§1. Proprietary alloys and the Alcoa–Boeing contract [00:02]
An aluminum company can develop their own alloys, but if they want to register it, they register it with the Aluminum Association, a trade organization based in Washington DC. Then it becomes a national standard, and eventually some of these become international standards. There's a whole hierarchy out there of buying and selling.
Student: [Question about proprietary alloys.]
I'm involved right now in a problem between Alcoa and another company that stole fifty percent of Alcoa's business for aircraft wings. Alcoa is not happy. They laid off 600 people on January first because they lost fifty percent of their business. They make the wing spars for Boeing, for Airbus, essentially everything. They had a semi-monopoly. There are two primary alloys, one's a 2000 series, one's a 7000. The 7000 series they actually use, they sell as 7150 — you can look on the web and find out the aircraft wings are 7150 alloy, which is an Aluminum Association designation. But Alcoa has a tighter internal spec that they actually make them to, and that's a trade secret.
That's true in a lot of these cases. A lot of these specs in the ASTM standards came about sixty, seventy years ago, and we've learned to improve our metallurgy in the plant. We can make things to tighter compositional control than we ever did before. A lot of companies do that — it gives them an edge in the marketplace, and they hold that as a trade secret. But by the same token, if you want to order a 5/16 steel for a pressure vessel, you put it on the purchase order, and you don't have to write all the spec into your specification — you just reference it in one line, and you increase your specification by four pages. It's incorporated in your purchase order.
To give you a little more detail — this is not the proprietary part — Boeing and Alcoa have a contract, just like Airbus and Alcoa have a contract, and they have to keep that information confidential. Alcoa can't tell Airbus what Boeing is telling them, because that's Boeing's trade secret, and they can't tell Boeing what Airbus is saying. So they negotiate with each other. Boeing has to be able to audit — and you guys aren't SUPSHIP, you're auditors of the purchase that NAVSEA went out and made, and you're there to do the quality assurance audit. If it's Bath Iron Works, they're doing the quality control as the fabricator, responsible for building the ship and making sure it's of the proper quality. But you've got a whole building up there in Bath for the Navy officers assigned there and the civilians working for the Navy, who are going out and looking over the shoulders of the quality control inspectors at the Iron Works. They spend their life in meetings arguing with each other, trying to decide who's right and who's wrong. Ultimately the Navy's got the checkbook and they're always right. But they can burn you, so you might as well try to cooperate. It's this interesting dynamic. I'm glad you guys get to spend your life in meetings like that rather than me.
But in any case, you can specify these things. What was your question about the aluminum again?
Student: [Follow-up question about what happens with proprietary information.]
In this particular case, Boeing is allowed to designate one person who gets to know everything about Alcoa's process. That person cannot tell all the other Boeing people, unless they're in a meeting with Alcoa and Alcoa approves it. That person is the auditor, because Boeing has to know what's going on, but only one person, in theory, knows everything from Boeing's side about what Alcoa does.
§2. FAA, MMPDS, and the privatization of standards [05:25]
There are other players. There's the Federal Aviation Administration, and the FAA has a handbook which used to be known as Mil Handbook 5. It's about 12 or 13 volumes, takes about 12 inches of shelf space. I have it. A few years ago you could download the digital copy for free, and you could purchase the hard copy from the government printing office for $110, because that's what it costs to print it. By law, if you're building a commercial aircraft, the FAA requires you to use the material properties as defined in that handbook. It will have about 20 pages on the properties of 7150 alloy. Whatever Alcoa and Boeing choose to use, if they're going to tell the FAA they're using 7150 alloy, it has to fall within that universe of properties defined by that specification, which is now MMPDS — Metallic Materials Properties Development and Standardization. In the dual-use stuff, they got rid of Mil Handbook 5. It had originally been developed by the Defense Department and the FAA in cooperation, and now the Defense Department's turned over management of it to the FAA. Now you can buy it for $800, or maybe $850 for a hard copy, $800 for the digital copy, which costs whoever is doing it nothing to produce.
This is another problem with codes and standards in the last fifteen or twenty years. The government is not allowed to make a profit on these things, but what they've done is turn it over to a commercial outfit to manage the dissemination, and those people are making a fortune. They're raping the rest of American industry. And it's written in the law. Alcoa, if they tell the FAA they've certified this aircraft of 7150 alloy, the Aluminum Association alloy, they can have a tighter spec between Alcoa and Boeing, but it has to fit within the 7150 umbrella. So it all makes sense.
Student: [Question about how the spars are formed.]
Because these are complex shapes, they do it by stretch forming of the spars. Alcoa does that. Boeing owns the dies — this is not proprietary. Alcoa stretches the aluminum extrusion over the dies to a particular shape, and when it gets to Boeing it fits right into their jigs and fixtures for machining. If it didn't, they'd be like a shipyard, where you've got all these jacks and come-alongs to get the thing to fit. This is fairly precise machining on an aircraft wing — actually not more precise than a ship.
Did you know that building a ship is the most precise manufacturing operation we have in terms of parts per thousand? If you tried to machine a one-inch diameter shaft in a regular machine shop, like we have down the hall here, being able to machine to half a thousandth is one part in two thousand. What's the tolerance in building a ship that's a hundred yards long, three hundred feet? It's one part in twenty thousand. About ten times more precise. The only thing that rivals it is building a semiconductor chip, where we have nanometer technology, and that gets down nowadays to about one part in twenty thousand. So really small or really big.
You calculate what happens when the sun comes out from behind the cloud down at Ingalls, and the ship is changing its dimension as the sun heats one side of the ship and not the other. All of a sudden those things you just aligned within a quarter of an inch are now off by half an inch. And it's just the sun coming up. When it goes back behind the cloud, it shrinks back the other way. There are some things you have to worry about, and most people don't realize that shipbuilding is one of the most precise industries. If you go through it, you don't really think it's so precise. But you start calculating what you have to have for tolerances...
What I was looking for — I thought I transferred it over to my Dropbox, but obviously I didn't. Spec Van Lambing, one of the N students who was coordinating things, did an inspection report on modular submarine construction. It was just a bunch of pictures, but it showed the modular construction — a cylinder of a sub that had been built up at Quonset Point being taken down to Newport News on a barge, and then finally put together. Submarine construction, and ship construction generally, is more precise than most people think.
So have I answered your question about where the numbers come from? It's a very complex process. That's what my codes and standards module is about — how some things have force of law, some are consensus standards, and others, like Pirates of the Caribbean, are just guidelines really. You don't have to follow it; they have recommended practices. American Petroleum Institute's very big on recommended practices.
§3. Brian's case: under-deposit corrosion of cast iron after Hurricane Sandy [12:21]
So Brian, you want to tell your story about under-deposit corrosion and Hurricane Sandy?
Student (Brian): So there's cast iron, which is iron with four percent carbon... [continues describing the case]. You take out the cast iron component and we cover the cross-section, look at it, and all you see is a spongy material. So what you were getting there was a graphitic corrosion. Very similar to — they check the alloy, they see with the acid, dezincification. But this is for cast iron. What happened was you had seawater, and so it was eating away at all the iron.
Ordinarily the cast iron will last for a hundred years or more if you maintain it. Cast iron forms essentially a silicon dioxide — a glass surface, the passive film. Cast iron is iron, four percent carbon, three percent silicon, and all that silicon forms a glassy layer on the surface. So cast iron in aqueous service has better corrosion resistance than steel, which rusts. Steel doesn't have this thick glassy layer; cast iron does. But chlorides will eat through that when you start getting under-deposit attack. So one rule, which I didn't put in my list of rules on corrosion, is: cleanliness is next to godliness. It helps to keep things clean.
§4. Fracture toughness recap and the mystery rubber [14:25]
We had talked before about fracture toughness, and I had neglected to bring my rubber pieces. Remember I talked about — you put a little notch in something and it reduces strength by eighty percent? Well if it's a ductile material like rubber and you pull on it, it doesn't just propagate that crack, it blunts the crack. You don't get a sharp notch for a stress concentration; it just deforms. That's what ductile steel does. There's also brittle steel.
I gotta analyze this someday. A dozen years ago I had someone in the lab get me a sheet of metal — I wanted a sheet of rubber to test something. This stuff is the toughest rubber I've ever found. I pulled on this as far as I can, I can't get this thing to propagate. It's all stretched. I've spent several hundred dollars buying all kinds of rubber to get something as good as that. This red one I could actually get to break if I pulled on it hard enough. But this other one, boy, it's tough, and I haven't figured out what it is. I got to do an analysis on it sometime, but rubber analysis is not easy to do.
§5. Nalco guides: heat-exchanger tube materials [15:59]
What I wanted to do — this is more like a recitation day in some ways, because I wanted to pick up on a few things. There are two books, and you guys have a better allowance than people did five or ten years ago in your class.
[Tom holds up the Nalco Guides.] These are the Nalco Guide to Cooling Water Failure Analysis and the Nalco Guide to Boiler System Failure Analysis. Nalco's the world's largest water treatment company — they're four billion dollars a year. They go around, and if you've got a boiler, they'll take the contract and maintain your water chemistry. Navy does that themselves; they don't contract out to Nalco. You've got enough water systems that you can do it yourself. These are now in the second edition, published by McGraw-Hill. I had the first edition before. I just ordered this one, and it came in yesterday — I saw it at my secretary's desk after class. There were enough things in here that had to do with yesterday's lecture that I thought I'd bring it in and we could talk about some of them as a recitation.
We talked about condenser tubes — I mentioned some things about condenser tubes — and these are some of the materials from the very beginning of the book, alloys for heat exchanger tubing, condenser tubes. Copper's got the best thermal conductivity. You're trying to transfer heat, so you'd love to use copper. If you've got really clean water with no solid particles in it that are going to cause erosion, copper would be great. Aluminum is good but it's pretty soft and would erode away — although aluminum has very good corrosion resistance in sewage. The sewage treatment plants here in Boston are all made out of aluminum, because the pH is somewhere near seven, and aluminum is very good there. We'll see why a little bit later.
Yellow brass — that's what people used to use. Number three in thermal conductivity. Admiralty brass and aluminum brass are alloyed a little bit, we lose a little, but the aluminum brass has this good aluminum oxide skin, so it has good erosion resistance. That sample passed around, that prototype propeller — manganese aluminum bronze is what the Navy uses for ship propellers. Admiralty brass is what they were using thirty or forty years ago at Baltimore Gas and Electric, where they had the under-deposit attack I was talking about. You can use nickel — kind of pricey, but it's still got pretty good corrosion resistance. Carbon steel, we all know, rusts. The Navy likes to use cupro-nickel — 70/30s, or the Monels, which are 30/70 cupro-nickel. Now you're really biting the dust. You can use stainless steel, but it tends to pit. Titanium is better than stainless and has fantastic corrosion resistance. If you have to go to really high temperatures, well beyond water, you can use some of the nickel-based alloys. There's a whole host of materials. If you're talking about condenser tubes or boiler tubes, these are the types of materials you use. If you're talking nuclear reactors, boiling water reactors typically use stainless steels; pressurized water reactors, we're getting into the Inconels. It has to do with the operating temperatures and conditions.
Cupro-nickel is just copper-nickel alloy. It's ninety percent copper, ten percent nickel. Monel 400 is thirty percent copper, seventy percent nickel. So if you go across the copper-nickel phase diagram, you've got 90/10 and 70/30. The Navy uses lots of Monel in their seawater piping. Good corrosion resistance, not fantastic. I told you the story about the aquarium — stainless steel wasn't good enough, Monel was intermediate in maintenance, and titanium would have been the best, but it's pricey. So it's a complete trade-off between price and performance and maintenance.
The Navy is starting to go more and more towards titanium for piping, because now you have fifty-year ships. When you had thirty-year ships, you would use carbon steel seawater piping. Now they've started going to some of these higher alloys — the copper-nickel alloys. They never would have gone stainless steel, because it'll pit like mad if you get any dirt in there. That's the problem at the New England Aquarium, in a sense. But now they're even going titanium, because we're looking for fifty-year lives.
Student: [Question about why titanium, given its cost.]
You can use less of it because it doesn't corrode. You don't have to have extra corrosion allowance on your thickness — you can use half the thickness. It's also stronger, so you can use less thickness; maybe you can use a third of the thickness in some cases. But you're talking about material that's going to cost you $70 a pound as opposed to $7 a pound for the alternatives, as opposed to carbon steel at $0.70 a pound.
More and more industry is now going that way. There are lots of things you can do in design to compensate for the lower thermal conductivity. You can have plates, turbulators — because it's not all the material conductivity, it's the boundary layer inside the tubes. This is where your fluid flow comes in, guys. You start to turbulate the inside of the tubes, putting little ribs in there, so you get turbulence.
If you got deposits in there, you could have a problem. But titanium doesn't care about the deposits. Those tubes in cross-section have little fingers on them. I've got one — this happens to be copper in an air-conditioning system at a hospital down south of here. This might be half a millimeter, and these little ribs might be a third of a millimeter, and they might be half a millimeter high. They're just fins for better heat transfer.
§6. The Mormon Temple heat exchanger [24:51]
So we can do a lot geometrically. This is just material property. There's a lot you can do — you change the diameter of the tubes, you can change the water flow rate through the tubes. I had a heat exchanger up at the Mormon Temple in Belmont. They had originally put in copper as the heat exchanger material. After one year they knew they had a little leak in there somewhere, so they took a paint stop off, and up at the tube sheet — the plate that holds all the tubes — where everything had turned because of the higher speed, they had erosion corrosion on the copper because of the solids in the water. I wanted to go to 70/30 cupro-nickel, but they couldn't get it. They had a two-week shutdown — they discovered it on the third day of the shutdown, the two-week shutdown — they had to get a new heat exchanger in so they could be up and running in another ten days.
It turns out the people who originally charged them sixteen thousand dollars for this tube sheet — for the set of tubes; it was a small heat exchanger, about the size of one of these desks — they found out the manufacturer down here in Rhode Island would sell them one for six thousand, even though the original people charged them sixteen. A little bit of a markup there. They had in stock down in Rhode Island the 90/10, and I said, well, you've got to get up and running, let's put in the 90/10. The 90/10 has worked just fine for the last fifteen years now. I haven't looked at it since then. I was being a little conservative when they said how do we solve this problem, I said you need to go to 70/30, because I knew from the velocity data that 70/30 would have had enough safety factor. Couldn't get it, used 90/10, and it's worked. The real thing is it only cost them six thousand dollars for the replacement, which I thought — well, if you want to use that contractor next time, mark it up ten thousand dollars. That was not a part. Other questions?
§7. Corrosion rates, the Phoenix warehouse, and stainless steel grades [27:09]
Here's another thing we'll go over when I get back to the corrosion lecture. This is a nice slide, I'll incorporate it into my thing. This is water chemistry versus steel corrosion rate in thousandths of an inch — your mils per year. I always say — in fact I think you'll see my presentation later — typical corrosion rate of steel in water is four thousandths of an inch per year. I've got lots of stories. Once a year someone comes to me and says, our steel rusted. And we need you to tell us, did this occur in two weeks, six months, three years, twenty years? What I do is I divide the thickness by four — four thousandths — and I say, wow, this took about — if they lost a hundred thousandths, I say this took about twenty-five years. They say, how did you get that? Well, I divided a hundred thousandths by four, you get twenty-five. I learned this in elementary school, folks. It's not that difficult.
This is actually a more detailed plot — corrosion curves for steel as a function of alkalinity and calcium concentration. This is the hardness of your water, and this is the amount of sodium and CO2 in your water chemistry. You can see there's your average, and I always say it's four thousandths, but if you look in the literature for corrosion in some place like Arizona, it may only be two, and in a really bad industrial atmosphere — the water — it might be ten. One time in Arizona I saw it was twenty, but that was up in the roof of a building.
It was a big warehouse, about a million square foot warehouse, and they had a big rainstorm in Phoenix while they were building this, and all the insulation got wet before they put the decking on. So what did they do? They put the decking on, and they had a bottom seal, so they had about an 8-inch height in that ceiling, of a little greenhouse. The sun would come out — did you know the sun comes out in Phoenix, Arizona? — and it gets hot. Every day and every night they basically would just bathe this in wet, hot water. They actually had twenty thousandths of an inch a year. The galvanizing went away in a few months. Then this sheet metal, which was only sixteenth of an inch thick — twenty thousandths of an inch a year, within three years they had holes in their sheet metal. A leaky roof — it's only a million square feet of roof. And it's because it rained, terrible downpour. Anybody ever seen Phoenix a week after the rain? Beautiful desert blooms. I was in Phoenix once about a week after a rain, and it's beautiful. All it needs is a little water, and that's what they had — their little water.
So that's just corrosion rates. You asked me about stainless steels. This is a list of commonly used grades of stainless steel. You don't have to worry about this, but these S numbers are the international UNS — Unified Numbering System — for alloy numbers. Originally you had 304, which a hundred years ago the American Iron and Steel Institute said 18-8 stainless steel is 304, and 18-10 with two percent molybdenum is 316. About twenty-five years ago they came up with a UNS alloy number, and in Europe they have an International Standards Organization number. There's big competition between North America and Europe for who's going to control the standards, because it makes a big difference economically. If people in India are ordering a boiler and pressure vessel to the American Society of Mechanical Engineers standards, they're going to end up using the American standards for the steel specifications, and that gives the American steel companies an advantage. Actually gives them no advantage, because it's going to be coming from Japan anyway. But that's another story.
In any case, here are a lot of the grades of stainless steel, and they will give you the compositions and the ranges. Your 304 is 18 or 19, 8 or 9. These are the ranges — and silicon — it gives you the compositional limits. There are dozens. If you look in the Metals Handbook, it will give you the non-standard grades, which are the proprietary grades. Sandvik, a Scandinavian company, makes a lot of stainless steel — especially stainless steels — and they haven't submitted to American Iron and Steel Institute for a designation. They'll have a UNS designation, but they may not have the old AISI.
There are different types of stainless steels — martensitic, like this. [Tom produces a pair of surgical scissors.] This is thirty years ago, a three-hundred-dollar pair of scissors made by Johnson & Johnson for surgeons. It's a fifty-dollar pair of scissors with cobalt inserts brazed in for wear resistance, so it keeps a nice sharp edge. Gold-plated handles, because you're going to charge three hundred dollars to the surgeon, and two hundred and seventy dollars worth of product liability insurance, in case you get sued. About ninety percent of the cost of a lot of medical products is insurance for the lawsuits that are going to come.
Duplex stainless steels, for example. Anybody know where the Navy uses duplex stainless steels? On aircraft carriers, need really high strength. The duplex stainless steels' base is basically a mixture of ferrite and austenite. You can get very high strength, very good salt water corrosion resistance, and that's what you need for the catapults. I don't know what's the psi on catapults — three or four thousand psi or something?
Student: [Correction about steam vs hydraulic pressures.]
Yeah, that's maybe what I'm getting confused with. The steam may be four or five hundred, but some of the hydraulics are pretty high. Whatever it is, there are different pressures and different things. I had a problem once for Ferralium 255, or 2205 — maybe it's 2205 actually.
§8. The Miami Art Museum hanging garden [34:37]
If we want to talk about stainless steels and corrosion resistance, we can talk about the Miami Art Museum, which is a fun story. So Miami's building an art museum, about a hundred yards from Biscayne Bay, and it's a hanging garden. Some guy in Switzerland is designing this, and up six stories in the air they have these big concrete beams. Since mold and green things grow all over in Miami, they decided to make it part of the architecture. This is supposed to be like the Babylonian Hanging Gardens — people could sit out there in the verandah and sip their mint juleps or whatever, with these concrete beams shielding them from the sun.
The only problem was, as they're building this, they used a martensitic stainless steel. Someone specified a 410 stainless steel because you get very high strength. Without getting all the metallurgy, yes, they give very high strength, but they actually were in a range that gave it low fracture toughness. Then they had to tack all these things, and they did a tension connection. Why some idiot created a tension connection where you could have had something in a compression connection just laying on top — they put the beam that could fall underneath, and then stainless steel bolts holding it on. If the bolt failed, whoa, you've got twenty tons of concrete landing on your julep, and maybe on you.
This was during construction. They erect these things — they had some tack welds on there during the erection — and about a week later a couple of them come crashing down. It's during construction, no one got hurt. The architect says, well, why'd you weld them? The guy says, well, look at your drawing, it had a weld. This martensitic stainless steel's not very good for welding. So they started looking around and decided, oh, we'll use 316 stainless steel.
This was in the fall, they had the failure, they had to have everything completed in this building by the next December for their grand opening. I got a phone call one Wednesday in March of whatever the year was, a few years ago. The general contractor wanted to hire me to look over the shoulder of the people joining the precast concrete and doing the erection with the stainless steel. He sent me some documents Thursday, he was in my office at 7:00 a.m. Friday morning, the first day I could meet with him. I said, well, 410 was a terrible choice — absolutely terrible choice for a saltwater environment, and welding, and everything else. He says, well, they're going to 316. The construction contract had a one-year warranty. I said, 316 should last one year. May not. He says, well, last ten. What's ten years? That's the statute of repose — when you can no longer sue. I said, well, I don't know if you're going to make ten years on 316. Just shooting from the hip.
He asked me why, and I explained there's something called stress corrosion cracking. We're supposed to be talking about corrosion, so there is some corrosion in this story. He goes back and tells them, and they tell the engineer of record, and the engineer says, well, this is not a saltwater environment. It's a hundred yards from Biscayne Bay, which is salt water. They said, well, it's not a saltwater environment. Excuse me? It's a hundred yards from the bay. Well, you don't have to be immersed in salt water. And did you know they had hurricanes in Florida? Every now and then the oceans come in horizontal at you. It's called rain.
We had this big meeting, and the owners and the architects and the engineers were all shooting at me, because I was telling them they had a problem they didn't want to hear about. But I stood my ground. The one good thing was the architects' expert was a guy I had worked with for years here in Boston, with a firm here in Boston, and he's a concrete expert. He knows a lot more than me about concrete, but he accepts that I know more metallurgy than he does, and so he was telling them, listen to this guy. I originally said — I didn't know how many bolts they had — if you just replace them with one of the nickel-based superalloys, it turns out they had three million dollars worth of stainless steel, and this would have been thirty million dollars, adding about twenty-seven million dollars to a sixty-million-dollar building, which was not a good idea.
So they started doing things. They went back to Sandvik in Sweden, and Sandvik said, use the duplex stainless steel. They said, well, can we use the duplex stainless steel? I said, sure, as long as you don't have any weld details. Can't weld this without real problems. It's good for pipe, and oil wells, and sour oil wells where you have hydrogen sulfide. We went through several things, and again, everybody's asking me — even though I was the idiot who didn't know what I was talking about when we had the first meeting — everyone is now asking me to approve the design. I said, nope, I'm not being paid to take the liability for this building, particularly if it comes crashing down like the Hyatt Regency collapse, which killed a couple hundred people. I would give advice, but I was not going to sign off. There is an engineer of record who has to sign off legally on this design, and I was not going to assume that responsibility for the peanuts they were paying me for consulting. If someone's going to allow me to make a couple million dollars profit, I can afford to go buy liability insurance. But anyway, those are other stories.
So 2205 is what they have, and it's actually working, and they made their deadlines, and it only cost them about seven or eight million dollars increase on the sixty-million-dollar building.
§9. Pitting, dealloying, exfoliation, and the chloride scale [41:41]
This is pitting of stainless steel. This was in the first edition but they didn't tell me much about it. There's a quarter — this is just a stainless steel beaker which perforated in sixty-four hours. It's 304 stainless steel. That's a hundred and four millimeters per year. It's a tenth of a meter — that's three inches a year. The resistance is due to exposure to water — it contained a chloride tablet, thousands of parts per million. But it's not a huge amount.
Here is chloride pitting resistance versus chloride. This is 304, 316, 2205, and pitting. This is what we ended up using at the Miami Art Museum, significantly better than 316 or 304. AISI 304 is down here, 316 is in here, and this is a 2304 which I don't even know what it is, and then 2205.
This is under-deposit attack. Here we've got some crevice, you've got a rivet, you have different ions in one area than another, and this is what the localized wastage looks like. The 316 stainless steel heat exchanger tube, due to the pitting corrosion. And there's the pitting corrosion that goes on, in the localized chemistry.
I told you that you need oxygen to have corrosion. Here's a little thing out of Herb Uhlig's handbook. Herb Uhlig was the guy who started the corrosion research here at MIT. You go up the second floor, it's the Uhlig Corrosion Lab, which doesn't exist anymore, but they have a plaque on the wall. Concentration of dissolved oxygen versus corrosion rate in milligrams per square decimeter per day. Don't ask me why these people use these units, but that's what they used back in the 1930s. Effect of oxygen concentration on corrosion of mild steel in slowly moving water containing 165 parts per million chloride. Pretty good correlation. You need oxygen to corrode.
This is some dealloying on a condenser tube — remember, this is a book on cooling water systems. This is the dealloyed area — pitting fouling, dezincified in this area, and then all the way through here. It gets spongy, porous.
These are pictures of exfoliation corrosion we talked about — in 7075 aluminum. Just flaking off in layers. This is a vertical view of exfoliation corrosion, and you can see it really is like a Greek pastry.
This is a wedge-shaped fissure due to corrosion fatigue. The crack opens up on the first stress cycle, it starts to oxidize, next stress cycle. As the crack grows, you get a wedge shape. Whether it's thermal fatigue or corrosion fatigue, you see these little wedge-shaped cracks full of oxide, and that tells you the mechanism.
Here's another thing — effect of chloride concentration on susceptibility of type 304. Test time in hours versus the crack specimen cumulative. As you march up from 10 parts per million chloride — if you condense the air in Cambridge, Massachusetts, because I've done it, and measure the chloride, it's 5 ppm chloride. The drinking water, which is pretty good in this area except in the winter when they're putting salt on the roads, is around 10 or 20 parts per million chloride. 100 parts per million chloride, you're getting to the EPA drinking water level. 1500 parts per million chloride, we're getting into the brackish water area. Sea water is like 30,000. You can see test time — this is a log scale — your stainless steel — this is actually our hot dog cooker, in a sense. Lots of chlorides in that, and those dogs and 304 just couldn't cut it at the high chlorides, and failed within days.
So I got back to my office, had this new book, flipped through it, put all these post-it notes on, and it just says all kinds of things that reinforce the types of things we were talking about yesterday. Chlorides are bad for lots of materials, not just stainless steels. Any questions? You need to ask more questions. Anyway, tomorrow we'll go back to the can slides.
This is the type of lecture I used to give, using my little thing, throwing books up here, and wandering around. I will appreciate your feedback on which one you prefer, or whether it makes a difference. He was incoherent anyway, so what do we care, we're not being graded. Thanks.