Making a New Metal

Once we have a recipe for a new metal, how do we find out if it lives up to its initial promise? We find out how metallurgists forge, treat and test a novel composition, and why...
20 May 2012

Interview with 

Dr Nick Jones, Cambridge University


Ben -   Once we have a recipe for a new metal, how do we find out if it lives up to its initial promise?  Dr. Nick Jones from Cambridge University invited me along to the Department of Materials and Metallurgy to find out more about the process and he also joins us in the studio.  Now Nick, thank you ever so much for coming in.

Nick -   Absolute pleasure.

Ben -   So, we've heard from Roger that you get given this sort of recipe.  Where do you then start?

Nick -   Well, as Roger said, they can be either given me on a bit of paper as a weight percent or atomic percent.  If they're an atomic percent, then I need to convert them into something a bit more useful for me which is weight percent.  And then I'm into the lab to go our raw elements cupboard where we have a whole variety of little pots containing pure elemental metals and then I start weighing them out, just like you would for baking in a kitchen, only with a more accurate set of scales.

The microstructure of a superalloyBen -   How important is it that it's accurate?  If you're getting atomic percentages, then presumably, your measuring has to be incredibly accurate.

Nick -   Well obviously, the more accurate we can be with the measurements I make, before I start making anything any further, the more likely we are to actually be on the composition that's been specified.  So, if I muck it up at this stage, very much like a cake, it's not going to be what they ask for.  At which point, someone like Roger gets probably quite annoyed with me and I have to have another go.

Ben -   And you took me to see some of the equipment that you actually use.  So, once you've got your measurement, you then need to melt it.  And for this, you took me to see an arc melter.

Nick -   So what we're looking at here is, this is an arc melter.  We use this for actually melting up some of our compositions.  So, we have our raw elements which we have weighed out and we place them in one of these little receptacles, so this finger-shaped recess here.  Having put them all into that recess, we would put the whole system under vacuum.  Oxygen goes into metals very easily particularly when they're hot.  We don't really want that, so we'll pump that down a reasonably high vacuum and then before we actually start melting these, we backfill the system with argon which is an inert gas before striking an arc, which is a bit like welding.  Then we use that to vary the current, the strength of the arc, to actually melt the elements together.  Things mix very well as a liquid, much better than they do when they're solid.  So we want to get the thing nice and hot, into a liquid form so everything mixes together and we get one uniform solid.

Ben -   So, that was your arc melter which works in the same way as arc welding, you're actually using a very large electric current and that generates lots of heat.  So why would you want to do it that way instead of using maybe induction - where you set up an electromagnetic field and that encourages the metal itself to get hot - or even in a more traditional oven?

Nick -   I think in many ways, on some of the elements we're trying to melt, we want to get some quite high temperatures.  Roger mentioned that in some of these Nickel based superalloys that he's designing, there are some refractory metal elements in there, and they have very high melting temperatures.  And one thing the arc gives us the ability to get to these very high temperatures.  So, we can make small amounts, and these are quite small, about the size of your little finger, to try out these materials.  Now, if we wanted to make bigger quantities, we might well use a different type of furnace like an induction furnace, but we may have to melt a couple of those elements together first to get them to a lower temperature.  So it's easier to melt them in these other types of furnace.

Ben -   And do you have any problem whereby if you're passing a current, you are obviously also creating magnetic fields around it?  Does that affect the structure that you end up with?  I mean, would you get a different structure if you had used a different way of heating it?

Nick -   I think that these things cool and that gives us the structures they form.  I've never actually personally looked at whether we get an influence of magnetic field.  I know some people do look at things in steels where they try and use magnetic fields while the material is at certain temperatures, but in a lot of the alloys that certainly I deal with, they don't have magnetic properties so I'd get away from that one, so I guess I get off lightly.

Ben -   So what do you do next?  You've obviously melted it.  You've allowed it to cool.  You have your little test sample.  What's the next step?

Nick -   Well one of the first things I would do is take a little slice of that material and I prepare it, and then actually go and put it into a scanning electron microscope.  And from that, I can use a technique of having a very quick check on the composition to see whether I'm actually in the right ballpark.  And from there, we would actually go and get it sent off to have a more accurate compositional analysis done.  If I'm a long way off the composition, like I've just weighed something out wrong because I'm having a bad day, and those happen!  Then I need to have another go.  There's no point continuing what could be a quite lengthy process of fully characterising and doing these very long heat treatments as we saw in our introduction.  I need to make sure I'm right to start with.

A micrograph of bronze revealing a cast dendritic structureBen -   So, you've said that these are samples of metal about the size of a finger.  If you're then cutting off bits, what size samples are we dealing with?

Nick -   Well that can vary, but typically a few millimetres in length and if you just imagine taking off the right tip of your finger, it could be something as small as that.  The amounts of material we can look at can be quite small actually and this gives us a representative idea of what might happen with these materials.

Ben -   You've just mentioned heat treatment and this was the next thing that you took me along to see.  So, let's hear a bit of that.

Nick -   So what we're looking at now is a box furnace.  This happens to one that goes to 1100 degrees centigrade.  They are, as they say, just boxes with heating elements down either side and then ceramic walls all around them.  We would use these for putting in our samples to give them heat treatments to allow lots of diffusion to occur when they're at high temperatures, and this makes sure we can assess the crystallographic phase that they will form, the different phases.  We do this for a long period of time.  We've got two more furnaces next to us that a colleague of mine is currently using.  There are at 800 and 1,000 degrees centigrade and they're both on for a thousand hours.  So quite long heat treatments to allow us to get these materials to equilibrium and make sure we understand what their equilibrium phases are.

(Door shut)

Ben -   Sounds that that could do with oiling I think, but what's the point of the heat treatment?  What changes would you expect to happen in that thousand hours at 1,000 degrees C?

Nick -   Well, it allows all the elements to move around within the solid.  So these things are still solid at these temperatures and as I said, the elements can diffuse in that material, move around to where they'd like to be, their lowest energy positions within that solid piece of metal.  As Roger mentioned early on, this could actually lead to the formation of precipitates in some case, which as he said, makes the material behave more like a composite.  And we may want to form these because they could strengthen the material.  So we need to form these precipitates and allow this diffusion to happen.  We need to give them the ability to do that which means high temperature and then the time to actually do it.  It doesn't happen very quickly.  Diffusion, particularly of some of the refraction metals, is very slow.

Ben -   So we're going to end up ultimately with a very different metal compared to when it was first melted.  When you heat treat it for this long, the internal crystal structures are very different and so, the properties should be different.

Nick -   Well that's very much the case and they can be.  Sometimes, what you get out when you've just cast it isn't quite what you want and hopefully, by giving these long heat treatments, we get to the equilibrium phase that they naturally want to form.  And then if we change those temperatures, we can induce these precipitates to form and therefore, yes, we can start producing a material with very different properties than would've been there in the as-cast material.

Ben -   And what then do you need to do?  So you've already taken an initial sample from the immediate as-cast metal and you know the structure of that.  Once you've done the heat treatment, do you need to do the same sorts of tests again?

Nick -   Absolutely.  We need to see what that microstructure now looks like.  So this is up from the atomic scale, what it actually looks like on the microscopic level.  We can see quite a lot from how these precipitates form, we have to go and have a look at them because the size they adopt and the morphology, so that's the shape, can be quite important.  And these things are being designed in.  As Roger said, they're actually looking to design these features in, so I need to check whether we've thought about it right and actually try to manufacture it, and if we haven't.  It's time to another go again.

Ben -   But as well as the chemical structure of it, presumably, you need to work out if the properties of that metal are the properties that you were looking for.  There must be mechanical tests that you need to do.

Nick -   Well precisely.  We use a range of mechanical tests.  This can be from something as simple as a little diamond indenter going in to give us some idea of the hardness.  Roger mentioned strength so we could assess that with the tensile test and we have machines which can do that at higher temperatures to give us some indication of whether the room temperature properties might be good, but is it also going to perform at the high temperatures that you find in a gas turbine engine?  Other things Roger mentioned, he said chromium in there for corrosion protection and so, we might look at oxidation and these things get very hot as I said in my piece about the arc melter.  Oxygen absorbs into the metals and then we've got oxides forming at the surface and that's bad because it's consuming your component.  So we might want to test how good the oxidation resistance of these alloys are because that's something that's key to their design.


Add a comment