Page 105 - Nanotechnology an introduction

by the initially uniform cells (see [110] for the modeling of neurogenesis). On a yet larger scale, it is likely that the construction of nests by social
insects such as ants and wasps relies on simple rules held and enacted by individual agents (the insects) according to local conditions; this
process has been called stigmergy and is evidently conceptually related to programmable self-assembly %modelled by graph grammars (Section
8.2.8).

Reproducibility is interpreted somewhat differently by living processes in comparison with the mass, standardized manufacturing processes of the
Industrial Revolution paradigm. Although the basic building blocks (e.g., proteins) of living organisms are identical, templated from a master
specification encoded in DNA (see Section 4.1.4), organisms are not identical in the way that very large-scale integrated circuits (VLSIs) are. What
is specified (genetically) is at most an algorithm (subject to local environmental influence) for constructing an organism, or maybe just an algorithm
for an algorithm. Reproducibility exists at the functional level: every termite's nest is able to protect its denizens from a wide range of hazards under
varied local conditions of topography, soil type and vegetation; every dog has a brain able to ensure its survival for a certain period; and so forth.
This concept of an algorithm specifying how the construction should take place is what is used for building the nests of social insects, which are
constructed stigmergically—each insect is armed with rules specifying what to do under a variety of local circumstances.
There are some hitherto relatively unexploited niches for creating nano-objects via biological growth. For example, the magnetic protein ferritin,
which is constituted from an iron oxide core surrounded by protein, can be made on a large scale by low-cost biotechnological manufacturing
routes and due to its strict monodispersity can be used, after eliminating the protein matter, in magnetic memory devices [163].

8.2.13. Self-Assembly as a Manufacturing Process

Practical, industrial interest in self-assembly is strongly driven by the increasing difficulty of reducing the feature sizes that can be fabricated by
semiconductor processing technology. Pending the ultimate introduction of productive nanosystems based on bottom-to-bottom fabrication
(Section 8.3), self-assembly is positioned as a rival to the “top–down” processes that currently constitute the majority of nanofacture. For certain
engineering problems, such as membranes for separating valuable resources (such as rare earth ions) from a matrix in which they are present in
very diluted form, self-assembly may already be useful to manufacture regular structures such as dots or stripes over an indefinitely large area;
passive self-assembly might be able to produce such structures with feature sizes at the molecular scale of a few nanometers.
As well as the mimicry of natural surfaces for biological molecular recognition for sensing and other nanomedical applications (Chapter 4), self-
assembly fabrication techniques should be generally applicable to create feature sizes in the nanometer range (i.e., 1–100 nm), which is still
relatively difficult to achieve, and certainly very costly (limiting its application to ultrahigh production numbers) using conventional top–down
semiconductor processing techniques. Furthermore, self-assembly is more flexible regarding the geometry of the substrata to which it can be
applied; e.g., there is no restriction to planar surfaces. This is particularly advantageous if it is (for example) desired to engineer the surfaces of
intricately curved objects such as vascular stents in order to increase their biocompatibility.

A current challenge of “bottom–up” nanotechnology is the formulation of design rules. Generally we ask: “how to design X to carry out a desired
function Y?” Design means, essentially, specifying structure, hence the question can be reworded as: “what structure will give function
Y?”—“function” being interpreted as properties and performance; and structure being constituted from certain numbers of different types of entities,
connected together in a certain way. Micro (and macro) engineering benefits from vast experience; i.e., a look-up table with structure in the left-
hand column and function in the right. There is less experience in the nanoworld, but if a nanostructure is simply a microstructure in miniature (often
it is not), this experience can be transferred. Undoubtedly one needs to ask whether the properties of matter change at the nanometer scale
(Chapter 2); i.e., do we need a new set of structure–property relations? These relations may also affect the fabrication process.
Regarding the formulation of assembly rules, again current experience is exiguous and rules come mainly from knowledge of certain biological
processes. Typically they are either very general (such as the Principle of Least Action, see Section 8.2.11), without any specific indication of how
to apply them to a particular case, or very specific and possibly quite esoteric (e.g., relying on chemical intuition). Challenges lie in the adaptation of
general principles to specific cases, and in the formalization and generalization of known specific heuristic rules (or intuition). Indeed, the main
disadvantage of bottom–up is that the process is not well understood theoretically. Hence although we need to be able to, at present we cannot
design the starting objects (precursors) to achieve a specified final device.
8.3. Bottom-to-Bottom Methods

Essentially, the contribution of nanotechnology to the effort of ever improving machining accuracy is simply to take it to the ultimate level, in the spirit
of “shaping the world atom-by-atom” (the subtitle of a report on nanotechnology prepared under the guidance of the US National Science and
Technology Council Committee on Technology in 1999). Rather like the chemist trying to synthesize an elaborate multifunctional molecule (e.g.,
adding conjugated olefins to provide color, and hydroxyl groups to provide Lewis acid/base interactions), the materials nanotechnologist aims to
juxtapose different atoms to achieve multifunctionality. This approach is known as mechanosynthetic chemistry or, in its large-scale industrial
realization, as molecular manufacturing. The essential difference from chemistry is the eutactic environment, in which every atom is placed in a
precise location.
The famous experiment of Schweizer and Eigler, in which they rearranged xenon atoms on a nickel surface to form the logo “IBM”[151],
represented a first step in this direction. Since then there has been intensive activity in the area, both computational and experimental, but it still
remains uncertain to what extent arbitrary combinations of atoms can be assembled disregarding chemical concepts such as affinity and potential
energy surfaces, and whether the process can ever be scaled up to provide macroscopic quantities of materials.
Also known as molecular manufacturing or mechanosynthesis or “pick and place” chemistry, bottom-to-bottom methods literally construct things
atom-by-atom. In other words it is chemistry with positional control; i.e., taking place in a eutactic environment. This is what is sometimes called
“hard” or “true” nanotechnology (in the sense of being uncompromisingly faithful to Feynman's original vision [56]).
A specific realization based on carbon (mainly diamondoid) structures has been elaborated by Drexler and others. This envisages an appropriately
functionalized molecular tool driven by mechanical forces (such as the tip of a scanning probe microscope), which abstracts hydrogen from
passivated surfaces to form radicals (“dangling bonds”), where other atoms can be added. Hence this approach is also called “tip-based
nanofabrication”.