It is often desirable to ascertain by preliminary experiments the quantity of alkali which will be needed to neutralize acidity in any particular ore. In this connection a distinction is sometimes made between “ free ” and “ latent ” acidity.
The free acidity is that due to substances soluble in water, and is determined by agitating a known weight (say half a pound or 200 grams) in a closed bottle with a sufficient quantity of water (say 4 or 5 parts water to 1 of ore) for half an hour or so. After settling, filter off a sufficient portion of the liquid, take an aliquot part (say ¼ or 1/10 of the whole liquid), and determine the acidity by titrating with standard alkali. In some cases this acidity may be determined by direct titration of the filtrate with standard caustic soda and phenol-phthalein indicator, but the results are usually indefinite owing to the precipitation of metallic salts which obscure the end-point. A better method is to add a decided excess of alkali; filter, and titrate the filtrate or an aliquot part of it with standard acid, using one or two drops of a 0.1 per cent, solution of methyl orange as indicator.
The latent acidity may be taken to mean the consumption of alkali due to the combined effect of substances insoluble in water. Thus many ores contain insoluble sulphates of iron, etc., which are decomposed by lime, caustic soda, etc., and thus neutralize alkali.
The total acidity (free and latent together) is determined by agitating a weighed quantity of the substance with an excess of standard caustic soda or clear lime water, using a measured quantity of known strength. After agitation for half an hour, or one hour, the liquid is filtered and a measured portion titrated with standard acid, using methyl orange as indicator. A convenient proportion for this test is two parts liquid to one of ore. The standard alkali solution may be so adjusted that the consumption per ton of ore may be obtained with little or no calculation. Thus, if 200 grams ore be used, it is agitated with 400 cc. of 0.1 per cent. NaOH or CaO, containing, therefore, 0.4 gram of the alkali. Say, for example, that 100 cc. of the filtrate are titrated and that V cc. of standard acid are required, equivalent to V cc. of standard alkali; then the entire 400 cc. of liquid would have required 4 V cc., equivalent to 0.004 V gram alkali remaining unconsumed. The alkali consumption is therefore 0.4 — 0.004 V gram on 200 grams of ore, or 4 (1 — 0.01 V) lb. per ton of 2000 lb.
The latent acidity may be found, if required, by taking the difference between free and total acidity. It should be remembered that the determination must generally be made on a moist sample of material, as drying vitiates the test by changing the conditions. It must therefore generally be accompanied by a moisture determination, made on a separate portion of the material, as described in Section I, and the results calculated to dry weight. It is also necessary to point out that various soluble salts (ferrous sulphate, magnesium sulphate, alum, etc.) which are not in themselves acid to test paper, may nevertheless constitute the whole or part of the acidity when determined as described.
When an approximate idea of the acidity of an ore has been obtained by tests such as those just described, it is often possible to determine more exactly the quantity which would be needed in practice by agitating or percolating a number of equal portions of the ore with water to which different amounts of lime or caustic soda have been added. By this means the minimum alkali necessary to neutralize acidity may be ascertained, the solution drawn off from each test being titrated with standard acid. When lime is used, sufficient time must be allowed for the reaction, owing to its slight solubility in water.