The following experiments show the effect of variation in the conditions of the t.i.tration:--

~Effect of Varying Temperature.~--The solution should be t.i.trated while boiling. This is especially necessary for the last few c.c. in order to get a decided and fixed finishing point.

Temperature 15 C. 30 C. 70 C. 100 C.

"Uranium" required 18.0 c.c. 19.2 c.c. 19.0 c.c. 18.9 c.c.

~Effect of Varying Bulk.~--

Bulk 50.0 c.c. 100.0 c.c. 200.0 c.c. 300.0 c.c.

"Uranium" required 18.8 " 18.9 " 19.0 " 19.3 "

Variation in bulk affects the results; therefore, a constant bulk should be adhered to.

~Effect of Varying Sodium Acetate and Acetic Acid Solution.~--

Sodium acetate and acetic acid solution 0.0 c.c. 1.0 c.c. 5.0 c.c. 10.0 c.c. 20.0 c.c.

"Uranium" required 18.9 " 18.9 " 19.0 " 18.8 " 17.5 "

As in the t.i.tration with a.r.s.enates, an excess is dangerous to the a.s.say; a definite quant.i.ty (5 c.c.) should, therefore, be used.

~Effect of Foreign Salts.~--Besides the sodium acetate, &c., added, the only salts likely to be present are those of ammonia and magnesia. In three experiments, in one of which no foreign salts were introduced, while in the other two 5 grams of ammonic chloride and of magnesium sulphate respectively were added, there were required:--

With ammonic chloride 18.8 c.c. "Uranium" solution With magnesium sulphate 19.0 " "

Without foreign salts 18.9 " "

~Effect of Varying Phosphate.~--

"Phosphate" solution added 10.0 c.c. 20.0 c.c. 50.0 c.c. 100.0 c.c.

"Uranium" required 9.8 " 18.9 " 47.6 " 94.5 "

The quant.i.ty of phosphoric oxide in the a.s.say solution for the conditions of t.i.tration should not be much less than 0.05 gram. For smaller quant.i.ties the uranium solution should be diluted to half its strength, and the a.s.say solution concentrated by reducing its bulk to 50 c.c. and using 2.5 c.c. of the sodium acetate and acetic acid solution.

~Determination of Phosphoric Oxide in Apat.i.te.~--Weigh up 0.5 gram of the dried and powdered sample, and dissolve it in 5 c.c. of hydrochloric acid. Evaporate to a paste, add 5 c.c. of the sodic acetate and acetic acid solution, dilute to 100 c.c. with water, boil, and t.i.trate with uranium acetate solution.

In an example, 0.5 gram of apat.i.te required 37.4 c.c. of uranium acetate solution (standard equal to 0.5291 gram of phosphoric oxide). The sample therefore contained 0.1979 gram of P_{2}O_{5}, equal to 39.58 per cent.

~Determination of Phosphoric Oxide in an Iron Ore.~--Take 10 grams, boil with 50 c.c. of hydrochloric acid, and evaporate to a paste; take up with 10 c.c. of dilute hydrochloric acid, and dilute with water to 400 c.c. Pa.s.s sulphuretted hydrogen for nearly a quarter of an hour; warm, and filter. Boil off the excess of gas; cool, add ammonia till nearly neutral, and then a few drops of ferric chloride solution, and 4 or 5 grams of sodium acetate, with a drop or two of acetic acid. Boil and filter. Dissolve the precipitate in hot dilute hydrochloric acid, and add citro-magnesia mixture and ammonia; allow to stand overnight; filter, ignite, and weigh.

In an example, 10 grams of ore gave 28.5 milligrams of magnesic pyrophosphate, which is equivalent to 0.18 per cent. of phosphoric oxide.

~Determination of Phosphorus in Iron.~--Take from 2 to 10 grams (according to the amount of phosphorus present), and dissolve in aqua regia, keeping the nitric acid in excess; evaporate to dryness and take up with hydrochloric acid, boil, dilute, and filter. Add 10 c.c. of nitric acid, nearly neutralise with ammonia, render acid with 3 or 4 c.c. of nitric acid, and add 10 or 20 c.c. of ammonic molybdate solution. Heat for some time, allow to settle, filter, and wash the precipitate with a solution of ammonic nitrate. Dissolve the precipitate in dilute ammonia, nearly neutralise with dilute hydrochloric acid, and add first "magnesia mixture," and then ammonia; allow to stand overnight; filter, wash with dilute ammonia, dry, ignite, and weigh as magnesic pyrophosphate. Calculate to phosphorus.

PRACTICAL EXERCISES.

1. Ten grams of an iron yielded 12 milligrams of pyrophosphate of magnesia. What percentage of phosphorus did the metal contain?

2. Ten grams of an iron ore gave 12 milligrams of pyrophosphate. What percentage of phosphoric oxide did it contain?

3. What weight of apat.i.te 3Ca_{3}(PO_{4})_{2}.CaClF would require 50 c.c. of standard uranium solution (100 c.c. equal to 0.5 gram of P_{2}O_{5})?

4. You have reason to believe that a precipitate which has been weighed as magnetic pyrophosphate contains some a.r.s.enate. How would you determine the amount of phosphate really present?

5. Twenty c.c. of a solution of sodic phosphate containing 0.100 gram of P_{2}O_{5} was found to require a solution containing 0.700 gram of hydrated uranium acetate in a t.i.tration. The precipitate contains 80.09 per cent. uranium oxide and 19.91 per cent. of phosphoric oxide. What percentage of uranium oxide was contained in the uranic acetate?

NITROGEN AND NITRATES.

Nitrogen occurs in nature in the free state, and forms about four-fifths of the atmosphere. In combination, as nitrate, it is found in nitre (KNO_{3}), and Chili saltpetre (NaNO_{3}), minerals which have a commercial importance. The latter occurs in beds, and is extensively worked for use as a manure and in the preparation of nitric acid.

Nitrogen is mainly characterised by negative properties, although many of its compounds are very energetic bodies. It is a gas, present everywhere, but so inactive that the a.s.sayer can always afford to ignore its presence, and, except in testing furnace gases, &c., he is never called on to determine its quant.i.ty.

The nitrates are an important cla.s.s of salts, and may be looked on as compounds of the bases with nitric pentoxide (N_{2}O_{5}). They are, with the exception of a few basic compounds, soluble in water, and are remarkable for the ease with which they give up their oxygen. The alkaline nitrates fuse readily, and lose oxygen with effervescence forming nitrites; while at a higher temperature they yield more oxygen and lose their nitrogen, either as a lower oxide or as nitrogen. The nitrates of the metals, on heating, leave the oxide of the metal. It is as yielders of oxygen that nitrates are so largely used in the manufacture of explosives. Gunpowder contains from 65 to 75 per cent. of pota.s.sium nitrate (nitre).

Nitrates are best detected and determined by their yielding nitric oxide when treated with sulphuric acid and a suitable reducing agent, such as ferrous sulphate, mercury, or copper. Nitric oxide is a colourless gas very slightly soluble in water. It combines at once with oxygen, on mixing with the air, to form brown "nitrous fumes," and dissolves in a solution of ferrous sulphate, producing a characteristic blackish-brown colour. It is this colour which affords the best and most easily-applied test for nitrates. The substance suspected to contain nitrates is dissolved in about 1 c.c. of water, and treated with an equal volume of strong sulphuric acid. After cooling, a solution of ferrous sulphate is poured on its surface, so as to form a layer resting on it. On standing, a brown or black ring is developed where the liquids join, if any nitrate or nitrite is present. Nitrites are distinguished from nitrates by effervescing and yielding brown fumes when treated with a little dilute sulphuric acid.

The separation of nitrates is in many cases difficult. Generally, on treating the substance with water, the nitrate will be in the solution, and is filtered off from any insoluble matter. In the exceptional cases it is got into solution by treating with a boiling solution of sodium carbonate; the nitrate will contain it as an alkaline nitrate.

Since, however, in their determination, nitrates are never separated and weighed as such, the difficulty of separating them has little importance. Usually, the determination can be made on the original aqueous solution, and it is never necessary to do more than remove any special substance which has a bad effect; and this is easily done by the usual reagents.

GRAVIMETRIC DETERMINATION.

It follows from what has been said that there is no direct gravimetric determination. The percentage of nitrogen pentoxide (N_{2}O_{5}) in a comparatively pure nitrate is sometimes determined indirectly in the following way:--Place in a platinum-crucible 4 or 5 grams of powdered and cleaned quartz. Ignite, cool in a desiccator, and weigh with the cover. Mix 1 gram of the dried and powdered salt with the quartz in the crucible by stirring with a stout platinum-wire. Cover the crucible, and heat in a Bunsen-burner flame at scarcely visible redness for half-an-hour. Cool and weigh. The loss in weight gives the amount of nitrogen pentoxide. Sulphates and chlorides in moderate quant.i.ty do not interfere. The following is an example of the process:--

Crucible and sand 26.6485 grams Nitre taken 1.0000 "

------- 27.6485 "

Weight after ignition 27.1160 "

------- Loss on ignition 0.5325 "

This is equal to 53.25 per cent. of nitrogen pentoxide.

VOLUMETRIC DETERMINATION.

This is based on the oxidising action of nitric acid, or of nitrates in acid solutions on ferrous salts. The pentoxide (N_{2}O_{5}) of the nitrate is reduced to nitric oxide (NO), so that 336 parts of iron peroxidised represent 108 parts of nitric pentoxide as oxidising agent.[112] The quant.i.ty of iron peroxidised is determined by taking a known quant.i.ty of ferrous salt, oxidizing with a weighed sample of nitrate, and then determining the residual ferrous iron by t.i.tration with bichromate or permanganate of pota.s.sium solution. The difference between the ferrous iron taken and that found, gives the amount oxidized by the nitrate. The speed with which nitric oxide takes up oxygen from the air, and thus becomes capable of oxidising more iron, renders some precautions necessary; ferrous chloride should, therefore, be used, since it is easier to expel nitric oxide (by boiling) from solutions of a chloride than it is from those of a sulphate. The process is as follows:--Dissolve 2 grams of thin soft iron wire in 50 c.c. of hydrochloric acid in a flask provided with an arrangement for maintaining an atmosphere of carbon dioxide. When the iron has dissolved, allow the solution to cool, and add 0.5 gram of the nitrate.

Heat gently for a few minutes, and then boil until the nitric oxide is expelled. An atmosphere of carbon dioxide must be kept up. Dilute with water, and t.i.trate the residual iron with standard solution of bichromate of pota.s.sium. The standard "bichromate" is made by dissolving 17.5 grams of the salt (K_{2}Cr_{2}O_{7}) in water, and diluting to 1 litre: 100 c.c. equal 2 grams of iron. Deduct the weight of iron found from the 2 grams originally taken, and multiply by 0.3214. This gives the weight of the pentoxide in the sample. In an example, 0.5 gram of nitre was taken, and 59.4 c.c. of the "bichromate" solution were required. The 59.4 c.c. thus used are equivalent to 1.198 gram of iron.

This leaves 0.822 gram as the quant.i.ty oxidised by the nitre, which, multiplied by 0.3214, gives 0.2642 gram for the nitrogen pentoxide, or 52.8 per cent.

GASOMETRIC METHOD.

This is based upon the measurement of the nitric oxide evolved on shaking up a weighed quant.i.ty of the nitrate with sulphuric acid over mercury in a nitrometer. Each c.c. of nitric oxide obtained, when reduced to normal temperature and pressure, is equivalent to:--

0.627 milligram of nitrogen.

1.343 " of nitric oxide.

2.418 " of nitric pentoxide.

2.820 " of nitric acid.

3.805 " of sodium nitrate.