AM_F2019_02

Today I'm going to start talking about additive manufacturing, but that's not really the topic. The real topic is more one of why people get on bandwagons of things. And I have to give you my bias. If you serve on a National Research Council committee in Washington, they have to have a bias discussion of all the members of the committee. It takes about two hours and it seems like a colossal waste of time. But do you even know what the National Research Council is? Have you heard of the National Academy of Sciences?

So the National Academy was chartered by Abraham Lincoln in 1863 to give advice to the government on science. And then in 1964 [1964] some engineers got together and said, well, engineering is important too, we should have a National Academy of Engineering. So they formed the National Academy of Engineering. And a few years later the medical doctors got together and said, well, we should have an Institute of Medicine. And there's only about 2,000 members of the National Academy of Sciences, 2,000 members of the National Academy of Engineering, and about 2,000 — even less, it used to be less, but I think they're getting up to around 2,000. So really you're talking about people who, after being recognized by their peers by being elected to the National Academy, they're kind of one in a thousand, okay, because there's about two million engineers in the country, and if there's only two thousand members of the National Academy of Engineering, 2,000 over two million — I did that in my head, okay — it's one in a thousand.

If you've heard of the Royal Society in Britain, okay, well, it's the U.S. equivalent of the Royal Society, which are advisors to the government. And their building is right across from the Lincoln Memorial on the mall in Washington DC. They are not a government agency, but they're sort of quasi-governmental in the sense there's very little money they can take from anything except the government to do paper studies. They do the paper studies through the National Research Council. And so a lot of — they will put together a team. I worked on a team most recently. It had to do with the Department of the Interior, and it had to do with bolts, bolted connections in undersea oil production. They didn't want another Macondo failure. Did you know we've had about four near-miss Macondo failures in addition to the Macondo in the last 15 or 20 years? It's all public knowledge, but most people don't know that. But the Department of the Interior [Tom said "charrier"] knows it because they're responsible for it. And so they put a bunch of people on the committee to try to figure out what they should do to try to make sure we don't have another failure like that.

I served on one when they wanted to redo the Alvin submarine, you know, the deep-diving submersible they have at Woods Hole that goes down and finds the Titanic and things like that. And the National Science Foundation had 25 million dollars to build a new one, and they wanted to figure out, well, what features should it have, and can we afford to build it, and things like that. So you sit down a committee, you don't get paid but they pay your travel expenses, and you meet a lot of interesting people, okay. So they have a bias discussion.

So my bias discussion here is, I consider additive manufacturing is sort of a fad, okay. Now most of you students don't — how many of you are working in additive manufacturing? One, two, okay. Many people actually been working in additive manufacturing for many years, and we'll go through that. But I consider it a fad, just like high-temperature superconductors a few years ago were the fad. That's all you could get research money in. Now it's sort of calmed down. Structural ceramics were a fad 35 years ago. There's lots of fads that come along. Nanotechnology is a fad. What do you call the little fibers — nanotubes of carbon? Carbon nanotubes, that's a fad, okay. And I tend not to like to work on fads.

And so I actually had worked on some of these things before they were popular and had the name of additive manufacturing. And I worked in manufacturing, but I didn't call it additive manufacturing. But when it got to be a fad I really didn't want to have anything to do with it. And a company called Digital Alloys came to me — or they called me up — and said, we'd like you to be on our science advisory board. And I said, no thanks. And they said, well, can't you at least come for a visit? I said, where are you? They said, Burlington, which is all of a half an hour away, there's only about ten minutes from my house. I said, nope. And they called me back, and anyway, in the meantime I learned that one of the — it wasn't my student, but a graduate student who used to do the IT for my lab — was working there. And I said, okay, since Sal's working for you I'll listen to your story, but you're gonna have to come to my house, okay, I'm not going to travel to see you.

And so the guy came and we sat down at the dining room table, and he showed me some things they were doing. I said, oh that's interesting. He said, well, will you come for a visit? I said, okay, if Sal's there I'll come for a visit. So I went to see what they're doing, and I was impressed that they had an advantage over a lot of these other technologies. And so I agreed to be on the advisory board. And Professor Slocum and Professor Hart and Professor Sadoway and I are the four faculty members from MIT on the advisory board of this company. We're going to use Digital Alloys' website, so I'm telling you this — it's full disclosure as sort of the course materials here.

And in general I think that most of the stuff on advanced — on additive manufacturing is sort of schlock, if you know Jewish slang. I mentioned it yesterday to a colleague in the department. He called some of this stuff a scam, okay. We're gonna have a guest lecturer come — well, not a guest lecturer, but a guest tomorrow who was my house tutor when I was a freshman. One of the three brightest people I've ever known. He was a vice president at Honeywell before he was 40 years old, a Fortune 50 company. He became CEO of several chemical companies, and now he's retired. And he and his wife live here in Boston during the summer because of health problems they live in Florida and a resort area other times of the year.

And so I asked Dave as we were having dinner, I said, well, why don't you come do a lecture, my guest lecturer of the class. He says, what are you doing? I said, I'm — just this semester I'm gonna do additive manufacturing. He says, that's a hoax, okay. So not everybody believes additive manufacturing is great.

But I saved this from last week. This is Advanced Materials and Processes, a magazine for materials engineers, and here's the whole — our whole magazine is on additive manufacturing. "Metal printing trends and tactics. Unless you've been enjoying an extended off-the-grid sabbatical, you already know that metal additive manufacturing is continuing to post double-digit growth year after year. Every materials engineering magazine, website and technical conferences addressing this topic," blah. That's the editorial. "Revenue from metals for additive manufacturing grew 42% in 2018 to roughly 262 million, up from 183 million in 2017, 126 million" — blah, okay.

That can sound impressive — a quarter billion dollars. What's the value of manufacturing in the United States? Two trillion. So you can do the math of how much a quarter billion is compared to two trillion. Still a relatively small fraction. And to show that you have double-digit growth in a very small sector is a relatively small — not necessarily the greatest thing in the world. So there's a lot of hype and things, and we'll go through some more of that.

But to start out, I want to point out — by the way, I got to pass around this sign-in sheet, so take attendance, make sure everybody's attended things. There are generally two — well, the Japanese divide materials into two classes: structural materials, which is course is supposed to be about, and functional materials, which is what 75% of the department works on. Chemical, magnetic, electrical, all kinds of specific properties — in many cases medical, okay, type things. But structural materials, you're talking about mechanical properties: strength, toughness, corrosion resistance, aesthetics, and things like that. Structural materials are used in huge volumes. Most people don't have — have a hard time even conceiving of the volumes of these materials.

There was a guy, Jack Westbrook at General Electric, he graduated this department in 1962. He put together this curve when he was at General Electric for a secret General Electric report. This is pounds per year versus dollars per pound. I've actually redone this thing as of about five years ago, it still looks the same, and I talked about in some other courses. Here are some of the materials. The material that's used most as a structural material is stone, a very important material. But not used in very large volumes but very expensive as diamond down at this end, so far as that goes.

And I like to give two quotes. One's from Bob Sprague, who's passed away now, but he used to be head of materials for GE Aircraft Engines. "Whenever you first hear the properties of a new material, write it down, because those are the best properties the material will ever have," okay. And then Jim Williams, who replaced Bob Sprague — and he was dean of engineering at Ohio State — a titanium, one of the world's titanium experts: "Whenever you hear the cost of a new material, write it down. That's the lowest price the material will ever have," okay. So people are always touting new materials, they oversell new materials. Dave Hill's going to come tomorrow — 25 years ago told me it's "glam tech," okay. People tell you how wonderful these new materials are, and if you really know something about them you realize, oh, they're interesting, but they don't meet all the properties you really need in practice.

So what's the history of additive manufacturing? Well, I found in the Welding Journal of 1922 this additive manufactured part, whereas people were laying weld beads down on cast-iron — weld beads. There are little cracks on either side of this thing, but they were essentially building up pads and doing additive manufacturing. So that's the earliest modern example. Can anyone think of an earlier example of additive manufacturing? Yeah.

Student: Bricks.

All right, yeah, you're absolutely right. You got it. You hit it. Pottery, 25,000 BC, okay. That's additive manufacturing, isn't it? May not be metals, but certainly additive manufacturing. You take clay and you, you know, build it up and shape it to what you want, okay.

If we look at structural materials today, there's what I call the billion-ton-per-year club. There are only four structural materials. It's actually only four materials in the world, unless you include water and air, okay, which would be at the top of the list. But structural materials, there's only four materials. I already told you about stone — leads the list at 53 billion tons a year. Now there's lots of stone everywhere in the world, different types of stone. You don't transfer it very far because if you look at Jack Westbrook's plot, it's worth about two cents a pound, and you can't transfer it a lot of stuff. So you tend to generate crushed stone somewhere close to where you are.

Concrete is tied for number two. I like to pass around this piece of concrete. [Tom hands a sample around the class.] When you grab that, what's the first thing you feel about it? It's super light. That is a piece of fire brick, for lining — or fire concrete, for lining a chimney. And you want something that's got lots of pores in it so it's a good insulator, okay. They can engineer concrete in all kinds of ways. That one was specifically — like, it's easy to bring to class because it's light, and it's light — it's amazingly light when you pick it up.

Number — tied for number two is engineered wood products. I'm not including all wood. There's a much larger billions of tons a year wood if we include wood as a fuel. But as an engineered wood product, we can make better I-beams out of wood by essentially making them like we do plywood, an engineered wood product.

Student: [inaudible question about chipping]

Yep, it's easy to chip because it's full of pores. You just — think of anything else that's full of porosity, it's going to be on a — if you use what's called the rule of mixtures, if it's 90 percent pores compared to a piece of concrete that's 10 percent pores, it has one-tenth the strength. That's the rule of mixtures. Sometimes it's worse than that. This is engineered wood products. If you look at it, what they do is they take a log and they essentially shave it like taking off a jelly roll, okay. And they glue it together. You look at the pieces in there, you'll see they got them crossed at 45-degree angles and stuff to get rid of the fibrous nature of normal wood. And they don't use much adhesive in there because that's the expensive product. But engineered wood products are a huge business.

The only metal — and people always joke, particularly the mechanical engineers I'm told over there, the students like to joke that "Tommy Eagar thinks steel is wonderful." It's not that I think it's wonderful, but among the metals it's far above anything else at 1.6 billion tons. And in fact, if you go look at the next closest metal, it's called aluminum, at sixty million tons, okay. Steel is 95% of all metals made. Why? It's got great toughness, it's got great strength, and compared to the other structural materials, steel has better toughness and strength than anything else.

In fact, what do we do with reinforced concrete? We reinforce it — rebar. You put rebar inside the concrete because the concrete's good in compression, lousy in tension, okay. And the oldest reinforced concrete structure in the country is right up the road, the Harvard football stadium, 1910. That's when they started putting steel rod in concrete to give it strength in tension, okay. And in fact, even today, concrete is much — I just showed you the data — concrete is three times the amount of steel. And in fact if you look and think on the volume, it's lighter, so it's more like six times the volume. That's why in certain parts of the world you can't afford steel, okay. You can only afford concrete. Although you can ship concrete around the world as far as that goes. Any questions? Yeah.

Student: [inaudible question, apparently about availability of materials for concrete]

In most places. But for example, Hawaii has not — soft volcanic soil. But you're right. But it turns out one of the biggest costs in producing concrete is the energy cost of taking the calcium carbonate and heating it up to a thousand degrees centigrade, burning off the CO2, and making burnt lime, which is calcium oxide, which you mix with water to get calcium hydroxide, which with the aluminates and silicates is concrete, okay. And if you use some types of volcanic ash, you get Roman concrete, which is some of the best in the world, okay. I mean, Professor Masic over — Admir Masic — yeah, in civil, he's studying Roman concrete. But people have been studying it for 2,000 years. It's still some of the best concrete in the world, in part because of the other things they add to it. There's — concrete chemistry is too complex for most material scientists. That's why the civil engineers do it. They know no chemistry, so they do it, okay.

No, but Professor Oral Büyüköztürk [Tom said "or B is austere"] over there has a 25 million-dollar program from Kuwait to design the next type of concrete structures. And just like they can tailor concrete to make it ultra light, they can make it dense, they can fill it up with things so that it has radiation shielding for certain types of, you know, if your nuclear reactors and things like that. You can specialize concrete in ways that most of us don't think about, okay. Other questions? Yeah.

Student: [inaudible question about how the light concrete is made]

I don't know for sure, but I bet that they basically are taking something — adding something that gives off CO2 when it mixes with water or whatever, so that you basically — it's like opening a soda and having all the fizz and stuff. So you're gonna generate bubbles, okay, gas bubbles inside, but probably by a reaction that gives off CO2, okay. Just like maybe making whipped cream out of a can, right? I mean, it's the same thing, right? So that's nitrous oxide. Anyway, good question.

Yeah, this is very light. I've never seen concrete this light before. But you can do all kinds of things. I bet just that one's eighty percent air, okay. But there are other cases — you can get anywhere from, you know, 20 percent concrete all the way up to 95 percent dense, or even fully — even more completely dense, okay. The concrete usually has some pores, so far as that goes. It's hard to make it without pores. You think it doesn't have pores, but if you put too much water in it — well, you can get rid of those pores if you fill it up with water, but then it gets weak, okay.

Concrete's also an interesting material in that it — it's an exothermic reaction. When you mix it with water it actually will get up to about 140 degrees Fahrenheit. I had a hockey rink that was put in Connecticut in August, and they used polyethylene — black polyethylene tubing — and they had the tubing all laying there in the rink to pour the concrete on. And of course the hot summer day, the black was hot, they poured the concrete on top, and as it sets, concrete is not a very good thermal conductor, and so it basically heated up the pipes and they collapsed inside the concrete. The whole thing was — they had to rebuild it. You need to keep it cool, or you have to cool it. Many times in concrete curing, if it's a very thick slab, they actually have to run some pipes through it to keep it cool, okay. 'Cause it does heat up pretty much. There's lots of little nuances of things like that. Yes.

Student: [inaudible question about specific properties / steel vs aluminum]

Yeah, we often talk about specific properties being — you normalize them to the density of water, okay. And aluminum is stronger than steel in specific properties, okay, in many cases. High-strength aluminum is stronger than a lot of steel in specific properties. But then you actually have to get to the real structural application. If I'm just building a huge stationary breakwater in a port, okay, where ships are going to dock, I'm interested in the lowest cost per volume, okay, that will survive in the water and not corrode. And concrete doesn't really corrode in the water, okay, for a long period of time. So specific — per the volume is the most important thing there. But in other cases I need the structural strength, and so I measure the strength of the thing, okay, and steel is stronger than concrete on a strength basis, certainly in tension, okay, and many times even in compression, okay. So you have to look at the actual application to decide what your quality metrics should be. It's not always the same, okay. And in fact, that's one of the fallacies — and we'll get to that later — where sometimes you assume that the quality metric is one thing and you think something is wonderful because of that, and that happens in 3D printing. Other questions?

Okay, the next thing I want to talk about is over time — oops, well, that's interesting, it's still on. Oh no, it's not — there's — oh, my computer's not — just unplugged it, okay. Let's see if it's happy now. Okay.

So if you look at what was the principal non-monetary thing that people were buying and based their trade on, in the old days it was food. How many bushels of wheat or barley or corn or whatever. That's because up until — well, in the 1770s, 97 percent of the workforce was engaged in agriculture, just to eat. Only 3 percent of the people were lawyers and doctors and things like that. From 1870 to 1970 you had Henry Bessemer and then Andrew Carnegie. Steel and the railroads took over, okay, because it was possible to produce steel after Bessemer showed us how. And Andrew Carnegie became the richest man in the world. He was richer than Bill Gates or Carlos Slim or any of those other people in today's dollars, okay.

And from 1970, or the early 1970s, after the Arab oil embargo in '73, oil became the monetary thing. And in one of my lectures on steel I go through how that has changed and why that's changed in terms of what it takes to make steel. It used to take a lot of labor 400 years ago. Today it takes mostly energy and the raw materials, and the labor is a small fraction of production of a metal today. And really, oil has been supplanted by other forms of energy, whether it's solar or wind or whatever. But energy is still the thing that you measure the value of something in materials.

And so you can look at the energy content of materials, and this is megajoules per litre, so there's your volume, and this is megajoules per kilogram, so this is density and stuff. But way up here getting — I mean, hydrogen's over here and aluminum's up there, and everything else is down in here, okay. So aluminum is sometimes called "canned electricity," okay. When peace broke out with the Soviet Union — well, with Gorbachev in 1985 and then into early '90s, the Soviets had tremendous electric capacity in Siberia. You can't transport electricity more than about a thousand miles. Why? Anybody know why? Yeah.

Now, you can build the lines, we have lines all over the country. The problem is when you have AC electricity, you get a phase between your — a phase difference between your current and your voltage, and when that phase difference gets to be too great, all of a sudden you're just heating up everything around you and you're not transporting the energy, you're just heating up everything, losing the energy along the way. If you could do DC transmission, you could go for several thousand miles, okay. Yeah.

It's impedance losses, if you will. If I — can I use a fancy word that I don't understand? But basically it is — it's not resistive losses, it's impedance losses. But as you say, inductance and capacitance, because of the phase difference between the voltage and the current. But you have to get into complex numbers to explain all this, you know, square root of minus one, complex numbers, okay. But it all makes sense to the electrical engineers. Doesn't make that much sense to me.

But what they could do is they could produce aluminum and transport it by rail better than they could transfer all that hydro power. And so they started dumping aluminum on the world market and destroyed the business of making aluminum for Alcoa and Norsk Hydro and Alcan and all these other people. They basically got out of the business, okay, because the Soviets — that's how they're transporting their energy from Siberia, and there's a lot of it there.

But anyway, you can look at — you know, I'm not going to go through all these things, there are some of my other lectures, I just stole these things. Structural materials: steel is at the top, plastics are number two. This is all plastics, everything from polyethylene to some $400-a-pound plastic used in aircraft, okay. But all plastics are only 1/5 the volume of steel. Aluminum is way down here, and then you've got some things like scandium — we only produce 10 tons a year of scandium, and you can ask me later what scandium is used for. And the dollars per ton are all over the map, okay. Steel is fairly cheap. Plastics are actually fairly expensive. Aluminum's fairly expensive compared to steel. Copper and a lot of these other metals — the reason they're not used as much is because they're a lot more expensive than steel. But you get to some of these other things like gold, the platinum group metals, or scandium and stuff, they get pretty pricey, okay.

I'll tell you what scandium could be used for. If you could put a tenth of a percent scandium in aluminum alloys, you can increase the aluminum strength by 10 to 15 percent. Boeing would love you, okay. But the problem is, to put a tenth of a percent scandium in aluminum doubles the price of the aluminum, okay. It may only be a tenth of a percent. So we actually use scandium — I should have brought my sample — anybody know where we use scandium in aluminum alloys? Baseball bats, okay. You can get a hundred thousand ksi strength in scandium aluminum baseball bats. Anyway, so we don't — this is relative cost, and Nestlé — this gets to your question, lots of metrics, okay, of these things.

Now, I gave you on Stellar — I mentioned this, "Materials for the 21st Century: Research Needs" — and this is if you went back 20 years ago, I kind of had a lecture on structural materials, and this is a summary of the whole course, okay, in four pages. But in there I point out that the value of a pound saved in an automobile over the life, which is a hundred thousand miles or so, it's two dollars a pound. If you ship something by rail or by boat, it's only 20 cents a pound. So I can transport concrete around the world at 20 cents a pound and still pay for the energy savings, okay. Spacecraft, twenty thousand dollars a pound. So sometimes people talk about advanced materials or additive manufacturing and they'll give you examples from the aircraft or spacecraft business, but only because that's the easy example — you don't have to talk about — cost is not the driver there.

But let's look at — oh, I didn't put it in here. Oh, if you have another slide — they talked about what's the volume of the business, and it turns out automobiles is the big business. That's six hundred billion dollars a year, if you multiply the number of ships and things by the — that are built per year and stuff. Automobiles just swamp everything else. These other three industries are nothing compared to steel — are not steel, but automobiles, okay. So why are we interested in advanced additive manufacturing of metals? Well, we're not just interested in additive manufacturing of metals, but most of the additive manufacturing that's been done, starting with pottery twenty-five thousand years ago, is for nonmetals.

And if you watch the lecture from spring of 2015 that I suggested, I give an example of what the problems are with additive manufacturing of metals. Things like surface tension — metals have a higher surface tension than any other material. Yeah, well, has to do with the chemical bonding. They have surface energies four times that of — sometimes ten times that of — most ceramics, and therefore they want to ball up. And once they start to ball up, it's very difficult when you're doing this laser surface melting or electron beam melting — that was electron beam melting — it's very difficult to get rid of the defect, okay. So there's problems with additive manufacturing of metals.

One of the things I talked about — one of the things that I wanted to do back thirty years ago when Ely Sachs and Mike Cima invented the term "3D printing" — and they called it 3D printing because they started with an inkjet print head, and all they did was take an inkjet printer and spray the binder on a layer of ceramic powder. Wasn't really fancy, but they were able to make things immediately. That's not the first additive manufacturing. What was the first additive manufacturing? About nineteen eighty-five down in Texas — selective laser sintering, where you take two laser beams in a polymer liquid, and where they intersect you get a high energy density because of the beams intersecting, and that would cause the liquid polymer to polymerize into a solid, and you make this very friable, weak plastic. But people jumped on that in the late 1980s for prototyping, to make parts so they could see what they would look like in building an automobile or something, okay. It was slow, and it kind of looked like your water bottle there — it's kind of a gold clear plastic. And they still use it for a lot of prototyping. But you never could make a structural material because it was a very specific type of polymer, that was a photopolymer that the laser reacted to.

And so Mike Cima and Ely Sachs came along with an inkjet print head and laying down a layer of ceramic — very thin layer, printing the adhesive on top, putting another layer of ceramic, printing the thing, and just building it up. A lot of the stuff in the early '90s is developing the computer codes to make the complex structures. What were the early applications? One of the earliest applications was making casting molds for artificial hips. Now, or — dentists, it was actually too expensive for a dentist early on. But an artificial hip is a 50,000 dollar operation. If you can make that metal hip personalized for the patient so it fits right, okay, custom fit, fifty thousand dollar operation, who cares? You make the mold, you cast the alloy in a regular casting metal casting shop, but you make the mold. That's how they make the Super Bowl rings, too, still today. They're only going to make a couple hundred Super Bowl rings or World Series rings. You go down to Tiffany's down here in Rhode Island, and you'll see they got about six 3D printers that are binding ceramics for the mold, and they can customize the design and make, you know, 200 molds for your rings, and then they destroy that, and you have to win another Super Bowl and get another design, okay.

So the other thing that they found, it was very good for making pills for medicine. One of the problems with making pills is most of your pills that you take — not an aspirin, where it's all solid silicon — but most of them are just something like limestone, okay, or magnesium sulfate, or something that's not toxic. And the medicine component is like a tenth of a percent. And so how do you mix that in and get the constant concentration? Well, they found you could just go along and put your liquid on top of your drop and you get a very constant concentration. That was a great idea, okay.

And then people have been trying to use it for all kinds of things, but they had a lot of problems with trying to do it with metals. And that's what — my, if you watch that video of my one-hour lecture, why metals are a problem in trying to make the whole process work. But in fact, we have been doing it — was the end of that lecture — we've been doing it for a hundred years with metals. I've been to steel plants where they're rebuilding the steel rolls to roll the steel for continuous casting — size of a football field, and nothing but a bunch of lathes with the rolls being refurbished and just laying metal down to resurface them, okay. This is additive manufacturing. You can't afford to throw away these huge rolls. So we've been doing it for years, okay.

But what we are going to do for the next few lectures is, I'd like you to go through the Digital Alloys guide to metal additive manufacturing. Alex Huckstepp is — there, vice president of business development. He will actually go through, and because he's trying to sell this technology, but he did a series — and I think it's a very fair series — comparing the 15 or 20 different types of additive manufacturing of metals, okay. And so it is a guide to metal additive manufacturing, okay. And first he talks about the business values. Anybody had a chance to look at any of these? Yeah. They're not very long, okay. We will, in some of the next classes — tomorrow, next class, we're not going to talk about this, we're gonna have Dave Hill here — but it's got applications, economics. He actually gives a very bright picture, and he talks about, I think, dollars per cubic centimeter, okay, and they're fair numbers.

But in fact, to make a part twice the size of this out of titanium might be a $20,000 part — whoops, no wonder I'm only talking about spacecraft and aircraft, okay. To make a part like this — [Tom holds up a sample part.] This part is actually being made by Digital Alloys. They're making about a thousand of these. This one I took because it looks like the World Trade Center with a thing on the top. It shows you the surface texture. It is 100% dense titanium. But why are they making thousands of them for Boeing? Because Boeing can't use that unless they have the statistics that meet the Metallic Materials Property Development and Standardization, MMPDS-05. This used to be Mil Standard 05. This is actually — this is volume tenth — or chapter nine. It's a series of books that takes up that much shelf space, okay.

In my office, back eight years ago — 2010 — I paid a hundred and ten dollars from the government printing office for this. The government no longer sells it, even though it's a government publication. You can buy it for 1,100 now from a private source that the government has outsourced to. So they're making money — the government is not making money, but that company is. But by law, this was written by the Defense Department and the Federal Aviation Administration — actually they didn't write it, they basically compiled it, and the data comes from, for the aluminum alloys, that comes from Alcoa, Boeing, Alcoa — users and manufacturers of aluminum alloys.

And it will have tables and tables of the properties, and those properties have to, by the guidelines, meet certain statistical requirements. You're gonna find an aircraft, they want to know within ninety-five percent confidence that piece of aluminum or titanium has properties at least of this minimum strength, or toughness or fatigue resistance design. I'm gonna read: "design minimum mechanical properties tabulated in the MMPDS" — which is this — "are calculated by direct or indirect statistical procedures. The minimum sample size required for the direct computation of T99" — which is 99 percent confidence limits, which is for spacecraft — "and T90 values," which are for most other things, "is 100." You have to have a hundred samples, okay. "These 100 observations must include data from at least 10 heats," which means 10 different lots of the production material. Each one of those is a separate lot inside additive manufacturing, so you don't have a problem there. And 10 lot is defined in the next paragraph. You have to have a hundred samples from 10 different lots and statistically average these. You have to have that data before you can ever put it in an aircraft.

So just because you can make a part, do you really want to make a hundred of them at 20 thousand dollars a piece, to just measure — tear up and measure and throw away — before you can actually put one in an aircraft? So there are codes and regulations you have to deal with. Yes.

Student: [inaudible question about scope of the standard]

This is for the structural — this is not for the seats you ride in on the inside of the aircraft. This is the airframe structure or the power plant, okay. So it is for structural materials, okay. But there are lots of things — lots of weight on the aircraft is not structural, right? So it's only for the structural components, the beams and things like that.

And for example, I'll give you an example. Right now I'm involved in a lawsuit of Alcoa against another aluminum manufacturer for ten years. This other aluminum manufacturer, which is a little extrusion plant — they were hiring people away from Alcoa's Lafayette Indiana facility, okay, and then all of a sudden ten years later they get 45% of Boeing's wing business for the spars for the wings. That and the landing gear are the two critical parts of designing an aircraft. Those are the things Alcoa or Boeing will never give up. How did they get a hundred spars from ten different heats? Well, the answer is they didn't. And now they're in a half-billion-dollar lawsuit where we're trying to show that they stole it, they stole the data, and it was Alcoa's data that they got, okay. I don't want to happen — that's the whole story about how they did it, but that's what the lawsuit's all about. They claim, oh, you could just find this in literature.

Student: [inaudible — apparently asking whether you literally have to build 100 airplanes]

Right, you're missing the material shape so — right. You still have to have a tensile bar, yeah. You don't have to build a hundred airplanes, you're right, okay. You only have to have — if I'm doing a wing spar for an aircraft, I only have to have a hundred pieces of that extrusion produced in the same general way, okay. Has to be produced by the same process, okay. It's not just the material composition, it's also — if you look in here, it talks about product form: sheet, extrusions. So yes.

Student: [inaudible — apparently asking about whether properties change with geometry]

I don't think this has ever caught up with that. Oh, you're right. Okay, you can get different properties in different geometries. But this will only have two — they're only making these so they can make tensile bars, okay. The part — titanium part they may want to make may be much bigger, okay. Well, you know, they have to have real tensile bars, they can't just do a hardness test and infer the strength, okay. So — but your question — they haven't gotten that far of how do you incorporate additive manufacturing into those things.

The point I want to make is there are lots of hurdles to incorporating a part, whether it's additive manufacturing or any other new manufacturing process, that limit your ability to introduce that part into the marketplace, okay. A lot of these — that's — I have a lecture series on codes and standards that I've only taught once. The students aren't that interested in it, but it's actually one of the biggest barriers that you have to introducing a new material into production. It's the regulations that are out there that — things like, you got to have a hundred different tests, okay, on that material. If you're gonna put it in an aircraft — you want to put it in a car, General Motors will decide that. But the FAA regulates Boeing. Now if you're the Air Force, an Air Force general can tell the FAA "go jump in a lake," but he won't — he's still going to use this because the Department of Defense helped generate this, okay. I'm just trying to say there's other factors other than just the technical ability to make a complex part, okay.

With that, we kind of used up today, but what we're gonna do tomorrow, I'll give you a heads-up. Whoops, I'm going wrong way. How come I'm going the wrong way?

Okay. Why the interest in additive manufacturing? Why do venture capitalists frequent MIT? Should you invest in AM startups? Under what conditions? That's what we're going to discuss. And not just for additive manufacturing. But twenty-five years ago, would you have invested in high-temperature superconductors for high-field magnets and magnetically levitated trains, which have not come to occur in the last twenty-five years, by the way? Could you figure that out, okay? Would you have invested in structural ceramics in 1985 that everybody was investing in, but they haven't come about, okay? Can you figure out ahead of time, as a critically thinking engineer, whether additive manufacturing makes sense for a given type of part? And the bottom line is, yes, for some parts, but for mass manufacturing, no, it's not going to be — well, you'll see some of the quotes in the next thing, okay. That's enough. I can keep going, but you got to go to class, okay.