2025 · Nanoscale Advances
Advances in core technologies for semiconductor manufacturing: applications and challenges of atomic layer etching, neutral beam etching and atomic layer deposition
Verdict
SCIENTIFIC
The study follows the scientific method, and the paradigm it assumes passes. Both layers must pass. The verdict is scientific.
1·Study methodology · inside the paradigm
SCIENTIFIC
Follows the scientific method within Classical Mechanics / Engineering.
2·Paradigm · Classical Mechanics / Engineering
SCIENTIFIC
Passes all three tests.
How the verdict is decided
The verdict grades both layers: the paradigm this study assumes, and how the study was carried out inside it. Both must be scientific for the whole thing to be scientific. A clean method inside an unscientific paradigm is unscientific. An unscientific method inside a better paradigm is unscientific. A split on either layer is mixed. The worse layer decides.
1·Study methodology
Did this study test its claim with methods that are independent, falsifiable, and non-circular?
SCIENTIFIC
This study claims that atomic layer etching, neutral beam etching, and atomic layer deposition processes achieve "atomic-scale precision", enabling fabrication of next-generation miniaturized semiconductor devices. All three tests pass.
Independently Verifiable
Pass
Chips fabricated with the processes this paper describes work as predicted in phones, computers, and other electronics, and a person who rejects the atomic model can still observe that the device functions.
Falsifiable
Pass
If ALE, NBE, and ALD did not achieve the precision this paper claims, the devices they produce would not function and would not be sold.
Non-Circular
Pass
The processes use gases, plasmas, and chemical reactions to remove or deposit material on wafers, and the result is validated by whether the device works, not by a model-dependent instrument confirming the process.
Why
This paper describes processes that use gases, plasmas, and chemical reactions to remove or deposit material on semiconductor wafers in controlled increments. What the model calls atomic layer etching alternates between adsorption and reaction steps, removing material one layer at a time. Neutral beam etching uses neutralized particles to avoid the charging and radiation damage that plasma methods cause. Atomic layer deposition builds material onto surfaces through similar self-limiting reactions. The paper traces how these processes enable fabrication of three-dimensional transistor structures, advanced memory, and other devices at the 5 nm node and below. Nobody has seen an atom. The theory uses empirical corrections throughout. But the output is a chip that switches, computes, and holds data. Chips made with these processes work as predicted in phones, computers, and other electronics. If the processes did not achieve the precision claimed, the devices would not function and would not ship. The validation is the working product, not the theory.
2·Paradigm · Classical Mechanics / Engineering
Does the framework this study assumes pass the three tests?
SCIENTIFIC
Independently Verifiable
Pass
The engineering framework produces physical devices whose function anyone can verify without accepting the model's claims about atoms or electrons.
Falsifiable
Pass
The framework depends on working outputs to survive, and a fabrication process that produces non-functional chips gets revised because the product must work to be sold.
Non-Circular
Pass
The theory was built on top of observations that were already empirical, and the processes this paper describes are validated by device performance, not by circular reference to the theory.
Why
The engineering framework started with things you could hold and observe. A transistor that switched when you applied voltage was the first observation. Theory was added on top of what already worked. This paper sits at the top of that accumulation. It describes processes that achieve "atomic-scale precision", but the precision is validated by device performance, not by direct observation of atoms. The field depends on working outputs to survive. A company cannot sell a chip that does not turn on. When a process produces a device that fails, the process gets revised. The theory does not absorb failures as missing heritability or novel variants. It revises. That is the difference between a framework built on empirical accumulation and one built on model-dependent interpretation. The ontology may be incomplete. The empirical corrections prove it is. But the outputs match predictions and improve every year, and that is what validates a theory in engineering.
From the study
ALE alternates between adsorption and reaction steps to remove the material one atomic layer at a time
NBE uses neutral beams to avoid charge build-up and photon-radiation damage typically caused by plasma etching, making it ideal for ultra-high-density integrated circuits
ALE operates through a series of self-limiting reactions, ensuring that each cycle removes only a single atomic layer of material. This minimizes the risk of over-etching and reduces surface roughness, which is essential for maintaining the integrity of fine features.