and fuses to a brown, uncrystallisable mass (caramel); long boiling with water increases its colour, and lessens its tendency to crystallise; its aqueous solution dissolves alkalies, earths, and many metallic oxides, with facility. The presence of cane sugar in solutions containing certain metallic salts prevents the precipitation of their oxides by alkalies. The oxides of copper and iron are amongst those thus kept in solution. Sugar also possesses the power of effecting the partial or complete reduction of many metallic oxides, if boiled with their salts; the first results is exemplified in the case of the chromates; for if a chromate be added to a solution of sugar, and to the mixture a few drops of free acid, the chromic acid suffers reduction to chromic oxide, which, dissolving in the excess of acid, imparts a green colour to the liquid. Mercuric salts become reduced to mercurous, whilst the salts of gold throw down a precipitate of the metal in fine powder. The action of strong oil of vitriol on cane sugar is very energetic. The sugar is instantly reduced to a black charred mass, whilst carbonic and formic acids are given off. The same effects are produced by exposing it to dry chlorine at a temperature of 212° F. By nitric acid of sp. gr. 1·25, cane sugar is converted into saccharic acid; if a stronger acid be employed, oxalic acid is produced. When a mixture of concentrated nitric and sulphuric acids is poured on to cane sugar, an explosive compound, resembling gun cotton, is produced. This body is known as ‘nitro-sugar.’ Weak syrups take up about half as much hydrate of calcium as they contain sugar; when slowly crystallised, it assumes the form of oblique 4-sided prisms, terminated by 2-sided summits. Sp. gr. 1·60 (1·577—Ure).
Pur. Moist or muscovado sugar and crushed lump sugar are occasionally adulterated with chalk, plaster, sand, potato-flour, and other fecula; but frequently, and in certain neighbourhoods constantly, with starch sugar or potato-sugar.[203] These frauds may be detected as follows:
[203] See further on.
Tests.—1. Pure cane sugar dissolves freely and entirely in both water and proof spirit, forming transparent colourless solutions, which are unaffected by either sulphuretted hydrogen or dilute sulphuric acid.—2. Its solution bends the luminous rays in circumpolarisation to the right, whereas grape and fecula sugars bend it to the left.[204]—3. (Chevallier.) Boiled for a short time in water containing 2 or 3% of caustic potassa, the liquid remains colourless; but it turns brown, which is more or less intense, according to the quantity, if starch sugar is present. Even 2 or 3% of starch sugar may be thus detected.—4. (E. Krantz.) A filtered solution of 33 gr. of cane or beet sugar
in 1 fl. oz. of water, mixed with 3 gr. of pure hydrate of potassium, and then agitated with 11⁄2 gr. of sulphate of copper in an air-tight bottle, remains clear, even after the lapse of several days; but if starch sugar be present, a red precipitate is formed after some time; and if it is present in considerable quantity, the copper will be wholly converted into oxide within 24 hours, the solution turning first blue or green, and then entirely losing its colour.—5. (Trommer’s test.) A solution of cane sugar is mixed with a solution of sulphate of copper, and hydrate of potassium added in excess; a blue liquid is obtained, which, on being heated, is at first but little altered; a small quantity of red powder falls after a time, but the liquid long retains its blue tint. When grape sugar or fecula sugar is thus treated, the first application of heat throws down a copious greenish precipitate, which rapidly changes to scarlet, and eventually to dark red, leaving a nearly colourless solution. This is an excellent test for distinguishing the two varieties of sugar, or discovering an admixture of grape sugar with cane sugar. The 1⁄1000th part of grape sugar may be thus detected. The proportion of oxide of copper produced affords a good criterion, not only of the purity of the sugar, but also of the extent of the adulteration.—6. (Ure.) Dissolve a little sulphate of copper (say 20 gr.) in a measured quantity of water, and add to it, in the cold, a solution of hydrate of potassium, until, by testing with turmeric paper, the liquid appears faintly alkaline, shown by the paper becoming slightly brown. If a small quantity of this test-liquor (previously well shaken) be added to an aqueous solution of the sugar, and the whole boiled, the solution becomes at first green, and then olive-green, if dextrin is present; but if it contain grape sugar, the salt of copper is immediately reduced into the state of orange and oxide; whilst a solution of pure sugar undergoes no change, or is scarcely altered.—7. M. Riffard,[205] taking advantage of the fact that sugar, like tartaric, malic, citric acid, and albumen, prevents the precipitation of iron by ammonia, employs iron as a means for estimating sugar. A solution containing sugar and iron in a certain proportion, when saturated with ammonia, will form a compound of a fine red colour, which remains clear if no alkaline earthy metals are present. M. Riffard has applied to sugar the method proposed by M. Juette for the estimation of tartaric acid. He observed that a neutral or acid solution of crystallised perchloride of iron, when heated for a considerable time to 100° C, requires 2·710 grams of sugar, if 100 milligrams of iron are to remain in solution in the presence of ammonia. If, on the other hand, the solution is prepared simply by dissolving crystallised perchloride of iron in pure water, without the addition of an acid, 100
milligrams of iron only require 2·587 grammes of sugar to remain dissolved. In this case the liquid is perfectly clear, and remains so; but if a smaller quantity of sugar be added, it is turbid, and deposits peroxide of iron. To estimate the sugar by this process, 25·870 grammes of the substance to be tested are dissolved, the solution mixed with a few drops of oxalate of ammonia to precipitate the lime, filtered and made up with water to 250 c.c., 25 c.c. of this mixture require the addition of as many milligrams of iron as there are per cents. of pure sugar in the sample under examination, and by two tests the following results will be arrived at:—With n milligrams of iron the solution is clear. With n + 1 milligrams of iron the solution is precipitated. n representing the number of per cents. of sugar contained in the sample.—8. M. Perrot’s method for the determination of sugars by means of normal solutions is as follows:—He prepares a standard solution of copper by dissolving 39·275 grams of sulphate of copper, very pure, and dried between several folds of filtering paper, and makes it up with distilled water to 1000 c.c. Each c.c. of this solution contains 0·01 grams of copper. On the other hand, he dissolves about 25· grams of pure cyanide of potassium in 1 litre of distilled water. Of this solution 10 c.c. are taken and put in a flask, to which about 20 c.c. of ammonia are added, and the liquid is kept at a temperature of 60° or 70°. He pours in the copper solution drop by drop by means of a burette graduated into tenths of a c.c., until there appears the blue tint characteristic of salts of copper in an ammoniacal solution. The number of degrees of the burette are then read off, and indicate the quantity of copper which has been required to produce the reaction. The solution of the sugar in question (previously inverted if it is required to determine crystalline sugar) is then placed in contact with an excess of Fehling’s liquor, and reduced in the water-bath. The whole is filtered in order to collect the precipitate of suboxide, which is first well washed with hot water, and dissolved in nitric acid, diluted with an equal volume of water, and a few fragments of chlorate of potassa are added. This solution is effected on the filter, which is then carefully washed in acidulated water. The filtrate to which the washings are added is then mixed with water enough to make up 100 or 150 c.c., and is then poured by means of the burette into 10 c.c. of cyanide, mixed with 20 c.c. of ammonia as above, stopping when the blue colour appears, and reading off the quantity of copper employed. From the former experiment it is known how much copper 10 c.c. of the cyanide solution require. Hence it is easy to calculate the total amount of copper which has been present as suboxide. The amount of sugar is then found from the data that 9298 parts of copper equal 5000 of crystalline sugar, or 5263 of
glucose.[206]—9. The specific gravities and crystalline forms offer other means of distinguishing the varieties of sugar.
[204] Of late years, owing to the little difference in price between the two, this form of adulteration has been abandoned.—Ed.
[205] ‘Journ. de Pharm. et de Chimie,’ 1874, 49 (‘Pharm. Year Book,’ 1874).
[206] ‘Comptes Rendus Hebdomadaire des Sciences’ (‘Chem. News’), January 5th, 1877.