Research project descriptions: Our REU will focus on fundamental research experiences related to
Design and manufacturing of sensors, actuators and smart materials
Advanced imaging and diagnostic systems and
Big data approaches to design and analysis of sensing systems.
Selected projects are listed below – these feature a new slate of projects compared to our previous offerings. The REU students will be active members of interdisciplinary research groups and will interact with faculty, post-docs, graduate students, and industrial collaborators. To aid in participant preparation, faculty will make available background reading material and if appropriate, tutorials/videos for each project for the participants to access and digest prior to arriving on campus.
Design and manufacturing of sensors, actuators and smart materials
1) Portable Sensors for Chemical Detection (Shrotriya):
The Shrotriya research group has recently developed a miniature differential sensor (Fig. 2) consisting of two adjacent micromachined cantilevers (a sensing/reference pair) for detection of chemical and biological species. The sensing strategy involves coating one surface of a microcantilever with a receptor species that has high affinity for the analyte molecule of interest. The presence of the analyte is detected by resolving the surface stress change associated with absorption/adsorption of analyte molecules on sensing cantilever. Example REU Project: An REU student will work with a graduate student on validating novel competition sensing mode-based handheld sensors that enable detection with high sensitivity and specificity using aptamer functionalized microcantilever. The REU student will aid in validating a portable device design (Fig. 2) for testing specific common food pathogens but it also can be applied to other instances of analyte detection as the need arises.
2) Real-time Actuation based Assembly of Nanomaterials (Juárez):
Self-assembly strategies routinely use micro- or nanoscopic building blocks such as small particles, polymers or proteins to fabricate new types of materials for energy storage (e.g., solar cells) or drug delivery (e.g., encapsulation of a pharmaceutical). The Juárez lab addresses this technical challenge by relying on real-time actuation of these building blocks using electric fields (Fig. 3). Real-time image analysis using a feedback control algorithm devised with the aid of a physical model is used as a sensor to help decide when the actuation is executed. This allows us to the tune the assembly in an informed manner to produce the highest quality structures (e.g., crystals for optical computing). Example REU Project: An REU student will design and operate a platform for manipulating and controlling assembly of small particles. The student will work with a senior graduate student to design a platform for assembling crystals. The platform that the student designs will be tested on an optical microscope. Video microscopy results from the student-driven experiments will be analyzed using codes written in MATLAB or FORTRAN.
3) Actuator Design for Bioprosthetics using Isogeometric Analysis (Hsu):
Heart valves consist of thin, flexible leaflets that actuate passively, in response to blood flow and serve to ensure unidirectional flow of blood through the circulatory systems of humans. In aortic heart valves, the leaflets may become diseased and often, valves must be replaced by prostheses.. Hsu’s group is developing a parametric design and optimization framework based on immersogeometric fluid-structure interaction (FSI) for the simulation of heart valve function over a complete cardiac cycle under realistic physiological conditions (Fig. 4). This research will allow us to study detailed information of hemodynamics and cyclic stresses developed in the leaflets, and provide a unique approach to the rational development of new BHVs with improved performance and durability. Example REU Project: An REU student will be involved in designing new geometric and material parameters and cost functions for heart actuator (valve) optimization problem. The new BHV design will be tested numerically using our isogeometric design-through-analysis platform, including dynamic and FSI simulations, to investigate its performance and durability. The student will have the opportunity to learn the impact of design variables to the functionality of the BHVs including interaction between leaflets and surrounding blood flows.
4) Graphene Biosensor Project (Claussen):
Ecosystem health, including human health, is highly dependent on soil and food health. Key biogeochemical indicators of soil health (e.g., nitrogen, and phosphorus) are spatially and temporally dynamic, and vary widely with soil type and regional conditions. Likewise, foodborne pathogens foodborne pathogens (e.g., Salmonella) that severely compromise food safety can significantly vary between food batches or lots and consequently often go undetected. Hence the ability to ensure soil health and food safety requires continual, high-fidelity monitoring of farm fields and food products. In other words, large networks of low-cost, but highly selective and sensitive biosensors need to be developed.
This project is focused on developing graphene-based biosensors that can electrochemically monitor fertilizer ions in soil and bacteria in food product products. Graphene inks comprised of low-cost exfoliated flakes of graphite or even coal (i.e., the cheapest material on Earth: ~ $10-$60 per ton) are inkjet printed onto thin sheets of polymers and functionalized with appropriate ion selective membranes or antibodies for fertilizer ion and Salmonella detection respectively. Various surface chemistry techniques are used to ensure the sensors are resilient to false positive signals. The sensors are coupled with 3D printed microfluidic cartridges as needed to deliver test solutions directly to the sensors surfaces. Students working on this project can expect to learn how to formulate nanoparticle inks, print these inks onto thin polymer sheets, biologically functionalize the printed graphene circuits for biosensing, and perform electrochemical biosensing experiments. Students interested in advanced materials manufacturing, fluid dynamics and mass transport, as well as surface chemistry may find this project particularly compelling.
Advanced imaging and diagnostic systems 5) Probe-based Nanomechanical Property Quantification and Imaging of Soft Materials (Ren):
Nanoscale morphological characterization and mechanical property quantification of soft and biological materials play an important role in areas ranging from nano-composite material synthesis and characterization, cellular mechanics to drug design. Ren, who recently joined our faculty, has created a set of control-based approaches and adaptive imaging modes to enable accurate nanomechanical quantification on soft materials for both broadband imaging and force-curve measurements (Fig. 5). Example REU Project: An REU student will work with a graduate student towards investigating the correlation between the bio-morphological change and nanomechanical property alteration using AFM-based approaches. Specific tasks may include: (1) Test and calibrate a new AFM device; (2) Optimize current imaging and force-curve measurement approaches on AFM to achieve simultaneous accurate surface quantification and nanomechanical mapping on live cells. The optimized measurement tools can be used to study the morphological and nanomechanical variations of healthy human endothelial cells treated by different chemical compounds.
6) Micro-scale imaging of thermophysical phenomena in biomass feedstocks (Michael):
Biomass pyrolysis offers the potential to produce carbon-neutral or carbon-negative liquid fuels for transportation. Michael’s group is investigating the physical and chemical pathways and mechanisms of pyrolysis using laser-based measurements. The reactors used are designed to replicate the thermal characteristics of pyrolysis reactors which are typically harsh environments without optical access, making in situ measurements difficult. Example REU Project: The study’s objective will be the use of laser-based imaging to determine chemical species present during the rapid heating process of biomass fast pyrolysis. The REU student will work a graduate student and Dr. Michael to set up and analyze Raman and fluorescence spectra to determine important intermediates and products in biomass conversion processes. The REU student will process these spectra and correlate these with imaging of biomass pyrolysis phenomena while using spectral libraries and image processing to analyze experimental data.
7) In situ measurements of nanoscale energetic material interfaces during reaction (Sippel):
Nanoscale energetic materials whose energy release characteristics are highly controlled by material interfaces rather than bulk material properties can overcome challenges associated with slow energy release rates and high reaction thresholds of currently available materials. These materials have applications in energy dense pumps for MEMS devices, gas generators (air bags, propulsion), thermal energy sources (heating, remote power generation), and pyrotechnics. Sippel’s group attempts to obtain a fundamental understanding of the initiation and reactivity of nanoscale energetics via high spatial resolution study of reactive interfaces during thermal loading (Fig. 7). Example REU Project: Guided by a graduate student, an REU student will perform experiments to investigate both the condensed phase and the gas phase reaction onset occurring during heating. In doing so, an REU student will learn to perform one or more of the following tasks: 1) High spatial resolution interface observations using hot stage electron microscopy; 2) X-ray photoelectron spectroscopy, and/or bulk crystal phase structure via heated x-ray diffraction. These techniques, combined with bulk combustion measurements and observation of reaction/oxidation mechanisms reveal to the student how nanostructure and interfaces affect energy release and reactivity.
8) Ultrasound for Imaging and Therapy (Bigelow):
It may be possible to significantly reduce the ~35,000 surgeries each year needed to replace infected surgical meshes following abdominal hernia repair. Mesh infections require reoperation for mesh removal ~70% of the time resulting in the potential for hernia reoccurrence and the need for additional operations. Therefore, there is a critical need to develop new methods to noninvasively treat mesh infections without removing the mesh. Otherwise, the treatment of mesh infections will remain highly invasive and costly. Our goal is to develop ultrasound cavitation-based histotripsy to treat infections on medical implants. Example REU Project: This study’s objective is to optimize the treatment of surgical mesh infections by exploring the synergistic relationship between cavitation damage to cells and radiation. The REU student will work with Dr. Bigelow to grow bacteria on the mesh samples. They will then assist in the ultrasound exposure experiments as well as the radiation exposure experiments. Lastly, they will assist in determining the number of colony forming units surviving the treatments.
9) Microfluidically Produced Polymeric Microfibers (Hashemi): Engineering 3D biomimetic scaffolds that incorporate both biochemical and mechanical properties required for cell culturing is critical for many biotechnology applications. Hydrogel-based scaffolds are widely used due to their biocompatibility, tunable biochemical properties, and tissue-like water content. In contrast to hydrogels, microfibers have high mechanical strength and are used as the building blocks to create highly porous scaffolds.
Example REU Project: At the Hashemi Lab, we are interested in exploring different fabrication processes for the development of the microfibers. We will study the effect of fabrication parameters such as ejection rate, temperature, and time of pre-fixation on microfiber properties. The resulting scaffold morphology will be characterized using scanning electron microscopy and compression test.
10) Advanced Additive Manufacturing of Soft Electronics (Montazami):
Soft electronics can have a wide range of applications such as wearable electronics, personal gadgets, conformal electronic devices, and epidermal electronics and sensors. Fabricating electronics on soft substrates is very challenging as conductive electronic patterns frequently crack and delaminate under even slight deformations rendering the device useless. At Montazami Lab we aim to utilize advanced additive manufacturing techniques such as 3D printing, electrohydrodynamic jet printing, and self-assembly to fabricate and characterize a wide range of patterns and structures. And, gain a fundamental understanding of mechanics of the printed materials by investigating surface properties of, and interfacial bonding between the printed layers.
Example REU Project: Mentored by a graduate student and the PI, the REU student will learn to design and conduct experiments; and, analyze experimental and analytical data to characterize surface and interfacial properties of single and multilayered printed patterns. Two major areas of study will be 1) investigating the effects of inks and substrates’ chemical properties on their interfacial bonding and consequently, mechanical integrity of the printed structures; and 2) establishing the relationship between printing parameters and bonding at the ink-substrate interface. A high-fidelity model will be developed to describe correlations between interfacial chemical bonding and mechanical properties of printed structures.
11) Multi-Scale Models for Personalized Cardiovascular Simulations (Krishnamurthy):
The main objective of this research is the advancement of the state-of-the-art in translational medicine with the help of computational tools for patient-specific cardiovascular simulations. Personalized medicine, especially in designing custom pacemaker therapies for optimizing patient response, is gaining popularity due to advancements in medical imaging and computer simulations. However, current methods of personalized cardiovascular model generation are tedious and time consuming, requiring expert knowledge of computational tools. As part of our research, we plan to take advantage of computer aided-design (CAD) methodology and tools to accelerate the personalized-model creation process. Example REU Project: Patient-Specific Cardiac Mesh Generation from Image Data: One of the requirements for computational modeling is the generation of patient-specific finite-element mesh of the ventricles based on data from cardiac computed tomography (CT) or magnetic resonance (MR) images with minimal user intervention. A REU project will consist of using our mesh generation framework to generate patient-specific cardiac meshes. The project will expose the student to different mesh generation tools and the research methodology involved in patient-specific heart modeling.
Data driven approaches to Design and Reliability of Sensors and Devices
12) Data-Driven Health Management of Lithium-Ion Rechargeable Batteries in Implantable Medical Devices (C. Hu):
Reliability of Lithium-ion (Li-ion) batteries used in implantable medical devices has been recognized as of high importance from a broad range of stakeholders, including medical device manufacturers, regulatory agencies, patients and physicians. Hu’s group aims to develop a robust data-driven prognostic method for improving the operational reliability and safety of Li-ion batteries used in implantable medical devices. The methods developed in this project will utilize large volumes of measurement data to estimate the capacity of a battery and predict its remaining useful life (Fig. 8), while taking into account various sources of uncertainty (e.g., cell-to-cell variations and future loading conditions), and will offer personalized prediction of battery health for individual patients. Fast and efficient machine learning techniques will be employed to speed up data processing for on-board applications. Example REU Project: An REU student will work closely with a graduate student on two main tasks. First, (s)he will develop and implement a data-driven prognostic method that utilizes real-time voltage and current measurements from a Li-ion battery to infer the capacity of the battery at every charge/discharge cycle. Second, the student will help develop a pseudo-two-dimensional (P2D) physics-based model that simulates the dynamic behavior of a battery under faulty conditions. The simulation data will be used to demonstrate the effectiveness of the developed method in fault detection and failure prevention. In addition to these two tasks, the REU student may also help conduct experiments with defects intentionally seeded into implantable-grade Li-ion batteries and acquire experimental run-to-failure data that can be used to validate the developed method.
13) Machine Learning, Pattern Discovery and Data Analytics for the design of microfluidic devices (Ganapathysubramanian and Sarkar):
Controlling the shape and location of a fluid streams provides a fundamental tool for cost-effective, efficient, and scalable creation of microfluidic lab-on-chip and similar diagnostic devices. Ganapathysubramanian and coworkers have demonstrated the ability to engineer the cross-sectional shape of a fluid using the notion of inertial flow deformations induced by sequences of pillars that disrupt the flow (see image). The basic idea is to use the inertial flow deformations associated with the flow around a library of single cylindrical pillars at specific positions within a microfluidic channel as fundamental operation. Using this framework, pillar programming has been successfully employed for shaping polymer precursors for streams and particles, reducing inertial flow focusing positions, and solution transfer around particles. Example REU project: The REU project will tackle the problem of designing a sequence of pillars that result in a set of desired flow transformations. While the forward problem (of predicting flow transformations given a sequence of pillars) is well studied, the design problem (of identifying a sequence given a flow transformation) is an open problem. This REU project will leverage recent advances in machine learning and data analytics to explore solution strategies for the design problems. The student will work on using hierarchical feature extraction (e.g., Deep Neural Networks) and probabilistic graphical model learning (e.g., Causal Networks) algorithms to discover the patterns in the large data sets (that are obtained by parametric sweeps of the forward problem) in a robust and scalable manner (Fig. 9) with high degree of automation.
Additional REU projects and the faculty mentors are listed in Table 2 below.
Table 2: Additional research projects and faculty mentors
(Design and Manufacturing of Sensors and Actuators)
Graphene based high performance electrochemical sensors
Nanocomposite manufacturing for sensors and actuators