ρ = pM/(RT): density is proportional to pressure and molar mass, inversely to absolute temperature. Right to a fraction of a per cent for air at everyday conditions.
A compressibility factor Z = 1 + Bp/(RT) from each gas’s second virial coefficient (Pitzer–Abbott). With it every density matches published values at 0°C to within 0.05%; without it CO₂ is 0.7% light and propane 2.2%.
Lee–Kesler vapour pressure from the same three constants, and Buck’s formula for water. Humid air treats dry air and vapour each as an ideal gas at its own partial pressure.
Speed of sound is √(γRT/M); viscosity follows Sutherland’s law with the constants published by White, only for the gases that table covers.
Density is proportional to pressure, and pressure falls about 11% in the first kilometre you climb. A cold sea-level day is about 15% denser than a hot one; Denver is about 15% thinner than sea level in the standard atmosphere, and a jet at 11 km flies through air less than a third as dense as at the runway.
Engines, propellers and wings all depend on the mass of air they move, so they lose power and lift as density falls. That is why aircraft performance charts use density altitude, and why a hot afternoon at a mountain airfield can leave a light aircraft unable to climb out.
At the same temperature and pressure, a cubic metre of any gas holds the same number of molecules (Avogadro’s law), so density is proportional to molar mass. Carbon dioxide is half as heavy again as air and pools in low places; methane is just over half and rises.
Propane in a cylinder is a liquid with gas above it, because at room temperature it condenses at about 8 bar; its pressure tells you the temperature, not how full it is. Carbon dioxide does the same at 57 bar. Nitrogen and oxygen cannot be liquefied at room temperature at any pressure, which is why their cylinders are pure gas at 200 bar.
It is not a full equation of state. Past about ten bar, or close to the point where a gas condenses, it is no longer enough, and the page warns when you are there. Pressure stops at 1,000 kPa for that reason: a cylinder at 200 bar needs a real equation of state.
For air, nitrogen and the noble gases that is excellent; for carbon dioxide, propane and sulfur hexafluoride it falls as they warm, so their speed of sound is a few per cent out at the hot end of the range. Helium and hydrogen are quantum gases the correlation fits poorly, but their correction moves density by less than 0.05%.
Humid air uses Buck’s saturation pressure over liquid water and ignores the small enhancement factor, a 0.4% effect on the vapour pressure that changes the density by far less. Air is treated as a fixed mixture; real air carries 0.04% carbon dioxide and a little argon variation, both below anything shown here.
About a third of a per cent for every degree near room temperature, because at fixed pressure density goes as 1/T in kelvin, and 1/293 is 0.34%. Across the year that adds up: air at −5 °C is about 15% denser than air at 35 °C. Pressure matters just as directly — density is proportional to it — which is why altitude, not the weather, makes the biggest difference.
Because a water molecule (18 g/mol) is lighter than the nitrogen (28) and oxygen (32) it replaces. At the same pressure and temperature a fixed volume holds the same number of molecules, so swapping some for water vapour lowers the mass. Saturated air at 30 °C is about 1.6% lighter than dry air, which is part of why humid air rises and why aircraft and engines lose a little performance in it.
The International Standard Atmosphere defines air as an ideal gas, and 1.2250 kg/m³ is what the ideal gas law gives at 15 °C and 101.325 kPa. Real air is very slightly more compressible than an ideal gas at that temperature, so its true density is 0.04% higher. For air the difference never matters; for carbon dioxide it is 0.7%, which is why the correction is here.
Because the gas would no longer be a gas. Below a certain temperature at a given pressure — propane below about −42 °C at one atmosphere, steam below 100 °C — it condenses, and a gas-density formula says nothing about a liquid. The map shows that region shaded and the line where it begins. Helium, hydrogen, nitrogen, oxygen, argon, methane and air cannot condense anywhere in this range of temperatures at any pressure here.
The opposite of a liquid, and for a different reason. In a gas, viscosity is momentum carried between layers by molecules flying across them; hotter molecules move faster and carry more, so it rises, roughly as T0.7. In a liquid it is molecules having to break free of their neighbours, which gets easier as they warm. The kinematic viscosity of a gas rises much faster still, because its density is falling at the same time.
This page has changed in 63 archived releases, but each of those was a site-wide pass, so none is attributable to this tool on its own and none is listed here. That is not a claim that the tool never changed — a release that reworked many pages at once may well have altered this one too. The changelog has them.