Many‐Body Molecular Interactions in a Memristor

Abstract

Electronic transitions in molecular‐circuit elements hinge on complex interactions between molecules and ions, offering a multidimensional parameter space to embed, access, and optimize material functionalities for target‐specific applications. This opportunity is not cultivated in molecular memristors because their low‐temperature charge transport, which is a route to decipher molecular many‐body interactions, is unexplored. To address this, robust, temperature‐resilient molecular memristors based on a Ru complex of an azo aromatic ligand are designed, and current–voltage sweep measurements from room temperature down to 2 K with different cooling protocols are performed. By freezing out or activating different components of supramolecular dynamics, the local Coulombic interactions between the molecules and counterions that affect the electronic transport can be controlled. Operating conditions are designed where functionalities spanning bipolar, unipolar, nonvolatile, and volatile memristors with sharp as well as gradual analog transitions are captured within a single device. A mathematical design space evolves, thereof comprising 36 tuneable parameters in which all possible steady‐state functional variations in a memristor characteristic can be attainable. This enables a deterministic design route to engineer neuromorphic devices with unprecedented control over the transformation characteristics governing their functional flexibility and tunability.

Linear, symmetric, self-selecting 14-bit molecular memristors

“…3d), where the fast electron transfer occurs but the counterions remain in their original positions from state-31. The 2231 state is metastable (or volatile) and depending on the applied pulse width, a fraction of this will undergo a counterion relaxation 26,27 to form 2222, while the remainder reverts to 3131 (consistent with Raman measurements, Fig. 3a, 3b).…”

“…We fabricated the largest to date molecular memristor 64×64 crossbar (Fig. 1a) structure comprising a 60 nm thick film of [Ru II L2](BF4)2 (L = 2,6-bis(phenylazo)pyridine) 26,27 sandwiched between top and bottom gold electrodes (Fig. S1).…”

Memristor-Based Artificial Chips

“…For the nanodevices and nanotechnology, the memristor-based nanodevice primitives should be optimized to meet the AI application requirements of computing chips, and developing integration technology is beneficial to the design of future large-scale computing chips. Through integration with a high-density memristor-based crossbar synapse array, future computing chips should be more efficient in terms of area and should reach a much larger integration scale. − In the future development direction, it is expected that there will be general brain-like chips based on memristors, which will have a unified architecture design based on the most advanced technology. − At the same time, these universal chips should achieve high energy efficiency while achieving computational accuracy which surpasses that of traditional chips.…”

Energy and Space Efficient Parallel Adder Using Molecular Memristors

“…We have taken a rational middle ground. Using our robust molecular platform, [ 46–52 ] we fabricated the largest functional molecular memristor crossbar (comprising 128 cross points) demonstrated to date for an organic component to show that our circuit elements are consistent enough to pursue translation. Subsequently, we used realistic data to simulate a commercially competitive platform that can be benchmarked against an existing technology.…”

“…We have taken a rational middle ground. Using our robust molecular platform, [46][47][48][49][50][51][52] we fabricated the largest functional molecular memristor crossbar (comprising 128 cross points) demonstrated to date for an organic component to show that our circuit elements are consistent enough to pursue Adv. Mater.…”