METHOD FOR DETERMINING SMALL QUANTITIES OF CARBONATES. 81 X -A Method for Detem&ing Small Quantities of Carbonates. By ALFRED DANIEL HALL and EDWARD JOHN RUSSELL, THE determination of small quantities of carbonates in material like soil is attended with many difficulties, owing t o the solubility of the carbon dioxide in the acid used for decomposing the carbonate. When the soil contains 2 per cent. or more of carbonates, calculated as calcium carbonate, Scheibler’s apparatus may be used, and the empirical correction for solution of the carbon dioxide (Warington, Chern. News, 1875, 31, 253) will not introduce a greater experimental error than attaches to the natural variation of the sample for analysis. But with small proportions of calcium carbonate, 0.5 per cent. and VOL.LXXXI. G82 HALL AND RUSSELL: A METHOD FOR below, Scheibler’s apparatus becomes unworkable, for all the gas pro- duced remains in the reacting acid. Gravimetric methods, where the carbon dioxide is either weighed directly or by difference, require very refined manipulation when 100 grams of soil have to be attacked by acid and the mixture boiled, &c., to obtain quantities like 0.1 gram of carbon dioxide. Working in a vacuum by ordinary gas analysis methods, the carbon dioxide can be collected and measured, but there are, again, difficulties due to solution which render the process tedious and susceptible of error. The suggestion has often been made that the soil should be treated with a known volume of standard acid, and the amount of calcium carbonate present calculated from the acid neutralised.This process, however, always gives results which are too high, owing to the fact that various humates, silicates, and in some cases ferric oxide, are also attacked by the acid without liberating any acid which affects the indicator. Stutzer and Hartleb (Zeit. angew. Chem., 1899, 12, 448) have pro- posed to distil the soil with a solution of ammonium chloride; the calcium carbonate present forms ammonium carbonate by double de- composition ; this dissociates, and the ammonia is caught by standard acid and titrated. This method is open to all the sources of error indicated above (compare Schutte, Zeit. angew. Chem., 1899, 14, 854 ; Woy, Chem. Centr., 1899, ii, 847 ; and ImmendorB, Zeit. angew. Chem., 1900, 15, 1177).I n searching for a more workable process, the authors have devised the apparatus described below, by means of which the main source of error in determinations of carbon dioxide, its solubility, is eliminated. The process is also reasonably rapid and requires no special skill in manipulation. ( A ) is the reaction bulb, about 60 C.C. in capacity. It is connected from below with tho small funnel (B), carrying the stopcock (a). ( A ) is connected to the rest of the apparatus by a cup joint at ( b ) . (C) is a second bulb, rather smaller than ( A ) (in the apparatus actually used, its capacity was 42.5 c.c.); on the tube connecting (C) to the rest of the apparatus is a stopcock (c). The stopcocks and cup joint must be well ground and lubricated so as to maintain a vacuum.( D ) is a capillary tube 800 mm. long, dipping into a small reservoir of mercury and serving as a manometer; a third stopcock (d) is placed between the manometer and the pump. The bulbs ( A ) and (C) can be enclosed in a water-bath. Before t h e apparatus is fixed on the stand, the capacity of the bulb (C) must be ascertained with accuracy ; this may be done by filling the bulb with mercury and then weighing the mercury when shaken out and collected. The apparatus is figured on p. 83.DETERMINING SMALL QUANTITIES OF CARRONATES. 83 Two to 10 grams of the substance in a finely powdered state are introduced into ( A ) and covered with water, the cup joint is wiped, well lubricated, and ( A ) then joined to the rest of the apparatus. The cup joint is sealed with a little mercury, a little also being poured into the funnel (B), so that the bore of the tap is quitefilled.The stop- cock ( c ) is opened, and connection made to a good pump until approxi- mately a vacuum is established inside the apparatus. Entire freedom from air is not necessary, but when determining very small quantities of carbon dioxide (1 to 5 C.C. from 10 grams of soil), the pressure should be run down until the manometer indicates little more than the P 6- vapour pressure of water within the apparatus. When dealing with larger quantities of gas, for example, 10 to 20 c.c., a mercury pump is not necessary, it is sufficient to use a good water pump or hand air pump that will establish an internal pressure of 50-60 mm. of mercury. The stopcock (d) is closed, the height of the mercury in (D) and the temperature of the water-bath are read ; this reading = R.Stopcock ( c ) is then closed, a well-boiled and cooled mixture of equal volumes of sulphuric acid and water is placed in the funnel (B), and a few C.C. introduced into the reacting bulb. Since it enters from below, the liquid and soil get well stirred up ; the mixture is left For a few minutes G 284 HALL AND RUSSELT,: A METHOD FOR to cool down t o the temperature of the bath, then the apparatus is shaken to expel the carbon dioxide present in excess in the solution, and allowed to stand, with further occasional shaking, until the gauge shows a constant reading. The gas evolved causes an increase of pressure inside the apparatus, and the manometer column is read agnin=R1.Communication is now made with the bulb (C) by opening the stopcock (c), the gas ex- pands again into (C) and the mercury rises again in (D). A little time and shaking cause the gas dissolved in the liquid i n ( A ) t o come into equilibrium with the gas above at the new pressure; the mano- meter column is then read when constant, and = B,. Assuming the temperature, t, of the water-bath has remained constant, and calling d the difference in mm. between R and R,, d' the difference in mm. between R and R,, and C the volume of C.d.d' (d - d')760 the bulb (C), then the volume of gas evolved a t N.T.P. = .* The operation amounts t o finding ac unknown volume of 273 2 m gas in ( A ) by the change in pressure produced when it expands by a known volume.The advantage of the method lies in the fact that the volume of soil, liquid, &c., which may have been introduced into ( A ) is immaterial, and does not appear in the calculation, and especially * The complete proof of the formula given is as follows : Let x = the p. v. of the carbon dioxide evolved at the given temperature. A = the volume of the apparatus excluding (C) and the liquid in ( A ) . C = the volume of (C). P = barometer reading. n = tension of aqueous vapour at the given temperature, k = the volume of carbon dioxide soluble in the liquid in ( A ) a t unit R, R, and R, = the readings as above. pressure, At starting, the apparatus contains some air = ( A +B)(P - R - a). ( C ) is shut off and x of carbon dioxide evolved. Then : z+A(P-R-a) = A ( P - R , - a ) + k ( R - R , ) X = A + k . .. . . . [l] €2-€2, The gas is then allowed to expand into (C), when x + ( A + C ) ( P - R- a ) = ( A + C)(P -3, -n) + k ( R - XJ x. = A + C + k . . . . . . [2] A-8, Combining [I] and [2] x - C+Z - - ~ R-R, R-R,DETERMINING SMALT, QUANTITIES OF CARBONATES. 85 that the effect of the gas remaining dissolved in the liquid in ( A ) is also eliminated. The liquid is saturated by the gas, so that the gas within and without the liquid is in equilibrium. When the volume is increased by opening the stopcock to ( C ) , an amount of gas, propor- tional to the reduction in pressure, escapes from the liquid. I n brief, the gas contained in the liquid of ( A ) obeys the same laws as the gas above the liquid, and the liquid becomes practically only a portion of the gas-filled space of ( A ) .It is necessary to have some solid particles like soil or glass beads in ( A ) , otherwise the liquid becomes, and remains, obstinately super- saturated with carbon dioxide, nor can the excess be shaken out. This tendency to supersaturation forms the chief difficulty in working with the apparatus; the amount of substance taken should be such that the pressure of the carbon dioxide liberated does not exceed 100 or 150 mm., or the time required t o obtain equilibrium becomes very great, and may even amount to 2 or 3 hours. The lower the pressure, or, in other words, the smaller the amount of carbonate present, the easier the determination is t o carry out ; the limit is fixed only by the accuracy with which the gauge can be read. Appended are a few numbers obtained with the apparatus i n the case of pure sodium carbonate and Iceland spar, the bulb ( A ) being half filled with glass beads : Number. Substance taken. 0'000624 gram Na,CO, 0.00125 ,) ,, 0 0025 ,, ,> 0.005 2 , 7 , 0.010 ,) ,> 0.020 ,, 9 , 0.0503 ) ) CaCO, CO, a t N.T.1'. (calc.). 0.13 C.C. 0'26 ,) 0'53 ,, 1.06 ,, 2.11 ,, 4'22 ), 11.26 ,) :O, a t N.T. 1'. (foulid). 0'15 C.C. 0'26 ,, 0'51 ,, 4'38 ,, 1'00 ), 2.00 ), 11'18 ,, The apparatus may be conveniently applied to any reaction involv- ing the measnrement of a gas evolved from a liquid. SOUTH E-4STElt.N AGRICULTURAL COLLEGE, WYE.