# Thermodynamics of thermometer

Mercury is used in thermometers because it increases in length significantly due to rise in temperature, However, mercury has high density relative to water, which means more inter-molecular forces which makes it harder to expand than water. Then,why is mercury preferred to water in a thermometer?

• Your claims are incorrect, as R Millikan points out. – Carl Witthoft Jan 8 '14 at 14:30

The density of mercury (13.534 g/cm^3) does not imply high intermolecular forces. It simply reflects that the mercury atom is much more massive than a water molecule. The atomic weight of mercury is 200.6, while the molecular weight of water is about 18, so mercury atoms take up $\frac {200.6} {18\cdot 13.534}=0.823$ as much volume as a water molecule. This doesn't say anything about the forces between the atoms.

Probably the worst thing about water for a thermometer (assuming you keep it from freezing) is that the expansion is not linear. Water hits a maximum density at 4C, so contracts as it drops in temperature from 0 to 4. Above 4C, it expands quite slowly for a while, then more quickly as it heats up. A plot of the volume of water as a function of temperature is below. It makes thermometers much easier if the curve is linear, which this is not. Over a small range, it is close to linear. It would work well for medical thermometers with a range of $35-42 C$, say

I don't design thermometers and don't know exactly what the engineering tradeoffs are, but here are some advantages of mercury over water that pop to mind:

1. Mercury doesn't turn to a solid and expand at 0°C.

2. Mercury is not a clear liquid, which makes seeing it in a clear glass tube a lot easier.

3. Mercury has much lower vapor pressure than water, so won't evaporate and condense at other parts of the tube.

4. Mercury is a lot more linear in expansion as a function of temperature than water. Water isn't even monotonic a few degrees above freezing. Non-monotonicity is a obvious show stopper for a thermometer material.