Scientia Sinica

No. 1 HUANG, YANG: ACTIVITY COEFFICIENTS OF NONELECTROLYTES 67

and heated to 45°C. Then it is sealed according to the procedure described above and put into the thermostat at 25°C. It is nrst clamped to the wingplate in a position so that its axis is perpendicular to the rotating axle. This keeps the content in violent stirring. After one hour, the mixture becomes turbid. Then it is clamped so that its axis becomes earallel to the rotating axle and rotated for two hours more. Finally let it stand vertically for two or three hours in the thermostat. The last two steps make the turbid mixture separate again into two clear liquid layers.

Since #-valeric acid forms the upper layer of the content, direct pipetting of the salt solution in the lower layer causes some valeric acid to be carried out on the tip of the pipette and brings in error in analysis. To avoid this, the following procedure is adopted. Let the solubility bottle stand vertically in the thermostat with its neck above water. The glass stopper is replaced by a cork fitted with a glass tubing, one end of which, already blown into a thin bulb, almost touches the bottom of the solubility bottle. In the near neighbourhood of this bottle, put in a large test tube with its mouth above water. It is closed up by a rubber stopper ‘fitted with two bent glass tubings, one of which is connected with that of the solubility bottle by a short rubber tube. When it is found by close observation that no #-valeric acid adheres to the glass bulb, break it by pressing it against the bottom. By sucking the other bent tubing of the large test tube, the aqueous salt solution in the solubility bottle is transferred, without contact with the #-valeric acid layer, to the large test tube.

20 cc of the salt solution is pipetted out, (the pipette being at the same temperature as the thermostat, 1e. 25°C), and its #-valeric acid content is analyzed by titration with a standard NaOH solution. For every salt solution, two or three titrations are made.

Data anp CALCULATIONS

In Table 1, the first column gives the cobaltammines, the second, their concentrations, C,, in mols per litre; the third, the solubilities of #-valeric acid, S, in grams per litre; the fourth, the activity coefficient of #-valeric acid, f, as defined by S,/S; the fifth, log f; the sixth, 1/f. All these quantities are values at 25°C, S,=37.43 g/litre. According to Lieben and Rossi!”!, it requires 27 cc of water to dissolve 1 cc of n-valeric acid at 16°C. On the basis of this datum, L.C.T. gives S9>=35 g per 1,000 g of solution at 16°C. No datum at 25°C is available for comparison. We also find that for »-valeric acid at 25°C, D?= 0.9979. Since m-valeric acid is a weak electrolyte, our f is not exactly the same as the activity coefficient of real nonelectrolyte which involves no ions. Our f is only an approximate expression, which other authors also used for treatment of solution of weak eletrolytes!”!.