BHATNAGAR STUDIES IN EMULSIONS. PART 11. 61 V.-Studies Eniulsions. Part II. The Reversal of Phases by Electrolytes and the Eflects of F r e e Fatty Acids and Alkalis on Emulsion Equilibrium. By SHANTI SWARUPA BHATNAGAR. THE study of the effects of electrollytes on emulsion equilibrium is undoubtedly of greatl importance for the elucidation of the mechanism of emulsification but unfortunately very little quanti-tative data is available on the subject. The work of Clowes (6. Physical Chem. 1916 20 445) is restricted to! a particular typ 62 BHATNAGIAR STUDIES IN EMULSIONS. PART 11. of emulsion in which the volumes of the two phases are always equal. I n a previous paper (T. 1920 117,549) this work of Clowes has been discussed. Subsequent work on the problem has shown that the procedure of using alkaline solutions and free fatty acids in oils employe’d by Clowes is not quite satisfactory.When dilute sollutions of alkalis are used a comparatively large amount of free fatty acid in the) system is left unneutralised. The presence of free fatty acid or free alkali makes the system more complex. I n addition they seem t o have a definite effect on the emulsion equilibrium as will be shown later. Really corn-parable results can only be1 obtained and repeated by employing neutral oil and soap solutions. EXPERIMENTAL. Tn all experiments described in this paper various soaps were used instead of different alkalis. Sodium oleatel and potassium stearate were obtained from Kahlbaum. The lithium stearate was a sample prepared by Sir William Ramsay and preserved in this laboratory.Sodium lindeate and other soaps were prepared in 1 he manner advocated by MacBain ( Y ’ r u ~ s . b”imtZay SOC. 1913, 9 99). The electrolytes were1 pure chemical reagents (excepting nickel nitrate) and were recrystallised before making up the solutions. I n order to! ensure1 identical conditions for repeating the observ-ations the time for shaking the volume of the phases and the size and other conditions of vessel were always kept alike. It was found that under these conditions with carefully cleansed vessels, the observations could be repeated with a fair degree of accuracy. The emulsions were prepared in wide-mouthed bottes and the type noted by examining small portions of emulsions in a T-shaped glass container according to the method described in a previous paper (Zoc.c i t . ) and with frequent testings under the microscope. The two1 methods gave identical results for all fine-grained emulsions. Eqneriments with E’lectrolytes. Volume of oil phase = 10 C.C. Volume of aqueous phase=10 C.C. I2 = Inversion point BIIATNAGAR STUDlES CN EMULSIONS. PART 11. 63 TABLE I. Soul3 Used Sodium Oleate. The figures givea in the columns represent% the amount of the respective salts in niillimols at R1. Amount of soap in milli-mol. Ba(N03),. Sr(N03)2. Pb(NO,),. Ni(N03)2. A12(S04),. Cr,(S04),. 0.080 0.0398 0.0398 0.0396 0.036 0.014 0.016 0.101 0.0500 0.052 0.504 0.040 0.017 0.019 0.150 0.080 0.084 0.082 0.060 0.025 0.027 0.162 0.084 0.090 0.088 0.076 0.027 0.030 0.210 0-112 0.116 0.110 0.0798 0-036 0.039 TABLE 11.Soap Potassium Stearate. Soap in milli-mol. €3a(N0,)2. SI!(NO,)~. Ni(N0,)2. Pb(NO,),. Al,(SO,),. Cr2(S04),. 0.086 0.044 0.046 0.040 0.040 0.018 0.018 0.10 0.0508 0.0512 0.050 0.052 0.026 0.022 0.15 0.084 0.086 0-078 0.088 0.027 0.026 0.25 0.132 0.136 0.128 0.140 0.048 0.048 0.30 0.156 0.160 0.150 0.154 0.056 0.057 TABLE I 11. Soap Lithium Stenrate. Soap. Ba(NO,),. Sr(N03)3. Ni( NO3),. Pb(N0,)3. Al,(SO,),. Cr2(S04),. 0.09 0.040 0.040 0.036 0.042 0.015 0.016 0.12 0.058 0.058 0.050 0.056 0.021 0.023 0.16 0.090 0.094 0*080 0.094 0.028 0.029 0.20 0.102 0.104 0.096 0.106 0.03 0.032 TABLE. IV. Soap Sodium Linoleate. Soap. Ba(NO,),. Sr(NO,),. Ni(N03), Pb(NO,),. Al,(SO,),. Cr(SO,),. 0.083 0.068 0.068 0.058 0.072 0.028 0.028 0-112 0.100 0.106 0.090 0.116 0.038 0.037 0-125 0.128 0.134 0.120 0.136 0.042 0.043 0.18 0.198 0.20 0.178 0.200 0-061 0-063 The results shown in tables I 11 111 and IV indicate in a marked manner the difference in the amounts of various electro-lytes required to bring about the reversal of phases.The effect of the valency of the electrolyte is worth noting. The amounts of the tervalent electrolytes aluminium and chromium sulphates, required to bring about the reversal of phases are less t'han thos 64 BHATNBGAR STUDIES IN HIKULSIONY. PART 11. od bivalent elsctrolytw. The polwer oif reversing the phases in these electrolytes is in the order a1uminium)chr~mium) nickel>lead>barium>strontium. Calcium bivalent iron and magnesium are found to( have practically the same values as strontium and are therefofre not shown in the tables.The amount 09 electrolytes required to1 bring about the reversal of phases differs with different soapsl but the valency effect of the electrolyte still holds goold. Effect of Changz'n)g the Volume-ratio of t 7 ~ e Phases. I n all previous experiments the volumes of the oil and aqueous phases have been equal. Keeping all other conditions constant, some experiments were tried in order to find how tbhe action of ,electrollytes on the inversion point is affected by changing the volumerat,iol of the two phases. The results' are shown in table V TABLE V. Tot,al volume of emulsion always 20 C.C. Soap Lithium Stearate 0.12 ?tLillirtLoL. B. P.parafh in C.C. 2 5 8 10 11 13 15 Aqueous phase in C . C . 18 15 12 10 0 7 5 Ba(NO,) in millimol in the aqueous phase at R,. 0.078 0.070 0.060 0.058 0.054 0.044 0.040 Al,( SO& in millirnol at R,. 0.030 0.028 0.024 0.021 0.018 0.012 0.010 It is interesting to note that the amount of multivalent elwtro-lyte requireid to bring about the inversion of phases increases as the volume of the aqueous phase is increased and diminishes as it is reduced. A corresposnding increase in the oil phase has a reverse effect. Thus in table V the ratio1 between lithium stearate and barium chloride a t the inversion point was as 0.24 0.058 when the volumes of the two phases were as 10:lO. When the volume of the aqueous phase was increased so that the ratio between them was 15 :5 the amount of barium nitrate a t R increased to 0.070 millimol and was reduced to 0.04 when the ratio became 5:15, that is oln proportionately increasing the oil phase.Similar rwdts were olbtainable with other soaps and electrolytes used in this investigation BHATNAGAR STUDIES I N EMULSIONS. PART 11. 65 Effect of DiJution 0% the Reversal of Phases by Electrolytes. Emulsions can be diluted infinitely by the addition of the con-tinuous phase. Itl was considered desirable to find how dilution affects the actioln of electrolytes on the inversion point. A large quantity od fine-grained emulsions was prepared by vigorously shaking in a mechanical shaker some oil to which the aqueous phase was gradually added until the total volumes of the aqueous and oil phases became equal.This standard emulsion was divided into four equal portions of 20 C.C. each. One portion was kept un-diluted and the other three portions were differentally diluted with solutions of soap and different amounts of electrolytes added previous to a second shaking in the manner already described. Some of the results are given in tableu VI. TABLE VI. Soap Sodium Oleate. Amount of soldium odeate in each sample 0.101 millimol. BaCl in millimol at the inversion A12(S0,)S in milli-mol at the inver-20 C.C. of emulsion. point. sion point. 1. Standard emulsion ... ........ . ..... . ... 0.050 0.014 2 . SY , twice diluted ...... 0.0798 0.025 3. ? Y , thrice diluted. .. .. . 0-1 12 0.040 4.9 ) , four times diluted 0.156 0.057 The remlts in table VI indicate that the greater the dilutJon or the distance between the oil particles in an emulsion the larger is the amount of the' multivalentl electrolyte required tol bring about the reversal of phases. These results are in keeping with the previously observed effect of dilution on the rate of coagulation of colloidal sols by electrolytw (T. 1919 115 462). A few observations were also made to ascertain how the rate of coagulation of the typical natural emulsion milk is affected on dilution. The procedure consisted in preparing several different solutions of milk and water. Equal volumes of these solutions were placed in clean testitubes a definite volume of 0'4906N-sulphuric acid was added the tube well shaken and the times f o r the first perceptible change and complete ccagulation were noted.Table VII indicates some of the results o b t a i n d VOL. CXIX 66 BHATNAGAR STUDIES IN EMULSIONS. PART 11. TABLE VII. Time for first perceptible change on the addition of a known volume of sulphuria Volume of solution always 20 C.C. acid. Pure milk ............... Immediately after adding. Twice diluted ............ Thrice diluted ......... 12 minutes. Four times diluted ... 15 minutes 3 seconds. Five times diluted ...... 25 minutes 2 seconds. Eight times diluted ... 50 minutes. Ten times diluted ...... 70 minutes. 9 minutes 2 seconds. Another set of experiments was performed to find the strength of a known volume of acid which will produce a perceptible change immediately on addition to the solutions and it was found that the greater the dilution the stronger was the acid required to produce an immediate coagulation effect.It is t*hus evident that the reversal of phases by electrolytes in emulsio,ns the precipitation of colloidal sols in suspensions and the coagulation olf substances like milk in which the acids used act oln the casein membrane round the f a t globules are similarly affected on dilution. It is probable t(ha6 the cause or causes producing these identical effects are also identical and a thorough study of any one of these phenolmena may lead to the elucidation of the problem of coagulation. Effects of Free Fatty Acids and Alkalis on the Emulsion. Equili bm'um. The procedure adopted fosr investigating the irregularit'ies observed when free fatty acids and alkalis were used for emulsifi-cation consisted in preparing three sets of emulsions.All con-tained the same amount of sodium oleate and the same volume ratio the only difference being that in one set the oil contained free fatty acid in the second the aqueous phase contained free alkali and in the third set' neutral oil and soap solutions alone were used. The following table represents the results which show TABLE VIII. Soap Used Sodium Oleate. Emulsions Free containing otassium 0.162 millimol Eydroxide. of soap. Gram. 20 C.C. 0.01 SP 0.02 1 as 0 88 0 P? 0 Ba(NO)S in acid. required at 0 0.09 0 0.096 0.003 0.078 0*008 0.070 0 0.084 Free fatty millimol El* Gram BHATNAGAR STUDIES IN EMULSIONS.PART 11. 67 that the free fatty acids and1 free alkalis shift the inversion point in opposite directions. During the microscopic exaniination of these three sets of emulsions it was found that the average size of globules of oil was greater in emulsions containing free fatty acid and smaller in emulsions containing free alkali in the system than the average size for identical and equal shakings when neutral oil and soap solutions alone were used. Itl is interesting to note that the addi-tion of a little free alkali to a soap solution also decreases the size of the particles and results in turning a slightly turbid solution into a perfectly clear one whilst the addition of free fatty acid increases the size of the particles and1 makes the solution more turbid.This may possibly account for the difference in the sizes of particles in the above-mentioned emulsions. It is to be noted that as the degree of dispersion is an important factor in the stability of an ernulsion all attempts to arrange various emulsifying agents in order of their efficiency for emulsifi-cation without taking into account the size of the particles of these agents are defective. The size of the particles of an emulsifying agent has also been found by Pickering (T. 1907 91, 2001) to have a definite effect on the degree of dispersion. Discussiorz of Results. The results indicated in tablm I 11 111 and IV show that the tervalent electrolytes have a greater effect on the inversion of the emulsion than the bivalent ones.A similar conclusion has been drawn by Clowes (Zoc. c i t . ) . The effect of dilution and of increas-ing t’he distance between the oil particles in an emulsion is essenti-ally similar in nat’ure to the effect of dilution on colloidal sols. A complete discussion of the results obtained here will be reserved for a later contribution. The difference in the amount of electrolytes required to bring about the reversal of phases with different soaps points t o the probability of a difference in their protective actions. The results indicate tchat the soaps can be arranged in order of their protective action as potassium stearate>sodium stearate>sodium and potassium palmitates>potassium o1eate)sodium oleate for B .P. paraffin oil.The results have an important bearing on the cleansing power of soaps. The old hypothesis of Chevreul that the cleansing power of soap is due to the free alkali liberated is now discarded. The modern conception attributes the washing power of soap to its “emulsifying efficiency” and to its protective action which keeps the dirt and grease bolund in the form of an 0 68 BHATNAGAR STUDIES IN EMULSIONS. PART 11. emulsion. On this view the cleansing properties of maps will be considerably affected by the electrolytic impurities in water. Sodium linoleate and soaps with a little free alkali will be gmd for cleansing purposes as they give finer-grained emulsions than the sodium oleate solutions when pure water is employed for wash-ing. Large amounts of calcium or barium salts or salts of ter-valent metals in water will be injurious in these cases as well as when neutral or acid soap solutions are eniployed for the latter are also easily transformed into water-in-oil emulsions which are very sticky and difficult to remove by water.It is intended to study the varioius physical properties particu-larly viscosities of emulsions of two immiscible liquids sf equal densities in order to obtain more evidence as to the mechanism of the reversal of phases in emulsions and further work on the renewal of phases in emulsions prepared by insoluble emulsifiers is in progress. Summary. (1) It has been shown that tervalent electrolytes are more effective in bringing about the reversal of phases than bivalent ones. (2) Soaps and electrolytes have been arranged in the order of their emulsifying and precipitating powers respectively for B .P. paraffin oil emulsions. (3) It has been sholwn that the reversal of phases in emulsions, the precipitation of sols in suspensions,' and the coagulation of natural emulsions like milk are similarly affected by dilution. (4) The bearing of the results on the chemical interpretation of the washing power of soap is briefly discussed. (5) The effects of free alkalis and free fatty acids on soap emulsions have been investigated. The author takes 6his opportanity of t,hanking Prof. F. G. Donnan F.R.S. for the help afforded him during this investiga-tion. His thanks are also due to Dr. J. C. Ghwh and Mr. J. N. Mukherjee. CHEMICAL LABORATORY. UNIVERSITY COLLEGE, LONDON W.C.l. [Received October 12th 1920.