Research group

Our group's research concerns the theoretical and computational rheology (deformation and flow properties) of soft materials such as colloids, emulsions, foams, microgels, surfactants, liquid crystals, and polymers. A particular current focus is on uncovering general principles to underpin a new understanding of how these materials yield from an initially solid-like to finally liquid-like state. In particular, we seek to understand how strain localises during yielding, in both shear and extensional flows, destabilising processing flows and sometimes causing catastrophic material failure. Research methods used in this work range from direct particle simulations through mesoscale statistical mechanical models to continuum fluid-dynamical approaches. A broader aim is to cross fertlise the understanding that we develop here in the context of soft materials to help understand strain localisation during yielding in plastic solids such as metallic and polymeric glasses, in geophysical systems such as avalanches, mudslides, earthquakes and debris flows, and in the way biological tissue reshapes itself under the internal stresses generated by cell division. We are also interested in flow instabilities more generally, such as at free surfaces and in the way fluids demix in an applied flow.

The work forms several distinct yet related strands:

  1. Shear banding in complex fluids.
  2. Extensional necking instabilities in complex fluids.
  3. Edge fracture in complex fluids.
  4. Fluid-fluid demixing.
  5. The transition to viscoelastic turbulence.
  6. Biologically active suspensions.
  7. Glassy ageing dynamics of disordered soft materials.
Further details on each of these topics are below.

Current group members are:

  • Suzanne Fielding, Professor
  • Ewan Hemingway, Research Associate
  • Rahul Chacko, PhD student
  • Clara Despard, PhD student
  • Hugh Barlow, PhD student
  • Gerhard Jung, PhD student, visiting us from Mainz, 2018.

    1. Shear banding in complex fluids

    Complex fluids such as surfactant solutions, polymers and liquid crystals commonly show flow instabilities when subject to shear. Often these instabilities lead to the formation of macroscopic "shear bands" of differing viscosity and internal structure.


    This effect can be viewed as a non-equilibrium (flow-induced) phase transition. Indeed, much of the phenomenology mirrors that of conventional (equilibrium) phase transitions, and I have pursued this analogy to study the kinetics of band formation, as well as to construct flow phase diagrams for the ultimate banded state. At the same time, there are fundamental differences from the equilibrium case, for example in the way the coexistence state is selected in the absence of a free energy minimisation principle.

    Particularly exciting developments have concerned the chaotic dynamics of shear banded flows, studying both the possibility of a bulk instability of one band, as shown in this shear rate greyscale:

    Chaotic bands

    and of an instability of the interface between the bands:


    The most recent focus has been on establishing that, besides those fluids that show shear banding as their ultimate long-time response to a steadily imposed shear flow, many (indeed possibly even all) complex fluids have a tendency to display banding in flows with a strong enough time-dependence, even if they are incapable of supporting it as their ultimate response to steady shear. This tendency can manifest itself transiently (for example in a shear startup protocol, during the process whereby a steady flowing state is estabilished out of an initial rest state), or in a sustained way in imposed flows (such as large amplitude oscillatory shear) that themselves have a sustained time-dependence.


  • Dr. James Adams (University of Surrey)
  • Prof. Peter Olmsted (Georgetown University)
  • Prof. Patrick Tabeling (ESPCI, Paris)
  • Prof. Helen Wilson (University College London)

    3. Extensional necking instabilities

    Besides shear flows, complex fluids are also widely subject to extensional deformations. In an industrial context these form the basis of spinning polymeric materials into fibres for textiles, for example. In fluid dynamical terms, extensional flows cause material elements to separate exponentially quickly and so subject the underlying macromolecules to greater stretching and reorientation than shear. Indeed extensional flow response is very sensitive to underlying molecular details (linear vs. branched polymer chains, for example), and many nonlinear flow effects manifest themselves only in extension.

    Experimentally, notable extensional devices include the filament stretching rheometer in which a cylindrical sample of fluid is sandwiched between concentric circular end-plates that are then pulled apart to draw the sample out in length. Though the aim in such devices is to achieve homogeneous extensional flow for convenient comparison with theory, more complicated features often arise. Particularly serious is the widespread occurrence of extensional necking instabilities in which any small indentations in cross-sectional area become ever more pronounced until part of the sample thins to the point of failures. Recent work has focused on deriving a general criterion for the onset of necking, applicable to a broad class of viscoelastic materials. Phot


  • Prof. Ole Hassager (DTU Copenhagen)

    2. Edge fracture in complex fluids

    Edge fracture is a free surface instability that arises almost ubiquitously when a viscoelastic material is sheared in an open flow cell. It has been cited as the most significant limiting factor in experimental shear rheometry. The aim of this work is to elucidate a full mechanistic understanding of this phenomenon in a combined theory/simulation study, and to suggest a way experimentalists might mitigate it, potentially enabling them to achieve and measure faster flows than hitherto.



  • Prof. Dimitris Vlassopoulos (FORTH, Crete).
  • Prof. Roger Tanner (University of Sydney)

    4. Fluid-fluid demixing

    When an initially homogeneous mixture of two fluids (A and B) undergoes a deep temperature quench into the spinodal regime, it phase separates into well defined domains of A-rich and B-rich fluid. These then slowly coarsen in time through the action of the surface tension in the interfaces that separate them, such that the excess interfacial energy of the system progressively relaxes towards its minimal equilibrium value. This coarsening process proceeds through three distinct regimes that are successively dominated by diffusive, viscous and inertial dynamics. In the limit of an infinite system size, the typical domain size perpetually increases without bound: the system never globally equilibrates, even in the limit of infinite time.

    Here, we consider systems that are both undergoing phase separation and simultaneously subject to an applied shear flow. The main question that we address is whether shear interrupts domain coarsening to give a nonequilibrium steady state with a typical domain size set by the inverse of the applied shear rate; or whether coarsening persists indefinitely, up to the system size, as in zero shear. For systems with inertia we have reproduced the nonequilibrium steady states reported in a recent lattice Boltzmann study by the group of Cates. The domain coarsening that would occur in zero shear is arrested by the applied shear flow, which restores a finite domain size set by the inverse shear rate. For inertialess systems, in contrast, we have found no evidence of nonequilibrium steady states free of finite size effects: coarsening persists indefinitely until the typical domain size attains the system size, as in zero shear.

    Chaotic bands

    5. Transition to viscoelastic turbulence

    In simple fluids, hydrodynamic instabilities and turbulent flows have long been known to arise at high Reynolds number, due to the nonlinearity inherent in the Navier Stokes equation. In the viscoelastic fluids and flow regimes of interest here, the Reynolds number is negligible. Despite this, the inherent nonlinearity of a polymeric material's constitutive response can cause purely viscoelastic instabilities. In curved flow devices, these include an inertialess Taylor-Couette instability. In plate-plate flow, a pathway through to fully developed viscoelastic turbulence has recently been demonstrated experimentally by the group of Larson.

    Laminar viscoelastic flows with parallel streamlines are known to be linearly stable: that is, able to resist perturbations of tiny amplitude. Despite this, a nonlinear instability with respect to perturbations of a finite amplitude was recently predicted by Morozov et al. My work has focused on testing this prediction numerically within a popular model of polymeric flows.

    6. Rheology of soft glassy materials

    This work brings a unifying approach to the rheological (deformation and flow) behaviour of a broad class of soft materials, including:

  • foams
  • dense emulsions
  • colloidal suspensions
  • surfactant onion phases
  • gel bead suspensions

    The essential hypothesis is that all these materials share the basic features of disorder and metastability, which result in an underlying glassiness in the rheological response. This approach has successfully explained a broad class of existing rheological data. Furthermore it has led to new predictions of rheological ageing (slow progression towards a more elastic state), now widely being studied experimentally.


  • Prof. Mike Cates, FRS (Cambridge University).
  • Prof. Peter Sollich (King's College London).

    7. Biologically active suspensions

    In addition to fluids that are taken far from equilibrium by external driving (shearing applied at the boundaries, as in the above example), another even more challenging class of complex fluids concerns those that are driven out of equilibrium by an activity inherent within their own bulk. Examples include swarms of self-propelled bacteria or protozoa; and the viscoelastic matrix of the biological cell, in which molecular motors at cross-links between polymeric strands render the network as a whole capable of mechanical motion (in cell division or amoebic crawling). Here each mesoscopic substructure (bacterium/motor) itself individually consumes energy, and so can actively propel itself: ``swimming'' through the suspending fluid, or ``marching'' along a neighbouring cytoskeletal filament. Their collective dynamics is thus inherently far from equilibrium, even without boundary driving. Indeed in this new class of fluids the driving away from equilibrium occurs from within the volume of the fluid itself. Emergent phenomena include hydrodynamic instabilities in which an initially quiescent fluid gives way to shear banded (establishing a link to (i) above) or turbulently swirling patterns, as seen experimentally and reproduced in my simulations below.

    Phot Phot


  • Prof. Mike Cates, FRS (Cambridge University).
  • Prof. Ramin Golestanian (Oxford University).
  • Prof. Tanniemola Liverpool (Bristol University).
  • Prof. Davide Marenduzzo (Edinburgh University).
  • Prof. Sriram Ramaswamy (IISc Bangalore).
  • Prof. Julia Yeomans, FRS (Oxford University).

  • Home Biography Publications Presentations Teaching

    Last updated: 3th September 2018