Gallium Alloys and their melting points?

[some bloke]

TL;DR

I am interested in creating gallium alloys with the specific purpose of manipulating their melting points. Is there a known process to follow, or is this going to be trial and error?

I am looking to try to create an alloy of gallium (probably with Tin) with a goal of increasing the melting point from 29°C to around 40°C. My current theory is to simply try to average the two melting points out by mass, which my calculations imply means 95% Gallium to 5% tin.

my calculation is (% used x melting point), averaged. so ((0.95*29.76)+(0.05*231.9))/2 = 39.86.

(interestingly, I changed it to use Kelvin instead of °C, and it gave the same result)

Is this method going to work in any capacity, or is the science behind this significantly more challenging than I am anticipating? Any advice on Alloys and manipulating their melting points is greatly appreciated!

 

[sysprog]

Gallium has the interesting property of being solid on the table and liquid in your hand (please don't put it in your bare hand $-$ although the free metal is not dangerous to touch, some gallium compounds are toxic), and its most used alloy partner, in the semiconductor metal gallium arsenide, is the extremely toxic metallic element arsenic. Maybe on the path of trying to produce new gallium alloys, you might look at what's already been done, and investigate why and how gallium has its orthorhombic structural nature.

 

[snorkack]

Summary:: I am interested in creating gallium alloys with the specific purpose of manipulating their melting points. Is there a known process to follow, or is this going to be trial and error?

I am looking to try to create an alloy of gallium (probably with Tin) with a goal of increasing the melting point from 29°C to around 40°C. My current theory is to simply try to average the two melting points out by mass, which my calculations imply means 95% Gallium to 5% tin.

my calculation is (% used x melting point), averaged. so ((0.95*29.76)+(0.05*231.9))/2 = 39.86.

(interestingly, I changed it to use Kelvin instead of °C, and it gave the same result)

Is this method going to work in any capacity, or is the science behind this significantly more challenging than I am anticipating? Any advice on Alloys and manipulating their melting points is greatly appreciated!

It is significantly more challenging. There are rules of thumb, but these are summaries of trial and error. So a logical direction is trial and error... or look up literature, which is other people´s trials and errors written down.
One insight: a molten alloy is a solution. Thus freezing of an alloy is freezing of a solution, and same rules apply.
Another: a lot of substances mix better when molten and poorly or in limited amount when solid. In these cases, adding a solute, though with a higher melting point, does instead lower the melting point.
Melting point can be and is increased by adding a solute when the solute is so well suited that it forms a good solid solution (and is higher melting than solvent). A classical such case is Ag-Au: they form a perfect solid solution, and addition of Au raises melting point of Ag.
But it seems that nothing forms a good solid solution in Ga, and therefore anything will lower the melting point of Ga. Well, the one thing that does dissolve in solid Ga is its own isotopes (it has 2 stable ones - 69 and 71), but isotopic melting point effects are minor for all elements after H.
For a good summary of binary phase diagrams, see:
http://www.himikatus.ru/art/phase-diagr1/diagrams.php
The frame is in Russian and Cyrillic, but elements are alphabetized under Latin symbols and the accompanying texts are English. Alphabetized under the alphabetically first element of the pair, so the elements before Ga are each under its own symbol and those after Ga are all under Ga.

 

[some bloke]

Thankyou for the replies and for the link & rundown of the complexities of alloying!

I had a look through that site and found this graph for Ga: Sn:

Am I correct that the angled line represents the melting point (going entirely off of logic, as it links pure galliums melting point to pure tins meltipng point)? If so, it looks like 18% tin will be about the right combination to achieve a circa 40°C melt temperature. Am I reading this correctly?

 

[TeethWhitener]

The L in the diagram represents the liquid phase. The point at 8.4% Sn is the eutectic point (lowest possible melting point for a Ga:Sn combination). But I’m not seeing why you think 18% Sn will give you a melting point of 40C. It looks closer to 60-70C with my eyes.

 

[snorkack]

Note different scales at bottom and top. Ga and Sn have different atomic masses, and thus atomic and weight percentages differ. The atomic percentage is shown at bottom (and is presented as linear), the weight percentage at top (nonlinear).
But the upper line is the liquidus line. Below that line, to solidus line, two phases - liquid and solid - coexist.
This means that a Ga-Sn alloy, so long as the Ga content exceeds the solid solubility of Ga in Sn (6,4 atom %) will partially melt at 20,5 Celsius, separating into the Ga-Sn melt (initially 8,4 atom % Sn) and impure solid Sn (initially 6,4 atom % Ga). As the melt is warmed in contact with the solid, more Sn dissolves into the melt so the Sn concentration in melt increases. It looks a plausible reading that at 40 C, the saturated solution of Sn in Ga is 18 weight % Sn.

 

[some bloke]

Thanks for the further replies!
I was going by weight, as that is what I will be able to work with easiest for my backyard experiments!

It's hard to gauge because of the irregular tickmarks & the resolution of the image but it looks like 16-18% by weight should get somewhere in the right region - from there I can go for trial & error!

Next step is research on making an alloy - I think it's a case of heating Gallium to over the melting point of tin, mixing the two thoroughly so the tin dissolves, and then allowing it to cool - but I will go for a bit more research before I actually do it!

 

[TeethWhitener]

Next step is research on making an alloy - I think it's a case of heating Gallium to over the melting point of tin, mixing the two thoroughly so the tin dissolves, and then allowing it to cool

Honestly, whenever I’ve made a galinstan alloy, I just melt the gallium and add the other metals. They’ll dissolve relatively easily to give the alloy. In the case of galinstan, this is really easy to see since the melting point is so low. I don’t imagine it’s too different for Ga:Sn.

 

[smplcrtrs]

Next step is research on making an alloy - I think it's a case of heating Gallium to over the melting point of tin, mixing the two thoroughly so the tin dissolves, and then allowing it to cool - but I will go for a bit more research before I actually do it!

Think of it as dissolving a substance (tin) in a solvent (gallium). You don't need to heat the gallium much to make it liquid, and then adding the tin slowly to allow it to dissolve should be fine. An (oversimplified) analogy would be dissolving salt in water - you'd add the solid salt, not molten salt.

Not sure if there's any issue with tin having an oxide layer through passivation - maybe scratch the tin with a sharp knife just before putting into your gallium?

 

[cormsby]

Looking at the graph with my old eyes, 40°C is at 12% atomic ratio Sn
$$\frac{0.12\times 118.7}{0.12\times 118.7+0.88\times 69.72}=0.188$$
...so 18.8% Sn by weight.