Temperature derived from current, then plotted against V?

[evanghellidis]

I'm not sure if I'm posting in the right forum, please excuse if misposting.

I have the following situation: pieces of conductive nanostructured material are attached to different substrates and heated resistively. We can only measure voltage and current through the piece of material (and respective contacts). The temperature of the nanostructured layer is derived using the temperature coefficient of resistivity, assuming linearity and that the piece of material behaves as a classical conductor.

The same type of temperature derivation is then used to plot temperature against "on-off" cycles, where several voltage pulses are applied to the material. That is, 2 volts are applied for a few minutes, then a lower voltage for another few minutes. The current variation is used to calculate resistance and then temperature. The plot shows the square voltage pulses in time, overlapped with the corresponding, derived temperature pulses. This is done for each substrate.

The cycling plots are meant to show that the material can heat up repeatedly and consistently, and to discuss differences in its heating speed given by the different substrates. My question: is this is a valid discussion, given that temperature is not measured directly?

 

[siddharth23]

If I understood correctly, you're measuring temperature by measuring the change in resistance, right?

I think its ok to do so. I don't know how accurate the readings will be (probably depends on the setup and the least count of your instruments). But the method seems alright to me.

 

[Stephen Tashi]

My question: is this is a valid discussion, given that temperature is not measured directly?

Can you check it by heating a sample to a known temperature and the taking a quick voltage reading across it?

 

[evanghellidis]

I could use a heating plate, yes. It'll still be imprecise, but at least it gives temperature data points which are experimentally independent of the electrical data points. That's a good idea, thank you!

 

[evanghellidis]

If I understood correctly, you're measuring temperature by measuring the change in resistance, right?

I think its ok to do so. I don't know how accurate the readings will be (probably depends on the setup and the least count of your instruments). But the method seems alright to me.

Yes, temperature measurement through change in resistance. The electrical measurements are fairly accurate, it just seems to me that there's something wrong about using just one type of experimental data in an experiment which is meant to show the relation between two different physical phenomena (i.e. electrical conduction and heating). More so when you're already making very broad approximations regarding the sample.

This situation has occurred in the lab I work. I would've preferred using a temperature probe but the PI said there's no need. I hope the reviewers will see things the same way.