Interaction Energy of Emulsion Droplets and the Influence of Adsorbed Layers on it BY H. SONNTAG J. NETZEL AND B. UNTERBERGER Deutsche Akademie der Wissenschaften zu Berlin Zentralinstitut fur Physikalische Chemie Berlin-Adlershof Deutsche Received 9th March 1970 The equilibrium distance contact angle and formation velocity of black films between model droplets in aqueous solution of surfaceactive agent were measured. These variables produced information about the forces of interaction between the droplets. The concentration of the surface- active agent was increased to determine the effects of surface concentration upon the above forces for small and wide particle spacings with reference to the thickness of the adsorbed layers. All experi- ments proved that by stepping up the concentration of the surface active agent the attraction between the particles was increased and consequently the stability reduced.The stability of colloids and the formation of coagulation networks in concentrated dispersions may be modified within wide limits by adsorbed layers of surface-active materials. They are capable of changing both the electrostatic repulsive forces and the dispersion forces. Other factors that have to be taken into account under conditions of small particle spacings are the dipole forces the hydrogen bonds of water molecules adsorbed according to a given orientation and the steric hindrance of the adsorbed layers likely to occur in their mutual penetration or compression. While for practical purposes there is a common use of such modification of interaction forces by adsorbed layers little quantitative work has so far been performed on this aspect.This paper will be limited to results obtained from investigations of non-ionic surface active materials which are less complex since adsorption-dependent changes of the charge may be neglected. The influence of non-ionic surface active agents on the stability of sols has already been described by some authors. 1-6 However the influence of surface concentrations of surface-active agents upon interaction has not yet been studied in connection with model tests of foams and emulsions. An attempt is made to close this gap in this paper which reports the effects undergone by the interaction forces between oil droplets which are separated by an aqueous film of surface-active agents. EXPERIMENTAL MATERIALS Spectroscopically-pure n-octane and doubledistilled water were used.The electrolytes were molten prior to use. The non-ionic surface-active agent used was nonylphenol with 20 moles of ethylene oxide (NP 20) a preparation made in the laboratory. MEASUREMENT OF THE EQUILIBRIUM DISTANCE The interaction forces between emulsion droplets may be characterized by the equilibrium spacing desn resulting from an equilibrium of repulsive with attractive forces which is reached 57 58 INTERACTION ENERGY OF EMULSION DROPLETS with a certain electrolyte concentration. The distance between microscopic emulsion droplets were measured by interferometry. The radius of the plane-parallel contact zone was 0.01 cm. The apparatus used is described in detail in ref. (7). For emulsions the calculation of droplet spacings from the intensity of light reflection proved to be easier than for foam laminae since it could be based on the single-layer model.The alkyl groups of the surface-active material were found in the oil phase. with their refractive index to first approxi- mation being identical with that of octane. The greatest change of the refractive index was found to take place at the boundary between the oily phase and the hydrated polar groups so that the percentage of the adsorbed layers formed by the polar groups will be included simultaneously by the optical measurement of oil droplet spacing. MEASUREMENT OF THE CONTACT ANGLE A porous annular glass cell g as shown in fig. 1 was used to measure the contact angle between the black film and the volume phase. The surface of the glass frit was melt-sealed t c \ 9 \ \ \ m \ \ \ 1777777772 FIG.1 .-Apparatus for contact angle determination. so that the drainage of the fluid took place only inside the conical bore 0.08 cm in diameter. The drainage proper was effected by means of the micrometer screw m. The glass frit was inserted in a cuvette c with a plane-parallel bottom which was filled with the oil phase. The glass tube t shown to be fused on to the glass frit above bore level was not needed for emulsion testing but would be required to test the heterosystem air water oil not considered in this context. The detector was mounted on an incident-light microscope and the droplets were photo- graphed prior to and after black film formation. The contact angle was calculated from the spacing of Newton’s rings or from the growth of the contact area following black film forma- tion according to the method of Sheludko and co-workers.* VELOCITY OF BLACK FILM FORMATION The velocity of black film formation was characterized by measuring the time that elapsed from the appearance of the first black spots to the completion of a homogeneous black film.g H .SONNTAG J . NETZEL AND B . UNTERBERGER RESULTS The equilibrium spacings of octane droplets in NP-20 solutions of different concentrations were measured in electrolyte concentrations of 0.001-0.01 m KCI. The results are given in fig. 2. It may be seem from the graph that up to the critical 59 500 400 - 3 0 0 s 1 2 0 0 I00 0 FIG. 2.-Equilibrium distance deqU(A) as a function of the KC1 concentration c (rnol/l) for different surface-active agent concentrations 0 7.5 x mol/l ; x 7.5 x mol/l ; v 1 x mol/l.mol/l ; 0 7.5 x mol/l ; 0 2.5 x concentration of micelle formation (c.m.c.) which was reached at 1.4 x mol/l the equilibrium films decreased in thickness with the electrolyte concentration remaining constant. The influence of the surface-active agent grows as the droplet distances are reduced. While above the c.m.c. level no further change of the equilibrium spacings takes place the influence of the surface-active agent can be observed already at lower electrolyte concentrations. These experimental findings might be explained by both a decrease of the repulsive forces or an increase of the attractive forces along with growing adsorption of surface active agent. Equilibrium spacings were measured only under conditions were they were much greater than the layer thickness of the adsorbed polar groups.Contact angle measurements were performed to find out whether with reference to the sign a further change in the influence of the surface-active agents on the interaction forces was possible even with small spacings. The droplet distances chosen for that purpose were so small that the intensity of light reflection coincided with that of the minimum range. They were smaller than 40 A. The experimental results are collected together in table 1. While the influence of the surface active agent concentration upon the contact angle was small a growth of the attractive forces rather than if the repulsive forces was evident and it was even more obvious by comparing the time intervals tA that elapsed from the appearance of the first black spots to the formation of homogeneous black films.These intervals 60 INTERACTION ENERGY OF EMULSION DROPLETS TABLE CONTACT ANGLE 8 AND TIME t OF FORMING BLACK FILMS IN EQUILIBRIUM WITH SOLUTIONS CONTAINING m AND 7.5 x lod3 m NP 20 AS A FUNCTION OF ELECTROLYTE CONCENTRATION electrolyte 10-1 m KCl 0.01 38 0.03 25 1 m KCI 0.00 30 0.04 15 2 m KCl 0.01 10 0.07 tl 5 x m MgS04 0.02 61 0.03 23 5x 10-1 m MgS04 7.5 . 1 8.0 tl show a clear-cut declining trend despite the increased concentration of the surface- active agent in the film. The result that the growth in surface-active agent concentration was accompanied for all distances by a growth (or decrease) of the attractive forces (or repulsive forces) is in a contradiction to the flocculation tests applied to silver iodide sols and latices by Ottewill 1* who found that with in the c.m.c.range any addition of surface-active agent entailed an increase of stability. Slight stability decrease was observed by Glazman in silver iodide and arsenic sulphide sols containing small amounts of surface-active agent whereas in the c.m.c. range the stability was found to grow strongly These results were confirmed by Srivastava who referred to antimonic and arsenic sulphide sols. With reference to peptization of clay with smaller quantities of surface-active agent Schott found a stability decrease for these smaller concentrations and a stability rise for higher concentrations. No change in stability after the addition of non-ionic surface-active agent was established from experi- mental sedimentation of kaolin performed by Lange.Unfortunately there have been no systematic studies into foam films as yet. However it is believed that the authors' results provide an explanation for the different layer thicknesses found by van der Waarde lo and Sheludko l1 who tested macroscopic and microscopic foam films with one and the same surface-active material. The macroscopic films were stabilized at surface-active agent concentra- tions above the c.m.c. and gave equilibrium thicknesses much smaller than those of microscopic films in which the surface-active agent concentrations were smaller by two orders of magnitude. DISCUSSION The only factors which have to be considered in calculating the interaction forces per unit area are the electrostatic repulsive forces (He,) the dispersion forces @ID) and the capillary pressure (ITu) since in all measurements the equilibrium spacings were bigger than the thicknesses of the adsorbed layers.The following equation has to be satisfied for the equilibrium spacings n, = n,+n,. The two forces may be analysed separately if the influence of the surface-active agent is attributable mainly to the change of II, or to that of II,. While nu is also dependent on the given surface-active agent concentration it can be determined separately by measurement of the interfacial tension. The inter- facial tension and consequently the capillary pressure will decrease approximately by a factor of two until the c.m.c. is reached. This will cause a reduction of the attractive forces and therefore cannot explain the reduction of the equilibrium H .SONNTAG J . NETZEL AND B. UNTERBERGER 61 spacings. Since the magnitude of the capillary pressure is identical with that of the dispersion forces for larger distances a change of the former could even lead to a compensation of the measured effect. The dispersion forces are calculated according to the model proposed by Vold.12 Two particles with the Hamaker constant Al interact in a medium with the Hamaker constant Ao. The adsorbed layer is subdivided into two parts,13 the non-polar part of thickness 6 and the Hamaker constant A2 on the one hand and the polar groups of thickness ij3 and the Hamaker constant As on the other. In earlier work evidence was produced to the effect that surface-active agents with equal alkane chain length but different polar group would affect the dispersion interaction due to the structure of this group and that the hydrocarbon chains (solute in oil) did not contribute to the energy of di~persi0n.l~ This would support the conclusion that to a first approxima- tion A is equal to A l .The dispersion forces may then be expressed by the following equation 1 (A8 - A$)2 + (A4 - 2(A$ - A$)(A$ - A t ) + d 3 (d- 6,13 n,=- 6n (d - 263)3 Y where d is the distance determined by interferometry. If A3 differs from Ao smaller distances will always result in increased attraction no matter whether A3 is smaller or bigger than A . and A l since the first and third terms of eqn (2) will be very large. A decrease of dispersion energy may result from larger spacings,13 provided that A3 is smaller than both Al and Ao. The results obtainable from the above equation cannot be evaluated unless at least A.and At are known from other measurements. erg found for microscopic foam films with low surface-active agent concentrations by Shelduko and co-workers l 5 ; A = 8.5 x 10-13 erg is substituted for the oil phase on the basis of measurements on latices l6 and emulsions with low surface-active agent concentra- tions. The adsorbed layers of surface-active agents will be capable of affecting the electrostatic forces in two ways either by changing the structure of the double layer or by changing the potential of the diffuse double layer ($8). Since structural changes would be detectable mainly along with a close approach of the particles this factor should be neglected for the purpose of equilibrium distance measurement (d> B3).The potential at the oil/water interphase may be generated by adsorption of OH’ from the water or by adsorption of the surface active-agents proper. Under the first assumption it would be possible that by increased adsorption of surface- active materials the potential-determining ions would be displaced whch would result in a reduction of the potential as such and consequently of the electrostatic interaction. This was tested by measuring equilibrium distances at constant surface- active agent concentration and variable pH values. The tests have shown that the equilibrium thicknesses remain constant in the measured pH interval of 3-7. Measurements of the interfacial tension performed concurrently have shown that within the above pH range the adsorption of surface-active agents remained un- a1 tered.The evaluation of the experimental results was to produce information as to the contribution made by each of the various factors (A3 a3 and to the influence of surface-active agents. The following equation is obtained for an 1 1-electrolyte [c(mol/l)] and 25°C by introducing in addition the various distance functions into Let us substitute for A. the Hamaker constant of 3 . 5 ~ 62 INTERACTION ENERGY OF EMULSION DROPLETS eqn (1) exp (19.83 1 exp (19.83 + 1 exp (-0.329 x 1O8deqnJc = 1.59 x 1 0 9 ~ 1 (A8 - At)' (A$ - At)2 2(AB - A$)(A$ - Af) + + + n o (3) 6.n(d,,,-263)3 dZqn (deqn - 6)3 6 and A3 may be calculated for different surface-active agent concentrations from the (desu,c) functions determined experimentally. The results are given in table 2.TABLE 2.-INTERFACIAL TENSION 8 POTENTIAL OF THE DIFFUSE DOUBLE LAYER $8 HAMAKER CONSTANT AJ AND THICKNESS 6s OF THE POLAR GROUPS AS A FUNCTION OF CONCENTRATION OF NP 20 surfactant concentration @6 A3 [mol/l.] [dyGcm] [mV] 1013 [erg] $1 7 . 5 x 19.1 32 0.7 22 7.5 x 10-5 1 0 . 5 19 0.7 22 2 . 5 ~ 10-4 10.1 12 1 . 6 38 7.5 x 10-4 9.0 12 1 . 6 38 The interfacial tension and consequently the capillary pressure decreases with increasing surfactant concentration. This will cause a reduction of the attractive forces and therefore cannot explain the decline of equilibrium spacings. The double layer potential decreases up to the saturation concentration of the adsorbed surfactant molecules. On the other hand the Hamaker constant and the thickness of the adsorbed layer increases up to the c.m.c.These results show that the influence of non-ionics on the interaction energy is complex and that each of the various factors tend to decrease equilibrium spacings. An interesting result would be obtained by comparison of the thickness of the polar groups with the structure of polyoxyethylene. Staudinger postulated two structures for polyoxyethylene the zigzag and the meander configuration. Up to a degree of polymerization of about 9 the chain exhibits in a zigzag structure whereas at higher degree of polymerization a meander structure is observed. For the oxyethylene unit in the meander-type chains a value of 2 A is obtained and therefore the length of the oxyethylene chain in NP 20 should have a value of 40 A. This value agrees well with the measured thickness of the polar groups of 38 8 in the saturated adsorbed layer.More intricate problems were faced with regard to contact angle measurement since with the small particle spacings some additional forces of interaction with their orders of magnitude and distance functions still being unknown had to be taken into account. Therefore an analysis of the contact angle measurements and fornia- tion velocity is impossible. The only possible statement is that these measurements too showed an increased attraction of the particles with increasing surface concentra- tions. The hydrate shells often postulated as the cause of stability did not play any role at least in our example. In a theoretical study by Havemann 17a evidence was produced to the effect that water molecules adsorbed by a specified orientation (the dipoles being orientated parallel and arranged vertically relative to the interphase) cause the appearance of repulsive forces almost equal to that of the dispersion forces only if the degree of orientation is above 65 "/o.Such a high degree of Orientation was not reached in our experiments. The fact that a finite particle distance is obtained may be attributed to the steric hindrance of the adsorbed layers which cannot be displaced from the interphase due H . SONNTAG J . NETZEL AND B . UNTERBERGER 63 to the insolubility of the given surface-active agent in the oil phase. Compression or mutual penetration is encountered by the adsorbed layers producing a force which can be compensated by the dispersion forces. This force was measured directly in non-polar media by compression of monomolecular and multimolecular adsorbed layers on mercury droplets.18 The reversible change of distance between two flocculated mercury droplets was determined under conditions under which the droplets were pressed against each other by different pressures.The elasticity (vertical relative to the interphase!) derived therefrom for adsorbed layers of oleic acid 54 A in thickness was lo6 dyn cm-2. Such a value was sufficient for compensa- tion of the dispersion forces. The " mechanical " strength of the adsorbed layers under load relative to the interphase in a vertical manner is believed by the authors to be the cause of the coalescence stability. The latter has often been attributed to the strength in the adsorbed layer (mechanical properties of the layer measured parallel with the interphase).Measurements of both the shear viscosity and elasticity by the method of wave damping have revealed what has become established know- ledge viz. that in monomolecular adsorbed layers of surface-active agent agreement between the coalescence stability on the one hand and the mechanical properties measured parallel with the interface on the other would occur only in few selected instances. This seems to support the conclusion that a breaking-up of the adsorbed layer i.e. the tearing open of a hole is not the decisive step towards coalescence. The question why the stability may be both increased or reduced by non-ionic surface-active agents depending on the nature of the disperse systems studied still remains unanswered. The studies conducted by Cockbain l9 and Lemberger 2o may be considered as confirmations of our own results.In emulsions 1-2p in particle size increased adsorption of ionic surface-active agents will lead to stronger aggregation as supported by the creaming volume quoted in ref. (19) and direct coulter counter determination reported in ref. (20). The same result was achieved by using macrodisperse solid di~persions.~. 21 Yet it seems not plausible that particle size alone should be responsible. K. G. Mathai and R. H. Ottewill Trans. Faraday SOC. 1966 62 759. R. H. Ottewill and T. Walker Kolloid-2.2. Polymere 1968 227 108. H. Lange Kolloid-Z. Z . Polymere 1966 211 106. M. Glazman Jr. Disc. Faraday SOC. 1966 42,255. H. Schott J. Colloid Interface Sci. 1968 26 133. K. L. Daluja and S. N. Srivastava Indian J. Chem. 1969 7 790.H. Sonntag J. Netzel and H. Klare Kolloid-2. 2. Polymere 1966 211 121. A. D. Scheludko B. Radoev and T. Kolarov Trans. Faraday Soc. 1968 64,2213. H. Sonntag and J. Netzel Tenside 1966 3 296. D. Exerowa and A. D. Scheludko Proc. IV. Int. Coizgr. Surface Active Substames (Briissell 1964) vol. 2 p. 1097. l o Van der Waarde private communication. l2 M. J. Vold J. Colloid Sci. 1961 16 1. l3 H. Sonntag and K. Strenge Koagulation und Stabilitat disperser Systertie VEB Deutscher l4 H. Sonntag Tenside 1968 5 188. l 5 A. D. Scheludko and D. Exerowa Kolloid-Z. 1960 168 24. l6 A. Watillon and A. M. Joseph-Petit Disc. Faraday SOC. 1966 42 143. l7 J. Netzel Diss. (Humboldt-Universitat Berlin 1968). l8 K. Strenge and H. Sonntag Tenside 1969 6 61. 2o A. P. Lemberger and N. Mourad J. Pharm. Sci. 1965 54 229. 21 Th. Steudel Forschung Fortschritt 1964 38 201. Verlag der Wissenschaften Berlin 1970 p. 43. 7a U. Havemann unpublished see ref. (1 3). Cockbain E. G. Trans. Faraday Soc. 1952 48 185.