Protection of foreign investment has long been an issue facing China.  The newly adopted Foreign Investment Law (FIL) and Implementation Regulations not only unify the foreign investment regulations but also reformulate the regulatory regime that governs foreign investment in the country.  In response to the mounting criticism of the practices in China that damage the interests of foreign investors, including, among others, forced technology transfer and commercial theft, the FIL is purposed to build a better environment so that foreign investment will be more effectively protected.

The FIL changes the main themes of China’s regulation of foreign investment and puts new market access rules and measures in place on the foreign investment horizon.  The FIL Implementation Regulations intend to fill the gaps left in the FIL.  Still, many questions remain unanswered.  Both the broadness and vagueness of the FIL require further clarification and specific measures in different aspects.  The Supreme People’s Court is expected to issue judicial interpretations on various practical matters.

The FIL is charged with the mission to even the playing field by providing fair treatment to foreign investors in the country, but the challenges encountering foreign businesses seeking establishment in China remain.  The clear rules of implementation aside, an effective enforcement mechanism is essential to the achievement of the intended goals of the FIL.

This dissertation investigates the mechanics of semiflexible filament network and the effects of polydispersity, anisotropy, and labile crosslinks through numerical simulations.

This study starts with exploring the effects of filament length polydispersity on the mechanical properties of semiflexible network. Extending previous studies on monodisperse network, the responses of networks of bimodal and exponential length distributions are tested numerically. It is found that in polydisperse networks, mixtures of long and short filaments interact cooperatively to generally produce mechanical response closer to the affine prediction than comparable monodisperse networks of either long or short filaments. Overall, length polydispersity has the effect of sharpening and shifting the affine/nonaffine(A/NA) transition to lower network densities.

The effect of adding long, stiff filaments to a semiflexible network is studied next. It is shown that the addition of a small fraction of longer and stiffer filaments (microtubules) to a nonaffine network (actin filaments) leads to a significant increase in its overall elastic moduli, even though the long filaments do not form a stress bearing network by themselves. Moreover, there is a strong negative correlation between long filament density and local geometric measures of nonaffinity.

Anisotropic, monodisperse network is then investigated and like isotropic networks, it undergoes an A/NA transition controlled by the ratio of the filament length to the nonaffinity length $\lambda$. Deep in the nonaffine regime, however, these anisotropic networks exhibit a vanishing linear response regime for highly ordered networks and a dependence of the shear modulus on shear direction. These features can be understood in terms of a generalized floppy modes analysis of the nonaffine mechanics and a type of cooperative Euler buckling, which are discussed in this study, too.

Lastly, by allowing crosslinks to break and rebind, network's dynamic response is studied. It is shown that force-induced crosslink breakage leads to significant creeping and the network retains part of its elastic modulus even after significant plastic flow. The spatial correlation of crosslink breakage is also studied.

This dissertation focuses on human behavior modeling and incentive designs for online platforms, specifically in two application domains: collective intelligence platforms and healthcare intervention platforms. In the last two decades, wide adoption of internet facilitates the emergence of online collective intelligence platforms. In the mean time, the fast developing mobile technology enables real-time healthcare interventions. This dissertation is a step towards implementing human behavior modeling through machine learning and optimization techniques to enhance the efficacy of these online platforms. Case studies show that human behavior modeling combined with smart incentive designs have the potential to improve the performance of such online platforms.

In the first part of this dissertation, I present two collective intelligence platforms: M-CAFE and DebateCAFE. M-CAFE is a mobile-friendly platform that encourages students to check in weekly to numerically assess their course performance, provide textual ideas about how the course might be improved, and rate ideas suggested by other students. For instructors, M-CAFE displays ongoing trends and highlights potentially valuable ideas based on collaborative filtering. M-CAFE is complementary to existing platforms, such as Piazza and stackExchange and allows students to step back and consider their own performance and the performance of their instructors, filling the gap between voluminous transcripts from existing platforms and a one-time-only, end-of-course evaluation. DebateCAFE is an online deliberation platform that introduces a novel incentive mechanism to encourage participants to articulate persuasive arguments on both sides of a complex issue. It uses a combination of uncertainty sampling and collaborative filtering to mitigate bias from selective exposure and highlight/rank the persuasive arguments. Furthermore, DebateCAFE assigns a score to each participant based on the lower of the Wilson scores of the two arguments entered to encourage strong arguments for opposing opinions. Both platforms were built upon the CAFE framework, which is based on Opinion Space.

In the second part of this dissertation, I describe a novel mobile phone application (app) called CalFit and introduce the Discontinuation Prediction Score (DiPS) for non-adherence prediction. CalFit implements important behavior-change features like dynamic goal setting and self-monitoring. Specifically, CalFit uses a reinforcement learning algorithm to generate personalized daily step goals that are challenging but attainable. Two empirical studies with university staff and students indicate that dynamic goal setting is effective in promoting physical activity. Despite the success, I identify that low adherence is a major drawback for mobile-based interventions like ours. Therefore, I further developed the Discontinuation Prediction Score (DiPS), which uses objectively measured past data (e.g., steps and goal achievement) to provide a numerical quantity indicating the likelihood of exercise relapse for the upcoming week for each subject. I present two versions of DiPS using logistic regression and support vector machine methods to demonstrate that DiPS has potential to accurately predict exercise relapse and efficiently allocate resources to improve compliance.

The cement industry is one of the most difficult sectors to decarbonize. The manufacturing of Portland cement – the most common type of cement used for structural concrete – presently accounts for roughly 2% of global energy usage and 8% of CO 2 emissions. The current manufacturing process has not changed substantially over the past 140 years and requires temperatures above 1450 °C with heat provided by the combustion of fossil fuels. The massive CO2 emission originates from: 1) fossil fuel combustion needed to heat the raw materials ultimately to ~1450 °C for calcination and sintering, and 2) decomposition of CaCO3 in the kiln. About 1 ton of CO2 is released for every ton of Portland cement produced in this manner. To date, most of the efforts to reduce CO2 emissions have focused on the use of supplementary cementitious materials. Other approach such as capturing CO2 from fossil fuel power plants and sequestering the carbon in building materials or geologically stable substances requires the building of vast infrastructure and remains to be proven. Future manufacturing approaches that can decarbonize cement and cement-based materials are needed.

This dissertation study establishes a new paradigm to decarbonize cement and cementitious materials through integrating novel manufacturing with cement chemistry and micromechanics. The new paradigm enables the novel design, manufacturing and validation of cement and cementitious materials with much reduced material-production-stage CO2 as well as life-cycle CO2 : (1) cement produced based on a clean electrochemical path instead of conventional fossil fuel combustion process; (2) ductile ultra-low-binder-content cementitious materials to reduce lifecycle CO2 by 65% and energy consumption by 70% of transportation infrastructure; and (3) an alternative low-temperature low-CO 2 “cement” clinker with additional CO2 sequestration functionality, which can be synthesized at room temperature and has potential of 4 times of CO2 uptake than conventional Portland cement. This research advances knowledge in understanding the reaction kinetics of alite and belite syntheses from electrochemically produced Ca(OH)2 and waste precursors, in tailoring the rheology, compaction self-assembly and micromechanics of fiber-reinforced low-binder cementitious materials, and in elucidating the chemistry, microstructure and properties of Ca(OH)2 -activated cementitious materials.

This study focuses on understanding the electrochemical behavior of cement-based materials. Electrochemical impedance spectroscopy (EIS) was conducted on various cement-based specimens to reveal how material age, chemical composition, and the incorporation of carbon black nanoparticles and polymeric fibers affect the AC impedance spectra under a wide range of AC probing frequency from 1 Hz to 10 MHz. Through EIS, the frequency-dependent electrical behavior of the cementitious materials was represented by an idealized model circuit consisting of discrete electrical components. Analyzing the circuit model parameters revealed the inner changes of electrical microstructure during the process of hydration. What can be improved in the experiment of this study was discussed at the end of the article.

Recent large-scale text-to-image (T2I) diffusion models have achieved remarkable success, enabling the generation of complex and realistic images from any text prompt that describes the target concept. Despite the significant advantages, the T2I diffusion model suffers from poor spatial controllability solely from text description. This thesis focuses on improving the pre-trained T2I diffusion models with additional support to take spatial reference.

The first part of this thesis proposed FreeControl, a training-free and guidance-based approach for controllable T2I generation that supports multiple conditions, architectures, and checkpoints simultaneously. FreeControl enforces structure guidance to facilitate the global alignment with a guidance image, and appearance guidance to collect visual details from images generated without control. Extensive qualitative and quantitative experiments demonstrate the superior performance of FreeControl across a variety of pre-trained T2I models. In particular, FreeControl enables convenient training-free control over many different architectures and checkpoints, allows the challenging input conditions on which most of the existing training-free methods fail, and achieves competitive synthesis quality compared to training-based approaches.

The second part of this thesis presents Ctrl-X, a training-free and guidance-free method that supports structure and appearance customization from a large spectrum of mage modalities. Ctrl-X designs feed-forward structure control to enable the structure alignment with a structure image and semantic-aware appearance transfer to facilitate the appearance transfer from a user-input image. Extensive qualitative and quantitative experiments illustrate the superior performance of Ctrl-X on various condition inputs and model checkpoints.

Damage tolerance and self-healing are naturally occurring phenomena found in many biological systems, optimized through millennia of evolution. Man-made materials, however, generally do not possess such attributes. Concrete is the most heavily used man-made material, but is inherently brittle and inevitably suffers from cracking under various mechanical and environmental loading conditions. Cracking impairs concrete mechanical properties, causing local stress concentration, and stiffness and strength reduction in structural members. Cracking also compromises concrete transport properties, accelerating other deterioration mechanisms such as chloride diffusion, moisture penetration, and reinforcement corrosion. These challenges can be potentially addressed by a new generation of self-healing cementitious materials, which can autogenously regain its transport property as well as mechanical capacity after damage.

While self-healing phenomena has been observed in cementitious materials, questions arise as to whether robust self-healing can occur reliably. The goal of this dissertation is to generate fundamental understandings of the chemical, physical and environmental mechanisms that control the self-healing process in cementitious materials. It is hypothesized that these mechanisms determine what, how and to what extent healing products form along crack walls. Understanding these mechanisms will lead to new material tailoring strategies to achieve robust and intrinsic self-healing in cementitious materials.

In this dissertation, we first establish an experimental framework to accurately probe the level and quality of self-healing within a crack, to analyze the chemical composition of the healing products, and to quantify the recovery of material transport and mechanical properties. Through experiments, we uncover and quantify the nonhomogeneous self-healing ratios and varying chemical compositions along the crack depth. Second, we formulate a coupled transport-thermodynamics model to predict crack profile evolution with time due to self-healing. The model captures three mechanisms that dynamically control the healing process: the advective transport of ions as reactants caused by the flow of the aqueous solution, the molecular diffusion of dissolved reactants due to the concentration gradient, and the kinetics of chemical reactions occurring at the water-material interfaces controlled by thermodynamics. We test and validate the model through a set of designed experiments with well-controlled chemical and physical parameters. Third, we extend the experimental framework and the model to study the effects of different chemical variables (e.g. binder chemistry, age, ion concentration), physical properties (e.g. crack width, geometry, transport properties), and environmental conditions (e.g. wet and dry, flow rate, pH) on the self-healing extent and properties. The results elucidate the healing mechanisms and governing parameters, laying the groundwork for designing robust self-healing into cementitious materials. Fourth, we develop new self-healing cementitious composite materials by satisfying a combination of criteria: (a) the presence of a multitude of essential chemical species or ions that can react with natural actuators upon cracking, (b) distributed damage behavior that sequentially and spatially activates healing reactants, and (c) self-controlled tight crack width that can promote robust self-healing without consuming a large amount of healing reactants. The experiments validate the improved three-dimensional healing extent of the new materials, and the recovery of transport and mechanical properties. Finally, the dissertation explores the multifunctionality of new cementitious materials by coupling the cracking and healing processes with material frequency-dependent electrical response.