DOI
https://doi.org/10.25772/X27R-5126
Author ORCID Identifier
https://orcid.org/0000-0003-0974-0227
Defense Date
2023
Document Type
Dissertation
Degree Name
Doctor of Philosophy
Department
Nanoscience and Nanotechnology
First Advisor
Indika Arachchige
Abstract
Semiconductor materials are a fundamental component of many important devices and technologies, with applications ranging from light harvesting and photovoltaics,1 to catalysis,2 diodes,3 and bioimaging agents.4 Among the semiconductors which have been characterized, many are made from highly toxic elements (CdSe, PbS, GaAs, etc.), thus limiting their practical applicability. The group IV semiconductors, however, Si and Ge, are far less toxic,5 and also at relatively high abundance,6 making them more environmentally friendly and economically feasible in comparison to other materials. Early solid-state processing of Si and Ge formed the basis for much of our modern understanding of materials science. However, the utility of Si and Ge are hindered due to their indirect fundamental bandgap,7 which limits their absorption and luminescence properties. Experimenters have addressed this by discovering that Ge undergoes an indirect-to-direct transition upon being alloyed with ~6% Sn.8 This was first accomplished in thin films grown by chemical vapor deposition9 and molecular beam epitaxy,10 where Sn compositions of up to 15% were reached without β-Sn segregation. Additionally, this creates an avenue for bandgap tunability, as a higher Sn content will result in a narrower bandgap.11 Regarding group IV semiconductors and their alloys, there are several advantages gained in the transition from bulk to nanoscale crystals. When the spatial dimensions of a semiconductor crystal are less than the Bohr exciton radius, the electronic structure changes from a conduction band/ valence band, and begins to approximate a HOMO/LUMO structure.12 This eliminates the disadvantages from an indirect bandgap, as rare photon-phonon interactions are no longer required to make a transition. Additionally, semiconducting nanocrystals (NCs) display size-based energy gap tunability as a consequence of quantum confinement phenomena.13 This opens up another route by which the properties of the material can be adjusted through variations in experimental procedure. In recent years, Ge1-xSnx NCs have been thoroughly characterized through a series of theoretical and experimental studies,14-16 and can be synthesized at diameters as low as 1.4 nm,17 and with Sn compositions as high as 95%.18 This results in a range of photoluminescence tunability from 1.31-2.05 eV,19 extending from the NIR range into low-visible light. Further alloying with Si has been theorized to increase the range of energy gap tunability deeper into the visible light range,20,21 allowing for additional compatibility with a wider variety of electronic and optical technologies. First, we report a colloidal synthetic procedure for bulk-like (≥ 10 nm) and quantum confined (≤ 4.5 nm) GeSiSn alloy NCs. This is accomplished through the reduction of halide precursors by n-butylithium in oleylamine. The resulting NCs have tunable Si content of 0.9- 16.1%, and Sn content from 1.8-14.9%. Particle morphology and elemental distribution is analyzed through electron microscopy and elemental mapping methods, where structural homogeneity is confirmed. These, along with X-ray diffraction (XRD) verify the absence of β-Sn, GeO2, and Si impurities. Ge-Ge Raman peaks range from 285.0-291.4 cm-1 (small NCs) and 283.4-288.8 cm-1 (large NCs), redshifted from the pure Ge vibration at 300 cm-1 due to the combined effects of Sn alloying and phonon confinement. A Ge-Si vibration also appears at ~430 cm-1, with intensity increasing with Si content. Finally, solid state absorption spectra of the smaller NCs are used to evaluate the composition-tunable energy gaps. The GeSiSn alloys have an energy gap range of 1.21-1.94 eV, increasing in a manner directly proportional to Si content. This range is greater than that of GeSn NCs at a similar diameter (0.75-1.29 eV), thus representing an expansion in the energy gaps which are achievable for group IV nanomaterials. Although GeSn NCs have seen significant advancements in synthetic methods and photophysical properties, their practical applicability is limited due to problems pertaining to surface chemistry. Because atoms on the surface of a nanocrystal are at a relatively high energy, they have a tendency to break off, leaving defects and cations in their place.22 These cations have a very high affinity for excited electrons, and thus, enable nonradiative decay pathways which can be deleterious to the optical properties of the crystal.23 A solution to this is to passivate the surface with a sturdy, inorganic material which will saturate surface cations and prevent defects or oxides from forming on the core particle surface.24 Silica, SiO2, is a great candidate for this, because not only does its amorphous nature eliminate complications from lattice mismatch, but also, silica-coated nanomaterials are colloidally stable in polar solvents at physiological conditions, allowing for potential applications in bioimaging.25,26 The second project published herein is the improved surface passivation of GeSn NCs through amorphous SiO2 shell growth. This proceeds through the base-catalyzed hydrolysis and condensation of tetramethyl orthosilicate (TMOS). The organically passivated core particles are first dispersed in 9 mL of toluene. 0.9 mL DMSO is added, and at these volumes, the two solvents are entirely immiscible and phase separate. Next, the TMOS and base catalyst are added, and the shell growth reaction is allowed to proceed. Afterwards, the core-shell products are collected in the DMSO phase, where they are colloidally stable, as the core particle’s surface functionality has changed. Here, the particles can be precipitated for solid-state analysis. We characterize both larger, bulk-like and smaller, quantum confined GeSn@SiO2 core-shell nanocrystals. Electron microscopy allows us to visualize the crystalline core and amorphous shell, as well as confirm colloidal stability in DMSO. We see that upon shell growth, the core crystal structure is maintained, although the diameter has reduced due to etching effects. The SiO2 shell is thin, with an average diameter of ~1.8 nm. STEM-HAADF elemental mapping is used to differentiate the GeSn cores from the silica shells. XRD of the larger NCs shows a slight reduction in crystallite size (-1.2 nm) due to the etching, although there is an absence of GeO2 or β-Sn impurities. The larger NC core-shell Raman peak is also redshifted 3 cm-1 relative to the cores, because of enhanced phonon confinement effects in etched particles. The smaller GeSn@SiO2 core-shell NCs are prepared at four different compositions: 3.1%, 4.3%, 6.4%, and 8.9%. These Raman peaks range from 288.0-288.9 cm-1 (cores), and 286.2-288.0 cm-1 (core-shells), shifted from the pure Ge-Ge peak at 300 cm-1 due to the combined effects of alloying and phonon confinement. XRD shows that the diamond cubic crystal structure is maintained for all of the compositions studied, and is not disrupted upon growth of the silica shell. Solid-state absorption spectra display composition-based energy gap tunability in the range 0.91-1.67 eV for the cores, and 1.23-2.07 eV for the core-shells, proving that important optical properties of the core are maintained after shell growth. Finally, XPS wasperformed on the core and core-shell samples to show a reduction in surface cations and the overall binding energy of the atoms, verifying the improved passivation of the nanocrystal surface. Core-shell XPS for both larger and smaller NCs showed a significant reduction in Ge2+ intensity, as well as a significant recovery in Ge0 . Additionally, the core-shell samples showed no presence of Ge4+, further confirming that the shell growth procedure does not oxidize the surface Ge. Together, these results affirm the importance of Ge-based alloy semiconductor NCs in forming new and practical photonic and electronic technologies.
Rights
© The Author
Is Part Of
VCU University Archives
Is Part Of
VCU Theses and Dissertations
Date of Submission
5-4-2023