Modeling microstructure during heat treatment: simulating local microstructure could provide a tool for optimizing part geometry and casting operations to achieve desired mechanical properties at desired locations.
While casting designers and procurers primarily care about the residual stress condition of the final, machined casting and its properties prior to assembly, those properties change throughout the manufacturing process. During the solution treatment process step, near the solidus temperature, copper and magnesium-rich phases derived from the casting process go into solution again. Subsequent quenching should be controlled so it does not create high residual stresses and to assure copper and magnesium stay in solution. Copper/magnesium is needed in the aging process to modify the microstructure in a controlled way through the precipitation of intermetallic phases.
Advancements in casting process modeling include the ability to simulate local microstructure through casting and heat treatment. Authors Marc Schneider and Chxistof Heisser, Magma Foundry Technologies, Schaumburg, Ill., recently wrote a paper, "Modeling of Microstructure and Mechanical Properties of Aluminum Alloys During the Casting and Heat Treatment Process," based on research conducted to investigate how local microstructure information can be used as input parameters to simulate further process steps, such as lifetime prediction, as well as to determine how to achieve desired mechanical properties within economical boundaries.
Can process simulation, compared with measured data, be used to predict local microstructure through each casting process and the entire heat treatment?
A T6/T7 heat treatment consists of solution treatment, quenching and aging. The changes within the microstructure are solid phase transitions and primarily diffusion-driven processes. The hypoeutectic AISiMg, AISiCu and AISiMgCu alloys create [Mg.sub.2]Si or [Al.sub.2]Cu phases at the end of the solidification process, at which point the phases contain the bulk of the magnesium or copper, depending on the initial melting composition.
The size and distribution of these phases in metalcasting are not optimal for desired mechanical properties, and heat treatment is used to bring the phases into solution again and precipitate them in a desired size and distribution. This leads to the precipitating hardening effect, which results in better mechanical properties.
Solution heat treatment dissolves the [Mg.sub.2]Si or [Al.sub.2]Cu phases and creates a condition where empty spots are present within the crystal structure as much as possible. The higher the empty spot density, the faster the diffusion processes are performed.The diffusion velocity is driven by the local dendrite arm spacing and the fraction of [Mg.sub.2]Si or [Al.sub.2]Cu phases. Starting condition is the phase distribution calculated in the solidification simulation.
The initial concentration profile calculation considers the back diffusion of copper or magnesium during solidification. The concentration at the outer rim of a cell uses values based on the thermodynamic equilibrium. The results of a solidification simulation that started by measuring copperdistribution has shown good correlation with copper distribution at several solution treatment points for smaller secondary dendrite arm spacing.
Quenching creates a state of saturation of the magnesium/copper concentration, as well as the empty spot concentration. The material must be cooled as fast as possible to avoid unwanted precipitations or crystal structure changes. Cooling that is too slow creates precipitants that are neither optimally sized nor in sufficient numbers for the precipitation-hardening process to provide the intended increase in strength.
Previous experiments confirmed peak yield strengths after aging depend on the cooling rate during quenching. The simulation model used in the current research uses data derived from quenching experiments to determine the consumed amounts of magnesium and copper. According to the authors, additional experiments are needed to better determine the consumed amounts, particularly for low cooling rates.
Aging provides a controlled process to generate the desired number and size of precipitants. The over-saturated empty spot concentration results in an acceleration of the process. Similar to the solution treatment, modeling in the current experiment was based on the calculation of magnesium/copper diffusion in solution in the crystal matrix to spherical precipitants, with the radius and the concentration and their subsequent growth. The concentration at both sides of the boundary between the particle and the matrix equate to the thermodynamic equilibrium concentration. The cell's size is determined through the number of inoculation sites and the amount of magnesium or copper available to create precipitants.
When all the magnesium and copper is consumed, precipitations only cluster together, so the total fraction of precipitants increase their individual size. In this experiment, modeling the mechanical properties was based on the size and volume fraction of the particles, the average distance between them and the magnesium/copper content of the crystal matrix.
The strength profile for AlSiMg alloys with magnesium concentrations have shown a steep increase in yield strength through the precipitation hardening, a subsequent plateau with the final peak value, followed by a slow decrease in yield strength due to over aging, according to measured and simulated results (Fig. 1).
[FIGURE 1 OMITTED]
Within the European research project NADIA, which aims to promote the design and manufacture of automotive components in lightweight alloys, a cylinder head was poured using several melt compositions, cooling conditions and degrees of porosity (Fig. 2). The castings then were exposed to a T6 or T7 heat treatment. Temperature instruments placed at several locations on the castings recorded temperature profiles during filling, solidification and heat treatment. Additionally, sensors documented the stress profile during heat treatment. Researchers performed microstructure analysis in the area of the firing deck of the cylinder head.
[FIGURE 2 OMITTED]
The measured temperature profiles were used to calibrate the boundary conditions in the simulation software for all process steps, partially by using inverse optimization technology to correctly determine the conditions for air and water quenching.
Secondary dendrite arm spacing is an important microstructure component after solidification, according to the authors. The shorter the solidification time, the more refined the microstructure (Fig. 3). The simulated values matched the measured ones well.
[FIGURE 3 OMITTED]
The simulated fraction of the [Mg.sub.2]Si phase had a relatively constant value of approximately 0.4% evenly distributed throughout the part with some small, cooling condition-dependent variations. The pore volume was in the range of 0.04% to 0.57% and at a hydrogen content in the melt (prior to pouring) of 0.15ml/100g.
Measurements on the firing deck in areas impacted by the effect of cooling channels found values around 0.12% pore volume.
The simulation of the heat treatment provides the degree of dissolution as a result of the solution treatment, which can be used to evaluate whether its duration and temperature were sufficient to dissolve the copper and magnesium. Additionally, the concentrations of copper and magnesium were predicted.
3 Results and Conclusions
Simulation numbers for the concentration of magnesium in a cylinder head after solution treatment at 936F (530C) for 240 minutes correlated with previously established research (Fig. 4). Quench rates of the part exposed to a 150F (60C) water quench bath were around 122F/second (50C/second) for the casting's outer areas and 50F/second (10C/second) for the inner areas. According to the authors, due to the ideal test conditions, the maximum value is considerably higher than what could be achieved in a production environment, when many castings are quenched together in a basket.
[FIGURE 4 OMITTED]
Measurements for yield strength after aging found values around 260 MPa in the outer areas, which were in good correlation with the simulation. However, the simulated values for the interior of the part were about 25MPa below the measured values.
The peak yield strength of the part was achieved at approximately the same time as simulated--about one hour--and did not change much after 140 minutes of aging (Fig. 5). The researchers found the simulated distribution of the yield strength shows the quenching process has the biggest impact on this value, and casting geometry and its orientation during quenching are the most important parameters. The measurements show a smaller variation in the yield strength as a function of its location than was simulated.
[FIGURE 5 OMITTED]
Overall, the results of the cylinder head tests showed the as-cast microstructure, including phase and porosity distribution and mechanical properties after heat treatment, can be predicted. The models can be used to evaluate and optimize alloy composition and process parameters with metallurgical relevance.
By predicting local mechanical properties, designers and metalcasters can optimize the casting and heat treatment processes and part geometry. The simulation results can be used in lifetime optimization tools to evaluate thermomechanical fatigue in areas exposed to high thermal load conditions.
The paper (12-028) on which this article is based was first presented at the 2012 American Foundry Society Metalcasting Congress in Columbus, Ohio.
RELATED ARTICLE: ADDING IT ALL UP
Breaking down the industry's latest research papers is as easy as 1-2-3.
Modeling of Microstructure and Mechanical Properties of Aluminum Alloys During the Casting and Heat Treatment Process.
Marc Schneider, Magma Giessereitechnologie GmbH, Aachen, Germany, and Christof Heisser, Magma Foundry Technologies Inc., Schaumburg, III.
1 Background--The strength profile for AlSiMg alloys with magnesium concentrations show a steep increase in yield strength through the precipitation hardening, a subsequent plateau with the final peak value, followed by a slow decrease in yield strength due to over aging, according to actual measurements and software simulations.
2 Procedure--Researchers attempt to predict local microstructure and its affect on mechanical properties for a cylinder head casting using advancements in modeling techniques and abilities
3 Results and Conclusions--Overall, the researchers believe the results of the cylinder head tests show the as-cast microstructure, including phase and porosity distribution and mechanical properties after heat treatment, can be predicted.
To read the full paper on which this article was based, visit www.moderncasting.com.
A MODERN CASTING STAFF REPORT
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||TESTING 1-2-3|
|Date:||Sep 1, 2012|
|Previous Article:||Which ERP system is right for you? When two metalcasting facilities decided to upgrade their ERR software, one chose in-house and the other chose to...|
|Next Article:||AFS unveils brand, logo renovation.|