118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13. THE INFLUENCE OF ANIONS IN THE COAGULATION OF A NEGATIVE COLLOIDAL SOL. BY D. C. HENRY AND V. A. MORRIS. ( A Paper read Before THE FARADAY SOCIETY, Monday, February 18th, I 924, SIR ROBERT ROBERTSON, K.B.E., F.R.S., PRESIDENT, in the Chair.) Received November 2 7 th, I 9 2 3. ABSTRACT. The coagulating power of an electrolyte for a lyophobe sol is largely determined by the nature of the ion of opposite sign to the colloidal particle, but is also influenced by the ion of the same sign, which exerts a stabilising action.Experiments have been carried out on the coagulating powers for a (negative) gold sol of a series of salts of the same cation (sodium), with the object of determining the relative stabilising powers of the different anions. For each salt the curve was traced connecting electrolyte con- centration with the time required to reach a definite arbitrary stage of coagulation, estimated by a colorimetric method. If the logarithms of the electrolyte concentrations and the corresponding coagulation times are plotted, the curves obtained are either linear or of small curvature.The results indicate the following sequence of stabilising power of various anions when associated with sodium ion in the coagulation of a gold sol : oxalate > HPO,” > Cog”,> OH’, citrate > HCO,’ > Br’, 1’, acetate, valerate > butyrate, CNS’ > SO,” > C1 , benzoate. I. Introductory.-The coagulation of a lyophobe sol by the addition of electrolytes is commonly attributed to the partial or complete neutralisation of the electric charge of the colloid particle by the adsorption of ions bearing a charge of opposite sign. Little attention has hitherto been paid to the influence of the ion which has a charge of the same sign as that of the colloid although, as several writers1 hsve pointed out, its effect must be considerable, since the coagulating powers of various salts, all containing the same ‘‘ coagulating ion,” differ considerably among themselves.2 The experiments reported below were undertaken in order to obtain semi-quantitative indications of the stabilising effect of various anions when associated with one and the same cation in the coagulation of a negative gold sol.Sodium was selected for the invariable cation on account of the solubility of its salts. In measuring the coagulative powers of a series of salts no attempt was made to determine a “threshold value ” of the concentration, below which the sol remained stable, and above which slow coagulation ensued. In- stead of this, measurements were made of the time required for the sol to reach a definite stage of coagulation in the presence of various concentra- tions of electrolyte.The coagulative powers of the various electrolytes are deduced from a comparison of the respective time-concentration curves thus obtained. ’Ostwald, Koll. Zeitschr., 26, 28, 69, 1920; Bancroft, Second Report on Colloid Chemistry, pp. 9, 11 ; Weiser and Nicholas, yourn. Phys. Chem., 25, 742, 1921 ; Bach, Joum. de Chem. Phys., 24, 701, 1920. 2 See, for example, the table of coagulation values collected in Burton’s Physical Properties of Colloidal Solutions, p. 158. 30ANIONS AND A NEGATIVE COLLOIDAL SOL 9-80 9-21 10.40 9'57 10.73 9'77 11-47 10.32 11-87 10.56 12.80 11-26 11.11 10'11 31 16.2 8.0 6'0 4'1 3'1 2'2 2'0 2. ExperimentaL-The gold sols used in the experiments recorded were made by Zsigmondy's " nucleus method," and were dialysed in collodion thimbles against conductivity water till the conductivity of the sol fell below I O - ~ reciprocal ohms.A colorimetric method of following the coagulation was employed. Definite volumes of gold sol and of electrolyte solution of suitable con- centration were simultaneously poured from separate test tubes into one cup of a Dubosc colnrimeter in such a manner as to ensure complete and rapid mixing. The depth of the layer to be examined was adjusted to a fixed value, and the time noted at which the sol matched a standard tint placed in the other limb of the colorimeter. The standard tints were pre- pared of stained gelatin cemented in glass, and were protected from light when not in use. No great precision was obtainable in the evaluation of the coagulation time ; the average deviation from the mean of duplicate determinations was about 8 to 10 per cent.All solutions were made up with water of conductivity about 10 - 6 re- ciprocal ohms, prepared in a still with silver condenser ; they were stored in Jena glass vessels cleaned with caustic soda and Beckmann's mixture. 3. Experimentad ResuZts.-In the table of results given below, c1 de- notes the final concentration of the electrolyte, after mixture with the sol, in millimols per liter, c2 denotes the corresponding concentration of sodium ions in milligram-ions per liter. In view of the low precision in the values of the coagulation time, it has been considered adequate to calculate cz from c1 by interpolation from the following typical '' degrees of dissocia- tion " : for uni-univalent salts, at 0.01 N, a = 0.94 ; at 0.05 N, a = 0.87, for uni-bivalent salts ,, ,, a = 0.87 ; ,, ,, a = 0-78, for uni-trivalent salts, ,, ,, a = 0.82 ; ,, ,, a = 0.70.t denotes the time in minutes required to reach the standard stage of coagulation, and is the mean of zz determinations. 85.0 37'5 13.1 5.1 1'5 2'2 1'0 TABLE I. 10 C.C. of sol mixed with 5 C.C. of electrolyte solution. Series 1.-Sol B, Standard Tint B. -~ __ c1. 1 c.3 j t . I I t . 2 2 2 2 2 2 2 ~ _ _ _ _ ( I ) Sodium Chloride. 34'70 36-23 36'70 37-00 37-88 55'52 57'97 58-72 59-20 60.61 21.52 22'28 22'88 25-04 27-07 27'97 29-28 19-52 20'16 20'66 22-53 24-28 25.04 26-12 5'55 6'66 6'93 7'23 7'56 8.3 3 (2) Sodium Sulphate. 9'99 I 1'85 12-33 12'72 13-31 14.66 29-0 6.7 3'8 10'0 2'1 0.8 (4) Sodium Oxalate.13.7 7'5 2'9 0'4 I 'I 2 3 3 3 3 2 2 2 2 2 232 THE INFLUENCE OF ANIONS IN THE COAGULATION Series 11.-Sol D, standard tint D. 5 C.C. of sol mixed with 2 C.C. of electrolyte solution. n. 2 2 3 3 3 3 3 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 2 2 2 3 3 2 j) Sodium Chloride. 6) Sodium Bromide. 40.81 41-71 42-85 43'95 45-11 46-34 47'62 49'27 35'87 37'58 38'50 39'47 40.50 41-53 42'92 36-62 26'0 18.0 9'9 5'5 3 '5 2'5 1 '5 0.8 13-18 13-85 14-20 14-58 15-40 16-30 17-36 17-93 30.0 20.5 13'7 9'9 6.0 4'1 2'1 1'2 14-29 15'04 15'44 16-81 19.05 19-70 15-87 17-85 37'53 39-09 40.80 42'65 44-68 46'91 49'38 55.20 (7) Sodium Iodide. (8) Sodium Thiocyanate. 29'31 30'44 31-66 32'9l 34'42 3 5'97 37'69 39'58 26-18 27-14 28-18 29-26 30'5 7 33'29 31-83 34'83 33-06 34'40 35-86 37'41 39-10 40'95 43*0= 47'75 33'5 24 '5 15'5 9'5 4'7 2'5 0.7 2'1 31'0 16.0 9'0 4'7 3'1 2'2 1'5 0 -6 (9) Sodium Hydroxide.10) Sodium Benzoate. 60'00 61-92 64-00 66-20 68'56 71-12 80.00 73'84 75-80 51'90 53'44 55-11 56'87 58-76 60.82 62-99 64'51 67-76 24.0 16.5 12.4 8.8 7'0 4'5 2'5 0'4 1'0 13'79 14'57 15'00 15-45 15'94 16-30 I 7 -00 17-59 18'89 12-71 13.42 13-80 14-20 14-63 14'95 15-56 16-08 17.21 23.0 11.9 9'0 7'0 4 '5 3'9 3'0 2'0 1'0 (I I) Sodium Bicarbonate. (12) Sodium Carbonate. 38-54 38.93 39'37 40'22 42-05 44'04 46'25 60.90 61-52 62-13 63.40 66.1 I 63-97 72-15 30'5 16.1 5 -6 10'0 3'0 0.9 2'1 50'55 9-42 52-24 53'92 55'72 59'69 44'03 44'69 45'29 46-70 48-14 ' 51'46 32'5 14.0 7'9 3'8 2.4 0'6 (13) Di-sodium Hydrogen Phosphate.(14) Sodium Citrate. 26-0 20'0 12'0 8.0 3'7 1 '9 0'5 1'1 69-80 70.83 72-00 73-20 74-28 76'85 79'59 20.3 9'1 6 '5 3'0 2'4 0.5 1'0 52-18 52-83 53'52 55-01 56-51 58.18 59'87 61-69 44'63 45'35 46'16 46-97 47.80 49-58 51-48 -- 22-62 22-96 23-29 24-00 24'75 25'55 26-40 27-31OF A NEGATIVE COLLOIDAL SOL 33 15-89 14'59 36.7 17-21 15-76 8.0 18.77 17'12 3'9 20.65 18.77 2'0 22'95 20'75 1'0 16.39 15-03 17'0 Series 111.-Sol E, standard tint D. 5 C.C. of s.31 mixed with 2 C.C. of electrolyte solution. 2 2 2 2 2 2 18 16 14 Sol E / I. (17) Sodium Valerate. 1 19-81 20'44 2 I -96 22'75 23-71 24'70 26-95 IS-03 18.60 19.90 20.57 21-41 22-26 24.15 25-47 30.0 16.0 7'5 6.0 4'1 3'0 1 '4 0'9 was a portion of sol D which had remained standing for some weeks after its use in Series 11.Its properties had changed in-the mean- time, as is shown by a comparison of experiments ( 5 ) and ( I s). 4. DiscussLm-The time-concentration curves are of the usual type (Fig. I), showing at low concentrations a rapid increase of coagulation time with increase in electrolyte content, and at higher concentrations a tendency for the coagulation time to approach a lower limit. This limit, which corresponds to the region of "rapid coagulation," was not nearly reached in any of the experiments carried out. For purposes of comparison, it is convenient to plot log c, against log # (Fig. z ) , since not only is the whole diagram thus reduced to manageable size, but also the resulting graphs are found to be lines of small curvature. I n fact for more than half the electrolytes employed the logarithmic curves are linear within the error of experiment, corresponding to the relation C, = ktp, where k and p are constants specific to the electrolyte, p being always less than unity, This relation cannot have any extended validity, since it allows neither for a '' threshold value," nor for a region of " rapid coagula- tion," where the time is independent of the electrolyte concentration within34 THE INFLUENCE OF ANIONS I N THE COAGULATION wide limits.’ With the exception of that for sodium hydroxide, those curves which are not linear are convex to the log t axis, a behaviour which 1 ‘9 1.8 = ‘3 I ‘2 I ’I \ NaSQ I -6 0’0 1‘0 log t 2’0 FIG.2.-Collected logarithmic curves, all reduced to standard of Series 11.Series I., 1.1; Series lI., 0, 0 ; Series III., A. is no doubt characteristic of the complete log c,-log t relation (6 Paine and Evans,2 who have measured the rate of coagulation of a gold sol over Zsigmondy, Zeitschv. F. Elektrochem., 23, 148,1917; ‘ I Kolloidchemie” (xgzo), p. 66. Paine and Evans, Trans. Faraday SOC., 19, 649, Feb. 1924.OF A NEGATIVE COLLOIDAL SOL 3 s wide ranges). The sodium hydroxide curve shows a decided bend in the opposite direction at higher concentrations, indicating that this electrolyte in sufficient quantity tends to act as a stabilising agent rather than as a coagulant. This is not an entirely unexpected conclusion in view of the exceptional adsorbability of the hydroxyl ion. The results of Series I., 11. and 111. are not directly comparable, being made with different sols; the experiments performed in each series with sodium chloride provide, however, a rough basis for intercomparison, if we make the plausible assumption that, for a given coagulation time, the ratio of the corresponding concentrations in different series will be the same for all electrolytes.This assumption is probably justifiable for electrolytes of the same type,l but is more doubtful when applied to electrolytes of different types.2 The transformed results for sodium sulphate and oxalate should therefore be accepted with reserve, and are bracketed as doubtful in the final table of this paper. In Fig. 2, all the curves have been adjusted in the above manner to the standard of Series 11. It is interesting that, whereas the correcting ratio, deduced from the sodium chloride curves, for the comparison of Series I.and 11. varies with the coagulation time, that for Series 11. and 111. is independent of the latter, a circumstance which may very likely be connected with the fact that these two series differ only in the initial states of the sols, E being probably a partially coagulated specimen of D. From the logarithmic curves of Fig. z it is evident that there is no invariable order of coagulating power among the salts used, since some of the curves intersect. By grouping together substances whose curves are close together we obtain the following sequence of anions in descending order of stabilising power. The numerical factors given are the rounded- off values of the concentration of sodium ions required to bring about coagulation (on the standard of Series 11.) in ten minutes in the presence of the equivalent quantity of the respective anions, and may be taken as approximate measures of the stabilising powers, per equivalent, of the latter.The sequence of anions obtained above agrees neither with the orthodox Hofmeister adsorption series, nor with the order of coagulative power of anions for positive sols, as determined by Freundlich and TABLE 11. Ion. [Oxalate . . Hydrogen phosphate Carbonate . . Hydroxyl . . Citrate . . . Bicarbonate . . Bromide . . Iodide . . . Acetate . . . Valerate . . Butyrate . . Thiocyanate . . [Sulphate . . Chloride . . Benzoate . . . . . . . . . * . . . . . . . . . . . . . . . . . . . . . . 851 70 63 55 45 35 28 181 14 1 See, e.g., Bach, loc. cit. 2 See, for example, Burton and Bishop, yourn. of Phys. Chem., 24, 701, 1920.36 ANIONS AND A NEGATIVE COLLOIDAL SOL others. on the coagulation of arsenious sulphide, Bancroft has deduced that the stabilising power of the anions is as follows : citrate > acetate > formate > sulphate > nitrate = chloride. This sequence is in complete agreement with that of Table II., so far as the two overlap. The various sequences of ions deduced by different workers for adsorp- tive and coagulative processes indicates, however, that ionic adsorption is highly specific to the adsorbent, and that no invariable “Hofmeister series ’’ is to be expected. The approximate rule that multivalent ions have a greater influence on coagulation than univalent is roughly borne out by our measurements. This work is being continued, both with a view to obtaining a higher degree of accuracy, and with the object of throwing light on the process of ‘‘ slow coagulation ”. From the measurements, however, of Freundlich and Schucht The authors wish to express their thanks to Dr. Powell White, Director of the Helen Swindells Research Laboratory, for the loan of the colorimeter used in this research. 1 Freundlich and Schucht, Zeitschr. f. Phys. Chew., 80, 564, 1912. Bancroft, Second Report (iit Colloid Chemistry, p. 11. Thomas Graham CoZZoid Research Laboratory, Victoria University of Manchesfer.