InSb - Dry Etching

InSb - Dry Etching

Material Name: InSb
Recipe No.: 8516
Primary Chemical Element in Material: In
Sample Type: Thin film
Uses: Etching
Etchant Name: None
Etching Method: Dry etching
Etchant (Electrolyte) Composition: For all of the investigated etch masks, the best resultswere obtained with an ICP power of 600W, an RF power of 150W, a chamber pressure of 15 mT, and a table temperature of 110 °C. These parameters provided a DC bias of-315 V.
Procedure (Condition): The most widely reported chlorine-free process suitable for dry etching of InSb was based on CH4/H2/Ar therefore the same gas mixture was chosen as starting point for process development. The tool used was the Plasmalab 100 Inductively Coupled Plasma (ICP) 180 etching tool from Oxford instruments. All samples etched were mounted on a 6 inch silicon carrier wafer by means of thermal grease, to ensure a good thermal contact. Irreversible temperature indicatorswere placed on the carrier tomonitor themaximumtemperature reached in the chamber due to the etching plasma. Samples prepared for etching all underwent the same cleaning procedure consisting of OptiClear, acetone, isopropyl alcohol soak and de-ionised (DI) water rinse prior to the definition of the mask. Ultrasonic baths were not used during the cleaning procedure in order to avoid inducing damage to the material. Tests were carried out using three types of masks: Silicon nitride (SiNx), deposited at room temperature in an ICP reactor; Hydrogen silsesquioxane (HSQ), a negative-tone resist patterned by electron beam lithography; and Shipley AZ4562 resist, a positive-tone photolithography resist. The outcome of each etch test was assessed in the first instance by a profilometer, then samples were cleaved and their cross-section examined by Scanning Electron Microscope (SEM).
Note: We present a new chlorine-free dry etching process which was used to successfully etch indium antimonide grown on gallium arsenide substrates while keeping the substrate temperature below 150 °C. By use of a reflowed photoresist mask a sidewall with 60 degree positive slope was achieved, whereas a nearly vertical one was obtained when hard masks were used. Long etch tests demonstrated the non-selectivity of the process by etching through the entire multi-layer epitaxial structure. Electrical and optical measurements on devices fabricated both by wet and dry etch techniques provided similar results, proving that the dry etch process does not cause damage to the material. This technique has a great potential to replace the standard wet etching techniques used for fabrication of indium antimonide devices with a non-damaging low temperature plasma process.
Reference: Vincenzo Pusino, Chengzhi Xie, Ata Khalid, Iain G. Thayne, David R.S. Cumming, Development of InSb dry etch for mid-IR applications, Microelectronic Engineering 153 (2016) 11–14.