Alcoholic strength
Unit outside the SI · quantity: the volume fraction of ethanol in a water-alcohol solution
In the English Exchequer of the seventeenth century strength was tested with gunpowder: a spoonful of spirit was poured over a pinch of powder and set alight. If the powder flashed once the flame had died away, the spirit was held to be proved; if the water left behind drowned the powder and it merely hissed, the goods fell short of the standard and were charged at a different rate. The test looks crude, yet it rested on real physics: ethanol burns for as long as there is not too much water in the solution, and that threshold falls at roughly fifty-seven per cent by volume.
Out of that gunpowder threshold the modern scales grew, drifting apart from one another over the better part of a century. Britain fixed «a hundred degrees proof» at precisely the strength at which the powder still caught, so that the unit of her scale came out at seven quarters of the volume per cent; America, unwilling to bother with fractions, declared proof to be simply twice the per cent, and a hundred there became half a bottle of pure spirit. Gay-Lussac went the simplest way of all and called the volume per cent itself a degree, and for that his scale outlived both the others and became the basis of the present rules.
01 · Definition
Strength in per cent by volume is the ratio of the volume of pure ethanol to the volume of the whole solution, both volumes being taken separately at twenty degrees Celsius. The proviso about separately is essential: spirit and water contract on mixing, and a litre of spirit with a litre of water yields about 1.93 litres of mixture rather than two, so the fractions are reckoned before mixing and not after.
Any instrument will serve for such a quantity, provided it can tell densities apart: spirit is lighter than water by nearly a fifth, so a float sits noticeably deeper in vodka than in wine. Hence both the design of the alcoholmeter, graduated directly in per cent, and the obligatory temperature correction, since a warm solution seems to the instrument stronger than it is.
The three historical scales differ from one another by a factor alone, and so convert exactly and without any tables. Gay-Lussac set his degree equal to the volume per cent, the American proof is twice the per cent, and the British took the gunpowder threshold for its hundred — the strength at which the solution still ignites powder — and so obtained a unit of seven quarters of a per cent. Britain abolished her scale in 1980, but the inscription «70° proof» on bottles of an old bottling still calls for conversion.
Mass fraction lives alongside and is confused with the volume fraction constantly, though the two differ substantially: in forty-per-cent vodka the spirit by mass amounts to only about thirty-three per cent, because it is lighter than water. Duty, incidentally, is reckoned from neither, but from the litres of pure alcohol in a consignment, so that for the bookkeeping strength matters only as a factor.
Drive the strength with the slider and watch the spoon: while there is spirit enough, the flame burns down to dry powder and the powder flashes; the moment water passes the threshold, what is left drowns the powder and instead of a flash there comes a hiss. The three scales on the right read one and the same liquid, and they diverge by exactly as much as their factors differ.
The solid line shows how the density of a water-alcohol mixture falls as strength rises, and the dotted one how it would fall if the liquids simply added up. The lines part because molecules of spirit and water pack more closely than either does on its own, and that dip is deepest around forty per cent. It is precisely on its account that the alcoholmeter has to be graduated from tables rather than with a ruler.
02 · Conversion
Enter what is written on the label
Between the three scales of strength the conversions are exact, since all of them differ by a factor alone. Approximation appears only where density enters into it: mass fraction and the weight of a consignment are reckoned from tabulated values for water-alcohol mixtures.
The British scale is shown in two forms: in ordinary degrees proof and in the old Sikes notation, where the departure from a hundred was counted upward or downward. The second form can be hidden in the fine settings of the sheet.
Temperature matters for strength no less than for viscosity, though it looks different: the fraction of spirit itself does not change with heating, but the density falls, and the alcoholmeter begins to overstate its reading by about a third of a per cent for every degree. The verification tables are therefore drawn up for twenty degrees Celsius, and in American practice for sixty Fahrenheit.
| Value | Note | ||
|---|---|---|---|
| {u} | {name} | {value} | {note} |
03 · Orders of magnitude
from alcohol-free to rectified spirit04 · Measuring instruments
What strength is measured with
A spoonful of spirit was poured over a pinch of powder and set alight, and by whether the powder flashed once the flame had died or hissed in the water left behind, the strength was judged. The test told apart only two states, yet it required neither instrument nor tables, and for that it lasted in the Exchequer for nearly a century and a half.
A brass float with a set of hanging weights, given legal force by the British Parliament in 1816: the weight was chosen so that the instrument sank into the graduated region, and the strength was read as a departure from a hundred, upward or downward. It was this that turned the powder's yes-or-no into a number that could be entered on an invoice.
A glass float graduated directly in volume per cent from the tables of water-alcohol mixtures — the very instrument still in use both in the laboratory and in the home distiller's kitchen. Its reading holds only for a pure mixture of spirit and water, and so sugar in a cordial or extract in whisky throws it off entirely.
The only way to learn the strength of a drink that holds something besides spirit and water: the sample is separated on a column and the area under the ethanol peak is measured. Liqueurs, infusions and everything sweet are checked this way, where a float is useless, and methanol is caught at the same time, which an honest product should not contain.
05 · Writing rules
Proof without a country means nothing
The volume fraction is written as a number, a space and the per cent sign, with the qualifier «vol» placed after it, since without it the fraction reads two ways. For proof the country must always be named: one and the same figure means a different strength on either side of the Atlantic, and the difference is no trifle — a hundred British degrees give fifty-seven per cent, a hundred American exactly fifty.
The first notation is the colloquial «forty degrees», which has no place in a document, since strength was called a degree only by Gay-Lussac and Tralles, and their symbol is its own. The second does not say whether the fraction is by volume or by mass, and for forty per cent the difference between them runs to a full seven points. The third names no country and therefore cannot be converted unambiguously. The fourth is simply wrong: by mass the same vodka holds about thirty-three per cent of spirit.
06 · Neighbouring units
Alongside stand all the scales by which concentration is read through the density of a solution: sugar in wort is measured in degrees Brix, extract in brewing in degrees Plato, and acids and alkalis in the works in degrees Baumé. They are built alike: a float sees the density, and the glass carries the figures of whichever trade the instrument was made for.
in the volume of solution
per cent by mass
density of the original wort
Of the Simetrium data sheets nearby stand degree Brix and degree Plato, the hydrometer scales °Bé / °Tw / °API, where the same density is read the oilman's way, and density in kg/m³, without which volume fraction cannot be tied to mass fraction.
07 · Historical section
Proof by gunpowder
The English excise service levied duty on spirit at two rates, and the goods had to be told apart right there in the port, without any instruments. The test they devised was simple: spirit was poured over powder and set alight, and by whether the powder flashed or hissed the goods were counted as proved or as weak. The word proof has meant strength ever since, though literally it is «a proving».
Parliament approved the instrument of Bartholomew Sikes and with it gave legal force to the definition itself: proof spirit is a solution whose density amounts to twelve thirteenths of the density of water. The figure came out at about fifty-seven per cent by volume, that is, the same gunpowder threshold, only measured with a float instead of a match.
The American Treasury inherited the word proof but would not trouble itself with British fractions and declared that a hundred proof is exactly half spirit by volume. The link with gunpowder was broken there, but the scale became one that could be reckoned in the head, and to this day «100 proof» on a bottle of bourbon means fifty per cent and not fifty-seven.
When trade within the Common Market called for one scale for all, the simplest was chosen: the volume per cent after Gay-Lussac. Britain abolished her scale on the first of January 1980, and since then bottles have been marked alike from Lisbon to Helsinki, while the old «70° proof» is met only on labels that have lain in a cellar for more than forty years.
A test able to answer only yes or no
The gunpowder trial looks like a curiosity, yet in substance it is the most honest instrument of all described here: it never pretended to measure a quantity. It answered a single question — is the strength above the threshold or below it — and it answered that question reproducibly, which is more than can be said of many a graduated scale. The whole later history of the unit is the history of turning that yes-or-no into a number.
Replacing fire with a hydrometer was not merely a refinement but a change in the very nature of the quantity: in place of a threshold there appeared a scale, and in place of two states a continuum. From that moment strength could be entered on an invoice, taxed by degrees and disputed in court over tenths — none of which the powder allowed. The price of it was the tables: a float sees density, not spirit, and only a table turns the one into the other.
Catalogue · units of measurement
A data sheet for every quantity
Seven SI base units, twenty-two derived ones with names of their own, and the non-SI quantities that neither science nor daily life does without. Each gets its own sheet: definition, conversion, instruments, writing rules, history. In 36 languages.