CRS_CUHK401/22
Superradiant Time-crystal Maser and Applications in Quantum Metrology

Hong Kong Project Coordinator Prof Renbao Liu / The Chinese University of Hong Kong
Mainland Project Coordinator: Prof Xing Rong / Tsinghua University

We will demonstrate a novel class of out-of-equilibrium quantum matters – superradiant time-crystal maser and develop ultra-precise quantum metrology using the maser. Laser and maser, with many photons in coherent states, are important quantum matters, and, owing to their excellent coherence, are the basis of many metrological technologies. In conventional laser and maser, the emitters are usually incoherent. Advances in quantum information technologies make available emitters that have good coherence. Coherent emitters can collectively radiate photons, leading to the so-called superradiance. Superradiant laser or maser has coherence time increasing with the density of emitters. However, the interactions and disorder in materials with high-density emitters would destroy the coherence. To address this issue, we propose to develop superradiant time-crystal maser. Time-crystalline phases occur in quantum many-body systems under periodic driving when the discrete time-translation symmetry is spontaneously broken. Time crystals feature long-lived collective coherence, and more importantly, the interactions and disorder can contribute to sustaining the quantum oscillations, turning a disadvantage into an advantage.

To realize the superradiant time-crystal maser, we will consider interacting nuclear spins and color-centre spins in solids under periodic driving and coupled to high-quality resonators. To achieve the gain conditions for masing, optical pump and polarization transfer will be employed. We will model the coupled systems using the Floquet-state formalism and identify the conditions for the onset of discrete time crystals and then the superradiant time-crystal maser. The photon statistics of the maser, especially the high-order correlations beyond the conventional coherence theory, will be calculated and measured. We will utilize the exotic photon statistics for detecting correlations in quantum many-body systems that are inaccessible to classical lights. We will develop ultra-sensitive magnetometers that exploit the superb coherence and unique photon statistics of the superradiant time-crystal maser and apply them to discovering fundamental physics such as dark matter.

This project is based on long-term, fruitful collaborations. The two teams were awarded the NSFC-RGC grants twice and published jointly in top journals (including Nature, Nature Physics, and Physical Review Letters). The theoretical team in Hong Kong comprises of leaders in diamond maser, quantum sensing, quantum nonlinear spectroscopy, and quantum matters; the experimental team in Hefei pioneers in Floquet maser, spin-based quantum simulations, and detecting dark matter using spin resonance spectroscopy.

The discovery of new out-of-equilibrium quantum matters and the development of unprecedented quantum metrology will have long-term impacts on fundamental physics and quantum information technologies.

CRS_CUHK405/22
Measuring China’s Industrial Policies and Evaluating Their Effects in a Unified Quantitative Framework

Hong Kong Project Coordinator Prof Michael Zheng Song / The Chinese University of Hong Kong
Mainland Project Coordinator: Prof Chong-en Bai / Tsinghua University

Industrial policies have been extensively employed by governments in many developed and developing countries with the objective of promoting economic growth. Yet, the existing research on industrial policy is surprisingly thin. This project is to initiate a comprehensive quantitative analysis on industrial policy. We will take China as an ideal laboratory, where a large number of industrial policies have been carried out by the central and local governments in various forms and the effects of some industrial policies have spilled over to the rest of the world.

The project consists of three parts. We will first scrape all the online documents released by the major departments in the central and local governments and adopt textual analysis methods to extract the relevant information on industrial policy. This will constitute a novel dataset that provides, for the first time, a panoramic view of all kinds of industrial policies planned by the Chinese government. The second part is to develop a unified quantitative framework that can quantify various industrial policies actually implemented from firm-level data in an internally consistent way. The unified framework builds a general theoretical foundation for industrial policy. The framework can structurally interpret some key moments in the firm-level data as the magnitude of industrial policy. The findings will provide an external validity check of the consistency between the planned and actual industrial policies.

The last part is to take the measurements to evaluate the effects of industrial policy in the unified framework. An important ingredient of the evaluation is that we take into account the fact that many industrial policies are enacted and implemented by local government. Some local industrial policies may improve the local welfare but harm the aggregate welfare of the country. Likewise, the effects of purely domestic industrial policies might spill over to the rest of the world, filling the global economy with consternation. We will rank the industrial policies by their local, national and global welfare effects, respectively. The results will help open the black box of industrial policy and make its welfare implications highly transparent. Finally, we will discuss the implications of this study in the development of the Greater Bay Area and the policies targeting major technology bottlenecks.

CRS_HKBU201/22
Understanding the Molecular Mechanism Linking mRNA Decay and Capping with Post-transcriptional Gene Silencing

Hong Kong Project Coordinator Prof Yiji Xia / Hong Kong Baptist University
Mainland Project Coordinator: Prof Hongwei Guo / Southern University of Science and Technology of China

Genes determine traits of living organisms. A gene is used as a template to produce RNA in a process termed transcription. Messenger RNA (mRNA) is then used as a template for protein synthesis in a process termed translation. To function properly, mRNAs are modified in different ways during and after synthesis, including splicing and adding a cap and a tail. mRNA quality is closely monitored and abnormal or unneeded mRNAs can be decayed by nucleases to prevent formation of abnormal or unneeded proteins. If the RNA decay process becomes abnormal, mRNA might trigger formation of small interfering RNA (siRNA). siRNA can act through the post-transcriptional gene silencing (PTGS) pathway to inhibit protein translation. However, inappropriate biogenesis of siRNA is often detrimental to living organisms.

In this study, we will use Arabidopsis as a model organism to study biogenesis of ct-siRNA (a type of siRNA) from mRNA when RNA decay becomes abnormal or the RNA cap is not properly formed. Our preliminary results indicate that biogenesis of ct-siRNAs is dependent on features in the gene/mRNA sequences that might affect mRNA translatability. We will use genetics, molecular biology, bioinformatics, and other methods to reveal how mRNA sequence features and cap modifications affect translatability and ct-siRNA biogenesis. We will also study how plants use these processes to regulate gene expression in response to environmental stresses. This project will advance our understanding of the molecular mechanism that integrates these gene regulation processes to mediate normal growth and development.

CRS_HKU703/22
Investigation of Developmental Potency Association with Chromatin Structure in Preimplantation Embryos and Expanded Potential Stem Cells by Single Cell Multi-omics

Hong Kong Project Coordinator Prof Pengtao Liu / The University of Hong Kong
Mainland Project Coordinator: Prof Baofa Sun / Nankai University

In recent years, multi-omics approaches have greatly sped up the discovery in biomedical research fields. However, the cellular and molecular mechanisms in mammalian development from a fertilized egg to a fully functional individual remains to be fully elucidated. Cells are the basic building blocks of all independent living organisms. The developmental process involves in cell fate determination and cell lineage development and is tightly regulated at multiple levels such as chromatin, transcriptional, translational and post-translational activities.

This project uses single-cell multi-omics technology to systematically investigate, at single cell level, the similarities and differences between normal in-vivo embryonic cells and in-vitro cultured expanded potential stem cells (EPSCs) of multiple mammalian species established in our laboratory, including mouse, porcine, bovine, and human. The findings from the proposed study is expected to reveal how the 3-dimention chromatin structure regulates specific sets of gene expression and subsequently determines cell fate in early embryonic development, and to discover why EPSCs have expanded development potential. The knowledge learned can provide a better understanding of basic human biology and enable development of an improved stem cell system for fundamental research, for regenerative medicine and other healthcare services and for biotech and agriculture.

CRS_HKUST601/22
Event-triggered Learning for Neuromorphic Sensing with Application in Robotics

Hong Kong Project Coordinator Dr Ling Shi / The Hong Kong University of Science and Technology
Mainland Project Coordinator: Dr Dawei Shi / Beijing Institute of Technology

This project is oriented on the frontier area of neuromorphic sensing and information processing, and aims to explore the key research question of “how to achieve real-time dynamic target feature reconstruction with neuromorphic event sensing information”. Considering the event-driven asynchronous update, ultra-sparse representation, and large-scale parallelism characteristics of the event sensing data, we aim to analyze the generation mechanism and mathematical representation of event streams, and design new observability and identifiability notions that ensure the solvability of the feature reconstruction problems; construct recursive algorithms for event-triggered estimation and learning and propose the corresponding low-complexity fast implementation methods; characterize the sufficient/necessary conditions that are needed to ensure the stability of the estimation error dynamics and asymptotic learning performance; and complete the application and verification of the theoretical method in remote pose estimation for a wheel-legged robotic system. The project is expected to develop a relatively systematic feature reconstruction approach that has both theoretic performance guarantee and engineering implementability for event sensor feature reconstruction tasks, and will provide theoretical foundation and methodological support for the development of the interpretable neuromorphic signal processing technology.

CRS_HKUST603/22
Native AI Empowered Next Generation Wireless Networks

Hong Kong Project Coordinator Prof Jun Zhang / The Hong Kong University of Science and Technology
Mainland Project Coordinator: Prof Shi Jin / Southeast University

Wireless communication technologies have profoundly changed every aspect of our daily lives and largely reshaped various industries. Their impact has been accelerated by the recent deployment of 5G networks. Qualcomm estimates that 5G will increase the global economic value of the digital economy by US$13.2 trillion by 2035. These great successes and substantial impacts drive the further evolution of wireless networks. To support a plethora of emerging applications (e.g., augmented/virtual reality, smart city, and industrial automation) and to fulfil the increasing user demands for ultra-low latency, ultra-high data rates, and a truly immersive user experience, next-generation wireless networks will become ever more powerful and complex. This trend will result in formidable design challenges, which cannot be addressed by conventional techniques and methodologies, and place a tremendous burden on network operating and capital expenses. Inspired by recent breakthroughs in artificial intelligence (AI), there has been intense interest among industry and academia in developing AI-enabled techniques for the automatic design and cost-effective operation of wireless networks. Meanwhile, AI as a service (AIaaS) has been envisioned as a critical new application scenario for beyond 5G networks, which are expected to provide seamless support for ubiquitous AI-enabled services. AIaaS will facilitate exciting use cases (e.g., smart healthcare, autonomous vehicles, and service robots) and innovative business opportunities, while giving rise to new performance requirements for wireless networking.

Led by world-renowned researchers in wireless communications, this collaborative research project will develop an AI-native air interface with deep learning-based physical (PHY) and medium access control (MAC) layer technologies for future wireless networks, while providing native support for wireless AI applications. Unlike existing studies that follow a plug-and-play approach and directly adopt AI models developed in other domains, we will develop AI-based methods specifically for wireless networks. Special attention will be paid to the important practical aspects including generalization, scalability, adaptivity, and robustness. We will develop model-driven deep learning methods for the PHY layer by exploiting the algorithmic and problem structures, and design learning-based radio resource management algorithms for the MAC layer by exploiting the characteristics of the wireless network topology. Moreover, as conventional data-oriented design principles are no longer effective for wireless AI applications, we will design learning-driven networking to support distributed training and propose task-oriented communication strategies to enable low-latency collaborative inference. Finally, simulation platforms and prototype systems will be built to validate the proposed methods and algorithms.

CRS_HKUST604/22
Development of Sulfide-based High-Energy Density Solid-State Lithium Batteries

Hong Kong Project Coordinator Dr Minhua Shao / The Hong Kong University of Science and Technology
Mainland Project Coordinator: Dr Yong Yang / Xiamen University

Solid-state batteries (SSBs) have attracted great attention owing to their safety and high energy densities that can potentially reach 600 Wh kg-1. However, the energy density of current SSBs based on lithium (Li) metal oxide cathode materials is lower than 400 Wh kg-1 at best. This collaborative project aims to develop sulfide-based SSBs with unprecedented performance with an energy density higher than 500 Wh kg-1. The outcomes of this project will have a great impact on the development of next-generation SSB technology and take it closer to commercialization than ever before.

CRS_HKUST605/22
Investigation of Molecular Classification and Clonal Evolution of Diffuse Glioma through Quantitative Proteomics and Pan-omics Integration

Hong Kong Project Coordinator Prof Jiguang Wang / The Hong Kong University of Science and Technology
Mainland Project Coordinator: Prof Tao Jiang / Beijing Neurosurgical Institute

Diffuse gliomas are the most common and highly heterogeneous malignant brain tumors in adults. The latest 2021 World Health Organization (WHO) classification of tumors of the central nervous system classifies adult diffuse gliomas into three molecular subtypes with distinct survival outcomes primarily based on histopathology and genomics. Yet this classification has limited impact on clinical intervention and the fact that aggressive gliomas remain incurable does not change.

Our inter-disciplinary team has been focusing on the identification of clinically relevant glioma drivers and actionable drug targets in East Asian populations and has achieved fruitful results, such as the identification and clinical application of METex14 (Nat Genet 2018, Cell 2018) and the elucidation of MGMT gene fusion driving chemoresistance (Nat Commun 2020). Empowered by the maturing quantitative proteomics technologies, proteomic profiling will provide new opportunities to investigate how proteomes and metabolisms are altered in cancer development and progression and how they can be used to improve glioma classification and intervention.

Leveraging the well-established collaboration between the Hong Kong team and the mainland team, we propose to cross-sectionally and longitudinally characterize gliomas at the multi-omics level to identify clinically relevant subtypes, novel biomarkers, and non-stochastic patterns of cancer evolution thereby improving diagnosis and therapeutics. Firstly, matched proteomic, genomic, and transcriptomic profiling of at least 500 Chinese glioma patients will be analyzed to refine disease subtypes based on network integration of pan-omics. Secondly, innovative methods will be deployed to identify single-molecular biomarkers that constitute conventional and de novo peptides, as well as network biomarkers that correlate with patient prognosis and/or treatment response. Thirdly, multi-sectional and longitudinal data will be used to track the dynamic changes of recurrent gliomas, especially in the proteome, to better understand clonal evolution under treatment. Finally, translational research back to the bedside will be explored by combining well- calibrated machine learning methods, our key findings at the bench, and close collaboration with the patient care clinical team at Beijing Tiantan Hospital.

There is no doubt that this project will provide invaluable data resources to the glioma research community, advance our biological knowledge about glioma development and evolution, and improve clinical stratification of gliomas for better patient management and treatment. If successful, this project has the potential to refine the WHO classification of adult brain tumors and bring us one step closer to precision neuro-oncology.

CRS_PolyU502/22
Monolithically Integrated Electronics with Two-Dimensional Semiconductors – From Controllable Growth to Device Integration

Hong Kong Project Coordinator Dr Yang Chai / The Hong Kong Polytechnic University
Mainland Project Coordinator: Prof Wenjing Zhang / Shenzhen University

In the past decades, state-of-the-art silicon-based integrated circuits have been kept shrinking for high density and high performance. However, the silicon-based electronics are running into their physical limits due to significant carrier scattering at nanoscale, which requires new semiconductor materials for more efficient operation. Atomically thin two-dimensional (2D) layered materials are considered as promising semiconductors for making energy-efficient nanoscale devices. However, the systematic research from controllable materials growth to large-scale device integration is still relatively unexplored at the current stage. To realize large-scale monolithic integrated circuits based on 2D semiconductors, we propose to study the controllable large-area growth, fundamental understanding of the device physics, and large-scale integration technology. First, we aim to grow wafer-scale, single-crystalline and continuous monolayer 2D semiconductors and develop non-destructive method to transfer wafer-scale semiconductors to targeted substrates. Second, we will optimize the contact and interface of the device to further improve the performance. Third, we will develop the processing technology for large-scale device integration and realize both memory and computation functions.

CRS_PolyU504/22
Long-cycle and High-energy-density Flexible Li Batteries Using Hollow Multishelled Structure and Hierarchical Composite Electrode

Hong Kong Project Coordinator Prof Zijian Zheng / The Hong Kong Polytechnic University
Mainland Project Coordinator: Prof Mei Yang / Institute of Process Engineering, Chinese Academy of Sciences

The rapid development of wearable electronics has enabled a wide range of future applications such as point-of-care health monitoring and medicine, interactive virtual reality sensing and actuation, soft robotics, and Internet of Things (IoTs). These emerging applications impose unprecedented demand on flexible batteries, which are indispensable to realize seamless and low-profile design form factors. Because of the limited space allowed to accommodate the battery and the frequent need for deformation during wear, these flexible batteries are expected to possess not only high energy density, but also outstanding electrochemical and mechanical stability. Nevertheless, commercial Li-ion batteries (LIBs) show little flexibility due to the poor elasticity of the battery components. Although much effort has been spent in developing flexible LIBs in the past decade, the flexibility was achieved at a huge cost of sacrificing the energy density and stability of the battery.

To address the critical tradeoff, this collaborative project aims to develop long-cycling and high energy-density flexible Li batteries by rational design of three-dimensional (3D) hierarchical composite electrodes, so as to replace conventional lamellar-like electrodes stacked on metal foils. The study will be conducted based on the complementary expertise and the well-established collaborative track of record between the two teams from Hong Kong and the mainland. The Hong Kong team will focus on the development of ultra-lightweight, ultrathin, and highly flexible 3D scaffolds. The mainland team will synthesize high-capacity and stable cathodic and anodic materials by infiltrating high-capacity electrode materials into hollow multishelled (HoMS) structures and modifying the surface with catalytic dopants and/or single atoms. We will then fabricate composite electrodes via coupling the 3D scaffolds with HoMS-based electrode materials, and study the interfacial, chemical, electrical, electrochemical, and mechanical properties of the developed materials and electrodes via both experimental and computational approaches. Finally, we will deliver flexible Li batteries with high energy density, long cycling life, and outstanding flexibility. The success of the project will make a paradigm shift in the materials and structure design of batteries, and improve the performance of flexible batteries by at least 2 folds in comparison to the state of the art. This performance will satisfy most of the energy needs for future flexible and wearable electronics. The acquired knowledge is likely to be applicable to other research fields such as supercapacitor and catalysis.