1968 when I came as a freshman, and he had a technician, Tony Zona, and Tony didn't want to go to Wisconsin with Professor Adams, so he stayed at MIT. He didn't really have anything to do, so he started teaching an art sculpture class, and they would do oxyacetylene welding.

§3.p4

So I was walking through my lab twenty years later, and I had four or five of my students all standing around one of these tables with the oxyacetylene torch. I said, what are you doing? They were trying to teach themselves to oxyacetylene weld. They looked terrible, like turds on a plate. I said, that's not how you do it — give me that. I took the torch and I put down an oxyacetylene bead, and it was a very nice looking weld bead. I told them how to do it and walked away. What they didn't know is that was the first time in my life I had ever oxyacetylene welded. But I did understand heat flow and welding, and I understood the thermal time constant for oxyacetylene welding as opposed to arc welding. I knew you had to go slow, and I actually knew how slow you had to go. Most people try to rush it.

§3.p5

I also do a thing about reaction times. If you go to Wentworth Institute of Technology, they will probably teach you welding by starting you with oxyacetylene. The reason is the thermal time constant is on the order of three to five seconds of that weld pool, so you've got to go really slow — three to five seconds in one spot. If you do that you get a good weld. For an arc weld, the thermal time constant is on the order of a tenth to three-tenths of a second. Ten times faster, because the heat intensity is ten times greater. This makes sense if you take the heat transfer part of my course.

§3.p6

It's because of reaction time. You don't start out on the highest speed of a video game — you have to build up your skill and your reflexes, and a welder has to do the same thing. There are many places where they give people one week of instruction in welding, start them with arc welds, they make terrible welds for several days, and then they send them out to make terrible welds in production. But if you go through a four-month class, they'll start you with oxyacetylene for several weeks and then move you to arc welding. It's hand-eye coordination — let's face it, if you're manually welding. So yes, I do know how to weld. Am I the best welder in the world? No. If I welded a pressure vessel, not only would it not pass inspection most likely, but I wouldn't want to stand next to it when it's pressurized. You really need to do it on a regular basis. The codes — depends on which code — require that if you have not welded in the past year, you must be requalified, because you haven't kept up your skills. There's skill, but I don't practice sufficiently.

§3.p7

All the rest of you could weld if you wanted to learn. Many students come to me and say, I'd like to learn to weld, so I send them off with one of the technicians, and usually after about one or two hours of making lousy welds they say okay. Very few students over the years have ever gone for more than a few hours' worth of practice.


§4. Why there are so many steels: alloying economics [17:44]

§4.p1

Getting back to steels: there's hardness, hardenability, and hardening ability. Hardness is merely how strong it is. Hardenability is how deep you can get a hardness. Hardening ability is what the maximum strength is that you can get under the best conditions.

§4.p2

We have many, many steels because they're used in such high volume that people will literally demand, I want a tenth of a percent nickel more, or a tenth of a percent nickel less. That doesn't sound like much, but a tenth of a percent nickel is two pounds in a ton. At $10 a pound, that's $20 a ton, so it makes a difference, particularly if you're buying 100,000 tons — you'd be talking $2 million. So there are many steels.

§4.p3

Does that answer the question? I think it was your question several weeks ago — I said I would cover why we have so many steels. I also talked about the Achilles heel of steel being corrosion, and we'll get to that. The easiest way is to jump into stainless steels as the next class of material, separate from the carbon and low alloy steels.


§5. Stainless steel genealogy and 304 [19:16]

§5.p1

The stainless steels — this comes out of a book by John Sedriks called Corrosion of Stainless Steels, and he has a genealogy of stainless steels. The first stainless steel was basically what we call 304 stainless steel, discovered around 1907 in Sweden. Back then they were just learning how to make steels of different compositions and adding different things. You may have heard of 18-8 stainless steel — basically, instead of 19 chrome 10 nickel, 18 chrome 8 nickel — it's all within the range of 304 stainless. It's either 40% or 60% of all stainless steel. I think it's 40% of all stainless steel and 60% of all the austenitics.

§5.p2

There are different types of stainless steel: austenitic, ferritic, martensitic, duplex, precipitation hardening. The austenitics have face-centered cubic crystal structure, they're non-magnetic, and 304 is the most common of all. You'd like to improve on that alloy that's been around for 100 years. One thing they do is add sulfur or selenium for machinability. We also add sulfur to plain old carbon steels for machinability. You end up with sulfur inclusions, and when the chips are coming off the lathe or milling machine, those inclusions — which tend to form stringers — cause the chips to break up. You don't get this long pigtail corkscrew that'll whip around and cut people's arms off.


§6. Manganese, sulfur, and the early steel industry [21:51]

§6.p1

Student: Why is manganese in steel?

§6.p2

It's not stainless steels per se — steels in general have to have manganese for sulfur control. Steel cannot be made without manganese. Although we do make some now that have extremely low manganese.

§6.p3

To make steel in the old days — actually even today — you take iron ore, coal, and limestone, put them in a blast furnace, you make cast iron. The problem is the coal may have two or three percent sulfur in it, and if it does, the sulfur ends up getting in the steel. Iron sulfide is a very low melting alloy, and if you take that steel without any manganese and try to roll it — when you heat it up to the hot working temperature of 1,800 to 2,000 degrees Fahrenheit and try to roll it, it will just shatter, because it has liquid grain boundaries of iron sulfide. That's what happened in the 1880s, and they didn't understand it was sulfur until they were using different ores in different parts of the world and some people had this problem and some didn't. They started analyzing it and they found manganese — the steels with three-tenths of a percent or more manganese would not shatter.

§6.p4

Then people started doing metallography — they had learned about metallography in the 1880s — to look at where the sulfur was, and they found you formed manganese sulfides. If you look at the periodic table, the only two elements that have higher melting sulfides than iron are manganese and molybdenum. When you add molybdenum to iron, you get a low melting sulfide. When you add manganese, you get a high melting sulfide that melts above the melting temperature of the iron. The only element in the periodic table that does that. There is no other element that will tie up the sulfur impurity. So you have to either get rid of your sulfur impurity, or you have to put manganese in the steel.

§6.p5

If you didn't have manganese, the steel would not be hot-workable. It's like if you were trying to hot-work a Slurpee — a liquid on top of solid ice particles. The steel would have that kind of consistency; when you compress it, it would actually crack apart. Not exactly like a Slurpee, but you basically have liquids along the grain boundaries.

§6.p6

A steel officially is an iron-carbon alloy, but it always has some manganese — usually about three-tenths of a percent up to two percent — and it has some silicon, quite often. The silicon is there to get some of the oxygen out; manganese will take some of the oxygen out too. I always talk too much about steel. Historically they would say the manganese had to be 8 to 10 times the sulfur content because you have to tie up the sulfur. Nowadays we've learned to get the sulfur out of the steel while we make it, using the right types of fluxes. One of the guys who taught the world how to do this using the principles of physical chemistry was John Chipman of the Chipman room. The world learned how to make steel from John Chipman. He and Nick Grant, who was a professor starting here in 1946 — the two of them published some of the first articles on control of sulfur in steel in the 1930s. People knew you put manganese in, but they didn't know how much, and they were wasting a lot of money on manganese. Virtually all the manganese in the world goes into alloying for steel.

§6.p7

Next thing they did was teach people how to get rid of the phosphorus, how to control it. A lot of the iron ores in Europe are high in phosphorus, and phosphorus can be good in steel in small quantities, but bad in large quantities. There's a lot about the alloying elements of steel. People teach whole courses at some universities on this — not much anymore, but there's still a few.


§7. 303, 304, and the Sub-Zero refrigerator hinge [27:20]

§7.p1

You start with 304, and you can add sulfur — in stainless steels you add selenium because you can afford it. You can't afford it in most carbon steels. Some carbon steels have selenium, but only if you can't have sulfur for some other reason. Selenium is beneath sulfur on the periodic table. So all you've done is go down the periodic table, and you get these inclusions in the steel that help break up the chips. When we machine brass, we used to put lead in because it would break up the chips. Free-machining brass is a leaded brass — it forms lead inclusions rather than sulfur inclusions. That's 303 stainless. What do we lose when we do that? It's not weldable anymore. You can't weld 303 stainless. Same composition as 304, just has high sulfur.

§7.p2

I meant to bring in a hinge from a Sub-Zero freezer. Anybody know what a Sub-Zero refrigerator is? It's an $11,000 home refrigerator. Very high-end. Obviously none of you have millionaires for parents. I have a son who just bought a house and it has a Sub-Zero refrigerator, and a Wolf stove — goes for only five or six thousand. Very high-end kitchen appliances. I have a Sub-Zero hinge because someone made a mistake. They had a little pin that was supposed to be made out of carbon steel and welded to a carbon steel plate to make the hinge, and someone used 303. When they machined these little pins, and when they went to weld them, they had cracks. They didn't see the cracks. They plated them, put them on the refrigerator, and people bought $11,000 refrigerators and the doors started falling off. They were not happy. And these are the type of people who have enough money to sue you. So if you're going to shaft somebody, do it to poor people — they can't afford attorneys. That actually cost about $7 million worth of repairs.


§8. Duplex, precipitation hardening, and high-manganese stainlesses [29:46]

§8.p1

Duplex stainless steels: they increase the chromium and lower the nickel to get higher strength. You can double or triple the strength by going to a duplex. A duplex stainless is merely a mixture of about 50/50, or 30/70, or 70/30, of austenite and ferrite — ferrite being a body-centered cubic magnetic phase. [Tom holds up a piece of duplex stainless steel.] This came off a centrifuge that was used to separate starch from other things. They started selling a lot of these centrifuges when they started making methanol from corn, because they had to separate the starch from all the other liquid. The centrifuge weighs about a ton, spins around like 3,000 RPM, about four feet in diameter. Made out of cast duplex stainless steel because they needed high strength.

§8.p2

If you're not careful with duplex stainless steels — you have higher chromium and lower nickel, and the ferrite can get hydrogen in it. But the real problem is you can get iron-chrome intermetallic phases that embrittle it. You want to see a brittle steel — here it is. [Tom indicates the fracture edge.] It did have a little bit of ductility. If you look right on this edge, you can see the curvature of the centrifuge, and you can actually see a slight change in the curvature, where the large grains of this casting started to stretch — what we call orange peel, because it looks like a surface of an orange. It can be extremely brittle if you don't do it properly, but you can get 100 KSI strength rather than 30 or 40 KSI for the 304. So you can triple the strength.

§8.p3

They use duplex now — it's become very popular in the last 15 or 20 years — for downhole sour gas applications where you need high strength in deep wells in the oil business. Before that it was very specialized; they used it on the catapults of Navy aircraft carriers. High strength corrosion resistant alloy. Better chloride stress corrosion resistance than 304, so if you had a chloride environment and you wanted high strength, you could go to duplex stainless. Hard to weld, but high strength and some interesting mechanical properties. You pay a price in some other things — can be toughness, can be a welding price.

§8.p4

Precipitation hardening stainlesses have been around for a long time. [Tom produces a pair of medical-grade scissors.] Here's a $300 pair of scissors — actually probably $600 now. Medical grade. They actually have a nickel-base hard alloy insert brazed in them. Because they're medical, instead of being a $40 pair of scissors, 25 years ago when I got them they were $300, I bet they're $600 now. Precipitation hardened stainless. You wouldn't do that for a regular pair of scissors — what do you care about it on the handle — but the doctors want to know that if they're paying $600 for a pair of scissors, they're getting the best quality. Why it's precipitation hardened — if you're going to put inserts in the cutting edge, I don't know why you need to precipitation harden the handle, but anyway. The Navy used to use these on hydrofoils, on the foils that go through the water. They make the base ship out of aluminum for lightweight, but the foils need corrosion resistance — aluminum would just erode away. The America's Cup boats now are basically hydrofoils. I have no idea what they use — they might be using a precipitation hardened stainless. You can get 180 KSI out of these steels. You have to go through a fancy heat treatment, harder than regular carbon or alloy steels, but you can get tremendous strength.

§8.p5

201 and 202 add manganese and nitrogen, and lower the nickel, to get higher strength — 80 to 100 KSI, and it's cheaper. Manganese instead of chrome. Manganese is a lot cheaper than chrome. I don't actually see these that often. If you go really high manganese — like 12 to 13% — you get what they call Hadfield's manganese steel, which is not officially a stainless steel, but it's an alloy that work-hardens so rapidly that you literally cannot cut it with a saw. It's easily formable in its annealed condition, but you go to saw it and as you're cutting, those chips harden — about triples its strength — and it will dull saw blades. One of the uses of Hadfield's manganese steel was prison bars. You got a file? Yeah, you want to dull your file. Try to file through Hadfield's manganese steel, because it transforms as you're work-hardening it to this hard martensitic steel.


§9. Martensitics, crystal structures, and hydrogen embrittlement [35:25]

§9.p1

The martensitic steel — you take the nickel out, lower the chrome down to around 10 or 12%, and you get the 403, 410, and 420 steels. Many of the scissors and medical instruments are actually made out of the martensitics. They're magnetic. Martensite is basically the same as ferrite, except you've trapped carbon in the lattice and it becomes body-centered tetragonal — but very barely body-centered tetragonal rather than body-centered cubic. Austenite is face-centered cubic; ferrite is body-centered cubic; martensite is body-centered tetragonal. Duplex is FCC plus BCC, that's why you call it duplex. And precipitation hardening is basically FCC with precipitates, which do the hardening.

§9.p2

In general strength levels: austenitic about 40 KSI. Ferritic might be 40 to 80, depending on how work-hardened. Martensitics could be heading up toward 160 — if you take piano wire, deform it into wire, you can get up to 300 KSI; body-centered tetragonal you can be up to 180 KSI. But you really like to be less than 120 for hydrogen embrittlement reasons. Duplex typically about 100 KSI, good ductility, good chloride resistance. And precipitation hardening 180 KSI with good ductility and reasonable hydrogen embrittlement resistance.

§9.p3

The only ones that hydrogen embrittle are the ones that are BCC. Frankly it's still a matter of opinion among metallurgists what causes hydrogen embrittlement. There are four or five theories — strain in the lattice — I suspect when we get really good at quantum mechanics on a larger crystal, rather than just a few atoms, we might be able to figure out what hydrogen does in terms of embrittling the body-centered cubic structure. You can embrittle the austenitic, FCC structure, but it takes 50 to 100 times as much hydrogen to do it, so we don't usually talk about hydrogen embrittlement unless it's ferritic. The austenitics and the precipitation-hardened grades are much more resistant.


§10. Molybdenum, 316, and the Fenway Frank hot dog cooker [38:44]

§10.p1

If you have corrosion resistance in aqueous systems, particularly with chlorides, you add molybdenum for pitting resistance. You might have 2% molybdenum in a 316 stainless steel. The 304 stainless steel cost you $2,000 a ton, with 18% chrome and 8% nickel, rather than $400 a ton for carbon steel. So 304 stainless is just as expensive as aluminum. For 316, you add 2% molybdenum. Molybdenum is going for $20 or $30 a pound. 40 pounds for 2% per ton — you're adding $1,200 a ton. Big jump in price to go to 316. You can also go to 317, which they used to use for a lot of medical applications — that's 4% molybdenum, so you're talking $24,000 a ton.

§10.p2

To show you the difference — this is the most dramatic. [Tom holds up a cracked, rusted stainless steel component.] People use 316 all the time because it's weldable, it's got stress corrosion cracking resistance, but you have to be careful — these are all cracked. You can see the cracks in this thing, it's all rusty. This came out of the Fenway Frank hot dog cooker over here in Everett. The hot dog cooker is about the size of this room — it's a big steam oven. They chop up all the types of meat you'd never want to eat, make a slurry out of it, extrude it, put it inside the casing, with a lot of salt — that's why you need chloride resistance. The original hot dog cooker was all made out of 316 stainless steel, had been around for 10 or 20 years no problems. They had to modify some things, it was all specified to be 316, and they put the new stuff in, and within several weeks the new stuff was getting rusty. It was cracking. They called me in, and I actually got to see how you cook hot dogs. Lots of interesting things in this work. They asked me to figure out why. We brought it back, turns out it's 304. Someone used 304 rather than 316. Maybe a harmless mistake, because 40 to 60% of all austenitic stainless steel is 304, and 316 is a lot less in tonnage. Just a mix-up. But that 2% molybdenum went from something that would look stainless for 15 or 20 years or more in this environment, to something that could only last a couple of weeks. That's the most dramatic example I've seen. Plenty of failures where people use 304 rather than 316, but nothing this dramatic.


§11. Active vs. passive stainless: the New England Aquarium and Diet Coke [42:05]

§11.p1

People will often bring me a piece of steel and say it's supposed to be stainless and it can't be, because it's rusty. Well, I'm sorry — it's stainless. There are two types of stainless: rusty and unrusted. More specifically, if you look at the galvanic series — which I didn't bring with me — the galvanic series is what dissimilar metals will attack one another in a corrosion series.

§11.p2

The galvanic series has magnesium as the most electroactive and carbon as the least. Everything else in between. Platinum is close to carbon, zinc is close to magnesium, copper is somewhere in the middle, iron somewhere in the middle. Stainless steel comes in two varieties: active and passive. What gives stainless steel its corrosion resistance is a protective oxide — for our purposes call it Cr₂O₃·nH₂O — a complex oxide on the surface that has very good aqueous and high temperature corrosion resistance, except in the presence of chlorides. Active is when you take off that protective oxide, and now you're exposing a steel without one. Passive: no rust. Active: rust. There are lots of ways to take it off.

§11.p3

Student: What about fluorides?

§11.p4

Fluorides are terrible — worse than chlorides — but there aren't a lot of fluorides we expose things to usually. Chlorides are everywhere, fluorides are not. The best way to take this chrome oxide off is not hydrochloric acid, it's hydrofluoric [hydrochloric] acid. That's the most effective.

§11.p5

50 nanometers — very thin. Doesn't take much to take it off. Stick it in hydrochloric acid, you take a passive stainless and turn it into an active stainless within minutes. 316 with the molybdenum is resistant to three times the chloride concentration of 304. 317 is probably resistant to five times. Typical atmospheric corrosion resistance of 304 might be a thousand parts per million chlorine — if you were raining or spraying salty water. 316 would be 3,000 PPM. 317 might be even higher.

§11.p6

The reason you could use it in the body — anybody know what the salt content of the body is? Seawater is about 30,000 parts per million, and your blood is about 30,000 parts per million. When I was in fourth grade they taught me that's proof we evolved from the fishes, because our blood content was the same as saltwater — if you believe that. A little oversimplified, but basically inside our body is like seawater in terms of chlorides. One important difference: we don't have the same oxygen content as at the surface of the ocean. We might have an oxygen content similar to the bottom of the ocean. I guess that proves we started out not a surface fish but a deep sea fish. We were deep sea predators.

§11.p7

In any case, this is better and better chloride resistance. It really depends on not just the chlorine but the oxygen content, because the chlorine screws up the protective oxide layer when it forms. The nice thing is, usually even in 304, if you break off that protective oxide layer — which is very easy to do — it will reform just in the air as a protective layer, unless there are some chlorides around.

§11.p8

If you want to go over to the New England Aquarium — how many people have been to the aquarium? What's the outside made out of? Stainless steel. What was the specification? They wanted to make the outside look like fish scales, and they were going to use 304 stainless panels, and they were to be passivated. I read the spec. Passivation means in the mill they took the coils of steel before they had been stamped into fish scales, and they passed them through a bath of nitric acid — about a 5 or 10% solution, not concentrated. It's an oxidizing acid. You clean the stainless steel in the nitric, but you also oxidize it to form the protective oxide. It's now passivated. You can passivate 304 stainless in 15 minutes in nitric acid.

§11.p9

So they bought the passive stainless steel, cut it into panels the size of fish scales, then took a grinder and brushed on this texture. What happens when you brush on the texture? You're taking off the protective oxide. What happens when you then put it in a marine environment, Boston Harbor? You get oxidation right along those little scratch marks. If you go look carefully — very few people do, but I had to, I got paid for it — you will find rust. If you go around the side to the loading dock where the trucks come in to drop off the food, you'll see where they actually scraped up against the stainless steel with their bumpers, and you have big bands of rust, because they mechanically abraded away the protective scale. In a chloride environment, the protective scale did not reform like it would inland. But right there on the coast.

§11.p10

We had to figure out how to clean it. I tried citric acid — that will work like nitric acid. We had to have something that was not going to destroy Boston Harbor, because all the environmentalists. One of the things I tried was Diet Coke. It didn't work all that well, but Diet Coke is phosphoric acid, and it also is a cleaning solution. Who could complain about dropping Coke into Boston Harbor? Not that kind of coke — the drink.

§11.p11

I had a case like that once, in New York City — one of these 20-story buildings above the approach to the George Washington Bridge — and they had a weld crack. When the cab driver let me off, I didn't know the addresses in New York City, and he dropped me off and there were all these little empty capsules all over the sidewalk, which were basically crack cocaine capsules. They told the homeowners, the condo owners, at a meeting before I was there, that they had a crack in the basement, and after the meeting a bunch of the homeowners wanted to go down there and get some of the crack. True story. I got to see the crack — eight feet long. Had a piece of it in my lab. I didn't eat it.


§12. The L grades and the AOD revolution [51:15]

§12.p1

You have this interaction between oxygen and chlorine in stainless steels. Chlorine is not good for the steels. You can add molybdenum at a tremendous cost, but they do all the time. Then there's the L grade of these — all three of them — which means extra low carbon. That means you can weld it without getting chrome carbides that also destroy the corrosion resistance. I'll talk about that next time, but let me tell you that most of the stainless steel we make nowadays is made by a process that gives us very low carbon — less than 300 parts per million carbon. Which prevents these chrome carbides and loss of corrosion resistance that occurs during welding.

§12.p2

The process is called argon-oxygen decarbonization. It's now the process by which every bit of stainless steel and most nickel-based superalloys in the world are made. It was developed right down here in the basement of Building 8 by a guy named Krivsky, who was doing his doctoral thesis under John Chipman. They were doing basic research, bubbling argon through molten steel and looking at the carbon monoxide reaction — carbon and oxygen in the steel — finding out what would happen. They found if you bubbled argon rather than air, you would have a low enough oxygen potential in argon — the oxygen potential is very low — you would basically pull carbon and oxygen out. It used to cost a small fortune to make 304L stainless. Nowadays with AOD steelmaking, you get low carbon without any difficulty, such that we hardly ever make 304 stainless anymore.

§12.p3

Back in 1950 it was a super premium to get 304L, because it doubled your steelmaking time — you had to have multiple slags to get the carbon out. The process wasn't patented at MIT. Krivsky was doing this basic science research, then went to a company and realized this process could be used commercially to get the carbon out of stainless steel. Since about 1960, virtually every stainless steel in the world is made by argon-oxygen decarbonization. You just bubble the argon through and take the oxygen and carbon out. I once estimated that's worth about a billion dollars a year to the world steel industry. I'll see you tomorrow. I thought I was going to finish up steel — well.

Cases referenced

  • America's Cup 4340 welding procedure development §2.p4

    Used to introduce the hardenability concept. A four-inch-thick keel beam couldn't be made from HY-130 (too soft, plastically deforming in trials) so they wanted 4340, but didn't know how to weld it in that section thickness. Tom showed them how.

  • Sub-Zero refrigerator hinge failure §7.p2

    Illustrates the consequence of 303 vs. 304 mix-up. Manufacturer used 303 (sulfur-bearing, non-weldable) instead of 304 for a pin-to-plate weld. Cracked at the weld, plated over, doors fell off in service. ~$7M in repairs.

  • Duplex stainless steel centrifuge brittle failure §8.p1

    Physical demonstration. Cast duplex stainless centrifuge (~1 ton, 3000 RPM, 4 ft diameter) for separating starch in corn-methanol production. Embrittled by iron-chrome intermetallic phases. Shows orange-peel surface where large casting grains stretched before fracture.

  • 316 stainless steel heat exchanger tube pitting failure (Fenway Frank application) §10.p2

    Dramatic 304-vs-316 substitution case. Original 316 cooker ran 10-20 years with no problems. Replacement parts mis-specified as 304 cracked and rusted within weeks in the high-salt steam environment.

  • New England Aquarium exterior panel corrosion §11.p8

    Passivation defeated by post-passivation mechanical abrasion. Mill-passivated 304 panels were ground to create fish-scale texture, which removed the protective chrome oxide. In marine environment (Boston Harbor chlorides), the oxide could not reform. Rust appeared along grinding marks and along bumper scrapes at the loading dock. Tom tested citric acid and Diet Coke (phosphoric acid) as environmentally acceptable cleaning agents.

  • New York City building crack near George Washington Bridge §11.p11

    Brief anecdote — 8-foot weld crack in a 20-story building. Tom uses it for the cocaine-capsule sidewalk joke connected to the Diet Coke punchline of the preceding aquarium case. The technical content is minimal in this lecture.

  • Argon oxygen decarbonization (AOD) process development §12.p2

    Closing case of the lecture. MIT doctoral student Krivsky, working under John Chipman in the Building 8 basement, doing basic carbon-monoxide-reaction research. Discovered low oxygen potential of argon would pull carbon and oxygen out of molten steel. Not patented at MIT; Krivsky commercialized it at a company. Now used for virtually all stainless steel and most nickel superalloys worldwide. Tom estimates ~$1B/year value to the world steel industry. The case Tom uses to illustrate "MIT lost the patent" lessons elsewhere in the corpus.

  • Infinium magnesium extraction process development §1.p3

    Forward-reference to Adam Powell's upcoming guest lecture. Spinoff from MIT via Boston University; ~$10-20M in funding for an environmentally clean metals extraction process.

  • Japanese steel soda cans §1.p4

    Policy-protected industrial choice — Japanese government protected domestic steel by keeping vending-machine soda in steel cans rather than aluminum, since Japan has no indigenous aluminum industry. Aluminum has now displaced steel as the protection eased. Used as the lecture-opening illustration of how non-technical factors drive materials selection.

  • Steel vs. aluminum automobiles (US vs. Europe market split) §1.p5

    Fuel-price economics would favor aluminum cars in Europe, but European manufacturers must build for the US market where fuel prices favor steel. Illustrates how exporting drives materials selection.

  • Hadfield's manganese steel (prison bars application) §8.p5

    12-13% Mn steel that work-hardens during machining, making it nearly impossible to saw or file. Used historically for prison bars. Distinct alloy class from the 200-series stainless steels Tom is discussing. ## Figures and recurring statistics

Layer 2 — cleanup edit
p1 00:02

Tell me about potential projects or presentations. I mean I say project, presentation, um.

p2 00:19

Yeah, well if it's not well known to you necessarily, I mean, the purpose is for you to present something that you're interested in. Fiberglass is fine, if it's a little broad to just say fiberglass everywhere because it's been around for so long, you might pick an application and say okay, people use fiberglass in small craft, boat hulls. Why don't they make them out of wood, because they do make some of them out of wood. Why don't they make them out of another type of plastic? Well hopefully you may have learned some things about properties of plastics, why you want to use a composite in that application, but composites are more expensive than steel or aluminum. People make boats out of aluminum, but, and why would you make it out of steel? People make boats out of steel. So there's lots of materials and you could say okay here's an application, someone has chosen fiberglass, why in that application?

p3 01:17

Okay so, but I, you take small boats or take tennis rackets, or get it, focus it a little bit more, or you're just going to be, well fiberglass is a material, glass fiber in a plastic resin. Well okay, I want a little more depth than that, I want a little more thought into, well why, you know, in a case like that, why would someone choose fiberglass in this application.

p4 01:46

In fact, depending, I should finish up this week, we've got a guest lecturer Adam Powell on Thursday who's going to talk about um Infinium, uh, which is a new process that started out here at MIT and then the professor got, went over to Boston University, and now they started this company that's going through about ten or twenty million dollars of funding for a new environmentally clean way to make all kinds of metals, almost everything on the periodic table. Um, and um, anyway so Adam will be giving a guest lecture. But on Friday, I've been, if I have time, I've been thinking about doing something like comparing soda bottles, or you know, beverage containers. What are the materials of choice? Are we on the thing? Okay that's fine. So what are the materials of choice in beverage containers? I mean, you know, there is aluminum.

p5 02:48

Plastic. Paper, or actually most paper is a composite even if it is just the uh plastic-coated paper of milk cartons. But if you go to those drink boxes that's seven-layer composite okay, so it's pretty complex. Anyway, what did you say, glass, okay that's right. So why would someone, there's also one other. Steel. Uh, it turns out in Japan, anybody ever lived in Japan for a little while, visited Japan? Yes. And you go and you use the vending machine and you get a soda, is that an aluminum can?

p6 03:26

Depends, nowadays it depends. When I lived there thirty years ago it was always steel, and it was a political thing. Japan doesn't have an indigenous aluminum industry, but their government wanted to protect their steel industry and give them a bigger market. So the Japanese used steel, not because it was cheaper, not because it was better, but because the government was trying to protect their steel, or give a market to the steel companies, okay. But anyway, so I might go through something like that. Um, not that the steel companies really needed it in 1985, but they still had it. But in talking to people, I haven't been back to Japan for ten years now, but I do understand that yes they are using more aluminum cans, and that's because they don't need to protect their steel industry anymore. Okay, um, and it gets down to economics. Okay, so steel is sort of going away, but there was this one thing where, geez they use steel in Japan, why do they do that, but anyway.

p7 04:26

Uh, what do they make cars out of in Europe, steel or aluminum? In general steel. But at the price of fuel in Europe, aluminum should be cheaper. So why do they make steel cars? Because everybody wants to sell into the United States, okay, and at the price of fuel in the United States, steel is cheaper. Okay, but actually in Europe, the Europeans, if they were only selling in their own market, all their automobiles would be aluminum, okay. But they couldn't sell in the United States then, they'd be priced out of the market anyway.

p8 05:06

Okay, um, anybody have any questions? I was kind of asking questions about the presentations, if people have ideas and stuff. Probably one day this week I might ask you to start writing down what your topics are and stuff, or letting me know, so think about it. Uh, remember you've only got about ten minutes for the presentation, that means ten overheads max. Um, and you probably ought to practice your presentation in front of a mirror or something to make sure you can do it in ten minutes. If you can do it in ten minutes with no audience, you will probably do it in twelve minutes with an audience, okay. Just making eye contact, or if someone raises a question in the middle of it for something, okay.

p9 05:53

Um, I remember as a young assistant professor giving my first few talks at some conference, I would go over that talk fifteen times and literally stand in the hotel room in front of the mirror and give the talk over and over. Now I don't have to do that anymore, but uh, maybe I should. But um, nonetheless, if you haven't had a lot of experience giving presentations, you need to practice. You don't just kind of walk in and do it, okay.

p10 06:27

Um, any questions, concerns? The um, we started talking about steels last time and hopefully we're going to finish up here in just a minute. I mentioned that they're made everywhere, it's the most, one of the most technologically advanced materials, if not the most technologically advanced. We know more about it than anything else just because of the size of the market. There are many types of steels, and I gave a handout. If you didn't get it last time you can see Jerry, cuz I gave her all the leftover handouts.

p11 07:00

Um, but this was one of the handouts and it talks about, oh that's various hardnesses of steels, but it talks about, oops, plain carbon steel parts, uh, and all the different types of steels. One of you asked, well why are there so many steels, and I pointed out last time that the carbon content of the steel determines its strength to a first approximation. And we go from a uh, this 1050 would be a uh half-percent carbon steel, and they don't have a 1010 on here, here's 1010, uh, for a carburized part that's got a tenth of a percent. The last two digits are hundredths of a percent of carbon, and typically most carbon steels are less than um, uh one percent carbon. So here's a stabilizer bar which is a 1090 steel, okay. Lawnmower blade of 1060. Uh, a spring, or yeah, spring at 1080, okay. High strength high carbon, low strength low carbon.

p12 08:06

But then we also have alloy steel parts that have typically different amounts of chrome, nickel, molybdenum and vanadium, and there's a whole wide range of these. And I pointed out the difference, there's a difference between uh, hardness, hardenability and hardening ability. If I have that, yeah I've got it right here. Um, I actually gave you a part of a dictionary of materials, where hardness is just the resistance to indentation. It's a direct measure of the strength. In fact the hardness measurement, if you measure it in kilograms per square millimeter or KSI, is about three times the tensile strength of the material, whether it's steel or copper or aluminum. Uh, the hardness is about three times the tensile strength for a ductile material.

p13 09:00

Then there's hardenability, which is how deep can you harden this. And I just mentioned this thing where I was asked by a professor over, would be now mechanical engineering but at the time he was naval architecture, he's working on the America's Cup. And he had a four-inch-thick, what do you call that, it's not a keel, the, it's a, it okay, maybe it's a keel beam, but it's on a big sailboat. Like that, the keel beam is not just on a big ship, the keel is, you know, the thing you build everything up from. In fact there's a keel beam on a 747 for example, it's a big aluminum beam and they build everything up from that. But in a sailboat, or the, this, that particular America's Cup, they had a big steel thing that created the resistance to being blown sideways, okay. And this was supposed to be four inches thick.

p14 10:05

HY-130 was a very weldable steel, or is, well it's more weldable than most steels, still got problems, but the Navy had developed it in the 1950s and 60s for submarine hulls and it needed to be weldable. And so it's got less than, it's got about two-tenths-percent carbon. Well it turns out, this was four inches thick, it had to have enough hardenability to be able to transform to martensite four inches deep, so you get uniform strength through the thickness. Whereas carbon steel, oh you might be able to get, depending on the carbon steel, a good eighth of an inch or quarter of an inch thick before you lose your ability to quench it fast enough. But when you start putting these alloying elements in there, it slows down the transformation and you can heat something and quench it such that you get hardness four inches, eight inches, even deeper than that in some steels, okay. Not that we need that in most things, but for this keel beam, they, this was supposed to be as I remember about four inches thick, and they actually were bending it in some of the trials, it was plastically deforming. Uh, it's better than snapping, but anyway, uh, they also wanted weight down in the bottom of the ship okay, uh, they want light weight, but they also want weight in the, if you're going to have weight you want it in the bottom. In fact they used to put lead in these ships in the bottom, which you're paying a huge penalty, but they just needed the weight towards the bottom.

p15 11:40

In any case, hardenability is the depth and distribution of hardness that can be induced by quenching. Hardening ability is what the maximum hardness can be, and it turns out depending on your carbon content, at about 0.6 carbon you get to the maximum hardness that you're going to get in most steels. Well, it turns out HY-130 only had about two-tenths-percent carbon, and it only had 130 KSI yield. If you went to 4340, which is readily available in big heavy sections, we use it to make all kinds of pressure vess—not pressure vessels necessarily, but components, high strength components in thicknesses up to ten inches thick. But it's got 0.4 carbon, 4340, okay. The 40 is four-tenths-percent carbon, and you can get much higher hardness, much higher strength, 180 KSI rather than 130. So they wanted to use 4340 but they didn't know how to weld it in that thickness, okay. So that's why he called me up one morning. Um, and I showed them how to weld 4340. It's not easy, not cheap, but who cares if you're building America's Cup, it's only one of a kind. You may only be sailing one ship, but they actually have multiple hulls okay on these things, it's a very interesting business. Um, we're not doing it right now, but yeah.

p16 13:06

Can I weld really well? Not particularly, I don't do it every day. I can weld, I was taught when I was an engineer at Bethlehem Steel by the technicians. I said I want to learn how to weld, you know, and they took me out and I did a lousy job. Uh, I always remember one time though, I was walking through my lab, not like it looks like now, but they used to have a bunch of boxy welding setups. And just like they have the um, the blacksmith shop in the basement, they used to have a guy, um, it started when I was a student. Um, not that I had anything to do with it, but my thesis advisor became the head of this little thing, but it was to help a technician who's the guy who had the welding lab before me left in 1968 when I came as a freshman. And he had a technician, Tony Zona, and Tony didn't want to go to Wisconsin with Professor Adams, so he just stayed at MIT, but he didn't really have anything to do, so he started teaching an art sculpture class okay, and they would do oxyacetylene welding.

p17 14:10

So I was walking through my lab this twenty years later, and I had four or five of my students all standing around one of these tables with the oxyacetylene torch. I said what are you doing, and they were trying to teach themselves to oxyacetylene weld. I said, they look like terrible, they look like turds on a plate, okay. And I said that's not how you do it, and I said give me that. So I took the torch and I put down an oxyacetylene bead, and it was a very nice looking weld bead, and I told them how to do it and stuff, and I walked away. What they didn't know is that was the first time in my life I had ever oxyacetylene welded, okay. But I did understand heat flow and welding, and I understood the thermal time constant for oxyacetylene welding as opposed to arc welding. If you take my welding thing I might even tell the same story. And so I knew you had to go slow, and I actually knew how slow you had to go, and if you realize that most people try to rush it, okay. But anyway, that's another story.

p18 15:10

Turns out, um, I also do a thing about reaction times. Uh, a lot of schools, if you go to Wentworth technology, they will probably teach you welding by starting you with oxyacetylene. And the reason is the thermal time constant is on the order of three to five seconds of that weld pool, so you got to go really slow okay, three to five seconds in one spot. But if you do that you get a good weld. For an arc weld, the thermal time constant is on the order of a tenth to three-tenths of a second. It's ten times faster, because the heat intensity is ten times greater. This all makes sense if you actually take the heat transfer part of my course. Um, and uh, I pointed out that was because of reaction time. When someone's starting, you don't start out on the highest speed of a video game right, you have to build up your skill and your reaction reflexes to various things, and a welder has to do the same thing.

p19 16:10

There are many places where they start people out and they give them one week worth of instruction in welding and they start them out with arc welds, and they make terrible welds for several days, and then um, they send them out there to make terrible welds in production. Um, but if you go through a four-month class, they will start you out with oxyacetylene for several weeks and then move you on to arc welding. So it's like learning how to play, because it's hand-eye coordination, let's face it, if you're manually welding. So yes I do know how to weld, am I the best welder in the world, no, okay. If I welded a pressure vessel, not only would it not pass the inspection most likely, but um, I wouldn't want to stand next to it when it's pressurized if it ever did get pressurized. Uh, um, but it's partly because you really need to do it on a regular basis. A typical um welder, I think the codes require, depends on which code it is, require that if you have not welded in the past year you must be requalified, because you haven't kept up your skills, okay. So there's skill, but I don't practice sufficiently.

p20 17:23

All the rest of you could weld if you wanted to learn. Many students come to me and say oh I'd like to learn to weld, so I send them off with one of the technicians, and usually after about one or two hours of making lousy welds they say okay. Very few students over the years have ever gone for more than uh, a few hours worth of practice.

p21 17:44

But getting back to steels, there's hardness, there's hardenability and there's hardening ability. Hardness is merely how strong it is, hardenability is how deep can you get a hardness, and hardening ability is how, what's the maximum strength you can get under the best conditions, okay. Um, and because of that I pointed out, we have many many steels because they're used in such high volume that people will literally demand, I want a tenth of percent nickel more or a tenth of percent nickel less in the steel, and that, am, sound like much, but a tenth of a percent nickel is 10 pounds, 10 pounds in 20—pounds in a ton. And at $10 a pound, that's $200 a ton, so it makes a difference, particularly if you're buying 100,000 tons, you would be talking $20 million. So there are many steels.

p22 18:47

That, uh, does that answer the question? I think it was your question several weeks ago, I said I would cover why we have so many steels. I also talked about um, the Achilles heel of steel being corrosion, and we'll get to that. Um, more, but I think the easiest way is to actually jump into stainless steels as the next class of material, separate of the carbon and low alloy steels.

p23 19:16

The stainless steels, and this comes out of a book by John Sedriks [Sedrik] called Corrosion of Stainless Steels, and he has a genealogy of stainless steels. The first stainless steel was basically what we call 304 stainless steel, was discovered like 1907 or so I think, in Sweden. Um, they were back then they were just learning how to make steels of different compositions, and they were adding different things. Uh, you may have heard of 18-8 uh stainless steel. Basically instead of 19 chrome 10 nickel, 18 chrome 8 nickel, it's all within the range of this 304 stainless. Uh, it's either 40% or 60% of all stainless steel is 304 stainless steel. I think it's 40% of all stainless steel and 60% of all the austenitics. I actually, I probably tell you about the different types of stainless steel. There is austenitic, ferritic, martensitic, duplex, precipitation hardening.

p24 20:53

And this actually goes through all of those. So the austenitics have face-centered cubic crystal structure, they're non-magnetic. Um, and like I say 304 is the most common of all the stainless steels. Uh, but you'd like to improve on that, uh that alloy that's been around for 100 years. And so one of the things they do is they might add sulfur or selenium for machinability. We actually also add sulfur um to plain old carbon steels for machinability. You end up with sulfur inclusions, and when the chips are coming off the lathe or the milling machine, the, those inclusions which will tend to form stringers will cause the chips to break up, and you don't get this long pigtail corkscrew that'll whip around and get in things and cut people's arms off and things like that.

p25 21:51

Yeah, sulfur, for sulfur control, that's in steel in general, okay. Um, steel cannot be made without manganese. Um, although we do make some now that have extremely low manganese, but to give you the history on that, um, but when, finish asking your question. If you had a question about why is manganese in steel, it's not stainless steels per se, it's actually steels have to have manganese for sulfur control. But was that your question? Yeah. Okay, I'm looking for, here it is. Well there's nothing, one I wanted, but anyway I can use this, it's all marked up, but um I should have another one in here somewhere, it's not all marked up.

p26 22:43

Um, in any case, the um, in uh, the, you, to make steel in the old days, actually even today, you take iron ore, coal, and limestone and you put them in a blast furnace, you make cast iron, okay. The problem is the coal may have two or three percent sulfur in it, and if it does the sulfur ends up getting in the steel. Well iron sulfide is a very low melting alloy, and if you take that steel without any manganese around and you try to roll that steel, when you heat it up to the hot working temperature of 1,800 to 2,000 degrees Fahrenheit and you try to roll it, it will just shatter, because it has liquid grain boundaries of iron sulfide along the grain boundaries. And so that's not good, um, but that's what happened in the 1880s, and they didn't understand that it was sulfur until they were using different ores in different parts of the world and some people had this problem, some people didn't. They started analyzing it and they found manganese, the steels with three-tenths of a percent or more manganese would not shatter.

p27 23:58

And then people started analyzing it, they had learned about metallography in the 1880s, they started doing metallography to look at where the sulfur was, and they found you formed manganese sulfides. If you look at the periodic table, all the elements in the periodic table, the only element, the only two elements that have higher melting sulfides than iron are manganese and molybdenum. And it turns out when you add molybdenum to iron, it turns out you get a low melting sulfide. When you add manganese to iron, you get a high melting sulfide that melts above the melting temperature of the iron. The only element in the periodic table that does that. There is no other element that will tie up the sulfur impurity. So you have to either get rid of your sulfur impurity, or you have to put manganese in the steel.

p28 24:53

And so I guess um, one of the questions is why does steels contain manganese? Well they contain manganese because if you, here's the clean periodic table anyway, if you didn't have manganese, the steel would not be hot-workable. It would just, it's like if you were trying to hot-work a Slurpee, which is a liquid on top of a solid ice particles okay. The steel would have that kind of consistency, when you compress it would actually crack apart, okay. So, it's not exactly like a Slurpee, it's not the best analogy, but you basically have liquids along the grain boundaries.

p29 25:30

So um, a steel officially is an iron-carbon alloy, but it always has some manganese, usually about three-tenths of a percent up to two percent, and it has some silicon, quite often, doesn't have to. The silicon is there to get some of the oxygen out, uh manganese will take some of the oxygen out. I don't want to get into talking too much about steel because I always talk, already talk too much about steel. But there is some basic metallurgy in the manganese. Historically they would say the manganese had to be 8 to 10 times the sulfur content because you have to tie up the sulfur. Nowadays we've learned to get the sulfur out of the steel while we make the steel, using the right types of fluxes. And one of the guys who taught the world how to do this using the principles of physical chemistry was John Chipman of the Chipman room. I mentioned the other day that the world learned how to make steel from John Chipman. Well, he was the one, he and Nick Grant who was a professor starting here in 1946, the two of them basically published some of the first articles on control of sulfur and steel in the 1930s. People knew you put manganese in, but they didn't know how much manganese, and they were wasting a lot of money on manganese. Virtually all the manganese in the world goes into alloying for steel.

p30 26:53

Um, uh, then they also next thing they did is they taught people how to get rid of the phosphorus in steel, how to control the phosphorus. A lot of the iron ores in Europe are high in phosphorus, and phosphorus can be good in steel in small quantities, but it's bad in large quantities. So anyway, there's a lot about the alloying elements of steel. People teach whole courses at some universities on this, not much anymore, but there's still a few universities.

p31 27:20

In any case, you start with 304 and you can add sulfur, and in stainless steels you add selenium because you can afford it, you can't afford it, most carbon steels. They have some carbon steels with selenium, but only if you can't have sulfur for some other reason. Selenium, if you look at the periodic table, is beneath sulfur isn't it, yeah. So all you've done is go down the periodic table, and you get these inclusions in the steel that help break up the chips. When we machine brass, we used to put lead in a lot because it would break up the chips. Free-machining brass is a leaded brass, and it forms lead inclusions rather than sulfur inclusions, uh, in steel. But anyway, so that's 304 stainless. What do we lose when we do that? It's not weldable anymore, you can't weld the 304, 303 stainless. But it's the same composition, just has high sulfur.

p32 28:21

I meant to bring in, I have a, actually it's a carbon steel uh hinge from a Sub-Zero freezer. Anybody know what a Sub-Zero refrigerator is? Yeah. Uh, well actually, can I guess, it's an $11,000 home refrigerator okay. Very high-end. Obviously none of you have millionaires for parents or multi-millionaires for parents. Um, I have a son who just bought a house and it has a Sub-Zero refrigerator, and a Wolf stove goes for only five or six thousand. Um, very high-end uh kitchen appliances. But I have a Sub-Zero hinge because someone made a mistake. They had this little pin that was supposed to be made out of carbon steel and welded to a carbon steel plate to make the hinge, and someone made a mistake, they used 303. When they machined these little pins that had to be welded, and when they went to weld them they had cracks. They didn't see the cracks, they plated them, put them on the refrigerator, and people bought $11,000 refrigerators and the doors started falling off. They were not happy, okay. And these are the type of people who have enough money to sue you, okay. So if you're going to shaft somebody, do it to poor people, they can't afford attorneys. Um, but in any case, uh, that actually cost about $7 million worth of repairs.

p33 29:46

Duplex stainless steels, they increase the chromium and lower the nickel to get higher strength. You can double or triple the strength by going to a duplex. A duplex stainless is merely a mixture of about, let's call it 50/50 or 30/70 or 70/30, of austenite and ferrite, ferrite being a body-centered cubic magnetic phase. So I probably don't have a magnet with me, but this is a piece of duplex stainless steel. It came off a centrifuge that was used to separate starch from other things. Um, when they started, well, one of, they started selling a lot of these centrifuges when they started making methanol from corn, okay, because they had to separate the starch from all the other liquid and the corn and stuff. And so they used a big centrifuge, and this thing weighs about a ton and it spins around like 3,000 RPM, and it's about four feet in diameter. It's made out of cast duplex stainless steel because they needed high strength.

p34 30:52

Turns out, if you're not careful in the duplex stainless steels, you have higher chromium and lower nickel, and you have the ferrite which can get hydrogen in it. But the real problem is you can get iron-chrome intermetallic phases which embrittle it. You want to see a brittle steel, here it is. Did have a little bit of ductility, if you look on this edge right here you'll see a little, if you hold it up you can see the curvature of the centrifuge, but right on this edge you can actually see um a slight change in the curvature. If you look carefully you actually see the grains where the large grains of this casting started to stretch, what we call orange peel okay, because it looks like a surface of an orange, sort of, in some cases. Um, can be extremely brittle if you don't do it properly, but you can get 100 KSI strength rather than 30 or 40 KSI strength for the 304. So you can triple the strength.

p35 31:50

What do we use it for? Uh, they use it now, it's become very popular in the last 15 or 20 years for downhole sour gas applications where you need high strength deep wells in the oil business. Before that it was very specialized, they used it on the catapults of Navy aircraft carriers, okay. Need a high strength corrosion resistant alloy. It has better chloride stress corrosion resistance than 304, and so if you had a chloride environment, uh, and you wanted high strength, you could go to duplex stainless. Hard to weld, but high strength and uh some interesting mechanical properties. But you pay a price in some other things, can be toughness prices, and can be a welding price.

p36 32:38

Precipitation hardening stainlesses have been around for a long time. I thought I had it, yeah, I do have it with me. Um, so here's a $300 pair of scissors, actually probably $600 now. These are medical grade scissors, they actually have a nickel-base uh hard alloy insert brazed in them. Um, but because they are medical, instead of being a $40 pair of scissors, they are a, well, 25 years ago when I got them they were a $300 pair, I bet they're $600 now okay. Precipitation hardened stainless. Okay you wouldn't do that for a regular pair of scissors, what do you care about it on the handle, but you know, the doctors want to know if they're paying $600 for a pair of scissors they're getting the best quality. Why it's precipitation hardened, if you're going to put inserts in the cutting edge, I don't know why you need to precipitation harden it, but anyway. The Navy used to use these on hydrofoils okay, for the ships, need corrosion resistance, aluminum would just erode away. They make the base ship out of aluminum for lightweight, but the foils that are going through the water. I have no idea on the America's Cup because those are hydrofoils now basically, I have no idea what they use, but they might be using a precipitation hardened stainless. You can get 180 KSI out of these steels okay. Uh, you have to go through a fancy heat treatment, harder than regular carbon steels or alloy steels, but you can get tremendous strength.

p37 34:11

Um, 201 and 202 add manganese and nitrogen, and lower the nickel, to get higher strength, so you can get uh 80 to 100 KSI, and it's cheaper, manganese instead of chrome, okay. Manganese is a lot cheaper than chrome. Um, I don't actually see these that often, but in certain applications, if you go a really high manganese, like 12 to 13%, you get what they call Hadfield's manganese steel, which is not officially a stainless steel, but it is an alloy that work-hardens so rapidly that you literally cannot cut it with a saw. Not that it's not easily formable in its annealed condition, but you go to saw it and as you're cutting those chips it hardens and about triples its strength, it will dull saw blades. One of the uses of Hadfield's manganese steel was prison bars, okay. You got a file, yeah you want to dull your file okay, try to file through a Hadfield's manganese steel, because it transforms as you're work-hardening it to this hard martensitic steel.

p38 35:25

Uh, in fact the martensitic steel, you take the nickel out, lower the chrome down to around uh 10 or 12%, and you get the 403, 410 and 420 steels. Many of the uh scissors and um medical instruments are actually made out of the martensitics. Yeah, they're magnetic. Martensite is basically the same as ferrite, except you've trapped carbon in the lattice and it becomes body-centered tetragonal, but very barely body-centered tetragonal rather than body-centered cubic. This is face-centered cubic, this is body-centered cubic, this is body-centered tetragonal. Duplex is FCC plus BCC, that's why you call it duplex. And precipitation hardening is basically FCC with precipitates, which do the hardening, okay.

p39 36:24

If you want to talk general strength levels, you might be talking, uh, let's say 40 KSI. Ferritic might be 40 to 80, depending on how work-hardened they are. Um, these actually could be heading up towards 160, if you take piano wire, you take this and you deform it and deform it into wire, you can get up to, actually you can get up to 300 KSI, but body-centered tetragonal you can be up to 180 KSI, but you really like to be less than 120 for hydrogen embrittlement reasons. Duplex typically about 100 KSI, good ductility, good chloride resistance. And precipitation hardening 180 KSI with good ductility and reasonable hydrogen embrittlement resistance.

p40 37:21

Um, pardon me, the only ones that hydrogen embrittle are the ones that are BCC. That's the, uh, frankly it's still a matter of opinion among metallurgists what causes hydrogen embrittlement. There's four or five theories out there, strain in the lattice, um, I suspect when we get really good at, um, oops, what's going on here, oh, that's okay. Um, when we get really good at quantum mechanics, um, on a larger crystal okay, rather than just a few atoms, we might be able to figure out what hydrogen does in terms of embrittling the body-centered cubic structure. You can embrittle the austenitic structure, the FCC structure, but it takes 50 to 100 times as much hydrogen to do it, so we don't usually talk about hydrogen embrittlement unless it's ferritic. So hydrogen embrittlement problem, hydrogen embrittlement problem. These top and bottom basically are much more resistant to hydrogen embrittlement.

p41 38:44

So um, uh, then if you have corrosion resistance in aqueous systems, particularly with chlorides, you add molybdenum for pitting resistance, and so you might have 2% molybdenum in a 316 stainless steel. Well, let's say the 304 stainless steel cost you $2,000 a ton, with 18% chrome and 8% nickel, rather than $400 a ton for carbon steel. So is it, now, this is just as expensive as aluminum, now, 304 stainless. Well 316, you can add 2% molybdenum. I don't know what molybdenum is going for now, $30 a pound, $20 a pound. 40 pounds for 2% per ton, okay, and you're now adding $1,200 a ton, okay, to what cost you, you know, $2,000 a ton, $14,000 a ton versus $2,000 a ton. Big jump in price to go to 316. But you can also go to 317, which they used to use for a lot of medical applications, it's 4%, so you're talking $24,000 a ton.

p42 40:01

And to show you the difference, this is the most dramatic. People use 316 all the time because it's weldable, it's got stress corrosion cracking resistance, but you have to be careful with these, because they're all cracked. Okay, you can see the cracks in this thing, it's all rusty. This came out of the ballpark, uh, not ballpark, Fenway Frank hot dog cooker over here in Everett, okay. So the hot dog cooker is about the size of this room, it's just a big steam oven okay. They, you know, chop up all the types of meat you'd never want to eat and make a slurry out of it and extrude it, put it inside the casing, then they have to, with a lot of salt, okay, that's why you need chloride resistance. And the original hot dog cooker was all made out of 316 stainless steel, had been around for 10 or 20 years no problems. They had to modify some things, it was all specified to be 316, and they put the new stuff in, and within several weeks it was getting rust, the new stuff was getting rusty. It was cracking, and they called me in, and I actually got to see how you cook hot dogs okay. Anyway, it's a lot of interesting things in doing this type of work. Um, and they asked me to figure out why. So we brought it back, turns out it's 304. Someone used 304 rather than 316. Maybe a harmless mistake because, you know, 40 to 60% of all austenitic stainless steel is this, and these things are obviously a lot less, it could be any one of these others, but this is a lot less in tonnage, and there's just a mix-up. But that 2% molybdenum went from something that would look stainless for 15 or 20 years or more in this environment, to something it could only last a couple of weeks. That's the most dramatic example I've seen. Plenty of failures where people use 304 rather than 316, but nothing this dramatic, okay.

p43 42:05

Uh, so uh, the other thing is, people will often bring me a piece of steel and they say it's supposed to be stainless and it can't be because it's rusty. Well, I'm sorry, it's stainless okay. There are two types of stainless, okay, rusty and unrusted okay. Uh, and actually more specifically, if you look at the galvanic series, which I didn't bring with me, actually I probably did, I probably have it in this book right here. Um, I don't know if I can find it quickly, but anyway, um, galvanic series is what dissimilar metals will um attack one another in a series, uh corrosion series.

p44 43:00

And I'm sure I can't find this that quickly, it is in here. But anyway, um, the galvanic series basically, he's got magnesium is the most electroactive and carbon as the least electroactive, and you got everything in between. Platinum is close to um uh carbon, zinc is close to magnesium, copper is somewhere in the middle, iron somewhere in the middle. Turns out stainless steel comes in two varieties, there's active and passive. What gives stainless steel its corrosion resistance is a protective oxide, and I will just call it for our purposes Cr₂O₃·nH₂O, okay, it's a complex oxide on the surface of the stainless steel that has very good aqueous and high temperature corrosion resistance in most cases, except in the presence of chlorides. Active is one where you take off that protective oxide, and now you're exposing a steel without a protective oxide. This one, active, rust, no rust. So there's lots of ways to take it off.

p45 44:26

Yeah, fluorides. Fluorides are terrible, worse than chlorides, but there's not a lot of fluorides that we expose things to usually, okay. Chlorides are everywhere, fluorides are not everywhere. Is that your question? Okay, but yes, fluorides are worse than chlorides, we just don't have them that often. In fact, the best way to take this chrome oxide off is not hydrochloric acid, it's hydrofluoric [hydrochloric] acid, okay. That's the most effective.

p46 45:03

Yeah, you had a question back in the back. Uh, 50 nanometers, very thin okay, doesn't take much to take it off. Stick it in hydrochloric acid, you take a passive stainless and turn it into an active stainless, okay, within minutes. Yeah, it can be either one okay, depends on whether it's got the chrome oxide. 316 with the molybdenum in it is resistant to three times the chloride concentration of 304. 317 is probably resistant to five times the chloride concentration. So, uh, typical atmospheric corrosion resistance of 304 might be a thousand parts per million chlorine, okay, in an atmospheric sort of, if you were raining salty water or spraying salty water. 316 would be 3,000 PPM chlorine, 317 might be even higher.

p47 46:10

The reason you could use it in the body, the, anybody know what the salt content of the body is, the body fluids? 150 what, millimolar? Yeah, that's probably his, millimolar, but I don't know millimolar in PPMs. Seawater is about 30,000 parts per million, and your blood is about 30,000 parts per million. When I was in fourth grade they taught me that's proof that we evolved from the fishes, because our blood content was the same as saltwater, if you believe that. Uh, I think it was a little oversimplified, but nonetheless, basically inside our body is like seawater in terms of chlorides. There's one important difference, we don't have the same oxygen content that they have at the surface of the ocean. We might have an oxygen content similar to the bottom of the ocean. I guess that proves that we started out not a surface fish, but a deep sea fish. At least, you know, we were deep sea predators, top predators or something.

p48 47:12

You, um, in any case, this is better and better chloride resistance. It really depends on not just the chlorine but the oxygen content, because the chlorine screws up the protective oxide layer when it forms. The nice thing is, usually even in 304, if you break off that protective oxide layer, which is very easy to do, it will reform just in the air as a protective layer, unless there are some chlorides around.

p49 47:40

So if you want to go over here to the New England Aquarium, how many people have been to the aquarium? Anybody looked, what's the outside made out of? Stainless steel. And what was the specification? They wanted to make the outside look like fish scales, and they were going to use 304 stainless panels, and they were to be passivated. I read the spec, okay. And passivation means in the mill they took the coils of steel before they had been stamped into fish scales okay, um, and they passed them through a bath of nitric acid, a solution of, not concentrated, but about a 5 or 10% solution of nitric acid. It's an oxidizing acid. So you take a stainless steel, you clean it in the nitric basically, but you also oxidize it to form the protective oxide. It's now passivated. You can passivate 304 stainless in 15 minutes in nitric acid.

p50 48:42

So they bought the passive stainless steel, they cut it into panels the size of these fish scales, they then took a grinder and they brushed on this texture. What happens when you brush on the texture? You're taking off the oxide scale, protective oxide. What happens when you then go put it in a marine environment, Boston Harbor? You get oxidation right along those little scratch marks. And if you go look carefully, very few people do, but I had to, well, I had to, I got paid for it. Um, you will find rust, okay. In fact, if you go around the side to the loading dock where the trucks come in to, you know, drop off the food and stuff, um, you'll see where they actually scraped up against the stainless steel with their bumpers, and you have big bands of rust, okay, on the stainless steel, because mechanically abraded away the protective scale. In a chloride environment, the protective scale did not reform like it would if it was inland, okay. But right there on the coast.

p51 49:50

And so we had to figure out, one, we had to figure out how to clean it. So I tried um citric acid, that will also just like nitric acid. We had to have something that was not going to destroy Boston Harbor because all the environmentalists. So one of the things I tried was Diet Coke. Didn't work all that well, but Diet Coke is phosphoric acid okay, and it also is a cleaning solution. But who could complain about dropping Coke into Boston Harbor, right, not that kind of coke, the kind of drink okay.

p52 50:23

Uh, I had a case like that once, it was in New York City in one of these building, this 20-story buildings, above the approach to the George Washington Bridge, and they had a weld crack. And when the cab driver let me off, I didn't know the addresses in New York City, he dropped me off, there's all these little empty capsules all over the sidewalk okay, which were basically cocaine capsules right. Uh, and uh, they told the homeowners or the condo owners, uh, at a meeting before I was there, that they had a crack in the basement, and after the meeting a bunch of the homeowners wanted to go down there and get some of the crack, okay. It's a true story, okay. Uh, I got to see the crack, it's 8 feet long, had a piece of it in my lab here anyway. Uh, but I didn't eat it.

p53 51:15

Um, in any case, you have this interaction between oxygen and chlorine in stainless steels. Chlorine is not good for the steels, you can add molybdenum at a tremendous cost, but they do all the time okay. Then there's the L grade of these things, all three of these, which means extra low carbon, and that means that you can weld it without uh getting chrome carbides that also destroy the corrosion resistance. And I'll talk about that next time, but let me tell you that most of the stainless steel we make nowadays is made by a process that gives us very low carbon, less than 30 parts per million chlor—300 parts per million carbon. Um, which prevents these chrome carbides and loss of corrosion resistance for this reason that occurs during welding, and I'll talk about it some more.

p54 52:09

But um, the process for that, called argon-oxygen decarbonization, is now the process by which every bit of stainless steel, most nickel-based super alloys uh in the world is made. And it was developed right down here in the basement of building 8 by a guy named Krivsky [Krivsky], who was doing his doctoral thesis under John Chipman. Um, and they were basically doing basic research, they were bubbling argon through molten steel and looking at the carbon monoxide reaction, carbon and oxygen in the steel, and finding out what would happen. And they found if you bubbled argon rather than air, you would have a low enough oxygen potential in argon, the oxygen potential is very low, you would basically pull carbon and oxygen out. And it used to cost a small fortune to make 304 L stainless. Nowadays with AOD steelmaking, where they bubble argon through there, you get low carbon without any difficulty, such that everybody, we hardly ever make 304 stainless okay anymore.

p55 53:24

But back in 1950 it was a super premium to get 304 L, because it doubled your steelmaking time, you had to have multiple slags to get the carbon out. Um, but, so the process wasn't patented at MIT. Krivsky was doing this basic science research, he went to a company and he realized that hey, this process could be used commercially to get the carbon out of stainless steel. And since about 1960, virtually every st—the steel in the world is made by argon-oxygen decarbonization, where you just bubble the argon through and take the oxygen and carbon out of the steel. So, and I once one time estimated that's worth about a billion dollars a year to the U.S. steel industry, or to the world steel industry, okay. I'll see you tomorrow. Well, I thought I was going to finish up steel, well.