# News this Week

Science  08 Apr 2011:
Vol. 332, Issue 6026, pp. 154
1. # Around the World

1 - Fukushima, Japan
Nuclear Crisis Drags On
2 - Haiti
New Studies Support Asian Source of Cholera
3 - Daresbury, U.K.
Scientists Take New Accelerator on Test Run
4 - South Atlantic Ocean
Oceanographers Find Failed Flight
5 - Brasília, Brazil
Alleged Fraud Spurs New Scientific Integrity Commission

## Fukushima, Japan

### Nuclear Crisis Drags On

As Science went to press, engineers were trying to staunch the flow of highly radioactive water from the devastated Fukushima Daiichi Nuclear Power Plant. Alarming concentrations of radioactive iodine-131 and cesium-137 have been detected around the plant and in nearby seawater. Authorities suspect that water pumped in to cool the unit two reactor is leaking into a cracked concrete utility pit, from which it is seeping into the ground and then into the ocean. “We have to do everything we can to stop the leakage as soon as possible,” Noriyuki Shikata, a government spokesperson, said at a press conference on 4 April. In addition, there will be an intentional release of 11,500 tons of slightly radioactive water from a holding tank into the sea to make room to store more dangerously contaminated water.

The 11 March tsunami knocked out fishing fleets in the area near the power plant. Japan's Fisheries Agency says it has detected no significant radiation in catches from neighboring areas so far. Meanwhile, plant workers have been trying to restore cooling systems to stabilize the nuclear fuel in the reactor cores and spent fuel pools.

## Haiti

### New Studies Support Asian Source of Cholera

Two recently published studies support the theory that Haiti's cholera outbreak was introduced from South Asia—but neither specifically fingers the U.N. peacekeeping forces from Nepal that some suspect inadvertently brought the disease to the island last October (Science, 28 January, p. 388).

The two genetic analyses—one published in the April issue of The Lancet Infectious Diseases, the other online 18 March in Emerging Infectious Diseases—both found that Haiti's strain most closely resembles cholera samples collected in South Asia, providing “independent corroboration” that the microbes were imported from that part of the world, says Matthew Waldor of Harvard Medical School in Boston, who drew the same conclusion in a January paper in The New England Journal of Medicine. To prove a Nepalese connection, however, researchers would need to compare Haitian microbes with a much larger set of samples from various Asian countries, he adds. Waldor and other cholera scientists are eagerly awaiting the report by a U.N. expert panel on the outbreak, slated for release within 3 weeks.

## Daresbury, U.K.

### Scientists Take New Accelerator on Test Run

A new type of particle accelerator, which promises to provide particle beams more cheaply and efficiently than existing types for applications ranging from cancer treatment to nuclear power, passed a major milestone last week when it circulated a beam all the way around its circumference and accelerated it to 18 megaelectron volts.

The accelerator, known as EMMA (Electron Machine with Many Applications) and built at the Daresbury Laboratory in the United Kingdom, is the world's first nonscaling fixed-field alternating-gradient accelerator, or nonscaling FFAG. Such machines don't have the accelerating power of CERN's Large Hadron Collider. But their use of smaller, simpler magnets allows them to produce particle beams that are more affordable for applications such as proton beam cancer therapy, scanning cargo for explosives, and inherently safe nuclear reactors known as accelerator-driven systems.

“Now that we know that the basic idea works, we can go on to develop the applications of this new technology,” says team member Ken Peach, director of the John Adams Institute for Accelerator Science at the University of Oxford and Royal Holloway University of London.

## South Atlantic Ocean

### Oceanographers Find Failed Flight

Only days into the fourth search for an ill-fated 2009 flight, oceanographers have found the remains of Air France Flight 447 3900 meters deep in the Atlantic, about 500 kilometers off northeast Brazil. The three autonomous underwater vehicles (AUVs), operated by oceanographers from Woods Hole Oceanographic Institution in Massachusetts, first located debris on relatively smooth seafloor using sonar; an AUV then returned to take photographs, which provided definite identification. The wreck isn't far from the spot that French authorities had eventually identified as the most likely resting place for the crash. A follow-up investigation using a remotely operated vehicle able to grapple objects will attempt to recover the plane's flight recorders.

## Brasília, Brazil

### Alleged Fraud Spurs New Scientific Integrity Commission

Brazil's science ministry will create a commission on scientific integrity after Elsevier said it would retract 11 papers, the senior author of which is Claudio Airoldi, 68, a chemist at the State University of Campinas, known as Unicamp.

Elsevier launched an investigation after receiving complaints about the work; a review by three independent experts concluded the papers contained forged nuclear magnetic resonance (NMR) spectra. Both Airoldi, who has co-authored over 400 papers, and Denis Guerra, the first author on the retracted papers, denied the accusations. Guerra, now at the Federal University of Mato Grosso, was quoted in the Brazilian press as saying, “There is no need for a Unicamp researcher to fabricate data. It's an absurd accusation.”

Although the retraction of the thinly cited papers is unlikely to have much scientific impact, it could change the way fraud allegations are handled in Brazil. Brazil's key funding agency, the National Council for Scientific and Technology Development, has so far left fraud probes to universities. Now, officials in Brasilia say they will create their own investigative body. Airoldi, who holds the agency's highest ranking, was suspended from reviewing federal grants until investigations are complete.

2. # Newsmakers

## Three Q's

Caltech neuroscientist Christof Koch, famous for studying the biological nature of consciousness, is moving from Pasadena to Seattle to become chief scientific officer of the Allen Institute for Brain Science. The institute is best known for creating atlases of gene expression in the brain.

Q:Does this mark a turning point for the Allen Institute?

Until now, they've focused on these atlases, which you can think of as static information. Now what we want to do is much more dynamic, we want to record the electrical activity of large numbers of neurons in the brain during certain well-specified behaviors, using a range of techniques. The mission is to understand how information is encoded and processed in the brain.

Q:How will you decide which questions to ask with this technique?

Delicately. Nobody has done this kind of high-throughput electrophysiology. Some people will say it can't be done, but I don't think that's true. The difficulty is picking goals and milestones that are achievable in the next 10 years.

Q:What's one question you're considering?

One example is 40 hertz gamma oscillations [of electrical activity in the brain]. They were discovered a long time ago, but in 2011 if you ask a bunch of neuroscientists what they do or are they even there, you get totally different answers. Why? Because there are no unified standards for experiments like in other areas of science. People are more likely to use each other's toothbrush than to use each other's protocols. So even for basic questions like this, we still don't have an answer.

## Astrophysicist Honored For Asking ‘Vital Questions’

The Templeton Foundation has awarded its annual prize to astrophysicist Martin Rees for his “profound insights on the cosmos [that] have provoked vital questions that speak to humanity's highest hopes and worst fears.” Currently Master of Trinity College at the University of Cambridge in the United Kingdom, Rees has a seat in the House of Lords and served as president of the Royal Society from 2005 until November 2010. He has written numerous books and papers on the philosophical questions raised by the physics of the universe's beginnings, as well as how human activities will determine the Earth's future.

The $1.6 million prize is the largest award for an individual, and honors a person who has “made exceptional contributions to affirming life's spiritual dimension.” “I hadn't thought I had the basic entry qualifications, looking at the previous winners,” Rees says. “I'm proud to be joining that roll call.” The foundation's mission has concerned some scientists, who have accused it of blurring religious and scientific values. Unlike many past awardees, Rees is not religious but says he is “inspired” by religion's contributions to the arts and feels that philosophy and ethics are important to scientific inquiry. 3. # Random Sample ## Really Big Dinosaurs Take Over New York The biggest animals ever to walk the Earth are about to invade the Big Apple. From 16 April to 2 January 2012, the American Museum of Natural History (AMNH) in New York City is hosting an exhibit devoted to “The World's Largest Dinosaurs,” the gargantuan, plant-eating sauropods that roamed the planet 200 million to 65 million years ago. The show's centerpiece is a life-size model of Mamenchisaurus, who weighed 20 metric tons and whose 9-meter-long neck made up half of its body length. (Some of its sauropod cousins were up to four times heavier.) Parts of the model will be cut away to reveal its skeleton, muscles, heart, and digestive system. These dinosaurs may be ancient, but the science will be new. AMNH paleontologist Mark Norell, curator of the dino show, says that the exhibit will focus on recent findings by paleontologist Martin Sander of the University of Bonn in Germany, whose team has been studying why dinosaurs got so huge in the first place (Science, 10 October 2008, p. 200). “This exhibit is about … the evolutionary solutions they developed to cope with their enormous size,” Norell says. ## By the Numbers 1.21 billion — India's population, according to last week's census results. That's up 181 million since 2001, the equivalent of moving nine out of every 10 Brazilians to India. 9.57 — Amount, in zettabytes (1021 bytes), of business-related information processed by the world's computer servers in 2008, according to a report by scientists at the University of California, San Diego. ## Death of a Polar Bear A new scanning-and-prototyping technique has helped to explain why Knut, the world-famous polar bear, died suddenly in front of hundreds of shocked visitors to the Berlin Zoological Garden last month. Thomas Hildebrandt and colleagues at the Leibniz Institute for Zoo and Wildlife Research (IZW) in Berlin used their bear-sized computed tomography scanner (Science, 16 July 2010, p. 261) to scan Knut's body. Then, with Ben Jastram and Joachim Weinhold of the Technische Universität Berlin's 3D Lab, they made hyperaccurate replicas of his skull, brain, and face (pictured). The necropsy conducted by pathologist Claudia Szentiks at IZW identified encephalitis, possibly viral, as the cause of Knut's apparent seizure and fall into the water in his enclosure, where he drowned. The scans and models showed that the brain was structurally normal, ruling out an inborn defect and explaining why Knut showed no symptoms before his deadly fall. “It helped to show that the zoo couldn't have known he was sick,” Hildebrandt says. Polar bears in the wild sometimes fight to the death and so have an evolutionary reason to hide any symptoms of disease or other weakness, he says. The researchers are also using data from the scans to create a virtual tour of Knut's insides. 4. Agbiotech # Can Biotech and Organic Farmers Get Along? 1. Erik Stokstad The U.S. Department of Agriculture has been taking a closer look at the impact of biotech crops on organic farms. Research is providing tools to help them thrive side by side, but the politics are tricky. For proponents of Agricultural Biotechnology, getting new crops to farmers has become a lot more complicated in the past few years. Take alfalfa, genetically modified (GM) to resist herbicides. This long-awaited variety was introduced in 2005, and within a year, farmers snapped up all the available seed and sowed 250,000 hectares. But in 2007, all planting was halted after opponents persuaded a federal judge to order the U.S. Department of Agriculture (USDA) to conduct a more comprehensive environmental review—one that was completed only this past December. And then there are sugar beets. USDA approved herbicide-resistant sugar beets in 2005. About 3 years later, environmental groups and seed producers sued USDA, again claiming an inadequate environmental impact assessment. In August 2010, the court prohibited planting until the agency completes a more detailed review next year. In February, the court lifted the ban, but the confusion and uncertainty have thrown the industry into turmoil. Both of these cases have meant extra regulatory work at USDA, contributing to a growing backlog of applications for biotech approval. Other lawsuits have hit the industry directly in the wallet: Last year, a court ordered Bayer CropScience to pay$48 million in damages after experimental transgenic rice accidentally entered the food supply.

This terrain is a stark contrast to the boom years of the 1990s, when varieties of GM crops quickly spread across the United States. Although opponents raised concerns about safety and ecological impacts, USDA approved variety after variety. Philosophical objections to biotech crops haven't diminished, but the current spate of lawsuits and delays now hinges on economic damages; the issue is how to keep the transgenes in GM crops from spreading to organic crops, which can be costly for organic farmers to prevent. Failure can mean they give up the premium price that organic products fetch.

Enter Tom Vilsack, head of USDA, whose new mantra is “coexistence.” In his vision, biotech continues to innovate, organic farmers are protected from unwanted transgenes, and no one is tied up in costly litigation. In December, Vilsack electrified the debate by proposing the agency's first restrictions on biotech growers—specifying isolation distances between GM and organic alfalfa to prevent gene flow—and hosting a high-level discussion of how to achieve coexistence. For the organic community, the moves were long overdue. Although to the community's dismay, USDA subsequently reapproved GM alfalfa without any restrictions, the agency pledged to fund research on coexistence and to reinstate a key advisory committee on biotechnology to provide advice on the issue.

Experts say without question that GM and organic farmers can coexist—if there is an achievable, agreed-upon standard of how pure is pure enough to receive the organic label or to be accepted by overseas markets. At this stage, with so much of U.S. fields planted with GM crops—93% for soybeans—everyone agrees it's impossible to completely exclude transgenes from organic fields, but they can be kept to minimal levels. “As long as zero tolerance is insisted upon, you can forget about coexistence,” says Patrick Byrne, a plant geneticist at Colorado State University in Fort Collins.

With a defined threshold, say the presence of 0.1% or 0.9% transgenic seeds, depending on the crop, scientists can figure out the appropriate distances between fields to minimize gene flow. In the future, computer models of pollinator behavior may help provide recommendations tailored to particular landscapes. Another approach to prevent the spread of transgenes is to breed crops that can't be fertilized by transgenic pollen; the first commercial varieties of corn with this protection should be released this fall.

Harmony isn't likely anytime soon, however. The sides remain split on key issues. Organic groups demand more government oversight, that the biotech industry share the cost of preventing gene flow, and the creation of a compensation fund for damages if their crops cannot be sold as organic. The biotech industry opposes all of these goals. So far, USDA seems to continue to lean toward the industry in how it approves GM crops. “The public face of USDA is changing, but their actions are business as usual,” says A. Bryan Endres, an expert in agriculture law at the University of Illinois, Urbana-Champaign.

## A changed landscape

Coexistence has been a part of agriculture for a long time. Neighboring farmers have had to make sure that they keep enough distance when they plant crops that shouldn't cross-pollinate, such as popcorn and sweet corn, lest they end up with something unsalable. The issue is even more important when growing seeds for planting, because seed purity affects the quality and consistency of the crops. For seed growers, USDA and certifying organizations have for decades provided guidance on best practices, including isolation distances for fields.

But the advent of biotech changed the game because DNA tests make it easy to find GM seeds—at very low concentrations. “The transgenes have given a great opportunity for finding that contamination,” says Carol Mallory-Smith of Oregon State University (OSU) in Corvallis, whose research has shown just how easily transgenes move around (see p. 168). And given the strong opposition to GM crops in some export markets, Japan and Europe, especially, farmers are worried about not being able to sell specially grown organic crops if just the tiniest trace of transgenes is found. “The organic industry is extremely concerned” about the spread of transgenes, says Christine Bushway, executive director of the Organic Trade Association in Washington, D.C.

There are few data on the economic impact of approved GM crops on organic farmers, but they say the burden can be steep. Organic farmers bear the costs of preventing gene flow into their crops, which they do by planting buffer strips, for example, and testing for transgenes. Economists point out that these costs are generally compensated by the premium price fetched by organic crops. However, organic farmers can face other risks as well; planting at different times than their neighbors, a tactic to avoid gene flow, can lower yields and raise the odds of crop loss from bad weather.

Coexistence strategies are based on how pollen moves, which varies dramatically by crop. Soybeans, for example, are self-pollinating, and insects only rarely spread their pollen. Corn, which is outcrossing and wind-pollinated, is another matter entirely. To figure out what isolation distances and farming practices to recommend to farmers dealing with GM crops, scientists typically sow a field with plants that have a genetic marker, such as herbicide tolerance. Then they plant nontransgenic seeds in “study plots” at various distances. The transgenic crops will survive when sprayed with herbicide; if any plants in the study plot are still living, then they have acquired the resistance gene from the original field.

Although such experiments are straight-forward in principle, they rarely encompass the diversity and complexity of agricultural and environmental conditions. Farming practices, geography, and weather can all influence the spread of pollen. This is especially true for the dynamics of insect-pollinated crops, such as alfalfa. Seed producers may bring in hives full of honeybees or leafcutter bees, which fly different distances, to pollinate their crops. Native pollinators such as bumblebees may also influence how far the pollen travels. Researchers with the University of California, Davis, led by Larry Teuber, and USDA recently spent 3 years tracking pollinators (see photos, above) and sampling seed to measure gene spread via pollen. The results are being used to verify and refine isolation distances for biotech alfalfa.

In a different project, Johanne Brunet, an evolutionary biologist with the USDA's Agricultural Research Service in Madison, Wisconsin, is building a model to predict gene flow based on pollinator behavior and the agricultural landscape. “My hope is that it might help us design better ways to prevent gene flow,” Brunet says. Ultimately, the model might provide isolation distances and other guidance customized for particular farms and crops.

## Pollen barriers

Isolation distances can't prevent gene flow entirely, however. Wind and insects can take pollen surprisingly far, albeit in small amounts (Science, 28 June 2002, p. 2314). So rather than simply focusing on changing farming practices, some scientists are investigating another strategy: changing the plant itself. The hope is that by designing plants that can't be fertilized by transgenic pollen, researchers can “fence out” GM crops.

For instance, researchers have known for decades that some varieties of tropical corn can't be pollinated by other varieties, a phenomenon known as cross incompatibility. Fertilization of these varieties can take place only if the pollen and the plant receiving it have genes called GA1 and GA2. About 10 years ago Tom Hoegemeyer, now at the University of Nebraska, Lincoln, began breeding this trait into regular field corn, which he dubbed PuraMaize. After 4 years of further breeding, Blue River Hybrids Organic Seed, a company in Kelley, Iowa, plans to start selling three commercial hybrid varieties of PuraMaize to organic farmers this fall. “We believe it should be effective,” says company founder Maury Johnson, who has done small-scale tests.

But it's unlikely to be a long-term solution because even though PuraMaize plants are protected from foreign genes, their own pollen is able to fertilize other varieties. This means that the GA1 and GA2 genes can spread, mixing into the DNA of other corn. If that happens with transgenic corn, it would gain the ability to pollinate PuraMaize. It's also possible that GA1 and GA2 could slip into future kinds of GM corn from other research varieties used during breeding. “This [PuraMaize] system will probably break down,” predicts Major Goodman, a corn breeder at North Carolina State University in Raleigh, who is nevertheless working on adding the trait into other varieties of corn.

Another source of cross incompatibility is teosinte crossing barrier-1 (TCB). This locus, which behaves similarly to GA1 and GA2, was first researched in the 1970s by Jerry Kermicle of the University of Wisconsin (UW), Madison. Kermicle has taken a first step at breeding TCB into regular corn, but so far there hasn't been much interest from corn breeders. “There's nothing in it for the biotech firms,” he says, “and the small players in the organic community, they're not organized or centralized.”

Efforts are also under way to modify biotech crops so that they can't spread pollen. This would essentially “fence in” the transgenes. Several approaches are being investigated in labs, such as using RNAi to make transgenic plants sterile. But such approaches are fairly far away from development and implementation, largely for commercial reasons, says Henry Daniell, a molecular biologist at the University of Central Florida College of Medicine in Orlando: “The biotech companies have no incentive unless they get a strong product advantage.”

At least one company is pursuing a technology that would, as a side benefit, prevent gene flow in pollen. That approach, which Bayer CropScience has researched in soybeans and other plants, is to engineer chloroplasts to express transgenic traits. Because chloroplasts are maternally inherited, pollen very rarely carries the transgene. And traits placed into chloroplasts can be expressed at levels higher than those put into the nuclear genome, Daniell says, although it is a more difficult approach. Steven Strauss of OSU notes that this approach would be most useful as part of a suite of containment tools for biopharmceuticals rather than food crops.

Some in the organic community are skeptical in any case. “We do not believe there exists a technological silver bullet to the multiple challenges that genetically engineered crops present,” says Kristina Hubbard, director of advocacy for the Organic Seed Alliance in Port Townsend, Washington.

## Entrenched positions

Even if gene flow from GM to organic crops could be prevented in the field, there are ample opportunities for it to occur after harvest, for instance, through accidental mixing. This occurred with dramatic consequences in September 2000 with biotech corn called StarLink. Approved only for animal feed, StarLink was detected in taco shells and then found in more than 10% of the corn supply. Massive recalls resulted. The debacle was estimated to have cost Aventis, the developer of StarLink, hundreds of millions of dollars.

Commingling of approved biotech varieties won't lead to recalls of organic food in the United States. That's because, even though most consumers are unaware, organic products in the United States are allowed to contain transgenes. To earn the organic label, USDA specifies that farmers can't plant GM crops. But if transgenes accidentally end up in their fields, the harvest can still be called organic.

That's unacceptable to some buyers, who have zero tolerance for GM organisms (GMOs). But others recognize that a little contamination is inevitable, which underscores the importance of agreeing on quantitative standards for how pure is pure enough.

Such standards can be set in a variety of ways. Some countries decide what percentage of transgenes is acceptable in imported commodities, say 1% in soybeans. But in the United States, the government has not set a standard for organic labels.

In the absence of a federal standard, a private labeling scheme, the Non-GMO Project, set its own targets of 0.25% for seeds and 0.9% for food and ingredients in 2008 for all participating products. Manufacturers who are certified as testing their ingredients for transgenes can use the Non-GMO Project label. “Doing this proactively with a high level of transparency was [decided to be] the best way to win and sustain consumer confidence” in organic products, says Charles Benbrook of the Organic Center in Boulder, Colorado, who is a technical adviser to the Non-GMO Project.

In addition, some agricultural trade groups in the United States have agreed on realistic standards for specific crops. Before GM alfalfa was first approved, for example, industry funded research into gene flow and determined that organic seed producers would be able to easily achieve a goal of less than 0.1% transgenes in their fields. Forage Genetics International (FGI) in Nampa, Idaho, which commercialized biotech alfalfa with Monsanto, will require farmers who buy its seed to agree to its list of best practices. “The stewardship programs are mandatory,” says FGI President Mark McCaslin.

But such agreements aren't enough for some organic groups, which would like to see the government, relying on a scientific advisory committee, set such standards and create penalties for breaking them. USDA doesn't seem headed in that direction. In its January decision on alfalfa (Science, 4 February, p. 523), the agency opted to leave regulation up to industry, as it did the next month when it approved the first corn variety designed for ethanol producers. “At the place we are now politically, I don't think we'll see strict standards put in place” by USDA, says Alison Peck of the West Virginia University College of Law in Morgantown.

The organic community may yet get at least a discussion of its most controversial wish: a liability fund that would compensate growers who can't sell to markets that reject biotech. This idea is anathema to the biotech industry, which argues that shifting this cost to technology developers would stifle innovation. Vilsack, however, has asked the newly reestablished USDA Advisory Committee on Biotechnology and 21st Century Agriculture to explore the concept.

As for Vilsack's goal of reducing litigation, most observers say there's probably no way to stop the lawsuits against USDA, which are filed by advocacy groups—such as the Center for Food Safety, the lead plaintiff in the alfalfa and sugar beet lawsuits—that remain vehemently opposed to new GM crops and want existing varieties yanked as well. “Expect to see more litigation,” the University of Illinois's Endres says.

Even so, many are hopeful that the new discussions, taking place among farmers, seed producers, and buyers, will lead to concessions that lower the risk for organic farmers. “I'm optimistic that a lot of people in places of influence are thinking about this,” says Manjit Misra of Iowa State University in Ames. Molly Jahn of UW Madison is also sanguine, based on recent discussions with stakeholders involved in the alfalfa guidelines. “It's possible for reasonable people to commit to a way forward, grounded in science, with reasonable standards,” she says.

5. Profile: Carol Mallory-Smith

# Scientist in the Middle of the GM-Organic Wars

1. Dan Charles*

For Carol Mallory-Smith of Oregon State University, the migration of genes in agricultural crops is not just a research topic or a matter of policy debate. It's the cause of a vexing quarrel among her neighbors: the farmers of Oregon's Willamette Valley.

For Carol Mallory-Smith of Oregon State University in Corvallis, the migration of genes in agricultural crops is not just a research topic or a matter of policy debate. It's the cause of a vexing quarrel among her neighbors: the farmers of Oregon's Willamette Valley.

This valley, because of its mild climate and ample water for irrigation, is one of the world's great seed-growing centers, supplying farms, gardens, and golf courses worldwide. Because pollination is at the heart of seed production, the valley is now the scene of heated debates—and one far-reaching lawsuit—over the consequences of genetically modified pollen or seeds drifting into fields filled with sexually compatible non-GM crops. Organic farmers who fear contamination of their crops are on one side; the growers of seed for genetically engineered sugar beets are on the other. And Mallory-Smith is in the middle. Her publications have found an avid readership among biotech industry lawyers and activists opposed to GM crops. “The research part has been fantastic,” she says with a wry smile. “The politics of it is difficult, to say the least.”

Mallory-Smith came to the topic of migrating modified genes through her research on cross-pollination between wheat and one of its close relatives, a weed called jointed goatgrass. The two species can produce hybrids, and Mallory-Smith wanted to know whether this would allow new herbicide-resistance genes that had been introduced into wheat to migrate into the crop's weedy relative.

She extended that research to turfgrass in 2001, when the Scotts Miracle-Gro Co. began experimental field trials of glyphosate-resistant, sometimes called “Roundup Ready,” creeping bentgrass near the town of Madras, in central Oregon. Mallory-Smith began monitoring areas near the fields and, along with other researchers, documented a large-scale genetic migration: Hundreds of glyphosate-resistant grass plants evaded all the company's efforts to confine them. This grass still has not been approved for unrestricted cultivation, so the company is supposed to find and destroy any such plants that show up outside of its trial plots. Some GM grass was found 20 kilometers from the fields where it belonged. It was destroyed, but years later, researchers continue to find unapproved GM grass plants growing wild near the former test plots in Madras.

Much of this gene jailbreak probably happened on one day in 2003, when an unexpected windstorm blew away swaths of seed that had been left out to dry at the Scotts research facility. Mallory-Smith says this event should not have been a surprise, because freak weather is a natural part of farming: “This is what happens in production agriculture.”

That episode cemented Mallory-Smith's reputation as an expert on migrating transgenes, especially in the Willamette Valley. When sugar beet seedlings turned up in topsoil that was sold at a garden supply store in Corvallis, residents brought them to Mallory-Smith for testing. The seedlings turned out to be glyphosate-resistant.

Last fall, farmers in eastern Oregon noticed that some grass in their irrigation ditches seemed immune to Roundup. They, too, sent samples of the grass to Mallory-Smith, who confirmed that the grass contained an inserted glyphosate-resistance gene. She also flew out to take a look for herself. “The plants have obviously been there for a while. They're large. They've gone to seed. My guess is, they're 3 or 4 years old,” she says.

Such grass still cannot be legally planted without a permit from the U.S. Department of Agriculture (USDA). Most likely, it spread from another set of field trials conducted by Scotts Miracle-Gro just across the border in Idaho.

The Scotts Miracle-Gro Co. tried to eliminate the rogue grass over the past few weeks with a herbicide that state regulators approved temporarily for this purpose. (Glyphosate had been the only weed killer allowed in irrigation ditches.) But according to Jay Chamberlin, manager of the Owyhee Irrigation District in Nyssa, Oregon, no more spraying will be allowed this spring because water is about to reenter the ditches. Chamberlin says it is likely that some unapproved turfgrass will remain in the ditches through the summer, and it may spread farther. “The more they look, the more they find,” he says.

These experiences convinced Mallory-Smith that USDA regulators haven't fully understood the dynamics of gene flow, at least when it comes to turfgrass. “They knew a lot about corn, soybeans, and cotton. But now we're dealing with a perennial, with lots of relatives. And it's weedier; it does survive outside of cultivation. They didn't have an understanding of that kind of cultivation,” she says.

Generally, she says, regulators and biotech companies have been overly confident that they can prevent modified genes from spreading. “When you put them out there, you have to accept the fact that you're not going to contain them; you're not going to retract all these genes,” she says.

For all the shortcomings that Mallory-Smith sees in GM regulation, she doesn't really fit in the anti-GM camp. “She's an honest broker, if there are any honest brokers in all this,” says Steven Strauss, a colleague at Oregon State. Mallory-Smith says she's sympathetic to both sides. “I'm a public servant, and that sometimes means that you end up in the middle of these factions,” she says. But she's hoping that both organic farmers and growers of genetically engineered beets can thrive: “We need all our industries.”

• * Dan Charles is a writer based in Washington, D.C.

6. American Physical Society Meeting

# Electrons Surf Sound Waves To Connect the Quantum Dots

Physicists have used sound waves to transport individual electrons over significant distances between artificial atoms on a chip, they reported at the American Physical Society meeting.

Physicists at the University of Cambridge in the United Kingdom have used sound waves to transport individual electrons over significant distances between artificial atoms on a chip. The technique might someday help scientists to shuttle information around within a quantum computer, although such a device is still far from reality. More immediately, the technique represents a striking confluence of electronics, acoustics, and nanotechnology.

The mock atoms are quantum dots, which can be either nanometer-sized bumps of semiconductor or simply spots on the surface of a chip ringed by electrodes that control them. As in an atom, an electron in a dot can occupy only a few quantum states of definite energy. Scientists can use a dot as a quantum bit, or “qubit,” to store information. For example, they can trap an electron in one state and use its spin to encode information: a “0” if the electron's spin points up, a “1” if it points down, or—thanks to the weirdness of quantum mechanics—both 0 and 1 as the electron spins in both directions at once.

Researchers have used such qubits to perform elementary logic calculations. But they still face the problem of making the dots communicate with one another. One scheme involves getting an electron from one dot to “tunnel” into a neighboring dot. But that requires placing the two dots within a couple of hundred nanometers of each other, says Hendrik Bluhm, a physicist at RWTH Aachen University in Germany. “Moving quantum information around over long distances is a big problem for the field,” he says.

Now Robert McNeil, Christopher Ford, and colleagues at Cambridge have found a way to transport individual electrons much farther using sound waves. On a wafer of gallium arsenide, they laid out electrodes 4 micrometers apart that defined two quantum dots and also a channel between them. They also laid out a ladderlike electrode that, when charged, causes the gallium arsenide to contract, setting off a sound wave that ripples across the surface. The sound wave in turn produces a corrugated electric field that travels with it and can predictably pick up an electron from one dot and transfer it to the other. Literally surfing along at the speed of sound, the electron covers the distance between the dots in 1.5 nanoseconds, McNeil reported. “This procedure can be repeated hundreds of times with the same electron,” he says.

The next challenge is to ensure that the sound wave transporting the electron does not obliterate the delicate quantum state of the electron's spin. The electron acts a bit like a gyroscope, and if the process is gentle enough, then transporting it across the semiconductor should simply make the direction of its spin turn, or “precess,” by a predictable amount, Bluhm says. McNeil says he and his colleagues are now trying to verify that it does.

But even without answering that question, the Cambridge team has performed a very pretty experiment, Bluhm says: “Just the bare technique is pretty impressive.”

7. American Physical Society Meeting

# Ice Is Predicted to Be Weirder Still

A team of theorists reported at the American Physical Society meeting that a film of ice only two molecules thick may form the oddest of all crystalline structures, a so-called quasicrystal, which lacks the exact repeatability of an ordinary crystal structure but preserves other symmetries of a crystal.

Plain old water continues to surprise physicists. For example, researchers have found that the molecules in pressurized ice can arrange themselves in at least 15 distinct crystalline patterns. Now a team of theorists says a film of ice only two molecules thick may form the oddest of all crystalline structures, a so-called quasicrystal, which lacks the exact repeatability of an ordinary crystal structure but preserves other symmetries of a crystal.

A quasicrystal is nature's way of finessing a problem in geometry. Rows of squares, regular triangles, or regular hexagons can tile a plane. But equal-sided pentagons need not apply, as the five-sided figures leave gaps between them.

In the 1960s, mathematicians found a surprising way to skirt this rule with not-quite-regular patterns containing pentagons. The patterns don't repeat from row to row. However, as with squares, triangles, and hexagons, the patterns look the same when rotated by a certain amount—in this case, a fifth of a turn. In the 1980s, scientists found alloys such as aluminum lithium copper that form such quasicrystal structures. Now, Valeria Molinero, a theoretical chemist at the University of Utah in Salt Lake City, and colleagues predict that a layer of ice under the right conditions should also form the bizarre structure.

Ice on surfaces has pulled some tricks before. For example, in bulk ice, water molecules arrange themselves in a three-dimensional pattern. But theorists had predicted that when confined to two layers on a surface, the molecules would bend so that they could lay down completely flat in a hexagonal pattern. And that's what experimenters observed in 2009.

This time, Molinero and colleagues considered two layers of water molecules that are confined between two generic metallic surfaces and subjected to low temperatures and high pressures. Near the surface, the V-shaped molecules, with an oxygen atom between two hydrogen atoms, must balance several influences. The oxygen atoms prefer to sit atop the atoms in the surface. The molecules also try to maintain the lengths of their oxygen-hydrogen bonds and to arrange themselves so that none of those bonds is left dangling. Simulations that account for those factors reproduce the known phases of ice on surfaces and predict the existence of a quasicrystal, Molinero said at the meeting.

The prediction seems plausible to Angelos Michaelides, a chemist at University College London. In fact, in 2009 he and his colleagues observed isolated chains of water molecules arranged in pentagons on a copper surface. “Increasingly, I'm starting to think that in two dimensions, the pentagons are more stable than the hexagons,” he says.

Greg Kimmel, a physicist at Pacific Northwest National Laboratory in Richland, Washington, says the key to making the quasicrystal will be finding a surface that holds the water tight enough that there's no need for a second surface pushing down from above. That's how his team produced the hexagonal pattern, which was also predicted assuming a squeeze between two surfaces.

8. American Physical Society Meeting

# One Cool Way to Erase Information

While erasing information, a tiny system can sometimes generate less than the minimum amount of heat required by a principle of thermodynamics, physicists reported at the American Physical Society meeting.

When your computer erases a bit of information—a “0” or a “1” stored in some electronic switch—it must generate a minimum amount of heat. Or so says a principle of thermodynamics. However, a tiny system can sometimes dip below that limit while erasing information, physicists from Canada report.

The finding won't lead to infinitely cool computers, as on average the tiny bit still obeys the rule. But the exceptional events underscore the importance of fluctuations in very small systems and could point to a mechanism that cells might naturally exploit.

“That's fantastic!” says Eric Lutz, a theorist at the University of Augsburg in Germany, who with a colleague predicted 2 years ago how the thermodynamic limit might be evaded. “I tried to convince a number of people to do it.”

Disorder and information are the yin and yang of thermodynamics. Whenever a device destroys information, the disorder in the device and its surroundings must increase. That increase in disorder, or entropy, equals the release of heat. According to the so-called Landauer limit, erasing a bit at room temperature generates at least 0.003 billion-billionths of a joule of heat.

Strictly speaking, however, the Landauer limit applies to devices so large that the random jiggling of their atoms and molecules can be neglected when analyzing how they function. Things aren't so simple in very small widgets. To prove it, Yonggun Jun and John Bechhoefer of Simon Fraser University in Burnaby, Canada, fashioned a bit out of a 200-nanometer-wide polystyrene bead suspended in water and surrounded by four electrodes.

Using a microscope, they tracked the position and speed of the bead, which was so small that it was buffeted by the jiggling water molecules. The electrodes effectively produced a landscape of voltage resembling a valley with two low spots (see diagram). If the bead started out in the left dip, then the bit was set to 1. If the bead started in the right dip, then the bit was set to 0.

Starting with random settings, the researchers repeatedly erased the bit by lowering the voltage hump between the dips, tilting the whole valley to the right and restoring the original configuration, which forced the bead into the 0 spot. By tracking the bead, they could calculate the heat it generated, Jun reported. As Lutz predicted, during some cycles it generated less heat than the Landauer limit demands because the water molecules helped nudge it along.

What's it good for? One problem in applying the result is that it's impossible to predict when the bit will beat the Landauer limit. However, Bechhoefer speculates that cells might exploit such fluctuations to drive the “molecular motors” within them more efficiently.

Lutz, who was not at the meeting, notes that today's computer bits generate at least 100 times more heat than the Landauer limit. But if computer bits continue to shrink, then 20 or 30 years from now they may approach the limit, he says. Then hardware engineers will have to worry about thermal fluctuations spontaneously erasing information.

9. American Physical Society Meeting

# Snapshots From the Meeting

Snapshots from the American Physical Society meeting include lowering the energy of a vibrating widget enough to achieve the least motion allowed by quantum mechanics—the so-called ground state of motion—and a network model that demonstrates that if 10% of the members of a group hold an unshakable conviction, their view will eventually win out.

Bang the drum slowly. For the second time, physicists have lowered the energy of a vibrating widget enough to achieve the least motion allowed by quantum mechanics: the so-called ground state of motion. Researchers had previously cooled a tiny diving board–shaped gizmo to a few thousandths of a degree with a liquid helium refrigerator. This time, John Teufel of the National Institute of Standards and Technology in Boulder, Colorado, and colleagues borrowed a technique from laser physics and used microwaves to draw energy out of a tiny drumhead. The new widget oscillates 1000 times more slowly than the first quantum machine and should remain in a quantum state of motion much longer.

The power of true believers. If 10% of the members of a group hold an unshakable conviction, their view will eventually win out, according to a network model. Theorist Sameet Sreenivasan and colleagues at Rensselaer Polytechnic Institute in Troy, New York, find that the structure of the group's social connections does not greatly affect that threshold. Beyond the threshold, as the size of a group grows extremely large, only then do belief and disbelief coexist—assuming there are no unpersuadable disbelievers. Beate Schmittmann, a theorist at Virginia Polytechnic Institute and State University in Blacksburg, says the analysis underscores that “if you want to have a diverse society, it has to be a certain size.” The U.S. civil rights movement took off as the percentage of African Americans approached 10% of the population, Sreenivasan notes.

10. News

# Superconductivity's Smorgasbord of Insights: A Movable Feast

Resistance-free electric currents and their eventual explanation have influenced thinking far beyond solid state physics.

On 8 April 1911, physicist Heike Kamerlingh Onnes scrawled in a notebook misdated 1910, “Mercury practically zero.” Lost in a page of text, that cryptic phrase marks perhaps the most important discovery in the physics of materials. Kamerlingh Onnes meant that when he and his team at Leiden University in the Netherlands cooled a sample of mercury to within 3° of absolute zero, or 3 kelvin, its electrical resistance vanished so that current flowed through the metal with no energy to push it. Kamerlingh Onnes had discovered superconductivity.

The terse note gives no clue that the researcher had discovered one of the most bizarre tricks of nature. Physicists struggled for nearly 50 years to explain the phenomenon. When they finally did, the resulting theory would prove to be far more than just the explanation for one weird property of some metals. “It's much bigger than that, much bigger,” says Frank Wilczek, a particle physicist at the Massachusetts Institute of Technology (MIT) in Cambridge who shared the 2004 Nobel Prize in physics.

Physicists have applied the theory of superconductivity directly to nuclear matter, liquid helium, and ultracold atomic gases. Historically, insights from superconductivity convinced theorists of the importance of symmetries and the ways in which a physical system can muddle or “break” them. The concept of “spontaneous symmetry breaking” now undergirds theory in many fields, especially particle physics. “It was not a way that people were thinking, certainly not in elementary particle physics,” says Gordon Baym, a theorist at the University of Illinois, Urbana-Champaign. Superconductivity, he says, “changed the way people thought in different fields.”

That's not bad for an advance whose discoverer recorded it without so much as an exclamation point. Why did Kamerlingh Onnes make so little fuss over his observation? The short answer is that he was expecting the resistance of mercury to go to zero. (Technically, a material's intrinsic ability to impede current is its “resistivity” a sample's “resistance” depends on its size.) But he expected that for the wrong reasons, and at first he overlooked the all-important way mercury's resistance dropped to zero.

A few years earlier, Kamerlingh Onnes had embraced the opposite idea: that a metal at zero temperature would have infinite resistance. In 1902, British physicist William Thomson, a.k.a. Lord Kelvin, argued that at very low temperatures, the electrons in a metal would freeze to the ions in the material. So a metal's resistance should at first fall with temperature but then bottom out and climb again. In 1906, Kamerlingh Onnes cooled gold to 14 kelvin with liquid hydrogen. “A minimum of resistance seems not to be far off,” he reported to the Royal Netherlands Academy of Arts and Sciences.

Then he proved himself wrong. In 1910, Kamerlingh Onnes used liquid helium, which he had achieved 2 years earlier, to measure the resistance of platinum down to 1 kelvin. It did not climb but simply leveled off. Data for gold suggested that the eventual level depended on purity. Kamerlingh Onnes realized that those data jibed with the ideas of German theorist Paul Drude, who argued that the electrons in a metal form a gas and ping off the vibrating ions but not one another. If so, Kamerlingh Onnes theorized, a pure metal's resistance should vanish at zero temperature as its ions stop vibrating.

To prove it, he needed a pure metal. Mercury was the obvious choice; at room temperature it's a liquid that can be distilled. But Kamerlingh Onnes predicted that mercury's resistance should fall gradually. In experiments in May and October it abruptly vanished at 4.2 kelvin. “His model couldn't make sense of this jump,” says Dirk van Delft, a physicist and director of the Boerhaave Museum in Leiden. How does a metal's resistance just disappear?

## The hardest problem in solids

That mystery would stump the likes of Werner Heisenberg, Albert Einstein, and Richard Feynman. At the same time physicists explained myriad other properties of solids using quantum theory invented in the 1920s. “Superconductivity was driving people crazy,” says N. David Mermin, a theorist at Cornell University. “It was the one thing that quantum mechanics didn't seem to clear up.”

Whereas Drude had treated the electrons like marbles, physicists realized that they should be described by quantum waves that ripple with fixed energy and momentum. Also, electrons act like tiny tops that spin with angular momentum equal to half a fundamental measure called Planck's constant. Particles with 1/2, 3/2, 5/2 … units of angular momentum are known collectively as “fermions,” and a law of nature forbids two identical fermions from occupying the same quantum state.

So in a metal, the electrons stack into the lowest energy states. In an abstract three-dimensional “momentum space,” they fill a volume called the “Fermi sea.” Only the states near the surface aren't crammed full and effectively clogged, so only they contribute to the flow of electrons. At the same time, the solid's ions form an array known as a “lattice” whose vibrations are themselves quantum waves called “phonons.” Such theory explained many of the thermal, electrical, magnetic, and optical properties of solids. But not superconductivity.

Then in 1955, John Bardeen of the University of Illinois, Urbana-Champaign, set graduate student Robert Schrieffer and postdoc Leon Cooper on the problem. “I wasn't totally aware of the number of people who had failed, but you can't overestimate the arrogance of a young theoretical physicist,” says Cooper, now at Brown University. “If you're not arrogant at some level, you can't be a theoretical physicist.” Within 2 years the trio would show that superconductivity arises when the electrons in a metal form pairs that cannot be deflected without severing their bonds. That takes energy that's not available at low temperatures, so the pairs glide unimpeded.

The “BCS theory” came together in a sequence of keen insights. In 1950, experimenters found that the precise temperature at which mercury becomes a superconductor depends on the isotope of mercury and hence the mass of ions in it, suggesting the ions play a key role. In 1955, Bardeen's postdoc David Pines had shown that in a metal, two electrons, which ordinarily repel each other, could in fact attract each other by tugging on the ions and creating phonons that draw them together—much as two people lying on a waterbed create dips that draw them together (see diagram).

A year later, Cooper showed that the feeble pull could bind two electrons surfing on opposite sides of the Fermi sea. That wasn't obvious, Mermin says, because in empty space two particles must pull with a certain strength to pair. Cooper showed that the mere presence of the other electrons in the Fermi sea lowered the binding threshold for the two surfers to zero. “You can form pairs no matter how weak the interaction. That was Cooper's great insight,” Mermin says.

Explaining how all of the electrons in the Fermi sea could pair took more doing. The theorists labored for months to find a quantum state that would achieve that while meeting the restriction that the electrons avoid each other. In January 1957, Schrieffer guessed the state's mathematical form—while riding the New York City subway. “For what Bob did, there was no logical progression,” says Pines, now at the University of California, Davis. “It was a true ‘Aha!’ moment.”

The theorists now had a rigorous theory. At low temperatures, the electrons in some metals form a macroscopic quantum wave of intertwined pairs. The theory immediately yielded precise calculations and in 1972 earned Bardeen, Cooper, and Schrieffer a Nobel Prize.

Superconductivity still holds some mystery. In 1986, physicists discovered compounds of oxygen and copper that are superconducting at temperature up to 138 kelvin. Most physicists agree that phonons pull too weakly to explain that. But nobody doubts that in such “high-temperature superconductors” electrons pair, Pines says.

## Pairs, pairs everywhere

The BCS model was more than a one-trick pony. Bardeen, Cooper, and Schrieffer had based it on just two assumptions: that the particles are fermions and that they attract each other. So “it was obvious to all of us” that the theory would apply to other particles interacting through different forces, Cooper says.

First came applications to atomic nuclei. In the summer of 1957, before the BCS theory was published, Pines visited the University of Copenhagen. There he, Aage Bohr (Niels Bohr's son), and Ben Mottelson found they could explain long-standing puzzles, such as why nuclei with an even number of protons and even number of neutrons are particularly tightly bound. The protons and neutrons, also fermions, independently pair. The work helped Aage Bohr and Mottelson win shares of a Nobel Prize in 1975.

In 1959, Russian theorist Arkady Migdal predicted that a stellar corpse known as a neutron star—a ball of neutrons bound by its gravity—should contain pairs of neutrons that form a free-flowing “superfluid.” The model can explain why rapidly spinning, energy-emitting neutron stars called pulsars sometimes suddenly speed up as swirling whirlpools migrate en masse out of the star's superfluid interior. A pulsar then slows down again only gradually over months, says Illinois's Baym. “That's very hard to do without a superfluid,” he says.

The BCS theory may apply to neutron stars in a weirder way. Since the early 1970s, physicists have known that protons and neutrons consist of particles called quarks, also fermions. In the ultradense heart of a neutron star, the neutrons may dissolve into these constituents, which would then be free to pair. Scientists don't yet have experimental evidence for such pairing, says MIT's Wilczek. Still, he says, “I think the theory is pretty firm. It's just a question of whether the density gets high enough.”

Closer to home, physicists have applied the BCS theory to liquid helium. In the 1960s, theorists predicted that magnetic forces might cause atoms in liquid helium—specifically, the isotope helium-3, which is a fermion—to pair and flow without resistance at temperatures as high as 80 thousandths of a kelvin (millikelvin). But experimenters failed to spot pairing, and many wrote off the prediction as “a theorist's pipe dream,” says Douglas Osheroff of Stanford University in Palo Alto, California. Still, in 1972, as a graduate student at Cornell University, Osheroff spotted superfluid helium-3 at 2 millikelvin while trying to study magnetism in solid helium.

As predicted, pairing in helium-3 proved more complex than in an ordinary superconductor. In the superconductor, electrons with opposite spins join in the simplest way. In helium-3, paired atoms dance around each other and can spin in the same or opposite directions, producing three distinct superfluid “phases.” “Helium-3 allowed us to extend our understanding of the implications of the BCS theory,” says Osheroff, who shared the 1996 Nobel Prize with his advisers, Robert Richardson and David Lee.

Now, physicists are pushing the bounds of the BCS theory even further in experiments with ultracold gases of atoms. Since the 1990s, scientists have used laser light, radio waves, and magnetic fields to trap and cool atoms to billionths of a kelvin. In 2005, Deborah Jin and colleagues at JILA, a laboratory run jointly by the University of Colorado, Boulder, and the National Institute of Standards and Technology, used the technique to achieve a BCS state of paired atoms of potassium-40. Other groups quickly followed suit.

“Cold fermionic gases are ideal implementations of the ideas of BCS,” says Wolfgang Ketterle, an experimenter at MIT. And they let researchers test the theory in new ways. For example, they have resolved a debate about the relationship between the BCS state and another form of superfluidity. In 1938, physicists discovered that, when liquefied, the heavier isotope of helium, helium-4, formed a free-flowing superfluid below 2.2 kelvin. Helium-4 atoms have no spin and, like all particles with 0, 1, 2 … units of spin, are known as bosons. Unlike fermions, bosons prefer to crowd into a single quantum wave. So superfluidity in helium-4 stems from a phenomenon called Bose-Einstein condensation, in which particles cram into one quantum wave with no need for pairing. In 1995, physicists achieved Bose-Einstein condensation in cold-atom experiments, too.

Are the BCS mechanism and Bose-Einstein condensation related? Bardeen argued they weren't; others reasoned they were. Now, cold-atom experiments have settled the issue, as experimenters can tune a gas of fermions between a BCS state of overlapping pairs and a Bose-Einstein condensate of tightly bound bosonic molecules. “It's not black and white; Bose-Einstein condensation and BCS are two ends of a spectrum,” says Ketterle, who shared the 2001 Nobel Prize for achieving Bose-Einstein condensation in cold atoms.

The BCS theory has even seeded breakthroughs in particle physics. It underscored the importance of spontaneous symmetry breaking, which occurs when the basic interactions in a system possess a symmetry that's lost when the system settles into its lowest-energy state. For example, the forces between atoms in a magnet do not favor any one direction, yet all the atoms spontaneously align to minimize their energy. In a superconductor, the mere emergence of a quantum wave of electron pairs breaks a more abstract symmetry called “gauge invariance.”

Particle physicists have employed this idea to surprising ends. In 1961, Yoichiro Nambu of the University of Chicago used it to explain the masses of and interactions between hefty protons and neutrons and light particles called pions. Following the BCS theory closely, Nambu imagined that space is filled with a quantum wave of proton-antiproton pairs and neutron-antineutron pairs. The wave, or condensate, then pulls on a proton or neutron to give it mass, whereas the particles in the condensate form the pions. Such work won Nambu a share of the 2008 Nobel Prize.

In 1967, Steven Weinberg similarly argued that the electromagnetic force and the so-called weak nuclear force are two aspects of one thing. That's not obvious, as the electromagnetic force extends infinitely far and the weak force reaches only across the nucleus. Weinberg imagined a condensate that drags on the particles that convey the weak force, which he dubbed the W and Z bosons, to make them massive and short-ranged, whereas the photon, which conveys the electromagnetic force, remains massless. “Nambu was inspired by the ideas of superconductivity; I was inspired by Nambu,” says Weinberg, now at the University of Texas, Austin, who shared the Nobel Prize in 1979.

That theory proved on the mark in 1983 when the W and Z bosons were discovered with the predicted masses. And the simplest form of Weinberg's condensate would be a quantum field made up of the famed Higgs boson—the most sought-after prize in particle physics. The notion of symmetry breaking has become so essential to particle physics that “all attempts to go beyond the standard model play with symmetries, some of which are broken and some of which are not,” Weinberg says.

Superconductivity didn't entirely alter the views of one prominent physicist. Kamerlingh Onnes, who died in 1926 at age 72, never quite accepted that the resistance of a superconductor went completely to zero or fully embraced quantum mechanics, says Van Delft. “He was more or less frozen in the classical world and unable to go down that path,” Van Delft says. Still, after a century, one can only marvel at how far Kamerlingh Onnes's discovery has led others.

11. News

# Search for Majorana Fermions Nearing Success at Last?

1. Robert F. Service

Researchers think they are on the verge of discovering weird new particles that borrow a trick from superconductors and could give a big boost to quantum computers.

It happens over and over again in particle physics: Theorists predict the existence of a particle and then, sometime later, experimenters find it. Neutrons, positrons, neutrinos, pions, W and Z bosons, and other subatomic denizens all existed on paper before researchers spotted them, and right now physicists at particle accelerators at Fermi lab in Illinois and CERN near Geneva, Switzerland, are hoping that the long-sought Higgs boson will follow suit.

But this well-trodden path is virtually unknown among those scientists who study condensed matter, the stuff of our everyday world. In most materials, defects, impurities, and other imperfections generate too much noise for researchers to spot the vanishingly small signals of ephemeral particles. Even so, one of the most fleeting of all may be on the point of discovery, more than 70 years after it was first proposed.

In recent years, theoretical physicists have suggested that a handful of exotic materials could give rise to this never-before-seen type of particle, known as a Majorana fermion. Now experimental groups around the world are racing to spot it, using devices made in most cases with superconducting materials. And it looks as if some groups are closing in fast or may even have bagged Majoranas already.

“My bet is that if there is something to find, we'll see it within the next 5 years,” says David Goldhaber-Gordon, an experimental physicist at Stanford University in Palo Alto, California. Adds Michael Freedman, a mathematician turned theoretical physicist at Station Q, a collaborative research center between Microsoft and the University of California (UC), Santa Barbara: “This is the decade for Majorana fermions. I am extremely optimistic they will be found.”

If they exist, the novel particles are expected to display fundamentally new properties that could open a new window into the mysterious world of quantum mechanics. Their behavior is also expected to make Majorana fermions ideally suited to be stable bits of information in a quantum computer, something that has eluded researchers for decades. “It's the most exciting thing that has happened in fundamental physics in a long time,” says Leo Kouwenhoven, an experimental physicist at Delft University of Technology in the Netherlands.

## Mind-bending math

The notion of Majorana fermions arose as quantum mechanics took shape in the early 20th century as a way to explain the seemingly contradictory behavior of elementary particles, which behaved both as particles and as waves. Researchers had shown that elementary particles have intrinsic properties called charge and spin. Charge is just the familiar electrical property that makes electrons negative, protons positive, and atoms neutral. Spin is a type of rotational momentum related to the magnetic behavior of charged particles. Physicists at the time also knew that elementary particles come in two families: bosons, such as photons, and fermions, such as electrons, that have different groupings of spin.

In 1926, Austrian physicist Erwin Schrödinger came up with an equation that describes how quantum matter changes over time. Two years later, a young English physicist named Paul Dirac tweaked Schrödinger's equation to make it apply to fermions, such as electrons, that move at speeds near that of light. The expansion integrated quantum mechanics for the first time with Einstein's special theory of relativity.

Dirac's new equations also implied the existence of antimatter, matching each fundamental particle with an antiparticle that would annihilate it if the two should ever meet. To their surprise, physicists realized that certain particles, including some photons, could serve as their own antiparticles and annihilate themselves. But fermions weren't thought to be among them.

Then the story took a twist. In some cases, Dirac's equations produced results involving imaginary numbers, which some physicists considered inelegant. That's where a young, gifted Italian physicist named Ettore Majorana entered the picture. In 1937, Majorana modified Dirac's equations in a way that banished imaginary numbers but had its own mind-bending implications. Most importantly, it allowed for the existence of an entirely new class of fermions—the Majorana fermions—that, unlike traditional fermions, would be their own antiparticles.

## Closing in

Majorana never knew what became of his idea. A year after publishing his paper reworking Dirac's equation, he disappeared mysteriously and was never heard from again. His own favorite candidate for possible Majorana particles, neutrinos, remains a tantalizing possibility today. Massive detectors, such as one deep in the Apennine Mountains of the Abruzzo region in Italy, have been running for years in the hope of spotting a so-called neutrino double-beta decay, which could clinch neutrinos' Majorana status.

Now condensed-matter physicists are getting in on the hunt and relishing the chance to discover new particles. “That would be pretty cool if we could out-CERN CERN,” says Charles Marcus, a condensed-matter physicist at Harvard University.

It won't be easy. Majorana fermions can't exist just anywhere. To harbor them in all but a few esoteric cases, the material must allow the particles to behave as if they had no spin. It must also force electrons in the materials to pair up as they do in superconductors, in which the pairing allows the electrons to surf through the material without electrical resistance. But whereas superconductors can “glue” electrons together in several different ways, Majoranas can use only one type: an attraction known as chiral p-wave symmetry, which causes electrons with the same spin to pair up. Finally, the electron pairs will hold together only if the material exists as either two-dimensional sheets or one-dimensional wires.

All known fermions have spin, and only a handful of materials meet the other criteria. So for decades the prospects for finding Majorana fermions in materials looked slim at best. “It was a very hard sounding problem,” says Jason Alicea, a theoretical physicist at UC Irvine.

Help came from physicists who study how electrons behave in groups. Usually, electrons are like pedestrians on a New York City street, traveling every which way. But like a flash mob that breaks out into a coordinated song and dance routine, groups of electrons can correlate their behavior due to simple interactions between the spins and charges of neighbors. Those interactions can give rise to emergent particles—also called quasiparticles—with characteristics very different from the simple, well-known properties of single electrons, notes Amir Yacoby, an experimental physicist at Harvard. “When you put many together, the collective behavior is fundamentally different,” he says.

In some semiconductors, for example, electrons moving through strong magnetic fields can jointly create quasiparticles with a charge less than that of an electron—a phenomenon physicists discovered in 1982 and called the fractional quantum Hall effect (FQHE). Initially, those fractional charges all had odd-numbered denominators, such as 1/3 or 3/7. In 1987, however, researchers led by Robert Willett at Bell Laboratories in Murray Hill, New Jersey, discovered an FQHE quasiparticle with a 5/2 charge in a semiconductor crystal of gallium arsenide. Four years later, a pair of theorists at Yale University suggested that this 5/2 quasiparticle could be a Majorana fermion.

Condensed-matter physicists sensed a discovery in the wings. And now a handful of groups appear to be close to nailing it down. Two years ago Willett, a physicist working with crystal growth experts Loren Pfeiffer and Ken West of Princeton University, reported measuring signals like those expected from Majorana fermions in the 5/2 FQHE state. They had built a tiny device known as an interferometer on a gallium arsenide chip tailored to generate the 5/2 state. Interferometers measure the interference pattern of electromagnetic waves, often visualized as a set of light and dark stripes produced by interfering beams of photons. Willett's team looked at the interference patterns produced by emergent particles in the 5/2 state.

Theory suggested that Majoranas should appear to have a charge 1/4 that of an electron (e/4): the value researchers see in electrical readings from chiral p-type super conductors. Also in theory, this e/4 signal should oscillate and should be interrupted by patches without any electrical activity at all. In a pair of papers published in 2009 and 2010, Willett and his colleagues reported that they had seen the oscillating e/4 signal. But in places where the bare patches should be, they saw instead an oscillating charge of e/2.

Freedman calls Willett's work “beautiful physics.” But he and others say the data remain noisy. “Bob is shaking the right tree,” Marcus says. “Something fell out of the tree. But is it an apple?” More results are needed, Marcus says.

Yacoby agrees. He, too, has teamed up with Pfeiffer and West and also appears to be closing in on Majoranas in the 5/2 state. Using a layered gallium arsenide chip but a different monitoring method, the group reported in Nature earlier this year that they had also spotted an e/4 signature. That's promising, but more evidence is needed, Yacoby says: e/4 charge alone is “necessary but not sufficient” to prove the existence of the 5/2 state and the presence of Majorana fermions.

## Hot pursuit

The search for Majorana particles is heating up in other solid-state systems as well. In 2008, theorists Liang Fu and Charles Kane of the University of Pennsylvania argued that depositing a widely available type of superconductor known as an s-wave superconductor atop a novel material known as a topological insulator would be a simpler way to make them appear.

Topological insulators conduct electricity only on their surfaces but act as insulators throughout their bulk; shaped into two-dimensional sheets, they conduct only at their edges. Fu and Kane's calculations suggested that the interactions between the two materials would force quasiparticles to form with p-wave glue and behave as if they had no spin—the exact recipe needed to make Majoranas. “That idea jump-started the field,” says Marcel Franz, a theoretical physicist at the University of British Columbia, Vancouver, in Canada.

Sankar Das Sarma, a theorist at the University of Maryland, College Park, and others ran with Fu and Kane's ideas. They soon discovered that topological insulators aren't necessary; a particular type of semiconducting nanowire placed atop an s-wave superconductor would work just as well. The key feature needed in the nanowires was a property known as high spin-orbit coupling, in which an electron's orbital motion as it circles an atomic nucleus is strongly linked to its spin.

As fortune would have it, Kouwenhoven's lab at Delft had experimented with exactly that setup. The group members were experts at making pristine nanowires from a semiconductor alloy called indium arsenide (InAs), whose ability to create near-perfect electrical contacts with certain metals makes it an excellent test bed for studying how electrical charges move between different materials.

In 2005, Kouwenhoven's postdoc Silvano De Franceschi had placed an InAs nanowire on top of a superconductor and measured how currents flowed through the two materials. The results fell in line with what was expected, so the researchers published their results and moved on. “We thought that was the end of the work,” Kouwenhoven says.

Das Sarma's paper, along with another by a team led by Yuval Oreg of the Weizmann Institute in Rehovot, Israel, made them look again. Because InAs has strong spin-orbit coupling, Kouwenhoven realized, the new work suggested that a setup like De Franceschi's ought to be a fruitful hunting ground for Majorana fermions. “It put us at the center of attention,” Kouwenhoven says.

Kouwenhoven's team raced to recreate its earlier nanowire experiments and carefully added a strong magnetic field. Too high a field would destroy a material's ability to superconduct. But one just below that level, according to Fu and Kane's theory, should produce Majorana fermions at the ends of the wires.

If the exotic quasiparticles are there, theory suggests physicists should be able to detect them by measuring (with, say, the metal tip of a scanning tunneling microscope) how readily they conduct electrons. Normal particles have a small electrical resistance, whereas Majoranas should have none. Physicists should also see unique electronic signatures when Majorana fermions are created in the heart of so-called superconducting quantum interference devices.

Kouwenhoven says he and his group are looking for such signatures. They haven't spotted them yet in InAs and are now testing nano wires with higher spin-orbit coupling, eager for the right mix of properties. “We hope to hit a sweet spot,” Kouwenhoven says.

## A Majorana computer?

They aren't alone. Alicea, Franz, and others who are plugged into the field say proposals from other academic groups are pouring in to spot Majorana fermions in nanowire setups and a handful of other promising solid state systems. If any of the techniques succeed, researchers will gain the ability to play with particles with quantum behavior unlike anything else ever seen.

Most notably, Majorana fermions are expected to break new ground in an area called quantum-mechanical exchange statistics. The phrase describes how the quantum state of two identical particles changes as the particles change places. All known bosons and fermions behave in a way that, in effect, erases the history of the motions; Majorana particles, by contrast, will leave tracks.

To visualize how this works for bosons, Marcus says, imagine holding two pegs, each with a string dangling from it (see figure, top of page). Move the two pegs around each other to trade places and the strings follow. Swap them again and the pegs and strings end up back where they started, leaving no clue as to their earlier path. Things are slightly more complex with standard fermions, but after two exchanges they return to their original state as well.

Not so with Majorana fermions. In their case, it's as if each string starts out taped to the floor below its peg. If you move the two pegs past each other in the same direction, the strings will wind around each other—and anyone who examines them can tell whether the pegs went clockwise or counterclockwise. Similarly, when Majoranas move, they change their quantum state in a manner that reflects the path they've taken. The result, researchers say, could be an entirely new way of studying the subtleties of quantum mechanics. “This is a really big fish out there waiting for us to find it,” Alicea says.

It may have practical applications, too. Several years ago, Freedman and his colleagues reasoned that such unique exchange behavior could be used to encode information. In that case, Majorana fermions could serve as quantum bits of data, or qubits, in a quantum computer. Qubits are quantum analogs to the binary 0s and 1s in your standard desktop, laptop, or cell phone. Unlike conventional bits, qubits store information not as either a 0 or a 1, but as a superposition of both 0 and 1. As a result, a quantum computer containing only a few hundred qubits can carry out certain kinds of calculations faster than the most powerful supercomputers.

Real-world qubits, however, are extremely delicate. The slightest hint of heat or other perturbation can destroy the superposition, rendering them useless. That fragility has made complex quantum computers difficult to create. The braided Majorana particles, by contrast, are expected to shrug off such environmental wiggles. “In theory, they should be more robust qubits,” Franz says.

Freedman, Alicea, and others have designed systems of nanowires and other devices to manipulate Majorana fermions and encode information. But few physicists expect to see working models anytime soon. That's just fine with Marcus, who says he looks forward to the challenge. The hunt for Majorana fermions “is not a Holy Grail problem, where once you get it there's nothing left to do,” he says. Researchers who discover them will still face the arduous task of learning to control them well enough to build a quantum computer. “I'll see you in 20 years when we get all this to work,” Marcus says.