As the demand for data-intensive computing grows, so too does the need for next-generation memory technologies capable of delivering speed, energy efficiency, and scalability. Memristors—resistive memory devices that store and process data simultaneously—are considered promising candidates for next-generation in-memory and neuromorphic computing system.

Yet their widespread implementation remains limited by manufacturing constraints, particularly the trade-off between device performance and fabrication cost.

In the International Journal of Extreme Manufacturing (IJEM, IF: 21.3), Prof. Joohoon Kang and his team at Yonsei University examine this promising frontier: using solution-processed two-dimensional (2D) materials to fabricate memristors. The article offers the first in-depth survey dedicated specifically to memristors created using solution-based 2D materials, from synthesis and device engineering to switching mechanisms and applications.

“While 2D materials have long shown promise in electronic devices, their high manufacturing costs have hindered practical applications,” said Prof. Kang. “Our review spotlights recent breakthroughs demonstrating that solution-based techniques can bring these materials closer to scalable, real-world use.”

Why Solution-Processed 2D Materials?

Traditional memristors are often based on transition metal oxides, which provide good performance but can be difficult and expensive to manufacture at scale. Two-dimensional materials—such as MoS₂ or other layered crystals—are being studied as alternatives, thanks to their potential for low power consumption, fast switching, and mechanical flexibility.

However, the most common production techniques for 2D materials, such as mechanical exfoliation or chemical vapor deposition (CVD), face limitations in terms of yield, cost, and equipment requirements.

To address these limitations, researchers are turning to solution-based approaches, such as liquid-phase exfoliation, which can produce large quantities of 2D material dispersions at relatively low cost. These dispersions can then be deposited onto surfaces using familiar manufacturing methods such as spray-coating, inkjet printing, or spin coating.

“Solution processing allows 2D materials to be produced and applied in ways that are more compatible with large-area and high-throughput manufacturing,” Prof. Kang explained.

Recent Breakthroughs

Despite its scalability, solution processing historically suffered from serious drawbacks. Nanosheets produced via liquid exfoliation tend to be small (typically <100 nm) and defective, resulting in high intersheet resistance and non-uniform films that degrade device performance.

Kang’s team highlights how recent innovations—particularly electrochemical intercalation followed by mild sonication—are overcoming these barriers.

This technique produces larger, less-damaged nanosheets with improved film continuity and drastically reduced junction resistance. As a result, memristors fabricated using these materials are now approaching the performance levels of those made via CVD or mechanical exfoliation.

“This approach has revitalized the field,” noted co-first author Kijeong Nam. “We're now seeing electronic performance metrics that make solution-based 2D memristors competitive for the first time.”

Remaining Challenges

Although the progress is significant, the authors caution that several technical challenges must be addressed before commercial adoption. Chief among these are the need for:

• Achieving uniform film thickness and low surface roughness
• Reducing device-to-device variability
• Improving the size and structural uniformity of the nanosheets
• Integrating the materials into large-scale device arrays

“Achieving reproducible, large-area arrays with tight performance variation remains a critical hurdle,” said Dr. Dongjoon Rhee, another co-first author. “But the interdisciplinary nature of this field, such as chemistry, materials science, device physics, and AI, makes it an exciting one to watch.”

Outlook

As artificial intelligence, edge computing, and IoT applications continue to proliferate, the pressure to develop efficient and cost-effective memory solutions is mounting. The review by Kang and colleagues offers a timely and comprehensive look at a class of materials that may hold the key to scalable, high-performance memristive computing.

“The future of solution-processed 2D memristors,” Prof. Kang added, “will depend on how well we can marry low-cost fabrication with device reliability at scale. There is still work to do, but the potential benefits of these materials make them worth pursuing.”