Chemical Forums

Chemistry Forums for Students => Undergraduate General Chemistry Forum => Topic started by: labeled on October 08, 2019, 08:18:30 AM

Title: Yttrium-Arsenic-Selenium
Post by: labeled on October 08, 2019, 08:18:30 AM
I would like to make this because it allegedly has a low Seebeck coefficient.
Aside from the solid elements, what is the right ratio to achieve this?
What tools do I need to produce it?  I only know that the Arsenic is ≤1%.

I'd like to test the concept of making a functional Seebeck generator.
The Seebeck difference between Silicon and Bismuth is ~500.
As alternative to Silicon, and I'm trying Copper and Yttrium-Arsenic-Selenium.

At 600*C there is a ~300-350 difference in coefficients and that is satisfactory.
Title: Re: Yttrium-Arsenic-Selenium
Post by: Enthalpy on October 08, 2019, 01:30:16 PM
Welcome, labeled!

Combining a semiconductor with a metal is simpler but less efficient than two semiconductors, because the metal can leak much heat from the heat source to the sink depending on its shape, and its Fermi level is stuck so it contributes little to the voltage. The best Seebeck modules combine two semiconductors, not just to produce the biggest voltage, but also to convert more heat power into electric power. Just in case:

From the report available as Google book
Thermoelectric Properties of Yttrium- Group V and Yttrium- Group VI Binary and Ternary Compounds
By Timothy J. Driscoll, Lindsay D. Norman
I see on page 7
(YAs)·(Y3Se4)x and
where x varies between 1 and 2.25
Very probably, said materials are polycrystalline or amorphous, less difficult to produce. BUT Yttrium melts at 1800K. where As (subl 887K) and Se (bp=958K) have >>1bar vapour pressure, and both are toxic. That's a very serious problem. I suppose a melting pot won't do it.

The report is available in print for 10 bucks
and apparently for free on the Web

The compound is biphasic, As is <1% in one phase but ~25% in the other. Their table 5 gives proportions.
Title: Re: Yttrium-Arsenic-Selenium
Post by: labeled on October 09, 2019, 04:08:16 AM
Thank you for your detailed response; if I have complete control over the shape of the metal down to 10 microns, would it mostly mitigate this problem of heat leaking?

Is there any other semiconductor that I can make that has a ≤300 Seebeck coefficient @600*C with a melting pot that is non-toxic?
Title: Re: Yttrium-Arsenic-Selenium
Post by: Enthalpy on October 09, 2019, 06:49:44 AM
Shape of the metal : sure. Make it thin, like a wire, and long, if necessary as a helix, and then it can match the heat and current conductivity of the semiconductor. A small optimisation tells you how thin. The supporting material, if any, must not itself leak heat.

Other materials: i'm too far outside my knowledge here. Wiki knows it better than I do.

Did you check what commercial components use? Peltier modules are available, the end plates are of alumina, both materials are semiconductor. If you're lucky, you might use one module as is, or maybe reuse the plates and semiconductors, and replace the brazing alloy.