Studying Nanotechnology
Foresight Briefing #1
originally published in 1988
© Copyright 1988, The Foresight Institute.
All rights reserved.
Many students have asked what they should study to prepare for
careers in nanotechnology. Giving a decent answer requires
outlining the different fields of research that fall under the
nanotechnology umbrella and describing the background knowledge
required to work in them. It also seems wise to say something
about the different levels of knowledge and modes of learning
that are relevant to such a broad, interdisciplinary area. The
following is a personal view, based on what I have learned (and
wished I had learned), and on how learning in these areas seems
to work best.
|
| One can't master everything
relevant to so broad a field. |
|
Nanotechnology will mean complete control of the structure of
matter, building complex objects with molecular precision. It
doesn't exist yet, because we don't have molecular assemblers
yet. Work related to nanotechnology accordingly falls into two
broad areas: the study of nanotechnology itself (which must
remain theoretical, for the time being) and research on enabling
technologies leading toward assemblers and nanotechnology (which
can be theoretical in part, but which also has an experimental,
developmental component).
The theoretical study of nanotechnology involves exploratory
engineering work in any of several of areas. It includes basic
studies in nanomechanical engineering (the study of molecular
machines) and nanoelectrical engineering (the study of molecular
and atomically-precise nanometer scale electronic systems). It
also includes studies of complex systems, such as assemblers,
replicators, and nanocomputers. More broadly, it includes studies
of non-nanoscale applications, such as large systems built by
teams of assemblers.
Because we lack the tools to do real nanotechnology today,
these theoretical studies amount to building castles in the air.
Accordingly, there is little funding for such efforts and
frequent skepticism about their value. Nonetheless, such studies
can be pursued with intellectual discipline, yielding firm
results and a better understanding of our choices as a society.
They have been my main focus and have spawned the current
interest in nanotechnology—including the interest in giving
these theoretical castles hardware foundations.
Studying what can be done with assemblers yields more
foresight than it does progress; working to develop assemblers
yields more progress than it does foresight. Inevitably, more
resources will go into development than into theory, because
technology development will yield practical, short- term results
on the way to long-term objectives. It makes no practical sense
to try to build an assembler today, but it does make sense to
build tools today that will make it easier to build assemblers
tomorrow. These tools are termed "enabling
technologies."
Promising enabling technologies fall into several familiar
categories. These include:
- protein engineering (involving efforts to develop
techniques for designing molecular devices made of
protein),
- general macromolecular engineering (involving efforts to
develop techniques for designing and synthesizing
molecular devices made of more tractable materials)
- micromanipulation techniques (involving efforts to extend
the technology of scanning tunneling and atomic force
microscopy to chemical synthesis, and then to the
construction of molecular devices).
These approaches have differing strengths and weaknesses.
Protein engineering can draw on a host of examples and prototypes
from nature, and can exploit existing self-replicating machines
(bacteria) to make products cheaply—a major consideration,
where short-term payoffs are concerned. General macromolecular
engineering avoids the major problem with protein engineering
(proteins, not having been designed for designability, are hard
to design), but at the cost of moving away from natural
prototypes and requiring more expensive chemical synthesis
techniques for making near-term products (thus reducing the
potential market). Micromanipulation techniques promise to ease
design problems by allowing direct construction of molecular
objects, but they suffer from higher costs: a chemical reaction
typically makes many trillions of molecules at once, while a
manipulator would make but one, hence manipulator-made products
can be expected to cost trillions of times more, dramatically
reducing the potential market. Also, as of this writing [1988],
micromanipulation has not achieved even a single
chemically-specific step in molecular synthesis, while chemists
have built specific molecules containing thousands of atoms.
All the above areas bear watching, and all will be pursued to
some extent, regardless of which ultimately proves to have the
biggest payoff. Hybrid approaches, combining techniques from
several of these areas (e.g., micromanipulation of molecular
tools), seem promising. Finally, improved computational modeling
of molecular systems is a generic enabling technology, relevant
to all these approaches.
|
| Nanotechnology is fundamentally a
branch of engineering. |
|
There are, as yet [1988], no college curricula aimed at
preparing students for work in nanotechnology. My own course at
Stanford provided at best an overview of the field. Rather than
seeking courses (and books, and journals) in nanotechnology, one
should seek courses (etc.) in the broad field of molecular
science and technology.
Unfortunately, there are, as yet, few (if any) schools that
treat molecular science and technology as a unified field. (A
note to curriculum reformers: developing a program having this
focus makes sense in terms of current science and technology, and
would provide a natural home for early studies in
nanotechnology.) Students aiming to gain a solid background in
areas important to nanotechnology should be prepared to shop
around from department to department. The following section lists
some of the important topics and some of the departments in which
they are frequently taught.
To study science and technology in a serious way, one must
have an adequate background in mathematics. Basic calculus is
essential, and differential equations and linear algebra are
widely used. Problems in nanotechnology vary widely in the
mathematical sophistication required for their solution.
The study of physical systems is founded on physics. A
knowledge of basic classical mechanics and electromagnetism is
essential, as is a knowledge of at least the rudiments of quantum
mechanics. Anyone aiming to do any sort of sophisticated work in
chemistry and molecular machines can benefit from deeper
knowledge of quantum mechanics; anyone interested in molecular
electronics should make quantum mechanics a chief focus of study.
"Quantum mechanics" is a broad area, however. The
quantum mechanics of interest here is not quantum
electrodynamics, quantum chromodynamics, or superstring theory,
but the garden-variety quantum mechanics of electrons in matter,
the sort studied by chemists and solid-state physicists. Both
quantum chemistry and solid state physics are topics of great
relevance to nanotechnology.
As with mathematics, so with physics: problems in
nanotechnology vary widely in the sophistication needed for their
solution.
Nanomachines and nanoelectronic devices are often greatly
influenced by thermal noise. To understand its effects, one needs
knowledge of thermodynamics and of statistical mechanics.
Thermodynamics deals with the flow of energy and heat in matter
in bulk; its principles constrain all physical systems and its
subject matter is regarded as a prerequisite for the study of
statistical mechanics, which describes much the same territory in
a more detailed, molecular fashion. These topics are often taught
in chemistry and physics departments.
Nanotechnology can be viewed as an outgrowth of chemistry, the
leading science in the field of molecular devices and molecular
manipulation. Anyone planning serious work in nanotechnology
should seek at least a basic background in chemistry, focusing on
its structural, molecular aspects. Those interested in assemblers
and molecular mechanical devices should study organic chemistry,
and those interested in the chemical-synthesis path to
nanotechnology should study synthetic organic chemistry, and
learn the arts of the chemistry lab.
Many specific fields have special relevance. Chemical kinetics
and reaction transition-state theory is of special relevance to
assembler theory. Molecular mechanics is fundamental to any sort
of molecular machine design. Studies in materials science (often
considered closely allied to chemistry) are also of value;
materials scientists consider the mechanical behavior of larger
systems of bonded atoms than chemists typically contemplate.
Biology is the leading science in the study of existing
molecular machines. Here, biochemistry is central: enzyme
reaction mechanisms provide examples of what many nanomachines
will need to do; the folding of proteins and the self-assembly of
protein systems provide examples of how complex first generation
molecular machines may be made. Familiarity with these fields is
of considerable importance to anyone interested in enabling
technologies.
Although nanotechnologists will need a thorough grounding in
relevant scientific principles, nanotechnology is fundamentally a
branch of engineering. To work as an engineer, one must learn to
think as an engineer, and that means studying (and doing) design.
Nanosystems will be systems, and so the principles of systems
engineering apply. Many nanosystems will be mechanical, and so
the principles of mechanical engineering apply. Studies in solid
mechanics, system dynamics, mechanisms, and control theory all
are relevant to both nanotechnology and enabling technologies.
Engineering departments often teach more specialized topics of
relevance to nanotechnology, such as VLSI circuit design
(relevant to nanocomputer design) and microfabrication (relevant
to possible enabling technologies). The principles of
conventional electronic circuit design are applicable to
moderately large nanoelectronic systems, and the principles of
quantum electronics are applicable to the smallest systems.
Software systems will be vital to nanotechnology and to
enabling technologies along the way. A basic introduction to
computers and software will be of value to anyone in any area of
science or technology. Those interested in software related to
nanotechnology should pay special attention to numerical
simulation methods for molecular mechanical and quantum
electronic systems, and to the design of programs for highly
parallel computer systems, since this is the direction hardware
will be moving in the coming years. Parallel systems will help
designers develop nanotechnology, and nanocomputers will later be
used to build massively parallel (trillion processor and up)
computer systems. Finally, if powerful systems are to be useful
in molecular design, they will need to be accessible through
fast, clean, intuitive interfaces that let designers see and
manipulate model molecules.
|
| Learn the fundamentals of molecular
science and technology. |
|
"In short, to do good work in nanotechnology, one must
master everything relevant to the physics, chemistry, and
engineering of molecules, from quantum mechanics to advanced
software architectures." Fortunately, this isn't true. Of
course, the more you know, the better you'll do (within
limits—studying mustn't completely displace doing), but one
can't master everything relevant to so broad a field.
What one can and should do is try to master some areas and
know a lot about the others. Real molecular devices can do many
different things: they can vibrate, pull apart, shake apart,
deform, transform, photolyse, or pop from state to state—any
of these behaviors can occur in a simple mechanical part, and any
can make it fail. Real physical systems will do something
when used, and if what they will do is strikingly
different from what you think they will do, then the
work you're doing may be a waste of time for you and for anyone
who listens to you. It's much better to be right about what will
work, and this means knowing enough to steer clear of potential
problems.
It makes sense to think in terms of three levels of knowledge
about a field:
- Knowing what a field is about—knowing what sorts of
physical systems and phenomena it deals with, and what
sorts of questions it asks and answers.
- Knowing the content of a field in a qualitative
sense—having a good feel for what sorts of phenomena
can be important in what circumstances, and knowing when
you need answers from work in that field.
- Knowing how to get those answers yourself, based on
personal mastery of enough of the field's subject matter.
If one has enough knowledge at levels (1) and (2) in enough
fields, then one can steer clear of problems in those fields
while doing work in a related field where you have knowledge at
level (3). And this is a good thing, because knowledge at levels
(1 ) and (2) takes far less time to acquire. But to make proper
use of knowledge at levels (1) and (2) requires a harsh
discipline: attempt to assume the worst about what you don't
know. Don't assume that a poorly-understood physical effect will
somehow save your design; do assume (until finding otherwise)
that it may utterly ruin it. Without this discipline, you'll
become an intellectual hazard. With it, you'll be able to make a
real contribution.
|
| Ideas about real systems must
somehow be disciplined by reality. |
|
How can one get this sort of general knowledge of a field?
Courses can help, but they tend to focus on mastery of a narrow
range of knowledge, rather than familiarity with a wide range of
knowledge. One can gain this familiarity by reading magazines and
journals that offer broad coverage of science and technology:
good choices include Science, Nature, Science
News, Scientific American, and IEEE
Spectrum. Another good tactic is to skim a wide range of
books on the new books shelf of a science library, on a regular
basis, and to do likewise with a wide range of technical
journals.
To do all this properly requires the discipline to read what
you don't understand—despite the school-induced reflex which
says "Oh, no! I don't understand, so I'll fail the
test—maybe I should drop this subject!" By reading what
you don't understand, you gain a sense of the patterns of the
field—the terms and abstract relationships, the kinds of
problems being addressed, and the kinds of knowledge required to
understand more. And this adds up to an important sort of
understanding. Later, this familiarity makes it much easier to
consult the literature: one knows which disciplines deal with
what problems, and what one needs to study to gain a deeper
understanding. Also, it fills your mind with questions, so that
you can later recognize the answers and have your mind seize them
more firmly.
For a thorough grounding in a basic field, classes can be
excellent. If classes aren't available, textbooks can often serve
well, especially if you work many of the problems.
In any evolving, interdisciplinary field, you must learn to
learn from books and journals. Learn to use libraries (as
horrible as they are, compared to tomorrow's hypertext publishing
systems). Learn to read skeptically—it is a rare book or
journal that doesn't have a few serious errors, and occasionally
publications are utter bilge, especially in interdisciplinary
fields (which too often lack any discipline at all).
Finally, tackle problems. If you can find a professor doing
good, interesting work, consider becoming an apprentice
researcher. If not (or in addition), pursue technical problems
that interest you. The best way to learn is to seek answers to
questions that interest you, and there is no other way to make an
original contribution.
Learn to criticize ideas, especially your own. Most new ideas
are wrong or inadequate. If you don't reject most of your ideas
promptly, then you're almost surely fooling yourself, and if you
also spread them, you're almost surely polluting the intellectual
world. But if an idea really seems to stand up under testing, try
filling in more details, and criticizing it again.
Get criticism from others. Learn to present ideas in
discussions, papers, and talks, and listen to the responses,
especially from people who know relevant fields. If they
disbelieve your idea and tell you why, either understand and
refute their criticism, or consider working on a different idea.
If they look at you oddly and change the subject, consider
whether you are perhaps overlooking a really big, basic
problem—are you really familiar with the relevant fields? If
they disbelieve you at first, but can be persuaded,
congratulations! You've probably got hold of something
interesting, perhaps even new and important.
Always remember that ideas about real systems must somehow be
disciplined by reality. Experimental work brings its own
discipline from nature, if the experimenter uses good technique.
This discipline is direct and hard to escape. Theoretical work,
in contrast, must be disciplined by knowledge of experimental
results and natural law; this discipline doesn't impose itself,
it must be sought out and largely self-applied. To be a careful
thinker, try to understand things in more than one way: if you
get the same answer from physical calculations and from
analogies to known machines and from analogies to
biology, then you're probably right. If all you have is a rough
analogy or a crude calculation, you may well be wrong.
Seek out weaknesses in ideas, and build only on ideas that
pass rigorous tests, or you may see the foundations of your
thinking later crumble and dump a year's (or a decade's) work
into the trash. Beware of those who have neither experimental
results nor a theoretician's voluntary discipline; expect them to
spout great streams of plausible nonsense, unconstrained by
reality. Don't become one of these, even if you find that many
(ignorant) people are intrigued and entertained by your wilder
imaginings.
In short, learn the fundamentals of molecular science and
technology. Survey other relevant knowledge. Learn to learn from
books and journals. Pursue problems, think critically, and learn
more. Design and calculate or experiment. Publish your
contribution and add to the world's knowledge. Good luck.
To supplement this Briefing on studying nanotechnology, Tanya
Jones has collected the following links to material available on
the Web (completed early 1998).
Textbook recommendation
The standard text in the field is Nanosystems: molecular
machinery, manufacturing and computation by K. Eric Drexler. Buy a copy and
study it.
Other pages on studying for nanotechnology
Educational Opportunities
Rice Univerisity
USC
Indiana University-Purdue
Related Topics
nanoelectrical
engineering
general
macromolecular engineering
quantum
chemistry
thermodynamics
organic
chemistry
synthetic organic
chemistry
chemical
kinetics
reaction
transition-state theory
molecular
mechanics
materials
science
VLSI circuit design
microfabrication
nanofabrication
numerical simulation methods
protein engineering
quantum
mechanics
Nature
Science
Science News
Scientific American
IEEE Spectrum
|