Highly Selective Wet Etching of Ge Over Ge_{1- x} Sn _x Enabling Fully Released Group IV Microstructures

Selective etching between closely related group-IV semiconductors is a critical yet challenging requirement for micro and nanofabrication of advanced electronic, photonic, and thermoelectric devices. Isolating Ge_1-x Sn _x epilayers from underlying Ge buffers is hindered by their similar chemical behavior and by the mechanical coupling imposed by heteroepitaxial growth on silicon. Here, we report a highly selective isotropic wet-etch process that enables the controlled removal of Ge relative to Ge_1-x Sn _x , facilitating the fabrication of fully released Ge_1-x Sn _x microstructures. Using an ammonia-peroxide-water etchant, we demonstrate a strong composition dependence of the etch rate, with pure Ge etching at ∼150 nm/min while Ge_1-x Sn _x containing ≥6.8% Sn exhibits etch rates below 17 nm/min, yielding selectivity up to ∼1:30. For lower Sn contents, selective protection via aligned lithography enables robust undercutting across the full compositional range studied. Integration with anisotropic Si etching using KOH or TMAH allows complete release of the Ge_1-x Sn _x layer from the substrate. Raman spectroscopy confirms a substantial release of the initial compressive strain following removal of Ge buffer layer, with the corresponding strain relaxation estimated to be ε ≈ 0.5-0.55%. Complementary AFM analysis reveals the disappearance of cross-hatch morphology after complete suspension while maintaining a low root-mean-square surface roughness around 1 nm, confirming that the release process preserves excellent surface quality. This scalable and lithography-compatible micromachining strategy provides a robust route for selective material removal in Ge_1-x Sn _x /Ge/Si heterostructures, enabling the fabrication of strain-free, fully suspended GeSn microstructures. Beyond strain engineering, this approach offers a versatile platform for probing the intrinsic properties of GeSn alloys without substrate-induced constraints. The demonstrated methodology therefore opens new opportunities for the integration of suspended group-IV materials in next-generation electronic, photonic, and quantum device architectures.