ETdECTRICAT~ CONDUCTIVITIES OF HYDROGEN CHLORIDE ETC. 8923 CCCCV.-The Electrical Conductivities of Hydrogen Chloride and Potassium Chloride in Water and Acetone- Water Mixtures. By THOMAS KERFOOT BROWSON and FRANK MAURICE CRAY. THE electrical conductivities of a large number of salts in acetone-water mixtures have been measured especially by Jones and his co-workers (Jones Bingham and McMaster 2. physikd. Chew ., 1906 57 193,257 ; Jones and Mahin Carnegie Institute of Wmhiszg-ton Publication 1913,180 193). The present investigation of the electrical conductivities of hydrogen chloride and potassium chloride in water and acetone-water mixtures was carried out to obtain information in connexion with a research upon the preparation of solutions of standard, hydrogen-ion concentration and the measurement of indicator ranges in an acetone-water mixture containing 10% by volume of wafer (Cray and Westrip Trans.F a r A y Soc. 1925 21). Owing to the nature of the investigation it was necessary to obtain a high order of accuracy. A detailed survey has been made of the alteration in the electrical conductivities and the degree of dissociation of these electrolytes over a wide range of dilutions in solvents ranging from pure water to the acetonewater mixture containing only 5 volumes of water in 100 volumes of the mixed solvent. Conductivity measurements have been wade for hydrogen chloride in water and eight mixtures of acetonewater and for potassium chloride in water and six acetone-water mixtures at 30" and 25" and at dilutions ranging from 10 to 10,OOO litres per gram-molecule.E X P E R I M E N T A L. Materials .-Acetone was dehydrated over fused calcium chloride and distilled fractionally at least twice immediately before use, care being taken to avoid contamination from the air. The acetone distilled between 56.2" and 56.3" a t 760 mm. and none with a specific conductivity greater than 0.5 x 10-7 mho. at 20" was used. This acetone compares favourably with that obtained by other investigators (e.g. Dutoit and Levier J . Chim. Phys. 1905 3 435; K 0.5-2.0 x 10-7 mho. at' 20". Benz Dissertation Lausanne, 1905; K 0.22 x mho. at 18"). The water was prepared from high grade distilled water by a double distillation and had a specific conductivity less than 1-2 x 10-6 mho at 20" 2924 BROWNSON AND CUY THE ELECTRICAL CONDUCTIVITIES The hydrochloric acid was prepared in normal solution and its strength determined gravimetrically from time to time.The potassium chloride (A.R. quality) was recrystallised three times from conductivity water heated for some hours at 120" and kept in a vacuum desiccator. Standard solutions were made up as required by direct weighing and standardised against silver nitrate. Prepration of Solutions.-All mixtures of acetone and water were made up to contain the specified volume of water in 100 volumes of the mixed solvent the final volume adjustment being carried out at 15". Great care was paid to this dilution on account of the considerable contraction in volume (Reilly Proc. Roy. Dublin Soc., 1919 15 43) and change in temperature which take place upon mixing acetone and water.In each series of conductivity measurements the most con-centrated solution and occasionally the more dilute solutions of hydrogen chloride and potassium chloride were prepared in a manner similar to the above the requisite volume of acid or salt solution replacing an equivalent volume of water. Subsequent dilution was carried out using specially standardised flasks and pipettes. This method was considered preferable to direct weighing on account of the vohtility of the acetone. All measurements were carried out on the same day as the solutions were made up and these were kept in the dark until actually required. The range of dilutions covered in the case of potassium chloride was limited by its sparing solubility especially in solvents rich in acetone.The electrical conductivities were measured at dilutions up to 10,OOO litres except in solvents with a high water content; the highest dilution was then 2,500. A pawatus.-The usual type of Kohlrausch apparatus modified as described below was used with a thermionic valve oscillator as the source of alternating current and with the telephones across the ends of the bridge wire (Schlesinger and Reed J. Amer. Chem. SOC., 1919 41 1'727). The bridge wire calibrated by the method of Strouhal and Barus and of accurately measured resistance could be extended a t either end by means of non-inductive standard resistance coils. The other arms of the bridge contained a standard resistance box and the electrolytic cell respectively whilst two variable air-condensers were arranged in parallel so that they could be connected in parallel across either arm of the bridge as required, in order to balance the capacities in the system.All connecting wires were of stout copper and of known resistance, which was corrected for when necessary OF HYDROQEN OHLORIDE AND POTASSIUM CHLORIDE ETC. 2925 The thermostats were regulated to & 0.05" by means of Lomy regulators with electrical control and all metal parts were earthed. Source of Current.-Taylor and Acree ( J . Amer. Ckm. SOC. 1916, 38,2415) have shown that the resistance of an electrolyte in aqueous solution measured between platinised electrodes of half-inch diameter alters with frequency up to 600 cycles but that there is no change of resistance at high frequencies which were investigated up to 2,000 cycles.Preliminary work on these acetonewater mixtures with an induetion coil as the source of alternating current showed that accurate reproducible results were unobtainable on account of the lowness and inconstancy of the frequency together with the unsym-metrical and irregular nature of the current. This source of current was therefore discarded in favour of a Sullivan thermionic ealve oscillator the frequency of the alternating current being readily adjustable by means of the anode condenser. The frequency used in this investigation was 1060 cycles which was sufficiently high to reduce the possibility of polarisation in the cell and also gave a note readily detected in the telephones.A cathode ray oscillograph was used to ascertain whether the current had a pure sinusoidal form. No difEculty wzts experienced even at the highest dilutions in obtaining an excellent minimum in the telephones when the resist-ances and capacities were accurately balanced. Conductivity CeZZs.-Taylor and Acree (Zoc. cit.) have shown that in aqueous solution whilst there is no change of resistance with change in frequency when platinised electrodes are used at fre-quencies above 600 cycles there is a fall in resistance with increasing frequency when plain platinum electrodes are used. When these results are extrapolated to infinite frequency the resistance measured with plain electrodes is identical with that found at the lower frequencies with platinised electrodes.A preliminary survey of the behaviour of electrodes in acetone-water mixtures showed that the resistances of solutions measured with plain electrodes were higher than the resistances of the same solutions measured with platinised electrodes but that the dis-crepancy between them decreased as the resistance of the solution being measured increased. Thus in the acetonewater mixture containing 10% by volume of water the results obtained at the higher dilutions with either electrode surface were substantially identical. The possibility of error due to the catalytic action of the platinum black on the acetone was investigated in a series of experiments in which the specifh conductivities of hydrogen chloride solution containing one gram-molecule in 20 500 aiid 5,000 litres of 96 5 acetone-water were measured a t 20" for periods up to 50 minutes, during which time the current was passing through the cell.The maximum alteration in specific conductivity of these solutions over this period was & 0.15y0 at the highest dilution but as the changes were in the direction of both increased and decreased specific conductivity any alteration was ascribed to temperature fluctu-ations and experimental error in determining the bridge setting. The conclusion is drawn that the use of platinised platinum electrodes is not attended by objection in acetone-water mixtures on account of catalytic action of platinum black on the solvent. Three cells having cell constants 0.5703 0.4686 and 0.1686 were used for the measurement of the specific conductivities of the electrolyte solutions.The standard solution for determining these constants was 0-02N-aqueous potassium chloride ; its specific conductivity at 25" being taken from the results of A. C. Melcher (Noyes and Falk J. Amer. Chem. Soc. 1912 34 454). The cell constants were checked a t frequent intervals during the course of the research and wherever possible the conductivity of a solution was measured in two cells having respectively low and high cell constants. The cells were all of the same type consisting of a tubular borosili-c a b glass vessel closed by means of a stopper and fitted in each case with lightly platinised platinum disk electrodes 1 cm. in diameter mounted vertically on stout platinum leads which were carried through the walls of the vessel into side tubes containing mercury.Solvent Correction and Calculation of the EquiGalent Conductivity at Infinite Dilution.-The corrections which should be applied to values of the specific conductivities of electrolytes on account of the conductivity of the solvent itself have been discussed by several investigators. Kendall ( J . Amer. Chem. Soc. 1917 39 7) has summarised the position and concludes that if the solvent is of sdliciently high degree of purity no correction need be applied in the case of acids stronger than acetic acid throughout the ordinary range of dilution but that where the electrolyte is the salt of a strong acid and strong base substantially accurate values are obtained by the procedure of Kohlrausch namely direct subtraction of the whole of the solvent conductivity.The dissociation constant calculated for hydrogen chloride from conductivity measurements falls as the percentage of acetone in the mixed solvent increases but even in the solvent containing 5 volumes of water in 100 volumes of acetone-water it is approx-imately 6 >- at 2.5" compared with 1.8 x 10-5 a t 25' for aceti OF HYDROGEN CHLORIDE AND POTASSIUM CHLORIDE ETC. 2927 acid in water. Consequently no correction has been applied on account of the conductivity of the solvent itself to the values obtained for hydrogen chloride in these acetonewater mixtures but in the case of potassium chloride the whole of the solvent conductivity has been aubtracted. In all cases the specific conductivity of the solvent was small in comparison with that of the electrolytes even a t the highest dilutions.The values for the equivalent conductivity at infinite dilution have been calculated by the method suggested by Washburn ( J . Amer. Ckm. Soc. 19!8 40 122) from the equivalent con-ductivities at the highest dilutions measured the value of the mass-action expression A ~ / A ~ ( A ~ - &)?I being plotted against the concentration for various assumed values of Am and that figure for A being taken as the most probable which led to no abrupt rise or fall in the curve at the highest dilution. The degree of dissociation of the electrolytes has been obtained from the expression a = h/h, no account being taken of the change of viscosity with dilution.RWUlts. The influence of variation in the composition of the solvent will be shown to be extremely marked in acetone-water mixtures of high acetone content especially in the case of the more concentrated electrolyte solutions. Thus alteration in the water content from 5 to 10% by volume is accompanied by a lOOyo change in the equivalent conductivity at dilution v = 50 and by a 2.5% change atv = 10,OOO. The measurements in all acetone-water mixtures were repeated several times with different samples of acetone. Thus, in the case of hydrogen chloride in the solvent containing 10% by volume of water nine separate series of measurements at all dilutions were made with different samples of acetone. The results given in all cases are the mean values.In the solvent mentioned the variation from the mean of these nine series of results over all dilutions wm & 0.5% and the reproducibility in solvents richer in water was of considerably higher degree of accuracy. In the solvent containing 50% by volume of water the variations were only 3 0.2% the results in the different series being in the same relative order a t all dilutions. A series of at least fen measurements with different bridge settings was made for every solution the maximum divergence from the mean being generally less than 5 o.05y0. Owing to the considerable difficulties in obtaining really accurate values in pure acetone due particularly to the marked effect o 2928 BROWNSON AND CRAY THE ELECTRICAL CONDUCTIVITIES TABLE I. Equivalent conductivities of hydrogen chloride in metone-water mixtures cat 25".Volume yo of water V 10 20 25 M) 100 250 500 1,000 2 m 5,000 10,000 K of solvent mho x lo-. Q) 5 10 - 2049 13.76 - - 30-81 19.70 38.93 26.15 48.42 37.50 63-40 48.08 73-96 60.39 83-13 77-47 93.16 88.41 98.22 97.61 101-5 012.2) (105.5) h 0.054 0-11 20 51-49 63-93 72.88 80.55 91.40 96-66 -100.8 104.9 106-4 107-7 (109.1) 0.20 35 50 99.44 149.7 110.1 158-0 116.9 162-7 122.6 167-3 129.0 172-5 132.2 174.2 133.8 175.4 135-8 175.9 137.0 176.3 137.7 176.5 038.5) (176.8) 0.30 0.44 - -65 202.1 211-0 215.9 219.7 224.3 225.8 226-6 227.3 227.5 (227.8) 0.55 --80 2 70.1 281.6 286-8 290.3 293.3 295.0 296.4 297.3 -- -(297.9) 0.75 -.90 100 321.3 389.9 332.0 401.0 336.0 406.4 341.0 411.0 344.2 415.3 346-6 418.0 348.0 419.5 348.9 420.7 - -- - - -(349.6) (422.0) 0.85 1.0 TABLE II. The equivalent conductivities of Aydrogen chloride in acetone-water mixtures at 20". Volume % of water. / 1 \ 2). 5. 10. 20. 35. 50. 80. 100. 10 - 19.70 47.86 90-77 135.7 247.0 362.5 20 13.21 -25 - 28.77 59.02 100.4 143.1 257.6 372.5 50 18.97 36.55 67-20 106.4 147.1 262-4 376.9 100 25.05 45-37 74-24 111.7 161.1 265.1 380-6 260 35.90 59.29 84.10 117.3 155.6 267-8 384-5 500 45-88 69.00 88.60 119-9 157.1 269.4 387.2 1,OOO 67-54 77.28 92.35 121.3 158.1 270.4 388.1 2,500 73.55 86-50 96-10 123.0 158.6 271.3 389.6 - - - - -6,000 83.80 91-25 97.25 123.7 158.9 - -10,OOO 92.56 94.40 98-40 124.5 159.1 - -in mhox 10" a (106.4) (98.0) (99-7) (125.2) (159.3) (272.0) (390.6) K of solvent 0.049 0.09 0.18 0.26 0.40 0.68 0.90 TABLE 111.The equivalent conductivities of potcassiunt chloride in acetone-water mixtures at 25" Volume yo of water. 0. - _ _ _ ~ U. 5. 10. 20. 35. 50. 65. 10'0. 50 - - 53-20 59-75 67-45 80.40 138.6 100 - 58.90 60.32 64.47 70-80 82.09 141.6 250 - 72.95 68.50 68-49 74-02 85.20 145.2 500 - 82.50 73-20 70-95 75-70 86-00 146.7 1,000 98.39 89.00 76-54 73.00 76-61 86.70 147.7 2,000 108.1 2,500 - 98-80 79-80 74-65 78.00 87-76 149.1 5,000 117.9 99.15 81-30 75-90 78.64 88.40 -10,000 122.4 101.3 82.50 - - - -al (127-7) (103.7) (84-50) (77.30) (79.40) (89.10) (150.1) K of solvent 0-054 0-11 0.20 0.30 0-44 0.66 1-0 in mho x l(r - - - - - OF HYDBOGEN CHLORIDE AND POTASSIUM CHLOBLDE ETC.2929 TABLE TV. The equicalent conductivities of potassium ch,?rn.de in acebne-woter nixtures at 20". Volume yo of water. r V. 5. 10. 20. 36. 60. 66. 100: 50 - - 48.81 53-81 60-00 71-27 125-5 100 - 55.06 55.20 57-81 63.02 72.78 128.2 250 - 68.13 62.60 61.39 65.80 75.28 131.0 500 - 76-40 66.85 63.60 67.10 76-05 132.1 1,000 92.31 82-28 69.91 65.42 68.01 76.70 132.9 2,000 101.5 -2,500 - 88-40 72-75 66.60 69.03 77.50 134.3 5,000 110.2 91.70 74.00 67.70 69.65 78.00 -- - - - -10,000 114-3 93.60 75.30 - - - -00 (119.1) (95.7) (76.90) (68.90) (70-3) (78.60) (135.2) K of solvent 0-049 0-09 0.18 0.26 0.40 0.50 0-90 in d o x 10-6 TABLE V.The epuiz.alent corzductivities of bydrogen c h i d e in water at 18" and 25" and of ptmsiurn chlm'de at 25". Hydrogen chloride. 2). 10 25 50 100 250 500 1000 2500 M Goodwin and Haskell This (inter- This rewarch. polatad). research. 351.7 351.4 389-9 362.0 - 401.0 364.9 365.5 406.4 368.5 369.2 411-0 372.3 373.6 415.3 374.9 375.0 418.0 375.8 375-9 419-5 377.2 - 420.7 (378.2) - (422-0) Bray and Hunt (intar-polated). 390.4 400.7 406-7 41 1.6 416-4 418-6 419.0 -Potassium chloride. 25O. w-Lorenz This (inter-research. Melcher. polated). - 129.0 -135.4 - 135.1 138.6 138-65 138.6: 141.6 141.4 141-6 145.2 - 144.75 146.7 146.5 146.55 147.7 - 147-76 149.1 - 149.22 (150.1) (150.6) -even the slightest trace of water this solvent has not been studied in this investigation.Results can be extrapolated for pure acetone from the values given but it is hoped in the future to investigate the problem experimentally as the only data available are very incomplete (Carrara Gazzetta 1897 27 i 207; Sackur Ber. 1902, 35 124s). The values for hydrogen chloride are all observed values being uncorrected and those for potassium chloride corrected by sub-traction of the solvent conductivity. Kendall ( J . Amer. C h m . Soc., 1917 39 7) gives 379.1 as the most probable value of A for hydrogen chloride in water at 18" and 422.7 at 25". References.-Goodwin and Haskell Physical Rev. 1904 19, 380. Melcher Bray and Hunt J . Amer. Chem. Soc.1911,33 781 2930 BROWNSON AND GRAY THE ELECTRICAL CONDUCTIVITIES see Noyes and Falk ibid. 1912 34 154. 1921,116 161. Lorenz 2. avgew. Cherr~., Discussion. The Influence of the Solvent on the Equident Conductivity of PotcGssiurn ClUoride and Hydrogen ChEoride.-The influence FIG. 1. I I I I 0 10 20 30 40 f of the ~ ~ i 30 100% acetone. Volume yo of water. 0% acetone. solvent upon the equivalent conductivity of hydrogen chloride in these acetone-water mixtures is extremely marked (fig. 1). Accord-ing to the classical theory of electrolytic dissociation this arises from two causes ; first the effect of the solvent upon the migration velocities of the individual ions and secondly upon the degree of dissociation of the electrolyte. The equivalent conductivity at infinite dilution falls rapidly as water is replaced by acetone up t OF HYDROGEN CHLORIDE AND POTASSIUM CHLORIDE ETC.3931 about 85% of acetone by volume followed by an increase of con-ductivity in acetonewater mixtures richer in acetone. The effect of alteration in the degree of dissociation of the electrolyte is shown by the equivalent conductivities at finite dilutions and accounts for the pints of idexion in the equivalent conductivity-composition of the solvent curves (Fig. 1) at the acetone-water mixture containing 10% by volume of water where the solvent has a marked effect iipon the degree of dissociation of the electrolyte (Fig. 3). FIG. 2. t i Ti A 1 90 loo D I 0% a,cetone. The nature of the equivalent conductivity-composition of the solvent curves for potassium chloride in acetone-water mixtures differs greatly from the case of hydrogen chloride.The equivalent conductivity at infinite dilution falls to a minimum in the solvent containing 40% by volume of water (Fig. 2) but owing to the effect of the fall in degree of the dissociation of this electrolyte with increase in acetone content of the solvent (Fig. 4) this minimum changes with dilution and indeed in the case of dilutions v = 50; v = 100 no minimum is observable within the limited range of acetone-water mixtures which can be studied at those concen 2932 BROWNSON AND CRAY THE ELECTRICAL CONDUCTIVITIES trations on account of the low solubility of potassium chloride in mixtures rich in acetone. Numerous investigators have attempted to show the dependence of the equivalent conductivity at infinite dilution on the physical properties of the solvent.Hartley Thomas and Applebey (J. 0 10 20 30 40 50 60 70 80 90 100 100Yo acetone Volume yo of water. 0% acetone. 1908 93 538) on the assumption of the applicability of Stokes' law to an ion moving with a spherical solvent atmosphere sur-rounding it have shown that if the ionic radii of the anion and kation respectively do not vary with the composition of the solvent, then A,q = constant where 7 is the viscosity of the solvent. Walden (2. physikd. Chem. 1906 52 242; 1911 78 273 278, etc.) has shown that the product A,? is constant and independen OF HYDROQEN CHLORIDE AND POTASSITJM CHLORIDE ETC. 2933 of the temperature for tetramethylammonium iodide tetrapropyl-ammonium iodide potassium iodide and other organic salts in a large number of organic solvents but he has also shown that solvents with large association fact,ors and high dielectric constants give variations from this rule.Creighton ( J . Franklin Inst. 1919 187, 33) found that deviations were shown by trimethyl-p-tolylammonim iodide in several organic solvents. The viscosities of acetone-water mixtures rise to a maximum in the solvent containing approximately 60% by volume of water (Davis Hughes and Jones 2. physikd. Chm. 1913 85 535), whereas the equivalent conductivity at infinite dilution of potassium FIG. 4. 100 yo acetone. Volume yo of water. 0% acetone. chloride in acetone-water mixtures has been shown in the present research to be a minimum in the solvent containing 40% of water by volume.No constancy of the expression 1-7 caa be expected. The equivalent conductivity of the electrolyte at inhite dilution is therefore dependent not only on the viscosity of the solvent but also on its other physical properties for example its state of associ-ation and dielectric constant. In aqueous solutions the conductivity of hydrogen chloride is exceptional owing to the high value of the mobility of the hydrogen ion. In acetone-water mixtures of high acetone content however, the mobility of the hydrogen ion does not d8er greatly from that of the potassium ion as the equivalent conductivity of hydrogen chloride a t infinite dilution is only slightly higher than that o 2934 BROWNSON AND CRAY THE ELECTRICAL CONDUCTrVlTIES potassium chloride in the acetone-water mixture containing 10% by volume of water.In the solvent containing 5% by volume of water it is actually smaller. The Influence of the Solvent on the Degree of Dissociation.-The influence of the solvent upon the degree of dissociation of electrolytes has been formulated in the well-known Nernst-Thomson rule. Bruhl (2. physih?. Chem. 1899 30 1) has pointed out that no absolute proportionality can exist between the dissociating power and the dielectric constant of the solvent as the latter varies greatly with temperature and also with frequency. The values of the degree of dissociation of hydrogen chloride and potassium chloride calculated from the expression a = &,/A are in Table VI and the influence of the solvent is shown in Figs.3 and 4. The influence of temperature upon the degree of dissociation of the electrolytes in acetone-water mixtures is as would be expected, more marked the greater the concentration and the smaller the degree of dissociation of the electrolyte and the dielectric constant of the solvent. TABLE VI. The degree of dissociation of hydrogen chloride and phsirr.m chloride Hpdrogen Chloride. in acetone-water mixtures at 20" and 25". Volume of water ~ - - - - - __ - - __ __ - r-5 I0 00 35 50 80 100 SOo 25" 20" 25' -00" "5" 20" 25' 00" 25O 20" 35O. 20'. 25O 10 - - 0.201 0-198 0-480 <-472 0.725 0-718 0-852 0-847 0.908 0-906 0-928 0.925 25 - - 0.295 0.29% 0.592 0-586 0.802 0.795 0.899 0.894 0-947 0.945 0.953 0.950 50 0.178 0.176 0-373 0.369 0.674 0.668 0.850 0.844 0-923 0-920 0.965 0.963 0.965 0.963 100 0.236 0.233 0.463 0.459 0-744 0.738 0.892 0.885 0.949 0.946 0.975 0.974 0.974 0.974 250 0.337 0.334 0.605 0.601 0.843 0.838 0.937 0.931 0.977 0.975 0.984 0.984 0.984 0-984 500 0431 0.428 0.704 0.701 0.889 0.885 0.958 0.954 0-986 0.985 0.991 0.990 0.991 0.991 1,000 0.541 0.538 0-789 0.788 0.926 0.924 0.969 0.966 0.993 0.992 0.994 0-995 0-994 0.994 2,500 0-691 0.690 0-883 0.883 0.964 0.962 0-982 0.980 0-996 0.995 0.997 0.998 0.997 0.997 5y000 0-787 0.788 0.931 0.931 0-975 0-975 0-989 0.989 0.998 0.997 - - - -10,000 0.870 0.870 0.963 0.962 0.987 0-987 0.994 0-994 0-999 0.998 - - - -2' A - 7\ +- 4 Potanssium Chlm.de.Volume "/b of water. 50 100 250 500 lyoOO 2,500 8,000 1!J,Vo0 I 0.775 0.925 0.960 -050 800 250 yo0 "0 - - _ 0.635 0-630 - U*t75 0-568 0.718 0.714 - U-rl2 0-704 0-814 0.811 - 0.798 0.796 0-869 0-866 0-770 0-860 0.858 0.909 0.906 - 0.924 0.924 0-946 0.944 0.923 0-958 0.956 0-962 0.962 0.958 0.978 0.977 0.979 0.980 U-891 0.886 0.936 0.923 0-918 0.954 0.949 0.944 0.967 0-967 0.966 0.982 0.983 0.988 0.991 - - -250 0.850 0.892 0-938 0.954 0-965 0-982 0.990 -20" 25" 0.907 0.902 0.926 0.921 0.958 0-956 0.968 0-965 0.976 0.973 0.986 0.985 0.992 0-992 - -'OD 250 0.928 0.923 0.948 0.943 0.969 0.967 0.977 0.977 0-983 0.983 0-994 0.993 - -- -The values of the Ostwald expression G/(l - a)v show consider-able alteration with dilution in water or acetone-water solvents an OF HYDROGEN CHLORIDE AND POTASSIUM CHLORIDE ETC.2935 in no case is a true constant obtained but in mixtures rich in acetone the change with dilution is much less marked. Summary. (1) The electrical conductivities of hydrogen chloride and potas-sium chloride have been measured over a wide range of dilution at 20" and 25" in acetone-water mixtures containing from 5 to 100% of water by volume. The equivalent conductivity of hydrogen chloride a t infinite dilution falls sharply to a minimum in the solvent conhining about 86% of acetone after which the change is small an increase being noticed in solvents very rich in acetone. In other solutions the equivalent conductivity falls as the acetone content increases as the result of the influence of the solvent upon the degree of dis-sociation of the electrolyte as well as upon the migration velocities of the individual ions. The equivalent conductivity of potassium chloride a t infmite dilution falls to a well-defined minimum in the solvent containing approximately 40% by volume of water but a t other dilutions, owing to the change in degree of dissociation of the electrolyte this minimum shifts and in more concentrated solutions w = 100 and u = 50 no minimum is observable in any acetone-water mixture in which the solubility of this electrolyte allows the measurement to be made. (2) It is shown that the use of platinised platinum electrodes is not attended by error due to catalytic influence upon the acetone in the mixed solvents studied. (3) The influence of temperature upon the degree of dissociation of these electrolytes is more marked the higher the acetone content of the solvent and the higher the concentration of the solution. ( 4 ) The Ostwald dilution law does not hold fully in any acetone-water mixture investigated but as the acetone content increases, gives values more nearly approaching a constant a t all dilutions. We wish to express our indebtedness to Dr. G. Rotter C.B.E., and Dr. J. N. Pring for t-heir interest in this work which is published by permission of the Director of Artillery War Office. RESEARCH DEPARTMENT, ROYAL ~ E N A L WOOLWICH. [Received Azrgwrt Zlet 1925.