Materials Science Forum Vols. 575-578

Abstract: Mullite interlayer of the carbon fiber reinforced Si-C-N matrix composite (C/Si-C-N) was fabricated by liquid precursor impregnation and pyrolysis (PIP). The mixture of aluminium secbutoxide (A1OBu) and tetraethoxysilicate (TEOS) were used as starting materials. The monophase polymer gel used as mullite precursor could be formed from the mixture in the semi-closed air. A mullite interlayer with a thickness about 500nm is obtained at 850°C- 1200°C. The fiber/interlayer bonds are tight. The carbon fibers are not damaged in the fabrication procedure.

Abstract: A coupled thermo-mechanical model is employed to analyze the thermo-mechanical behavior of a widely used laminated composite subject to temperature decrease at service conditions. Three sets of governing equations, i.e. heat transfer, thermo-mechanical deformation and damage evolution are respectively described in the model. These equations are then assembled into a coupled matrix equation using finite element formulation and then solved simultaneously at each time interval. A numerical model of two layered composites with some preexisting equal-spacing cracks along the interface in the lower layer is set up to investigate the thermal induced crack propagation due to temperature decrease. Results are presented in the form of crack propagation process in stress profiles and discussed. Numerical simulations show that the crack propagation behavior of the composites is closely dependent on the physico-mechanical properties of two layers and preexisting cracks. It is found that thermal induced cracks penetrate into the upper layer and grow in the upper layer due to the low strength of the upper layer when the model is subject to uniform temperature decrease.

Abstract: Basing on the experimental results of the hardenability investigations, which employed Jominy method, the model of the neural networks was developed and fully verified experimentally. The model makes it possible to obtain Jominy hardenability curves basing on the steel chemical composition. The modified hardenability curves calculation method is presented in the paper, initially developed by Tartaglia, Eldis, and Geissler, later extended by T. Inoue. The method makes use of the similarity of the Jominy curve to the hyperbolic secant function. The empirical formulae proposed by the authors make calculation of the hardenability curve possible basing on the chemical composition of the steel. However, regression coefficients characteristic for the particular steel grade must be known. Replacing some formulae by the neural network models is proposed in the paper.

Abstract: Two mild steels with Nb-microalloying and Nb-free were smelted. After single-pass hot compression of various processes conducted on a Gleeble-2000 thermomechanical simulator, the effects of deformation conditions and Nb-microalloying on microstructural evolution kinetics, including the volume fraction, grain size and grain number per unit area were investigated through quantitative metallographic analysis. The experimental results show that, grain number per unit area of DIF increases with decreasing deformation temperature or increasing deformation amount; the grain size of DIF is not very sensitive to the deformation conditions; the volume fraction of DIF increases due to the increased grain number per unit area. The microalloying element of Nb dissolved in austenite inhibits the grain growth and reduces the volume fraction of DIF.

Abstract: Improved methods to estimate the kinetics of dynamic and static recrystallization are proposed in this paper. The kinetics for dynamic and static recrystallization can be evaluated by inverse analysis of the flow curves obtained using the single-hit and double-hit hot compression tests carried out on Gleeble 3500. The dynamic and static recrystallization volume fractions can be seen as functions of plastic strain and time, respectively. The mathematical formulations between the recrystallization volume fractions and the dislocation density related to the flow stress are used in incremental forms in the study. The methods are applied to the hot compression tests of plain carbon steel and the kinetics of dynamic and static recrystallization are gained successfully for some conditions at elevated temperature. The results are clarified by comparing them with those reported in previous investigations. It is confirmed that the present methods can provide accurate kinetics for dynamic and static recrystallization with shorter time for experiment and computation.

Abstract: Microstructures and toughness of simulated coarse grain heat-affected zone of hot continuously rolled copper-bearing steel were investigated using physical simulation. The results showed that brittlement is easy to happen in the region of CGHAZ with slower thermal cycles (t8/5≥45s). Granular bainite transformed from austenite led to brittlement. The dimensions of granular martensite and austenite (M-A) constituents are main factors influencing the impact toughness. There is no visible effect on the toughness when the dimensions of M-A constituents are less than 1μ m. However, the toughness decreases greatly once the dimensions exceed 1μ m. Therefore, decreasing the dimensions of M-A constituents by controlling weld heat input will do good to improve the impact toughness of copper-bearing steel.

Abstract: This paper presents our effort to reduce thermo-mechanical failures related IC packaging reliability problems. These reliability problems are driven by the mismatch between the different material properties. First of all, finite element analysis are adopted to investigate the influence of encapsulation structures and material properties on the stress distribution in dielectric layers and plastic strain distribution in die for a typical stacked package of IC microstructures. Results show that thinner package body, lower thermal stress and smaller plastic strain can be realized in this micro-structure with the different design geometry parameters and materials properties. Secondly, it was also found that the transition layer of TiN between the die and dielectric layer has a pronounced influence on decreasing the local stress in the passivation layer by comparison analysis. These studies would be beneficial to improve the reliability of stacked IC micro-structure packages.

Abstract: Interfacial delamination is a recognized failure mode in Integrated circuits (ICs). A major cause for this failure is the mismatch of Thermal Expansion Coefficients, Young’s modulus, and Poisson’s ratios of the package materials. Here, the influence of delamination between epoxy and dielectric layers on pattern shift and passivation cracking in IC package under aeronautical conditions, mainly temperature and load cycles, is studied by maximum plastic strain and maximum principal stresses theory using a certain 2D FEM model with different delamination length “L_c.right”. Delaminations are easy to introduce more dangerous impact to the package, because the IC microstructures endure serious thermo-mechanical loading under aeronautical working conditions. The method can be used to find the dangerous designed structure schedules and will provide a basis for selecting passivation materials of aeronautical IC packages.

Abstract: The Kurdjumow -Sachs mechanism of martensite transformation is investigated in detail in this paper. The distortion matrixes of the first shear, the second shear and lineal adjustment according to the K-S mechanism are calculated. From the lattice distortion matrix, we can obtain invariant normal planes. According to the assumption that the habit plane is constructed with some of the invariant normal planes, the macroscopic habit plane (225)A is also discussed. The results show that the K-S mechanism can account for the relief effects, habit plane, orientation relationship and the change of structure.

Abstract: Carbon black (CB) filled rubber is microscopically heterogeneous although homogeneous on a macroscopic scale. CB particles are generally in the form of aggregates, which form the CB network in the rubber matrix. In this work, the junction width between CB aggregates is modeled as a contact resistor and the tunneling conduction mechanism is taken into account, and then an infinite circuit consisting of numerous contact resistors, interconnected with each other, is proposed to simulate the CB network in filled rubber. Prior to determination of the junction width distribution, CB spheres equivalent to CB aggregates in volume is assumed in a specifically random arrangement. Thus, the effect of CB aggregate distribution on the electrical resistivity is discussed. It is found that, for CB (N330) filled natural rubber with volume fraction of 27.5%, the simulated electrical resistivity at a standard deviation of 0.1 mean junction width is in good agreement with the experimental data available in the literature.