Grain refinement of commercial pure Al treated by Pulsed Magneto-Oscillation on the top surface of melt

Commercial pure Al can be refined by Pulsed Magneto-Oscillation(PMO) treatment applied via a plate induction coil above the top surface of the melt. The proportion of the equiaxed zone area increases with decreasing Height to Diameter(H/D) ratios from 3.5 to1.8 and further to 1.0. Meanwhile, it increases and then decreases with increasing peak current for the three kinds of ingots with H/D ratios of 3.5, 1.8 and 1.0, respectively. However, when the H/D ratio decreases to 0.44, the area proportion of equiaxed zone can reach the maximum value with a lower peak current. FEA software simulation indicates that smaller H/D ratio results in larger current density, electromagnetic force and convection on the top surface of the melt, favoring nucleation and subsequent grain formation. Through evaluating Joule heating effect by PMO, it was found that the proper amount of Joule heating benefits grain refinement. Excessive Joule heating can reduce the size of the equiaxed zone and change the growth morphology of the grains.

Separate Neural Networks for Gains and Losses in Intertemporal Choice

An important and unresolved question is how human brain regions process information and interact with each other in intertemporal choice related to gains and losses. Using psychophysiological interaction and dynamic causal modeling analyses, we investigated the functional interactions between regions involved in the decision- making process while participants performed temporal discounting tasks in both the gains and losses domains. We found two distinct intrinsic valuation systems underlying temporal discounting in the gains and losses domains： gains were specifically evaluated in the medial regions, including the medial prefrontal and orbitofrontal cortices, and losses were evaluated in the lateral dorsolateral prefrontal cortex. In addition, immediate reward or pun- ishment was found to modulate the functional interactions between the dorsolateral prefrontal cortex and distinct regions in both the gains and losses domains： in the gains domain, the mesolimbic regions; in the losses domain, the medial prefrontal cortex, anterior cingulate cortex, and insula. These findings suggest that intertemporal choice of gains and losses might involve distinct valuation systems, and more importantly, separate neural interactions may implement the intertemporal choices of gains and losses. These findings may provide a new biological perspective for understanding the neural mechanisms underlying intertemporal choice of gains and losses.