Hydrocyanic Acid Determination.-This of course need only be looked for in solutions containing no protective alkalinity.

First determine free cyanide. Then take another portion, add a little solution of potassium or sodium bicarbonate free from monocarbonate, and titrate again for free cyanide. (Free carbonic acid must be absent.) The difference in the reading between the two titrations gives the equivalent in KCN of the hydrocyanic acid present (Bettel).1

Method for Detecting Traces of Cyanide in Water.2-Evaporate 500 cc of the solution with 3 or 4 drops of ammonium sulphide. Bring to dryness on the water bath and take up with a small quantity of water or water and alcohol. Filter, and add a drop of ferric chloride solution. A red coloration indicates the presence of cyanide. By careful evaporation 1 part in 100,000 can be detected. The author states that by evaporating solutions containing known quantities of cyanide in this manner he has found it possible to obtain a rough colorimetric estimation of the amount of cyanide present.

TABLE 1.-SHOWS VARIOUS STANDARD STRENGTHS OF SILVER NITRATE AND VARYING FACTORS

1 Proceedings Chem., Met. and Min. Soc. of South America, Vol. I, page 165. 2 Gerard W. Williams, Journal Chem., Metall. and Min. Soc. of S. A., May, 1904.

Protective Alkali Determination. This term is usually defined as "the alkaline hydrates and half the monocarbonates," whose action is to protect the cyanide from decomposition, by acids developed in the ore and by atmospheric carbon dioxide.

Either oxalic acid or a mineral acid may be used as a standard solution. Objections have been raised to the use of oxalic on the ground that it is unstable in weak solutions. This is no doubt true where standard solutions are to be kept in stock for an indefinite time, and where extreme accuracy is required, but it has been proved by the writer that decinormal oxalic acid may be kept in a bottle 34 full of air in a tropical climate for at least 4 weeks without showing any appreciable deterioration when checked against carefully standardized decinormal alkali. The advantage of oxalic is that if a good brand be used and the bottle of crystals be kept well stoppered, it may be weighed up, dissolved, and used without standardization. Of course, theoretical accuracy is not attained without standardization on account of uncertain hydration of the crystals, but with the precautions mentioned such a solution is perfectly adequate for all purposes of alkali control in plant solutions.

The decinormal solution is made by dissolving 6.3 grams of the oxalic acid crystals in distilled water and making up to a litre. Decinormal solution of nitric or some other mineral acid may be used if preferred, but it is necessary that this should be standardized against a standard alkali. For this purpose Sutton recommends sodium carbonate as being the most useful. To make the decinormal solution take 10 or 12 grams of pure sodium carbonate, ignite gently, and cool under the desiccator. Then quickly weigh up 5.3 grams of this and dissolve in hot distilled water. When cool put into a litre flask and fill up to the mark with distilled water.

A decinormal nitric acid solution contains 6.3 grams of HNO3 per litre. If the acid is about 1.4 sp. gr. put 6.5 cc of colorless acid into a litre flask and fill up to the mark with distilled water. A given volume of the decinormal soda solution is taken (say 50 cc), and a few drops of methyl orange (not phenolphthalein)

indicator added. The nitric acid solution is then run in from a burette until the color of the indicator shows that the solution is neutral. If the nitric acid solution is decinormal it should take exactly 50 cc of it to neutralize the soda. If it needs more or less than 50 cc it is incorrect, and the strength may be rectified by adding water or acid, as the case may be, after calculating the amount necessary. It should then be verified by another titration.

As an indicator for determination of protective alkali phenolphthalein is generally used, as it gives a value for monocarbonates corresponding with the definition of protective alkali already mentioned. Moreover, methyl orange cannot be used as indicator with oxalic acid. Dissolve about 0.5 gram of phenolphthalein in a little alcohol and add water up to 100 cc. a milkiness form on adding water, more alcohol is needed. Two methods are in general use for determining protective alkali, (1) Clennell's, and (2) Green's.

Should

Clennell's Method for Protective Alkali.1-This is carried out on the same portion of solution that was used for the free cyanide test, method No. 2. After the necessary amount of silver nitrate has been run in to determine free cyanide2 a few drops of the phenolphthalein indicator are added, giving a rose pink color to the solution if free alkali is present; it is then titrated with the decinormal acid until the pink color is dispelled. Taking 25 cc of original cyanide solution, the number of cc of 0.016 = % free alkali in terms of NaOH, or × 0.0112 terms of CaO. If 10 cc of original cyanide solution was taken,

N
10

N
10

acid used X

=

% in

multiply the number of cc of acid used by 0.04 to get percentage

of alkalinity in terms of NaOH or by 0.028 to get percentage of alkalinity in terms of CaO.

H2C2O4. 2H2O + 2NaOH = Na2C2O4 + 4H2O

1 Chemistry of Cyanide Solutions, page 63 (Second Edition).

2 Clennell's statement that “addition of even a considerable quantity of silver in excess of the necessary amount does not materially affect the result" is not true of solutions containing zinc.

Atomic weights: 126 oxalic = 80 sodium hydroxide.

Decinormal oxalic acid = 6.3 grams per litre, equivalent

N

If 25 cc of mill solution be taken for the test then 1 cc

acid

10

N

=0.004 gm. NaOH = 0.016%, therefore No. of cc = % NaOH.

N

Special standard acid is frequently used in preference to 10° (a) To Read in Terms of NaOH.-If 3.937 grams of oxalic acid crystals be dissolved and made up to 1 litre with water, then, taking 25 cc of original cyanide solution the number of cc of standard acid used X 0.01 % free alkali in terms of NaOH, or X 0.1 its equivalent in kilos of NaOH per metric ton or X 0.2 = its equivalent in lb. of NaOH per ton of 2000 lb.

=

If 10 cc of original cyanide solution be taken, the special solution is made by dissolving 1.575 gm. of oxalic acid in water and making up to a litre In this case 1 cc of acid 0.01% NaOH.

=

(b) To Read in Terms of Ca0.-If 5.625 grams of oxalic acid per litre be used, then the number of cc X 0.01% of free alkali in terms of CaO, or X 0.1 = its equivalent in kilos per metric ton or X 0.2 its equivalent in lb. of CaO per ton of 2000 lb.

=

If 10 cc of original cyanide solution be taken the special solution is made by dissolving 2.25 gm. oxalic acid per litre, 1 cc of which 0.01% CaO.

=

This method is perfectly satisfactory in solutions containing zinc, provided that the preliminary titration for free cyanide given under method No. 2 has been carefully performed and addition of silver has been stopped before more than the merest trace of zinc has been precipitated. Green has shown that as soon as sufficient silver has been added to throw down a precipitate of zinc there is a formation of acid zinc nitrate which destroys the reliability of the test for alkali.

Green's Method for Protective Alkali.1-In order to avoid the risk just mentioned Green devised the following method.

First determine total cyanide, as described, recording the number of cc of silver nitrate used. Take a new 25 cc of the solution to be tested and add an excess of potassium ferrocyanide (usually 5 to 10 drops of 10% solution will suffice), then run in from the burette the number of cc of silver nitrate recorded in the preliminary test, and a few drops over. Then add phenolphthalein indicator, and titrate with standard or decinormal acid. Calculate protective alkali as in Clennell's method.

TABLE 2.-SHOWS VARIOUS STANDARD STRENGTHS OF OXALIC ACID SO LUTION, AND VARYING FACTORS

Available Alkalinity of Lime.2-The sample as delivered for analysis is contained in an air-tight vessel, having been passed

1 Proceedings I. M. M. (London), Oct., 1901.

2 Abstract from report of South African Engineering Standards' Committee. Mining and Scientific Press, May 3, 1913.