However, even though such devices or the features on such devices may be microscopic, they are not atomically-precise nor are they of the scale of single atoms.
But, self-assembly is limited in the structures that can be created by the need to design around the shape and charge requirements of the individual units.
However, none of these techniques can provide atomic accuracy while manufacturing devices with diverse functions, out of a wide range of materials.
However, while these terms are used for clarity, defining one structure as the tip and another as the workpiece can be arbitrary in certain circumstances.
While atomic microscopy equipment is exceptionally accurate, no equipment is perfect.
Such simulations, for any large number of atoms, require substantial computer processing power.
Even on powerful computers, simulating large numbers of atoms at high levels of detail can be extremely computationally-demanding.
However, an entire mechanosynthetic system need not be simulated at a high level of detail.
It should be noted that in some cases, primarily with respect to hydrogen due to its low atomic mass, tunneling can contribute to reaction error.
However, thermal energy is nonspecific and facilitates desired and undesired reactions alike.
However, the experimental examples are generally limited to modifying surfaces rather than building complex or three-dimensional structures, lack separation of feedstock, presentation surface and workpiece (that is, the presentation surface often serves as all three), teach only a small, non-generalizable set of tools and reactions, and use atomically-imprecise tips with no bootstrap process to facilitate the transition to atomically-precise tips.
Obviously, this limits the versatility of the products that can be manufactured since it constrains the elements used in reactions and the workpieces to which they are applied.
However, due to insufficient simulation detail, lack of a bootstrap sequence, lack of a comprehensive set of reactions and tips, or other drawbacks, previous work has not been directed to a system that can be implemented using existing technology, capable of a large set of reactions that can be used to create complex atomically-precise structures.
However, this was not via purely mechanical means but rather used an electrically-pulsed STM tip.
However, none of the tools described previously, alone or in combination, could practically provide a bootstrap process, a set of tools exhibiting closure (that is, a set of tools that could build themselves), a versatile set of reactions, a set of reactions of known reliability, nor were they directed to a system for three-dimensional fabrication, among other drawbacks.
Use of the presentation surface as the feedstock depot, feedstock, and workpiece places limitations on what workpieces may be built, as workpieces are thus limited to being made out of the same element(s) as the presentation surface, among other drawbacks.
The prior art does not anticipate being able to extend atomically-precise mechanosynthetically-created structures into three dimensions.
The prior art is frequently limited to the removal of a single adatom (a surface atom), or the insertion of a single atom into a vacancy left by the removal of such an adatom, often using a single element and involving a very specific crystal structure.
Note that the experimental setup in this example does not demonstrate a robust set of reactions applicable to building complex structures.
To the best of our knowledge the prior art does not address this issue.
The mechanosynthesis prior art generally does not address the issue of designing for reaction reliability.
They did not attempt to design a system ahead of time with a known level of reliability.
Even where modeling is performed in the prior art, modeling of an atomically-imprecise tip is unlikely to accurately represent the actual experimental system due to lack of knowledge of the exact structure of the tip.
Prior art using large (compared to atoms) building blocks is not an appropriate parallel to positioning, and making and breaking bonds, at the atomic or molecular level.
And, the experimental mechanosynthetic reactions found in the prior art do not appear to have been engineered in advance for versatility or reliability using computational chemistry techniques.
Another drawback of the prior art is that the presentation surface also frequently serves as the feedstock depot, feedstock and workpiece, such as with the “vertical manipulation” prior art, of which Oyabu, Custance et al.
Without separating the presentation surface, feedstock and workpiece, the ability to create diverse structures can be limited.
And, the prior art contains no teachings as to how one might generalize the mechanosynthetic techniques presented to other elements and reactions, or to construct complex, three-dimensional workpieces.