Page 243 - Mechanics of Asphalt Microstructure and Micromechanics
P. 243

Models for  Asphalt Concrete   235


                 Where W i  = dissipated energy per cycle; W n  = dissipated energy at cycle n; and
               n
              ∑ W  = total sum of dissipated energy up to cycle n. Then fatigue life, based on the DER
                  i
              i=1
              criterion, can be represented as:
                                            N =  K (  1  ) K 1                  (7-100)
                                              f   2  W
                                                     i
                 Where K 1  and K 2  are the slope and the Y intercept, respectively, of the fitted fatigue
              curves (W i versus N p ) for a given asphalt binder. The slope and the Y intercept from fa-
              tigue curves obtained from different types of fatigue testing were used in the equation
              to determine the number of cycles to failure (N f ).
                 The major limitations for all the fatigue models include the criteria for fatigue life;
              the testing boundary conditions; and the lack of fundamental mechanism at micro-
              scopic levels.


        7.11  Cohesive Zone Models for Numerical Simulations

              7.11.1 General Ideas
              The idea to describe fracture as a material separation across a surface was started by
              Barenblatt (1962). Generally, it is called a cohesive zone model. The cohesive zone is a
              surface in a bulk material where displacement discontinuities occur. Continuum is ex-
              tended with discontinuities in the form of displacement jump which requires addi-
              tional constitutive description. Equations relating normal and tangential displacement
              jumps across the cohesive surfaces with the proper tractions define a specific cohesive
              zone model.
                 Based on the elementary functions used in cohesive zone models, they can be clas-
              sified as (1) multilinear, (2) polynomial, (3) trigonometric, and (4) exponential. Almost
              all these models are constructed as follows: tractions increase, reach a maximum, and
              then approach zero with increasing separation. Needleman (1987) introduced the cohe-
              sive zone models in computational practice. Since then, cohesive zone models are used
              widely in finite element simulations of crack tip plasticity and creep; crazing in poly-
              mers; adhesively bonded joints; interface cracks in bimaterials; delamination in com-
              posites and multilayered materials; fast crack propagation in polymers, and so on.
                 Four typical cohesive zone models are briefly introduced here. They present rela-
              tionships between surface tractions T and displacement jumps Δ across the cohesive
              zone. Maximum surface traction T max  is the cohesive strength. The corresponding dis-
              placement jump is Δ max . Dimensionless parameters used are:

                                              σ =  T                            (7-101)
                                                  T
                                                   max
                                                   Δ
                                              δ =  Δ                            (7-102)
                                                   max
                 The work of separation is:
                                              J =  ∫ TdΔ                        (7-103)

                 Or dimensionless:
                                             φ =   J                            (7-104)
                                                T  Δ
                                                 max  max
   238   239   240   241   242   243   244   245   246   247   248