Quantificaion of ion diffusion in gallium arsenide-based spintronic Light-Emitting Diode devices using time-of-flight secondary ion mass spectrometry
Cogswell, Jeffrey Ryan
MetadataShow full item record
Depth profiling using Secondary Ion Mass Spectrometry (SIMS) is a direct method to measure diffusion of atomic or molecular species that have migrated distances of nanometers/micrometers in a specific material. For this research, the diffusion of Mn, sequentially Ga ions, in Gallium Arsenide (GaAs)-based spin Light Emitting Diode (LED) devices is studied by quantitative Time-of-Flight (ToF) SIMS. The goal is to prove conclusively the driving force and mechanism behind Mn diffusion in GaAs by quantifying the diffusion of these ions in each device. Previous work has identified two competing processes for the movement of Mn in GaAs: diffusion and phase separation. The process is dependent on the temperature the sample is exposed to, either by post-annealing, or during the molecular beam epitaxy (MBE) growth process. The hypothesis is that Manganese Arsenide (MnAs) is thermodynamically more stable than randomly distributed Mn ions in GaAs, and that by annealing at a certain temperature, a pure MnAs layer can be produced from a GaMnAs layer in a working spin LED device. Secondly, the spin efficiencies will be measured and the difference will be related to the formation of a pure MnAs layer. The first chapter of this dissertation discusses the history of spintronic devices, including details on the established methods for characterization, the importance for potential application to the semiconductor industry, and the requirements for the full implementation of spintronic devices in modern-day computers. MnAs and GaMnAs devices are studied, their preparation and properties are described, and the study's experimental design is covered in the latter part of Chapter 1. Chapter 2 includes a review of diffusion in semiconductors, including the types of diffusion, mechanisms they follow, and the different established experimental methods for studying diffusion. The later sections include summaries of Mn diffusion and previous studies investigating Mn diffusion in different materials. They describe the preliminary annealing experiments and the reasons behind choosing an annealing temperature of 250°C for this research, along with adjustments to the experimental plan and the reasons behind them. The second portion of this research is described in Chapters 3 and 4. This involves detailing the conversion of the x-axis from time to depth (Chapter 3) and the y-axis from intensity to concentration (Chapter 4) for the quantification of the ToF-SIMS depth profiles. Chapter 4 also combines both axes and details the conclusions from the research. To change the x-axis required the consideration that relatively thin layers (less than 100nm) of the different materials complicated direct comparison of pre- and post-annealed depth profiles. Instability in the current of the sputtering ion guns and the effects of annealing at different temperatures are common for SIMS depth profiling. The thin layers of the working spin LED led to problems and the establishment of new procedures for the correct conversion to depth. A quantitative SIMS depth profile has the y-axis in concentration of a particular atom per cubic centimeter (atoms/cm 3 ). To change the original y-axis from signal intensity requires the establishment of relative sensitivity factors (RSFs). This chapter details the reasons for choosing each relative sensitivity factor (RSF) used for potential conversion of the y-axis. The specific RSF values and their %Relative Standard Deviation (%RSD) are 1.38x10 23 ±5.4% (atoms/cm 3 ) for Mn in MnAs, 4.58x10 23 ±3.86% (atoms/cm 3 ) for Mn in Al 0.15 Ga 0.85 As, and 2.98x10 22 ±9.56% (atoms/cm 3 ) for Ga in MnAs. It was determined that no significant Mn diffusion was occurring in these devices at an annealing temperature of 250°C. From the quantitative depth profiles of Ga in the MnAs layer there was a 25.5% increase of the Ga concentration after the device was annealed at 250°C ± 1°C for 1 hour.