Or the iron may be separated from the copper in the tin by the following process:--Dissolve 5 grams of metal in 30 c.c. of hydrochloric acid and 5 c.c. of nitric acid, and evaporate to dryness. Take up with 5 c.c. of dilute hydrochloric acid, add 10 grams of potash dissolved in 30 c.c. of water, and warm till the tin is dissolved. Pa.s.s sulphuretted hydrogen, boil, cool, and filter. The iron and copper will be in the precipitate.
They are separated in the ordinary manner.
PRACTICAL EXERCISES.
1. Calculate from the following determinations the percentages of ferrous, ferric, and total iron in the sample of ore used.
1 gram of ore dissolved and t.i.trated required 26.7 c.c. of bichromate of pota.s.sium solution.
1 gram of ore dissolved, reduced, and t.i.trated required 43.5 c.c. of bichromate of pota.s.sium solution.
Standard = 1.014.
2. One gram of an ore contained 0.307 gram of ferrous iron and 0.655 gram of total iron. The iron existing as oxide, what are the percentages of ferrous oxide (FeO) and ferric oxide (Fe_{2}O_{3}) in the ore?
3. One gram of brown iron ore dissolved in hydrochloric acid required 59.2 c.c. of stannous chloride (standard = 0.930). Another gram dissolved in acid and t.i.trated with "permanganate" required 8.2 c.c.
(standard = 0.4951). Calculate the percentages of ferrous, ferric, and total iron.
4. Another gram of the same ore, roasted, dissolved and t.i.trated with stannous chloride, required 63.5 c.c. To what extent does this result confirm the others?
5. Two grams of a metal were dissolved and diluted to 100 c.c. Five c.c.
were taken for a colorimetric determination, and required 4.5 c.c. of the standard ferric chloride solution. What is the percentage of iron in the metal?
NICKEL.
Nickel and cobalt are closely related in their chemical properties, and may best be considered together. Nickel is the commoner of the two, and is met with in commerce alloyed with copper and zinc as German silver; as also in the coinage of the United States and on the Continent. It is used for plating polished iron and steel goods, forming a coating little liable to rust and taking a good polish. The ores of nickel are not very common. Kupfernickel and chloanthite are a.r.s.enides of nickel with, generally, more or less iron and cobalt. Noumeite and garnierite are hydrated silicates of nickel and magnesia. The chief sources of nickel are these silicates, which are found in large quant.i.ty in New Caledonia; and a pyrites found in Norway, containing three or four per cent. of the metal. In smaller quant.i.ties it is more widely distributed, being frequently met with in copper ores; consequently, commercial copper is rarely free from it.
Nickel is readily soluble in moderately concentrated nitric acid. Its salts are mostly green, and soluble in excess of ammonia, forming blue solutions; in these respects it resembles copper. The acid solutions, however, are not precipitated by sulphuretted hydrogen, although in alkaline solutions a black sulphide is formed which is insoluble in dilute hydrochloric acid. If the sulphide is formed in a solution containing much free ammonia, the precipitation is incomplete, some sulphide remaining in the solution and colouring it dark brown. These reactions serve to distinguish and separate nickel from other metals, except cobalt. If the separated sulphide be heated in a borax bead, the colour obtained will be a sherry brown in the outer flame, and grey or colourless in the inner flame if nickel only is present. In the presence of cobalt these colours are masked by the intense and characteristic blue yielded in both flames by that metal.
DRY a.s.sAY.
The dry a.s.say of nickel (cobalt being at the same time determined) is based on the formation of a speise which will carry the cobalt, nickel, copper, and some of the iron of the ore in combination with a.r.s.enic. A speise of this kind, fused and exposed at a red heat to air, first loses a.r.s.enide of iron by oxidation. It is only when the iron has been oxidised that the a.r.s.enide of cobalt begins to be attacked; and when the removal of the cobalt is complete, the nickel commences to pa.s.s into the slag, the copper being left till last. The changes are rendered evident by fusion in contact with borax. The process is as follows:--Weigh up 5 grams of the ore, and calcine thoroughly on a roasting dish in the m.u.f.fle. Rub up with some anthracite, and re-roast. Mix intimately with from 3 to 5 grams of metallic a.r.s.enic, and heat in a small covered clay crucible at dull redness in a m.u.f.fle until no more fumes of a.r.s.enic come off (about 15 minutes). Take out the crucible, and inject a mixture of 20 grams of carbonate of soda, 5 grams of flour, and 2 grams of fused borax. Place in the wind furnace, and raise the temperature gradually until the charge is in a state of tranquil fusion. Pour; when cold, detach the b.u.t.ton of speise, and weigh.
Weigh out carefully a portion of about 1 gram of it. Place a shallow clay dish in the m.u.f.fle, and heat it to bright redness; then add about 1.5 gram of borax gla.s.s wrapped in a piece of tissue paper; when this has fused, drop the piece of speise into it. Close the m.u.f.fle until the speise has melted, which should be almost at once. The a.r.s.enide of iron will oxidise first, and when this has ceased the surface of the b.u.t.ton brightens. Remove it from the m.u.f.fle, and quench in water as soon as the b.u.t.ton has solidified. The borax should be coloured slightly blue.
Weigh: the loss is the a.r.s.enide of iron. Repeat the operation with the weighed b.u.t.ton on another dish, using rather less borax. Continue the scorification until a film, green when cold, floating on the surface of the b.u.t.ton shows that the nickel is beginning to oxidise. Cool, separate, and weigh the b.u.t.ton as before. The loss is the a.r.s.enide of cobalt.
If copper is absent, the speise is now a.r.s.enide of nickel.
The weight of nickel corresponding to the a.r.s.enide got is calculated by multiplying by 0.607; and, similarly, the weight of the cobalt is ascertained by multiplying the loss in the last scorification by 0.615.[71] It must be remembered that the nickel and cobalt so obtained are derived from a fraction only of the speise yielded by the ore taken, so that the results must be multiplied by the weight of the whole of the speise, and divided by the weight of the fragment used in the determination. As an example, suppose 5 grams of ore gave 3.3 grams of speise, and 1.1 gram of this gave 0.8 gram of nickel a.r.s.enide. Then--
0.80.607 = 0.4856 gram of nickel 0.48563.3/1.1 = 1.456 gram of nickel
And this being obtained from 5 grams of ore is equivalent to 29.12 per cent.
When copper is also present, weigh up accurately about 0.5 gram of gold, and place it on the scorifier with the b.u.t.ton of nickel and copper a.r.s.enide, using borax as before. Scorify until the b.u.t.ton shows the bluish-green colour of a fused gold-copper alloy. Then cool, and weigh the b.u.t.ton of copper and gold. The increase in weight of the gold b.u.t.ton gives the copper as metal. The weight of the copper multiplied by 1.395 is the weight of the copper a.r.s.enide (Cu_{3}As) present. The difference will be the nickel a.r.s.enide.
The student should enter the weighings in his book as follows:
Ore taken -- grams Speise got -- "
Speise taken -- grams a.r.s.enides of cobalt, nickel, and copper -- "
" nickel and copper -- "
Gold added -- "
Gold and copper got -- "
Showing Cobalt -- per cent.
Nickel -- "
Copper -- "
WET METHODS.
~Solution and Separation.~--Two or three grams of a rich ore, or 5 to 10 grams if poor, are taken for the a.s.say. If much a.r.s.enic is present (as is usually the case), the ore must be calcined before attacking with acids. Transfer to a flask; and boil, first with hydrochloric acid until the oxides are dissolved, and then with the help of nitric acid, until nothing metalliferous is left. Dilute, nearly neutralise with soda, and separate the iron as basic acetate,[72] as described in page 233.
Through the filtrate pa.s.s sulphuretted hydrogen till saturated. Allow to settle (best overnight), filter, and wash. Transfer the precipitate to a beaker, and dissolve in nitric acid. Dilute with water, pa.s.s sulphuretted hydrogen, and filter off the precipitate, if any. Boil off the gas, add ammonia until a precipitate is formed, and then acidify somewhat strongly with acetic acid. Pa.s.s sulphuretted hydrogen in a slow stream until any white precipitate of zinc sulphide, there may be, begins to darken. Filter; to the filtrate add ammonia, and pa.s.s sulphuretted hydrogen. The precipitate will contain the nickel and cobalt as sulphides.
Where small quant.i.ties of nickel and cobalt are present, and an approximate determination is sufficient, they can be concentrated as follows:--Remove the copper, &c., by pa.s.sing sulphuretted hydrogen through the acid solution and filtering; add ammonia to the filtrate, and again pa.s.s sulphuretted hydrogen; then heat nearly to boiling, and filter. Dissolve the precipitate off the filter with dilute hydrochloric acid; the residue will contain nearly all the nickel and cobalt as sulphides.
~Separation of Nickel and Cobalt.~--Dissolve the sulphides separated as above in nitric acid; render alkaline with a solution of potash, then acidify with acetic acid; add a concentrated solution of _nitrite_ of potash. The liquid after this addition must have an acid reaction. Allow to stand for 24 hours in a warm place. Filter off the yellow precipitate of nitrite of potash and cobalt, and wash with a 10 per cent. solution of acetate of potash. The cobalt is determined in the precipitate in the way described under _Cobalt_. The nickel is separated from the solution by boiling with sodic hydrate, filtering, and dissolving the precipitate in nitric acid. The solution will contain the nickel.
GRAVIMETRIC DETERMINATION.
The solution, which contains the nickel free from other metals, is heated, and a solution of sodic hydrate added in slight excess. The precipitate is filtered off, washed with boiling water, dried, ignited at a red heat, and weighed when cold. The ignited substance is nickel oxide (NiO), and contains 78.67 per cent. of nickel. The oxide is a green powder, readily and completely soluble in hydrochloric acid, and without action on litmus paper. It is very easily reduced by ignition in hydrogen to metallic nickel.
[Ill.u.s.tration: FIG. 56.]
Nickel is also determined by electrolysis, as follows:--The nitric acid solution is rendered strongly ammoniacal, and placed under the electrolytic apparatus used for the copper a.s.say. Three cells (fig. 56), however, must be used, coupled up for intensity, that is, with the zinc of one connected with the copper of the next. The electrolysis is allowed to go on overnight, and in the morning the nickel will be deposited as a bright and coherent film. A portion of the solution is drawn off with a pipette; if it smells of ammonia, has no blue colour, and gives no precipitate with ammonic sulphide, the separation is complete. Wash the cylinder containing the deposited metal, first with water and then with alcohol, as in the copper a.s.say. Dry in the water oven, and weigh. The increase in weight is metallic nickel.
As an example:--There was taken 1 gram of a nickel alloy used for coinage. It was dissolved in 10 c.c. of nitric acid, and diluted to 100 c.c. with water. The copper was then precipitated by electrolysis. It weighed 0.734 gram. The solution, after electrolysis, was treated with sulphuretted hydrogen, and the remaining copper was thrown down as sulphide, and estimated colorimetrically. This amounted to 3-1/2 milligrams. The filtrate was evaporated, treated with ammonia, warmed, and filtered. The ferric hydrate was dissolved in dilute acid, and reprecipitated, dried, ignited, and weighed. Its weight was 0.0310 gram.
The two filtrates were mixed, and reduced in bulk to about 50 c.c.; a considerable excess of ammonia was added, and the nickel precipitated by electrolysis. It weighed 0.2434 gram. These quant.i.ties are equivalent to:
Copper 73.75 per cent.
Nickel 24.34 "
Iron 2.17 "
------ 100.26
VOLUMETRIC DETERMINATION.
An alkaline solution of pota.s.sium cyanide, to which a little pota.s.sium iodide has been added, can be a.s.sayed for its strength in cyanide by t.i.trating with a standard solution of silver nitrate. Nickel interferes with this a.s.say, doing the work of its equivalent of silver; and the quant.i.ty of nickel present can be calculated from the amount of its interference in the t.i.tration. A volumetric a.s.say for nickel is based on this. It has the disadvantage of all indirect t.i.trations in that it requires two standard solutions. On the other hand it gives good results even under unfavourable conditions, and is applicable in the presence of much zinc. Small quant.i.ties of cobalt will count as so much nickel, but larger quant.i.ties make the a.s.say unworkable. Some of the other metals--lead for example--have no appreciable effect; but practically the solution demands a preliminary treatment which would result in their removal. Nevertheless it is a very satisfactory method and makes the determination of nickel quick and comparatively easy in most cases.
_The standard solution of silver nitrate_ is made by dissolving 14.48 grams of recrystallised silver nitrate in distilled water and diluting to 1 litre: 100 c.c. of this solution are equivalent to 0.25 gram of nickel.[73]
_The standard solution of pota.s.sium cyanide_ should be made so as to be exactly equal to the silver nitrate solution. This can be done as follows: Weigh up 12 grams of good pota.s.sium cyanide (95 per cent.), dissolve in water, add 50 c.c. of a 10 per cent. solution of sodium hydrate and dilute to 1 litre. Fill one burette with this and another with the solution of silver nitrate. Run 50 c.c. of the cyanide into a flask; add a few drops of pota.s.sium iodide solution and t.i.trate with the standard silver nitrate until there is a distinct permanent yellowish turbidity. The t.i.tration is more fully described under _Cyanide_, p.
165. The cyanide solution will be found rather stronger than the silver nitrate; dilute it so as to get the two solutions of equal value. For example, 51.3 c.c. of silver nitrate may have been required: then add 1.3 c.c. of water to each 50 c.c. of the cyanide solution remaining. If the full 950 c.c. are available, then add to them 24.7 c.c. of water.
After mixing, take another 50 c.c. and t.i.trate with the silver nitrate; the two solutions should now be exactly equal. The cyanide solution, being strongly alkaline with soda, keeps very well; but its strength should be checked from time to time by t.i.trating with silver nitrate; should there be any slight inequality in the strengths of the two solutions it is easily allowed for in the calculations.