NOTE: This version has several huge changes in numerics. Calculations may take longer time as usual.

The following new features are implemented in MatCalc versions 5.61.nnnn:

• Improved treatment of primary precipitates. The virtual treatment procedure is obsolete. Instead, size distributions can be generated directly from any existing phase, e.g., as calculated from a Scheil-Gulliver analysis.
• New FSK trapping routines, which can be applied to equilibrium and kinetic calculations.
• New couples and pairs model for solute trapping, which can describe the thermodynamics of atomic couples in solution phases. Presently, only the equilibrium implementation is operative.
• New treatment of grain boundary precipitation. User can now select whether gb precipitates are treated with gb diffusion geometry or with classical spherical diffusion fields as in the original SFFK model.
• Substantial changes in the substructure evolution models. The ABC and 2ABC models are working rather well already.

Below are details describing the various changes and new features of the MatCalc versions 5.61.xxxx.

• Several improvements in ABC and 2ABC model for substructure evolution.
• Implemented user-defined functions in treatment of coalescence.
• Fixed a potential bug in calculation of diffusion coefficients of ordered phases.
• Implemented subgrain boundary mobility into ABC and 2ABC model. Have decoupled it from C-parameter.
• Subgrain/cell size can now be controlled independently with A' parameter in ABC model.
• Fixed a bug in single-class grain size model, where grains could shrink in special cases.
• Fixed a problem with comments at the end of FOR command.
• Updated template scripts to comply to latest changes.
• Added variable for the Read-Shockley model for geometrically necessary dislocations in a substructure network of given subgrain size (DD_EQU_RS$prec_domain_name).
• Removed Sherstnev-Koze advanced model.
• 2ABC model: Wall dislocation density now converges towards a separate equilibrium wall dislocation density.
• Fixed a problem with negative solid solution strengthening coefficients.
• Changed terminology from co-clusters to couples and pairs.

• Added the possibility of defining a fixed number of nucleation sites.
• Added the functinality to suppress nucleation of a complex phase when the supersaturationi is so high that the critical radius is below the minimum radius of the crystal structure (i.e., below the size of the unit cell. This feature will avoid that all different types of phases immediately nucleate with very small critical radius during fast quenching processes.
• Added a variable to display phase fraction of only incoherent precipitate population (F_PREC_INC$phase_name). This feature can be used to analyse partial phase fractions, i.e., phase fractions of a given size range.
• Re-organization of the precipitation domain dialog.
• First implementation of co-cluster thermodynamics.
• Added a possibility to take into account that plastic deformation can break up particles.
• Added feature to automatically change grain boundary precipitates from gb diffusion geometry to classical SFFK with spherical diffusion fields.
• Modified treatment of the equilibrium cluster distribution function (ECDF) when switching from one heat treatment segment to another.
• Fixed several minor bugs.

• Modified the numerics for precipitation calculation to improve stability and performance. New implementation should also show less oscillations in chemical potential and other variables.
• Modified treatment of Clemm-Fisher model for interfacial energy reduction at grain and subgrain boundaries. MatCalc now automatically adds the structural part
• Implemented a size-dependent inner particle diffusion coefficient factor (ipdcf). This feature (ipdcf radius exponent) makes it possible to account for easier chemical composition changes in large precipitates. The assumption is that the balance of radius and composition change for the precipitate made by the thermodynamic extremal principle (TEP) is now determined for a small shell at the precipitate boundary instead of the entire precipitate volume. It gives good results for para-cementite precipitation in martensite for a value of 2. Default is 0, which means it is not used at all. This ensures compatibility with older versions. The exponent enters the evaluation of the diffusion coefficient inside the particle proportional to the (scaled) radius to the power of exponent.
• Added a variable for nucleation site efficiency. In contrast to the nucleation constant, which enters only the nucleation rate expression, this efficiency also limits the total number of nucleation sites. With this feature, you can efficiently control the total number of sites.
• Fixed a bug in treatment of SFAK excess vavancy evolution model introduced earlier in this version.
• Modified some variable names for strength: precipitate strength now consistently refers to shear stress, whereas precipitation domain strength is mostly expressed in terms of yield stress.
• Introduced strengthening contribution model instead of only “ignore for strengthening” switch.
• Implemented co-cluster strengthening model of Starink et al. This is needed for a better representation of natural aging of Al-alloys.
• Several improvements in SFK trapping implementation. Modified treatment of available trap sites and changed several variable names and types.
• Implemented routine for sorting of size distributions, both, precipitate and grain size distributions. We now always maintain descending order in size variable.
• Fixed a “factor 10” typo in solid solution strengthening parameters in defaults.txt file.
• Changed model for chemical strengthening: Now the effective interfacial energy is used instead of the calculated planar sharp IE.
• Fixed a crash when inspecting the yx_nucl_comp$prec variable.
• The precipitate size distribution is now displayed AND evaluated depending on the x-axis setting of the histogram plot: “lin” will evaluate in linear size intervals, “log” in logarithmic intervals.
• Fixed a bug in treatment of excess vacancies on restarting precipitation calc from loaded state.
• In tabulate precipitate distribution window: append not prepend tabulate variables, now. This means, variables being dropped into the window are now added in the rightmost column.
• Fixed a few issues with multiple selection in the phase status dialog.

• Variables, strings, functions and tables can now be defined as being either global or local by using the prefix “::”. Global variables are saved in the default-Gibbs node. This is a useful feature when accessing one single variable from extra Gibbs nodes, such as, for instance, in cell simulations. Example: “my_variable” exists only in the particular node in which it was created. Several “my_variable” can exist in different nodes. The variable “::my_variable” is always created and accessible in the default node, irrespective of the selection of current active node.
• Fixed a bug in the evaluation of the maximum nucleation rate nucleus composition.
• Fixed a problem with excess vacancies and recovery in ABC substructure model.
• Renamed several variables related to dislocation densities and substructure. This process is ongoing and will probably continue some more time.
• Fixed a problem with saving last result of kinetic simulation during continuous cooling.
• Fixed a crash in table editor.
• The default value for “ignore misfit stress during deformation” is now set on FALSE. The user must explicitely check this option to use it.

• Fixed mistake in calculation routine of mean distances for precipitation strengthening.