Physica A 631 (2023) 128653

Contents lists available at ScienceDirect

Physica A
journal homepage: www.elsevier.com/locate/physa

Introduction to viscoelasticity and plasticity, and their relation

to the underlying microscopic dynamics in soft matter systems
∗
Cerbino Roberto a , , Trappe Véronique b
a
Faculty of Physics, University of Vienna, Boltzmanngasse 5, Vienna, 1090, Austria
b
Department of Physics, University of Fribourg, Chemin du Musée 3, Fribourg, 1700, Switzerland

article info a b s t r a c t

Article history: Understanding the rheological behavior of complex fluids, and in particular its relation to
Available online 9 March 2023 the dynamics of the elementary microscopic constituents, is one of the open challenges
Keywords:
in soft matter science. In this contribution, we give a brief introduction to linear shear
Soft glassy materials rheology and passive microrheology of soft amorphous materials, and we then focus
Non-linear rheology on the transition to non-linearity during oscillatory rheology tests performed on soft
Yielding glassy materials. These materials exhibit solid-like properties at rest, but they display
Shear induced dynamics characteristics of fluids when a sufficiently large shear is applied. Significant progress in
Plastic rearrangements understanding such ‘fluidization’ has been recently achieved by combining rheometry
Microrheology with techniques capable to capture the microscopic dynamics induced by shear, and in
Large amplitude oscillatory shear
particular by oscillatory shear. We here summarize some of these achievements, as an
introduction to the field for a non-expert reader.
© 2023 The Author(s). Published by Elsevier B.V. This is an open access article under the CC
BY license (http://creativecommons.org/licenses/by/4.0/).

1. Introduction

When a mechanical load is applied to a generic atomic or molecular solid, it causes a deformation. For each specific type
of load/deformation (e.g. shear, tension, compression, etc.), it is generally true that for small enough loads the deformation
is proportional to the load, and that it goes back to zero when the load is removed (see Fig. 1, elastic regime). If the load
force is normalized with a suitable area, and the deformation with a suitable reference length, one obtains the stress and
the strain, respectively. For a fixed type of load/deformation, the stress–strain curve represents the mechanical fingerprint
of a given material; in particular, the proportionality constant between stress and strain in the linear elastic regime is the
elastic modulus of the material.
If the load is further increased, one may first encounter a non-linear elastic regime, but eventually enters the plastic
regime (see Fig. 1, plastic regime), characterized by irreversible, permanent deformation of the solid after removal of
the load. The yield point (Fig. 1, A) marks the onset of the plastic regime, which extends up to the fracture point (Fig. 1,
B), where the material eventually breaks. There are many different types of fractures that can occur in solids, including
tensile fractures, compressive fractures and shear fractures, among others. The specific type of fracture that occurs in a
given material will depend on the properties of the material and the nature of the applied force. Moreover, depending on
the material one may observe either brittle or ductile fracture. Brittle fracture occurs when a material fractures without
significant plastic deformation, while ductile fracture occurs when a material undergoes significant plastic deformation
before fracturing. The ductility of a material depends on its physical state: glass is notoriously brittle at low temperature
but can be mechanically worked and shaped without fracturing when heated; on the contrary, metals are normally ductile

∗ Corresponding author.
E-mail address: roberto.cerbino@univie.ac.at (C. Roberto).

https://doi.org/10.1016/j.physa.2023.128653
0378-4371/© 2023 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/
licenses/by/4.0/).
C. Roberto and T. Véronique Physica A 631 (2023) 128653

Fig. 1. Schematic idealized load deformation curve for a generic atomic/molecular solid: the elastic (continuous black line) and plastic (dashed red
line) regime are separated by the yield point A, whereas B is the fracture point. The details of these curve will depend on the rate at which the
load or deformation are imposed.

but can be made brittle by lowering their temperature. These are known facts to bike thieves, who use liquid nitrogen to
lower the temperature of metal locks before easily breaking them with a hammer [1].
In crystalline solids, plasticity and fracture are rather well understood, with topological defects in the crystalline lattice
(dislocations) playing a key role. Edge dislocations, corresponding to one missing crystalline plane in the lattice, and screw
dislocations with a lattice plane shifted by one layer have been long identified as the key ingredients leading to plastic
deformation in crystals [2]. When atomic, crystalline solids are continuously sheared, local yield events (dislocations) are
collective and very often self-organize into avalanches that exhibit power-law distributions of characteristic variables [3,4].
Similar observations were also recently made for colloidal crystals [5]. By contrast, the absence of a symmetric reference
structure in amorphous glassy systems makes it difficult to identify a suitable counterpart of topological defects.
Consequently, identifying the mechanism according to which plastic deformation occurs in amorphous solids is currently
one of the major unsolved problems in condensed matter physics.
To gain further insight, an increasingly popular class of materials to study is that of amorphous soft materials, such as
concentrated colloidal suspensions, polymer solutions, polymer melts, gels, emulsions, foams, and many others. All these
materials are characterized by a microstructural lengthscale that is way larger than the typical scale of atomic/molecular
systems, which comes with important experimental advantages: first, the larger size of the elementary constituents
facilitates their study with optical microscopy and light scattering; in addition, it is ultimately responsible for the elastic
modulus to be way smaller (hence soft materials) than the typical values (10–100 GPa) observed for atomic and molecular
solids. A paradigmatic example is a dense suspension of colloidal particles interacting solely by excluded volume; for
such colloidal hard sphere glass, the microstructure is solely governed by entropy [6], such that the elastic modulus is
∼ kB T /d3 , with kB the Boltzmann constant, T the absolute temperature, and d ∼ [10, 1000] nm is the size of the colloidal
particles, so that the elastic modulus is typically below 1 kPa.
For sufficiently small loads and deformations, amorphous soft materials also exhibit a linear regime. However, this
regime is not purely elastic as it is for the idealized atomic/molecular solid portrayed in Fig. 1: instead, a combination of
a solid-like and fluid-like responses across timescales is observed, which is why this linear regime is referred to as linear
viscoelastic regime [7]. In this regime, it is useful to conceptually distinguish between two types of viscoelastic materials:
the viscoelastic fluid and the viscoelastic solid. The distinction between the two relates to their long-time behavior, the
fluid being characterize by a long-time relaxation and the solid by a long-time elastic modulus. A wide range of viscoelastic
solids belong to the class of so-called yield stress materials [8]1 : they deform in a reversible, elastic fashion upon applying
a small external force, a condition in which the material microstructure is not altered by the load; however, when the
force is sufficiently increased they will eventually exhibit a constant deformation rate, which is a characteristic of a fluid.
Although, as exemplified above for the colloidal glass, elasticity can be understood in terms of the material microstructure,
it is not clear whether and how the full macroscopic rheological properties of yield stress materials can be mapped onto
the structure and dynamics of the material elementary constituent. In particular, yielding and the subsequent plastic
deformation are determined by a series of microscopic events whose nature has not yet been fully elucidated.
During the last years, important progress has been made thanks to the combination of studies on hard and soft
amorphous materials [9]. Theoretical work and simulation suggested that plastic deformation in amorphous systems

1 For simplicity, we consider here only the ideal case of simple yield stress materials, exhibiting no flow until a critical stress is reached, at which
point they will eventually exhibit a steady state shear rate; this choice leaves out important other cases such as thixotropic yield stress materials,
which are characterized by shear history dependent behavior, with viscosity and yield point depending on time and shear; interested readers can
refer to Ref. [8] and references therein.

2
C. Roberto and T. Véronique Physica A 631 (2023) 128653

Fig. 2. A sheared material is subjected to a pair of opposite forces of amplitude F that are exerted on two opposing surfaces with area A; the shear
stress is defined as σ = F /A, and the shear strain as γ = ∆x/y. The mechanical response function of an ideal (Hookean) solid is described by
σ = Gγ (G is the shear modulus, units Pa), that of an ideal (Newtonian) liquid is described by σ = ηγ̇ (η is the shear viscosity, units Pa s, and
γ̇ = dγ /dt).

results from the microscopic occurrence of local rearrangements involving few particles (of the order of a few units)
in well-defined regions, the so-called shear transformation zones (STZ). Such rearrangements trigger anisotropic stresses
(strains), known as Eshelby stresses (strains), that exhibit characteristic quadrupolar spatial correlations. A simple yet
paradigmatic example of Eshelby events is found in foams, where they are more commonly referred to as T1 or neighbor
switching events [10]. These perturbations cause deformations over large distances, which in turn can trigger avalanches
that give rise to power-law behavior [9,11–14]. As far as experiments are concerned, imaging atoms or molecules under
mechanical stress is generally beyond reach, since the constituents are too small and move too quickly. For this reason,
most experimental results have been obtained with colloidal particles that enable direct visualization of the particle
rearrangements occurring during shear [15–18]. These experiments confirmed that hard-sphere colloidal glasses, when
extremely slowly sheared (γ̇ ∼ 10−5 s−1 ), exhibit fourfold patterns in the spatial autocorrelation of the local strain field,
identifying shear transformation zones as the main mechanism for deformation in glasses [11]. In fact, STZ behave like
local regions of large plastic strain that couple to the surrounding material with long-range elastic interactions, producing
long-range correlations in the strain field and collective rearrangements. In particular, the continuously sheared glass
was found to grow clusters of particles that exhibit non-affine rearrangements; the size of the clusters diverges upon
approaching the yielding transition, resembling a percolation transition [18].
So far, this introduction focused on results obtained upon applying a constant stress or continuous shear. However,
there are several interesting experimental, theoretical, and numerical results that have been obtained in oscillatory shear,
a condition that has attracted much interest over the years for a variety of reasons. One of them is that a periodic
mechanical perturbation of the sample at the macroscopic scale, offers a quite convenient setting to explore microscopic
rearrangements occurring within the sample at each perturbation cycle, a condition that is increasingly exploited in
experiments.
These lecture notes attempt to provide a self-contained presentation of these oscillatory shear studies to a non-
specialized audience, in primis to the students of the XV Summer School Fundamental Problems in Statistical Physics
(Bruneck, 11–24 July 2021) for whom they were originally conceived. The reader should take into account that, mainly for
space limitations, several aspects of the treatment of the different topics have been simplified. As an example, a correct
treatment of rheological problems makes use of second-rank tensors (stress, deformation, deformation rate) that capture
the anisotropy of the rheological properties of the material and non-trivial coupling between stress in one direction and
deformation in a different one. For such tensorial treatment, which goes beyond the scope of the present article, we refer
the reader to classical books such as the one by Makosco [19].

2. Linear shear rheology and microrheology

2.1. Elasticity and viscosity

According to its founding father E.C. Bingham (1878–1946), rheology is ‘‘the science of deformation and flow‘‘. Typical
rheology experiments make use of instruments named rheometers, which can apply controlled stress or strain to the soft
material to be investigated and in turn measure the resulting strain or stress, respectively. One of the most widespread
choices is to make use of the so-called rotational shear rheometers, advanced instruments widely used in academic and
industrial laboratories. In a rotational shear rheometer, the sample is typically confined between a fixed surface and a
surface that can rotate and impose a shear deformation to the material sample comprised between the two surfaces.
The typical effect of shearing is illustrated in Fig. 2, where we take a small cubic portion of our soft material, such that
the shear can be considered uniform across the cube. The force F applied over the surface area A of the cube defines the
shear stress σ = F /A, while the shear strain is defined by γ = ∆x/y.
3
C. Roberto and T. Véronique Physica A 631 (2023) 128653

Fig. 3. (Left column) Creep and recovery test: (a0) a step stress is imposed at t0 and subsequently released at ts ; (a1) strain response of an
ideal elastic solid; (a2) strain response of an ideal viscous fluid; (a3) strain response of an ideal viscoelastic solid; (a4) strain response of an ideal
viscoelastic fluid. (right column) Stress relaxation test: (b0) a step strain is imposed at t0 ; (b1) stress response of an ideal elastic solid; (b2) stress
response of an ideal viscous fluid; (b3) stress response of an ideal viscoelastic solid; (b4) stress response of an ideal viscoelastic fluid.

For a shear deformation, the simplest material response is that of an ideal (Hookean) solid, which behaves like a spring
by deforming elastically and reversibly for arbitrarily large shearing amplitude. With the definitions given above, there
is a simple proportionality relation between the stress and the strain (for Hooke solids): σ = Gγ , where G is the shear
modulus (SI units Pa). As we saw in Fig. 1, such ideal solid does not exist in practice; nevertheless, a hard solid in the
linear elastic regime behaves to a good approximation as a Hookean solid. Another useful limit behavior is that of an ideal
(Newtonian) liquid: when a constant shear stress is applied to it, a linear increase of the strain is observed, where the
strain rate γ̇ = dγ /dt is directly proportional to the stress: σ = ηγ̇ , with η the dynamic (shear) viscosity (SI units Pa s).
While an ideal elastic solid stores energy that is returned when the shear force is removed, for an ideal viscous liquid,
energy is irreversibly dissipated during shearing.

2.2. Viscoelasticity: Creep, recovery and stress relaxation

The simple behavior of these ideal materials can be used to understand the more complex response of soft materials
exhibiting viscoelasticity. The concept is illustrated in Fig. 3, where we model elastic contributions by springs and viscous
contributions by dashpots. In this figure, we present two standard rheological tests: the creep and recovery test performed
by applying a constant stress at some time t0 , and subsequently setting the stress to zero at some time ts while monitoring
the strain response of the material in time (left column in Fig. 3); the stress relaxation test performed by applying a step
strain at some time t0 while monitoring the stress response in time (right column in Fig. 3).
4
C. Roberto and T. Véronique Physica A 631 (2023) 128653

Subjecting a Hookean solid to a step-stress results in an instantaneous strain γ (t0 ) that remains constant until the stress
is set to zero, whereupon the strain is fully recovered, as shown in Fig. 3(a1). By contrast, the response of a Newtonian
fluid to a step stress is a linearly increasing strain (Fig. 3(a2)); upon setting the stress to zero at ts the strain remains at
the value γ (ts ).
The situation is more complex for a viscoelastic material. A prototypical viscoelastic solid is described by the Kelvin–
Voigt model in which a spring and a dashpot are connected in parallel, such that the long-time behavior of the system is
governed by the spring. Because of the presence of the dashpot, the response of a viscoelastic solid to a step-stress is not
instantaneous, as shown in Fig. 3(a3). The strain increases slowly to eventually reach a constant strain corresponding to
the extension of the spring expected for the imposed stress. Correspondingly, the strain goes back to zero during recovery,
but it takes some time to do so.
Another type of viscoelastic material is the viscoelastic fluid, characterized by a fluid-like response at long times.
The Maxwell model, consisting of a serial connection of a spring and a dashpot, describes the general behavior of these
materials. Upon applying a step stress, the spring is instantaneously stretched to γ (t0 ), whereupon the strain linearly
increases due to the viscous contribution described by the dashpot. Upon releasing the stress, the spring instantaneously
recoils while the strain acquired during the extension of the dashpot is not recovered, as shown in Fig. 3(a4).
While the strain response of a material to a given stress is uniquely defined by the magnitude of the stress applied,
the relation of stress and strain are linearly related as long as the applied stress remains sufficiently small to not induce
structural relaxation processes in addition to those occurring at rest. In this linear regime, the mechanical response solely
reflects the quiescent properties of the system. Varying the stress merely leads to a change of the magnitude of the
strain or strain-rate, but not to a general change of the time dependent behavior. It is thus generally more useful to
describe the material response function by the compliance, which is the strain normalized by the applied (constant) stress,
J(t) = γ (t)/σ . Any test series probing the strain response at different applied stresses collapses to a unique curve when
reported in terms of a compliance, as long the stresses applied are within the linear regime.
Equivalently, experiments probing the stress response to a step-strain within the linear regime, collapses to a unique
curve when reported in terms of a shear modulus, which is the stress normalized by the applied strain, G(t) = σ (t)/γ , also
known as stress-relaxation modulus. For a Hookean solid, applying a step strain as in Fig. 3(b0) leads to an instantaneous
stress that remains constant as long the strain is applied (Fig. 3(b1)). By contrast, a Newtonian fluid subjected to a step
strain at t0 experiences a non-zero strain rate only for t = t0 . This causes a very short, instantaneous stress peak for t = t0
that immediately relaxes to zero again, as denoted in Fig. 3(b2).
Likewise, imposing a step-strain to a Kelvin–Voigt element leads to an instantaneous large increase of the stress due to
the presence of the dashpot, which then relaxes immediately to the stress amplitude expected for the prescribed extension
of the spring (Fig. 3(b3)). However, the most meaningful stress relaxation test is that performed with Maxwellian fluids.
After the application of a step-strain at t0 , such viscoelastic fluid initially reacts like a solid to then show an exponential
stress relaxation, as shown in Fig. 3(b4).
More complex models combining springs and dashpots are typically used to get a description of the rheology of real
soft materials. More details can be found in advanced rheology books [19].

2.3. Viscoelasticity: Linear oscillatory tests and microrheology

Beyond investigating the material response to a sudden onset of a constant stress or strain, another test that is
particularly instructive for investigating the linear rheology of soft materials is the small amplitude oscillatory shear (SAOS)
test. In these tests, a sinusoidally varying shear strain γ (t) = γ0 sin ωt is imposed and the resulting sinusoidal shear stress
σ (t) = σ0 sin (ωt + δ ) is measured. For the ideal Hookean solid, the stress linearly relates to the strain, such that the
oscillating stress will be in phase with the strain (δ = 0) independent of the frequency ω applied. By contrast, for the
ideal Newtonian fluid, where the stress linearly relates to the strain rate, the stress will be in quadrature of phase with
the strain (δ = π /2), as illustrated in Fig. 4.
For viscoelastic systems, δ will adopt values between 0 and π/2, which will depend on frequency. In this case,
it is useful[ to decompose the stress ]response in two parts: the in-phase and the quadrature-of-phase component,
σ (t) = γ0 G′ (ω) sin ωt + G"(ω) cos ωt , where the storage (or elastic) modulus G′ (ω) relates to the energy stored per
unit volume and the loss (or viscous) modulus G"(ω) is proportional to the rate at which energy is dissipated [20]. In
a frequency-sweep experiment, one fixes the magnitude of the sinusoidal strain γ0 in the linear regime, and performs a
systematic investigation of the storage and loss moduli as a function of the oscillation frequency. For the two prototypical
viscoelastic models, describing a viscoelastic solid (Kelvin–Voigt model) and a viscoelastic fluid (Maxwell model) the
frequency dependences of G′ (ω) and G"(ω) are shown in respectively Fig. 4(b) and (c). Characteristic of the Kelvin–Voigt
model is that the storage modulus is frequency independent, while the loss modulus linearly increases with frequency.
This necessarily implies that the storage modulus will dominate in the low-frequency range and that it is finite at ω = 0,
which characterizes a solid system. By contrast, for the Maxwell system G′ (ω) dominates only at high frequencies, while
at low frequencies G"(ω) becomes the dominant modulus, in addition to exhibiting the linear frequency dependence
characterizing viscous fluids.
In real viscoelastic systems, the response functions are never fully described by either of these models. This is illustrated
in Fig. 5, where we show the experimental data obtained for a dense emulsion [21] in (a) and an entangled worm-
like micellar system [22,23] in (b). Dense emulsions are generally considered to be soft glassy materials and as such
5
C. Roberto and T. Véronique Physica A 631 (2023) 128653

Fig. 4. Oscillatory strain test: (a0) an oscillatory stain is imposed; (a1) oscillatory stress response of an ideal elastic solid; (a2) oscillatory stress
response of an ideal viscous fluid; (a3) oscillatory stress response of a viscoelastic system. Frequency dependence of G′ (red continuous line) and G"
(black dashed line) for (b) an ideal viscoelastic solid and (c) an ideal viscoelastic fluid.

to belong to the class of viscoelastic solids. Indeed, the almost frequency independent behavior of G′ (ω) is reminiscent of
the characteristics of the Kelvin–Voigt model, G"(ω) is however clearly not a linear function of frequency. Instead, G"(ω)
exhibits a shallow minimum and develops almost parallel to G′ (ω), indicating that some dissipative process contributes to
the low frequency response of the system. Similarly, the worm-like micellar system, considered to be a typical viscoelastic
fluid, exhibits the characteristics of a Maxwell fluid only at low frequencies. Upon increasing the frequency sufficiently
G"(ω) increases again due to dissipation induced by fluctuations of segments of the micelles. Clearly, to predict and
interpret the frequency dependent response of a system one needs to consider all possible dynamic contributions of
the system at hand.
6
C. Roberto and T. Véronique Physica A 631 (2023) 128653

Fig. 5. Frequency dependence of G′ (red squares) and G" (black triangles) obtained within the linear range of viscoelasticity for (a) a dense
emulsion [21] and (b) entangled wormlike micelles [22].

2.4. Passive microrheology

To perform a classical linear rheological experiment, we generally need to ensure that the mechanical perturbation
imposed to probe the mechanical response of a system is small enough to not induce any structural rearrangement within
the system. This condition is naturally fulfilled when using passive microrheology as a means to probe mechanical response
functions. The principle of passive microrheology is based on the relation between the thermal motion of a probe particle
and the mechanical properties of the material in which it is embedded [24].
To set the stage, let us first consider two classical examples [25]. The first is the case of a freely diffusing spherical
particle of mass m and radius R embedded in a Newtonian fluid with viscosity η. We know from Stokes-law that such
a particle would experience a friction force F⃗f = −ζ v ⃗ , where v⃗ is the particle velocity, and ζ = 6πηR is the friction
coefficient. At the same time, the molecules of the fluid collide with the particle, which results in a fluctuating random
force F⃗r (t) exerted on the particle. The resulting motion of the particle can be described by a Langevin equation that in
one-dimension (1D) reads mv̇ = −ζ v + Fr (t). For a delta-correlated random force, and for t ≫ m/ζ , one obtains the key
result ⟨∆x2 ⟩ = 2D0 t for the particle mean-square displacement, where the diffusion coefficient D0 of the particle is given
by the Stokes–Einstein relation D0 = kB T /ζ . This well known result implies that measuring the diffusion coefficient D0 of
a colloidal particle of known radius allows us to determine the viscosity of the medium via the relation η = kB T /(6π D0 R).
The second example is obtained when the particle, still embedded in the same viscous medium, is also attached to
a spring of elastic constant k. At first approximation, one can imagine that this elastic constant is representative of the
elastic modulus of the material ( [24]. The
) particle in such a harmonic trap is still described by a Langevin equation, whose
solution gives ⟨∆x2 ⟩ = kB 1 − e−t /τ for the mean-square displacement [25]. The emerging picture is here that for times
2k T

smaller than the characteristic time τ = ζ /k the particle is freely diffusing with ⟨∆x2 ⟩ = 2D0 t, whereas for large times the
limit ⟨∆x2 ⟩max = kB T /k is reached. We can thus simply consider that the behavior of a particle in an ideal Hookean solid
(η → 0) can be well represented by a constant mean-square displacement ⟨∆x2 ⟩HS = 2kB T /k, which looks surprisingly
similar to the behavior expected for a particle embedded in a Newtonian liquid ⟨∆x2 ⟩NL = (2kB T /ζ )t. As noted in [24], the
analogy can be brought even further when the shear modulus G is much smaller than the bulk compressional modulus
K . Under this condition one can write k ≃ 6π GR as the analog of ζ = 6πηR, which connects the shear modulus G of the
Hookian solid to the properties of the motion of embedded particles.
At this stage the reader may have already noticed, that the mean-square displacement of a particle embedded in a
Newtonian liquid and in a Hookean solid resembles expectations for the compliance J(t) (the response of a material to a
unit step-stress): the compliance for a Hooke solid is constant in time (Fig. 3a1), whereas it increases linearly with time
for a Newtonian liquid (Fig. 3a2). This resemblance is not incidental, as it can be proven that the probe particle is actually
a stress-controlled rheometer operating at the microscale and fueled by the random force F⃗r , as first suggested in Ref. [26].
For instance, the motion of the Brownian particle attached to a spring described above, is the microscopic equivalent of
the Kelvin–Voigt model (see Fig. 3a3). More generally, it is possible to show that for a generic viscoelastic material the
relation
3π R
J(t) = ⟨∆ x 2 ⟩ (1)
kB T
holds in 1D (in 2D one should include a factor two at the denominator, and in 3D a factor three).
Equivalently, instead of focusing on the compliance J(t) one can numerically transform the experimentally determined
mean-square displacement to obtain the frequency-dependent viscoelastic moduli G′ (ω) and G"(ω), which can be
considered as real and imaginary parts of the so-called complex modulus G∗ (ω) = G′ (ω) + iG"(ω). Mathematically, one
can write a generalized Stokes–Einstein relation (GSER)
⏐
kB T
G (ω ) =
∗
⏐
⏐ (2)
3π Rs⟨∆x2 (s)⟩ ⏐s=iω
7
C. Roberto and T. Véronique Physica A 631 (2023) 128653

Fig. 6. Example of passive microrheology experiment taken from Ref. [28]. The sample is a polyethylene oxide solution in water, which is seeded
with tracer particles of radius R. The mean-square displacement of the tracer particles (multiplied by R) shown in (a) can be converted to the
viscoelastic moduli G′ (red full squares) and G" (black full triangles) shown in (b). The data obtained in standard rheology, shown as open symbols,
compares well with those obtained in microrheology.

where ⟨∆x2 (s)⟩ is the Laplace transform of the mean-square displacement and the substitution s = iω corresponds to
analytical continuation. An example of such transformation is shown in Fig. 6. The interested reader can refer e.g. to
Ref. [24] for a more detailed treatment, as well as for a thorough description of the conditions under which passive
microrheology is expected to provide a meaningful characterization of the moduli of the viscoelastic materials. Here, it
shall be sufficient to point out that the probe particle must be large compared to any relevant characteristic length scale
of the material to ensure that the positional fluctuations of the particle are representative of the material loss and storage
moduli rather than the material microstructure. A typical case in which the working hypotheses of microrheology are
violated is represented by a probe particle that moves in the pores of a porous material. Also, it must be pointed out that
the inversion procedure that starting from the compliance (or mean-square displacement) obtains the complex modulus
is mathematically subtle, and may lead to a variety of known artifacts if not executed properly (see e.g. [24,27]).
Passive microrheology can be implemented by using different experimental techniques allowing for the determination
of the mean-square displacement of one or several probe particles embedded in a material with unknown rheological
properties. This is commonly done by optical means, for example by using video microscopy to track the position of the
particles in real space and calculating the mean-square displacement from the particle trajectories [29]. Particle tracking
(PT) is a very powerful approach that allows to go well beyond the mean-square displacement and to isolate effectively
anomalous and heterogeneous particle dynamics [30]. Alternatively, one can obtain the mean-square displacement from
the inversion of dynamic light scattering (DLS) or diffusing wave spectroscopy data (DWS) [24]. DLS is a very popular
optical scattering technique that relies on the fact that the motion of the particles causes fluctuations of the intensity
of the scattered light at a fixed angle in the far-field of the sample; these fluctuations can be statistically quantified
via an intensity autocorrelation function that, under suitable hypotheses, can be inverted to give the particle mean-
square displacement [31]. DLS is intrinsically a multi-scale technique, as changing the scattering angle corresponds to
changing the scattering wave-vector q at which the particle dynamics is probed. In contrast to DLS, DWS relies on
strongly multiply scattered light, which lost any directional information while traveling through the sample; due to the
large number of scattering events that light undergoes before being detected, DWS is sensitive to very small changes in
the sample configuration and can provide access to very small particle displacements [32]. More recently, differential
dynamic microscopy (DDM) [33] was also used to perform passive microrheology experiments [28,34,35]. Similar to
particle tracking, DDM is based on videos obtained with a microscope; however, DDM first highlights the motion of the
particles by taking the difference between two images acquired at different times. Once the difference image is available,
its spatial Fourier transform is calculated and the procedure is repeated for pairs of images that are separated by a different
time delay; this analysis provides q-resolved results that are very similar to those obtained with DLS.

3. Non-linear oscillatory tests and shear induced dynamics

3.1. Non-linear oscillatory tests

The non-linear rheological behavior of soft materials (and in particular their yielding behavior) can be probed with a
variety of rheological tests [19]. In this manuscript we restrict our discussion to non-linear oscillatory strain experiments,
where the typical test is the strain-sweep (amplitude sweep). In this test the frequency is kept fixed ω = ω0 , and the storage
and loss moduli G′ (ω0 ) and G"(ω0 ) are measured as a function of the strain amplitude. A typical result is schematically
shown in Fig. 7(a). In the low amplitude oscillatory shear (LAOS) regime, the moduli are independent of the amplitude
and the stress remains sinusoidal (Fig. 7((b)). In this regime, oscillatory strain measurements probe the structure and the
interactions of soft materials without perturbing them. However, upon increasing the amplitude to the large amplitude
8
C. Roberto and T. Véronique Physica A 631 (2023) 128653

Fig. 7. (a) Schematic illustration of a strain sweep at fixed frequency. This test can be used to determine the transition from the linear to the
nonlinear viscoelastic regime, i.e. the transition from the small amplitude oscillatory shear (SAOS) to the large amplitude oscillatory shear (LAOS)
regime. In the SAOS regime, the storage (G’ — red continuous line) and loss (G’’ — black dashed line) moduli are independent of the applied strain
amplitude and the stress response is a sinusoidal wave (panel b). In the LAOS regime, the storage and loss moduli become a function of the strain
amplitude and the stress response is distorted (panel c). Correspondingly, the Lissajous plot, representing stress as a function of strain, changes from
an ellipse in the SAOS regime (panel d) to a distorted ellipse in the LAOS regime (panel e).

oscillatory shear (LAOS) regime, the moduli become amplitude dependent and the stress response to the sinusoidal strain
becomes distorted (Fig. 7((c)). Because of the emergence of higher-harmonics in the stress response, the mechanical
moduli are not precisely defined anymore. However, it is often acceptable to use the moduli extracted from the first-
harmonic component of the stress as measures of energy storage and dissipation for the description of the viscoelastic
properties of the system [36].
Most arrested soft matter materials, such as dense emulsions and microgels, foams, colloidal glasses and self-assembled
polymer gels, display a behavior similar to that sketched in Fig. 7(a), termed type-III behavior [36]. The loss modulus
exhibits an overshoot and becomes the dominant modulus at some critical strain. Beyond this critical strain both moduli
decrease with increasing strain, albeit following different power-laws, the ratio G′′ /G′ being an increasing function of
strain.
These characteristics bear a strong connection to plasticity and yielding [20]. Several system-specific models have
been proposed to explain such behavior; they include strain dependent effects such as structural changes in the material,
changes in the interaction potential, forced strain relaxation, and others [36]. However, the generic type-III features seem
to be robust to changing sample characteristics, such as the compressibility of the constituent or the interaction potential.
9
C. Roberto and T. Véronique Physica A 631 (2023) 128653

Fig. 8. Strain dependence of G′ and G′′ for (a) polysaccharidede gel formed by the association of 4w t% Xanthan gum in water and (b) a dense
packing of micelles formed by 20w t% PEO–PPO–PEO triblock copolymers in water (PEO-polyethylene oxide; PPO-polypropylene oxide). Both aqueous
solutions display LAOS type III behavior. However, the shapes of the nonlinear stress waveforms are markedly different. The oscillating stress data
and the Lissajous patterns are shown in (c) for the Xanthan gum system and in (d) for the PEO–PPO–PEO triblock copolymer system. Reproduced
by permission of Hyun et al. [36].

This points to an insensitivity of the first-harmonic of the stress to details in the origin of yielding, which could be encoded
in the higher-harmonics [36]. For this purpose, closed loop graphical representations of the stress versus strain (Lissajous
curves) are often used. In this representation, the response of an ideal viscous fluid to an oscillatory stain is a perfect
circle, while it is a diagonal line for the response of a perfect elastic system. Examples of Lissajous curves of the response
of a viscoelastic system in the SAOS and LAOS regime are shown in Fig. 7(d) and 7(e), respectively.
Indeed, investigations of systems, exhibiting similar type-III characteristics in the first harmonics, have uncovered
distinct behavior in the higher order harmonics. Such example is shown in Fig. 8, displaying the results obtained for
a gel formed by the association of polysaccharides (panel a) and for a dense packing of micelles formed by a triblock
copolymer (panel b). While considering the analysis of the first harmonics the yielding characteristics of the two systems
seem indistinguishable, the higher harmonics reveal distinct yielding pathways. A detailed discussion of these differences
and peculiarities goes beyond the aim of this introduction, it is however instructive to make some brief comments: for
increasing values of the applied strain γ0 , both the temporal stress dependence and the Lissajous plot for the Xanthan
gel show a clear stress overshoot at the beginning and upon reversal of the strain cycle. This overshoot is reminiscent of
that observed in shear start-up experiments, where a constant strain rate is imposed to an initially unperturbed sample.
It suggests that Xanthan gel (Fig. 8, a) yields as a brittle solid, whereas the micelles (Fig. 8, b) exhibit a smoother behavior
that is more typical of ductile yielding [37].

3.2. Shear induced dynamics

To fully establish the microscopic origin of yielding in oscillatory strain tests, investigations of the structure and in
particular the dynamics under shear are essential. By dynamics, we here generally understand the displacements of the
system’s constituents that are not trivially related to the strain-induced ‘affine’ displacements. As sketched in Fig. 9, the
non-trivial ‘non-affine’ displacement can be obtained by subtracting off the ‘affine’ contribution to the real displacement.
In oscillatory shear experiments, we have a convenient alternative to obtain the non-affine contributions. Indeed, we can
exploit the periodic return of the strain to the same value to assess the irreversible non-affine displacements without any
need of accounting for the affine displacements that contribute only when the strain is distinct.
This trick is exploited in echo-techniques, where a microscopic configuration obtained at time t is compared to that
obtained for t + nT , with T the period of the oscillatory strain and n an integer. In the absence of any rearrangements, the
structural configuration is identical after each T ; any deviation from this case can thus be directly connected to particle
rearrangements. The first echo-experiment has been introduced by Hebraud et al. [38], who used echo DWS to explore
the dynamics of dense emulsions subjected to oscillatory shear. Using this technique, the authors uncovered that yielding
only involves the reconfiguration of a small fraction of the sample.
10
C. Roberto and T. Véronique Physica A 631 (2023) 128653

Fig. 9. Illustration of non-affine displacement under shear. The non-affine displacement is defined as the difference between the real displacement
and the affine displacement.

Fig. 10. (a–b) Data obtained for an emulsion with volume fraction 0.88 [39]. While the amplitude sweep denotes a rather smooth transition from
the SAOS to the LAOS regime (a), the displacement as measured in echo particle tracking displays a sharp increase at the onset to non-linearity
(b). (c–d) Data obtained for a dense packing of Carbopol microgels seeded with tracer particles [40]. Amplitude sweep (c) exhibits features similar
to that obtained for the emulsion. Integrated dynamic activity maps (d) reveal dramatic differences in the shear induced dynamics when strain
amplitudes of respectively 20% and 80% are applied. The scale bar in (d) corresponds to 35 mm, and the horizontal (vertical) axis represents the
shear (vorticity) direction. In (a) and (c) G′ is represented with red squares and G" with black triangles.

Since then, the use of echoes has been extended to dynamic light scattering and particle tracking microscopy
techniques [39–46], leading to a better understanding of the transition from the SAOS to the LAOS regime. As an example,
we show in Fig. 10(a) and (b) results obtained by Knowlton et al. [39], who studied yielding of concentrated emulsions
using echo-PT. These emulsions exhibit the rheological signature typical of type-III materials, as shown in Fig. 10(a). At
the onset of non-linearity, the irreversible non-affine displacements are found to display a very sharp increase (Fig. 10(b)),
denoting that yielding occurs at strains where G’ is still the dominant modulus.
More recently, Edera et al. demonstrated that echo-DDM may be particularly suitable for the study of the irreversible
non-affine displacements [40]. They performed experiments with a dense packing of Carbopol microgels exhibiting type-
III behavior (Fig. 10(c)), and showed that shear induced dynamics display characteristics of diffusion, with diffusion
11
C. Roberto and T. Véronique Physica A 631 (2023) 128653

coefficients increasing by three orders of magnitudes across the yielding point. In addition, using dynamic activity maps
they show that the portion of the system involved in plastic irreversible rearrangements strongly increases at the yield
point (Fig. 10(d)): for strains below yielding (20%), the rearrangements are localized and occur intermittently, whereas for
large enough strain amplitudes (80%) the rearrangements are more homogeneously distributed both in space and time.
However, the remarkable increase of irreversible particle rearrangements at a given strain contrasts with the rather
smooth transition observed in rheology, which indicates that yielding involves more than only irreversible plastic
rearrangements. Indeed, recent experiments probing rearrangements during the oscillation cycle denoted that yielding
involves both irreversible and reversible rearrangements [47]. Equivalently, experiments probing the recoverable strain
at different moments during the cycle showed that the loss modulus at the yielding transition contains both reversible
(solid) and irreversible (fluid) components [20].
Clearly, to fully elucidate yielding of amorphous soft matter systems, more work combining measurements of the
macroscopic rheology and microscopic dynamics are needed and, we believe, will likely appear in the next future.

4. Concluding remarks

Understanding the mechanisms of mechanical failure is of fundamental importance to material sciences. In this respect,
studying the microscopic origin of the yielding transition in soft matter systems is not only of interest to the field of soft
matter itself. Indeed, soft matter systems can be considered as convenient model systems to investigate the more general
characteristics of mechanical failure. The main reason lies in the fundamental building blocks of soft matter having a
characteristic mesoscopic lengthscale. This not only enables the study of their structure with optical means, but also
makes them slow enough to easily capture their dynamics and, of course, soft enough to conveniently measure their
mechanical properties without the need of applying exceedingly large forces. In this context, rheo-optical investigations,
in which the sample is optically monitored and characterized during the controlled application of forces and deformations,
are acquiring increasing relevance. They lead to an improved understanding of the relation between the microscopic
structural and dynamical features of a sample and its macroscopic rheological characteristics, in particular across the
yielding transition. In this paper, we focused mostly on the latter, by giving a brief and self-contained introduction to
the wide area of the main rheo-optical tools that can be used. We chose to give particular emphasis to oscillatory shear
tests, as they give direct access to non-affine displacements. However, schemes to assess non-affine displacements in the
presence of affine displacements exist and are quickly developing [48–50], such that the microscopic origin of yielding
and fracture can also be assessed using other tests. It is thus likely that significant progress in understanding mechanical
failure, as well as progress in developing concepts to prevent such failure, will be made in a near future, and as we hope
by some of the readers of these notes.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have
appeared to influence the work reported in this paper.

Data availability

The authors do not have permission to share data.

Acknowledgments

Open Access funding provided by University of Vienna. The authors thank Paolo Edera, Fabio Giavazzi, Nikos Kalafatakis,
George Petekidis for useful discussions. VT acknowledges support from the Swiss National Science Foundation (Grant No.
200021_197340).

References

[1] M. Eberhart, Why Things Break: Understanding the World by the Way It Comes Apart, Crown, 2007.
[2] J.-C. Tolédano, Physical Basis of Plasticity in Solids, World Scientific, 2011.
[3] M. Miguel, A. Vespignani, S. Zapperi, J. Weiss, J.-R. Grasso, et al., Intermittent dislocation flow in viscoplastic deformation, Nature 410 (6829)
(2001) 667–671.
[4] D.M. Dimiduk, C. Woodward, R. LeSar, M.D. Uchic, Scale-free intermittent flow in crystal plasticity, Science 312 (5777) (2006) 1188–1190.
[5] A. Pertsinidis, X.S. Ling, Video microscopy and micromechanics studies of one-and two-dimensional colloidal crystals, New J. Phys. 7 (1) (2005)
33.
[6] G.L. Hunter, E.R. Weeks, The physics of the colloidal glass transition, Rep. Progr. Phys. 75 (6) (2012) 066501.
[7] D.T. Chen, Q. Wen, P.A. Janmey, J.C. Crocker, A.G. Yodh, Rheology of soft materials, Annu. Rev. Condens. Matter Phys. 1 (1) (2010) 301–322.
[8] D. Bonn, M.M. Denn, L. Berthier, T. Divoux, S. Manneville, Yield stress materials in soft condensed matter, Rev. Modern Phys. 89 (3) (2017)
035005.
[9] A. Nicolas, E.E. Ferrero, K. Martens, J.-L. Barrat, Deformation and flow of amorphous solids: Insights from elastoplastic models, Rev. Modern
Phys. 90 (4) (2018) 045006.
[10] I. Cantat, S. Cohen-Addad, F. Elias, F. Graner, R. Höhler, O. Pitois, F. Rouyer, A. Saint-Jalmes, Foams: Structure and Dynamics, OUP Oxford, 2013.

12
C. Roberto and T. Véronique Physica A 631 (2023) 128653

[11] M.L. Falk, J.S. Langer, Dynamics of viscoplastic deformation in amorphous solids, Phys. Rev. E 57 (6) (1998) 7192.
[12] M.L. Falk, J.S. Langer, Deformation and failure of amorphous, solidlike materials, Annu. Rev. Condens. Matter Phys. 2 (1) (2011) 353–373.
[13] T. Voigtmann, Nonlinear glassy rheology, Curr. Opin. Colloid Interface Sci. 19 (6) (2014) 549–560.
[14] B. Sun, H. Yu, W. Jiao, H. Bai, D. Zhao, W. Wang, Plasticity of ductile metallic glasses: A self-organized critical state, Phys. Rev. Lett. 105 (3)
(2010) 035501.
[15] P. Schall, D.A. Weitz, F. Spaepen, Structural rearrangements that govern flow in colloidal glasses, Science 318 (5858) (2007) 1895–1899.
[16] V. Chikkadi, G. Wegdam, D. Bonn, B. Nienhuis, P. Schall, Long-range strain correlations in sheared colloidal glasses, Phys. Rev. Lett. 107 (19)
(2011) 198303.
[17] K. Jensen, D.A. Weitz, F. Spaepen, Local shear transformations in deformed and quiescent hard-sphere colloidal glasses, Phys. Rev. E 90 (4)
(2014) 042305.
[18] A. Ghosh, Z. Budrikis, V. Chikkadi, A.L. Sellerio, S. Zapperi, P. Schall, Direct observation of percolation in the yielding transition of colloidal
glasses, Phys. Rev. Lett. 118 (14) (2017) 148001.
[19] C.W. Macosko, Rheology: Principles, Measurements, and Applications, VCH, 1994.
[20] G.J. Donley, P.K. Singh, A. Shetty, S.A. Rogers, Elucidating the G ’’ overshoot in soft materials with a yield transition via a time-resolved
experimental strain decomposition, Proc. Natl. Acad. Sci. 117 (36) (2020) 21945–21952.
[21] T. Mason, J. Bibette, D. Weitz, Elasticity of compressed emulsions, Phys. Rev. Lett. 75 (10) (1995) 2051.
[22] A.K. Gurnon, C. Lopez-Barron, M.J. Wasbrough, L. Porcar, N.J. Wagner, Spatially resolved concentration and segmental flow alignment in a
shear-banding solution of polymer-like micelles, ACS Macro Lett. 3 (3) (2014) 276–280.
[23] S.A. Rogers, M.A. Calabrese, N.J. Wagner, Rheology of branched wormlike micelles, Curr. Opin. Colloid Interface Sci. 19 (6) (2014) 530–535.
[24] E.M. Furst, T.M. Squires, Microrheology, Oxford University Press, 2017.
[25] M. Doi, Soft Matter Physics, Oxford University Press, 2013.
[26] T.G. Mason, D.A. Weitz, Optical measurements of frequency-dependent linear viscoelastic moduli of complex fluids, Phys. Rev. Lett. 74 (7)
(1995) 1250.
[27] M.G. Smith, G.M. Gibson, M. Tassieri, i-RheoFT: Fourier transforming sampled functions without artefacts, Sci. Rep. 11 (1) (2021) 1–12.
[28] P. Edera, D. Bergamini, V. Trappe, F. Giavazzi, R. Cerbino, Differential dynamic microscopy microrheology of soft materials: A tracking-free
determination of the frequency-dependent loss and storage moduli, Phys. Rev. Mater. 1 (7) (2017) 073804.
[29] J.C. Crocker, D.G. Grier, Methods of digital video microscopy for colloidal studies, J. Colloid Interface Sci. 179 (1) (1996) 298–310.
[30] K.A. Rose, M. Molaei, M.J. Boyle, D. Lee, J.C. Crocker, R.J. Composto, Particle tracking of nanoparticles in soft matter, J. Appl. Phys. 127 (19)
(2020) 191101.
[31] B.J. Berne, R. Pecora, Dynamic Light Scattering: With Applications to Chemistry, Biology, and Physics, Courier Corporation, 2000.
[32] D.J. Pine, D.A. Weitz, P.M. Chaikin, E. Herbolzheimer, Diffusing wave spectroscopy, Phys. Rev. Lett. 60 (12) (1988) 1134.
[33] R. Cerbino, V. Trappe, Differential dynamic microscopy: Probing wave vector dependent dynamics with a microscope, Phys. Rev. Lett. 100 (18)
(2008) 188102.
[34] A.V. Bayles, T.M. Squires, M.E. Helgeson, Probe microrheology without particle tracking by differential dynamic microscopy, Rheol. Acta 56 (11)
(2017) 863–869.
[35] M. Escobedo-Sánchez, J. Segovia-Gutiérrez, A. Zuccolotto-Bernez, J. Hansen, C. Marciniak, K. Sachowsky, F. Platten, S. Egelhaaf, Microliter
viscometry using a bright-field microscope: η-DDM, Soft Matter 14 (34) (2018) 7016–7025.
[36] K. Hyun, M. Wilhelm, C.O. Klein, K.S. Cho, J.G. Nam, K.H. Ahn, S.J. Lee, R.H. Ewoldt, G.H. McKinley, A review of nonlinear oscillatory shear tests:
Analysis and application of large amplitude oscillatory shear (LAOS), Prog. Polym. Sci. 36 (12) (2011) 1697–1753.
[37] H.J. Barlow, J.O. Cochran, S.M. Fielding, Ductile and brittle yielding in thermal and athermal amorphous materials, Phys. Rev. Lett. 125 (16)
(2020) 168003.
[38] P. Hébraud, F. Lequeux, J. Munch, D. Pine, Yielding and rearrangements in disordered emulsions, Phys. Rev. Lett. 78 (24) (1997) 4657.
[39] E.D. Knowlton, D.J. Pine, L. Cipelletti, A microscopic view of the yielding transition in concentrated emulsions, Soft Matter 10 (36) (2014)
6931–6940.
[40] P. Edera, M. Brizioli, G. Zanchetta, G. Petekidis, F. Giavazzi, R. Cerbino, Deformation profiles and microscopic dynamics of complex fluids during
oscillatory shear experiments, Soft Matter 17 (37) (2021) 8553–8566.
[41] G. Petekidis, A. Moussaid, P. Pusey, Rearrangements in hard-sphere glasses under oscillatory shear strain, Phys. Rev. E 66 (5) (2002) 051402.
[42] N.C. Keim, P.E. Arratia, Yielding and microstructure in a 2D jammed material under shear deformation, Soft Matter 9 (27) (2013) 6222–6225.
[43] M. Laurati, S. Egelhaaf, G. Petekidis, Plastic rearrangements in colloidal gels investigated by LAOS and LS-echo, J. Rheol. 58 (5) (2014) 1395–1417.
[44] M.C. Rogers, K. Chen, L. Andrzejewski, S. Narayanan, S. Ramakrishnan, R.L. Leheny, J.L. Harden, Echoes in X-ray speckles track nanometer-scale
plastic events in colloidal gels under shear, Phys. Rev. E 90 (6) (2014) 062310.
[45] M.C. Rogers, K. Chen, M.J. Pagenkopp, T.G. Mason, S. Narayanan, J.L. Harden, R.L. Leheny, Microscopic signatures of yielding in concentrated
nanoemulsions under large-amplitude oscillatory shear, Phys. Rev. Mater. 2 (9) (2018) 095601.
[46] V.V. Vasisht, S.K. Dutta, E. Del Gado, D.L. Blair, Rate dependence of elementary rearrangements and spatiotemporal correlations in the 3D flow
of soft solids, Phys. Rev. Lett. 120 (1) (2018) 018001.
[47] K.L. Galloway, X. Ma, N.C. Keim, D.J. Jerolmack, A.G. Yodh, P.E. Arratia, Scaling of relaxation and excess entropy in plastically deformed amorphous
solids, Proc. Natl. Acad. Sci. 117 (22) (2020) 11887–11893.
[48] S. Aime, L. Cipelletti, Probing shear-induced rearrangements in Fourier space. I. Dynamic light scattering, Soft Matter 15 (2) (2019) 200–212.
[49] S. Aime, L. Cipelletti, Probing shear-induced rearrangements in Fourier space. II. Differential dynamic microscopy, Soft Matter 15 (2) (2019)
213–226.
[50] J.A. Richards, V.A. Martinez, J. Arlt, Characterising shear-induced dynamics in flowing complex fluids using differential dynamic microscopy,
Soft Matter 17 (39) (2021) 8838–8849.

13