Table of contents

A striking number of new spatial structures or modes of self-organization in liquids have recently been discovered. Such a new structure can be expected to give rise to further new properties in the liquid, which are not apparent from the structure itself. To illustrate such 'hidden consequences', two instances of heterogeneous polymer systems are reviewed. The first is the associating polymers, whose heterogeneity causes them to stick together at specific points. This gives rise to unusual consequences in the solubility of the polymers and in the rheology of their solutions. The second instance is diblock copolymers, whose heterogeneity gives rise to extended periodic domains. The elastic free energy stored in the elongation of the polymers in these domains qualitatively alters the motion of these chains and the response of the liquid to applied stress.

Chemical dynamics is the dynamics associated with the chemical rearrangements of atoms to form products from reactants. Such processes are infrequent but the important ones to understand are the kinetics of chemistry. Transition state theory provides a classical perspective with which one may focus on these rare but important events. Through two examples, this paper describes some of what can be learned with this perspective and its dynamical and quantum-mechanical generalizations.

Reviews some recent theoretical and computer simulation studies of simple atomic fluids adsorbed at structureless substrates. Emphasis is placed on phase transitions, especially the various types of wetting transition. Criticality is associated with capillary-wave-like fluctuations in a continuously growing wetting film. This is of a subtle nature, which is best understood in terms of the pairwise correlation function of the fluid. Other surface phase transitions, such as prewetting and layering, occur out of bulk coexistence. Theory suggests that for sufficiently attractive substrates a sequence of first-order transitions, corresponding to the growth of new adsorbed liquid layers, should occur as the pressure of the bulk gas increases towards saturation at temperatures not too far above the bulk triple point. The extent to which such behaviour is found in adsorption experiments is discussed. The authors also argue that a simple fluid confined between two parallel hard-walls can exhibit surprisingly rich equilibria.

Experimental results for fluid metals near the liquid-vapour critical point shown that profound changes in the electronic structure of fluid metals occur in that region. A gradual transition from metallic to non-metallic behaviour occurs with decreasing density, which manifests itself in a correspondingly strong thermodynamic state dependence of the effective interparticle interaction. It is shown that this strong dependence noticeably influences the thermodynamic and kinetic features of the vapour-liquid phase transition of fluid metals. Special emphasis is given to the interplay between the critical point density fluctuations and the change in the electronic properties in the course of the metal-non-metal transition.

The onset of chaotic behaviour and space patterning in chemically reacting systems is analysed. Special emphasis is placed on the specificities of chemical dynamics as compared to other branches of physical sciences giving rise to chaos and self-organization and on the role of the size of the system as a universal bifurcation parameter.

It is argued that powders have so many particles per unit volume that they can be treated in a manner similar to conventional liquids. They have an entropy S(V,N), but as energy is not important the place of temperature delta E/ delta S is taken by delta V/ delta S. Equations capable of giving plug flow are derived. It is argued that highly viscous liquids approaching the glass transition can have a structural order that differs from that of equilibrium at the ambient temperature, defined by the other majority degrees of freedom. The ideas from powder theory enable one to derive the dependence of the glass temperature on the cooling rate.

The relaxation of electric birefringence is experimentally investigated in three distinct systems: critical binary mixtures, polydisperse micellar solutions and dilute polyelectrolyte solutions. The asymptotic behaviour is shown to be consistent with a stretched-exponential form exp(-(t/ tau )^alpha ). Most of the data are quantitatively explained by a simple model of parallel relaxation of many independent processes characterized by a wide distribution of sizes.

The basic mechanisms which take place during the displacement of immiscible fluids in porous media have been observed in micromodels and have been modelled. At the pore level, in drainage, the invading fluid chooses the largest throat. In imbibition, the displacement depends on the local geometry. For a large pore-to-throat ratio (aspect ratio), the main mechanism is the collapse of the invading fluid in the smallest channel, without entering the pore. For a small aspect ratio, the wetting fluid invades the pore first, and then the adjacent channels. From observations at the pore level, the author has modelled the displacement on a large scale in some extreme cases by using statistical theories. The different behaviours are then displayed as domains in three phase diagrams: one for drainage and two for imbibition (large and small aspect ratios). At a high rate, when viscous forces are dominant, all the diagrams show a stable domain (described by anti-DLA) and a viscous fingering domain (DLA). In drainage, low capillary numbers lead to capillary fingering represented by invasion percolation. In imbibition, the capillary domain is described either by a compact cluster growth (small aspect ratio) or percolation theory (large aspect ratio). In addition the possibility of flow by film along the roughness of the walls leads to disconnected structures.

Using the surface forces apparatus technique the authors have measured the dynamic properties of ultrathin liquid films between two molecularly smooth solid surfaces sliding past each other. The results on several different liquids are reviewed; these reveal film properties that are profoundly different from those of the bulk liquids once the film thickness falls below five molecular diameters. For example, the liquids can now support a normal hydrostatic pressure as well as shear stresses, exhibiting upper and lower yield points (stick-slip friction). Certain molecular rearrangements can take 10¹⁰ times longer in a 10 AA film than in the bulk liquid. The experimental results are reproduced by computer simulations which indicate that liquid molecules in ultrathin films become ordered and solid-like, 'freezing' into discrete layers which also have lateral order. On applying a shear force, the film undergoes a melting transition from 'solid' to 'liquid' at the yield point. But even during slip some order remains within the film which therefore becomes more like a liquid crystal than a simple liquid. These phenomena and thin film properties do not appear to be readily describable in terms of mechanisms or concepts applicable to bulk liquids or solids.

During the last decade, important new information has been gained on the details of the dynamics in atomic liquids and gases on the molecular scale. This has primarily been the result of an intensive interplay between theory and improved inelastic neutron scattering (INS) experiments and computer molecular dynamics (CMD) simulations. A fruitful step in a better description of the dynamics of fluids has been the introduction of the concept of effective eigenmodes that can be determined from either hydrodynamic or kinetic theory or from a fit to observed (INS) dynamic structure factor or (CMD) time correlation functions. In this case a density fluctuations is built up from five separate contributions (lines) that take into account all possible correlations between five microscopic quantities (density n, velocity u, temperature T, longitudinal stress tensor sigma , and heat flux vector q). The application of this model is discussed for INS results on argon and CMD results of a Lennard-Jones fluid, recent INS experiments on fluid helium (at 13-39 K and 40-200 bar), where the dynamics are more complicated and strongly temperature dependent, and also recent neutral Brillouin scattering data on low-density ³⁶Ar gas.

Quasi-elastic incoherent neutron scattering data on liquid water are analysed in terms of two dynamical processes. They give the residence time and the characteristic lifetime of a hydrogen bond. Percolation theory for low temperature water relates the Arrhenius behaviour of the hydrogen bond temperature dependence to the anomalous transport properties of liquid water.

Femtosecond impulsive stimulated scattering (ISS) experiments were performed on liquid carbon disulphide as a function of pressure at room temperature in a diamond anvil cell. The subpicosecond response of the liquid becomes weakly oscillatory as pressure is raised from one atmosphere to several kilobars, indicating microscopic dynamics that is vibrational in character at short times. Molecular dynamics simulations were performed to study the contribution to the ISS signal from single-molecule reorientational motion. Even at very high densities the reorientation of single molecules in the computer liquid is insufficient to explain the oscillations in the ISS signal.

A review is given of how molecular dynamics methods have been modified to perform simulations in the constant-temperature condition. One usually considers a system which is thermally connected with a huge external system (a heat reservoir) to describe a canonical ensemble in statistical mechanics. The way in which this situation is reflected is a key factor for simulations under isothermal conditions. The total kinetic energy is kept to a constant value in constraint methods. In stochastic methods, interactions with a heat bath are treated as random collisions with hypothetical atoms or random forces acting on particles. In the extended-system method, a degree of freedom which mimics a heat bath is introduced, and the total energy of a physical system is allowed to fluctuate.

The authors discuss a differential approach to the theory of fluids, the hierarchical reference theory, which, above the critical temperature, has been shown to be (i) as accurate as the most widespread theories of liquid state in the high density region and (ii) able to reproduce the renormalization group results in the critical region. In this region it predicts both the universal and the non-universal quantities. The authors have studied the Lennard-Jones fluid in detail but the method can be directly applied to more realistic interactions between molecules. The treatment of temperatures below the critical one presents some additional difficulties due to the presence of the inhomogeneous two-phase region. Preliminary results indicate that the theory gives the coexistence curve with the correct scaling behaviour without any need for an ad hoc Maxwell construction. The extension of the formalism to binary mixtures is under way.

By performing careful observations very near the critical point of binary fluids or microemulsions using optical microscopy it is possible to obtain a resolution of the order of the correlation length and observe fluctuations in the order parameter (concentration). The origin of these fluctuations is discussed by comparing the picture element to a spin block variable within 3D Ising model. It follows that the free energy of the configuration can be obtained from the histogram of the fluctuation amplitudes. Black and white domains can be defined by clipping these fluctuations relative to a mean value. Domains are seen to be self-similar in shape, with a fractal dimension of 2.8. The origin of this self-similarity is discussed and a possible relation with percolation model is envisaged.

The generalized effective liquid approximation (GELA) to the density functional theory of classical nonuniform systems reproduces all the formal properties of the free energy and requires only the direct correlation function of the uniform system as input. In the case of the freezing of hard spheres very accurate free energies, pressures and fluid-solid coexistence data can be obtained from the GELA. The theory predicts, besides the equilibrium FCC solid, metastable BCC and SC phases also.

The density functional theory of freezing has been recently extended to treat quantum liquids. The authors consider three approximations of the weighted-density type for the freezing of jellium at T=0. The agreement with a simulation of the predicted freezing parameters varies from satisfactory to very good, depending on the approximation.

The theoretical calculation of the properties of ionic solutions is discussed from the viewpoint of numerical simulations. The authors consider the simple case of the mixture of a solvent of polar molecules and ions and also, briefly, the case where both solvent molecules and ions are polarizable.

Recent experiments aimed at enhancing the understanding of aqueous ionic solutions are summarized, and some possibilities for future studies are outlined.

The structure and dynamics of nonaqueous solutions are discussed. Specific properties of these solvents are investigated, especially by neutron scattering experiments, and details are given of their technological applications.

The authors present new data on H₂+Xe mixtures that support the existence of a fast sound mode in binary gas mixtures. In addition they show that using an appropriate scaling, the behaviour of the slow sound mode in systems with different mass ratios, i.e. H₂+Xe, H₂+Ar and He+Xe, is identical.

A very short review is given of the methods and models applied in the numerical simulation of hydrated electrons. The results for the energetics, the diffusion and the transient dynamics are compared with experimental data. While there is good qualitative agreement, the quantitative discrepancies are still serious. One of the several effects, which are treated only in an approximate way in most calculations, is considered more closely, namely the coupling to the collective induced polarization of the solvent. The author presents new results for the structure and energy, indicating that a self-consistent treatment of electronic polarization is important but is not likely to resolve the discrepancy with experiment.

The chemical potential is a key quantity in the theory of chemical equilibria and solvation processes in the liquid phase. Several approaches, all equivalent in principle, were proposed in the literature to evaluate this quantity by computer simulation. It is shown for some examples that these methods converge to the same values for molecular liquids but yield different values for ionic solutions. This discrepancy is discussed in detail.

The elucidation of detailed mechanisms of ultrafast events that occur in molecular charge transfer or reaction dynamics has been made possible by recent advances in spectroscopy techniques that use ultrashort laser pulse generation. Ultrashort laser pulses (100 femtoseconds duration, 1 fs=10^-15 s) allow initiation of selective photochemical processes (photoejection of epithermal electrons), and obtaining of unique information on the dynamics of primary steps of radical reactions involving ultrafast electron or proton-transfer: formation of the hydration cage around an electron, encounter pair formation, ion-molecule reactions. Recent investigations on the non-equilibrium reaction processes in the universal protic solvent (water) are presented.

The Raman spectra of reacting molecules in liquids can yield information about various aspects of the reaction dynamics. The author discusses the analysis of Raman spectra for three prototypical unimolecular reactions, the rotational isomerization of n-butane and 1,2-difluoroethane, and the barrierless exchange of axial and equatorial hydrogens in cyclopentane via pseudorotation. In the first two cases the spectra are sensitive to torsional oscillations of the gauche conformer, and yield estimates of the torsional solvent friction. In the case of cyclopentane, the spectra can be used to discriminate between different stochastic models of the pseudorotation dynamics, and to determine the relevant friction coefficients.

Raman depolarized and isotropic spectra of water at various temperatures in the 0-1600 cm^-1 range are discussed in terms of interaction-induced mechanisms involving the polarizabilities of an isolated molecule. The comparison between experimental and computer simulation results is also shown.

The aim of the work is to characterize by quasielastic neutron scattering (QNS) not only the usual transport processes occurring in aqueous solution but also something specific to the high proton conductivity in acidic solutions. The addition of a strong acid such as H₂SO₄ or HNO₃ to water provokes a marked slowing down of the mean translational diffusive motions and, to a lesser extent, of the rotational ones. Furthermore, the authors have evidenced a weak and broad additional QNS component which is not present in pure water. However, the presence of this component also in concentrated salt solutions eliminates the hypothesis of a fast proton transport process and indicates that some kind of relaxation process of the ionic environment could be involved.

The dynamic properties of water molecules coordinated to simple ions (alkali and alkali-earth halides) and hydrophobic ions (tetraalkylammonium (TAA) halides) in supercooled solutions have been investigated with NMR. The study of spin-lattice relaxation rates and self-diffusion coefficients as functions of temperature, pressure Larmor frequency and concentration reveals characteristic features of molecular motions close to the low-temperature limit of the metastable phase (percolation transition and glass transition) and provides certain details of the local arrangement of water molecules in the coordination sphere of these ions. The intramolecular flexibility of the alkyl chains of the hydrated TAA cations has been investigated also. The dynamics of the methyl group reorientation provide a sensitive probe of structural differences in these solutions.

A mode-coupling theory for supercooled liquid dynamics exhibits bifurcation singularities which cause a temperature T_c for the crossover from liquid to glass behaviour. Near T_c the long-time relaxation is a two-step process related to the experimentally known alpha - and beta -relaxation in structural glass formers. The dynamical anomalies predicted by the theory for the two processes are reviewed.

The various routes by which a liquid may form a glass have recently been related to the structural properties of the glass, at one extreme represented by network glasses of strong covalent bondings and at the other by typical non-network systems of weak van der Waals or Coulomb bonding types. The authors examine the success of such an approach by comparing data on structural relaxations in glass-forming liquids chosen to cover a wide range of behaviour from non-network ionic and molecular systems to a network forming oxide glass. They also include data of a glass-forming polymeric system. While there seems to be a strong correlation between increasing fragility (i.e. decreasing temperature and resistance in the medium-range order) and increasing departure from Arrhenius behaviour of the relaxation time, there is no simple relation between the degree of fragility and non-exponentiality as has been suggested.

A series of neutron diffraction experiments have been used to investigate the properties of liquid, solid and amorphous sulphur. The liquid was studied from 131 degrees C to 230 degrees C, i.e. at conditions around the λ -transition (T_λ=159 degrees C). The amorphous sample was studied at -196 degrees C and the crystalline solid at room temperature. The data have been transformed to give an accurate pair correlation function g(r) from which the absolute numbers of atoms in given molecular configurations may be determined for distances less than 8 AA. A comparison of the solid and liquid g(r) data indicates which intermolecular bonds are broken on melting. The liquid data seem to be inconsistent with the picture of 'crown'-shaped S₈ molecules below T_λ , and with the proposed S₈-to-polymer transition at T_λ. Also it has been found that the diffraction pattern may be analyzed as though the liquid was an assembly of roughly spherical molecular units. In this case each unit would contain about six atoms, and the λ-transition could be related to the percolation limit for these units. The amorphous material was studied at the quenching temperature of -196 degrees C and its structure is intermediate between that of the parent liquid and the crystalline solid. It is concluded that a fundamental diffraction analysis of the states of sulphur has thrown new light on long standing problems.

A scheme to invert the conductivity and thermopower data to yield the fundamental parameters for liquid semiconductors is presented. As an illustrative example the method is applied to liquid thallium chalcogenides. Some comments on the thermoelectric figure of merit are included.

The authors describe a recently proposed molecular dynamics scheme, which allows simulations using forces obtained from accurate quantum mechanical calculations. They apply this scheme to molten KSi, which is known to form Zintl-like ions (Si)₄^4- in the solid state. It is shown that these complexes tend to lose their identity in the liquid phase, and that Si atoms form an extended network with threefold-coordinated sites.

The first-order difference method of neutron diffraction is applied to molten GeSe₂ at 784+or-3 degrees C. The results show that Ge correlations contribute to the first sharp diffraction peak of the total structure factor, give a nearest-neighbour distance r_GeSe=2.40+or-0.02 AA and give an average coordination number of 3.7+or-0.2 selenium atoms around a germanium atom.

The authors present first-principles calculations of the electronic structure of molten simple and transition metals.

The Knight shift K and quadrupolar relaxation rate R_q in liquid metallic systems, in which effects of bonding become increasingly prominent, are surveyed. In Rb, a theoretical calculation of R_q, including mode-coupling theory for the liquid, and the r-dependent Sternheimer factor, predicted closely the recent experimental redetermination. In Ge and in Cu-Ge and similar nearly free-electron systems, the quantitative analysis of K still poses problems, while qualitatively K(x) displays clearly a correspondence to the resistivity maximum. In metallic alloys with compound forming tendency, models based on an association (A+B from or to AB) connect K and R_q quantitatively with the heat of mixing, but the microscopic foundation of the association ansatz is uncertain.

In a neutron scattering experiment on liquid caesium a positive dispersion of the collective modes has been observed, indicating the existence of shear relaxations in the liquid. The measured solid-like high-frequency sound propagation with velocity c_x>c_s-with c_s being the well known adiabatic velocity of sound-yields the experimental determination of the shear modulus G_x in the fluid for the first time and confirms the evidence that the observed positive dispersion of collective modes is due to viscous shear relaxation beyond the hydrodynamic limit of a dense liquid near the melting point.

The physical processes responsible for the 'interaction-induced' polarizabilities in ionic melts are investigated by electronic structure calculations. These are built into a computationally tractable model for the polarizability of the melt whose correlation functions are then calculated by molecular dynamics simulation. The experimental light scattering data are quantitatively reproduced. A theory of the lineshape is verified with the simulation data. The influences on the lineshape of the various dynamical modes of the ionic system (e.g. the plasmon mode) are thereby assessed.

The modification of a technique that was developed to study time correlations in lattice-gas cellular automata to facilitate the numerical simulation of chain molecules is described. As an example, the calculation of the excess chemical potential of an ideal polymer in a dense colloidal suspension is discussed.

Phase transitions among paraelectric, ferroelectric, ferrielectric and antiferroelectric phases in a chiral smectic liquid crystal, MHPOBC, have been investigated by means of circular dichroism (CD). It was found that the helix undergoes a complex variation through the transitions such as discontinuous jumps and handedness change. By considering the twisting power, the ferrielectric phase was found to be composed of ferroelectric and antiferroelectric structures in the ratio of 3:7. The finite induced CD signal in SmC_alpha * and SmA in the vicinity of the transition point suggests the helical structure in these phases.

The authors consider a tilt grain-boundary with small angle theta in an incommensurate smectic A_ic liquid crystal with ordering wavenumbers q₁, q₂ and sample thickness h in the direction normal to the layers. For q_i theta h<<1, a limit well within the reach of experiment, they find that the dislocation array which forms such a boundary has a spacing D approximately h^1/5 theta ^-4/5. This prediction, which is qualitatively different from that obtained for periodic smectics by Nallet and Prost, should serve as a decisive test for the existence of phasons and hence for the truly incommensurate nature of these systems.

The anisotropy of the viscosity and the flow alignment of nematic and nematic discotic liquid crystals as well as that of oriented ferro-fluids are described by the same set of viscosity coefficients. Results obtained from NEMD computer simulations are reported for completely aligned nematics and for perfectly oriented ferro-fluids.

The authors have previously found that saturated phospholipids such as phosphatidyl-ethanolamines can, in certain cases, adopt as many as three different inverse bicontinuous cubic phases in water, of probable space groups Ia3d (No 230), Im3m (No 229) and Pn3m (No 224). They found that these cubic phases could be induced to appear by reducing the chain length or by increasing the hydrophilicity of the headgroup of the phospholipid molecule. All of these cubic phases are located in the phase diagrams between the lamellar and the inverse hexagonal (H_II) phases. They now report the observation of a novel inverse face-centred cubic phase, of probable space group Fd3m (No 227), in two different systems of hydrated binary lipid mixtures. One of these systems consists of mixtures of phosphatidylcholine with diacylglycerol; the other is an acid-soap mixture of an unsaturated fatty acid with its alkali salt. This Fd3m cubic phase in both systems occurs between the inverse hexagonal (H_II) phase and the inverse micellar solution (L₂), with increasing concentration of the lipid component with the less strongly hydrophilic headgroup. They surmise that the average mean curvature of the polar/nonpolar interface in this Fd3m cubic phase is more negative than that of the neighbouring H_II phase; this is quite different from the inverse bicontinuous cubic phases, where it has a value intermediate between those of the lamellar and H_II phases. They conclude that the structure of this Fd3m cubic phase most probably consists solely of closed inverse micellar aggregates.

The authors have demonstrated that it is possible to include tiny magnetic particles into different types of lyotropic phases, such as sponge, microemulsion or lamellar phases. The first point of interest in these results is to prove the compatibility between solid colloids and organized liquids. As for the hybrid lamellar phase, they have studied its phase diagram versus the smectic period and the particle concentration-which are the two relevant parameters-and deduced its range of stability. Moreover, this ferrosmectic phase exhibits original features when subjected to a magnetic field even when it is very low: the lamellae orientate in the direction of the field. The detailed mechanism of this strong coupling between the spherical particles, the flexible membranes and the magnetic field is not fully understood, and deserves further experimental and theoretical study.

The authors review some recent MD simulations for polymer melts. For the melt chains of up to six entanglement lengths are taken into account. The data strongly support the reptation picture.

Branched polymers are most usually synthesized in the vicinity of a sol-gel transition and are very polydisperse. The number distribution was shown experimentally to be the same as in percolation. This may be described by two characteristic masses, N_z and N_w, that diverge with different exponents. Similarly, one finds that there is a continuous distribution of relaxation times, varying as a power law cut-off at large times by an exponential decay. In the reaction bath, this distribution is related to the viscoelastic properties. In a dilute solution it may be related to the distribution of masses. In both cases, two diverging times may be defined. The author reviews the properties of these distributions of relaxation times, and considers the consequences on the relaxation of branched polymers in the reaction bath and in solution.

The depolarized Rayleigh spectra I_VH( omega ) of the solvents di-2-ethylhexylphthalate (DOP) and polychlorinated biphenyl (A1248) in undiluted state and of the solutions A1248/polystyrene (PS) (c_PS up to 15%) and DOP/PS (c_PS up to 75%) were measured over the temperature range 30-140 degrees C using Fabry-Perot interferometry (FPI). The experimental I_VH( omega ) of the undiluted solvents were well represented by a single Lorentzian whereas two Lorentzians were found necessary to fit the I_VH( omega ) of the PS solutions. The variation of the intensity ratio of the two Lorentzians with temperature and the increase of the orientation time tau ₀ of the solvents, upon addition of PS, show evidence of polymer-induced modification of solvent mobility. The rate of change of tau (c,T)/ tau ₀ with c_PS, which is a measure of the solvent local friction assumes similar values for both DOP/PS and A1248/PS solutions at high temperatures.

The recent analysis of osmotic pressure in a semi-dilute solution of strongly stretched polymers is used to study the uniaxial stretching of a swollen polymer gel, in a good solvent. The resulting steady-state stress-strain relation complements Daoudi's analysis of the transient deformation of polymer gels.

The structure of layers of poly(dimethylsiloxane) terminally grafted on silica has been studied by small angle neutron scattering. The authors have observed the influence of the polymer molecular weight, the grafting density and the solvent quality. At a high grafting density, in good and in bad solvents, the chains are stretched and the thickness of the layers is proportional to the polymer mass. In a good solvent the inner structure of the layers is similar to that of a semi-dilute solution of free polymer at the same local concentration, in agreement with the description of Alexander.

The author discusses the nature of the initial interactions as two polymer-bearing surfaces in a good solvent approach each other, as measured directly by the mica force apparatus. It is suggested that, while for grafted chains and for adsorbed chains at low surface coverages these initial interactions are probably well described by constrained-equilibrium models, this may not be the case for adsorbed chains at high surface coverages.

The formation of a small daughter vesicle from a large mother vesicle is considered, the two being connected by a narrow constriction. This budding is achieved by either decreasing the enclosed volume or increasing the membrane spontaneous curvature of a nearly spherical vesicle while the other quantity and the membrane area are constant. The activation energy of the discontinuous transition is identified and found to decrease to zero as the respective parameter change progresses.

The authors present experimental results on microemulsions made with water-alkane nonionic surfactant systems that confirm the close relationship between the maximum characteristic size in the microemulsion and the persistence length of the surfactant layer. The microemulsion structures are found with surfactants that form films of small bending elasticity. When the bending elasticity is too large, ordered lamellar phases are obtained. When it is too small, the surfactant film cannot form, and the medium is a structureless molecular mixture. This evolution is associated with a wetting transition.

Surfactant molecules in dilute solution may aggregate reversibly into extended structures. For suitably chosen molecules, the preferred packing involves a locally flat bilayer which tends to wander entropically at large distances. At low temperatures (and/or high concentrations) the system forms a stack of flat sheets with one-dimensional quasi-long range order (a smectic liquid crystal), but at high temperatures or low concentrations, the stack can melt into a random surface structure that resembles a multiply connected labyrinth or 'sponge' of bilayer in a sea of solvent. Recent theoretical and experimental progress in understanding the properties of the sponge is reviewed. The authors argue that the sponge phase may provide a good system for the study of various liquid-state critical phenomena.

The bending elastic modulus of monolayers is deduced from the study of their thermal fluctuations on a low scale (about 100 AA) by optical techniques: X-ray reflectivity measurements and ellipsometry. The author explains why ellipsometry is more sensitive than reflectivity at small scales.

The authors have performed an extensive series of dynamic light scattering measurements of a three-component water-in-oil microemulsion system, consisting of AOT/water/decane, throughout the isotropic one-phase region at water to AOT molar ratio X=40.8, covering wide ranges of volume fraction phi and temperature T above and below the percolation threshold. The droplet density correlation function at long time has a nonexponential form for phi >0.4 and below the percolation threshold. When fitted to a stretched exponential function the apparent exponent beta , starting at about 0.7 at the lowest temperatures, increases continuously and approaches unity at the percolation temperature and above. The initial and the average decay rates of the correlation function, as a function of temperature, show an abrupt change of slope at the percolation threshold.

The authors examine theoretically the effect of electrostatics on the self-assembly of charged cylindrical micelles which behave as living polymers. The growth of micelles as a function of increasing surfactant and/or electrolyte concentration exhibits three distinct regimes. The most striking feature of the growth law is the existence of a dilute regime, (i), in which the average micelle size varies slowly with concentration. At high concentrations, regimes (ii) and (iii) are characterized by more rapid growth than for neutral micelles. This may be responsible, in part, for the anomalous scaling of rheological properties, as observed in recent experiments.

The authors report on synchrotron X-ray studies of the nematic (N) and the smectic-A (SMA) phases under nonequilibrium 'steady state' shear flow conditions. Under shear, the presence of SMA fluctuations leads to a novel flow-induced fluctuation force on the nematic director n which alters its equation of motion. This leads to rich behaviour where the nematic phase exhibits a sequence of regimes in which the orientational phase space (OPS) explored by n evolves as the N-SMA transition is approached. They directly observe the critical slowing down of the SMA order parameter fluctuations through the X-ray profiles which give the intensity map of the time- and space-averaged OPS traversed by n. The data are consistent with the classical Ericksen-Leslie-Parodi theory of nematodynamics away from the immediate vicinity of the N-SMA transition temperature. Closer in, however, fluctuation effects dominate and a model of critical nematodynamics has to be considered. The experiments demonstrate that synchrotron scattering techniques may be used as effective structural probes of dynamical systems.

Various aspects of the behaviour of essentially hard spherical colloidal particles, suspended in a liquid, are outlined. The authors consider the phase behaviour and crystal structure of one- and two-component suspensions and the glass transition of a one-component system.

Structural and dynamic properties of suspensions of colloidal particles have been studied extensively by static and dynamic scattering experiments. The theoretical analysis has been performed by applying and extending equilibrium and non-equilibrium theories of simple liquids to the case of interacting Brownian particles. A fairly well developed understanding of monodisperse systems has emerged for highly charged colloids and for systems with very short-range interactions. In this contribution two extensions will be discussed: (i) static scattering from polydisperse charge-stabilized systems, and (ii) the collective diffusion of weakly charged monodisperse particles, for which hydrodynamic 3s well as electrostatic interactions are of importance.

The authors measure the instantaneous atomic structure and dynamics of a three dimensional (3D) fluid near crystallization in contact with a smooth wall. The 'atoms' are monodisperse highly charged submicrometre spheres in colloidal suspension in water. The authors simultaneously follow the positions of about 2000 spheres using video microscopy, taking snapshots at intervals close to the collision time of the fluid. They find that the dense fluid is layered in the direction perpendicular to the wall, with at least four distinct layers. They compare the microscopic particle dynamics of the first layer of fluid with those of a two dimensional (2D) layer of spheres at the same density but rigidly confined between two smooth walls, on a time scale at which over 96% of the spheres remain in the first 3D layer. They find dramatic differences in the correlation lengths and times, the defect structure and dynamics, the particle trajectories and the self-diffusion times for the two cases.

Hard spheres serve as one of the basic models in classical equilibrium statistical mechanics, as well as, in fluid mechanics. Sterically stabilized polymethylmethacrylate (PMMA) spheres suspended in certain organic solvents have interparticle interactions which approximate the hard sphere interaction. Thus the equilibrium properties of PMMA particle suspensions should correlate with theoretical and computer simulation results of pure hard sphere systems. However, non-equilibrium properties must be compared with theories which include the hydrodynamic effects of the suspending medium. Experimental results are presented which suggest hard sphere behaviour: a solid-liquid phase transition, equilibrium crystal structure, liquid and crystal sedimentation velocities. Non-equilibrium microstructure in steady and oscillatory shear flows is then examined using light scattering from samples where the solvent index of refraction matches that of the particles. Oscillatory shear flows of the proper strain amplitude can shake crystal-like order into an equilibrium liquid-like sample. These results may be understood in terms of a simple hard sphere model.

While static and dynamic light scattering experiments have been a major source of information about the structure and dynamics of dilute colloidal systems or macromolecular solutions, their application to more concentrated samples is severely limited by multiple scattering. The authors' dual-colour photon correlation experiment achieves very efficient isolation of singly scattered light by cross-correlation over the full range of scattering angles between 20 degrees and 140 degrees . The dual-colour cross-correlation technique immediately yields dynamic structure factors, which are not affected by multiple scattering. A second use of dynamic dual-colour cross-correlation measurements is in estimating the ratio of singly and multiply scattered light. Finally the experiment allows very successful recovery of static information from dynamic light scattering data on concentrated samples with central two-body interaction forces.

Using a surface force apparatus, the authors observed the lubrication of a colloidal suspension using spherical and planar geometry on a nanometric scale. They have shown that the behaviour of the layer, and in particular the friction instabilities can be explained in terms of the compaction of adsorbed colloidal particles.

Monolayers of amphiphilic molecules spread at the air-water interface have been studied using X-ray reflectivity, a technique which allows an independent determination of their thicknesses and densities (i.e. structural parameters), as well as their roughnesses (due to thermally excited capillary waves). The phase diagrams of C₁₅, C₂₁ and C₂₉ fatty acids and of the phospholipid L- alpha -dipalmitoylphosphatidylcholine (DPPC) have been investigated and their phase transitions characterized. Evidence for the liquid-expanded-liquid-condensed transition is given by an abrupt increase in the thickness of the aliphatic medium, and the structure is characteristic of the liquid-condensed phase below and above the triple point. A strong decrease in the roughness is observed at the transition to the solid state. This striking feature is attributed to the bending rigidity of the monolayer in the solid phase, whose value has been determined for different lengths of the aliphatic chains.

The specular reflection of neutrons is now finding widespread application in the study of a range of problems in surface chemistry. In recent studies, using hydrogen-deuterium contrast variation, it has been possible in the investigation of surfactant adsorption at the air-liquid interface to determine both the surfactant surface excess and the detailed surface structure of the surfactant molecules. The authors present results that extend these studies to the adsorption of mixed surfactants at the air-liquid interface. For the mixed system sodium dodecyl sulphate and dodecanol, they have determined the surface excess and surface structure of each component in the mixed surface layer.

The author reviews the thermodynamic and structural behaviour of interfaces between two coexisting phases in 3D. In particular the authors looks at fluid-fluid interfaces in ternary systems. The (microscopic) interface between two phases may (under some circumstances) exhibit structural changes that will drastically change the thickness of the transition region. Examples of these are wetting transitions. Here, the author reviews theoretical as well as experimental results for wetting transitions in binary and ternary (surfactant) mixtures.

The authors present a detailed study of the macroscopic spreading kinetics of a non-volatile liquid, polydimethylsiloxane, on model solid surfaces of continuously adjustable spreading power, with special attention paid to the immediate vicinity of the zero spreading parameter value. They establish the existence of new spreading regimes which are, they think, the signature of non-monotonic interactions versus thickness laws.

The authors report the first systematic study of the wetting behaviour in fluid alkali metal-alkali halide systems on the metal-rich side of the phase diagram. To this end the interface of fluid sample-inert substrate (sapphire) has been probed by ellipsometry. Of particular interest is the influence of differences of the bulk phase diagram on the wetting characteristics. If the bulk fluid phase exhibits homogeneous miscibility like Cs-CsCl the optical reflectivity changes continuously with composition consistent with metallic Drude type behaviour. However, in systems with a critical and a triple point like K-KCl and Na-NaCl a wetting transition is observed. This occurs in metal-rich solutions approaching the triple point along the phase boundary. In K-KCl a salt-rich wetting film of approximately=100 nm thickness and composition corresponding to K_0.1KCl_0.9 has been determined. This is the film thickness in thermal equilibrium as has been found by vigorous ultrasonic stirring.

The authors show that diffusing-wave spectroscopy can be used as a non-invasive probe of the bulk properties of three-dimensional foams. A new picture accounting for the origin of the temporal fluctuations of multiply scattered light is developed and corroborated with direct observations through a microscope. The interpretation and measurements yield the growth law for the coarsening of foam bubbles and new insight into their dynamics.

The authors present experimental studies of the heterogeneities of porous media by tracer dispersion performed by using either a classical transmission technique or an echo technique; in the latter, the tracer is first injected into the sample and then pumped back. When large channel- or strata-like heterogeneities are present, measuring the degree of reversibility allows one to estimate the size of the heterogeneities and the permeability variations. In homogeneous media such as sphere packings, dispersion becomes irreversible for penetration depths larger than a few diameters. An electrochemical technique is described and applied to dispersion measurements with a resolution better than one grain size.

The author reviews the state of the art in flow instabilities in displacement processes in porous media. Both miscible and immiscible displacements are considered. Effects of mobility ratio, capillary and Peclet numbers are explored. The author juxtaposes continuum and discrete methods and discusses results and limitations. Emphasis is placed on heterogeneity and its interplay with flow instabilities.

The author discusses those properties of porous media which can be deduced from experiments using measurements of superfluid 1st, 2nd, 4th, and 3rd sound; he also explores the transferability of these results to other transport experiments, especially the acoustic properties of porous media saturated with Newtonian fluids. Many of the relevant geometrical parameters are those which arise in a canonical electrical conductivity problem in which the porous solid is insulating, the pore fluid is conducting, and there is an additional surface conductivity lining the walls of the pore space. The most important geometrical parameters are the three-dimensional tortuosity of the pore space, alpha ₃, the two-dimensional tortuosity of the pore/grain interface, alpha ₂, and Lambda , which is a well-defined measure of dynamically connected pore sizes.

Capillary pressure, hydraulic conductivity and the capillary dispersion coefficient have been observed to obey power laws in the wetting phase saturation. The authors relate power-law behaviour at low wetting phase saturations, i.e. at high capillary pressures, to the thin-film physics of the wetting phase and the fractal character of the pore space of natural porous media.

The authors consider Ginzburg-Landau-type models for localized structures observed in the vicinity of subcritical bifurcations to cellular flows, where two metastable homogeneous states coexist in an interval range of the control parameter. A localized structure consists of a small region in the bifurcated state surrounded by the basic state. The authors show how non-variational effects, i.e. the absence of a free energy to minimize, can explain the stability of these structures, contrary to the case of droplets in first-order phase transitions.

A generation of large-scale helical vortices resulting from the instability of small-scale helical turbulence with respect to two-scale disturbances is considered. To investigate such an instability the authors consider an incompressible liquid containing rigid particles. An equation describing the evolution of mean disturbances is derived and the instability increment is obtained.

Recent experimental and theoretical efforts have revealed the existence of a fingering instability at the moving front of thin liquid films forced to spread under gravitational, rotational or surface shear stresses, as for example by using the Marangoni effect. The authors describe how the presence of a precursor film in front of the spreading macroscopic film, whether it is by prewetting the substrate or by surface diffusion or multilayer absorption, can prevent the development of the instability.

The transition to spatiotemporal chaos in thermal convection is described and analysed by studying the properties of local and global variables. Beyond the transition point for spacetime chaos the system displays thermodynamics properties in Fourier space. It is shown that a suitable free energy accounts for the experimental results.

Research into fluid physics under reduced gravity conditions has been stimulated by basic research and numerous applications. For example, research includes the shape and stability of liquid surfaces; the bifurcation of convective flows; the stability of Marangoni flows; and with respect to applications, surface tension tanks; spacecraft dynamics; crystal growth; the production of composite materials and the extremely large field of life sciences. The stability criteria of liquid bridges between coaxial circular discs are discussed. The stability of liquid volumes at edges is considered. Respective experiments performed in parabolic flights of a KC-135 aircraft, in sounding rockets and in the Spacelab are reported. Special attention is paid to the various effects of Marangoni convection on crystal growth and on the separation of binary liquid mixtures exhibiting a miscibility gap.

The Coulomb force exerted by an electric field on any charge present in a dielectric liquid may induce fluid motion. At high applied electric fields in an industrial grade insulating liquid, charge carriers are created at metallic/liquid interfaces, a process referred to as ion injection, and result from electrochemical reactions. Recently it has become possible to reproduce these injection processes and carefully designed experiments may be related to analytical models. In this review the author focuses attention on the electrohydrodynamic instabilities and chaos induced by unipolar charge injection.