Acetylene solubility in acetone why the 300 volumes figure is a Henry's-law shortcut and what was actually measured

Everyone in the trade knows that acetone dissolves 25 times its own volume of acetylene per atmosphere, and that at 12 atm it therefore holds 300 volumes. The first half is a measurement; the second half is an extrapolation that was never true at ordinary Indian temperatures. The figure read from the measured table at 20 °C and 12 atm total pressure is 247 volumes per volume of acetone, and it falls to about 196 at 30 °C. The 300 is reached only at about 12 °C.

This page sets out what was measured, by whom, and the three physical reasons the shortcut fails: the acetone's own vapour pressure, the swelling of the solvent as it loads, and the departure of the system from Henry's law. It matters because every cylinder-filling calculation, every contents estimate and every "how much gas does my cylinder hold" conversation rests on this number.

Claude and Hess reported in 1897 (Compt. rend. 124, 626) that acetylene was extraordinarily soluble in acetone, and Berthelot and Vieille showed in 1900 that the solution did not decompose under pressures up to 10 atm. Those two findings, with Le Chatelier's porous mass and Janet's combination of the two in 1901, are the dissolved acetylene cylinder; the story is told in why acetylene is dissolved in acetone.

The early solubility determinations were confused, chiefly because authors did not say whether their "pressure" was the total pressure over the solution or the acetylene's partial pressure. The definitive measurements are those of Hölemann and Hasselmann at the Nordrhein-Westfalen research programme between 1952 and 1959, published in the Forschungsberichte des Landes and in Chemie-Ingenieur-Technik. They measured the partial pressure of acetone over the loaded solution first, then the solubility twice, once against partial pressure below 1 atm and once against total pressure above it, with the solvent density and dilation needed to convert between mass and volume. Everything below is from that series.

The volume figures are KIGL's conversion: g/kg divided by the NTP density of acetylene (1.17167 g/l) and multiplied by the acetone density at that temperature (0.7958 g/ml at 15 °C, 0.7899 at 20 °C, 0.7840 at 25 °C, 0.7782 at 30 °C). The C.P.I. handbook of 1968 gives the same picture in its own units: a Bunsen coefficient of 26.6 for acetone at 15 °C and about 1 atm, falling to 20.9 at 25 °C and 14.4 at 40 °C, and an engineering coefficient S′ of 24.75 at 13.78 kg/cm² and 15 °C.

Note the last column of the row that matters. At 20 °C and 12 atm the solvent holds 247 volumes per volume of acetone, not 300. Expressed per litre of charged solution (the figure a filler actually needs, because a cylinder is a fixed volume), it is 151 volumes, because the solution now occupies far more space than the acetone did.

A gauge on a cylinder reads total pressure. Part of that is acetone vapour. At 20 °C and a total of 1 atm, the acetone contributes 0.23 atm and the acetylene is at only 0.77 atm; at 50 °C and 1 atm the acetylene partial pressure is just 0.22 atm. This is why the solubility "at 1 atm" is 36.0 g/kg against partial pressure but 27.9 g/kg against total pressure, a 22 % difference before anything else is considered.

At cylinder pressures the effect shrinks, because the acetone's mole fraction in a heavily loaded solution has fallen: at 20 °C and 15 atm the acetone contributes only 0.18 atm, about 1.2 % of the total. But at the end of a cylinder's life, as the pressure falls towards 1 atm, the acetone share climbs back to 23 % of the withdrawn gas, which is the mechanism behind solvent loss covered in acetone carry-over and residual pressure.

Dissolving acetylene expands the acetone. Each gram dissolved adds about 1.6 ml to the solution volume, slightly more than the 1.26 ml the same mass would occupy as liquid acetylene. At 15 °C the relation is V = V₀ (0.981 + 1.58 s), where s is grams of acetylene per gram of acetone. At the standard filling condition, s = 0.546, the solution occupies 1.84 times the volume of the unloaded acetone. A cylinder charged with 16 litres of acetone contains nearly 30 litres of liquid at 15 atm.

That is why "volumes per volume of acetone" and "volumes per volume of solution" differ by a factor of 1.6 or more, and why the cylinder's porous-mass void, its solvent charge and its maximum filling pressure cannot be set independently. Overcharging solvent, not overpressure, is the classic cause of liquid carry-over at the valve. The internal design is described in inside a dissolved acetylene cylinder.

Below 1 atm acetone obeys Henry's law almost exactly: 0.36 g/kg at 0.01 atm, 3.6 at 0.1 atm, 36.0 at 1 atm, a constant 36 g/kg per atmosphere at 20 °C. Above about 5 atm the solubility coefficient does not stay constant, nor does it fall as a simple saturation model would predict; it rises.

At 30 atm and 20 °C the liquid is 1,297 g of acetylene per 1,000 g of acetone, 56 mol % acetylene; it is no longer a dilute solution but an acetylene–acetone mixture. Hölemann and Hasselmann fitted the 5–30 atm region with M = P(a − b log P), where M is the acetylene mole fraction, P the partial pressure and a and b temperature-dependent constants, accurate to about 1 %. The point for the trade is simpler: no single "volumes per atmosphere" number describes the system between 1 and 15 atm, and multiplying the 1 atm value by the pressure is not a calculation but a guess that happens to land near the truth at 12 °C.

A persistent piece of sales copy says acetylene is "chemically combined" with the acetone. It is not. The acetylene–acetone phase diagram (Baume and Germann, 1911–14) shows putative compounds whose freezing points are so flat that their existence is doubtful and their stability negligible. Infra-red spectra show two C–H stretching bands in mixtures of acetone with inert solvents, one for free acetylene and one for acetylene hydrogen-bonded to the carbonyl oxygen of the acetone (Shuvalova 1959; Iogansen and Kurkchi 1962). The acidic terminal hydrogen of acetylene bonds to electron-donor solvents, and the stronger the donor the greater the solubility: dimethylformamide shifts the C–H band twice as far as acetone and holds more gas, as acetone vs DMF explains. This is physical solution, reversed by lowering the pressure, and it releases heat on filling (684–757 kJ per kg of acetylene dissolved, Schay, Székely, Rácz and Traply 1958) and absorbs it again on withdrawal.

Take an illustrative cylinder, not a KIGL specification, holding 7.0 kg of acetylene at 15 atm and 15 °C. From the measured 546 g/kg it needs 7,000 ÷ 546 = 12.8 kg of acetone, which is the charge the trade in fact uses; the measurement and the practice agree. The same cylinder at 30 °C and the same 15 atm would hold only 364 × 12.8 = 4.7 kg; the rest has left solution and raised the pressure. Two consequences:

• Contents must be checked by weight. Pressure tells you the temperature of a correctly filled cylinder, not how much gas it holds; see checking contents by weight.
• The filling limit is defined at a temperature. IS 14814:2023 cl 7.2.2 sets the maximum working pressure of a cylinder charged with acetylene at 15 °C at 15.495 atm with acetone as solvent, the same 15 kg/cm² settled-pressure rule the C.P.I. recorded in 1968, and IS 308:2024 assumes acetone as the solvent for industrial dissolved acetylene (see the dissolved acetylene standards).