Session A12: Undergraduate Research II

Simulating Noise in Superconducting Qubits Using Qiskit

Qubits are quantum systems analogous to classical bits. Classically, errors in bits arise from environmental factors like external magnetic fields, and intrinsic factors like dissipation in resistors. To correct for these errors, bits are copied for detection and corrected. In qubits, errors can result from internal factors such as improper state preparation from inaccurate gates, or coupling with the environment, which leads to energy relaxation and decoherence. Error correction in qubits requires a new approach as qubits cannot be copied, and because they exist in superposition, phase information must also be preserved.

Through IBM Qiskit, we simulated different noise models such as bit and phase flips, energy relaxation, and decoherence errors to analyze their effect on a singlet state. We generated T₁ and T₂, times, which denote when the system exponentially decays to 63% of its initial state through energy relaxation. Then, we compared the impact of these models on the singlet state and the preparation process. Lastly, we analyzed the extent to which these models cause a complete loss of information in the qubit state and whether this can be detected and partially reversed.   

Noise Temperature Measurements of Josephson Parametric Amplifiers for the Axion Search at IBS/CAPP

Josephson parametric amplifiers (JPAs) are promising devices for amplifying the weak signal of the axion in haloscopes due to their low noise amplification and tunable resonant frequency. For the axion experiment, the noise temperature of the JPA needs to be determined at various frequencies, which requires altering the noise source temperature. Optimizing the time required to collect this data is crucial for this experiment. Two methods were used to estimate the noise temperature of the JPA and calculate the standard deviation of the fitted data. First, a modified Y-factor method was developed to calculate the noise temperature using only two noise source temperatures. Then, a method using curve fitting to estimate the noise temperature was applied, utilizing all fifty-three noise source temperature spectra, and the standard deviation was calculated. The fitting method provided much more accurate results, but the measurement time was significantly longer than what would be required for the Y-factor method.   

Integrating Molecular Rectifiers in AC Circuits

The field of molecular electronics exploits single molecules as basic components. Envisioned to provide a route to the continued satisfaction of Moore's Law, these nano-scale components exploit quantum effects to provide functionality at a length scale impossible with current technologies. This presentation focuses on molecular rectifiers based on single molecular layers sandwiched between two electrodes, operating similarly to solid-state diodes. We have developed a new molecule, (E)-1-(4-cyanophenyl)-N-(3-(triethoxysilyl)propyl)methanimine, that allows current rectification with rectification ratios greater than 2500. We further tested the functionality of molecular rectifiers made from this molecule in a standard AC circuit to create a DC rectifier. A test circuit was built where molecular diodes were connected in series with a 1 MΩ resistor and 100 nF smoothing capacitor, using a specially designed 3D-printed low-noise integration system. The rectification ratio of each diode was tested before circuit integration to associate this value with circuit performance. DC voltages measured for a variety of input frequencies show low ripple voltage, the mark of a quality DC rectifier, indicating that the experimental molecular layer acts as a stable diode in AC circuitry.   

Gating dynamics in ionic-liquid-gated FeS2 single crystals

Ionic liquid (IL) gating has proven remarkably effective in voltage control of superconductivity, insulator-metal transitions, and magnetism. This is in large part due to its high electric fields, and thus large accumulated surface charge densities (> 10¹⁴ cm^-2). Recent studies, however, emphasize the importance of distinguishing electrostatic from electrochemical gating mechanisms in such devices. Here, we present a detailed study of the transport dynamics of IL-gated FeS₂ single crystals, where a positive gate voltage is observed to induce a remarkable insulator-metal and diamagnetic-ferromagnetic transition. This transition is found to be highly reversible in transport, which, given the delicate nature of surface conduction in FeS₂ [1], strongly evidences an electrostatic gating mechanism. Hysteretic gate voltage sweeps suggest the electrostatic electron accumulation and depletion to be spatially non-uniform, with a sweep-direction-dependent percolation transition. The observation of reversible electrostatic response is discussed in terms of the formation enthalpy and diffusivity of S vacancies in FeS₂.
[1] Walter et al., Phys. Rev. Materials, 1, 065403 (2017).
  

Electrical characterization of a tungsten diselenide/silicon heterostructure

A pn diode was fabricated by using a simple method of transferring a p-type two-dimensional WSe₂ film onto a cleaved n-Si/SiO₂ wafer. The current-voltage characteristics of the device were measured, and the conduction mechanisms analyzed over a temperature range of 80 K–300 K. At high temperatures, the current-voltage characteristics of the diode show that thermionic emission transport dominates. However, tunneling also contributes at low temperatures. To explain the transport behavior of the heterojunction, a model that takes into consideration both thermionic emission and tunneling will be presented. Furthermore, the device was tested as a half-wave rectifier at room temperature at low frequencies. The rectification ratio and low turn-on voltages of the diode make it suitable for optoelectronic applications.   

Measuring the Dielectric Constants of Perovskite Nanoparticles in a Polymer Matrix

Barium titanate (BTO) is a perovskite material used in energy storage applications due to its high dielectric constant [1, 2]. Interestingly, Wada et al. found that BTO nanoparticles ranging from 17 to 500nm exhibited a dependence of their dielectric constant on particle size [2]. Particles with sizes over 300 nm exhibited a dielectric constant of 4000, but a sharp increase in dielectric constant to over 15,000 was observed at a BTO size of 70 nm [2]. To investigate the relationship between the particle size and dielectric constant of BTO, we developed an injection molding process to fabricate polymer-matrix nanocomposites containing BTO powders. Here we present our novel fabrication method and the results of our investigation, which was focused on determining the dielectric constants of nanocomposites containing BTO nanoparticles with sizes ranging from 50 to 500 nm. We will discuss our methods of (1) measuring the dielectric constants of nanocomposites and (2) extracting the dielectric constants of the BTO nanoparticles from those measurements and relating them to particle size and volume loading.

[1] Park et al. Journal of Korean Physical Society, Vol. 49 (2006): S680-S683.
[2] Wada et al. Japanese Journal of Applied Physics 42.Part 1, No. 9B (2003): 6188-195.   

Electronic Properties in Strained and Suspended Sheet of MoS2 on Anodized Aluminum Oxide.

Two-dimensional materials are as thin as physically possible and have tunable electronic properties that can be useful for the development of faster and smaller electronic devices. One way to tune the electronic properties of 2-D materials is to induce strain by deforming the lattice. In addition to strain impact on the electronic properties, a suspended sheet of 2-D material exhibits higher electrical conductivity. In this project, I induce strain by deforming a semiconductor, MoS₂, lattice with a nanopatterned substrate, and characterize the strain by atomic force microscopy (AFM) and Raman spectroscopy. Here I present the results of patterning strain and suspension into MoS₂, and their impact on the electronic properties of MoS₂. Single layer MoS₂ is isolated via mechanical exfoliation and transferred onto anodized aluminum oxide (AAO) - a substrate with nanoscale valleys and hills. The average strain in the MoS₂ sheet is 0.23% and 0.29%, which obtained from the Raman spectroscopy and AFM data, respectively. AFM techniques such as KPFM, C-AFM, and TERS are used in this project to characterize the electronic properties in the strain-textured MoS₂.   

Understanding the Effects of Controlled Strain on Low Dimensional Material Properties

A material is typically considered low dimensional if electron movement within the material is limited, for example, to within a particular ionic plane or along an axis. When strain is introduced to an ideal material, lattice deformations may cause substantial changes in structural and electronic properties. Studying the effects strain have on a material can lead to basic insights into the physics governing the material as well as an understanding of how a material’s properties can be manipulated for use in applications. Strain can be introduced to a low dimensional material through processes such as sample cleaving, elemental doping, chemical pressure, and external mechanical application. First, we detail our efforts in characterizing the expected strain delivered to the sample from our external application as a function of temperature. We then present resistivity measurements showing the effects strain has on the bulk phase transition temperature of a charge density wave compound. We will discuss how this system can be used in further studies that characterize strain on the nanoscale.   

Studying the Crystallization of ALD-Deposited Doped Nb2O5 into NbO2 for Next Generation Electronics Applications

Phase change materials have a variety of applications across many disciplines. They can be used in homes to produce more energy-efficient refrigerators and freezers, or in clothes to help regulate body temperature in varying climates. NbO₂ is one such material, and can be used in a variety of electronic applications. In this project, we studied the conversion of amorphous Nb₂O₅ thin-films doped with either Al₂O₃ or ZnO (supplied by our collaborators at the Naval Research Laboratory) to crystalline NbO₂. Previous studies show Nb₂O₅ can be converted into NbO₂ by annealing to temperatures around 800 C. Systematic anneals in a tube furnace showed how the dopants affect the conversion of Nb₂O₅ into NbO₂. Our results indicate that samples of Nb₂O₅ doped with Al₂O₃ convert much slower than pure Nb₂O₅, and those with ZnO appear to convert faster.   

A Graphene Encapsulated Growth Method to form Ultra-thin Magnesium Diboride (MgB2)

There is interest in coupling 3D topological insulators (TI) with s-wave superconductors (SC) for realizing topological superconductivity (TSC) for quantum computing applications. MgB₂ is an intriguing superconductor for this application as it has T_c = 39K and a hexagonal crystal structure, compatible with common Bi-based TIs. However, MgB₂ oxidizes in air making it difficult to form high quality TI/SC interfaces. Here, we propose a novel approach to synthesize ultra-thin MgB₂ via intercalation between epitaxial graphene (EG) and SiC. EG is used to protect the MgB₂ surface while enabling epitaxial growth of the TI on top. Initial studies investigated Mg intercalation between EG and SiC using a tube furnace. X-ray photoelectron spectroscopy (XPS) and cross section transmission electron microscopy-energy dispersive spectroscopy (TEM-EDS) confirmed Mg intercalation between EG and SiC. Current studies are being performed to optimize process conditions to intercalate Mg. The results of planned investigations to convert Mg to MgB₂ via annealing in B₂H₆ will also be discussed.   

Temperature Dependent Switching Dynamics in BaTiO3 at ns Speeds

BaTiO3 (BTO) is a model ferroelectric, which has been studied extensively, both in bulk crystals and thin films. We have made significant progress, achieving ultra-low coercive fields in thin film BTO samples that hold promise for low-power microelectronics. While there has been extensive work studying switching kinetics in BTO, relatively few studies have performed time-domain experiments on timescales in the low ns regime, since they are typically limited with instrumental constraints (e.g., rise time of oscilloscope or the voltage source). We are able to access this regime by using ultra-low coercive field BTO as a model system and measuring the response to voltage pulses with ~100ps rise time. We present data revealing switching speeds faster than previous reports for BTO, on the order of 1ns. Further, with this setup we study switching dynamics as a function of temperature as we approach the ferroelectric phase transition. The data reveals a decrease in switching time with increasing temperature, which we analyze in the context of classical ferroelectric switching theory.   

Detection of Phase Transitions in Phase Separated (La1-yPry)1-xCaxMnO3 Thin Films

Phase separated manganites have competing phases that each have unique electronic, magnetic, and structural properties. The similar free energies these phases possess allow for the manipulation of phase transitions with the application of external stimuli. Here we report the role of an external magnetic field and a change in chemical doping ratios on the phase transitions in (La_1-yPr_y)_1-xCa_xMnO₃ thin films grown on NdGaO₃. These thin films showed that the nucleation and growth of ferromagnetic metallic regions in an anti-ferromagnetic charge ordered insulating background occurs at a higher temperature when an external magnetic field is applied. At a high enough field strength, these ferromagnetic metallic regions hinder the phase transition from paramagnetic insulating to anti-ferromagnetic charge ordered insulating phase which occurs at a relatively high temperature (close to 200 K). The value of the activation energy for the anti-ferromagnetic charge ordered insulating phase calculated by fitting the resistance vs. temperature data to the Arrhenius equation agrees with direct measurements taken using scanning tunneling spectroscopy.   

Doublons Instability in the Presence of Impurities

In the presence of strong interactions, particles can bind in pairs forming what became known as doublons. A doublon moves together as a single particle, but contrary to it, doublons move slowly. They emerge in common models of condensed matter physics, such as the anisotropic Heisenberg model and the Bose-Hubbard model. We show that doublons are very stable and do not split up even when this would not violate energy conservation. We illustrate our results using one- and two-dimensional spin models in the presence of impurities.   

Localization in two-dimensional trivial and Chern insulators

We investigate the localization properties of two-dimensional tight-binding models using transfer matrix methods. In particular, we numerically determine the localization length for cylindrical lattices with single-particle hamiltonians that have nearest-neighbor and next-nearest-neighbor hopping terms, both in the trivial and Chern insulator phases. We consider the role of disorder strength and particle energy on localization, and construct phase diagrams of conductivity properties. In addition, we use these results to guide an exploration of the critical behavior of localization-delocalization transitions in these models.