§1. The cost of corrosion [00:05]
The corrosion engineers were thought leaders back in the 1970s. They did a study showing that it cost the United States seventy-five billion dollars a year for replacing things that corroded — replacing automobiles every fifteen years, bridges in a hundred years, ships every thirty years — and you come up with a big number. They went to Congress with this number and said, you need to fund more corrosion research, and actually Congress did. Then all the welders and all the other people in different fields came along and said, the cost of bad welds is such-and-such, and Congress isn't buying. You have to be the first one.
There are studies and the corrosion people still do them. This one is probably about the 1990s, and they actually broke it down by different sectors. The biggest sector for corrosion — and this is replacement of things — is drinking water and sewer systems, thirty-six billion a year. Motor vehicles is twenty-three billion a year because cars rust out. Defense, that you should have some interest in, is about twenty billion a year. Twenty billion a year is a lot more than the budget of most of the militaries of anybody in the world except us, because we're spending about two-thirds of all the world's military budget. Of course China gets a little more for their dollars.
Since this is a Navy course, there's a book on case histories in marine corrosion — Illustrated Case Histories in Marine Corrosion. I found this on my bookshelf; I didn't even know I had it. There's a paper in here that goes through forms of corrosion. Remember, I told you there are eight forms of corrosion. It says there's a theory of marine corrosion. Well, the theory of marine corrosion is exactly the same as the theory of most other corrosion. It takes moisture and it takes oxygen. And it talks about cathode and anode areas and types of corrosion, uniform attack. Here's a galvanic series in here, which would be in some of the other books.
§2. The galvanic series and graphite-magnesium composites [02:39]
Anybody know what the galvanic series is? It's the difference between metals. The farther apart they are, the greater the potential reaction — the voltage you can get between two metals. That voltage is a function of the free energy of formation of those materials. You can prove it from chemical thermodynamics. This is MIT; we go back to the fundamental principles. The most noble metal is actually a non-metal — it's graphite. Then platinum. They don't have gold on here; they list titanium, and I'm going to talk about why titanium [is] down here later. The least noble — the one that wants to corrode the most — has a negative 1.6 voltage. Platinum and graphite have a 0.2 voltage. The most corrosive of the ones listed here is magnesium. Actually beryllium would be on here, on a practical structural materials plot.
Anybody been to Port Hueneme? Port Hueneme, about thirty-five years ago, decided to get into the metal matrix composite business. They were making graphite-magnesium composites — very lightweight. Graphite's lightweight, magnesium's lightweight. These are interesting because if you're making a sample and trying to polish it to look at the microstructure, if you polished it in water it would be corroding in the polishing machine faster than you could polish it. Because you've got the biggest voltage difference between those two.
So they started thinking, we do diving around here, and they came up with two applications for this magnesium-graphite composite. One: they could wrap it in a plastic bag, and the diver could take it down and put a little rubber tent over whatever they wanted to salvage, and then take a knife and pop the plastic bag, and the thing would corrode so fast it would create a hydrogen bubble underneath the plastic bag, and you can float it to the surface with hydrogen. Just don't have any ignition sources. Hydrogen's not that bad, it's just a big poof.
The other was, they put it in a plastic bag, strap it to the diver's belt, and if he broke it and was in cold water it would generate enough heat to keep him warm. Cheaper than a battery. It's just a big battery. In fact, we have batteries — typical dry cell batteries, one and a half volts — it's graphite and manganese, which is not on here, but you get one and a half volts. These are all the galvanic potentials for what we call the thermodynamic standard state, and not everything is at the standard state concentration. You can have things larger than the aqueous standard state, but typically it's very difficult to get more than about one and a half volts out of a battery.
Just think of your car battery, which is a lead-acid battery, lead sulfate battery — you get about one and a half volts per cell. That's what people do: they find different chemical systems that will give you about one and a half volts. It's almost impossible to get a voltage of two volts in a simple single-cell battery. By the same token, it only takes one or two volts to corrode something. If I want to do cathodic protection of steel — iron, mild steel is right here — I need 0.85 volts impressed on that to keep it from corroding.
§3. Cathodic protection in practice [06:50]
So they build whole pipelines, and in the areas where they have AC electricity, they put in a rectifier, create DC electricity, put electrodes in the ground, and they actually make the whole pipeline 0.85 volts to a reference electrode. And it will keep the thing from corroding for a hundred years, a thousand years. When they're somewhere where they don't have electricity — I was out in west Texas where they're putting in some transmission towers — they use zincs. At the very back of this handout, they talk about marine corrosion and how they use zinc as the sacrificial anode, and the sacrificial anode corrodes rather than the ship's hull. Or they put on an electrical system — they have an insoluble anode and insulation and everything else, but they essentially are impressing electric current. That's what we do for nuclear subs.
You have to maintain 0.85 volts; you can't go above 1.1 volts. You might know why you can't go above 1.1 volts. Because you now start turning sea water into hydrogen and oxygen. You weren't doing that in high school — it wasn't sea water, you took water and you electrolyzed it and you got pure hydrogen and pure oxygen. You can do that with magnesium and graphite because it's 1.8 volts, and that's more than 1.1, and you actually start generating hydrogen bubbles and oxygen bubbles. So that's some corrosion protection as far as that goes. This is just a general handout on marine corrosion. If you didn't want to read anything else on corrosion and just make your presentation, you could just read this handout on illustrated case histories of marine corrosion.
§4. Pourbaix and the thermodynamic atlas [09:32]
Now, let's get to the thermodynamics. There was a guy in Belgium after World War II, Marcel Pourbaix. No one had any money to do research after World War II. They didn't even have any money to buy food, because they didn't have any food — pretty desperate time. But Pourbaix was a scientist at a research lab in Belgium, and he decided he could do calculations; he didn't need equipment or things like that. He came up with the Atlas of Electrochemical Equilibria in Aqueous Solutions. Really catchy title — I kind of like my Fifty Years one better. But he came up with a book where he calculated the thermodynamic stability of basically every element in the periodic table.
If you look at the table of contents, they've got H2O, hydrogen peroxide, hydrogen, lithium, sodium, potassium, rubidium, cesium — he's going down the periodic table. Second column: beryllium and magnesium, calcium. He groups them like the periodic table, and he's calculated everything. He even calculated Tc. What's Tc? It's technetium. Doesn't exist on this earth in a natural form, although you nuclear guys know that we use it for stress tests. The reason it doesn't exist on this earth is the elements were formed in supernovas, when stars blow up. You take the hydrogen and the helium and they make heavier elements. The most stable element in the periodic table is something called iron. That's why there's a lot of iron in the universe.
Technetium is formed, and you can see it spectroscopically in supernovas, but its longest half-life is about ten hours. We actually form it in nuclear reactors. There's a reactor in Canada and one in the United States that make technetium. Technetium's a silvery-white metal if you make it in metal form. It's extremely toxic. They basically use jets to fly it around the country, and when people are having a stress test, you act as a living autoradiograph. They put you in a room that is going to look at all the radiation coming through, and they do a stress test and take a picture of your heart pumping as you're stressing — running on a treadmill or whatever. They're taking this real-time x-ray. Then they come out and inject you with some technetium, which is right underneath magnesium and manganese — manganese is non-toxic, it actually is essential for health.
So they give you a little radioactive technetium, and it's decaying rapidly enough that they can see where the blood vessels are, because the technetium's running through your blood. If you have a blockage in your heart, they can say, we saw something before and there was a blood vessel there, and now we don't see the blood vessel, which means blood is not flowing through. I had a case once where a school burned down in Kentucky, and the people did an analysis in the scanning electron microscope and came up with an analysis that said there was fourteen percent technetium in the roofing nails. I thought, well, we can settle this case — we just sell the nails for the technetium. It just shows you how good some chemical analyses are. Thanks for that expert.
§5. Thermodynamic versus practical nobility [13:41]
I gave you the handout on the Pourbaix diagram. Let's go over some of the stuff in Pourbaix's book. He has a chapter at the beginning about corrosion, because corrosion is an electrochemical process. The whole chart is on page 80 of Pourbaix's book. I'm sure Pourbaix has passed away now, but the whole research center in Belgium — Cebelcor — is the Pourbaix Laboratory. He lists forty-three of the elements in the periodic table in terms of their thermodynamic nobility and their practical nobility.
We've already talked about which one is best in terms of thermodynamic nobility and resistance to corrosion — it's gold. Number two is iridium, number three is platinum. That's why when I was a graduate student about to get married, I electron-beam melted my wife's engagement ring, and then machined it. They couldn't afford the gold, actually. But I wanted to have something a little different. I also almost got electrocuted — that would have been a shorter wedding.
There's also what they call practical nobility. Thermodynamic nobility means it just doesn't want to form an oxide or a sulfide — it's not going to react with a lot of things. Gold is the best; iridium and platinum are near the top. Most of you don't know rhodium and ruthenium and palladium, but those are platinum group metals. Mercury is oxidation resistant; silver too. Mercury actually does react with sulfur very strongly. Osmium is another platinum group. Selenium is sort of a near metal underneath sulfur. Tellurium underneath selenium. Polonium, which is only something you poison officials in Ukraine with.
Did you know that story? That wasn't that long ago — two or three years ago. The leader of Ukraine had some enemies, and some of those enemies had some good friends who used to be part of KGB, and so they got some radioactive polonium, which is very difficult to detect, and they were poisoning him slightly over time. He was losing his hair and all these other things because he was radiation poisoned. But this is why he was the president of Ukraine. So let's teach you guys sneaky tricks — who's going to have to be analyzing for polonium. Most people can't even test for activity.
In terms of practical nobility, rhodium is better than gold; niobium and tantalum are better than gold. Titanium's pretty good. Some of these things that were way down in thermodynamic nobility — aluminum comes shooting up to number nineteen from thirty-nine. Magnesium shoots all the way up from forty-three to forty-one. Manganese — they used manganese in their batteries, graphite-manganese, so manganese tends to corrode pretty easily anyway. Niobium and tantalum were thirty-three and thirty-four; they shoot up to two and three. Why are they so good? It turns out these things have practical nobility because they form a protective oxide skin.
§6. Titanium: pacemakers, jet fires, and a heat exchanger in New York [17:38]
[Tom holds up a titanium pacemaker case.] This is a very old one — I think I was an assistant professor when I got this. Now this one has some sharp edges because they cut it. So you're going to carry a pacemaker in someone's chest. Titanium will not corrode in the body. I told you the biggest voltage you can get is less than two volts in an aqueous solution. Titanium can take five volts; it won't pit, won't corrode. It's more stable with its oxide than just about anything else, unless you get rid of its oxide. It's not easy to get rid of its oxide. Above about 900 degrees centigrade — I have to check, I think it's 900 — it will dissolve its own oxide. And so they've had jet engine fires due to titanium catching on fire. At that point you have a huge flare, only it's called your engine, and when it consumes itself you don't go very fast anymore in that jet engine.
I had a case where some guys in New York City, just after 9/11, in another skyscraper, had a titanium heat exchanger. You can corrode titanium. In this case they got some silt — someone opened the wrong valve, and they got all the East River silt into their heat exchanger, which is a real problem. You get what we call under-deposit corrosion attack, and titanium can corrode under some of those conditions, and it did. They had to replace this heat exchanger. They hired immigrants who couldn't speak English, because they were the only ones who would just do this and wouldn't ask questions. They went in with oxy-acetylene torches. You can flame-cut titanium — you'll learn about that when you take my flame's department — lots of smoke.
We were cutting in the lab here about 25 or 30 years ago, and the secretary across the hall — where my secretary is now; I didn't have this office then — complained to the MIT environmental people. They came down and they saw the smoke in the lab and said, what's going on here? He said, what type of lab is it? I said, it's a welding lab. He said, but what do they expect to come out of the welding lab? These are the old days thirty-five years ago, and it's okay to have a little smoke in the hallway. It's not good — the paint on the wall, it's tight.
§7. Tantalum: sulfuric acid plants and brain plates [20:30]
[Tom holds up a tantalum tube.] Titanium forms a very protective oxide skin. Tantalum forms a very protective oxide skin. The problem with tantalum is, it's a little bit pricey — it costs about the same as silver. One of the world's largest tantalum producers is H.C. Starck, right over here in Newton, Massachusetts. Right among all those homes. When H.C. Starck built their factory over there, it was all rural — this is the late 1940s. It's now owned by Bayer Chemical, which is number one or two with DuPont, the world's largest chemical company. About half of all the world's tantalum is processed in that factory, with all these little residential homes around it sixty years later. I don't like this factory right in the middle of that, but that's where I got a tantalum tube from them.
They put it in heat exchangers if you're going to use sulfuric acid. The most common inorganic acid industrially is sulfuric acid. The most common organic acid is acetic acid — they make acetate sheet plastics out of it. Sulfuric acid is used for lots of things. All your gasoline uses sulfuric acid as a catalyst, to absorb water in the reaction when they break up the heavy hydrocarbons into octane. The only things that will not corrode in concentrated sulfuric acid are tantalum or graphite. So they make great big heat exchangers and columns of graphite. You can't afford to make the whole thing out of tantalum, so you clad the tantalum onto carbon steel — you take a clad material, and you're only using tantalum where you need it. You can use very thin tantalum because it just doesn't corrode.
[Tom holds up a tantalum brain plate.] You also use tantalum as implants in the brain. This is a brain plate, made out of tantalum — won't corrode in the body, because it has extremely high practical nobility, as does titanium. So we make implants out of titanium. Get an artificial knee, artificial hip — some of your grandparents or your parents may have them. They're usually made out of titanium because they won't corrode in the body. That's gas tungsten arc weld quality. They take sheet, they roll it; they actually melt the whole thing in a great big electron beam furnace. Tantalum and niobium can be cold-worked at room temperature all the way down. You don't have to anneal — they're just so soft. You can anneal them, but they usually don't. You can start with an ingot this thick and roll it all the way down maybe with one anneal in between. You can make a thin sheet and then weld it. They work very easily. They're kind of pricey.
So you can talk about thermodynamic nobility and practical nobility, and practical nobility is what really counts. But the problem with practical nobility: if you lose your protective oxide skin on some of these things that are the best, like titanium — I told you, you can dissolve titanium's own oxide at 900 C. The other thing, early on, they had some very serious explosions where titanium was being used when they first developed it, because it had such good corrosion resistance, and they had a problem with anhydrous ammonia. Anhydrous ammonia, NH3 — if it has no moisture in it, like less than a hundred parts per million of moisture, the titanium will react. Titanium plus NH3 reacts to form titanium nitride, which is very stable, plus titanium hydride, which is very stable. There are some x's and y's and z's in there to balance the equation. Titanium will react with anhydrous ammonia and cause stress corrosion cracking, and they had some really exciting explosions. They were using some of this with the rocket programs early on and blew up some of the launch pads.
§8. Reading the Pourbaix diagram [25:08]
So there's practical nobility and thermodynamic nobility. If you go to Pourbaix diagrams — let's look at the Pourbaix diagrams. This is a Pourbaix diagram; you should have a copy of this for water. It's a diagram of potential in volts. Minus 1.6 is where magnesium would be. They're Europeans, so it's V in volts. And it's pH down here. Seven is neutral pH. A and B — A is the line at which, at that voltage in that pH, you will form hydrogen at the cathode, and here you will form oxygen.
As you change the voltage — this is cathodic down here, this is anodic up here. Down here you have hydrogen ion stable; up here you'll form hydrogen peroxide in your aqueous solution. Of the gaseous substances, you have oxygen up here, hydrogen down here. This is kind of the average where your water is forming. You can have ozone way up here. This is thermodynamic stability. Down here at the cathode, up here you're going to have to form some oxide skin for something to be protective in the water environment. If you're worried about nuclear reactors, the only place where water is stable is between these two bold lines. You're actually forming gaseous species when you're above or below either hydrogen or oxygen.
The same type of plot — you can think of liberation of oxygen and you acidify the water, liberation of hydrogen and you turn it into a caustic type of water. The hydrogen that you release up here combines with H2O to form a hydroxyl radical. Here's thermodynamically stable water. You can have oxidizing and acidic, oxidizing and alkaline, and similar things down here. So if you want to dissolve gold, what type of water do you use? The alchemists named it aqua regia. Anybody ever heard of aqua regia? The king of the waters. It's a mixture of nitric acid and hydrochloric acid, either three to one or one to one. It's oxidizing and very acidic — the concentrated acid. And it's got high chloride content. Gold chloride is very slightly stable.
§9. Attleboro and South Africa: stealing gold [28:25]
I used to work for a gold company down south of here in Attleboro, where they make most of the jewelry and gold jewelry in the United States. I used to get some really good stories about people trying to steal gold. The company I was working for was called Leach and Garner. They had built a plant in 1899. All the other jewelry manufacturers built up around there because Leach and Garner was the mill. They would take gold bullion and melt it. They had two continuous casters for making gold alloys, and they would continuously cast strips of the whole, and then they would make it into everything you could think of: tubes, sheet, foil, bar, wire. They sold it to the jewelry people, who'd take the wire and make it into eyeglass flex frames or whatever.
They would also take some of that alloy and grain it — they call it graining — and they would spray it into a bath of water and you get these little pebbles. They sell that to people making class rings or whatever, who would take that alloy and melt it and make the rings. Right through the middle of town in Attleboro there's a little stream. One of the companies was finding they were missing some of the gold rings. They found that people were throwing them out the window into the river, and then on the weekend they'd go diving for the rings. So the solution was to purchase the mineral rights to the stream from the town. That way they could prosecute the people for stealing their gold. But then they had a problem: they caught one guy trying to steal, and they fired him, and they were still missing some rings. Turns out his girlfriend was still working, and she was just flushing them down the toilet to spite the company.
My favorite story comes out of South Africa. One guy was working in one of the gold mines in South Africa in a chem lab, and he would bring in a hard-boiled egg with his lunch and a little salt shaker for lunch. But when he went home, he actually disposed of the salt and filled it up with gold chloride — a white salt that looks like sodium chloride. He was carrying out about an ounce of gold a day. People are ingenious — now they steal gold in their lunches.
Oxidizing and acidic will dissolve gold. Very reducing — like using ammonia, where you have no oxygen potential, you've got rid of all the H2O and you're way down here — you can dissolve away the titanium dioxide, no longer have the practical nobility, and things go poof. There's a very rapid reaction.
§10. Pourbaix diagrams for engineering metals [31:52]
You can look at other things in the corrosion area for Pourbaix diagrams. Beryllium, in neutral waters between about three and a half and ten and a half, is really insoluble. Acidic or caustic solutions will dissolve it either way. Aluminum is similar. As long as you don't get too acidic or too caustic, you could actually clean aluminum industrially using caustic solutions. You can also use acidic solutions, but it's usually hydrogen fluoride, which is sort of nasty all by itself, so you usually use caustic solutions for cleaning aluminum. Here's silver, has very good resistance — lots of oxides and bases. Here's gold, stable everywhere, which we know. Titanium has practical stability.
Zirconium. What's our major use of structural zirconium? The nuclear guys — it's cladding for fuel rods. You put your uranium oxide or plutonium oxide, whatever your fuel is, inside a zirconium tube. Why do you use the zirconium tube? Because what's the cross section of zirconium for neutrons? It's one of the lowest. It's transparent to neutrons. You don't want to absorb neutrons — if neutrons are coming out of the fuel, you want them to get out, do their work, and then catch them into your water or whatever you have, however you design this reactor. Boron is the exact opposite — it absorbs neutrons better than anything else. So use borated water — just put borax in water, and you can stop that whole nuclear reactor from reacting.
The zirconium has to be very pure and almost no hafnium, because hafnium and zirconium are right above each other on the periodic table, and hafnium has a fairly high neutron absorption. So you have to have hafnium-free zirconium for nuclear service. The other use of zirconium is for making acetic acid. I told you tantalum is what we use for sulfuric acid. For acetic acid, they make reactors out of zirconium. If you want to learn how to weld a titanium submarine — titanium and zirconium are right below each other on the periodic table. The people down in Texas City, Texas, where they make huge acetic acid plants, build zirconium reactors to make their acetic acid. If you go over to China, they've also learned how to build their own zirconium reactors. That's another story we can get into sometime, about how the British tried to fool the Chinese and the Chinese snookered them, by developing a better technology.
Here's titanium again. Chromium is the basis of stainless steels, and chromium has a fairly wide range. It goes to very caustic solutions; it doesn't like acidic solutions. Later I'll probably show you a plot of chromium with chlorine, and you see this band just shrinks down to nothing. That's why stainless steel doesn't always do well with saltwater. Tin is fairly stable. Iron — one of the Achilles heels of iron is its susceptibility to corrosion. Lead can actually be fairly corrosion resistant. Silver, copper, and so on.
Pourbaix can tell you the thermodynamics of corrosion, but what we really want to know in most cases is the kinetics of corrosion. Pourbaix developed his atlas for aqueous corrosion — wet, the world we live in, where we have water. But at Pourbaix's research center, which is now called Cebelcor, the Belgian Center for Corrosion Study, they've also worked on an Atlas of Electrochemical Equilibria in the Presence of a Gaseous Phase rather than an aqueous phase. Because it's more recent, they have nice color diagrams in some cases. It's a similar type of diagram, but here we have hydrogen chloride.