(698g) Pit Rim Decomposition into Multiple Quantum Dots on Surfaces of Epitaxial Thin Films Grown on Pit-Patterned Substrates

Semiconductor nanostructures such as quantum dots (QDs) and nanorings enable numerous applications in electronic and optoelectronic device technologies, as well as data storage systems. The Stranski-Krastanow (SK) growth instability is a common approach to trigger the formation of such nanostructures on surfaces of thin films. These films are deposited epitaxially on thick semiconductor substrates and growth instabilities are driven by the induced biaxial stress in the growing film because of the lattice mismatch between the deposited film and substrate materials. However, the QDs forming through SK growth instabilities are randomly arranged on the film surface and their size and distribution are not uniform, while uniform positioning and ordering of quantum dots is required for device fabrication purposes. Recently, numerous strategies have been developed for guiding the growth of QDs that are uniformly arranged and consistently sized. Experimental studies have shown that, among such strategies, depositing thin films epitaxially on properly engineered pit-patterned substrate surfaces is a promising, effective approach toward the assembly of ordered nanostructures.

Here, we report results of the surface morphological evolution of coherently strained Ge thin films grown epitaxially on pit-patterned Si {100} substrates. Our analysis is based on self-consistent dynamical simulations according to an atomistically-informed, 3D continuum-scale epitaxial film surface evolution model that has been validated by comparisons of its predictions with experimental observations on Ge/Si and InAs/GaAs heteroepitaxial systems employing pit-patterned substrates. We focus on patterned substrates with inverted truncated pyramidal pit geometries and examine the effects of varying the relevant geometrical design parameters on the resulting nanopatterns that are formed on the rim of the pits in the epitaxial film surface. These geometrical parameters include the pit opening dimensions, length and width, the pit wall inclination, and the pit depth. Varying the pit opening length and width leads to the formation of nanostructures on the rim of the pits, ranging from elongated QDs on each side of the pit rim to 1D arrays of multiple QDs as a result of the decomposition of each of these elongated QDs. While varying the pit wall inclination and the pit opening dimensions have direct impact on the number of QDs forming on the rim, varying the pit depth only affects the size of the generated QDs. The results of our computational analysis are in good agreement with the predictions of linear and nonlinear morphological stability theories for the number of QDs resulting from the pit rim decomposition and the dimensions of these QDs. This agreement between simulation results and theoretical predictions establishes the observed nanopattern formation upon decomposition of the pit rims on the epitaxial film surface as the outcome of a “tip-splitting” instability that accompanies the SK growth instability.

Furthermore, we have studied the effect of pit-pit interactions on the deposited film surface morphology for films grown epitaxially on pit-patterned substrates. We find that the pit-pit interaction energetics follows a power law, with the interaction energy decreasing with increasing pit separation distance (i.e., pit-pattern period), and with the interaction strength increasing with increasing pit size. This pit-pit interaction energetics provides a theoretical explanation for the film surface patterns predicted by our simulations. Our findings have important implications for designing optimal surface patterns of ordered semiconductor nanostructures in coherently strained epitaxial thin films toward enabling engineering strategies for future nanofabrication technologies by exploiting surface morphological instabilities.