Nanotech: The Next Generation

September 15th, 2009

Views: 325

Posted by ChrisG at 4:08 pm

At this event last week, I had to leave early and so missed Andrew Maynard’s contribution to the closing debate on the difficulties of regulating the “next generation” of nanotechnology. The grand vision that has been promulgated for a long time by advocates of long-term revolutionary nanotech is for atomically precise control over matter, with the aim of producing systems which can engineer functional structures of various kinds on the nanoscale. Among the functions and capacities which might be possible (it’s said) is autonomy, ranging in scope from the ability of a system to modify itself in response to environmental changes to more refined forms of artificial intelligence and self-directed activity. Instances of the former already exist, in the form of nano and micro scale encapsulation of drugs, producing tiny drug-transporting “bombs” (possibly the most popular metaphor for this tech among researchers, though I’m open to correction) which can be conveyed to a particular site in the body and then triggered with e.g. infrared light.

Maynard has very helpfully written up his presentation from the day, which revolves around the observation that a the problem of how to regulate passive nanomaterials (like the sort of things that end up in toothpastes, facecreams and socks) has to date been “how materials engineered at a nanometer scale might behave differently to more conventional materials, and how this might affect their safe use.”

And indeed, this is at the heart of the continuing vicious regulatory circle in which nanotech finds itself, where regulators wait for more data from researchers and companies so they can frame laws, and researchers and companies wait for regulatory guidance before they develop new technologies and release data.

Looking ahead (in the same vein as this report from colleagues of his at PEN), Maynard suggests that the world which is “on the way” with the advent of active nanotechnologies would be a “post-chemical” one, because the regulatory issues which would surround such technologies would no longer have to do with a toxicological concern with substances, but a broader concern with functions. He develops this in terms of a distinction between intrinsic and extrinsic properties.

In asking “what is different?” it is useful to distinguish between the intrinsic and extrinsic properties of material that has been engineered at the nanoscale. In essence, to differentiate between what it is, and what it does. Again, this is something of a simplification, but is useful for getting a handle on what might be important here. Intrinsic properties can be seen as those that associated with the material itself, rather than how it is being used. For instance, chemical composition leads to intrinsic properties. Size and shape can also underpin some intrinsic properties.

[…] At this point, the extrinsic properties of the “stuff” that we build become critical—the functionality associated with a carefully engineered collection of chemicals and components (what it does) becomes more than just the sum of its parts.

[...] On the other hand, it does seem that increasingly sophisticated nanotechnologies are going to present a major challenge to regulations that are built on assessing and managing risk associated with what they are made of, rather than what they do.

This distinction between intrinsic and (essentially) emergent properties seems to me to be incorrect, so expressed, since these “extrinsic” properties of active nanotechnologies seem to be a subset of the relational properties of nanosystems and nanomaterials in general. It’s these relational properties that are the subject of concern, and which underlie Alfred Nordmann’s argument that nanotechnologies cannot be regulated in any meaningful sense, and which make temporal considerations such as lifecycle assessment crucial to understanding the ethical issues which surround nanotechnology – from exposure and safety right up to higher level considerations in e.g. environmental ethics.

The capacity of nanomaterials to interact with the environment in which they are employed, released or ultimately disposed of (whether this is the human body, or e.g. a soil or aquatic system) has been at the heart of research on nanoparticle safety for a long time. Studies on the possible effects of nanomaterial exposure on plant root growth and gene expression in some vertebrate species are good examples. The problem, as the philosopher Geoffrey Hunt has pointed out (in the first of his essays in this volume), is that the already “technologised” environment in which nanomaterials are being employed (one in which a variety of synthetic chemicals are already distributed) the synergistic effects of interactions between complex systems and nanomaterials with novel properties are beyond the ability of traditional approaches to toxicology to assess. Looking at the lifecycle of a nanomaterial, and trying to understand how it may be disaggregated from any matrix in which it is fixed, before being diffused within a given environmental context, may reveal a variety of concerns, as is indeed happening with respect to nanoscale metal oxides and nanotubes at the moment.

There is a natural tendency to assume that, if we can’t see the physical form and complexity of something, its form and complexity don’t matter. As a consequence, most substance-related regulations—irrespective of the country or region they apply to—focus on the intrinsic properties of materials—which usually means focusing on their chemical composition.

Maynard makes a good point here, one which put me in mind of Alfred Nordmann’s essay on the uncanniness of nanotechnology [PDF}. But the extreme “relationality” of nanotechnology, e.g. the capacity of even first generation passive nanomaterials for entering into complex and unforeseen interactions with all manner of living and non-living systems, means that the regulatory difficulties he envisages for the next generation of nanotech are already here. The added functionality of “active” nanosystems will simply intensify these issues, rather than adding a qualitatively new set of problems.

The problem really comes down to the impact of technological evolution upon natural evolution, and the dissonance between, on the one hand, the timescales at which natural selection incrementally creates relationships of “fitness” and, on the other, the speed with which technological innovation can proceed (recalling Virilio's idea of "transplantation"). Often, advocates of revolutionary nanotechnology note that “nature is a nanoengineer”, i.e. that nature works at the level of nanoscale systems. Consequently, they deduce, there is no problem in principle with advanced nanoscale engineering of autonomous systems, as nature already does it.

The Foresight Institute's Christine Peterson [PDF, subscription needed]:

A certain fraction of researchers in the physical sciences and engineering assume that, since molecular machine systems exist (and are powerful) in nature, the proposal that artificial ones could someday be built, and be even more powerful, is rather obvious.
p. 12

The NSTC’s infamous Building the World Atom by Atom brochure:

What could we humans do if we could assemble the basic ingredients of the material world with even a glint of nature’s virtuosity? What if we could build things the way nature does—atom by atom and molecule by molecule?

But such assumptions confuse two problems of control. There is, first of all, the question of whether it is physically possible to control matter at the nanoscale. Advocates of revolutionary nanotech are fond of quoting Richard Feynman’s seminal remarks from way back in 1959 in this context: “[t]he principles of physics, as far as I can see, do not speak against the possibility of manoeuvring things atom by atom”. But this aside, applying nanoscale science for technological ends raises another problem: how to understand the actual environment within which new artefacts will be made, will “live”, “die” and be disposed of. “Control” of this environment is impossible: it includes, for example, organisms which are products of evolutionary processes that have taken many millions of years to unfold, in parallel, achieving along the way a fragile yet mutual accommodation. The sudden introduction of new quasi-autonomous nanosystems, fully autonomous abiotic machines, or even passive nanomaterials into unprepared complex systems could cause serious harm.

The problem here is one of time (or rather, temporality) and hard-to-accommodate novelty, not one of intrinsic and “extrinsic” properties.