## Etching test for GeSn For RIE etching test, we will use $\ce{Cl2}$ as the etching gas for the initial test, with proper combination with $\ce{O2}$ and $\ce{N2}$. This will be carried out in the Sentech ICP-RIE, which allows $\ce{Cl2}$ as the etching gas. Many papers (e.g., [[Investigation of GeSn as Novel Group IV Semiconductor for Electronic Applications]], [[Fabrication of Germanium Tin Microstructures Through Inductively Coupled Plasma Dry Etching]], etc.), intended to fabricate nanowires or microdisks, are interested in the selective etching of $\ce{GeSn}$ over $\ce{Ge}$. ### Process gas and condition ### Evaluation parameters [^1] #### Etch rate The etch rate should be not too big (to make the process controllable) or too low (to ensure not consuming too much time). The etch rate is defined as the etched thickness per unit time, with a unit $\mathrm{Å}/\rm{min}$. #### Uniformity The etch rates should be uniform in ideal case. Uniformity is defined as the percentage variation of the etch rates across the wafer. #### Selectivity The process gas should be selective to the material that would like to be etched, and not affect others. #### Anisotropic Isotropic means the material is etched down in all directions, while anisotropic means the directional etching. This is more problematic for wet etching, while for RIE, anisotropic is mostly achieved from the mechanism, while profiles could be not desirable. #### Microloading Dense/isolated structures have different etch rate. Somethings this refers to RIE-lag (smaller etch rate in smaller features), although they have different mechanism although in many cases different to tell. #### Profiles The patterns obtained after etching may deviate from the ideal anisotropic profile, exhibiting distortions such as bowing, undercutting, tapered sidewalls, or more localized effects near the bottom, such as footing and notching. #### Critical dimensions and CD bias Critical dimension (CD) is the width/length/space of one or between two patterns. Critical dimension bias (CD bias) is the difference between the CD measured after the etching and CD measure after the photolithography. With this parameter know one could correction the actual etched device dimension, it is $CD_{bias}=CD_{etch}-CD_{photo}$ ### Test structure design To evaluate how reliably the etching process can transfer features of various sizes, a set of resolution test structures is introduced. These structures help assess CD control, microloading effects, and pattern fidelity across a range of critical dimensions. #### Linewidth resolution - **Linewidth sweep**: Isolated lines of varying widths are used to test CD control and profile stability. - **Pitch sweep**: Constant-width lines with decreasing spacing are used to assess microloading effects and RIE lag. - **Line/space pairs**: Dense periodic structures evaluate the resolution limit and pattern fidelity during etching. #### Etch anisotropy and aspect ratio dependence #### Minimum feature size and pattern fidelity [^1]: This part is reported in the thesis [[Characterization and optimization of high density plasma etching processes for advanced memories application]].