Why the alcohol thermometer is less precise than the mercury one?

[ORF]

Hi,

Why the alcohol thermometer is less precise than the mercury one?

If I remember correctly, it is because the thermal expansion coefficient of alcohol is not constant, but I am not sure.

Thank you for your time.

Regards,
ORF

 

[kuruman]

I think it's because alcohol wets the walls of the thermometer making the reading less accurate.

 

[gmax137]

My first thought was coefficient of expansion; maybe mercury is higher so a bigger change in volume per change in temperature. But this is wrong. A little internet searching shows volumetric coefficients for "ethyl alcohol" as 1100 (x10-6 per degree C), compared to 180 for mercury. So the alcohol expands ~6 times as much.

 

[phinds]

Why the alcohol thermometer is less precise than the mercury one?

Do you know for sure that it is? I thought the reason mercury is used is because it allows for reasonable sized thermometers whereas alcohol ones would be awkwardly large to get an error range similar to a mercury one.

 

[DaveE]

I think it's because alcohol wets the walls of the thermometer making the reading less accurate.

Do you mean that the meniscus is variable, or that it's more prone to humans misreading it consistently?

 

[kuruman]

Mostly the meniscus.

 

[DaveE]

Since both have to be read by a human (I assume), then I'm not sure your accuracy question is complete. You are not addressing the biggest error source, IMO. This is one of many reasons everyone uses electronics to collect data these days.

Anyway, why would there not be a deterministic equation that relates temperature to liquid volume in a sealed glass tube? Granted, it might not be linear, but does that matter?

A question for those who know materials science better than I (a pretty low bar, actually). What can you do to change the volume of a liquid at a fixed thermal equilibrium and constant pressure (because the glass is thick)? Those would be the error sources, right?