Building machines from the atoms up

Building machines from the atoms up

Photo by Laura Ockel on Unsplash

Originally published 25 July 2000

The pop star physi­cist, Richard Feyn­man, had lots of clever ideas in his life­time, includ­ing a few that won him a Nobel Prize, but what he might be best remem­bered for 100 years from now is some­thing he tossed off almost in pass­ing back in 1959.

All arti­facts of human tech­nol­o­gy, from Stone Age chop­pers to auto­mo­biles, have been fash­ioned by tak­ing a lump of pre­ex­ist­ing amor­phous mat­ter and chip­ping, carv­ing, bend­ing, or bash­ing it into shape.

But there’s anoth­er way, sug­gest­ed Feyn­man, and that’s start­ing from the bot­tom and assem­bling arti­facts atom by atom. Back in 1959, it was hard to imag­ine how this might be done, but there is no phys­i­cal law that would pre­vent it, he said.

Nowa­days, Feyn­man’s idea is all the rage. It’s called nan­otech­nol­o­gy (from the nanome­ter, a unit of mea­sure­ment a bil­lionth of a meter long). A recent sto­ry in US News & World Report hailed nan­otech­nol­o­gy as the “next indus­tri­al revolution.”

The mag­a­zine quotes Mihail Roco, a senior advis­er for nan­otech­nol­o­gy at the Nation­al Sci­ence Foun­da­tion: “Because of nan­otech, we will see more change in our civ­i­liza­tion in the next 30 years than we did dur­ing all of the 20th century.”

That’s quite a claim, and we’ll have to wait and see if it hap­pens. Grant­ed, com­put­er cir­cuits con­tin­ue to shrink, and gears and motors have been devised that could fit on the head of a pin, but these have most­ly been cre­at­ed in the old-fash­ioned way, by etch­ing a block of amor­phous mat­ter — usu­al­ly silicon.

Which is not to say that atom-by-atom fab­ri­ca­tion is all talk. Sin­gle atoms can be arranged on a flat sur­face with a device called a scan­ning tun­nel­ing micro­scope. And three-dimen­sion­al struc­tures, such as com­put­er cir­cuits, can be built up by deposit­ing suc­ces­sive lay­ers of mol­e­cules that are only a sin­gle mol­e­cule thick.

So how long until we see the famous­ly pre­dict­ed “util­i­ty fog,” a mist of micro­scop­i­cal­ly-small mechan­i­cal robots (nanobots), fab­ri­cat­ed atom by atom, that can assem­ble them­selves in their teem­ing num­bers into any­thing from a com­fy chair to a com­put­er? How long until we inject nanobots into the blood­stream by hypo­der­mic to seek out can­cer cells and dis­as­sem­ble them atom by atom? How long until we send invis­i­ble clouds of nanobots into an ene­my’s ter­ri­to­ry to dis­as­sem­ble every­thing and every­body into a “gray goo?” Longer, I sus­pect, than the cheer­lead­ers (or doom-mon­gers) of nan­otech would have us believe.

But, of course, Feyn­man was right: There is no law of physics to keep it from hap­pen­ing. We know that incred­i­bly sophis­ti­cat­ed machines can be fab­ri­cat­ed mol­e­cule-by-mol­e­cule on a micro­scop­ic scale. Nature does it every day.

As I write, a vast army of red spi­der mites is encamped out­side my win­dow — on the sill and win­dow panes. They are tak­ing advan­tage of the dry, bright weath­er to feast on the plants in the win­dow box. At first glance, they appear to be specks of red ink, about the size of the peri­od at the end of this sen­tence; but then you notice that the specks are mov­ing. Under a mag­ni­fi­er they become gor­geous scar­let ani­mals, with six artic­u­lat­ed legs and wav­ing antennae.

Where did these lit­tle fel­lows come from? Cer­tain­ly, nature did­n’t carve them from an amor­phous block of mat­ter, the way we make arrow­heads and com­put­er chips. Nature assem­bled them atom by atom — eyes, mouth, bel­ly, anus, gen­i­tals, heart, brain, ner­vous sys­tem — some­thing like 10 quadrillion atoms in all, every one in its prop­er place.

Red spi­der mites are vis­i­ble to the unaid­ed eye, if only bare­ly. The dust mites in my bed are too small to see with­out a micro­scope, but they are there by the tens of thou­sands (don’t smirk, they are in your bed, too). Elec­tron micro­scope pho­tographs of dust mites show every bris­tle and hair. Quel mon­stre! These all-nat­ur­al nanocrea­tures go about their busi­ness, eat­ing the skin flakes that fall from our bod­ies, and there is almost noth­ing we can do about them.

Will nano­engi­neers cre­ate mite-sized R2D2s — micro­scop­ic mechan­i­cal robots that will go into our beds and do bat­tle with dust mites? And then, hav­ing cleared our beds of dust mites, and skin flakes, too, march out onto the win­dow sill to gang up on the spi­der mites and save the plants? That’s what the nanogu­rus are telling us.

Don’t hold your breath. Maybe in 100 years, but not 30.

Mean­while, we will cer­tain­ly see nanocrea­tures doing use­ful things by human design (we have them already). They won’t be micro­scop­ic mechan­i­cal robots assem­bled atom by atom with tech­nolo­gies like the scan­ning tun­nel­ing micro­scope. Rather, they will be ordi­nary organ­isms, bac­te­ria most like­ly, but maybe dust mites, too, that have been genet­i­cal­ly engi­neered to make them do things we want them to do — like eat bad germs, clean up oil spills, or take up res­i­dence in some­one else’s bed.

What nature does, we also can do — by bor­row­ing nature’s DNA technology.

Share this Musing: