Background
As an emerging non-volatile device, memristors have garnered significant attention due to their low power consumption, high density, and multifunctionality. In modern computing systems, traditional logic gates based on complementary metal-oxide-semiconductor (CMOS) technology face challenges such as high power consumption, large integration area, and frequent data transfer between storage and computation during processing—known as the "von Neumann bottleneck." These limitations hinder their further development in high-computational power and energy-efficient scenarios like artificial intelligence and the Internet of Things. Memristors, with their unique resistive switching characteristics, can integrate data storage and logical operations within the same device, providing a revolutionary solution for in-memory computing. Therefore, exploring the design and application of memristor-based logic gates is of great significance for building next-generation high-performance, low-power computing architectures.
Research Progress
To address these challenges, the research team led by Researcher Tianyu Wang from the School of Integrated Circuits at Shandong University has systematically reviewed the latest advances in emerging memristors for in-memory computing applications. This review summarizes key breakthroughs and challenges in material design, device performance, and circuit implementation of memristors in logic gate applications, covering various material systems including two-dimensional materials, perovskite materials, and optoelectronic materials, as well as novel structures such as array architectures and wearable textile memristors, and evaluating their suitability for achieving stable and efficient logic operations (Fig. 1).
In terms of logic function implementation, the research team systematically elaborated on how to directly utilize the resistive states of memristors to realize basic Boolean logic (e.g., AND, OR). Using principles such as voltage division or material implication logic (IMP), memristors can construct compact logic gates. Furthermore, by combining memristors with CMOS devices or designing pure memristor stateful logic (e.g., the MAGIC architecture), complex logic functions such as NAND and NOR can be efficiently realized, which significantly reduces circuit complexity and power consumption (Fig. 2).
The study also highlights the development of optoelectronic logic gates. By introducing optical signals as inputs, memristor logic gates can leverage photogenerated carriers or light-induced ion migration effects to achieve fast, low-power, and reconfigurable logic operations (Fig. 3). Some devices can even integrate multimodal inputs such as electrical, optical, and humidity signals to perform environment-aware complex logic decisions, showcasing their potential in smart sensing and edge computing applications.
At the combinational logic level, memristor arrays have been successfully used to construct basic arithmetic units such as XOR gates, half-adders, and full-adders, and further applied to more complex structures like parallel prefix adders (Fig. 4). Through designs such as majority gates or threshold logic, memristor arrays can achieve highly parallel and energy-efficient in-memory arithmetic operations. These functionalities have also been extended to hardware encryption, where memristor arrays enable efficient encryption and decryption of data and images, validating their potential in integrated logic-storage and secure computing systems.
Future Prospects
Looking ahead, memristor-based logic gate technology holds broad prospects across multiple fields. In computing architecture, memristor logic is expected to drive the development of in-memory computing and neuromorphic computing hardware, enabling the construction of high-efficiency, low-power brain-inspired intelligent chips by emulating biological synapses and neuron behaviors. In edge computing and IoT applications, their low power consumption, high integration density, and multimodal signal processing capabilities make them highly suitable for smart sensors, wearable devices, and similar scenarios.
However, to achieve large-scale practical application of memristor logic, key challenges remain, including device performance uniformity, multi-state stability, scalability of array size, and compatibility with existing CMOS processes. Future research needs to focus on developing high-performance, highly reliable memristor materials and integration processes, while designing robust circuit architectures and error-correction mechanisms to address device non-idealities. With collaborative innovation in materials science, device physics, and circuit design, memristor logic gates are poised to become a key enabling technology for breaking through traditional computing paradigms and ushering in an era of efficient and intelligent computing in the post-Moore era.

The complete study is accessible via DOI:10.34133/research.0916