NSF Award
2430828
Future electronics manufacturing will not only encompass devices like chips and chiplets but also integrate passive components, sensors, and Radio Frequency (RF) front ends with power amplifiers into high-density heterogeneous packages. Currently, package components are tested individually and reassembled on high-density interposer substrates. However, future trends indicate that the interconnect pitch between devices will scale down to less than a micron. This shift towards high-density integration necessitates more sophisticated methods to mitigate electromagnetic interference (EMI) and ensure electromagnetic compatibility (EMC) within the overall package. With this in mind, this research focuses on thin film magnetodielectric materials and electronic package topologies, coupled with Multiphysics digital twin models, aimed at suppressing: 1) jittering and degradation of bit error rates, 2) switching noise in digital integrated circuits due to emissions or crosstalk, 3) close proximity coupling and radiation due to high-speed digital or analog interconnects, and 4) transmission line crosstalk due to conduction or ground bounces of electromagnetic fields. The primary goal of this project is to develop new techniques for predicting and suppressing signal interference in future high-density electronic packages. Concurrently, the project will develop a comprehensive multiphysics toolset to integrate circuit extraction methods and materials for EMI suppression with popular chip and electronic/RF circuit toolsets seamlessly. This research is expected to impact the rapidly growing U.S. semiconductor industry and enable the packaging of reliable, vertically integrated electronics. Additionally, it will substantially enhance the training of a diverse group of students by leveraging Florida International University's unique position as the only Majority-Minority Carnegie R1 Research University in the continental U.S. This research will further advance educational efforts to broaden participation of women and other underrepresented groups in STEM through curriculum development and REU programs.<br/><br/>The aim of this research is to pioneer disruptive EMI/EMC mitigation techniques in high-density heterogeneous semiconductor device packages through digital twin emulations. The goal is to advance knowledge and enable optimal embedded packaging with heterogeneous active and passive components to ensure future electronic package reliability and performance. To address these challenges, the project will investigate: a) new EMI/EMC ferromagnetic composite materials, b) novel packaging topologies, and c) advanced computational and circuits extraction methods. A number of innovations are expected: 1) multilayered films formed of cobalt nickel-iron alloy (CoNiFe) and copper (Cu) layers, but still only 5 micrometers thick, to achieve maximum shielding with minimal film thickness. As much as 30 dB more shielding can be attained using these composite films; 2) New shielding topologies for reliable interference suppression using prefabricated forms to enable rapid electronic package formations. These are aimed at suppressing high-power device interference and high-speed circuit radiation at the chip interconnects (as much as 40 dB or more); 3) New class of multiphysics toolsets that cut across electromagnetic, thermomechanical and thermal designs to provide system performance and reliability. Such multiphysics models are of critical importance as interactions among various physics domains imply design trade-offs needing quantification; 4) AI multiphysics models that combine model-agnostic meta-learning and physics-informed learning approaches using measured data as well as analytical and semi-analytical models for accurate and robust modeling; 5) Novel S-parameter extraction methods that uniquely account for coupling and external fields using new port excitations within the circuit network of the multiphysics model.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
