2788 COLVIN THE IONIC ACTIVITY PRODUCT CCCLXXX1V.-The Ionic Activity Product of Water in Glycerol- Water Mixtures. By JAMES COLVIN. MEASUREMENTS have shown that a very marked increase occurs in the activity of hydrogen ion derived from hydrogen chloride in aqueous solution in the presence of a solvent-displacing agent, generally a non-electrolyte such as sucrose or glycerol (Moran and Lewis J. 1922,121 1613). It is of interest therefore to determine the activities of hydrogen and hydroxyl ions in a solution containing a eubstance such as glycerol with a view to obtain the value of the ionic activity product of water i.e. the product of the hydrogen-and the hydroxyl-ion activities. Should the product remain constant independent of the concentration of glycerol it would permit of the determination of hydroxyl-ion activities by the hydrogen electrode which is in many cases more convenient to use than a direct hydroxyl-ion electrode such as the Donnan-Allmand electrode.Measurements of the activities of hydrogen and hydroxyl ions were therefore carried out in solutions of N/lOO-sodium hydroxide, containing various amounts of glycerol. EX P E R I M E N T A I,. The sodium hydroxide solution was prepared from metallic sodium and water free from carbon dioxide. The glycerol had been previously employed for conductivity measurements. For the measurements of the E.M.P. of the cells a Cambridge-Paul potentiometer reading to 0.1 millivolt was used in conjunction with a mirror galvanometer. To avoid stray currents the potentio-meter and galvanometer were mounted on glass insulated from earthed iron sheets by paraffin wax.The cells were maintained in a thermostat a t a temperature of 25" & 0.05" OB WATER. ULYCERObWATEB &IXTfTRES. 2789 The Activity of Hydrogen Ion in Aqueous Sdutions of Sodaum Hydroxide and GZycerd.-The cell employed wa,s of the following type: pt N/lm-Na*H I S a t . ~ a N-Calomel + H I Glycerol I I -The electrode consisted of platinum foil coated with platinum black and was placed for 10 minutes in boiling water before use, following the procedure of &and (J. 1909 95 2151). The hydrogen gas was prepared by the electrolysis of a solution of cawtic soda and was washed by paming through a wash-bottle containing the solution under investigation. The gaa thus being saturafed with water vapour at the same pressure as the vapour pressure of the solution in the cell no concentration changes could occur in the solution due to evaporation.As some doubts have been expressed as to the trustworthiness and reproducibility of the hydrogen electrode in alkaline solution, it was thought advisable to carry out some preliminary measure-ments on the cell, N/lO-NaOH Sat. KC1 N-Calomel + I I I The following results were obtained : cell ............... 1 2 3 4 5 6 E.M.P. (volts) 1.0412 1.0415 1.0411 1-0412 1.0415 1-0411 The mean value of the E.M.P. 1.0413 volts is in satiafactory agreement with the value 1-0416 volts obtained by Harned ( J . Amer. Chem. SOC. 1916 37 2460). As the reproducibility of the hydrogen electrode was thus satis-factorily demonstrated measuremenfs were made using N/100-sodium hydroxide in presence of various amounfs of glycerol.As the quantity of glycerol in the solution increased a longer time w m required for the electrode to attain equilibrium ; in all cams, however the values could be reproduced. The activities of the hydrogen ion were calculated by means of the formula xH. = 0.282 + 0.059 log,,pX.. The absolute potential of the calomel elec-trode ww taken to be 0.56 volt at 18". The temperature coefficient of the electrode is 040068 volt per degree hence the value at 25" is 066443 volt (compare Chroustchov and Sitnikov Compt. rend., 1889 108 941). The results obtained are in Table I. The glycerol produces a very marked increase in the activity of hydrogen ion in the solution of alkali.6BS 2790 COLVIN ! THE 3ONIC ACTIVITY PRODUCT TABLE I. G. of glycerol E.M.P. in per 100 C.C. volt. - TH'. QH' x 1012. 0-0 0.9870 0-4222 1.16 12.6 0.9661 0.4013 2.62 20-0 0-9580 0.3932 3-60 25.2 0.9530 0.3882 4.40 40.0 0.9406 0.3758 7-06 The Activity of Hydroxyl Ion in Aqueous Solutiom of Sodium Hydroxide and Glycerol.-The electrode employed was that described by Doman and Allmand (J. 1911 99 845) and consists essentially of a layer of mercuric oxide superimposed on a layer of mercury, the electrode vessel then being Med with the solution under investigation. The electrode reaction is given by the equation HgO + H,O + 2E = Hg + 20H', so that the equation for the electrode potential becomes where E' is t.he elect,rode potential in solutions where all the activities are equal to unity.Since the activities of the mercury and solid mercuric oxide are constant they may be included in the k t term on the right-hand side. For dilute solutions the activity of the water remains sensibly constant. Hence the equation for the electrode potential in dilute solutions may be written The value of the term E was determined experimentally by the following method. - Hg HgO I N/lO-KOH I Sat. KC1 i N-Calomel + was found to be 0.1178 volt at 25". This gives the value 0.4470 volt as the electrode potential in X/lO-potassium hydroxide. The activity of hydroxyl ion in the solution can be obtained from the data of Knobel ( J . Amer. Chem. Soc. 1923 45 70) for the mean activity of potassium hydroxide in aqueous solution.On the assumption that the potassium ion has the same activity in solutions of p o w -ium hydroxide and of potassium chloride of the same concentration, by employing the data of Noyes and MacInnes ( J . Amer. Ckm. SOC., 1920,42,239) for the activity of potassium ion in potassium chloride solutions the value 0.0825 is obtained for the activity of hydr-oxyl ion in N/lO-potassium hydroxide. By putting this value in the expression for the electrode potential the value 0.3823 volt is obtained for E,. x O H ~ = E - 0.059 log, aoH. The E.M.F. of the cell OF WATER IN GLYCEROGWATER MIXTURES. 2791 When glycerol is present however it is no longer justisable to The expression for the equate the activity of the water to unity.electrode potential becomes in which Eo has the value found above. Measurements &th the Donnan-Allmand Electrode.-The mercury used in the cells was distilled in a vacuum. The mercuric oxide was prepared by heating the nitrate until no more brown fumes were evolved. The nitrate itself was prepared by acting on re-distilled mercury with nitric acid (d 1-2) and recrystallising from nitric acid. The product thus obtained gave very reproducible results. At firs% a large electrode vessel wfts used but was soon discarded in favour of a smaller type as equilibrium was then more readily attained. xOH 2 Eo + (0*059/2) log,o aHaO - 0.059 log^^ uOH,, The cells set up were of the type In all cases readings of the E.N.F. were taken at intervals of 24 hours from the time of setting up the cell.In general with N/lOO-sodium hydroxide alone present equilibrium was attained within 48 hours. With glycerol present the E.M.F. declined with time hence a series of readings taken at intervals of 24 hours for the first' 3 days were averaged to gke the E.M.F. of the cell. As five cells for each concentration of glycerol were set up and the readings averaged considerable confidence is placed in the results, which are in Table 11. The activity of the water in glycerol-water mixtures may be obtained from the vapour pressure data of Perman and Price (T'runs. Furaduy SOC. 1912 8 $4) at 70". The activity of the water which may be taken as the ratio of the vapour pressure of the solution to the vapour pressure of the solvent is regarded as independent of the temperature (Lewis and Randall, " Thermodynamics," p.349). TBBLE 11. G. of glycerol per 100 C.C. E.M.F. TOH'. UH&. aOH' x 10'. 0 0-0604 0-5044 1-00 0.883 12.6 0-0394 0-5254 0-98 0.373 20-0 0-03 14 0.5334 0.96 0-270 25.2 0.0266 0.5382 0.94 0.221 40.0 0.0147 0.5501 0.89 0.135 The glycerol produces a marked decrease in the activity of the hydroxyl ion 2792 C O L ~ THE IOHIC ACTIVITY PRODUCT OF WATER ETC. The I m i ~ Activity Product of Wuter.4ufficient data have now been obtained to calculate the ionic activity product and also the dissociation constant of water in presence of glycerol. The results are in Table III. TABLE III. G. of glycerol uE* x UOEP x 1014 per 100 C.C. a H - x 10l2. U O H ~ x 10'. am,-,. ap* XUOHJ x 1014.aE10 0 1.16 0.883 1.00 1.02 1.02 12-6 2.62 0.373 0.98 0.98 1-00 20.0 3.60 0.270 0.96 0-97 1.01 25-2 4.40 0.221 0.94 0-97 1-03 40.0 7-06 0-135 0-89 0-95 1.07 The value of the dissociation constant of water i.e. aH.x a,=,/ aHtO remains practically constant over the whole range up to 40% of glycerol the deviations from constancy being very small in com-parison with the changes in the activities of the individual ions. At the same time since the changes in the activity of the water effected by the presence of the glycerol are relatively small the ionic activity product of the water maintains a reasonably constant value. Summary. 1. Electrometric measurements of the activities of hydrogen ion in aqueous solutions of sodium hydroxide in presence of glycerol have been made at 25". 2. The Donnan-Allmand electrode has been used to determine the activity of hydroxyl ion in solutions of sodium hydroxide containing glycerol. 3. The dissociation constant of water remains constant for the solutions containing up to Myo of glycerol; the ionic activity product does not exhibit appreciable change over the same range. The author wishes to acknowledge his indebtedness to the Department of Scientific and Industrial Research for a grant which enabIed him to carry out this investigation. Muspaam LABORATORY OF PHYSICAL AND ELECTRO-CHEMISTRY, U m ~ s r r y OF LIVERPOOL. [Received September 28th 1925.