Answers are grounded in this paper and cite their sources.
Paper 5 of the Angelica Paper Series. The Angelica foundation grows GaN on Si(111) through an epitaxial Gd₂O₃ base. Gd₂O₃ dissolves in dilute hydrochloric acid, but GaN and Si do not. Using the Angelica engine, with nothing measured, this paper computes what it takes to use that base as a removable foundation: grow the GaN on silicon, dissolve the base, and bond the device film to diamond or another carrier. Base: Of 19 Si-matched candidates from the Alchemy map, all dissolve at about the same mean rate. Pure Gd₂O₃ wins on cleanliness; (Gd,Sm)₂O₃ is the backup. Release: Etch holes at a 51 µm pitch release the film in about one hour if the lateral etch rate is 10 nm/s. That rate is the first quantity to measure. Film and bonding: Without its oxide, the film curls, so it must stay on a handle throughout. Bonding temperature limits are given for each stack. Voltage: A field plate graded in steps lifts the 6 µm stack to 800 V. Gain: On diamond, the power for the same channel temperature rise is 2.0–3.6× that on Si, and substrate loss at 40 GHz vanishes. For 650 V power conversion, diamond must cost 6–24 USD/cm² to compete. Every result is tagged Implementable, Testable or Concept, and six test vehicles cover what computation cannot. The record contains the paper in English, Traditional Chinese and Japanese, a primer in all three languages, and the full code and sources. Licence: text and figures CC BY 4.0; code Apache-2.0. Defensive publication: disclosed as prior art, free for anyone to use and not to be patented by others.
No takes yet. Share an insight, caveat, or question.
Tung Ning Liu (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: