Thiosulphuric Acid, H₂S₂O₃

Sodium thiosulphate was obtained probably for the first time by Chaussier in 1799 when endeavouring to prepare sodium carbonate by heating sodium sulphate with charcoal. The name hyposulphurous acid for the corresponding acid was suggested by Gay-Lussac in 1813, who wrongly regarded the acid as representing an intermediate stage of oxidation between sulphur and sulphurous acid. In 1877 Wagner recommended the more correct description "thiosulphuric acid," which has almost entirely displaced the earlier name from chemical literature, although "hyposulphite" is still used by photographers.

The free acid is exceedingly unstable and, at best, is obtainable only in very dilute solution; many of the older methods described for the synthetic production of aqueous solutions of the acid probably yielded only one or more of the polythionic acids.

Under suitable conditions, however, sulphurous acid appears to be capable of combining directly with sulphur, giving some thiosulphuric acid, the conditions being most favourable in alcohol solution at the ordinary temperature.

A more satisfactory method is to pass dry hydrogen sulphide into alcohol containing lead thiosulphate in suspension; after filtering off the lead sulphide formed, excess of hydrogen sulphide is removed from the solution by a current of air. The solution obtained decomposes in the course of a few days.

The acid appears to be formed when a few drops of a concentrated solution of sodium thiosulphate are added to a few cubic centimetres of fuming hydrochloric acid. Sodium chloride is precipitated and a clear solution obtained which remains stable for about an hour.

The salts are much more stable and can be prepared by various processes; the sodium salt is the one most commonly manufactured.

Sulphites of the alkali metals when heated in aqueous solution with sulphur or polysulphides are converted into thiosulphates; the sulphur should be in excess and the mixture kept well stirred. A similar conversion can be effected by careful heating in the absence of a solvent,

Na₂SO₃ + S = Na₂S₂O₃.

A similar result is achieved by treating an alkali sulphite in aqueous solution with hydrogen sulphide, the formation of sulphur probably occurring as the first stage.

The interaction of aqueous alkalis and sulphur, and of fused alkalis or alkali carbonates with the same element, also produces some thiosulphate.

Alkali sulphides and polysulphides on suitable oxidation give rise to thiosulphates; the formation of thiosulphate by the action of sulphur dioxide on an alkali sulphide may be considered as a special case of this class or a modification of the method given first above. If aqueous solutions of sodium hydrogen sulphide and sodium hydrogen sulphite in the molecular proportion 1:2 are mixed, sodium thiosulphate is obtained in a high degree of purity:

2NaHS + 4NaHSO₃ = 3Na₂S₂O₃ + 3H₂O.

Also, in the preparation of the thiosulphate from sodium sulphide and sulphur dioxide, or from sodium sulphite and hydrogen sulphide, as already described, if sodium hydroxide is first added to the solution in such quantity as to lead to the ultimate formation of the hydrogen sulphide and hydrogen sulphite salts in the proportions mentioned, then the reaction proceeds very smoothly and almost without separation of sulphur.

The decomposition of the polythionic acids (q.v.) in the presence of alkali, the decomposition of hydrosulphites and the hydrolysis of nitrogen sulphide by water or aqueous alkali are also processes in which salts of thiosulphuric acid are formed. Certain micro-organisms capable of converting sulphur to thiosulphate have been isolated from soil cultures.

Of especial interest from the point of view of the molecular constitution of the thiosulphates, is the possibility of preparing the alkali salts from the corresponding sulphites and sulphides by the action of iodine:

Na₂SO₃ + Na₂S + I₂ = 2NaI + NaHSO₃,

or by electrolysis, in which case thiosulphate is produced at the anode:

SO₃'' + S'' + 2⁺ = S₂O₃''.

For details of the commercial processes for the manufacture of thiosulphates, the volumes of this series dealing with the particular salts in question should be consulted.

As is to be expected from its unstable nature, thiosulphuric acid is endothermic when referred to aqueous sulphur dioxide and free sulphur or even to sulphur dioxide, water and sulphur, although, on account of the high value of the heat of formation of water, the heat of formation of the acid from its elements is a positive quantity, amounting to 137.83 Calories per gram-molecule in aqueous solution.

When solutions of the salts are acidified there is a gradual deposition of sulphur, the rate being dependent on the concentration of the solutions as well as on other factors. No turbidity is observable at first, and the interval is sometimes ascribed to the sulphur remaining in colloidal solution for a time, subsequent neutralisation not preventing the later separation of at least some of the sulphur; also, the presence of colloidal particles may be detected by the ultra-microscope. With solutions of one gram-molecule of sodium thiosulphate and one gram-molecule of hydrochloric acid, each in 16 litres of water at 12° C., a mixture of equal volumes gives no turbidity until after one and a half minutes.

The foregoing view of the cause of the delayed turbidity is correct only in part, considerable evidence pointing to at least a definite although short existence of thiosulphuric acid itself; for instance, the decomposition of thiosulphuric acid is a slightly reversible process; also, sodium thiosulphate can be titrated satisfactorily with iodine in dilute acid solution, and an acidified solution of a thiosulphate will reduce Methylene Blue in aqueous alcohol solution, whereas sulphurous acid will not. During the decomposition of the thiosulphuric acid, however, secondary reactions occur, resulting in the formation of polythionic acids, and considerable investigation has been directed towards this subject in recent years, the main conclusions from which will now be given.

Aqueous solutions of thiosulphuric acid and its salts are not very stable, but tend to decompose, yielding sulphurous and polythionic acids and a deposit of sulphur. The decomposition may take place in at least three different ways:

1. H₂S₂O₃ ⇔ H₂SO₃ + S,
2. 2H₂S₂O₃ ⇔ H₂S + H₂S₃O₆,
3. 2H₂S₂O₃ ⇔ H₂O + H₂S₄O₅.

Of these, the first reaction is the one usually recorded, and in the majority of cases it is the one that occurs to the largest extent. The presence of alkali pushes the equilibrium well to the left, so that in alkaline solution the thiosulphate is stable. This explains the formation of thiosulphates on boiling alkaline sulphite solutions with sulphur. The laws of chemical equilibria, however, demand the presence of perfectly definite although perhaps very small quantities of sulphite and free sulphur in solution, and if alkaline thiosulphate solutions containing alkali sulphide are boiled in the absence of air, they become deep yellow, owing to polysulphide formation, the extra sulphur for which is obtained from the thiosulphate. Assuming the sodium derivatives to be used, the equation may be written

Na₂S + Na₂S₂O₃ ⇔ Na₂S₂ + Na₂SO₃,

and higher polysulphides may result as well.

This equilibrium affords an explanation for the fact that a trace of hydrogen sulphide accelerates the reaction between sodium sulphite and sulphur; sometimes there is a long delay before sulphur begins to dissolve in boiling sodium sulphite solution, and in such cases the effect of passing a few bubbles of hydrogen sulphide into the solution is very marked; when once the reaction begins it proceeds smoothly.

Although the reaction (a) is usually written in the manner indicated it appears to be bimoleeular, and may be more correctly represented as

2H₂S₂O₃ ⇔ 2H₂SO₃ + S₂.

This seems to suggest that the sulphur unit is S₂. This unit is capable of uniting with hydrogen sulphide to form the trisulphide, H₂S₃; but sodium sulphite can only combine with one atom of sulphur to yield thiosulphate. The catalytic activity of the hydrogen sulphide would thus appear to be due to its ability to absorb a whole sulphur unit, S₂, and subsequently to give up, on reduction, each atom of sulphur separately, thus:

H₂S + S₂ ⇔ H₂S₃,
H₂S₃ + Na₂SO₃ ⇔ Na₂S₂O₃ + H₂S₂,
H₂S₂ + Na₂SO₃ ⇔ Na₂S₂O₃ + H₂S.

The manner in which the S₂ unit is eliminated from thiosulphuric acid remains to be considered. Bassett and Durrant point out that when the known weakness of the second stage ionisation of sulphurous acid is considered in conjunction with the known tendency for sulphur to become co-ordinated with four atoms or groups, it would appear that the direct loss of sulphur by thiosulphuric acid is largely due to a hydrogen atom taking the place of the escaping sulphur atom, thus:



the sulphurous acid produced having, in the first instance at any rate, the sulphonic structure.

From spectrophotometric measurements, Jablcynski and Rytel, however, maintain that the decomposition is unimolecular, monatomic sulphur being first produced, which itself retards the reaction. As the opalescence is due to the formation of polyatomic sulphur aggregates, the latter act as an autocatalyst by withdrawing the atomic sulphur from the solution.

Turning now to equation (b), namely

2H₂S₂O₃ ⇔ H₂S + H₂S₃O₆,

it has been known for many years that, upon acidification, solutions of thiosulphates invariably evolve hydrogen sulphide. Water alone suffices to cause evolution of the gas from sodium thiosulphate. The gas is evolved also if carbon dioxide is bubbled through the solutions, or if the thiosulphate is warmed with boric acid, whilst stronger acids liberate it readily. Various explanations have been offered for this, some attributing the presence of the gas in standard sodium thiosulphate solution to bacterial action, whilst Foerster and his co-workers suggest hydrolysis, according to the equation

H₂S₂O₃ + H₂O ⇔ H₂SO₄ + H₂S.

Against this latter view is the fact that little or no sulphuric acid is formed unless the mixture is boiled for a long time. In alkaline solution, alkali sulphide and trithionate react to form thiosulphate,but alkaline solutions of sulphate and sulphide do not. It would appear, therefore, that the correct explanation lies in the reversible equation (b). This receives support from the fact that when lead thiosulphate is boiled with water it yields, in the first instance, lead sulphide and lead trithionate. Conversely, lead sulphide on digestion with potassium trithionate yields lead thiosulphate. Further, weakly alkaline solutions of sodium thiosulphate itself yield, on boiling, sodium sulphide and sodium trithionate, with only a trace of sulphate; if boiled with sodium plumbite, lead sulphide is precipitated and sodium trithionate remains in solution.

When acids act on thiosulphates, polythionic acids are formed. This is explained by reactions (b) and (c). The fact that the amount of hydrogen sulphide liberated is very small in proportion to the amount of polythionic acid formed is attributed to the fact that reaction (a) is that which normally occurs to the greatest extent when acid acts upon thiosulphate; the hydrogen sulphide is thus liberated in the presence of a large amount of sulphurous acid and hence rapidly destroyed. Trithionic acid is thus a primary product of thiosulphate decomposition, and in its turn decomposes.

In very acid solution, Bassett and Durrant believe that reaction (c) takes place, with formation of a new acid, di-thio-pyrosulphuric acid, H₂S₄O₅:

2H₂S₂O₃ ⇔ H₂O + H₂S₄O₅.

This reaction is clearly effective in removing hydrogen ions from solution. The acid is presumed to be formed by the removal of a molecule of water from two hydroxyl groups of two molecules of thiosulphuric acid-not from one hydroxyl and one thiol group-in which case its structure may be written as:

(i) or (ii)

From the second formula it is clear that three tautomeric structures are possible, according as both hydrogen atoms are ionisable (as in the formula), or one or both atoms are absorbed into the complex, being attached by co-valencies to either a sulphur or an oxygen atom.

When ice-cold concentrated hydrochloric acid is added to a thiosulphate solution, a colourless solution is obtained similar to that resulting from the action of sulphur dioxide, no sulphur being precipitated. After fifteen hours no thiosulphate can be detected in the solution.

Although the foregoing explanation may be valid, it has been suggested that the compound formed in solution is of the type H₂S₂O₃.SO₂.

Sulphurous acid causes the decomposition of thiosulphuric acid to take place more in accordance with equation (b); this it does in three ways, namely:

i) By retarding reaction (a), of which it is a product:

(a) 2H₂S₂O₃ ⇔ 2H₂SO₃ + S₂;

ii) by forming a relatively stable additive product, H₂S₂O₃.SO₂;

iii) by accelerating reaction (b):

b) 2H₂S₂O₃ ⇔ H₂S + H₂S₃O₆,

through removal of hydrogen sulphide, with which it interacts, liberating sulphur, which may, under favourable conditions, react with more sulphurous acid to regenerate thiosulphuric acid. It is thus possible to convert a thiosulphate almost quantitatively into trithionate, as, for example, by acting upon a saturated solution of the potassium salt with concentrated sulphurous acid at 30° C., thus:

K₂S₂O₃ + 4SO₂ + H₂O = K₂S₃O₆ + H₂S₃O₆.

The reaction, of course, will not end there; continued action of sulphurous acid results in the hydrolysis of the trithionate, with formation of tetra- and pentathionates, and ultimately of sulphate and sulphur.

If gaseous sulphur dioxide is passed into a solution of thiosulphate, a yellow solution is formed which on keeping becomes colourless; it then yields a precipitate of sulphur when treated with formaldehyde and sodium hydroxide, but no polythionate can be detected. When the colourless solution is neutralised with sodium hydroxide, it is found to contain sulphite and thiosulphate, but sulphur is not precipitated. These results may be due to the formation of an additive compound such as that mentioned in (ii) above.

Free thiosulphuric acid solutions are rendered more stable by the addition of alcohol; decomposition is facilitated by exposure to direct sunlight, and its course is dependent on the conditions prevailing at the time. Thus, acids generally accelerate the decomposition, as also do charcoal and finely divided (colloidal) sulphur in the presence of acid. Many salts retard the change, although some, such as mercuric and bismuth chlorides, lead acetate, sodium tungstate and sulphide, have no marked influence on the decomposition of the sodium salt by means of hydrochloric acid. Protective colloids also retard the reaction. The presence of alkaline substances has a stabilising effect, and the addition of sodium carbonate (about 0.2 gram per litre) to standard volumetric solutions of sodium thiosulphate will preserve them for a considerable time. Such solutions should be made with boiled distilled water in order to minimise the decomposing action attributed to bacteria.

According to Mayr and Kerschbaum bacterial action is the principal cause of the instability of thiosulphate solutions, and the protective action of alkali is ascribed to its restraining effect on the growth of the bacteria, a pH value of 9 to 10 being most effective. The presence of copper accelerates decomposition only when bacteria are also present. Thiosulphate solutions may be almost completely sterilised by the addition of amyl alcohol (1 per cent, by volume) or of mercuric cyanide (0.01 per cent, by weight).

The addition of formaldehyde to sodium thiosulphate solution prevents its decomposition with precipitation of sulphur on subsequent addition of hydrochloric acid; this effect is to be attributed to the formation of a condensation product analogous to that formed between formaldehyde and sodium hydrogen sulphite.

From the fact that sodium thiosulphate is but slightly affected by acetic acid, it is probable that thiosulphuric acid is a comparatively strong acid.

The claims of two formulae for the constitution of thiosulphuric acid have in the past called for serious consideration, the formulae being and .

The latter is of more recent date and the evidence in its favour is not very considerable. From the fact that sulphuryl chloride is not produced by the action of phosphorus pentachloride on sodium thiosulphate, the absence of the grouping has been argued; the reaction between an alkali thiosulphate and potassium cyanide with formation of alkali sulphite and potassium thioeyanate has also been adduced as an argument in favour of the dihydroxy constitution, as also has the apparently analogous instability of thiosulphuric and sulphurous acids as compared with sulphuric acid.

The fact that the ultimate products of the atmospheric oxidation of sodium thiosulphate are sulphur dioxide and sodium sulphate may be explained by assuming a primary decomposition into sulphite and sulphur (traces of the latter always being present in the salt), and subsequent oxidation of these.

On the other hand, much stronger evidence is available in favour of the earlier formula, which accords well with the relationship between the acid and the polythionic acids. The formation of sodium thiosulphate by Spring's synthesis from sodium sulphide and sodium sulphite is definitely favourable to this constitution, as also especially is the fact that an alkali thiosulphate will react with only an equimolecular proportion of an organic (alkyl) halide, the product most certainly having the constitution , where M represents the alkali metal and R the newly introduced organic radical. For example, in sodium ethyl thiosulphate, the juxtaposition of the ethyl radical and the sulphur atom is clearly proved by the facts that acids cause the production of ethyl hydrogen sulphide, C₂H₅.SH, whilst electrolytic reduction or the action of alkali yields ethyl disulphide and oxidation produces ethyl sulphonic acid, C₂H₅.SO₂.OH.

In view of such evidence it appears impossible to avoid the conclusion that the two hydrogen atoms in thiosulphuric acid must be differently linked to the central atom, namely by a sulphur atom and an oxygen atom, respectively. The formula is also stated to agree better with the strength of the acid, as shown by the relative inertness of the alkali salts towards acetic acid, and with the electrolytic conductivity of the salts.

Indications have been obtained of isomeric salts of the constitutions and , and if such a result could be placed beyond doubt, convincing evidence of the correctness of this formula for the acid would be forthcoming; the result, however, needs careful re-examination.

Such a constitution suggests that thiosulphuric acid is a mixed anhydride of sulphuric acid and hydrogen sulphide, thus:



and in aqueous solution one would expect decomposition to produce sulphuric acid and hydrogen sulphide rather than sulphurous acid and sulphur. Piccard and Thomas, by allowing sulphur trioxide and hydrogen sulphide to react in carbon dioxide solution at the temperature of liquid air, obtained a product which they presumed to be the true mixed anhydride. This compound contained a sexavalent central sulphur atom, the added sulphur atom being in the place of a negative bivalent oxygen atom of sulphuric acid. Ordinary thiosulphuric acid was considered to be an electronic isomeride of this compound, with a quadrivalent central atom, the added sulphur atom being neutral. The differences are expressed in the following co-ordinative formulae:

" True " thiosulphuric acid	Ordinary thiosulphuric acid.

The decomposition of thiosulphates by means of hydrochloric acid to yield sulphur dioxide with separation of sulphur serves as a primary identification test, the limit of sensitiveness being about 0.1 mg. S₂O₃ per c.c. The alkali thiosulphates produce with silver nitrate solution a white precipitate of thiosulphate which gradually turns yellow, then brown, and finally black, due to the formation of sulphide; the change is accelerated by warming:

Ag₂S₂O₃ + H₂O = Ag₂S + H₂SO₄.

This test is 100 times more sensitive than the acidification test.

Precipitation with copper sulphate also affords a test which is considerably more sensitive than the acidification test, and the result is not affected by the presence of polythionates.

Lead salts similarly yield a precipitate of thiosulphate, soluble in excess of alkali thiosulphate, which also blackens on warming, but the decomposition of the lead thiosulphate is less straightforward, a considerable quantity of sulphur being present in the greyish product. Barium chloride with a concentrated solution of an alkali thiosulphate forms a crystalline precipitate of sparingly soluble barium thiosulphate, of which one part dissolves in 480 of water at 18° C.

A solution of a thiosulphate also exhibits certain striking colour reactions. With ferric chloride solution a transient violet colour is obtained. With sodium nitroprusside which has been exposed to the atmosphere until it has become brown, a blue coloration is produced, whilst the thiosulphate solution, after reduction with sodium hydroxide and a little aluminium, gives with a fresh nitroprusside solution the violet colour characteristic of a sulphide. Very small quantities of a thiosulphate are sufficient to give a blue "ring test" when the solution is carefully poured on to a mixture of ammonium molybdate solution with concentrated sulphuric acid.

Other reactions readily available for the detection of a thiosulphate include reduction to hydrogen sulphide by most reducing agents, for instance by zinc and hydrochloric acid, oxidation to sulphuric acid or a sulphate, and formation of a thiocyanate on warming with an alkaline solution of a cyanide:

Na₂S₂O₃ + KCN = Na₂SO₃ + KCNS.

When solutions of iodine and sodium azide are mixed, no reaction occurs, but in the presence of thiosulphate vigorous evolution of nitrogen occurs. The reaction is extremely sensitive, the thiosulphate being effective at a dilution of 1 in 6×10⁶, and is not brought about by elementary sulphur or by any other sulphur compound except sulphide and thiocyanate.

The most trustworthy method for the gravimetric estimation of a pure thiosulphate is oxidation to sulphuric acid, for example by means of chlorine or bromine, or by the addition of an alkali salt of a halogen oxy-acid, and then precipitation with barium chloride.

Of volumetric processes, titration with iodine is the most commonly applied:

2Na₂S₂O₃ + I₂ = 2NaI + Na₂S₄O₆.

An alternative principle for the volumetric determination is to apply an acidimetric process. The thiosulphate may be oxidised with hydrogen peroxide in the presence of a known quantity of an alkali in excess and the excess of alkali measured by titration with standard acid, or the oxidation may be effected with hydrogen peroxide alone and the resulting acidity directly titrated with a standard alkali:

Na₂S₂O₃ + 4H₂O₂ = Na₂SO₄ + H₂SO₄ + 3H₂O.

The acidimetric procedure may be applied, however, in quite a different manner; mercuric chloride reacts with thiosulphate solutions giving rise to a white precipitate of a "thiochloride," Hg₃S₂Cl₂ or HgCl₂.2HgS, the equation being:

2Na₂S₂O₃ + 3HgCl₂ + 2H₂O = 2Na₂SO₄ + 4HCl + HgCl₂.2HgS.

By titration of the resulting acid the original quantity of thiosulphate can be calculated. This method has the especial advantage of being applicable in the presence of sulphides, which give a similar precipitate, but do not produce acidity. A similar process has been proposed using silver nitrate as precipitant:

2AgNO₃ + Na₂S₂O₃ + H₂O = Ag₂S + H₂SO₄ + 2NaNO₃.