# What are the ideal pair of (everyday) elements to pair in a thermocouple

I want to create a thermocouple (to carry out experiments on thermoelectricity).

I want to use the pair of everyday* elements (e.g $\ce{CuO}$ and $\ce{Al}$ wires) to create the thermocouple. The temperature range of the heat source will be between 300° C - 500° C

My (little remaining) knowledge of physical chemistry informs me that the periodic table would provide useful clues as to which elements to use (that's why I posted this question here, rather than in physics or engineering) - but unfortunately, it's not clear to me how I may use the periodic table in this case, to select the two elements that meet my twin requirements of:

1. Maximum voltage and current generated (via the Seebeck Effect) within the operating temperature indicated above
2. Both elements are "everyday" elements

Note: By everyday elements, I mean elements that I can buy off the shelf (preferably, in wire or sheet form), and are relatively inexpensive.

• Have you checked a thermoelectric series? See if you can find common materials in it. Commented Jun 2, 2017 at 9:57

My book has a few metals/alloys listed in a Seebeck Thermoelectric Series:

Antimony, Nichrome, Iron, Zinc, Copper, Gold, Silver, Lead, Aluminum, Mercury, Platinum-Rhodium, Platinum, Nickel, Constantan, Bismuth

There's also the Seebeck formula for emf of a thermocouple:

$$E_{\text{AB}} = a_{\text{AB}}\theta + \frac{1}{2}b_{\text{AB}}{\theta}^2$$

Here, $a_{\text{AB}}$ and $b_{\text{AB}}$ are constants that depend on the pair of metals $\text{A}$ and $\text{B}$ chosen. $\theta$ is the temperature difference between the junctions of the thermocouple.

There's another defined quantity called thermoelectric Power, which is defined as $\frac{dE_{\text{AB}}}{d\theta}$.

The graph of $E_{\text{AB}}$ with respect to $\theta$ would be a parabola. If $a_{\text{AB}}$ and $b_{\text{AB}}$ are of opposite signs, then the parabola crosses the $x$-axis twice, and thus, the thermo-emf becomes zero for two values of $\theta$ and maximum for one value of $\theta$.

From the above thermoelectric series, a iron-nickel might seem like a good option to consider building a thermocouple with, considering they're quite far away in the series and are easily accessible. For this couple, we get the values of thermoelectric constants as $a_{\ce{Fe-Ni}}=\pu{2.5 \mu V/°C}$ and $b_{\ce{Fe-Ni}}≈0$.Thus, you get:

$$E_{\ce{Fe-Ni}}=2.5\theta$$

If you're using your temperature range of 300°C-500°C for the hot junctions and room temperature (25°C) for the cold junction, you would get your thermo-emf in the range $\pu{0.687-1.1875mV}$.

If you don't want to reinvent the wheel, use iron / constantan. Good for those temperatures and voltage/temperature tables are readily available.

• Completely agree. I primarily used Fe - Cu/Ni for casual stuff ($\pu{−180 - +800 ^\circ C}$), and Pt/Rh - Pt for higher temperatures (up to $\pu{1550 ^\circ C}$). These two were enough practically for all applications. Commented Jul 5, 2017 at 14:28