On this page
Why can a cylinder of carbon dioxide hold liquid on a cool day, while oxygen in a cylinder stays a gas however hard it is pressed? The answer is a single temperature for each substance, first measured carefully by Thomas Andrews in Belfast. This post tells his story, explains what he saw in his glass tubes, and works through his own numbers.
Who was Thomas Andrews?
The memoir printed in The Scientific Papers of the late Thomas Andrews (1889) records that he was born at 3 Donegall Square, Belfast, on 19 December 1813, the eldest son of a linen merchant, and died on 26 November 1885. He trained as a physician and practised in Belfast while teaching chemistry. The Royal Society gave him a Royal Medal in 1844 for work on the heat released in chemical reactions. In 1845 he became Vice-President of the new Queen's College, Belfast, and its professor of chemistry, a post he held until 1879.
What did people know about liquefying gases?
Andrews opened his 1869 lecture with the history. In 1822 Cagniard de la Tour saw liquids such as ether vanish into vapour when heated in sealed tubes. In 1823 Michael Faraday liquefied chlorine and other gases by pressure alone. But Andrews had found earlier that oxygen, hydrogen, nitrogen and other gases did not liquefy even when cold and pressure reduced them to a small fraction of their normal volume. Why would some gases give in and others refuse?
What did Andrews see in his tubes?
He sealed very dry, pure carbon dioxide in thick glass tubes with a fine capillary bore, and pressed it through a column of mercury, with the pressure raised by turning a steel screw into a water-filled copper tube. A first account, from a letter of 1863 quoted in the lecture, describes what happened when he partly liquefied the gas and slowly warmed it. The boundary between liquid and gas grew fainter, lost its curve and disappeared. Above that temperature he could not produce a liquid even at 300 or 400 atmospheres.
In the full paper, On the continuity of the gaseous and liquid states of matter, published in the Philosophical Transactions in 1869, he fixed this critical point for carbon dioxide in three trials at 30.92 °C, or 87.7 °F. Above it, he wrote, the most careful examination found no separation into two states. Nitrous oxide, hydrochloric acid, ammonia, ether and carbon disulphide behaved the same way, each with its own critical point.
How can a gas become a liquid without boiling?
This was his deepest point. Andrews described compressing carbon dioxide at a temperature above 31 °C, then cooling it while keeping the pressure above 76 atmospheres. The gas ends as a liquid, but at no moment is there a surface between two states or a sudden change of volume. He said one would never suspect it had become liquid, except that it boiled when the pressure was released. Gas and liquid, he concluded, are "only distant stages of the same condition of matter", able to pass into one another by continuous change.
Worked example 1: how far from ideal is carbon dioxide?
Andrews reported that at 13.1 °C and about 49 atmospheres, carbon dioxide occupies little more than three-fifths of the volume a perfect gas would.
Ideal volume for 1 mol. V = RT/P = 0.0821 × 286.25 / 49 ≈ 0.479 L.
Real volume. A little more than 3/5 of this: about 0.6 × 0.479 ≈ 0.29 L.
Compressibility factor. Z = PV/nRT = V_real / V_ideal ≈ 0.6.
A value of Z well below 1 means attractions between molecules are pulling the gas together, the same attraction that, below 30.92 °C, finally makes it collapse into liquid. Our guide to real gases and the van der Waals equation shows how the constants a and b describe this.
Worked example 2: gas or vapour?
Use 30.92 °C as the critical temperature of carbon dioxide.
A cylinder at 25 °C. Below 30.92 °C, so the carbon dioxide is a vapour. Enough pressure gives liquid and gas together, with a visible surface.
The same cylinder at 35 °C. Above 30.92 °C. No pressure gives a separate liquid; the contents are one uniform fluid.
Units check. 30.92 × 9/5 + 32 = 87.66 °F, matching Andrews' 87.7 °F.
Andrews' own comparison. He called ether a vapour below 200 °C and a gas above it. By his rule, what decides "gas" or "vapour" is the critical point, not whether the boiling point is above or below room temperature.
Common mistakes with critical temperature
Thinking enough pressure liquefies anything. Above the critical temperature, it does not.
Mixing critical temperature and boiling point. Boiling point depends on pressure; the critical temperature is one fixed value.
Treating Z below 1 as an error. It means attractions dominate; Z above 1 means molecular size dominates.
Forgetting kelvin. Use T in kelvin in PV = nRT: 13.1 °C is 286.25 K.
Real gases and van der Waals' equation are in the chemistry section of the JEE (Advanced) 2026 brochure, as "deviation from ideality, van der Waals equation". The current NCERT Class 11 Chemistry book no longer has a states-of-matter unit; its Unit 5 is now thermodynamics. Check the current year's documents. If you want to see how isotherms change shape near the critical point, the Fermi Simulation Lab has interactive simulations for ideas that are easier to see than read.
What can a student take from how Andrews worked?
Purify before you measure. He spent hours drying and purifying the gas and still allowed for the tiny trace of air left.
Question the categories. He asked whether "gas" and "vapour" were really different, and found they were not.
Write the definition the facts support. His vapour rule came directly from what he saw.
Practice routine: find the ideal volume of 1 mol of gas at 21.5 °C and 60 atm, decide whether carbon dioxide at 28 °C and at 40 °C can be liquefied by pressure, and draw a pressure-volume path that turns gas into liquid without crossing a boundary.
Next step: for any gas question with "liquefy" in it, first write down the critical temperature and compare it with the given temperature.
