SSW_S2013_06

Is a we on? Yeah okay, this is a piece of polyethylene pipe. Uh, the gas industry was tired of all the steel pipe corroding in the ground after fifty, sixty years and they have to replace everything. Um, plus they didn't like it when houses blew up. Um, so they wanted something that was corrosion resistant, and polyethylene will be around for millennia, okay. Just not a lot of bugs that eat it. Uh, but how do you put it together? You actually, they weld it. It's hot pressure welding. Um, in this case they take the two ends of the pipes, they machine them flat, they put a heater in between, and then they pull the heater out when it's hot and they just squeeze it together. And you extrude out the uh the contaminated surface, cuz it is still a contaminated polyethylene surface. Although polyethylene doesn't, we'll see later it has a very low surface energy, and so it doesn't want to attract very many things. If it's got a lower surface energy to begin with, then it's not going to get contaminated.

We talked about, metals have a high surface energy yesterday inherently, because they have long range electron bonding and stuff. Well uh, plastics have very low surface energies and so they're not as easily contaminated, but they still will have water vapor and oils potentially on the surface. If you look very carefully at this you actually can see that there's some residual stresses in this, because when you heat it up you're only heating up this end and the pipe expands. When you squeeze it together and then it cools back down, you'll see it's slightly dished in. It's got a shape with the extrusion and the joints here. Um, this is the center line, so it's slightly dished in. And because of that, while these things are strong enough, they're not great for impact resistance. But if you're going to bury it in the ground nobody should be hitting it hard, although people come along with backhoes and things like that and it's not a good thing. You should call Dig Safe uh before you dig. Um, if you don't, there will be an explosion and someone will call in someone like me to say oh, you broke the pipe.

Um, yeah, two questions. The first one is about uh shielding gas, do, is it used or? No, just straight air, because you like to use, you like to have dry air, you don't want to be doing this in the rain. I mean it's done outdoors right. Okay, and they just go down in the pit where they're laying this pipe with a heater, electric heater, and it takes, if you go through the thermal conductivity of the plastic, it takes three or four minutes to heat this thing up back a quarter of an inch, or you know 3/8 of an inch on each side, and then you just extrude it together. It's basically a forge weld. Weld in plastic.

Student: If you were able to put a twist, would it be more like a friction stir weld? Would that improve the bond?

Well, it's not going to improve the bond, because you, the twist is not going to push the contamination out. You want to get the contamination out, and you've extruded it out by just squeezing it together. If you twist it you're just smearing it around right. You get new material, you want new material, okay right. You want a new surface. You want shearing. Wow, you want shearing in the proper direction. You're right, I can get shearing by twisting, but that doesn't get the junk out. You will need to get the junk out, and then twist.

Student: Because what I'm thinking about is avoiding a mold line.

That's what I, oh well in fact you do have a mold line, and the reason this is weak is there's no polymer chains going across that interface. And so in terms of strength it's not bad, but in terms of impact energy it's lousy because there's nothing going across here, okay. Small right.

Well in fact that's the problem in joining plastics in general: you can't get polymer chains going across the interface. They don't diffuse. They're too big to diffuse, and so you can't, you always end up with a strong demarcation which is just van der Waals bonding. Now we know in theory van der Waals bonding can give you tremendous strengths, but in fact for all the other reasons you get imperfections and voids and whatnot at the interface, and it doesn't give you great strength. But in fact the strength of that, if you do a tensile test, you can get those, set with parameters such that you'll probably fail in the base material or you'll fail at a substantial fraction of the yield stress of the material. But impacts are still a problem, okay. Creep probably, uh creep, but yes, if you put this under tension. But you know it's supposed to be buried in the ground, so this application is not a bad application. And I, frankly, I don't, I know they're concerned about impacts, but I have never seen a failure of this stuff. Most of the gas pipe failures are old cast iron pipes that were put in a hundred years ago that have just corroded, okay. Um, or there's steel pipes, smaller diameter steel pipes that were put in sixty years ago, okay. And in fact those steel pipes that were put in sixty years ago probably perforated about thirty or forty years ago, but the soil, I mean it's in moist soil, and the soil basically plugged the leak.

And anytime you ever dig in, dig up around where, you know, they're gas pipes, you know that you're going to open up leaks. So you know, twenty years ago one Sunday morning at 7:00 my neighbor calls me and tells me there's water bubbling up between the sidewalk and the driveway in front of my house. And sure enough they had planted a tree on top of the water box back sixty years before, and eventually the roots of the tree and the corrosion caused the water box to break. So I call up the town water department and they come out on Sunday morning, and I'm standing there at 10:00 and they're digging. And as they're digging in the ground I start smelling gas. And just then, it was Boston Gas back then, the gas truck pulls up. And as I'm smelling gas they had just uncovered the pipe. One of the guys with the backhoe, next to the backhoe operator, jumps down in the hole, takes some chewing gum out of his mouth and he plugs the hole in the gas pipe. I said, how do you guys know to call Boston Gas? They said, we call them whenever we start the work, cuz we know when we uncover the pipe we're going to uncover a leak. Okay, so all the gas pipe out there is leaking.

And in fact a few years later they came down, they come by like every three years, and they have to do, they stick probes in the ground to see how much natural gas is in the ground. And my house and another house about, or seven houses down, they found so much gas that they came in and they sleeve the steel pipe with the plastic pipe, okay. Uh, and they have these little connectors that they've developed over time, um, to be able to connect steel to copper, steel. They're just manual connectors with the uh compression seals and stuff between the metal and the plastic, and they usually work pretty good. There's another story that goes on with that but I won't tell it right now. Um, but it did fill up my house with gas one time, but it didn't blow, okay, or I wouldn't have that house. Um, anyway, um, so that's another form of pressure welding. It's hot, it's not necessarily cold.

Um, I did want to talk about metals and what metals are easiest to cold weld. There's a guy Tylecote [Tylecote] who back in 1954 in the British Welding Journal did some studies on what metals are easiest to cold bond, basically with just normal compression hardness where you're extruding things out, you don't have a lot of shear, okay. And he found that indium, he only needed a 10% deformation of two indium pieces before he could get a fairly strong bond, okay. Uh, there's data on this, this is in your notes um that I handed out. And aluminum is the next easiest, and then tin, and then you get down here for more structural materials like iron and copper and stuff. In any case, indium is by far and away the easiest material to cold bond by just squeezing it together. Um, and the reason he found was, or he plotted oxide to metal hardness ratio. And if you think about it, the harder the oxide, the less you're going to smear the oxide across the surface. Copper, which has a relatively soft oxide, and silver, which really doesn't have the oxide but has a sulfide, silver sulfide is very soft, and basically if you try to squeeze two pieces of copper or two pieces of silver together, that silver sulfide or the copper oxide just smear across the surface. You might be creating new surface with your deformation because down here at 80% deformation you've increased the surface by fivefold, but all you've done is smear that layer of oxide thinner and thinner.

It turns out indium, which has a, well indium is a very soft metal, I wouldn't say it's oxide is extremely hard, but it's not soft. It has a very high oxide to metal hardness ratio. And what happens is when you squeeze it together you deform the indium and these little islands of hard oxide break up exposing indium between the cracks, and the oxide metal contact. So it turns, indium is the easiest. Aluminum, which has a very high hardness oxide, uh sapphire or corundum is number nine on the Mohs scale of naturally occurring materials. It's number two behind diamond, okay. Aluminum, this is pure aluminum, and this aluminum with some alloy aluminum, and you can see that the harder the oxide and the softer the base material the easier it is to deform the base material without just smearing your oxide across the surface, okay.

Um, so why is that important? Well, it's not as important now as it used to be, but about fifty, sixty years ago when they were still making single transistors, or even sometimes today in critical applications like we'll say some nuclear weapons which we don't know about, um, when they have to bond something at room temperature and they actually have to get a true metal seal, they will often use indium. And uh, I don't have any but they used to have single transistors, when they went from vacuum tubes to transistors, they'd have a single silicon chip with several leads coming in the base, and they'd have a little metal can on the top. That little metal can was just squeezed on, and you had a little bead of indium, and so you had indium, and you were making an indium solder bond at cold, at room temperature, okay. So it was used many years ago.

So now I want to go to adhesive bonding. And adhesive bonding is different than these two rules that I told you. If you're going to get a good bond, you have to get rid of surface roughness and you got to get rid of surface contamination. Now adhesive bonding, whether you believe it yet or not, I'm going to try to prove to you that we get rid of the surface roughness by interposing basically a liquid. Now that liquid may be a fairly high viscosity liquid initially, but usually it turns out it's a lower viscosity liquid, okay. Um, but so you get rid of the surface roughness by interposing a liquid that fills the valleys and the, the hills and the valleys um of the uh surface roughness. Adhesive bonding is unique in that it does not get rid of the surface contamination. It just buries it beneath the surface. As a result, it turns out, you've got to clean the surfaces. If you want to get good adhesive bonds, you got to get off those easy things to get off, like grease, that it's not, you know, get off with the detergent. You may not be able to get rid of the oxides, you may not be able to get rid of some things like the oxides, there other contaminants on the surface, but you want to get the highest surface energy possible and get rid of all those low surface energy greases and other things um to get a good bond. I mean obviously anybody knows if you put oil on the surface and try to stick Scotch tape to it, doesn't bond, okay, because oil has a very low surface energy.

So there are people who start thinking about adhesion, and this book was written, by edited by people who worked at Xerox Corporation research labs in Webster, New York. It's called Fundamentals of Adhesion. And they start kind of where I started with, um, actually they start even before that, um. But on page six they start with the Lennard-Jones potential and bringing two atoms together, and that's for a perfectly clean surface. So they're starting as if you had a perfectly clean surface, and they don't, you know, they're not really starting where you need to on adhesive bonding, which is where you have an oxide surface, a dirty surface. But they go back even further. On page four they have gluons and adhesion of nuclei. The name gluon, created by physicists, gluons are those particles binding quarks together. And quarks have been postulated to be constituents of protons, neutrons. Do I really care, okay. So typically that's where that book goes, okay. Um, and this is what you often loves books, well I'm going to retrieve it cuz I can't afford, I got to use it next year you know, next time I teach the class. You don't know how many times that book has been in the trash, okay. But this is sort of the approach often of some scientist who starts all the way back at gluons.

We're going to start at the engineering end of, we're going to assume that we have a real, relatively clean but still oxide contaminated surface, and we're going to look at what really creates the strength of an adhesive bond, okay. And some of this is sort of my terminology that I've developed over the years, but I have two types of adhesive bonding. Um, it's not my ideas necessarily, but it's kind of, when I talk about type one and type two, that's definitely kind of Tom Eagar. This comes out of a book, um, second edition Handbook of Adhesives. I can't find my second edition which got a blue cover, I think I loaned it to somebody and they kept it on the no return plan. This is the third edition of the Handbook of Adhesives by Irving Skeist [Skeist], who was a consultant in the adhesives business. Um, but in any case um, he def, I call him type one and type two.

So type one — [Tom writes on the board: type one = surface tension.] — and we're going to see how surface tension will actually hold materials together. And Shannon, you may have seen this before. Steve, I think you probably, you've seen me pull out the, your Johansson blocks before right? You haven't? Okay, so well gee, you haven't had all my glasses yet, all my little demos. [Tom locates a case.] I just got to figure out how to open it. Now, here ring, what did I do here? We go, I think I got it. Here we go.

So does anybody other than Matt know what a Johansson block is?

Student: I've seen it in your class only.

Okay. So Johansson blocks are length standards, and I have a little certificate here that traces the dimensions of these blocks back to the National Bureau of Standards or National Institute of Standards. This is the StarrettWebber Gauge Division. And somewhere on here um it'll say, all Masters and Grand Masters um are calibrated directly by the National, the Grand Masters are calibrated directly by the National Institute of Standards and Technology, okay, in Gaithersburg, Maryland, okay. Um, these are not Masters, they're not Grand Masters, but these are down accurate to about 50 microinches I guess. If I look at my certificate it might tell me that these things are accurate to .02 microns, well actually 2 to 4 microinches. These are 2-inch blocks I think. 1.7 micro plus or minus 1.7 microinches, which is plus or minus .04 microns, okay. So we're down around, I know you can convert that, that's 40 microns, 40, 1.7 microinches is um uh .50, .17, right okay. It's two-millionths of an inch roughly, plus or minus. But they have to tell you, and I think at 72° C, or 72° Fahrenheit, or 20° C, because if I change the temperature 2° they're that much longer or shorter. But the surfaces are polished, the end surfaces, and a machinist who wants to measure something better than he can with a micrometer will ring them together. And they're supposed to have a slight oil film.

First ones I bought were hardened steel, but they rust over time if you don't take care of them. But I now have bonded these together just because they have very flat surfaces. There's still surface roughness, but it's polished, so it's small surface roughness. And I've interposed a liquid film. Now typically I didn't do it, but well, they, it's not the strongest bond in the world, and in fact you will have the opportunity to calculate, if you did the problem sets, the adhesive time and strengths of these things. And it's directly proportional, as we'll see, cuz we'll go through and do the calculations. Um, or I'll show, give you the equation. Um, if you were doing the problem set, which I think most of you are probably going to do presentations now.

Um, yeah, to get a little grease from your skin and stuff, um, I actually have some lubricant here and I was just sort of using the residual lubricant. You have to ring them together and shear, and you have to be, I'm going to pass them around. You can try to do it if you want. [Tom passes the blocks.]

Anyway, so they don't, you have to have just a little bit of liquid in there, and the liquid forms, um, becomes the adhesive. This is the ideal adhesive joint: you have two solid adherends, and you have some liquid that you're going to interpose, and some big radius 2R, and some spacing D. And if you go through and analyze this in terms of surface tension — and this is, here, do you want to try the, there's some bigger ones, there's also some, if you want, here, here's a shammy, and there's some more oil. You're going to get more oil than you ever wanted, so you have to clean some of it off. Uh, but yeah, usually a machinist will actually take the oil from their hands. I mean maybe you can do it from your nose, but most people have enough oil in their hands. So you just get them off, clean, and here's the formula for type one adhesive bonding.

You got that same little diagram I showed you before. The change, difference in pressure between the inside and the outside of this adhesive, so out here in the air versus inside the adhesive, is just the surface tension of the adhesive times the curvature. So if I have an adhesive that wets the surface, there's some liquid-vapor surface energy here, and the curvature is just 1/r + 1/R where these are the two principal radii of curvature. But the little r is actually the, if this is the center of where I'm measuring R, this little r is negative if it's a wetting surface, okay. So it's 1 over Big R minus one over little r. 1 over Big R is very small compared to 1 over little r, and so the whole thing just becomes minus gamma over r. It's negative, so it's lower pressure inside in the blue than it is outside. You have a pressure differential that's holding these together, okay. This is type one adhesive bonding. We're not talking about any roughness of the surface. We're just talking about the surface energy and the distance of separation. The thinner the joint, the stronger the joint. And if you get a little bit too much dust on those, you can't ring them together close enough, they got to be dust free, okay. You're actually ringing those together and getting something on the order of a 5 micron or 10 micron thick joint, okay. Um, and in fact we may not be able to do, it may be just my chalk dust, I may have contaminated them enough, they both working, oh they're working now okay fine. One the big one okay fine.

Um, in any case, so that's bonding of Johansson blocks. And we'll talk about how type one adhesive bonding is strengthened. Type two adhesive bonding, and again this is Tom Eagar's type one and type two. Type two adhesive bonding is nothing more than very rough surfaces, and it's mechanical interlocking. So type one is surface tension, type two — [Tom writes: type two = mechanical interlocking.]

And I'll tell you that type two is the predominant type of adhesive bonding. Now it's not always the type of adhesive bonding. If I'm going to glue the mirror attachment to the inside of my windshield in my car, if I'm General Motors, that's going to be a type one adhesive bonding, cuz that glass is smooth, okay. And there's not any mechanical surface roughness to the glass. Now the other side, whatever you're doing, you might introduce some mechanical roughness. But many times, if you want to strengthen the bond, if it's type two bonding, you actually want to take some sandpaper and roughen the surface, okay. There are times when having a little bit of surface roughness even in type one, as long as you don't destroy the thinness of the joint, a little surface roughness increases the surface area and can enhance the effective surface energy of the joint.

But if I'm talking about bonding something like aluminum for aircraft, where did I do with my pointer now, oh here it is. Um, if I want to bond aluminum, this is anodized aluminum, says aluminum, series 2024 aluminum, that's typical standard aluminum sheet metal. Um, there's a, it says stereo micrograph but it's an SEM, that might be a stereo micrograph, this is SEM I think. Um, any case, and here's a drawing that someone has produced. Essentially when you anodize aluminum, you put aluminum into a uh a bath, in this case let's call it sulfuric acid bath, okay. There's chromic acids anodization of aluminum, sulfuric acid, I think chromic acid gives a little bit better bonding. Um, but you have to deal with hexavalent chrome which is a carcinogen. So anyway, when you do this you basically run the current and make your aluminum substrate the anode. That's why we call it anodizing. You're corroding the aluminum, you're bringing oxygen to the surface, you're growing an oxide. And we call it growing an oxide film, but actually it has to be a porous film, because aluminum oxide is a great insulator, and in fact there has to be some channels for the electrons to get in, so, and the, or the oxygen to get in and the electrons to flow in a liquid. So you're actually building this little cells of aluminum oxide. It's a very porous structure.

Now if I anodize aluminum and I want to use it for adhesive bonding I'm bowing, okay, here's my rough surface, I've increased the mechanical surface roughness. I slap my adhesive on there and I get a great bond strength. If I had not anodized, I just went with an as-sanded aluminum surface, I might get 1/5 the bond strength that I would get of doing adhesive bonding on an anodized surface. If I'm anodizing the aluminum because I want corrosion resistance, then I have to seal the aluminum, uh, the anodizing. And all you do, you seal it by, after you've anodized — and this is, we're not talking about adhesive bonding here, talking about some window frame or door frame that we're putting the anodizing on instead of paint — you got all these pores and you want to do something to seal it. And they call sealing the anodizing, you basically boil it in hot water, okay. And what happens, now I'm not trying to pass any current through it, I'm just boiling it in hot water, and I will grow and fill up those pores and get a continuous aluminum oxide coating. So you have to be careful about this, okay.

Um, the first time I was ever down on Cape Cod I went down there in the middle of a 6-inch snowstorm because some guy here in Boston had bought a nice new 40-foot sailboat. It had a nice wooden gunnel with anodized aluminum rails up above the wooden gunnel, this is the sides of the rail of the boat. And it was, the aluminum was pitting. It was only a year old, and it was going to cost a fortune cuz you had to remove all that beautiful mahogany woodwork in order to be able to lift the aluminum rail off. And he wanted to know why it was pitting. And I had to go down there in the middle of the snowstorm and look at it, and I determined that, well, whoever had anodized it forgot to seal it, okay. They didn't know that if you want corrosion resistance for your aluminum oxide you have to continue the process after you form the porous coating. You have to seal it by making continuous coating. But if you're going to um use it for an adhesive-bonded prep surface, you don't want to seal it. You want to leave it nice and porous, okay. So it's mechanical interlocking.

And well, actually here's some data from that same materials handbook. And here's chromic acid anodizing. This is crack length versus time. You're just loading it up to a high level and seeing how long it takes. This is um, oh, PAA process, I can't remember, um, phosphoric, it may be um. [Tom searches the page.] I have to go back and get the book out, but yeah it might be poly, it's different types of anodizing acids, okay. You can anodize lots of metals, unfortunately not iron, because if you try to anodize iron you're just going to form rust. And the rust has a larger volume than the underneath steel or iron, and the iron oxide, iron hydroxide that forms just flakes off. So you can only successfully anodize things where the oxide to metal volume ratio, what they call the Pilling-Bedworth ratio, is such that the oxide will not just flake off, or shrink. Some things actually shrink in size and they crack like a dried river bed, okay.

Um, but I drew niobium wire, my, and I used to anodize it in ammonium hydroxide solution. Just put a little ammonium hydroxide in a big long, had to anodize a long wire, had a long tube of glass, filled it up with a solution, and uh um basically used a 45-volt battery and just stuck it on. Get all this bubbling going on cause the oxidation, and pull it out. It's nice beautiful blue color. And that would hold the lubricant for the wire drawing very well, cuz I had this porous anodized surface on the niobium. Uh, twenty years ago, colored titanium jewelry was popular, okay. And I remember reading something from the Worshipful Guild of Silversmiths or Goldsmiths or something in England, jewelry makers basically. And they were saying that the best material to anodize titanium to get the different colors — and you get the different colors by just how long you anodize it, it's a different thickness of the inorganic or the, yeah the inorganic coating will refract the light differently — and so you put it in at different voltages, you'll get different thicknesses, and you can get um, on titanium you get brilliant yellows and some reds and eventually end up with blues and purples when it gets fairly thick. But in any case, their favorite material for the Worshipful Guilds was Coca-Cola, okay.

Now I never did ask whether they use Diet Coke or leaded Coke. Uh, but Diet Coke is nothing more than phosphoric acid, okay. They used to give, you know, Coke was developed at a drugstore in Atlanta, Georgia, and the guy uh basically took what were called phosphates, and people used to drink phosphates, sodas, and he added some cocaine to it. Uh, I'm sorry, yes he actually did. Actually he act, it wasn't cocaine like the purified stuff we have today, but he added uh, that's why it's called Coca-Cola. You can go look it up on Google with history of Coke if you want. Um, actually there's other soft drinks that came out, started out as drugs. Anybody know which one I'm thinking of? Seven Up, back in the '30s, was lithiated lemon-lime soda. So now lithium is, used as an antidepressant, but in the old days you could drink a Seven Up. And you can go on eBay and you can find labels from the '30s that you can purchase of Seven Up being a lithiated soda. I once mentioned this to a guy and he said, that's ridiculous, that can't be true, he bet me, he said I'll give you $500 if that's true. Anyway, so I went and I looked it up, sent him the links, and he sent me a check for $500 and I never cashed it. I sort of like Mitt Romney's $10,000 bet, okay.

Um, so if we want to understand, I just talked, well actually before I do that, well no let's go ahead and do this. Um, there's one guy that, where I first learned about what really makes adhesive joint, is there's a book called The Science of Adhesive Joints by J.J. Bikerman. Well this is 1961, but J.J. Bikerman was a professor over here in civil engineering, okay. And he's an engineer, he wasn't some guy who's starting with gluons and the adhesion of particles and uh of quarks forming protons and stuff. Uh, he actually was an engineer, and so he looked at the kinetics of how we form an adhesive joint, okay. I said we have a liquid, and it really doesn't matter whether it's type one, where we have surface tension giving us the strength due to the pressure difference between the inside and the outside, and you want a very thin joint, or it's mechanical interlocking, which is the more predominant type of adhesive bonding. You still are going to interpose some sort of liquid. Now that liquid could be a pretty thick liquid. If you bond bricks, mortar is your liquid, and your mechanical interlocks are pores in that brick. Now if you're bonding glass bricks, you got to have something else in your mortar to kind of make it stick very well, either that, or you got to sandblast the glass to get some surface roughness, okay. But that's how you do your surface prep for the two different types.

But the kinetics are, if I want to understand the spreading of a viscous liquid between two plates where I'm pushing down with one plate on the other, um, and you, I sometimes call this the Scotch tape lecture, okay. If you're trying to apply Scotch tape you press that tape onto the surface, and it has this sticky adhesive, but you want to understand how thin I can get that joint, because that gets to be critical particularly for type one. There's something called the Stefan equation, okay. And if you go through Bikerman's book — and I never have figured out if this is the same Stefan of Stefan-Boltzmann, but anyway — you got some velocity, he's going to do it as kinetics, velocity is, you have some H times the length, and that you're flowing it out, and you got some force necessary to do this. And he goes through and solves all this stuff. It ends up with a formula that says that the force times the time over which you're doing this, because you're basically, think of a shock absorber, you're pushing a viscous liquid, and so the faster you push the bigger the force has to be right, the slower you do it the lower the force. So it's the force-time product is equal to four times the viscosity, uh, let's look at it over here, well actually let me look at it over here. Force time is equal to 3 * the viscosity times V which is volume, over 1 over H to the 4th which is the initial thickness, minus 1 over H1 to the 4th which is the final thickness, okay. Um, anyway it's, I think this stuff is in your notes, oh wait a second it's in my notes here we go.

It's been a while since I've looked at this um. So it's the Stefan equation, the little f was a pressure, the big F is actually force, but and that's just got an area in it obviously. Force times time, or pressure times time, is 3/4 R² the radius of this thing, uh, times the viscosity, 1 over Hf² minus 1 over H₀², and the other one that had H to the 4th power was because I had to move an area over πr² right or whatever. Um, anyway, in all of this, is the thickness, this is the radius, this is the pressure and the time and the viscosity. So if I'm going to change some of these, I could change the thickness and get a thinner and thinner joint, just like those Johansson blocks. And I can, that's going to strengthen the joint as one over the thickness squared.

What can I do? I can't change R very much. I could try to apply more pressure, and you've seen people do that, if the Scotch tape falls off, the masking tape falls off the wall, they go up and they just push harder right, apply more pressure. And that means it will stay up longer right. It doesn't mean it won't, it'll stay up forever, because this is a viscous flow problem. That thing is just slowly creeping off the wall. In fact that's how all the adhesives work, um. It's basically interposing a viscous liquid between two solid surfaces. This is the kinetics of forming the joint. It's also the kinetics of breaking the joint.

Student: Have epoxy in that as well?

Yes, we're going to talk about that, okay. That's, you're just jumping about two minutes ahead. So, but what else can I change here? I can change the pressure. That's just linear. This goes as the square. What else can I change in here? Viscosity. How can I change the viscosity, and how much can I change the viscosity? I can change it by ten orders of magnitude. I can go from a liquid to a solid. And now, if this is the kinetics of forming the joint, and I put a low viscosity liquid in, I can form it very quickly. And then if that converts to a solid, then all of a sudden I have five or six orders of magnitude increase in time. How many seconds in a year folks? About 10 to the 8, okay. So if I get 10 orders of magnitude this thing could last for 10 years. Now many of you know adhesive joints under stress really don't usually last 10 years. They might last two or three years. Well you know, start looking at this right, it sort of works out.

All you're doing with adhesive bonding in most cases, you're trying to find adhesives that increase the viscosity. And Bikerman goes through and he gives some typical values of how long it takes to form a joint. This is actually right out of his book. Force and pressure, 14 PSI, 1.4 PSI, viscosity in gram-seconds per centimeter squared, which I think is uh um, well, centipoise, yes thank you, um. There's also Saybolts which are kinematic, etc. Anyway viscosity units are just as bad as slugs and pounds, I mean you know. Uh, anyway, radius 1, 1 cm, 10 cm, different thicknesses of joints, 10 mils, 1 mil initial thicknesses, hi I guess is initial, and the time to form the joint, 7 and 1/2 milliseconds, 75 seconds, or 7500 seconds, okay.

So now we get to something critical: putting the labels on beer bottles, okay. You have to have a very low viscosity liquid to be able to put the label on a beer bottle in the time you have available. How much time do you have available to slap that label on a beer bottle? They're going through at about a case a second through the beer lines, the bottling lines. Some of them are only 20 bottles a second, but typically, actually I take that back, the bottles are about 20 a second, the cans are about 24 a second, okay. Now they actually, I treated it a little bit, they're on a great big wheel almost the diameter of this, well 3/4 of the diameter of this, the length of this room, the width of this room. And so it actually, it takes more than 1/24th of a second to fill the liquid into that can or that bottle. They're actually under the filling spout for about half of a revolution or a third of a revolution. So it's taking a third of a second just to shoot that liquid in there. And you know, they got a person sitting there, and when the beer bottle isn't completely full, it got a lot of foam, they just sitting there and they pull it off the line, the conveyor line coming through, and they throw it over their shoulder into a dumpster. And you look at the dumpster, they have a 3/4 inch hole and the beer is just draining out of that hole about halfway into a drain pipe on the floor. And that's what happens to the beer. Cuz when I went through these plants, which was thirty years ago, uh to watch this, beer at that time cost 16 cents a gallon to brew, okay. Today maybe it's 32 cents a gallon. Beer is cheap.

The person there said, well I used to work for Welch's grape juice and the grape juice cost $9 a gallon, and whenever we had a packaging problem we had to figure out how to recover the grape juice cuz it had real value. The nice thing about making beer is, it's not much more valuable than water and you just run it down the drain. Now that's not true, beer actually cost more. Why? Marketing. You're paying for all those ads you see, okay. But that's another story I can tell you more about. But it turns out, in order to have an adhesive for something as simple as the labels on a beer bottle, it has to be very low viscosity in order to make the joint quickly enough, okay.

Well how do I increase the viscosity of something, if I, now I've made the joint? And there's three basic ways. Now viscosity can actually vary by about 20 orders of magnitude. Um, the highest viscosity ever calculated, cuz viscosity is just amount of displacement versus a force over time right. So you want to have the longest period of time, the smallest displacement, and the biggest force, to get a high, to be able to measure a high viscosity. And I mentioned Egon Orowan, who was one of the guys who had discovered dislocations by watching the maid curl the rug and kick it across the room. Professor Orowan, a mechanical engineer, he's passed away now, he'd be about 110 if he was still alive probably. But anyway, he once estimated the viscosity of the Finnish coastline, okay. All the rocks in Finland, at 10 to the 22nd poise. I don't know if it's Poise or centipoise, so but when you get 10 to the 22nd who cares, right? Um, but viscosity can change by about 20 orders of magnitude, from something that's rock solid like a piece of sapphire or diamond, to, you know, in terms of shear strength to something like a very low viscosity liquid. Water is about 1 centipoise roughly, but there are things that are more fluid than water. Well gases are about a thousand times more fluid, but talking about liquids, there are some things that are a little less in the centipoise, and then up to diamond and stuff. And if you start looking at the shear strength of diamond, you'll find it, it's about 10 to the 18th or something. I don't know how he got his 10 to the 22nd for the Finnish coastline, cuz who was, who had the force gauge there for the last million years, you know, I mean, anyway.

Uh well, you don't need to see this stuff, I meant to flip that over. Okay, so the first one, you may have seen it, how do you increase the viscosity? What do you, know think of an adhesive, you know, let's take a postage stamp. It's dry, okay. And so how did you lower the viscosity in order to use it? This is the old postage stamps that you had to lick, right? You interposed, you mixed it with water, you lower the viscosity, and then you let the water wick away through the paper. That's called solvent removal, okay. Postage stamps, okay. Elmer's Glue, remember eating the glue in kindergarten, you know, when you can stick your finger in the white glue, paste, if you left the top off the bottle, it hardened up right. So there's solvent removal, that's a common type.

Uh, you mentioned epoxies, right? Epoxies, polymerization, okay. You basically start with liquids, you make your joint, and then you allow some polymerization reaction to go on.

Student: Yeah, but epoxies and cyanoacrylates are really fundamentally different in terms of what you can do with them right?

Yeah okay.

Student: Which ways? Like cyanoacrylate, you can, if you're doing surface, you could never use a cyanoacrylate to, for example, the epoxies are used as, like, when you're building a structure for anchoring different components to each other, and you have trouble aligning them you just put them all in a fixture and just kind of, you know, squirt epoxy in joints to kind of fill in the space. So cyanoacrylates are really bad at kind of filling large gaps right?

And but you're also thinking of epoxies as sort of these sort of gummy things, you know, high viscosity. Some, oh, I've seen epoxies that are almost as fluid as cyanoacrylates — not as fluid okay, but like maple syrup, okay — when they're liquid. And they typically have a mixing gun, a little screw, okay, and they come squirting out and they're fairly fluid. You're right, when I'm trying to fixture something I actually want something that's got a little stiffness to, a higher viscosity. But I just gave up two or three orders of magnitude on strength, bonding strength, cuz I started out with something I'm not going to get this thin, I'm not going to use the thinness of the Stefan equation there. I'm actually going to do mechanical interlocking, okay, on a big, on a macro scale, is when I'm using that type of, putting, you know, doing something anchoring bolts in the wall, or like the OMAX water jets, most of those parts, a lot of that part, that machine is just epoxied together for vibration resistance right, probably fatigue and fretting and everything else. But it's epoxied together because they want everything stiff in terms of lack of fretting and things like that.

But cyanoacrylates, for those of you that are not familiar, Crazy Glue, okay. And it's a polymerization reaction. What is the catalyst for? Yeah, it is water, but what type of water specifically? Hydroxy radicals. It's not just water. Crazy Glue sticks better to paper than it does to wood, because paper is slightly basic. Sticks your fingers together really well because you have a slightly pH slightly above pH 7. For surgery, yeah, Dermabond, yeah, Derma bonding, okay. Because it's the OH that is the polymer, if you look at the, if you're a polymer chemist and you want to look at the polymerization reaction of cyanoacrylates, it's the OH. You have OH even in things that are acidic, I mean there's an equilibrium between water, it's just you have a lot more OH's in things that are basic. And therefore Crazy Glue doesn't work as well on most woods because woods are slightly acidic. Your fingers are basic and glues them together just great, okay.

So um, there's another one, Loctite. It's now a billion dollar corporation. What's the polymerization reaction with the Loctite? It's an anaerobic, if that helps you. You've heard of, have you done exercising, heard of aerobics? Okay, aerobics is with oxygen, and anaerobic is without oxygen. And so it turns out Loctite, when you put it down inside those little threads between the threads of a screw, there's no air around. And if you go looking at your Loctite adhesive, you'll find somewhere in that system they have a little hole so the air can get in but the liquid won't get out in your jar, because once you exclude the air this thing polymerizes, okay.

Student: Yeah, what about like also UV, is that in the same kind of, okay?

UV is um, actually I probably add a fourth, there are three common ones, but sandpaper for example. Um, the rate at which they make sandpaper, you cannot afford to slow down your line to bond the sand to the paper to wait for something, solvent removal or polymerization, these things are too slow. And the other one, which is melting, might as well tell you, melting. So I add UV or electron beam. [Tom writes on the board.]

It turns, in a sense they're, you can say they're part of polymerization, because all you're doing is breaking a bond with the electrons or the ultraviolet light, and you're allowing it to polymerize. But they actually, um, I guess since I've never seen it, um, there's a lot of proprietary technology goes in, but sandpaper is often electron beam. They put down, uh, you know, they put down wet sand basically with an adhesive, and then it goes through a high power electron beam which polymerizes it very rapidly because they can't wait for the chemical reaction of polymerization, of the kinetics of liquids getting together and reacting, the diffusion kinetics, or solvent removal evaporation. They could be sucking with a huge vacuum system, and anyway, plus how do you get, how do you seal the stuff coming into the vacuum system if you get sandpaper? So there are some, just like the speed at which you slap on labels on beer bottles is an issue, the speed at which you bond sand to sandpaper is an issue, and people do speed these things up by UV and electron beam curing. I've never heard of a laser cure process, but lasers don't have enough energies per, energy per photon, unless you had a UV laser. And if we had a UV laser we could do all kinds of neat things, okay. Uh, actually they have spent billions of dollars trying to develop x-ray lasers. Well that's another story, okay.

Um, so that's basically how you do it. Just increase the viscosity. Use the Stefan equation, well, whoever uses the Stefan equation. Um, I've only met one other joining engineer who even knew what the Stefan equation was, but it's fundamental to adhesive bonding.

Um, now let's talk, this comes out of the second edition of Skeist, um. And he's basically got a bunch of hydrocarbon compounds. From polyvinyl chloride, this is a carbon backbone, these are all carbon backbones, this has two chlorines and two hydrogens in the chain. And he lists the coefficient of friction, remember I told you friction or lubrication is the opposite of joining and stuff. So he's looking at coefficient of friction. Actually the guy who did this is, I can't remember his name cuz I don't have the second edition uh with me right now, but it was a guy at the Naval Research Lab who wrote this chapter, okay. Was a good chemist. And he was looking at the bonding, critical surface tension dynes per centimeter, which is ergs per square cm, which is .001 joules per square meter. 40, polyvinyl chloride. One chlorine rather than two drops to 39. Polyethylene, there's my pipe, my polyethylene pipe, 31, all hydrogens. Put a fluorine in here, polyvinyl fluoride, 28. Two fluorines, 25, polyfluoro, polytrifluoroethylene, so I got three, and it's 22. And I put four fluorines in there, Teflon, I dropped to 18. Water is, oh sorry, uh, I just, enjoy the Scotch tape, I call this the Scotch tape flexure, sorry, thanks. Um, so anyway, water, Teflon has the lowest surface tension of virtually any solid. They talk about water on ice, you know, water on ice and things, we'll talk about all that stuff on Monday, okay.