1 day 5.0 " 24.0 " 27.6 "

2 days 3.0 " 14.5 " 21.3 "

These indicate very clearly the increased susceptibility to oxidation in strongly acid solutions.

_A standard solution of ferric chloride_ is prepared in the same manner as that described under the experiments on the methods of reduction; but it should be of twice the strength, so that 100 c.c. may contain 1 gram of iron. This solution is used for standardising the stannous chloride when required; and must be carefully prepared; and tested for the presence of nitric acid.

The t.i.tration is more limited in its application than either of the oxidising processes because of the restrictions as to bulk, quality and quant.i.ty of free acid present, and other conditions of the solution to be a.s.sayed. The following experiments show the conditions necessary for a successful t.i.tration.

~Effect of Varying Temperature.~--Twenty c.c. of ferric chloride solution with 20 c.c. of strong hydrochloric acid, diluted to 50 c.c., gave the following results when t.i.trated:--

Temperature 15 30 70 100 Stannous chloride required 22.8 c.c. 22.0 c.c. 22.1 c.c. 22.0 c.c.

The finishing point, however, is more distinct the hotter the solution; so that it is best in all cases to run the standard into the boiling solution.

~Effect of Varying Bulk.~--Solutions containing the same quant.i.ty of iron and acid as the last, but diluted to various bulks, and t.i.trated while boiling, gave the following results:--

Bulk 30 c.c. 100 c.c. 500 c.c.

Stannous chloride required 21.5 " 21.7 " 24.3 "

~Effect of Varying Quant.i.ties of Hydrochloric Acid.~--In these experiments the bulk before t.i.tration was 50 c.c. except in the last, in which it was 70 c.c. With less than 5 c.c. of strong hydrochloric acid the finishing point is indistinct and prolonged.

Strong hydrochloric acid present 5 c.c. 10 c.c. 20 c.c. 30 c.c. 50 c.c.

Stannous chloride required 21.1 " 21.1 " 21.2 " 21.8 " 22.2 "

~Effect of Free Sulphuric Acid.~--In these experiments 20 c.c. of hydrochloric acid were present, and the bulk was 50 c.c.

Strong sulphuric acid present -- c.c. 3 c.c. 5 c.c. 10 c.c.

Stannous chloride required 21.6 " 22.3 " 22.9 " 23.1 "

This interference of strong sulphuric acid may be completely counteracted by somewhat modifying the mode of working. Another experiment, like the last of this series, required 21.6 c.c.

~Effect of Foreign Salts.~--Experiments in which 10 grams of various salts were added showed them to be without effect. The results were as follows:--

Salt present -- AmCl Am_{2}SO_{4} MgCl_{2} Stannous chloride required 21.6 c.c. 21.6 c.c. 21.6 c.c. 21.6 c.c.

Salt present CaCl_{2} FeCl_{2} Al_{2}Cl_{6} Stannous chloride required 21.8 c.c. 21.6 c.c. 21.6 c.c.

~Effect of Varying Iron.~--t.i.trating a solution (with 20 c.c. of hydrochloric acid) measuring 50 c.c., and kept boiling, the quant.i.ty of stannous chloride solution required is practically proportional to the iron present.

Ferric chloride added 1 c.c. 10 c.c. 20 c.c. 50 c.c. 100 c.c.

Stannous chloride required 1.1 " 10.5 " 20.6 " 51.4 " 102.6 "

The student, having practised some of the above experiments, may proceed to the a.s.say of an iron ore.

~Determination of Iron in Brown Iron Ore.~--Weigh up 1 gram of the dried and powdered ore, calcine in the cover of a platinum crucible, and dissolve up in an evaporating dish[69] with 20 c.c. of strong hydrochloric acid. When solution is complete, dilute to 50 c.c. after replacing any acid that may have been evaporated. Boil, and run in the stannous chloride solution until the colour is faintly yellow; boil again, and continue the addition of the stannous chloride solution, stirring continuously until the solution appears colourless. Note the quant.i.ty of the stannous chloride solution required. Suppose this to be 59 c.c. Take 60 c.c. of the standard ferric chloride solution, add 20 c.c. of hydrochloric acid, boil and t.i.trate in the same way as before.

Suppose this to require 61 c.c. Then as 61 is equivalent to 60 of the iron solution, 59 is equivalent to 58.13.[70] This gives the percentage.

It is not necessary to standardise the stannous chloride solution in this way with each sample a.s.sayed, the ratio 61: 60 would serve for a whole batch of samples; but the standardising should be repeated at least once each day.

COLORIMETRIC METHOD.

This method is valuable for the determination of small quant.i.ties of iron present as impurities in other metals or ores. It is based on the red coloration developed by the action of pota.s.sic sulphocyanate on acid solutions of ferric salts.

_Standard Ferric Chloride Solution._--Take 1 c.c. of the ferric chloride solution used for standardising the stannous chloride solution, add 2 c.c. of dilute hydrochloric acid, and dilute to 1 litre with water. 1 c.c. = 0.01 milligram.

_Solution of Pota.s.sic Sulphocyanate._--Dissolve 60 grams of the salt in water, and dilute to a litre. It should be colourless. Use 10 c.c. for each test.

The quant.i.ty of the substance to be weighed for the a.s.say should not contain more than a milligram of iron; consequently, if the ore contain more than 0.1 per cent. of that metal, less than a gram of it must be taken.

The method is as follows:--Weigh up 1 gram of the substance and dissolve in a suitable acid; dilute; and add permanganate of potash solution until tinted. Boil for some time and dilute to 100 c.c. Take a couple of Nessler tubes, holding over 100 c.c., but marked at 50 c.c.; label them "1" and "2"; and into each put 10 c.c. of the pota.s.sic sulphocyanate solution and 2 c.c. of dilute hydrochloric acid. The solutions should be colourless. To "1" add 10 c.c. of the a.s.say solution, and dilute to the 50 c.c. mark. To the other add water, but only to within 5 or 10 c.c. of this mark. Now run in the standard ferric chloride solution from a small burette, 1 c.c. at a time, stirring after each addition till the colour is nearly equal to that of the a.s.say (No. 1). At this stage bring the solution to the same level by diluting, and make a further addition of the standard ferric chloride solution till the colours correspond. The amount of iron will be the same in each tube; that in the standard may be known by reading off the volume from the burette and multiplying by 0.01 milligram.

If the 10 c.c. of the a.s.say solution gave a colour requiring more than 5 or 6 c.c. of the standard ferric chloride solution, repeat the determination, taking a smaller proportion.

The effect of varying conditions on the a.s.say will be seen from the following experiments:--

~Effect of Varying Temperature.~--The effect of increase of temperature is to lessen the colour; in fact, by boiling, the colour can be entirely removed. All a.s.says are best carried out in the cold.

1 c.c. at 15 would only show the colour of 0.75 c.c. at 45 2 " " " " 1.75 "

5 " " " " 4.0 "

~Effect of Time.~--The effect of increase of time is to increase the colour, as will be seen from the following experiments:--

2 c.c. on standing 10 minutes became equal to 2.25 c.c.

2 " 20 " " " 2.75 "

2 " 40 " " " 3.00 "

~Effect of Free Acid.~--If no acid at all be present, the sulphocyanate of pota.s.sium solution removes the colour it first produces, so that a certain amount of acid is necessary to develop the colour. The use of a large excess has a tendency to increase the colour produced.

5 c.c. nitric acid (sp. g. 1.4) read 3.7 c.c. instead of 2 c.c. with the dilute acid.

5 c.c. sulphuric acid (sp. g. 1.32) read 2.2 c.c. instead of 2 c.c. with the dilute acid.

5 c.c. hydrochloric acid (sp. g. 1.16) read 2.5 c.c. instead of 2 c.c.

with the dilute acid.

~Effect of Foreign Metals.~--Lead, mercury, cadmium, bis.m.u.th, a.r.s.enic, tin, antimony, nickel, cobalt, manganese, aluminium, zinc, strontium, barium, calcium, magnesium, sodium, or pota.s.sium, when separately present in quant.i.ties of from 100 to 200 times the weight of iron present, do not interfere if they have previously been brought to their highest oxidised condition by boiling with nitric acid or by treating with permanganate. a.r.s.enic and phosphoric acids interfere unless an excess of free hydrochloric or other acid is present. Oxalic acid (but not tartaric acid) in minute quant.i.ties destroys the colour. Nitrous acid strikes a red colour with the sulphocyanate of pota.s.sium; consequently, when nitric acid has been used in excess, high results may be obtained. Copper and some other metals interfere, so that in most cases it is advisable to concentrate the iron before estimating it. A blank experiment should always be made with the reagents used in order to determine the iron, if any, introduced during the solution, &c., of the substance a.s.sayed.

~Determination of Iron in Metallic Copper.~--This may be most conveniently done during the estimation of the a.r.s.enic. The small quant.i.ty of white flocculent precipitate which may be observed in the acetic acid solution before t.i.trating, contains the whole of the iron as ferric a.r.s.enate. It should be filtered off, dissolved in 10 c.c. of dilute hydrochloric acid, and diluted to 100 c.c.; 10 c.c. of this may be taken for the estimation. For example: 10 grams of copper were taken, and the iron estimated; 3.0 c.c. of standard ferric chloride solution were used, equivalent to 0.03 milligram of iron; this multiplied by 10 (because only 1/10th of the sample was taken) gives 0.3 milligram as the iron in 10 grams of copper. This equals 0.003 per cent.

In a series of experiments with this method working on 10-gram lots of copper, to which known quant.i.ties of iron had been added, the following were the results:--

Iron present 0.015% 0.070% 0.100% 0.495% Iron found 0.015" 0.061" 0.087" 0.522"

When no a.r.s.enic is present in the copper, the iron can be separated by fractionally precipitating with sodic carbonate, dissolving in ammonia, and filtering off the ferric hydrate. Coppers generally carry more iron the less a.r.s.enic they contain.

~Determination of Iron in Metallic Zinc.~--Dissolve 1 gram of zinc in 10 c.c. of dilute hydrochloric acid, adding a drop or two of nitric acid towards the end to effect complete solution. Boil, dilute, and tint with the permanganate of pota.s.sium solution; boil till colourless, and dilute to 100 c.c. Take 10 c.c. for the determination. Make a blank experiment by boiling 10 c.c. of dilute hydrochloric acid with a drop or two of nitric acid; add a similar quant.i.ty of the permanganate of pota.s.sium solution, boiling, &c., as before. The quant.i.ty of iron in zinc varies from less than 0.005 to more than 2.0 per cent. When 1 gram is taken and worked as above, each c.c. of ferric chloride solution required indicates 0.01 per cent. of iron.

~Determination of Iron in Metallic Tin.~--Cover 1 gram of tin with 5 c.c. of hydrochloric acid, add 1 c.c. of nitric acid, and evaporate to dryness. Take up with 2 c.c. of dilute hydrochloric acid, add 10 c.c. of the pota.s.sic sulphocyanate solution, and make up to 50 c.c. Probably the colour developed will be brown instead of red owing to the presence of copper; in this case, add to the standard as much copper as the a.s.say is known to contain (which must have previously been determined; see _Copper_); the t.i.tration is then carried out in the usual way.