Is the view worth the climb?

Richard has asked for ideas about what we could do if we had a matter compiler. There has been a range of suggestions but, to my industrial eyes, too high a fraction of “you could do this piece of science”. Jack points out the confusion between science and technology. To paraphrase a wittier quote “research is turning money into ideas and technology is turning ideas into money”.

Although there is a fair amount of activity that parades under the nomenclature of nanotechnology, I suspect much is driven by the availability of funding or the desire not to be left out! Industry does what it can identify the output of – and output that has value to society or individuals. We are surely aiming at having the output of the ideas factory accessible to society? Is there another route?

There are also a lot of questions about whether we want short or long term goals. If we are to realise the full potential of the goal we are setting ourselves this week, we need to identify both short and long term goals but – and to my mind this is the crux of the question – we need to start with is the final paragraph of Jack’s last post “What things – methods, products, outcomes, social needs – should we be imagining? And what questions should we have in our minds as we do so?” We can then prioritise them in terms of effort required and resultant benefit and decide where best to focus our efforts – over time. In other words we first need to ask what CAN we do and then what WILL we do?

David Bott

About these ads

7 Responses to “Is the view worth the climb?”

  1. 1 Phillip Huggan January 7, 2007 at 7:37 pm

    What interests me is a matter compiler whose product is deflationary to the cost of matter compilers; can output a cost-effective matter compiler component or feedstock “purification” procedure.

    Because a moiety is so small compared to an SPM, there aren’t any marketable products in the meantime.

    Is the question asked here: what can we market (new surface patterning techniques or perhaps new tools) with a partial matter compiler prowess? Or is the question: what can we build if we have a sort of mature matter compiler proficiency? The answer to the former question will be different for every substrate/lattice. In the future, N.Seeman’s DNA manipulations may be very important for biosciences (I’m guessing) but will be useless in advancing aerospace technologies.

  2. 2 Philip Moriarty January 7, 2007 at 11:32 pm

    “There has been a range of suggestions but, to my industrial eyes, too high a fraction of “you could do this piece of science”.”

    David, are you suggesting that pure science without technological application/ commercial exploitation is of no benefit to society? In my opinion, if the output of the sandpit at the end of the week is a scheme enabling us to “do this piece of [exciting, pioneering] science” I’d be delighted. I realise that this concept of “application-less” pure science is not far from outright heresy in the context of the Treasury’s view of the future of research funding (see this link ). Recent suggestions that all research proposals should be subject to assessment on the basis of potential economic impact (see, for example, RCUK’s recent consultation document on the future of peer review) concern me greatly. Not only will this approach stifle adventure, risk-taking, and innovation, but the idea that only science which has a measurable economic impact is worth funding worryingly puts me in mind of Charles Clarke’s “education for education’s sake is a bit dodgy” maxim of a few years ago.

    From one perspective the question of “What is science for?” is of course much bigger than the Software Compilation of Matter theme and one might argue that we’re perhaps moving outside the “remit” of the sandpit. On the other hand, the issue of basic vs applied research very much underlies the key question Jack poses: “What things – methods, products, outcomes, social needs – should we be imagining?”


  3. 3 Brian Wang January 8, 2007 at 1:07 am

    There are plenty of products and markets. However, which specific one you will be going after depends on timing (when you can deliver some kind of matter compiler) and specific capabilities.

    Here is the stuff happening now or is well on the way:

    Nanoscale, nanoparticle tech already has plenty of markets and applications.

    Nanograined metals are up to ten times stronger. They are still mostly in the lab but could revolutionize cars, vehicles and golf clubs.

    100 nm Nanospheres will begin human trials against cancer.

    Existing synthetic biology is at these three links

    Those components have about 900 base pairs (BP). A ribosome has 2.3 million base pairs. Just improving gene synthesis (specialized matter compiling) has big potential markets. (Cellulose ethanol)

    Nanopore sequencing of genes could lead to $1000 genome sequencing.
    $10 to sequence the 1% that is variable between people.
    Use that for personal medicine. Instead of 90% have good response with drug X. it is here the stuff to take that will work for you. Here is the stuff to get take to prevent what we know were hidden problems. Here is the stuff we are going to watch out for in your situation.

    do things the MEMS market ($5 billion) is doing but better.

    You should always be able to make better computers.

    There will always be a need and big market in better sensors. For health or homeland security.

    Can the matter compiler make better material in any kind of quantity?
    A nearer term target is to make some small quantities of precise things. So some kind of new enzyme/catalyst that makes something valuable in chemistry better.

    However: specifics go back to first figure out how to do some aspect of matter compiling a lot better than competing/available methods (which means doing that survey of what is competing/available first). If you are not doing anything better then you are looking for missed opportunities with existing tech (which you do not need a Sandpit on software compilation of matter).

  4. 4 Brian Wang January 8, 2007 at 1:34 am

    If you only look at incremental and “realistic” things, you will not only have a tough time making a difference but a tough time differentiating. Looking at small molecule targets for health care applications. Welcome to the world of pharmaceuticals. You have a bunch of multi-billion dollar capitalized competitors. How about the next nanoparticle ? The next NEMS application etc….? If it is a market in the 4-5 year time frame that can be accomplished with a few million dollars. You have to be reasonable sure that your hot new idea is not one of thousands getting Venture capital funding.

  5. 5 Chris Phoenix January 8, 2007 at 7:11 am

    To whatever extent you think about social and policy implications, such thinking will be tragically incomplete unless you consider the consequences not only of your work, but of the capabilities and technologies that your work could enable. A pathway leads from atom-precise 3D manufacturing to a flood of advanced products with immense consequences–and that pathway may be unexpectedly short and may start from surprisingly primitive technology. Phillip Huggan is absolutely right in asking about matter compilers that build components of themselves.

    The scientific and technical community is probably the only place where this issue can be adequately evaluated. I hope that some of you will be willing to take the time to grapple with this post–everything I have written up till now is subordinate to this argument.

    Most of you probably see a disconnect between your work this week and the ultimate goal of a “nanofactory” that can make a diamond-strong spacecraft out of hydrocarbon feedstock in an hour. From a technical point of view, you’re right. Your work would still be interesting and valuable even if such a nanofactory were impossible–even if the most you could ever hope to build was 3D plasmon-channeling diamond nanostructures at £1,000,000,000 per gram. And I honestly don’t know whether aiming the Ideas Factory at a nanofactory or a nanoscale plasmon channel would generate more creativity and productivity in the present time and context.

    But the gap between the first nanogram and the millionth nanofactory may be crossed by basic engineering, not cutting-edge science. The broad outlines of that engineering may already exist. The gap may be crossed more quickly than someone who hasn’t studied it would believe. And on the other side of that gap lie consequences that the world cannot afford to ignore–that we must begin to study and comprehend as soon as possible.

    Even if a nanofactory is not interesting to you today as a technical goal, it should be of interest to the world as a consequence of atom-precise manufacturing. Here are some of the calculated technical consequences of atom-precise nanoscale covalent-solid machines:
    * Micro-structured materials with near-theoretical (defect-free) strength.
    * Motors with a power density a million times greater than today’s.
    * World-class supercomputers occupying a cubic millimeter and using 1 watt.
    * Molecular sorting and processing systems with error rates of 10^-15.
    * Sliding and rotating bearings with zero wear and near-zero friction (superlubricity).
    * Scalable, programmable manufacturing systems that can fabricate atom-precise products of their own mass and complexity in about an hour.

    A system that can build nanogram atom-precise structures (perhaps picogram or even femtogram) may enable the rapid development and integration of designs with these capabilities. Preliminary architectures exist for fabrication systems based on these capabilities. How long will it take before the ideas are embodied?

    It appears that the integrated self-contained self-building molecular manufacturing approach can scale up to kilogram-scale or even ton-scale factories. If whatever technology you expect to develop will be able to build as wide a range of products, as inexpensively, and as rapidly, then molecular manufacturing nanofactories may never be interesting as a technical goal. But that would not change the fact today’s molecular manufacturing self-building factory calculations represent the current best estimate of a lower bound of product performance and fabrication speed, and an upper bound of manufacturing cost.

    To restate: You work will be advancing the time when a newly created blueprint could be converted into a rapid-prototyped product in an hour, with zero labor and near-zero capital cost. This could speed the product development cycle by an order of magnitude or more. As soon as the design was completed, a thousand or a million factories could build the exact same product in parallel, close to the point of use. The product would be orders of magnitude more lightweight and powerful than anything we could build today.

    Just to take one example, aerospace R&D would be revolutionized if a test vehicle could be built overnight, fully assembled, directly from raw materials, and 95-99% lighter. Failure would become affordable again, and tests could be run far more quickly and with far less planning. Lighter airframes and avionics would make even earth orbit accessible with a few hundred pounds of machine-built hardware. Sounds like science fiction, but there’s no known physical reason why it’s impractical. If it is possible, it will happen–and your work will bring it a lot closer.

    If this is at all plausible, it is vital to know how soon it might happen. I don’t have an answer to that–but I will note that a general-purpose manufacturing system may involve just a few dozen straightforward mechanical designs, a few dozen parameter sets for reactions, a computer (which may be mechanical), and some software. Most of the system may be built with mechanical and molecular structures that can be analyzed classically. And given a palette of nanoscale designs to re-use and re-combine, most product designers will not have to worry about the unfamiliar physics of the nanoscale. Manufacturing system and product designers may be on familiar mechanical ground, but aided by the superior uniformity of covalent construction.

    So what kind of arms race would this kind of capability generate? Could it possibly be stable? Could lives be secure with this technology in civilian hands? Could the technology be kept out of civilian hands without massive oppression or at least massive loss of opportunity for good? How would economies and global trade cope with a shift in manufacturing methods and raw material sources? How can the demands of intellectual property be balanced with the benefits of widely accessible invention? (Europe’s resistance to software patents may be highly relevant here.)

    It is not your responsibility to answer these questions, or to stop working until they are answered. And it is not your responsibility to verify that the claims above are true. But if it is even plausible that the claims may be true… then it may be your responsibility to raise the questions to those who can study them.


  6. 6 davidbott January 8, 2007 at 8:41 am


    Since I chose the dark path of industrial science almost 30 years ago, I am probably a lost soul. :-) I have to admit that I am bemused by the inference that all industrial scientists are grey, risk averse, uncreative drones. It does not fit with my experiences, so maybe I need to introduce you to some I know.

    To answer you specific question, I do not believe there is such a thing as pure science – merely science we haven’t worked out the apllication of yet! The reason I am pushing the link to application is partly contained in Chris’s response. Without a vision of where you are going, most people don’t get challenged. To bring together a disparate group of academic scientists and, in a week, come up with a small group of common causes seem to me to need a rallying vision. I believe that addressing societies increasingly acute needs in areas like energy, environmental impact and healthcare are valid goals. What we could usefully do this week is work out how what we use the £1.5m investment to start the process of addressing these needs. We will, necessarily, have to use “cutting edge science”.

    If pure science did exist, I would put it alongside great art and music – something which enriches society but doesn’t practically help it. If you can afford it, it is a nice to have – if not…..


  7. 7 brian wang January 8, 2007 at 4:59 pm

    I do not think industrial scientists are uncreative drones. That is why if the parameters of the effort are competing with those tens of thousands people who are focused upon the next improvement in various areas then the twenty people at the meeting will have a tough time doing something beyond those regular business or business research efforts.

    David: Perhaps you can start by suggesting what you think are worthwhile and achievable goals. What you think is already being developed elsewhere and what this group should try to target that meets the software programmable matter but with an application.

    Designer molecularly precise materials are being made now:
    M5 fiber
    Carbon nanotubes

    DNA synthesis

    There is work now to put nano and MEMS devices into the bloodstream. I mention these things below, since most people have the knee jerk reaction that nanobots in the bloodstream is fantastical/sci-fiction/impossible. The work below shows
    that a lot of it is already happening and will happen over the next 5 years. It means that the targets/applications that the programmable matter compiler needs to be beyond current/near term work. Although medical related things into a body have long and costly approval and testing cycles.

    A group of NASA-funded bioengineers at the Universities of Pennsylvania and Minnesota have created double-walled artificial cells, called polymersomes, that can potentially float through the bloodstream loaded with cargo: cancer-zapping drugs, imaging agents–and, yes, extra oxygen. Durable, chemically controllable and biocompatible, polymersomes can be dried and rehydrated, which makes them appealing as an artificial blood. Unlike Freitas’s O2-pumping nanobots, an oxygen boost from a polymersome would be a one-shot deal–one that would allow you to leave your diving equipment at home.

    Devices are being implanted into the body and connected to the brain

    MAGNETIC TORPEDOES Researchers such as Metin Sitti at the Carnegie Mellon NanoRobotics Lab envision nanobots powered and controlled by magnetic fields generated outside the body in machines akin to today’s MRI equipment. Propulsion might be generated by a mechanical flagellum. Other concepts call for screw propellers or flapping fins.

    BATTERY-POWERED CRAWLERS While many nanobots will swim through the bloodstream, cerebrospinal fluid or other liquid environments, other bots may perform biopsies, deliver drugs and conduct imaging in the intestines. This Gutbot, envisioned by Carnegie Mellon’s Sitti, creeps along on grippy feet modeled after a beetle’s. Somewhat larger than artery-borne bots, it can carry a battery.

    BACTERIA IN HARNESS While researchers are just beginning to devise methods for constructing minute propellers, nature has been working on the problem for eons. Some scientists propose attaching large numbers of E. coli bacteria, which move using whiplike flagella, to nanobots. Movement would be controlled by focusing a laser beam on the organisms, which are sensitive to light.

    STEALTH POLYMERS Not all the new bots will look like machines. For instance, the University of Michigan’s James Baker developed a dendrimer (a complex polymer) to fight cancer. In rat trials, the dendrimer used folic acid to trick cancer cells into letting it in. The dendrimers also carried a potent chemotherapy drug, which then poisoned the cell before being flushed out of the body. Such targeted chemotherapy may prove both safer and more effective than current treatments.

    The big societal impact application zones:
    Human health
    Life extension
    Solve Energy problems (solar solution probably, non-nano is mass produced nuclear that is safer)
    Solve space access cost problem
    Revolutionize production
    Revolutionize materials

    Progress is being made already. So target beyond the current work.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

%d bloggers like this: