A brain-inspired memory device is a programmable device that can improve computing power and speed while simplifying semiconductor circuit development.
Many electronic gadgets currently rely on semiconductor logic chips built on hard-wired gates that perform specified logic operations. Researchers from the National University of Singapore have invented a unique molecular memristor, or electronic memory device, with outstanding memory customization in collaboration with an international team of researchers.
The molecular device, unlike hard-wired traditional circuits, could be changed using voltage to integrate different computational processes. The new energy-efficient technology, which has increased processing speed and power, has the possibility to be used in hand-held devices and applications that require minimal power resources as well as in edge computing.
By altering the applied voltages, the memory device, like the synapses in the human brain, can be changed on the go for various processing applications. Furthermore, in the same way, those brain cells keep memories, the same device can store data for future recovery and analysis.
Dr. Sreebrata Goswami of the research team conceptualized and built a molecular system from the chemical family of phenyl azo pyridines, which feature a central metal atom bounded to organic molecules called as ligands. These molecules act as electron sponges, allowing around six electron transfers to occur, resulting in five distinct chemical states. Dr. Sreebrata Goswami stated that the interconnection between these states is the key to the device's customization.
Dr. Sreetosh Goswami developed a short electrical circuit with a molecular film of 40-nm placed between a gold top layer and a gold-infused nanodisc and indium tin oxide bottom layer. He saw an extraordinary current-voltage curve when he supplied a negative voltage to the gadget. Unlike traditional metal-oxide memristors, that can only be turned on and off at a single static voltage, these organic molecule devices can move between on and off states at a series of separate consecutive voltages. In contrast to the traditional approach of employing basic physics-based equations, the researchers defined the performance of the particles with the help of a decision tree algorithm using if-then-else expressions, which is utilized in the coding of various computer algorithms, mainly digital games.