N_CUHK404/14
Key Functional Photonic Elements for On-chip Mode-multiplexed Optical Interconnects

Hong Kong Principal Investigator: Prof Hon-ki Tsang (The Chinese University of Hong Kong)
Mainland Principal Investigator: Prof Daoxin Dai (Zhejiang University)

Optical interconnects for on-chip global interconnects have attracted much interest in recent years because they provide a potential solution to the problem of thermal and power consumption limitations of copper interconnects, which already severely constrain the maximum clock frequencies that may be used in microprocessors. While state of the art CMOS transistors can switch at frequencies of many tens of gigahertz, and CMOS RF systems operating at 60GHz have been available commercially for several years now, desktop microprocessors today are still limited to ~5GHz clock frequencies because of the extreme heat density and associated practical problem of dissipating the heat generated if the dense electrical interconnects were operated at higher frequencies. The use of dense wavelength division multiplexing (DWDM) can provide an order of magnitude improvement in on-chip data throughput at the same power consumption as electrical interconnects. The use of DWDM, rather than point to point interconnects using different single mode waveguides was found to be necessary because of the limited area available for on-chip optical interconnects and the high density of interconnects that are needed. However the high cost of implementation of DWDM optical network-on-chip, with tens of wavelength channels from an external array of precise wavelength lasers (requiring temperature control and wavelength selective elements and wavelength for each optical to electrical conversion), is one of the key obstacles preventing practical implementation of optical interconnects. In this project we shall develop the functional elements for implementing on chip optical interconnects based on an alternative approach which can operate with only a single wavelength source, and which uses the relatively unexplored concept of mode division multiplexing (MDM) for on-chip networks. We propose to develop the basic functional elements of MDM including the mode converters (equivalent to wavelength converters in DWDM systems), mode switches, grating arrays for optical fiber to MDM network interface, selective mode modulators and other functional elements that will be needed to construct high capacity on-chip MDM networks. We shall also study the theoretical constraints in communications capacity from inter-mode crosstalk from mode scattering and unintended mode coupling at bends and tapers, and constraints from nonlinear effects, such as two photon absorption, in the silicon waveguides. The project aims to demonstrate the feasibility of MDM for network on chips, advance the technology of on-chip optical interconnects using MDM, and ultimately enable future microprocessors to operate at much higher clock frequencies.

N_HKUST615/14
Key technologies for the next-generation real-time high-resolution minimally-invasive implantable Electrocorticography (ECoG) system

Hong Kong Principal Investigator: Prof Jie Yuan (The Hong Kong University of Science and Technology)
Mainland Principal Investigator: Prof Zhihua Wang (Tsinghua University)

Human brain holds the deepest mystery to our knowledge of the Nature. A major difficulty of brain research is the lack of an effective technological tool to access it. The invasiveness of brain surgeries is the biggest hurdle for their acceptance. The complication of inflection caused by these invasive methods prevents the diagnosis and treatment of many neurological diseases, such as epilepsy.

Brain machine interface (BMI) is a technology emerging worldwide in recent years. Existing BMI technologies are mainly non-invasive EEG technologies. By combining other sensing modality, such as the blood oxygen level, BMI has been demonstrated to sense some cognitive and motor ideas in the cortex with good accuracy. Nonetheless, as it does not access the brain, non-invasive BMI has fundamental difficulty to serve as effective tool for brain study and neurological disease diagnosis.

Instead, minimally-invasive ECoG implant could be the tool to access the brain while minimizing the trauma caused by the implantation. Although preliminary research has been pursued worldwide, major technological breakthrough in the following four aspects is needed to turn the ECoG implant into a viable tool for clinical usage.
1. Invasiveness: Existing ECoG implants are big with huge power consumption, which is due to the many chips and passive components involved. A future ECoG implant should integrate all functions in one chip and package the chip and passive components on flexible bio-compatible substrate;
2. Implantable: Future ECoG implant should be powered wirelessly from outside. As a result, the energy for the implant is very stringent. Considering the multi-channel signal acquisition and high-speed data transmission tasks, the power needs to be used very efficiently in the implant;
3. Spatial resolution: Many neurological diseases are due to cellular changes. It is important for the ECoG implant to access the cortical tissue down to the cellular level;
4. Multiple modalities: Besides the field potential, our brain is closely influenced by many factors, such as neurotransmitters, oxygen levels. Future ECoG implant should be able to record these modalities simultaneously.

In this project, we will develop novel technologies on the four aspects for the next-generation of ECoG implant and lead to a prototype ECoG implant to test on free-moving rats. This inter-disciplinary project is collaborated by two teams of experienced researchers with strong track records of working at the frontier of bio-medical circuits and systems. The team also has the access to advanced facilities on micro-fabrication, circuit design and testing, biological experiments. Besides, technologies in this project have strong commercial prospects. The project team has the strong support from a world-leading foundry.

N_HKUST627/14
III-V-on-Silicon Coupled-Resonator-Optical-Waveguide Lasers for Direct-Modulated Multi-Wavelength Emission and Active Mode-Locking

Hong Kong Principal Investigator: Prof Andrew W O Poon (The Hong Kong University of Science and Technology)
Mainland Principal Investigator: Prof Yong Zhen Huang (Institute of Semiconductors, Chinese Academy of Sciences)

Due to the rise of Big Data, cloud computing, and many fast-emerging mobile and web-based datacom applications, demands on data capacity keep growing rapidly. Conventional copper electrical interconnects between computers within a datacenter and inside a computer are becoming bandwidth-limiting and causing excessive electrical power consumption. One potential solution is the optical interconnects technology, which offers many key advantages including a high data bandwidth and potentially a low-power consumption. An integrated on-chip light source is a key component to implement such a technology with mass production and low cost.

In this project, we aim to develop a novel coherent array of coupled semiconductor laser sources in a compact footprint suitable to be integrated on a silicon chip, based on our current research on passive coupled oscillators and hybrid silicon lasers. The proposed laser sources can potentially emit at multiple wavelengths, with multiple spatial output ports and can be directly modulated at a high data rate. It can also potentially work in a pulsed mode with regular pulse trains or pseudo-random waveforms. Such on-chip laser sources, if successfully developed, can serve as a multi-functional building block to enable novel functionalities or network architectures in next-generation high-performance computing systems and data communications.

N_HKU728/14
Understanding the evolution and interspecies transmission of betacoronaviruses by structural and biophysical approaches

Hong Kong Principal Investigator: Dr Kwok Yung Yuen (The University of Hong Kong)
Mainland Principal Investigator: Dr George Fu Gao (Chinese Academy of Sciences)

Animals are important source of emerging infections. The source of human SARS coronavirus has been successfully traced to Civets and finally bats. The interaction between virus attachment protein and host cell receptor is the key event governing the jumping of interspecies barrier. We will use technologies of molecular and structural biology to understand the interaction between bat and human virus and their host receptors. Findings will help to predict what coronavirus may jump from animals to human in the future.

N_HKU736/14
Schizophrenia-related de novo and compound heterozygous mutations

Hong Kong Principal Investigator: Prof Pak Chung Sham (The University of Hong Kong)
Mainland Principal Investigator: Prof Tao Li (Sichuan University)

Schizophrenia (SZ) is a chronic and disabling mental disorder affecting affecting approximately 1% of the population worldwide. Genetic variation accounts for much of individual difference in susceptibility to SZ, with over 100 common variants being recently discovered by genome-wide association study (GWAS). This study aims to uncover rare variants that increases predisposition to SZ, through whole-exome sequencing (WES) of in case-parent trios (i.e. affected patients and both unaffected parents). This study design will allow both de novo mutations and compound heterozygous mutations to be detected. We propose to conduct WES on 100 Han Chinese case-parent trios (50 from Chengdu and 50 from Hong Kong). The proposed study will be among the first to systematically screen the entire exome for de novo and compound heterozygous mutations that confer increased risk to SZ in the Han Chinese population. Compared to loci discovered by GWAS, these mutations are likely to have a much bigger impact on disease risk. The nature of the genes located at these mutations will throw light on the pathophysiological processes underlying SZ.

N_HKU740/14
Nur77: new insights in signaling and mechanism of activation of epithelial-mesenchymal transition and tumor metastasis

Hong Kong Principal Investigator: Dr Alice Sze Tsai Wong (The University of Hong Kong)
Mainland Principal Investigator: Dr Jin Zhang Zeng (Xiamen University)

Most cancer deaths result from tumor metastasis, and understanding the underlying mechanisms is of obvious importance. Epithelial-mesenchymal transition (EMT) is a first and rate-limiting step in the metastatic cascade. Recently, we show for the first time a role for Nur77 in the loss of E-cadherin expression and acquisition of N-cadherin expression that promotes EMT and colon cancer cell metastasis. Interestingly, Nur77-dependent EMT is possibly mediated through a novel Nur77-mTORC1 signaling. mTORC1 is frequently activated in a variety of tumors and represents one of the most common dysregulated signaling in cancers. Our pilot experiments show that Nur77 may regulate mTORC1 activity through interaction with FKBP38 and Rheb, which can be regulated by honokiol and its derivatives. In this proposal, we will employ honokiol and its derivatives to investigate the role and regulation of Nur77-mTORC1 signaling axis in EMT and tumor metastasis. This study will provide an important understanding of the novel mechanisms of Nur77 involved in tumor metastasis, and also insight on the development of new and more efficacious therapeutic targets.

N_HKU726/14
Fatty acid binding protein-4 as a mediator of autoimmune diabetes: from molecular mechanism to clinical significance

Hong Kong Principal Investigator: Prof Aimin Xu (The University of Hong Kong)
Mainland Principal Investigator: Prof Zhiguang Zhou (Central South University)

Type 1 diabetes (T1D) is the most common chronic diseases in children and adolescents and the most severe type of diabetes. This disease is not curable at this stage, and T1D patients need lifelong medication with insulin. T1D is caused by self-destruction of insulin-producing £]-cells in pancreas, resulting in the lack of insulin secretion and elevation of blood glucose.

We have recently found that fatty acid binding protein-4 (FABP4), a lipid-binding chaperone with pro-inflammatory properties, is an important mediator in inflammation and autoimmune diabetes. In both T1D patients and NOD mice (a well-established animal model of T1D), a marked elevation of FABP4 levels was found in circulation compared to their healthy controls. Delayed diabetes onset time and decreased T1D incidence were also observed in NOD mice after chronic administration with the pharmacological inhibitor of FABP4. Furthermore, the production of macrophage-derived pro-inflammatory cytokines induced by endotoxin or toxic lipids was compromised by genetic ablation or pharmacological inhibition of FABP4. These findings suggest that elevated FABP4 may contribute to the onset and/or progression of insulitis and £]-cell destruction by promoting the infiltration and activation of macrophages and/or by mediating the immune cells crosstalk leading to ultimate immune system imbalance.

In the present proposal, we plan to use a FABP4-deficient mouse model, in combination with adoptive immune cell transfer and ex vivo co-culture approaches to investigate what pathophysiological roles of FABP4 play in T1D and to dissect the molecular pathways FABP4 involved in and how it triggers the autoimmune destruction of pancreatic £]-cells. In addition, our mainland collaborator will use several unique clinical study cohorts to further explore the clinical relevance of FABP4 in the onset and progression of T1D. The results of this study will enhance our understanding towards the pathogenesis of autoimmune diseases and in the long-term will help to validate whether FABP4 is a potential biomarker for prognosis and a target for therapeutic intervention of T1D.

N_HKU735/14
Orexin-induced modulation of activity-dependent synaptic plasticity is critical for the maturation of vestibular circuitry and functions

Hong Kong Principal Investigator: Prof Ying Shing Chan (The University of Hong Kong)
Mainland Principal Investigator: Prof Jian Jun Wang (Nanjing University)

The key to learning and memory is change in plasticity of synapses in the brain. Neuropeptide orexin is known to play parts in motor control. How orexin modulates the plasticity of our balance system, thus affecting related motor function is not known. Our team has recently discovered that orexin excites brainstem vestibular neurons that have motor function in adult rats. Also, we revealed that postnatal refinement of synaptic plasticity in the vestibular network is crucial for establishment of a spatial map and behaviorial expression. To demonstrate importance of the orexinergic regulation, we will adopt a postnatal intervention approach to perturb orexin action and then to monitor for changes in (i) spatial map formation and (ii) spatial learning behavior. Results will illustrate how neuromodulators tune synaptic plasticity in the maturation of vestibular functions, thereby offering a platform for therapeutic strategies that could rescue synaptic disorders.

N_CityU116/14
Design and Optimizing Laser-Based 3D Printing of Metallic Glass: A Systematic Study of the Joining and Crystallization Mechanisms of Amorphous Structures under Laser Irradiation

Hong Kong Principal Investigator: Dr Yong Yang (City University of Hong Kong)
Mainland Principal Investigator: Prof Wei Hua Wang (Chinese Academy of Sciences (CAS))

Since the 1960s, metallic glass (MG) has been attracting tremendous research interest because of its unique combination of structural/functional properties, such as superb strength, high elastic limit, excellent thermoplastic formability, superior biocompatibility and etc. After decades of enduring efforts, today the research of MGs is mostly centered in finding new applications. In the literature, the new use of MGs has been frequently reported for a wide range of applications, such as renewable energy, healthcare, thin film technology, nano-devices and biomedical implants. Although the applications of MGs start booming, however, some longstanding issues still remain, among which one is the limitation of the obtainable size of bulk metallic glass (BMG) and another is the room-temperature brittleness. These 'roadblocks' have been hurdling the use of BMGs, particularly for structural applications in which BMGs outweigh most existing structural materials in strength.

To solve these problems, one possible solution that has been enthusiastically discussed recently is the use of the 3D printing technique on MGs. Through this layer additive technique (3D printing), we may build up BMGs in any shape/size with small-sized MG powders. Furthermore, it is envisioned that we may gain a control of the amorphous microstructure by carefully selecting the operational parameters of 3D printing such that the brittleness issue of BMGs can be resolved. Despite its promise, the research in 3D printing of MGs is still in its infant stage. There are quite a few fundamental/technical issues yet to be solved, such as the design of a MG alloy suitable for 3D printing, understanding the crystallization dynamics of small-sized MG powder under laser irradiation and etc. The proposal of the joint research between CityU and CAS is aimed to address these issues with their respective strength in alloy design, theoretical modeling, nanomechanical characterization and numerical simulations. For this joint research, our goal is to develop a 3D printing protocol and also a suitable MG alloy which fit each other in the production of strong yet ductile 3D printed MG samples. In the long run, a platform, as initiated through this project, will be established to bridge the two institutions, based respectively in Hong Kong and mainland China, in the MG-based 3D printing research for the support of the local manufacturing industries.

N_CUHK440/14
Metal/Oxide Nanostructures as Plasmonic Catalysts for the Synthesis of Organic Molecules

Hong Kong Principal Investigator: Prof Jianfang Wang (The Chinese University of Hong Kong)
Mainland Principal Investigator: Prof Chun-hua Yan (Peking University)

The aim of this proposal is to (i) synthesize (metal nanocrystal core)/(oxide semiconductor shell) nanostructures as plasmon-enhanced photocatalysts for organic transformations under visible/NIR light, (ii) understand plasmon-induced hot-carrier injection and photocatalysis in the synthesized core/shell nanostructures, and (iii) optimize the nanostructures for photocatalytic organic transformations.

Worldwide efforts are being made to intensively develop low-cost, highly efficient technologies for harvesting renewable energy resources and converting them into electricity. These efforts are also strongly stimulating changes and revolutions in many other fields. In chemistry, increasing efforts are being made on developing economical, energy-sustainable, and environment-friendly methods for organic synthesis with high activity and selectivity. The most attractive approach is to design new, highly active and selective, recyclable, and environmentally-benign catalysts. Such catalysts are required to not only satisfy the traditional needs of making valuable chemicals, fuels, and pharmaceuticals, but also fulfill the goal of green chemistry in atomic efficiency, waste minimization, energy saving, and efficient catalyst recovery.

Herein we propose to integrate plasmonic metal nanocrystals with metal oxide semiconductors for plasmon-enhanced chemical reactions under visible/NIR light. Metal oxide semiconductors are abundant, diverse, chemically and thermally stable, resistant to photo-corrosion, and environmentally benign. The band gap energies of most oxide semiconductors are in the UV region, while UV photons only accounts for a small fraction of sunlight. In addition, the product selectivity in UV-induced reactions with metal oxides is often unsatisfactory. Localized plasmon resonance endows noble metal nanocrystals with very strong optical responses to visible and NIR light. Plasmon excitation can generate hot charge carriers for catalysis. The integration of plasmonic metal nanocrystals with oxide semiconductors can achieve photocatalysis in the visible/NIR regions with high activity and selectivity, and therefore address the challenges of green chemistry. We will synthesize (metal core)/(oxide shell) nanostructures in our studies. The core/shell architecture increases the interfacial area between the two materials, which is beneficial to plasmon-enhanced/enabled processes. It can also protect the metal core from reshaping, aggregation, and chemical corrosion. We will strive to understand the plasmon-induced hot-carrier injection and plasmon-enhanced catalysis by performing extensive nanostructure syntheses, photoelectrochemical measurements, and catalytic reactions. According to the acquired knowledge and understanding, we will further optimize the core/shell nanostructure photocatalysts for different chemical reactions.

The success of this proposal will lead to the design of efficient visible or NIR metal/semiconductor photocatalysts for organic synthesis, the improvement in understanding plasmon-enhanced light harvesting in many photovoltaic and photocatalytic applications.

N_PolyU517/14
Photovoltaic, Spin Field Effect Transistor and Sensing Devices Based on Polar Oxide Heterostructural Two-dimensional Electron Gas

Hong Kong Principal Investigator: Prof Jiyan Dai (The Hong Kong Polytechnic University)
Mainland Principal Investigator: Prof Changgan Zeng (University of Science and Technology China)

Polar oxide heterostructural two dimensional electron gas (2DEG)refers to the conduction electrons at polar/polar or polar/nonpolar oxide heterostructure interface such as LaAlO3/SrTiO3 (LAO/STO). In this proposed project, we focus on: (1) study of photovoltaic effect in polar oxide heterostructural 2DEG, (2) study of the device physics of LAO/STO based magnetic spin-FET, and (3) sensing related new phenomena in such system and physics behind. We will achieve the following objectives: (1) to achieve deeper understanding on orientation and structure-dependent transport properties of polar oxide heterostructures; (2) to realize polar oxides heterostructural electronic devices from photovoltaic to spintronics and sensors based on the 2DEG. The outcomes from this project will not only promise the potential interest for understanding the oxide interfacial 2DEG but also its application in all-oxide devices, thus open a new route to complex oxide physics and ultimately for the design of devices in oxide electronics.

N_PolyU533/14
Development of Multifunctional Nanocomposite Particles for Imaging and Gene Therapy in Cancer Treatment

Hong Kong Principal Investigator: Prof Pei Li (The Hong Kong Polytechnic University)
Mainland Principal Investigator: Prof Yongsheng Li (East China University of Science and Technology)

The application of nanoparticles in the biomedical field is an emerging research area nowadays because of their exciting performances in bio-imaging, targeted drug and gene deliveries, sensors and so on. The nano-scaled size and high surface to volume ratio of nanoparticle are the key features which make them especially useful in medical treatment because of their many new properties, ease of functionalization, conjugation of biomolecules etc. Hence, this project aims to design and synthesize novel multi-functional bimetallic/polymer nanocomposite particles as novel and efficient types of nanomaterials for parallel MR/CT dual-modality imaging, MRI-guided gene delivery and stimulus-response tumor therapy. This All-In-One multiple purpose agents are constructed through a combination of distinctive features of superparamagnetic iron oxide nanoparticles with high saturation magnetization, gold nanoparticles, pH- and temperature responsive polymers and nano-scaled particles. Project objectives include: 1) Development of biocompatible bimetallic/polymer nanocomposite particles as dual-mode MR/CT contrast agents fortumor imaging; 2) Development of dual-functional core-shell nanocomposite particles for MR/CT imaging-guided gene therapy; 3) Development of intelligent nanocomposite particles for MR/CT imaging-guided gene therapy; 4) Evaluation of intelligent nanocomposite system for MR/CT imaging-guided responsive (pH/temperature) gene therapy.

N_HKUST604/14
Development of Multi-modality AIE Nanoprobes for Targeted Detection of Drug Resistant Gene AXL in Lung Cancer and their Preclinical Application

Hong Kong Principal Investigator: Dr Yuning Hong (The Hong Kong University of Science and Technology)
Mainland Principal Investigator: Prof Zhenfeng Zhang (Sun Yat-sen University)

Lung cancer has become the leading cause of cancer death in China due to the large population of smokers and severe environmental pollution. Since there are no symptoms at early stages of most lung cancers, they are usually diagnosed at an advanced stage and have a poor therapeutic efficacy. The difficulty lies in that almost all the patients acquire drug resistance after a short period of treatment with the anti-cancer drugs. In this project, we will collaborate with Prof. Zhenfeng Zhang, who has indentified a drug resistant gene in lung cancer, in Sun Yat-sen University Cancer Center. We will develop new tools with improved sensitivity for the early and differential diagnosis of lung cancer as well as for monitoring the drug resistance during the treatment of lung cancer. The same strategy will be applied for other cancers including breast, colon, and ovarian cancers.

N_HKUST610/14
Nanostructured n-type Photoanodes and p-type Photocathodes for High Performance Water Splitting Photoelectrochemical Whole Cells

Hong Kong Principal Investigator: Prof Shihe Yang (The Hong Kong University of Science and Technology)
Mainland Principal Investigator: Prof Yexiang Tong (Sun Yat-Sen University)

Photoelectrochemical cell (PEC) water splitting over semiconductors is one of the most viable approaches to produce hydrogen for sustainable energy applications because sunlight is clean, CO2 emission free, abundant and renewable. This proposed project aims to develop a PEC whole cell by combining deliberately nanostructured p-type photocathodes and n-type photoanodes through the joint efforts of the Hong Kong team and the Mainland team. Our PEC whole cell design emulates the so-called Z-scheme adopted by nature in photosynthesis. We focus on the engineering of 1D n-type and p-type nanomaterials and their arrays, which are earth-abundant and stable, with a view to enhancing light harvesting, charge separation and PEC surface reaction kinetics, as well as the complementarity and synergy of the materials for a high performing PEC whole cell. The perfect combination of our expertise in nanostructured materials synthesis and photoelectrochemical characterization is essential to synergistically develop the nanostructured photoelectrodes for PEC applications.

N_HKUST617/14
Thermo-mechanical coupling and spatiotemporal effects in phase transitions of shape memory materials

Hong Kong Principal Investigator: Prof Qingping Sun (The Hong Kong University of Science and Technology)
Mainland Principal Investigator: Prof Yongzhong Huo (Fudan University)

Shape memory alloys (SMAs) and shape memory polymers (SMPs), are increasingly being used in many fields due to their excellent and unique properties. One of the most frequent working modes of these materials in services is the cyclic deformation through reversible phase transformation between parent and new phases under external cyclic loadings. Thermomechanical coupling and spatiotemporal heterogeneities at different time and length scales are the typical features of the phase transition behaviors of the materials. Understanding these spatiotemporal effects and their consequences to the fatigue behavior of the materials are of great importance in the applications. In this project, we will conduct a systematic investigation on two issues: (1) the effects of thermomechanical coupling on the spatiotemporal behavior of commercial coarse-grained (with GS=80~100nm) polycrystalline NiTi SMA under cyclic loading; (2) the effects of GS on the thermomechanical coupling, spatiotemporal and fatigue behaviors of the nano-structured NiTi SMA materials (GS from 100 nm to 10 nm). The effects of internal (GS) as well as external parameters on the cyclic responses of the materials will be examined and their impacts on the spatiotemporal dynamics will be revealed. The project is motivated by both the academic importance and the practical need to design and search materials with improved properties in applications.

N_HKU713/14
Rationalizing scaffold design with optimal cell niche for mesenchymal stem cell (MSC)-based therapy in disc degeneration

Hong Kong Principal Investigator: Dr Barbara Pui Chan (The University of Hong Kong)
Mainland Principal Investigator: Prof Yanan Du (Tsinghua University)

Intervertebral disc (IVD) degeneration is a common disease. Recent attempts to treat disc degeneration with mesenchymal stem cells (MSCs) in scaffolds showed encouraging results in delaying disease progression but real regeneration has not been achieved. A primary reason is the unfavorable microenvironment that MSCs expose to when they are delivered. Therefore, systemic investigation on the appropriate microenvironment stimulating disc regeneration is warranted.

Cell microenvironment consists of multiple interactive factors including extracellular matrix (ECM), mechanical signals, topological features and soluble growth factors. Owing to its complexity, high throughput screening technologies (HST) are necessary to investigate multiple factors in combinations. Ideally, simple, rapid, affordable, efficient and highly controllable HSTs able to simultaneously incorporate multiple niche factors should be developed.

We recently developed a multiphoton-based microfabrication platform, which is able to "freely write or print" complex 3D user-defined protein micro-structures and micro-patterns with sub-micron features. Morphological, topological, matrix, mechanical and porosity properties of protein structures can be precisely controlled using this technology. Protein micropatterns such as micropillar array showed excellent cytocompatibility and responded differently to micropatterns with varying mechanical and topological properties.

In this study, we hypothesize that an optimal cell microenvironment with specific soluble signals, matrix components, topological features and mechanical properties can be engineered to maintain the phenotype and function of disc cells, and incorporated rationally into the scaffold design for MSC-based therapy in disc degeneration. Specifically, we aim to (1) screen for the optimal cell microenvironment maintaining the disc cells in physiological manner; (2) study the fate of MSCs upon exposure to such cell microenvironment; and (3) incorporate the optimal cell niche into scaffold design for IVD tissue engineering.

This project will delineate the optimal composition of disc cell microenvironment, establish novel biomedical applications of the 3D multiphoton-based micro-fabrication platform and rationalize the scaffold design for IVD tissue engineering. In the long run, this study will contribute to development of stem cell-based therapies for disc degeneration.

N_HKU715/14
Organic memory array fabricated under ambient air environment: from polycrystalline thin film to single crystal devices

Hong Kong Principal Investigator: Dr Paddy Kwok Leung Chan (The University of Hong Kong)
Mainland Principal Investigator: Prof Hanying Li (Zhejiang University)

The rapid development of organic electronic devices sparks brand new applications of flexible electronics like wearable electronics or artificial electronic skin. Among these devices, organic memory transistor plays a very important role on the information storage. Today, most of scientific investigations of these devices like charge retention time or carrier mobility are mainly based on individual devices. Although these studies can provide useful information on the performance of single memory transistors, they cannot capture their performance in real applications where a network of memory transistors are interconnected to achieve the desired memory capacity. More importantly, most of these organic devices are relying on the thin film structure deposited under vacuum environment. A significant manufacturing and maintenance cost goes to the vacuum equipment. It would be an important step towards to the mass production of organic memory devices if they can be fabricated under ambient air environment without vacuum requirements. In the current proposal, we plan to use organic single crystal and polycrystalline crystal developed by solution processing under ambient air environment. These devices will be scale up into an organic array for practical applications.

N_CityU120/14
Approximation Analysis of Information Theoretic Learning and Ranking Type Learning Problems

Hong Kong Principal Investigator: Prof Ding Xuan Zhou (City University of Hong Kong)
Mainland Principal Investigator: Prof Zongmin Wu (Fudan University)

Theory of learning with the classical least squares loss has been well developed in mathematics, based on probability analysis, statistics, and approximation theory. Information theoretic learning is a different learning framework using descriptors from information theory to substitute the conventional statistical descriptors of variance and covariance in the least squares method for processing non-Gaussian noise. Minimum error entropy is such a principle using entropies in this framework. Ranking type learning problems aim at efficient algorithms involving sample pairs, which is different from methods for regression or classification. The purpose of this project is to develop rigorous mathematical analysis for some problems in these two topics by methods and ideas from approximation theory and wavelet analysis. We shall first establish error analysis for minimum error entropy algorithms in both empirical risk minimization and regularization settings. Analysis for information theoretic learning algorithms induced by Wasserstein metric will also be provided by scaling-based approximation schemes. We shall then make Fourier analysis for some ranking type learning algorithms. Minimizers of the generalization errors associated with the scoring function-based ranking losses and with the metric and similarity learning will be characterized in terms of the generalized Fourier transform by ideas from the study of radial basis functions. Robustness analysis of ranking type regularization schemes will be carried out. Finally some interesting approximation theory problems arising from learning theory involving correntropy, additive models, Wasserstein metric-based approximation, and positive definite kernels will be investigated.

N_CityU132/14
Programmable and Integrated Fabrication of Nano-material Devices by Optically - Induced Force Field

Hong Kong Principal Investigator: Prof Wenjung Li (City University of Hong Kong)
Mainland Principal Investigator: Prof Yuechao Wang (Chinese Academy of Sciences (CAS))

An optically-induced electrokinetics (OEK)-based integrated fabrication method for nano-material devices will be developed in this project in order to overcome the limitations, such as low efficiency, limited selectivity, and complexity, of the existing techniques used to fabricate nano-material based sensors and electronic devices. Through this project, our joint team will explore the fundamental mechanisms behind using OEK phenomena to manipulate, separate, and assemble ions, molecules and nano-materials. Specifically, we will focus on demonstrating three core technologies: 1) OEK-based in situ forming of metal or conductive polymer electrodes; 2) OEK-based high-throughput and parallel assembly of low-dimensional nano-materials, e.g., zinc oxide nanowires, carbon nanotubes, and graphene, to rapidly construct nano-sensors and electronic devices; 3) OEK-based in situ packaging of nano-devices using polymers. In order to demonstrate these core technologies, we need to overcome several technical challenges: i) generate dynamic, and reconfigurable electric field to induce localized OEK forces to assemble nano-materials down to 1nm length-scale; ii) develop models and simulations to understand the fundamental principles behind the various OEK related phenomena in manipulating ions, molecules, and nano-materials; iii) develop advanced algorithms for an OEK platform to automate the process of nano-material manipulation and assembly. Our joint team will be the first in the world to demonstrate a rapid and scalable technology to fabricate nano-devices using dynamically reconfigurable and controllable electric field if the technical objectives are successfully achieved. This accomplishment would be a significant advancement towards the eventually realization of large-scale fabrication and integration of nano-sensing devices.

N_CUHK442/14
Synthesis and Enantioselective Transformation of Carborane-fused Cyclobutenes and Alkenylcarboranes

Hong Kong Principal Investigator: Prof Zuowei Xie (The Chinese University of Hong Kong)
Mainland Principal Investigator: Prof Yong Tang (Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences)

Alkenes are basic chemicals feedstock materials. Their asymmetric transformations such as enantioselective Diels-Alder reaction, cyclopropanation, aziridination, hydrogenation and H-E (E = heteroatoms) addition are of great importance in both academia and industrial processes, which are generally achieved via the action of chiral catalysts consisting of metals and chiral ligands. Though these are extensively investigated processes, the enantioselective transformation of alkenes bearing carborane substituent is virtually unexplored.

Carboranes are a class of boron hydride clusters in which one or more of the BH vertices are replaced by CH units. They constitute a class of structurally unique molecules with exceptionally thermal and chemical stabilities and the ability to hold various substituents. However, the chemistry of chiral carborane derivatives is in its nascent stage due to very limited methods available for the preparation of these asymmetric molecules. In view of their unique structural and electronic properties, chiral carborane derivatives can be employed as chiral building blocks in boron neutron capture therapy (BNCT) agents, in supramolecular/drug design, and in organometallic/coordination chemistry. In this connection, we plan in the current joint proposal to combine the expertise of both Hong Kong and mainland teams to develop a toolbox for the preparation of chiral carborane derivatives via enantioselective transformations of alkenylcarboranes and carborane-fused cyclobutenes. The resultant chiral molecules bearing carborane moiety can be used as chiral ligands for asymmetric catalysis or bioactive compounds for drug screening. The screening results will offer important information on new drug design and structure-property relationships. In addition, it is anticipated that new chiral ligands and new methods developed in this research would be used in the enantioselective transformations of other types of alkenes.

N_CUHK443/14
On Mathematical Theory of the Compressible Fluid-dynamical Equations

Hong Kong Principal Investigator: Prof Zhouping Xin (The Chinese University of Hong Kong)
Mainland Principal Investigator: Prof Hai-liang Li (Capital Normal University)

The nonlinear partial differential equations of compressible Navier-Stokes system and Euler system, or Boltzmann equation and their variants are the fundamental governing equations in continuum mechanics based on the Newton's laws to model fluid motions. For instance, the compressible Navier-Stokes (CNS) system models the motions at macroscopic scale of viscous compressible fluids such as viscous polytropic gases and viscous Saint-Venant model; while the motions of inviscid compressible fluid at macroscopic scale are governed by the compressible Euler system simulates, which is the protype of multi-dimensional hyperbolic conservation laws; also the Boltzmann equation and its variants model motions of dilute particles obeying binary elastic collisions at mesoscopic scale, among which the Vlasov-Poisson (Maxwell)-Boltzmann system is the governing equations for the transportations of charged particles forced by electric field or electromagnetic filed in plasma physics and semiconductor devices. Since their derivations in 19th century, these system have been at the center of studies in the area of nonlinear partial differential equations and applied analysis.

However, despite the huge important progress made in past hundreds of years, the mathematical studies on the well-posedness theory and qualitative behaviors of solutions to these nonlinear partial differential equations in multi-dimension remains far from being completed and satisfied, and there are still many significant and challenging problems to be solved, due to the essential difficulties such as the strongly hyperbolic-parabolic coupling, multi-scale behaviors, mixed-type and singularities, and possible degeneracy in the appearance of vacuum, etc.

In this project, we shall study in three basic aspects of the compressible fluid-dynamical equations: the well-posedness and behaviors of the compressible Navier-Stokes system with various boundary conditions such as non-slip boundary and free boundary for general initial data possibly containing vacuum; the short time well-posedness and vacuum problem of the compressible Euler Equations in general bounded domain; and the spectrum analysis and asymptotical behaviors of Vlasov-Poisson (Maxwell)-Boltzmann system. These are fundamental and frontier issues in this area, and we expect that the study on these open problems can develop new methods and techniques, make important breakthrough on these basic issues of great mathematical challenges, and enrich the pre-existing mathematical theory and qualitative understandings of these fundamental systems.

N_PolyU504/14
Sparse Optimization: Algorithms and Theories

Hong Kong Principal Investigator: Prof Xiaojun Chen (The Hong Kong Polytechnic University)
Mainland Principal Investigator: Prof Ya-xiang Yuan (Chinese Academy of Sciences)

Many interesting models arising in a variety of application domains such as information theory, image processing, statistical learning, medical imaging, bioinformatics, electronic commerce, computer vision, share the feature of seeking a solution with the sparsity structure of an optimization model in a vector, matrix or tensor space. Consequentially, a number of sparse optimization models arise widely in the literature. Mathematically, these models may have the difficulty of non-smoothness, non-Lipschitzness, and even non-convexity. Moreover, in the era with big data where the dimensionality of models increases extremely rapidly, these sparse optimization models usually have another common difficulty of the high dimensionality. All these difficulties critically urge optimizers to work intensively on the algorithmic aspects of these sparse optimization models and thus develop efficient numerical algorithms with affirmative theoretical supports; this is the main purpose of this project. We will systematically investigate how to develop efficient numerical algorithms for some sparse optimization models with strong application backgrounds; and analyze the corresponding theoretical properties such as the convergence and convergence rates of these algorithms. The algorithms will also be applied to solve some real-life applications. This project represents an endeavor to enhance the interaction between optimization and other disciplines; and to provide faster algorithms for practitioners in the mentioned areas.