Past Awards

4th-8th Award Winners 2024.pdf

HS Award Winners 2024.pdf

2024 ISEF Finalists From CKRSEF

Please note that some of these finalists qualified to ISEF from the KY State Science Fair

MATS014 - Doping in Organic Electrochemical TransistorsOrganic electrochemical transistors (OECTs) are among the most powerful transistors to date by combining both ionic and electronic conduction. This unique mechanism makes these devices desirable for implantable bioelectronics in the human body, particularly as biosensors, deep brain stimulators, pacemakers, and artificial muscles, reducing reliance on invasive medical procedures. However, electron-conducting (n-type) OECTs, suitable for bioelectronic applications, historically suffer from instability and low performance, limiting their practical application. Chemical doping was investigated as a novel and cost-effective method to enhance device performance, successfully identifying a new dopant and unique doping techniques. Various organic salts were explored as potential dopants because of their strong Lewis base makeup. Current-voltage tests and electrochemical impedance spectroscopy characterized and compared undoped and salt-doped n-type OECT performance. Doping with tetrabutylammonium chloride (Bu4NCl) salt improves the device’s transconductance, mobility, signal-to-noise ratio, threshold signal, and capacitance. The Bu4NCl dopant’s statistically significant improvements in metrics yield a 97% improvement in amplification and a 77% improvement in switching speed and charge storage. Doping with Bu4NCl is optimized at a concentration of 20 molar percentage and a solvent blend of chlorobenzene to chloroform at a 1:5 ratio, enabling the salt to promote charge transfer and delocalization in the polymer network. Bu4NCl is identified as a new dopant to fabricate n-type OECTs with high performance. This study is the first to explore dopants, doping concentration, and solvent design in conjunction, advancing chemical doping in tailoring OECTs to become commercially viable.

ETSD035 - An Improved Color Filter Array Design Color image sensors typically rely on absorbing filters arranged in the Bayer pattern (red, 2×green, and blue). Dye-and pigment-based filter arrays have several disadvantages such as photo- and thermal- instability, a multi-step fabrication process, and reduced camera sensitivity from absorption. These disadvantages have motivated the investigation of alternatives to reduce fabrication cost and enhance camera sensitivity. Most approaches focus on structure-based filters. Here I present an alternative, structure-based CFA that combines dielectric phase plates and pillar-based metasurfaces. The CFAs were fabricated using two-photon lithography (Nanoscribe Photonic Professional GT) with IP-Dip resist. CFAs have metafilter sizes of 13.2 × 13.2 μm, a pillar periodicity of 1.2 μm, and a focal length of 18 μm. This design allows the zero and first order focal spots to be captured by separate pixels on the image sensor. The total efficiency of the sensor, which is close to 60% across much of the visible region, significantly exceeds most Bayer arrays. The current design also offers a flatter spectral response, which may be desirable under some lighting conditions.. The maximum color error of the filters is 5.5∆E and the average error is 2.9∆E in CIE xy coordinates. These results are comparable or better than the errors of common cellular phone and DSLR digital cameras.

ROBO060 - Lane Detection With Enhanced ReproducibilityIn 2019 alone, there were reportedly over thirty-thousand driver related deaths and over three-million driver related injuries. Interestingly, over 60% of these deaths and around 200,000 of these injuries could have been prevented with some type of advanced driver assistance system (ADAS) onboard. Lane detection, a critical component of ADAS technology, aims to accurately identify road lane boundaries from camera or other inputs to enable functions like lane keeping assistance and autonomous navigation. However, existing state-of-the-art lane detection models still struggle in challenging real-world conditions such as temporary occlusions, changing environments, and complex urban scenarios. This project presents TempURRA-NET, a novel neural network architecture that integrates a backbone-supported U-NET with iterative refinement, recurrent components, and temporal data handling to enhance lane detection robustness, consistency, and prediction stability. Additionally, to address the lack of reproducibility and access to easy model evaluation in this field, Lane.EV was created: a standardized model evaluation platform that streamlines the testing process of lane detection and computer vision models. TempURRA-NET achieves a max average accuracy of ~98% on the TuSimple benchmark, outperforming previous SOTA methods while Lane.EV ensures the reproducibility and sustainability of these results.

MATH014 - Twisted Homogeneous RacksAn important step towards the classification of finite-dimensional pointed Hopf algebras, which are algebraic structures with a wide range of fundamental applications to mathematics and physics, is the classification of finite-dimensional Nichols algebras arising from braided vector spaces of group type. This question is fundamentally linked with the structure of simpler algebraic objects called racks. Of particular interest to this classification is the type D condition on racks, a sufficient condition for a rack to not be the source of a finite-dimensional Nichols algebra. In this project, the type D condition in simple racks arising from the alternating groups was studied. Several families of twisted homogeneous racks arising from alternating groups are shown to be of type D, expanding upon previous work in this direction. This is important progress towards a general classification of twisted homogeneous racks of type D, and ultimately towards a classification of finite-dimensional pointed Hopf algebras.

Page updated

Google Sites

Report abuse