Research

Photonic sensors, which use light to detect environmental changes, are critical for applications in deep-earth drilling, deep-sea exploration, nuclear reactors, and outer space. Conventional platforms such as silica fibers and silicon photonics often degrade under high temperatures or in corrosive environments, limiting their reliability. Silicon carbide (SiC) offers a promising alternative due to its thermal stability and chemical resistance, but current SiC photonics face challenges including integration of crystalline SiC, limited sensing resolution at high temperatures, and insufficient characterization under extreme conditions.

At AMPS Lab, we are advancing amorphous SiC (a:SiC) photonic sensors that can be fabricated with standard foundry processes, integrated with functional materials, and characterized for reliable operation in harsh environments. Using sputtering, we synthesize thin-film a:SiC and fabricate devices such as directional couplers, ring resonators, and interferometers with electron-beam lithography. When combined with phase-change or magneto-optical materials, these devices become the foundation for resonators, polarimeters, and spectral filters used in thermography, gas sensing, magnetometry, and LIDAR. This research addresses key challenges in fabrication precision, optical losses, materials integration, and extreme-environment passivation—laying the groundwork for next-generation multifunctional sensors. We are looking for collaborators working in coatings and passivation layers that protect devices from high temperatures. 

The increasing prevalence of data-intensive applications propelled by advancements in machine learning and artificial intelligence gives rise to an unparalleled need for rapid data transfer and exceptionally fast data processing capabilities. These requirements place an ever-greater strain on the energy and link budgets of current computing systems, often yielding suboptimal remedies for this burgeoning challenge. Although conventional electronic systems are diligently evolving to meet the escalating data demands, substantial enhancements can only be realized by pushing computational frontiers beyond the confines of electrons.

Beyond electrons, photons present compelling opportunities to enhance energy efficiency and reduce latency in photonic computing architectures. Notably, consolidating processor and memory units into a single in-memory computing unit can eliminate energy costs associated with data transfer. Our research endeavors to create an integrated photonic in-memory processor by addressing fundamental questions about:

1. Identifying materials and their properties that enable in-memory computing.
2. Defining data storage formats, associated read-write techniques, and characterizing long-term bit-stability and reproducibility.
3. Designing photonic architectures that facilitate simultaneous amplitude, phase, and polarization manipulation.

To this end, our current projects involve the investigation of materials chemistry that extends traditional phase change materials like GeTe to also exhibit magneto-optical properties, which will enable simultaneous amplitude and polarization modulation. We are also developing directional couplers with ferroelectric material cladding (with collaborators who can deposit such materials using ALD) to achieve non-volatile hybrid processing units (HPU) in an electro-optical platform. We are actively looking for collaborations to design and fabricate photonic devices and circuits, which can be combined with the materials from our lab to achieve in-memory computing operations. 

Highly efficient electric power conversion is critical for enabling all-electric propulsion systems and lossless energy transmission from solar and nuclear sources—technologies that are essential for future advanced air mobility. At the core of power conversion are soft magnetic materials, whose high saturation induction and low coercivity allow compact, low-loss components such as inductors and transformers. State-of-the-art amorphous and nanocomposite materials show promise, but magnetic cores in megawatt- and gigawatt-scale converters often face rapid high-temperature cycling, which degrades efficiency over time.

Our research addresses this challenge by exploring transition-metal and rare-earth dopants in cobalt–iron–boron alloys to enhance thermal stability and magnetic performance. For example, we study Zr- and V-doped alloys to raise phase transition temperatures and reduce BH losses, fabricating thin films and scaling promising compositions into bulk melt-spun ribbons with collaborators. We are also investigating rare-earth additions to promote antiferromagnetic exchange interactions, enabling higher-frequency operation and reducing the size and power density limits of inductors. Ultimately, our goal is to realize integrated inductors that operate reliably under rapid thermal cycling and harsh conditions representative of advanced power systems.

Harnessing shorter wavelength regimes such as millimeter waves enables high-capacity communication channels with bandwidths above 10 GHz and data rates over 100 Gbps. Yet, most existing electronics operate only at fixed or narrow frequency ranges due to structural resonance limits. At AMPS Lab, we explore magnetic materials with tunable ferromagnetic resonance to overcome this barrier and create widely adaptable wireless devices. One project investigates yttrium iron garnet (YIG) integrated with patch antenna arrays to form a reconfigurable intelligent surface (RIS), capable of selectively reflecting different frequencies by dynamically tuning the magnetic resonance condition. We are actively seeking collaborations with groups that can provide expertise in the following areas –  1. Thick-ferrite materials (1-10 um), 2. FMR characterization of novel magnetic thin films and 3. High-frequency characterization of devices between 10-50 GHz and beyond.

Cytochrome P450 (CYP450) enzymes are responsible for metabolizing a vast number of therapeutic compounds, but with over 200,000 isoforms, they can produce highly variable drug responses across individuals. This variability complicates dosing, as current pharmacogenetic tests require extensive preclinical and clinical validation for each drug–enzyme combination and fail to account for dynamic factors such as infections, comorbidities, or nutrition. Without a way to separate signals from specific CYP–drug interactions, monitoring phenotypic drug metabolism remains a challenge, often leading to adverse effects in slow metabolizers or diminished therapeutic outcomes in fast metabolizers.

At AMPS Lab, we are developing multiplexed biosensors that can distinguish electrochemical signals from different CYP–drug interactions without prior genetic knowledge. Our approach addresses major bottlenecks in protein immobilization and electron transfer by using nanomagnetic entrapment and spin-selective interfaces. One project explores nanoparticle linkers to orient proteins and improve electron transfer efficiency, while another investigates spin-selective electrodes that leverage the principle of chiral-induced spin selectivity (CISS). These strategies aim to increase sensitivity, improve signal-to-noise ratios, and enable reliable real-time monitoring of drug metabolism—paving the way for truly personalized medicine. If you are in life sciences, biomedical engineering or have experience with protein electrochemistry, we would love to discuss these ideas for a potential collaboration.