Programmable Optoelectronic Matter: From 2D Materials to Photonic Intelligence

Jinyong Wang

doi:10.63823/pm20260004

Precision Manufacturing ›› DOI: 10.63823/pm20260004

review-article

Programmable Optoelectronic Matter: From 2D Materials to Photonic Intelligence

Jinyong Wang ^*

Author information +

History +

PDF (709KB)

Abstract

Modern computing is built upon the separation of sensing, memory, and computation—an architecture that increasingly limits intelligent systems due to the energy and latency cost of moving data between physically distinct units. Here, we introduce programmable optoelectronic matter, a framework in which optical, electronic, ionic, and interfacial degrees of freedom are coupled and dynamically reconfigured to sense, store, process, and learn within the same physical substrate. Two-dimensional (2D) van der Waals materials provide a uniquely powerful platform for this paradigm, enabling strong coupling among light, charge, ions, and interfaces across multiple spatial and temporal scales. The field is evolving from individual optoelectronic devices toward integrated photonic-electronic systems. Programmable optoelectronic matter may provide a pathway toward photonic intelligence, where information processing emerges from the programmable dynamics of matter itself.

Graphical abstract

Cite this article

Download citation ▾

Jinyong Wang. Programmable Optoelectronic Matter: From 2D Materials to Photonic Intelligence. Precision Manufacturing DOI:10.63823/pm20260004

登录浏览全文

4963

注册一个新账户忘记密码

1. Introduction

The history of information technology has been shaped by functional separation: sensors acquire data, memories store it, and processors perform computation [1,2]. This architectural principle has driven the digital revolution, yet it imposes a fundamental bottleneck—information must be continuously transferred between physically distinct hardware units. As modern intelligent systems increasingly rely on massive sensory data and large-scale models, the energy and latency associated with data movement are becoming dominant constraints on performance [3,4].

Biological intelligence operates differently. Sensing, memory, and computation are not strictly separated but emerge from the collective dynamics of adaptive physical networks [5 -9]. This contrast raises a fundamental question: can intelligence emerge directly from the dynamics of matter itself? Grounded in foundational ideas in neuromorphic engineering [1,3] and inspired by the biological model of co-localized sensing, memory, and computation, we introduce programmable optoelectronic matter: a conceptual framework in which optical, electronic, ionic, and interfacial degrees of freedom are coupled and dynamically reconfigured to sense, store, process, and learn within the same physical substrate. This perspective connects the proposed framework to broader concepts of physics-based and matter-centric computing [10,11], while positioning 2D materials as a uniquely enabling platform for the next generation of intelligent hardware.

2. A multidimensional state-space framework

Among available material platforms, 2D van der Waals materials are uniquely positioned to realize programmable optoelectronic matter. Their atomic-scale thickness, interface engineering capability, and rich coupled physical dynamics enable the convergence of light, charge, ions, and interfaces within a single system [12 -14]. Optical excitations, electronic charge states, ionic and defect dynamics, and interfacial phenomena can coexist, interact, and evolve in a highly correlated manner, creating forms of information processing that are difficult to achieve in conventional semiconductor systems [15 -17].

Formally, programmable optoelectronic matter can be represented as a dynamical system composed of coupled state variables:

(1)

$S\left(t\right)\mathrm{ }=\mathrm{ }\left\{O\right(t),\mathrm{ }E(t),\mathrm{ }I(t),\mathrm{ }F(t\left)\right\}$

where O(t), E(t), I(t), and F(t) denote the optical, electronic, ionic/defect, and interfacial states, respectively. Their collective evolution governs information processing, not any single transport mechanism.

Optical states. Photoexcitation generates excitons and free carriers (hν → X → e⁻ + h⁺), redistributing charge and energy within the material. In 2D semiconductors, photoexcited excitons may either dissociate into free carriers or recombine, depending on material composition, dielectric environment, and device conditions [6,18,19]. Optical excitation thus serves not merely as an input signal but as a driving force that pushes the system toward new physical states.

Electronic states. Carrier transport, trapping, and recombination continuously reshape conductivity and local electric fields, while simultaneously influencing ionic motion and interfacial energetics [20,21].

Ionic and defect states. Mobile ions, vacancies, and defect species introduce memory and history dependence. Their evolution follows the Nernst-Planck equation:

(2)

${J}_{i}={-D}_{i}{\nabla }_{{c}_{i}}+{\mu }_{i}{c}_{i}E$

where J_i denotes ionic flux, D_i and μ_i the diffusional and field-driven transport coefficients, c_i the local concentration, and E the electric field. Defect generation, annihilation, and electrochemical reactions can introduce source/sink terms beyond the standard drift-diffusion description [22 -24]. Such stimulus-dependent evolution endows programmable optoelectronic matter with adaptive, non-volatile, and reconfigurable functionalities absent in static semiconductor systems.

Interfacial states. Interfaces provide an additional layer of programmability. The Schottky barrier height (Φ_B = Φ_M − χ) is not fixed but evolves under charge trapping, polarization fields, defect reconfiguration, and ionic redistribution [25,26]—continuously modifying carrier injection pathways and coupling directly to memory formation. Interfaces therefore act not merely as material boundaries but as dynamic state variables.

The defining feature of 2D van der Waals materials is the strong mutual coupling among these states. Optical excitation generates carriers; electronic transport redistributes charge; ionic motion reconfigures local energy landscapes; defects preserve the history of external stimuli; interfaces regulate charge separation and transport. These processes operate over vastly different temporal and spatial scales yet remain continuously interconnected. The system trajectory is therefore governed by nonlinear coupling:

(3)

$dS/dt\mathrm{ }=\mathrm{ }F(O,\mathrm{ }E,\mathrm{ }I,\mathrm{ }F)$

transforming the material from a passive medium into an active computational system.

3. Why two-dimensional materials are central

The importance of 2D materials extends far beyond their atomic thickness. Their true significance lies in the ability to engineer and couple multiple physical degrees of freedom within a single programmable platform [12 -14,27,28]. Several properties make them uniquely suited for programmable optoelectronic matter:

Atomic-scale thickness and interface dominance. In 2D materials, virtually every atom is at or near a surface. This means that interfaces, with substrates, gate dielectrics, capping layers, or adjacent 2D sheets, profoundly influence electronic, optical, and ionic properties. Interface engineering therefore provides a direct handle on all four state variables simultaneously [29,30].

van der Waals heterostructure engineering. 2D materials can be stacked in arbitrary sequences without lattice-matching constraints, enabling the construction of designer heterostructures where each layer contributes a distinct functionality, such as photodetection, charge storage, synaptic plasticity, or ionic gating [13 -15]. This deterministic assembly capability creates design spaces inaccessible to epitaxial semiconductor systems.

Rich multi-physical coupling. In a single 2D material or heterostructure, optical, electronic, ionic, and interfacial degrees of freedom are strongly coupled. Light can gate ionic motion; ionic redistribution can shift optical absorption; defect dynamics can modulate Schottky barriers. This cross-coupling is not a nuisance to be engineered away but the core physical mechanism that enables computation to emerge from material dynamics [16,17,31,32].

Wafer-scale synthesis and CMOS integration. Advances in chemical vapor deposition and atomic layer epitaxy are enabling wafer-scale growth of MoS₂, WS₂, WSe₂, and related materials. Heterogeneous integration with CMOS back-end-of-line processes is already demonstrated, providing a manufacturable pathway for 2D-material-based intelligent hardware [33,34].

4. Optoelectronic memory as a computational primitive

The field is evolving beyond artificial synapses toward a more fundamental concept: optoelectronic memory [35,36]. Unlike conventional memory, which stores information passively, optoelectronic memory directly couples optical perception, state retention, and information processing within the same material system. Light does not merely provide an input signal; it reshapes the physical state of the material and thereby influences future computation [37 -39]. Memory becomes an active participant in information processing rather than a separate storage unit [11,18,31].

By merging sensing, memory, and computation into a common physical substrate, optoelectronic memory establishes a new computational primitive that may underpin future in-sensor computing and photonic intelligence systems. Fully integrated multi-mode optoelectronic memristor arrays have recently demonstrated the viability of this approach for diversified in-sensor computing [7].

5. Heterogeneous integration and photonic intelligence

The greatest opportunity for programmable optoelectronic matter lies not in improving individual device metrics, but in integrating diverse functionalities into a unified physical architecture. As illustrated in Figure 1, vertically integrated van der Waals heterostructures provide a framework for combining optical transmission, photodetection, optoelectronic memory, and CMOS control within a three-dimensional platform [40,41]. Information is sensed, stored, communicated, and processed through coupled photonic and electronic layers, creating a direct pathway toward in-memory photonic computing.

This convergence defines photonic intelligence: a regime in which photons actively participate in memory formation, adaptation, and decision-making rather than merely carrying information [42 -44]. Computation begins before digitization, emerging directly from coupled optical, electronic, ionic, and interfacial dynamics, thereby blurring the distinction between hardware and algorithm [26,45].

6. Outlook

Five emerging opportunities may define the evolution of programmable optoelectronic matter. First, establishing predictive physical frameworks that link atomic-scale dynamics to system-level functionality, moving from phenomenological descriptions toward quantitative models capable of rational materials design. Second, transitioning from demonstrating isolated device functionalities to constructing integrated intelligent systems in which sensing, memory, computation, and adaptive learning are co-localized. Third, translating the unique properties of 2D materials into manufacturable platforms through advances in wafer-scale synthesis and heterogeneous integration. Fourth, developing photonic-electronic convergent architectures that combine electronic memory, optical communication, and ionic adaptation across multiple scales, moving beyond von Neumann constraints. Fifth, and perhaps most transformatively, advancing toward matter-centric intelligence: systems in which sensing, memory, computation, and learning emerge directly from the intrinsic dynamics of physical matter, rather than from algorithms executing on passive hardware.

Practical realization of these opportunities will require addressing key engineering bottlenecks: achieving reliable and reproducible large-area synthesis of 2D materials, developing robust benchmarking methodologies and system-level performance metrics, and ensuring long-term device stability under thermal cycling and electrical stress. Reliability and manufacturability must be addressed in parallel with conceptual advances if programmable optoelectronic matter is to mature into a deployable technology platform.

7. Conclusion

Programmable optoelectronic matter represents a conceptual shift from computing with devices to computing with matter. By coupling optical, electronic, ionic, and interfacial degrees of freedom within the same physical substrate—a capability enabled by the unique properties of 2D van der Waals materials—it provides a framework in which sensing, memory, computation, and learning can emerge from the evolution of physical states rather than from isolated hardware modules. The ultimate significance of this paradigm lies not in enabling better individual devices, but in establishing a new physical basis for intelligent systems. As the field matures from empirical phenomena toward predictive frameworks, and from laboratory demonstrations toward manufacturable platforms, programmable optoelectronic matter may provide a pathway toward photonic intelligence and a new physical foundation for artificial intelligence itself.

Declaration of Competing Interest

The author declares that he has no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]

Jian Zhao, Cameron S. Jorgensen, Krishnamurthy Mahalingam, Cynthia Bowers, Wataru Sugimoto, Kai Ito, Seung Ju Kim, Ruoyu Zhao, Yichun Xu, Han-Ting Liao, Rajiv K. Kalia, Aiichiro Nakano, Kohei Shimamura, Fuyuki Shimojo, Priya Vashishta, Ajit K. Roy, Ning Ge, Miao Hu, R. Stanley Williams, Qiangfei Xia, Sabyasachi Ganguli, J. Joshua Yang. High-temperature memristors enabled by interfacial engineering. Science 2026, 392, 771-779. https://doi.org/10.1126/science.aeb9934.

[2]	Renji Bian, Ri He, Er Pan, Zefen Li, Guiming Cao, Peng Meng, Jiangang Chen, Qing Liu, Zhicheng Zhong, Wenwu Li, Fucai Liu. Developing fatigue-resistant ferroelectrics using interlayer sliding switching. Science 2024, 385, 57-62. https://doi.org/10.1126/science.ado1744.

[3]	Guy Rachmuth, Harel Z. Shouval, Mark F. Bear, Chi-Sang Poon. A biophysically-based neuromorphic model of spike rate- and timing-dependent plasticity. Proceedings of the National Academy of Sciences of the United States of America 2011, 108, E1266-E1274. https://doi.org/10.1073/pnas.1106161108.

[4]	Jinyong Wang, Nasir Ilyas, Yujing Ren, Yun Ji, Sifan Li, Changcun Li, Fucai Liu, Deen Gu, Kah-Wee Ang. Technology and integration roadmap for optoelectronic memristor. Advanced Materials 2024, 36, 2307393. https://doi.org/10.1002/adma.202307393.

[5]

Jiangang Chen, Chao Zhu, Guiming Cao, Haishi Liu, Renji Bian, Jinyong Wang, Changcun Li, Jieqiong Chen, Qundong Fu, Qing Liu, Peng Meng, Wei Li, Fucai Liu, Zheng Liu. Mimicking neuroplasticity via ion migration in van der Waals layered copper indium thiophosphate. Advanced Materials 2022, 34, 2104676. https://doi.org/10.1002/adma.202104676.

[6]

Chunmei Li, Jinyong Wang, Dongyang Li, Nasir Ilyas, Zhiqiang Yang, Kexin Chen, Peng Gu, Xiangdong Jiang, Deen Gu, Fucai Liu, Yadong Jiang, Wei Li. An oxide-based heterojunction optoelectronic synaptic device with wideband and rapid response performance. Journal of Materials Science and Technology 2022, 123, 159-167. https://doi.org/10.1016/j.jmst.2021.11.082.

[7]

Heyi Huang, Xiangpeng Liang, Yuyan Wang, Jianshi Tang, Yuankun Li, Yiwei Du, Wen Sun, Jianing Zhang, Peng Yao, Xing Mou, Feng Xu, Jinzhi Zhang, Yuyao Lu, Zhengwu Liu, Jianlin Wang, Zhixing Jiang, Ruofei Hu, Ze Wang, Qingtian Zhang, Bin Gao, Xuedong Bai, Lu Fang, Qionghai Dai, Huaxiang Yin, He Qian, Huaqiang Wu. Fully integrated multi-mode optoelectronic memristor array for diversified in-sensor computing. Nature Nanotechnology 2025, 20, 93-103. https://doi.org/10.1038/s41565-024-01794-z.

[8]	Guobin Zhang, Qi Luo, Jiacheng Yao, Shuai Zhong, Hua Wang, Fei Xue, Bin Yu, Kian Ping Loh, Yishu Zhang. All-in-one neuromorphic hardware with 2D material technology: current status and future perspective. Chemical Society Reviews 2025, 54, 8196-8242. DOI https://doi.org/10.1039/D5CS00251F.

[9]

Feichi Zhou, Zheng Zhou, Jiewei Chen, Tsz Hin Choy, Jingli Wang, Ning Zhang, Ziyuan Lin, Shimeng Yu, Jinfeng Kang, H.-S. Philip Wong, Yang Chai. Optoelectronic resistive random access memory for neuromorphic vision sensors. Nature Nanotechnology 2019, 14, 776-782. https://doi.org/10.1038/s41565-019-0501-3.

[10]	M. A. McEvoy, N. Correll. Materials that couple sensing, actuation, computation, and communication. Science 2015, 347, 1261689. https://doi.org/10.1126/science.1261689.

[11]	Abu Sebastian, Manuel Le Gallo, Riduan Khaddam-Aljameh, Evangelos Eleftheriou. Memory devices and applications for in-memory computing. Nature Nanotechnology 2020, 15, 529-544. https://doi.org/10.1038/s41565-020-0655-z.

[12]	Feichi Zhou, Yang Chai. Near-sensor and in-sensor computing. Nature Electronics 2020, 3, 664-671. https://doi.org/10.1038/s41928-020-00501-9.

[13]

Dan Cai, Jinyong Wang, Tianchen Zhao, Miao Shen, Yunbo Liu, Tieyi Zhang, Fangjie Zhang, Yang Wang, Yadong Jiang, Deen Gu. IGZO-based first spike timing tactile encoders and coupling-enhanced transistor synapses for efficient spiking neural networks. Advanced Science 2026, 13, 2511168. https://doi.org/10.1002/advs.202511168.

[14]

Yun Ji, Baoshan Tang, Jinyong Wang, Haofei Zheng, Yangwu Wu, Yihao Wu, Sifan Li, Aaron Voon-Yew Thean, Kah-Wee Ang. High-speed and low-energy resistive switching with two-dimensional cobalt phosphorus trisulfide for efficient neuromorphic computing. ACS Nano 2025, 19, 722-735. https://doi.org/10.1021/acsnano.4c11890.

[15]	Yun Ji, Lin Wang, Yinfeng Long, Jinyong Wang, Haofei Zheng, Zhi Gen Yu, Yong-Wei Zhang, Kah-Wee Ang. Ultralow energy adaptive neuromorphic computing using reconfigurable zinc phosphorus trisulfide memristors. Nature Communications 2025, 16, 6899. https://doi.org/10.1038/s41467-025-62306-8.

[16]	Dan Cai, Yunbo Liu, Jinyong Wang, Tianchen Zhao, Miao Shen, Fangjie Zhang, Yadong Jiang, Deen Gu. BCM-inspired synapses constructed with barrier-modulated coupling junctions for enhancing speech recognition. Advanced Functional Materials 2024, 34, 2314660. https://doi.org/10.1002/adfm.202314660.

[17]	Chunmei Li, Nasir Ilyas, Jinyong Wang, Yanan Li, Haolin Luo, Dongyang Li, Deen Gu, Fucai Liu, Yadong Jiang, Wei Li. Nanostructured CuAlO₂@ZnO optoelectronic device for artificial synaptic applications. Applied Surface Science 2023, 611, 155682. https://doi.org/10.1016/j.apsusc.2022.155682.

[18]

Jinyong Wang, Yujing Ren, Ze Yang, Qiaoya Lv, Yu Zhang, Mingyue Zhang, Tiancheng Zhao, Deen Gu, Fucai Liu, Baoshan Tang, Weifeng Yang, Zhiqun Lin. Synergistically modulating conductive filaments in ion-based memristors for enhanced analog in-memory computing. Advanced Science 2024, 11, 2309538. https://doi.org/10.1002/advs.202309538.

[19]	Shreyash Hadke, Min-A Kang, Vinod K. Sangwan, Mark C. Hersam. Two-dimensional materials for brain-inspired computing hardware. Chemical Reviews 2025, 125, 835-932. https://doi.org/10.1021/acs.chemrev.4c00631.

[20]	Vinod K. Sangwan, Mark C. Hersam. Neuromorphic nanoelectronic materials. Nature Nanotechnology 2020, 15, 517-528. https://doi.org/10.1038/s41565-020-0647-z.

[21]	Yuan Liu, Nathan O. Weiss, Xidong Duan, Hung-Chieh Cheng, Yu Huang, Xiangfeng Duan. Van der Waals heterostructures and devices. Nature Reviews Materials 2016, 1, 16042. https://doi.org/10.1038/natrevmats.2016.42.

[22]	K. S. Novoselov, A. Mishchenko, A. Carvalho, A. H. Castro Neto. 2D materials and van der Waals heterostructures. Science 2016, 353, aac9439. https://doi.org/10.1126/science.aac9439.

[23]

Sidong Lei, Fangfang Wen, Bo Li, Qizhong Wang, Yihan Huang, Yongji Gong, Yongmin He, Pei Dong, James Bellah, Antony George, Liehui Ge, Jun Lou, Naomi J. Halas, Robert Vajtai, Pulickel M. Ajayan.Optoelectronic memory using two-dimensional materials. Nano Letters 2015, 15, 259-265. https://doi.org/10.1021/nl503505f.

[24]	Qinqi Ren, Chaoyi Zhu, Sijie Ma, Zhaoqing Wang, Jianmin Yan, Tianqing Wan, Weicheng Yan, Yang Chai.Optoelectronic devices for in-sensor computing. Advanced Materials 2025, 37, 2407476. https://doi.org/10.1002/adma.202407476.

[25]

Dongyang Li, Jinyong Wang, Yujing Ren, Bo Wu, Tiancheng Zhao, Xun Cao, Deen Gu, Ming Xu, Jian Ma, Zhiqun Lin. Unraveling sub-nanostructure variability in amorphous silicon: mechanisms of short-range order and defect dynamics via in situ Raman spectroscopy. Advanced Functional Materials 2025, 35, 2423198. https://doi.org/10.1002/adfm.202423198.

[26]	Jinyong Wang, Yujing Ren, Baoshan Tang, Changcun Li, Fucai Liu, Deen Gu. Integrated optoelectronic memory array with enhanced broadband Si-based lines for advanced visual systems. CLEO: Applications and Technology, ATh5A.7, 2023. https://doi.org/10.1364/CLEO_AT.2023.ATh5A.7.

[27]	Jinyong Wang, Yangming Leng, Tiancheng Zhao, Chunmei Li, Deen Gu, Wei Li. SnO₂-based optoelectronic synapses for artificial visual applications. Journal of Physics: Conference Series 2023, 2524, 012011. https://doi.org/10.1088/1742-6596/2524/1/012011.

[28]	Chunmei Li, Yanan Li, Haolin Luo, Yangming Leng, Jinyong Wang, Wei Li. Light-modulated synaptic plasticity of SrTiO₃:Ru/CuAlO₂ bilayer based memristors for artificial visual system. Proceedings of SPIE 2022, 12351, 123510W. https://doi.org/10.1117/12.2652380.

[29]	Yanan Li, Haolin Luo, Yangming Leng, Chunmei Li, Jinyong Wang, Wei Li. Optoelectronic synaptic performance of NiO/ZnO bilayer based memristors. Proceedings of SPIE 2022, 12351, 123510T. https://doi.org/10.1117/12.2652368.

[30]	Haolin Luo, Yanan Li, Yangming Leng, Chunmei Li, Jinyong Wang, Wei Li. Synaptic properties of monolayer CuO_x-based optoelectronic devices. Proceedings of SPIE 2022, 12351, 123510U. https://doi.org/10.1117/12.2652375.

[31]	Ruoyu Zhao, Seung Ju Kim, Yichun Xu, Jian Zhao, Tong Wang, Rivu Midya, Sabyasachi Ganguli, Ajit K. Roy, Madan Dubey, R. Stanley Williams, J. Joshua Yang. Memristive ion dynamics to enable biorealistic computing. Chemical Reviews 2025, 125, 745-785. https://doi.org/10.1021/acs.chemrev.4c00587.

[32]	Jinyong Wang, Yujing Ren, Fenghua Xu, Yufei Shi, Jing Gao, Yu Zhang, Kah-Wee Ang. Van der Waals integrated flexible multi-mode optoelectronic memristor for bio-neuromorphic systems. CLEO 2025, SS143_4, 2025. https://doi.org/10.1364/CLEO_SI.2025.SS143_4.

[33]	Tianyi Xiong, Changwei Li, Xiulan He, Boyang Xie, Jianwei Zong, Yanan Jiang, Wenjie Ma, Fei Wu, Junjie Fei, Ping Yu, Lanqun Mao. Neuromorphic functions with a polyelectrolyte-confined fluidic memristor. Science 2023, 379, 156-161. https://doi.org/10.1126/science.adc9150.

[34]	Yaqi Hou, Xu Hou.Bioinspired nanofluidic iontronics. Science 2021, 373, 628-629. https://doi.org/10.1126/science.abj0437.

[35]

Hanwool Yeon, Peng Lin, Chanyeol Choi, Scott H. Tan, Yongmo Park, Doyoon Lee, Jaeyong Lee, Feng Xu, Bin Gao, Huaqiang Wu, He Qian, Yifan Nie, Seyoung Kim, Jeehwan Kim. Alloying conducting channels for reliable neuromorphic computing. Nature Nanotechnology 2020, 15, 574-579. https://doi.org/10.1038/s41565-020-0694-5.

[36]

Maolin Chen, Yinchang Ma, Nabeel Aslam, Chen Liu, Yiqiang Chen, Linqu Luo, Xiaowen Zhang, Kairan Mai, Han Xiao, Kaichen Zhu, Osamah Alharbi, Dongxing Zheng, Xiangming Xu, Hanguang Liao, Yiming Yang, Heng Wang, Zhican Zhou, Hanwen Wang, Bo Tian, Junzhu Li, Xin He, Kai Chang, Yating Wan, Atif Shamim, Husam N. Alshareef, Mario Lanza, Thomas D. Anthopoulos, Zheng Han, Fei Xue, Xixiang Zhang.Ultrawide-bandwidth boron nitride photonic memristors. Nature Nanotechnology 2025, 20, 1633-1640. https://doi.org/10.1038/s41565-025-01991-4.

[37]	Nasir Ilyas, Jinyong Wang, Chunmei Li, Dongyang Li, Hao Fu, Deen Gu, Xiangdong Jiang, Fucai Liu, Yadong Jiang, Wei Li. Nanostructured materials and architectures for advanced optoelectronic synaptic devices. Advanced Functional Materials 2022, 32, 2110976. https://doi.org/10.1002/adfm.202110976.

[38]	Chaoliang Tan, Xiehong Cao, Xue-Jun Wu, Qiyuan He, Jian Yang, Xiao Zhang, Junze Chen, Wei Zhao, Shikui Han, Gwang-Hyeon Nam, Melinda Sindoro, Hua Zhang. Recent advances in ultrathin two-dimensional nanomaterials. Chemical Reviews 2017, 117, 6225-6331. https://doi.org/10.1021/acs.chemrev.6b00558.

[39]	Houk Jang, Henry Hinton, Woo-Bin Jung, Min-Hyun Lee, Changhyun Kim, Min Park, Seoung-Ki Lee, Seongjun Park, Donhee Ham. In-sensor optoelectronic computing using electrostatically doped silicon. Nature Electronics 2022, 5, 519-525. https://doi.org/10.1038/s41928-022-00819-6.

[40]	Shuiyuan Wang, Xiaoxian Liu, Peng Zhou. The road for 2D semiconductors in the silicon age. Advanced Materials 2022, 34, 2106886. https://doi.org/10.1002/adma.202106886.

[41]	Deji Akinwande, Cedric Huyghebaert, Ching-Hua Wang, Martha I. Serna, Stijn Goossens, Lain-Jong Li, H.-S. Philip Wong, Frank H. L. Koppens. Graphene and two-dimensional materials for silicon technology. Nature 2019, 573, 507-518. https://doi.org/10.1038/s41586-019-1573-9.

[42]	Bhavin J. Shastri, Alexander N. Tait, T. Ferreira de Lima,Wolfram H. P. Pernice, Harish Bhaskaran, C. D. Wright, Paul R. Prucnal. Nature Photonics 2021, 15, 102-114. https://doi.org/10.1038/s41566-020-00754-y.

[43]	J. Feldmann, N. Youngblood, C. D. Wright, H. Bhaskaran, W. H. P. Pernice. All-optical spiking neurosynaptic networks with self-learning capabilities. Nature 2019, 569, 208-214. https://doi.org/10.1038/s41586-019-1157-8.

[44]	Elena Goi, Xi Chen, Qiming Zhang, Benjamin P. Cumming, Steffen Schoenhardt, Haitao Luan, Min Gu. Nanoprinted high-neuron-density optical linear perceptrons performing near-infrared inference on a CMOS chip. Light: Science & Applications 2021, 10, 40. https://doi.org/10.1038/s41377-021-00483-z.

[45]	P. Robin, T. Emmerich, A. Ismail, A. Niguès, Y. You, G.-H. Nam, A. Keerthi, A. Siria, A. K. Geim, B. Radha, L. Bocquet. Long-term memory and synapse-like dynamics in two-dimensional nanofluidic channels. Science 2023, 379, 161-167. https://doi.org/10.1126/science.adc9931.