shall correspond to a rate of flow of about 30 c.c. per minute. This rate can be judged from the information given in the following table:

Inside diameter of inlet tube Size of bubble when

Bubbles

c.c. gas

After air has been drawn through for fifteen minutes, turn off the suction, disconnect the tubes G and H, closing their stopcocks at the same time. Wipe the tubes G and H carefully and weigh them one at a time, taking all the necessary precautions with respect to the charge on the tubes due to the wiping (see §79). Reconnect the tubes G and H in position and draw air through the system for thirty minutes, and again determine the weights of G and H; these weights should not differ individually by more than 0.5 mg. from the preceding; if they do, the blank is unsatisfactory and tubes must be again reconnected and the aspiration repeated.

355. Running the Unknown. When a satisfactory blank has been obtained, connect the tubes G and H in their proper positions, and into the flask C transfer a weighed sample of the powdered limestone (about 1 gram), and add sufficient water to wet it. Open the stop-cocks of G and H, shut off the water pump, and close the stop-cock of the dropping funnel. Disconnect the guard tube A, and place in the dropping funnel 50 c.c. of 6 molar hydrochloric acid. Put the guard tube A back in place and gradually open the stop-cock of the dropping funnel so as to allow the hydrochloric acid to fall drop by drop upon the limestone (all the stop-cocks of the absorbing train must be turned in the "running position"). If the carbon dioxide is being evolved too violently, cut down the rate at which the acid is being added, otherwise some of the carbon dioxide may escape being absorbed by the soda lime tubes. When all but a few c.c. of the acid has been added, close the stop-cock of the dropping funnel and gradually heat the solution in the flask C to boiling and keep it at this temperature for two or three min

Measured at 760 mm. Hg pressure and a temperature of 25°.

utes. The suction pump is now started, the flame removed from under the flask C and the stop-cock of the dropping funnel opened. Air is sucked through the apparatus at the rate of 30 c.c. per minute for thirty minutes; at the end of this time, the suction pump is stopped, all the glass stoppers shut off, and the tubes G and H disconnected. If the tubes G and H are still warm, allow them to cool to room temperature after wiping them free from any adhering dirt. Before weighing them open the stopcocks momentarily to equalize the inside pressure with that of the atmosphere. The gain in weight is carbon dioxide. From this weight calculate the percentage of carbon dioxide in the sample. 356. Remarks. The ionic equilibria involved in the reaction may be written as follows:

Solid CaCO3

CaCO3 diss'd⇒ Ca++ + CO2

The addition of H+ removes carbonate ion to form first the slightly dissociated bicarbonate ion and finally molecular carbonic acid which is itself slightly ionized.

The forward action is further assisted by the fact that carbonic acid is unstable and breaks down into water and carbon dioxide. The carbon dioxide, as soon as it exceeds its solubility in water, escapes as gaseous carbon dioxide. At the beginning of the determination, before the addition of acid, we may consider the solution saturated with respect to calcium carbonate. Then we have

CCa++ X Cco =S.P. CaCO3.

On the addition of H+ the carbonate ion is removed as indicated above, and the ion product above tends to fall below S.P.Cacor But with the solid phase present this cannot be, because more calcium carbonate will go into solution tending to keep the product

As the H+ constantly removes carbonate ion and suppresses at the same time the solid calcium carbonate must gradually pass into solution until it has completely dissolved. The carbo

nate ion concentration at the end of the reaction is extremely minute, due to the presence of such a high concentration of hydrogen ion. This minute concentration is reduced to zero upon boiling the solution which removes the carbon dioxide dissolved in the water, driving the following reaction to completion: diss'd

CO2+H+CH2CO2H2O + CO2⇒ CO2 gas.

The absorption tubes become hot during the determination, due to the heat of reaction between carbon dioxide and the soda lime which is a mixture of sodium hydroxide and calcium hydroxide. The water which is formed in the reaction plus that aspirated away from the soda lime is retained in the tube by the calcium chloride placed over the soda lime.

357. — The same apparatus and method may be used for the determination of available carbon dioxide in baking powders, except that the dropping funnel should be filled with water instead of with hydrochloric acid.

358. Exercise No. 35. Determination of Total Carbon in Steel by Direct Combustion. In steel carbon exists in combination with the iron in the form of different carbides, of which Cementite is one of the most important. In cast iron the carbon, in addition to being in combination with the iron, also exists as graphitic carbon. When steel or cast iron is dissolved in acid, more or less of the combined carbon is liberated as a volatile hydrocarbon and thereby lost. To determine the total carbon, which is a matter of great technical importance, the sample of steel must either be burned directly in a rapid stream of oxygen and the resulting carbon dioxide collected, or less preferably the sample may be dissolved in a solution which contains per liter 1.7 moles potassium chloride, 1.7 moles cupric chloride and 0.1 mole hydrochloric acid, as this solution dissolves the iron and leaves the carbon, which is then filtered off, washed, ignited to carbon dioxide and the carbon dioxide collected and weighed. The direct combustion method is applicable to all kinds of steels, namely, plain carbon steels and special alloy steels, while the solution method is applicable only to plain carbon steels. Because of its general applicability we shall describe the first method

in detail, while for the details of the second method we will refer the reader to the following works: Blair, loc. cit., § 13, p. 156; Griffin, ibid. p. 130; Johnson, ibid. p. 246.

359. — The percentages of carbon that may be expected in various kinds of iron and steel products are as follows:

360. The direct combustion of plain carbon steels requires a temperature of at least 1000°. For alloy steels such as those containing chromium, molybdenum, tungsten, vanadium, etc., the temperature must be higher.3 The greater the percentage of these elements the more difficult is it to burn the carbon completely. In such cases where a sufficiently high temperature cannot be obtained for these special alloy steels, the sample is mixed with a fluxing material like red lead. In the presence of this substance the combustion can be made complete at a temperature of 1100°. If red lead is used, however, a blank must be run on it when mixed with a steel of known carbon content, because red lead is rarely free from carbon; for example, a carbon determination on a "C. P." sample of red lead, showed for a 4-gram sample as much as 0.0042 gram of carbon dioxide.

361. The state of subdivision of the sample itself is important. The drillings should be fine and only those which pass a 20 mesh screen should be used. Any thick pieces should be discarded because there is danger of incomplete combustion during the short time ordinarily required for the complete decarbonization of the sample. The sample must of course be free of lint and oil.

362. The actual determination requires about twenty-five to thirty minutes. A two-gram sample will be decarbonized completely in three minutes at a temperature of 1000°. In 12 to 15 minutes more the tube in which the combustion has

J. R. Cain and H. E. Cleaves, J. Wash. Acad. Sci. 4, 393 (1914), state that for chromium, tungsten, and titanium steels a temperature of 1500° and a time of thirty minutes is necessary to oxidize the carbon by direct combustion.

taken place can be freed completely of carbon dioxide. The balance of the time is taken up in weighing.

363. Apparatus and Assembly. The apparatus and its assembly are shown in Fig. 49. A is a steel tank which contains oxygen under pressure, and which is provided with a reducing valve, a high-pressure gauge, a low-pressure gauge, and a needle valve. B is an electric resistance furnace which is used as a preheater to remove any hydrocarbons that may be present in the oxygen by oxidizing them to carbon dioxide and water. It should be capable of giving a temperature of 500°-600° and should be provided with a suitable rheostat R. The tube inserted in the pre-heater should be of pyrex glass about 36 cm. long with an inside diameter of 1.5 cm. The edges, if rough, should be fire-polished in the blast flame. The copper oxide spiral which is to go into this tube should be about 12 cm. long, and may be easily made by tightly winding 40 mesh copper gauze about a stout copper wire and folding one end of the wire up about the edge of the gauze and the other into a small loop close to the gauze. This roll is inserted in the middle of the Pyrex tube and is oxidized when the pre-heater is heated. In case it is certain that oxygen which is being used for combustion purposes is free of hydrocarbons, the pre-heater may be omitted from the combustion train. C is a U-tube with limbs about 12 cm. long by 2 cm. outside diameter. It is filled about three-quarters with soda lime (12 mesh) containing 15% water, and one-quarter with calcium chloride as described in § 352. The purpose of C is to absorb any carbon dioxide resulting from the oxidation of hydrocarbons in the pre-heater. If the pre-heater is omitted from the train, C is still retained. D is an electric furnace capable of giving a sustained temperature in the neighborhood of 1100° within the silica tube in which the steel is burned. It should be about 25 to 30 cm. long and should have an opening about 4 cm. in diameter in order to accommodate without difficulty a silica tube having an outside diameter of 3 cm. A rheostat R' should be in series with the furnace in order to regulate the current and

4 The student is referred to the catalogue issued by the Hoskins Manufacturing Co., Detroit, Mich., for a detailed description of their electric combustion furnaces which are suitable for this purpose.