Scaling description of the yielding transition in soft amorphous solids at zero temperature
Yield stress materials flow if a sufficiently large shear stress is applied. Although such materials are ubiquitous and relevant for industry, there is no accepted microscopic description of how they yield, even in the simplest situations where temperature is negligible and where flow inhomogeneities such as shear bands or fractures are absent. Here we propose a scaling description of the yielding transition in amorphous solids made of soft particles at zero temperature. Our description makes a connection between the Herschel-Bulkley exponent characterizing the singularity of the flow curve near the yield stress , the extension and duration of the avalanches of plasticity observed at threshold, and the density of soft spots, or shear transformation zones, as a function of the stress increment beyond which they yield. We argue that the critical exponents of the yielding transition can be expressed in terms of three independent exponents , and , characterizing respectively the density of soft spots, the fractal dimension of the avalanches, and their duration. Our description shares some similarity with the depinning transition that occurs when an elastic manifold is driven through a random potential, but also presents some striking differences. We test our arguments in an elasto-plastic model, an automaton model similar to those used in depinning, but with a different interaction kernel, and find satisfying agreement with our predictions both in two and three dimensions.
Submitted to Proceedings of the National Academy of Sciences of the United States of America www.pnas.org/cgi/doi/10.1073/pnas.0709640104 \copyrightyear2008 \issuedateIssue Date \volumeVolume \issuenumberIssue Number
Submitted to Proceedings of the National Academy of Sciences of the United States of America
Yield stress solids flow if a sufficiently large shear stress is applied. Although such materials are ubiquitous and relevant for industry, there is no accepted microscopic description of how they yield. Here we propose a scaling description of the yielding transition which relates the flow curve, the statistics of the avalanches of plasticity observed at threshold, and the density of local zones that are about to yield. Our description shares some similarity with the depinning transition that occurs when an elastic manifold is driven through a random potential, but also presents some striking differences. Numerical simulations on a simple elasto-plastic model find good agreement with our predictions.
Many solids will flow and behave as fluids if a sufficiently large shear stress is applied. In crystals, plasticity is governed by the motion of dislocations [1, 2]. In amorphous solids there is no order, and conserved defects cannot be defined. However, as noticed by Argon , plasticity consists of elementary events localized in space, called shear transformations, where a few particles rearrange. This observation supports that there are special locations in the sample, called shear transformation zones or STZ’s , where the system lies close to an elastic instability. Several theoretical approaches of plasticity, such as STZ theory  or Soft Glassy Rheology (SGR)  assume that such zones relax independently, or are coupled to each other via an effective temperature. However, at zero temperature and small applied strain rate , computer experiments [6, 7, 8, 9, 10, 11] and very recent experiments [12, 13] indicate that local rearrangements are not independent: plasticity occurs via avalanches in which many shear transformations are involved, forming elongated structures where plasticity localizes. If conditions are such that flow is homogeneous (as can occur for example in foams or emulsions), one finds that the flow curves are singular at small strain rate and follow a Herschel-Bulkley law [14, 15]. These features are qualitatively reproduced by elastoplastic models [16, 17, 18, 19, 20] where space is discretized. In these models, a site that yields plastically affects the stress in its surroundings via some interaction kernel , argued to decay as a power-law of distance and to display a four-fold symmetry , as supported by observations [22, 23, 24]. This perturbation can trigger novel plastic events and lead to avalanches. However, even within this picture, the relationship between the avalanche dynamics and the singularity of the flow curves remains debated [7, 25].
It is tempting to seek progress by building a comparison between the yielding transition and the much better understood depinning transition that occurs when an elastic interface of dimension is driven in a random environment [2, 26]. The role that transverse displacements play in depinning corresponds to the local accumulated plastic strain and the total plastic strain can be identified with the center of mass of the interface, as illustrated in Fig.(1). Both phenomena display very similar properties: near the depinning threshold force , the velocity vanishes non-analytically and the interplay between disorder and elasticity at threshold leads to broadly distributed avalanches corresponding to jerky motions of the interface. Much more is known about the depinning transition: it is a dynamical critical point characterized by two independent exponents related to avalanche extension and duration [26, 65]. These exponents have been computed perturbatively with the functional renormalisation group [28, 29, 30] and evaluated numerically with high precision .The comparison between these two phenomena has led to the proposition that the yielding transition is in the universality class of mean-field depinning [32, 33]. However, experiments find a reological exponent against the predicted for elastic depinning, and numerical simulations display intriguing finite size effects that differ from depinning [9, 10, 34, 35, 36].
Formally, elastoplastic models are very similar to automaton models known to capture well the depinning transition , the key difference relies in the interaction kernel , long-ranged and of variable sign for elastoplastic models while is essentially a laplacian for depinning with short-range elasticity. We have recentely shown  that in presence of long-ranged with variable sign interactions the distribution of shear transformations at a distance from instability, , is singular with , unlike in depinning for which . As we shall recall this singularity naturally explains the finite size effects observed in simulations. In this letter we argue that once this key difference with depinning is taken into account, the analogy between these two phenomena is fruitful, and leads to a complete scaling description of the yielding transition. In particular we find that the Herschel-Bulkley exponent is related to avalanche extension and duration via Eq.(9) and that the avalanche statistics can be expressed in term of three independent exponents: , and , characterizing respectively the density of shear transformations, the fractal dimension of the avalanches, and their duration.
3 Definition of exponents
Several studies (see Table 2) have characterized the yielding transition with several exponents, which we now recall.
Flow curves: Rheological properties are singular near the yielding transition. Herschel and Bulkley noticed  that for many yield stress materials, , where is the macroscopic strain rate and is the external shear stress. By analogy with depinning we instead introduce the exponent , such that \be ˙γ ∼(Σ-Σ_c)^β. \eeIn contrast to depinning, one finds in the yielding transition, as we explain below. Our analysis below focuses on the regime ; effects not discussed here are expected to affect the flow curves at larger stresses[38, 39].
Length scales: near the yielding transition the dynamics becomes more and more cooperative, and is correlated on a length scale : \be ξ∼—Σ-Σ_c—^-ν \ee
Avalanche statistics: At threshold , the dynamics occur by avalanches whose size we define as , where is the plastic strain increment due to the avalanche, and is the volume of the system. The normalized avalanche distribution follows a power-law: \be ρ(S)∼S^-τ \eeIn a finite system of size , this distribution is cut-off at some value , where the linear extension of the avalanche is of order , enabling to define the fractal dimension : \be S_c∼L^d_f \ee
A key exponent relates length- and time- scales. characterizes the duration of an avalanche whose linear extension is : \be T∼l^z \ee
Density of shear transformations: If an amorphous solid is cut into small blocks containing several particles, one can define how much stress needs to be applied to the block before an instability occurs. The probability distribution is a measure of how many putative shear transformations are present in the sample . Near the depinning transition, a similar quantity can be defined, and in that case it is well known that . We have argued  that it must be so when the interaction kernel is monotonic, i.e. its sign is constant in space. For an elastic interface this is the case, as a region that yields will always destabilize other regions. This implies that locally the distance to instability always decreases with time, until when the block rearranges. Thus nothing in the dynamics allows the block to forecast that an instability approaches, and no depletion nor accumulation is expected to occur near . By contrast for the yielding transition, the sign of varies in space. Thus locally jumps both forward and backward, performing some kind of random walk. Since acts as an absorbing boundary condition (as the site is stabilized by a finite amount once it yields), one expects that depletion can occur near [20, 36, 41]. In  we argued that must indeed vanish at if the interaction is sufficiently long-range (in particular if , as is the case for the yielding transition), otherwise the system would be unstable: a small perturbation at the origin would cause extensive rearrangements in the system. Thus the yielding transition is affected by an additional exponent that does not enter the phenomenology of depinning problem: \be P(x)∼x^θ \eewith . Using elastoplastic models we previously measured for and for , as we confirm here with improved statistics.
4 Scaling relations
We now propose several scaling relations, which essentially mirror arguments made in the context of the depinning transition (see supplementary information (S.I.)), with the additional feature that is singular.
Stationarity: Consider applying a quasi-static strain in a system of linear size , as represented in Fig.(2). The stress fluctuates due to avalanches: an avalanche of size leads to a stress drop proportional to the plastic strain , of average . Using Eqs.(3,3) and assuming one gets , so that the average drop of stress is of order .
In between plastic events, the system loads elastic energy, and stress rises by some typical amount . is limited by the next plastic event, and is thus inversely proportional to the rate at which avalanches are triggered. Although one might think that if the system is twice larger, a plastic event will occur twice sooner (implying ), this is not the case and depends on system size with a non-trivial exponent [9, 10, 34, 35]. As argued in [34, 36], one expects to be of order , the weakest site in the system. If the are independent this implies that . In  it was argued based on local considerations that . Instead our recent work  implies that is governed by the elastic interactions between plastic events, and remains a non-trivial exponent that depends on interaction range and spatial dimension.
Dynamics: A powerful idea in the context of the depinning transition is that avalanches below threshold and flow above threshold are intimately related . Above threshold, the motion of the interface can be thought as consisting of a number of individual avalanches of spatial extension , acting in parallel. We propose the same image for the yielding transition. If so, the strain rate in the sample is simply equal to the characteristic strain rate of an avalanche of size , leading to:
implying our second scaling relation, which to our knowledge was not proposed in this context:
Statistical tilt symmetry: If flow above consists of independent avalanches of size , then the avalanche-induced fluctuations of stress on that lengthscale, , must be of order: \be δΣ∼S_c/ξ^d∼(Σ-Σ_c)^ν(d-d_f) \eeOne expects that the fluctuations of stress on the scale must be of order of the distance to threshold . Eq.(4) then leads to: \be ν=1d-df \eeIt was suggested in  that Eq.(4) may apply at the yielding transition. A similar relation holds for depinning of an interface if the elasticity is assumed to be linear, a non-trivial assumption underlying Eq.(4). In that case it can be derived using the so-called statistical-tilt-symmetry. In S.I., we discuss evidence that linearity applies at the yielding transition, enabling us to use this symmetry to derive Eq.(4).
|exponent||expression||relations||2d measured/prediction||3d measured/prediction|
5 Elasto-plastic model
The phenomenological description proposed above may apply to real materials with inertial or over-damped dynamics, as well as to elasto-plastic models, although the yielding transition in these situations may not lie in the same universality class [35, 43]. In what follows we test our predictions in elasto-plastic models, implemented as in , whose details are recalled here.
We consider square (d=2) and cubic (d=3) lattices of unit lattice size with periodic boundary conditions, where each lattice point can be viewed as the coarse grained description of a group of particles. It is characterized by a scalar stress , a local yield stress , and a strain . The total stress carried by the system is . The elastic strain satisfies . The plastic strain is constant in time except when site becomes plastic, which occurs at a rate if the site is unstable, defined here as . For simplicity we consider that does not vary in space, and use it to define our unit stress . is the only time scale in the problem, and defines our unit of time.
When plasticity occurs, the plastic strain increases locally and the stress is reduced by the same amount . We assume that where is some random number, taken to be uniformly distributed between and . would correspond to imposing zero local stress after a plastic event (a choice that we avoid as it sometimes leads to periodic dynamics). When a site relaxes it affects the stress level on other sites immediately, such that: \beδx_j= -G(→r_ij)δx_i \eewith in an infinite two-dimensional system under simple shear, and where is the angle between the shear direction and . In a finite system depends on the boundary conditions . At fixed stress, by definition and stress conservation implies that the sum of on any line or column of the lattice is zero. At fixed global strain however, one plastic event reduces the stress by . When desired, we model this effect by modifying the interaction kernel as follows: .
In our model the average plastic strain is defined as , and the strain rate simply follows , where is the heaviside function. Above , the system will reach a steady state with a finite . Below or in the vicinity of however, the system can spontaneously stop. When this happens, to generate a new avalanche, we trigger the dynamics by giving very small random kicks to the system (chosen to conserve stress on every line and column) until one site becomes unstable.
This elasto-plastic model is essentially identical to the automaton models introduced in  in the context of the depinning transition, where the role of the plastic strain is played by the transverse displacement of the elastic interface . The only qualitative difference is the form of .
6 Numerical estimation of critical exponents
Flow curves and length scales: We first implement the extremal dynamics protocol: the average stress decreases by after each plastic event during avalanches, and increases again to generate a new active site at the beginning of a new avalanche. The corresponding stress-plastic strain curves shown in Fig.(2) allows us to estimate the critical stress and the correlation length exponent from the fluctuations of at different sizes:
where is the mean stress and the standard deviation at a given size , and are non universal constants. From our data (see S.I.), we obtain , for d=2 and , for d=3. These quantities can also be reliably extracted from finite strain rate measurements, as shown in S.I.
We then compute the flow curve at fixed strain rate. The stress is adjusted in order to keep the fraction of unstable sites fixed. The determination of the exponent is very sensitive to the value of . Using the values obtained from the previous analysis we find for and for , as shown in Fig.(3).
Avalanche statistics: Avalanche statistics can be investigated using extremal dynamics and Fig.(4). As documented in S.I. this method leads for our largest system size to for and for . This measure appears to have large finite size effects however. We find that such effects are diminished if we work instead at constant stress , and consider as . In S.I., we find using this method that for , and for .
Next we evaluate the fractal dimension and the dynamical exponent using extremal dynamics. Here the avalanche cut-off corresponds to avalanche of linear extension , so that for large systems one expects , where is some function and . This collapse is checked in Fig.(4) and leads to for and for (for the collapse we used the values of measured with the constant stress protocol). Error bars are estimated by considering the range of exponents for which the collapse is satisfactory. To measure we record the duration of each avalanche, and compute the duration distribution for different system size. These distributions are cut-off at some , corresponding to the duration of avalanches of spatial extension , so that . As shown in the right panels of Fig.(4) we indeed find a good collapse with , for and , for .
Density of shear transformations: In elastoplastic models it is straightforward to access the local distance to thresholds and to compute its distribution . Here we recall these results with improved statistics. We fix the stress at , and let the system evolve for a long enough time such that . The dynamics occasionally stops; at that point we measure , and average over many realizations. As shown in Fig.(5), we find for , for , where the error bar is from the error estimation of linear fit. Although in experiments is hard to access, the system size dependence of the average increment of stress where no plasticity occurs should be accessible, and follows . In the insets of Fig.(5) is computed via extremal dynamics, leading to slightly smaller exponents for and for , a difference presumably resulting from corrections to scaling.
6.1 Theory vs numerics
|exponent||values(2d)||values(3d)||lattice model||molecular dynamics||experiments|
|1.52||1.38||1.78||2, 2.33, 3(3d)||2.22(3d), 2.78, 2.22(3d)|
|0.72||0.53||0.5, 0.6, 1||0.5, 0.43, 0.33(3d)|
|1.36||1.43||1.34,1.25||1.3 , 1.3(3d)||1.37–1.49(3d), 1.5(3d)|
|1.1||1.5||1.5, 1.5, 1||0.9, 1.1(3d),1, 1.5(3d), 1.6(3d)|
|0.7||0.8||0.75||1, 0.6, 0.8(3d)|
|0.57||0.35||0.54, 0.43(3d), 0.5,0.5(3d)|
7 Comparison with MD and experiments
Although elastoplastic models are well-suited to test theories, they make many simplifications, and thus may not fall in the universality class of real materials. One encouraging item is our estimate of , which is very similar to the value extracted from finite size effects in MD simulations using overdamped dynamics, as reported in Table.2. This is consistent with our finding  that (and ) is independent of the choice of dynamical rules in our model, that can however dramatically affect the dynamics. Concerning the latter, our choice that the interaction is instantaneous in time, while still being long-range, is likely to affect the exponents and . We expect that if a more realistic time-dependent interaction kernel is considered (a costly choice numerically), the exponent will satisfy . According to Eq.(9), this will lead to larger values of , in agreement with experiments.
The scaling relations for and in Table 1 appear to be supported by MD simulations. In  for overdamped dynamics, and for , whereas and for , leading to in both and , which compares well to their measured value . In , all numerics [35, 52] report , leading to as observed in . In , there is some disagreement on the value of : [52, 53] report as we do in our elasto-plastic model, in disagreement with  for which , implying that . It would be useful to resolve this discrepancy, since in the depinning problem, when another length scale enters in the scaling description, which affects in particular finite size effects [44, 54]. In this situation however, we expect our scaling description to be unchanged if is meant to characterize the correlations of the dynamics for .
We have proposed a scaling description of stationary flow in soft amorphous solids, and it is interesting to reflect if this approach can apply to other systems. Plasticity in crystals shares many similarities with that of amorphous solids, and the far-field effect of a moving dislocation is essentially identical to the effect of a shear-transformation . Thus we expect that the stability argument of  on the density of regions about to yield also applies in crystals, leading to a non-trivial exponent in that case too. Our scaling relations may thus hold in crystals, although the formation of structures such as domain walls could strongly affect the yielding transition.
Avalanches of plasticity are seen in granular materials where particles are hard [12, 13]. However, we believe that at least for over-damped systems this behavior is only transient, and that the elasto-plastic description does not apply for such materials under continuous shear. Some of us have argued that in that case, a picture based on geometry applies [55, 56], which also leads to a diverging length scale, but of a different nature .
The scaling relations proposed here do not fix the values of the exponents, in particular that of . To make progress, it is tempting to seek a mean-field description of this problem, that would apply beyond some critical dimension. Current mean-field models in which the interaction is random and does not decay with distance lead to [41, 20]. However, the anisotropy is lost in this view, and the fact that diminishes as increases when anisotropy is considered suggests that a mean-field model that includes anisotropy is needed. Such a model would be valuable to build a hydrodynamic description of flow, that would apply for example for slow flow near walls [58, 59], a problem for which current descriptions do not include the role of anisotropy [60, 25].
Acknowledgements.We thank Eric DeGiuli, Alaa Saade, Le Yan, Gustavo Düring, Bruno Andreotti, Mehdi Omidvar, Stephan Bless, and Magued Iskander for discussions. MW acknowledges support from NYU Poly Seed Fund Grant M8769, NSF CBET Grant 1236378, NSF DMR Grant 1105387, and MRSEC Program of the NSF DMR-0820341 for partial funding.
-  Miguel MC, Vespignani A, Zapperi S, Weiss J, Grasso JR (2001) Intermittent dislocation flow in viscoplastic deformation. Nature, 410:667–671.
-  Zaiser M (2006) Scale invariance in plastic flow of crystalline solids. Advances in Physics, 55(1-2):185–245.
-  Argon AS (1979) Plastic deformation in metallic glasses. Acta Metallurgica, 27(1):47–58.
-  Falk ML, Langer JS (1998) Dynamics of viscoplastic deformation in amorphous solids. Phys Rev E, 57(6):7192-7205.
-  Sollich P (1998) Rheological constitutive equation for a model of soft glassy materials. Phys Rev E, 58(1):738-759.
-  Maloney CE, Robbins MO (2009) Anisotropic Power Law Strain Correlations in Sheablue Amorphous 2D Solids. Phys Rev Lett, 102(22):225502.
-  Lemaitre A and Caroli C (2009) Rate-Dependent Avalanche Size in Athermally Sheablue Amorphous Solids. Phys Rev E 103(6):065501.
-  Karmakar S, Lerner E, Procaccia I, Zylberg J (2010) Statistical physics of elastoplastic steady states in amorphous solids: Finite temperatures and strain rates. Phys Rev E, 82(3):031301.
-  Salerno KM, Maloney CE, Robbins MO (2012) Avalanches in strained amorphous solids: Does inertia destroy critical behavior? Phys Rev Lett, 109(10):105703.
-  Maloney C, Lemaitre A (2004) Subextensive scaling in the athermal, quasistatic limit of amorphous matter in plastic shear flow. Phys Rev Lett, 93(1):016001.
-  Gimbert F, Amitrano D, Weiss J (2013) Crossover from quasi-static to dense flow regime in compressed frictional granular media. Europhys Lett, 104(4):46001.
-  Amon A, Nguyen V, Bruand A, Crassous J, Clément E(2012)Hot spots in an athermal system. Phys Rev Lett, 108(13):135502.
-  Antoine LB, Axelle A, Sean M, and Crassous J(2014) Emergence of cooperativity in plasticity of soft glassy materials. Phys Rev Lett, 112(24):246001.
-  Höhler R, Cohen-Addad S (2005) Rheology of liquid foam. Journal of Physics: Condensed Matter, 17(41):R1041.
-  Roberts GR, Barnes HA (2001) New measurements of the flow-curves for carbopol dispersions without slip artefacts. Rheologica Acta, 40(5):499–503.
-  Talamali M, Petäjä V, Vandembroucq D, Roux S (2011) Avalanches, precursors, and finite-size fluctuations in a mesoscopic model of amorphous plasticity. Phys Rev E, 84(1):016115.
-  Baret JC, Vandembroucq D, Roux S (2002) Extremal model for amorphous media plasticity. Phys Rev Lett, 89(19):196506.
-  Martens K, Bocquet L, Barrat JL (2011) Connecting diffusion and dynamical heterogeneities in actively deformed amorphous systems. Phys Rev Lett, 106(15):156001
-  Picard G, Ajdari A, Lequeux F, Bocquet L (2005) Slow flows of yield stress fluids: Complex spatiotemporal behavior within a simple elastoplastic model. Phys Rev E, 71(1):010501.
-  Lin J, Saade A, Lerner E, Rosso A, Wyart M (2014) On the density of shear transformations in amorphous solids. Europhys Lett, 105(2):26003.
-  Picard G, Ajdari A, Lequeux F, Bocquet L (2004) Elastic consequences of a single plastic event: A step towards the microscopic modeling of the flow of yield stress fluids. The European Physical Journal E, 15(4):371–381.
-  Maloney CE, Lemaitre A (2006) Amorphous systems in athermal, quasistatic shear. Phys Rev E, 74(1):016118.
-  Chattoraj J, Lemaitre A (2013) Elastic signature of flow events in supercooled liquids under shear. Phys Rev Lett, 111(6):066001.
-  Schall P, Weitz D, Spaepen F (2007) Structural rearrangements that govern flow in colloidal glasses. Science, 318(5858):1895–1899.
-  Bocquet L, Colin A, Ajdari A (2009) Kinetic theory of plastic flow in soft glassy materials. Phys Rev Lett, 103(3):036001.
-  Fisher D (1998) Collective transport in random media: from superconductors to earthquakes. Physics Reports, 301(1-3):113–150.
-  Kardar M (1998) Nonequilibrium dynamics of interfaces and lines. Physics reports, 301(1-3):85–112.
-  Nattermann T, Stepanow S, Tang L, Leschhorn H (1992) Dynamics of interface depinning in a disordeblue medium. Journal de Physique II, 2(8):1483–1488.
-  Narayan O, Fisher D (1993) Threshold critical dynamics of driven interfaces in random media. Phys Rev B, 48(10):7030.
-  Chauve P, Le Doussal P, Wiese KJ (2001) Renormalization of pinned elastic systems: how does it work beyond one loop? Phys Rev Lett, 86(9):1785.
-  Ferrero EE, Bustingorry S, Kolton AB (2013) Nonsteady relaxation and critical exponents at the depinning transition. Phys Rev E, 87(3):032122.
-  Tsekenis G, Uhl JT, Goldenfeld N, Dahmen KA (2013) Determination of the universality class of crystal plasticity. Europhys Lett, 101(3):36003.
-  Dahmen KA, Ben-Zion Y, Uhl JT (2011) A simple analytic theory for the statistics of avalanches in sheablue granular materials. Nature Physics, 7(7):554-557.
-  Karmakar S, Lerner E, Procaccia I (2010) Statistical physics of the yielding transition in amorphous solids. Phys. Rev. E, 82(5):055103.
-  Salerno KM, Robbins M (2013) Effect of inertia on sheablue disordeblue solids: Critical scaling of avalanches in two and three dimensions. Phys Rev E, 88(6):062206.
-  Lemaître A, Caroli C (2007) Plastic response of a 2d amorphous solid to quasi-static shear: II-dynamical noise and avalanches in a mean field model. arXiv preprint arXiv:0705.3122.
-  Herschel WH, Bulkley R (1926) Consistency measurements of rubber benzene solutions. Koll. Zeit, 39:291.
-  Bonnecaze RT, Cloitre M (2010) Micromechanics of soft particle glasses. Adv Polym Sci, 236:117-161.
-  Olsson P, Teitel S (2007) Critical Scaling of Shear Viscosity at the Jamming Transition. Phys Rev Lett, 99(17):178001.
-  Rosso A, Hartmann AK, Krauth W (2003) Depinning of elastic manifolds. Phys Rev E, 67(2):021602.
-  Hébraud P, Lequeux F (1998) Mode-coupling theory for the pasty rheology of soft glassy materials. Phys Rev Lett, 81(14):2934–2937.
-  Aragón LE, Jagla EA, Rosso A (2012) Seismic cycles, size of the largest events, and the avalanche size distribution in a model of seismicity. Phys Rev E, 85(4):046112.
-  Nicolas A, Martens K, Bocquet L, Barrat JL (2014) Universal and non-universal features in coarse-grained models of flow in disordeblue solids. soft matter 10(26):4648-4661.
-  Narayan O, Middleton AA (1994) Avalanches and the renormalization group for pinned charge-density waves. Phys Rev B, 49(1):244-256.
-  Chaudhuri P, Berthier L, Bocquet L (2012) Inhomogeneous shear flows in soft jammed materials with tunable attractive forces. Phys Rev E, 85(2):021503.
-  Cloitre M, Borrega R, Monti F, Leibler L (2003) Glassy dynamics and flow properties of soft colloidal pastes. Phys Rev Lett, 90(6):068303.
-  Möbius ME, Katgert G, van Hecke M (2010) Relaxation and flow in linearly sheablue two-dimensional foams. Europhys Lett, 90(4):44003.
-  Bécu L, Manneville S, Colin A (2006) Yielding and flow in adhesive and nonadhesive concentrated emulsions. Phys Rev Lett, 96(13):138302.
-  Martens K, Bocquet L, Barrat JL (2011) Connecting diffusion and dynamical heterogeneities in actively deformed amorphous systems. Phys Rev Lett, 106:156001.
-  Budrikis Z, Zapperi S (2013) Avalanche localization and crossover scaling in amorphous plasticity. Phys Rev E, 88(6):062403.
-  Sun B, Yu H, Jiao W, Bai H, Zhao D, Wang W (2010) Plasticity of ductile metallic glasses: A self-organized critical state. Phys Rev Lett, 105(3):035501.
-  Arévalo R, Ciamarra MP (2014) Size and density avalanche scaling near jamming. Soft Matter, 10(16), 2728-2732
-  Bailey NP, Schiotz J, Lemaître A, Jacobsen KW (2007) Avalanche size scaling in sheared three-dimensional amorphous solid. Phys Rev Lett, 98(9):095501.
-  Myers CR, Sethna JP (1993) Collective dynamics in a model of sliding charge-density waves. ii.finite-size effects. Phys Rev B, 47(17):11194-11203.
-  Lerner E, Düring G, Wyart M (2012) Toward a microscopic description of flow near the jamming threshold. Europhys Lett, 99(5):58003.
-  Lerner E, Düring G, Wyart M (2012) A unified framework for non-brownian suspension flows and soft amorphous solids. Proceedings of the National Academy of Sciences,109(13):4798–4803.
-  Düring G, Lerner E, Wyart M (2014) Length scales and self-organization in dense suspension flows. Phys Rev E, 89(2):022305.
-  Forterre Y, Pouliquen O (2008) Flows of dense granular media. Annual Review of Fluid Mechanics, 40:1–24.
-  Goyon J, Colin A, Ovarlez G, Ajdari A, Bocquet L (2008) Spatial cooperativity in soft glassy flows. Nature, 454(7200):84–87.
-  Kamrin K, Koval G (2012) Nonlocal constitutive relation for steady granular flow. Phys Rev Lett, 108(17):178301.
-  Antonaglia J, Wright WJ, Gu X, Byer RR, Hufnagel TC, LeBlanc M, Uhl JT, Dahmen KA (2014) Bulk Metallic Glasses Deform via Slip Avalanches. Phys Rev Lett, 112(15):155501.
Appendix A Numerical evaluation of the critical stress
Using the extremal dynamics protocol, the system evolves to the critical point with an average stress and stress fluctuations in the stationary state with a dependence of system size as Eq.(13) in the main text. In the insets of Fig.(6), we plot out the as a function of , and the critical stress in the thermodynamic limit is just the intersection of the curves with axis, and we get for and for . From the dependence of on , shown in Fig.(6), we also extract in and in .
Appendix B Fixed stress protocol
At fixed stress in a finite size system, the dynamics will eventually stop. To trigger a new avalanche we give random kicks to all sites, of amplitude , while keeping fixed. We consider two methods. In the first one, a site is chosen randomly, and the amplitude of the kicks follows:
where is a constant adjusting the amplitude of kicks. Data presented in the text correspond to , but choosing smaller values of such as did not affect the results, see Fig.(9). Eq.(5) ensures that the stress is constant. If no sites become unstable, another site is chosen randomly and another set of kicks following Eq.(5) are given. In this method, the site that eventually becomes unstable was typically close to an instability before the random kicks were given. However is not necessarily the weakest site in the entire system.
In the second method, the dynamics is triggered by imposing that the weakest site (i.e. for all ) yields. According to our automaton model this leads to a change of local distance to instability everywhere in the system, which follows
and can lead to avalanches. We find that these two methods give consistent results for , as shown in Fig.(9).
Appendix C Finite size collapse of the flow curve
Our estimations of the threshold and the correlation length exponent are obtained in the main text using the extremal dynamics protocol. We obtain the same results if we use the fixed stress protocol, with which we can compute the size-dependent flow curve relating the strain rate, , as a function of the external stress, . From general arguments of finite size scaling, we expect:
To test the consistency of our methods, in Fig.(7) we collapse the different flow curves using Eq.(7) and the value of , and initially obtained with the extremal dynamics protocol. We observe a satisfying collapse without any free parameter.
Appendix D Avalanche statistics
To extract the avalanche distribution exponent accurately, we compare two protocols: (i) constant stress at and (ii) extremal dynamics, as shown in Fig.(8). It turns out that the avalanche distributions in extremal dynamics have stronger finite size effects than at constant stress. It is thus difficult to extract the avalanche exponent accurately using extremal dynamics. From the inset of Fig.(8)(right), doesn’t change significantly with system sizes in the constant stress method, in contrast to the estimate of that increases with in extremal dynamics. To extract accurately, we fix the stress at to collect the avalanche statistics, and we find in , and in , and the value of is the same for the two methods of fixed stress protocol, and also insensitive to the value of in the first method, shown in Fig.(9). The error associated to the exponent is estimated by varying the range of avalanche sizes considered in the fit.
Appendix E General scaling relations
The three scaling relations derived in the main text for the critical exponents of the yielding transition are similar but not identical to the scaling relations obtained for the depinning transition of an elastic interface. In the following we derive three more general relations, namely
that hold both for yielding and depinning. Here is the dimension of the interaction kernel . In the context of the yielding transition and so that . In the context of the depinning transition and , the dimension of the interaction kernel is for short range elasticity and for the long range elasticity of the contact line of a liquid meniscus  or of the crack front in brittle materials [63, 64].
Note that the relations (9) and (10) are expected to be very general, while the first relation is guaranteed only in presence of statistical-tilt-symmetry, hence only when the interactions are linear. For example, it is known that the non-harmonic corrections to the elastic energy can modify the universal behaviour of the depinning transition with critical exponents that violate the relation (8) . For the yielding transition the validity of (8) is supported by recent molecular dynamics simulations  that show that the stress decay during an avalanche is proportional to the energy jump, a scaling consistent with linear elasticity. Such linearity is assumed a priori in elasto-plastic models, and is required for the statistical-tilt-symmetry to apply, see below.
e.1 From the elasto-plastic automaton to the continuum model
The dimensional elasto-plastic model studied in this paper is a discrete automaton. Its continuum limit gives the time evolution of the strain field in each point of the space:
The first term of the equation describes the interactions between the different parts of the system. Note that the interactions are linear in the strain field , and governed by a time-independent interaction kernel, . As discussed in the main text, for elastic depinning models the kernel is monotonic, while for amorphous materials it is non-monotonic, anisotropic and can be conveniently written in the Fourier space:
The other two terms are the external stress, , and the quenched disorder, , which takes into account the inhomogeneities of the local yield stress. In the automaton model the scalar stress corresponds to the sum of the first two terms, and is assumed to be a collection of narrow wells randomly located along . The parameters , and are related respectively to the well depth, to the distance between consecutive wells and to the time needed to move from an unstable well to a stable one.
Below threshold, , the local strain fields are pinned inside a set of narrow wells. If a small perturbation is applied (e.g. a little change in the well locations), the local strain field responds either (almost everywhere) linearly simply readjusting its value inside the well, either (when a well becomes unstable) with a large modification accompanied by a stress release that can be the seed of a large avalanche. This non-linear response gives a singular contribution to the susceptibility which becomes important close to . Note that in presence of a non-monotonic interaction kernel, the avalanche size can be positive or negative, however the positive external stress strongly suppress negative avalanches that can, in practice, be neglected.
e.2 The statistical tilt symmetry
We now focus on the response of the system when we add to Eq.(11) a tilt, , namely an inhomogeneous local stress of zero spatial average. In presence of linear interactions, the tilt can be absorbed in a new strain filed defined as
and governed by the following evolution equation
The latter equation points out that the effect of the tilt can be absorbed with a shift of the location of the narrow wells. Thus, once the average over disorder is taken, the tilt disappears from Eq.(14) if the correlation only depend on . For example in the steady state, when the system becomes independent of the initial conditions, the average response of to a tilt acting on the mode , is
This exact expression should be compared with the scaling behaviour of the singular part of the susceptibility governed by the characteristic scale . In this regime the strain field grows as and noting that the tilt has the dimension of a stress, we expect that the singular part of the susceptibility scales as , which gives , namely Eq.(8). Here is the dimension of the kernel . For short range elastic depinning , for long range depinning , while the anisotropic kernel one has .
Concerning the other two scaling relations: Eq.(9) is identical to the one derived in the main text and Eq.(10) is still a consequence of the stationarity of the avalanche dynamics. In general an avalanche of size S leads to a stress drop which is not simply proportional to the plastic strain, but rather to , so that the average stress drop induced by avalanches scales as
On the other hand the stress injection before observing a new avalanche scales as , so that
Finally, using we obtain Eq.(10).
-  Joanny JF, De Gennes PG (1984) A model for contact angle hysteresis , J. Chem. Phys., 81:552.
-  Gao H, Rice JR (1989) A First-Order Perturbation Analysis of Crack Trapping by Arrays of Obstacles, J. Appl. Mech., 56: 828.
-  Bonamy D, Bouchaud E (2011) Failure of heterogeneous materials: A dynamic phase transition?, Phys. Rep., 498(1):1–44
-  Kardar M (1998) Nonequilibrium dynamics of interfaces and lines, Phys. Rep.,301(1):85–112 .
-  Salerno KM, Robbins MO (2013), Effect of inertia on sheared disordered solids: Critical scaling of avalanches in two and three dimensions, Phys. Rev. E, 88(6):062206.