Taotao Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Taotao Li has a total of 355 co-authors. Here are the top co-authors by number of shared publications:
- Xinran Wang
National Laboratory of Solid-State Microstructures, School of Electronic Science and Engineering and Collaborative Innovation Center of Advanced Microstructures, Nanjing University, Nanjing 210093, Jiangsu, China
19 shared publications - Yi Shi
National Laboratory of Solid-State Microstructures, School of Electronic Science and Engineering and Collaborative Innovation Center of Advanced Microstructures, Nanjing University, Nanjing, China
17 shared publications - Weisheng Li
School of Integrated Circuits, Nanjing University 2 , Suzhou 215163
11 shared publications Lei Liu
Suzhou Laboratory
10 shared publications- Zhihao Yu
College of Integrated Circuit Science and Engineering, Nanjing University of Posts and Telecommunications 1 , Nanjing 210023
9 shared publications - Hao Qiu
School of Electronic Science and Engineering and the Engineering Research Center of Opto-Electro Materials and Chip Techniques, Nanjing University, Nanjing, China
7 shared publications
Latest Publications
- Open Access
- 18 July 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1) , 8802
Wavelength-encoded neuromorphic inference enabled by microcavity MoS2 photodetector arrays
Biological vision systems tightly couple spectral sensing with temporal integration to extract task-relevant information with minimal data movement. In contrast, conventional optoelectronic vision hardware typically separates photodetection from electronic computation, incurring substantial latency and energy costs. Optical neural networks can alleviate this bottleneck, but many implementations are difficult to scale. Here, we propose a bio-inspired optoelectronic inference architecture based on a Fabry-Perot microcavity-integrated MoS2 photodetector array, in which sensing, weighting, and accumulation are unified within each pixel. Cavity-engineered wavelength selectivity encodes neural-network weights in the spectral domain, while the finite carrier lifetime of MoS2 enables analog temporal accumulation without external memory. The system achieves test accuracies of 99.6%, 94.8% and 94.0% on MNIST, CIFAR-10 and the Free Spoken Digit Dataset, respectively. Post-training optical Hessian pruning further reduces optical complexity while maintaining robustness. This architecture provides a compact route toward wavelength-aware in-sensor neuromorphic inference.
- 18 July 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1) , 8802
Open AccessWavelength-encoded neuromorphic inference enabled by microcavity MoS2 photodetector arrays
Biological vision systems tightly couple spectral sensing with temporal integration to extract task-relevant information with minimal data movement. In contrast, conventional optoelectronic vision hardware typically separates photodetection from electronic computation, incurring substantial latency and energy costs. Optical neural networks can alleviate this bottleneck, but many implementations are difficult to scale. Here, we propose a bio-inspired optoelectronic inference architecture based on a Fabry-Perot microcavity-integrated MoS2 photodetector array, in which sensing, weighting, and accumulation are unified within each pixel. Cavity-engineered wavelength selectivity encodes neural-network weights in the spectral domain, while the finite carrier lifetime of MoS2 enables analog temporal accumulation without external memory. The system achieves test accuracies of 99.6%, 94.8% and 94.0% on MNIST, CIFAR-10 and the Free Spoken Digit Dataset, respectively. Post-training optical Hessian pruning further reduces optical complexity while maintaining robustness. This architecture provides a compact route toward wavelength-aware in-sensor neuromorphic inference.
- Open Access
- 7 July 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Large-area two-dimensional MoO3 as a high-κ dielectric for van der Waals integration
Large-area high-quality dielectrics are essential for the integration of two-dimensional (2D) transistors. However, this integration remains challenging, especially for conventional dielectrics that require deposition processes. Adopting van der Waals (vdW) dielectrics is attractive. Molybdenum trioxide (α-MoO 3 ) has been demonstrated as a vdW dielectric in 2D field-effect transistors (FETs), but previous demonstrations showed limited performance and small size. Here, we grow large-area single-crystalline MoO 3 with uniform thickness and explore its properties as a dielectric. Single-crystalline MoO 3 reaches the millimeter scale and maintains a high dielectric constant of ~ 18.5, low gate leakage of approximately 10 −2 A cm −2 at 3.0 MV cm −1, and a large breakdown field of ~ 26.75 MV cm −1 . Top-gated molybdenum disulfide (MoS 2 ) transistors exhibit an average subthreshold swing of ~ 71 mV dec −1 , and an on/off current ratio exceeding 10 8 . Furthermore, we demonstrate MoO 3 -based integrated circuits and low-power complementary metal-oxide-semiconductor (CMOS) inverters.
- 7 July 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Open AccessLarge-area two-dimensional MoO3 as a high-κ dielectric for van der Waals integration
Large-area high-quality dielectrics are essential for the integration of two-dimensional (2D) transistors. However, this integration remains challenging, especially for conventional dielectrics that require deposition processes. Adopting van der Waals (vdW) dielectrics is attractive. Molybdenum trioxide (α-MoO 3 ) has been demonstrated as a vdW dielectric in 2D field-effect transistors (FETs), but previous demonstrations showed limited performance and small size. Here, we grow large-area single-crystalline MoO 3 with uniform thickness and explore its properties as a dielectric. Single-crystalline MoO 3 reaches the millimeter scale and maintains a high dielectric constant of ~ 18.5, low gate leakage of approximately 10 −2 A cm −2 at 3.0 MV cm −1, and a large breakdown field of ~ 26.75 MV cm −1 . Top-gated molybdenum disulfide (MoS 2 ) transistors exhibit an average subthreshold swing of ~ 71 mV dec −1 , and an on/off current ratio exceeding 10 8 . Furthermore, we demonstrate MoO 3 -based integrated circuits and low-power complementary metal-oxide-semiconductor (CMOS) inverters.
- 1 July 2026
- Published by IOP Publishing in Journal of Semiconductors
- Vol. 47 (7) , 070401
In-situ TEM unveils the secrets of two-dimensional material nucleation
In-situ TEM unveils the secrets of two-dimensional material nucleation , Liu, Lei, Zhang, Xiaotian, Li, Taotao, Wang, Xinran In-situ TEM unveils the secrets of two-dimensional material nucleation , Liu, Lei, Zhang, Xiaotian, Li, Taotao, Wang, Xinran
- 1 July 2026
- Published by IOP Publishing in Journal of Semiconductors
- Vol. 47 (7) , 070401
In-situ TEM unveils the secrets of two-dimensional material nucleation
In-situ TEM unveils the secrets of two-dimensional material nucleation , Liu, Lei, Zhang, Xiaotian, Li, Taotao, Wang, Xinran In-situ TEM unveils the secrets of two-dimensional material nucleation , Liu, Lei, Zhang, Xiaotian, Li, Taotao, Wang, Xinran
- 27 May 2026
- Published by Royal Society of Chemistry (RSC) in Nanoscale
- Vol. 18 (25) , 13313-13327
CrossMicroNet: a cross-scale small-sample image restoration framework for two-dimensional material microscopy imaging
High-quality microscopy is central to resolving the growth behavior, morphology, lattice organization and defect landscape of two-dimensional (2D) materials. Yet microscopy data acquired across scales are degraded by fundamentally different mechanisms: in situ optical microscopy (OM) is often compromised by motion blur, defocus, vibration-induced smearing and illumination inhomogeneity, whereas scanning transmission electron microscopy (STEM) is strongly affected by beam-induced amorphous carbon contamination. Here we introduce CrossMicroNet, a unified cross-scale image-clarification framework that couples a shared restoration front end with modality-adaptive refinement. The restoration module integrates contrast-limited adaptive histogram equalization, mild non-local means denoising, blind deconvolution, ringing suppression and wavelet-domain spatial-channel enhancement. For OM, this front end directly sharpens domain boundaries from small-sample data without requiring paired optical ground truth. For STEM, the restored output is further processed by a lightweight contamination-suppression branch that combines conservative structural guidance with smooth fusion to attenuate diffuse background haze while preserving lattice periodicity. Evaluated on 28 OM/SEM region pairs, where SEM serves only as an approximate structural reference, CrossMicroNet reduces the apparent edge-transition width to 0.22 μm and yields the most favorable overall trade-off among NIQE, LPIPS and PSNR-like structural-reference metrics. On the STEM benchmark, it achieves the best learning-based performance, with a PSNR of 20.50 dB, SSIM of 0.85, LPIPS of 0.08, VIF of 0.92 and FSIM of 0.93. Fourier-domain analysis further confirms suppression of low-frequency contamination while retaining lattice-frequency features. These results establish CrossMicroNet as a practical cross-scale clarification strategy for linking growth-scale OM with atomic-scale STEM in 2D-material research.
- 27 May 2026
- Published by Royal Society of Chemistry (RSC) in Nanoscale
- Vol. 18 (25) , 13313-13327
CrossMicroNet: a cross-scale small-sample image restoration framework for two-dimensional material microscopy imaging
High-quality microscopy is central to resolving the growth behavior, morphology, lattice organization and defect landscape of two-dimensional (2D) materials. Yet microscopy data acquired across scales are degraded by fundamentally different mechanisms: in situ optical microscopy (OM) is often compromised by motion blur, defocus, vibration-induced smearing and illumination inhomogeneity, whereas scanning transmission electron microscopy (STEM) is strongly affected by beam-induced amorphous carbon contamination. Here we introduce CrossMicroNet, a unified cross-scale image-clarification framework that couples a shared restoration front end with modality-adaptive refinement. The restoration module integrates contrast-limited adaptive histogram equalization, mild non-local means denoising, blind deconvolution, ringing suppression and wavelet-domain spatial-channel enhancement. For OM, this front end directly sharpens domain boundaries from small-sample data without requiring paired optical ground truth. For STEM, the restored output is further processed by a lightweight contamination-suppression branch that combines conservative structural guidance with smooth fusion to attenuate diffuse background haze while preserving lattice periodicity. Evaluated on 28 OM/SEM region pairs, where SEM serves only as an approximate structural reference, CrossMicroNet reduces the apparent edge-transition width to 0.22 μm and yields the most favorable overall trade-off among NIQE, LPIPS and PSNR-like structural-reference metrics. On the STEM benchmark, it achieves the best learning-based performance, with a PSNR of 20.50 dB, SSIM of 0.85, LPIPS of 0.08, VIF of 0.92 and FSIM of 0.93. Fourier-domain analysis further confirms suppression of low-frequency contamination while retaining lattice-frequency features. These results establish CrossMicroNet as a practical cross-scale clarification strategy for linking growth-scale OM with atomic-scale STEM in 2D-material research.
- 26 May 2026
- Published by Springer Nature in Nature Electronics
- Vol. 9 (8) , 887-896
A bit-parallel molybdenum disulfide computer built through multi-level co-optimization
- 26 May 2026
- Published by Springer Nature in Nature Electronics
- Vol. 9 (8) , 887-896
A bit-parallel molybdenum disulfide computer built through multi-level co-optimization
- Top Cited Papers
- 29 January 2026
- Published by American Association for the Advancement of Science (AAAS) in Science
- Vol. 391 (6784) , 494-498
Kinetic acceleration of MoS 2 growth by oxy-metal-organic chemical vapor deposition
Kinetics determine the growth behavior of thin films, particularly for atomically thin transition-metal dichalcogenides. Metal-organic (MO) chemical vapor deposition (CVD) offers promise for scalable growth, but the reactions are kinetically limited, leading to nanometer-scale domain size and carbon contaminations. Here, we unveil the fundamental kinetic limitations and overcome them by introducing oxygen-assisted MOCVD (oxy-MOCVD) technology. By tuning reactions with oxygen, MO precursors are converted into high-purity transition-metal oxides and chalcogens, producing aligned molybdenum disulfide (MoS 2 ) domains with a size and growth rate that are orders of magnitude larger than conventional MOCVD. The MoS 2 is free of carbon impurities and exhibits average mobility exceeding 100 square centimeters per volt per second. The scalability of oxy-MOCVD is demonstrated by 150-millimeter single-crystal MoS 2 wafers, proving the feasibility of industrial-scale production.
- 29 January 2026
- Published by American Association for the Advancement of Science (AAAS) in Science
- Vol. 391 (6784) , 494-498
Top Cited PapersKinetic acceleration of MoS 2 growth by oxy-metal-organic chemical vapor deposition
Kinetics determine the growth behavior of thin films, particularly for atomically thin transition-metal dichalcogenides. Metal-organic (MO) chemical vapor deposition (CVD) offers promise for scalable growth, but the reactions are kinetically limited, leading to nanometer-scale domain size and carbon contaminations. Here, we unveil the fundamental kinetic limitations and overcome them by introducing oxygen-assisted MOCVD (oxy-MOCVD) technology. By tuning reactions with oxygen, MO precursors are converted into high-purity transition-metal oxides and chalcogens, producing aligned molybdenum disulfide (MoS 2 ) domains with a size and growth rate that are orders of magnitude larger than conventional MOCVD. The MoS 2 is free of carbon impurities and exhibits average mobility exceeding 100 square centimeters per volt per second. The scalability of oxy-MOCVD is demonstrated by 150-millimeter single-crystal MoS 2 wafers, proving the feasibility of industrial-scale production.
- Open Access
- 13 January 2026
- Published by Springer Nature in Nature Electronics
- Vol. 9 (1) , 116
Author Correction: Scaled crystalline antimony ohmic contacts for two-dimensional transistors
- 13 January 2026
- Published by Springer Nature in Nature Electronics
- Vol. 9 (1) , 116
Open AccessAuthor Correction: Scaled crystalline antimony ohmic contacts for two-dimensional transistors
- 17 December 2025
- Published by Springer Nature in Nature Electronics
- Vol. 8 (12) , 1191-1200
Scaled crystalline antimony ohmic contacts for two-dimensional transistors
- 17 December 2025
- Published by Springer Nature in Nature Electronics
- Vol. 8 (12) , 1191-1200
Scaled crystalline antimony ohmic contacts for two-dimensional transistors
- Top Cited Papers
- 23 October 2025
- Published by American Association for the Advancement of Science (AAAS) in Science
- Vol. 390 (6771) , eaea0849
Robust epitaxy of single-crystal transition-metal dichalcogenides on lanthanum-passivated sapphire
Two-dimensional (2D) transition-metal dichalcogenide (TMDC) semiconductors are promising materials for beyond-silicon electronics, but the growth of single-crystalline TMDCs has been limited to small wafer sizes in laboratory settings. We report the epitaxy of 150-millimeter single-crystalline TMDC wafers on lanthanum-passivated c -plane sapphire. The single atomic layer of lanthanum reduces the surface symmetry and increases the energy difference between antiparallel domains by as much as 200 times, leading to unidirectional domain alignment. We grew single-crystalline molybdenum disulfide (MoS 2 ), molybdenum diselenide (MoSe 2 ), tungsten disulfide (WS 2 ), and tungsten diselenide (WSe 2 ) by means of both chemical vapor deposition (CVD) and metal-organic CVD processes. Wafer-scale spectroscopies and device measurements demonstrate the exceptional quality and uniformity of 150-millimeter TMDCs, with average mobility of 110 and 131 square centimeters per volt per second for MoS 2 and WSe 2 , respectively, at room temperature.
- 23 October 2025
- Published by American Association for the Advancement of Science (AAAS) in Science
- Vol. 390 (6771) , eaea0849
Top Cited PapersRobust epitaxy of single-crystal transition-metal dichalcogenides on lanthanum-passivated sapphire
Two-dimensional (2D) transition-metal dichalcogenide (TMDC) semiconductors are promising materials for beyond-silicon electronics, but the growth of single-crystalline TMDCs has been limited to small wafer sizes in laboratory settings. We report the epitaxy of 150-millimeter single-crystalline TMDC wafers on lanthanum-passivated c -plane sapphire. The single atomic layer of lanthanum reduces the surface symmetry and increases the energy difference between antiparallel domains by as much as 200 times, leading to unidirectional domain alignment. We grew single-crystalline molybdenum disulfide (MoS 2 ), molybdenum diselenide (MoSe 2 ), tungsten disulfide (WS 2 ), and tungsten diselenide (WSe 2 ) by means of both chemical vapor deposition (CVD) and metal-organic CVD processes. Wafer-scale spectroscopies and device measurements demonstrate the exceptional quality and uniformity of 150-millimeter TMDCs, with average mobility of 110 and 131 square centimeters per volt per second for MoS 2 and WSe 2 , respectively, at room temperature.