The gold standard against which chemical sensors are compared is the dog?s nose. However, dogs are expensive to train and can only be used a few hours per day. By detecting the electrical signals produced by olfactory sensory neurons (OSNs), it should be possible to achieve high-sensitivity, high-specificity, high-speed, stand-off detection of trace amounts of compounds associated with volatile human compounds characteristic of gender, stress, individual ?fingerprint?, and various medical conditions. The team plans to develop a miniaturized system for human biometric characterization using, initially immortalized cell lines for detection of human compounds and then developing techniques for direct detection of airborne odorants using artificial mucous, thin membranes, continuous perfusion with water or a combination. Cell-based chemical sensors will have broad societal benefits through diverse applications, outside of biometric detection: explosives detection, monitoring food and air, odor-based medical diagnosis, drug detection in airports, and screening of pharmaceuticals, to name a few. This project will support two full time graduate research assistants. The PIs are actively engaged in educating graduate and undergraduate students and interdisciplinary curricular development. The PIs have demonstrated strong track records of mentoring and supporting students from underrepresented groups and of developing and supporting programs that train and promote these students.
The goal of this work was to do basic science answering fundamental questions about the use of olfactory sensory neurons (OSNs) for biometric assessment of human subjects, and to develop methods and technology for achieving this. The project has produced a number of significant technology developments that enable future progress towards the goal of an electronic olfactory transducer based on OSNs. The required technology developments were not anticipated at the outset of the project, however, and delayed demonstration of OSN-based odorant sensing. Most labs-on-a-chip (LOCs) today are still actually chips in labs, used with benchtop equipment; the technology that we are developing will eventually be stand-alone and portable. Three critical technologies came out of this work that should allow realization of cell-based sensors and true lab-on-a-chip devices. 1) Packaging of small, commercially-produced IC chips for operation in fluidic environments. 2) Design approaches and insights for bio-micro-electro-mechanical (bioMEMS) systems on chip. 3) Culture of electrically active cells (and OSNs in particular) on chips and sensor arrays. Packaging Packaging for operation in bio-fluids is the most important technical hurdle for hybrid bioMEMS systems-on-chip. This is because the wires that carry the electrical signals and power to/from the chip must remain dry while at the same time the nearby electrodes (25 mm distant) on the surface of the chip must be exposed to a fluidic cell culture environment. Failure of the package results in immediate system failure. Most research chips are large, so they can be packaged simply by gluing a cm-scale well to the surface to hold the fluid, and outside of that make use of wire bonding to form electrical connections. For the small (3x3 mm or less in area) state-of-the-art complementary metal oxide semiconductor (CMOS) integrated circuit (IC) chips used in our work, this approach cannot be applied. We also learned during this project that packaging methods that encapsulate wire bonds fail, typically within 1-14 days after immersing the chips into cell medium; some encapsulant materials fail sooner than others. We explored several alternative packaging methods and found one that not only works extremely well with foundry-produced CMOS dies, but is easily implemented so that it can be adopted by others. The die is first embedded in a polymer handle wafer, resulting in a planar surface that allows subsequent post-processing with standard micro-fabrication techniques. Electrical connections and insulation can be formed directly on this surface, and then standard microfluidics integrated on top. BioMEMS Systems CMOS ICs are a key emerging technology for LOCs. It is well known that ICs can perform computational tasks. They can also be used as transducers for directly sensing various types of stimuli (optical, electrical, etc.). This combination of capabilities is powerful and allows for new sensing capabilities in several ways: by having computational tasks closely tied to sensors, by allowing multiple sensors to measure different aspects of the same phenomenon at nearly the same location, and by offering high spatial density in an addressable array of sensors. The value of CMOS in LOCs can be further increased by the addition of surface micro-machined structures, including devices that can be electrically connected to the internal IC circuitry. This allows for the fabrication of complex systems that can transduce physical phenomena beyond those sensed by CMOS alone. This project has contributed towards better understanding of design considerations for bioMEMS systems on chip. We developed approaches to address thermal effects (to avoid damaging the cells), floor-planning (to be compatible with cell culture and surface microfabrication), signal coupling (altered by the fluid environment), system modeling (using a multi-step approach), and operation in cell culture conditions. We also developed an IC that combines these roles of sensing and computation; it simultaneously serves as an active microelectrode array for long term recordings of cultured neural cells, while locally detecting spike activity and communicating only spiking activity off-chip for further analysis. This reduces the required communication bandwidth by a factor of 1600, making way for higher density neural recording arrays in the future. Cell Culture Olfactory sensory neurons do not adhere and grow well on bare metal microelectrode arrays; therefore an adhesion promoter such as laminin must generally be used in order to ensure cell health and adhesion. Mammalian olfactory neurons remained viable for only a few hours after harvesting. Amphibian cells were more resilient, showing activity after nine days of storage in a refrigerator. Additionally, achieving high quality recordings requires low impedance electrode materials. In this project we found that the conductive polymer poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate), commonly known as PEDOT:PSS, offers lower impedance than materials such as gold, platinum, or even platinum black. The packaging method described above allows this cell-compatible coating to be applied over the electrodes.
