C.H.J. Davies

Growth of nuclei in a cellular automaton simulation of recrystallisation

Cellular automata (CA) have been extensively used to simulate solidification processes, but have received limited attention in the simulation of solid-state transformations. The attractiveness of CA simulation is that it captures the elegance of Monte Carlo simulations, but combined this with the practicality of models based on empirical measurements. The simulation of recrystallization by the CA method is demonstrated by Hesselbarth and Gobel, and Pezzee and Dunand. The limiting features of these works are that grain structure is not tied to any real microstructure (i.e., no scale is ascribed to the microstructures), and, more importantly, the recrystallization kinetics is not calibrated to real time. Such deficiencies necessarily prevent the use of the technique for quantitative simulation which can be applied to industrial processes. The primary aim of this paper is to address the second of the two deficiencies. The kinetics of a three dimensional CA simulation is related to real time using two growth models. Current growth algorithms used in non-CA models are derived directly from experimental data: Juul Jensen notes that before applying such a model, exhaustive experimental investigations are required. Whilst not disputing the capabilities of such an approach, in this paper this recursive method is compared tomorexa0» an a priori calculation of a growth velocity. The results of the simulations are compared to each other, and to the experimental data and analytical solution of Vandermeer and Rath. The relative merits of the two growth models are discussed.«xa0less

The prediction of the hot strength in steels with an integrated phenomenological and artificial neural network model

Abstract The hot torsion data of a commercial 304 stainless steel has been analysed with an integrated phenomenological-artificial neural network model (IPANN), developed from the Estrin–Mecking (EM) phenomenological model and a back-propagation artificial neural network (ANN) model. In order to predict the flow stress in this model, the work-hardening coefficient and its product with the stress were used as inputs, along with strain, temperature and strain rate. The Pearson correlation coefficient was used to evaluate the performance and terminate the simulation of the IPANN model, whilst the standard errors were employed to quantitatively compare the accuracy of different models. The IPANN model is able to predict the distribution of flow stress more accurately in the work-hardening and dynamic recrystallisation regimes in comparison with the original EM and ANN models. The training speed is significantly improved and the test of the model is satisfactory, if reasonable training data is provided. In addition, by using the phenomenological model as training data, the IPANN model may be used for extrapolation.

The cellular automaton simulation of static recrystallization in cold-rolled AA1050

As the latitude for property optimization in conventional engineering alloys diminishes, models of microstructural evolution have become more sophisticated in attempts to more accurately reproduce the behavior of a metal during processing. Of particular importance for metal processing is the effective modelling of static recrystallization; this paper reports the implementation of one of several parallel methodologies for simulating recrystallization: the cellular automaton technique. The cellular automaton simulation described in this paper combines the attractive probabilistic nature of the Monte Carlo method with the practical advantage associated with the geometrical approach of being readily calibrated to experiments. The study shows that recrystallization behavior can be simulated with reasonable accuracy without having to explicitly encode the form of the recrystallization kinetics, and builds on previous work by incorporating crystallographic texture into the simulation. Simulated grain size distributions and means were found to be comparable to experimental values for each of three texture classes, as well as for the combination of the three classes, although some anomalies were observed. The kinetics of the simulation were in reasonable agreement with the experimentally-determined kinetics.

Constitutive behaviour of composites of AA6061 and alumina

Abstract Metal-matrix composites are being used increasingly for the fabrication of engineering components, yet the response of such materials to deformation is still poorly characterised. In an attempt to further the optimisation of processing, the constitutive behaviour of the composites 6061/Al 2 O 3 /10p and 6061/Al 2 O 3 /20p has been investigated; for comparison purposes, the unreinforced alloy was tested also. Compression testing was carried out at strain rates of between 0.01 and 10 s −1 , at temperatures of between 300 and 550°C. The resulting data were fitted to the hyperbolic sine relationship and an internal stress relationship. It was found that the constitutive behaviour of unreinforced alloy could be characterised adequately by a single relationship over the whole temperature range, whereas the behaviour of the composites was best described in two temperature ranges, above and below about 425°C. The hyperbolic sine relationship was found to be the more accurate in each case; however, for extrapolating to lower temperatures, the internal stress relationship was found to be more reliable for the composites. The change in constitutive behaviour at 425°C is related to the controlling mechanism in the two temperature regimes: precipitation of β′ at low temperatures, and dynamic recrystallisation at high temperatures. This observation may need to be accounted for in the commercial production of metal-matrix composites.

Mechanical, microstructure and texture properties of interstitial-free steel and copper subjected to equal channel angular extrusion and cold-rolling

Interstitial-free steel and OFHC copper were subjected to 8 passes, route BC room temperature ECAE followed by cold-rolling up to 97.5% thickness reduction. Uniaxial tensile tests and Electron Back-Scattering Diffraction were used to characterise the evolution in mechanical properties, microstructure refinement and micro-texture. IF-steel showed continuous increase in strength whereas Cu returned reduced strength and a small gain in ductility at 97.5% reduction. In both metals substructure refinement was accompanied by an increase in high-angle boundary fraction, average misorientation and a slight increase in Σ3 boundaries. An evolution of crystallographic orientations from negative shear to predominantly cold-rolled textures after 95% and 97.5% reduction was observed in both metals.

Rationale for composition-specific ageing schedules for particle-reinforced metal-matrix composites

Abstract Optimised age-hardening treatments are vital if aluminium-based particle-reinforced metal-matrix composites (PRMMCs) are to achieve widespread use as an engineering material. This paper shows that for some PRMMCs ageing may not be optimised, and suggests why this is the case. Composites of a 7xxx-series aluminium alloy containing a number of volume fractions of SiC were investigated. On ageing, the materials were found to fall into two groups: (i) the 10% SiC composite, which followed an ageing sequence similar to the unreinforced alloy, and (ii) those containing greater than 15% SiC, which followed a different sequence. The implication of this work is that substantially different ageing schedules need to be devised for the two different groups of materials if the full potential of age-hardenable composites is to be realised in the higher volume-fraction materials.

Effect of Processing Parameters on the Magnetic Properties and Macrotexture of a Nd 13.5 Fe 73.8 Co 6.7 B 5.6 Ga 0.4 Alloy Processed by Equal Channel Angular Pressing With Back Pressure

Equal channel angular pressing (ECAP) is a well-established thermo-mechanical processing technique, which could induce the c-axis texture of Nd₂Fe₁₄B in a melt-spun Nd_13.5Fe_73.8Co_6.7B_5.6Ga_0.4 alloy. However, the effects of ECAP processing parameters, such as temperature, back pressure (BP), and multiple-pass ECAP routes, remain unknown for this alloy. In this paper, we have investigated the effects of these processing parameters on the c-axis texture formation. It is found by X-ray diffraction macrotexture analysis that the maximum intensity of (001) pole figures for the tetragonal-Nd₂Fe₁₄B phase (I_max) shows an increase from 2.7 to 4.1 m.r.d. (multiples of random distribution) by increasing the ECAP temperature from 723 to 823 K, while the difference in remanent magnetization between easy and hard directions (AMr) rises from 24.0 to 41.5 Am²/kg. When the BP was increased from 0.25 to 0.5 GPa at 823 K, I_max showed an increase from 2.8 to 4.1 m.r.d. However, I_max saturated for BPs above 0.5 GPa, suggesting that BP has limited effect on the texture formation, although it is necessary for the compaction of the alloy powders. Two multiple-pass ECAP routes conventionally known as routes A and C were employed for two-pass ECAP at 823 K. It is found that route A processing is effective in enhancing the texture formation, while the texture is lost by a subsequent pressing when adopting route C. Therefore, the compaction of Nd_13.5Fe_73.8Co_6.7B_5.6Ga_0.4 alloy powder using route A ECAP passes with 0.5 GPa BP at 823 K results in pronounced texture, which is beneficial for anisotropic hard magnetic properties.