Table of contents

This special issue of Journal of Physics: Condensed Matter contains the Proceedings of the 6th Liquid Matter Conference held in Utrecht, The Netherlands, 2–6 July 2005.

The three-yearly Liquid Matter Conference is organized by the Liquids Section of the Condensed Matter Division of the European Physical Society. This series of meetings began in Lyon in 1990. The most recent meeting was held in 2003 in Konstanz. The aim of the Liquid Matter Conferences is to bring together scientists working on the liquid state of matter. This rapidly growing field includes the physics, chemistry, biology and chemical engineering of liquid matter as well as various applied research areas. In fact, the Utrecht meeting had, for the first time, a special session devoted to Fundamental Challenges in Applied Liquid Physics and Microfluidics.

The Utrecht meeting had 760 registered participants from four continents. An important event at this meeting was the award of the First Liquid Matter Prize of the European Physical Society to Professor Jean-Pierre Hansen FRS, of Cambridge University. In addition to a plenary speech by the recipient of the Liquid Matter Prize, the scientific programme consisted of 10 plenary lectures, 117 symposia talks, 25 of which were keynote lectures and some 650 poster contributions. The meeting also hosted a one-day symposium of the Division of Liquids and Interfaces of the Chemical Sciences division of NWO. This special issue of Journal of Physics: Condensed Matter contains 61 of the oral communications.

Liquid state physics is at the interface of many fields of research. As a consequence, many of the attendants come from adjacent fields and find in the Liquid Matter Conference a forum to meet experts from other areas of research. This aspect of the Liquid Matter Conference makes it an exciting meeting as it not only offers the participants an up-to-date picture of the status of research into the liquid state of matter, but it also allows them to establish new (and often unexpected) trans-disciplinary contacts for joint scientific endeavours. This applies in particular to the area of soft condensed matter such as colloidal suspensions, polymeric systems and biological materials.

The conference was held at the Uithof, the campus of the University of Utrecht. The organizers gratefully acknowledge the generosity of the University and City of Utrecht, which enabled us to stage both the scientific part of the conference and several festive and cultural events in some of the most attractive venues of the Netherlands. We were also delighted by the substantial contributions offered by the sponsors of the 6th Liquid Matter Conference. With this support it became possible to support a large number of scientists who would otherwise not have been able to attend.

Finally, we owe a great debt of gratitude to the secretarial staff of the conference and the many students, postdocs and other colleagues who helped tirelessly (and very efficiently) to make the conference run smoothly.

The Board of the Liquids Section of the European Physical Society decided that the 7th Liquid Matter Conference will be held in Lund (Sweden). The tentative dates are Friday 27 June 2008 to Tuesday 1 July 2008.

A coarse-graining strategy for dilute and semi-dilute solutions of interacting polymers, and of colloid–polymer mixtures, is briefly described. Monomer degrees of freedom are traced out to derive an effective, state-dependent pair potential between the polymer centres of mass. The cross-over between good and poor solvent conditions is discussed within a scaling analysis. The method is extended to block copolymers represented as 'necklaces' of soft 'blobs', and its success is illustrated here in the case of a symmetric diblock copolymer which exhibits microphase separation.

Recently we found that viscoelastic phase separation (VPS) is observed not only in polymer solutions, but also in protein solutions and colloidal suspensions. This suggests that VPS may be universally observed in any dynamically asymmetric mixture composed of slow and fast components. Such mixtures are ubiquitous in soft matter. The common feature of VPS is the formation of a transient gel made of the slow components just after a temperature quench into an unstable region of a phase diagram. Transient gelation is induced by deformation fields induced by phase separation itself; thus, it is a self-induced shear-thickening phenomenon. The connectivity of a transient gel causes the stress against diffusion-type deformation upon phase separation and thus leads to a strong coupling between the mechanical stress and the diffusion of particles: phase separation (or diffusion) can proceed only with the local breakup of a transient gel and the resulting loss of the connectivity. We argue that VPS is the fracture process of a transient gel induced by the self-generated mechanical stress due to interparticle attractive interactions. A simple toy model supports this scenario.

A review of recent simulation work in the area of phase transitions in ionic systems is presented. The vapour–liquid transition for the restricted primitive model has been studied extensively in the past decade. The critical temperature is now known to excellent accuracy and the critical density to moderate accuracy. There is also strong simulation-based evidence that the model is in the Ising universality class. Discretized lattice versions of the model are reviewed. Other systems covered are size- and charge-asymmetric electrolytes, colloid–salt mixtures, realistic salt models and charged chains. Areas of future research needs are briefly discussed.

We use femtosecond two-colour mid-infrared spectroscopy to study the dynamics of aqueous solutions of salts in HDO:D₂O. We find that the lifetime of the O–H stretch vibration of HDO molecules in the solvation shell of the halogenic anions Cl⁻, Br⁻, and I⁻ is much longer than the lifetime of the O–H stretch vibration of the HDO molecule in bulk D₂O solution. This difference in lifetime allows for a clear separation of the response of the solvation shell from that of the bulk liquid. We observe that the solvating HDO molecules show much slower hydrogen-bond and reorientation dynamics than bulk water molecules. In contrast, the dynamics of the water molecules outside the first ionic hydration shell are observed to be negligibly affected by the ions.

With the completion of the human genome project, the biosciences community is beginning the daunting task of understanding the structures and functions of a large number of interacting biological macromolecules. Examples include the interacting molecules involved in the process of DNA condensation during the cell cycle, and in the formation of bundles and networks of filamentous actin proteins in cell attachment, motility and cytokinesis. In this proceedings paper we present examples of supramolecular assembly based on proteins derived from the vertebrate nerve cell cytoskeleton. The axonal cytoskeleton in vertebrate neurons provides a rich example of bundles and networks of neurofilaments, microtubules (MTs) and filamentous actin, where the nature of the interactions, structures, and structure–function correlations remains poorly understood. We describe synchrotron x-ray diffraction, electron microscopy, and optical imaging data, in reconstituted protein systems purified from bovine central nervous system, which reveal unexpected structures not predicted by current electrostatic theories of polyelectrolyte bundling, including three-dimensional MT bundles and two-dimensional MT necklaces.

At a sufficiently large Reynolds number the flow of a classical fluid becomes turbulent. Typically, turbulent energy is injected into large-scale eddies, from which it flows through the action of the non-linear term in the Navier–Stokes equation into smaller and smaller eddies until the scale of the motion is small enough that the energy can be dissipated by viscosity. In a superfluid (liquid ⁴He or liquid ³He), which has no viscosity, this source of dissipation is absent, so that perhaps the flow of turbulent energy persists down to atomic scales; furthermore, the absence of viscosity seems to imply that the Reynolds number is infinite. However, turbulent motion in a superfluid is restricted by quantum effects, associated with the quantization of angular momentum, and it will be shown how these effects change this picture and lead to an effectively finite Reynolds number, with dissipation that mimics the effect of viscosity and arises from the radiation of sound by special types of quantized motion.

The fluctuation theorem (FT) quantifies the probability of Second Law of Thermodynamics violations in small systems over short timescales. While this theorem has been experimentally demonstrated for systems that are perturbed from an initial equilibrium state, there are a number of studies suggesting that the theorem applies asymptotically in the long time limit to systems in a non-equilibrium steady state. The asymptotic application of the FT to such non-equilibrium steady-states has been referred to in the literature as the steady-state fluctuation theorem (or SSFT). In 2005 Wang et al demonstrated experimentally an integrated form of the SSFT using a colloidal bead that was weakly held in a circularly translating optical trap. Moreover, they showed that the integrated form of the FT may, for certain systems, hold under non-equilibrium steady states for all time, and not just in the long time limit, as suggested by the SSFT. While demonstration of the integrated forms of these theorems is compact and illustrative, a proper demonstration shows the theorem directly, rather than in its integrated form. In this paper, we present experimental results that demonstrate the SSFT directly, and show that the FT can hold for all time under non-equilibrium steady states.

We describe a simulation study of the liquid–vapour phase behaviour of a model polydisperse fluid. Particle interactions are given by a Lennard-Jones potential in which polydispersity features both in the particle sizes and the amplitude of their interactions. We address the computational problem of accurately locating the cloud curve for such a system using Monte Carlo simulations within the grand canonical ensemble. The strongly nonlinear variation of the fractional volumes of the phases across the coexistence region precludes naive extrapolation to determine the cloud point density. Instead we propose an improved estimator for the cloud point location and use scaling arguments to predicts its finite-size behaviour. Excellent agreement is found with the simulation results. Application of the method reveals that the measured cloud curve is highly sensitive to the presence of large particles, even when they are extremely rare. This finding is expected to have implications for the reproducibility of experimentally measured phase diagrams in colloids and polymers.

Consider a fluid composed of two species of particles, where the interparticle pair potentials . On confining an equal number of particles from each species in a cavity, one finds that the average one body density profiles of each species are constrained to be exactly the same due to the symmetry, when both external cavity potentials are the same. For a binary fluid of Brownian particles interacting via repulsive Gaussian pair potentials that exhibits phase separation, we study the dynamics of the fluid one body density profiles on breaking the symmetry of the external potentials, using the dynamical density functional theory of Marconi and Tarazona (1999 J. Chem. Phys.110 8032). On breaking the symmetry we see that the fluid one body density profiles can then show the phase separation that is present.

Carbonyl sulfide (OCS) molecules in superfluid helium-4 clusters have been studied by path integral hybrid Monte Carlo methods. A new technique was developed to treat the quantum rotational degree of freedom of the molecules in hybrid Monte Carlo methods; this is referred to as a 'Legendre potential technique'. Then, our method was applied to OCS-doped helium clusters. It was found that although the molecule is solvated inside the cluster, the calculated orientational correlation function exhibits free-rotor-type behaviour. The estimated effective rotational constant was in good agreement with the experimental value.

We have performed classical molecular dynamics simulations to investigate the structure and dynamics of water clusters in aqueous methanol across a range of solution compositions. Investigation of the effect of different model potentials and cluster membership criteria indicates that the qualitative trends in cluster size distributions are preserved. Characterization of the water molecules on the interior of such clusters using a tetrahedral order parameter previously used to study pure water suggests that they are over-structured compared to the pure liquid for a broad range of cluster sizes and at different compositions. Calculation of the hydrogen-bond dynamics indicates that the water–water hydrogen-bonds in the solution are longer lived than those in the pure liquid for all solutions studied and increase almost linearly as a function of methanol concentration.

Most classical computer simulations of water use an effective potential which reflects the local symmetry of the water molecule's nuclei. This in turn gives rise to a symmetric distribution of neighbouring molecules on average. Recent data from x-ray absorption spectroscopy however are interpreted as implying that there is local asymmetry in water structure with some strong and some weak hydrogen bonds. Here, existing neutron and x-ray diffraction data are interpreted with empirical potential structure refinement using an asymmetric water potential. It is found that quite accurate representations of the diffraction data are possible by this means. This does not prove that the asymmetric model is correct, nor that the symmetric model is incorrect, but that x-ray and neutron diffraction data on water are rather insensitive to these details of the water interaction potential.

The melting point of ice I_h for the TIP3P, SPC, SPC/E, TIP4P, TIP4P/Ew and TIP5P models has been determined by computer simulation. It has been found that the melting points of ice I_h for these models are 146, 190, 215, 232, 245 and 274 K respectively. Thus from the models of water available so far only TIP5P reproduces the experimental melting point of water. The relative stability of ice II with respect to ice I_h at the normal melting point has also been considered. Ice II is more stable than ice I_h for the TIP3P, SPC, SPC/E and TIP5P models. Only for the TIP4P and TIP4P/Ew models is ice I_h more stable than ice II at low pressures. The complete phase diagram for the SPC/E, TIP4P and TIP5P models has been computed. It has been found that SPC/E and TIP5P do not correctly describe the phase diagram of water. However, TIP4P provides a qualitatively correct description of the phase diagram of water. A slight modification of the parameters of the TIP4P model yields a new model, denoted as TIP4P/ice, which reproduces the experimental melting point of water and provides an excellent description of the densities of all ice phases.

Oscillatory wetting–dewetting instabilities at the liquid/vapour interface of Ga–Pb and, for the first time, of Ga–Bi alloys have been studied employing second harmonic generation and kinetic ellipsometry. Alloy samples prepared at x–T conditions inside the respective miscibility gap (Ga_0.8Bi_0.2: T = 520 K; Ga_0.95Pb_0.05: T = 595 K) show oscillations on the timescale of typically hours. The surface transforms back and forth from a Ga-rich non-wetting state to a complete wetting state where the Ga-rich bulk is covered with a Bi-rich (or Pb-rich) film. In the case of Ga–Bi the change of the wetting film thickness was characterized quantitatively with the aid of the ellipsometric measurements. It is consistent with a wetting–dewetting mechanism whereby the wetting film varies periodically between macroscopic and microscopic values. The phenomenon can be understood taking different emissivities of the wet and the nonwet state into account. They cause at otherwise constant conditions different temperatures of the sample, which has to adopt its phase compositions according to the respective phase diagram. This results in mass transport to and away from the surface. A model calculation on Ga–Bi taking these effects into account recovers in good quantitative agreement the oscillation period and the oscillations of the temperature as well as of the film thickness. Similarities and differences between Ga–Bi and Ga–Pb are finally considered.

The influence of ion size asymmetry on the properties of ionic liquid–vapour interfaces is investigated using molecular dynamics simulations of the soft primitive model. Ion size asymmetry results in charge separation at the liquid–vapour interface and therefore in a local violation of the electroneutrality condition. For moderate size asymmetries the electrostatic potential at the interface can reach values of the order of 0.1 V. Size asymmetry plays a very important role in determining ion adsorption at the liquid–vapour interface of ionic mixtures. The interfacial adsorption of the bigger component results in an increase of the electrostatic potential, and a reduction of the interfacial surface tension. Our results show that ionic mixtures provide a very efficient way to tune the electrostatics and surface properties of ionic liquid–vapour interfaces.

Two methods for calculating the effective charge of colloidal particles inside a salt free suspension are presented and analysed using Monte Carlo simulations. The first method is based on the Alexander prescription, in which the exact electrostatic potential is matched asymptomatically to the solution of the linearized Poisson–Boltzmann equation. The second method relies on a dynamical criterion, which defines the condensed counterions through a bound on their total energy. Although the two methods appear to be quite different, they lead to identical values of the effective colloidal charge, even for suspensions with multivalent counterions for which the Poisson–Boltzmann equation fails.

Neutron diffraction measurements, and structural modelling by reverse Monte Carlo simulation, have been performed for liquid Rb–Se mixtures in the Se-rich concentration range. In the liquid mixtures some of the Se chains are strongly bent to form ring-like Se chains with Rb atoms at the chain ends, creating a large amount of voids. The shortened Se chains link up to form a large cluster, and Rb atoms sit around it near the boundary of the miscibility gap at low temperature. The structure factor S(Q) exhibits a small pre-peak at ∼1.3 Å⁻¹. From the void analysis it has been revealed that the observed pre-peak corresponds to a pre-peak in the concentration–concentration structure factor due to the chemically ordered packing of the interstitial voids and Se chains in the structure.

We report measurements of the surface tension and capillary wave spectra at the liquid/vapour interface of the ionic liquids [C₄mim]⁺X⁻ (X⁻ = Cl⁻,AlCl₄⁻) and compare these with recent results of [C₄mim]⁺PF₆⁻. Special attention is given to the temperature dependence of the surface dipole moment density γ(T), which has been derived from the capillary wave spectra with the aid of the properly extended dispersion relation. For [C₄mim]⁺PF₆⁻ an order–disorder interfacial phase transition with a critical temperature of T_c = 385 K is found, which has not been reported before. At higher temperatures a first order ferroelectric surface phase transition is indicated in γ(T) of both [C₄mim]⁺PF₆⁻ and [C₄mim]⁺Cl⁻ melts. Increasing the anion size in [C₄mim]⁺AlCl₄⁻, these interfacial phase transitions are suppressed.

Several striking phenomena found recently by isoplethal and hygroscopic methods are reviewed briefly: rich and poor facetings at cubic/isotropic interfaces, growth by redistribution process in cubic crystals, faceting by the ratchet-like effect in a saw-tooth temperature cycling, asymmetric shapes of cubic crystals induced in thermal gradients by the Ludwig–Soret effect and anisotropic and facet-by-facet surface meltings. We also report on the discovery, by hygroscopy, of a new phase that has been so far missed by isoplethal methods. The constellation of facets of this ephemeral phase suggests a random hexagonal close packed structure similar to the one found recently in soft polymeric colloids.

We study model membranes in the form of giant unilamellar vesicles (GUVs) composed of two saturated lipids with different hydrophilic headgroups or different hydrophobic chain lengths. Lateral phase separation in the lipid bilayer causes solid-like 'gel' domains to nucleate and grow in the fluid membrane. We study the shape and size of these domains as well as their growth and interactions.

There is a great need for understanding the link between the detailed chemistry of surfactants, forming wormlike micelles, and their macroscopic rheological properties. In this paper we show how this link may be explored through particle simulations. First we review an existing bead–spring model. We find that shear flow enhances the formation of rings at the expense of linear chains. The shear viscosity of this model is dominated by solvent contributions, however, and the link with the chemistry of the surfactants is missing. We introduce a more realistic Brownian dynamics model, the parameters of which are measured from atomistic molecular dynamics simulations.

We present an experimental study dedicated to understanding the behaviour of polyelectrolyte chains when salt goes into a polyelectrolyte brush. We use the ability of asymmetric neutral-charged diblock copolymers to anchor to a selective interface or to self-assemble in water, to examine polyelectrolyte brushes both in planar and in spherical geometries. Using neutron reflectivity, the monomer profile of planar brushes has been found to switch from a Gaussian profile to a parabolic profile, showing that at large spatial scale a salted planar brush behaves like a neutral one in good solvent. Using small angle neutron scattering, it is found that spherical brushes exhibit the same behaviour at a large spatial scale although polyelectrolyte chains remain rod-like at a small spatial scale whatever the salinity is. The charged chains inside a polyelectrolyte brush could be viewed as 'surveyor's chains', which fold their rod-like segments of persistence length l_p. In the last part, spherical polyelectrolyte brushes in contact have been examined. Their behaviour is discussed in terms of interdigitation or contraction of the brushes.

We examine theoretically the structural behaviour of dilute solutions of multiarm star polymers under the addition of smaller homopolymer chains. The approach is based on effective interactions, for which the star-polymer centres and the centres of mass of the linear chains are used as effective coordinates. We find that addition of linear chains first leads to a softening of the star-polymer repulsions and, at higher chain concentrations, to the formation of stable, multi-star clusters. We accompany the theoretical approach with dynamical light scattering measurements in real systems, finding agreement between theory and experiment. We rationalize our findings by deriving the chain-induced, one-component effective potential between stars, which features an attractive well followed by a repulsive barrier. We discuss the dependence of these characteristics on the size and concentration of the homopolymer additives and relate the present system to recent models that also display cluster formation.

We report how large spherical impurities affect the nucleation and growth of hard sphere colloidal crystals. Both the impurities and the colloids are fluorescently labelled polymethylmetacrylate particles and are dispersed in an optically and density matching solvent mixture. Crystal growth, initiated either at the impurity surface, or at the sample bottom, was studied by imaging sequences of two-dimensional xy-slices in the plane of the impurity's centre of mass with a laser scanning confocal microscope. At least two factors determine whether a large impurity can function as a seed for heterogeneous nucleation: timescales and impurity curvature. The curvature needs to be sufficiently low for crystal nuclei to form on the impurity surface. If bulk crystal growth has already approached the impurity, bulk growth is dominant over growth of crystallites on the impurity surface. Such surface crystallites eventually reorient to adapt to the overall bulk crystal symmetry.

The behaviour of colloidal mono- and bilayers in external magnetic fields and under shear is discussed and recent progress is summarized. Superparamagnetic colloidal particles form monolayers when they are confined to a air–water interface in a hanging water droplet. An external magnetic field allows us to tune the strength of the mutual dipole–dipole interaction between the colloids and the anisotropy of the interaction can be controlled by the tilt angle of the magnetic field relative to the surface normal of the air–water interface. For sufficiently large magnetic field strength crystalline monolayers are found. The role of fluctuations in these two-dimensional crystals is discussed. Furthermore, clustering phenomena in binary mixtures of superparamagnetic particles forming fluid monolayers are predicted. Finally, we address sheared colloidal bilayers and find that the orientation of confined colloidal crystals can be tailored by a previously applied shear direction.

Within a general framework we study the effective, deformation-induced interaction between two colloids trapped at a fluid interface. As an application, we consider the interface deformation owing to the electrostatic field of charged colloids. The effective interaction is attractive and overcomes the direct electrostatic repulsion at large separations if the system is not mechanically isolated. Otherwise, a net attraction seems possible only for large enough colloidal charges.

We show how to implement stick boundary conditions for a spherical colloid in a solvent that is coarse-grained by the method of stochastic rotation dynamics. This allows us to measure colloidal rotational velocity auto-correlation functions by direct computer simulation. We find quantitative agreement with Enskog theory for short times and with hydrodynamic mode-coupling theory for longer times. For aqueous colloidal suspensions, the Enskog contribution to the rotational friction is larger than the hydrodynamic one when the colloidal radius drops below 35 nm.

We studied the structure of a colloid–polymer mixture, in the one-phase region, as a function of polymer concentration. Using confocal microscopy, and Fourier transformation of the image data, we accessed the system in real and reciprocal space simultaneously, at the single particle level. We found no change in the colloidal fluid structure as a function of polymer concentration. The fluid phase persisted to higher concentrations of polymer than expected, before an abrupt gelation transition. Furthermore, we found a dramatic change in the local structure around each particle at the onset of gelation. We attribute these phenomena to a screened electrostatic repulsion between the colloids, which, although weak, suppressed the depletion attraction due to the polymer. The repulsive and attractive interactions have a similar range, which results in behaviour distinct both from pure depletion and the recently observed cluster phase formed in the presence of long-ranged electrostatic repulsions. Instead, these results are qualitatively similar to the competition between van der Waals and electrostatic interactions in the Derjaguin–Landau–Verwey–Overbeek theory. We compare our experiments with Monte Carlo simulation and find agreement with a combination of the Asakura–Oosawa–Vrij and Yukawa potentials.

We consider solutions that consist of solute and solvent molecules of arbitrary sizes, shapes and internal charge distributions. The free energy of interaction (the potential of mean force) between the molecules is analysed in terms of a screened Coulomb potential and renormalized charge distributions of the molecules. The emphasis in this work is to bring out a simple physical interpretation of the theory, but the treatment is based on exact statistical mechanical theory without any approximations. The results thereby are the true properties of the system for given pair interaction potentials between the constituent particles. In general, the electric potential from any molecule in the solution can be exactly obtained for all distances by using a (generalized) screened Coulomb potential, provided the source charges constitute a renormalized charge distribution of the molecule. When charge renormalization is done consistently, macroions, small ions and solvent molecules are treated in fundamentally the same manner and all particles acquire renormalized charge distributions that generally are different from their actual (bare) distributions. The electrostatic free energy of interaction is given by the interaction between the renormalized charge distributions of the molecules as mediated by the screened Coulomb potential. The exact formalism is also used for the primitive model of electrolytes. The concepts in the general theory are illustrated by expressing the Poisson–Boltzmann and hypernetted chain (HNC) approximations in this alternative framework. Conditions are given under which the exact theory predicts the existence of attractive electrostatic interaction between two identical particles at large distances from each other.

The formation of equilibrium clusters in colloidal particle suspensions, due to the balance of a short-range attraction and a long-range repulsion interaction, has been recently widely discussed, considering its important role in many areas of material science and soft-matter physics. In this work, we describe a simple phenomenological model in which particles interact with a screened electrostatic repulsive interaction modelled by a Yukawa potential and with a short-ranged attractive potential ('charge patch' attraction) due to the strong lateral correlations among the adsorbed polyions at the surface of the approaching particles. This 'correlated adsorption' generates a more or less ordered 'patchwork-like' surface structure with excess negative charge domains (adsorbed polyelectrolyte domains) and excess positive charge domains (polyelectrolyte free domains), which give rise to a short-range attractive potential. In low-density colloidal liposome suspensions, particles form stable equilibrium clusters. We investigated the cluster formation and the growth in the cluster size in cationic liposomal particle suspensions (built up by DOTAP and DOTAP/DOPE mixtures), by means of a dynamic light scattering technique, in several experimental conditions (adsorbing polyions of different molecular weights, liposomes of different surface charge densities, presence of added salt), and a generally good agreement with the model predictions has been found.

We present confocal microscopy studies of novel particle-stabilized emulsions. The novelty arises because the immiscible fluids have an accessible upper critical solution temperature. The emulsions have been created by beginning with particles dispersed in the single-fluid phase. On cooling, regions of the minority phase nucleate. While coarsening, these nuclei become coated with particles due to the associated reduction in interfacial energy. The resulting emulsion is arrested, and the particle-coated interfaces have intriguing properties. Having made use of the binary-fluid phase diagram to create the emulsion we then make use of it to study the properties of the interfaces. As the emulsion is re-heated toward the single-fluid phase the interfacial tension falls and the volume of the dispersed phase drops. Crumpling, fracture or coalescence can follow. The results show that the elasticity of the interfaces has a controlling influence over the emulsion behaviour.

The deformation of vesicles in flow is studied by a mesoscopic simulation technique, which combines multi-particle collision dynamics for the solvent with a dynamically triangulated surface model for the membrane. Shape transitions are investigated both in simple shear flows and in cylindrical capillary flows. We focus on reduced volumes, where the discocyte shape of fluid vesicles is stable, and the prolate shape is metastable. In simple shear flow at low membrane viscosity, the shear induces a transformation from discocyte to prolate with increasing shear rate, while at high membrane viscosity, the shear induces a transformation from prolate to discocyte, or tumbling motion accompanied by oscillations between these two morphologies. In capillary flow, at small flow velocities the symmetry axis of the discocyte is found not to be oriented perpendicular to the cylinder axis. With increasing flow velocity, a transition to a prolate shape occurs for fluid vesicles, while vesicles with shear-elastic membranes (like red blood cells) transform into a coaxial parachute-like shape.

Poly-vinlyacetate (PVAc) forms very stable and reproducible monolayers on the surface of water, a model system to understand polymer physics on two dimensions. A recently introduced technique is applied here to to study viscoelasticity of PVAc monolayers. The method is based on measurement of surface tension in two orthogonal directions during anisotropic deformation. Compression and shear moduli are explored over a very large concentration range, highlighting a series of four different regimes. At low concentration the polymers are in a dilute gas. Above the overlap concentration Γ^* there is a fluid semi-dilute region, where the monolayer properties are described by scaling laws. At a threshold concentration Γ^**, a decrease in the gradient of pressure with concentration is observed, and we argue that there is still a large fraction of free area on the surface. Compressing further, we then identify close packing as the point where the pressure gradient rises sharply and a shear modulus emerges. This is interpreted as a transition to a soft-solid due to the kinetic arrest of close-packed monomers. The rheological properties of PVAc above Γ^** have not been studied previously. Discussion includes possible explanations for the observed behaviour in terms of both equilibrium and non-equilibrium conditions, and the relation to microscopic chain properties. Temperature dependent effects around Γ^** are also observed and described.

We present experimental results on the Stokes experiment performed in a 3D dry liquid foam. The system is used as a rheometric tool: from the force exerted on a 1 cm glass bead, plunged at controlled velocity in the foam in a quasi-static regime, local foam properties are probed around the sphere. With this original and simple technique, we show the possibility of measuring the foam shear modulus, the gravity drainage rate and the evolution of the bubble size during coarsening.

We study the curvature dependence of thermodynamical quantities such as the surface tension γ and the excess adsorptions Γ_i of fluid mixtures in contact with curved walls. If mixtures with short-ranged interaction at state points away from any phase separation are considered, all intrinsic length scales remain small, and we can generalize to mixtures recent findings for the one-component fluid. In addition, we relate the excess surface grand potential to B₂, the effective second virial coefficient between big particles of a mixture after integrating out the degrees of freedom of the smaller components.

We have compared calorimetric measurements with the nonlinear optical technique vibrational sum frequency scattering to investigate interface–solvent effects in colloidal gelation transitions. This allows us to explain the difference in gelation behaviour between dispersions of stearyl-coated silica particles in n-hexadecane and benzene or toluene. In n-hexadecane dispersions, an anomalous heat effect is observed, due to the formation of an ordered interface layer (that is not confined to the first monolayer and is composed of ∼1/3 surface crafted chains and ∼2/3 solvent molecules). For solvents that cannot interdigitate with the surface chains this transition does not occur and consequently no heat effect is observed.

An account is given of density functional theory (DFT) for quenched–annealed fluid mixtures that are used to model fluids adsorbed in random porous matrices. The theory is based on the replica trick and allows the treatment of situations where the quenched random matrix as well as the annealed fluid are inhomogeneous on average. Applications of the framework include investigation of the adsorption properties of hard spheres and model colloid–polymer mixtures in bulk matrices and at matrix surfaces, and the influence of the quenched disorder on phase transitions like fluid demixing, isotropic–nematic ordering, and freezing. Particularly rich wetting behaviour was found for colloid–polymer mixtures adsorbed against a porous wall.

We show that the 3D wedge filling transition in the presence of short-ranged interactions can be first order or second order depending on the strength of the line tension associated with the wedge bottom. This fact implies the existence of a tricritical point characterized by a short-distance expansion which differs from the usual continuous filling transition. Our analysis is based on an effective one-dimensional model for the 3D wedge filling, which arises from the identification of the breather modes as the only relevant interfacial fluctuations. From such analysis we find a correspondence between continuous 3D filling at bulk coexistence and 2D wetting transitions with random-bond disorder.

This paper presents recent advances in the characterization of the intrinsic structures in computer simulations of liquid surfaces. The use of operational definitions for the intrinsic surface, associated with each molecular configuration of a liquid slab, gives direct access to the intrinsic profile and to the wavevector dependent surface tension. However, the characteristics of these functions depend on the definition used for the intrinsic surface. We discuss the pathologies associated with a local Gibbs dividing surface definition, and consider the alternative definition of a minimal area surface, going though a set of surface pivots, self-consistently chosen to represent the first liquid layer.

We investigate critical drying for a model system in which fluid–fluid and wall–fluid interatomic potentials decay as −r⁻⁶. For a spherical substrate of radius R and liquids at bulk coexistence we show, by means of an effective interfacial potential approach, that at the planar drying transition temperature the modulus of the adsorption per unit area diverges as R^1/4 and the wall–liquid surface tension has a term in R^−3/4. The results of microscopic density functional calculations confirm this nonanalytic dependence on the curvature (R⁻¹) and point to the possibility of layers of depleted fluid density developing at a solvophobic substrate.

We study the adsorption and wetting behaviour of colloidal hard spheres and colloidal rod-like particles at a planar hard wall using computer simulations. Our results show complete wetting by the hard-sphere crystal at the wall–fluid interface, and complete wetting by the nematic phase of the hard-rod fluid at the wall–isotropic fluid interface. In addition, we investigated the effect of polymer-mediated effective interactions on the wetting behaviour of colloidal spheres. The many-body character of the effective colloidal interactions yields an adsorption behaviour that differs enormously from those found for pairwise simple fluids; e.g., far from the triple point, we find three layering transitions in the partial wetting regime prior to a transition to complete wetting by the colloidal liquid.

In bulk fluids hydrodynamic Navier–Stokes equations are proven to be valid down to the nanometre scale. However, during the dewetting process of thin liquid films of nanometre thickness the interplay of surface tension γ, substrate potential and thermal noise can lead to qualitatively different behaviour on laterally much larger scales up to microns. By deriving a stochastic thin film equation with a conserved noise term we show that the spectrum of capillary waves changes from an exponential decay to a power law k_BT/(γq²) for large wavevectors q due to thermal fluctuations at temperature T. Also the time evolution of film roughness σ(t) and of the typical wavevector k(t) of unstable perturbations changes qualitatively. Whereas a deterministic Navier–Stokes equation in the lubrication approximation predicts in the linear regime a constant k(t) = k₀, one finds a coarsening due to thermal noise.

We propose a method for direct calculation of an effective interface potential from computer simulation. The method allows us to locate the wetting transition, determine its order and calculate the concomitant contact angle of macroscopic drops. We illustrate the method by studying the wetting behaviour of a polymer melt on a brush consisting of chemically identical grafted polymers. At low grafting densities, the melt dewets; at intermediate grafting densities the liquid of free chains wets the brush. On further increasing the grafting density we encounter dewetting (autophobicity). When long-range forces are considered, autophobic behaviour is also observed, but the wetting transition is shifted to larger grafting densities and becomes first order.

A liquid drop can be kept bouncing at the surface of a bath of the same liquid for any length of time if the bath is kept oscillating vertically. Several regimes can be observed. For a liquid of moderate viscosity, the bouncing of the drop generates damped capillary waves with a wavelength corresponding to the forcing frequency. Therefore when several identical drops are placed on the oscillating surface, the interaction of their waves leads to the self-organization of the drops with a 2D triangular lattice. Another remarkable regime is observed when the forcing amplitude is increased close to the Faraday instability threshold: the drop starts moving in the horizontal plane at a constant velocity. We have studied the movement of one 'walking drop' as well as the possible interactions of several of these particular drops placed on the surface of the liquid, leading to their complex self-organization. These drops can collide via their waves and in certain situations attract each other and start orbiting.

Topological versus chemical ordering in disordered binary systems is investigated by using the Bhatia–Thornton formalism. The moments of the partial pair-distribution functions are considered and are related to relevant thermodynamic and other properties of the system. The results are illustrated by using the measured partial-pair distribution functions for the network forming glasses ZnCl₂ and GeSe₂.

We study the dynamic behaviour of concentrated colloidal hard spheres using time resolved correlation, a light scattering technique that can detect the slow evolution of the dynamics in out-of-equilibrium systems. Surprisingly, equilibrium is reached a very long time after sample initialization, the non-stationary regime lasting up to three orders of magnitude more than the relaxation time of the system. Before reaching equilibrium, the system displays unusual ageing behaviour. The intermediate scattering function decays faster than exponentially and its relaxation time evolves non-monotonically with sample age.

We study the formation and growth of equilibrium clusters in a suspension of weakly charged colloidal particles and small non-adsorbing polymers. The effective potential is characterized by a short-range attraction and a long-range repulsion. The size, shape and local structure of the clusters are studied using three-dimensional particle microscopy. We observe a rapid growth in the mean cluster size and the average number of nearest neighbours approaching the gel boundary.

We study the dependence of the dynamics on the size of particle clusters that grow by stepwise aggregation in a two-component reactive mixture. The data reveal the cluster property involved in the glasslike arrest and its quantitative link with the structural relaxation time. Specifically, we measured the number-average cluster size x_n—i.e., the average number of bonded monomers per molecule—and, independently, the characteristic time τ for decay of the photon correlation function, throughout the reaction. We find that x_n diverges as the system freezes at the glass transition, x_n and τ being exponentially related. The comparison with the gel transition that occurs at the divergence of the weight-average cluster size provides evidence that the two transitions are associated to distinct properties of the same underlying clustering process.

The dynamics of confined glass-forming liquids is discussed on the basis of the recent extension of the mode-coupling theory for the liquid–glass transition to the model of the quenched–annealed binary mixture. It is in particular shown that, in confinement, the collective density correlation functions always decay to a non-zero infinite time value, even in the fluid state, and some clarification is given about the question of the relation between structure and dynamics in confined fluids.

Kinetically constrained spin models are schematic coarse-grained models for the glass transition which represent an efficient theoretical tool to study detailed spatio-temporal aspects of dynamic heterogeneity in supercooled liquids. Here, we study how spatially correlated dynamic domains evolve with time and compare our results to various experimental and numerical investigations. We find that strong and fragile models yield different results. In particular, the lifetime of dynamic heterogeneity remains constant and roughly equal to the alpha relaxation time in strong models, while it increases more rapidly in fragile models when the glass transition is approached.

According to the Kosterlitz–Thouless–Halperin–Nelson–Young (KTHNY) theory a two-dimensional (2D) crystal melts when bound pairs of dislocations dissociate into single dislocations. The temperature at which dislocation pairs unbind can be identified by means of the Young's modulus K passing through 16π. In a previous paper (von Grünberg et al 2004 Phys. Rev. Lett.93 255703) we analysed video-microscopy data of 2D colloidal crystals, obtained elastic constants from normal-mode band-structures, and confirmed that the unbinding temperature is indeed close to the melting temperature. Processing the same data we now obtain elastic constants directly from an analysis in real space by computing relative mean-square displacements as a function of the lattice site separation. We have also carried out Monte Carlo simulations to check for the thermal softening of the colloidal crystal. Both studies confirm our previous results. We finally compare our results to the corresponding curves for a 2D electron solid on the surface of liquid helium.

Inside every living cell is the cytoplasm: a fluid mixture of thousands of different macromolecules, predominantly proteins. This mixture is where most of the biochemistry occurs that enables living cells to function, and it is perhaps the most complex liquid on earth. Here we take an inventory of what is actually in this mixture. Recent genome-sequencing work has given us for the first time at least some information on all of these thousands of components. Having done so we consider two physical phenomena in the cytoplasm: diffusion and possible phase separation. Diffusion is slower in the highly crowded cytoplasm than in dilute solution. Reasonable estimates of this slow-down can be obtained and their consequences explored; for example, monomer–dimer equilibria are established approximately 20 times more slowly than in a dilute solution. Phase separation in all except exceptional cells appears not to be a problem, despite the high density and so strong protein–protein interactions present. We suggest that this may be partially a by-product of the evolution of other properties, and partially a result of the huge number of components present.

Two scenarios are proposed for explaining the thermodynamic anomalies of supercooled water, but they cannot be directly tested in this region because of the onset of homogeneous crystallization. They also give two distinct temperature dependences of the limit of metastability of water stretched beyond its boiling curve: either monotonic, as in simple liquids, or exhibiting a minimum. Therefore cavitation experiments could allow one to distinguish between the competing pictures. We review previous experiments and describe our preliminary results.

Entrapment of gas bubbles at the liquid/solid interface plays a crucial role for fluidic systems when the volume to surface ratio goes down. Here, we report an investigation on the presence of air on hydrophobic surfaces in water. The gas pockets on such walls are expanded, and thus made visible, by lowering the liquid pressure, a phenomenon known as heterogeneous cavitation. The investigation is extended to the study of the formation and the dynamics of bubbles generated from well controlled nuclei.

We investigate the kinetics of phase separation for a mixture of rod-like viruses (fd) and polymer (dextran), which effectively constitutes a system of attractive rods. This dispersion is quenched from a flow-induced fully nematic state into the region where the nematic and the isotropic phase coexist. We show experimental evidence that the kinetic pathway depends on the overall concentration. When the quench is made at high concentrations, the system is meta-stable and we observe typical nucleation-and-growth. For quenches at low concentration the system is unstable and the system undergoes a spinodal decomposition. At intermediate concentrations we see the transition between both demixing processes, where we locate the spinodal point.

We present a detailed analysis of the structure of liquid carbon near the freezing line. The results are obtained by molecular simulation using a recently developed state-of-the-art bond order potential. We find that along the melting line the liquid is predominantly threefold coordinated up to pressures far beyond the liquid–graphite–diamond triple point. We find no sign of a first-order liquid–liquid phase transition when, at 10 500 K and ∼300 GPa, the local structure of the liquid along the melting line changes dominant coordination from three- to fourfold.

The stationary, shear rate dependent flow curves of yielding glasses are discussed within a mode coupling theory approach. Asymptotic formulae for the shear stress at the transition point are obtained that take the form of generalized Hershel–Bulkeley constitutive equations, and enable one to find dynamical yield stresses by extrapolation.

We study the coupled rotational diffusion in a two-particle chain on the basis of a Smoluchowski equation and calculate time correlation functions that are measurable in an experiment. This approach might be used to explore hydrodynamic interactions in the limit where lubrication theory is valid.

We present a general description of thermophoretic phenomena in dilute suspensions of spherical colloids which lie at the basis of the Ludwig/Soret effect. We first consider how a thermal gradient affects a homogeneous fluid showing that, to linear order, no momentum flux is generated in the bulk. The presence of a colloidal particle, however, modifies the pressure tensor of the fluid, thereby creating a velocity field in the neighbourhood of the surface of the colloid, which can be calculated by use of the Navier–Stokes equation. As a consequence, momentum transfer to the colloid takes place. When this effective force is substituted into a Smoluchowski equation for the particle motion, we arrive at a simple and general expression of the Soret coefficient in terms of the colloid–liquid surface tension.

In aqueous solutions and for high concentrations, tri-block copolymers are known to aggregate. As a critical volume fraction of micelles is reached, they crystallize. We have studied the temperature dependence of the structure formed by Pluronic P85 micelles solved in deuterated water close to an attractive and a repulsive wall. We find that the cubic structure is suppressed for the repulsive surface and for both interfaces during cooling.

The high-molecular-weight paraffinic ('wax') fraction separates from crude oils at low temperatures, a process that can lead to a sol–gel transition when the mass of wax solids exceeds 1–2%. Attractive interactions between the micron-size wax solids suspended in the non-polar medium have been suggested to be responsible for gel formation. The present study reports an optically transparent model oil system, based on a mixture of linear and branched paraffins. Rheological measurements and optical microscopy show that the model system reproduces essential features of crude oil gels. Small-angle light scattering studies conducted at temperatures intermediate between the cloud point (58 °C) and sol–gel transition (39 °C) show that phase separation and wax solid aggregation are rapid processes, leading to the formation of dynamically arrested structures well above the sol–gel transition determined rheologically. Analysis of gravity settling effects has provided a rough estimate for the yield stress of the wax particle network formed (greater than 0.7 Pa at 45 °C and 0.07 Pa at 55 °C). Clusters formed by the aggregated wax solids possess a fractal dimension of about 1.8, consistent with diffusion-limited cluster–cluster aggregation.