Abstract: A process is provided for producing an aluminium-titanium-boron grain refining master alloy containing titanium boride and titanium aluminide particles, the process comprising melting aluminium in a silicon carbide crucible in a medium frequency induction or an electric resistance furnace, adding to the melt at a temperature between 750 degrees Celsius and 900 degrees Celsius, KBF4 and K2TiF6 salts, pre-mixed in proportions to obtain a Ti/B ratio of 5 in the melt, gently mixing the salt mixture with the melt without introducing any stirring, transferring the molten alloy to an electric resistance furnace maintained at 800 degrees Celsius, decanting the K—Al—F salt, the by-product of the salt reaction, thoroughly stirring the molten alloy in the SiC crucible with graphite rods before finally casting the molten alloy into cylindrical molds in the form of billets and finally extruding the billet into 9.5 mm rods.

Abstract: Novel random number generation methods and novel random number generators based on continuous-time chaotic oscillators with dual oscillator architecture are presented. Numerical and experimental results not only verify the feasibility of the proposed circuits, but also encourage their use as a high-performance IC TRNG. In comparison with RNG's based on discrete-time chaotic maps, amplification of a noise source and jittered oscillator sampling, which are advantageous in the sense that true random behavior can be mathematically proven thanks to an analytical model that has been developed, it is seen that RNG's based on continuous-time' chaotic oscillators can offer much higher and constant data rated without post-processing. The proposed innovation increases the throughput, maximizes the statistical quality of the output sequence and is robust against against external interference, parameter variations and attacks aimed to force throughout. The proposed circuits can be integrated on today process at GHz range.