The idea of Stevenson is supported by an old but still very interesting Ahern patent
"As an alternative to mechanical and metallurgical techniques for producing nanometric surface features, lithographic wet-etch techniques may be used. For example, referring to FIG. 4A, in a first lithographic process, a bare substrate 30 of a selected host lattice material, for example, nickel, is provided with a selected crystallographic orientation, for example, the 110 or 100 orientation. The 110 crystal faces are favored in the case of a nickel host lattice substrate because the 110 planes support the highest solubility of hydrogen isotopes of any crystallographic planes.
As shown in FIGS. 4B, 4C and 4D, photoresist 32 is spun on the substrate and exposed using a patterned
lithographic mask 34 having a selected pattern of sub-micron sized geometries. Preferably, the maximum pattern dimension, d, or "duty cycle" of repeated pattern is about 0.2 μm in length. Such nanoscale features require the use of thin, state of the art photoresists. The unexposed resist is then removed using standard techniques to produce a photoresist etch mask. As shown in FIGS. 4E and 4F, the underlying substrate is then anisotropically etched using an appropriate etch to produce grooves in the substrate surface having a depth, h, of less than about 1 μm. Grooves of a greater height are less preferable because they would allow the prismatic faces of grooves exceeding about 1 μm to reconstruct to a more harmonic, high atomic coordination state. After removing the resist etch mask using standard photoresist processing techniques, the substrate 30 is provided with a topology of steps 36 which all ideally exhibit sharp corners and straight walls."
This is an interesting patent because it shows how not to write a patent. Instead of looking at the process from an atom's point of view while keeping the conditions as general as possible, an imagined model is applied to limit the conditions and behavior. Many methods are available to create gaps in a material. The successful patent will focus on the result, not on the method. The NAE has to have several unique features. These features need to be part of the patent. These unique features are presently unknown. I have identified only one of them as being the size. Obviously, several other conditions are required. The challenge is learning the nature of these conditions.