The Ewoldt Research Group
University of Illinois at Urbana-Champaign
Department of Mechanical Science and Engineering








See publications for the most up-to-date representation of research activity.

Engineers will increasingly deal with soft, viscoelastic, and complex materials, including biological and non-Newtonian fluids. Such complex fluids are both unavoidable and opportunistic for novel functionality. The wide-ranging applications of complex fluids, from energy to biomedicine to robotics, motivate fundamental studies of fluid mechanics and rheology, which is the core of our research. In common terms, this is the study of how materials squish, ooze, stretch, flow, and deform.

Many biological systems involve a complex combination of viscous (fluid) and elastic (solid) material behavior for their function, including the locomotion of snails, the locomotion of ulcer-causing bacteria, and the predatory defense mechanism used by hagfish. New engineered functionality is now being achieved by deliberately using viscoelastic materials in engineered systems; wall-climbing robots and tunable magnetic fluid adhesion are examples. Future developments will be enabled by a better understanding and description of soft materials and complex fluids. Our research spans from engineering applications of nonlinear viscoelastic materials to the theoretical foundations for describing and characterizing these ubiquitous but technologically under-utilized materials.

Biological Fluids and Systems
hagfish, microalgae, snails, bacteria

Engineered Fluids and Systems
yield stress fluids, magnetorheological fluids, gel networks, adhesion and interfaces

General Rheology Tools
large amplitude oscillatory shear (LAOS), creep ringing



Hagfish slime: A volume-expanding
self-defense gel

Ocean-dwelling hagfishes are well known for their ability to produce large amounts of slime when they are provoked or stressed. They do this by ejecting a small amount of exudate which can expand into a large volume via turbulent surrounding seawater, forming a mucus-like cohesive mass. The resulting material is hypothesized to serve as a defense mechanism against predators. Here we are interested in the nonlinear mechanical properties of such a resulting cohesive mass of hagfish slime.

With biologists Prof. Doug Fudge and PhD student Tim Winegard (University of Guelph), we have reported the first experimental measurements of nonlinear rheological material properties of hagfish slime, a hydrated biopolymer/biofiber gel network, and developed a microstructural constitutive model to explain the observed nonlinear viscoelastic behavior. The linear elastic modulus of the network is observed to be G' ~ 2 Pa for timescales of 0.1s to 10s, making it one of the softest elastic biomaterials known. Nonlinear rheology is examined via simple shear deformation; by exploiting inertio-elastic ringing (see below), we experimentally observe a secant elastic modulus which strain-softens at large input strain while the local tangent elastic modulus strain-stiffens simultaneously. This juxtaposition of simultaneous softening and stiffening suggests a general network structure composed of nonlinear elastic strain-stiffening elements, here modeled as Finite Extensible Nonlinear Elastic (FENE) springs, in which network connections are destroyed as elements are stretched. We simulate the network model and show that it captures the simultaneous softening of the secant modulus and stiffening of tangent modulus as the model enters the nonlinear viscoelastic regime.

Ongoing work will examine the role that turbulent flow plays in the formation of the gel structure, the nonlinear rheology of sub-comonents of the hagfish exudate, and bioinspired materials that mimic the fascinating features of this material.

R. H. Ewoldt et al., Int. J. Nonlin. Mech., 2011
D. S. Fudge et al., Integr. Comp. Biol. 2009



All hagfishes, like this Pacific hagfish (Eptatretus stoutii), release exudate when provoked
(image courtesy Fudge et al. 2009)

Exudate (A) contains a mixture of mucin vescicles (arrowheads) and bundled threads (arrows) which, under flow, expand, stretch, and form an elastic network (B). In (C) the gelled material is loaded into a rheometer for testing with a concentric cylinder geometry.
(A and B images from Fudge et al. 2005)



Microalgae suspensions

The photosynthetic cells of algae produce an oily goo that can be converted to biofuels. Microalgae are considered one of the most promising feedstocks for biofuels, but challenges remain in the economical scale-up, including cost-effective growth chambers and efficient harvesting of oil from the cells. Algae processing is influenced by the viscosity and rheological properties of microalgae suspensions. This motivates a fundamental scientific question: how do actively swimming particles change suspension viscosity differently than passive particles?

With undergraduate student Lucas Caretta, and in collaboration with PhD student Anwar Chengala (Saint Anthony Falls Laboratory) and Prof. Jian Sheng (University of Minnesota), we are using a rotational rheometer to experimentally visualize and measure the flow properties for motile and non-motile suspensions of unicellular green algae (Dunaliella primolecta, a biflagellated ``puller''). The low viscosity biological samples require careful experimental protocols to avoid settling and flow-induced migration, and to minimize precision error. With these protocols in place we can distinguish the intrinsic viscosity of the suspensions, allowing us to put the results in the context of more traditional theories (e.g. Einstein 1906) related to the intrinsic viscosity of passive suspensions, and alo compare with recently proposed dilute-regime theories which predict that ``pullers'' should have a higher viscosity than non-motile suspensions.

R. H. Ewoldt et al., in preparation
A. A. Chengala et al., Biotechnol. Bioeng., 2010



Algal biofuels: fuel source of the future?
(image from Nature News)

Dunaliella Primolecta
algae are motile,
bi-flagellated, unicellular green algae
(image courtesy Anwar Chengala)


Bioinspired snail-like locomotion (retired)

Snails can climb walls and ceilings because they excrete and crawl upon a fluid with nonlinear rheology: a yield stress fluid. The remnants of this adhesive fluid can be seen as the trail left behind a crawling animal. The slime acts like a solid glue at rest, but flows as a fluid when an adequate stress is applied (a stress exceeding the apparent yield stress). When the stress is removed, the slime quickly re-solidifies. Such a material is known as a yield stress fluid. By exploiting this reversible solid-to-liquid transition, a snail can keep part of its foot stuck to the wall while another part moves forward. The picture on the lower-right shows the bottom of a Leopard Slug, Limax maximus, during locomotion.

Robosnail and slime simulants
We have demonstrated an engineered system which successfully utilizes the adhesive locomotion strategy of snails. A slime simulant with suitable rheological properties was developed, and used in conjunction with a robotic crawling mechanisms built by Brian Chan (Ph.D., MIT 2009). The system performed inclined and inverted traversals of a flat substrate, proving the feasibility of this concept. Non-Newtonian fluid design rules were developed to guide future optimization and development of this bioinspired innovation.

M.W. Denny, Nature, 1980
B. Chan et al., Phys. Fluids A, 2005
R.H. Ewoldt et al., Soft Matter, 2007
R.H. Ewoldt et al., Integr. Comp. Biol., 2009



Inverted locomotion of snails

Robosnail (by Brian Chan) successfully
mimics snail locomotion using an appropriately
designed slime simulant (by Randy Ewoldt)

MOVIE: Leopard_Slug.wmv (1.12 MB)
personal image
video courtesy of Brian Chan


Ulcer-causing bacteria locomotion (retired)

In collaboration with Dr. Jon Celli and the Biological Physics research group at Boston University, rheological measurements were used to reveal the pH-dependent sol-gel transition of mucin, the principal glycoprotein component of mucus. We have found that mucin solutions exhibit nearly critical gel behavior near pH4.

These rheological results enabled an additional study to identify the manner in which the ulcer-causing gastric pathogen Heliobacter pylori moves through the viscoelastic mucus gel that coats the stomach. Our study shows that the common perception of the helical bacterium moving in a corkscrew like manner through a viscoelastic gel is wrong. Instead, H. pylori produces the enzyme urease, which catalyzes hydrolysis of urea to yield ammonia thus elevating the pH of its environment. The elevated pH reduces mucin viscoelasticity into a sol phase, allowing free swimming.

J.P. Celli et al., Biomacromolecules, 2007
J.P. Celli et al., Proc. Natl. Acad. Sci., 2009



Image credit: Zina Deretsky,
National Science Foundation


Yield stress fluids

We use the term ``yield stress fluid'' pragmatically to refer to any material or model which exhibits a dramatic and reversible change in viscosity (orders of magnitude) over a small range of applied stress (Barnes and Walters 1985; Barnes 1999), i.e. a material that is ``solid'' at low stress and ``fluid'' at larger stress. Examples include whipped cream, peanut butter, toothpaste, and hair gel.

Yield stress fluid behavior is ubiquitous in food and personal care products, since it inhibits settling, provides functional use, and appeals to sensory perception. Many of the projects listed on this page involve yield stress fluids, including (snails), (bacteria), (magnetorheological fluids), and (large amplitude oscillatory shear - laos).



Examples of yield stress fluids include (A-D) whipped cream, peanut butter, toothpaste, and hair gel, all of which could be used as slime simulants to biomimic
snail locomotion



Tunable adhesion with magnetorheological yield stress fluids

We have demonstrated experimentally that field-responsive magnetorheological fluids can adhere to non-magnetic substrates. The tunable adhesive performance is measured experimentally with pull-off tests (a.k.a. probe-tack experiments) in which the external magnetic field and fluid geometry are varied.

The peak adhesive force is predicted by a lubrication model which treats the adhesive as a yield stress fluid with inhomogeneous yield stress (caused by the inhomogeneous magnetic field strength). The peak adhesive force, the 'work of adhesion' and the mode of failure are all controlled by the field-responsive nature of the magnetorheological fluid forming the adhesive layer.

Adhesive locomotion with a magnetorheological fluid could result in a robotic snail with virtually no trail left behind.

R.H. Ewoldt et al., Physics of Fluids, in press
R.H. Ewoldt et al., Bull. A.P.S., 2008 (APS-DFD)


Magnets can adhere to non-magnetic surfaces
using an intermediate field-responsive fluid



Gel networks

Gluten dough
New LAOS characterization measures have guided the development of an appropriate mathematical model for a representative critical gel: gluten dough. The observation of simultaneous softening (of average elasticity G1') and intra-cycle elastic stiffening lead to the development of a multimode transient FENE network model that is able to predict correctly the nonlinear stress waveforms over a range of oscillation frequency and strain amplitude in the Pipkin space (see image at right). This work was in collaboration with Dr. Trevor Ng and Prof. Gareth McKinley (M.I.T.)

Our experimental measurements of hydrogel rheology have contributed to the understanding of injectable hydrogels for ultrasound-triggered on-demand drug delivery (in collaboration with Dr. Hila Epstein-Barash, Prof. Daniel Kohane, and Prof. Robert Langer - Harvard/M.I.T.). Injectable hydrogels have applications to may other biomedical applications in which the rheology of the gel plays an important role.

Triblock copolymers
With Luca Martinetti (PhD student, University of Minneosta), we are developing mathematical models for the nonlinear rheology of other networks, such as those based on triblock copolymers.

T. S. K. Ng et al., J. Rheol., 2011
H. G. Epstein-Barash et al., Biomaterials, 2010




Lissajous-Bowditch curves for a gluten gel; experimental data (open symbols) are compared with a transient nonlinear network model (lines)


Adhesion and interfaces (retired)

Immiscible polymer interfaces
Adhesion can be promoted between two immiscible polymer melts with an interfacial copolymer reaction. We have mathematically modeled the transport problem of an interfacial reaction under coextrusion flow. Experimental results (Jie Song and Chris Macosko, University of Minnesota) show that compressive flow towards the interface increases the effective interfacial reaction rate, in a way that standard modeling does not predict.

Soft material adhesion
Ongoing work is examining the adhesion of nonlinear viscoelastic materials to real world substrates such as concrete.

J. Song et al., AIChE J., in press


Schematic of a coextrusion die
(from Song et al. in press)


Describing nonlinear viscoelasticity: LAOS

Viscosity and elastic modulus are meaningful ways to describe the mechanical behavior of fluids and solids, respectively. But for nonlinear viscoelastic materials the description is more complicated. We have developed a framework, or ontology, for interpreting nonlinear material behavior using Large Amplitude Oscillatory Shear (LAOS) deformation.

For many systems the common practice of reporting only "viscoelastic moduli" as calculated by commercial rheometers (typically the first harmonic Fourier coefficients G1' , G1") is insufficient and/or misleading in describing the nonlinear phenomena. Although the higher Fourier harmonics of the material response capture the mathematical structure, they lack a clear physical interpretation.

Part of our framework gives a physical interpretation to the third-order Fourier coefficients. We build on the earlier geometrical interpretation of Cho et al. (2005) which decomposes a nonlinear stress response into elastic and viscous stress contributions using symmetry arguments. We then use Chebyshev polynomials (closely related to the Fourier decomposition) as orthonormal basis functions to further decompose these stresses into harmonic components having physical interpretations.

We also introduce new measures for reporting the first-order (linear) viscoelastic moduli. These measures give deeper physical insight than reporting only the first harmonic Fourier coefficients G1' , G1", and reduce to the linear viscoelastic framework of G', G" at small strains.

Software is available for analyzing raw data with this framework. To request MITlaos (free, but requires a MATLAB installation), please contact

K.S. Cho et al., J. Rheol., 2005
R.H. Ewoldt et al., J. Rheol., 2008
P.B. Winter et al., MITlaos User Manual
R.H. Ewoldt et al. Rheol. Acta, 2010
R.H. Ewoldt and G. H. McKinley, Rheol. Acta, 2010


3-D Lissajous-Bowditch curve representation
of large amplitude oscillatory deformation


Software developed for data analysis


Inertio-elastic ringing (retired)

The finite rotational inertia of a stress-controlled rheometer introduces experimental artifacts during creep tests that are often misunderstood. We have reviewed these artifacts and summarized the experimental techniques available to extract viscoelastic material parameters from this oft-discarded data.

R.H. Ewoldt and G.H. McKinley, Rheol. Bull., 2007