Tribologia.

© All Rights Reserved

Просмотров: 1

Tribologia.

© All Rights Reserved

- A Study on the Tribological Characteristics of a Magneto-Rheological Elasteomer(Journal of Tribology)-Libre
- ASTM G133 - 2005
- Mae 493n 593t Lec1
- About Friction
- 9M303.22
- Friction Coefficient in Hot Compression of Cylindrical Specimens
- 9. What if i Never Change the Oil in My Car
- HOW to Design a Pump System
- 2-01_Kim
- 130_jul_92
- Course Outline
- Abrasive and Diffusive Tool Wear Fem Simulation
- 1-s2.0-S0301679X10000162-main
- Groz Beckert Needle Oil Detail
- Bff2821 Test
- (Tribology Series 44) Gwidon W. Stachowiak_ Andrew W. Batchelor and Grazyna B. Stachowiak (Eds.)-Experimental Methods in Tribology-Academic Press_ Elsevier (2004).pdf
- 06075676
- RE
- Physics D DCP CE Lab- Friction
- Chapter 5

Вы находитесь на странице: 1из 11

Tribology International

journal homepage: www.elsevier.com/locate/triboint

Mathieu Renouf , Francesco Massi, Nicolas Fillot, Aurélien Saulot

Université de Lyon, CNRS, INSA-Lyon, LaMCoS UMR5259, F-69621 Villeurbanne, France

a r t i c l e i n f o a b s t r a c t

Article history: Tribologists are confronted on a daily basis by the need to understand the causes and consequences of

Received 22 June 2010 friction on the behaviour of bodies in contact. Understanding contact behaviour is not only a scientiﬁc

Received in revised form curiosity but the key to solving numerous industrial issues. Numerical tools have been developed to

24 December 2010

overcome the problems encountered in experiments due to limitations in the local dynamic analysis of

Accepted 15 February 2011

Available online 2 March 2011

multi-scale systems (mechanisms, bodies in contact, interfaces). More than an exhibition of numerical

results, the present paper proposes reviewing the literature on the numerical tribology of dry contacts

Keywords: by analysing the different scales involved.

Tribology & 2011 Elsevier Ltd. All rights reserved.

Modelling

Dry friction

Multi-scale

1. Tribological context One of the main reasons for the difﬁculties encountered so far is

that problems involving friction and wear are multi-scale and multi-

Friction is part of our everyday lives, in common actions such physical by nature, as well as being strongly coupled. From the

as holding a pencil, walking, etc. From a pragmatic point of view, mechanical viewpoint, constraints at contact level depend on the

it is also the key issue of many industrial problems (mechanism contact itself as well as on the whole mechanism. From the material

lifetime [1–3], different types of wear [4,5], brake squeal noise [6], standpoint, the materials forming the bodies in contact (ﬁrst bodies)

vibration and noise reduction [7], etc.) and economic challenges and the interface (third body [14]) respond to both temporal and

(energetic expense, warming, etc.). These are just a few of the local spatial distributions of stresses, by exhibiting complex beha-

reasons, why, from the beginning of history (e.g. tribology on viour. In addition, friction and wear models are confronted by the

Egyptian frescoes [8]), human beings have tried to minimise or great complexity of mechanical, thermal and physicochemical

maximise friction according to their needs. phenomena occurring within contacts and affecting the different

Since the Renaissance, with the research of Leonardo Da Vinci parameters used in the models. This complexity results from the

(1453–1519), followed by that of Guillaume Amontons (1663–1705) large number of strongly correlated parameters.

and Charles Augustin de Coulomb (1736–1806), researchers have This complexity, involving different scales and physics, has led to

attempted to interpret friction by formulating basic equations. The the appearance of several approaches to studying tribological pro-

ﬁrst contact model was proposed in 1881 by Hertz [9] while the ﬁrst blems that range from experimental validations to numerical simula-

wear laws were proposed by Archard, who developed pin on disk tions. The innovative concept of ‘‘tribological triplet’’, proposed in the

experiments in 1953 [10–12]. These involved total wear volume, 1980s by Godet and Berthier [4,14], takes a different path from that

normal load, running distance and an additional physics—less of conventional schemes, as it underlines the importance of solving a

parameter k. Since Archard’s pioneering works and other directly tribological problem by simultaneously considering the bodies in

related studies, more than 100 wear models have emerged [13]. contact, the mechanism (mechanical system) that contains them, and

These models are based on the fact that the more parameters they the interface that separates them; the latter is called the third body

contain, the easier it is to match them to experimental results. To which refers to the so-called ‘‘ﬁrst bodies’’ that it separates.

overcome this generality, Lim et al. [13] deﬁned wear maps From an experimental viewpoint, at least four main difﬁculties

incorporating velocity and contact pressure, making it possible to can be identiﬁed when analysing a contact locally:

characterize the nature of wear (soft, hard, delamination, etc.).

Nevertheless, it was soon apparent that this kind of approach could the conﬁnement of the contact makes it difﬁcult (if not

not be transferred from one contact condition to another. impossible) to perform measurements ‘‘in situ’’ without dis-

turbing the contact itself;

the three components of the tribological system (mechanism,

Corresponding author. ﬁrst bodies and third body) act simultaneously, making it

E-mail address: mathieu.renouf@insa-lyon.fr (M. Renouf). difﬁcult to determine the pilot component and its degree of

0301-679X/$ - see front matter & 2011 Elsevier Ltd. All rights reserved.

doi:10.1016/j.triboint.2011.02.008

M. Renouf et al. / Tribology International 44 (2011) 834–844 835

interaction, bringing to mind the image of the snake swallow- area divided into sliding/sticking sub-regions depending on the

ing its own tail; type of stress involved. These models have been used since the

the local response of the material (work-hardening, superﬁcial ﬁrst simulations of railway dynamics and can be found in several

tribological transformation (STT) [5,15,16], etc.) to tribological multi-body software applications dedicated to wheel–rail con-

stresses (quasi-hydrostatic pressure, shearing gradient, etc.); tacts. Rolling contacts are still the subject of many works and the

tribometric measurements are difﬁcult to transpose to real reader can consult speciﬁc works for more details [33–36].

mechanisms because they always correspond to speciﬁc Due to the huge number and variety of contact and friction

mechanical loads and physicochemical contact conditions. models available in the literature, it is not possible here to provide

a complete review. Therefore, those readers who are interested

To overcome such experimental difﬁculties, tribologists were can refer to the texts of Bowden and Tabor [37], Rabinowicz [38],

quick to recognise the potential of numerical tools for overcoming Suh [39], Kragelskii [40], and Dowson [8] for further details and

them; nowadays, simulations can be carried out at different references.

scales, taking advantage of the fact that it is possible to vary

one parameter at a time, something that is almost impossible

experimentally. Following a guideline based on the concept of the

tribological triplet, approaches at the scales of the mechanism, 3. Modelling mechanism and ﬁrst bodies

the bodies in contact and the interface are dealing with in the

following sections. After presenting the ﬁrst contact models 3.1. Modelling reality

in Section 2, Sections 3 and 4 present numerical tools used at

the scale of the mechanism and the ﬁrst bodies and at the scale of Although previous analytical models were meaningful tools that

the interface, while Section 6 presents more recent advances in led to understanding of the phenomena involved in contact

dry numerical tribology. For each section, the variety of numerical mechanics and tribology, their limits quickly became apparent when

methods used in tribology at each scale is presented, the hypoth- dealing with more realistic issues in which the role of the mechan-

esis retained to consider the other scales (above or below) is ism, together with the contact, has to be accounted for. For example,

described, and a relevant example is given. predicting the lifetime of a mechanical component like a bearing,

blade disk or brake system must take into account both the dynamic

analysis of the mechanism up to its end-of-life and knowledge of the

2. First contact models local transformations of the bodies in contact (local deforma-

tions [15,41], surface degradation [42,43], etc.).

For many decades, researchers in the ﬁeld of contact mechanics Thus additional modelling of both the dynamics of the whole

have tried to use simple models to describe the behaviour of the system (mechanism) and the local behaviour of the bodies in

relative motion between two bodies. Hertz was undoubtedly the contact (ﬁrst bodies) is necessary. Nevertheless, the difference

ﬁrst to propose a contact model [9] that initially comprised contact between the spatial and temporal scales of the two aspects of the

between a cylinder and a ﬂat surface and then between two spheres problem implies that developing a single model that includes

of different radii. The model offered a simpliﬁed representation of both the global behaviour of the mechanism and the detailed

the local deformation of the body, depending on the elasticity of the description of the bodies in contact results in cumbersome

materials in contact, and provided the stress ﬁeld in the contact area operations that are often impossible to achieve.

relating to the normal force, the radii of curvature of the two bodies Deﬁning objectives as a function of the problem to be solved.

and their elastic modulus in a static conﬁguration. This model is In the case where the model is designed to simulate the dynamic

valid for frictionless contacts, assuming that the contact area is small response of a mechanical system, the contact (either sliding or static

in comparison to the size of bodies in contact and that the stresses contact joints) is accounted for by reduced parameters that are

localised at the contact zone are less than the limit of elasticity. It assumed to represent the effect of the contact on the system’s

also assumes a continuous and smooth distribution of the pressure dynamics. These parameters generally consist of global contact

in the contact area. stiffness factors [44,45], contact damping factors [46] and energy

This theory formed the basis of many models, some of which loss factors if the model includes energetics [47]. Moreover, to

used the hypotheses of the original model. Extensions were overcome the complexity and the non-deterministic effects of the

proposed for other regular geometries [17,18], more complex nonlinearities of contact joints on the dynamics of a mechanism,

geometries and for accounting for the presence of friction and statistical techniques dealing with uncertainties have been devel-

inelastic behaviour [19]. oped, such as statistical energy analysis (SEA) [48]. In such models

In his model, Hertz considered perfectly smooth contact surfaces. the contact between the subsystems (components) of a complex

In the 1960s, Greenwood proposed a static contact model for rough mechanical system are represented by energy loss factors account-

surfaces [20]. This generalisation was to be the basis of many other ing for the loss of energy at the joint, and energy coupling factors

models proposed to describe contact surfaces by more or less accounting for the energy transfer between the subsystems. An

complex functions developed by the combination of trigonometric extensive review of these approaches and the huge amount of

functions [21], by the use of fractal model [22] or by numerical related literature is beyond the scope of this work, although readers

models involving ﬁnite element (FE) approaches [23–28]. will ﬁnd the articles mentioned in the references.

Hertz’s theory was extended to models that took into account On the other hand, when dealing with the contact itself, two

dynamic aspects. Carter [29] proposed a model of a line contact to different approaches have been proposed over recent decades:

describe the contact between wheel and rail, accounting for analytical or semi-analytical models, and numerical models

longitudinal slip only. He highlighted the coexistence of areas of (ﬁnite element). These approaches may appear contradictory

sticking and sliding within the contact zone. Carter’s simpliﬁed although they sometimes complement each other.

model was then extended to circular contacts [18] and elliptic Table 1 shows a comparison between the different approaches,

contacts [30], allowing the consideration of transverse sliding. emphasizing their respective advantages, drawbacks and several

Based on the work of Johnson, Kalker proposed a set of algorithms typical applications. Readers new to the ﬁeld can take advantage

using more or less complex models [31,32]. The contact area was of this initial comparison for choosing the approach to be used for

no longer seen as a uniform sliding or sticking region but as an each speciﬁc problem (Table 2).

836 M. Renouf et al. / Tribology International 44 (2011) 834–844

Table 1

Comparison between methods for modelling mechanism and ﬁrst bodies.

Semi-analytical Transmission systems Low computational cost Lost of local contact information Mechanical system scale

Full bearing systems Full system modelling Simpliﬁed ﬁrst body modelling Large time and space scale

Vehicle dynamics Long time simulation Application depending modelling

Energetical formulation Vibration analysis of jointed Low computational cost No local contact information First body scale or simple

structures Interaction between different Simpliﬁed ﬁrst body modelling system scale

Fluid–structure coupling Behaviour of ﬁrst bodies need of pre-modelling Only stationary states

Standard F.E. Transient model analysis of simple Local contact information Large computational cost result First body scale or simple

systems Full ﬁrst body modelling precision depending on the contact system scale

Contact instabilities Availability of robust general modelling no third-body effects Contact space scale

First body contact analysis formulations Short time scale

Substructured F.E. Transient model analysis of simple Local contact information Reduced ﬁrst body modelling result First body scale or simple

systems Reasonable computational cost precision depending on the contact system scale

Contact instabilities Robust formulation availability modelling Contact space scale

First body contact analysis No third-body effects Relative short time scale

Table 2

Comparison between continuous, discrete element (D.E.), coarse grain (CG), united atom (UA), molecular dynamics (MD), Monte Carlo (MC) and ab initio approaches for

modelling the third body and smaller system.

Continuous Powder lubrication Low computational cost Unknown a priori Interface scale

Rheological model interface behaviour

No wear modelling

No localisation phenomena

D.E. Powder lubrication No a priori behaviour Large computational cost Part of the interface

Wheel–rail contact Wear modelling Unknown a priori

Machining Localisation phenomena Interaction laws

C.G. Polymer modelling Large number of atoms represented by Difﬁcult to ﬁnd equivalent interaction Systems with very

an equivalent entity potential long molecules

U.A. Alkane modelling Few atoms grouped in equivalent energy Simpliﬁed molecule shape and Nanometer scale

sites interaction potentials

Biological molecules

M.D. Nano-indentation Deterministic simulation of atom motion Very high computational cost Nanometer scale

M.C. Solids at rest Fast to ﬁnd equilibrium positions No explicit time evolution Nanometer scale

Ab Initio Atomic interactions Includes electron modelling Long computation for only a few atoms Atomic/ subatomic

determination scales

3.2. Semi-analytical models wear are often implemented; these models are then assembled

together and dialog with each other through their respective

Semi-analytical (SA) models were the ﬁrst tools to be used inputs and outputs.

either for simplifying contact problems by trying to account for SA models have been used for understanding wear phenomena

mechanism behaviour or for understanding the behaviour of like fretting [1,50], in simpliﬁed elasto-plastic and wear models [49].

solids subjected to contact stresses over long periods of time. Other researchers have developed semi-analytical methods for

The latter approach is that most commonly adopted for modelling modelling ball bearing behaviour [2,51,3], by simplifying the inter-

the behaviour of mechanical systems over several hours opera- action between the bearing elements with visco-elastic contact

tion. In such cases computing-time becomes one of the main laws. The literature on semi-analytical models is extremely rich

constraints, forcing researchers to develop simpliﬁed models of and is growing rapidly every year, with speciﬁc branches that each

both the dynamic behaviour of system components and contact deal with particular issues, thus it is not possible to give a complete

stresses. These approaches have become efﬁcient tools for solving overview of existing works and those readers interested should refer

well-deﬁned problems more than general tools for understanding to more detailed reviews that deal with speciﬁc applications.

contact problems [1,49], because the phenomena and the scales of Nevertheless, it is worth mentioning the works of Nélias [52,1,49]

the mechanism and the contact are completely uncoupled. Con- concerning dry contact problems and the works of Wensing [7,3]

sequently one of the main drawbacks of these models, as pointed dealing with lubricated contact problems. A detailed review of

out in Table 1, is their speciﬁcity with respect to the problem to friction models for friction compensation methods in mechanical

be solved. There is no general formulation capable of being used systems with contact is provided by Armstrong-Héloury [53]. A

for several different problems. bridge between semi-analytical and continuous models is given by

The SA feature of the models described above can be used for the review presented by Oden and Martins [54].

time discretization in analytical models to simulate the evolution An interesting improvement and analysis of SA models can be

of a solution to a problem through time. Different simpliﬁed found in the works of Gallego et al. [49]. The authors propose an

analytical models accounting for friction, plastic deformation or alternative method to the FE method to perform wear computations

M. Renouf et al. / Tribology International 44 (2011) 834–844 837

on dovetail joints. Although an FE model is used to solve the effort and are thus still preferred for prediction over longer periods

structural problem, the SA model focuses on each contact area and of time.

its speed allow computing cyclic wear easily. Fig. 1 shows the wear An example of more phenomenological approaches is the effort

depth proﬁles after one fretting cycle with unworn surfaces (i.e. employed by researchers in modelling brake systems by FE meth-

after the ﬁrst cycle). The solid line in black corresponds to the FE ods, in order to investigate the brake squeal phenomenon. A review

results used as a reference case while the lines with symbols of recent literature [57] dealing with the numerical analysis of brake

correspond to the results provided by the SA model with different squeal shows how FE transient analysis has been employed for

levels of complexity. reproducing and investigating squeal vibrations (Fig. 2) and local

contact stresses, while FE complex eigenvalue analysis has been

developed for predicting brake dynamic instabilities (Fig. 3). While

3.3. Numerical methods for continuous modelling FE models have been of fundamental importance for understanding

the mechanism at the origin of the self-excited friction vibrations

Finite element modelling is the most commonly used numer- that cause squeal noise, their limits have been highlighted as they

ical method for solving contact problems with friction between cannot be used to develop a predictive FE tool. The lack of

deformable bodies [55]. Initially, researchers in tribology were information at the scales of the third body and of the global

not attracted by this numerical tool. One of the main reasons was mechanical system, together with limitations on model size and

the extremely long computational time required in comparison the huge computational efforts demanded for transient analysis,

to more efﬁcient analytical or semi-analytical approaches (see prohibit the development of a quantitative FE model of such systems

Table 1). Nevertheless, numerical approaches using ﬁnite element without updating the parameters through experimental comparison.

methods were given increasing attention by tribology journals by Therefore, despite several efforts, FE models remain more useful as

the end of the 1980s [56]. Nowadays, they have become an tools for understanding than for predicting.

essential method for solving contact problems where mechanism Several methods have been developed over recent decades for

and ﬁrst body behaviour is the main subject of analysis. Notwith- solving contact problems between two deformable bodies. They are

standing, it is important to pay attention to the different appro- generally classiﬁed as a function of the contact algorithm and the

aches and methods that can be adopted. Indeed, both the time time integration scheme used. The contact algorithms most com-

integration method and the modelling of contact elements can monly used employ either Lagrange multiplicators [59–61], the

considerably affect the results of ﬁnite element simulations. regularization parameters [62], or LCP (Linear Complementary

Transient analysis by the ﬁnite element method is better adapted Problems) [63]. Time integration schemes can be either explicit or

to the analysis of local phenomena over short periods (from a few implicit for describing the motion of a structure and for dealing with

milliseconds to several seconds, according to the application) rather a contact [61,64,65,62].

than the simulation of the transient behaviour of a complex system As asserted for the semi-analytical models, it would be difﬁcult

over long periods (several hours). The computational effort, as and, in any case, beyond the scope of this review to cite all the

opposed to semi-analytical methods, is no longer considered to be works dealing with contact problems and ﬁnite element methods.

a constraint, despite the fact that it can be cumbersome. Despite this The combination of these issues covers theoretical works dealing

expense in terms of computational effort, these models have been with convergence and the unicity of numerical solutions [66], works

proved as useful tools for understanding problems. Moreover, this dedicated to developing efﬁcient algorithms [67,63,68,69], and an

limitation is being constantly reduced by numerical optimizations extremely wide range of different applications (e.g. [70,34]). With

that use clusters and parallel calculation. Nevertheless, semi-analy- respect to tribological issues, i.e. understanding the causes and

tical models continue to be more efﬁcient concerning computational effects of friction in bodies in contact, the works of Ham [56] and

Fig. 1. Wear depth proﬁles after one fretting cycle with unworn surfaces (i.e. after the ﬁrst cycle). The solid line in black corresponds to the FE results while the lines with

symbols correspond to the results provided by the SA model with various levels of complexity [49].

838 M. Renouf et al. / Tribology International 44 (2011) 834–844

Contact forces between pad ans disc x 104 Y-acceleration of pad edge

250

1

200

Acceleration [m.s2]

Normal force 0.5

150

Force [N]

100 0

50

-0.5

0 friction force

-1

-50

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08

Time [s] Time [s]

Fig. 2. Transient FE analysis of brake squeal: (a) normal and tangential forces between brake pad and disc; (b) tangential acceleration calculated at the brake pad. FE

simulation makes it possible to reproduce the vibrations triggered in the contact [58].

Fig. 3. Complex eigenvalue FE analysis of brake squeal: (a) natural frequencies of the brake system cross each other and can become unstable; (b) locus plot of system

eigenvalues: the real part of one eigenvalue become positive, i.e. the system is unstable.

Baillet [59,60] deserve mention. Others works have aimed at under- case of a ﬂuid lubricant, and the same idea can be generalised to

standing speciﬁc contact phenomena, for example the behaviour of solid interfaces. Indeed, Godet [14] and Berthier [4] showed that this

composite materials subjected to tribological stresses [71], the more or less heterogeneous and discontinuous interface called

development of contact instabilities [72], wear laws in biomecha- ‘‘third body’’ can have different thicknesses (ranging from several

nics [73], the occurrence of rail defects [34], dynamic instabilities nanometers [82] to several tens of micrometers [83]) and ﬂow with

leading to squeal noise [58,74], contact rupture [75,76], contact an as yet unknown rheology. This discontinuous interface is present

wave generation and propagation [77–79]. The reader will ﬁnd a in an artiﬁcial way in lubrication powder (e.g. MoS2 for space

more complete bibliography concerning each speciﬁc domain in the lubrication, etc.) and, in this case, the properties of the media are

works cited. assumed to be known. Nevertheless, the evolution of such media

As highlighted in Table 1, recent advances in substructuring and their behaviour in response to strong pressures and shear rates

methods make it possible to consider a new approach halfway could change, making the interpretation of results complex. There-

between semi-analytical and FE models [80,81]. Substructured FE fore having a continuous view of this kind of interface does not

modelling allows accounting for contact distribution, the connec- always ensure that the reality of the contact is apparent, especially

tions between substructures, and it drastically reduces the num- when the wear rate is accounted for, or when the third body ﬂows

ber of degrees of freedom of the global FE model. First, single are not in steady state (which is rarely the case).

substructures are modelled and the matrix reduction technique is Two kinds of approaches are commonly used to model the third

used to reduce the system matrices to a smaller set of degrees of body ﬂow: continuous and discrete approaches. Continuous

freedoms. Then the reduced matrices of the single substructures approaches offer extensions of models derived from continuum

are assembled together to account for the general boundary mechanics and are essentially used in applications involving lubri-

conditions and the contact stresses at the contact interfaces cation powder, i.e. using an artiﬁcial third body. In the case of a

between single substructures. This approach dramatically reduces natural third body (resulting for the most part from the bodies in

computational effort at the expense of simplifying (reducing) contact), continuity hypotheses cannot be retained and, under such

modelling of the ﬁrst bodies (substructure) response. conditions, it is necessary to verify whether this approach is still

valid and what its limits are. This is the main reason for using

discrete element approaches to describe the evolution of a discon-

4. Within the tribological interface tinuous medium and to infer rheological behaviour.

4.2. Continuous approaches

When focusing on the interface that separates two bodies in

contact, ﬂows of the latter can be observed under certain conditions There is no universal method for describing lubrication powder or,

(high pressures and shear rates). This is easily understandable in the more generally, granular ﬂows. Therefore many approaches attempt

M. Renouf et al. / Tribology International 44 (2011) 834–844 839

to describe such ﬂows that are often reduced to speciﬁc geometries. interaction forces between particles must satisfy the mechanical

In the granular tribology community, the most widely used equations equilibrium of the system and use local laws with a physical

are those proposed by Haff [84] or Lun [85]. They are used in many sense involving the fewest possible calibration parameters.

applications involving bearings [51], Couette geometries (annu- Ten years after the ﬁrst discrete element simulations in ﬂuid

lar) [86], parallel plates [87,88] and converging plates [89]. lubrication [103,104], discrete element methods were used for

The equations of conservation (mass and momentum) are used the ﬁrst time to simulate lubrication by bearing powder [105].

to describe the continuous behaviour of the medium [90]. These The third body is represented by a collection of rigid spheres

equations are similar to the energetic equations used for ﬂuids in moving between two rough inclined planes. This work paved the

which variations of granular temperature are a balance between way for many succeeding studies. The ﬁrst were very exploratory

the energy injected into the system and the energy dissipated by and concerned the inﬂuence of different numerical [100,106]

friction and collisions. In this set of equations, the system is and geometric parameters (size of samples and particles) [107]

closed by a constitutive equation that connects the stress tensor on the mechanical response of the medium. Then more repre-

to the strain tensor. Regarding the available models satisfying sentative laws involving phenomena within an interface were

such a condition, the reader can refer to the work of Hesh- used (JKR contact model [108]) [109] and the inﬂuence of particle

mat [91,92], Tichy [93,94,89] or those of Khonsari [88] involving size and the cohesion between particles on the ﬂow behaviour were

different geometries. studied.

These methods were then extended to the study of wear

4.3. Discrete approaches mechanisms with simple shear [110,111], offering wear laws

based on discrete approaches. Other works concerned third body

Discrete element approaches allow representing the interface ﬂows in fretting-like simulations [112].

as a collection of more or less heterogeneous particles indepen- More recently some authors have accounted for the deformation

dent of each other. This type of method emerged in the early of particles in simple shear simulations [101,113]. The models are

1970s to model the evolution of rockfalls [95,96]. It was quickly not representative of dense granular ﬂows observed experimentally,

adapted to modelling numerous problems involving media with but they propose a new approach in the simulation of interfaces that

divided features (geophysical structures [97], ballast [98], masonry accounts for particle deformation. Fig. 4 presents the sample used in

[99], etc.). the works of Kabir et al. [101] using an explicit dynamic ﬁnite

The equations of continuum mechanics are no longer used to element model of parallel shear cells, containing 52 particles and

describe the evolution of systems, instead the equations of motion consisting of round and multishaped granules (diamond, triangle,

and laws managing the interactions between particles are imple- and rectangle). From a different perspective this kind of approach is

mented. They play the role of conservation and constitutive equations used to revisit the theory of Hertz in a medium with inclusions

respectively. Although there are many variations of the discrete [114].

element method (DEM) used in tribology, they all rely on three key Although several improvements have been performed in the

ingredients: last decade, three questions remain unanswered relating to

modelling the third body by using the discrete element method.

the time integration of the equation of mechanics; The ﬁrst question is that of how to choose the size of a single

contact detection between particles; particle, since numerical particles do not represent real particles.

computation of interaction forces between particles. The shapes used in models remain geometrically simple (disks or

spheres according to the dimension) and exhibit bulk behaviour

Time integration must not modify the mechanical balance of the (essentially rigid even if recent improvements take into account

system. Indeed, if a given energy is injected into the system, it is elastic deformation). Consequently, numerical particles play the

not dissipated numerically by the numerical scheme used role of (representative?) elementary volume when used with

(Euler [100], Newmark [101], theta-method [64], Verlet [102], homogenisation techniques, while the law that governs their

etc.). Contact detection must be accurate to properly describe the interaction represents the integration of local phenomena.

connections between the particles of the system. Although this The determination of such laws is the second main open

appears easy for spherical particles, it is more difﬁcult when question. How can local elasticity, plasticity, viscosity, damage,

bodies with complex geometries are used. Finally, computation of etc. be accounted for? Models use phenomenological laws based

Fig. 4. Two-dimensional 52 particle granular ﬂow shear cell ﬁnite element model [101]. (a) Round particles and (b) Multi-shaped particles.

840 M. Renouf et al. / Tribology International 44 (2011) 834–844

on basic assumptions. This is due to the lack of experimental MC approach solves the Boltzman equation where the motions

results concerning real interactions of third body particles. There- of atoms are simulated by statistical laws. This method proves to

fore current models offer only phenomenological results and be very efﬁcient for ﬁnding equilibrium positions but is unable to

should not be considered as generic. obtain an explicit evolution for atoms through time. On the

The last open question is related to spatial and temporal contrary, MD simulation sums up inter-atomic forces, and then

scales. Simulating third body ﬂows takes into account only a solves the classical Newton equation of motion over each atom,

small part of the contact interface within a short time. How can providing a deterministic way of simulating their displacement

these local results be extended to the evolution of a whole through time. Nevertheless they are very computationally costly

contact? At present mean local values are used by the upper and simulate only several nanometers. A variety of adaptations of

scale. The Section 6.1 will propose a few alternatives. molecular dynamics can be found in the literature to increase the

size of the system, ranging from nanoscale to higher scales. Atoms

can be grouped together, represented in only one point of space

5. Nanometric characterization of interfaces and their interaction potentials merged together in order to

deﬁne an equivalent interaction site (united atoms, coarse grain

One of the major problems for discrete simulations of third methods). However, when going up in scale, as soon as geometry

body ﬂows at the mesoscale (because of the complexity of matter) and potentials are simpliﬁed in the models (as matter becomes

is the quantitative deﬁnition of the interaction forces between increasingly complex) the latter looses its precision.

particles. Nevertheless, at the scale of the atoms and molecules Concerning ‘‘nano-tribology’’, early models for atomic friction

composing this medium, it is possible to deﬁne reliable inter- such as the Tomlinson model [117] or Frenkel–Kontorova model [118]

atomic laws from quantum calculations [115]. Thus two main involve two crystalline surfaces in direct contact, assuming that each

kinds of method are used to describe the motion of atoms: the surface atom is linked only to the center of the mass of the rubbing

Monte Carlo (MC) approach and molecular dynamics (MD) [116]. body [117] or that the surface atoms are linked only to their nearest

Fig. 5. Snapshots showing the evolution of a Cu(1 0 0) tip on a Cu(1 0 0) substrate during sliding to the left [119].

M. Renouf et al. / Tribology International 44 (2011) 834–844 841

neighbour by springs [118]. With the development of computers and At the scale of the interface, models offer richer descriptions

the use of both MC and MD methods, these friction models have been by mixing discrete and continuous approaches in a single simula-

developed over the years to represent a crystalline layer of adsorbed tion [134,133]. The inﬂuence of the deformation of the bodies in

atoms, including rearrangements of atoms due to the equilibrium contact is directly accounted for by the rheology of the interfaces.

condition (i.e. temperature) or sliding. Other models provide a dialogue between different scales by

To investigate atomic scale asperities and ﬁnite contact areas, applying an iterative method [134,135]. The friction at the scale of

many authors (such as [119], see Fig. 1) have modelled AFM-like the bodies in contact is the result of a simulation performed at the

tips-on-substrate contacts. Since the shearing velocity in classical scale of the interface (third body), while the boundary conditions

MD simulations is very high compared to real AFM experiments, of the simulations at the interface scale are provided by the local

large scale molecular dynamics simulations are still under devel- dynamics at the scale of the ﬁrst bodies.

opment, involving massive parallel programming [120] (Fig. 5). Some models couple the ﬁnite element method and molecular

Many tip-substrate simulations focusing on indentation were dynamics to account for a richer description of the interface and

performed in the 1990’s [121,122]. They showed that when the surface condition [136,137], while other methods couple

approaching the two surfaces (usually metallic surfaces) to each discrete models at the scale of the material’s structure to reﬂect

other, the normal force remains almost null, after which it suddenly local degradation in gear dynamics [138].

becomes extremely attractive with a jump to contact. This phenom-

enon was not so clear when approaching two diamond surfaces in 6.2. Towards multi-physics models

the case where at least one was hydrogen-terminated [123].

Regarding sliding, nanoscale simulations often highlighted inter- Multi-physics models are commonly used at the scale of the

mittent motion, with the system alternatively sticking and slipping ﬁrst bodies. Indeed, thermo-mechanical models based on robust

forward, as described by Rabinowicz [38] and Bowden and formalism have been implemented [139].

Tabor [124]. The stiffness of the contacting bodies and the commen- Concerning solid interfaces, multi-physics models have been

surability of the surface, i.e. the way atoms are arranged regularly (or developed only recently. Models coupling mechanical and electrical

not) depending on the crystalline lattice orientation [125–127], play aspects have been developed to understand phenomena such as

an important role in the dynamic friction process. shunting in the wheel–rail contact [140]; thermo-mechanical mod-

The ﬁrst tribochemical reactions in molecular dynamics simu- els have been developed to describe energy dissipation within the

lations were observed by Harrison and Brenner in 1994 [128], contact interfaces [141–143]. This work allowed a new approach to

who used reactive potentials that allow for the breaking and the concept of friction by incorporating thermal discontinuous

formation of chemical bounds. This brought to light the origin of aspects and updating the Blok theory [144].

the formation of molecular wear debris. More recent models are capable of coupling mechanical,

An impressive compilation has been proposed in [129], where the thermal and physical chemistry within the interface. This kind

reader will ﬁnd many complementary aspects related to this section. of approach allows linking every physical feature within a single

simulation [145,146]. The limits of these approaches depend on

the experimental data used feed the models.

6. Numerical tribology today

During the last decade, numerical tribology has evolved con- 7. Discussion and conclusion

siderably. It has taken advantage of the numerous numerical

developments achieved in the domain of contact mechanics and The numerical tools used at the scales of the mechanism, the

adapted them to its own needs. Numerical tribology has now bodies in contact and the interface have been presented following

become an indispensable tool for studying contact problems. a guideline based on the concept of the tribological triplet. The

Nevertheless, the need for in-depth understanding of the semi-analytical model, ﬁnite element method, discrete element

phenomena involved is leading to the development of models method and molecular dynamics approach appear to be the most

that must be as accurate as possible. They are neither longer commonly used tools.

purely mechanical, nor based on only one scale, as they incorpo- At the scale of the mechanism and the ﬁrst bodies, semi-

rate different physics and consider several scales. analytical models and ﬁnite element methods appear to be the most

appropriate tools for describing the behaviour of an entire system

and that of each component. SA models can be used to simulate the

6.1. Towards multi-scale models long-term evolution of full systems at low computational cost.

Nevertheless, they entail a loss of local contact information, the use

In the past, most of the numerical work in tribology was of simpliﬁed ﬁrst body modelling, and models dedicated to only

performed at a single scale, accounting for the other scales by one application. The ﬁnite element method made full ﬁrst body

boundary conditions or local conditions. Although the idea of modelling and the measurement of local contact characteristics

coupling the different scales is not a new one [5,130], it is only possible. Moreover, robust general formulations are available. On

recently, in tribology, that different approaches have tried to the other hand, the approach is computationally costly when the

account for the different scales of the tribological triplet within a results depend on contact modelling and it is not possible to simulate

single simulation, either directly or indirectly. the long-term evolution of a contact, even when using substructured

For example, certain works couple the mechanism and bodies approaches.

in contact to analyse bearings [51,131]. In the ﬁrst case, a nodal At the scale of the interface, rheological models are computa-

reduction is performed on the whole system and coupled with tionally inexpensive, enabling description through wear model-

condensation techniques. In the second case, the mechanism ling and reproducing localisation phenomena. Finally, they can be

results from a multi-body model and injected as boundary used to postulate an a priori unknown interface behaviour. On the

conditions into the ﬁnite element simulation of the contact. contrary, the discrete element method can be used to consider

Other works couple smaller scales (the scale of the bodies in wear and localisation processes but suffers from high computa-

contact and the interface) to account for the effect of local damage tional costs and is based on a priori unknown interaction laws (in

on the behaviour of bodies in contact [132,133]. most cases).

842 M. Renouf et al. / Tribology International 44 (2011) 834–844

At the nanoscale, molecular dynamic approaches provide good [25] Pei L, Hyun S, Molinari JF, Robbins MO. Finite element modeling of elasto-

descriptions of matter and are well-suited to nano-applications, plastic contact between rough surfaces. J Mech Phys Solids 2005;53:

2385–409.

but in the case of more microscopic problems, mapping informa- [26] Persson BNJ. Elastoplastic contact between randomly rough surfaces. Phys

tion becomes complex. Rev Lett 2001;87:116101.

Although many studies at different scales have led to clarifying [27] Persson BNJ, Alboht O, Tartaglino U, Volokitin AI, Tosatti E. On the nature of

the concept of friction, there is still some way to go before reliable surface roughness with application to contact mechanics, sealing, rubber

friction and adhesion. J Phys: Condens Matter 2005;17:R1–62.

predictive models of friction and wear can be obtained. Until we [28] Persson BNJ, Sivebaek IM, Samoilov VN, Khao K, Volokitin AI, Zhang Z. On

are able to model a full system, friction will remain a scientiﬁc the origin of amonton’s friction law. J Phys: Condens Matter 2008;20:

smokescreen. A great deal of research work is currently taking 395006.

[29] Carter FW. On the action of a locomotive driving wheel. Proc R Soc London,

this direction and integrating different physics, with dialogue

Ser A 1926;112:151–7.

between different scales. In conclusion, it is possible to assert [30] Haines J, Ollerton E. Contact stress distribution on elliptical contact surfaces

that, thanks to numerical tribology and its interactions with subjected to radial and tangential forces. Proc Int Mech Eng 1963;177:

experimental tribology, tribologists are wearing friction down. 95–114.

[31] Kalker JJ. The computation of three-dimensional rolling contact with dry

friction. Int J Numer Methods Eng 1979;14(1293–1307):1293–307.

[32] Kalker JJ. Mathematical models of friction for contact problems in elasticity.

Acknowledgements Wear 1986;113:61–77.

[33] Colette C. Usure ondulatoire des rails dans les transports urbains. Master’s

thesis, Université libre de Bruxelle; 2003.

Part of this work is funded by the Agence National de la [34] Saulot A, Baillet L. Dynamic ﬁnite element simulations for understanding

Recherche in the framework of the ANR-JCJC08-0020 DiNEET wheel–rail contact instability states occurring under sliding conditions.

ASME J Tribol 2006;128:761–70.

project (Dialogue Numérique Entre Echelles Tribologiques). [35] Andersson C, Johansson A. Prediction of rail corrugation generated by three-

dimensional wheel–rail interaction. Wear 2004;257(3–4):423–34.

[36] Telliskivi T, Olofsson U. Wheel–rail wear simulation. Wear 2004;257(11):

References 1145–53.

[37] Bowden FP, Tabor D. The friction and lubrication of solids. 2nd corrected ed.

Oxford University Press, Clarendon Press; 2001. [oxford classic text in the

[1] Gallego L, Nélias D, Deyber S. A fast and efﬁcient contact algorithm for fretting

physical sciences edition].

problems applied to fretting modes i, ii and iii. Wear 2010;268(1–2):208–22.

[38] Rabinowicz E. Friction and wear of materials. New York: Wiley; 1965.

[2] Morales-Espejel G. Semi-analytical methods for surface life analysis in

elastohydrodynamically lubricated and dry contacts; 2009. [39] Suh NP. Tribophysics. PrenticeHall; 1986.

[3] Wensing JA. On the dynamics of ball bearings. PhD thesis, University of [40] Kragelskii IV. Friction and wear. London: Butterworth; 1965.

Twente; 1998. [41] Carroll RI, Beynon JH. Rolling contact fatigue of white etching layer: Part 1:

[4] Berthier Y. Experimental evidence for friction and wear modelling. Wear crack morphology. Wear 2007;262(9–10):1253–66.

1990;139:77–92. [42] Cristol-Bulthe A-L, Desplanques Y, Degallaix G. Coupling between friction

[5] Berthier Y. Handbook of materials behavior models. In: Background on physical mechanisms and transient thermal phenomena involved in pad-

friction and wear. Lemaı̂tre Academic Press; 2001. p. 676–99. disc contact during railway braking. Wear 2007;263(7–12):1230–42.

[6] Massi F, Giannini G, Baillet L. Brake squeal as dynamic instability: an [43] Kasem H, Bonnamy S, Berthier Y, Jacquemard P. Fiber-matrix unbonding

experimental investigation. J Acoust Soc Am 2006;120(3):1388–99. and plastic deformation in c/c composites under tribological loading. Wear

[7] Tillema H. Noise reduction of rotating machinery by viscoelastic bearing 2010;269(1–2):104–11.

supports. PhD thesis, University of Twente; 2003. [44] Nobari AS, Robb DA, Ewins DJ. Model updating and joint identiﬁcation, their

[8] Dowson D. History of tribology. 2nd ed. London and Bury St. Edmunds, ranges of application, restriction and overlap. IJAEMA 1993;8(1):93–105.

England: Professional Engineering Publishing; 1998. [45] Liu W, Ewins DJ. Substructure synthesis via elastic medi part i: joint

[9] Hertz H. Über die berührung fester elastischer körper (sur le contact entre identiﬁcation. In: Proceedings of the 18th international modal analysis

corps élastiques). J für reine und angewandte Mathematik 1881;92:156–71. conference, San Antonio, USA, vol. II; 2000. p. 1160–6.

[10] Archard JF. Contact and rubbing of ﬂat surfaces. J Appl Phys 1953;24(8): [46] Gaul L, Bohlen S. Identiﬁcation of nonlinear structural joint models and

981–8. implementation in discretized structure models. In: Proceedings of the 5th

[11] Archard JF. Elastic deformation and the contact of surfaces. Nature international modal analysis conference, vol. I. London; 1987. p. 86–91.

1953;172(4385):918–9. [47] Ji L, Mace BR. Statistical energy analysis modelling of complex structures as

[12] Archard JF. Elastic deformation and the laws of friction. Proc R Soc London, coupled sets of oscillators: ensemble mean and variance of energy. J Sound

Ser A 1957;243:190–205. Vib 2008;317:760–80.

[13] Lim SC, Ashby MF, Brunton JH. Wear-rate transitions and their relationship [48] Culla A, Sestieri A, Carcaterra A. Energy ﬂow uncertainties in vibrating

to wear mechanisms. Acta Metall 1987;35:1343–8. systems: deﬁnition of a statistical conﬁdence factor. Mech Syst Signal

[14] Godet M. The third-body approach: a mechanical view of wear. Wear Process 2003;17(3):635–63.

1984;100:437–52. [49] Gallego L, Fulleringer B, Deyber S, Nélias D. Multiscale computation of

[15] Busquet M, Descartes S, Berthier Y. Formation conditions of mechanically fretting wear at the blade/disk interface. Tribol Int 2010;43(4):708–18.

modiﬁed superﬁcial structures for two steels. Tribol Int 2009;42(11–12): [50] Fouvry S, Kapsa Ph, Vincent L. An elastic-plastic shakedown analysis of

1730–43. [Special issue: 35th Leeds–Lyon symposium]. fretting wear. Wear 2001;247(1):41–54.

[16] Eleod A, Oucherif F, Devecz J, Berthier Y. Conception of numerical and [51] Weinzapfel N, Sadeghi F. A discrete element approach for modeling cage

experimental tools for study of the tribological transformation of surface

ﬂexibility in ball bearing dynamics simulations. ASME J Tribol 2009;131:

(tts). In: Dowson D, editor. Lubrication at the frontier, proceedings of the

021102.

25th Leeds–Lyon symposium on tribology. Tribology series, vol. 36. Else-

[52] Boucly V, Nélias D, Liu SB, Wang QJ, Keer LM. Contact analyses for bodies

vier; 1999. p. 673–82.

with frictional heating and plastic behavior. ASME J Tribol 2005;127(2):

[17] Gladwell GML. Contact problems in the classical theory of elasticity. Alphen

355–64.

aan den Rijn: Sjthoff and Noordhoff; 1980.

[53] Armstrong-Hlouvry B, DuPont P, Canudas de Wit C. A survey of models,

[18] Johnson KL. The effect of a tangential contact force upon the rolling motion

analysis tools and compensation methods for the control of machines with

of an elastic sphere on a plane. ASME J Appl Mech 1958;25:339–46.

[19] Barber JR, Ciavarella M. Contact mechanics. Int J Solids Struct 2000;37: friction. Automatica 1994;30(7):1093–138.

29–43. [54] Oden JT, Martins JAC. Models and computational methods for dynamic

[20] Greenwood JA, Williamson JBP. Contact of nominally ﬂat surfaces. Proc R friction phenomena. Comput Methods Appl Mech Eng 1985;52:527–634.

Soc London, Ser A 1966;295:300–19. [55] Zienkiewicz OC, Taylor RL. The ﬁnite element method in engineering

[21] Ciavarella M, Delﬁne V, Demelio V. A new 2d model with interaction for science. London: McGraw-Hill; 2000.

studying the contact of multiscale rough random proﬁles. Wear 2006;261: [56] Ham GT, Rubin CA, Hahn GL, Bharghava GV. Elasto-plastic ﬁnite element

556–67. analysis of repeated, two-dimensional rolling sliding contacts. ASME J

[22] Ciavarella M, Demelio G. Elastic multiscale contact of rough surfaces: Tribol 1988;110:44–9.

Archard’s model revisited and comparisons with modern fractal models. [57] Ouyang H, Nack W, Yuan Y, Chen F. Numerical analysis of automotive disc

ASME J Appl Mech 2001;68:496–8. brake squeal: a review. Int J Vehicle Noise 2005;1:207–31.

[23] Karpenko YA, Akay A. A numerical model of friction between rough [58] Massi F, Baillet L, Giannini O, Sestieri A. Brake squeal: linear and nonlinear

surfaces. Tribol Int 2001;34:531–45. numerical approaches. Mech Syst Signal Process 2007;21(6):2374–93.

[24] Pearson JD, Zikry MA, Wahl K. Computational design of thin-ﬁlm nano- [59] Baillet L, Sassi T. Simulations numériques de différentes méthodes d’élé-

composite coatings for optimized stress and velocity accommodation ments ﬁnis pour les proble mes de contact avec frottement. C R Mecanique

response. Wear 2009;267:1137–45. 2003;331(11):789–96.

M. Renouf et al. / Tribology International 44 (2011) 834–844 843

[60] Baillet L, Linck V, D’Errico S, Laulagnet B, Berthier Y. Finite element [95] Cundall PA. A computer model for simulating progressive large scale

simulation of dynamic instabilities in frictional sliding contact. ASME J movements of blocky rock systems. In: Proceedings of the symposium of

Tribol 2005;127(July):652–7. the international society of rock mechanics, vol. 1; 1971. p. 132–50.

[61] Carpenter NJ, Taylor RL, Katona MG. Lagrange constraints for transient ﬁnite [96] Cundall PA, Strack ODL. A discrete numerical model for granular assemblies.

element surface contact. Int J Numer Methods Eng 1991;32:103–28. Géotechnique 1979;29(1):47–65.

[62] Wriggers P. Finite element algorithms for contact problems. Arch Comput [97] Renouf M, Dubois F, Alart P. Numerical investigations of fault propagation

Methods Eng 1995;2:1–49. and forced-fold using a non smooth discrete element method. Rev Euro

[63] Chabrand P, Dubois F, Raous M. Various numerical methods for Method Num 2006;15:549–70.

solving unilateral contact problems with friction. Math Comput Modell [98] Saussine G, Cholet C, Dubois F, Bohatier C, Gautier PE, Moreau JJ. Modell-

1999;28(4–8):97–108. ing ballast behaviour under dynamic loading. Part 1: a 2d polygonal

[64] Jean M. The non smooth contact dynamics method. Compt Methods Appl discrete element method approach. Comput Methods Appl Mech Eng

Math Eng 1999;177:235–57. 2006;195(19–22):2841–59.

[65] Klarbring A. A mathematical programming approach to three-dimensional [99] Acary V, Jean M. Numerical modeling of three dimensional divided

contact problems with friction. Comput Methods Appl Mech Eng 1986;58: structures by the non smooth contact dynamics method: application to

175–200. masonry structure. In: Topping BHV, editor. The ﬁfth international con-

[66] Basseville S. Dynamique et stabilité d’un syste me discret en présence de ference on computational structures technology, 6–8 September. Civil-

contact et de frottement. PhD thesis, Université d’Aix-Marseille II; 2004. Comp Press; 2000. p. 211–22.

[67] Alart P. Méthode de newton généralise en mécanique du contact. J Math [100] Lubrecht AA, Berthier Y. Granular lubrication; a simple model and trends.

Pures Appl 1997;76:83–108. Tribol Ser 1995;30:53–62.

[68] Jourdan F, Alart P, Jean M. A Gauss–Seidel like algorithm to solve frictional [101] Kabir A, Jasti VK, Higgs III CF, Lovell M. An evaluation of the explicit ﬁnite

contact problem. Comp Meth Appl Mech Eng 1998;155:31–47. element method approach for modelling dense ﬂows of discrete grains in a

[69] Renouf M, Alart P. Conjugate gradient type algorithms for frictional multi- coquette shear cell. I Mech E 2008;222:715–23.

contact problems: applications to granular materials. Comput Methods [102] Iordanoff I, Se ve B, Berthier Y. Solid third body analysis using a discrete

Appl Mech Eng 2004;194(18–20):2019–41. approach: inﬂuence of adhesion and particle size on the macroscopic

[70] Busquet M, Baillet L, Bordreuil C, Berthier Y. 3D ﬁnite element investigation behavior of the contact. ASME J Tribol 2002;124:530–8.

on the plastic ﬂows of rolling contacts—correlation with railhead micro- [103] Lane JE, Spurling TH, Freasier B, Perram JW, Smith ER. Inconsistency of the

structural observations. Wear 2005;258(7–8):1071–80. density-functional theory of adsorption when using computer simulations.

[71] Peillex G, Baillet L, Berthier Y. Dynamic behavior of a homogenized Phys Rev A 1979;20-5:2147–53.

composite under contact with friction loading. C R Mecanique 2007;335: [104] Magda JJ, Tirrell M, Davis HT. Molecular dynamics of narrow, liquid-ﬁlled

394–400. pores. J Chem Phys 1985;63-4:1886–901.

[72] Linck V, Baillet L, Berthier Y. Modeling the consequences of local kinematics [105] Elrod HG, Brewe DE. Numerical experiments with ﬂows of elongated

of the ﬁrst body on friction and on the third body sources in wear. Wear granules. Tribol Ser 1991;21:219–26.

2003;255:299–308. [106] Lubrecht AA, Chan-Tien C, Berthier Y. A simple model for granular lubrica-

tion inﬂuence of boundaries. In: Dowson D, editor. The third body concept.

[73] Jourdan F. Numerical wear modeling in dynamics and large strains:

Tribology series, vol. 31; 1995.

application to knee joint prostheses. Wear 2006;261(3–4):283–92.

[107] Ghaouti A, Chaze M, Dubujet P, Sidoroff F. Particulate and granular

[74] Massi F, Berthier Y, Baillet L. Contact surface topography and system

simulation of the third body behaviour. Tribol Ser 1996;31:355–65.

dynamics of brake squeal. Wear 2008;265(11–12):1784–92.

[108] Johnson KL, Kendall K, Roberts AD. Surface energy and the contact of elastic

[75] Dang Van K, Maitournam MH. On some recent trends in modelling of

solids. Proc R Soc London A 1971;324:301–13.

contact fatigue and wear in rail. Wear 2002;253(1–2):219–27.

[109] Seve B, Iordanoff I, Berthier Y. A discrete solid third body model: inﬂuence

[76] Baietto MC, Rannou J, Gravouil A, Pelletier H, Gauthier C, Schirrer R. 3D

of the intergranular forces on the macroscopic behaviour. Tribol Interface

crack network analysis during a scratch test of a polymer: a combined

Eng Ser 2001;39:361–8.

experimental and multigrid X-fem based numerical approach. Tribology

[110] Fillot N, Iordanoff I, Berthier Y. A granular dynamic model for the degrada-

International, doi:10.1016/j.triboint.2010.04.014.

tion of material. ASME J Tribol 2004;126(3):606–14.

[77] Ben-Zion Y. Dynamic ruptures in recent models of earthquake faults. J Mech

[111] Fillot N, Iordanoff I, Berthier Y. Simulation of wear through a mass balance

Phys Solids 2001;49:2209–44.

in a dry contact. ASME J Tribol 2005;127(1):230–7.

[78] Adams GG. An intersonic slip pulse at a frictional interface between

[112] Renouf M, Saulot A, Berthier Y. Third body ﬂow during a wheel–rail

dissimilar materials. J Appl Mech 2001;68:81–6.

interaction. In: Mota Soares CA, editor. III European conference on

[79] Di Bartolomeo M, Meziane A, Massi F, Baillet L, Fregolent A. Dynamic

computational mechanics solids; 2006.

rupture at a frictional interface between dissimilar materials with aspe-

[113] Kabir A, Lovell M, Higgs III CF. Utilizing the explicit ﬁnite element method

rities. Tribol Int 2010;43(9):1620–30.

for studying granular ﬂows. Tribol Lett 2009;29(2):85–94.

[80] Zhaojun B, Ren-Cang L. Substructuring, dimension reduction and applica- [114] Raje N, Sadeghi F, Rateick Jr. RG, Hoeprich MR. Evaluation of stresses around

tions: an introduction. In: Dongarra J, Madsen K, Wasniewski J, editors. inclusions in hertzian contacts using the discrete element method. ASME J

Applied parallel computing. Lecture notes in computer science, vol. 3732. Tribol 2007;129:283–91.

Berlin, Heidelberg: Springer; 2006. p. 266–6. [115] Doucet J-P, Weber J. Computer-aided molecular design: theory and applica-

[81] Magoule s F. Substructuring techniques and domain decomposition meth- tions. Elsevier; 1996.

ods. Kippen: Saxe-Coburg Publications; 2010. [116] Allen MP, Tildesley DJ. Computer simulations of liquids. Oxford: Clarendon

[82] Descartes S, Berthier Y. Rheology and ﬂows of solid third bodies: back- Press; 1987.

ground and application to an mos1.6 coating. Wear 2002;252(7–8):546–56. [117] Tomlinson GA. A molecular theory of friction. Phil Mag Ser 1929;7:905–39.

[83] Berthier Y, Descartes S, Busquet M, Niccolini E, Desrayaud C, Baillet L, [118] Frenkel YI, Kontorova T. On the theory of plastic deformation and twinning.

Baietto-Dubourg MC. The role and effects of the third body in the wheel– Zh Eksp Teor Fiz 1938;8:1340.

rail interaction. Fatigue Fract Eng Mater Struct 2004;27:423–36. [119] Sorensen MR, Jacobsen KW, Stoltze P. Simulations of atomic-scale sliding

[84] Haff PK. Grain ﬂow as a ﬂuid-mechanical phenomenon. J Fluid Mech friction. Phys Rev B 1996;53:2101–13.

1983;134:401–30. [120] Martini A, Dong Y, Perez D, Voter AF. Low-speed atomistic simulation of

[85] Lun CKK. Kinetic theory for granular ﬂow of dense slightly inelastic, slightly stick–slip friction using parallel replica dynamics. Tribol Lett 2009;36:

rough spheres. J Fluid Mech 1991;233:539–59. 63–8.

[86] Yu C-M, Craig K, Tichy JA. Granular collision lubrication. J Rheol [121] Landman U, Luedtke WD. Nanomechanics and dynamics of tip-substrate

1994;38(4):921–36. interactions. J Vac Sci Technol B 1991;9:414–23.

[87] Hou K, Kalousek J, Magel E. Rheological model of solid layer in rolling [122] Harrison JA, Stuart SJ, Brenner DW. Atomic scale simulation of tribological

contact. Wear 1997;211:134–40. and related phenomena. In: Handbook of micro/nanotribology. CRC Press;

[88] Zhou L, Khonsari MM. Flow characteristics of a powder lubricant sheared 1999. p. 525.

between parallel plates. ASME J Tribol 2000;122(1):147–55. [123] Harrison JA, White CT, Colton RJ, Brenner DW. Nanoscale investigation of

[89] Sawyer WG, Tichy JA. Lubrication with granular ﬂow: continuum theory, indentation, adhesion and fracture of diamond (111) surfaces. Surf Sci

particle simulations, comparison with experiment. ASME J Tribol 2001;123: 1992;271:57–67.

777–84. [124] Bowden FP, Tabor D. The friction and lubrication of solids. 2nd corrected ed.

[90] Wornyoh EYA, Jasti VK, Higgs III CF. A review of dry particulate lubrication: Oxford University Press, Clarendon Press; 1986. [oxford classic text in the

powder and granular materials. ASME J Tribol 2007;129:438–49. physical sciences edition].

[91] Hesmat H. The rheology and hydrodynamics of dry powder lubrication. [125] He G, Müser MH, Robbins MO. Adsorbed layers and the origin of static

Tribol Trans 1991;34(3):433–9. friction. Science 1999;284:1650–2.

[92] Hesmat H, Godet M, Berthier Y. On the role and mechanism of dry [126] Hirano M, Shinjo K. Atomistic locking and friction. Phys Rev B 1990;41:

triboparticulate lubrication. Lubric Eng 1995;51:557–64. 11837–51.

[93] Higgs III CF, Tichy JA. Granular ﬂow lubrication: continuum modeling of [127] Hirano M, Shinjo K. Superlubricity and frictional anisotropy. Wear

shear behaviour. ASME J Tribol 2004;126:499–509. 1993;168:121–5.

[94] Higgs III CF, Tichy JA. Effect of particle and surface properties on granular [128] Harrison JA, Brenner DW. Simulated tribochemistry: an atomic scale view of

lubrication ﬂow. I Mech E 2008;222(703–713). the wear of diamond. J Am Chem Soc 1994;116:10399–402.

844 M. Renouf et al. / Tribology International 44 (2011) 834–844

[129] Robbins MO, Müser MH. Computer simulations of friction, lubrication, and [138] Sfantos GK, Aliabadi MH. Multi-scale boundary element modelling of

wear. In: Modern tribology handbook. CRC Press LLC; 2001. p. 717–65. material degradation and fracture. Comput Methods Appl Mech Eng

[130] Sham TL, Tichy JA. A scheme for hybrid molecular dynamics/ﬁnite element 2007;196:1310–29.

analysis of thin ﬁlm lubrication. Wear 1997;207:100–6. [139] Maugin GA. The thermomechanics of nonlinear irreversible behaviors.

[131] Massi F, Rocchi J, Culla A, Berthier Y. Coupling system dynamics and contact World scientiﬁc series on nonlinear science, series A, vol. 27. World

behaviour: modelling bearings subjected to environmental induced vibrations Scientiﬁc; 1999.

and ‘false brinelling’ degradation. Mech Syst Signal Process 2010;24(4):1068–80. [140] Renouf M, Fillot N. Coupling electrical and mechanical effects in discrete

[132] Bucher F, Dmitriev AI, Ertz M, Knothe K, Popov VL, Psakhie SG, Shilko EV. element simulations. Int J Numer Methods Eng 2008;74:238–54.

Multiscale simulation of dry friction in wheel/rail contact. Wear [141] Richard D, Iordanoff I, Berthier Y, Renouf M, Fillot N. Friction coefﬁcient

2006;261:874–84. as a macroscopic view of local dissipation. ASME J Tribol 2007;129(4):

[133] Temizer I, Wriggers P. A multiscale contact homogenization technique for 031404.

the modeling of third bodies in the contact interface. Comput Methods Appl [142] Richard D, Iordanoff I, Renouf M, Berthier Y. Thermal study of the dry sliding

Mech Eng 2008;198:377–96. contact with third-body presence. ASME J Tribol 2008;130(3):031404.

[134] Berthier Y, Cao H-P, Nhu V-H, Renouf M, Saulot A. Couplage et dialogue [143] Renouf M, Berthier Y. Modélisation numérique discréte des matériaux

entre méthode discre te et continue pour la tribologie numérique. In: Actes granulaires, Modélisation numérique de la génération et transmission de

du 9e Colloque National en Calcul des Structures, vol. 1; 2009. p. 155–60. la chaleur, Lavoisier; 2010. p. 381–400.

[135] Saulot A, Renouf M, Berthier Y. Fem-dem dialogue for tribological under- [144] Blok HA. Theoretical study of temperature rise at surface at actual contact

standing. In: Proceedings of the ASME/STLE international joint tribology under oilness lubricating conditions. I Mech E 1937;2:222–35.

conference; 2008. [145] Champelovier T, Renouf M. Mode les multiphysiques par éléments discrets -

[136] Anciaux G, Molinari JF. Contact mechanics at the nanoscale, a 3d multiscale application a la modélisation du troisie me corps. In: Actes du 9e Colloque

approach. Int J Numer Meth Eng 2009;79:1041–67. National en Calcul des Structures, vol. 2; 2009. p. 665–70.

[137] Luan BQ, Hyun S, Molinari JF, Bernstein N, Robbins MO. Multiscale modeling [146] Renouf M, Cao H-P, Nhu V-H. Multiphysical modeling of third-body

of two-dimensional contacts. Phys Rev E 2006;74:046710. rheology. Tribol Int 2011;44(4):417–25.

- A Study on the Tribological Characteristics of a Magneto-Rheological Elasteomer(Journal of Tribology)-LibreЗагружено:Anish Pai
- ASTM G133 - 2005Загружено:Denise Roth
- Mae 493n 593t Lec1Загружено:kostas.sierros9374
- About FrictionЗагружено:Rian Steve
- 9M303.22Загружено:Manoj
- Friction Coefficient in Hot Compression of Cylindrical SpecimensЗагружено:Ch V S Raju
- 9. What if i Never Change the Oil in My CarЗагружено:Velluto Khod
- HOW to Design a Pump SystemЗагружено:Pierre Norris
- 2-01_KimЗагружено:Jigar M. Upadhyay
- 130_jul_92Загружено:RigobertoGuerra
- Course OutlineЗагружено:Parth Jaggi
- Abrasive and Diffusive Tool Wear Fem SimulationЗагружено:Oakme
- 1-s2.0-S0301679X10000162-mainЗагружено:Joel E Valencia
- Groz Beckert Needle Oil DetailЗагружено:adali2020
- Bff2821 TestЗагружено:Aidil Shafiza
- (Tribology Series 44) Gwidon W. Stachowiak_ Andrew W. Batchelor and Grazyna B. Stachowiak (Eds.)-Experimental Methods in Tribology-Academic Press_ Elsevier (2004).pdfЗагружено:er_manojpanchal
- 06075676Загружено:mani manis
- REЗагружено:hameedirfan
- Physics D DCP CE Lab- FrictionЗагружено:Pappesh Patil
- Chapter 5Загружено:rasprasad
- Erdemir 1 - Tribological and Solid Lubrificant Coatings.pdfЗагружено:MÔNICA SUEDE S. SILVA
- ASTM D3702-94 Thrus iugougЗагружено:Taufik Hidayatullah
- Synthetics Lubricant Basestock Brochure - ExxonMobilЗагружено:Rafael Nakazato Recio
- Technical Information Sheet Bushings Lubrication ENB 04 0653Загружено:Duniwanhi Suchiang
- Measuring Wear and Friction Using TribometerЗагружено:Priyank Sharma
- Effect of Nanosilica Filled Polyurethane CompositeЗагружено:Huy Heineken
- lom-SP_001.docЗагружено:Siddharth Jain
- 3300008Загружено:chandu
- Investigation of Friction Mechanisms in Finger Pad Sliding Against SurfacesЗагружено:suyash gairola
- Caracterização Abrasion Wear Characterization of NaturaЗагружено:Juliano Tessinari Zagôto

- Permeabilidade de Leitos 1.pdfЗагружено:MÔNICA SUEDE S. SILVA
- Climate ChangeЗагружено:MÔNICA SUEDE S. SILVA
- Electrode PerformanceЗагружено:MÔNICA SUEDE S. SILVA
- Elkem ElectrodeЗагружено:MÔNICA SUEDE S. SILVA
- Smelting ProcessЗагружено:MÔNICA SUEDE S. SILVA
- Extractive Metallurgy of NickelЗагружено:MÔNICA SUEDE S. SILVA
- Mecanismos de DesgasteЗагружено:MÔNICA SUEDE S. SILVA
- Edital UESB 2016Загружено:MÔNICA SUEDE S. SILVA
- Powder MetallurgyЗагружено:MÔNICA SUEDE S. SILVA
- 671-VonKrugerЗагружено:Bagus Tri Ardaya Ardaya
- electric arc furnaceЗагружено:Vijay Choudhary
- Blast Furnace & Sintering March 2016Загружено:MÔNICA SUEDE S. SILVA
- Modelling of Tribofilm Sliding BehaviourЗагружено:MÔNICA SUEDE S. SILVA
- Erdemir 1 - Tribological and Solid Lubrificant Coatings.pdfЗагружено:MÔNICA SUEDE S. SILVA
- 633776345920000000_1Загружено:Azrul Zafrie
- Effect of Lubricants on Warm Compaction Process of Cu-based CompositeЗагружено:MÔNICA SUEDE S. SILVA
- Microstructure and Wear of MaterialsЗагружено:MÔNICA SUEDE S. SILVA
- Heat Generation Modeling of Ball Bearing Based on Internal Load DistributionЗагружено:MÔNICA SUEDE S. SILVA
- CFD Analysis_Hydrodynamic Lubrification and Bingham LubrificantЗагружено:MÔNICA SUEDE S. SILVA
- Automotive tribology overview of current advances and challenges for the future.pdfЗагружено:MÔNICA SUEDE S. SILVA
- 1-s2.0-S0261306913002379-mainЗагружено:krishna
- Rapid SinteringЗагружено:MÔNICA SUEDE S. SILVA
- C-DICЗагружено:MÔNICA SUEDE S. SILVA
- Ultra-high Cycle Fatigue Behaviour of Warm Compaction Fe_Cu_Ni_Mo_CЗагружено:MÔNICA SUEDE S. SILVA
- Interface Analysis and Wear Behavior of Ni Particulate ReinforcedЗагружено:MÔNICA SUEDE S. SILVA
- CorrosionAndHeatResistantNickelAlloys_10086Загружено:FILO74

- Intra PreneurЗагружено:Roonak
- Global Marketing-products n Services for ConsumersЗагружено:Reyad
- Lattice Data Structure - ThesisЗагружено:anant_nimkar9243
- 10th Grade Literature Syllabus[1]Загружено:David Velez Gonzalez
- Lesson Plan(1)(1)Загружено:Yehya El Arabi
- Historical Introductions to the Lutherian ConfessionsЗагружено:3am
- Convergence Chart-ElectricityЗагружено:kmotyka1
- (Guitar Lesson) Building Solos Using the Essential Jazz Lines Series (Extracto)Загружено:OrsoBalu
- HolySpiritWind Prophetic Words, Dreams and Visions.pdfЗагружено:OBEY GOD
- Japanese Adverbs.pdfЗагружено:User_07
- Hunt_1997_Moral Panic and Moral Language in the MediaЗагружено:Danijela Vojnovic
- 4Загружено:Akram Khan
- Lamberg Karlowski 2002 Archaeology and Language DiskusijaЗагружено:Svarogns
- africon04_whsЗагружено:Mostafa Annaka
- Assignment 2.1Загружено:Deborah Muhammad
- [Tigli Et Al. 94] Use of Multi Agent Systems for Mibile Robotics ControlЗагружено:Pedro Laureles
- (Wit+Transactions+on+Engineering+Sciences)+C.+A.+(Editors)+Brebbia,+D.+De+Wrachien-Monitoring,+Simulation,+Prevention+and+Remediation+of+Dense+and+Debris+Flow+III+(Wit+Transactions+on+Engineering+ScieЗагружено:André Toniati
- Brigid Daniel Resilience 2003Загружено:Naomi Hender
- Textbook2 TalebЗагружено:Dom DeSicilia
- Ubd Lesson PlansЗагружено:Ryan VanBogaert
- Lectures on Classical Mechanics-1Загружено:Abhijit Kar Gupta
- Dan Graham. Sound Is MaterialЗагружено:limboboy
- On BrentanoЗагружено:Anonymous LnTsz7cp
- Project listЗагружено:Arul Jothi
- Li Cent AaaaaaaЗагружено:Irina Dragut
- Louis XIV&GildingThesis 170598Загружено:TrickyDicky2
- Film & Psychoanalysis BibliographyЗагружено:nighb
- Efectos en la percepción del clima motivacional generado por los entrenadores y compañeros sobre la.pdfЗагружено:Jaime Cortez
- RTL Design 3year Exp M.tech VLSI Resume Akhil v MenonЗагружено:Akhil Menon
- Neural Network Radial Basis Function classifier for earthquake data using aFOA .Загружено:IJAR Journal

## Гораздо больше, чем просто документы.

Откройте для себя все, что может предложить Scribd, включая книги и аудиокниги от крупных издательств.

Отменить можно в любой момент.