Uttam Singisetti

University at Buffalo
Country flagUnited States

Research Interests

ApplyKite TwinLearns as you apply

Make your next move.

Prepare your application

Review drafts before you send them.

Explore related searches

Articles (6)

Gallium oxide HV switches to turbo charge the AI data center Prof. Singisetti at the University at Buffalo explains why gallium oxide semiconductor could be a key player in the AI transformation of society. The release of ChatGPT in 2022 has led to an explosive increase in Artificial Intelligence (AI) applications and investments. The large language model (LLM) AI has proven to be extremely capable of performing complex tasks across a broad range of topics, yielding enormous economic benefits. This has led to massive investments in AI data centers. However, a significant challenge is the substantial energy demand of AI to perform even basic tasks. The anticipated demand for AI energy has bolstered nuclear power as a reliable energy source. The increased demand has also led to plans for high-voltage transmission lines in the US. As the AI data centers proliferate, power semiconductors will play a crucial role in integrating the new energy sources into the grid.

Year:

2025

β -(AlxGa1−x)2O3 (AlGaO) is a promising ultra-wide bandgap semiconductor (UWBG) that can be utilized in AlGaO/Ga2O3 heterostructure devices for a wide range of electronic applications. To fully exploit the benefits of AlGaO alloys, it is important to understand the fundamental transport properties of the materials. Thus, in this work, the Effective Phonon Dispersion (EPD) of β-(AlxGa1−x)2O3 alloys is investigated using the phonon unfolding formalism. To capture the true randomness of the system, supercells (SCs) are designed using the technique of special quasirandom structures (SQS). The phonon unfolding technique is then used to obtain the EPD for aluminum fractions of 18.75% and 37.5% with respect to gallium atoms. Using the EPDs, the sound velocity of the alloys was calculated to gain further insight into the thermal conductivity. The disorder’s impact on the high frequency and DC dielectric constant is also investigated. The unfolding procedure was also applied to the dipole moment, and an approach to theoretically predict the IR spectra in disordered systems was also proposed. Using an SC approach, the low-field transport in β-(AlxGa1−x)2O3 alloys was investigated to understand the effects of electron–phonon scattering and alloy scattering. The calculated properties give us valuable insight into calculating the transport properties in these alloy semiconductors for device applications while capturing the true randomness of the system.

Year:

2025

Year:

2024

A systematic investigation of the electrical characteristics of β-Ga2O3 Schottky barrier diodes (SBDs) has been conducted under high-dose 60Co gamma radiation, with total cumulative doses reaching up to 5 Mrad (Si). Initial exposure of the diodes to 1 Mrad resulted in a significant decrease in on-current and an increase in on-resistance compared to the pre-radiation condition, likely due to the generation of radiation-induced deep-level acceptor traps. However, upon exposure to higher gamma radiation doses of 3 and 5 Mrad, a partial recovery of the device performance occurred, attributed to a radiation annealing effect. Capacitance–voltage (C–V) measurements showed a decrease in net carrier concentration in the β-Ga2O3 drift layer, from ∼3.20 × 1016 to ∼3.05 × 1016 cm−3, after 5 Mrad irradiation. Temperature-dependent I–V characteristics showed that 5 Mrad irradiation leads to a reduction in both forward and reverse currents across all investigated temperatures ranging from 25 to 250 °C, accompanied by slight increases in on-resistance, ideality factors, and Schottky barrier heights. Additionally, a slight increase in reverse breakdown voltage was observed post-radiation. Overall, β-Ga2O3 SBDs exhibit high resilience to gamma irradiation, with performance degradation mitigated by radiation-induced self-recovery, highlighting its potential for radiation-hardened electronic applications in extreme environment.

Year:

2024

To investigate the transport properties in random alloys, it is important to model the alloy disorder using supercells. Although computationally expensive, the local disorder in the system is accurately captured as translational symmetry that is imposed on the system over larger length scales. Additionally, in supercells, the error introduced by self-image interaction between the impurities is reduced. In this work, we have investigated the Effective Phonon Dispersion (EPD) and transport properties, from first principle calculations using supercells in AlxGa1−xN alloy systems. Using an in-house developed code for phonon-band unfolding, the EPD of AlGaN is obtained and the individual phonon modes are identified with good agreement with experimental values. Moreover, we report an in-house developed method to calculate low-field transport properties directly from supercells without phonon band unfolding. First, to validate our methods, we have solved the Boltzmann transport equation using Rode’s method to compare the phonon limited mobility in the 4 atom GaN primitive cell and 12 atom GaN supercell. Using the same technique, we have investigated the low field transport in random AlxGa1−xN alloy systems. The quadrupole interaction is included for transport properties of GaN and AlN to accurately capture the physics in these materials. Our calculations show that along with alloy scattering, electron–phonon scattering may also play an important role at room temperature and high-temperature device operation. This technique opens up the path for calculating phonon-limited transport properties in random alloy systems.

Year:

2024

This study provides a comprehensive analysis of the dominant deep acceptor level in nitrogen-doped beta-phase gallium oxide (β-Ga2O3), elucidating and reconciling the hole emission features observed in deep-level optical spectroscopy (DLOS). The unique behavior of this defect, coupled with its small optical cross section, complicates trap concentration analysis using DLOS, which is essential for defect characterization in β-Ga2O3. A complex feature arises in DLOS results due to simultaneous electron emission to the conduction band and hole emission to the valence band from the same defect state, indicating the formation of two distinct atomic configurations and suggesting metastable defect characteristics. This study discusses the implications of this behavior on DLOS analysis and employs advanced spectroscopy techniques such as double-beam DLOS and optical isothermal measurements to address these complications. The double-beam DLOS method reveals a distinct hole emission process at EV+1.3 eV previously obscured in conventional DLOS. Optical isothermal measurements further characterize this energy level, appearing only in N-doped β-Ga2O3. This enables an estimate of the β-Ga2O3 hole effective mass by analyzing temperature-dependent carrier emission rates. This work highlights the impact of partial trap-filling behavior on DLOS analysis and identifies the presence of hole trapping and emission in β-Ga2O3. Although N-doping is ideal for creating semi-insulating material through the efficient compensation of free electrons, this study also reveals a significant hole emission and migration process within the weak electric fields of the Schottky diode depletion region.

Year:

2024

Collaborators (4)

Steven A. Ringel

Research Scientist

Ohio State University

UNITED STATES
View details

Man Hoi Wong

Associate Professor

The Hong Kong University of Science and Technology

HONG KONG
View details

A F M Anhar Uddin Bhuiyan

Assistant Professor

University of Massachusetts Lowell

UNITED STATES
View details

Steven A. Ringel

-

UNITED STATES
View details