Day 5 - Atomic Fabrication with STEM (guest lecture) - Roccapriore

Day 5 - Atomic Fabrication with STEM (guest lecture) - Roccapriore

🎙 Kevin Roccapriore 👥 1K 📅 July 18, 2026 ⏱ 52 min 👁 21 📄 expert opinion 🧭 2026-08-16
Available in: English (current) Français

Keywords

STEMatomic fabricationdeterministicmachine learningquantum

Summary

Kevin Roccapriore, founder and CEO of AtomQ, presents a lecture on deterministic atomic manufacturing using Scanning Transmission Electron Microscopy (STEM). He begins by outlining motivations in quantum computing and semiconductor industries, emphasizing the need for scalable atomic-level fabrication. He reviews historical developments in atomic manipulation, from early observations to recent demonstrations in graphene and TMDs. The core of the talk focuses on transitioning from manual manipulation to automated, data-driven approaches, using machine learning to identify atoms and plan beam placements. He discusses challenges such as beam-induced distortions, the need for precise beam positioning (<20 picometers), and the difficulty of maintaining structures outside the microscope. He highlights his work on custom scan controllers and the use of feedback to improve precision. The talk concludes with a vision for commercializing this technology, acknowledging remaining hurdles but expressing optimism about its potential.

139 words

Critical Evaluation

Value of the Information & Strength of the Argument

The talk provides valuable insights into the state-of-the-art and challenges of atomic fabrication using STEM. The speaker’s firsthand experience lends credibility, and he presents concrete examples of automated manipulation and the importance of data-driven approaches. The argumentation is logical, progressing from motivations to technical details to future prospects. However, some claims are made without rigorous evidence, and the talk is more of an expert opinion than a systematic review. The discussion of machine learning applications is particularly valuable, highlighting the potential of small data and domain-specific models.

Scientific Rigor, Source Quality, Title Accuracy

The talk is based on the speaker’s own research and experience, but specific sources are not explicitly cited. The description mentions no external links. The title accurately reflects the content. The speaker references work by other groups (e.g., Oak Ridge, Vienna) but does not provide specific citations. The lack of formal references reduces the scientific rigor, but the speaker’s expertise and the technical detail compensate somewhat. The title is appropriate and not misleading.

175 words

Title / Content Match

The title accurately reflects the content: a guest lecture on atomic fabrication using STEM, delivered by Kevin Roccapriore.

Quality & Reliability

7/10

The speaker is a domain expert (founder/CEO of AtomQ, former staff at Oak Ridge National Laboratory) and presents technical details of atomic manipulation using STEM, including specific examples and challenges. However, the talk is largely a personal account and vision, with limited formal citations or peer-reviewed references, and some claims are presented without detailed evidence.

Key Moments

Contribution & Novelties

The talk provides a unique perspective on the commercialization of atomic fabrication, emphasizing the transition from manual to automated, data-driven methods. It highlights the importance of precise beam positioning and the use of machine learning for real-time decision-making. The speaker’s experience in founding a startup based on this technology adds practical insights not commonly found in academic presentations.

Pour aller plus loin :

94 words

Radar Profile

The radar profile shows high scores in technical level and information quality, reflecting the expert-level content and detailed explanations. The lower score in information quantity suggests the talk is focused rather than broad. Overall, the talk is technically strong but could benefit from more explicit sourcing.

Reliability 7/10