_Pota.s.sium_ may be determined by precipitation as pota.s.sium platino-chloride thus:--Dissolve 0.5 gramme in a small quant.i.ty (say 10 c.c.) of water, and carefully acidulate with hydrochloric acid, evaporate the resultant liquor to dryness in a tared platinum basin, and heat the residue gradually to dull redness. Cool in a desicator, weigh, and express the result as "mixed chlorides," _i.e._ chlorides of soda and potash. To the mixed chlorides add 10 c.c. water, and platinic chloride in excess (the quant.i.ty may be three times the amount of the mixed chlorides) and evaporate nearly to dryness; add 15 c.c. alcohol and allow to stand three hours covered with a watch-gla.s.s, giving the dish a gentle rotatory movement occasionally. The clear liquid is decanted through a tared filter, and the precipitate well washed with alcohol by decantation, and finally transferred to the filter, dried and weighed. From the weight of pota.s.sium platino-chloride, K_{2}PtCl_{6}, is calculated the amount of pota.s.sium oxide K_{2}O by the use of the factor 94/488.2 or 0.19254.

_Chlorides_, determined with N/10 silver nitrate solution, and calculated to KCl.

_Sulphates_, estimated as barium sulphate, and calculated to K_{2}SO_{4}.

_Sodium Carbonate_, found by deducting the K_{2}CO_{3} corresponding to the actual pota.s.sium as determined above, from the total alkali.

_Iron_, precipitated with excess of ammonia, filtered, ignited, and weighed as Fe_{2}O_{3}.

SODIUM CHLORIDE (COMMON SALT).

This should be examined for the following:--

_Actual Chloride_, either t.i.trated with N/10 silver nitrate solution, using neutral pota.s.sium chromate solution as indicator, or, preferably, estimated gravimetrically as silver chloride by precipitation with silver nitrate solution, the precipitate transferred to a tared filter paper, washed, dried and weighed.

_Insoluble matter_, remaining on dissolving 5 grammes in water, and filtering. This is washed, dried, ignited and weighed.

_Moisture._--5 grammes are weighed into a platinum crucible, and heat gently applied. The temperature is gradually increased to a dull red heat, which is maintained for a few minutes, the dish cooled in a desicator, and weighed.

_Sulphates_ are estimated by precipitation as barium sulphate and calculated to Na_{2}SO_{4}.

_Sodium._--This may be determined by converting the salt into sodium sulphate by the action of concentrated sulphuric acid, igniting to drive off hydrochloric and sulphuric acids, and fusing the ma.s.s until constant in weight, weighing finally as Na_{2}SO_{4}.

POTa.s.sIUM CHLORIDE.

This should be examined, in the same way as sodium chloride, for chloride, insoluble matter, moisture, and sulphate. The pota.s.sium may be determined as pota.s.sium platino-chloride, as described under carbonate of potash.

SILICATES OF SODA AND POTASH.

The most important determinations for these are total alkali and silica.

_Total alkali_ is estimated by dissolving 2 grammes in distilled water, and t.i.trating when cold, with N/1 acid, using methyl orange as indicator.

_Silica_ may be determined by dissolving 1 gramme in distilled water, rendering the solution acid with HCl, and evaporating to complete dryness on the water-bath, after which the residue is moistened with HCl and again evaporated, this operation being repeated a third time. The residue is then heated to about 150 C., extracted with hot dilute HCl, filtered, thoroughly washed, dried, ignited in a tared platinum crucible, and weighed as SiO_{2}.

ESSENTIAL OILS.

As already stated, these are very liable to adulteration, and an examination of all kinds of oil is desirable, while in the case of the more expensive varieties it should never be omitted.

_Specific Gravity._--As with fats and oils, this is usually taken at 15 C., and compared with water at the same temperature. In the case of otto of rose and guaiac wood oil, however, which are solid at this temperature, it is generally observed at 30 C. compared with water at 15 C.

The specific gravity is preferably taken in a bottle or U-tube, but if sufficient of the oil is available and a high degree of accuracy is not necessary, it may be taken either with a Westphal balance, or by means of a hydrometer.

_Optical Rotation._--For this purpose a special instrument, known as a polarimeter, is required, details of the construction and use of which would be out of place here. Suffice it to mention that temperature plays an important part in the determination of the optical activity of certain essential oils, notably in the case of lemon and orange oils.

For these Gildemeister and Hoffmann give the following corrections:--

Lemon oil, below 20 C. subtract 9' for each degree below, above 20 C.

add 8' for each degree above.

Orange oil, below 20 C. subtract 14' for each degree below, above 20 C. add 13' for each degree above.

_Refractive Index._--This figure is occasionally useful, and is best determined with an Abbe refractometer, at 20 C.

_Solubility in Alcohol._--This is found by running alcohol of the requisite strength from a burette into a measured volume of the oil with constant agitation, until the oil forms a clear solution with the alcohol. Having noted the quant.i.ty of alcohol added, it is well to run in a small further quant.i.ty of alcohol, and observe whether any opalescence or cloudiness appears.

_Acid_, _ester_, and _saponification values_ are determined exactly as described under fats and oils. Instead of expressing the result as saponification value or number, the percentage of ester, calculated in the form of the most important ester present, may be obtained by multiplying the number of c.c. of N/1 alkali absorbed in the saponification by the molecular weight of the ester. Thus, to find the percentage as linalyl acetate, the number of c.c. absorbed would be multiplied by 0.196 and by 100, and divided by the weight of oil taken.

_Alcohols._--For the estimation of these, if the oil contains much ester it must first be saponified with alcoholic potash, to liberate the combined alcohols, and after neutralising the excess of alkali with acid, the oil is washed into a separating funnel with water, separated, dried with anhydrous sodium sulphate, and is then ready for the alcohol determination.

If there is only a small quant.i.ty of ester present, this preliminary saponification is unnecessary.

The alcohols are estimated by conversion into their acetic esters, which are then saponified with standard alcoholic potash, thereby furnishing a measure of the amount of alcohol esterified.

Ten c.c. of the oil is placed in a flask with an equal volume of acetic anhydride, and 2 grammes of anhydrous sodium acetate, and gently boiled for an hour to an hour and a half. After cooling, water is added, and the contents of the flask heated on the water-bath for fifteen to thirty minutes, after which they are cooled, transferred to a separating funnel, and washed with a brine solution until the washings cease to give an acid reaction with litmus paper. The oil is now dried with anhydrous sodium sulphate, filtered, and 1-2 grammes weighed into a flask and saponified with alcoholic potash as in the determination of ester or saponification value.

The calculation is a little complicated, but an example may perhaps serve to make it clear.

A geranium oil containing 26.9 per cent. of ester, calculated as geranyl tiglate, was acetylated, after saponification, to liberate the combined geraniol, and 2.3825 grammes of the acetylated oil required 9.1 c.c. of N/1 alkali for its saponification.

Now every 196 grammes of geranyl acetate present in the acetylated oil correspond to 154 grammes of geraniol, so that for every 196 grammes of ester now present in the oil, 42 grammes have been added to its weight, and it is therefore necessary to make a deduction from the weight of oil taken for the final saponification to allow for this, and since each c.c. of N/1 alkali absorbed corresponds to 0.196 gramme of geranyl acetate, the amount to be deducted is found by multiplying the number of c.c. absorbed by 0.042 gramme, the formula for the estimation of total alcohols thus becoming in the example given:--

9.1 0.154 100 Per cent. of geraniol = ---------------------- = 70.2 2.3825 - (9.1 0.042)

The percentage of combined alcohols can be calculated from the amount of ester found, and by subtracting this from the percentage of total alcohols, that of the free alcohols is obtained.

In the example quoted, the ester corresponds to 17.6 per cent. geraniol, and this, deducted from the total alcohols, gives 52.6 per cent. free alcohols, calculated as geraniol.

This process gives accurate results with geraniol, borneol, and menthol, but with linalol and terpineol the figures obtained are only comparative, a considerable quant.i.ty of these alcohols being decomposed during the acetylation. The aldehyde citronellal is converted by acetic anhydride into isopulegol acetate, so that this is also included in the determination of graniol in citronella oil.

_Phenols._--These bodies are soluble in alkalies, and may be estimated by measuring 5 c.c. or 10 c.c. of the oil into a Hirschsohn flask (a flask of about 100 c.c. capacity with a long narrow neck holding 10 c.c., graduated in tenths of a c.c.), adding 25 c.c. of a 5 per cent.

aqueous caustic potash solution, and warming in the water-bath, then adding another 25 c.c., and after one hour in the water-bath filling the flask with the potash solution until the unabsorbed oil rises into the neck of the flask, the volume of this oil being read off when it has cooled down to the temperature of the laboratory. From the volume of oil dissolved the percentage of phenols is readily calculated.

_Aldehydes._--In the estimation of these substances, use is made of their property of combining with sodium bisulphite to form compounds soluble in hot water. From 5-10 c.c. of the oil is measured into a Hirschsohn flask, about 30 c.c. of a hot saturated solution of sodium bisulphite added, and the flask immersed in a boiling water bath, and thoroughly shaken at frequent intervals. Further quant.i.ties of the bisulphite solution are gradually added, until, after about one hour, the unabsorbed oil rises into the neck of the flask, where, after cooling, its volume is read off, and the percentage of absorbed oil, or aldehydes, calculated.

In the case of lemon oil, where the proportion of aldehydes, though of great importance, is relatively very small, it is necessary to first concentrate the aldehydes before determining them. For this purpose, 100 c.c. of the oil is placed in a Ladenburg fractional distillation flask, and 90 c.c. distilled off under a pressure of not more than 40 mm., and the residue steam distilled. The oil so obtained is separated from the condensed water, measured, dried, and 5 c.c. a.s.sayed for aldehydes either by the process already described, or by the following process devised by Burgess (_a.n.a.lyst_, 1904, 78):--

Five c.c. of the oil are placed in the Hirschsohn flask, about 20 c.c.

of a saturated solution of neutral sodium sulphite added, together with a few drops of rosolic acid solution as indicator, and the flask placed in a boiling water-bath and continually agitated. The contents of the flask soon become red owing to the liberation of free alkali by the combination of the aldehyde with part of the sodium sulphite, and this coloration is just discharged by the addition of sufficient 10 per cent. acetic-acid solution. The flask is again placed in the water-bath, the shaking continued, and any further alkali liberated neutralised by more acetic acid, the process being continued in this way until no further red colour is produced. The flask is then filled with the sodium sulphite solution, the volume of the cooled unabsorbed oil read off, and the percentage of aldehydes calculated as before.

_Solidifying Point, or Congealing Point._--This is of some importance in the examination of anise and fennel oils, and is also useful in the examination of otto of rose. A suitable apparatus may be made by obtaining three test tubes, of different sizes, which will fit one inside the other, and fixing them together in this way through corks.

The innermost tube is then filled with the oil, and a sensitive thermometer, similar to that described under the t.i.tre test for fats, suspended with its bulb completely immersed in the oil. With anise and fennel, the oil is cooled down with constant stirring until it just starts crystallising, when the stirring is interrupted, and the maximum temperature to which the mercury rises noted. This is the solidifying point.

In the case of otto of rose, the otto is continually stirred, and the point at which the first crystal is observed is usually regarded as the congealing point.