Nano & Nature

Time Line - when is all this going to happen?

Estimates run from 15 to 50 years depending on who is talking and what they are predicting. I worked with Dr. Drexler for almost a year designing and animatating “Productive Nanosystems” to show a theoretical design for his nanofactory. The nanofactory is a desktop device that will manufacture any physical object that can be described in software for the machine. It uses a chemical feedstock and produces drinking water as a byproduct of the manufacturing process. Dr. Drexler is using the 5 minute animation to educate the public and to communicate important concepts.

You can see the animation hosted as streaming video at http://www.kurzweilai.net/news/frame.html?main=news_single.html?id%3D4809 thanks to the people at kurzweilai.net

Once we have the technology to build this nanofactory, we can build almost anything. Some of the early benefits are tiny medical sensors and probes that can be injected to kill cancer cells directly. http://www.actionart3d.com/im_nano/CancerKillerCpyright.jpg you can see one design for such a probe. It is rather large and probably travels around by means of the blood supply. Once it is carried into a known tumor, it would begin killing cancer cells by rupturing their cell walls. It also allows a doctor to monitor blood chemicals and general health of the patient. Although it looks somewhat forbidding, the long body allows it to attach itself in one location and move it's head about to accomplish a task. The head contains both DNA sensors and sharp edges suitable for tissue work. Probably a doctor would inject a tracer substance that binds to a cancer cell and then the chemical serves as a marker for the robot. Any cell surface covered with the tracer molecule would be attacked by the head of the robot.

Another, more simple nanobot can be seen at
http://www.actionart3d.com/im_nano/BuffBot_C_900x.jpg
It shuts off the blood supply inside a cancer tumor. Requires very little intelligence since a doctor signals it to inflate once it enters the tumor. It is the same size as a red blood cell so it circulates over and over through the body until it does pass through the tumor. A few thousand could really cripple a tumor.

There are several steps that must be accomplished to get to this nanofactory and it's production of wonderful medical tools. We have to build the tools that build the tools that build the nanofactory. Right now, we can't pick up and place the atoms and molecules that are needed to build the nanofactory. They are too small. So we need smaller tools.

I know one scientist (Dr. Robert A. Freitas Jr., Senior Research Fellow at the Institute for Molecular Manufacturing ) who is working on a chemical pathway to manufacture the pyramid tool tips that are shown in the animation. They will be used to pick up individual atoms and molecules, move them around and place them where the atoms are needed. He expects to have the pathway defined by 2006 and if the process proves to be useful, we might have a few tool tips by 2007 or 2008. That depends on the funds available to support Dr. Freitas and to finance his research in the lab. You can contribute to that fund at www.imm.org.

So let's assume we have useful tool tips by 2008. A lab then needs to attach a good tip to the needle of a scanning microscope and manually use it to pick and place atoms. After a lot of testing and modifications, that effort should produce a tiny laboratory that allows you to build a few simple things by hand. Believe me, you don't want to build much that way because it takes a LOT of atoms to fill in a volume of any size whatever. A block of atoms only 10x10x10 is a thousand cycles of put and place. I expect you can see the need to teach a computer how to run the equipment and to be able to tell it to build you a block of diamond with dimensions of such and such. Hopefully it will be able to build arbitrary shapes of carbon. Let's give them two years after they have the tool tips to have them under computer contol for this initial simple assembly. So, in 2010 we could expect to be assembling simple tools and the first pieces of the nanofactory. Or at least some small version that can produce tiny parts one at a time.

A really cool company, Nanorex Inc., run by Mark Sims, is creating a software tool to design such shapes. Mark made a challenge grant to this project and 21 other people made matching grants to make sure “Productive Nanosystems” was finished. I owe all these people for that support. With this tool, nanoEngineer, which is in alpha release in 2005, we should be able to design carbon tools atom by atom and export that information to a file to tell these computers how to build whatever we want. At first it will be just carbon, but other atoms will be added and new tool tips will be added to handle those additional atoms.

Once we can build simple tools by hand out of carbon, we can start combining those small parts to reach the goal of a small manufacturing facility.

My own idea is to put all this onto a self contained IC wafer with an on-chip vacuum system. The type of vacuum pump used for the final stages of a pump down are called ion pumps and they have no moving parts. Seems like we could reduce that technology until it fits into a square cm or so. We need six or more vacuum pumps to handle multiple doors in the airlock tunnel. The volume to be evacuated is very small since everything on the wafer is small. I'm working on an animation to show how the input stages of such a wafer would work. Silicon micromachines move the products from an input port into a vacuum tunnel and then into the working section of the wafer. There, a true, computer controlled factory takes shape.

First you have to build the tool set that can build any arbitrary shape found in the nanofactory. Then you have to make enough of them to build a small nanofactory - say only 30 x 30 microns of active surface. That will be a huge machine to build by hand. The smallest parts are only 10 nm across ( 1% of a micron). So you need a lot of them to build one machine which is still so small you can't see it. Maybe you could see it as a tiny dot. In good light.

So, when do we get this small machine working. Give them 5 years to build it.

We are at 2015 and we have the first microscopic machine which can build subsystems of itself - and therefore can build itself. We take these systems by hand and start them building more of themselves. Still a manual job to set each one up. But after a while you have a whole lab full of IC wafers which are building tiny nanofactories.

Now you need to bring all these together into one big unit. Now we are getting close to a real nanofactory that can build a wrist watch or a hand computer. Once you assemble all these little ones into one big one, you can build self contained big ones. So a month after you build the first full size nanofactory, you can have as many of them as you want.

And a month after that, you can have a nanofactory big enough to extrude a car. Sometime around 2018. Say, between 2015 to 2020. Course a lot of people must have the resources to actually work on these bits and pieces. Probably needs government grants, but private money would help too.

---