Page:Encyclopædia Britannica, Ninth Edition, v. 5.djvu/559

OEOANIC.] CHEMISTRY 547 so as to expel the last traces of nitrogen. The C0 2 pro duced by the combustion and from the NaHCO 3 being absorbed by the potash solution, the graduated receiver contains only the whole volume of pure nitrogen. The receiver and its contents are accordingly transferred to a vessel of water the mercury and potash solution allowed to be replaced by water, the receiver is raised till the pressure is equalized by the water being at the same level both inside and outside, and the volume of gas is read off, the temperature of the air of the room and the height of the barometer being at the same time noted. If N = weight of nitrogen in grammes, t the tempera ture of the air, b the height of the barometer, e the tension of aqueous vapour at the temperature t, and V the volume of nitrogen in cubic centimetres 0-0012562 (l + 0-00367f)760 (0-0012562 being wt. of 1 c.c. of N" at C. and 760 mm. bar.). Determination of the Halogen Elements, Sulphur, and Phosphorus. The halogens are sometimes determined as silver salts, by burning a known weight of the substance with pure quicklime in a combustion tube, dissolving in dilute nitric acid, and adding silver nitrate. Sulphur and phosphorus may be determined by fusing a known quantity of the substance with a mixture of potassium hydroxide and nitrate in a silver dish. The sulphur is by this means oxidized to sulphuric and the phosphorus to phosphoric acid, and, on dissolving the fused mass and acidulating, these acids can be estimated. By the method of Carius the halogens, sulphur, and phosphorus can be determined, if necessary, in one opera tion. A known weight of the substance is sealed up in a strong glass tube with about 20 times its weight of nitric acid ,(sp. gr. 1 4), and the tube then heated for some hours in an oil-bath to a temperature of 140 - 300 C, The sub stance is completely oxidized by this operation, the sulphur and phosphorus being converted into their respective acids, so that their determination then becomes an operation of inorganic analysis. If halogens are present, it is custo mary to add a few crystals of silver nitrate before sealing up the tube. After the operation the haloid silver salt is filtered off, the excess of silver is removed from the fil trate by HC1, and H 2 S0 4 or H 3 P0 4 is determined in the ordinary way. Determination of Oxygen. Although several processes have been devised for the direct estimation of oxygen (by Baumhauer, Maumene, and Mitscherlich), they are seldom employed in laboratories. This element is usually deter mined by " difference," i.e., by adding the percentages of the other elements, and subtracting the result from 100. Formulae and Constitution of Organic Compounds. Empirical Formulae. Having by the above methods of analysis arrived at the percentage composition of a sub stance, the next step is to determine its formula. The empirical formula is obtained by dividing each percentage number by the atomic weight of its respective element. Thus, supposing an analysis of common alcohol gave the following percentage numbers : Carbon .................................................. 52 15 Hydrogen ................................................. 13 06 Oxygen (by difference) ................................ 3479 100-00 Dividing these by the respective atomic weights c _52-15_ 4 . 3 H _ -_- 3,H _ 13 . Q _ 0,0 - 34-79 These numbers show that the atoms of C, H, and are present in numbers having the ratios 2:6:1, since by the atomic theory the atoms of each element must exist in a compound in integral numbers. The differences between the integral and fractional numbers are justly assignable to the unavoidable " experimental errors " of analysis. Thus the simplest formula that can be given to alcohol from the foregoing analysis is C 2 H 6 O, and it is usual to express the results in the following manner : Theory C 2 H 6 0. Found. C 2 =24 = 52-17 C 52-15 H 6 = 6 = 13-04 H 13-06 =16 = 34-79 {by difference) 34 -79 46 100-00 100-00 This imaginary example may serve to show that the determination of empirical formulae cannot be made accord ing to any fixed set of rules. The errors of experiment are seldom so small as in the supposed illustration, and in cases where these are large, and where the substance contains a large number of atoms in its molecule, great difficulty is often experienced. Molecular Formulae. The formulae obtained by the method just described express simply the ratios existing between the numbers of atoms in the molecule of a siib- stance, without regard to the actual number of atoms in such a molecule. Reasoning downwards from the law of Avogadro, which has been explained in the foregoing por tions of this article, it will be seen that the volume of such molecules as do not undergo dissociation when heated is always equal "bo the volume of the molecule (2 parts by weight) of hydrogen. To ascertain molecular formula, therefore, all that is necessary is to determine the vapour- density of the substance as referred to hydrogen. Thus the analysis of benzene, a hydrocarbon obtained from coal tar, leads to the formula CH, but there is no evidence to show whether its molecular formula is CH, C 2 H 2 , C 3 H 3 , C 4 H 4 , C 5 H 5 , or C.H,,. By experiment its vapour density is found to be 39, so that its molecular weight is 78. Dividing this number in the ratio C : H, i.e., 12:1, we obtain 72 : 6 as the actual ratio of the weights of C and H existing in the molecule. Seventy- two parts of C correspond to -|-| = 6 atoms, and 6 parts by weight of hydrogen correspond to |- = 6 atoms, so that the molecular formula of benzene is C 6 H 6 . Again, with respect to alcohol. The vapour-density (H = l) is 23, so that its molecular weight is 46. This number, however, agrees with the molecular weight of a substance having the formula C 2 H 6 0, since 12x2 + 6 + 16 = 46 ; hence this formula must be assigned to alcohol, and we have an illustration of a case in which the empiri cal and molecular formulae are identical. Determination of Vapour-density. In practice, the vapour-density is determined by the methods of Dumas or Hofmann. Dumas s Method. In Dumas s process the weight of a known volume of vapour is ascertained in the following manner (see fig. 8). A globular glass flask, as light as possible, with a neck fused into it, is first provided. The capacity of the globe may vaiy from ^ to J litre, according to the amount of substance to be operated upon. The neck is drawn out in the blowpipe flame to a capillary termi nation of about one millimetre diameter, and then bent up so as to project above the surface of the liquid of the bath in which the globe is to be immersed. The globe is first weighed full of air the temperature and height of the barometer being noted. By warming the globe, and plunging the point of the neck into some of the liquid of which the vapour-density is to be determined, a few grammes of the latter are introduced. The globe and its con tents are then plunged into a bath of water, paraffin, or fusible metal kept at a constant temperature, at least 20 or 30 C above the boiling-point of the substance. As soon as the vapour ceases to rush out of the capillary orifice of the neck, the point is sealed hermetically by the blowpipe flame, the height of the barometer

I and the temperature of the bath being observed.