Tuesday, October 23rd, 2012

Time	Event
4:00a	Tiny pores in graphene could give rise to membranes Much has been made of graphene’s exceptional qualities, from its ability to conduct heat and electricity better than any other material to its unparalleled strength: Worked into a composite material, graphene can repel bullets better than Kevlar. Previous research has also shown that pristine graphene — a microscopic sheet of carbon atoms arranged in a honeycomb pattern — is among the most impermeable materials ever discovered, making the substance ideal as a barrier film. But the material may not be as impenetrable as scientists have thought. By engineering relatively large membranes from single sheets of graphene grown by chemical vapor deposition, researchers from MIT, Oak Ridge National Laboratory (ORNL) and elsewhere have found that the material bears intrinsic defects, or holes in its atom-sized armor. In experiments, the researchers found that small molecules like salts passed easily through a graphene membrane’s tiny pores, while larger molecules were unable to penetrate. The results, the researchers say, point not to a flaw in graphene, but to the possibility of promising applications, such as membranes that filter microscopic contaminants from water, or that separate specific types of molecules from biological samples. “No one has looked for holes in graphene before,” says Rohit Karnik, associate professor of mechanical engineering at MIT. “There’s a lot of chemical methods that can be used to modify these pores, so it’s a platform technology for a new class of membranes.” Karnik and his colleagues, including researchers from the Indian Institute of Technology and King Fahd University of Petroleum and Minerals, have published their results in the journal ACS Nano. Karnik worked with MIT graduate student Sean O’Hern to look for materials “that could lead to not just incremental changes, but substantial leaps in terms of the way membranes perform.” In particular, the team cast around for materials with two key attributes, high flux and tunability: that is, membranes that quickly filter fluids, but are also easily tailored to let certain molecules through while trapping others. The group settled on graphene, in part because of its extremely thin structure and its strength: A sheet of graphene is as thin as a single atom, but strong enough to let high volumes of fluids through without shredding apart. The team set out to engineer a membrane spanning 25 square millimeters — a surface area that is large by graphene standards, holding about a quadrillion carbon atoms. They used graphene synthesized by chemical vapor deposition, borrowing on expertise from the research group of Jing Kong, the ITT Career Development Associate Professor of Electrical Engineering at MIT. The team then developed techniques to transfer the graphene sheet to a polycarbonate substrate dotted with holes. Once the researchers successfully transferred the graphene, they began to experiment with the resulting membrane, exposing it to flowing water containing molecules of varying sizes. They theorized that if graphene were indeed impermeable, the molecules would be blocked from flowing across. However, experiments showed otherwise, as researchers observed salts flowing through the membrane. As another test, the team exposed a copper foil with graphene grown on it to a chemical agent that dissolves copper. Instead of protecting the metal, graphene let the agent through, corroding the underlying copper. To test the size of the pores within graphene, the group attempted to filter water with larger molecules. It appeared that there was a limit to the size of the pores, as larger molecules were unable to pass through the membrane. As a final experiment, Karnik and O’Hern observed the actual holes in the graphene membrane, looking at the material through a high-powered electron microscope at ORNL in collaboration with Juan-Carlos Idrobo. They found that pores ranged in size from about 1 to 12 nanometers — just wide enough to selectively let some small molecules through. “Right now we know from this characterization how the graphene behaves, and what kind of intrinsic pores it has,” Karnik says. “In some sense it’s the first step to practically realizing graphene-based membranes.” Karnik adds that a near-term application for such membranes may include a portable sensor in which a layer of graphene “could shield the sensor from the environment,” letting through only a molecule or contaminant of interest. Another use may be in drug delivery, with graphene, dotted with pores of a determined size, delivering therapies in a controlled release. “We’re right now in the process of transferring more graphene to different substrates and making holes of our own, making a viable membrane for water filtration,” O’Hern says. Scott Bunch, an assistant professor of mechanical engineering at the University of Colorado, says the group’s results are the first demonstration that graphene bears defects. The membrane developed by the group “has the potential to be a revolutionary membrane” that separates particles at the molecular scale. “The issue that now needs to be addressed is whether one can discriminate between smaller molecules,” Bunch says. “Once this happens, graphene membranes will eventually live up to the truly remarkable properties that they promise.” Other researchers involved in the work are Cameron Stewart, Michael Boutilier, Sreekar Bhaviripudi, Sarit Das, Tahar Laoui and Muataz Atieh. This work was funded by the King Fahd University of Petroleum and Minerals through the Center for Clean Water and Clean Energy at MIT and KFUPM, and was also supported by the ORNL ShaRE program. (Comment on this)
3:03p	Scholars ponder better ways to elect a president The upcoming presidential election appears to be so close that either President Barack Obama or his challenger, former Massachusetts Gov. Mitt Romney, could lose the popular vote, yet still gain the White House by winning a majority of Electoral College votes. With this in mind, should the United States choose its president through the Electoral College, as it does now, or through another method, such as a national popular vote? And how should people be allowed to register and vote? Those questions formed the dual focal points of a spirited conference on election systems held at MIT on Friday, Oct. 19. “One person, one vote should be the norm of a modern democracy,” said John Koza, the Stanford University computer scientist who is a leader of the group National Popular Vote, which aims to elect presidents through a popular vote. By contrast, as Koza pointed out, the structure of the Electoral College, which allocates each state’s electoral votes according to its number of congressmen, including senators, means that voters in smaller states have far more electors per capita than do voters in big states. Koza also criticized the Electoral College system for creating campaigns in which candidates fixate on issues that most directly concern swing-state voters. On average, Koza said, “We have a system that’s ignoring four out of five voters in the country.” Backed in part by the efforts of the National Popular Vote group, eight states — California, Hawaii, Illinois, Maryland, Massachusetts, New Jersey, Vermont and Washington — and the District of Columbia have all passed into law the National Popular Vote Interstate Compact. In the terms of that agreement, those states have pledged to award their Electoral College votes to the winner of the national popular vote, as soon as the number of states joining the compact collectively hold at least 270 Electoral College votes, the minimum number needed to win the presidency. The popular-vote movement had both supporters and opponents among the panelists. The concept “conforms with the popularly held conception of fairness,” said Alexander Keyssar, a professor at Harvard University’s Kennedy School of Government, who supports the idea. However, Alexander Belenky, a professor at the National Research University Higher School of Economics in Moscow, said the idea ignored the “will of the states,” and suggested that if any changes were necessary, state officials should examine ways of allotting electoral votes proportionately. “Almost any state can be changed into a battleground by its state legislators,” Belenky said. Other scholars challenged that reasoning, however. “States do not have a will,” said Jack Rakove, a historian and legal scholar from Stanford. That said, Rakove also thinks that even if enough states pass legislation to join the compact, it will face heavy legal and legislative obstacles. While popular-vote advocates cite a 1978 U.S. Supreme Court ruling on the steel industry as a precedent for the legality of interstate compacts, Rakove suggested that ruling was unlikely to be a robust precedent. “The idea that this is going to escape a challenge in Congress is utterly [wrong],” Rakove said. “Whatever the prior jurisprudence is, it’s not going to affect this Supreme Court or any Supreme Court. This is clearly a change of constitutional dimensions, done by compact or not.” Pick one candidate, or rank them? The event — titled “Does the Current Presidential Election System Serve America Well?” — featured almost 20 speakers, and took place in MIT’s Bartos Theater. Several speakers discussed alternate ways that voters could choose candidates on ballots: Instead of just selecting one candidate, for instance, voters could rank them, or rate them on a numerical scale. Eric Maskin, a Nobel Prize-winning economist from Harvard, gave a keynote talk exploring ways to avoid elections — such as the presidential elections from 1992 through 2000 — in which the winner receives only a plurality of the vote, and not an outright majority. One option Maskin finds valuable, attributed to the French Enlightenment thinker Condorcet, asks voters to rank candidates, and then selects as a winner the one candidate who would beat all others in the rankings on a head-to-head basis. “Voters under the current system are really not providing enough information about what they really want,” Maskin said.   Arnold Barnett, the George Eastman Professor of Management Science at the MIT Sloan School of Management, presented a modification of the U.S. system in which Electoral College votes would be awarded by multiplying each state’s popular-vote percentage by its number of electoral votes — which he said would also give candidates incentive to campaign nationally. “The candidates would rediscover the way to San Jose — and to Fort Worth, and to Brooklyn, and to Chicago,” Barnett said. The wave of voter ID laws Two of the seven panels at the event focused on the wave of voter-registration laws that have been passed in 33 states across the country. Many of them require voters to present photo identification at the polling place; some studies have shown that around 10 percent of the population does not possess such identification. The topic is the subject of bitter political dispute at the moment; Republicans have largely enacted the laws, which Democrats charge are intended to keep Democratic-leaning citizens from voting. Hans Von Spakovsky of the Heritage Foundation, a former Justice Department official and a central figure in the effort to enact voter-identification laws, presented a summary of the current legal status of the laws in many key states. In Pennsylvania and South Carolina, for instance, judges have ruled that the statutes are legal, but cannot be implemented in time to apply to the 2012 election. The U.S. Supreme Court, in 2008, also upheld a 2007 Indiana law requiring photo ID. According to Von Spakovsky, the Indiana law “has not prevented anyone from voting in the state,” and the lack of legal challenges to the way the state’s law has been administered in practice demonstrates its validity. Vikram Amar, a law professor at the University of California at Davis, said there was an “uncomfortable partisan pattern” to the voter-registration laws — even accounting for the fact that, as Von Spakovsky pointed out, liberal U.S. Supreme Court justice John Paul Stevens voted to uphold the Indiana law. Amar also suggested that it would take time to assess the empirical evidence about the effect the laws are having on particular blocs of voters. The conference was sponsored by the MIT Sociotechnical Systems Research Center and MIT's Center for Engineering Systems Fundamentals. (Comment on this)

<< Previous Day

2012/10/23 [Calendar]

Next Day >>