General Introduction BY B. A. PETHICA Unilever Research Laboratory Port Sunlight Cheshire England Received 14th October 1970 These introductory remarks are intended to serve two purposes. In the first place the reasons for holding this Special Discussion of the Faraday Society are set out. Secondly a review of current problems in research into thin liquid films and boundary layers is given. During the past year the Council of the Faraday Society decided to form an Industrial Sub-Committee of the Standing Committee on Conferences. This decision was a recognition of the fact that research in industry is making a substantial contri- bution to the advance of those basic sciences which the Faraday Society is concerned to foster. It was also recognized that the motivation of industrial research towards the direct use of its results constitutes a powerful current stimulus to the advance of fundamental science in a variety of areas.The principal purpose of the Industrial Sub-committee is to propose topics for Discussions or Symposia which are of scientific merit and timeliness and of industrial significance. These topics may be adopted as part of the Society’s customary programme or may be the basis for Special Discussions or Symposia of which this meeting is the first example. This decision by the Society marks no lessening of concern with its traditional objectives and standards ; rather it marks the Society’s reassertion of the professional objectives common to academic and industrial scientists alike and its responsiveness to developments on the borders of physical chemistry.One such border is that between physical chemistry and the engineering sciences as demonstrated at this Discussion. The physico-chemical problems of industry have always been of interest to the Society as is evident from its contributions in catalysis. Thus new venture is in some ways a return to an early tradition as an examination of the titles of the Society’s Discussions back to the 1920’s will show. In 1920 the Society discussed Basic Slags and in 1921 the subject for debate was The Failure of Metals. There followed Discussions on The Physical Chemistry of the Photographic Process (1923) and Textile Fibres (1924). Coming to more recent times one of the topics for 1954 was The Physical Clzemistry of Dyeing and Tanning ; and of direct interest to us at this meeting the Society in 1948 discussed The Interaction of Water and Porous Materials.That Discussion contained some remarkable foretastes of our debates at this 1970 meeting and I can hardly do better to illustrate the joint scientific and industrial value of the Thin Liquid Film topic than to quote from the 1948 Discussion a remark by the late Prof. D. H. Bangham then Director of the British Coal Utilization Research Associa- tion and formerly Professor of Physical Chemistry in Cairo. “ In view of the large number of papers written on the subject of adsorbed water it is astonishing how little experimental work is directed towards ascertaining its properties. Most workers are content with being able to give a self-consistent account of a very small range of facts of which more than one explanation is possible.There are however awaiting solution a number of technical problems which turn upon the behaviour of these 7 8 GENERAL INTRODUCTION films and it is important that their nature should not remain merely a matter of conjecture ”.I Bangham went on to give a highly topical example of these “ technical problems ” relating to the adhesiveness of dust particles. We can easily add examples to make a formidable list of industrial problems which depend for their solutions in part at least on our understanding the properties of thin liquid layers-exampIes drawn from such areas as lubrication corrosion flotation foaming emulsion formation colloid stability wetting etc. Bangham’s remarkable pioneering work on poly- molecular liquid layers on solids is not as well known as it should be and it is sad to note that after his death the subject has languished among physical chemists in Britain until recently.There were many earlier indications that liquid layers on solids can have unexpected properties-for example in the work of Hardy in 1913-but we may regard Bangham and Razouk’s analysis of the contact angle equilibrium in 1937 as a turning point showing as it did that for non-zero contact angles the polymolecular layer adsorbed on the solid beyond the boundary of a liquid drop at saturation vapour pressure does not simply have the properties of the bulk liquid. At almost the same time Frumkin was drawing very similar conclu- sions in the U.S.S.R. The proposal that the formation of structured liquid layers near to solid surfaces is of importance in lubrication processes also has a long history.Griffiths in 1920 suggested that liquid surface layers possess a special rigidity Bastow and Bowden in 1931 among others taking the opposite viewpoint. Here again workers in the Soviet Union notably FuksY7 have interested themselves deeply in the mechanical properties of liquid boundary layers. The current broad interest in the properties of liquid boundary layers and thin films owes much to the development of quantitative theories of colloid stability particularly in the Soviet Union and Holland by Deryaguin and LandauY8 and Verwey and O~erbeek.~ These theories have facilitated a vast volume of quantitative research into surface forces and to some extent the present debate on the reality and significance of special structural properties in thin liquid films and boundary layers is the result of a growing realization that the accumulated data will not be covered by the interplay of two sets of forces alone (the electrical double-layer interactions and van der Waals’ forces) and of a certain impatience among some colloid scientists with the opacity of the now complicated corrections to a formerly elegant theory particularly so far as the electrical double-layer is concerned.On the other hand despite the many quantitative studies directed to elucidating the special properties of liquids in boundary layers and thin films particularly by Deryaguin and his co-workers considerable doubt remains as to the generality and extent of the alteration of liquids in these films. It must also be admitted that quantitative predictions of the effect of these layers in influencing colloid stability for example are not yet available.Progress towards such quantitative prediction will require further phenomenological investiga- tions and many exact studies of molecular and thermodynamic behaviour in films and boundary layers. To the extent that in dilute electrolyte systems the long-range nature of surface effects due to electrical double-layers and van der Waals’ forces is commonly accepted the issue is not simply one of the distance over which surface forces act in liquid layers but rather as to whether or not the liquid itself can play a more direct role than that of providing a fluid dielectric with convenient dispersion characteristics. In principle the situation can be covered by using more refined two-force models.Such models will take account of saturation effects ion volumes dipole terms asymmetry of polarizability etc. and hence they will include more directly the involvement of the solvent. The corresponding calculations of disjoining pressure B. A. PETHICA 9 and other measurable quantities will then include the energy and entropic contribu- tions due to the solvent. Nevertheless the role of solvent orientation is sufficiently distinct for Deryaguin to introduce the concept of “ forces of the third kind ’,.lo From the experimentalist’s point of view the importance of this concept is that it suggests new experiments designed to reveal the molecular situation at the liquid boundary thereby guiding further theory with a wider range of measurements than hitherto have been available.It is remarkable for example that until recently there has been so little interest in temperature effects in lyophobic colloid systems. An early exception is again in the work of Bangham’s group (Bond Griffiths and Maggs)? Temperature variations play a minor role in the classical DLVO theory but a major role in any model explicitly involving entropic layers and long-range ordering in the liquid. Without doubt some of these recent experiments on tempera- ture effects have been inspired by the “ordered liquid” theories and represent attempts to break away from the two-force theory. The time has come to give up our addiction to precise thermostatting at 25°C. If the phenomena are not very temperature sensitive why bother? And if they are sensitive it is the temperature coefficient that is the most interesting variable.In this Discussion we have a variety of measurements of temperature effects of direct relevance to deciding the existence and role of solvent layers and liquid structure changes in boundary layers. In the paper by Prins and van den Tempel the study of the effect of temperature changes on a free “equilibrium” soap film suggests that disjoining pressure equilibrium may not be the controlling factor in apparent film stability. The paper of Clunie and his co-workers on temperature and salt effects on a non-ionic stabilized foam system raises substantial doubts as to the meaning of the derived Hamaker constants which appear to vary by a factor of two over a 10 K range of temperature. The effects of temperature on liquid layers on quartz as described by Adlfinger and Peschel suggest a strong correlation between the disjoining pressure and ordering in the liquids as do the effects of temperature on viscosity in thin capillaries recorded by Churayev Sobolev and Zorin.in the simple well-defined graphon +liquid alkane system described in the paper of Ash and Findenegg the heats and excess volumes of wetting show temperature coefficients that indicate significant liquid structural changes at the boundary and the data for the graphon/water interface suggest extensive liquid structuring. The paper by Vincent and Lyklema deserves the special attention of colloid chemists since it gives evidence through temperature effects of some structuring of water at the surface of silver iodide sols-perhaps the most exhaustively studied colloid system.On the other hand the 1i.m.r. experiments over a range of temperatures on the aqueous polymer latex system reported by Clifford Oakes and Tiddy and those on the water + vermiculite system reported by van Olphen and his co-workers strongly suggest tight binding of small amounts of water with little evidence of extended alteration in the Brownian motion deep into the liquid layer. Clifford’s results show that the extended structural effects in water near to polyvinyl acetate latex particles reported earlier from the same laboratory,12 are most likely caused by the fibrillar and porous nature of the surfaces of those latices. The polymer latex data show clearly the importance of the precise characterization of the solid surfaces contacting the liquid layers. Of the surface layers discussed in the various papers at this meeting the mica surface studied by Bailey Price and Kay the layered vermiculites and clays discussed in several papers the silver iodide sol surface and the graphon interface are perhaps the best characterized of the solid+ liquid systems.The free-standing liquid films have been customarily regarded as a reliable model system but Prins and van den Tempel may cause us to revise this 10 GENERAL INTRODUCTION opinion,. Surfaces of solid quartz steel and rubber are less well characterized. Quartz has been widely studicd but the surface characterization is ambiguous i n most cases. An interesting illustration of this point is that the heat of immersion into water of un-annealed Aerosil silicas with a range of surface hydrations shows a striking dependence on the temperature of immersion suggesting strongly that un-annealed silica can induce significant structural effects in the local water.When the silica powders are annealed and rehydrated at the surface the temperature variation of the heat of immersion is no longer present for silicas annealed in air (in the presence of a small vapour pressure of water) but partially remains for silicas annealed in vacuo.13 Results of this kind show that short-range surface effects have a profound influence in " triggering-off" structural changes in the local liquid layers and that we should be cautious of supposing that where these effects occur they result from long-range forces entirely. This same point comes out clearly from numerous results on the thermodynamic functions for the adsorption of vapours on solid surfaces.It is well known that the B.E.T. equation as commonly applied to vapour adsorption assumes that the second and successive layers of adsorbed vapour have bulk-liquid properties and to the extent that the B.E.T. equation is successful it gives support to the view that polymolecular liquid layers on many solids have no unusual structure. Even for systems in which the B.E.T. equation gives a good fit however the experimentally determined differential heats and entropies of adsorp- tion do not usually become indistinguishable from the bulk-liquid values until several layers are adsorbed. It is almost certain that increased accuracy in the determination of these parameters would show that very many layers are necessary before true bulk-liquid properties are reached and we should never forget that even if the deviations in molar functions are small the liquid concentrations are high.The detailed preparation of the solid surface has a profound effect on the energetics of the physical adsorption of vapours and necessarily therefore on the behaviour of polymolecular layers [see e.g. Holmes 14]. Another set of considerations are provoked by taking the view that the solid/liquid and fluid/liquid interfacial systems discussed at this meeting represent extreme examples of solutions in the liquid phase. The properties of the solvent take pride of place in this approach and many suggestive correlations come to mind from considerations of solution and liquid-state thermodynamics. This approach is illustrated in part by Adlfinger and Peschel and by Ash and Findenegg.The copious data on the properties of solutions of ions inert gases polar and amphipathic molecules suggest that a variety of structure-making and -breaking effects will be manifest at surfaces. Extrapolating from thermodynamic and n.m.r. studies of surfactant micelles for example one would not expect extensive long-range solvent orientation changes at the surface of oil in water emulsions stabilized by highly charged surfactants. The variations of water-structuring effects with surface charge reported by Vincent and Lyklema are strongly reminiscent of short-range solvent polarization effects in aqueous electrolyte solutions. Perhaps most interesting of the relevant speculations that arise from considera- tions of bulk liquid thermodynamic properties is that the term " liquid " includes " liquid crystals " and a variety of systems involving the reversible aggregation of amphipathic molecules.The phase-rule relationships in these bulk nematic and smectic systems have been worked out for many examples and the use of expressions such as " surface liquid phase " may in some instances be well justified. In this meeting Drauglis Lucas and Allen develop the smectic phase analogy for lubricating films of fatty acids in hydrocarbon oils and Clifford and his co-workers remind us B. A . PETHICA 1 1 of the relationships between lamellar soap phases and thin foam films. The paper from Haydon's laboratory is a valuable contribution on several scores-it provides data on thin oil layers stabilized by a chemically defined and thermodynamically characterized solute in the oil.A striking result is that this stabilizer glyceryl mono- oleate (which is incidentally related to well-known emulsion stabilizers used in the food industry) shows reversible micellar aggregation in hydrocarbon solutions. Furthermore these black oil films form typically at concentrations near to or above the inicelle point of the stabilizer which is a common situation also in the formation of black films from aqueous surfactant solutions. These considerations suggest that the sharp salt-induced transition (recorded by Clunie and co-workers) between the first and second black films stabilized by a non-ionic surfactant is in fact a phase change which should have a direct parallel in bulk solutions of the stabilizer. These suggestions take us directly to the point that even if the forces controlling the thinning of free-standing thin films are correctly represented by the interaction of electrical double layers and van der Waals' forces in planar (or laterally extended) geometric arrangements we must still enquire as to the causes of the planar arrange- ments.Smectic and nematic phases provide one of the keys to answering this question. The thermodynamic stability of liquids which are either themselves molecularly asymmetric and flexible or contain such asymmetric solutes involves a balance of conformational entropy and energy terms deriving in part from geometrical considerations and in part from " chemical " factors (asymmetric polarizabilities of different groups etc.). This balance is necessarily modified in the vicinity of the asymmetry we call a surface and our Discussion will go far in showing whether and where these structural changes are significant in a wide range of experimental situations.' Bangham Disc. Furuday SOC. 1948,3 102. Hardy Proc. Roy. Sac. A 1913,88 313. Bangham and Razouk Trans. Faruduy SOC. 1937,33 1459. Frumkin Zhur. Fiz. Khim. 1938,12 33. Bastow and Bowden Proc. Roy. Soc. A 1931 134,404. vol. 1 p. 79. ' Griffiths Phil. Trans. A 1920 221 163. ' Fuks Research in Surface Forces ed. Deryaguin (Consultants Bureau New York 1964) Deryaguin and Landau Actu plzysicochim. 1941,14,633. Verwey and Overbeek Theory of the Stability of Lyophobic Colloids (Elsevier Amsterdam 1948). ' O Deryaguin Research in Surface Forces ed. Deryaguin (Consultants Bureau New York 1964) '' Bond Griffiths and Maggs Disc. Faruday SOC. 1948,3,29. ' l 3 Tyler Taylor Pethica and Hockey Trans. Furaduy SOC. in press. l4 Holmes The Solid-Gas Interface ed. Alison Flood (Edward Arnold & Marcel Dekker 1967) *' Clifford and Pethica Trans. Faruduy Soc. 1965 61 182. vol. 1 p. 3. Johnson Lecchini Smith Clifford and Pethica Disc. Faraday SOC. 1966,42,120. vol. 1 p. 127.