By now, just about everyone with an interest in the field of nanotechnology has heard that Heinrich Rohrer, who won the 1986 Nobel Prize in Physics for his co-invention of the scanning tunneling microscope (STM), passed away this week at the age of 79 from natural causes.

It would be hard to overstate the impact that Rohrer and his colleague at IBM Zurich, Gerd Binnig, have had on the field of nanotechnology. The STM has become a cornerstone tool for characterizing and manipulating the world on the nanoscale. Through ever more refined iterations of the device, we are peering into the atomic scale with greater and greater clarity. Even the lay-est of laypersons can appreciate the STM’s feats of prowess when they're put on display in videos in which atoms are made to perform stunts as if they're children in a home movie.

For a description of how the STM came to be and how it works, IBM Zurich’s reporting on Rohrer’s life is both thorough and poignant and I recommend you take a look at it.

All I would add are my own personal recollections of Rohrer from a one-on-one interview I had with him and from joint interviews I and other journalists had with him and Binnig back in 2011 while attending the grand opening of IBM Zurich’s new nanotechnology research facility, which IBM aptly named the  “Binnig and Rohrer Nanotechnology Center.“

In these interviews, I was struck by three things.

First, Rohrer’s absolute humility in his role in the development of the STM. He characterized himself as simply wanting to see if it would be possible to eliminate approximations of inhomogenities on surfaces and measure them precisely. Beyond his genius of simply asking the right question, he also had the good sense to hire a brilliant young scientist—Binnig—who could help him in his quest.

Second, Rohrer was funny. Nearly everything he said during our brief time together had a wry twist of humor to it. It seemed to be humor borne of humility (not taking himself too seriously), pragmatism, and his sense that his role as a leader in a technology revolution was so unexpected that he just had to laugh at it.

Finally, I was struck by the chemistry between the two men. They expressed unflagging admiration for one another, despite being in some ways polar opposites. Rohrer was the pragmatist, while Binnig seems to have the touch of the poet. Interestingly, though, in the development of the STM those roles were reversed in that Rohrer was the idea guy and Binnig was the engineer who got the device built.

In any event, their contrasting personalities, humor, and chemistry were on clear display the day of the opening of the lab named after them.

After Binnig had carefully answered a question about their co-discovery of the STM, Rohrer quipped, "If you didn't quite understand what Gerd just told you, you are not alone."

The audience laughed with relief that it was okay that they didn’t understand the carefully thought out explanation—I among them. But the truth was that Rohrer understood Binnig’s explanation perfectly and said that to put the audience at ease. Rohrer was both a great scientist and a true gentleman.

Image: IBM

Technology writers often hear complaints from readers that go something like: “All you ever talk about is this technology 'would,' 'could,' or 'might'.” Fair enough. But when the field is an emerging one, such as nanotechnology, most of the good stories are about just that—a development in the lab, or just coming out of it, that may or may not have an impact in the years to come.

Let's face it. A tech blog isn't the Daily Racing Form, and even in horseracing, good breeding is no guarantee of crossing the finish line first. First comes the struggle to secure funding, and then come any number of opportunities for management to make some tragic blunder or to fail to dislodge the incumbent competition, which often successfully blocks the technology from ever coming to market. The bottom line is that not only is success in the marketplace the exception and not the rule, but discerning the few winners from the many losers at a technology's earliest stages can make picking the ponies feel like child's play.

Then there's the frustration of time. Going to the racetrack offers immediate gratification, but handicapping high-tech requires quite a bit of patience. I have been writing about emerging technologies for over 15 years. In that time, I've chronicled some successes and failures and a common rule of thumb I picked up early on was that it typically takes seven years to bring a laboratory technology to market.

The seven-year rule is something of a shibboleth. Try as I might, I have not been able to determine where that notion originates, but I thought I should at least try to see how accurate it is as a barometer as to whether a new technology can make a commercial impact.

Thus this post starts a new series within The Nanoclast that looks back on some of the technologies that we have covered with words like “would”, “could” or “might.” How far along have On-Off Super Glue or Junctionless Transistors, to name just two of my favorites, progressed?

We're calling the series “Seven or Never,” a reference to the seven-year time-to-market timescale shibboleth. But first, I thought I might see if that timeline really holds true by asking a couple of tech jockeys who have put in their time riding fast horses down a seven-furlong track.

The first is Adrian Burden, who, while a researcher at the Institute of Materials Research and Engineering (IMRE) in Singapore, discovered a method for using nanomagnets as an anti-counterfeiting measure, especially for pharmaceuticals. In 2005, he launched a company called Singular ID based on that technology that was later acquired by Bilcare Research. He is currently  Technical Director of Key IQ, a business and technology catalyst based in the UK.

Like me, Burden was mystified by the origin of the “seven-year” idea, but he saw that there were some pretty common sense reasons for why it might have come into being.

“Certainly it usually takes much longer than you originally anticipate to bring a technology to market,” says Burden. “When people put together business plans or roadmaps, it is very difficult to see beyond three to five years, so seven years takes you nicely over the horizon.”

Still, he thinks that seven years is a pretty tight timeline.

“I personally don't think that the seven-year rule holds,” says Burden. “I think it is much more instructive to say it will be much longer than you anticipate. There is a wide variety of technological developments that can be brought to market and their complexity will really govern the timelines, along with a good dose of luck or timeliness.

I also spoke to Peter Dobson. He's the director of the University of Oxford's Begbroke Science Park, which has not only spun out companies from laboratory research that went on to become multi-million-dollar operations, but also now guides others on how to do it. His insights into rapidly transferring laboratory technologies to the marketplace are sought after around the world.

Dobson similarly doubts that even seven years is enough time to bring a new technology to market. “I think it is longer," he says. "And if anyone gets it below seven for a tangible product as opposed to an Information and Communications Technology (ICT) product, they are either lucky or very well organized.”

While Dobson acknowledges too that there are many complexities to launching a new technology commercially, he thinks there are different approaches that can shorten the time frame—at the cost of some tradeoffs. “One has to take account of the times taken to build production facilities, form partnerships for manufacture or sales, progress through regulations, and so on,” he says. “If partnering is adopted it is possible to get the period down to less than seven years.” However, Dobson notes that when you start dividing up company, you won't reap the full fortunes of the company--if there are any.

I asked Burden about the timeline itself—where do we begin the countdown? Do we start at the first paper published, first patent filed, or first series of funding?

“I think you have to start the clock when you consider the discovery or technological development to be of commercial significance,” he says. “So, the very basic blue-sky ground work can be discounted.  As such, publishing a paper may not be the starting point, but certainly an internal invention disclosure, internal project proposal describing commercial applications or a patent drafting would be a useful marker.”

Then too, some fields have timelines that are inherently lengthy.

”Medical developments that require animal testing, clinical trials and approvals could take much longer than seven years to come to market,” says Burden. “Aviation or automotive components that require testing and approval, as well as being incorporated into designs that may be on the drawing board many years in advance also could take longer to be adopted.  And almost all manufactured products needs to obtain approvals for sale in specific markets—CE Marking being an example. These can be costly and time consuming to obtain, and moreover restrict improvements in design that may help accelerate adoption by the market.”

There are two parts to “Seven or Never,”—the second being that if a technology hasn’t made a commercial impact within that time, we aren’t likely to ever see on the market. To this, Burden adamantly disagrees.

“I think there is always an opportunity for prototypes or concepts to lie dormant and then re-emerge when the time is right,” says Burden. “This is sometimes because other enabling technologies come along or become mainstream - for example RFID-enabled phones will lead to a proliferation of near-field communication (NFC) products and services that have otherwise languished until now.”

So how do we rejuvenate languishing technologies, and keep others from languishing in the first place? According to Dobson, some “robust tweaking” is in order, and he cited some examples in the U.K, like the Knowledge Transfer Networks  (KTN) set up by the Technology Strategy Board in the UK. Organizations like the KTN, and others such as the new Catapult Centres and High Value Manufacturing Centres, Dobson believes have the right mix for promoting innovation that includes close cooperation between government, industry and research centers. Dobson also believes that there needs to be an increased emphasis at universities on the concept of innovation and skills in entrepreneurship with courses to support that emphasis. "All of these can improve the situation," Dobson notes.

One approach that has been growing increasingly popular over the last decade is innovation parks—miniature Silicon Valleys devoted to a specific emerging technology. (Nanotechnology seems to be a favorite, attracting the attention of both New York and the U.K.) To be sure, there are no guarantees they will produce the economic impact some regions are counting on. But Burden, who came out of Singapore’s model of a tightly grouped collection of research institutes, believes it may be the best plan right now for shortening what's often more like a 10 to 15 year timescale for bringing an emerging technology to market.

“Although I don't believe you can force the next Silicon Valley to happen, I do believe that you can encourage it with the right environment,” says Burden. “So, by having vibrant tech parks interacting with local communities in which value chains can be established, entrepreneurs can share experiences and technologists from different disciplines can interact all help bring new technologies to market more quickly.”

He adds, “Ultimately, if you have the right conditions you can accelerate time to market. [Having] people who have done it before helps, [as does] being near companies that need your product, being able to quickly hire (and fire) people with the skills you need, and being able to raise finance in a timely manner to keep the momentum going.”

In summary, it's easy to call into question both the seven and the never part of Seven or Never. Nonetheless, we're going to stay with the name. In the months to come, I'm going to look at some of the technologies I've written about, but I'm also interested in the technologies and companies you've followed—or better still, worked with—as they run the ponderous gauntlet toward commercial viability. Maybe we'll learn something about handicapping the next generation of nanotechnologies.

Image Art: Mark Montgomery

Just three years after announcing a huge capacity increase to its multi-walled carbon nanotube (MWNT) production, Bayer Material Science has announced that it will completely close down its MWNT production to focus on its core business.

This is no surprise since there was a huge glut of product resulting in industry utilization rates that must have been in the single digits. This oversupplied market was the result of a MWNT capacity arms race that started in the mid-2000s. While this steep ramping up of production capacity reduced pricing from $700/kg in 2006 to below $100/kg in 2009—with some estimates putting the price at $50/kg as of last year—the problem seemed to be that no matter how cheap you made the stuff nobody was buying it because there were no applications for it. This resulted in stories, at once humorous and worrisome, of big chemical companies that had gotten themselves caught up in this arm race making desperate phone calls to laboratory researchers pitching application ideas for the material.

While some observers believed that this price cut would result in the applications being developed, most people recognized that this was a case of putting the cart before the horse, or “technology push” ahead of the preferable “market pull.”

This is not to say strategically it was wrong for a company like Bayer Material Science to build out capacity for a product that nobody seemed to want at that moment but may in the future. A company like Bayer can ramp up production with relatively little capital cost and manage to price everyone else out of the market. It was worth the risk.

However, hindsight makes it pretty clear that MWNTs applications were never really going to materialize as had been hoped. This became painfully clear when after a few years into production one of the target applications being touted for the material was the blades of large wind turbines. That announcement smacked of desperation.

Despite this, the story of MWNT capacity growth has been very instructive for how the so-called “nanotechnology industry” will shake out.

First, it’s clear that small operations that have found a way to produce a nanomaterial cheaply will have a difficult time competing with large chemical companies. This is not because they can’t produce the material more cheaply or at a better quality, but because they do not have the supply chain that well-established chemical companies have.

Second, you don’t want to be in the business of producing a nanomaterial that serves just to make some other product. You want to be making the final product. Many small start-ups no longer exist because they figured that they could just license their technology to a company that would make a product from their nanomaterial.

Bayer Material Science is in the position where it can just mothball its production without too much pain, but there may be some other companies that are less diversified for which that may not be an option. Sometimes when one domino falls the rest go in quick succession. So this should be an area to watch in the near future.

Image: Martin McCarthy/iStockphoto

Quantum dots have been promoted as a technology that is poised to transform the LCD (liquid-crystal display) market for years now. This promise looked to be taking shape when California-based Nanosys Inc. announced last year that it had worked out a deal with the Optical Systems Division of 3M Company to produce an LCD capable of displaying 50 percent more color.

The Nanosys/3M pairing was intended to improve the color and performance efficiency of LCD displays by using the quantum dots as an improved back light.

In the current display market landscape, LCDs are both inefficient and don’t produce the vibrant colors of organic light-emitting diodes (OLEDs). However, LCDs are far cheaper to produce in large screen sizes, and consumers often choose the right price over the right color. Quantum dots were supposed to give us the best of both worlds.

In work that appears to tip the scales further for quantum dot-enabled LCDs, researchers at the University of Illinois at Chicago (UIC) have developed a method for doping quantum dots that will give LCDs a color vibrancy not seen before.

In research published in the ACS journal Nano Letters ("Cluster-Seeded Synthesis of Doped CdSe:Cu4 Quantum Dots"), the UIC team reveal a method for introducing precisely four copper ions into each and every quantum dot. This doping with copper ions opens up the potential for fine-tuning the optical properties of the quantum dots and producing extraordinarily bright colors.

“When the crystallinity is perfect, the quantum dots do something that no one expected—they become very emissive and end up being the world’s best dye,” says Preston Snee, assistant professor of chemistry at UIC and principal investigator on the study, in a press release.

Whether UIC's doped quantum could be a compliment to the Nanosys/3M technology or a competition is not known. Likewise, it remains to be seen if they can keep LCDs at or near their current price point while bringing picture quality up to that of OLEDs. In other words, it'll take a few more years worth of Consumer Electronics Shows to sort out the winners and losers.

Image: University of Illinois, Chicago

Researchers at MIT and Sandia National Laboratory have made some long-awaited progress in lithium-air batteries. The research has provided insight into the electrochemical reactions that occur when they are being charged.

Lithium-air batteries promise five to 10 times greater storage capacity than traditional lithium-ion batteries, leading many to believe that they may hold the key to turning electrical vehicles from a niche market to a much larger segment of the automotive industry.

There’s no question that electric vehicles continue to grab headlines, like the Tesla Model S,  which recently won Consumer Reports' highest rating yet for an automobile. I suppose the Consumer Reports editors must not be bothered by its 425 kilomter range or the hours-long refueling.

To bring EVs more in line with what people expect from their fossil-fuel-powered cars—namely, a 650-kilometer driving range and a about 2 minutes to fill it up again for the next 650 kilometers—there will need to be some significant improvements to the batteries that power these all-electric vehicles.

While there have been improvements to the lithium-ion (Li-ion) batteries that are now used to power EVs, it’s long been rumored that technologically speaking we may have been barking up the wrong tree with Li-ion batteries.

To get to the point where batteries have the energy density to compete head-to-head on performance with fossil fuels they will need to reach around 10 00 Wh/kg. If you improved Li-ion batteries to twice their capabilities of where they are today, they would still only reach 400 Wh/kg. As former Secretary of Energy Stephen Chu outlined, battery technology will need to have six to seven times higher storage capacity than today’s batteries to be competitive with the internal combustion engine.

This is where lithium-air batteries step in with their ability to offer up to ten times the storage capacity of the Li-ion battery. However, to date these batteries have significant challenges for use in anything outside of highly-controlled laboratory environments.

The MIT and Sandia National Lab researchers, in work that was published the ACS journal Nano Letters (“In Situ Transmission Electron Microscopy Observations of Electrochemical Oxidation of Li₂O₂ ), used transmission electron microscope (TEM) to peer into one of the trouble spots in the further development of these batteries: the reaction known as oxygen evolution.

It was in this reaction that the researchers observed for the first time the oxidation of lithium peroxide, which is the byproduct material created during the discharge of a lithium-air battery. The observations revealed that the lithium peroxide forms primarily at the interface of the substrate, which is made of multiwalled carbon nanotubes.

In this location, the lithium peroxide acts as resistance to the flow of electrons and handicaps the charging of batteries. However, the researchers also discovered that during charging, when the electrons are passing through the carbon nanotubes, the lithium peroxide particles that had formed during discharge began to shrink. This means that if electron transport for these batteries can be improved these batteries could be made to charge much more quickly than previously had been thought.

"This work has identified the key limiting condition, electron transport … providing a critical contribution,” says Jie Xiao, a researcher at Pacific Northwest National Laboratory. This is a great example of how fundamental research can significantly improve our understanding to resolve challenges in practical devices. The information provided in this paper will benefit the rational design of the air electrode of lithium-air batteries.

While this research still doesn’t show a clear path for these batteries out of their current laboratory use, it may make better sense to pursue a battery technology for electric vehicles that has at least the promise of making them performance competitive with fossil-fuel-powered automobiles.

Photo: Creative Commons

Researchers from both the University of Madrid Complutense and Universidad Autonoma working together at the IMDEA-Nanociencia Institute in Spain have for the first time given graphene magnetic properties,opening up the potential that the material can find new applications in future spintronic devices.

Unlike electronics in which an electron’s charge-carrying capabilities are exploited to create circuits, spintronics involves the quantum mechanical property of electrons to spin, which creates a magnetic moment that makes the electrons behave briefly like magnets. When in the presence of a magnetic field the spin of the electrons moves either into a parallel or antiparallel position in relation to the field. This positioning can be translated into a binary signal (1 or 0).

The trials and tribulations trying to make graphene applicable to electronics despite its lack of an inherent band gap have been well documented. However, what many have overlooked in the quest to bring graphene to electronics is that it doesn’t really lend itself very well to spintronics either.

Since 2007, researchers have looked at graphene as the material for channels in spintronic devices. At this function, it appears to excel. In fact, just this year record distances were achieved for carry information using the spin of electrons.

Unfortunately, when two-dimensional graphene is laid out flat, the motion of electrons moving through the material doesn’t influence the spin of other electrons that they pass. Instead the direction and the spin of electrons remain random rather than patterned.

More than two years ago, researchers at the University of Copenhagen discovered that that all changed if you curved the graphene into a cylinder. In that shape, the movement of electrons did influence the spin of other electrons, opening the door to their potential in spintronics.

In order for a material to be have magnetic properties a majority of the electrons in the material must be spinning in the same direction. Despite the work of the Copenhagen researchers and many others, it has remained a challenge to get graphenes’ electrons to spin in the same direction instead of just randomly. But the Spanish researchers believe they have accomplished it.

“In spite of the huge efforts to date of scientists all over the world, it has not been possible to add the magnetic properties required to develop graphene-based spintronics. However these results pave the way to this possibility," says Prof. Rodolfo Miranda, Director of IMDEA-Nanociencia, in a press release.

The research, which was published in the journal Nature Physics ("Long-range magnetic order in a purely organic 2D layer adsorbed on epitaxial graphene"), first grew ultra pure graphene film over a crystal inside of a vacuum. While still in the vacumm, the researchers evaporated molecules of a semiconductor on the graphene’s surface. When they observed the material with a scanning tunneling microscope, they were surprised to discover that the semiconductor molecules were organized and regularly distributed across the surface of the graphene and its crystal substrate.

Since spintronics hasn't really progressed beyond its application into devices beyond hard-disk drives, the ability to give graphene magnetic properties likely won't bring spintronic devices into other applications any sooner. But these kinds of breakthroughs do have a way of opening up unexpected possibilities.

Photos:  IMDEA-Nanoscience

Last year, Amin Salehi-Khojin, assistant professor of mechanical and industrial engineering at the University of Illinois at Chicago, discovered that he could make highly sensitive chemical sensors from graphene. He also determined why they were so sensitive: Defects.

The research unveiled not only highly sensitive sensors capable of detecting a single molecule of a chemical, but also that the sensitivity, which was directly tied to defects around the edges of the graphene, would be lost if those defects were to be removed.

When Salehi-Khojin and his colleagues looked a little deeper into the need for defects to maintain sensitivity in graphene nanosensors, they found something remarkable: The graphene could be free from defects and still be a highly sensitive sensor as long as the substrate it was on was a little ragged around edges.

“This was a very surprising result,” Salehi-Khojin said in a press release. “[The results] will open up entirely new possibilities for modulation and control of the chemical sensitivity of these sensors, without compromising the intrinsic electrical and structural properties of graphene.”

The research, which was published in the ACS journal Nano Letters (“The Role of External Defects in Chemical Sensing of Graphene Field-Effect Transistors”), revealed that the poor sensitivity of pristine graphene in terms of electrical conductivity is not necessarily intrinsic to the material but instead can be affected and approved upon by the underlying substrate.

“We could now say that graphene itself is insensitive unless it has defects—internal defects on the graphene surface, or external defects on the substrate surface,”  noted UIC graduate student Poya Yasaei in the press release.

Now that graphene-based field effect transistors (FET) have been with us for a couple of years, this latest research opens up the potential for graphene-based chemFET sensors to be engineered for a number of various applications.

Photo: Roberta Dupuis-Devlin

The fate of polymer solar cells in the marketplace has been tied to three main factors: Lifespan in outdoor environments, the cost of materials that make up the modules (namely indium tin oxide, or ITO), and power-conversion efficiency. These three issues remain the keys to unlocking the commercial potential of polymer solar cells to being someday rolled out like plastic tarps to power our homes cheaply and reliably.

Nanotechnology has been trying to address all three of these issues, but perhaps none of them more than improving the power-conversion efficiency, which has lingered at around five to seven percent. Now researchers at the Ulsan National Institute of Science and Technology (UNIST) in Korea have used metal nanoparticles to achieve the highest yet reported power conversion efficiency for plasmonic polymer solar cells, reaching 8.92 percent. While polymer solar cells have been reported as high as 10.6 percent for polymer solar cells with more than one p-n junction, the UNIST researchers believe that their device, which reached nearly 9 percent using a single junction, could exceed 10 percent in commercial products.

The research, which was published in the ACS journal Nanoletters (“Multipositional Silica-Coated Silver Nanoparticles for High-Performance Polymer Solar Cells”) focused on polymer solar cells enhanced by plasmonics. Plasmonics exploits the phenomenon of "photons striking small, metallic structures to create plasmons, which are oscillations of electron density in the metal," as Neil Savage explained here on the pages of Spectrum.

The Korean researchers were able achieve high light absorption despite thinning out the films that make up the active layer of the solar cell by using silver nanoparticles. These nanoparticles provided the metal in the material that allowed for the exploitation of the surface plasmon resonance effect.

“This is the first report introducing metal NPs between the hole transport layer and active layer for enhancing device performance,” says Jin Young Kim, associate professor at UNIST and a leader of the research, in a press release. “The multi-positional and solutions-processable properties of our surface plasmon resonance (SPR) materials offer the possibility to use multiple plasmonic effects by introducing various metal nanoparticles into different spatial location for high-performance optoelectronic device via mass production techniques.”

While conversion efficiency seems to have been improved in the lab with this research, it will need to be demonstrated this new material will not deteriorate in the environment as some nanomaterial-enabled polymer solar cells have in the past. However, if it can exhibit this kind of robustness and is coupled with a suitable material to replace the expensive ITO, it may indeed be an important commercial step in the polymer solar cells.

Photo: Ken Fields/Creative Commons

This blog has chronicled many nanomaterials suggested for cleaning up oil spills over the years, with the most recent being an aerogel developed in China that the researchers claim to be the lightest ever produced and capable of soaking up a rather astounding 900 times its own weight in oil. This compares favorably to the current mainstay for oil spill remediation, hay, which only absorbs 3 to 15 times its weight in oil.

Now researchers at Deakin University in Australia have developed a nanosheet made of a porous boron nitride that can soak up 33 times its own weight in oil. While this weight-to-oil-ratio figure doesn’t seem to stack up favorably to some other technologies, like the aerogel above, it does have some side benefits that are lacking in some of the other solutions.

Notable among them is that once the nanosheets have soaked up their share of oil, they can be cleaned and ready to be used again by merely letting them heat in ambient air for two hours. They also are hydrophobic, meaning they repel water, which allows them to float on the surface of the water and be available for easy retrieval during a clean up.

The nanosheets, which are fully described in the journal Nature Communications (“Porous boron nitride nanosheets for effective water cleaning”),  were fabricated by mixing boron oxide powder and guanidine hydrochloride with methane and then heated at 1100 C for several hours in nitrogen gas. In this process, the guanidine hydrochloride decomposes to release several gasses that tunnel out, which results in the formation of the holes in the nanosheets.

So it sounds like a solution to oil spills is at hand—in fact, the Deakin University nanosheets have attractive characteristics for not just oil spill remediation but water purification in general. In fact, there are a variety of nanomaterials for these applications—so many of them that there are catalogues to guide you through them.  But not so fast. As yet, no one is bothering to commercialize them so that they are available for the next oil spill.

Today is three years to the day since I first highlighted this critical point that the nanotechnologies exist but not the commercial interest in making them available for the next oil spill, not much has changed. Perhaps the only way to ensure that these superior technologies are available to clean up the next inevitable oil spill is to institute government regulations requiring them, as IEEE Spectrum editor, Steven Cherry, suggested on his podcast, also nearly three years ago. Sometimes you have to force markets to adopt technologies when doing so may not help the bottom line, but keeps our planet habitable.

Image: Weiwei Lei

If there were a Nanoscale category in the Academy Awards, the 2013 winner would surely be a movie made by IBM Research that has carbon and oxygen atoms of carbon monoxide molecules being moved around on a copper surface with a scanning tunneling microscope. The 250-frame stop-motion film, entitled “A Boy and His Atom” (which you can see below), uses discrete atoms to draw a stick-figure-like boy that bounces on a trampoline and plays catch with an individual atom "ball."