Exploring the vast space has always been one of the greatest dream for all humanbeings. The International Space Station (ISS) is currently working as a test platform for experimenting various essential equipment for further deep-space exploration and colonization. Because of limited space for storing water on ISS, Water Recovery System (WRS) is heavily depended to produce recycled drinkable water from collected wastewater. As one important component of WRS, the thermal catalytic reactor is designated to remove all Volatile Organic Compounds (VOCs) which is difficult to be dissolved by other treatments. However, the high temperature and pressure working environment makes some of its fragile parts need continuous replacements leading to increased cost and potential risks. Herein, we propose a photocatalytic microfluidic reactor (PMFR) which can be operated at normal temperature and pressure to replace the thermal catalytic reactor. For this titanium (Ti) PMFR, micropillar arrays covered with nanoporous titania (NPT) are designed to attain increased surface area to volume ratios, reduced diffusion lengths, and greater uniformity of irradiation. For the first time, the microelectromechanical systems vii i (MEMS) techniques are used to fabricate a Ti-based microfluidic reactor. For the first degradation study with organic dye Methylene Blue, the micropillar reactor (MP) was found to dramatically outperform its planar counterpart and the 50 μm deep MP (MP50) presented at least two times higher degradation activity than other reported PMFRs. The second study introduced a VOC half-life estimation model designed based on the steady-state concentration of hydroxyl radicals (·OH). This radical was considered as the main reactant of the photocatalytic degradation of VOCs, and its concentration was measured by using nitrobenzene as probe. Half-lives for all VOCs existed in the Ersatz wastewater proposed by NASA were estimated, and results indicated that the MP50 was capable of dissolving most of these VOCs at or under around 100 seconds. The third and last study looked into the MP50 degradation performance with two common stubborn VOCs – ethanol and 1,4-dioxane. Results showed that MP50 eventually fully mineralized about 11.78% of ethanol. As for 1,4-dioxane, MP50 successfully degraded around 78.83% of it with a reaction rate constant two to three orders of magnitude higher than that of other bulk reactors reported in literatures. These findings demonstrate a great potential for the micropillar PMFR to replace thermal catalytic reactor on ISS and even be used for more varieties of water purifications.

Flow diverting stents have transformed the treatment for intracranial aneurysms. These tightly braided metallic wire cylinders when deployed endovascularly at the neck of an aneurysm have been shown to drastically reduce intra-aneurysmal hemodynamic velocity and aneurysm lumen wall shear stress. These devices reduce the chance of rupture and thereby hemorrhagic stroke while simultaneously inducing thrombogenic cues for eventual aneurysm absorption. Current clinically available devices achieve this change in IA by manipulating device porosity and pore density to reduce shear into the aneurysm. Herein, it is demonstrated that device porosity, pore density and shear are not the only design parameters and physical mechanism to consider for flow diversion. 2D and 3D numerical simulations are compared to experimental devices deployed across the neurovasculature with varying diverter strut aspect ratios. Device designs whose struts have an aspect ratio greater than 1 are shown to be effective at inducing hemodynamic transition into aneurysmal flow regimes associated with good clinical outcomes. High aspect ratio devices allow for higher porosities and reduce the IA hemodynamic response to potential changes in porosity. While addressing numerous clinical complications such as in-stent stenosis and edema this design simultaneously widens the range of devices indicated for use in high risk areas such as the posterior communicating artery and basilar trunk by altering the balance of energy transfer through the device from parent anatomy to aneurysm. These results show the potential of more robust devices of higher porosity and a reimagining of the physical forces being manipulated when flow diversion is considered in the clinic.

Flow diverters have generally been successful in the treatment of intracranial aneurysms; however, their utility has been limited by the device’s tendency to divert flow from healthy structures in the vicinity of the lesion site. Here, we show the potential for achieving flow diversion in a selective manner by utilizing an alternative approach in which high aspect-ratio microscale structures are intentionally designed to work congruently with an intermediate porosity flow diverter. We demonstrate this design principle through the realization of a microfabricated flow diverter with finely controlled cross-sectional elements. Using an in vitro basilar trunk model, we demonstrate comparable intra-aneurysmal flow reduction to commercially available flow diverters, while also demonstrating that such devices minimally affect flow into a “jailed” perforating artery (i.e., nearly-eight-fold lower perforator flow reduction than commercial device). This may provide an additional design parameter by which the indication of flow diversion devices may be extended into more delicate structures in the neurovasculature which currently lack a treatment option.

Producing enough drinkable water for long-term deep space missions is a great challenge. A small scale photocatalytic microreactor with nano-porous TiO2 coated titanium micropillar arrays is designed to solve the problem. The COMSOL Multiphysics is used to simulate several parts of the microreactor. The first objective is to model the tree-branched bifurcating flow distribution design and the diamond-shaped flow distribution design, then subsequently compare the pressure drop between the two. Based on simulation results, the tree-branched bifurcating flow distribution design can provide a longer residence time for the waste liquid due to the lower pressure drop inside the distribution channels which increases the photocatalytic efficiency. However, the extension of channels increases the area that this design takes up, thus reducing the size of the reactor chamber. The diamond-shaped distributor is more compact in size because of its geometry which gives more space to the reactor chamber. However, the pressure drop in the microreactor increases about 256% compared to the pressure drop of tree-branched bifurcating distributor which reduces the residence time. The second objective is to simulate photon interaction among micropillars and light intensity along the micropillars under UV irradiation, then find how the micropillar height influences the light intensity on the micropillar. According to simulation, light intensity on the 50 µm height micropillar creates an ideal photocatalytic efficiency. For the 100 µm height micropillar, 50% of the micropillar has low light intensity which causes low photocatalytic efficiency. And 66.67% of the 150 µm height micropillar’s surface has low light intensity. As the height of the micropillar increases, a larger percentage of the micropillar’s surface will have low light intensity. This leads to a decrease in photocatalytic efficiency.

Many applications in cell biology, genetic engineering, cell-based therapeutics, and drug discovery require precise and safe methods for introducing membrane-impermeable molecules into cells. This can be implemented satisfactorily by microinjection. However, disadvantages of traditional manual microinjection include high degree of operator skill, low throughput and labor-intensiveness. Many studies have focused on developing automated and high-throughput systems for microinjection to address these limitations. However, none have provided sufficient throughput for applications such as ex vivo cell therapy, where manipulation of many millions of cells is required. Herein, we propose an ultrahigh throughput (UHT) mechanoporation concept that seeks to address these limitations. The mechanoporation device is a massively-parallelized MEMS-based platform for passively delivering molecules into living cells via mechanical cell membrane penetration. Studies focusing on device design, fabrication and validation at the proof-of-concept level are presented in this dissertation.

Detailed system concept and design is introduced, which integrates functions of cell transfer, capture, penetration and release into a single piece of instrumentation using a microfluidic approach. System operating parameters are analytically analyzed and numerically simulated. Results from these studies agree with previous studies by others in related applications, and suggest reasonable operation feasibility without detrimental effect on cells. Those estimated operation parameters also provide basis to develop test models in practical cell studies. The device fabrication utilized conventional silicon MEMS technologies, and we successfully produced millimeter-scale device chips containing an array of ten thousand hemispherical capture wells with monolithically integrated solid penetrators. A flow circuit system involving a syringe pump, pressure transducer, and fixture set supporting the device chip was developed, and preliminary functional testing was carried out. Device validation tests using K562 cells obtained about 15% average penetration efficiency of live cells after manipulation. Subsequent testing with fluorescent beads and Mouse Embryonic Fibroblast (MEF) cells identified several key issues responsible for the lower-than-expected efficiency, thus suggesting that improved performance may be possible with further system and operation optimization.

The UHT mechanoporation device developed in this effort shows promise for providing an efficient and safe method for introducing membrane impermeable molecules into cells with ultrahigh throughput. Moreover, these studies also represent key steps towards our long-term goal of developing instrumentation capable of UHT cellular manipulation via active microinjection. This new instrumentation will have broad potential for advancing understanding of fundamental cellular processes, as well as facilitating clinical translation of cell-based therapies.

The use of microelectromechanical systems (MEMS) technology in medical and biological applications has increased dramatically in the past decade due to the potential for enhanced sensitivity, functionality, and performance associated with the miniaturization of devices, as well as the market potential for low-cost, personalized medicine. However, the utility of such devices in clinical medicine is ultimately limited due to factors associated with prevailing micromachined materials such as silicon, as it poses concerns of safety and reliability due to its intrinsically brittle properties, making it prone to catastrophic failure. Recent advances in titanium (Ti) micromachining provides an opportunity to create devices with enhanced safety and performance due to its proven biocompatibility and high fracture toughness, which causes it to fail by means of graceful, plasticity-based deformation. Motivated by this opportunity, we discuss our efforts to advance Ti MEMS technology in two ways: 1) Through the development of titanium-based microneedles (MNs) that seek to provide a safer, simpler, and more efficacious means of ocular drug delivery, and 2) Through the advancement of Ti anodic bonding for future realization of robust microfluidic devices for photocatalysis applications.

As for the first of these thrusts, we show that MN devices with in-plane geometry and through-thickness fenestrations that serve as drug reservoirs for passive delivery via diffusive transport from fast-dissolving coatings can be fabricated utilizing Ti deep reactive ion etching (Ti DRIE). Our mechanical testing and finite element analysis (FEA) results suggest that these devices possess sufficient stiffness for reliable corneal insertion. Our MN coating studies show that, relative to solid MNs of identical shank dimension, fenestrated devices can increase drug carrying capacity by 5-fold. Furthermore, we demonstrate that through-etched fenestrations provide a protective cavity for delivering drugs subsurface, thereby enhancing delivery efficiencies in an ex vivo rabbit cornea model. Collectively, these results show the potential embodied in developing Ti MNs for effective, minimally invasive, and low-cost ocular drug delivery.

Additionally, or the second of these thrusts, we report the development of an anodic bonding process that allows, for the first time, high-strength joining of bulk Ti and glass substrates at the wafer-scale, without need for interlayers or adhesives. We demonstrate that uniform, full-wafer bonding can be achieved at temperatures as low as 250°C, and that failure during burst pressure testing occurs via crack propagation through the glass, rather than the Ti/glass interface, thus demonstrating the robustness of the bonding. Moreover, using optimized bonding conditions, we demonstrate the fabrication of rudimentary Ti/glass-based microfluidic devices at the wafer-scale, and their leak-free operation under pressure-driven flow. Finally, we demonstrate the monolithic integration of nanoporous titanium dioxide within such devices, thus illustrating the promise embodied in Ti anodic bonding for future realization of robust microfluidic devices for photocatalysis applications. Together, these results demonstrate the potential embodied in utilizing Ti MEMS technology for the fabrication of novel drug delivery and microfluidic systems with enhanced robustness, safety, and performance.

Flow diversion devices have generally been successful in the treatment of intracranial aneurysms. However, clinical complications have often been linked to complex and counter-intuitive mechanical responses which arise from the device’s braid-based design (e.g. poor wall apposition, excessive foreshortening, etc.). Herein, as a potential response to these issues, we demonstrate the feasibility of a novel microfabricated balloon-deployable flow diverter with finely controlled orthogonal cross-sectional elements which are monolithically integrated into a solid construct. Previous studies have mostly focused on characterizing flow diversion as a result of porosity in highly symmetric devices. By contrast, here we show that similar results can be achieved in more sparse and heterogeneous structures by utilizing basic principles of aerodynamics in micro-scale design (i.e. intentionally designing structural elements to induce drag). Computational and experimental models suggest the potential for such treatment, noting an average intra-aneurysmal velocity reduction of 86% and 88% ± 4% (n=3), respectively and reductions in average wall shear stress of 87% and 88% ± 4% (n=3). Experimental models showed agreement with computational models with the model’s predictive metrics for aneurysm occlusion falling within the 95% confidence interval for the experiments.

Physical methods of intracellular delivery based on microfluidic approaches have emerged as a promising alternative to conventional delivery and cellular engineering techniques with many possible applications, particularly in the gene therapy sphere. No matter the physical poration method, there are tradeoffs between delivery efficiency and cell viability, and these differ depending on the manner of membrane poration. For several physical delivery methods, the process is difficult to control and random, which can limit their effectiveness and utility. Herein, we propose a novel silicon-based ultrahigh throughput deterministic mechanoporation (DMP) device platform for nanomechanical gene delivery into cells to address the current gap in knowledge and performance in current gene delivery techniques. This platform functions through microfluidic flow control, inducing the impingement of a single cell upon a solid needle resulting in a membrane pore after the cells are released, but on a massively-parallelized scale. This temporary pore allows diffusion of genetic cargo into the cell before membrane repair.In these studies, the performance of DMP delivery of model genetic cargos into human T cells was investigated, as well as how experimental design and analysis influences these types of experimental results. Next, characterization of how T cells interact with and are morphologically impacted by the DMP device was performed along with reassessment of device performance with small molecule cargo delivery. Finally, a preliminary investigation of how larger, complex gene editing constructs can be delivered to cells through conventional nonviral techniques was completed with a novel genetic construct. Through future design and operational parameter optimization, DMP aims to serve as a critical enabler for gene editing applications, given their dependence on successful delivery of these large genetic cargos and the survivability of edited cells for expansion and eventual treatment of patients with gene therapies.

Drug-eluting stents have revolutionized the field of interventional cardiology by reducing incidence of restenosis through local delivery of drugs that inhibit inflammation caused by implantation-induced injury. However, growing evidence suggests that this may also inhibit reestablishment of the endothelium, thus delaying healing and increasing potential for thrombogenic stimulus. Herein, we discuss progress towards realization of next-generation titanium (Ti) stents that seek to mitigate adverse physiological responses to stenting via rational design of stent surface topography at the micro- and sub-micrometer scale.

To better understand the effect of surface topography on cells, we evaluate the in vitro response of EA926 endothelial cells (EC) to variation in precisely-defined, micrometer to sub-micrometer scale groove-based topography, with groove widths ranging from 0.5 to 50 µm. Silicon (Si) and Ti materials are chosen for these studies due to their relevance for implantable microdevice applications, while grating-based patterns are chosen for their potential to induce elongated and aligned cellular morphologies reminiscent of the native endothelium. We show significant improvement in cellular adhesion, proliferation, and morphology with decreasing feature size on patterned Ti substrates. Moreover, we show similar, yet attenuated, trending on patterned Si substrates as compared to patterned Ti substrates. These results suggest promise for sub-micrometer topographic patterning to enhance endothelialization and neovascularisation for implantable microdevice applications.

We also discuss: 1) advances which now allow patterning of features in Ti substrates down to 150 nm, which represents the smallest features achieved to date using our novel Ti deep reactive ion etching (Ti DRIE) technique; 2) creation and evaluation of balloon-deployable, cylindrical, surface-patterned stents from micromachined planar Ti substrates; and 3) integration of these processes to produce a device platform that allows, for the first time, evaluation of surface patterning in more physiologically relevant contexts, e.g. in vitro organ culture. Using elasto-plastic finite element analysis, we also explore planar stents with novel locking mechanisms intended to address radial stiffness deficiencies observed in earlier studies. Collectively, these efforts represent key steps towards our long-term goal of developing a new paradigm for stents in which rationally-designed surface patterning provides a physical means for complementing, or replacing, current pharmacological interventions.

Stroke is the fourth leading cause of death in the United States with a high morbidity

rate for both ischemic and hemorrhagic strokes. Although the majority of stokes

that occur are ischemic, these patients have a high chance of survivability if treated.

However, 50% of patients with hemorrhagic stroke, prognosis of death within the

first six months is high. Hemorrhagic strokes typically occur due to rupture of an

intracerebral aneurysm (i.e., a portion of a neurovasculature that has ballooned out

due to disease-induced weakening). Current treatment methods are moving toward

less invasive techniques to treat aneurysms before rupture occurs, one particularly

compelling example of which is the flow diverter (FDs). These devices redirect

flow away from the aneurysm sac, and they have been shown to allow for healing

of the diseased tissue. Flow diverters are currently fabricated by braiding individual

wires into a mesh-like structure. However, due to this design, coupled with the

typical location of the aneurysms being treated, there is a high chance of occluding

small perforator arteries as well. Our lab is developing a new concept for FD design,

enabled by novel MEMS fabrication techniques we have pioneered, which seeks to

address this limitation via use of high-aspect-ratio (height-to-width ratio) struts.

The study described herein, has sought to explore the effect of device placement

and strut thickness on flow diversion performance within this context.