ICF10B

      ⋅  =      = β σ β α YS SS E CTOA J T J J exp exp 0 0 0 (3), where J0 is the initial value of the J integral at crack initiation, T0 is the tearing modulus, α ≈ 1, β = 5.1 from the mechanics description, E and σys are modulus and yield strength. With 0.01 radians average value of the CTOA, a modulus of 120 GPa, a yield strength of 1 GPa and β = 5.1, one finds: { } SS 0 0 J J exp 0.23 1.27J = ≈ (4), which means that during slow crack growth the strain energy release rate has barely increased for this 120 nm thick copper film. This again agrees with the actual measured value of ~0.9 J/m2 for the strain energy release rate. CARBON CONTAMINATION SOURCE There are three possible sources for carbon: adhesive tape, the diamond indenter and hydrocarbons from the atmosphere. In our previous studies the first two sources were eliminated [13-15], and hydrocarbons from the atmosphere were proposed as a source of fiducial mark formation. Figure 3. SEM micrograph and corresponding Carbon Auger map of a GaAs fracture surface upon TiWXNY film removal. During the course of this study it was found that a similar type of contamination is present in a different film system of TiXWNY on GaAs, where the telephone cord delaminations formed due to the high residual stress relief (Figure 3). The carbon traces noted both on the film and substrate surfaces mimic the original telephone cord delamination pattern. Fiducial crack arrest marks are like those observed in the Cu/SiO2 system. Figure 3 also shows a carbon Auger map, where brighter areas correspond to

RkJQdWJsaXNoZXIy MjM0NDE=