[Barbs]

Hi,

I am having some problems in my work in which I use a low melting point allow (melts at about 100C) to create radiation shielding. With a new batch of alloy I have struck a problem where the metal tends to crystalise as it hardens/cools.

The shields I am making are poured into moulds of high density polystyrene foam and are generally about 3 x 3cm to 10 x 10cm and are all 6.5-7.5cm high. The top cm or so of the shields is subject to the crystalisation which unfortunately in some cases makes them unuable.

I am working on the assumption that the crystalisation is a result of cooling time/speed but in all my trial and error or different techniques I have been unable to come to any conclusion as to what the cause is. I am forced to use this new batch of alloy because it contains less cadmium and is therefore safer to use.

Not being a chemist I am running out of ideas as to what I could try next to avoid this problem.

The closest I have come to an answer was when I sped up cooling from both the bottom and the top of the shields (remember they are 6.5-7.5cm high), but in this case the top and bottom were fine, but the middle crystalised.

I hope someone can provide me some eduction on why this happens and it may lead me to try something I have not thought of yet.

Kind Regards
Rob

[jdurg]

Crystallization will happen no matter what you do, but it's only noticeable when you have large crystals.  ALL metals crystalize as they solidify from the liquid state.  The thing is, a hunk of metal that we are used to seeing is composed of ungodly amounts of small crystals that are interwoven within each other.  This gives a solid, smooth surface which we see as one lump of metal.  In reality, it's millions upon millions of tiny little crystals that have formed.  Crystal size increases as the purity increases and the time it takes to cool down increases.  So to minimize the size of the crystals, you need a relatively impure sample of metal that cools rapidly.  You may want to introduce a rough surface in your mold which will create many "seed" points that allow crystals to develop.  This will aid in increasing the number of crystals thus decreasing the apparent crystals in your sample.  (I know this sounds counter-intuitive, but the more crystals you have the more 'metal' the sample will look and the less "crystally" it will look).

[Barbs]

Hi,

Thanks for your reply.  I have learnt some of what you say (about all metals cooling into crystals) during my research today.  And what you describe is exactly what is happening.  The top portion of the (lets call them) "blocks" is crystalising into larger crystals than is acceptible, producing a clearly visible matrix which includes holes.  The holes are the unacceptible aspect as we do not want these shields to have holes in them as that will reduce their effectiveness as a radiation shield.

I am experimenting further into speeding up the cooling process at the moment, with some success.  However, I have also noticed that a variation in the cooling rate is also causing problems.  Perhaps you are able to explain what is happening below.

The blocks cool on a chiller plate, that cools to <4C to speed up the cooling process.  The alloy that is immediately on top of that cooling plate is best (has the smoothest finish) and the 1-1.5cm of alloy on top of the block (ie 5cm or so away from the cooling plate) is usually the worst.  I experimented with using a fan to cool the top portion of the block which resulted in the top portion of the block being much much better.  However about 1cm below the surface, the crystal structure was larger and contained holes.  This layer under the surface does not appear when the alloy is only cooled from one direction (the cooling plate it sits on).

The above paragraph is what put the doubts in my head about cooling rate...and ultimately led me here.  Why would this change the way in which the allow just under the surface cools?

I will investigate use of a rough surface on the mould, but I think that might affect our end result in another unacceptible way.

Thanks for your time.

[jdurg]

What's happening is that the metal itself is cooling off and not transmitting heat in a consistent manner.  So the metal that is in contact with the cooling plate is able to transfer its heat quite rapidly to the cooling plate while that layer further above is unable to.  What you may want to consider doing is taking liquid nitrogen and submerging the entire thing into a bath of liquid nitrogen.  That will instantly cool everything off resulting in VERY tiny crystals.  The only problem is if liquid N2 gets trapped in the metal itself.  To avoid that, simply bathing the mold in the liquid nitrogen on the outside may be the best thing to try.